European Journal of Physical and Rehabilitation

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, CHARCOT-MARIE-TOOTH 1A EVALUATION MORI cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. 10-MWT 10-MWT 10-MWT 6-MWT BBS SPPB 10-MWT 6-MWT 6-MWT Walk-12 SPPB Walk-12 Figure 1.—Primary outcome measures significant correlations. Table IV.—Correlations between SF36 and motor tests. SF36 components 10MWT Walk12 6MWT BBS SPPB Foot dorsal Foot plantar Physical functioning -0.55 -0.60 0.51 0.48 0.54 0.32 0.29 Role, Physical P=0.0017* P<0.001* P<0.001* P=0.0013* P<0.001* P=0.05* P=0.11 Bodily pain General health -0.41 -0.56 0.25 0.27 0.41 0.23 0.18 Vitality P=0.0085* P<0.001* P=0.13 P=0.083 P=0.007* P=0.17 P=0.37 Social functioning Role emotional -0.16 -0.59 0.14 0.19 0.33 0.18 0.16 Mental health P=0.29 P<0.001* P=0.41 P=0.26 P=0.017* P=0.26 P=0.40 -0.35 -0.48 0.24 0.29 0.34 0.33 0.19 P=0.019* P<0.001* P=0.14 P=0.051 P=0.017* P=0.05* P=0.37 -0.27 -0.50 0.18 0.16 0.30 0.20 0.06 P=0.07 P<0.001* P=0.31 P=0.29 P=0.039* P=0.24 P=0.76 -0.29 -0.58 0.32 0.30 0.38 0.07 0.06 P=0.062 P<0.001* P=0.049* P=0.051* P=0.017* P=0.67 P=0.76 -0.36 -0.45 0.15 0.16 0.24 0.15 0.17 P=0.019* P=0.0017* P=0.39 P=0.29 P=0.097 P=0.35 P=0.37 -0.25 -0.25 0.09 0.15 0.20 0.04 0.03 P=0.097 P=0.097 P=0.56 P=0.32 P=0.17 P=0.76 P=0.90 P values for significance of correlation coefficients were corrected for multiple tests according to the false-discovery rate approach. *Statistically significant correlation. Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 51

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, MORI CHARCOT-MARIE-TOOTH 1A EVALUATION cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. of SF36 and with the SPPB, which had a moderate cor- balance in the quantitative evaluation of functional impair- relation with all subscales except for Role Emotional and ment in CMT1A patients. Mental Health. The Mental Health parameter showed no correlation at all. The negative correlation between 6MWT and 10MWT suggests that better spatial performance corresponds to bet- Main component analysis ter temporal performance. In fact, whereas the 6MWT is a test that evaluates the distance covered in a fixed time, the A total of three main components were identified. Globally 10MWT evaluates the time taken to tread a specific distance. they explained 68% of total variance. In Supplementary Digital Material 1 (Supplementary Figure 1), the screen The negative correlations of Walk12 with other motor plot to define the number of main components was re- test, particularly with 6MWT, BBS and SPPB, attests that ported. The first main component explained 33.6% of the subjects who do not experience motor skills limitations whole variance and was represented by almost all SF36 have improved performances in balance tests and gait pa- subscales except for physical activity. The second com- rameters, as proposed in previous studies.43 The low-mild ponent, accounting for 17.9% of total variance, was rep- correlation of the Walk12 test with the duration of disease resented by Walk12 and 10MWT with a negative correla- may reflect the degree of impairment: the longer the dura- tion (i.e. increasing values of these outcomes correspond tion of the disease is, the higher are patient’s impairment to a decrease of this component) and by 6MWT and BBS and perception of limitations in physical activities. with a positive correlation. The third component instead, accounting for 16.5% of total variance, showed a negative Relatively to balance assessments, it is not surprising correlation with CMTNS, while a positive correlation with that SPPB and BBS interrelate. Moreover, the correlation SPPB and the foot flexions. observed between SPPB and all the other motor assess- ments testifies the validity of this OM and suggests its use Discussion in future trials on CMT disease. Particularly, to the best of our knowledge, SPPB was never used before as OM in We analyzed the baseline data of a large population of subjects affected by CMT. We suggest its use because of CMT1A patients collected in the context of a randomized, its rapidity to assess balance and gait disorders in clini- single blinded, rehabilitative trial (the TreSPE), to further cal settings. BBS is a largely used scale in experimental address the issue of OM in ambulatory patients affected paradigms, but its administration can take between 6 and by hereditary neuropathies and to define the best measures 30 minutes. Both SPPB and BBS also correlate with all to be used in clinical trials, especially rehabilitative ones. the other OMs, more precisely with walking tests. This proves that patients with better balance control walk at The mean score of 10 obtained by our patients at the higher speed and perceive less limitations in their physical CMTNS, testifies that they had to be included in the mild activities or motor skills. As we previously observed for to moderate category.40 This result fits with the intent of the BBS, they strongly correlate with disease disability and study to treat a group of patients who are able to perform with distal muscular weakness.45 an intense aerobic training. In our sample, the CMTNS and the disease duration showed only low-mild correla- This can be reconducted to the fact that ankle stabil- tion with SPPB. This data may suggest that the endurance ity depends upon different factors such as anatomy integ- parameters are not well assessed by the CMTNS or may be rity, muscle strength and proprioception. Hence the risk of due to a selection bias towards the less affected subjects. falls is not only due to balance disorders, but also to distal weakness. This observation is in line with a recent study, The core OMs in our study were walking tests. Not which applied a protocol of open-ended questions and sub- surprisingly, 6MWT, 10MWT and Walk12 significantly jective scales to show that fatigue, poor balance, muscle correlated with foot strength, especially with foot dorsal weakness and pain are the most important limitations to flexion, even if correlation were frequently < 0.40. Our pa- physical activity in patients affected by CMT.46 Further- tients showed better performances at the 6MWT compared more, studies performed by comparing subjects affected to literature.43, 44 The most important observation of this by CMT and healthy subjects demonstrated that patients study is, however, the significant and moderately strong affected by CMT show altered balance and impaired pos- correlation between 6MWT, 10MWT, SPPB and BBS. tural stabilization that depend from proprioceptive deficits This result unequivocally demonstrates the importance of and distal muscle weakness.47, 48 In our study, as well as in these tests in order to investigate the walking ability and previous works, all motor tests mainly correlate with foot dorsal flexion strength. The explanation of this feature is 52 European Journal of Physical and Rehabilitation Medicine February 2019

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, CHARCOT-MARIE-TOOTH 1A EVALUATION MORI cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. that in patients affected by CMT lower limbs dorsal flexor Limitations of the study muscles are usually weaker than plantiflexor muscles. The main limitation of this study is that, due to the strict The issue of OMs in clinical trials for CMT has been ad- inclusion criteria, patients were recruited between subjects dressed by two Cochrane reviews, one focused on random- that could deal with the protocol. This led to exclude most ized trials for foot drop in neuromuscular diseases11 and of the compromised subjects. Another limitation is that the other on treatment for patients affected by CMT.4 Al- this is not a longitudinal study. Nonetheless, all data of though these reviews and other studies suggest that walk- the TreSPE study will soon be presented in another paper. ing tests may be sensitive, reliable and practical measures Finally, we do not show control group data, as the aim of to assess disability in CMT, there still is no consensus on the main study was to evaluate two rehabilitation proto- the best OM to evaluate the efficacy of rehabilitative or cols in a population of CMT1A subjects in order to verify pharmacological trials in these patients.12, 18, 49-51 Accord- whether the use of treadmill could lead to an improvement ingly, additional studies were performed and showed that compared to standard treatment. Therefore, the creation of three clinical OMs (the 10MWT, 9 hole-peg test, and foot a control group was not necessary to this purpose. More- dorsal flexion dynamometry) further improve discrimina- over, in a previous study based on different OMs, we dem- tion between severely and mildly affected patients and that onstrated that CMT1A patients had worse walking and the 6MWT may be a valid and reliable OM for patients balance performances compared to a control group.8 affected by CMT.6, 13 As walking test require a long time to be performed, despite their usefulness we suggest that Conclusions the SPPB is the most advisable outcome measure since it includes gait and balance evaluations. In conclusion, we confirm that the use of further clinical OMs in addition to the CMTNS is important to quantify As regards the subjective evaluation of the QoL, it is the clinical impairment of subjects affected by CMT1A clear that patients with a rare disease that leads to disability and offers different advantages in studies focused on re- in walking demonstrate slightly lower scores than normal at habilitation. Among the different OMs, we recognize that the SF36 since the disease results in a compromised motor 6MWT, 10MWT, BBS and SPPB are those that better performance. This feature is confirmed by a previous study evaluate the functional impairment and that provide more of our group.52 In this study, QoL investigated in a CMT1A information on motor skills, walking ability and balance. population appeared to correlate best with the ability to However, since BBS and walking tests are scales that en- walk autonomously and with axonal loss, therefore suggest- tail a long time to be carried out, we highlight the fact that ing that improvement of gait and of the ability to ambulate SPPB is a fast and simple test to assess balance in CMT1A autonomously may be the most important factor to improve patients. patients’ QoL. Other recent studies investigated the QoL in subjects affected by CMT. In 2013 Taniguchi submitted a References population of CMT1A patients to a survey with the SF36, reporting that the majority QoL impairments occurred in 1.  Casasnovas C, Cano LM, Albertí A, Céspedes M, Rigo G. Charcot- the social and emotional domains, although also functional Marie-tooth disease. Foot Ankle Spec 2008;1:350–4. capacity tended to be significantly affected.53 In 2014 Cor- 2.  Schenone A, Nobbio L, Monti Bragadin M, Ursino G, Grandis M. In- deiro et al. revealed that subjects affected by CMT exhibit herited neuropathies. Curr Treat Options Neurol 2011;13:160–79. an increased trend towards depressive symptoms compared 3.  Tozza S, Aceto MG, Pisciotta C, Bruzzese D, Iodice R, Santoro L, et with the general population and that they are exposed to a al. Postural instability in Charcot-Marie-Tooth 1A disease. Gait Posture higher risk of reduced QoL and significant sleep impair- 2016;49:353–7. ment.54 Still in 2014 Johnson et al. investigated the QoL in 4.  Boffeli TJ, Tabatt JA. Minimally Invasive Early Operative Treatment a large sample of subjects affected by CMT, highlighting of Progressive Foot and Ankle Deformity Associated With Charcot-Ma- that the most disabling symptoms are foot and ankle weak- rie-Tooth Disease. J Foot Ankle Surg 2015;54:701–8. ness also whereas impaired balance and that the prevalence 5.  Young P, De Jonghe P, Stögbauer F, Butterfass-Bahloul T. Treat- and impact of these symptoms gradually increases with age ment for Charcot-Marie-Tooth disease. Cochrane Database Syst Rev and duration of symptoms.55 Furthermore, in our study the 2008;2008:CD006052. Physical Functioning component of the SF36 correlates 6.  Padua L, Pazzaglia C, Schenone A, Ferraro F, Biroli A, Esposito C, et with every motor test and this is not surprising since they al. Rehabilitation for Charcot Marie tooth: a survey study of patients and consider and investigate motor functions. familiar/caregiver perspective and perception of efficacy and needs. Eur J Phys Rehabil Med 2014;50:25–30. 7.  Sman AD, Hackett D, Fiatarone Singh M, Fornusek C, Menezes MP, Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 53

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, MORI CHARCOT-MARIE-TOOTH 1A EVALUATION cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. Burns J. Systematic review of exercise for Charcot-Marie-Tooth disease. 26.  Bladh S, Nilsson MH, Hariz GM, Westergren A, Hobart J, Hagell P. J Peripher Nerv Syst 2015;20:347–62. Psychometric performance of a generic walking scale (Walk-12G) in mul- 8.  Maggi G, Monti Bragadin M, Padua L, Fiorina E, Bellone E, Grandis tiple sclerosis and Parkinson’s disease. J Neurol 2012;259:729–38. M, et al. Outcome measures and rehabilitation treatment in patients af- 27.  Brogårdh C, Flansbjer UB, Lexell J. Self-reported walking ability in fected by Charcot-Marie-Tooth neuropathy: a pilot study. Am J Phys Med persons with chronic stroke and the relationship with gait performance Rehabil 2011;90:628–37. tests. PM R 2012;4:734–8. 9.  Toots A, Littbrand H, Lindelöf N, Wiklund R, Holmberg H, Nordström 28.  Pagliano E, Moroni I, Baranello G, Magro A, Marchi A, Bulgh- P, et al. Effects of a High-Intensity Functional Exercise Program on De- eroni S, et al. Outcome measures for Charcot-Marie-Tooth disease: pendence in Activities of Daily Living and Balance in Older Adults with clinical and neurofunctional assessment in children. J Peripher Nerv Syst Dementia. J Am Geriatr Soc 2016;64:55–64. 2011;16:237–42. 10.  Khan F, Amatya B. Rehabilitation in Multiple Sclerosis: A Systematic 29.  Paltamaa J, Sarasoja T, Leskinen E, Wikström J, Mälkiä E. Measures Review of Systematic Reviews. Arch Phys Med Rehabil 2017;98:353–67. of physical functioning predict self-reported performance in self-care, 11.  Sackley C, Disler PB, Turner-Stokes L, Wade DT. Rehabilitation in- mobility, and domestic life in ambulatory persons with multiple sclerosis. terventions for foot drop in neuromuscular disease. Cochrane Database Arch Phys Med Rehabil 2007;88:1649–57. Syst Rev 2007;(2):CD003908. 30.  Qutubuddin AA, Pegg PO, Cifu DX, Brown R, McNamee S, Carne W. 12.  Reilly MM, Shy ME, Muntoni F, Pareyson D. 168th ENMC Interna- Validating the Berg Balance Scale for patients with Parkinson’s disease: a tional Workshop: outcome measures and clinical trials in Charcot-Marie- key to rehabilitation evaluation. Arch Phys Med Rehabil 2005;86:789–92. Tooth disease (CMT). Neuromuscul Disord 2010;20:839–46. 31.  Blum L, Korner-Bitensky N. Usefulness of the Berg Balance Scale 13.  Mannil M, Solari A, Leha A, Pelayo-Negro AL, Berciano J, Schlot- in stroke rehabilitation: a systematic review. Phys Ther 2008;88:559–66. ter-Weigel B, et al.; CMT-TRIAAL/CMT-TRAUK Group. Selected items 32.  Matjacić Z, Zupan A. Effects of dynamic balance training during from the Charcot-Marie-Tooth (CMT) Neuropathy Score and secondary standing and stepping in patients with hereditary sensory motor neuropa- clinical outcome measures serve as sensitive clinical markers of disease thy. Disabil Rehabil 2006;28:1455–9. severity in CMT1A patients. Neuromuscul Disord 2014;24:1003–17. 33.  Rose KJ, Hiller CE, Mandarakas M, Raymond J, Refshauge K, Burns 14.  Piscosquito G, Reilly MM, Schenone A, Fabrizi GM, Cavallaro T, J. Correlates of functional ankle instability in children and adolescents Santoro L, et al.; CMT-TRIAAL Group; CMT-TRAUK Group. Respon- with Charcot-Marie-Tooth disease. J Foot Ankle Res 2015;8:61. siveness of clinical outcome measures in Charcot-Marie-Tooth disease. 34.  Palumbo P, Palmerini L, Bandinelli S, Chiari L. Fall Risk Assessment Eur J Neurol 2015;22:1556–63. Tools for Elderly Living in the Community: Can We Do Better? PLoS One 15.  Murphy SM, Herrmann DN, McDermott MP, Scherer SS, Shy ME, 2015;10:e0146247. Reilly MM, et al. Reliability of the CMT neuropathy score (second ver- 35.  Chiles NS, Phillips CL, Volpato S, Bandinelli S, Ferrucci L, Guralnik sion) in Charcot-Marie-Tooth disease. J Peripher Nerv Syst 2011;16:191–8. JM, et al. Diabetes, peripheral neuropathy, and lower-extremity function. 16.  Sadjadi R, Reilly MM, Shy ME, Pareyson D, Laura M, Murphy S, et J Diabetes Complications 2014a;28:91–5. al. Psychometrics evaluation of Charcot-Marie-Tooth Neuropathy Score 36.  Motl RW, Learmonth YC, Wójcicki TR, Fanning J, Hubbard EA, (CMTNSv2) second version, using Rasch analysis. J Peripher Nerv Syst Kinnett-Hopkins D, et al. Preliminary validation of the short physical per- 2014;19:192–6. formance battery in older adults with multiple sclerosis: secondary data 17.  Padua L, Pazzaglia C, Pareyson D, Schenone A, Aiello A, Fabrizi analysis. BMC Geriatr 2015;15:157. GM, et al.; CMT-TRIAAL Group. Novel outcome measures for Charcot- 37.  Farag I, Sherrington C, Hayes A, Canning CG, Lord SR, Close JC, Marie-Tooth disease: validation and reliability of the 6-min walk test and et al. Economic evaluation of a falls prevention exercise program among StepWatch(™) Activity Monitor and identification of the walking features people With Parkinson’s disease. Mov Disord 2016;31:53–61. related to higher quality of life. Eur J Neurol 2016;23:1343–50. 38.  Taricco M, Dallolio L, Calugi S, Rucci P, Fugazzaro S, Stuart M, 18.  Solari A, Laurà M, Salsano E, Radice D, Pareyson D; CMT-TRIAAL et al.; Esercizio Fisico di Gruppo/2009 Investigators. Impact of adapted Study Group. Reliability of clinical outcome measures in Charcot-Marie- physical activity and therapeutic patient education on functioning and Tooth disease. Neuromuscul Disord 2008a;18:19–26. quality of life in patients with postacute strokes. Neurorehabil Neural Re- 19.  Ries JD, Echternach JL, Nof L, Gagnon Blodgett M. Test-retest reli- pair 2014;28:719–28. ability and minimal detectable change scores for the timed “up & go” test, 39.  Evans JD. Straightforward statistics for the behavioral sciences. Pa- the six-minute walk test, and gait speed in people with Alzheimer disease. cific Grove, CA: Brooks/Cole Publishing; 1996. Phys Ther 2009;89:569–79. 40.  Shy ME, Blake J, Krajewski K, Fuerst DR, Laura M, Hahn AF, et al. 20.  Steffen T, Seney M. Test-retest reliability and minimal detectable Reliability and validity of the CMT neuropathy score as a measure of dis- change on balance and ambulation tests, the 36-item short-form health ability. Neurology 2005;64:1209–14. survey, and the unified Parkinson disease rating scale in people with par- 41.  Apolone G, Mosconi P. The Italian SF-36 Health Survey: translation, kinsonism. Phys Ther 2008;88:733–46. validation and norming. J Clin Epidemiol 1998;51:1025–36. 21.  Scivoletto G, Tamburella F, Laurenza L, Foti C, Ditunno JF, Molinari 42.  Vinci P, Serrao M, Millul A, Deidda A, De Santis F, Capici S, et al. M. Validity and reliability of the 10-m walk test and the 6-min walk test in Quality of life in patients with Charcot-Marie-Tooth disease. Neurology spinal cord injury patients. Spinal Cord 2011;49:736–40. 2005;65:922–4. 22.  Fulk GD, Echternach JL, Nof L, O’Sullivan S. Clinometric proper- 43.  Ramdharry GM, Pollard A, Anderson C, Laurá M, Murphy SM, ties of the six-minute walk test in individuals undergoing rehabilitation Dudziec M, et al. A pilot study of proximal strength training in Charcot- poststroke. Physiother Theory Pract 2008;24:195–204. Marie-Tooth disease. J Peripher Nerv Syst 2014;19:328–32. 23.  van Loo MA, Moseley AM, Bosman JM, de Bie RA, Hassett L. Test- 44.  Ramdharry GM, Pollard AJ, Marsden JF, Reilly MM. Comparing gait re-test reliability of walking speed, step length and step width measure- performance of people with Charcot-Marie-Tooth disease who do and do ment after traumatic brain injury: a pilot study. Brain Inj 2004;18:1041–8. not wear ankle foot orthoses. Physiother Res Int 2012;17:191–9. 24.  Knak KL, Andersen LK, Witting N, Vissing J. Reliability of the 45.  Monti Bragadin M, Francini L, Bellone E, Grandis M, Reni L, Can- 2- and 6-minute walk tests in neuromuscular diseases. J Rehabil Med neva S, et al. Tinetti and Berg balance scales correlate with disability in 2017;49:362–6. hereditary peripheral neuropathies: a preliminary study. Eur J Phys Reha- 25.  Graham RC, Hughes RA. Clinimetric properties of a walking scale bil Med 2015;51:423–7. in peripheral neuropathy. J Neurol Neurosurg Psychiatry 2006;77:977–9. 46.  Anens E, Emtner M, Hellström K. Exploratory study of physical ac- 54 European Journal of Physical and Rehabilitation Medicine February 2019

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, CHARCOT-MARIE-TOOTH 1A EVALUATION MORI cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. tivity in persons with Charcot-Marie-Tooth disease. Arch Phys Med Re- 51.  El Mhandi L, Millet GY, Calmels P, Richard A, Oullion R, Gautheron habil 2015;96:260–8. V, et al. Benefits of interval-training on fatigue and functional capacities 47.  Lencioni T, Rabuffetti M, Piscosquito G, Pareyson D, Aiello A, Di in Charcot-Marie-Tooth disease. Muscle Nerve 2008;37:601–10. Sipio E, et al. Postural stabilization and balance assessment in Charcot- 52.  Padua L, Shy ME, Aprile I, Cavallaro T, Pareyson D, Quattrone A, et Marie-Tooth 1A subjects. Gait Posture 2014;40:481–6. al. Correlation between clinical/neurophysiological findings and quality of 48.  Lencioni T, Piscosquito G, Rabuffetti M, Bovi G, Calabrese D, Aiello life in Charcot-Marie-Tooth type 1A. J Peripher Nerv Syst 2008;13:64–70. A, et al. The influence of somatosensory and muscular deficits on postural 53.  Taniguchi JB, Elui VM, Osório FL, Hallak JE, Crippa JA, Machado- stabilization: insights from an instrumented analysis of subjects affected de-Sousa JP, et al. Quality of life in patients with Charcot-Marie-Tooth by different types of Charcot-Marie-Tooth disease. Neuromuscul Disord disease type 1A. Arq Neuropsiquiatr 2013;71:392–6. 2015;25:640–5. 54.  Cordeiro JL, Marques W, Hallak JE, Osório FL. Charcot-Marie-Tooth 49.  Lindeman E, Leffers P, Spaans F, Drukker J, Reulen J. Deterioration disease, psychiatric indicators and quality of life: a systematic review. of motor function in myotonic dystrophy and hereditary motor and sen- ASN Neuro 2014;6:185–92. sory neuropathy. Scand J Rehabil Med 1995;27:59–64. 55.  Johnson NE, Heatwole CR, Dilek N, Sowden J, Kirk CA, Shereff 50.  Wright NC, Kilmer DD, McCrory MA, Aitkens SG, Holcomb BJ, D, et al.; Inherited Neuropathies Consortium. Quality-of-life in Charcot- Bernauer EM. Aerobic walking in slowly progressive neuromuscular dis- Marie-Tooth disease: the patient’s perspective. Neuromuscul Disord ease: effect of a 12-week program. Arch Phys Med Rehabil 1996;77:64–9. 2014;24:1018–23. Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript. Funding.—This work was supported by Telethon-UILDM (grant no.: GUP09013). Group name.—Members of the TreSPE Study Group qualified as contributors include: Susanna ACCOGLI (Department of Neuroscience, Rehabilitation, Ophthalmology, Genetics, Maternal and Child Health, University of Genoa, Genoa, Italy), Simone BOLLA (Department of Neuroscience, Rehabilitation, Ophthalmology, Genetics, Maternal and Child Health, University of Genoa, Genoa, Italy), Annalisa BRUGNERA (Neuromotor and Cognitive Rehabilitation Research Center, Department of Neurosciences, Biomedicine and Movement Sciences, University of Verona, Verona, Italy), Eleonora CASATI (Don Carlo Gnocchi Foundation and Nonprofit Organization, Milan, Italy), Davide CATTANEO (Don Carlo Gnocchi Foundation and Nonprofit Organization, Milan, Italy), Emanuele CRIMI (Unit of Respiratory Pathophysiology, Department of Internal Medicine and Medical Specialties, University of Genoa, Genoa, Italy), Carla FONTANA (Neuromotor and Cognitive Rehabilitation Research Center, Department of Neurosciences, Biomedicine and Movement Sciences, University of Verona, Verona, Italy), Luca FRANCINI (Department of Neuroscience, Rehabilitation, Ophthalmology, Genetics, Maternal and Child Health, University of Genoa, Genoa, Italy), Giovanni MAGGI (San Martino Polyclinic Hospital, Genoa, Italy), Lucio MARINELLI (Department of Neuroscience, Rehabilitation, Ophthalmology, Genetics, Maternal and Child Health, University of Genoa, Genoa, Italy; San Martino Polyclinic Hospital, Genoa, Italy), Angelo MONTESANO (Don Carlo Gnocchi Foundation and Nonprofit Organization, Milan, Italy), Margherita MONTI BRAGADIN (Department of Neu- roscience, Rehabilitation, Ophthalmology, Genetics, Maternal and Child Health, University of Genoa, Genoa, Italy), Daniele MUNARI (Neuromotor and Cognitive Rehabilitation Research Center, Department of Neurosciences, Biomedicine and Movement Sciences, University of Verona, Verona, Italy)Alex SALERNO (Department of Neuroscience, Rehabilitation, Ophthalmology, Genetics, Maternal and Child Health, University of Genoa, Genoa, Italy), Debo- rah SCORSONE (San Martino Polyclinic Hospital, Genoa, Italy), Riccardo ZUCCARINO (Service of Neuromuscular Rehabilitation, La Colletta Hospital, Arenzano, Genoa, Italy). Congresses.—Data from this study have been presented at the XVI SIRN Congress (Ascoli Piceno, 7th-9th April, 2016) and at the 44° S.I.M.F.E.R. congress (Bari, 23rd-26th October, 2016). Acknowledgements.—The authors wish to thank all the patients who participated in the trial. Article first published online: June 11, 2018. - Manuscript accepted: June 7, 2018. - Manuscript revised: April 16, 2018. - Manuscript received: December 11, 2017. For supplementary materials, please see the HTML version of this article at www.minervamedica.it Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 55

COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically © 2018 EDIZIONI MINERVA MEDICA European Journal of Physical and Rehabilitation Medicine 2019 February;55(1):56-62 or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access Online version at http://www.minervamedica.it DOI: 10.23736/S1973-9087.18.05191-2 to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. ORIGINAL ARTICLE Classes of vitamin D status and functional outcome after hip fracture: a prospective, short-term study of 1350 inpatients Marco DI MONACO 1 *, Carlotta CASTIGLIONI 1, Silvia DI CARLO 1, Elena LA MARMORA 2, Irena FILIPOVIC 2, Edoardo MILANO 1, Marco A. MINETTO 2, Giuseppe MASSAZZA 2 1Division of Physical Medicine and Rehabilitation, Osteoporosis Research Center, Presidio Sanitario San Camillo, Opera San Camillo Foundation, Turin, Italy; 2Division of Physical Medicine and Rehabilitation, Department of Surgical Sciences, University of Turin, Turin, Italy *Corresponding author: Marco Di Monaco, Division of Physical Medicine and Rehabilitation, Osteoporosis Research Center, Presidio Sanitario San Camillo, Opera San Camillo Foundation strada Santa Margherita 136, 10131 Turin, Italy. E-mail [email protected] ABSTRACT BACKGROUND: Vitamin D depletion is associated with unfavourable outcomes after hip fracture. However, the classes of vitamin D status currently in use, which are defined according to serum calcifediol levels, have not been validated for their predictive capability of the functional recovery. AIM: To investigate the association between serum calcifediol categorized into 4 classes and the functional recovery after hip fracture. DESIGN: Prospective, short-term observational study. SETTING: Rehabilitation hospital in Italy. POPULATION: We evaluated 1350 of 1412 inpatients with hip fracture. METHODS: Serum calcifediol was measured by an immunoenzymatic assay 14.7±4.4 (mean±SD) days after surgery and categorized into 4 classes: I class <12 ng/mL; II class 12-20 ng/mL; III class 21-29 ng/mL; IV class ≥30ng/mL. The functional outcome was assessed by using the Barthel Index. RESULTS: We found a significant difference in Barthel index scores at the end of inpatient rehabilitation across the 4 classes of vitamin D status: χ2 (3, N.=1350) 27.2; P<0.001. The difference persisted after adjustment for 8 covariates (P=0.004). By comparing pairs of classes, we found that Barthel index scores were lower in the 829 patients of the I class than in the 275 of the II (P=0.005) who had in turn Barthel index scores lower than the 132 patients of the III class (P=0.038). Conversely, no significant differences emerged between the patients of the III class and the 114 patients of the IV class (P=0.421). The results did not materially change when Barthel Index effectiveness was substituted for Barthel Index scores as the outcome measure. CONCLUSIONS: Calcifediol levels below 12ng/mL were associated with a worse recovery than those between 12 and 20ng/mL that were in turn associated with a worse recovery than those between 21 and 29 ng/mL. Conversely, no significant differences were found between the patients with calcifediol between 21 and 29ng/mL and those with calcifediol ≥30 ng/mL. CLINICAL REHABILITATION IMPACT: Despite caution due to the observational design, our study suggests that vitamin D depletion should be treated after hip fracture to optimize the functional outcome, with a target level for serum calcifediol of 21-29ng/mL and no further advantages associated with calcifediol levels of 30ng/mL or higher. (Cite this article as: Di Monaco M, Castiglioni C, Di Carlo S, La Marmora E, Filipovic I, Milano E, et al. Classes of vitamin D status and functional outcome after hip fracture: a prospective, short-term study of 1350 inpatients. Eur J Phys Rehabil Med 2019;55:56-62. DOI: 10.23736/S1973- 9087.18.05191-2). Key words: Hip fractures - Rehabilitation - Vitamin D. Hip fractures strongly impact on public health: they re- functional recovery.3-7 Vitamin D depletion is thought to sult in 8% to 36% excess mortality within one year and negatively affect ability to function following a fracture approximately 10-20% of hip fracture survivors require of the hip, given the significant association between se- long-term nursing home care, whereas only 40-70% fully rum calcifediol levels and measures of function shown in regain their pre-injury level of independence.1, 2 A num- 5 observational studies from 3 research groups.8-12 How- ber of independent variables have been associated with the ever, optimal levels of serum calcifediol have not been 56 European Journal of Physical and Rehabilitation Medicine February 2019

