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, 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. Limitations of the studies 13.  Johanson MA, Donatelli R, Wooden MJ, Andrew PD, Cummings GS. Effects of three different posting methods on controlling abnormal subta- There are some issues which could limit our findings. lar pronation. Phys Ther 1994;74:149–58, discussion 158–61. Participants of the study had no specific pathology but 14.  Vicenzino B. Foot orthotics in the treatment of lower limb conditions: flexible flatfoot deformity, the effect of insoles may dif- a musculoskeletal physiotherapy perspective. Man Ther 2004;9:185–96. fer in different pathologies. Also, we did not monitor the 15.  Zifchock RA, Davis I. A comparison of semi-custom and custom foot compliance to home based exercise program so we do not orthotic devices in high- and low-arched individuals during walking. Clin know whether all of them did exercises like we advised. Biomech (Bristol, Avon) 2008;23:1287–93. The young average age of our study population could limit 16.  Nielsen MD, Dodson EE, Shadrick DL, Catanzariti AR, Mendicino the generalizability of our results and also the two-month RW, Malay DS. Nonoperative care for the treatment of adult-acquired flat- follow-up interval was sufficient to see the benefit of in- foot deformity. J Foot Ankle Surg 2011;50:311–4. soles but may not be long enough to determine whether 17.  Alvarez RG, Marini A, Schmitt C, Saltzman CL. Stage I and II the benefit will last. posterior tibial tendon dysfunction treated by a structured nonoperative management protocol: an orthosis and exercise program. Foot Ankle Int Conclusions 2006;27:2–8. 18.  Banwell HA, Mackintosh S, Thewlis D. Foot orthoses for adults with CAD-CAM and conventionally designed insoles in con- flexible pes planus: a systematic review. J Foot Ankle Res 2014;7:23. junction with a home-based exercise program are both 19.  Healy A, Dunning DN, Chockalingam N. Effect of insole material more effective in controlling pain compared with sham on lower limb kinematics and plantar pressures during treadmill walking. insole and exercise in flexible flatfoot. Clinicians can pre- Prosthet Orthot Int 2012;36:53–62. scribe both types of semicustom insoles as a part of con- 20.  Tong JW, Ng EY. Preliminary investigation on the reduction of servative treatment in PFFF, instead of each other. plantar loading pressure with different insole materials (SRP—Slow Recovery Poron, P—Poron, PPF—Poron +Plastazote, firm and PPS— References Poron+Plastazote, soft). Foot 2010;20:1–6. 21.  Ki SW, Leung AK, Li AN. Comparison of plantar pressure distribu- 1.  Michelson JD, Durant DM, McFarland E. The injury risk associated tion patterns between foot orthoses provided by the CAD-CAM and foam with pes planus in athletes. Foot Ankle Int 2002;23:629–33. impression methods. Prosthet Orthot Int 2008;32:356–62. 2.  Thomas Haendlmayer K, John Harris N. Flatfoot deformity: an over- 22.  Crabtree P, Dhokia VG, Newman ST, Ansell MP. Manufacturing view. Orthop Trauma 2009;23:395–403. methodology for personalised symptom-specific sports insoles. Robot 3.  Staheli LT, Chew DE, Corbett M. The longitudinal arch. A survey of Comput-Integr Manuf 2009;25:972–9. eight hundred and eighty-two feet in normal children and adults. J Bone 23.  Ciobanu O. [The use of CAD/CAM and rapid fabrication technolo- Joint Surg Am 1987;69:426–8. gies in prosthesis and orthotics manufacturing]. Rev Med Chir Soc Med 4.  Beeson P. Posterior tibial tendinopathy: what are the risk factors? J Am Nat Iasi 2012;116:642–8. Podiatr Med Assoc 2014;104:455–67. 24.  Shih YF, Wen YK, Chen WY. Application of wedged foot orthosis ef- 5.  Beeson P. Plantar fasciopathy: revisiting the risk factors. Foot Ankle fectively reduces pain in runners with pronated foot: a randomized clinical Surg 2014;20:160–5. study. Clin Rehabil 2011;25:913–23. 6.  Lakstein D, Fridman T, Ziv YB, Kosashvili Y. Prevalence of ante- 25.  Karl B, Landorf JA. Radford, Minimal important difference: Values rior knee pain and pes planus in Israel defense force recruits. Mil Med for the Foot Health Status Questionnaire, Foot Function Index and Visual 2010;175:855–7. Analogue Scale. Foot 2008;18:15–9. 7.  Newman P, Witchalls J, Waddington G, Adams R. Risk factors associ- 26.  Budiman-Mak E, Conrad KJ, Roach KE. The Foot Function Index: ated with medial tibial stress syndrome in runners: a systematic review a measure of foot pain and disability. J Clin Epidemiol 1991;44:561–70. and meta-analysis. Open Access J Sports Med 2013;4:229–41. 27.  Kocyigit H, Aydemir O, Olmez N, Memis A. Reliability and validity 8.  Kaufman KR, Brodine SK, Shaffer RA, Johnson CW, Cullison TR. of the Turkish version of Short-Form-36 (SF-36). Turkish J Drugs Therap The effect of foot structure and range of motion on musculoskeletal over- 1999;12:102–6. use injuries. Am J Sports Med 1999;27:585–93. 28.  Saglam M, Arikan H, Savci S, Inal-Ince D, Bosnak-Guclu M, Kara- 9.  Marzano R. Nonoperative management of adult flatfoot deformities. bulut E, et al. International physical activity questionnaire: reliability and Clin Podiatr Med Surg 2014;31:337–47. validity of the Turkish version. Percept Mot Skills 2010;111:278–84. 10.  MacLean C, Davis IM, Hamill J. Influence of a custom foot orthotic 29.  Mündermann A, Nigg BM, Humble RN, Stefanyshyn DJ. Foot or- intervention on lower extremity dynamics in healthy runners. Clin Bio- thotics affect lower extremity kinematics and kinetics during running. mech (Bristol, Avon) 2006;21:623–30. Clin Biomech (Bristol, Avon) 2003;18:254–62. 11.  Castro-Méndez A, Munuera PV, Albornoz-Cabello M. The short-term 30.  de Morais Barbosa C, Barros Bértolo M, Marques Neto JF, Bellini effect of custom-made foot orthoses in subjects with excessive foot pro- Coimbra I, Davitt M, de Paiva Magalhães E. The effect of foot orthoses nation and lower back pain: a randomized, double-blinded, clinical trial. on balance, foot pain and disability in elderly women with osteoporosis: a Prosthet Orthot Int 2013;37:384–90. randomized clinical trial. Rheumatology (Oxford) 2013;52:515–22. 12.  Salles AS, Gyi DE. The specification of personalised insoles using 31.  Redmond A, Lumb PS, Landorf K. Effect of cast and noncast foot additive manufacturing. Work 2012;41(Suppl 1):1771–4. orthoses on plantar pressure and force during normal gait. J Am Podiatr Med Assoc 2000;90:441–9. 32.  Deland JT. Adult-acquired flatfoot deformity. J Am Acad Orthop Surg 2008;16:399–406. 33.  Kotrlik JW, Williams HA. The incorporation of effect size in informa- tion technology, learning and performance research. Info Tech Learn Perf J. 2003;21:1–7. 34.  Pfeffer G, Bacchetti P, Deland J, Lewis A, Anderson R, Davis W, et Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 101

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. al. Comparison of custom and prefabricated orthoses in the initial treat- comes of foot orthotic treatment and changes in rearfoot kinematics. J Am ment of proximal plantar fasciitis. Foot Ankle Int 1999;20:214–21. Podiatr Med Assoc 2007;97:207–12. 35.  Wrobel JS, Fleischer AE, Crews RT, Jarrett B, Najafi B. A random- ized controlled trial of custom foot orthoses for the treatment of plantar 38.  McCormick CJ, Bonanno DR, Landorf KB. The effect of custom- heel pain. J Am Podiatr Med Assoc 2015;105:281–94. ised and sham foot orthoses on plantar pressures. J Foot Ankle Res 36.  Martin JE, Hosch JC, Goforth WP, Murff RT, Lynch DM, Odom RD. 2013;6:19. Mechanical treatment of plantar fasciitis. A prospective study. J Am Podi- atr Med Assoc 2001;91:55–62. 39.  Esterman A, Pilotto L. Foot shape and its effect on functioning in 37.  Zammit GV, Payne CB. Relationship between positive clinical out- Royal Australian Air Force recruits. Part 2: Pilot, randomized, controlled trial of orthotics in recruits with flat feet. Mil Med 2005;170:629–33. Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript. Acknowledgements.—The authors acknowledge the valuable support of Mehtap Malkoç, the Department Chair of Eastern Mediterranean University, Faculty of Health Sciences, Department of Physiotherapy and Rehabilitation. Article first published online: March 16, 2018. - Manuscript accepted: March 15, 2018. - Manuscript revised: February 28, 2018. - Manuscript received: December 9, 2017. 102 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):103-12 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/S1973-9087.18.05093-1 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 Cardiorespiratory adaptation during 6-Minute Walk Test in fibrotic idiopathic interstitial pneumonia patients who did or did not respond to pulmonary rehabilitation Baptiste CHÉHÈRE 1, Valérie BOUGAULT 1 *, Cécile CHENIVESSE 2, Jean-Marie GROSBOIS 3, 4, Benoit WALLAERT 2, 3 1EA 7369 - URePSSS - Multidisciplinary Research Unit in Sport Health Society, University of Lille, Lille, France; 2Competence Center for Rare Pulmonary Diseases, Department of Immuno-Allergology and Respiratory Diseases, Lille University Hospital, University of Lille, Lille, France; 3Department of Respiratory Medicine, Germon and Gauthier Hospital, Béthune, France; 4FormAction Santé, Pérenchies, France *Corresponding author: Valérie Bougault, URePSSS - Multidisciplinary Research Unit in Sport Health Society, EURASPORT, 413 Avenue Eugène Avinée, 59120 Loos, France. E-mail: [email protected] ABSTRACT BACKGROUND: Pulmonary rehabilitation (PR) improves performance in the 6-min walk test (6MWT) in a subset of patients with fibrotic idiopathic interstitial pneumonia (f-IIP); however, a large proportion of patients does not respond to PR. AIM: To investigate the effects of a PR program on cardiorespiratory responses during a 6MWT and to identify the characteristics of patients who do not show improved performance after PR. DESIGN: An observational study. SETTING: Patients were recruited from the Competence Centre for Rare Pulmonary Diseases at Lille University Hospital, France and com- pleted an 8-week home-based PR program. POPULATION: A total of 19 patients with f-IIP; 12 with idiopathic pulmonary fibrosis (IPF) and 7 with fibrotic non-specific interstitial pneumonia. METHODS: Patients underwent spirometry and completed a 6MWT before and after an 8-week PR program. Gas exchange, heart rate, and pulse O2 saturation were measured continuously during the 6MWT. Quality of life, dyspnea, and anxiety/depression were assessed using the Short- Form 36 (SF-36), the baseline/transition dyspnea index (BDI/TDI), and the Hospital Anxiety and Depression Scale (HADS) questionnaires. RESULTS: Patients who did and did not improve the distance walked in the 6MWT by at least 30 m after PR were classified as responders (N.=9) and non-responders (N.=10), respectively. O2 uptake, ventilation rate, and distance covered during the 6MWT were significantly im- proved only in the responder group (P<0.05). Changes in SF-36, BDI/TDI, and HADS scores did not differ significantly between responders and non-responders. The non-responder group contained significantly more patients with IPF (P<0.05) and experienced greater arterial oxygen desaturation during the 6MWT compared with the responder group. CONCLUSIONS: Failure to improve performance in the 6MWT after PR was associated with a diagnosis of IPF, non-improvement in gas ex- change, and greater arterial oxygen desaturation. CLINICAL REHABILITATION IMPACT: Most f-IIP patients who did not respond to PR were diagnosed with IPF and displayed greater hypox- emia during exercise. Clinical practitioners should seek to determine why patients fail to improve exercise performance after PR and propose an alternative exercise regimen to these patients. (Cite this article as: Chéhère B, Bougault V, Chenivesse C, Grosbois JM, Wallaert B. Cardiorespiratory adaptation during 6-Minute Walk Test in fibrotic idiopathic interstitial pneumonia patients who did or did not respond to pulmonary rehabilitation. Eur J Phys Rehabil Med 2019;55:103-12. DOI: 10.23736/S1973-9087.18.05093-1) Key words: Exercise therapy - Cardiorespiratory fitness - Physiological adaptation - Walk test - Interstitial lung diseases - Idiopathic pulmonary fibrosis. Interstitial lung diseases (ILD) include various disorders brosis, decreased pulmonary capacity, and impaired gas characterized by alveolar and interstitial space dam- exchange.1 Patients generally present with exercise intol- age, pulmonary inflammation usually associated with fi- erance, dyspnea on exertion, and poor quality of life,2-4 Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 103

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, CHÉHÈRE CARDIORESPIRATORY ADAPTATION IN INTERSTITIAL PNEUMONIA 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 improved by pulmonary rehabilitation (PR) Materials and methods programs performed at home5, 6 or in outpatient or inpa- tient settings.7-11 The study was conducted in accordance with the Decla- ration of Helsinki. Approval for the use of patient data ILD patients with fibrotic idiopathic interstitial pneumo- was obtained from the Institutional Review Board of the nia (f-IIP) generally have a poorer exercise tolerance and French Learned Society for Pulmonology (CEPRO 2011- survival rate compared with non-fibrotic patients.7, 12 In 036). Written informed consent was obtained from each particular, the distance walked during the 6-Min Walk Test participant. (6MWT) is independently associated with mortality in these patients.13, 14 Bajwah et al. reviewed the effects of pharma- Patients with a diagnosis of f-IIP according to estab- cological and non-pharmacological interventions on chang- lished criteria26, 27 and followed at the Competence Centre es in the 6MWT distance (6MWD), and only pirfenidone for Rare Pulmonary Diseases, Lille University Hospital and PR showed strong evidence of a positive impact.15 Al- (Lille, France) were recruited by pulmonologists during though most studies have reported that PR does improve the the patients’ routine monitoring visits. Inclusion criteria 6MWD in f-IIP patients,5, 7-10, 16, 17 one retrospective study of were: 1) a diagnosis of fibrotic non-specific interstitial 599 ILD patients (46% with idiopathic pulmonary fibrosis pneumonia (f-NSIP) or IPF by high-resolution computed [IPF]) reported that 40% showed either no clinical improve- tomography or lung biopsy; 2) resting pulse O2 saturation ment or experienced worsening of the 6MWD after PR.18 ≥88%. Patients were excluded if they had participated in a PR program in the preceding year, were receiving con- Because changes in the 6MWD over time predict surviv- tinuous O2 therapy, had a comorbidity precluding exercise al in f-IIP patients,14 there is an urgent need to understand training or affecting test performance, had a forced vital why some patients fail to improve exercise performance af- capacity (FVC) <50% of the predicted value, or had a dif- ter PR. In these patients, exercise limitations may be due to fusing capacity of the lung for carbon monoxide (DLCO) pulmonary and/or gas exchange impairments (diminished <25% of the predicted value. lung volume, oxygen desaturation), circulatory limitations (pulmonary hypertension, cardiac dysfunction), and/or mus- Enrollment (N.=34) cular dysfunction.4, 19 However, few studies have examined the effects of PR on gas exchange during submaximal ex- Excluded (N.=13) ercise tests, such as the 6MWT, in patients with respiratory - Did not meet inclusion criteria diseases. Chronic obstructive pulmonary disease (COPD) patients generally show increased peak oxygen uptake (VO- (N.=3) 2peak) in the post-PR 6MWT.20-23 In contrast, several stud- - D eclined to participate in PR ies have detected no significant change in VO2peak after PR, despite an increase in 6MWD, f-IIP patients.7, 10, 24, 25 This (N.=10) discrepancy may be due to inter-study differences in the proportion of patients who did or did not show improved Assessed for eligibility (N.=21) 6MWD (i.e., responders and non-responders). For example, patients whose pathology worsened during the PR program Excluded (N.=2) may have shown no improvement in either the 6MWD or - S topped the PR program (N.=1) VO2. Alternatively, patients with greater arterial desatura- - Refused the second evaluation tion during exercise, reflecting more limited ventilatory function may have benefited less from the PR program. (N.=1) Based on these possibilities, we hypothesized that f- Completed the PR program (N.=19) IIP patients who displayed an increase in exercise perfor- mance (i.e., distance walked) during a post-PR 6MWT Responders (N.=9) Non-responders (N.=10) would show a concomitant increase in VO2). The aim of Δ6MWD≥30 m Δ6MWD<30 m the present study was to investigate changes in cardio- respiratory responses during the 6MWT in f-IIP patients Did not complete gaseous exchange analysis (N.=6) after an 8-week home-based PR program, and to identify - Stopped during the 6MWT (N.=l) characteristics that differentiate between responders and - Technical problems during the test (N.=2) non-responders. - D id not tolerate the face mask (N.=2) - Low back pain during the second evaluation (N.=l) Responders with complete Non-responders with complete gaseous exchange datasets (N.=6) gaseous exchange datasets (N.=7) Figure 1.—Flow chart of patient selection and management. ∆6MWD: pre- vs. post-PR change in 6MWT distance; 6MWT: 6-Minute Walk Test; PR: pulmonary rehabilitation. 104 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, CARDIORESPIRATORY ADAPTATION IN INTERSTITIAL PNEUMONIA CHÉHÈRE 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. Thirty-four patients with f-IIP were recruited between gen was provided during the training if pulse O2 saturation September 2014 and June 2016 and offered the home- <88%. Finally, patients were encouraged to increase their based PR program. Ten patients declined to participate and physical activity duration in daily life. Each patient was 3 were excluded; 1 with a DLCO <25% of predicted and 2 asked about his/her adherence to the non-supervised ses- with severe orthopedic or neurological comorbidities. Fi- sions but details of their physical activities and exercise nally, a total of 21 patients (12 with IPF, 7 with f-NSIP), were not recorded. were included in the study. A flow diagram outlining pa- tient recruitment is shown in Figure 1. Therapeutic education program and psychosocial sup- port Study design An educational assessment was performed at each pa- Patients completed an 8-week home-based PR program. tient’s home before starting the PR program to evaluate Once a week, a health professional with specific exper- his/her existing difficulties and to assess changes required tise in exercise training and therapeutic education super- in health-related behavior. Short, medium and long-term vised a 90-min session in the patient’s home. Patients goals were assessed to provide real-life motivation for the were encouraged to perform the same exercises indepen- PR program. Depending on the patient’s needs, the pro- dently, with a total target of five sessions per week. The gram addressed respiratory diseases and comorbidities supervised sessions included exercise training, therapeu- (e.g., diabetes, cardiovascular diseases, obesity, depres- tic patient education, and psychosocial support based on sion, anxiety), treatments, prevention and recognition of an educational needs assessment (see below). Pulmonary exacerbations, physical exercise, outings, sleep, sexuality, function tests and the 6MWT with measurement of gas breathing management, stress management, balanced diet exchange were performed at the Hospital before and after and weight control, smoking cessation, self-image, and the PR program. The mean time between the end of the self-esteem. This section of the program was administered home-based PR program and the post-PR evaluation was at each visit (usually in the presence of the patient’s spouse 8±5 days. or caregiver) in the form of interactive presentations, ques- tion and answer sessions, card games, and illustrated print- Exercise training program ed material. Each session included 30 min endurance training on a cy- Assessments cle ergometer (Domyos VM 200, Decathlon, Villeneuve- D’Ascq, France) and/or a stepper (Stepper Réglable Ath- Baseline lung diffusing capacity and volumes were ob- litech, Groupe Go Sport, Sassenage, France), which were tained by plethysmography. Forced expiratory volume in set up at the patient’s home at the start of the PR program. 1 seconds (FEV1), FVC, total lung capacity (TLC), and Training intensity was determined for each individual with DLCO were measured using a BodyBox 5500 (Medisoft, a specific functional goal. The exercise intensity for each Sorinnes, Belgium). Predicted normal values were derived patient at the start of the program was based on the aver- from standard equations recommended by the European age heart rate (±5 bpm) obtained during the last 3 min of Respiratory Society.29 the pre-PR 6MWT.28 The cycling workload and stepper movement speed were adjusted to target an exercise rate Quality of life was assessed using the Medical Out- at perceived exertion scores between 3 and 4 on the 0-10 comes Study Short-Form 36 (SF-36), which is composed Borg scale or between 11 and 13 on the Borg 6-20 Scale. of eight domains and two component summary scores If patients failed to complete the optimal duration of 30 (physical and mental).30 Dyspnea was assessed using the min continuous exercise, the duration was reduced to 10- baseline and transition dyspnea index (BDI/TDI) and the min periods of interval training, with a goal of gradually 10-point Borg scale at the end of the 6MWT.31, 32 Anxiety increasing to 30 min continuously over the 8-week pro- and depression were evaluated using the Hospital Anxiety gram. Strength training of upper and lower limb muscles and Depression Scale (HADS).33 was performed using body weight, dumbbell, and elastic band exercises, for 15 minutes per session. Exercise was The 6MWT was performed according to the American monitored and modified by an experienced physiotherapist Thoracic Society (2002) recommendations using a 30-m according to a standardized protocol. Supplemental oxy- corridor, without encouragement.34 At rest and during the exercise test, gas exchanges and ventilatory responses were measured continuously using a portable spiroergom- Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 105