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, VITAMIN D AND RECOVERY AFTER HIP FRACTURE DI MONACO cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. established in hip fracture patients, who commonly have concomitant diseases. The final study sample included severe vitamin D deficiency,13-15 and there are not target 1350 patients. We focused on white patients because few ranges for serum calcifediol agreed on to optimize vitamin non-white, elderly people live in our country. Our hospital D supplementation in everyday practice. is in a city with about 1 million inhabitants, and the pa- tients came from several orthopedic wards from various In the general population, vitamin D status has been hospitals. All patients were referred for acute inpatient re- differently categorized according to the levels of serum habilitation by the consultant physiatrists of the orthopedic calcifediol. In 2011 the Institute of Medicine (IOM) pro- wards. The criteria agreed upon for selecting patients to posed 3 classes to define vitamin D status: deficiency, risk undergo acute inpatient rehabilitation were health condi- of inadequacy and sufficiency (serum calcifediol <12 ng/ tions allowing a total of 3 hours of physical therapy and/or mL, 12-20 ng/mL and >20 ng/mL, respectively).16 In the occupational therapy daily, weight bearing to tolerance on same year the Endocrine Society (ES) suggested a differ- the fractured hip, and a potential high increase in the abil- ent categorization into 3 classes: deficiency, insufficiency ity to function in activities of daily living because of an in- and sufficiency (serum calcifediol <20 ng/mL, 21-29 ng/ tensive rehabilitation regimen. None of the 1350 inpatients mL and ≥30 ng/mL, respectively).17 The thresholds listed included in the study received vitamin D supplementation above may be used to define optimal targets for vitamin D between the hip-fracture occurrence and the blood sample supplementation after hip fracture, but there is the need of collection. All the patients gave their written informed studies specifically conducted in patients with hip fracture consent to participate in the study. The study protocol was to validate any suggested thresholds. Indeed, no attempts approved by the local ethics committee (protocol 52119; have been made so far to validate any calcifediol levels CS/209). as for their ability to predict the functional outcomes in hip-fracture survivors. In summary, clinicians know that Calcifediol measure and assessment of covariates severe vitamin D deficiency is associated with unfavorable outcomes after hip fracture, but they do not know which In the 1350 patients, a blood sample was collected in are the target levels of serum calcifediol to optimize re- the morning following an overnight fasting 14.7±4.4 covery. (mean±SD) days after surgical repair of the hip fracture. We evaluated calcifediol levels by an immunoenzymatic Our aim was to investigate the association between se- assay (coefficient of variation intrassay <8%; interassay rum calcifediol levels categorized into 4 classes and the <10%) (IDS Inc., Fountain Hills, AZ, USA). We used the functional recovery after a fragility fracture of the hip. The same laboratory assay to measure calcifediol levels dur- 4 classes were defined taking into account the thresholds ing the whole study period. We categorized the calcifediol pointed out by both IOM and ES (<12 ng/mL; 12-20 ng/ levels into 4 classes: I class <12 ng/mL; II class 12-20 ng/ mL; 21-29 ng/mL; ≥30 ng/mL).16, 17 We expected to ob- mL; III class 21-29 ng/mL; IV class ≥30 ng/mL. serve increasing levels of functional autonomy in activi- ties of daily living across the 4 classes of vitamin D status. Materials and methods Outcome measures and assessment of covariates Study population Functional evaluation, both at rehabilitation admission and at discharge from the rehabilitation hospital, was as- We evaluated 1412 consecutive inpatients with a hip frac- sessed by skilled physiatrists by using the Barthel Index ture admitted to our Physical Medicine and Rehabilitation (original version unchanged). The functional index as- division. Forty of the 1412 inpatients we evaluated were sesses basic activities of daily living; its score ranges from excluded from our study because their hip fractures re- 0 (total dependence) to 100 (total independence). Barthel sulted from either major trauma or cancer affecting bone. Index effectiveness was calculated using the following The remaining 1372 patients sustained fractures that either formula: (Change in Barthel Index score during rehabili- were spontaneous or resulted from minimal trauma (trau- tation/[100 - Barthel Index +score at admission to reha- ma equal to or less than a fall from a standing position). All bilitation]) x 100. It represented the proportion of potential the fractures were confirmed by X-ray examination and improvement actually achieved by each patient during the were surgically operated on. Twenty-two patients could rehabilitation course. The physiatrists were not aware of not complete the rehabilitation course because of acute the results of calcifediol assessment at the time of Barthel Index score evaluation. Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 57

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically DI MONACO COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, VITAMIN D AND RECOVERY AFTER HIP FRACTURE cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. In each patient we took into account the role of 8 covari- Table I.—D escription of the 1350 patients given as median (inter- ates that were shown to affect functional recovery after quartile range) unless otherwise indicated. acute inpatient rehabilitation following a hip fracture: age, sex, Barthel Index scores before rehabilitation, neurologic Age (years), mean (SD) 80.1 (7.7) impairment (impairment found at clinical examination due Sex (women/men), % 87/13 to neurologic diseases, i.e., Parkinson disease, stroke with Calcifediol (ng/mL) hemiplegia, paraparesis, monoparesis, tetraparesis, or cer- Barthel Index score before rehabilitation 9.1 (5.3; 16.0) ebellar syndrome), cognitive impairment (Mini-Mental Barthel Index score after rehabilitation 45 (30; 60) State Examination <24/30), infections (at least one infec- Barthel Index effectiveness 90 (70; 95) tion needing antibiotic therapy during the stay in hospital), Neurologic impairment, % 77.8 (53.2; 92.3) pressure ulcers (stage 2 or higher according to the classi- Cognitive impairment, % fication from the National Pressure Ulcer Advisory Panel) Infections during the stay in hospital, % 17% and hip-fracture type (either medial or lateral).3-12 Pressure ulcers, % 24% Fracture type (lateral/medial), % 54% Statistical analysis 26% 54/46 The two outcome measures (Barthel Index scores after rehabilitation and Barthel Index effectiveness) were non- the 4 classes of vitamin D status: χ2 (3, N.=1350) 27.3; normally distributed at a Shapiro-Wilk test. We compared P<0.001. By comparing pairs of classes, we found that the two outcome measures by a Kruskal-Wallis non-para- Barthel Index scores were significantly lower in the 829 metric test across the 4 classes of calcifediol levels. Pairs of patients of the I class (median 85; interquartile range=IQR classes were compared by a Mann-Whitney U Test. Addi- 65-95) than in the 275 of the II class (median 90; IQR tionally, the calcifediol levels categorized into the 4 classes 75-95): U=101,228; z=-2.81; P=0.005. The 275 patients of were included in linear multiple regression models as an the II class had in turn Barthel Index scores significantly independent variable together with age, sex, Barthel Index lower than the 132 patients of the III class (median 95; scores before rehabilitation, neurologic impairment, cogni- IQR 76.25-100): U=15888; z=-2.07; P=0.038. Conversely, tive impairment, infections, pressure ulcers and hip-frac- no significant differences emerged between the 132 pa- ture type. The dependent variable in the regression models tients of the III class and the 114 patients of the IV class was either the Barthel Index score at the end of inpatient whose median Barthel Index score was 90 (IQR 75-100): rehabilitation or Barthel Index effectiveness. Because the U=7087; z=-0.804; P=0.421. dependent variables were non-normally distributed, area transformation was performed, using the formula (r-1/2)/w, A linear multiple regression confirmed the significant where w is the number of observations and r is the rank.11 association between vitamin D status categorized into the Following area transformation of the dependent variable, 4 classes and the Barthel Index scores assessed at the end the residuals were normally distributed in the regression of postacute inpatient rehabilitation after adjustment for 8 models. Homoscedasticity was verified by plotting the re- covariates (P=0.004), as shown in Table II. siduals against the predicted values: the variance of the re- siduals looked homogeneous across levels of the predicted Barthel Index effectiveness was significantly different values. Collinearity diagnostics showed that the percent of across the 4 classes of vitamin D status: χ2 (3, N.=1350) variance in each predictor that could not be accounted for 24.05; P<0.001. By comparing pairs of classes, we found by the other predictors was always greater than 85% (no that Barthel Index effectiveness was significantly lower in redundant predictors were found). We had no missing data. the 829 patients of the I class (median 75; IQR 50-90) than in the 275 of the II class (median 81.82; IQR 53.33-94.44): The statistical package used was SPSS, version 17. U=102,746; z=-2.46; P=0.014. The 275 patients of the II class had in turn a Barthel Index effectiveness significantly Results lower than the 132 patients of the III class (median 87.5; IQR 61.54-100): U=15862; z=-2.08; P=0.037. Conversely, Descriptive statistics for the 1350 inpatients are shown in no significant differences emerged between the 132 pa- Table I. tients of the III class and the 114 patients of the IV class whose median Barthel Index effectiveness was 83.33 (IQR We found a significant difference in Barthel Index 60-100): U=7103; z=-0.771; P=0.441. scores assessed at the end of inpatient rehabilitation across A linear multiple regression confirmed the significant association between vitamin D status categorized into the 4 classes and the Barthel Index effectiveness (P=0.007), after adjustment for 8 covariates as shown in Table III. 58 European Journal of Physical and Rehabilitation Medicine February 2019

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, VITAMIN D AND RECOVERY AFTER HIP FRACTURE DI MONACO cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. Table II.—L inear multiple regression analysis model assessing the variables associated with the Barthel Index scores at the end of the rehabilitation course. Independent variables B (95% CI) Beta P (Constant) 0.553 (0.158; 0.948) 0.052 0.006 Vitamin D status (categorized into 4 classes) 0.049 (0.016; 0.082) 0.57 0.004 Barthel Index score before rehabilitation 0.028 (0.026; 0.030) -0.17 <0.001 Age (years) -0.020 (-0.024; -0.016) -0.05 <0.001 Sex -0.121 (-0.215; -0.026) -0.12 0.012 Neurologic impairment -0.311 (-0.409; -0.213) -0.13 <0.001 Cognitive impairment -0.320 (-0.412; -0.228) -0.04 <0.001 Infections during the stay in hospital -0.080 (-0.153; -0.007) -0.03 0.031 Pressure ulcers -0.066 (-0.152; 0.019) -0.01 0.129 Hip-fracture type -0.025 (-0.092; 0.042) 0.468 The dependent variable was the Barthel Index score at the end of the rehabilitation course (after normalization by area transformation). The independent variables were those listed in the Table. For each independent variable, unstandardized B coefficients with 95% confidence intervals, standardized Beta coefficients and P values are shown. Male sex, lateral fractures and the presence of neurologic impairment, cognitive impairment, infections during the stay in hospital and pressure ulcers were conventionally attributed a value of one (female sex, medial fractures and the absence of the conditions listed above were conventionally attributed a value of 0). R2=0.57; F=195.1; P<0.001. Table III.—Linear multiple regression analysis model assessing the variables associated with the Barthel index effectiveness. Independent variables B (95% CI) Beta P (Constant) 1.321 (0.839; 1.803) 0.06 <0.001 Vitamin D status (categorized into 4 classes) 0.056 (0.015; 0.096) 0.37 0.007 Barthel Index score before rehabilitation 0.019 (0.016; 0.021) -0.21 Age (years) -0.025 (-0.030; -0.020) -0.05 <0.001 Sex -0.127 (-0.242; -0.012) -0.15 <0.001 Neurologic impairment -0.391 (-0.510; -0.272) -0.16 Cognitive impairment -0.402 (-0.513; -0.290) -0.05 0.031 Infections during the stay in hospital -0.107 (-0.196; -0.018) -0.02 <0.001 Pressure ulcers -0.041 (-0.146; 0.063) 0.01 <0.001 Hip-fracture type 0.009 (-0.073; 0.091) 0.019 0.439 0.825 The dependent variable was the Barthel Index effectiveness (after normalization by area transformation). The independent variables were those listed in the Table. For each independent variable, unstandardized B coefficients with 95% confidence intervals, standardized Beta coefficients and P values are shown. Male sex, lateral fractures and the presence of neurologic impairment, cognitive impairment, infections during the stay in hospital and pressure ulcers were conventionally attributed a value of one (female sex, medial fractures and the absence of the conditions listed above were conventionally attributed a value of 0). R2=0.38; F=92.1; P<0.001. Discussion for serum calcifediol levels in everyday rehabilitation. We show that calcifediol levels below 12 ng/mL were associ- In our sample of inpatients with hip fracture, vitamin D ated with a worse recovery than those between 12 and 20ng/ status was significantly associated with the short-term mL that were in turn associated with a worse recovery than functional outcome. This result confirms previous studies those between 21 and 29 ng/mL. Conversely, no significant from 3 research groups performed after a fracture of the differences were found between the patients with calcife- hip Di Monaco et al. showed that calcifediol levels were diol between 21 and 29ng/mL and those with calcifediol significantly associated with the Barthel Index score as- ≥30 ng/mL. In the previous studies only LeBoff et al. cat- sessed at the end of inpatient rehabilitation.8, 9, 11 LeBoff egorized their patients according to a defined threshold of et al. showed that baseline calcifediol levels significantly serum calcifediol; however, they simply compared the pa- predicted legs performance and fall risk in women at a one- tients with levels either below or above 9ng/mL.10 There is year follow-up.10 Pioli et al. showed that baseline calcife- wide agreement on prevention and treatment of very low diol levels predicted mobility at a six-month follow-up.12 calcifediol levels, as those below 9ng/mL pointed out by LeBoff et al.16-20 Conversely, there is no consensus on the The novelty from this study is the attempt to validate 4 threshold defining “sufficient” vitamin D status, and opti- classes of vitamin D status16, 17 as for their predictive capa- mal levels >20 ng/mL,16, 18 ≥30 ng/mL,17, 19 or either >20 bility of the functional outcome following a hip fracture. ng/mL or ≥ 30ng/mL depending on patients characteristics20 The validation aims to individuate an optimal target range Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 59

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically DI MONACO COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, VITAMIN D AND RECOVERY AFTER HIP FRACTURE cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. have been pointed out. Despite caution due to the observa- that current studies on prognostic factors include multivar- tional design, our study suggests that no further advantages iate analyses with several variabless to predict a substantial seem associated with shifting the threshold for vitamin D proportion of outcome measures variance. Overall the pan- sufficiency from 20 ng/mL to 30 ng/mL at least to optimize el of prognostic factors we evaluated accounted for 57% of functional autonomy following a hip fracture. A relatively the variance in the Barthel Index scores assessed at the end low target level of serum calcifediol (20 ng/mL instead of of the post-acute inpatient rehabilitation. It is noteworthy 30 ng/mL) leads to two relevant consequences in clinical that the prognostic role of vitamin D was independent of practice. The first consequence is that the doses of vitamin 8 covariates included in the predictive regression model. D needed for supplementation are much lower. This is due We did not assess the mechanisms underlying the associa- to the non-linear response of serum calcifediol to vitamin D tion between vitamin D status and functional ability. Five intake: proportionately more vitamin D is required to con- lines of evidence support the view that vitamin D deple- tinue to increase serum calcifediol level after the threshold tion can cause muscle impairment, a condition that is in of 20ng/mL is reached (IOM).16 The second consequence is turn associated with disability following a hip fracture.28-32 that the possibility of achieving potentially toxic calcifediol First, clinical manifestations including muscle weakness, levels is dramatically lowered. Concern on toxic levels has muscle pain and gait impairment are commonly observed raised in the last years, because even modest serum calcife- in osteomalacia due to severe vitamin D deficiency.33 Sec- diol levels (40-50 ng/mL or higher) have been associated ond, direct actions of vitamin D mediated by its receptor, with unfavorable outcomes including both increased fall and predominantly indirect actions mediated by changes in risk in randomized controlled trials21-24 and enhanced mor- calcium and phosphate homeostasis have been document- tality in survey studies.25 Overall, a reverse-J-shaped curve ed in the muscle.34-36 Third, vitamin D can modulate the has been proposed to depict risks and benefits of increasing cross-talk between muscle and bone throughout the whole levels of serum calcifediol,24, 25 even at concentrations quite life of the musculoskeletal system.34 Fourth, several ob- lower than those associated with classical toxicity with hy- servational studies showed a positive association between percalcemia which are rarely found in clinical practice.16, 26 vitamin D status and both muscle strength and physical performance.36, 37, 38 Finally, several randomized controlled The absolute difference in median Barthel Index scores trials showed favorable effects of vitamin D administration between patients in the lowest class of serum calcifediol on muscle strength,39 body balance,40 and fall prevention,41 and those with calcifediol levels in the suggested optimal although conflicting studies exist.42 It is likely that the vita- range (21-29 ng/mL) was 10. Although we did not assess min D dosage and the vitamin D status of the treated popu- the minimal clinical important difference in our sample, a lation are largely responsible for the inconsistencies found 10-point change in Barthel Index scores is usually thought in the literature: fall-risk reduction may not be observed in to be clinically significant and beyond measurement error, low dose treatment regimens and in patients not deficient thus strengthening the clinical meaning of our findings.27 enough in vitamin D to benefit from supplementation.43, 44 Treating patients with unequivocal vitamin D depletion Besides calcifediol levels, 6 factors were independently and achieving optimal calcifediol levels seem to be the 2 associated with Barthel Index scores in our sample. The needs to appreciate the beneficial effects of vitamin D ad- presence of multiple independent prognostic factors in pa- ministration. The first need is virtually always met follow- tients with hip fracture is in agreement with the wider liter- ing a hip fracture, given the very high prevalence of severe ature.3-12 Negative prognostic roles of functional autonomy vitamin D deficiency shown consistently by our current before rehabilitation, advanced age, male sex, neurologic data and several previous reports.13-15 Our attempt to define impairment, cognitive impairment, and infections occur- the optimal calcifediol range may contribute to meet the ring during the hospital stay are in agreement with previous second need by elucidating the target for serum calcifediol reports.3-12 Our findings confirm the crucial role of patient- in rehabilitation. Benefits due to vitamin D supplementa- related factors in affecting the functional recovery: indi- tion have been actually shown by few randomized con- vidual characteristics (age, sex, comorbidity and functional trolled trials performed after a hip fracture. Harwood et al. ability) are the main determinants of the rehabilitation out- showed that either oral or injected vitamin D reduced fall comes after hip fracture.3, 28 Unfortunately, the number of risk after one year in 150 previously independent women.45 independent predictors pointed out by the literature is high Bischoff-Ferrari et al. showed that 2000 IU/d of cholecal- and we still lack simple predictive models of the functional ciferol reduced hospital readmissions (but not falls) in 173 recovery largely agreed on and robustly validated in differ- ent samples of patients with hip fracture.28 It is mandatory 60 European Journal of Physical and Rehabilitation Medicine February 2019

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, VITAMIN D AND RECOVERY AFTER HIP FRACTURE DI MONACO cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. patients (79.2% women) at a one-year follow-up.46 Mak References et al. showed that a single loading dose of cholecalciferol (250,000 IU) followed by daily 800 IU reduced the risk of 1.  Abrahamsen B, van Staa T, Ariely R, Olson M, Cooper C. Excess mor- falling versus the daily dose alone in 218 patients (77.1% tality following hip fracture: a systematic epidemiological review. Osteo- women) at a 26-week follow-up.47 poros Int 2009;20:1633–50. 2.  Dyer SM, Crotty M, Fairhall N, Magaziner J, Beaupre LA, Cameron Limitations of the study ID, et al.; Fragility Fracture Network (FFN) Rehabilitation Research Spe- cial Interest Group. A critical review of the long-term disability outcomes Three strengths of this study are blinding of the assessors, following hip fracture. BMC Geriatr 2016;16:158. adjustment of our data for 8 potential confounders, and 3.  Kristensen MT. Factors affecting functional prognosis of patients with large sample size. Our study has several limitations. We hip fracture. Eur J Phys Rehabil Med 2011;47:257–64. evaluated one sample of white inpatients, who were sur- 4.  Beaupre LA, Binder EF, Cameron ID, Jones CA, Orwig D, Sherrington gically operated on and who were referred for inpatient C, et al. Maximising functional recovery following hip fracture in frail rehabilitation in a single division. As a consequence, our seniors. Best Pract Res Clin Rheumatol 2013;27:771–88. results are not generalizable to the overall population of 5.  Benedetti MG, Ginex V, Mariani E, Zati A, Cotti A, Pignotti E, et al. hip fracture patients. We did not collect data on vitamin D Cognitive impairment is a negative short-term and long-term prognos- supplementation before fracture occurrence. Blood sam- tic factor in elderly patients with hip fracture. Eur J Phys Rehabil Med ples for laboratory assessment were collected around two 2015;51:815–23. weeks after surgery, so a role of hospitalization in wors- 6.  Gialanella B, Ferlucci C, Monguzzi V, Prometti P. Determinants of ening the deficiency of vitamin D cannot be excluded. outcome in hip fracture: role of daily living activities. Eur J Phys Rehabil However, changes in circulating levels of calcifediol are Med 2015;51:253–60. slow, because of both progressive cholecalciferol release 7.  Brown K, Cameron ID, Keay L, Coxon K, Ivers R. Functioning and from adipose tissue and muscle and slow calcifediol me- health-related quality of life following injury in older people: a systematic tabolism. We assessed the functional recovery by using the review. Inj Prev 2017;23:403–11. Barthel Index. This scale is validated and widely used for 8.  Di Monaco M, Vallero F, Di Monaco R, Mautino F, Cavanna A. Serum the assessment of ability to function in activities of daily levels of 25-hydroxyvitamin D and functional recovery after hip fracture. living (ADL), which plays a pivotal role in the functional Arch Phys Med Rehabil 2005;86:64–8. prognosis after a hip fracture. However, we considered 9.  Di Monaco M, Vallero F, Di Monaco R, Tappero R, Cavanna A. 25-hy- neither other scales assessing ADL, such as the Functional droxyvitamin D, parathyroid hormone, and functional recovery after hip Independence Measure scale, nor any tools to evaluate fracture in elderly patients. J Bone Miner Metab 2006;24:42–7. other outcomes. Particularly, we did not assess any mea- 10.  LeBoff MS, Hawkes WG, Glowacki J, Yu-Yahiro J, Hurwitz S, sures of participation or health-related quality of life. Our Magaziner J. Vitamin D-deficiency and post-fracture changes in lower follow-up was short. Finally, the study design does not extremity function and falls in women with hip fractures. Osteoporos Int prove causation: we show the predictive role of vitamin D 2008;19:1283–90. status, not cause-and effect relationships. 11.  Di Monaco M, Castiglioni C, Vallero F, Di Monaco R, Tappero R. Appendicular lean mass does not mediate the significant association be- Conclusions tween vitamin D status and functional outcome in hip-fracture women. Arch Phys Med Rehabil 2011;92:271–6. Despite caution due to the observational design, we drive 12.  Pioli G, Lauretani F, Pellicciotti F, Pignedoli P, Bendini C, Davoli two conclusions from this study. First, we confirm that vi- ML, et al. Modifiable and non-modifiable risk factors affecting walking tamin D depletion (a very common condition following a recovery after hip fracture. Osteoporos Int 2016;27:2009–16. hip fracture) should be treated to optimize the functional 13.  Bischoff-Ferrari HA, Can U, Staehelin HB, Platz A, Henschkowski outcome. Second, we provide the first evidence to support J, Michel BA, et al. Severe vitamin D deficiency in Swiss hip fracture a target level for serum calcifediol of 21-29 ng/mL follow- patients. Bone 2008;42:597–602. ing a hip fracture, with no further advantages associated 14.  Ramason R, Selvaganapathi N, Ismail NH, Wong WC, Rajamoney with calcifediol levels of 30 ng/mL or higher. Our sugges- GN, Chong MS. Prevalence of vitamin d deficiency in patients with hip tion of a defined target range may help clinicians to opti- fracture seen in an orthogeriatric service in sunny singapore. Geriatr Or- mize vitamin D supplementation and its expected benefits thop Surg Rehabil 2014;5:82–6. in the rehabilitation process of the patients who sustain a 15.  DI Monaco M, Castiglioni C, DI Monaco R, Tappero R. Time trend fracture of the hip. 2000-2013 of vitamin D status in older people who sustain hip fractures: steps forward or steps back? A retrospective study of 1599 patients. Eur J Phys Rehabil Med 2016;52:502–7. 16.  Institute of Medicine (US) Committee to Review Dietary Reference Intakes for Vitamin D and Calcium. Ross AC, Taylor CL, Yaktine AL, Del Valle HB, editors. Dietary Reference Intakes for Calcium and Vitamin D. Washington (DC): National Academies Press (US); 2011. 17.  Holick MF, Binkley NC, Bischoff-Ferrari HA, Gordon CM, Hanley DA, Heaney RP, et al.; Endocrine Society. Evaluation, treatment, and pre- vention of vitamin D deficiency: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 2011;96:1911–30. 18.  Francis RM, Aspray TJ, Bowring CE, Fraser WD, Gittoes NJ, Javaid MK, et al. National Osteoporosis Society practical clinical guideline on vitamin D and bone health. Maturitas 2015;80:119–21. Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 61

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically DI MONACO COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, VITAMIN D AND RECOVERY AFTER HIP FRACTURE cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. 19.  American Geriatrics Society Workgroup on Vitamin D Supplementa- 33.  Reddy Munagala VV, Tomar V. Images in clinical medicine. Osteo- tion for Older Adults. Recommendations abstracted from the American malacia. N Engl J Med 2014;370:e10. Geriatrics Society Consensus Statement on vitamin D for prevention of 34.  Girgis CM, Baldock PA, Downes M. Vitamin D, muscle and bone: falls and their consequences. J Am Geriatr Soc 2014;62:147–52. integrating effects in development, aging and injury. Mol Cell Endocrinol 20.  Rizzoli R, Boonen S, Brandi ML, Bruyère O, Cooper C, Kanis JA, 2015;410:3–10. et al. Vitamin D supplementation in elderly or postmenopausal women: a 35.  Dawson-Hughes B. Vitamin D and muscle function. J Steroid Bio- 2013 update of the 2008 recommendations from the European Society for chem Mol Biol 2017;173:313–6. Clinical and Economic Aspects of Osteoporosis and Osteoarthritis (ES- 36.  Olsson K, Saini A, Strömberg A, Alam S, Lilja M, Rullman E, et CEO). Curr Med Res Opin 2013;29:305–13. al. Evidence for Vitamin D Receptor Expression and Direct Effects of 21.  Smith H, Anderson F, Raphael H, Maslin P, Crozier S, Cooper C. 1α,25(OH)2D3 in Human Skeletal Muscle Precursor Cells. Endocrinol- Effect of annual intramuscular vitamin D on fracture risk in elderly men ogy 2016;157:98–111. and women—a population-based, randomized, double-blind, placebo- 37.  Bischoff-Ferrari HA, Dietrich T, Orav EJ, Hu FB, Zhang Y, Karlson controlled trial. Rheumatology (Oxford) 2007;46:1852–7. EW, et al. Higher 25-hydroxyvitamin D concentrations are associated 22.  Sanders KM, Stuart AL, Williamson EJ, Simpson JA, Kotowicz MA, with better lower-extremity function in both active and inactive persons Young D, et al. Annual high-dose oral vitamin D and falls and fractures aged > or =60 y. Am J Clin Nutr 2004;80:752–8. in older women: a randomized controlled trial. JAMA 2010;303:1815–22. 38.  Iolascon G, Mauro GL, Fiore P, Cisari C, Benedetti MG, Panella L, et 23.  Bischoff-Ferrari HA, Dawson-Hughes B, Orav EJ, Staehelin HB, al. Can vitamin D deficiency influence muscle performance in postmeno- Meyer OW, Theiler R, et al. Monthly high-dose vitamin D treatment for pausal women? A multicenter retrospective study. Eur J Phys Rehabil Med the prevention of functional decline: a randomized clinical trial. JAMA 2018;54:676–82. Intern Med 2016;176:175–83. 39.  Beaudart C, Buckinx F, Rabenda V, Gillain S, Cavalier E, Slomian J, 24.  Smith LM, Gallagher JC, Suiter C. Medium doses of daily vi- et al. The effects of vitamin D on skeletal muscle strength, muscle mass, tamin D decrease falls and higher doses of daily vitamin D3 in- and muscle power: a systematic review and meta-analysis of randomized crease falls: A randomized clinical trial. J Steroid Biochem Mol Biol controlled trials. J Clin Endocrinol Metab 2014;99:4336–45. 2017;173:317–22. 40.  Muir SW, Montero-Odasso M. Effect of vitamin D supplementation 25.  Pilz S, Grübler M, Gaksch M, Schwetz V, Trummer C, Hartaigh BÓ, on muscle strength, gait and balance in older adults: a systematic review et al. Vitamin D and Mortality. Anticancer Res 2016;36:1379–87. and meta-analysis. J Am Geriatr Soc 2011;59:2291–300. 26.  Italian Society of Osteoporosis. Mineral Metabolism and Skeletal 41.  Murad MH, Elamin KB, Abu Elnour NO, Elamin MB, Alkatib Diseases (SIOMMMS); Italian Society of Rheumatology (SIR), Varenna AA, Fatourechi MM, et al. Clinical review: The effect of vitamin D on M, Bertoldo F, Di Monaco M, Giusti A, Martini G, Rossini M. Safety pro- falls: a systematic review and meta-analysis. J Clin Endocrinol Metab file of drugs used in the treatment of osteoporosis: a systematical review 2011;96:2997–3006. of the literature. Reumatismo 2013;65:143–66. 42.  Bolland MJ, Grey A, Gamble GD, Reid IR. Vitamin D supplementa- 27.  Quinn TJ, Langhorne P, Stott DJ. Barthel index for stroke trials: de- tion and falls: a trial sequential meta-analysis. Lancet Diabetes Endocrinol velopment, properties, and application. Stroke 2011;42:1146–51. 2014;2:573–80. 28.  Di Monaco M. Rehabilitation after hip fracture in older people. Eur J 43.  LeBlanc ES, Chou R. Vitamin D and falls-fitting new data with cur- Phys Rehabil Med 2011;47:253–5. rent guidelines. JAMA Intern Med 2015;175:712–3. 29.  Mangione KK, Craik RL, Palombaro KM, Tomlinson SS, Hof- 44.  Gillespie LD, Robertson MC, Gillespie WJ, Sherrington C, Gates S, mann MT. Home-based leg-strengthening exercise improves function 1 Clemson LM, et al. Interventions for preventing falls in older people liv- year after hip fracture: a randomized controlled study. J Am Geriatr Soc ing in the community. Cochrane Database Syst Rev 2012;(9):CD007146. 2010;58:1911–7. 45.  Harwood RH, Sahota O, Gaynor K, Masud T, Hosking DJ; Notting- 30.  Sylliaas H, Brovold T, Wyller TB, Bergland A. Prolonged strength ham Neck of Femur (NONOF) Study. A randomised, controlled com- training in older patients after hip fracture: a randomised controlled trial. parison of different calcium and vitamin D supplementation regimens in Age Ageing 2012;41:206–12. elderly women after hip fracture: The Nottingham Neck of Femur (NON- 31.  Di Monaco M, Castiglioni C, De Toma E, Gardin L, Giordano S, Tap- OF) Study. Age Ageing 2004;33:45–51. pero R. Handgrip strength is an independent predictor of functional out- 46.  Bischoff-Ferrari HA, Dawson-Hughes B, Platz A, Orav EJ, Stähelin come in hip-fracture women: a prospective study with 6-month follow-up. HB, Willett WC, et al. Effect of high-dosage cholecalciferol and extended Medicine (Baltimore) 2015;94:e542. physiotherapy on complications after hip fracture: a randomized con- 32.  Di Monaco M, Castiglioni C. Weakness and low lean mass in wom- trolled trial. Arch Intern Med 2010;170:813–20. en with hip fracture: prevalence according to the FNIH criteria and as- 47.  Mak JC, Mason RS, Klein L, Cameron ID. An initial loading-dose vi- sociation with the short-term functional recovery. J Geriatr Phys Ther tamin D versus placebo after hip fracture surgery: randomized trial. BMC 2017;40:80–5. Musculoskelet Disord 2016;17:336. Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript. Article first published online: June 14, 2018. - Manuscript accepted: June 13, 2018. - Manuscript revised: May 23, 2018. - Manuscript received: January 20, 2018. 62 European Journal of Physical and Rehabilitation Medicine February 2019

COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically © 2018 EDIZIONI MINERVA MEDICA European Journal of Physical and Rehabilitation Medicine 2019 February;55(1):63-70 or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access Online version at http://www.minervamedica.it DOI: 10.23736/S1973-9087.18.05118-3 to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. ORIGINAL ARTICLE Train the brain! Immediate sensorimotor effects of mentally-performed flexor exercises in patients with neck pain A pilot study Konstantin BEINERT 1 *, Marc SOFSKY 1, Jörg TROJAN 2 1Department of Physiotherapy, University of Applied Health Sciences Rhein-Neckar, Mannheim, Germany; 2Department of Psychology, University of Koblenz-Landau, Landau, Germany *Corresponding author: Konstantin Beinert, Department of Physiotherapy, University of Applied Health Sciences Rhein-Neckar, Walter-Krause Straße 11, 68163 Mannheim, Germany. E-mail: [email protected] ABSTRACT BACKGROUND: Sensorimotor tests, like cranio-cervical flexion and cervical joint position sense tests, share a strong cognitive component during their execution. However, cognitive training for those tests has not been investigated so far. AIM: The aim of this study was to compare mental and physical exercises for improving the sensorimotor function of the cervical spine. DESIGN: A within-subject design with 16 participants. SETTING: Outpatient physiotherapy center. POPULATION: Patients with chronic neck pain. METHODS: Participants were instructed to perform specific active or mental exercises for the deep and superficial neck flexor muscles. The primary outcomes were cranio-cervical flexion test performance, postural sway, cervical joint position sense and pressure pain threshold. A mixed model analysis was used. RESULTS: The interventions improved cranio-cervical flexion performance (P<0.001), with no difference between actively or mentally-per- formed exercises. Postural sway increased after actively (P<0.01) and mentally (P<0.05) performed deep cervical neck flexor exercises, but not after superficial neck flexor exercises. Mentally-performed superficial neck flexor exercises improved cervical joint position sense when compared to mentally-performed deep cervical flexor exercises (P<0.05), and actively performed superficial neck flexor exercises were effective in improving cervical joint position sense acuity compared to mentally-performed deep cervical flexor exercises (P<0.05) for relocation tasks in the transverse plane. The pressure pain threshold at the cervical spine increased after active deep cervical flexor exercises (P<0.05) and after mental superficial neck flexor exercise (P<0.05). CONCLUSIONS: Mentally-performed deep cervical flexor exercises improved cranio-cervical flexion test performance and pressure pain threshold at the cervical spine, while postural sway increased. Mentally-performed superficial neck flexor exercises improved cervical joint position sense acuity more than mentally-performed deep cervical flexor exercises. CLINICAL REHABILITATION IMPACT: Mentally-performed exercises are recommended in the early stages of rehabilitation to counteract extensive muscle impairment, and these can be incorporated into daily routine. (Cite this article as: Beinert K, Sofsky M, Trojan J. Train the brain: immediate sensorimotor effects of mentally-performed flexor exercises in patients with neck pain. A pilot study. Eur J Phys Rehabil Med 2019;55:63-70. DOI: 10.23736/S1973-9087.18.05118-3) Key words: Neck pain - Neck muscles - Posture - Pain threshold. Patients with chronic neck pain display a large variety developed to evaluate DCF muscle performance.7 Within of sensorimotor impairments such as reduced cervi- this framework, a specific low-load exercise approach has cal joint position sense (CJPS) acuity,1-3 impaired postural been developed to restore DCF muscle function.8 This control4, 5 and reduced activity of the deep cervical flexor specific exercise approach has been demonstrated to be (DCF) muscles.6 The DCF muscles have been extensively effective for immediately increasing CCFT performance9 investigated, and their function can be tested clinically and pressure pain threshold (PPT).9, 10 In the long run, re- with the cranio-cervical flexion test (CCFT) which was storing DCF muscle function decreases pain and improves Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 63

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically BEINERT COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, MENTALLY-PERFORMED EXERCISES FOR NECK PAIN cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. quality of life.11 There is conflicting evidence concerning Table I.—Demographic data of the subjects. Mean value the effect of DCF muscle activation on CJPS,11-13 and it remains unclear whether activation of the DCF muscles Demographics 39±17 has an impact on postural control. An alternative approach 172±7 to actively performed movements is the combination of Age, years motor and action observation, so-called “mental inter- Body height, cm 78±14 vention,” which has gained little attention in the field of Body weight, kg 4.3±1.7 musculoskeletal rehabilitation to date. Mental intervention Usual pain (NRS 0-10) 4.9±5.8 has been demonstrated to improve CJPS in patients with Pain duration 47.4±12.8 chronic neck pain without any immediate effects on PPT.14 16.3±6.9 In patients with shoulder impingement, motor imagery was NPAD-d (0-100 points) found to be sufficient to reduce pain and improve range of CES-D motion.15 In healthy participants, mentally-performed bal- ance exercises improved balance performance.16 However, ria were any neurological or vestibular pathologies, acute it is still unclear whether mental intervention is able to im- ankle sprain, problems with hearing or vision and insuffi- prove DCF muscle performance. Therefore, the aim of the cient German language skills. Patients completed the Ger- present study was to explore potential differences between man version of the Neck Pain Disability Scale (NPAD-d) mentally and actively performed DCF exercises. We also to describe their level of disability and were screened for wanted to compare the effects of specific DCF exercises depressive disorders with the German version of the Cen- with general neck flexor exercises. We hypothesised that ter for Epidemiological Studies Depression Scale (CES- the effects of actively performed exercises would be great- D). Demographic data of the subjects are presented in er compared to mentally-performed exercises in terms of Table I. CJPS, CCFT and postural control. Regarding PPT, we hypothesised a superior outcome for actively performed Intervention exercises compared to mentally-performed exercises, with greater pain reduction for DCF exercises. Four interventions, two actively performed exercises and two mentally-performed exercises, were compared and Materials and methods performed by each participant in random order. Each inter- vention took 5 minutes and was instructed face to face by A within-subject design was chosen for this study. Each a trained physiotherapist with 5 years of professional ex- subject performed two mental and two physical exercis- perience. A break of 20 minute followed each intervention. es. Interventions (see below) were randomly assigned to subjects. Randomization was prepared electronically with Overall, the experimenal procedure took around two QuikCalc Software (GraphPad Software, La Jolla, CA, and a half hours from baseline measurement over all four USA). Subjects were recruited via local advertisement in interventions to the last outcome. Each neck pain patient private physiotherapy centers in the south-west of Ger- completed the trial without reporting any side effects. many. Data collection was performed in an outpatient re- habiliation center. Written informed consent was obtained Activation of deep cervical flexor muscles before participation, and the study was approved by the ethics committee of the University of Landau (Germany) Activation of deep cervical flexor (DCF) muscles was and adhered to the Declaration of Helsinki. performed as previously described by Jull et al.8 The key feature of this exercise is the low-load activation of DCF Subjects muscles by a cranio-cervical flexion motion in a supine position, which is accompanied by sliding of the back of Sixteen neck pain patients (mean age 39±17 years; five the head in the caudal and cranial directions to ensure sag- men and 11 women) participated in this study. Patients ittal rotation and avoid retraction (Figure 1, 2). Palpation were included if they were aged between 18 and 65 years of the superficial flexor muscles was performed to avoid old, had experienced more than nine unilateral or bilat- dominant use of these muscles during cranio-cervical flex- eral neck pain episodes during the past year and had not ion. Patients activated their DCF muscles for 4 seconds reported pain radiation into the arm. The exclusion crite- followed by a 2-second break. Each activity followed by a break (4+2 s) was considered one repetition. Subjects performed six repetitions over 36 s followed by a break for 24 s, which was considered one series. Five series of DCF muscle activation were performed over a time period of 5 minutes. 64 European Journal of Physical and Rehabilitation Medicine February 2019

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, MENTALLY-PERFORMED EXERCISES FOR NECK PAIN BEINERT cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. Figure 1.—Starting position for Figure 2.—End position for ac- Figure 3.—Starting position for ac- Figure 4.—End position for acti- activation of deep cervical flexor tivation of deep cervical flexor tivation of superficial neck flexor vation of superficial neck flexor muscles. muscles. muscles. muscles. Activation of superficial neck flexor muscles lowed by a break for 2 s. Each activity followed by a break (4+2 s) represented one repetition. Subjects performed six Superficial neck flexor muscles were activated in the su- repetitions over 36 seconds followed by a break of 24 s, pine position by lifting the head from the treatment table which was considered one series of mental activation. Five with the chin vertically orientated to avoid a cranio-cervical series of mental superficial neck muscle activation were flexion movement (Figure 3, 4). The head was lifted about performed over a time period of 5 minutes. 2 cm so that the therapist could place his hand between the patient’s head and the treatment table. Subjects acti- Outcome measures vated their superficial neck flexor muscles isometrically for 4 s followed by a break for 2 s. Each activity followed Subjects performed baseline measurements at the begin- by a break (4+2 s) was considered one repetition. Subjects ning of the session prior to any intervention. During the performed six repetitions over 36 seconds followed by a intervention, the outcome assessor left the treatment room. break of 24 s, which was considered one series. Five se- Immediately after intervention, the outcome assessor was ries of superficial neck flexor muscle activation were per- recalled to perform outcome measurement blinded to the formed over a time period of 5 minutes. previously-performed intervention. Mental intervention of DCF muscle activity Cranio-cervical flexion test Subjects were placed in a supine position. A laptop was Subjects were placed in the supine position for their initial placed over the head of the subjects at a distance of 30 cm introduction to the cranio-cervical flexion test (CCFT). from face to monitor (14” Latitude 6220, Dell, Frankfurt Information emphasising the low-load character of the am Mein, Germany). Patients were instructed to observe task was given to the subjects, and they were instructed and imagine the cranio-cervical flexion movement they to avoid superficial flexor muscle activity and retraction saw on the monitor without executing any movement. This movement. After this initial familiarisation, the CCFT mental activation of DCF muscles lasted for 4 s followed was performed. For this purpose, an air-filled pressure by a 2-s break. Each mental activation with a break (4+2 cushion (Stabilizer™, Chattanooga Group Inc., Hixson, s) was considered one repetition. Subjects performed six TN, USA) was placed under the cervical spine, inferior to repetitions over 36 seconds followed by a break of 24 s, the occiput. The air-filled pressure cushion was filled with which was considered one series of mental activation. Five air until 20 mmHg as a starting point. Subjects were in- series of mental DCF muscle activation was performed structed to increase the pressure by 2 mmHg and hold this over a time period of 5 minutes. pressure for 8 s. This stage was successfully passed if sub- jects were able to maintain the pressure without activat- Mental intervention of superficial neck flexor activity ing the superficial muscles, with a tonic contraction of the DCF muscles and with a visible cranio-cervical flexion Subjects were placed in the same position as for the men- movement. For the next stage, subjects had to increase the tal activation of DCF muscle exercise and were instructed pressure to 24 mmHg with the same technical criteria as to observe and imagine the head elevation they observed mentioned above. The test was stopped if subjects reached on the monitor without executing any movement. Subjects 28 mmHg. The CCFT shows intra-class-correlation of mentally activated their superficial neck flexors for 4 s fol- 0.63-0.86.17 Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 65

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically BEINERT COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, MENTALLY-PERFORMED EXERCISES FOR NECK PAIN cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. Cervical joint position sense pedal stance on the platform with their hands placed on their hips and their eyes closed. After familiarisation with To test the cervical joint position sense (CJPS), subjects the task, subjects were instructed to stand as still as pos- were seated in an unsupported upright position on a chair sible on the force platform for 15 s. Three trials of 15 s at a distance of 90 cm away from the target.18 The target were performed for the baseline measurement as well as (adjustable board of 1.5×1 m) was placed parallel to the directly after each intervention. Reliability of static pos- frontal plane of the patient with the center of the board tural assessment revealed good test-retest reliability (ICC positioned at eye level when looking straight. A cervical 0.38-0.90).20 goniometer (CMS; GMON, Stuttgard, Germany) with a laser pointer (P2; NOBO, Berlin, Germany) was placed on Pressure pain threshold the head of the participants. The head position with eyes looking straight was defined as the neutral head position The pressure pain threshold (PPT) was assessed by a (NHP). For testing CJPS, subjects were instructed to ro- trained assessor at the zygapophyseal joint of C2/C3 and tate their head while blindfolded from the NHP to the left at the trigger point of the levator scapulae muscle over the and the right before relocating the head into the NHP. Af- angulus superior of the scapula on the affected side us- ter assessing the NHP, eight repetitions of head relocation ing a pressure gauge device (Algometer, FDN200; Wagner were performed to ensure reliability. Data were obtained Instruments, Greenwich, CT, USA) and a probe area of 1 for rotation to the left, rotation to the right, extension and cm2 (probe diameter of 1.1 cm). The assessor gradually flexion.19 The first five repetitions were intended to get the increased the pressure over the above-mentioned targets subjects accustomed to the joint repositioning task, while until the subject called out “yes,” indicating the beginning the following three repetitions were performed to measure of painful pressure. Three repetitions where performed at the accuracy of joint position sense. The patients verbally the zygapophyseal joint of C2/C3 and at the levator scapu- indicated when they thought they had reached their NHP. lae muscle trigger point. Excellent intra-tester reliability The location of the laser point at that instant was marked, with intra-class-correlations ranging from 0.94-0.97 have and the deviation to the actual target position was mea- been previously reported for this test.21 sured in centimetres. The absolute errors were calculated. The CJPS was tested at baseline and directly after each in- Statistical analysis tervention. After each intervention, three trials were evalu- ated for each direction to avoid further learning. Measuring Statistical analyses were conducted with IBM SPSS Sta- cervical joint position sense has fair to excellent test-retest tistics v. 24 (Chicago, IL, USA). Sample size calculation reliability with an intra-class-correlation of 0.39 to 0.78.19 was based on a previous experiment in which mental inter- vention was performed in neck pain patients with cervical Postural control joint position sense and pressure pain threshold as primary outcomes.14 A Gaussian distribution of the data was con- Postural control was evaluated by measuring oscillations firmed with the Kolmogorov-Smirnov test. Mean values of the center of pressure on a force platform (Wellengang, were calculated for individual subjects then averaged Nürnberg, Germany). Subjects were positioned in a bi- within subjects for CCFT, postural sway, PPT and CJPS. Table II.—Results for each intervention. Parameter DCF Flex MIDCF MIF CCFT, mmHg 1.65±0.47** 1.06±0.47* 1.89±0.47*** 0.83±0.47 Postural sway, mm 39.38±10.73*** 11.06±9.66 23.76±10.14* 17.35±10.14 PPT C2, N 3.47±1.62* 3.06±1.54 1.69±1.57 3.45±1.57 PPT LS, N 4.84±3.77 CJPS transversal, cm 7.77±3.77 -0.84±0.72 6.13±3.77 4.05±3.77 CJPS sagittal, cm -0.11±0.82 -0.09±0.74 0.28±0.75 -1.17±0.75 -1.29±0.72 -0.63±0.73 -1.14±0.73 Data are presented as the mean difference between baseline and intervention and the standard deviation of the mean. DCF: actively performed deep cervical flexor exercise; Flex: actively performed superficial neck flexor exercise; MIDCF: mentally-performed deep cervical flexor exercise; MIF: mentally-performed superficial neck flexor exercise; CCFT: cranio-cervical flexion test; PPT C2: pressure pain threshold at zygapophyseal joint C2/3; PPT LS: pressure pain threshold at the trigger point of the levator scapulae; CJPS transversal: cervical joint position sense acuity for relocation after rotation right and left; CJPS sag: cervical joint position sense acuity for relocation after extension and flexion. *α<0.05; **α<0.01; ***α<0.001. 66 European Journal of Physical and Rehabilitation Medicine February 2019

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, MENTALLY-PERFORMED EXERCISES FOR NECK PAIN BEINERT cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. CCFT (mmHg) CJPS sagittal (cm) CJPS transversal (cm) BL aDCF aSNF mDCF mSNF BL aDCF aSNF mDCF mSNF BL aDCF aSNF mDCF mSNF PPT AS (N) PPT C2/C3 (N) PC (mm) BL aDCF aSNF mDCF mSNF BL aDCF aSNF mDCF mSNF BL aDCF aSNF mDCF mSNF Figure 5.—Results of mentally and physically performed exercises on cranio-cervical flexion task, cervical joint position sense, postural stability and pressure pain thresholds. CCFT: cranio-cervical flexion test; CJPS: cervical joint position sense; PPT AS: pressure pain threshold angulus superior; PPT C2/3: pressure pain threshold at zygapophyeal joint of the second to third cervical vertebrae; PC: postural control measured by oscillations of the center of pressure, BL: baseline; aDCF: actively-performed deep cervical flexor exercise; aSNF: actively-performed superficial neck flexor exercise; mDCF: mentally- performed deep cervical flexor exercise; mSNF: mentally-performed superficial neck flexor exercise. With respect to CJPS, data were calculated separately for Results relocation of the NHP in the sagittal and transversal planes and were averaged within subjects and across subjects. All 16 patients successfully completed the pre-test mea- To identify potential differences between measurements surements, interventions and post-test measurements. performed at baseline and post-intervention, as well as Neck pain patients showed moderate values on the Neck between interventions, a mixed model analysis was cal- Pain Disability Scale and had no depressive disorders (Ta- culated for each dependent variable with treatment type ble I). (DCF, Flex, MIDCF, and MIF) as the fixed factor and the order and subject as random factors. In case of significant The mixed model calculations revealed a significant F values, LSD-corrected post-hoc tests were performed. treatment effect of actively and mentally-performed ex- For all tests, significant differences were regarded as P val- ercises (F(4, 31.19)=6.36; P<0.001) for the CCFT. The LSD ues below 0.05. post-hoc analysis of each intervention showed a significant improvement in CCFT after performing the DCF exercise Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 67

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically BEINERT COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, MENTALLY-PERFORMED EXERCISES FOR NECK PAIN cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. (mean difference 1.65; P<0.01, 95% CI: 0.69 to 2.60) perficial flexor exercise increased acuity. The difference and improved CCFT after the superficial flexor exercise between exercises was significant (mean difference -1.45; (mean difference 1.06; P<0.05, 95% CI: 0.110 to 2.01). P<0.05, 95% CI: 0.37 to 2.53), as well as the difference Mentally-performed DCF significantly improved CCFT between the actively performed superficial neck flexor ex- (mean difference 1.89; P<0.01, 95% CI: 0.940 to 2.83), ercise compared to the mentally-performed DCF exercises but mentally-performed superficial flexor exercise had no (mean difference -1.12; P<0.01, 95% CI: -0.18 to -2.07) significant effect on CCFT (mean difference 0.83; P<0.1, (Table II, Figure 5). 95% CI: -0.11 to 1.78). Between exercise comparison re- vealed no significant differences between interventions for No treatment effect was observed for CJPS in the sagit- CCFT performance (Table II, Figure 5). tal plane (F(4, 22.39)=1.50; not significant). Furthermore, no significant effects were found for a single intervention or Regarding the effects on postural sway, a significant pairwise comparison (Table II, Figure 5). increase in postural sway was observed after treatment (F(4, 24.02)=4.01; P<0.05). The LSD post-hoc analysis of Discussion each intervention showed a significant increase in static postural sway after the DCF exercise (mean difference The aim of the present study was to explore potential dif- 39.58; P<0.01, 95% CI: -61.16 to -17.98), but the super- ferences between mentally and actively performed exer- ficial flexor exercise had no effect on postural sway (not cises, as well as the differences between superficial flexor significant, 95% CI: -8.48 to 30.60). After the mentally- exercise and DCF exercises on CCFT, static postural sway, performed DCF exercise, postural sway increased sig- PPT and CJPS in subjects with neck pain. With respect nificantly (mean difference 23.76; P<0.01, 95% CI: 3.34 to CCFT performance, both actively performed specific to 44.18). Mentally-performed superficial flexor exercise DCF exercises and superficial neck flexor exercises im- had no effect on postural sway (not significant, 95% CI: proved CCFT performance. Furthermore, while mentally- -3.06 to 37.76). Between exercise comparison revealed a performed DCF exercises improved CCFT performance, significantly greater increase in postural sway after DCF mentally-performed superficial neck flexor exercises did compared to the superficial flexor exercise (mean differ- not. No differences between actively and mentally-per- ence 28.52; P<0.05, 95% CI: 3.93 to 53.12). No further formed interventions were observed. Our initial hypoth- differences were observed between mentally and actively esis that actively performed exercises creates greater ef- performed exercises (Table II, Figure 5). fects compared with mentally-performed exercises must therefore be rejected. No treatment effects were observed for PPT at the level of C2 (F(4, 30.41)=1.98; not significant), nor for PPT of the The absence of differences between deep and super- levator scapulae muscle (F(4, 33.23)=1.66; not significant). ficial muscle action on CCFT performance is in contrast However, pairwise comparison revealed a significant an- to the results of Jull et al.,22 who demonstrated increased algesic effect of DCF compared to baseline (mean dif- DCF activity at all stages of the CCFT after DCF exercis- ference 3.47; P<0.05, 95% CI: 0.10 to 6.83) for PPT at es, an effect which was not observable in their superficial C2. Furthermore, the mentally-performed superficial neck exercise group. This difference may partly be explained by flexor exercise induced a significant increase in PPT at C2 the time period of the intervention. While we looked for compared to baseline (mean difference 3.45; P<0.05, 95% immediate effects, Jull et al. obtained results after 6 weeks CI: 0.18 to 6.73). When DCF and the mentally-performed with six sessions of exercise training.22 superficial neck flexor exercises were compared, there were no differences between treatments for PPT of C2 (not The present finding also demonstrates improved CCFT significant, 95% CI: -2.27 to 2.25) (Table II, Figure 5). performance after mentally-performed DCF exercise. Pairwise comparison of treatments for PPT at the levator It is well established that mentally-performed exercises scapulae muscle revealed no differences. and real task execution activate similar brain areas23 and share overlapping neural activation.24 The absence of Actively and mentally-performed exercises had an differences between physically executed exercises and overall treatment effect on CJPS in the transversal plane mentally-performed exercises may be partly explained (F(4, 26.28)=2.67; P<0.05). Furthermore, significant differ- by the proposed overlapping brain activity between task ences were observed between the treatments. The men- execution and mentally-performed exercises. Further, this tally-performed DCF exercise decreased CJPS acuity in result highlights the cognitive aspect of this specifically the transverse plane, while the mentally-performed su- designed DCF exercise and offers new possibilities for the 68 European Journal of Physical and Rehabilitation Medicine February 2019

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, MENTALLY-PERFORMED EXERCISES FOR NECK PAIN BEINERT cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. early stage of neck pain rehabilitation when head on neck close to a pain-triggering motion, while in the present movements are usually very painful. In these cases, men- study participants were positioned in supine which likely tally-performed exercises could counteract the extensive caused no interference to their pain-triggering motion. impairment of DCF muscles. The starting position and the relationship of the mentally- performed exercise to the pain-triggering movements of Postural sway increased significantly after the interven- the participants might be a critical factor determining the tions. Interestingly, both actively performed DCF exer- effect on PPT, especially as functional magnetic resonance cises and mentally-performed DCF resulted in significant imaging has shown activation of pain-related centers when increase in postural sway, whereas the superficial neck the pain-provoking movement was performed mentally.28 flexor exercise had no effect, irrespective of whether ac- The underlying mechanisms beyond the analgesic effect tively or mentally-performed. Fatigue of neck muscles of mentally-performed exercises remains speculative, but in healthy subjects has been reported to increase postural in the absence of motion execution, a top down inhibition sway,25, 26 but these results were always caused by neck might be probable. PPT at the trigger point of the levator extensor fatigue, not by neck flexor fatigue. Furthermore, scapulae muscle revealed no analgesic response to active- neither of the mentioned studies focused on superficial ly or mentally-performed exercises. versus deep muscles. In patients with neck pain, fatigue of neck flexor muscles increased postural sway.27 Patients in An overall treatment effect was found for CJPS acuity in the latter study performed isometric neck flexor contrac- the transverse plane, but no single intervention resulted in tions without emphasising superficial versus deep flexor significant difference compared to baseline performance. muscles. The present results question whether increased However, a closer look revealed significant differences postural sway after neck muscle fatigue is related to super- between the interventions. The mentally-performed su- ficial versus deep neck muscles. Furthermore, the role of perficial neck flexor exercise improved CJPS acuity in the muscle fatigue should be assessed in more detail as a key sagittal plane more than the active DCF exercises. Further- feature of activation of DCF muscles is the low-load ap- more, actively performed superficial neck flexor exercises proach, which should not cause fatigue to the same extent improved CJPS acuity more than mentally-performed DCF as high-load, exhausting muscle performance. exercises. Finally, mentally-performed superficial neck flexor exercise improved CJPS acuity while mentally- Mentally-performed DCF also increased postural sway. performed DCF decreased CJPS acuity. Thus, the results Whether this result relies on an altered head-to-trunk per- of the present study suggest no effect of DCF exercises ception or orientation in space due to the nature of the on CJPS, irrespective of whether they were performed ac- mentally-performed exercises remains unclear. tively or mentally. Mentally-performed neck rotation exer- cises have previously been demonstrated to improve CJPS No overall treatment effects were found for PPT at acuity,14 confirming the present results. No treatment effect C2; however, both actively performed DCF exercises and and no between-treatment differences were found for CJPS mentally-performed superficial flexor exercises signifi- acuity in the sagittal plane. This is somewhat surprising as cantly increased PPT at C2 compared to baseline. Thus, our interventions mainly involved actively and mentally- our initial hypothesis that actively performed exercises performed movements in the sagittal plane. exert greater analgesic effects compared to mentally-per- formed exercises has to be rejected. A trend was observed Limitations of the study for actively performed superficial neck exercises, but not for mentally-performed DCF. Regarding the actively per- A limitation of the present study is the within-subject de- formed DCF exercise, the present results are in line with sign. Further, the results must be considered with caution previous reports.9, 10 A new finding of the current study due to the limited sample size and due to the fact that only is the analgesic effect of mentally-performed superficial immediate effects were measured. neck flexor exercises on PPT. This is in contrast to a previ- ous experiment, in which mentally-performed CJPS test- Conclusions ing failed to demonstrate an analgesic effect on PPT.14 However, the mentally-performed exercises in the present Actively performed superficial neck flexor exercises as study differ from those performed in this previous study well as actively and mentally-performed specific DCF ex- with regard to the direction of motion and training posi- ercises are capable of causing an immediate improvement tion. In our previous study, participants performed the in CCFT performance. Mentally-performed DCF exercises mental exercise in a sitting position which may have been Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 69