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, CHÉHÈRE CARDIORESPIRATORY ADAPTATION IN INTERSTITIAL PNEUMONIA 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. eter (MetaMax 3B, Cortex Biophysik, Lepzig, Germany). viation. Data were analyzed using SigmaStat software v. VO2, carbon dioxide output (VCO2), minute ventilation 3.5. Univariate normality assumptions were verified with (VE), breathing frequency (Bf), tidal volume (VT), respi- the Kolmogorov-Smirnov test, and homogeneity of vari- ratory exchange ratio, respiratory equivalent ratios for O2 ances was verified using the Brown-Forsythe variation (VE/VO2), and carbon dioxide (VE/VCO2) were obtained of Levene’s test. A paired t-test and the nonparametric continuously. Data were recorded every 5 s and the av- Wilcoxon test were used to compare dyspnea, and SpO2 erage value per minute was used for statistical analysis. before and after the 6MWT. Repeated measures ANOVA Pulse O2 saturation (SpO2) was also recorded every minute (time and pre/post) was used to compare the change in car- using a pulse oximeter (Novametrix 513 Pulse Oximeter, diorespiratory parameters at rest and at each minute of the Wallingford, CN, USA), and heart rate (HR) was moni- 6MWT pre- and post-PR. When ANOVA was significant, tored continuously using a belt compatible with the gas Tukey’s post-hoc test for multiple comparisons was ap- exchange analyser (Polar FS3C, Oy, Finland). Resting plied. A paired t-test and the nonparametric Wilcoxon test (SpO2rest) and minimum SpO2 during exercise (SpO2nadir) were used to compare the questionnaire scores (HADS, were recorded, and ∆SpO2 was calculated as (SpO2rest − SF-36 and BDI/TDI) pre- and post-PR. SpO2nadir). Fisher’s test was used to compare the distribution be- Classification of responders and non-responders tween the responder and non-responder groups of f-IIP patients and of patients who achieved the MID of 30 m Patients who did or did not show a 6MWD increase ≥30 m for 6MWT. An unpaired t-test and the nonparametric Wil- after PR (considered the minimal important difference coxon test were used to compare values at rest and changes [MID])35 were classified as responders or non-responders, following PR for the responders and non-responders. Be- respectively. cause the number of subjects was small, the effect size was calculated for P values close to 0.15. Cohen36 previously Statistical analysis suggested that effect sizes of <0.2, 0.2-0.5, 0.5-0.8, and >0.8 reflected trivial, small, moderate, and large effects, Cardiorespiratory parameters (HR, VO2, VCO2, VE, VT, respectively.36 A P value <0.05 was considered statistically Bf, VE/VO2, VE/VCO2) during the 6MWT performed be- significant. fore (pre-PR) and after (post-PR) the PR program were compared. Values are expressed as mean and standard de- Table I.—C haracteristics of the 19 patients with fibrotic idiopathic interstitial pneumonia. Variable All (N.=19) IPF (N.=12) f-NSIP (N.=7) P value Men/women 15/4 10/2 5/2 0.60 65±9 66±7 62±12 0.53 Age, years 30±6 30±3 30±9 0.90 75±13 79±12 70±13 0.13 BMI, kg/m2 78±19 83±20 69±14 0.11 73±12 75±10 71±15 0.44 FVC, % predicted 40±8 39±8 42±10 0.43 15 (79%) 9 (75%) 6 (86%) 1.00 TLC, % predicted 0.18 7±2 7±2 6±2 FEV1, % predicted 0.18 DLCO, % predicted 425±57 438±59 401±49 0.12 85±5 84±6 88±4 Exercise O2 therapy support, N. (%) 0.37 BDI, score 11 (58%) 8 (67%) 3 (43%) 0.17 8 (42%) 7 (58%) 1 (14%) 0.15 6MWT 7 (37%) 6 (50%) 1 (14%) 0.62 6 (32%) 3 (25%) 3 (43%) 1.00 6MWD, m 2 (11%) 1 (8%) 1 (14%) SpO2nadir, % * Comorbidities, N. (%) Arterial hypertension, N. (%) Type 2 diabetes, N. (%) Cardiovascular disease, N. (%) Sleep-disordered breathing, N. (%) Gastroesophageal reflux, N. (%) Values are expressed as the mean±SD, number and percentage of patients, or number of patients. *The effect size for the SpO2nadir was 0.75 (moderate). 6MWD: 6-min walk test distance; 6MWT: 6-min walk test; BDI: baseline dyspnea index; BMI: body mass index; DLCO: diffusing capacity of the lung for carbon monoxide; FEV1: forced expiratory volume in 1s; f-NSIP: fibrotic non-specific interstitial pneumonia; FVC: forced vital capacity; IPF: idiopathic pulmonary fibrosis; SpO2nadir: minimum SpO2 recorded; TLC: total lung capacity. 106 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, CARDIORESPIRATORY ADAPTATION IN INTERSTITIAL PNEUMONIA CHÉHÈRE 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. Results were not significantly different pre- and post-PR (P=0.24, data not shown). Similarly, there was no significant effect of Twenty-one patients with f-IIP were enrolled and deemed the PR program on the HADS anxiety or depression scores eligible for the study, of whom 19 completed the PR pro- (P=0.53 and P=0.87, respectively; Table II). gram (Figure 1). A subset of 13 patients had complete (pre- and post-PR) datasets for the gas exchange analysis. Of Characterization of responders and non-responders the 19 patients, 12 (63%) had IPF and seven (37%) had f-NSIP. The main characteristics and comorbidities of the Of the 19 patients, nine (47%) showed a 6MWD improve- patients are summarized in Table I. ment of at least 30 m (the MID for this test35) and were considered responders; the remaining 10 patients were des- Following the PR program, the mean 6MWD for the en- ignated non-responders. The two groups showed no signif- tire cohort (N.=19) was significantly increased from 425±57 icant differences in the baseline (pre-PR program) 6MWD m to 448±68 m (P=0.01, Table II). Among the eight domains or other characteristics (Table III). However, the propor- of the SF-36 questionnaire, only the physical functioning tion of f-NSIP and IPF patients was significantly different score of quality of life was improved post-PR (P=0.004, between the groups, with the non-responder group contain- Table II). The BDI/TDI score showed an improvement of ing only one of the seven f-NSIP patients but 9 of the 12 dyspnea post-PR (P=0.01, Table II). However, the dyspnea IPF patients (14% vs. 75%; P=0.02). Notably, changes of score on the Borg scale assessed at the end of the 6MWT Table II.—Pre- vs. post-PR changes in dyspnea and quality of life scores in the f-IIP patients. Variable Pre-PR Post-PR P value Effect Size 6MWD, m 425±57 448±68 0.01 0.37 SF-36 scores 54±19 60±18 0.10 0.33 Physical summary score 55±23 63±24 0.004 0.33 Physical functioning score 60±21 66±21 0.14 0.28 Mental summary score 6.8±2.1 +0.9±1.3 0.01 0.39 BDI/TDI score HADS scores 7.6±4.8 6.7±4.1 0.53 Anxiety score 5.2±3.5 5.3±3.0 0.87 Depression score Values are expressed as the mean±SD. 6MWD: 6-Minute Walk Test distance; BDI/TDI: baseline/transition dyspnea index; HADS: hospital anxiety depression scale; PR: physical rehabilitation; SF-36: Short-Form 36. Table III.—Baseline characteristics of the responder and non-responder groups. Variable Responders Non-responders P value Effect size (N.=9) (N.=10) Pulmonary function 73±13 77±12 0.48 0.73 FVC, % predicted 70±12 82±20 0.16 0.69 TLC, % predicted 73±14 74±10 0.77 FEV1, % predicted 44±9 38±7 0.15 0.68 DLCO, % predicted 0.77 6WMT 422±66 428±51 0.83 87±4 83±6 0.14 6MWD, m 6±3 5±2 0.21 SpO2nadir, % 6±2 7±2 0.45 Dyspnea end, Borg 0-10 Scale BDI score 50±21 58±18 0.41 SF-36 score 54±18 65±22 0.27 Physical component score Mental component score 9±4 6±5 0.14 HADS score 7±4 4±5 0.10 Anxiety score Depression score Values are expressed as the mean±SD. 6MWD: 6-Minute Walk Test distance; BDI: baseline dyspnea index; DLCO: diffusing capacity of the lung for carbon monoxide; FEV1: forced expiratory volume in 1 s; FVC: forced vital capacity; HADS: hospital anxiety depression scale; SF-36: Short-Form 36; SpO2nadir: minimum SpO2 recorded; TLC: total lung capacity. Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 107

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, CHÉHÈRE CARDIORESPIRATORY ADAPTATION IN INTERSTITIAL PNEUMONIA 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 IV.—P re- vs. post-PR changes in the responder and non-responder groups. Variable Responders Non-responders P value Effect size (N.=9) (N.=10) ∆FVC, mL 0.31 0.82 ∆TLC, mL 2±119 -55±119 0.08 0.94 ∆FEV1, mL 131±305 -347±711 0.04 ∆DLCO, mL/min/mmHg 26±66 0.47 ∆TDI score 0.10±0.94 -85±131 0.93 ∆SF-36 physical component score -0.09±0.13 0.47 ∆SF-36 mental component score 1±1 0.86 ∆HADS anxiety score 9±17 1±1 0.23 ∆HADS depression score 6±17 4±16 0.41 -2±4 5±15 -1±3 0±3 1±2 Values are expressed as the mean±SD. DLCO: diffusing capacity of the lung for carbon monoxide; FEV1: forced expiratory volume in 1 s; FVC: forced vital capacity; HADS: hospital anxiety depression scale; SF-36: Short-Form 36; TLC: total lung capacity; TDI: transition dyspnea index. FEV1 and TLC values for the responder and non-responder We found that responders had higher VO2, VCO2, and VE groups were different (Table IV). Whereas, the respond- during the post-PR 6MWT, but no differences in cardio- ers showed increased (or at least the maintenance) FEV1 respiratory parameters were detected in non-responders and TLC following the PR program, both parameters were before vs. after the PR program. The two patient groups decreased in the non-responder group (P=0.04 for ∆FEV1, also showed no significant differences in pre- vs. post-PR P=0.08 for ∆TLC; Table IV). However, there were no changes in dyspnea, quality of life, or anxiety/depression significant differences between the responder and non- scores. The non-responder group was notable for the pre- responder groups in the changes in quality of life, anxiety, dominance of IPF patients (9/10) and the greater arterial depression, or dyspnea from pre- to post-PR. oxygen desaturation during exercise. Cardiorespiratory adaptation during the 6MWT Characterization of responders and non-responders Of the 19 patients who completed the PR program, 13 were Our findings are consistent with previous studies showing analyzed for gaseous exchange during the 6MWT (six re- the beneficial effects of a PR program for f-IIP patients in sponders and seven non-responders) (Figure 1). The mean the 6MWT.5, 7, 8, 10, 11, 25, 37 However, we also observed a cardiorespiratory values (HR, VO2, VCO2, VE, VT, and Bf) 53% non-responder rate, which matches previous reports for the group of 13 were not significantly different before that 58-60% of IPF patients failed to attain a MID of 28- vs. after the PR program or at rest vs. during the 6MWT. 34 m in a post-PR 6MWT.25, 38 One exception to this trend The change in the 6MWD from pre- to post-PR correlated was the study of Vainshelboim et al., who reported that with the changes in VO2 (r=0.59, P=0.03), in VE (r=0.69, 13 of 15 IPF patients reached a MID of 25 m in a post-PR P=0.01) and in Bf (r=0.55, P=0.05). 6MWT.10 This difference could be partly explained by the shorter MID selected in the latter study. In the responder group (N.=6), the post-PR 6MWT showed significantly increased VO2 and VCO2 during the The baseline severity of lung disease (FVC, SpO2nadir),38 last 3 min and significantly increased VE throughout com- dyspnea grade,37 and 6MWD11, 39 have been reported to pared with the pre-PR 6MWT (Figure 2A). In contrast, HR predict post-PR changes in the 6MWD in ILD patients. and ventilatory patterns were not significantly affected by In our study, the non-responder group contained more pa- the PR program. In the non-responder group (N.=7), none tients with IPF and with severe lung diffusion impairment, of the cardiorespiratory parameters evaluated were sig- as reflected by baseline DLCO level and arterial oxygen de- nificantly changed between the pre- and post-PR 6MWT saturation during the 6MWT, compared with the responder (Figure 2B). group. These results are consistent with the study of Hol- land et al., who found that exercise-induced desaturation Discussion was a significant predictor of the change in 6MWD in ILD patients after a 6-month PR program.38 This result should In this study, we showed that more than half (10/19, 53%) be interpreted with caution, however, since a similar study of f-IIP patients who underwent an 8-week home-based of ILD patients by Dowman et al. found that SpO2nadir did PR program did not reach the MID of 30 m in the 6MWD. not predict short- or long-term changes in 6MWD post- 108 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 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, 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. Vol. 55 - No. 1 B A COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA Figure 2.—Change in cardiorespiratory parameters during the 6MWT performed before and after a PR program: results for the 6 responders (A) and Heart rate (beats·min-1) VO2 (mL·min-1) Heart rate (beats·min-1) VO2 (mL·min-1) CARDIORESPIRATORY ADAPTATION IN INTERSTITIAL PNEUMONIA CHÉHÈRE 7 non-responders (B) who underwent gas exchange analysis. Data are presented as the mean±SD. P-ANOVA = 0.78 80 120 250 500 750 1000 1500 40 60 20 80 40 100 60 120 250 500 750 1000 1250 1750 6MWT: 6-Minute Walk Test; HR: heart rate; VCO2: carbon dioxide output; VO2: O2 uptake; VE: minute ventilation. 40 20 40 P-ANOVA = 0.14 1250 P-ANOVA = 0.03 1500 *P<0.05; **P<0.01 by repeated measures ANOVA. European Journal of Physical and Rehabilitation Medicine 109 R 1 2 3 4 5 6 60 100 0 Pre-PR R 1 2 3 4 5 6 0 Pre-PR R 1 2 3 4 5 6 Post-PR R 1 2 3 4 5 6 Post-PR 140 P-ANOVA = 0.65 VE (L·min-1) VO2 (mL·min-1) VE (L·min-1) VO2 (mL·min-1) P-ANOVA = 0.74 1250 0 1750 0 1000 R 1 2 3 4 5 6 P-ANOVA < 0.001 1500 R 1 2 3 4 5 6 750 1250 500 1000 250 750 P-ANOVA = 0.76 500 0 R 1 2 3 4 5 6 250 P-ANOVA = 0.01 60 0 R 1 2 3 4 5 6 80