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically BEINERT COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, MENTALLY-PERFORMED EXERCISES FOR NECK PAIN cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. may be of specific value in the early rehabilitation stages cal joint position sense: the effect of two exercise regimes. J Orthop Res when sensitised neck structures do not allow much active 2007;25:404–12. movement. Furthermore, mentally-performed exercises 12.  Armstrong BS, McNair PJ, Williams M. Head and neck position sense are easy to incorporate into daily office life, when people in whiplash patients and healthy individuals and the effect of the cranio- are working at a computer. Regarding postural sway, fu- cervical flexion action. Clin Biomech (Bristol, Avon) 2005;20:675–84. ture research should investigate potential differences in the 13.  Gallego Izquierdo T, Pecos-Martin D, Lluch Girbés E, Plaza-Man- influence of deep versus superficial exercises. The effects zano G, Rodríguez Caldentey R, Mayor Melús R, et al. Comparison of of mentally-performed exercises in patients with dizziness cranio-cervical flexion training versus cervical proprioception training in should also be established. patients with chronic neck pain: A randomized controlled clinical trial. J Rehabil Med 2016;48:48–55. Future studies should evaluate additional effects of 14.  Beinert K, Preiss S, Huber M, Taube W. Cervical joint position sense mentally-performed exercises. Because of their easy inte- in neck pain. Immediate effects of muscle vibration versus mental training gration into daily routine activities, the intervention time interventions: a RCT. Eur J Phys Rehabil Med 2015;51:825–32. could be increased. Furthermore, integration of the target 15.  Hoyek N, Di Rienzo F, Collet C, Hoyek F, Guillot A. The therapeutic muscle into daily activities could be facilitated, potentially role of motor imagery on the functional rehabilitation of a stage II shoul- leading to better outcomes. der impingement syndrome. Disabil Rehabil 2014;36:1113–9. 16.  Taube W, Lorch M, Zeiter S, Keller M. Non-physical practice im- References proves task performance in an unstable, perturbed environment: mo- tor imagery and observational balance training. Front Hum Neurosci 1.  Armstrong B, McNair P, Taylor D. Head and neck position sense. 2014;8:972. Sports Med 2008;38:101–17. 17.  Jørgensen R, Ris I, Falla D, Juul-Kristensen B. Reliability, construct 2.  de Zoete RM, Osmotherly PG, Rivett DA, Farrell SF, Snodgrass SJ. and discriminative validity of clinical testing in subjects with and without Sensorimotor Control in Individuals With Idiopathic Neck Pain and chronic neck pain. BMC Musculoskelet Disord 2014;15:408. Healthy Individuals: A Systematic Review and Meta-Analysis. Arch Phys 18.  Revel M, Andre-Deshays C, Minguet M. Cervicocephalic kines- Med Rehabil 2017;98:1257–71. thetic sensibility in patients with cervical pain. Arch Phys Med Rehabil 3.  de Vries J, Ischebeck BK, Voogt LP, van der Geest JN, Janssen M, 1991;72:288–91. Frens MA, et al. Joint position sense error in people with neck pain: A 19.  Pinsault N, Fleury A, Virone G, Bouvier B, Vaillant J, Vuillerme N. systematic review. Man Ther 2015;20:736–44. Test-retest reliability of cervicocephalic relocation test to neutral head po- 4.  Ruhe A, Fejer R, Walker B. Altered postural sway in patients suffering sition. Physiother Theory Pract 2008;24:380–91. from non-specific neck pain and whiplash associated disorder - A system- 20.  Quatman-Yates CC, Lee A, Hugentobler JA, Kurowski BG, Myer atic review of the literature. Chiropr Man Therap 2011;19:13. GD, Riley MA. Test-retest consistency of a postural sway assessment pro- 5.  Silva AG, Cruz AL. Standing balance in patients with whiplash- tocol for adolescent athletes measured with a force plate. Int J Sports Phys associated neck pain and idiopathic neck pain when compared with as- Ther 2013;8:741–8. ymptomatic participants: A systematic review. Physiother Theory Pract 21.  Walton DM, Macdermid JC, Nielson W, Teasell RW, Chiasson M, 2013;29:1–18. Brown L. Reliability, standard error, and minimum detectable change of 6.  Falla D, O’Leary S, Farina D, Jull G. Association between intensity of clinical pressure pain threshold testing in people with and without acute pain and impairment in onset and activation of the deep cervical flexors in neck pain. J Orthop Sports Phys Ther 2011;41:644–50. patients with persistent neck pain. Clin J Pain 2011;27:309–14. 22.  Jull GA, Falla D, Vicenzino B, Hodges PW. The effect of therapeutic 7.  Jull GA, O’Leary SP, Falla DL. Clinical assessment of the deep cervi- exercise on activation of the deep cervical flexor muscles in people with cal flexor muscles: the craniocervical flexion test. J Manipulative Physiol chronic neck pain. Man Ther 2009;14:696–701. Ther 2008;31:525–33. 23.  Lacourse MG, Orr EL, Cramer SC, Cohen MJ. Brain activation dur- 8.  Jull G, Sterling M, Falla D, Treleaven J, O’Leary S. Whiplash, Head- ing execution and motor imagery of novel and skilled sequential hand ache, and Neck Pain. Edinburgh: Chruchill Livingstone; 2008. movements. Neuroimage 2005;27:505–19. 9.  Lluch E, Schomacher J, Gizzi L, Petzke F, Seegar D, Falla D. Im- 24.  Hétu S, Grégoire M, Saimpont A, Coll MP, Eugène F, Michon PE, et mediate effects of active cranio-cervical flexion exercise versus passive al. The neural network of motor imagery: an ALE meta-analysis. Neurosci mobilisation of the upper cervical spine on pain and performance on the Biobehav Rev 2013;37:930–49. cranio-cervical flexion test. Man Ther 2014;19:25–31. 25.  Schieppati M, Nardone A, Schmid M. Neck muscle fatigue affects 10.  O’Leary S, Falla D, Hodges PW, Jull G, Vicenzino B. Specific thera- postural control in man. Neuroscience 2003;121:277–85. peutic exercise of the neck induces immediate local hypoalgesia. J Pain 26.  Gosselin G, Rassoulian H, Brown I. Effects of neck extensor muscles 2007;8:832–9. fatigue on balance. Clin Biomech (Bristol, Avon) 2004;19:473–9. 11.  Jull G, Falla D, Treleaven J, Hodges P, Vicenzino B. Retraining cervi- 27.  Cheng CH, Chien A, Hsu WL, Yen LW, Lin YH, Cheng HY. Changes of postural control and muscle activation pattern in response to external perturbations after neck flexor fatigue in young subjects with and without chronic neck pain. Gait Posture 2015;41:801–7. 28.  Beinert K, Mouthon A, Keller M, Mouthon M, Annoni JM, Taube W. Neural Correlates of Maladaptive Pain Behavior in Chronic Neck Pain - A Single Case Control fMRI Study. Pain Physician 2017;20:E115–25. Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript. Authors’ contributions.—Konstantin Beinert has provided idea, data analysis, and manuscript writing. Marc Sofsky has provided data collection and manu- script writing. Jörg Trojan has provided idea, data analysis and manuscript writing. Article first published online: May 9, 2018. - Manuscript accepted: May 8, 2018. - Manuscript revised: April 24, 2018. - Manuscript received: December 15, 2017. 70 European Journal of Physical and Rehabilitation Medicine February 2019

COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically © 2018 EDIZIONI MINERVA MEDICA European Journal of Physical and Rehabilitation Medicine 2019 February;55(1):71-8 or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access Online version at http://www.minervamedica.it DOI: 10.23736/S1973-9087.18.04977-8 to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. ORIGINAL ARTICLE Pain extent is more strongly associated with disability, psychological factors, and neck muscle function in people with non-traumatic versus traumatic chronic neck pain: a cross sectional study Inge RIS 1, 2 *, Marco BARBERO 3, Deborah FALLA 4, Mads H. LARSEN 1, Martin N. KRAFT 1, Karen SØGAARD 5, Birgit JUUL-KRISTENSEN 1, 6 1Research Unit of Clinical Biomechanics, Department of Sports Science and Clinical Biomechanics, University of Southern Denmark, Odense, Denmark; 2Metropolitan University College, Copenhagen, Denmark; 32rLab Rehabilitation Research Laboratory, Department of Business, Health and Social Care, University of Applied Sciences and Arts of Southern Switzerland (SUPSI), Manno, Switzerland; 4Center of Precision Rehabilitation for Spinal Pain (CPR Spine), School of Sport, Exercise and Rehabilitation Sciences, College of Life and Environmental Sciences, University of Birmingham, Birmingham, UK; 5Research Unit for Physical Activity and Health in Working Life, Department of Sports Science and Clinical Biomechanics, University of Southern Denmark, Odense, Denmark; 6Institute of Occupational Therapy, Physiotherapy and Radiography, Department of Health Sciences, Bergen University College, Bergen, Norway *Corresponding author: Inge Ris, University of Southern Denmark, Campusvej 55, 5230 Odense M, Denmark. E-mail: [email protected] ABSTRACT BACKGROUND: Neck pain is presented with a variety of symptoms. Pain drawings are used in the clinical assessment of people with neck pain. Pain extent based upon pain drawings can be associated with different factors. However, the relation between pain extent and function limitations in people with neck pain is unknown. AIM: The aim of this study was to explore the associations between pain extent extracted from pain drawings, and self-reported neck pain related disability, quality of life, depression, self-reported neck function, cervical muscle function, and range of motion in a chronic neck pain population and possible differences depending on the onset of pain being traumatic or not. DESIGN: Observational cross-sectional study. SETTING: Primary and secondary healthcare. POPULATION: People with chronic neck pain (N.=200) of traumatic (N.=120) or non-traumatic (N.=80) origin. METHODS: Outcome measures: Pain extent, Short Form 36 Health Survey Physical and Mental Component Summary (SF36-PCS/MCS), TAMPA Scale of Kinesiophobia (TSK), Beck Depression Inventory-II (BDI-II), Neck Disability Index (NDI), Craniocervical Flexion Test (CCFT), Cervical Extension Test (CE), and Cervical Range of Motion (ROM). Correlations were calculated using Spearman or Pearson correla- tion coefficients. Correlation between pain extent and outcomes were calculated for all participants collectively and then separately for those with a traumatic versus non-traumatic neck pain. RESULTS: Overall, significant positive correlations were observed between pain extent and NDI (r=0.33; P<0.001), BDI-II (r=0.29; P<0.001), CCFT (r=-0.24; P=0.001) and CE (r=-0.19; P=0.006). No difference was observed in pain extent between patients with traumatic (mean: 7.6±6.7%) and non-traumatic onset (7.4±6.8%). Pain extent correlated moderately with NDI, BDI-II, TSK, CCFT and CE in those with non- traumatic onset, but weakly with NDI, BDI-II, CCFT and CE in those with trauma-induced chronic neck pain. CONCLUSIONS: Pain extent is correlated with patient-reported neck function, depression and muscle test performance in people with chronic neck pain. These correlations were strongest in those with non-traumatic neck pain. CLINICAL REHABILITATION IMPACT: Patients presenting with larger pain areas show poorer psychological and physical function. Pain drawings can therefore indicate a need for addressing these functions in management of a person with chronic neck pain. (Cite this article as: Ris I, Barbero M, Falla D, Larsen MH, Kraft MN, Søgaard K, et al. Pain extent is more strongly associated with disability, psychological factors, and neck muscle function in people with non-traumatic versus traumatic chronic neck pain: a cross sectional study. Eur J Phys Rehabil Med 2019;55:71-8. DOI: 10.23736/S1973-9087.18.04977-8) Key words: Neck pain - Pain measurement - Wounds and injuries - Trauma centers. Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 71

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, RIS PAIN EXTENT IN CHRONIC NECK PAIN cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. Neck pain is a common condition ranking fourth high- groups (traumatic or non-traumatic), as well as the asso- est as a cause of years lived with disability.1 Patients ciations between pain extent and the selected variables for with chronic neck pain often present with a variety of other each group separately. signs and symptoms. These may include physical impair- ments as decreased cervical mobility and altered neuro- Materials and methods muscular function of the flexor2 and extensor3 muscles, as well as psychological distress, depression and fear avoid- Design ance,4 and reduced quality of life.5 The extent of some of these factors may depend on the origin of pain with differ- This study is an observational cross-sectional study and ences observed between those with traumatic versus non- conducted as secondary analysis of baseline data from a traumatic neck pain.6-8 For instance, certain sub-groups randomized paralleled two-group trial.15 The trial was reg- of patients with chronic neck pain typically present with istered at www.ClinicalTrials.gov (NCT01431261). The more widespread pain, i.e. those with neck pain induced Regional Scientific Ethics Committee of Southern Den- by a whiplash injury9 and those with neck pain with neu- mark approved the study (S-20100069) 21st February 2011 ropathic features. (Chairman Birger Møller). Participants were recruited in the period of March 2012 to September 2014. This study Pain drawings are widely used in the clinical assess- conformed to The Declaration of Helsinki 2008, fulfilling ment of people with neck pain10 as they allow the patient all ethical requirements and reporting of the present study to describe their perceived pain localization and pain ex- is guided by the STROBE checklist.17 Participants provid- tent. Larger pain extent may reflect biological factors (e.g. ed written informed consent. central sensitization, and inflammatory system response), and psychological factors, and may be associated with Settings poorer neck function. Recent work showed that pain extent is associated with greater disability, depression and lower The patients were recruited from both primary (eight phys- self-efficacy in individuals with whiplash-associated dis- iotherapy clinics) and secondary health care locations (two orders.11 Moreover, larger pain extent is associated with hospital spine centers, one hospital outpatient clinic and higher pain intensity in women with fibromyalgia.12 An as- one municipal rehabilitation center). The recruitment sites sociation between the extent of pain and pain catastroph- were in urban or rural areas and spread across the middle izing has also been observed in people with neck-shoulder and western parts of Denmark. Both spine centers and the pain.13 Pain extent is recorded as a factor related to poorer rehabilitation center were specialized in treating patients physical functioning.14 However, no study has evaluated with musculoskeletal spinal disorders. The hospital out- the relation between pain extent and function limitations patient clinic was specialized in treating neurological and in people with neck pain. Based upon our hypothesis that headache patients. Patients recruited from the physiother- larger pain extent would be associated with poorer psy- apy clinics had been referred from general practitioners chological and physical functioning in people with chronic for physiotherapy treatment. Patients recruited from sec- neck pain, we analyzed pain extent and its association with ondary health care centers had been referred from general psychological health, quality of life, and neck function as practitioners, chiropractors or specialized medical doctors a secondary analysis to a recently reported randomized to the involved centers. controlled study.15 Additionally, we hypothesized that pain extent would be larger and the associations stronger in Participants people with trauma-induced neck pain versus those with non-traumatic neck pain given that people with neck pain Patients were informed about the option of participating of traumatic onset typically report greater pain severity.8, 16 via in-clinic advertisements, by their treating clinician or Thus the first objective of this study was to evaluate as- at their first contact with the health care unit. Inclusion cri- sociations between pain extent and quality of life, fear teria were minimum age of 18 years, neck pain for at least avoidance, depression, self-reported and measured neck six months with either traumatic or non-traumatic onset, function, and cervical range of motion in people with neck reduced physical neck function (defined as Neck Disabil- pain regardless of the origin of their pain (traumatic and ity Index score of at least 10 out of 50), completed diag- non-traumatic onset of neck pain) and the second objec- nostic procedures (e.g. medical investigations, diagnostic tive to evaluate differences in pain extent between the two imaging), ability to read and understand Danish. Patients were allowed to have pain in other body regions as long 72 European Journal of Physical and Rehabilitation Medicine February 2019

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, PAIN EXTENT IN CHRONIC NECK PAIN RIS cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. as the primary pain area was their neck. Exclusion criteria Area 21,459 pixels were radiculopathies (clinically tested by: positive Spurl- (15% body chart) ing test, relief on cervical traction and positive brachial plexus provocation test on the affected side), engagement in experimental or progressive medical treatment, unstable social and/or working situation, pregnancy, known current fractures, severe depression (defined as Beck Depression Index score over 29),18 or other known co-existing medi- cal conditions. Assessment procedures Local physiotherapists screened patients with neck at their AB first contact with the clinic for eligibility as participant of the study using a checklist for inclusion and exclusion cri- Figure 1.—Illustration of the process of pain drawing digitalization teria as described above, being trained in the initial screen- starting from the paper body chart: A) an original pain drawing; B) final ing procedure by the first author (IR). After inclusion, two digitalized version used to calculate pain extent. physiotherapists assessed the patients. A standardized test protocol was used for all assessments; the physical tests body chart area (Figure 1). The current procedure of digi- were conducted in the same order for all patients followed talizing pain drawings has been previously described and by completion of the questionnaires within the same ses- its reliability confirmed.20, 21 Pain extent was computed us- sion. All clinical tests are described in detail as well as ing software that was developed and tested in a previous the inter- and intra-reliability, construct and discriminative study.10 validity and minimal detectable changes.19 Outcome measures Age, sex, duration of neck pain, employment status, edu- Patient-reported outcomes cational level and sleep pattern were registered as socio- demographic characteristics. Quality of life was measured with the Short Form 36 Health Survey, a generic questionnaire of functioning and Pain extent well-being, with high reliability and validity and tested both in general and disease-specific populations in Den- The patients completed a pain drawing by shading their mark.22, 23 The Standardized Physical Component Summa- perceived location and extent of pain on two paper body ry (SF36-PCS) and Mental Component Summary Scores charts (frontal and dorsal view of the body) using a pen- (SF36-MCS) were generated. cil. Patients were instructed to draw their perceived pain over the last two weeks. Subsequently, all completed paper Fear avoidance was measured with the TAMPA Scale body charts were digitized at 144 dpi using a commercial- of Kinesiophobia (TSK), a questionnaire with 17-items, ly available scanner. The shaded areas were encircled and measuring fear of movement, injury and re-injury (range: copied onto a digital body chart by two trained operators 17-68). A score above 37 indicates a high degree of kine- using image analysis software (Inkscape version 0.48.5). siophobia.24 TSK demonstrates adequate internal consis- The digital marking of the pain drawing was performed tency (Cronbach’s alpha range: 0.70-0.81) and good test- according to the following guidelines: an area marked retest reliability (r=0.78). The test shows acceptable con- digitally inside one continuous line indicated a pain area; current validity, correlating TSK with other self-reported in case of a single line, the area was marked by drawing measures of pain-related fear (r values ranging from 0.54 closely to the line on both sides. The density of shading to 0.60) and construct validity, tested among Danish work- of the marked area was not taken into account. Any area ers.25 of pain drawn outside of the body chart borders was not included in the analysis. Based on automatic pixel count- Depression was measured with the Beck Depression ing, the marked area was expressed as a percentage of total Inventory-II (BDI-II), a 21-item questionnaire of depres- sive symptoms including items assessing both cognitive and somatic complaints associated with depression. All the Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 73

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, RIS PAIN EXTENT IN CHRONIC NECK PAIN cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. scores are summed into one score, ranging from 0-63, with Cervical range of movement (ROM) was measured for a total score less than 14 indicating minimal or no depres- flexion-extension, right-left lateral flexion, and right-left sion, 14-19 mild depression, 20-28 moderate depression, rotation with the patient in upright sitting. An inclinometer and >28 severe depression.26 The BDI-II has high validity was used to measure flexion, extension and lateral flexion and reliability in measuring depressive symptoms, includ- in degrees. For rotation, a semi-circular goniometer was ing acceptable test–retest reliability (r=0.79) in a non-clin- placed upon the patients’ shoulders measuring cervical ro- ical population and outpatient population.27 tation to the nearest five degrees. Self-reported neck pain and disability was measured All of the above clinical tests are fully described else- with the Neck Disability Index (NDI); scores range from where, and their intra- and inter-examiner reliability valu- 0-50, with higher scores representing greater disability. ated in the settings used in this study (intraclass correla- The overall score (out of 50) was calculated by summing tions for CCFT: 0.63-0.86; CE: 0.75-0.90; ROM: 0.80- the responses to each individual item. The internal con- 0.94).19 sistency is acceptable (Cronbach’s alpha 0.864) and test- retest reliability high (intra-class correlation 0.94).28 The Statistical analysis questionnaire is found to be valid and reliable in primary care in Denmark.29 No formal a priori sample size calculation was conducted specifically for the present analyses, as the current data Clinical tests represent baseline data from a clinical trial exploring the effect of training and exercises in chronic neck pain pa- Cervical muscle function was tested with the craniocer- tients.15 A detailed description described how potential bi- vical flexion test (CCFT) to evaluate deep cervical flexor ases in data collection were addressed are described else- muscle function30 and the cervical extensor test (CE), to where.15 evaluate cervical extensor muscle function.19 All continuous data were checked for normality using The CCFT is a standardized test that consists of five Shapiro Wilks test and QQ plots. Data of SF36-MCS, incremental movements of increasing craniocervical flex- SF36-PCS and TSK were normally distributed, whereas ion range of motion. Performance was guided by visual data of BDI-II and NDI, and data of CCFT, CE, ROM, and feedback from an air-filled pressure sensor (Stabilizer™, pain extent were not normally distributed. Correlations Chattanooga Group Inc., Hixson, TN, USA) placed sub- were calculated using Spearman or Pearson correlation occipitally behind the subject’s neck and initially inflated coefficients depending on the distribution of the data. Cor- to a baseline pressure of 20 mmHg. The test was stopped relation between pain extent and SF36-PCS, SF36-MCS, at the point that the patient displayed compensatory strate- TSK, BDI-II, NDI, ROM, and CCFT and CE were calcu- gies (e.g. retraction or over activity of the superficial flex- lated for all participants collectively and then separately ors) and the pressure level below this level was registered. for those with a traumatic versus non-traumatic neck pain. The CCFT performance was divided into three categories: A weak statistical correlation was considered with correla- 22 mmHg, 24 mmHg, and 26-30 mmHg. tion coefficients <0.3, a moderate correlation from 0.3-0.5 and a strong correlation >0.5.31 Statistical significance was The CE test measured the duration that the patient could set to α=0.05. All statistical analyses were carried out with hold the head extended, while lying in prone with the head the Statistical Package for Social Sciences (v. 22.0.0, IBM, and neck over the edge of the bed. The participant per- New York, NY, USA). Missing data were not entered in formed cervical extension while keeping the craniocervical the analyses. region in a neutral position until extension in the middle or upper cervical column occurred. The participant wore a Results headband with a laser light placed on the top of their head. The laser beam was projected to a target on the floor and Patients’ characteristics the participant was instructed to hold the same position with the laser beam in the center of the target for as long as A total of 200 patients were included: 120 with neck possible. The test was terminated after 120 s or when the la- pain of traumatic origin and 80 with neck pain of non- ser beam moved away from the center of the target, and du- traumatic origin. The characteristics of both groups are ration was recorded in seconds. Each participant performed fully described in a former study.8 Briefly, the majority in the test three times with a rest period of less than five min- both groups were women (75%), those with neck pain of utes between repetitions and the best result was recorded.30 74 European Journal of Physical and Rehabilitation Medicine February 2019

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, PAIN EXTENT IN CHRONIC NECK PAIN RIS cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. traumatic onset were on average 4 years younger (mean: origin had an average pain extent of 7.6±6.7% whereas 43.5±11.4 years) compared to those in the non-traumatic those with non-traumatic origin had an average pain extent group (47.6±11.4 years) but this difference was not sig- of 7.4±6.8%. nificant. The average duration of symptoms was almost 9.0 years (107.8±101.8 months), however, those with neck For the total cohort, mean scores of the SF36-PCS and pain of traumatic origin had significantly (P<0.01) shorter SF36-MCS were 35.4±7.6 and 46.3±10.7, respectively. duration of symptoms (88.1±89.2 months) compared to BDI-II was 13.9±8.7 indicating mild depression, TSK those with non-traumatic origin (138.3±112.7 months). score was 37.8±6.9, indicating a high degree of kinesio- Both groups had the same educational level, with more phobia,32 and the average NDI score was 21.7±7.7, indi- than 80% skilled, 8% without education, and almost 11% cating moderate neck disability (Table I).33 Performance having an academic level. The unemployment rate was outcomes of the clinical tests for the entire cohort are pre- similar in both groups (8% and 9% for those with trau- sented in Table I. matic versus non-traumatic onset, respectively). In both groups, more than half were occupationally active: 54% Correlation between pain extent and other clinical fea- of those with neck pain of traumatic origin and 61% of tures those with non-traumatic origin. The remaining partici- pants were retired, on sick leave or undertaking education. Across the entire cohort, NDI was moderately correlated Across both groups, 67% experienced sleep disturbances. to pain extent (r=0.33; P<0.001) indicating that larger pain areas were associated with higher NDI scores. This cor- Across both groups the pain extent was 7.5±6.7% of the relation was stronger for the group with non-traumatic total body chart area; those with neck pain of traumatic neck pain (r=0.45, P<0.001) compared to those with neck pain of traumatic origin (r=0.22, P=0.02). Pain extent was Table I.—P atient characteristics self-reported outcomes and clini- also weakly correlated to both BDI-II and TSK (r=0.29; cal tests. P<0.001 and r=0.21; P<0.05, respectively), indicating that larger pain extent was associated with higher depression Parameters All patients Traumatic Non-traumatic and greater kinesiophobia scores across the entire group. (N.=200) onset (N.=120) onset (N.=80) Those with non-traumatic neck pain displayed a moder- ate correlation between pain extent and both variables Patient-reported outcomes 35.4±7.6 35.0±7.1 36.1±8.2 SF36-PCS 46.4±10.8 44.7±11.2 49.0±9.5 Table II.—C orrelation pain extent and self-reported outcomes and SF36-MCS 37.8±6.9 38.5±7.2 36.8±6.2 clinical tests. TSK 13.9±8.7 15.9±8.8 10.1±7.8 BDI-II 21.3±7.7 22.8±7.3 19.0±7.7 Parameters All patients Traumatic onset Non-traumatic NDI (N.=200) (N.=120) onset (N.=80) 39.9±15.2 37.9±15.8 42.9±13.7 Range-of-motion, degrees 44.1±19.1 41.2±20.3 48.6±16.2 Patient-reported outcomes Flex 58.1±16.3 56.3±17.6 60.9±14.0 Ext 60.1±15.2 59.1±15.4 61.8±14.8 SF36-PCS r=-0.03 (P=0.72)* r=-0.04 (P=0.69) r=-0.01 (P=0.96) Rot left 31.4±10.6 31.1±10.0 31.7±10.0 Rot right 34.1±12.1 33.6±12.9 34.8±10.8 SF36-MCS r=-0.15 (P=0.83)* r=-0.15 (P=0.12) r=-0.16 (P=0.15) LF left LF right 145 (72.5%) 92 (77%) 53 (66%) TSK r=0.21 (P=0.01)* r=0.12 (P=0.21) r=0.33 (P=0.01)* 36 (18.0%) 23 (19%) 13 (16%) Craniocervical flexion test 19 (9.5%) 5 (4%) 14 (18%) BDI-II r=0.29 (P<0.01)* r=0.21 (P=0.02)* r=0.40 (P<0.01)* 22 mmHg 24 mmHg 32 (16.0%) 22 (18%) 10 (12%) NDI r=0.33 (P<0.01)* r=0.22 (P=0.02)* r=0.45 (P<0.01)* 26-30 mmHg 55 (27.5%) 40 (33%) 15 (19%) 28 (14.0%) 16 (13%) 12 (15%) Range of motion Cervical extension test 85 (42.5%) 42 (35%) 43 (53%) 0-10 s 7.5±6.7% 7.6±6.7% 7.4±6.8% Flex r=-0.05 (P=0.50) r=-0.05 (P=0.63) r=-0.04 (P=0.75) 11-28 s 39-119 s Ext r=-0.05 (P=0.45) r=-0.01 (P=0.91) r=-0.12 (P=0.30) 120 s Rot left r=-0.10 (P=0.15) r=-0.08 (P=0.37) r=-0.11 (P=0.34) Mean pain extent (percentage of total Rot right r=-0.11 (P=0.15) r=-0.14 (P=0.15) r=-0.06 (P=0.62) body chart) LF left r=0.01 (P=0.94) r=-0.01 (P=0.98) r=-0.02 (P=0.87) LF right r=0.04 (P=0.59) r=-0.01 (P=0.99) r=0.09 (P=0.45) Muscle function CCFT r=-0.24 (P<0.01)* r=-0.18 (P=0.06) r=-0.31 (P=0.01)* CE r=-0.19 (P=0.01)* r=-0.12 (P=0.21) r=-0.29 (P=0.01)* SF36-PCS: Short Form-36, Standardized Physical Component Summary Score; SF36-PCS: Short Form-36, Standardized Physical Component Summary Score; SF36-MCS: Short Form 36, Standardized Mental Component Summary Score; SF36-MCS: Short Form 36, Standardized Mental Component Summary Score; TSK: Tampa Scale of Kinesiophobia; BDI-II: Beck Depression Inventory-II; TSK: Tampa Scale of Kinesiophobia; BDI-II: Beck Depression Inventory-II; NDI: Neck Disability Index; Flex: cervical flexion; Ext: cervical extension; Rot: NDI: Neck Disability Index; Flex: cervical flexion; Ext: cervical extension; Rot: cervical rotation; LF: cervical lateral flexion. cervical rotation; LF: cervical lateral flexion; CCFT: craniocervical flexion test; CE: cervical extension. *Statistically significant correlation. Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 75