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, CHÉHÈRE CARDIORESPIRATORY ADAPTATION IN INTERSTITIAL PNEUMONIA 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. PR.11 This discrepancy could be due to the higher propor- adaptation during a 6MWT in patients with pulmonary tion of patients with connective tissue disease-related ILD disease. Similar to the study of IPF patients by Jackson and the small number of patients requiring long-term or et al., we observed no differences in cardiorespiratory ad- exertional oxygen therapy in the study of Dowman et al.11 aptation following the PR program,41 despite significant As yet unidentified factors may also contribute to post-PR differences in performance. But, the responders and non- changes in 6MWD in ILD patients. For example, Holland responders differed in their VO2, VCO2, and VE during the et al. reported that desaturation may limit exercise inten- post-PR 6MWT. Our results also suggest that f-IIP patients sity during PR in ILD patients, thus reducing the potential with lower DLCO and greater desaturation are less likely beneficial effect of the program on the 6MWD.38 Patients to show improved 6MWD and cardiorespiratory perfor- in that study—as in ours—were permitted O2 therapy to mance after a PR program. However, we note that the maintain a SpO2 ≥88% during the PR program. Moreover, number of subjects compared is small. As proposed for a recent study found no difference in pre-6MWT SpO2nadir COPD patients,42 it is possible that the greater hypoxia ob- between ILD patients who did or did not improve their served in non-responders may limit muscular adaptation, exercise intensity during a PR program.11 and thus the post-PR 6MWD. Therefore, for ILD patients and especially f-IIP, further studies are warranted to ex- Interestingly, we did not observe significant differences plore the mechanisms, including muscular adaptation, un- between the responders and non-responders in pre- vs. derlying the post-PR differences between responders and post-PR quality of life, dyspnea, or anxiety and depression non-responders. scores. Spielmanns et al. similarly found no significant differences in SF-36 scores between responders and non- Limitations of the study responders in their study.18 These findings indicate that PR can have major benefits for f-IIP patients, independently One limitation of this study is the mild to moderate dis- of improvements in 6MWD. The educational therapy as- ease severity of the cohort. We excluded patients who re- pect of the program is likely to have played a part in the quired oxygen supplementation at rest for technical rea- improved symptom and quality of life scores, supporting sons related to the gas exchange analyzer. The fact that the previous recommendation40 that educational therapy the pre- and post-PR 6MWT were performed without oxy- and exercise training should be combined to optimize the gen may also explain the failure of some patients to show benefits of PR. Arizono et al. compared the effects of a PR overt improvement in the response. Even if some muscular program on the performance of f-IIP patients in various ex- adaptation had occurred, it may have been masked by a ercise tests, and they found that duration of the endurance worsening of pathology and consequent need for oxygen shuttle walk test (ESWT) was a more sensitive readout of therapy post-PR. In addition, the sample size was small, PR efficacy than the distance walked in the 6MWT.25 We and our results should be verified on a larger cohort. In note, however, that improvement in the 6MWD, but not particular, the small number of subjects per responder and the ESWT, requires the patient to walk faster after PR. It non-responder group limits the interpretation of the car- seems likely that it would be easier for a deconditioned diorespiratory outcomes of the 6MWT. However, we did patient to increase the duration, rather than intensity, of observe evidence of differences in respiratory adaptation physical activity during a home-based program, given between the responders and non-responders with moderate the lack of involvement of medical staff. In our study, for to large effect sizes. example, the non-responders may have shown improve- ments in exercise duration, following the PR program, but Conclusions this would not be detected by the 6MWT. Future studies should include a more detailed investigation of the re- More than 50% of the f-IIP patients in our cohort did sponders and non-responders, which could provide valu- not show an improvement in the 6MWD following a able information for clinicians to design different training PR program. Further studies should focus on the under- regimens for patients who are ‘non-responders’ using a lying mechanisms that explain the observed differences particular exercise test. between responders and non-responders, especially the role of chronic hypoxemia. This is a crucial point, be- Cardiorespiratory adaptation during the 6MWT cause patients who are non-responsive to the 6MWT may have poorer prognoses and survival rates compared with To the best of our knowledge, this is the first study to de- responders. Nevertheless, non-responders and respond- scribe the effects of a PR program on cardiorespiratory 110 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, CARDIORESPIRATORY ADAPTATION IN INTERSTITIAL PNEUMONIA CHÉHÈRE 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. ers showed similar benefits in quality of life and dyspnea, exercise training in patients with interstitial lung disease. J Cardiopulm suggesting that these parameters, or other exercise met- Rehabil Prev 2015;35:47–55. rics, may be more sensitive to PR than the 6MWD MID 17.  Dowman L, Hill CJ, Holland AE. Pulmonary rehabilitation for in- for some patients. For example, since our patient cohort terstitial lung disease. Cochrane Database Syst Rev 2014;10:CD006322. was asked to prioritize an increase in exercise duration 18.  Spielmanns M, Gloeckl R, Schmoor C, Windisch W, Storre JH, Boen- over intensity during the 8-week program, it is possible sch M, et al. Effects on pulmonary rehabilitation in patients with COPD that the non-responders in the 6MWT may have been or ILD: A retrospective analysis of clinical and functional predictors with responders in endurance tests performed at submaximal particular emphasis on gender. Respir Med 2016;113:8–14. intensity. 19.  Troy LK, Young IH, Lau EM, Corte TJ. Exercise pathophysiology and the role of oxygen therapy in idiopathic interstitial pneumonia. Respi- References rology 2016;21:1005–14. 20.  Gigliotti F, Coli C, Bianchi R, Romagnoli I, Lanini B, Binazzi B, et 1.  King TE Jr. Clinical advances in the diagnosis and therapy of the inter- al. Exercise training improves exertional dyspnea in patients with COPD: stitial lung diseases. Am J Respir Crit Care Med 2005;172:268–79. evidence of the role of mechanical factors. Chest 2003;123:1794–802. 2.  Markovitz GH, Cooper CB. Exercise and interstitial lung disease. Curr 21.  Puente-Maestu L, Abad YM, Pedraza F, Sánchez G, Stringer WW. Opin Pulm Med 1998;4:272–80. A controlled trial of the effects of leg training on breathing pattern and 3.  Chang JA, Curtis JR, Patrick DL, Raghu G. Assessment of health- dynamic hyperinflation in severe COPD. Hai 2006;184:159–67. related quality of life in patients with interstitial lung disease. Chest 22.  Skumlien S, Skogedal EA, Bjørtuft O, Ryg MS. Four weeks’ inten- 1999;116:1175–82. sive rehabilitation generates significant health effects in COPD patients. 4.  Holland AE. Exercise limitation in interstitial lung disease - mechanisms, Chron Respir Dis 2007;4:5–13. significance and therapeutic options. Chron Respir Dis 2010;7:101–11. 23.  Ramponi S, Tzani P, Aiello M, Marangio E, Clini E, Chetta A. Pul- 5.  Ozalevli S, Karaali HK, Ilgin D, Ucan ES. Effect of home-based pul- monary rehabilitation improves cardiovascular response to exercise in monary rehabilitation in patients with idiopathic pulmonary fibrosis. Mul- COPD. Respiration 2013;86:17–24. tidiscip Respir Med 2010;5:31–7. 24.  Vainshelboim B, Kramer MR, Fox BD, Izhakian S, Sagie A, Oliveira 6.  Rammaert B, Leroy S, Cavestri B, Wallaert B, Grosbois JM. Home- J. Supervised exercise training improves exercise cardiovascular function based pulmonary rehabilitation in idiopathic pulmonary fibrosis. Rev Mal in idiopathic pulmonary fibrosis. Eur J Phys Rehabil Med 2017;53:209–18. Respir 2011;28:e52–7. 25.  Arizono S, Taniguchi H, Sakamoto K, Kondoh Y, Kimura T, Kataoka 7.  Holland AE, Hill CJ, Conron M, Munro P, McDonald CF. Short term K, et al. Endurance time is the most responsive exercise measurement in improvement in exercise capacity and symptoms following exercise train- idiopathic pulmonary fibrosis. Respir Care 2014;59:1108–15. ing in interstitial lung disease. Thorax 2008;63:549–54. 26.  American Thoracic Society; European Respiratory Society. American 8.  Nishiyama O, Kondoh Y, Kimura T, Kato K, Kataoka K, Ogawa T, et Thoracic Society/European Respiratory Society International Multidisci- al. Effects of pulmonary rehabilitation in patients with idiopathic pulmo- plinary Consensus Classification of the Idiopathic Interstitial Pneumonias. nary fibrosis. Respirology 2008;13:394–9. This joint statement of the American Thoracic Society (ATS), and the Eu- 9.  Ryerson CJ, Cayou C, Topp F, Hilling L, Camp PG, Wilcox PG, et al. ropean Respiratory Society (ERS) was adopted by the ATS board of direc- Pulmonary rehabilitation improves long-term outcomes in interstitial lung tors, June 2001 and by the ERS Executive Committee, June 2001. Am J disease: a prospective cohort study. Respir Med 2014;108:203–10. Respir Crit Care Med 2002;165:277–304. 10.  Vainshelboim B, Oliveira J, Yehoshua L, Weiss I, Fox BD, Fruchter 27.  Travis WD, Costabel U, Hansell DM, King TE Jr, Lynch DA, Nich- O, et al. Exercise training-based pulmonary rehabilitation program olson AG, et al.; ATS/ERS Committee on Idiopathic Interstitial Pneu- is clinically beneficial for idiopathic pulmonary fibrosis. Respiration monias. An official American Thoracic Society/European Respiratory 2014;88:378–88. Society statement: update of the international multidisciplinary classifica- 11.  Dowman LM, McDonald CF, Hill CJ, Lee AL, Barker K, Boote C, et tion of the idiopathic interstitial pneumonias. Am J Respir Crit Care Med al. The evidence of benefits of exercise training in interstitial lung disease: 2013;188:733–48. a randomised controlled trial. Thorax 2017;72:610–9. 28.  Fabre C, Chéhère B, Bart F, Mucci P, Wallaert B, Grosbois JM. Rela- 12.  Travis WD, Matsui K, Moss J, Ferrans VJ. Idiopathic nonspecific tionships between heart rate target determined in different exercise testing interstitial pneumonia: prognostic significance of cellular and fibrosing in COPD patients to prescribed with individualized exercise training. Int J patterns: survival comparison with usual interstitial pneumonia and des- Chron Obstruct Pulmon Dis 2017;12:1483–9. quamative interstitial pneumonia. Am J Surg Pathol 2000;24:19–33. 29.  Quanjer PH, Tammeling GJ, Cotes JE, Pedersen OF, Peslin R, Yer- 13.  Lederer DJ, Arcasoy SM, Wilt JS, D’Ovidio F, Sonett JR, Kawut SM. nault JC; Official Statement of the European Respiratory Society. Lung Six-minute-walk distance predicts waiting list survival in idiopathic pul- volumes and forced ventilatory flows. Report Working Party Standardiza- monary fibrosis. Am J Respir Crit Care Med 2006;174:659–64. tion of Lung Function Tests, European Community for Steel and Coal. Eur 14.  Caminati A, Bianchi A, Cassandro R, Mirenda MR, Harari S. Walking Respir J Suppl 1993;16:5–40. distance on 6-MWT is a prognostic factor in idiopathic pulmonary fibro- 30.  Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health sis. Respir Med 2009;103:117–23. survey (SF-36). I. Conceptual framework and item selection. Med Care 15.  Bajwah S, Ross JR, Peacock JL, Higginson IJ, Wells AU, Patel AS, 1992;30:473–83. et al. Interventions to improve symptoms and quality of life of patients 31.  Mahler DA, Weinberg DH, Wells CK, Feinstein AR. The measure- with fibrotic interstitial lung disease: a systematic review of the literature. ment of dyspnea. Contents, interobserver agreement, and physiologic cor- Thorax 2013;68:867–79. relates of two new clinical indexes. Chest 1984;85:751–8. 16.  Keyser RE, Woolstenhulme JG, Chin LM, Nathan SD, Weir NA, 32.  Borg GA. Psychophysical bases of perceived exertion. Med Sci Connors G, et al. Cardiorespiratory function before and after aerobic Sports Exerc 1982;14:377–81. 33.  Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983;67:361–70. 34.  ATS Committee on Proficiency Standards for Clinical Pulmonary Function Laboratories. ATS statement: guidelines for the six-minute walk test. Am J Respir Crit Care Med 2002;166:111–7. Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 111

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, CHÉHÈRE CARDIORESPIRATORY ADAPTATION IN INTERSTITIAL PNEUMONIA 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. 35.  Holland AE, Hill CJ, Conron M, Munro P, McDonald CF. Small Pulmonary rehabilitation in interstitial lung disease: benefits and predic- changes in six-minute walk distance are important in diffuse parenchymal tors of response. Chest 2009;135:442–7. lung disease. Respir Med 2009;103:1430–5. 40.  Spruit MA, Singh SJ, Garvey C, ZuWallack R, Nici L, Rochester 36.  Cohen J. Statistical Power Analysis for the Behavioural Sciences, 2nd C, et al.; ATS/ERS Task Force on Pulmonary Rehabilitation. An official Edition. New York: Lawrence Erlbaum Associates; 1988. American Thoracic Society/European Respiratory Society statement: key 37.  Kozu R, Senjyu H, Jenkins SC, Mukae H, Sakamoto N, Kohno S. concepts and advances in pulmonary rehabilitation. Am J Respir Crit Care Differences in response to pulmonary rehabilitation in idiopathic pul- Med 2013;188:e13–64. monary fibrosis and chronic obstructive pulmonary disease. Respiration 41.  Jackson RM, Gómez-Marín OW, Ramos CF, Sol CM, Cohen MI, 2011;81:196–205. Gaunaurd IA, et al. Exercise limitation in IPF patients: a randomized trial 38.  Holland AE, Hill CJ, Glaspole I, Goh N, McDonald CF. Predictors of pulmonary rehabilitation. Hai 2014;192:367–76. of benefit following pulmonary rehabilitation for interstitial lung disease. 42.  Costes F, Gosker H, Feasson L, Desgeorges M, Kelders M, Castells Respir Med 2012;106:429–35. J, et al. Impaired exercise training-induced muscle fiber hypertrophy and 39.  Ferreira A, Garvey C, Connors GL, Hilling L, Rigler J, Farrell S, et al. Akt/mTOR pathway activation in hypoxemic patients with COPD. J Appl Physiol (1985) 2015;118:1040–9. 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.—Baptiste Chéhère made substantial contributions to study conception and design; acquisition, analysis, and interpretation of the data; and drafting of the manuscript. Valérie Bougault made substantial contributions to study conception and design, analysis and interpretation of the data, and drafting of the manuscript. Cécile Chenivesse made substantial contributions to acquisition of the data and drafting of the manuscript. Jean-Marie Grosbois made substantial contributions to study conception and design, acquisition of the data, and drafting of the manuscript. Benoit Wallaert made substantial con- tributions to study conception and design, analysis and interpretation of the data, and drafting of the manuscript. Acknowledgements.—The authors would like to thank the rehabilitation team (G. Tywoniuk, S. Duriez, F. Urbain, V. Wauquier, and M. Lambinet) and the staff of the Competence Centre for Rare Pulmonary Diseases who managed the patients during the home-based PR program and routine monitoring. The authors would also like to thank Adair, France Oxygène, Homeperf, LVL Medical, Orkyn, Santélys, SOS Oxygène, Sysmed, VitalAire, and ARS Nord-Pas- de-Calais for their support during the home-based PR program. Article first published online: June 14, 2018. - Manuscript accepted: June 13, 2018. - Manuscript revised: April 24, 2018. - Manuscript received: November 29, 2017. 112 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):113-22 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/S1973-9087.18.05156-0 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 Combined aerobic exercise and high-intensity respiratory muscle training in patients surgically treated for non-small cell lung cancer: a pilot randomized clinical trial Monique MESSAGGI-SARTOR 1, 2, Ester MARCO 1, 2, 3 *, Elisabeth MARTÍNEZ-TÉLLEZ 4, Alberto RODRIGUEZ-FUSTER 5, Carolina PALOMARES 6, Sandra CHIARELLA 2, Josep M. MUNIESA 1, 2, Mauricio OROZCO-LEVI 7, 8, Esther BARREIRO 8, 9, 10, Maria R. GÜELL 6 1Rehabilitation Research Group, Institut Hospital del Mar d’Investigacions Mèdiques (IMIM), Barcelona, Spain; 2Department of Physical Medicine and Rehabilitation, Parc de Salut Mar, Hospital del Mar, Hospital de l’Esperança, Barcelona, Spain; 3School of Medicine, Autonomous University of Barcelona, Barcelona, Spain; 4Department of Thoracic Surgery, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain; 5Department of Thoracic Surgery, Hospital del Mar, Barcelona, Spain; 6Unit of Pulmonary Rehabilitation, Department of Respiratory Medicine, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain; 7Department of Respiratory Medicine, Hospital del Mar, Barcelona, Spain; 8Department of Health and Experimental Sciences (CEXS), Universitat Pompeu i Fabra (UPF), Parc de Recerca Biomèdica de Barcelona (PRBB), Barcelona, Spain; 9Department of Pulmonology, Lung Cancer and Muscle Research Group, IMIM, Hospital del Mar, Barcelona, Spain; 10Centro de Investigación en Red de Enfermedades Respiratorias (CIBERES), Instituto de Salud Carlos III (ISCIII), Madrid, Spain *Corresponding author: Ester Marco, Cardiopulmonary Rehabilitation Unit, Physical Medicine and Rehabilitation, Parc de Salut Mar (Hospital del Mar - Hospital de l’Esperança), Institut Hospital del Mar d’Investigacions Mèdiques (IMIM), Autonomous University of Barcelona, Sant Josep de la Muntanya 12, 08024 Barcelona, Spain. E-mail: [email protected] ABSTRACT BACKGROUND: Lung resection surgery further decreases exercise capacity and negatively affects respiratory muscle function in patients with non-small cell lung cancer (NSCLC). The best design for exercise interventions in these patients has not been determined yet. AIM: To assess the impact of aerobic exercise and high-intensity respiratory muscle training on patient outcomes following lung cancer resec- tion surgery. DESIGN: Prospective, single-blind, pilot randomized controlled trial. SETTING: Outpatient cardiopulmonary rehabilitation unit of two university hospitals. POPULATION: Thirty-seven patients with NSCLC after tumor resection. METHODS: Patients were randomly assigned to exercise training or usual post-operative care. The training program consisted of aerobic ex- ercises and high-intensity respiratory muscle training (24 supervised sessions, 3 per week, 8 weeks). Primary outcome was exercise capacity assessed with peak oxygen uptake (VO2peak) during cardiopulmonary exercise test. Secondary outcomes included changes in respiratory muscle strength, levels of serum insulin growth factor I (IGF-I) and IGF binding protein 3 (IGFBP-3), and quality of life assessed with the European Organization for Research and Treatment of Cancer (EORTC QLQ-C30) questionnaire. RESULTS: The 8-week training program was associated with significant improvement in VO2peak (2.13 mL/Kg/min [95%CI 0.06 to 4.20]), maximal inspiratory and expiratory pressures (18.96 cmH2O [95% CI 2.7 to 24.1] and 18.58 cmH2O [95% CI 4.0 to 33.1], respectively) and IGFBP-3 (0.61 µg/mL [%95 CI 0.1 to 1.12]). No significant differences were observed in the EORTC QLQ-C30. CONCLUSIONS: An 8-week exercise program consisting of aerobic exercise and high-intensity respiratory muscle training improved exercise capacity, respiratory muscle strength, and serum IGFBP-3 levels in NSCLC patients after lung resection. There was no impact on the other outcomes assessed. CLINICAL REHABILITATION IMPACT: A combination of aerobic exercise and respiratory muscle training could be included in the rehabilita- tion program of deconditioned patients with NSCLC after lung resection surgery. (Cite this article as: Messaggi-Sartor M, Marco E, Martínez-Téllez E, Rodriguez-Fuster A, Palomares C, Chiarella S, et al. Combined aerobic exer- cise and high-intensity respiratory muscle training in patients surgically treated for non-small cell lung cancer: a pilot randomized clinical trial. Eur J Phys Rehabil Med 2019;55:113-22. DOI: 10.23736/S1973-9087.18.05156-0) Key words: Lung neoplasms - Cardiorespiratory fitness - Breathing exercises - Rehabilitation. Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 113