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, RIS PAIN EXTENT IN CHRONIC NECK PAIN cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. (r=0.42, P<0.01 and r=0.31, P<0.01 for BDI-II and TSK reporting a significant correlation between non-organic respectively), whereas those with neck pain of traumatic pain (determined from pain drawings) and pain catastro- origin had only a weak correlation between pain extent phizing.12 and BDI-II (r=0.22, P=0.02), and a non-significant corre- lation between pain extent and TSK. Correlations between A previous study, using the same body charts as this cur- pain extent and SF36-PCS, SF36-MCS and ROM were rent study, reported an association between pain extent and non-significant. NDI, depression and self-efficacy in people with whiplash associated disorders.10 This finding was confirmed in the For the clinical tests, correlations between pain extent current study with weak correlations observed between and CCFT and CE performance were weak and negative pain extent and NDI as well as BDI-II. Unique to this study (r=-0.23; P<0.01 and r=-0.20; P<0.01 respectively), indi- we also evaluated pain extent in people with neck pain of cating that larger pain extent was associated with poorer non-traumatic origin and showed that the correlation be- performance on both tests (Table II). tween pain extent and NDI, BDI-II and TSK was stronger compared to the correlations observed for the group with Discussion trauma-induced neck pain. The complexity and severity of symptoms is reported to be greater in those with neck pain The current study is the first to examine the relationship of traumatic origin including additional symptoms such as between pain extent and patient-reported outcomes on post-traumatic stress disorder,37 and central sensitization38 neck function, quality-of-life and psychological factors, which were not considered in the current study. in addition to clinical tests for neck muscle function in people with chronic neck pain. When considering the full This study also revealed significant correlations between sample of patients with either non-traumatic and trauma pain extent and performance on tests of neck muscle func- induced neck pain, moderate correlations were observed tion although these correlations were weak in those with between pain extent and perceived neck disability, and neck pain of traumatic origin and moderate in those with weak correlations were observed between pain extent and a non-traumatic origin. Previous work has shown stronger the level of depression, kinesiophobia as well as perfor- correlations between pain intensity and performance on mance on clinical tests for neck muscle function. This is the CCFT and CE.39, 40 Therefore, other dimensions such supported by a previous study, showing that pain extent as pain intensity may be stronger indicators of altered neck explained 5-6% of the variance of NDI scores in patients muscle function than pain extent. with chronic whiplash associated disorders.34 By compar- ing the subgroups in the current sample, it was shown that Methodological considerations correlation for those with non-traumatic neck pain were consistently on a moderate level for all outcomes whilst The current work included a large sample size and the they were only weak for those with trauma-induced neck study was conducted across multiple clinical centers add- pain, which is in contrast to our hypothesis. The mod- ing to the strengths of this study. The participants were erate correlation between pain extent and NDI indicates referred to the clinics via usual procedures and were part that other variables may explain the variance in pain ex- of the normal population of the centers. Those centers in- tent, such as pain intensity and kinesiophobia.34 Yet, only cluded both urban and rural areas, primary and secondary weak correlations were observed between pain extent care, as well as different geographic areas strengthening and depression as well as kinesiophobia. This finding is the external validity of the results. However, records of in line with a previous systematic review35 which con- those tested for eligibility but not entering the study were cluded that pain drawings (mainly low back pain) can- for practical reasons not possible to obtain. However, this not be used as an indicator of the patients’ psychological is not considered to influence the outcome of this current state. A more recent review confirmed that pain drawings study as this is a studying correlation between different from subacute/chronic low back pain patients cannot de- outcomes. Anecdoticly, the eligible patients excluded were tect psychological features.36 However, comparison with few. Therefore, we assume the external validity to be suf- other studies is challenging, as use of standardized body ficient to transfer the results of the study in daily clinical charts as well as methods for analyzing these are mostly practice. The data collection was based upon methods with lacking. This may be the reason for contrasting results a moderate to strong intra-class correlation and inter-tester with a recent study in patients with neck-shoulder pain reliability10, 13, 19, 24, 41, 42 and collected by the same two assessors according to a rigorous standardized protocol. Despite recruitment from different centers, data collection 76 European Journal of Physical and Rehabilitation Medicine February 2019

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, PAIN EXTENT IN CHRONIC NECK PAIN RIS cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. was performed with the same test equipment and the test 6.  Chien A, Sterling M. Sensory hypoaesthesia is a feature of chronic setting was similar across recruitment locations. The anal- whiplash but not chronic idiopathic neck pain. Man Ther 2010;15:48–53. yses of pain drawings were executed by two experienced 7.  Woodhouse A, Vasseljen O. Altered motor control patterns in whiplash investigators following established procedures. and chronic neck pain. BMC Musculoskelet Disord 2008;9:90. 8.  Ris I, Juul-Kristensen B, Boyle E, Kongsted A, Manniche C, Søgaard Limitations of the study K. Chronic neck pain patients with traumatic or non-traumatic on- set: differences in characteristics. A cross-sectional study. Scand J Pain Limitations of the study include the fact that the clinical 2017;14:1–8. tests and questionnaires used for inclusion of the partici- 9.  Curatolo M. Diagnosis of altered central pain processing. Spine pants were delivered by different physiotherapists at these 2011;36(Suppl):S200–4. different centers, which may have introduced some vari- 10.  Barbero M, Moresi F, Leoni D, Gatti R, Egloff M, Falla D. Test-retest ability; nevertheless, this variability was minimized since reliability of pain extent and pain location using a novel method for pain all physiotherapists were carefully trained in the proce- drawing analysis. Eur J Pain 2015;19:1129–38. dures by the same instructor. Also, there are no records of 11.  Falla D, Peolsson A, Peterson G, Ludvigsson ML, Soldini E, Schnee- those excluded after tested for eligibility, which weakens beli A, et al. Perceived pain extent is associated with disability, depression the external validity of the study. and self-efficacy in individuals with whiplash-associated disorders. Eur J Pain 2016;20:1490–501. Clinical interpretations 12.  Barbero M, Fernández-de-Las-Peñas C, Palacios-Ceña M, Cescon C, Falla D. Pain extent is associated with pain intensity but not with wide- The extent of pain, expressed in pain drawings, is asso- spread pressure or thermal pain sensitivity in women with fibromyalgia ciated with higher self-perceived neck disability and can syndrome. Clin Rheumatol 2017;36:1427–32. be an indicator for psychological screening, especially for 13.  Hayashi K, Arai YC, Morimoto A, Aono S, Yoshimoto T, Nishihara non-traumatic neck pain patients. M, et al. Associations between pain drawing and psychological character- istics of different body region pains. Pain Pract 2015;15:300–7. Conclusions 14.  Walker BF, Losco CD, Armson A, Meyer A, Stomski NJ. The associa- tion between pain diagram area, fear-avoidance beliefs, and pain catastro- Patients with chronic neck pain that describe larger pain phising. Chiropr Man Therap 2014;22:5. areas are likely to present with greater self-perceived dis- 15.  Ris I, Søgaard K, Gram B, Agerbo K, Boyle E, Juul-Kristensen B. ability, depression, kinesiophobia and poorer muscle func- Does a combination of physical training, specific exercises and pain edu- tion. These associations were strongest in those with neck cation improve health-related quality of life in patients with chronic neck pain of non-traumatic onset and only weak in those with pain? A randomised control trial with a 4-month follow up. Man Ther neck pain of traumatic onset. Pain drawings can therefore 2016;26:132–40. indicate a need for addressing these functions in manage- 16.  Anstey R, Kongsted A, Kamper S, Hancock MJ. Are People With ment of a person with chronic neck pain. Whiplash-Associated Neck Pain Different From People With Nonspecific Neck Pain? J Orthop Sports Phys Ther 2016;46:894–901. References 17.  von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Van- denbroucke JP; STROBE Initiative. The Strengthening the Reporting of 1.  Hoy D, March L, Woolf A, Blyth F, Brooks P, Smith E, et al. The Observational Studies in Epidemiology (STROBE) statement: guidelines global burden of neck pain: estimates from the global burden of disease for reporting observational studies. Epidemiology 2007;18:800–4. 2010 study. Ann Rheum Dis 2014;73:1309–15. 18.  Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory 2.  Falla D, O’Leary S, Farina D, Jull G. The change in deep cervical flex- for measuring depression. Arch Gen Psychiatry 1961;4:561–71. or activity after training is associated with the degree of pain reduction in 19.  Jørgensen R, Ris I, Falla D, Juul-Kristensen B. Reliability, construct patients with chronic neck pain. Clin J Pain 2012;28:628–34. and discriminative validity of clinical testing in subjects with and without 3.  Fernández-de-las-Peñas C, Albert-Sanchís JC, Buil M, Benitez JC, Al- chronic neck pain. BMC Musculoskelet Disord 2014;15:408. burquerque-Sendín F. Cross-sectional area of cervical multifidus muscle 20.  Dos Reis FJ, de Barros E Silva V, de Lucena RN, Mendes Cardoso in females with chronic bilateral neck pain compared to controls. J Orthop BA, Nogueira LC. Measuring the Pain Area: An Intra- and Inter-Rater Re- Sports Phys Ther 2008;38:175–80. liability Study Using Image Analysis Software. Pain Pract 2016;16:24–30. 4.  Sterling M, Carroll LJ, Kasch H, Kamper SJ, Stemper B. Prognosis 21.  Persson AL, Garametsos S, Pedersen J. Computer-aided surface es- after whiplash injury: where to from here? Discussion paper 4. Spine timation of pain drawings - intra- and inter-rater reliability. J Pain Res 2011;36(Suppl):S330–4. 2011;4:135–41. 5.  Yalcinkaya H, Ucok K, Ulasli AM, Coban NF, Aydin S, Kaya I, et al. 22.  Bjorner JB, Thunedborg K, Kristensen TS, Modvig J, Bech P. The Do male and female patients with chronic neck pain really have different Danish SF-36 Health Survey: translation and preliminary validity studies. health-related physical fitness, depression, anxiety and quality of life pa- J Clin Epidemiol 1998;51:991–9. rameters? Int J Rheum Dis 2017;20:1079–87. 23.  Bjorner JB, Damsgaard MT, Watt T, Groenvold M. Tests of data qual- ity, scaling assumptions, and reliability of the Danish SF-36. J Clin Epi- demiol 1998;51:1001–11. 24.  Cleland JA, Fritz JM, Childs JD. Psychometric properties of the Fear- Avoidance Beliefs Questionnaire and Tampa Scale of Kinesiophobia in patients with neck pain. Am J Phys Med Rehabil 2008;87:109–17. 25.  Jørgensen MB, Damsgård E, Holtermann A, Anke A, Søgaard K, Røe C. Properties of the Tampa Scale for Kinesiophobia across Workers with Different Pain Experiences and Cultural Backgrounds: A Rasch Analysis. J Appl Meas 2015;16:218–27. 26.  Beck AT, Steer RA, Carbin MG. Psychometric properties of the Beck Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 77

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, RIS PAIN EXTENT IN CHRONIC NECK PAIN cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. Depression Inventory: twenty-five years of evaluation. Clin Psychol Rev drawing literature: should pain drawings be used for psychologic screen- 1988;8:77–100. ing? Clin J Pain 2006;22:449–57. 27.  Aalto AM, Elovainio M, Kivimäki M, Uutela A, Pirkola S. The Beck 36.  Bertozzi L, Rosso A, Romeo A, Villafañe JH, Guccione AA, Pillas- Depression Inventory and General Health Questionnaire as measures of trini P, et al. The accuracy of pain drawing in identifying psychological depression in the general population: a validation study using the Com- distress in low back pain-systematic review and meta-analysis of diagnos- posite International Diagnostic Interview as the gold standard. Psychiatry tic studies. J Phys Ther Sci 2015;27:3319–24. Res 2012;197:163–71. 37.  Andersen TE, Karstoft KI, Brink O, Elklit A. Pain-catastrophizing 28.  McCarthy MJ, Grevitt MP, Silcocks P, Hobbs G. The reliability of the and fear-avoidance beliefs as mediators between post-traumatic stress Vernon and Mior neck disability index, and its validity compared with the symptoms and pain following whiplash injury - A prospective cohort short form-36 health survey questionnaire. Eur Spine J 2007;16:2111–7. study. Eur J Pain 2016;20:1241–52. 29.  Juul T, Søgaard K, Davis AM, Roos EM. Psychometric properties of 38.  Goldsmith R, Wright C, Bell SF, Rushton A. Cold hyperalgesia as a the Neck OutcOme Score, Neck Disability Index, and Short Form-36 were prognostic factor in whiplash associated disorders: a systematic review. evaluated in patients with neck pain. J Clin Epidemiol 2016;79:31–40. Man Ther 2012;17:402–10. 30.  Jull GA, O’Leary SP, Falla DL. Clinical assessment of the deep cervi- 39.  Chiu TT, Law EY, Chiu TH. Performance of the craniocervical flex- cal flexor muscles: the craniocervical flexion test. J Manipulative Physiol ion test in subjects with and without chronic neck pain. J Orthop Sports Ther 2008;31:525–33. Phys Ther 2005;35:567–71. 31.  Cohen J. Statistical power analysis for the behavioral sciences: Hills- 40.  Shahidi B, Johnson CL, Curran-Everett D, Maluf KS. Reliability and dale, NJ: L. Erlbaum Associates; 1988. group differences in quantitative cervicothoracic measures among indi- 32.  Vlaeyen JW, Kole-Snijders AM, Boeren RG, van Eek H. Fear of viduals with and without chronic neck pain. BMC Musculoskelet Disord movement/(re)injury in chronic low back pain and its relation to behav- 2012;13:215. ioral performance. Pain 1995;62:363–72. 41.  Bertilson B, Grunnesjö M, Johansson SE, Strender LE. Pain drawing 33.  Vernon H, Mior S. The Neck Disability Index: a study of reliability in the assessment of neurogenic pain and dysfunction in the neck/shoulder and validity. J Manipulative Physiol Ther 1991;14:409–15. region: inter-examiner reliability and concordance with clinical examina- 34.  Vernon H, Guerriero R, Kavanaugh S, Soave D, Moreton J. Psycho- tion. Pain Med 2007;8:134–46. logical factors in the use of the neck disability index in chronic whiplash 42.  Cleland JA, Childs JD, Whitman JM. Psychometric properties of the patients. Spine 2010;35:E16–21. Neck Disability Index and Numeric Pain Rating Scale in patients with 35.  Carnes D, Ashby D, Underwood M. A systematic review of pain mechanical neck pain. Arch Phys Med Rehabil 2008;89:69–74. Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript. Funding.—This study received funding from the Research Fund of the Region of Southern Denmark, the Danish Rheumatism Association, the Research Foundation of the Danish Association of Physiotherapy, Fund for Physiotherapy in Private Practice, and the Danish Society of Polio and Accident Victims (PTU). Congresses.—The study was presented as a poster presentation at the World Congress of the International Federation of Orthopedic Manipulative Physical Therapists in 2016. Article first published online: August 27, 2018. - Manuscript accepted: July 27, 2018. - Manuscript revised: June 27, 2018. - Manuscript received: September 7, 2017. 78 European Journal of Physical and Rehabilitation Medicine February 2019

COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically © 2018 EDIZIONI MINERVA MEDICA European Journal of Physical and Rehabilitation Medicine 2019 February;55(1):79-88 or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access Online version at http://www.minervamedica.it DOI: 10.23736/S1973-9087.18.05041-4 to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. ORIGINAL ARTICLE Effects of aquatic training and detraining on women with fibromyalgia: controlled randomized clinical trial Carolina P. ANDRADE 1 *, Antonio R. ZAMUNÉR 2, Meire FORTI 3, Nayara Y. TAMBURÚS 3, Ester SILVA 3 1Center of Health Sciences, Universidade do Sagrado Coração, USC, Bauru, Brasil; 2Department of Kinesiology, Catholic University of Maule, Talca, Chile; 3Department of Physiotherapy, Federal University of São Carlos, São Carlos, Brasil *Corresponding author: Carolina Pieroni Andrade, Department of Physical Therapy, Sacred Heart University, R. Irmã Arminda 10-50 - Jardim Brasil, 17011- 160 Bauru, São Paulo, Brazil. E-mail: [email protected]. ABSTRACT BACKGROUND: Aquatic physical training (APT) promotes improvement of clinical symptoms and aerobic functional capacity in women with fibromyalgia syndrome (FMS). However, there are controversial studies that refer to the maintenance of adaptations obtained from APT after a detraining period. AIM: To evaluate variables oxygen uptake (VO2) relative to lean body mass (LBM) and clinical symptomatology in women with FMS submitted to APT and after 16 weeks of detraining period, and to evaluate the association between the magnitude of VO2 improvement relative to LBM and clinical symptomatology. DESIGN: Blind randomized controlled trial. SETTING: Department of Physical Therapy of the Federal University of São Carlos. POPULATION: Fifty-four women with FMS were randomly assigned into trained group (27) and control group (27). METHODS: All women underwent cardiopulmonary exercise test (CPET) and body composition assessment to estimate VO2 and LBM respec- tively, and clinical symptoms were assessed before and after 16 weeks of training and detraining. trained group was submitted to APT program, performed twice a week for 16 weeks. RESULTS: After APT, trained group presented an increase in VO2 relative to LBM (P=0.01), in addition to an increase in pressure pain threshold (PPT) (P=0.02) and Visual Analogue Scale (VAS) pain (P=0.01), VAS well-being (P<0.01) well-being and lower Fibromyalgia Impact Question- naire Score (FIQ) (P=0.04). However, these improvements were not maintained after the 16-week detraining period (P>0.05). In addition, no significant correlations were observed between improvement of clinical manifestations and increased VO2 relative to LBM after APT (P>0.05). CONCLUSIONS: APT contributed both to increase VO2 at VAT and peak CPET, and improved clinical symptoms, but no association was ob- served. However, after 16 weeks of detraining, these variables were reduced near baseline. CLINICAL REHABILITATION IMPACT: The results of the present study suggest that APT should be continuously performed in order to im- prove clinical symptomatology and increase the aerobic functional capacity in women with FMS. (Cite this article as: Andrade CP, Zamunér AR, Forti M, Tamburús NY, Silva E. Effects of aquatic training and detraining on women with fibromyal- gia: controlled randomized clinical trial. Eur J Phys Rehabil Med 2019;55:79-88. DOI: 10.23736/S1973-9087.18.05041-4) Key words: Fibromyalgia - Cardiorespiratory fitness - Exercise - Physical therapy specialty. Clinical manifestations in patients with fibromyalgia proaches.10 Among non-pharmacological ones, individual- syndrome (FMS) are characterized by chronic and dif- ized aerobic exercise and aquatic physical training (APT) fuse musculoskeletal pain, fatigue and non-repairing sleep is strongly recommended for relieving pain, fatigue and with physical, psychological and social impairment.1-4 In sleep disorders, and improving aerobic functional capac- addition, studies have shown that these patients have low ity,6, 11-14 as well as psychological and social aspects.6, 11-13 capacity to perform physical activities, which contributes to low aerobic functional capacity that consequently wors- However, to our knowledge, no study has evaluated ens pain sensitivity and quality of life.5-9 whether the improvement in aerobic functional capac- ity, quantified by the increase in oxygen uptake (VO2) Currently, the clinical management of FMS is performed relative to lean body mass (LBM) obtained from APT is through pharmacological and non-pharmacological ap- maintained after a detraining period. A previous study Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 79

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically ANDRADE COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, EFFECTS OF TRAINING AND DETRAINING ON FIBROMYALGIA cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. showed that the relative VO2 corrected by the LBM al- Materials and methods lows obtaining more reliable values for the assessment of aerobic functional capacity in women with FMS than the Study design and participants traditional relative VO2 corrected by the total body mass (TBM), especially regarding peripheral responses that A single blinded controlled randomized trial was conducted lead to increased VO2.14 In addition, studies that reported at the Department of Physical Therapy of the Federal Uni- the duration of clinical improvement promoted by aquatic versity of São Carlos. The evaluators were blinded to the exercise in this population are controversial. Sañudo et protocol (i.e. they had no information whether the partici- al.15 and Gowans et al.16 observed that the improvement pants were assigned to the control group or to the trained in functional capacity, severity of depressive symptoms, group). Participants were recruited through posters and quality of life and the impact of fibromyalgia on quality of leaflets distributed at strategic points in the city (rheuma- life promoted by combined physical and aerobic training, tology, orthopedics and physiotherapy clinics and offices) respectively, were maintained after 6 months of detrain- from December 2013 to December 2014. One hundred and ing. On the other hand, Tomas-Carus et al.17 reported that twenty women with clinical diagnosis of FMS were inter- the improvement promoted after 12 weeks of aquatic ex- viewed, 54 were considered eligible and agreed to partici- ercises on quality of life, aerobic functional capacity and pate in the study. All participants should present clinical di- impact of fibromyalgia on quality of life in women with agnosis of FMS performed by a rheumatologist, according FMS did not remain after a 12-week detraining period. In to criteria established by the American College of Rheuma- addition, Murias et al.18 and St-Amand et al.19 observed tology.1, 26 Participants aged 30-60 years and had low level that after 14 days and 12 weeks of detraining, respectively, of physical activity according to the International Physical significantly decreased the maximal VO2 relative to the to- Activity Questionnaire (iPAQ).27 Volunteers with cardio- tal body mass (TBM) after an aerobic exercise program. vascular diseases, systemic arterial hypertension, arrhyth- mias, diabetes mellitus, musculoskeletal and neurological Therefore, elucidating this aspect is of great clinical rel- disorders that could directly interfere with assessments (for evance, since it allows clarifying if the reversibility prin- example, advanced joint diseases), presence of infections ciple of the physical exercise also applies to the clinical and any other rheumatic diseases (e.g., osteoarthritis, con- symptomatology of women with FMS. In this context, it is nective tissue disease, rheumatoid arthritis) were excluded. necessary to address the effects of training and detraining Volunteers were instructed not to take anxiolytic medica- after APT in this population. tions, antidepressants, analgesics and muscle relaxants for at least 48 hours before evaluations. Moreover, the association between aerobic functional capacity and clinical symptomatology in women with After baseline assessments, randomization was per- FMS is not well established. Some studies have shown that formed on a 1:1 ratio. Consecutively numbered, opaque aerobic functional capacity is inversely associated with envelopes were prepared ahead of time and randomly as- anxiety levels20, 21 and pressure pain threshold (PPT).22, 23 signed by a computer-generated table of random numbers On the other hand, other studies have reported that there (http://www.randomization.com). A research not involved is no association between these variables24, 25 nor with the in the study (masked to the study protocol) performed the pain assessed by a Visual Analogue Scale (VAS).23 In this randomization and provided the group assignment to the sense, Zamunér et al.6 and Valim et al.5 showed that the researchers responsible for delivering the therapy. Twenty- magnitude of improvement in clinical symptoms and is not seven participants were assigned to the trained group, which associated to the magnitude of increase in peak VO2 rela- was submitted to APT, and twenty seven were assigned to tive to the TBM. the control group, which received instructions for the main- tenance of daily life activities identified at baseline. Thus, since women with FMS show improvement in VO2 relative to the LBM after an individualized APT All participants read and signed the informed consent program,14 but little is known about the effects after a de- form before participating in the study, which was ap- training period, the aims of the present study were: 1) to proved by the Ethics Research Committee (protocol num- evaluate the VO2 variables related to LBM and clinical ber 112.508) registered at ClinicalTrials.gov under num- symptomatology in women with FMS after APT and after ber NTC01839305. a 16-week detraining period; 2) to evaluate the association between the magnitude of VO2 improvement relative to Figure 1 shows the flowchart of patients throughout LBM and clinical symptoms in this population. the study. Fifty-four women were considered eligible 80 European Journal of Physical and Rehabilitation Medicine February 2019

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, EFFECTS OF TRAINING AND DETRAINING ON FIBROMYALGIA ANDRADE cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. Enrollment Assessed for eligibility (N.=120) ity were evaluated from the submaximal cardiopulmonary exercise test (CPET). In the fourth step, after 16 weeks of Excluded (N.=66) follow-up (Post16), the third-stage experimental protocols - Not meeting inclusion criteria were repeated. In the fifth stage, volunteers who received intervention were again submitted to the third and fourth (N.=42) stage evaluations, but after a 16-week untreated period to - Declined to participate (N.=24) evaluate the detraining effect (Post32). Randomized (N.=54) All volunteers were evaluated in the morning (between 07:30 a.m. and 12:00 p.m.). The ambient temperature was Allocation maintained between 22 and 24° C and the relative air hu- midity between 40% and 60%. After baseline assessments, Allocation for intervention (APT) Allocation for follow-up blinded randomization was performed by a person not in- - T rained group (N.=27) - Control group (N.=27) volved in the study. Follow-up Clinical features Withdrawn (3) Withdrawn (3) Pain and fatigue intensity and well-being were evaluated by VAS from 0 to 100 mm. PPT was determined on the 18 - Gave up participating (N.=1) - Health problems (N.=2) tender points described by1 by means of a digital algom- - Unable to attend treatment (N.=2) - Personal problems (N.=1) eter (OE-220: Tissue Hardness Meter & Algometer, Ito Co., Japan). Tender points were considered active when - Per-protocol analysis (N.=24) Analysis the PPT reported by the participant was less than 4 kgf.1 - Intention to treat (N.=27) The mean PPT measured at 18 tender points was consid- Post 16 ered for analysis. - Per-protocol analysis (N.=24) The impact of fibromyalgia on quality of life was as- - Intention to treat (N.=27) sessed using the Fibromyalgia Impact Questionnaire (FIQ), in which the score ranged from 0 to 100. The higher Follow-up the score, the greater the impact of FMS on individuals’ quality of life.28 Withdrawn (2) - Gave up participating (N.=2) To assess the presence and severity of depression and anxiety symptoms, Beck’s Depression Inventory (BDI)29 Analysis and Beck’s Anxiety Inventory (BAI),30 respectively, were used. Post 32 Quality of life was assessed using the Short Form 36 - Per-protocol analysis (N.=22) (SF-36), divided into eight domains (functional capacity, - Intention to treat (N.=27) physical appearance, pain, general health status, vitality, social aspect, emotional aspect and mental health). The Figure 1.—Flowchart of patients who participated in the study. score of each domain ranges from 0 (worst state) to 100 (best state).31 and agreed to participate in the study. However, 3 trained group volunteers gave up participating in the APT and 3 Sleep quality was assessed by the Pittsburgh Sleep Qual- control group volunteers did not want to carry out reevalu- ity Index (PSQI),32 in which it is possible to evaluate sleep ations. Thus, 24 trained group volunteers and 24 control quality and disorders present in the period of one month group volunteers performed Post16 reassessments. Two prior to the date of its application. The PSQI consists of trained group volunteers continued APT and were not con- 19 items grouped into seven components, each scored on a sidered for Post32 reassessment. Thus, 22 trained group scale from 0 to 3. The components are respectively: 1) sub- women and 24 control group women were considered for jective sleep quality; 2) sleep latency; 3) sleep duration; 4) per-protocol analysis. usual sleep efficiency; 5) sleep disorders; 6) use of medi- cations for sleep; and 7) daytime dysfunction. The scores Measures of the seven components are summed to give an overall PSQI Score, which ranges from 0 to 21. Scores from 0 to Experiments were carried out in 5 stages. In the first stage, clinical and physiotherapeutic evaluations were performed in which the number of active tender points and anamne- sis were verified. In the second stage, familiarization with the equipment and experimental protocol was performed, and the maximum stress test for cardiovascular evaluation was performed. In the third stage (baseline), question- naires were applied and PPT and aerobic functional capac- Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 81