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, MESSAGGI-SARTOR AEROBIC EXERCISE IN NSCLC CANCER PATIENTS 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 with non-small cell lung cancer (NSCLC) may between lung cancer and circulating serum levels of have symptoms such as dyspnea, fatigue, anxiety, and IGF-I and IGFBP-3 revealed that IGF-3 acts as a tumor pain that contribute to physical inactivity, which can re- suppressor and is inversely correlated with lung cancer sult in further deconditioning.1 After lung resection sur- risk.13 Moreover, IGFBP-3 has been suggested as a sensi- gery, dyspnea, physical activity, exercise tolerance, stair tive marker of physical training in healthy rugby play- climbing performance, and quality of life may worsen and ers.14 remain impaired for 6 months or longer.2-6 Moreover, ad- juvant treatments (chemo- or radiotherapy) may adversely The main aim of this clinical trial was to assess the affect patients’ symptoms and physical performance, as impact of an 8-week intervention combining aerobic ex- reported in other cancer populations.3 ercise with inspiratory and expiratory muscle training (IEMT) on cardiovascular fitness and respiratory muscle Lung cancer treatment initiates a deconditioning storm dysfunction in NSCLC patients, considering changes in that further reduces the capacity to deliver and utilize oxy- exercise capacity, respiratory muscle strength, HRQoL, gen and metabolic substrates during exercise, contribut- and serum levels of IGF-I and IGFBP-3 levels. Second- ing to poor cardiorespiratory fitness.4 Exercise-based ran- ary objectives were to quantify the impairment in exercise domized clinical trials in cancer populations are scarce. A capacity and respiratory muscle function attributable to recent meta-analysis of eight controlled trials demonstrat- lung resection and to study the impact, if any, of the ex- ed the positive effects of exercise training to improve ex- ercise intervention on recurrences and death at two years ercise capacity and health-related quality of life (HRQoL) after surgery. after lung resection in patients with NSCLC stages I-IIIB, many of whom underwent adjuvant chemotherapy in ad- Materials and methods dition to lung resection.5 Most of the post-operative ex- ercise interventions were derived from existing training Design programs for patients with chronic obstructive pulmonary disease (COPD) that combined aerobic and resistance A two-center, prospective, single-blind, pilot randomized exercises.6, 7 However, the heterogeneity of participants controlled trial was designed, according to the Consoli- (lung cancer stage, adjuvant therapies) and interventions dated Standards of Reporting Trials Statements (CON- (components, modality, intensity and timing of exercise SORT),15 to determine the effectiveness of a pulmonary programs) in published studies makes it difficult to de- rehabilitation program in patients with NSCLC who are termine the optimal exercise intervention in surgically eligible for tumor resection. The clinical trial was ap- treated NSCLC patients.8 This lack of good-quality evi- proved by the local clinical research ethics committee, dence precludes drawing any conclusions or generaliza- registered in clinicaltrials.gov (NCT01771796), and per- tions about the best design for exercise intervention in formed in accordance with the Declaration of Helsinki. these patients.9, 10 Written informed consent was obtained from all partici- pants. The addition of respiratory muscle training to a gen- eral exercise program improves dyspnea and exercise Participants performance in patients with COPD;8 in surgically treated NSCLC patients, however, the effectiveness of respiratory Patients newly diagnosed with resectable NSCLC were muscle training has not been well addressed. In the only screened by the throracic surgery unit in two tertiary hos- randomized clinical trial reported to date in this population, pitals in the city of Barcelona and referred to the pulmo- two weeks of inspiratory muscle training in the immediate nary rehabilitation unit associated with these two hospi- postoperative period after lung resection improved oxy- tals. Inclusion criteria were age <80 years, diagnosis of genation but did not preserve respiratory muscle strength, stage I or II NSCLC, referral to lung cancer resection by compared with two weeks of standard physical therapy.11 muscle-sparing lateral thoracotomy or videothoracoscopy, and ability to understand and consent to the trial proce- The insulin-like growth factor system has a pivotal role dures. Exclusion criteria included adjuvant treatments in tumor biology and potential implications for health and (chemo- or radiotherapy, postoperative complications that fitness. High plasma levels of serum insulin growth fac- would prevent performing a maximal exercise test in the 6 tor I (IGF-I) and IGF binding protein 3 (IGFPB-3) are weeks after surgery, previous history of thoracic surgery, associated with good prognosis in patients with advanced and neurological and/or musculoskeletal comorbidities NSCLC.12 A meta-analysis evaluating the relationship 114 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, AEROBIC EXERCISE IN NSCLC CANCER PATIENTS MESSAGGI-SARTOR 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 prevented performing the study protocol. Patients Respiratory muscle strength, defined as the ability to without a baseline cardiopulmonary exercise test (CPET) develop a brief maximal respiratory effort, was assessed and muscle function assessments in the preoperative pe- through maximal inspiratory and expiratory pressures riod were also excluded. (PImax and PEmax, respectively). PImax was measured at the mouth during a maximum effort from residual volume Sample size against an occluded airway; to determine the PEmax, pa- tients performed a maximum expiratory effort from total Sample size was calculated to detect a change in the peak lung capacity with the occluded airway.16 The mouthpiece oxygen uptake (VO2peak) of 2.0 mL/Kg/min and standard used in the maneuvers had a small orifice to minimize deviation (SD) of 2.3 mL/kg/min, resulting in a required the participation of face and mouth muscles and was sample of 21 subjects per group to detect significant dif- connected to a MicroRPM pressure transducer (Micro ferences with an alpha-risk of 0.05 and a beta-risk of 20% Medical/Care Fusion, Kent, United Kingdom). A flanged on a bilateral contrast. The sample size was overestimated mouthpiece was used to create an optimal mouth seal in to allow potential losses of 10%. the presence of orofacial weakness. The highest value of three reproducible maneuvers (10% variability between Randomization and blinding values) was used for analysis. Reference values were those previously published for a Mediterranean popula- After receiving information about the study procedures, tion.17 eligible patients were randomly assigned to the exercise or control groups. Randomization was performed indepen- HRQoL was assessed using the European Organiza- dently by a staff member blinded to patient identity, who tion for Research and Treatment of Cancer questionnaire assigned the anonymized patient record to one of the study (EORTC QLQ-C30), specifically designed for use in clini- groups using a random number generator program. The re- cal trials. Single-item measures and scales range in score searchers assessing the main outcomes were blinded to the from 0 to 100. A high score for a functional scale rep- study group assignments. resents “a high/healthy level of functioning” and a high score for the global health status/quality of life represents Outcome variables a “high HRQoL”; in contrast, a high score for a symptom scale or item represents a “high level of symptomatology/ The outcome variables were the changes from baseline problems.”18 in exercise capacity, respiratory muscle strength, HRQoL, and prognostic biomarkers of NSCLC after the exercise Serum IGF-I and IGFBP-3 levels were analyzed using intervention. In a 2-year follow-up, lung cancer recur- ELISA according to manufacturer instructions. Before rences and deaths were prospectively assessed. The main and after exercise intervention, fasting peripheral venous outcome variable was exercise capacity, measured by blood (5 mL) was collected in the morning from all partici- CPET as VO2peak (mL/Kg/min). In addition, maximal pants. Blood samples were kept at room temperature for workload (watts), maximal ventilation (L/min), and maxi- 30 minutes. Serum was collected following centrifugation mal heart rate (beats per minute) in the CPET were regis- and frozen immediately at -20 °C until analysis. tered. All variables were also expressed as a percentage of the predicted values (%pred.). The test was performed on Long-term outcomes, such as lung cancer recurrence a cyclo-ergometer connected to a computerized analyzer and death, were also recorded from medical records at (Sensor Medics, Milan, Italy), using the breath-by-breath 2-year follow-up or caregiver telephone interview. method. Patients were monitored by blood pressure mea- surements every 2 minutes, SO2 measurement by pulse Exercise intervention oximeter, continuous 12-lead EKG, and exhaled O2 and CO2 breath-by-breath, measured at the patient’s mouth. The Template for Intervention Description and Replication The test began with a 2-minute evaluation of the patient (TIDieR) checklist was used to describe the exercise inter- at rest, followed by a warm-up period in which the patient vention. This intervention consisted of a combination of cycled freely for 2 minutes. Afterwards, the patient was two exercise modalities: continuous aerobic training and asked to pedal fast enough to maintain 55-65 rotations per IEMT (24 1-hour sessions, 3 times per week, 8 weeks). min at a work rate that increased 20 watts per minute until Continuous aerobic training was performed on an ergomet- exhaustion. ric bicycle; initial intensity was set at 60% of baseline peak workload and increased by 5 watts weekly if the patient was able to tolerate the set load for 30 min. IEMT consisted Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 115

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, MESSAGGI-SARTOR AEROBIC EXERCISE IN NSCLC CANCER PATIENTS 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 5 sets of 10 repetitions followed by 1-2 min of unloaded Assessed for eligibility (N.=70) recovery breathing (off the device), twice a day, 3 days per week, for 8 weeks. In the first session, training loads were Excluded (N.=33) set at 30% of PImax and PEmax; once the patient was fa- - Diagnosis other than NSCLC (N.=5) miliarized with the device, training load was set at 50% - Required chemotherapy and/or of PImax and PEmax, and adjusted weekly by 10 cmH2O if tolerated. Inspiratory and expiratory muscles were trained radiotherapy (N.=9) simultaneously using a respiratory muscle trainer (Orygen- - Post-operative complications (N.=2) Dual®, Forumed, Spain) at a rate of 15-20 breaths/min. All - Absence of exercise test before sessions included 5-min warm-up and 5-min cool down, and three series of bicep curls and chest and shoulder press surgery (N.=4) with a constant load of 0.5 kg. After completing the inter- - Declined participation (N.=6) vention, patients were encouraged to continue doing exer- - Other reasons (N.=7) cise at home or in community sports facilities. Randomized (N.=37) Allocation Allocated to exercise program (N.=16) Allocated to usual care (N.=21) Standard care adopted in the control group Pre-exercise intervention The control group received standard medical treatment and (4-6 weeks after surgery) periodic monitoring. Patients in the control group were ad- vised to perform physical activity, following WHO recom- Lost to follow-up (N.=0) Lost to follow-up (N.=0) mendations in the context of daily, family, and community activities: leisure time physical activity (walking, dancing, Analysis hiking, swimming), transportation (walking or cycling), occupational (if still in the workforce), household chores, Analyzed (N.=16) Analyzed (N.=21) play, games, and sports or planned exercise. Patients with a high level of physical activity were encouraged to con- Post-exercise tinue exercising. intervention (6 months after surgery) Lost to follow-up (N.=5) Lost to follow-up (N.=8) - Declined participation (N.=3) - Declined participation (N.=5) - Received chemotherapy (N.=2) - Received chemotherapy (N.=2) - Complications after surgery (N.=1) Study protocol Analysis Outcome variables were assessed by measuring maximal Analyzed (N.=11) Analyzed (N.=13) respiratory pressures and CPET at 6-8-weeks time (T1) af- ter surgery and after completing the exercise intervention Figure 1.—CONSORT 2010 Flow Diagram. (T2). Other data collected were age, sex, anthropometric characteristics, smoking history, comorbidities, and cancer analysis. Differences in outcome variables, adjusted by type. Only patients with pre-operative CPET and respira- baseline values, were assessed by linear regression model- tory muscle assessment were included, in order to quantify ing and expressed with the unstandardized beta coefficient changes from baseline (T0) data in exercise capacity and (B) and its standard error (SE); collinearity was checked respiratory muscle function attributable to lung resections. by testing the variance inflation factor. Changes in HRQoL A 2-year follow-up of the whole cohort was designed to parameters were assessed by analysis of variance using a determine mortality and recurrence. repeated-measures mixed design for intrasubject analysis and a one-factor design for intersubject analysis. Effect Statistical analysis size was calculated using the Cohen term d index. The level of significance was set at P≤0.05. Data analysis was Quantitative variables are presented as mean and stan- performed using the IBM SPSS Statistics v21. dard deviation (SD), unless otherwise stated. Univariate analysis was performed using x2, Fisher exact or Students’ Results t-tests, depending on the variables analyzed. The main outcome measures were analyzed per protocol. Students’ Figure 1 describes the flow of participants through each t-test for independent samples was used for inter-group stage of the study period. A total of 70 patients were analysis and t-test for repeated measures for intra-group screened to participate in the trial. Exclusion from ran- 116 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, AEROBIC EXERCISE IN NSCLC CANCER PATIENTS MESSAGGI-SARTOR 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 I.—Baseline demographic and clinical characteristics of participants. Age (years) Total sample Training group Control group P Sex: (N.=37) (N.=16) (N.=21) • Men >0.05 • Women 64.6 (SD 8.5) 64.2 (SD 8.1) 64.8 (SD 8.9) Body mass index (Kg/m2) 0.03 Smoking (%): 26 (70.3%) 8 (50%) 18 (85.7%) • Current smokers 11 (29.7%) 8 (50%) 3 (14.3%) >0.05 • Never smoked 27.4 (SD 3.8) 28.1 (SD 4.6) • Past smokers 26.8 (SD 3.1) >0.05 Smoking history (packs/year) 17 (45.9%) 5 (31.3%) >0.05 Presence of COPD 4 (10.8%) 3 (18.8%) 12 (57.1%) >0.05 Pulmonary function: 16 (43.2%) 8 (50.0%) 1 (4.8%) >0.05 • FEV1 (%pred.) 43.1 (SD 23.4) 38.4 (SD 23.1) 8 (38.1%) • FVC (%pred.) 27 (73%) 0.04 • FVC/FEV1 (%pred.) 75.5 (SD 19.1) 46.5 (SD 23.6) >0.05 • TLC (%pred.) 68.7 (SD 18.5) 85.2 (SD 18.3) >0.05 • DLCO (%pred.) 81.9 (SD 14.9) 69.3 (SD 8.1) 63.5 (SD 16.6) >0.05 Surgery: 66.3 (SD 14.7) 101.5 (SD 15.5) 79.4 (SD 11.8) >0.05 • Lobectomy 100.3 (SD 15.0) 70.5 (SD 14.2) 63.9 (SD 18.3) • Minor resection 68.4 (SD 15.8) 99.0 (SD 15.0) >0.05 • Pneumectomy 13 (81.3%) 66.3 (SD 17.4) >0.05 Surgery type: 27 (73.0%) 3 (18.8%) >0.05 • Thoracotomy 9 (24.3%) 0 14 (66.7%) • Videothoracoscopy 1 (2.7%) 6 (28.6%) >0.05 Histological feature: 14 (87.5%) 1 (4.8%) >0.05 • Adenocarcinoma 34 (91.9%) 2 (12.5%) • Squamous cell carcionoma 3 (8.1%) 20 (95.2%) >0.05 • Other 12 (75.0%) 1 (4.8%) >0.05 Length of hospitalization 24 (66.7%) 2 12.5% >0.05 Cardiorespiratory fitness: 8 (22.2%) 2 (12.5%) 12 (60.0%) >0.05 • Peak Oxygen Uptake (mL/Kg/min) 4 (11.1%) 6 (30.0%) • Maximal ventilation (L/min) 8.2 (SD 4.5) 6.9 (SD 1.7) 2 (10.0%) >0.05 • Maximal ventilation (%pred.) >0.05 • Peak work rate (watts) 14.9 (SD 2.1) 15.3 (SD 2.3) 9.1 (SD 5.7) >0.05 • Peak work rate (%pred.) 42.6 (SD 10.6) 41.9 (SD 13.0) >0.05 • Maximal heart rate (beats/minute) 68.4 (SD 17.1) 66.5 (SD 15.7) 14.5 (SD 1.9) >0.05 • Maximal heart rate (%pred.) 88.1 (SD 22.1) 80.4 (SD 16.4) 43.2 (SD 9.1) >0.05 Respiratory muscle strength: 71.8 (SD 22.6) 74.2 (SD 24.0) 69.8 (SD 18.4) >0.05 • PImax (cmH20) 122.8 (SD 16.9) 122 (SD 16.1) 93.9 (SD 25.3) • PImax (%pred.) 78.7 (SD 10.7) 78.1 (SD 9.4) 68.6 (SD 21.8) >0.05 • PEmax (cmH20) 123 (SD 17.9) >0.05 • PEmax (%pred.) 72.8 (SD 28.0) 69.4 (SD 25.0) 79.2 (SD 11.8) >0.05 Quality of Life EORTC QOL-C30: 69.9 (SD 26.9) 67.4 (SD 28.2) • Global Quality of Life* 98.8 (SD 27.9) 90.5 (SD 21.8) 75.4 (SD 30.4) >0.05 • Physical* 62.9 (SD 17.5) 62.8 (SD 16.6) 70.0 (SD 26.4) >0.05 • Emotional* 105.2 (SD 30.8) >0.05 • Fatigue** 65.3 (SD 20.0) 62.8 (SD 20.9) 62.9 (SD 18.6) >0.05 • Dyspnea** 94.4 (SD 7.2) 91.6 (SD 8.5) >0.05 Biomarkers: 73.0 (SD 23.9) 65.6 (SD 27.1) 67.1 (SD 19.1) • IGF-I, ng/ml 31.1 (SD 20.2) 33.3 (SD 16.3) 96.5 (SD 5.4) >0.05 • IGFBP-3, µg/ml 24.7 (SD 20.4) 26.2 (SD 23.3) 78.2 (SD 20.2) >0.05 29.6 (SD 22.6) 147.5 (SD 44.3) 150.7 (SD 58.9) 23.8 (SD 18.7) 3.9 (SD 1.4) 4.02 (SD 1.1) 143.5 (SD 52.1) 3.6 (SD 1.4) COPD: chronic obstructive pulmonary disease defined as FEV1/FVC <70% and FEV1<80% of predicted value; %pred, percentage of predicted value; FEV1: forced expiratory volume in the first second; FVC: forced vital capacity; TLC, total lung capacity; DLCO: diffusing capacity for carbon monoxide; PImax: maximal inspiratory pressure; PEmax: maximal expiratory pressure; EORTC QOL-C30: European Organization for Research and Treatment of Cancer (EORTC) Quality of Life; IGF-I: insulin-like growth factor-I; IGFBP-3: IGF binding protein-3. Data are presented as mean and standard deviation (SD) or N. (%). *Higher scores indicate better functioning (scaled from 0-100); **lower scores indicate fewer symptoms (scaled from 0-100). Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 117

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, MESSAGGI-SARTOR AEROBIC EXERCISE IN NSCLC CANCER PATIENTS 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. domization was based on diagnosis other than NSCLC 21 controls. Dropouts (5 in the intervention group and (N.=5), need for chemo- or radiotherapy (N.=9), post- 8 in the control group) were mainly related to issues of operative complications (N.=2), lack of baseline CPET transportation to the program location. Dropouts did not (N.=4), refusal to participate (N.=6), and other reasons significantly differ from the 24 remaining participants in (N.=7). Finally, 37 patients were randomly assigned to any baseline measurements. All 11 participants who per- the two study groups, 16 to the exercise intervention and sisted in the exercise group completed more than 80% of Table II.—C hanges in the main outcome variable after lung resection surgery. Before surgery After surgery Mean difference P value (N.=37) (N.=37) (95% CI) <0.001 Peak oxygen uptake (mL/kg/min) 17.3 (SD 2.8) 14.9 (SD 2.1) 2.39 (1.38 to 3.41) 0.03 Peak ventilation, (L/min) 49.2 (SD 11.7) 42.6 (SD 10.6) 6.34 (2.37 to 10.3) 0. 763 Peak ventilation (%pred.) 67.5 (SD 18.7) 68.6 (SD 17.3) 1.14 (-8.8 to 6.5) 0.113 Peak work rate (watts) 92.7 (SD 20.9) 87.5 (SD 22.2) 5.26 (-1.3 to 11.8) 0.247 Peak work rate (%pred.) 74.9 (SD 24.1) 71.8 (SD 22.7) 3.03 (-2.1 to 8.2) 0.685 Maximal inspiratory pressure (cmH2O)) 67.5 (SD 24.01) 68.8 (SD 26.8) 1.35 (-8.0 to 5.3) 0.022 Maximal expiratory pressure (cmH2O) 71.3 (SD 20.2) 62.8 (SD 17.5) 8.51 (1.2 to 15.7) %pred.: percentage of predicted value; PImax: maximal inspiratory pressure; PEmax: maximal expiratory pressure. Table III.—Pre- and postexercise intervention values for the study outcomes. Before exercise intervention (T1) After exercise intervention (T2) Between-group analysis Effect size at study completion (CI 95%) Training group Control group Training group Control group Mean difference P value 1.28 (0.3 to 2.1) (N.=16) (N.=21) (N.=11) (N.=13) (CI 95%)* 0.72 (-0.1 to 1.5) Peak oxygen uptake 14.1 (SD 1.5) 16.0 (SD 2.2) 16.7 (SD 2.14) 13.7 (SD 2.5) 3.08 (-5.1 to -1.0) 0.05 0.84 (-0.2 to 1.6) 1.2 (0.2 to -2.0) (mL/kg/min) 83.0 (SD 17.2) 14.8 (-32.9 to 3.1) 0.10 0.5 (-0.3 to 1.3) 66.9 (SD 26.2) 26.0 (-52.9 to 0.8) 0.05 Peak work rate (W) 85.6 (SD 15.1) 40.1 (SD 5.8) 103.8 (SD 18.1) 88.9 (SD 22.5) 12.02 (-20.8 to -3.2) 0.01 0.5 (-0.3 to 1.3) 71.6 (SD 32.9) 97.3 (SD 35.7) 71.3 (SD 26.4) 13.0 (-36.7 to 10.7) 0.26 Peak work rate (%pred) 77.8 (SD 25.4) 51.1 (SD 9.8) 39.1 (SD 10.1) 62.2 (SD 21.6) 83.0 (SD 21.2) 70.0 (SD 29.1) 9.5 (-27.0 to 7.8) 0.26 Peak ventilation (L/min) 44.7 (SD 12.0) Maximal inspiratory 68.5 (SD 27.5) pressure (% pred.) Maximal expiratory 64.5 (SD 19.5) 78.4 (SD 17.8) 68.8 (SD 20.1) pressure (% pred.) T: assessment timing; 95% CI: 95% confidence interval; %pred.: percentage of predicted value. *P value is that observed in the Student t-test for independent samples in the postintervention between-group analysis. Table IV.—Differences in main outcome variables adjusted by baseline values for the exercise intervention group, compared to control group. Exercise parameters: B SE (B) P value (95%CI) • ∆ Peak oxygen uptake (mL/kg/min) • ∆ Peak work rate (W) 2.13 0.99 0.044 (0.06 to 4.20) • ∆ Peak work rate (% pred.) 10.9 6.78 0.124 (-3.2 to 25.1) • ∆ Ventilation/minute (L/min) 10.63 6.28 0.106 (-2.4 to 23.7) Maximal respiratory pressures: 3.65 0.016 (2.0 to 17.2) • ∆ Maximal inspiratory pressure (%) 9.64 • ∆ Maximal inspiratory pressure (cmH2O) 7.10 0.015 (4.1 to 33.7) • ∆ Maximal expiratory pressure (%) 18.96 5.11 0.016 (2.7 to 24.1) • ∆ Maximal expiratory pressure (cmH2O) 13.42 5.00 0.023 (1.9 to 22.8) Biomarkers: 12.36 7.59 0.023 (2.9 to 34.6) • ∆ IGF-I (ng/mL) 18.76 • ∆ IGFBP-3 (µg/mL) 9.35 0.110 (-3.9 to 35.4) 15.75 0.24 0.021 (0.1 to 1.12) 0.61 B: unstandardized β coefficient; SE (B): B standard error; IGF-I: insulin growth factor I; IGFBP-3: insulin growth factor binding protein 3. 118 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, AEROBIC EXERCISE IN NSCLC CANCER PATIENTS MESSAGGI-SARTOR 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 V.—H ealth-related quality of life assessed with the European Organization for Research and Treatment of Cancer Quality of Life (EORTC QL-30) questionnaire (all differences between groups were non-significant at P>0.05). Before surgery Before exercise intervention After exercise intervention EORTCQL-30 Training group Control group Training group Control group Training group Control group Global Quality of Life* (N.=16) (N.=21) (N.=11) (N.=17) (N.=10) (N.=13) Physical* Emotional* 66.6 (SD 22.0) 77.2 (SD 24.6) 62.8 (SD 20.9) 67.1 (SD 19.1) 70.8 (SD 11.9) 74.3 (SD 20.2) Fatigue ** 96.4 (SD 7.9) 98.2 (SD 3.7) 91.6 (SD 8.5) 96.5 (SD 5.4) 96.0 (SD 5.6) 93.8 (SD 9.2) Dyspnea** 72.2 (SD 24.3) 83.9 (SD 20.9) 65.5 (SD 27.1) 78.2 (SD 20.1) 76.4 (SD 24.3) 73.7 (SD 25.2) Pain** 19.2 (SD 15.4) 33.3 (SD 16.3) 29.6 (SD 22.6) 28.9 (SD 16.7) 28.2 (SD 26.7) 11.1 (SD 16.2) 9.9 (SD 16.1) 26.2 (SD 23.3) 23.8 (SD 18.7) 18.5 (SD 24.2) 12.8 (SD 16.9) 16.6 SD 23.5 5.2 (SD 12.5) 27.7 (SD 30.0) 25.4 (SD 27.2) 16.6 (SD 27.6) 19.2 (SD 17.8) 11.4 SD 21.6 *Higher scores indicate better functioning (scaled from 0-100); **lower scores indicate fewer symptoms (scaled from 0-100). the training sessions and had no adverse effects during most of the HRQoL scores improved in both groups; no the trial. significant differences between groups were observed (Ta- ble V). Demographic and clinical characteristics of the sam- ple, including the main baseline outcomes under study, In the 2-year follow-up, 6 patients had lung cancer re- are summarized in Table I. The mean age was 64.6 (SD currence, which was fatal in 1 patient in the exercise group 8.5) years and there were 26 (70.3%) men and 11 (29.7%) and 3 patients in the control group. All 6 patients had women. No significant differences between the groups lower levels of IGF-I and IGFBP-3, but the small sample were observed, with the exception of a higher proportion precluded findings of any significant differences related to of men in the control group. these outcomes. Data on changes in the main outcome variables after Discussion lung resection are described in Table II. In all patients, a significant decrease was observed after lung resection sur- In this pilot study, an 8-week exercise program of aerobic gery in VO2peak, peak ventilation and PEmax. Lung resec- exercise and high-intensity IEMT, designed for patients tion surgery led to significant reductions in exercise capac- with resectable NSCLC at 6-8 weeks after lung resection, ity in all participants: mean reduction 2.2 (SD 3.0) mL/Kg/ significantly increased exercise capacity and respiratory min (95%CI 0.97 to 3.54). muscle strength. At 2-year follow-up, the exercise inter- vention was not associated with significant reductions in Table III shows the study outcomes in the pre- and post- recurrences and death; this finding was likely due to the exercise intervention period. At study completion, the ob- small sample size. served improvement was significant in VO2peak and peak ventilation, and marginally significant in maximal peak As the preferred method to assess the mechanisms of work rate (%pred) in the exercise group. exercise tolerance, CPET provides specific information on VO2peak and submaximal cardiopulmonary responses to Table IV shows differences in the main outcome vari- exercise, both of which are useful parameters to appropri- ables, adjusted by their baseline values for the exercise in- ately tailor individualized exercise and assess response to tervention group, compared to controls. Patients in the in- training. Poor exercise capacity (defined as VO2peak <15 tervention group had a significant increase in VO2peak (2.13 mL/kg/min) has been shown to be a major determinant of mL/Kg/min [95%CI 0.06 to 4.20] and maximal respiratory postoperative morbidity and mortality following lung re- pressures (PImax: 13.45 cmH2O [95%CI 2.7 to 24.1] and section surgery.19, 20 In our pilot study, VO2peak was 16.7 PEmax: 18.76 cmH2O [95%CI 2.9 to 34.6]). No differences (SD 2.14) mL/kg/min after the exercise intervention, com- were observed in the other outcomes studied. An increase pared to 13.7 (SD 2.51) in the control group (95%CI -5.1 of 0.61 µg/mL [95%CI 0.1 to 1.12] in the serum IGFBP-3 to -1.0, P<0.05). Ten patients in the intervention group, but levels for patients in the intervention group was also ob- just 2 patients in the control group, surpassed the threshold served. of 15 mL/kg/min for VO2peak. After lung resection and before beginning the exercise To our knowledge, only three studies21-23 assessing ex- intervention, patients reported a worsening of global qual- ercise capacity after lung resection surgery have reported ity of life, emotional state, and symptoms such as fatigue, pain, and dyspnea. After completing the study protocol, Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 119