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically ANDRADE COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, EFFECTS OF TRAINING AND DETRAINING ON FIBROMYALGIA cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. 4 indicate good sleep quality, from 5 to 10 indicate poor Ventilatory and metabolic variables quality and above 10 indicate sleep disturbance.32 Variables VO2 in absolute units and VO2 relative to TBM Evaluations were conducted by a researcher blind to the were obtained breath-to-breath through the entire CPET experimental protocol, but with training for application by means of an expired gas measurement system (Ul- and use of instruments. tima PFX system, Medical Graphics, St Paul, MN) duly calibrated before each test. The correction of VO2 relative The gains with aquatic training were determined ac- to LBM was obtained from the ratio between VO2 (mL / cording to OMERACT-10 FM response criteria. min-1) in absolute units and LBM. Body composition assessment Body composition was evaluated to estimate LBM accord- APT program ing to protocol described by Andrade et al.14 Briefly, Bio- dynamics® model 310 (Biodynamics Corporation, Seattle, The APT program was performed in a heated pool (30±2 WA, USA) analyzer was used according to the whole body °C). The protocol consisted of 32 sessions of 45 min, twice technique (hand-foot) using single frequency (50 kHz, 800 a week (alternating days) for 16 weeks. The sessions were μA). Four electrodes were placed on the dorsal surface of conducted in groups of up to 5 women and were super- the right hemisphere’s hand, wrist, foot, and ankle.33, 34 vised by three physiotherapists. The APT protocol has During the measurement, volunteers remained in the su- already been described in a previous study conducted by pine position on a non-conductive surface. our research group.14 The progression of aerobic exercises was adjusted throughout the sessions in order to maintain CPET HR and the subjective perceived exertion (RPE) reached at The CPET protocol was continuous ramp type, per- VAT level identified in the CPET. formed on cycloergometer with electromagnetic braking In all sessions, BP (auscultatory method), HR (Polar® (Quinton Corival 400, Seattle, WA, USA). monitor model FT1, Electro Oy, Finland), RPE (Borg CR- 10 scale) were evaluated during each session. CPET consisted of 1 min at rest in the sitting position on the cycle ergometer, followed by 4 min of warm-up at in- Detraining period tensity of 4 W of power. Power increase was then initiated until physical exhaustion (peak exercise), defined as the During the 16-week detraining period, participants from moment when the volunteers were unable to maintain the the trained group received regular phone calls (i.e. every pedals at 60 rpm or until the manifestation of some limit- 15 days) in order to check for their general health status ing symptom (e.g., pain, dizziness, nausea) or respiratory and to ensure they were keeping low levels of physical fatigue.35 Volunteers were instructed to maintain the pedal- activity (irregularly active or sedentary) as assessed by the ing rate at 60 revolutions per minute (rpm) and not to per- IPAQ Short Form.27 form isometric contraction of upper limbs during the test. Two volunteers were excluded because they met the The power increments were determined for each vol- “Active” classification according to the IPAQ27 (i.e. 3 or unteer according to the formula proposed by Wasserman more days of vigorous activity of at least 20 minutes per et al.:35 day or 5 or more days of moderate-intensity activity and/ or walking of at least 30 minutes per day or 5 or more days Power = {[(height – age) 0.14] – of any combination of walking, moderate-intensity or vig- [150 + (6 × body mass)]} / 100 orous intensity activities achieving a minimum of at least 600 MET-minutes/week). Thus, 22 volunteers underwent During CPET, electrocardiogram and HR were record- all the experimental procedure. For control group, the last ed in real-time beat-to-beat with CardioPerfect® Electro- evaluation was maintained for comparison purposes. cardiograph (Welch Allyn CardioPerfect Workstation, Skaneateles Falls, NY). Statistical analysis At the end of CPET, three trained observers identified Sample size calculation was performed using the GPower VAT by means of the visual graphic method to estimate software (version 3.1), based on a pilot study (N.=20). the disproportionate increase of ventilatory and metabolic For the analysis, effect size =0.70 was obtained for vari- variables during dynamic incremental exercise.36 The cri- able VO2. The suggested sample size was 20 participants terion adopted was the loss of parallelism between VO2 and carbon dioxide production (VCO2).36 82 European Journal of Physical and Rehabilitation Medicine February 2019

COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA EFFECTS OF TRAINING AND DETRAINING ON FIBROMYALGIA ANDRADE This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access in each group. The significance level was set at 5% with Table I.—Clinical characteristics of the study population. to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove,power of 80%. cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. TG (N.=27) CG (N.=27) The intention-to-treat analysis was performed using the VO2VAT_LBM multipleimputation method to impute values for all miss- Age (years) 48±8 47±8 (mL/kgLBM-1/min-1) ing data.37 Analysis using only patients with complete data SAP (mmHg) 27±3 30±5 was also performed (per-protocol analysis). DAP (mmHg) 121±12 121±11 VO2PEAK_LBM HR (bpm) 78±6 75±2 (mL/kgLBM-1/min-1) To evaluate data distribution and variance homogene- Duration of diagnosis (years) 77±9 73±10 ity, the Shapiro-Wilk and Levene tests, respectively, were Medications 10±5 applied. The sphericity assumption was evaluated by the 9±5 2 (7) Mauchly test and in case of violation, the Greenhouse- Anxiolytics (N./%) 3 (11) 12 (44) Geisser correction was used. For intergroup (control group Antidepressants (N./%) 8 (29) 0 (0) x trained group) comparisons related to hemodynamic AF (N./%) 2 (7) 17 (62) variables and age, the Student’s t-test for independent AN (N./%) 17 (62) 16 (59) samples was used. MR (N./%) 15 (55) Two-way mixed analysis of variance (ANOVA) was used Data are presented as mean±standard deviation, or counts (percentage). to evaluate variables clinical symptom and CPET of trained SAP: systolic arterial pressure; DAP: diastolic arterial pressure; HR: heart rate; group and control group before and after 16 and 32 weeks. AF: anti-inflammatory drugs; AN: analgesic drugs; MR: muscle relaxant drugs. When significant interaction was observed, the analysis of the main effects was disregarded and the multiple compari- Pre Post16 Post32 Pre Post16 Post32 B Pre Post16 Post32 Pre Post16 Post32 son test with Bonferroni adjustment was performed. A TG CG TG CG The association between variables VO2 related to LBM and the clinical characteristics of FMS was evaluated by the Figure 2.—Oxygen uptake (VO2, mL/kg-1/min-1) relative to lean body Pearson correlation coefficient. The correlation strength mass (LBM), obtained at ventilatory anaerobic threshold (VAT) (A) and was classified according to the following criteria: 0 with- peak CPET (peak), (B) of both groups (trained group and control group) out correlation, 0.1-0.3 weak correlation, 0.4-0.6 moderate at baseline (Pre), after 16 weeks of APT (Post16) and after 32 weeks of correlation, 0.7-0.9 strong correlation and 1.0 perfect cor- follow-up (Post32). relation. The effect size was calculated by Cohen’s d coef- *Statistically significant difference between Pre vs. Post16 (intragroup); ficient. Effect size greater than 0.8 was considered large, between Post16 vs. Post32 (intragroup) and between trained group vs. close to 0.5 moderate, and less than 0.2, small.38 control group (intergroup) (P<0.05). Analyses were processed using Statistical Package So- and time for VO2 related to LBM (F1.27 =2.06, P=0.01) at cial Science (SPSS Inc., Chicago, IL, USA) software ver- VAT. Intragroup analysis revealed that the trained group sion 20.0. The significance level was set at 5%. presented an increase in VO2 relative to LBM (P=0.04) at Results VAT after 16 weeks of APT. Higher VO2 relative to LBM (P<0.01) at VAT of trained group was also evidenced, Participants and baseline characteristics when compared to control group after 16 weeks of fol- There was no difference between results obtained by the intention-to-treat analysis and the per-protocol analysis. low-up. After the detraining period, intragroup analysis Therefore, only the results by intention-to-treat were pre- sented. revealed that the trained group presented reduction of VO2 relative to LBM at VAT (P<0.01) after 16 weeks of de- The baseline characteristics of volunteers are presented in Table I. training. In the intergroup analysis in pretraining and after Effect of APT and detraining on VO2 relative to LBM the detraining period, there was no significant difference In Figure 2A, B, VO2 obtained at VAT level and at peak for the trained group (P>0.05) as well as for the control CPET were presented in units relative to LBM (mL.kg-1 LBM.min-1). group, which did not present significant difference after Figure 2A shows significant interaction between group follow-up (P>0.05). In Figure 2B, significant interaction between group and time was observed for variable VO2 relative to LBM ob- tained at peak CPET (F1.27 =4.30; P=0.01). After 16 weeks of APT, it was observed that the trained group presented Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 83

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically ANDRADE COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, EFFECTS OF TRAINING AND DETRAINING ON FIBROMYALGIA cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. an increase of VO2 relative to LBM (P=0.04). In addition, Significant time effect was observed for variable VAS the trained group presented higher VO2 relative to LBM fatigue (F1.27 =4.68; P=0.03), components “limitation (P<0.01) at peak CPET, compared to the control group by physical appearance,” “pain,” “vitality” and “mental after the 16 weeks of follow-up. After the detraining pe- health” of the SF-36 questionnaire (F1.27 =4.58; P=0.03, riod, intragroup analysis revealed that the trained group F1.27 =10.69; P<0.01, F1.27 =6.21; P=0.01 and F1.27 =7.24; presented reduction of VO2 relative to LBM at peak CPET P=0.01 respectively). Thus, regardless of group, values (P<0.01) after 16 weeks of detraining. In the intergroup presented after 16 weeks of follow-up were smaller com- analysis in pretraining and after the detraining period, pared to values obtained at baseline. Significant group ef- there was no significant difference for the trained group fect for variables VAS fatigue (F1.27 =4.25, P=0.04), BAI (P>0.05), as well as for the control group, which did not (F1.27 =4.26, P=0.04), for components ‘functional capac- present significant difference after follow-up (P>0.05). ity’ (F1.27 =6.14; P=0.01), ‘pain’ (F1.27 =4.97, P=0.03) and ‘vitality’ (F1.27 =3.71; P=0.05) of the SF-36 questionnaire Effect of APT on clinical manifestations was also observed. Thus, regardless of evaluated moment, trained group presented lower fatigue, fibromyalgia im- Table II shows the clinical variables. There were signifi- pact on quality of life, anxiety, pain and greater well-being, cant interactions between group and time for variables functional capacity and vitality compared to control group. PPT (F1.27 =4.02, P=0.02), VAS pain (F1.27 =4.11, P=0.01), VAS well-being (F1.27 =5.51, P<0.01) and FIQ (F1.27 =3.29, According to OMERACT-10 FM response criteria, P=0.04). patients presented a 7% reduction in VAS pain, 192% in- crease in SF-36 physical function and a reduction of 19% The analysis of multiple comparisons revealed that after in FIQ, after 16 weeks of APT. APT, the trained group presented increase of PPT (P=0.05) and well-being (P<0.01), and reduction of FIQ (P<0.01) and Effect of detraining on clinical manifestations VAS pain (P=0.05). In the intergroup analysis, trained group presented higher PPT (P<0.01) and well-being (P=0.03), Multiple comparisons analyses revealed that after the de- and lower FIQ (P<0.01) and VAS pain (P=0.02) compared training period, the trained group showed a reduction of to control group after 16 and 32 weeks of follow-up. PPT (P<0.01) and well-being (P=0.01), and an increase in Table II.—Clinical variables: pressure pain threshold (PPT), VAS pain, VAS fatigue and VAS well-being and questionnaires of both groups (TG and CG) in the pre, after 16 weeks of APT (Post16) and after 16 weeks of follow-up (Post32). TG CG P Pre Post16 Post32 Pre Post16 Post32 G T I PPT (kg/cm2) 2.1±0.6 2.3±0.7* 2.0±0.7# 2.1±0.6 1.9±0.6† 1.9±0.6† NS NS 0.02 VAS pain (0-10 cm) 5.8±2.7 5.4±2.4* 5.7±1.8# 5.5±2.1 6.4±2.1† 6.4±2.1† NS NS 0.01 VAS fatigue (0-10 cm) 4.3±2.6 4.0±2.6 4.7±2.5 4.7±2.5 5.8±2.3 5.8±2.3 0.04 0.03 NS VAS well-being (0-10 cm) 5.5±2.5 6.9±2.3* 6.2±2.0# 5.7±2.1 4.7±2.2† 4.7±2.2† <0.01 NS <0.01 FIQ (0-100) 60.3±14.7 49.2±14.7* 54.5±15.2# 67.1±10.5 63.8±9.2† 63.8±9.2† <0.01 <0.01 0.04 BDI (0-21) 18.2±9.6 15.8±9.0 17.5±10.4 20.6±7.4 19.6±8.6 19.6±8.6 NS NS BAI (0-21) 16.1±9.1 15.3±9.1 15.1±9.1 21.2±9.1 19.5±9.0 19.5±9.0 0.04 NS NS SF-36 NS Functional capacity (0-100) 44.6±17.6 50.5±17.6 47.2±19.1 38.2±13.9 38.0±14.7 38.0±14.7 0.01 NS Physical appearance (0-100) 10.2±28.0 29.8±41.0 24.3±36.4 11.0±25.1 13.8±27.8 13.8±27.8 NS 0.03 NS Pain (0-100) 31.8±16.3 36.7±41.0 37.7±15.2 25.5±11.0 29.2±12.1 29.2±12.1 <0.01 NS General health status (0-100) 43.1±18.9 48.9±21.5 48.0±19.8 44.1±19.9 41.0±20.4 41.0±20.4 0.03 NS Vitality (0-100) 33.5±18.6 37.9±22.4 38.8±19.2 25.4±14.7 30.2±15.1 30.2±15.1 NS NS NS Social aspect (0-100) 48.1±17.9 54.3±22.2 47.8±19.0 44.5±20.2 45.4±23.0 45.4±23.0 0.01 NS Emotional aspect (0-100) 24.7±35.3 32.1±40.8 29.6±32.5 18.7±29.4 22.4±35.5 22.4±35.5 0.05 NS Mental health (0-100) 48.6±22.1 46.8±23.0 53.7±21.2 37.8±16.0 43.4±17.3 43.4±17.3 NS NS NS PSQI Total (0-21) 10.1±3.6 11.0±3.8 11.2±3.3 11.2±3.3 NS NS NS 9.4±4.3 8.8±4.4 NS 0.01 NS NS NS Data presented as mean±standard deviation. PPT: pressure pain threshold; VAS: Visual Analogical Scale; FIQ: fibromyalgia impact on quality of life; BDI: Beck’s Depression Inventory; BAI: Beck’s anxiety inventory; SF-36: quality of life; PSQI: Pittsburgh Sleep Quality Index. G: main group effect; T: main time effect: pre vs. post; I: interaction between group and time; NS: not significant. *P<0.05 vs. TG pre; #P<0.05 vs. TG post16; †P<0.05 vs. TG post16. 84 European Journal of Physical and Rehabilitation Medicine February 2019

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, EFFECTS OF TRAINING AND DETRAINING ON FIBROMYALGIA ANDRADE cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. FIQ (P<0.01) and VAS pain (P=0.02). In the intergroup Table III.—C orrelation between oxygen uptake (VO2) relative to analysis, no significant differences for the trained group lean body mass (LBM) at ventilatory anaerobic threshold (VAT) after the detraining period were observed (P>0.05) and the and peak CPET and clinical manifestations of FMS at baseline. control group did not present significant difference after follow-up (P>0.05). N.=54 Correlation between VO2 relative to LBM and clinical VO2 VO2 manifestations at baseline (mL.kg-1 MCM.min-1) (mL.kg-1 MCM.min-1) no LAV no PICO At baseline, significant correlations were observed be- tween clinical manifestations and VO2 relative to LBM rP rP at VAT and peak CPET of both groups (N.=54). A weak negative correlation was observed between VO2 relative to PPT (kg/cm2) -0.12 0.35 -0.24 0.07 LBM at VAT and FIQ (r = -0.27, P=0.04), BDI (r =-0.26, VAS pain (0-10 cm) -0.14 0.29 -0.11 0.39 P=0.04) and PSQI (r =-0.25, P=0.04). There was also a VAS fatigue (0-10 cm) -0.04 0.77 -0.08 0.55 weak positive correlation between VO2 relative to LBM at VAS well-being (0-10 cm) -0.09 0.49 -0.04 0.77 VAT and the ‘limitation by emotional aspects’ domain of FIQ (0-100) -0.27 0.04 -0.32 0.01 the SF-36 questionnaire (r =0.28, P=0.03). VO2 relative to BDI (0-21) -0.26 0.04 -0.16 0.22 LBM at VAT and the ‘limitation by physical appearance’ BAI (0-21) -0.18 0.18 -0.13 0.32 domain of the SF-36 questionnaire presented a moderate SF-36 positive correlation (r =0.40, P<0.01) and the VO2 relative 0.10 0.44 0.14 0.28 to LBM at VAT at peak CPET and FIQ presented moderate Functional capacity (0-100) 0.40 0.002 0.12 0.36 negative correlation (r =-0.32, P=0.01) (Table III). Physical appearance (0-100) 0.15 0.27 0.05 0.68 Pain (0-100) 0.09 0.50 -0.05 0.69 Correlation between VO2 relative to LBM and clinical General health status (0-100) 0.12 0.37 0.09 0.48 manifestations after APT Vitality (0-100) 0.13 0.32 0.08 0.52 Social aspect (0-100) 0.28 0.03 0.18 0.18 No significant correlations were observed between the Emotional aspect (0-100) 0.20 0.12 -0.08 0.56 magnitude of the improvement in clinical manifestations Mental health (0-100) -0.25 0.04 -0.10 0.61 PSQI total (0-21) Data presented as mean±standard deviation. VAS: Visual Analogical Scale; FIQ: fibromyalgia impact on quality of life; BDI: Beck’s Depression Inventory; BAI: Beck’s Anxiety Inventory; SF-36: quality of life; PSQI: Pittsburgh Sleep Quality Index. and the magnitude of improvement in VO2 relative to LBM for trained group after APT. Figure 3, 4 show the correlation analyses between the increase in VO2 relative ∆VO2VAT_LBM R=0.25 ∆VO2VAT_LBM R=-0.23 ∆VO2VAT_LBM R=-0.04 ∆VO2VAT_LBM R=-0.04 (mL/kgLBM-1/min-1) P= 0.19 (mL/kgLBM-1/min-1) P= 0.22 (mL/kgLBM-1/min-1) P= 0.81 (mL/kgLBM-1/min-1) P= 0.33 A ∆PPT (kg/cm2) B ∆Pain VAS (0-100 mm) A B∆Well-being VAS (0-100 mm) ∆FIQ (0-100) R=0.19 R=-0.22 R=-0.04 ∆VO2PEAK_LBM R=-0.10 P= 0.32 P= 0.26 P= 0.83 (mL/kgLBM-1/min-1) P= 0.58 ∆VO2PEAK_LBM ∆VO2PEAK_LBM ∆VO2PEAK_LBM (mL/kgLBM-1/min-1) (mL/kgLBM-1/min-1) (mL/kgLBM-1/min-1) C D∆PPT (kg/cm2) ∆Pain VAS (0-100 mm) C D∆Well-being VAS (0-100 mm) ∆FIQ (0-100) Figure 3.—Correlation between variation (ΔPost - Pre) of oxygen up- Figure 4.—Correlation between variation (ΔPost - Pre) of oxygen up- take (VO2) relative to lean body mass (LBM) at ventilatory anaerobic take (VO2) relative to lean body mass (LBM) at ventilatory anaerobic threshold (VAT) and peak CPET and clinical manifestations (ΔPost- threshold (VAT) and peak CPET and clinical manifestations (ΔPost- Pre) of FMS. ΔVO2 relative to LBM at VAT and pressure pain threshold Pre) of FMS. ΔVO2 relative to LBM at VAT and visual analogue scale (ΔPPT) (A); visual analogue scale (ΔVAS) pain (C). ΔVO2 relative to (ΔVAS) well-being (A); and on the fibromyalgia impact questionnaire LBM at peak CPET and ΔPPT (B) and ΔVAS pain (D). score (ΔFIQ) (C). ΔVO2 relative to LBM at peak CPET and ΔVAS well- being (B) and ΔFIQ Score (D). Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 85

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically ANDRADE COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, EFFECTS OF TRAINING AND DETRAINING ON FIBROMYALGIA cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. to LBM at VAT, reduction in PPT values (Figure 3A), VAS the peak of CPET after 14 days and 4 weeks of detraining pain (Figure 3C), VAS well-being (Figure 4A) and FIQ in healthy men, respectively. Thus, our results suggest that (Figure 4C) and between the increase in VO2 relative to these losses are related to the supply and use of oxygen LBM at peak CPET and reduction in PPT values (Figure for muscles in activity. Coyle et al.47 reported that after a 3C), VAS pain (Figure 3D), VAS well-being (Figure 4C) detraining period, there is a decline in mitochondrial enzy- and FIQ (Figure 4D). matic activity, with no structural changes at muscular and vascular levels. In the study by St-Amand et al.,19 when There were also no significant correlations between evaluating the effect of detraining on healthy men, they ob- VO2 relative to LBM at VAT and peak CPET and clinical served changes in genes related to oxidative phosphoryla- variables studied in the detraining period (P>0.05). tion (OxPhos). Therefore, these physiological changes sug- gest a lower supply of oxygen to the skeletal muscle, which Discussion may contribute to the reduction of VO2 after APT cessation. The main results of the present study showed that women Regarding the worsening of clinical symptoms after the with FMS undergoing 16 weeks of APT showed improved detraining period, our results can be justified according to aerobic functional capacity and improved clinical symp- the study of Smith and Lyle.43 The authors reported that tomatology such as pain, well-being and the impact of physical training, when continuously performed, stimu- fibromyalgia on quality of life. However, these improve- lates the release of endogenous opioids and, consequently, ments were not maintained after a 16-week detraining pe- increases the nociceptive threshold. However, these ben- riod. In addition, our results show that the magnitudes of efits are quickly lost with detraining.43 On the other hand, improvement in clinical symptomatology are not associ- our results are not in line with the study by Tomas-Carus,17 ated with the improvement in aerobic functional capacity. who performed a predominantly aerobic aquatic exercise program for 24 weeks and observed maintenance of pain The results of the present study corroborate previous reduction after 3 months of detraining. This can be attrib- studies that showed positive effects of APT on clinical uted to the fact that the training program is longer and the symptomatology in women with FMS, such as reduction detraining period is shorter when compared to our study, in number of tender points,6, 24, 39 in FIQ Score and VAS which may explain the difference between the results pain6, 39 and improvement in quality of life.24, 40 found. The improvement of clinical symptoms and aerobic The baseline results of the present study corroborate the gain in women with FMS can be understood by the periph- results of other studies,11, 23, 48 which observed significant eral adaptations promoted by aerobic exercise, which may associations between aerobic functional capacity and clin- include increase of oxidative enzymes at muscle level, in- ical symptomatology in women with FMS, such as FIQ, crease in mitochondrial capacity and density41 and higher BDI, PSQI, “limitation by physical appearance” and “lim- capillary density.42 In this context, it could be speculated itation by emotional aspects” of the SF-36 questionnaire. that increased muscle oxygenation as a result of peripheral Carbonell-Baeza et al.11 observed that the aerobic func- adaptations may have contributed to an increase in endog- tional capacity, as measured by the distance walked on the enous opioids43 and consequently reduction of pain44 and 6MWT, was inversely related to pain, as measured by the impact of FMS on quality of life. number of active tender points and by PPT in women with FMS. Similar results were reported by Soriano-Maldonado We should also highlight the acute benefits during exer- et al.,23, 48 who showed that high cardiorespiratory fitness cises in the aquatic environment, such as reduced gravity, is associated with low pain sensitivity and low impact of increased hydrostatic pressure, circulation and resistance fibromyalgia on quality of life; however, cardiorespiratory to movement.45 These factors may have contributed in an fitness is not associated with pain assessed by VAS pain.48 effective way to the improvement of momentary clinical On the other hand, it is noteworthy that these studies did symptoms, such as relief of pain, fatigue, and improved not evaluate the effect of any type of treatment for these well-being. patients, which may be a bias in the understanding of how these variables behave over time. Despite the interesting findings of the effects of APT, the gains obtained related to improved aerobic capacity and However, the results of the present study after APT symptomatology did not persist after 16 weeks of detrain- show a lack of association between the magnitude of im- ing and the values returned to near-baseline levels. These provement in clinical symptomatology and aerobic func- results corroborate studies by Mujika et al.46 and Murias et al.18 who also observed VO2 reduction in absolute units at 86 European Journal of Physical and Rehabilitation Medicine February 2019

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, EFFECTS OF TRAINING AND DETRAINING ON FIBROMYALGIA ANDRADE cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. tional capacity, even after correction of VO2 by LBM. Peak oxygen uptake and ventilatory anaerobic threshold in fibromyalgia. These findings agree with previous studies, which also J Rheumatol 2002;29:353–7. did not identify association between clinical improvement 6.  Zamunér AR, Andrade CP, Forti M, Marchi A, Milan J, Avila MA, et and increase in VO2 corrected by TBM after an exercise al. Effects of a hydrotherapy programme on symbolic and complexity program.5, 6, 44 This result reinforces the theory that FMS dynamics of heart rate variability and aerobic capacity in fibromyalgia is more complex than just a metabolic and sensory altera- patients. Clin Exp Rheumatol 2015 b;33(Suppl 88):S73–81. tion, and there is an important impairment of behavioral 7.  Mork PJ, Vasseljen O, Nilsen TI. Association between physical exer- and cognitive functions.5 Thus, it is necessary to follow cise, body mass index, and risk of fibromyalgia: longitudinal data from the recommendations proposed by the American Pain Soci- Norwegian Nord-Trøndelag Health Study. Arthritis Care Res (Hoboken) ety (APS)49 and by the Canadian Pain Society (CPS),50 in 2010;62:611–7. which physical exercise in fibromyalgic patients should 8.  Aparicio VA, Ortega FB, Heredia JM, Carbonell-Baeza A, Delgado- respect individuality and be performed in according to Fernández M. [Analysis of the body composition of Spanish women with their limits, since the focus primarily on the increase of fibromyalgia]. Reumatol Clin 2011;7:7–12. Spanish. VO2 will not necessarily reflect clinical improvement. 9.  Segura-Jimenez V, Aparicio VA, Alvarez-Gallardo IC, Carbonell-Bae- za A, Tornero-Quinones I, Delgado-Fernandez M. Does body composition Limitations of the study differ between fibromyalgia patients and controls? the al-Ándalus project. Clin Exp Rheumatol 2015;33(Suppl 88):S25–32. Among the study limitations, the lack of monthly reassess- 10.  Arnold LM, Clauw DJ. Challenges of implementing fibromyal- ments both in the APT period and in the detraining period gia treatment guidelines in current clinical practice. Postgrad Med due to the unavailability of volunteers who participated in 2017;129:709–14. the study stands out. This would better elucidate the mo- 11.  Carbonell-Baeza A, Aparicio VA, Ortega FB, Cuevas AM, Alvarez ments of gains and losses of the studied variables. More- IC, Ruiz JR, et al. Does a 3-month multidisciplinary intervention improve over, although medication doses were kept constant during pain, body composition and physical fitness in women with fibromyalgia? the entire study, future studies should account for a pos- Br J Sports Med 2011;45:1189–95. sible confounding effect on the outcomes. 12.  Bardal EM, Roeleveld K, Mork PJ. Aerobic and cardiovascular au- tonomic adaptations to moderate intensity endurance exercise in patients Conclusions with fibromyalgia. J Rehabil Med 2015;47:639–46. 13.  Sevimli D, Kozanoglu E, Guzel R, Doganay A. The effects of aquatic, It could be concluded that 16 weeks of APT was effective isometric strength-stretching and aerobic exercise on physical and psy- in promoting VO2 increase at VAT and peak CPET and chological parameters of female patients with fibromyalgia syndrome. J improved clinical symptomatology in women with FMS. Phys Ther Sci 2015;27:1781–6. However, after 16 weeks of detraining period, there was 14.  Andrade CP, Zamunér AR, Forti M, França TF, Tamburús NY, Silva a return near baseline. Thus, APT should be continuously E. Oxygen uptake and body composition after aquatic physical training performed in order to improve clinical symptomatology in women with fibromyalgia: a randomized controlled trial. Eur J Phys and increase the aerobic functional capacity of women Rehabil Med 2017;53:751–8. with FMS. 15.  Sañudo B, Carrasco L, de Hoyo M, McVeigh JG. Effects of exercise training and detraining in patients with fibromyalgia syndrome: a 3-yr References longitudinal study. Am J Phys Med Rehabil 2012;91:561–9, quiz 570–3. 16.  Gowans SE, Dehueck A, Voss S, Silaj A, Abbey SE. Six-month and 1.  Wolfe F, Smythe HA, Yunus MB, Bennett RM, Bombardier C, Gold- one-year followup of 23 weeks of aerobic exercise for individuals with enberg DL, et al. The American College of Rheumatology 1990 Criteria fibromyalgia. Arthritis Rheum 2004;51:890–8. for the Classification of Fibromyalgia. Report of the Multicenter Criteria 17.  Tomas-Carus P, Häkkinen A, Gusi N, Leal A, Häkkinen K, Ortega- Committee. Arthritis Rheum 1990;33:160–72. Alonso A. Aquatic training and detraining on fitness and quality of life in 2.  Häuser W, Thieme K, Turk DC. Guidelines on the management of fi- fibromyalgia. Med Sci Sports Exerc 2007;39:1044–50. bromyalgia syndrome - a systematic review. Eur J Pain 2010;14:5–10. 18.  Murias JM, Edwards JA, Paterson DH. Effects of short-term training 3.  Heymann RE, Paiva ES, Martineza JE, Jra MH, Rezende MC, Proven- and detraining on VO2 kinetics: faster VO2 kinetics response after one za JR, et al. Novas diretrizes para o diagnóstico da fibromialgia. Rev Bras training session. Scand J Med Sci Sports 2016;26:620–9. Reumatol 2017;57:S467–76. 19.  St-Amand J, Yoshioka M, Nishida Y, Tobina T, Shono N, Tanaka H. 4.  Talotta R, Bazzichi L, Di Franco M, Casale R, Batticciotto A, Gerardi Effects of mild-exercise training cessation in human skeletal muscle. Eur MC, et al. One year in review 2017: fibromyalgia. Clin Exp Rheumatol J Appl Physiol 2012;112:853–69. 2017;35(Suppl 105):6–12. 20.  Gavi MB, Vassalo DV, Amaral FT, Macedo DC, Gava PL, Dantas 5.  Valim V, Oliveira LM, Suda AL, Silva LE, Faro M, Neto TL, et al. EM, et al. Strengthening exercises improve symptoms and quality of life but do not change autonomic modulation in fibromyalgia: a randomized clinical trial. PLoS One 2014;9:e90767. 21.  Córdoba-Torrecilla S, Aparicio VA, Soriano-Maldonado A, Estévez- López F, Segura-Jiménez V, Álvarez-Gallardo I, et al. Physical fitness is associated with anxiety levels in women with fibromyalgia: the al-Ánda- lus project. Qual Life Res 2016;25:1053–8. 22.  Carbonell-Baeza A, Aparicio VA, Sjöström M, Ruiz JR, Delgado- Fernández M. Pain and functional capacity in female fibromyalgia pa- tients. Pain Med 2011;12:1667–75. 23.  Soriano-Maldonado A, Ortega FB, Munguía-Izquierdo D. Associa- tion of cardiorespiratory fitness with pressure pain sensitivity and clinical pain in women with fibromyalgia. Rheumatol Int 2015b;35:899–904. 24.  Munguía-Izquierdo D, Legaz-Arrese A. Assessment of the effects Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 87