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, MESSAGGI-SARTOR AEROBIC EXERCISE IN NSCLC CANCER PATIENTS 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. VO2peak increases (between 1.1 and 3.5 mL/kg/min) after ies with a larger sample of patients are needed to confirm an exercise intervention, compared to 2.13 mL/Kg/min in our results. our study. Several considerations should be pointed out regarding differences between the study by Edvardsen et Upregulation of IGF-I and downregulation of IGFBP-3 al. reporting 3.5 mL/kg/min22 and the present study. First, and IGFBP-7 have been suggested as potential diagnostic participants in the earlier study had better cardiorespirato- and prognostic biomarkers for NSCLC.12, 28 In our pilot ry fitness at baseline (19.2 mL/Kg/min vs 15.3 mL/kg/min study, patients in the intervention group showed signifi- in our study). Second, 52% were COPD patients, com- cantly greater increases in IGFBP-3 after training, com- pared to 80% in our sample. Third, their training protocol pared to controls. In a meta-analysis that aimed to deter- involved 20 weeks of high-intensity training (at 80-95% of mine the effects of exercise in modulating insulin-like maximal heart rate on a treadmill), compared to 8 weeks growth factor system in breast cancer survivors, exercise on a cyclo-ergometer in our intervention. interventions improved the serum levels of IGF-I, IGF-II, IGFBP-1 and IGFBP-3, which showed beneficial effects Preoperative inspiratory muscle training has been shown on tumor microenvironment, breast cancer recurrence, and to significantly improve respiratory muscle function and disease-free survival rate in these patients.29 Therefore, reduce the risk of postoperative respiratory complications considering the beneficial effects of our intervention on following cardiothoracic and upper-abdominal surgery.24 prognostic factors such as exercise capacity and IGFBP-3, However, the impact of inspiratory and/or expiratory mus- we conclude that exercise interventions should be highly cle training after lung resection surgery has not been fully recommended for these patients. addressed. Our pilot study showed strong improvement in respiratory muscle strength following high-intensity There are three key points to be highlighted. To our IEMT. An earlier study found that two weeks of respira- knowledge, this is the first exercise intervention trial to tory muscle training at a target intensity of 30% of PImax limit participation to surgically treated lung cancer pa- in the immediate postoperative period has no effect on re- tients who did not undergo adjuvant chemo- or radiother- spiratory muscle strength.11 Pain, a common symptom in apy. This exercise intervention included IEMT added to the first weeks after surgery, may prevent patients from aerobic training. Finally, changes in IGFBP-3, a marker of tolerating sufficiently high workloads to induce a training physical training in healthy subjects, suggest that an ap- effect. In our study, the exercise intervention started 6-8 propriate training intensity was selected for patients in this weeks after surgery, which could explain why high-inten- pilot study. sity IEMT was well tolerated and no adverse effects were observed. Limitations of the study Mean baseline scores in our cohort on the EORTC- Our study had some limitations that must be considered. QCL30 questionnaire were very similar to previously pub- First, the small sample limits the interpretation of our re- lished scores for the European population,25 but post-oper- sults; given the interesting findings, further studies with a ative global quality-of-life scores were lower in our study. larger population should be carried out. On this regard, we Higher levels of symptoms such as dyspnea, fatigue, and have calculated the effect side for the exercise interven- pain were observed in our cohort, compared to the European tion, since effect size is one important factor in determin- population. In line with previous studies that used generic ing the statistical power of analyses;30 this point would and/or specific questionnaires to assess HRQoL,22, 26, 27 we be very useful to designing a trial with a most appropri- observed no significant changes in HRQoL-specific scores ate sample size. Second, the attrition rate was higher than following the 8-week intervention; we would note, howev- expected, as 35% of participants were lost to follow-up, er, that our pilot study was not powered to detect changes and the analysis was underpowered as a result. Although in quality of life. Other studies that used HRQoL-specific losses to follow-up are common, not only in cancer stud- questionnaires also reported no changes in global health ies,31 but also in pulmonary rehabilitation studies,32 this after exercise training, but showed improvements in dys- circumstance could reflect a poor intervention feasibility. pnea scores following the intervention. In our study, the It is worth nothing that in our study, the losses of fol- similarity in dyspnea scores observed between the control low-up were larger in the control group, mainly because and intervention groups could be related to lower-intensity patients did not return for follow-up tests. Third, no spe- training protocols, a training modality that reduces ventila- cific dyspnea scale was included as an outcome variable tor response and dyspnea in COPD patients. Further stud- in the study design; some studies have reported improve- ments in dyspnea perception after IEMT.8, 33 Fourth, our 120 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, AEROBIC EXERCISE IN NSCLC CANCER PATIENTS MESSAGGI-SARTOR 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. intervention was performed in two different hospitals by 10.  Sasso JP, Eves ND, Christensen JF, Koelwyn GJ, Scott J, Jones LW. different physiotherapists; nonetheless, there were no A framework for prescription in exercise-oncology research. J Cachexia significant differences in participant characteristics or Sarcopenia Muscle 2015;6:115–24. outcomes between the two hospitals. Finally, a moderate- 11.  Brocki BC, Andreasen JJ, Langer D, Souza DS, Westerdahl E. Post- intensity training protocol was applied, which could jus- operative inspiratory muscle training in addition to breathing exercises tify the modest improvement in VO2peak observed in our and early mobilization improves oxygenation in high-risk patients after pilot study, compared to other reports. Further research lung cancer surgery: a randomized controlled trial. Eur J Cardiothorac is required to address these limitations and explore the Surg 2016;49:1483–91. effects of alternative training protocols (e.g. intervallic 12.  Han JY, Choi BG, Choi JY, Lee SY, Ju SY. The prognostic signifi- high-intensity protocols) in patients submitted to lung re- cance of pretreatment plasma levels of insulin-like growth factor (IGF)- section surgery. 1, IGF-2, and IGF binding protein-3 in patients with advanced non-small cell lung cancer. Lung Cancer 2006;54:227–34. Conclusions 13.  Chen B, Liu S, Xu W, Wang X, Zhao W, Wu J. IGF-I and IGFBP-3 and the risk of lung cancer: a meta-analysis based on nested case-control In conclusion, an 8-week exercise intervention plus high- studies. J Exp Clin Cancer Res 2009;28:89. intensity IEMT increased exercise capacity, respiratory 14.  Elloumi M, El Elj N, Zaouali M, Maso F, Filaire E, Tabka Z, et al. muscle strength and serum IGFBP-3 levels in this pilot IGFBP-3, a sensitive marker of physical training and overtraining. Br J study. No changes were observed in the other outcomes Sports Med 2005;39:604–10. measured. Considering the positive prognostic value of 15.  Turner L, Shamseer L, Altman DG, Weeks L, Peters J, Ko- VO2peak in patients with lung cancer, structured and super- ber T, et al. Consolidated standards of reporting trials (CONSORT) vised exercise interventions should be recommended for and the completeness of reporting of randomised controlled trials this population. (RCTs) published in medical journals. Cochrane Database Syst Rev 2012;11:MR000030. References 16.  Black LF, Hyatt RE. Maximal respiratory pressures: normal values and relationship to age and sex. Am Rev Respir Dis 1969;99:696–702. 1.  Rochester CL, Fairburn C, Crouch RH. Pulmonary rehabilitation for 17.  Morales P, Sanchis J, Cordero PJ, Díez JL. [Maximum static re- respiratory disorders other than chronic obstructive pulmonary disease. spiratory pressures in adults. The reference values for a Mediterranean Clin Chest Med 2014;35:369–89. Caucasian population]. Arch Bronconeumol 1997;33:213–9. 2.  Nagamatsu Y, Maeshiro K, Kimura NY, Nishi T, Shima I, Yamana H, 18.  Aaronson NK, Ahmedzai S, Bergman B, Bullinger M, Cull A, Duez et al. Long-term recovery of exercise capacity and pulmonary function NJ, et al. The European Organization for Research and Treatment of after lobectomy. J Thorac Cardiovasc Surg 2007;134:1273–8. Cancer QLQ-C30: a quality-of-life instrument for use in international 3.  Jack S, West MA, Raw D, Marwood S, Ambler G, Cope TM, et al. clinical trials in oncology. J Natl Cancer Inst 1993;85:365–76. The effect of neoadjuvant chemotherapy on physical fitness and survival 19.  Win T, Jackson A, Sharples L, Groves AM, Wells FC, Ritchie AJ, in patients undergoing oesophagogastric cancer surgery. Eur J Surg On- et al. Cardiopulmonary exercise tests and lung cancer surgical outcome. col 2014;40:1313–20. Chest 2005;127:1159–65. 4.  Handy JR Jr, Asaph JW, Skokan L, Reed CE, Koh S, Brooks G, et 20.  Brunelli A, Salati M. Preoperative evaluation of lung cancer: pre- al. What happens to patients undergoing lung cancer surgery? Outcomes dicting the impact of surgery on physiology and quality of life. Curr and quality of life before and after surgery. Chest 2002;122:21–30. Opin Pulm Med 2008;14:275–81. 5.  Ni HJ, Pudasaini B, Yuan XT, Li HF, Shi L, Yuan P. Exercise train- 21.  Jones LW, Eves ND, Peterson BL, Garst J, Crawford J, West MJ, et ing for patients pre- and postsurgically treated for non-small cell lung al. Safety and feasibility of aerobic training on cardiopulmonary func- cancer: A Systematic Review and Meta-analysis. Integr Cancer Ther tion and quality of life in postsurgical nonsmall cell lung cancer patients: 2017;16:63–73. a pilot study. Cancer 2008;113:3430–9. 6.  Cesario A, Ferri L, Galetta D, Pasqua F, Bonassi S, Clini E, et al. 22.  Edvardsen E, Skjønsberg OH, Holme I, Nordsletten L, Borchsenius Post-operative respiratory rehabilitation after lung resection for non- F, Anderssen SA. High-intensity training following lung cancer surgery: small cell lung cancer. Lung Cancer 2007;57:175–80. a randomised controlled trial. Thorax 2015;70:244–50. 7.  Spruit MA, Janssen PP, Willemsen SC, Hochstenbag MM, Wouters 23.  Cavalheri V, Jenkins S, Cecins N, Gain K, Phillips MJ, Sanders LH, EF. Exercise capacity before and after an 8-week multidisciplinary inpa- et al. Exercise training for people following curative intent treatment for tient rehabilitation program in lung cancer patients: a pilot study. Lung non-small cell lung cancer: a randomized controlled trial. Braz J Phys Cancer 2006;52:257–60. Ther 2017;21:58–68. 8.  Gosselink R, De Vos J, van den Heuvel SP, Segers J, Decramer 24.  Mans CM, Reeve JC, Elkins MR. Postoperative outcomes follow- M, Kwakkel G. Impact of inspiratory muscle training in patients with ing preoperative inspiratory muscle training in patients undergoing car- COPD: what is the evidence? Eur Respir J 2011;37:416–25. diothoracic or upper abdominal surgery: a systematic review and meta 9.  Crandall K, Maguire R, Campbell A, Kearney N. Exercise interven- analysis. Clin Rehabil 2015;29:426–38. tion for patients surgically treated for Non-Small Cell Lung Cancer 25.  Hinz A, Singer S, Brähler E. European reference values for the qual- (NSCLC): a systematic review. Surg Oncol 2014;23:17–30. ity of life questionnaire EORTC QLQ-C30: results of a German inves- tigation and a summarizing analysis of six European general population normative studies. Acta Oncol 2014;53:958–65. 26.  Cavalheri V, Tahirah F, Nonoyama M, Jenkins S, Hill K. Ex- ercise training for people following lung resection for non-small cell lung cancer - a Cochrane systematic review. Cancer Treat Rev 2014;40:585–94. 27.  Arbane G, Tropman D, Jackson D, Garrod R. Evaluation of an early exercise intervention after thoracotomy for non-small cell lung cancer (NSCLC), effects on quality of life, muscle strength and exercise toler- ance: randomised controlled trial. Lung Cancer 2011;71:229–34. Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 121

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, MESSAGGI-SARTOR AEROBIC EXERCISE IN NSCLC CANCER PATIENTS 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. 28.  Wang Z, Wang Z, Liang Z, Liu J, Shi W, Bai P, et al. Expression 31.  Kenny C, Gilheaney Ó, Walsh D, Regan J. Oropharyngeal dyspha- and clinical significance of IGF-1, IGFBP-3, and IGFBP-7 in serum and gia evaluation tools in adult with solid malignancies outside the head and lung cancer tissues from patients with non-small cell lung cancer. Onco- neck and upper GI tract: a systematic review. Dysphagia 2018;33:303–20. Targets Ther 2013;6:1437–44. [[Epub ahead of print]] 32.  Williams MT, Lewis LK, McKeough Z, Holland AE, Lee A, Mc- 29.  Meneses-Echávez JF, Jiménez EG, Río-Valle JS, Correa-Bautista JE, Namara R, et al. Reporting of exercise attendance rates for people with Izquierdo M, Ramírez-Vélez R. The insulin-like growth factor system is chronic obstructive pulmonary disease: a systematic review. Respirology modulated by exercise in breast cancer survivors: a systematic review and 2014;19:30–7. meta-analysis. BMC Cancer 2016;16:682. 33.  Marco E, Ramírez-Sarmiento AL, Coloma A, Sartor M, Comin-Co- let J, Vila J, et al. High-intensity vs. sham inspiratory muscle training in 30.  Weisz JR, Doss AJ, Hawley KM. Youth psychotherapy outcome re- patients with chronic heart failure: a prospective randomized trial. Eur J search: a review and critique of the evidence base. Annu Rev Psychol Heart Fail 2013;15:892–901. 2005;56:337–63. Conflicts of interests.—Authors state we have full control of all primary data and agree to allow the journal to review them if requested. Funding.—Supported by a grant from Societat Catalana de Pneumologia 2012. Authors’ contributions.—Monique Messaggi-Sartor: data collection, statistical analysis, interpretation of the data, drafting the manuscript. Ester Marco: de- sign and conceptualization of the study, analysis and interpretation of the data, revision of the manuscript for intellectual content. Elisabeth Martínez-Tellez, and Alberto Rodríguez-Fuster, contributed equally in recruitment of patients and revision of the manuscript. Carolina Palomares, and Sandra Chiarella: data collection, analysis and interpretation. Josep M. Muniesa, Mauricio Orozco-Levi, and Esther Barreiro, contributed equally in revision of the manuscript for intellectual contents. Maria Rosa Güell: design and conceptualization of the study, acquisition of funding, analysis and interpretation of the data, revision of the manuscript for intellectual content. Congresses.—Presented at the European Respiratory Society (ERS) International Congress. Amsterdam (The Netherlands), September 26-30, 2015. Acknowledgements.—The authors thank the study participants and other contributors, including but not limited to Sergi Mojal and Joan Vila for statistical support and Elaine Lilly, for English revision. Article first published online: July 6, 2018. - Manuscript accepted: July 5, 2018. - Manuscript revised: May 29, 2018. - Manuscript received: January 5, 2018. 122 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):123-30 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/S1973-9087.18.04904-3 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 Evaluation of hand function in patients with unilateral cerebral palsy who underwent multilevel functional surgery: a retrospective observational study Giovanna CRISTELLA 1 *, Maria C. FILIPPI 2, Maurizio MORI 2, Silvia ALBORESI 2, Adriano FERRARI 2 1Unit of Children Rehabilitation, IRCCS Fondazione Don Carlo Gnocchi, Florence, Italy; 2Unit of Children Rehabilitation, Arcispedale Santa Maria Nuova, Reggio Emilia, Italy *Corresponding author: Giovanna Cristella, Unit of Children Rehabilitation, IRCCS Fondazione Don Carlo Gnocchi, via di Scandicci 267, 50143 Florence, Italy. E-mail: [email protected] ABSTRACT BACKGROUND: Hemiplegia is the most common form of cerebral palsy. Upper limb is generally more affected than lower one. Indeed, hemiplegic children can spontaneously acquire standing and walking ability, while manipulation remains uncertain, with severe limitations in activity and participation, which define a child’s functional status (International Classification of Functioning [ICF]). Several non-surgical tools are currently available to approach upper limb impairments. Studies regarding upper limb multilevel surgery in Hemiplegic Cerebral Palsy are relatively few and inhomogeneous. AIM: The aim of this study is to propose a surgical approach based on upper limb functional level and manipulation strategy and establish wheth- er multilevel surgery can improve segmental alignment, performance and capacity, that ICF defines as activities and participation qualifiers. DESIGN: This study is an observational retrospective study. SETTING: This study involves patients who referred to the Unit of Children Rehabilitation of S. Maria Nuova Institute for Research and Care, in Reggio Emilia (Italy), over a four-year period. POPULATION: Children affected by hemiplegic cerebral palsy who underwent upper limb multilevel surgery. METHODS: For each patient, we previously defined functional use of affected upper limb applying the House classification and the Ferrari one of manipulation pattern. Patients are divided into three groups: synergic hand (House 4, 5), imprisoned hand (House 3), excluded hand (House 0). We recorded goals achievement through Goal Attainment Scale and unimanual and bimanual abilities through Melbourne Assessment of Unilateral Upper Limb Function and through Assisting Hand Assessment respectively. RESULTS: We recorded 16 upper limb multilevel surgical interventions in 13 children and report their results. CONCLUSIONS: This study suggests that surgery can induce a segmental and/or aesthetic and/or a functional change depending on manipula- tion pattern. It also underlines the importance to analyze results in term of spontaneous manipulation abilities and daily use. CLINICAL REHABILITATION IMPACT: This study provides a preliminary guide to plan surgery in relation to segmental deformities and overall manipulation pattern and describes their feasible improvement measures. It also suggests the most useful tools to record goal achieve- ments in modifying manipulation function. Further controlled, randomized and prospective studies are required to support this idea. (Cite this article as: Cristella G, Filippi MC, Mori M, Alboresi S, Ferrari A. Evaluation of hand function in patients with unilateral cerebral palsy who underwent multilevel functional surgery: a retrospective observational study. Eur J Phys Rehabil Med 2019;55:123-30. DOI: 10.23736/S1973- 9087.18.04904-3) Key words: Upper extremity - Hand deformities - Hemiplegia. Cerebral palsy (CP) is the principle cause of childhood CP in term infants (more than 50%) and the second one in physical disability in industrialized societies (1/500 preterm ones (20%).3 live births).1 Recently, Himmelman et al. recorded a turn- around in term of prevalence of CP forms: hemiplegic The Surveillance of Cerebral Palsy in Europe terms cerebral palsy is the most frequent (38%), followed by hemiplegic cerebral palsy as “unilateral cerebral palsy” diplegia (32%).2 Hemiplegia is the most common form of (UCP) and suggests the following diagnostic criteria: in- volvement of limbs on one side of the body and at least Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 123

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, CRISTELLA HAND FUNCTION IN UNILATERAL CEREBRAL PALSY 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. two among the following: 1) abnormal pattern of posture dynamic positional analysis, Pediatric Quality of Life In- and movement; 2) increased tone (not necessarily con- ventory (module domain of movement) and Canadian Oc- stant); 3) pathological reflexes.4 cupational Performance Measure score for satisfaction.11 This research shows surgery benefit in UL function, but The main clinical characteristic of hemiplegia is the re- mixes three different kinds of rehabilitation instruments duction of motor repertoire on the affected side in terms of that, according to us, have different indications: physio- modules (meant as the elementary components of move- therapy to modify function, botulinum toxin injection to ment the child is provided with), combinations (possibility reduce spasticity and surgery to reduce/modified muscle to organize the individual modules into different patterns retraction, bone and joint deformities. Smitherman et al. according to space relations), and sequences (ability to as- compare functional outcomes following UL surgery in a semble the individual modules according to different time retrospective case control series in hemiplegic CP, show- relations). These early clinical signs allow a prompt diag- ing a significant improvement in dynamic segmental align- nosis of UCP.5 Other clinical signs often reported in UCP ment and in spontaneous use.12 are: the presence of associated movements, sensory and perceptive defects, attention disorders, alteration of mus- The aim of this study is to establish whether multilev- cle tone, muscle retractions and (early or late) bone growth el UL surgery in children with UCP improves segmental alterations, elements more or less frequent, according to alignment, UL function (capacity) and UL daily use (per- the different clinical forms. formance) when indication for surgical treatment is per- formed in relation to the assessed functional level and not In UCP, the upper limb (UL) is generally more affected only to segmental deformity. Specific and individual goals than the lower one. Unimanual abilities of the affected were recorded through Goal Attainment Scale (GAS). UL and manipulation strategies are heterogeneous; con- When possible, we record unimanual and bimanual abili- sequently, activity and participation could be severely ties through Melbourne Assessment of Unilateral Upper impaired.6 For this reason, rehabilitation should focus on Limb (MUUL) and Assisting Hand Assessment (AHA) improving UL competence in executing a task or an action respectively. (capacity) and in daily activity (performance). UL spastic- ity and/or weakness, muscle contracture and/or retraction, Materials and methods limitation in joint range of motion, forearm, wrist and fin- ger deformation, poor dexterity and motor control, lack of Study design sensation and perception in affected limb, cause functional impairment that can be amplified by learned non-use, and This is a retrospective observational study conducted in auxiliary grips.7 the Unit of Children Rehabilitation of IRCSS S. Maria Nuova Hospital in Reggio Emilia, Italy, along a four years Several non-surgical tools are currently available; a re- period. Approval was obtained from the Research Ethics cent systematic review describes their efficacy.8 Committee. Studies about functional outcome after surgery in UCP Participants are relatively few and inhomogeneous. In a recent review, Van Muster indicates that many of them consist of case se- The inclusion criteria were: Clinical (UCP as defined by ries providing a low level of evidence. These studies sug- Surveillance of Cerebral Palsy in Europe, previously de- gest a positive effect on hand position, manipulation strat- scribed), and instrumental (brain MRI) diagnosis of UCP; egy, grip repertoire and spontaneous use, but it remains history of multilevel upper limb surgery performed at unclear whether surgery influences daily activity.9 S. Maria Nuova Hospital (Reggio Emilia, Italy); video- recording about spontaneous manipulation activity before Van Heest explores change in functional use by House and after multilevel surgery, to establish whether surgical classification in a heterogeneous group (spastic/athetoid goals were achieved. Patients with previous diagnosis of CP, quadriplegia/hemiplegia/triplegia) over a 25-year pe- mental retardation and/or behavioral disorder were ex- riod; this author reports a functional improvement for all cluded. patients and identifies good voluntary control as a posi- tive prediction factor.10 In a recent study, the same author While assessment and surgical interventions took place compares efficacy of tendon transfer surgery versus botu- in Unit of Children Rehabilitation, at S. Maria Nuova Hos- linum toxin injection and ongoing therapy and records an pital, the following rehabilitation program was delivered improvement at twelve months of follow-up for surgical group in Shriners Hospital Upper Extremity Evaluation, 124 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, HAND FUNCTION IN UNILATERAL CEREBRAL PALSY CRISTELLA 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 I.—H ouse Functional Classification. shown in the table, Synergic, Imprisoned, and Excluded hands have surgical indications. Grade Designation Activity level It is possible to establish a correspondence between 0 None None Ferrari manipulation pattern classification and House clas- 1 Poor passive assist Uses as stabilizing weight only sification: Integrated hand corresponds to House level 8, 2 Fair passive assist Can hold on to object placed in Semi-Functional hand to House level 7, Synergic hand to House levels 4 through 6, Imprisoned hand to House levels 3 Good passive assist hand 1 to 3, Excluded hand to House level 0. Can hold on to object and stabilize 4 Poor active assist Outcome it for use by the other hand 5 Fair active assist Can actively grasp object and hold Outcomes were recorded using GAS. GAS is a method of measuring individual progress towards specific, mea- 6 Good active assist it weakly surable, acceptable, relevant and time-related goals. It Can actively grasp object and is a 5-point scale, where 0 represents the expected level 7 Spontaneous use, partial of success, +1 and +2 the achievement of more than stabilize it well the expected, -1 and -2 a worse result than expected. 8 Spontaneous use, complete Can actively grasp object and then Each goal has a weight defined as importance × dif- ficulty.15, 16 manipulate it against other hand Can perform bimanual activities By assigning 0, GAS assumes the value 50. By assign- ing +1 and +2 it assumes values progressively greater than easily and occasionally uses the 50, while assigning -1 and -2 it assumes values progres- hand spontaneously sively less than 50.17 Uses hand completely independently without reference For each patient, we defined three goals: GASP (goal to the other hand concerning postural and segmental alignment), GASF (goal concerning spontaneous functional use of affected at the rehabilitation centers, where each child came from. UL) and GASA (goal describing patient satisfaction in UL multilevel surgery was followed by an individualized term of autonomy in daily activities). They are summa- physiotherapy treatment and/or personalized orthosis. rized in GASTOT, which expresses a weighted average of all three (in other words, an average that considers the For each patient, before surgery (T0), we defined ma- weight of each goal). nipulation ability referring to House classification, report- ed in Table I (House Functional Classification System).13 GAS was defined considering not only limitation in In recent years, we attempted to validate a classification of manipulation in UCP. It describes five patterns of ma- nipulation by analyzing hand kinematic profile and func- tional use as per Ferrari et al.:14 Integrated, Semi-Func- tional, Synergic, Imprisoned, Excluded. The main characteristics of each class and correspond- ing clinical approach are summarized in Table II. As Table II.—F errari manipulation pattern. Treatment Integrated Semi-functional Synergic Imprisoned Excluded Main core Subterminal/ Subterminal lateral Stereotypically Indirect grasping Functionally ineffective Physiotherapy terminal pinch, possible selective pinch with basically expressed grasping (passive loading) or negligible grasping finger movements, mastery of intrinsic adduced thumb, and release within Imprisoned thumb or motricity for manual exploration, Scarce/absent intrinsic flexion and extension positioned underneath Useful in early age motricity synergies. Active palm loading of object Useful through Useful for both affected Limited effectiveness Useless perceptive exercises Rarely required and unaffected hand Occasionally required Wrist splint are Botulinum toxin Rarely required Required to reduced Useful to inhibit sometimes required Occasionally nocturnal splint forearm pronation and spasticity hand Aesthetic indication No indication wrist ulnar deviation Orthoses Useless Occasionally thumb and Occasionally nocturnal wrist dynamic splint splints are useful are needed Multilevel upper limb Contraindicated Useful for muscle Useful for muscle surgery retraction retraction and bone deformity Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 125

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, CRISTELLA HAND FUNCTION IN UNILATERAL CEREBRAL PALSY 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. ROM and deformation, but also House level and Ferrari wrist deformities: patient #6 was subjected to epitrochlear manipulation pattern, and specific patient demand. muscle tenotomy and ulnar flexor carpi release, patients #7 and #11 to transfer of ulnar flexor carpi to extensor carpi We estimated goal achievement by calculating GAS af- radial brevis, patient #9 to carpus arthrodesis and ulnar ter three (T1) and twelve (T2) months. For some patients flexor carpi tenotomy first and to plate removal and flexor we have intermediate data. We compare mean values of pollicis brevis myotomy in a second time. GASP, GASF, GAS A and GASTOT. Imprisoned hand group As a secondary outcome, when possible, some patients were assessed through MUUL and AHA. MUUL18 and Patient #1 underwent surgical intervention in two time: AHA19 are scales frequently used in rehabilitation: the first first, flexor pollicis brevis, ulnar flexor carpi, radial flexor measures unilateral upper limb function while the second carpi, brachialis biceps release together with extensor bre- one measures bimanual activities. vis carpi radialis retention and in a second time, because of relapse in term of segmental alignment, adhesion release, Results adductor pollicis tenotomy and transfer of ulnar flexor carpi and radial flexor carpi to extensor brevis carpi ra- We recorded 16 UL multilevel surgical interventions in 13 dialis. Patient #8 required surgical approach to wrist flex- children with UCP. One patient underwent three surgical ion deformity through ulnar flexor carpi and radial flexor interventions, but we only have video-recording of two of carpi release, together with pronator teres and finger flexor them; another patient underwent two surgical interven- digitorum profundus release and palmar gracilis tenotomy. tions. Two patients were excluded due to mental retarda- The other patients with imprisoned hand were subjected tion, which required a different rehabilitation approach. to palmar gracilis tenotomy (patient #5) and to brachialis Children came from different Italian Rehabilitation cen- biceps fasciotomy (patient #10). ters. Sample characteristics are listed in Table III. Excluded hand group House level distribution (level: percentage): 8:0%; 7:0%, 6:0%, 5:15%, 4:31%, 3:39%, 2:0%, 1:0%, 0:15%. Patients #2 and #3 were respectively subjected to brachia- lis biceps fasciotomy, pronator teres and superficial finger Ferrari manipulation pattern (pattern: percentage): In- flexor release, transfer of ulnar flexor carpi to extensor car- tegrated hand: 0%, Semi-Functional hand: 0%, Synergic pi radial brevis, extensor ulnaris carpi brevis tenotomy the hand: 46%, Imprisoned hand: 36%, Excluded hand: 18%. first and metacarpophalangeal joint arthrodesis, brachialis biceps release, ulnar flexor carpi, flexor pollicis brevis, To follow, multilevel surgical interventions are briefly pronator teres release, palmar gracilis release the second. described group by group. We can notice that patient with synergic hand were sub- Synergic hand group jected to wrist and forearm surgery, but no surgery was performed for fingers, except for thumb. Imprisoned hand Patient #4 was subjected to pronator teres release. The oth- patients required surgical approach to wrist or thumb or er patients with synergic hand (patients #6, #7, #9.1, #9.2, finger deformities with a case of relapse when surgery is and #11) received a more substantial approach to their more conservative overall for thumb and wrist. Both pa- tients with excluded hand required intervention to obtain Table III.—Sample features. wrist alignment and to reduce forearm pronation. Patient # Age at Age at House Ferrari manipulation After surgery, physiotherapy and splinting were defined inclusion, years surgery, years grade pattern considering House level, Ferrari manipulation pattern, spasticity, weakness and surgical techniques. This consti- 1.1 17 13 3 Imprisoned tuted a first phase, based on daily specific stretching ma- 1.2 17 15 3 Imprisoned neuvers and a second phase based on therapeutic activities 2 19 16 0 Excluded consisting of a series of goal-directed actions, spontane- 3 16 13 0 Excluded ously and voluntarily executed by patients, under thera- 4 15 12 4 pist guide, in order to achieve functional solution in ma- 5 21 18 3 Synergic nipulation activity. Exercises take into account each child 6 16 14 4 Imprisoned 7 14 13 5 8 15 13 3 Synergic 9.1 24 19 4 Synergic 9.2 24 21 4 Imprisoned 10 11 10 3 Synergic 11 12 11 5 Synergic Imprisoned Synergic 126 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, HAND FUNCTION IN UNILATERAL CEREBRAL PALSY CRISTELLA 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 IV.—Data presented by group. Patient # T0 GASP GASF GASA T1 MUUL AHA T2 AHA MUUL AHA 53 GASP GASF GASA GASTOT MUUL 43.59 50.00 GASTOT 63.87% Synergic hand group (House 4, 5) 53.67 50.00 50.00 82.35% 79 50.00 62.40 43.80 52.1 4 42 62.35 60.87 31.64 47.84 6 31.34% 68.45 50.00 50.00 55.83 48.36% 76.52 50.00 61.64 63.27 7 73.64% 50.00 60.08 50.00 64.31 53.28% 50.00 60.00 50.00 53.33 9.1 70.00 73.27 50.00 50.00 9.2 62.27 56.30 46.94 57.55 63.57 57.63 61.74 62.27 73 11 70 61.12 64.91 60.02 57.50 54.29 57.74 Mean 63.39 43.20 56.74 Imprisoned hand group (House 3) 55.10 68.70 62.40 36.40 50.00 50.00 49.72 46.72% 1.1 41.8% 64.41 28.67 50.70 53.44 69.39 68.70 62.40 66.83 68.91% 1.2 46.72% 54.56 57.36 54.56 65.04 50.00 42.24 51.41 47.65 5 60.75% 54.91 56.20 62.40 43.93 74.54 47.55 54.56 58.61 8 58.26 61.99 54.65 55.55 38.06 31.39 37.60 35.75 10 57.45 58.94 53.68 47.97 51.19 51.71 Mean 43.80 39.43 55.83 Excluded hand group (House 0) 52.10 70.93 58.64 48.80 48.30 51.75 61.17 2 38.52% 50.75 57.37 49.04 47.77 44.78 43.96 70.17 49.56 3 51.43 59.79 46.79 46.13 60.96 55.37 Mean 53.78 coping solutions. When suitable, such as in case of tendon AHA variation score can be considered significant when transfer, arthrodesis and muscle weakness, in early stages, equal or greater than 4 in raw score.19 At T1, both patients continuous positional splint use was recommended. recorded a significant variation. Patient #11 repeated eval- uation in T2, recording a non-significant improvement. Physiotherapeutic program was planned by an experi- enced pediatric research physiotherapist and delivered to Into imprisoned hand group, postural goals were case physiotherapist. achieved for all patient in T1 and maintained at follow-up, with a GASP mean value greater than 50 at T1 and T2. The results for each group are provided in Table IV. Patient #1 needed a second-time surgery (#1.2) because of All patient with synergic hand achieved all goals (GASP, loss of segmental alignment after first surgical time (#1.1), GASF, GASA) in T1 (except patient #4) and maintained while maintaining good functional results. In other words, the results at follow up (T2) in fact GAS mean values are patient #1 required two surgical intervention to achieve greater than 50 for all GAS at T1 and T2, except for GASA segmental and functional goals. GASF mean value is at T1, influenced by negative result of patient #7, not con- very positive in T1 (overall because of the great results firmed at follow up. Patient #4’s negative result is non- of patient #1) but this result is not confirmed at T2, with acceptance of hand appearance, resulting in a very poor a negative GASF mean value. All patients’ expectation spontaneous hand use and compliance to physiotherapeu- was achieved at T2, except for patient #10 (this can be ex- tic program, instead of a good motor repertoire. In other plained by relapse of segmental deformities), with GASA words, patients’ expectations were satisfied, together with mean values greater than 50 at T1 and T2. GASTOT mean both better segmental alignment and improvement of ma- values confirm that preoperative goals were achieved. nipulation abilities. GASTOT mean values confirm that preoperative goals were achieved. Patient #1 was also evaluated by MUUL. After the first surgical session, the patient showed a positive but insig- We have data about MUUL at T0 and T1 for patients #6 nificant difference in T1 and T2. After second surgical and #7 but not about MUUL at T2. MUUL variation score session (#1.2), patient recorded a positive but not signifi- is considered significant when equal to 12% or more.18 An cant difference after three months; comparing this to the intermediate MUUL was recorded for patient 7 after six MUUL score in T1, the difference was 11.48%, very close months from surgery, its score is 92.44%. We recorded to 12% (limit of statistical significance). Patient #5 record- a positive but not significant variation for patient 7 after ed a positive score in MUUL in an intermediate time — six three months and a significant variation after six months. A months after surgery — (not shown in Table IV) confirmed significant modification was recorded in MUUL for patient at T2, but with a not significant score. #6 at T1. Patients #4 and #11 were evaluated with AHA. Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 127