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically ANDRADE COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, EFFECTS OF TRAINING AND DETRAINING ON FIBROMYALGIA cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. of aquatic therapy on global symptomatology in patients with fibromy- 38.  Cohen J. The concepts of power analysis. In: Cohen J, editor. Statisti- algia syndrome: a randomized controlled trial. Arch Phys Med Rehabil cal Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrence 2008;89:2250–7. Erlbaum Associates; 1988. p.1-17. 25.  Hooten WM, Rosenberg CJ, Eldrige JS, Qu W. Knee extensor strength 39.  McLoughlin MJ, Stegner AJ, Cook DB. The relationship between is associated with pressure pain thresholds in adults with fibromyalgia. physical activity and brain responses to pain in fibromyalgia. J Pain PLoS One 2013;8:e59930. 2011;12:640–51. 26.  Wolfe F, Clauw DJ, Fitzcharles MA, Goldenberg DL, Katz RS, Mease 40.  Rahman A, Underwood M, Carnes D. Fibromyalgia. BMJ P, et al. The American College of Rheumatology preliminary diagnostic 2014;348:g1224. criteria for fibromyalgia and measurement of symptom severity. Arthritis 41.  Cordero MD, De Miguel M, Moreno Fernández AM, Carmona López Care Res (Hoboken) 2010;62:600–10. IM, Garrido Maraver J, Cotán D, et al. Mitochondrial dysfunction and 27.  Craig CL, Marshall AL, Sjöström M, Bauman AE, Booth ML, Ain- mitophagy activation in blood mononuclear cells of fibromyalgia pa- sworth BE, et al. International physical activity questionnaire: 12-country tients: implications in the pathogenesis of the disease. Arthritis Res Ther reliability and validity. Med Sci Sports Exerc 2003;35:1381–95. 2010;12:R17. 28.  Marques AP, Santos AM, Assumpção A, Matsutani LA, Lage LV, 42.  Morf S, Amann-Vesti B, Forster A, Franzeck UK, Koppensteiner R, Pereira CA. Validation of the Brazilian version of the Fibromyalgia Im- Uebelhart D, et al. Microcirculation abnormalities in patients with fibro- pact Questionnaire (FIQ). Br J Rheumatol 2006;46:24–31. myalgia - measured by capillary microscopy and laser fluxmetry. Arthritis 29.  Gorenstein C, Andrade L. Validation of a Portuguese version of the Res Ther 2005;7:R209–16. Beck Depression Inventory and the State-Trait Anxiety Inventory in Bra- 43.  Smith MA, Lyle MA. Chronic exercise decreases sensitivity to mu zilian subjects. Braz J Med Biol Res 1996;29:453–7. opioids in female rats: correlation with exercise output. Pharmacol Bio- 30.  Cunha JA. Manual da versão em português das escalas Beck [Man- chem Behav 2006;85:12–22. ual of the Portuguese version of the Beck scales]. São Paulo: Casa do 44.  Hooten WM, Qu W, Townsend CO, Judd JW. Effects of strength vs Psicólogo; 2001. aerobic exercise on pain severity in adults with fibromyalgia: a random- 31.  Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health ized equivalence trial. Pain 2012;153:915–23. survey (SF-36). I. Conceptual framework and item selection. Med Care 45.  Mannerkorpi K. Exercise in fibromyalgia. Curr Opin Rheumatol 1992;30:473–83. 2005;17:190–4. 32.  Bertolazi AN, Fagondes SC, Hoff LS, Dartora EG, Miozzo IC, de 46.  Mujika I, Padilla S. Cardiorespiratory and metabolic characteristics Barba ME, et al. Validation of the Brazilian Portuguese version of the of detraining in humans. Med Sci Sports Exerc 2001;33:413–21. Pittsburgh Sleep Quality Index. Sleep Med 2011;12:70–5. 47.  Coyle EF, Martin WH 3rd, Bloomfield SA, Lowry OH, Holloszy JO. 33.  Heyward VH, Stolarczyk LM. Avaliação da composição corporal Effects of detraining on responses to submaximal exercise. J Appl Physiol aplicada. First edition. São Paulo: Manole; 2000. (1985) 1985;59:853–9. 34.  Lukaski HC, Bolonchuk WW, Hall CB, Siders WA, William AS. Vali- 48.  Soriano-Maldonado A, Henriksen M, Segura-Jiménez V, Aparicio dation of tetrapolar bioelectrical impedance method to assess human body VA, Carbonell-Baeza A, Delgado-Fernández M, et al. Association of composition. J Appl Physiol (1985) 1986;60:1327–32. Physical Fitness With Fibromyalgia Severity in Women: the al-Ándalus 35.  Wasserman K, Hansen JE, Sue D, Stringer W, Whipp B. Principles of Project. Arch Phys Med Rehabil 2015a;96:1599–605. exercise testing and interpretation. Philadelphia: Williams and Wilkins; 49.  Buckhardt CS, Goldenberg D, Crofford L, Gerwin R, Gowens S, 1999. Jackson K, et al. Guideline for the management of fibromyalgia syndrome 36.  Zamunér AR, Catai AM, Martins LE, Sakabe DI, Da Silva E. Identifi- pain in adults and children (Clinical practice guideline; no.4). Glenview, cation and agreement of first turn point by mathematical analysis applied IL: American Pain Society; 2005. to heart rate, carbon dioxide output and electromyography. Braz J Phys 50.  Fitzcharles MA, Ste-Marie PA, Goldenberg DL, Pereira JX, Ab- Ther 2013;17:614–22. bey S, Choinière M, et al.; National Fibromyalgia Guideline Advi- 37.  de Souto Araujo ZT, de Miranda Silva Nogueira PA, Cabral EE, de sory Panel. 2012 Canadian Guidelines for the diagnosis and manage- Paula Dos Santos L, da Silva IS, Ferreira GM. Effectiveness of low-inten- ment of fibromyalgia syndrome: executive summary. Pain Res Manag sity aquatic exercise on COPD: a randomized clinical trial. Respir Med 2013;18:119–26. 2012;106:1535–43. Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript. Funding.—This study received financial support from the São Paulo Research Foundation Support (FAPESP), process: # 2011 / 22122-5 and # 2013 / 17504- 1 and from the National Council for Scientific and Technological Development (CNPq), process: 307187 / 2013-6. Article first published online: July 6, 2018. - Manuscript accepted: July 5, 2018. - Manuscript revised: June 4, 2018. - Manuscript received: October 16, 2017. 88 European Journal of Physical and Rehabilitation Medicine February 2019

COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically © 2018 EDIZIONI MINERVA MEDICA European Journal of Physical and Rehabilitation Medicine 2019 February;55(1):89-94 or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access Online version at http://www.minervamedica.it DOI: 10.23736/S1973-9087.18.05314-5 to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. ORIGINAL ARTICLE Validity and reliability of Italian version of the Non-Communicating Children’s Pain Checklist: revised version Massimiliano MURGIA 1, Rosanna IZZO 1 *, Alessandra BETTINELLI 1, Chiara DI MAGGIO 2, Martina DE ANGELIS 2, Massimiliano MANGONE 1, Marco PAOLONI 1, Andrea BERNETTI 1, Andrea TORQUATI 1, Vincenzo LEUZZI 2, Valter SANTILLI 1 1Department of Physical Medicine and Rehabilitation, Sapienza University, Rome, Italy; 2Department of Pediatrics, Child Neurology and Psychiatry, Sapienza University, Rome, Italy *Corresponding author: Rosanna Izzo, Piazzale Aldo Moro 5, 00185 Rome, Italy. E-mail: [email protected] ABSTRACT BACKGROUND: The Non-Communicating Children Pain Check List-Revised (NCCPC-R) is a clinical assessment tool used to assess and measure pain in children aged 3 to 18 years, with mental and intellectual disabilities, incapable to speak. AIM: Aim of our study was to test the validity and reliability of the Italian version of the NCCPC-R in children with cognitive impairment, in order to obtain a valid tool for pain assessment in these children. DESIGN: Prospective observational study. SETTING: Pediatric Outpatient of Physical and Rehabilitative Medicine Department, and Pediatrics, Child Neurology and Psychiatry Depart- ment, Sapienza University, Rome. POPULATION: 55 non-communicating children, with severe intellectual disability, aged 3-18 years. METHODS: The guidelines for “translation, adaptation, and validation of instruments or scales for cross-cultural healthcare research” were used to translate the scale, which was administered by the parents/caregivers twice for 2 consecutive days, in association with NRS (Numerical Rating Scale). The reliability of the scale was evaluated using the intra and inter-class correlation coefficient (ICCs); Cronbach alpha coefficient was used to test the internal validity of the scale; “Receiver Operating Characteristic” (ROC) curves were used to compare pain-free scores with pain scores, determining threshold scores; Pearson correlation between NCCPC-R and NRS values was measured. RESULTS: The InterCC between the first and the second interviewer at T0 was 0,97, the IntraCC of the first interviewer at T0-T1 was 0,89, showing a high correlation; the Cronbach alpha coefficient at T0 was 0,97, showing a high scale’s validity. Pearson correlation between NRS and NCCPC-R values at T0 was 0,54 showing a medium level of agreement (P<0,0001). AUC (area under the curve) obtained by ROC curve at T0 was 0,807 (P=0,001), with sensitivity 95,2 and specificity 55,6, while a T1 AUC was 0,814 (P<0,001), with sensitivity 86,49, specificity 78,57. CONCLUSIONS: The italian version of NCCPC-R scale could be used to asses pain in non-communicating patients with mental and intellectual disabilities, showing a good correlation when compared to the NRS. CLINICAL REHABILITATION IMPACT: The use of NCCPC-R scale in daily life made parents/caregivers able to discriminate the presence/ absence of pain in non-communicating children, based on the scores obtained with the questionnaire. (Cite this article as: Murgia M, Izzo R, Bettinelli A, Di Maggio C, De Angelis M, Mangone M, et al. Validity and reliability of Italian version of the Non-Communicating Children’s Pain Checklist: revised version. Eur J Phys Rehabil Med 2019;55:89-94. DOI: 10.23736/S1973-9087.18.05314-5) Key words: Communication - Child - Parents - Pain - Rehabilitation. Pain is a stressful experience considered a global health Detection and management of pain in children with problem. Despite children are the most vulnerable and intellectual disabilities (ID) are relevant issues due to its under-served population, actually, painful experience in high prevalence and impact on quality of life.4 Children children is rarely understood and estimated, also for the with cognitive impairment have painful experiences more difficulty in its discrimination and objective evaluation.1 frequently than healthy children, due to their chronic con- Although scientific evidence about pediatric pain are re- ditions, disabilities and associated diseases (muscular con- cently increasing, this problem is still little faced in the tractures, spasticity, chronic constipation, gastro-esoph- daily clinical practice.2, 3 ageal reflux, hip-luxation, bone fractures, tooth decay); Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 89

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, MURGIA ITALIAN VERSION OF THE NON-COMMUNICATING CHILDREN’S PAIN CHECKLIST cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. furthermore they are subjected to frequent injuries and Psychiatry Department, Sapienza University, Rome, be- medical procedures.5 Painful experiences in these children tween September 2017 and February 2018. The study was often remain unrecognized and untreated, due to their in- approved by the ethical committee of Umberto I Policlin- adequate capacities to self-report pain. ico of Rome, and all parents gave their signed informed consent to the study. We recruited 55 non-communicating The Non-Communicating Children’s Pain Checklist – children, with severe ID evaluated with Vineland Adaptive Postoperative Version (NCCPC-PV), validated by Breau Behaviour Scales – Second Edition.13 Inclusion criteria of et al.6 is a clinical assessment tool used to assess and mea- the study were: age between 3 and 18 years, subjects not sure pain in children aged 3 to 18 years, with mental and able to communicate verbally, compliance of interview- ID, associated or not with physical disabilities, incapable ers in data collection and compilation of questionnaires, to speak. This instrument, intended for use by medical fig- interviewer’s ability to understand the items. Failure send- ures, was designed for evaluation of post-operative pain ing the completed questionnaire by the interviewers has in non-communicating children (especially for pain after been considered as exclusion criteria. For each patient, we surgery or pain due to procedures conducted in hospital). collected the following data: age, sex, disease, upper limb The NCCPC-PV scale consists of 27 items regarding vo- use, lower limb use, medical monitoring (weekly, monthly, cal, social, facial, activities, body and limbs movements six-monthly), and demographic information concerning and physiological status (total score 0-81). The NCCPC- the families (education, marital status, number of sons). PV has been validated in French7 and Swedish8 language. The guidelines for “the translation, adaptation, and valida- Recently, Zanchi et al.5 validated the Italian version of tion of instruments or scales for cross-cultural healthcare NCCPC-PV in 40 children (3-18 years of age) with se- research” have been used to translate the scale.14 In step 1 vere to profound ID, demonstrating that the consistency of two translators, individually, translated in Italian language evaluation between different observers was poor in a calm the NCCPC-R, producing two Italian versions of this situation and fair in a painful situation, with ICC values scale; successively in step 2 a different bilingual transla- similar to those found in the Swedish experience. tor, compared the two versions produced, and a third ver- sion of the scale in Italian was generated. In step 3 two The Non-Communicating Children Pain Check List- different English native speakers translated back to Eng- Revised (NCCPC-R) has been designed by Breau et al.9 lish the preliminary Italian version of the NCCPC-R, and to assess pain during daily life, intended for use by par- obtained two different back-translations of the NCCPC-R. ents or caregivers who usually care the disabled child, at In step 4 a multidisciplinary committee compared the two home or in long-term residential structures. The NCCPC- back-translations and the original form of English scale. R consists of 30 items regarding vocal, social, facial, ac- In step 5 a group of Italian health specialists preliminary tivities, body and limbs movements, physiological status tested in 15 target patients the NCCPC-R, evaluating the and introduces eating/sleeping parameters, not included in instructions, items and format clarity. In step 6, the multi- the Post-Operative version (total score 0--90).10 Currently, disciplinary committee approved the final Italian version there is no validated Italian version of the NCCPC-R; so of NCCPC-R, revising and refining the items of the instru- the absence of an Italian version of pain assessment tool ment. in non-communicating children makes difficult the evalu- ation of child’s pain, and its treatment, by physicians. The Italian version of NCCPC-R scale consists of 30 items regarding vocal, social, facial, activities, body and The aim of our study was to translate the NCCPC-R limbs movements, physiological status and eating/sleep- into Italian and to test the reliability and validity of the ing parameters; the time required to complete the NCCPC- Italian version of the NCCPC-R in children with cognitive R was two hours (approximately 120 minutes), in which impairment, by means of the intra- and inter-class correla- the observer (parent or caregiver) evaluated the child. tion (CC) coefficients, respectively,11, 12 in order to obtain During the two hours, for each item the observer defined a valid tool for pain assessment in these children and to how much a certain behaviour was present: 0=never pres- apply, in the future, this scale in clinical practice. ent, 1=only a little, 2=quite often, 3=very often, NA=not applicable because the child does not is able to perform Materials and methods the action (Figure 1). At time T0, both parents/caregivers observed their child for two consecutive hours, at home Our prospective observational study was conducted at the or in long-term residential structures, regardless of child’s Pediatric Outpatient of Physical and Rehabilitative Medi- cine Department, and Pediatrics, Child Neurology and 90 European Journal of Physical and Rehabilitation Medicine February 2019

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, ITALIAN VERSION OF THE NON-COMMUNICATING CHILDREN’S PAIN CHECKLIST MURGIA cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. NON-­‐COMMUNICATING  CHILDREN’S  PAIN  CHECKLIST  -­‐  REVISED  (NCCPC-­‐R)   tion; 0.50-0.69=moderate correlation; 0.70-0.89=high cor- relation; and 0.90-1=very high correlation. The IntraCC NOME     COGNOME   DATA:   has been calculated only in those patients showing a NRS OSSERVATORE:   ORA  D’INIZIO:   ORA  DI  FINE:   difference ≤-1 or ≥+1 between T0 and T1. The Cronbach Quante  volte  il  bambino  ha  mostrato  questi  comportamenti  nelle  ultime  due  ore?  Indicare  il  numero  per  ogni  voce.  Se   alpha coefficient12 has been used to test the internal valid- una   voce   non   è   applicabile   al   bambino   (per   esempio,   il   bambino   non   mangia   cibi   solidi),   allora   indicare   “non   ity (correlation of items between them) of the NCCPC-R applicabile”  per  quella  voce.     scale (with 95% confidence interval CI); in literature a val- 0=non  del  tutto                              1=un  po’                                  2=abbastanza  spesso                              3=molto  spesso                                  NA=non  applicabile     ue greater than 0.7 is considered acceptable. The standard error measurement (SEM)15 has been calculated to provide I.  ESPRESSIONE  VOCALE   0   1   2   3   NA   an absolute index of reliability. SEM has been estimated 1.  Gemiti,  lamenti,  piagnucolii  (piuttosto  deboli)…………………………………………………………….…..   0   1   2   3   NA   as follows: SEM=SD * √(1-ICC). The smallest detectable 2.  Pianto  (moderatamente  sonoro)……………………………………………………………………………………..   0   1   2   3   NA   difference (SDD), defined as the minimal change that falls 3.  Urla  (molto  forti)………………………………………………………………………………………………………………   0   1   2   3   NA   outside the measurement error in the score of an instru- 4.  Una  parola  specifica  per  il  dolore  (per  es.  una  parola,  un  lamento,  un  ghigno)…………………   0   1   2   3   NA   ment used to make a measurement, has been calculated as II.  SOCIALITA’   0   1   2   3   NA   1.96*√(2*SEM). The “Receiver Operating Characteristic” 5.  Non  collaborante,  nervoso,  irritabile,  triste………………………………………………………………………   0   1   2   3   NA   (ROC) curves have been used to compare pain-free scores 6.  Minor  interazione  con  gli  altri,  chiuso  in  sé  stesso……………………………………………………………   0   1   2   3   NA   with pain scores at T0 and T1, to determine the threshold 7.  Cerca  consolazione  o  contatto  fisico…………………………………………………………………………………   0   1   2   3   NA   scores; finally Pearson correlation has been used as a mea- 8.  E’  difficile  da  distrarre,  non  si  riesce  a  soddisfarlo  o  calmarlo……………………………………………   0   1   2   3   NA   sure of linear correlation between NCCPC-R values and III.  VOLTO     0   1   2   3   NA   NRS values. The level of statistical significance was set 9.  Fronte  aggrottata…………………………………………………………………………………………………………….   0   1   2   3   NA   at P<0.05. 10.  Cambiamento  nello  sguardo,  incluso:  occhi  strizzati,  sbarrati,  accigliati  ………………………..   0   1   2   3   NA   11.  Bocca  rivolta  all’ingiù,  non  sorride………………………………………………………………………………….   0   1   2   3   NA   For the estimation of the sample size we used the al- 12.  Labbra  “arricciate”,  serrate,  imbronciate  o  tremolanti…………………………………………………..   0   1   2   3   NA   gorithm proposed by Walter et al.16 Setting these values 13.  Serra  o  digrigna  i  denti,  mastica  o  mostra  la  lingua………………………………………………………..   0   1   2   3   NA   in the algorithm:• n (observer number)=2,• α (significance IV.  ATTIVITA’  MOTORIA   0   1   2   3   NA   level)=0.05,• (1-β) (power)=0.850,• ρ0 (acceptable reli- 14.  Non  si  muove,  è  meno  attivo,  silenzioso………………………………………………………………………..   0   1   2   3   NA   ability value)=0.70,• ρ1 (expected reliability value)=0.85,• 15.  Saltella,  è  agitato,  irrequieto………………………………………………………………………………………….   0   1   2   3   NA   percentage of Drop-out=10%, we obtain a value for sam- V.  CORPO  E  ARTI     0   1   2   3   NA   ple size equal to 55. 16.  Flosci……………………………………………………………………………………………………………………………..   0   1   2   3   NA   17.  Rigidi,  spastici,  tesi…………………………………………………………………………………………………………   18.  Indica  o  tocca  la  parte  del  corpo  che  duole……………………………………………………………………   19.  Protegge,  tutela  la  parte  del  corpo  che  duole………………………………………………………………..   20.  Sottrae  o  allontana  la  parte  del  corpo,  è  sensibile  al  tatto……………………………………………..   21.  Muove  il  corpo  in  un  modo  specifico  che  indica  dolore  (per  es.  porta  la  testa  indietro,     le  braccia  in  giù,  si  rannicchia)   VI.  STATO  FISIOLOGICO     0   1   2   3   NA   22.  Trema…………………………………………………………………………………………………………………………...   0   1   2   3   NA   23.  Cambiamento  del  colorito,  pallore………………………………………………………………………………..   0   1   2   3   NA   24.  Suda………………………………………………………………………………………………………………………………   0   1   2   3   NA   25.  Lacrima………………………………………………………………………………………………………………………….   0   1   2   3   NA   26.  Respiro  affannoso,  ansimante……………………………………………………………………………………….   0   1   2   3   NA   27.  Trattiene  il  respiro…………………………………………………………………………………………………………   0   1   2   3   NA   VII.  MANGIARE/DORMIRE   0   1   2   3   NA   28.  Mangia  meno,  non  prova  interesse  nel  cibo………………………………………………………………….   0   1   2   3   NA   29.  Dorme  di  più………………………………………………………………………………………………………………….   30.  Dorme  di  meno……………………………………………………………………………………………………………..     PUNTEGGIO  TOTALE:   CATEGORIA     I   II   III   IV   V   VI   VII   TOTALE                 PUNTEGGIO     Figure 1.—The Italian version of NCCPC-R Scale. health status (not necessarily in presence of pain); the same Results procedure has been repeated at time T1, the next day, 24 hours after the first observation, by the two parents/care- Fifty-five patients (mean age: 12.07 years) have been re- givers. Each observer also filled in at T0 and T1 the NRS cruited in our study; of these, 2 patients did not return the (Numerical Rating Scale) for pain assessment, ranged completed questionnaire, so 53 patients were included (22 from 0 (absence of pain) to 10 (the highest possible pain), women, 31 men). Collected data are reported in Table I. defining what, in its opinion, was the perception of pain in Cerebral palsy were diagnosed in 21 patients; 13 children the child. The filled questionnaires have been returned at had different specified complex syndromes (2 Down, 1 So- the end of the observation to our Department. tos, 1 West, 1 Zimmermann Laband, 1 Dravet, 1 Pelizaeus Merzbacher, 1 Glutaric Aciduria, 1 Ceroid-lipofuscinosis, Statistical analysis 1 Septo-optic dysplasia, 1 FORL1 gene deficit, 1 GNAO1 gene deficit, 1 AADC gene deficit), 8 unknown genetic Statistical analysis has been performed using the Med- syndromes; 5 epileptic syndromes, 3 Rett syndromes, 3 Calc Version 10.2.0.0 for Windows (https://www.medcalc. Angelman syndromes. All children had several mental and org). The reliability of the scale has been evaluated using ID, with Vineland score mean 37.95±5.15. The InterCC the intra and inter-class correlation coefficients (ICCs);11 between the first and the second interviewer at T0 was the ICC varies from 0 to 1, with 1 corresponding to to- 0.97, showing a very high correlation (95% CIs ranged tal reliability. The ICC has been interpreted as follows: from 0.95 to 0.98). The IntraCC has been calculated on 0.00-0.25=little or no correlation; 0.26-0.49=low correla- 26 patients and was 0.89 (95% CIs ranged from 0.75 to 0.95). The Cronbach alpha coefficient measured between Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 91

COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA MURGIA ITALIAN VERSION OF THE NON-COMMUNICATING CHILDREN’S PAIN CHECKLIST This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow accessTable I.—C ollected data patients, and families’ demographic in-the first and the second interviewer at T0 was 0.97, using to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove,formation. 95% lower confidence limit, showing a high validity of the cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. scale. The SEM mean value was 2.05 (SDD=3.96). The Number of patients Females: 22 ROC curve was obtained ad T0, comparing patients with SSeennssiittiivviittyyMean of age Males: 31 NRS=0 with the remaining patients who had a NRS value SSeennssiittiivviittyyVineland Score ≥1: the area under the ROC curve (AUC) obtained was a Mother education 12.07 years measure of how the scale can distinguish between children with pain and children without pain. The AUC at T0 was Father education 37.94±5.15 0,807 (P=0.001), with a sensitivity of 95.2 and a specific- ity of 55.6 and threshold value >9 (Figure 2). The same Number of children in family Less than high school: 18 curve was measured at T1: the AUC was 0,814 (P<0.001), High school: 23 with a sensitivity of 86.49 and a specificity of 78.57 and Marital status University: 12 threshold value >8 (Figure 3). Pearson correlation between Upper limb use NRS and NCCPC-R values was 0.54 (95% Cis: 0.41 to Less than high school: 22 0.64-P<0.0001), showing a medium level of agreement. Lower limb use High school: 19 University: 12 Discussion Requires medical monitoring 1 son: 13 According to our results, the Italian version of the NC- Diseases 2 sons: 26 CPC-R scale appears to be a reliable and valid tool to 3 sons: 10 assess pain of non-communicating children, when used 4 sons: 4 by their parents and/or caregivers. This is the first valida- tion study of an Italian version of the NCCPC-R scale, Married: 44 obtained after a careful and guideline-based translation Not married: 9 procedure from the original English-language version, which is potentially of great importance to assess pain in None: 16 Some: 29 Full: 8 None: 20 Some: 25 Full: 8 Six-monthly: 26 Monthly: 26 Weekly: 1 Cerebral palsy: 21 Unknown Genetic Syndrome: 8 Epileptic syndrome: 5 Rett Syndrome: 3 Angelman Syndrome: 3 Rare and complex Diseases: 13 NNCCCCPPCC--RR T1 NCCCCPPCC--RR 1000 10000 800 8800 600 6600 400 4400 200 2200 00 AUC==0.08,08707 00 AUC==0.081,8414 0 PP==0.000,0101 0 PP<<0.000,01 01 20 440 60 880 1100 20 40 660 80 100 110000--SSppeeccifiifciciittyy 100-Speecciifficity Figure 2.—The area under the ROC curve (AUC) at T0 was 0.807 Figure 3.—The area under the ROC curve (AUC) at T1 was 0.814 (P=0.001), with sensitivity 95.2, specificity 55.6, and threshold value >9. (P<0.001), with sensitivity 86.49, specificity 78.57 and threshold value >8. 92 European Journal of Physical and Rehabilitation Medicine February 2019

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, ITALIAN VERSION OF THE NON-COMMUNICATING CHILDREN’S PAIN CHECKLIST MURGIA cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. this particular population of young patients with chronic In the process of translation and validation of our scale, illnesses and disabilities, considering that currently there although we considered Zanchi et al.5 translation, we are no other pain assessment tools in non-communicating preferred drawing up a new version of the scale, show- children. Particularly, we found a high Inter and IntraCCs ing minimal semantic differences compared to the Italian in our population that, together with the high value of postoperative version, making our scale more personal- Cronbach alpha coefficient, indicate that this instrument ized and adaptable to the observer who reads it. shows high reliability and validity. Reliability and validi- ty are two important parameters of a scale: a high reliabil- The main difference between the postoperative ver- ity, evaluated through repeatability and reproducibility of sion6-8 and revised version10 is the possible application the instrument, is associated to a low random error, while of this scale in normal daily life, in home setting, when a high validity represents a good ability to evaluate the the child is not subjected to medical procedures. The daily condition for which the instrument is proposed. application of the scale might be important because the non-communicating child very often is not able to express In order to determine cut-off level, results obtained and quantify its painful symptomatology,18, 19 so the ques- with ROC curve at T0 show a high sensitivity (95.2), but tionnaire can be used by who, during the day, take care or a lower specificity (55.6), while at T1 a high sensitivi- interact with the child. ty (86.49) and specificity (78.57) can be observed, with threshold value >8: the results show that the best thresh- The use of this scale in clinical practice could be very old value to distinguish between children with pain and important also for physicians following the patient (not children without pain is a value of NCCPC-R>8. Our only the physiatrists, but also the pediatricians, the ortho- results are very similar to those obtained by Zabalia et pedists, the neurologists, the child neuropsychiatrists, the al.7 for the validation of French version of NCCPC-PV, in therapists and others medical/health figures following the which the threshold value obtained between the absence patient), that show often low attention to pain in children of pain and the presence of mild pain was 7, and sensitivi- with severe cognitive impairment; so a specific education- ty and specificity were 84 and 68 respectively; so parents/ al efforts should be done to improve the quality of care in caregivers of non-communicating children with suspected these patients, as reported also by Barbi et al.20 Further- pain, should be educated that the cut-off to suspect pain more, we could think in the future different applications in their children should be considered as >8 point on the of the scale, for example to evaluate the clinical efficacy NCCPC-R Scale. of pharmacological or rehabilitative treatments, for which these children are often subjected. Pearson correlation between the NCCPC-R and the NRS Scale showed a medium grade correlation, although Limitations of the study statistically significant: probably these results are due to the extremely subjective evaluation of pain performed by This study shows some limitations: restricted number of re- parents, using NRS Scale. These results agree with data cruited patients (N.53), the questionnaire’s administration already present in the literature: our results, in fact, are given only twice; furthermore, children were observed in similar to data showed by Crosta et al.,17 in which NC- their homes, by observers not physicians, without a stan- CPC-PV and NRS values are compared with a Pearson dardized pain stimulus, so, there was a high variability in correlation. quality and intensity of pain. Even if the original version of the NCCPC-R was de- Conclusions signed to assess the presence of pain,10 for the validation of the Italian version of NCCPC-R the parents were asked The assessment of pain in patients with mental and ID is to complete the questionnaire during normal daily life, re- very complex and difficult, due to their inability to commu- gardless of child’s health status. So, our study does not nicate pain: in this way, pain can be often underestimated compare a situation of pain with one of no-pain: we tried and untreated.21, 22 For these reasons, the use of this pain to identify the presence or absence of pain, basing on the assessment tool, translated and validated, could be very NCCPC-R questionnaire score, correlated with the NRS useful in daily clinical practice for pain management of values. Moreover, there was an intrinsic limitation in such these non-communicating patients. We can conclude that measurements since the painful subjective experience of the Italian version of NCCPC-R scale might be used to ass- non-communicating children was evaluated by the care- es pain in non-communicating patients with mental and ID, givers. showing a good correlation when compared to the NRS. Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 93

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, MURGIA ITALIAN VERSION OF THE NON-COMMUNICATING CHILDREN’S PAIN CHECKLIST cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. References properties of the non-communicating children’s pain checklist-revised. Pain 2002;99:349–57. 1.  Linhares MB, Doca FN, Martinez FE, Carlotti AP, Cassiano RG, Pfeif- 11.  Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater er LI, et al. Pediatric pain: prevalence, assessment, and management in a reliability. Psychol Bull 1979;86:420–8. teaching hospital. Braz J Med Biol Res 2012;45:1287–94. 12.  Reynaldo J, Santos A. Cronbach’s Alpha: A Tool for Assessing the 2.  Taylor EM, Boyer K, Campbell FA. Pain in hospitalized children: a Reliability of Scales. J Ext 1999;37:1–4. prospective cross-sectional survey of pain prevalence, intensity, assess- 13.  Sparrow SS, Cicchetti DV, Balla AD. Vineland Adaptive Behavior ment and management in a Canadian pediatric teaching hospital. Pain Res Scales-II. Second Edition. Survey Interview Form. Florence: Giunti Edi- Manag 2008;13:25–32. tori; 2016. 3.  Stevens BJ, Abbott LK, Yamada J, Harrison D, Stinson J, Taddio 14.  Sousa VD, Rojjanasrirat W. Translation, adaptation and validation of A, et al.; CIHR Team in Children’s Pain. Epidemiology and manage- instruments or scales for use in cross-cultural health care research: a clear ment of painful procedures in children in Canadian hospitals. CMAJ and user-friendly guideline. J Eval Clin Pract 2011;17:268–74. 2011;183:E403–10. 15.  Rousson V, Gasser T, Seifert B. Assessing intrarater, interrater and test- 4.  Massaro M, Pastore S, Ventura A, Barbi E. Pain in cognitively impaired retest reliability of continuous measurements. Stat Med 2002;21:3431–46. children: a focus for general pediatricians. Eur J Pediatr 2013;172:9–14. 16.  Walter SD, Eliasziw M, Donner A. Sample size and optimal designs 5.  Zanchi C, Massaro M, Ferrara G, Montico M, D’Osualdo F, Ruti- for reliability studies. Stat Med 1998;17:101–10. gliano R, et al. Validation of the Italian version of the Non-Communi- 17.  Crosta QR, Ward TM, Walker AJ, Peters LM. A review of pain mea- cating Children’s Pain Checklist-Postoperative Version. Ital J Pediatr sures for hospitalized children with cognitive impairment. J Spec Pediatr 2017;43:75. Nurs 2014;19:109–18. 6.  Breau LM, Finley GA, McGrath PJ, Camfield CS. Validation of the 18.  Massaro M, Ronfani L, Ferrara G, Badina L, Giorgi R, D’Osualdo Non-communicating Children’s Pain Checklist-Postoperative Version. F, et al. A comparison of three scales for measuring pain in children with Anesthesiology 2002;96:528–35. cognitive impairment. Acta Paediatr 2014;103:e495–500. 7.  Zabalia M, Breau LM, Wood C, Lévêque C, Hennequin M, Ville- 19.  Hunt A, Mastroyannopoulou K, Goldman A, Seers K. Not knowing— neuve E, et al. [Validation of the French version of the non-communi- the problem of pain in children with severe neurological impairment. Int J cating children’s pain checklist - postoperative version]. Can J Anaesth Nurs Stud 2003;40:171–83. 2011;58:1016–23. 20.  Barbi E, Badina L, Massaro M, Pisciotta G, Ventura A. Paediatricians’ 8.  Johansson M, Carlberg EB, Jylli L. Validity and reliability of a Swed- attitudes toward pain in children with severe cognitive impairment. A re- ish version of the Non-Communicating Children’s Pain Checklist—post- gional survey. Pediatr Med Chir 2014;36:94. operative Version. Acta Paediatr 2010;99:929–33. 21.  Krauss MW, Gulley S, Sciegaj M, Wells N. Access to specialty medi- 9.  Breau LM, McGrath PJ, Camfield C, Rosmus C, Finley GA. Prelimi- cal care for children with mental retardation, autism, and other special nary validation of an observational pain checklist for persons with cogni- health care needs. Ment Retard 2003;41:329–39. tive impairments and inability to communicate verbally. Dev Med Child 22.  Malviya S, Voepel-Lewis T, Tait AR, Merkel S, Lauer A, Munro H, et Neurol 2000;42:609–16. al. Pain management in children with and without cognitive impairment 10.  Breau LM, McGrath PJ, Camfield CS, Finley GA. Psychometric following spine fusion surgery. Paediatr Anaesth 2001;11:453–8. Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript. Article first published online: August 29, 2018. - Manuscript accepted: August 27, 2018. - Manuscript revised: July 11, 2018. - Manuscript received: April 11, 2018. 94 European Journal of Physical and Rehabilitation Medicine February 2019

COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically © 2018 EDIZIONI MINERVA MEDICA European Journal of Physical and Rehabilitation Medicine 2019 February;55(1):95-102 or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access Online version at http://www.minervamedica.it DOI: 10.23736/S1973-9087.18.05108-0 to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. ORIGINAL ARTICLE The effect of different foot orthoses on pain and health related quality of life in painful flexible flat foot: a randomized controlled trial Yasin YURT 1 *, Gül ŞENER 2, Yavuz YAKUT 3 1Department of Physiotherapy and Rehabilitation, Faculty of Health Sciences, Eastern Mediterranean University, Mersin, Turkey; 2Department of Physiotherapy and Rehabilitation, Faculty of Health Sciences, Hacettepe University, Ankara, Turkey; 3Institute of Health Sciences, Hasan Kalyoncu University, Gaziantep, Turkey *Corresponding author: Yasin Yurt, Eastern Mediterranean University, Faculty of Health Sciences, Department of Physiotherapy and Rehabilitation, Mersin, Turkey, 99628. E-mail: [email protected] ABSTRACT BACKGROUND: Foot orthoses are widely used in conservative treatment of painful flexible flatfoot (PFFF) however research is limited to choose the best option in orthotic treatment. AIM: We aimed to compare the effects of computer-aided design/computer-aided manufacturing (CAD-CAM) and conventional insole types in comparison with sham insole, on pain and health related quality of life in patients with PFFF. DESIGN: Randomized controlled trial. SETTING: Outpatient rehabilitation clinic. POPULATION: Sixty-seven people with PFFF, aged between 18 and 45 years, were assigned to CAD-CAM (N.=22), conventional (N.=22) or sham (N.=23) groups. METHODS: In addition to insoles, a home-based exercise program was followed by all participants for eight weeks. Foot pain intensity and quality of life were assessed at the initial evaluation and at the end of two-month follow-up. RESULTS: Pain intensity (Mean±SD, mm on VAS) was significantly lower in CAD-CAM (27.84±18.41) and conventional (27.05±16.82) insole groups than sham group (46.39±20.18) after two months (P<0.05), but there was no difference between conventional and CAD-CAM (P>0.05). All groups had significantly higher physical health scores at the second assessment (P<0.05), but there was no intergroup difference (P>0.05). CONCLUSIONS: CAD-CAM and conventionally designed insoles are both more effective than having sham insole in alleviating pain in PFFF. CLINICAL REHABILITATION IMPACT: CAD-CAM and conventionally designed semicustom insoles in conjunction with a home-based exercise program are both effective in controlling pain compared with sham insole and exercise in PFFF. Clinicians can prescribe both types of semicustom insoles as a part of conservative treatment instead of each other. (Cite this article as: Yurt Y, Şener G, Yakut Y. The effect of different foot orthoses on pain and health related quality of life in painful flexible flat foot: a randomized controlled trial. Eur J Phys Rehabil Med 2019;55:95-102. DOI: 10.23736/S1973-9087.18.05108-0) Key words: Computer-aided design - Flatfoot - Foot orthoses - Pain - Quality of life. Painful flexible flatfoot (PFFF) is a common orthope- Different types of insoles are frequently used as a dic problem that may results in disability,1, 2 and its part of conservative treatment in PFFF.9 Their types prevalence can be up to 20% in some populations.3 People vary significantly but can be broadly classified as pre- with flexible flatfoot deformity more likely to develop ten- fabricated, semicustom or custom-made. Prefabricated dinopathy of the tibialis posterior or Achilles, plantar fas- insoles are widely available and made of different ma- ciitis, medial tibial stress syndrome, patellofemoral pain terials like silicone or thermoplastic in different sizes syndrome or even lower back pain as a result of increased usually based on length of foot. Custom-made insole loading on soft tissues.4-8 production involves capturing an impression of the foot, Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 95

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, YURT FOOT ORTHOSES AND PAIN IN PAINFUL FLEXIBLE FLAT FOOT cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. which can be performed manually with plaster casting procedures were approved by the Clinical Research Eth- or foam impression or digitally with a 3D scanner.10-12 ics Committee of Hacettepe University at 08th of January Another cost-effective method, which is called semicus- in 2014, with GO 14/18-31 decision number and written tom, is usually performed simply by taking some met- informed consent was obtained from all participants. The ric measurements of the foot or adapting a prefabricated trial was registered with NCT02706327 registration num- insole to the patients feet with some modifications.13, 14 ber to ClinicalTrials.gov. Although capturing 3D shape of the foot provides a more accurate model for insole, Zifchock and Davis15 found Eligible participants were randomly assigned into three similar outcomes in terms of foot kinematics for custom parallel groups using simple randomization procedures. A and semicustom insoles. Different types of insoles are computer-generated list of random numbers was used for frequently prescribed but it is still not clear which one allocation with 1:1:1 ratio and each number were placed is better in alleviating symptoms. Whilst some studies into an opaque, unmarked envelope. All these procedures have shown positive effects of custom-made insoles on including enrollment of participants after envelope selec- pain and function in PFFF,16, 17 there is no worldwide tion, were performed by an independent researcher. After consensus on material and design used in production of each group reached at least 10 participants (convention- insoles.18-20 al-15, CAD-CAM-13, sham-10), we did a preliminary analysis and new participants were stratified according to Over the last decade, the computer-aided design/com- pain intensity (lower and higher 50 mm) to minimize inter- puter-aided manufacturing (CAD-CAM) method, integrat- group baseline difference, because mean of conventional ed with a pedobarograph or a scanner, has been commonly group was going notably higher than other groups. New used in insole production, but there is no research whether generated random numbers were used and participants it is more effective than the other methods.21-23 Our aim have selected envelopes just after they rated foot pain in- was to compare the effects of CAD-CAM and conven- tensity. tionally designed insoles and providing control condition of a flat sham insole on pain and health related quality Outcome measures of life in PFFF. Because of the well-known positive ef- fects of stretching and strengthening exercises in PFFF, Foot pain intensity was assessed using the 100 mm vi- we decided to give a home based exercise program to all sual Analog Scale (VAS) as the primary outcome. It participants.16, 17 There were two hypotheses: 1) given in is a commonly used valid tool for patients with foot conjunction with home based exercises; CAD-CAM or pain.25 Participants were asked to rate the maximum conventional insoles are more effective than sham insole level of foot pain on VAS after they were informed that in reducing pain and improving health related quality of 0 means no pain and 100 means unbearable pain they life in PFFF; 2) CAD-CAM insole would have better out- experienced in the last week. In addition, the Foot Func- comes than conventional insole. tion Index was used to measure pain intensity and to assess difficulties participating in different activities.26 Materials and methods The Short Form-36 was used for health-related qual- ity of life assessment.27 We analyzed the physical and Subjects mental health sections of it separately to assess the main effect. We also investigated the physical activity level This study was conducted between January 2014 and Jan- of participants because foot pain is strongly associated uary 2016. Patients who presented to an outpatient clin- with activity and changes in activity could mask our ic with foot pain of at least one month’s duration due to treatment effect. The International Physical Activity flexible flatfoot deformity were assessed. The eligibility Questionnaire-Short Form was used to determine each criteria included being aged between 18 and 45 years, a individual’s activity level with the exception of specific minimum of + 6 points on the Foot Posture Index,17 a min- foot exercises that we advised to do.28 All of these mea- imum 5 degrees standing tibiocalcaneal angle (measured surements were performed by the five years experienced with goniometer),24 no treatment of the foot for at least physiotherapist, at the initial evaluation and again after six months, no leg length discrepancy of more than 1 cm, two months of insole use. Participants were also queried no history of lower extremity surgery and no disease that about how many days a week they had used the insoles could affect lower extremity biomechanics. The research and were asked to rate their insole satisfaction using a visual analog scale. 96 European Journal of Physical and Rehabilitation Medicine February 2019

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, FOOT ORTHOSES AND PAIN IN PAINFUL FLEXIBLE FLAT FOOT YURT cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. Figure 1.—CAD-CAM (left), conventional (middle) and sham (right) the midfoot for medial longitudinal arch, the middle two- insoles. thirds of the forefoot for metatarsal pad, and two-thirds of the heel for medial heel wedge were used as the standard Insole production padding design. They were shaped with using a grinder by the same orthotist. The height of medial longitudinal arch All insoles (Figure 1) were made by same orthotist who pad was lessened laterally and anteroposteriorly beginning had six years’ experience about insole production, in or- from the navicular line to fill the arch. Metatarsal pad was der to eliminate personal differences. For CAD-CAM in- lessened in the posterior and mediolateral directions keep- sole design, participants walked on a five-meter platform ing the anterior side more prominent. Medial heel wedge equipped with a pedobarograph (Medilogic, platform ba- was shaped as a 60 rearfoot post as usual.31 After rectifi- sic, Germany). The most natural gait without any pause cation process of pads, they were located and glued on a or distinct asymmetric stepping and with clear pedobaro- 1-mm polypropylene sheet. The height and material fea- graphic measurement was selected by the orthotist after tures of the padding and covering were the same for both three repetitions. Measurement was recorded on computer the conventional and CAD-CAM designs (Figure 1). by the software and then we designed a corrective foot in- sole with the software using this foot pressure data. We The 3 mm, 15 Shore A hardness ethyl vinyl acetate that added 8 to 12 mm of height to the medial longitudinal arch was used to cover the main insoles, was also used as a pad, 4 to 6 mm metatarsal pad and 6° medial heel wedge flat sham insole for the third group (Figure 1). The insoles into the design for all participants, similarly to other pre- were implemented in a pair of standard walking-type, lace vious studies in literature.29, 30 Pads were located using up, sports shoes after removal of the main insoles from the pedobarographic pressure data. Completed design of the shoes and the participants were asked to wear the in- insole opened with software of the integrated computer sole-equipped shoes frequently, especially during active numerical control machine (PedCad, Germany) for pro- outdoor activities. The shoes were patients own and not duction; 35 Shore A hardness ethyl vinyl acetate was used completely standardized across subjects within this study. for the main insole, and 3 mm, 15 Shore A hardness ethyl The appropriateness and fitting of shoes were assessed by vinyl acetate was used for covering (Figure 1). the physiotherapist. For the conventional insole, the plantar surfaces of each Home-based exercise patient’s metatarsophalangeal joints were marked with a thick broad marker, and the participants were asked to stand A home-based written exercise program which contains on a clean paper. The borders of the foot were then drawn, gastrocnemius stretching, tibialis posterior and intrinsic and the medial longitudinal arch length was marked from muscles strengthening given to all participants as mostly the anterior aspect of the heel to the first metatarsopha- advised for flatfooted people.32 The gastrocnemius was langeal joint. We drew an insole design on paper accord- stretched in a standing position by bending forward toward ing to all of these measurements. The medial two-thirds of a wall with one leg in front and holding the position for 20 seconds. Body weight was used as resistance for the tibi- alis posterior muscle strengthening with heel rise. Towel grasping with toes was used for intrinsic muscle strength- ening. These three exercises were given to perform twice a day with 10 repetitions however no monitoring of compli- ance was performed within this study. Statistical analysis Analyses were performed with the Statistical Package for Social Sciences Version 18.0 (IBM, New York, USA) at the P≤0.05 significance level for all comparisons. Normality of the distribution was investigated by skewness-kurtosis, his- togram graphics, normality tests and plots before the statis- tical tests were performed. Group comparisons were tested with One-way analysis of variance when normality was Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 97

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, YURT FOOT ORTHOSES AND PAIN IN PAINFUL FLEXIBLE FLAT FOOT cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. achieved and Kruskal Wallis test when not. A post-hoc test Enrollment Assessed for eligibility (N.=190) was used with Bonferroni correction. Before and after treat- ment comparisons were tested using a paired sample t test Excluded (N.=123) or a Wilcoxon test, and effect sizes (Cohen’s d) were calcu- - Not meeting inclusion criteria lated for significant differences. An effect size between 0.2 and 0.5 was interpreted as a small effect, an effect size from (N.=115) 0.5 to 0.8 was interpreted as a medium effect, and an ef- - D eclined to participate (N.=8) fect size greater than 0.8 was interpreted as a large effect.33 Numeric results were reported as the mean and standard Randomized (N.=67) deviation (X±SD) along with frequencies by sample size and percentage (n/%). With an average expected value in Allocation the control group of 50 mm and a standard deviation of 20 mm on the visual analog scale, to detect a minimum clini- CAD-CAM Conventional Sham cally significant difference of 20 mm,25 each group needed (N.=22) (N.=22) (N.=23) minimum 21 individuals to achieve 80% power with 5% type I error level. An intention to treat analysis was per- Follow-up formed for participants who did not come to the second as- sessment without any reported reason by consuming their - 8-week follow-up with - 8-week follow-up with - 8-week follow-up with pre-treatment results also as their post-treatment results. insole and home based insole and home based home based exercise exercise exercise - Lost to follow-up (N.=2) - L ost to follow-up (N.=1) - Lost to follow-up (N.=1) (we could not contact) (we could not contact) (we could not contact) Results Analysis Eligible participants were recruited between January 2014 Analyzed (N.=22) Analyzed (N.=22) Analyzed (N.=23) and September 2015 and the trial was stopped with the last follow up ended in January 2016. An overview of (intention to treat) (intention to treat) (intention to treat) the study is presented in Figure 2. Age, body mass index, foot posture index and calcaneal valgus angle values were Figure 2.—Flow diagram of the study. similar for all groups (Table I). We found no significant differences in the frequency of insole use or activity lev- There were no between-group differences in terms of els between groups. Additionally, there were no significant the initial assessment of pain intensity, Foot Function In- differences for gender and pain localization distribution dex and health related quality of life (P>0.05, Table II). between the groups. Foot pain was rated as 59.7% on me- dial longitudinal arch and 40.3% on other areas such as the All groups had lower pain intensity, better Foot Func- metatarsal, heel and ankle (Table I). tion Index and physical quality of life scores after two months (P<0.001). Pain was significantly lower in CAD- CAM and conventional groups than sham group at second Table I.—Comparison of group characteristics. Age (years) Right CAD-CAM (N.=22) Conventional (N.=22) Sham (N.=23) P Body mass index (kg/m2) Left X±SD X±SD X±SD Foot Posture Index Right 0.251 Left 21.73±2.89 23.05±5.53 21.09±1.95 0.622 Calcaneal valgus angle 23.03±3.48 24.11±4.15 23.32±3.28 0.085 Female 7.50±1.73 8.59±1.76 8.48±2.04 0.172 IPAQ-SF (kcal/week) Male 7.68±1.75 8.45±1.73 8.57±1.95 0.428 Insole use frequency (days/week) MLA 6.36±1.64 6.68±1.67 6.22±1.67 0.198 Gender N. (%) Others 6.77±2.18 7.73±2.29 6.70±2.03 0.355 3438.81±3543.30 4731.27±5392.45 2824.54±2221.46 0.468 Foot pain localization N. (%) 3.73±1.42 3.97±1.51 4.21±1.33 0.907 13 (19.4) 12 (17.9) 14 (20.9) 0.219 9 (13.4) 10 (14.9) 9 (13.4) 11 (16.4) 12 (17.9) 11 (16.4) 10 (14.9) 17 (25.4) 6 (9.0) IPAQ-SF: International Physical Activity Questionnaire Short Form. 98 European Journal of Physical and Rehabilitation Medicine February 2019

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, FOOT ORTHOSES AND PAIN IN PAINFUL FLEXIBLE FLAT FOOT YURT cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. Table II.—Comparison of pain intensity, Foot Function Index, SF-36 and insole satisfaction between and within groups. CAD-CAM Conventional Sham P (N.=22) (N.=22) (N.=23) 0.960 X±SD P X±SD P X±SD P 0.001* (%95 CI) (ES) (%95 CI) (ES) (%95 CI) (ES) 0.227 0.518 Pain intensity (VAS, mm) BT 59.27±17.26 <0.001 60.32±16.82 <0.001 58.48±17.51 <0.001 0.352 0.547 Foot Function Index (51.62-66.93) (0.660)§ (52.86-67.78) (0.703)§ (50.91-66.05) (0.304)‡ 0.860 0.847 SF-36 Physical health AT 27.84±18.41 27.05±16.82 46.39±20.18 0.003* SF-36 Mental health (19.67-36.01) (18.29-33.21) (37.66-55.12) Insole satisfaction BT 29.95±14.47 0.016* 37.62±17.35 <0.001 30.09±13.34 0.017* (VAS, mm) (23.53-36.36) (0.293)‡ (29.93-45.31) (0.415)‡ (24.32-35.86) (0.130) Post-hoc analyses (Bonferroni correction) AT 21.81±11.94 24.11±11.70 26.50±13.91 Pain intensity P (16.51-27.11) (18.92-29.30) (20.48-32.51) Insole satisfaction P BT 44.76±7.24 0.001† 43.49±6.85 <0.001 45.60±7.16 0.015* (41.46-48.06) (0.387)‡ (40.37-46.60) (0.442)‡ (42.51-48.70) (0.134) AT 50.14±5.44 50.17±6.70 47.55±7.21 (47.72-52.55) (47.12-53.22) (44.43-50.67) BT 46.82±10.90 0.433 48.91±10.30 0.409 47.91±10.14 0.917 (41.85-51.78) (44.22-53.60) (43.53-52.30) AT 46.13±10.29 47.54±9.05 45.65±9.17 (41.57-50.70) (43.42-51.66) (43.68-51.62) 73.29±15.03 83.43±13.83 63.05±22.77 (66.44-80.13) (77.13-89.73) (52.68-73.42) CAD-CAM and Conventional CAD-CAM & Sham Conventional & Sham 0.690 0.003# 0.001# 0.020 0.182 0.002# CI: confidence interval, SF-36: Short Form 36, ES: Effect size; BT: before treatment, AT: after treatment, §medium effect, ‡small effect. *P<0.05, †P<0.01, #P<0.016. assessment (P<0.01), but there was no difference between insole had no superiority in comparison to conventional CAD-CAM and conventional (P>0.016, Table II). Both insole in terms of alleviating pain and so the second hy- CAD-CAM and conventional insoles had medium effect pothesis was rejected. sizes while sham insole had small effect size on pain re- duction (Table II). No significant difference was found for Whilst there are other studies that have investigated the Foot Function Index and physical quality of life between effect of foot orthoses on foot pain and function for dif- three groups after two months (P>0.05, Table II). Howev- ferent conditions, this study is the first to compare differ- er, the effect sizes on these two outcomes were both small ent types of insoles for PFFF. Pfeffer et al.34 investigated in CAD-CAM and conventional groups but better than the effect of different prefabricated and custom-made in- sham group (Table II). There was no significant satisfac- soles in conjunction with home-based stretching exercises tion difference between the CAD-CAM and conventional in patients with proximal plantar fasciitis. They obtained insole (P>0.016). Deteriorating of foot pain, measured better results for prefabricated insoles and stretching in by using VAS, was defined as a potential adverse event comparison to custom made after two months of use. On of treatment and none of the participants were reported the other hand, in two other studies; Wrobel et al.35 and worsened pain. Martin et al.36 showed positive and similar clinical out- comes for custom and prefabricated insoles in the treat- Discussion ment of plantar fasciitis. Contrary to our flatfooted sample group, participants had a specific pathology not deformity Using CAD-CAM or conventional insoles in conjunction in these three studies; however, patients with same pathol- with home based exercise were more successful than sham ogy may have different foot deformities and may respond insole and exercise in reducing pain, however none of in a different manner to orthotic treatment. Also, Wrobel them improved health related quality of life in PFFF. This et al.35 found no difference in Foot Function Index and partly confirmed our first hypothesis. The CAD-CAM SF-36 scores between custom, prefabricated and sham insoles after three months of treatment. They also gave a Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 99

This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, YURT FOOT ORTHOSES AND PAIN IN PAINFUL FLEXIBLE FLAT FOOT cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher. home-based self-care program in addition to insoles for ever, the sham insole got pretty high satisfaction score. all participants as we did in our study. Similarly, we did It is not surprising because McCormick et al38 found that not find a significant difference between groups in health- flat sham insole, same of which we used in our study, de- related quality of life and Foot Function Index, all groups creases peak plantar pressure under the heel. Shock ab- had better scores after two months but the calculated effect sorbing feature of the sham insole might be the reason of sizes were quite small. Positive changes of sham group in decreased pain in some cases. It could also be a result of both studies may be the results of home-based exercises or sports shoes worn more frequently than before. Prescribed sham insole or both of them. The main problem of these insoles are usually worn with a pair of athletic shoes in studies is the unknown compliance of patients to the ex- most of studies, but researchers usually concentrate on the ercises. In addition, Wrobel et al. also showed a greater effect of insoles not shoes. In a different study; Esterman improvement in spontaneous physical activity in custom and Pilotto39 researched the effect of a prefabricated in- group and suggested that traditional outcome measures sole in flat-footed air force recruits with lower extremity might not be sensitive or specific enough to detect main musculoskeletal pain and did not observe any significant effects of treatment. We did not investigate possible effects change. This study required participants to use orthoses of insoles on spontaneous physical activity level, because with their military boots but 50% of participants failed to in our opinion pain free life may be an important factor but use insoles. Their results may have been influenced by the could not be the only reason of increased physical activity. type of shoes used. This means insoles do not affect alone but with the shoes in which they are inserted. Zammit and Payne37 found approximately 20% im- provement in foot pain and foot function in symptomatic Most of patients came for the second assessment while flatfoot after four weeks follow-up with insole. Our foot some of them did not. They did not answer our call, but function changes were based on a different outcome mea- we did intention to treat analysis as recommended to do sure, however, the results were similar. Furthermore, they in randomized controlled trials and we think it is a strong used different insole types (18 custom made, 4 prefabri- part of the study. We do not know the actual reason of their cated) for participants and did not provide detailed infor- leave, but it might be having considerably less complaints mation about the insoles. Effect sizes of insoles were small than the other subjects. in both studies; this could be the limited effect of insoles on activity and participation in PFFF. It is the first study that has compared a novel method to a conventional for insole production in PFFF. Differ- According to results of above-mentioned studies, in- ent CAD-CAM systems with different types and costs are soles have positive clinical outcomes for relief of foot increasingly used for prosthetic and orthotic production pain but there are still some conflicting results whether but other conventional methods like we used in our study it should be prefabricated, semicustom or custom made, are still being used especially in developing countries. Our probably because of different sample groups of the stud- purpose was to investigate if there was an additional ef- ies. In our study we did not investigate this issue indeed. fect of this technology on treatment but that is not true Both of insoles had similar padding designs on 2D mea- to discuss the pros and cons of CAD-CAM insoles only sures of foot, therefore they both seem as semicustom in- in terms of treatment outcomes. Also, it should be noted sole. But using plantar pressure record in design, certain that CAD-CAM is an umbrella term of the methods used measures of pads and accuracy of milling makes CAD- in manufacturing with the aid of computer, our research CAM different than the conventional method. Therefore, was only about a small part of it. As we did not have the it was the study of comparison two different manufac- 3D measures of foot, we used similar pad heights in both turing techniques and results of our study should be dis- insoles. Therefore, this study has compared the manual cussed being aware of it. and computer aided methods in terms of design and shap- ing (milling) processes indeed. Although we have found An interesting result of this study was the significant no clinical difference between two methods, CAD-CAM improvement in sham group. Basic exercises were given is more likely to be used in this area because of its fast- to the sham group and also, they used a flat insole with er, more accurate and simple design and manufacturing sports shoes. However, it is hard to say that the positive features. In future, cost analysis of CAD-CAM systems changes of sham group resulted from the exercises or flat should be investigated by researchers. This could be help- insole. Because we could not monitor the compliance ful for companies to give a decision about their necessity to home exercises and most of participants were aware in production. it was not an insole that might support their feet; how- 100 European Journal of Physical and Rehabilitation Medicine February 2019