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, CRISTELLA HAND FUNCTION IN UNILATERAL CEREBRAL PALSY 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. About excluded hand group, both patients satisfied seg- GASA generally follows the positive trend of GASP and mental alignment goals in T1, but both recorded a relapse GASF (except for patient #4, who always refers trouble in in T2. Positive GASP and GASF mean value at T1 were accepting his disability). not confirmed at follow-up. GASA follows an opposite trend, with a very positive mean value at T2. Overall re- Into imprisoned hand group we describe high risk or re- sults (GASTOT) were satisfactory and were influenced currence and we obtain the best results in patient who un- overall by GASA positive values. Patient #2 was evalu- derwent surgery in two times. For this reason, it seems rea- ated by MUUL, recording a positive but not significant sonable to propose surgery as later as possible in order to improvement in T2. plan a one-time surgery by a “more aggressive” technique. Discussion Looking at patient #10, he records negative results at T2 in all three GAS and so in GASTOT. These negative Based on the results, some differences can be observed be- results could be explained by considering the “minimal” tween the groups. surgical approach that allowed a precocious recurrence. In other words, to prevent recurrence, tenotomy seems to be As shown, into synergic hand group, multilevel surgery better than fasciotomy, wrist transfer (for example trans- gives a gain in term of segmental alignment and manipula- ferring ulnar flexor carpi to extensor carpi radial brevis) tion ability. Fulcrum joint in synergic strategy manipula- seems to be better than flexor lengthening to obtain wrist tion seems to be wrist, so it’s important to obtain an opti- tendon-suspension. For example, patient #1.2 records a mal wrist alignment. In these patients, pattern manipula- small but gradual gain suggesting efficacy of transferring tion contains a substantial modifiability and surgery allows ulnar flexor carpi and radial flexor carpi to extensor carpi to express it. According to us, the better surgical approach radial brevis in getting wrist tendon- suspension. Looking to obtain a good wrist alignment and to express functional at surgery plan, we can suppose that thumb and wrist are modifiability is transferring ulnar flexor carpi to extensor the fulcrum of imprisoned hand manipulation ability. In carpi radial brevis (arthrodesis hinders wrist motion and fact, when thumb is imprisoned into palm, it induces grasp can make synergic strategy difficult). In fact, the results reflex and limits the possibility of passive hand grip. Good obtained in two patients with synergic hand (patients #7 wrist alignment puts fingers flexor in a more advantageous and #11) are remarkable. Probably this transfer gives a position to grasp. Both Patient 8 and 10 recorded a nega- better wrist alignment so that fingers flexor muscles are tive trend in GASF; this suggest that, even if it is possible in a favorable position to take advantage in synergic strat- to obtain a functional change in imprisoned hand, it is hard egy; furthermore, it breaks pathological synergies by put- to stabilize the results, suggesting changeability in seg- ting a muscle to function with its original antagonist and mental alignment but not in functional spontaneous use. maybe induces cortical reorganization by accessing to In particular, the outcome of patient #8, whose negative motor modules previously unused and reorganizing them trend in GASF is opposed to positive trend in GASP con- in more complex combinations and sequences. In other firms this: even if multilevel surgical approach can modify words, if there are grounds (modifiability, learning ability local alignment, the lack of motor repertoire, typical of and motivation), it allows the maximum expression of ma- imprisoned hand, severely conditions functional progno- nipulation function for each patient. This happens regard- sis. In other words, obtaining a good segmental alignment less of surgical technique, in fact in these two patients we does not ensure obtaining a manipulation ability gain be- used two different techniques: Carlson and Green transfer. cause of the narrow space of modifiability. In patient #1, Realistically, the grater is patient age, the less is function functional gain was probably due to adductor pollicis te- modifiability. It could be useful to carry out transfer in ear- notomy, which reduces self-stimulation of grasping, and lier age (our patients are #11 and #13) but there are two to ulnar flexor carpi and radial flexor carpi transfer onto main risks: manipulation mistakes are not yet permanent, extensor brevis carpi radialis because it puts finger flexor so results cannot be predictable, technique difficulties re- muscles in a more advantageous position to execute grasp. lated to small tendons and co-contraction in case of high We have also to underline that this patient showed a very level of spasticity. high level of compliance and motivation in rehabilitation program; those give a great guarantee of functional gain. Transfer is useful to modify function in patients with synergic hand but not in patient with imprisoned and ex- About GASA, its trend generally follows GASP: im- cluded hand because of their lower modifiability. prisoned hand has a strong visual impact and patients seem to be satisfied of getting it better. This remind us firstly 128 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, HAND FUNCTION IN UNILATERAL CEREBRAL PALSY CRISTELLA 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. the strong imprisoned hand aesthetic impact on patients, specialized team and patient motivation should be always probably more than functional ones (because of hyper- considered before surgery. specialization of unaffected hand), and secondly the great awareness of its poor modifiability. So, changing hand ap- Some observations can be made about the evaluation in- pearance could be a good reason to surgery. strument. We can suppose some differences between GAS, MUUL and AHA in term of recording changes following About excluded hand group, differently from the oth- treatment. Six patients were evaluated by MUUL. One of ers, we record no possibility to induce a functional long- them records a significant change in MUUL score, three term gain. GASF results confirm that is not possible to of them a better but no significant change, one of them a modify functional use because of poor motor and sensi- worse score (one patient did not perfume MUUL at follow- tive repertoire. The recurrence of segmental deformities up) whereas GAS values show the achievement of surgical suggests a more extended use of positional splint since goals (GASTOT) for all of them (except for patient #5 who early age and a more aggressive surgical approach (i.e. came back to pre-surgical situation but did not have a wors- wrist arthrodesis). Looking at GASA, we can say that pa- ening; patient #1.1 achieves goals by two-time surgery). In tient expectations are satisfied probably because of lower other words, this study suggests that MUUL has low speci- functional expectations and a greater awareness of narrow ficity in recording manipulation abilities change probably modifiability, compared to patients belonging to the other because it is made up of mono-manual tasks in which unaf- two groups. These patients indeed early develop very ef- fected UL is never involved. This manipulation strategy is fective strategies of manipulation by unaffected UL hyper- not congruent with hemiplegic child strategy: UCP child specialization. employs hemiplegic hand only in supporting other hand’s activity or when dominant hand is not available for manip- Comparing these groups, we can underline the positive ulation. Only two patients performed AHA; both recorded results of GASP mean values after three months and at fol- significant results in T1, and positive but insignificant result low-up (T2), except for excluded hand patient who record- in T2. AHA seems to be a more specific scale because it ed relapse and we can observe the grater results in synergic measures how effectively these children use the affected hand. Comparing GASF mean values between group, we hand in bimanual performance; indeed, it provides bimanu- can conclude that significant change at follow up is pos- al tasks and probes also affected UL initiative of use and co- sible only in synergic hand group. GASA mean values re- ordination. However, while the MUUL scale evaluates the cord negative results at T1 for synergic and excluded hand, capacity (the person’s ability to execute a task or an action maybe due to difficulties related to post-operative time; at on the highest probable level of functioning that a person T2, we record GASA mean values greater than 50 in all may reach in a standardized environment), AHA measures groups, with the best result in the excluded hand group, performance (the person’s ability to execute a task or an probably because of the great awareness of low modifi- action in a real-life environment, e.g. semi-structured play ability, as previously described. GASTOT mean records session). AHA tasks are always performed after examiner positive values at T1 and T2 for all groups, confirming request so do not describe spontaneous UL use. Hence, new goals achievement when they are established according to scales, based on spontaneous play observation, is needed. House level and Ferrari manipulation pattern. Videorecording spontaneous manipulation activities is a good instrument to evaluate impact on daily living. In conclusion, we can suppose that surgery can posi- tively modify segmental alignment and spontaneous hand GAS is a good evaluation instrument because it allows function in synergic hand group, sometimes leading to rehabilitation team to establish for each patient tailored unexpected results; surgery can also favorably influence goals, but it requires a proper training to avoid over-esti- imprisoned hand segmental alignment, carrying some- mation and under-estimation mistakes. Training on House times a narrow functional gain. Excluded hand group has and Ferrari classification is important to establish ma- no space of functional modifiability and has a high risk nipulation ability modifiability and the better instrument of segmental deformities recurrence. Generally patient (physiotherapy, botulinum toxin and multilevel surgery) to satisfaction is related to function in synergic hand and to obtain it. segmental alignment and aesthetic appearance in the other two groups; the grater is poorness in motor repertoire, the Limitations of the study greater is awareness of little modifiability. Sample size is little because of the recent experience of Compliance to rehabilitation program and its appro- our group in UL multilevel surgery. We lost some patient priateness affect final result; for this reason, it requires a Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 129

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, CRISTELLA HAND FUNCTION IN UNILATERAL CEREBRAL PALSY 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. at follow-up because patients referring to Children Reha- cerebral palsy in Sweden. X. Prevalence and origin in the birth-year pe- bilitation Unit at S. Maria Nuova Hospital, Reggio Emilia, riod 1999-2002. Acta Paediatr 2010;99:1337–43. come from all part of Italy, so some of them has organi- 3.  Hagberg B, Hagberg G, Olow I, von Wendt L. The changing panorama zational difficulties and were followed by local hospital. of cerebral palsy in Sweden. VII. Prevalence and origin in the birth year Some patient did not have MUUL or AHA evaluation be- period 1987-90. Acta Paediatr 1996;85:954–60. cause they consider them repetitive and boring and were 4.  Surveillance of Cerebral Palsy in Europe. Surveillance of cerebral less engage in them, with risk of a worse score. palsy in Europe: a collaboration of cerebral palsy surveys and registers. Surveillance of Cerebral Palsy in Europe (SCPE). Dev Med Child Neurol Instrument used for evaluation are sometimes little sen- 2000;42:816–24. sitive to record changes; this leads to the need of a more 5.  Ferrari A, Cioni G. Le forme spastiche della PCI. Milan: Springer; appropriate scale and to a more extensive use of narrative 2005. medicine. 6.  Sgandurra G, Ferrari A, Cossu G, Guzzetta A, Fogassi L, Cioni G. Randomized trial of observation and execution of upper extremity actions Conclusions versus action alone in children with unilateral cerebral palsy. Neurorehabil Neural Repair 2013;27:808–15. This study provides a preliminary guide to plan surgery as 7.  Ferrari A, Maoret AR, Muzzini S, Alboresi S, Lombardi F, Sgandurra a function of manipulation pattern and not only of segmen- G, et al. A randomized trial of upper limb botulimun toxin versus placebo tal deformities. injection, combined with physiotherapy, in children with hemiplegia. Res Dev Disabil 2014;35:2505–13. It suggests that surgery induces: 1) a segmental and func- 8.  Sakzewski L, Ziviani J, Boyd RN. Efficacy of upper limb therapies tional gain in synergic hand; 2) a segmental, aesthetic and for unilateral cerebral palsy: a meta-analysis. Pediatrics 2014;133:e175– sometimes minimal functional gain in imprisoned hand; 204. 3) a segmental, aesthetic gain in excluded hand. Surgery 9.  van Munster JC, Maathuis KG, Haga N, Verheij NP, Nicolai JP, Had- induces a positive change in capacity and performance in ders-Algra M. Does surgical management of the hand in children with synergic hand UCP and sometimes in imprisoned hand spastic unilateral cerebral palsy affect functional outcome? Dev Med group. However, hand appearance should be consider as Child Neurol 2007;49:385–9. indication for surgery. 10.  Van Heest AE, House JH, Cariello C. Upper extremity surgical treat- ment of cerebral palsy. J Hand Surg Am 1999;24:323–30. This study also underlines the importance to analyze 11.  Van Heest AE, Bagley A, Molitor F, James MA. Tendon transfer results in term of spontaneous manipulation abilities and surgery in upper-extremity cerebral palsy is more effective than botuli- daily use; for this reason, according to us, video-recording num toxin injections or regular, ongoing therapy. J Bone Joint Surg Am spontaneous hand use is the gold standard. 2015;97:529–36. 12.  Smitherman JA, Davids JR, Tanner S, Hardin JW, Wagner LV, Peace GAS is a good instrument to identify tailored goals for LC, et al. Functional outcomes following single-event multilevel surgery each patient and to establish if they are achieved or not. of the upper extremity for children with hemiplegic cerebral palsy. J Bone Other scales do not consider CP form and their modifiabil- Joint Surg Am 2011;93:655–61. ity space and sometimes they do not follow CP children 13.  House JH, Gwathmey FW, Fidler MO. A dynamic approach to strategy. the thumb-in palm deformity in cerebral palsy. J Bone Joint Surg Am 1981;63:216–25. Further controlled prospective studies are required to 14.  Ferrari A, Benedetti MG, Mori M, Alboresi S. L’arto superiore nella justify this kind of surgical approach. paralisi cerebrale infantile. Padua: Piccin Nuova Libraria; 2016. 15.  Kiresuk TJ, Sherman RE. Goal attainment scaling: A general method References for evaluating comprehensive community mental health programs. Com- munity Ment Health J 1968;4:443–53. 1.  Surman G, Hemming K, Platt MJ, Parkes J, Green A, Hutton J, et al. 16.  Steenbeek D, Ketelaar M, Lindeman E, Galama K, Gorter JW. Inter- Children with cerebral palsy: severity and trends over time. Paediatr Peri- rater reliability of goal attainment scaling in rehabilitation of children with nat Epidemiol 2009;23:513–21. cerebral palsy. Arch Phys Med Rehabil 2010;91:429–35. 2.  Himmelmann K, Hagberg G, Uvebrant P. The changing panorama of 17.  Forbes DA. Goal Attainment Scaling. A responsive measure of client outcomes. J Gerontol Nurs 1998;24:34–40. 18.  Johnson LM, Randall MJ, Reddihough DS, Oke LE, Byrt TA, Bach TM. Development of a clinical assessment of quality of move- ment for unilateral upper-limb function. Dev Med Child Neurol 1994;36:965–73. 19.  Holmefur M, Aarts P, Hoare B, Krumlinde-Sundholm L. Test-retest and alternate forms reliability of the assisting hand assessment. J Rehabil Med 2009;41:886–91. 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 27, 2018. - Manuscript accepted: July 27, 2018. - Manuscript revised: July 11, 2018. - Manuscript received: July 9, 2017. 130 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 © 2017 EDIZIONI MINERVA MEDICA European Journal of Physical and Rehabilitation Medicine 2019 February;55(1):131-5 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/S1973-9087.17.04970-X 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. SPECIAL ARTICLE METHODOLOGICAL NOTES ON APPLYING THE ICF IN REHABILITATION Graphical modelling: a tool for describing and understanding the functioning of people living with a health condition Cristina EHRMANN 1, 2 *, Jerome BICKENBACH 1, 2, 3, Gerold STUCKI 1, 2, 3 1Department of Health Sciences and Health Policy, Faculty of Humanities and Social Sciences, University of Lucerne, Lucerne, Switzerland; 2Swiss Paraplegic Research (SPF), Nottwil, Switzerland; 3ICF Research Branch, a cooperation partner within the WHO Collaborating Center for the Family of International Classifications in Germany (at DIMDI), Nottwil, Switzerland *Corresponding author: Cristina Ehrmann, Swiss Paraplegic Research (SPF), Guido A. Zäch Strasse 4, 6207, Nottwil, Switzerland. E-mail: [email protected] ABSTRACT Rehabilitation aims to optimize people’s lived experience of health or functioning. A comprehensive understanding of people’s functioning is thus fundamental for rehabilitation clinicians and scientists. Over the past ten years it has been shown that graphical modelling is a promising technique for modelling data on people’s functioning. It can contribute to our understanding of the complex associations between domains of functioning and the identification of potential targets for rehabilitation interventions both at the level of the person and the environment. The objective of this methodological note is to demonstrate how graphical modelling can be used by rehabilitation clinicians and scientists in the de- scription, understanding and influencing of people’s functioning. The application of graphical modelling and the interpretation of results is illus- trated using the Spinal Cord Injury Independence Measure – Self Report used in the Swiss Spinal Cord Injury Cohort Study. Finally, we discuss the potential of graphical modelling for the planning of studies that expand our understanding of functioning and for rehabilitation interventions. (Cite this article as: Ehrmann C, Bickenbach J, Stucki G. Graphical modelling: a tool for describing and understanding the functioning of people living with a health condition. Eur J Phys Rehabil Med 2019;55:131-5. DOI: 10.23736/S1973-9087.17.04970-X) Key words: International Classification of Functioning, Disability and Health - Rehabilitation - Spinal cord injuries. Rehabilitation aims to optimize people’s lived experi- Graphical modelling is one such approach; others in- ence of health or functioning.1-5 A comprehensive un- clude factor analysis, structural equation modelling and derstanding of people’s functioning is thus fundamental Bayesian networks.9 Graphical modelling has been de- for rehabilitation clinicians and scientists.6, 7 According veloped and first applied in genetics.10 In the health sci- to the World Health Organization’s International Classi- ences it was first applied in molecular biology.11 More fication of Functioning, Disability and Health (ICF), the recently, its application for the study of functioning and term ‘functioning’ represents the overall lived experience outcomes research has been introduced9, 12, 13 (Table of health, which comprises ‘biological health’ and ‘lived I).9, 12-19 It has been shown that graphical modelling can health’, in light of the person’s health conditions, given not only contribute to our understanding of the complex his/her resources, and in interaction with the environment.8 associations between domains of functioning but also the There are thus many intrinsic linkages between body func- identification of potential targets for rehabilitation inter- tions and structures and the many activities that humans ventions both at the level of the person and the environ- perform in interaction with the environment. Understand- ment.12, 14 ing the complexity of these interactions can be facilitated by the use of statistical approaches that allow us to study Currently, graphical modelling has been primarily used the complex associations among a large set of variables. in studies into the lived experience of people living with spinal cord injury.12, 14, 15 Applications in studying func- Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 131

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 EHRMANN 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, GRAPHICAL DATA MODELLING FOR FUNCTIONING 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 I.—C urrent research using graphical modelling into studying the functioning of people living with a health condition.9, 12-19 Type of study Study Summary Strobl et al. (2009)9 Methodological studies into This study introduced graphical modelling as an approach for the study of human the application of graphical Kalisch et al. (2010)12 functioning using data collected by means of the ICF. The least absolute shrinkage modelling for the study of and selection operator (LASSO) for generalized linear models were implemented functioning in the algorithm for constructing the graphs Fellinghauer et al. (2010)13 This study explored how graphical modelling may be used in the study of the ICF Strobl et al. (2012)18 data. The PC algorithm was implemented in the algorithm for constructing the Fellinghauer et al. (2013)19 graphs Studies examining functioning in Reinhardt et al. (2011)14 This study described the application of graphical modelling approach for an people living with SCI epidemiology of functioning Reinhardt et al. (2010)15 This study implemented the Bolasso algorithm for constructing graphs for high dimensional binary data Studies examining functioning in This study developed a new method called Graphical Random Forest for estimating other health conditions the association structure when studying human functioning of people living with a health condition Stroke Massa et al. (2015)17 This study showed differential association structures of ICF categories for lower Head and neck cancer Becker et al. (2011)16 and higher in all ICF categories. More specifically, a more fractionate graph in lower-resourced countries was identified This study showed that graphical modelling approach may be used to reduce the dimensionality of functioning. Most of the dimensions discovered were meaningfully from the clinical point of view Graphical modelling was used to understand the relations among cognitive problems following after stroke Using the algorithm developed by Strobl et al.9 in constructing the graphs, this study identified largely meaningful associations between the ICF categories tioning have occurred with other health conditions, includ- these variables. The two most common graphical models ing head and neck cancer16 and stroke17 (Table I). are called “directed,” which are based on directed acyclic graphs (DAGs), and “undirected” graphs, which are based In light of the potential of graphical modelling for func- on the skeleton of a DAG. The skeleton of a DAG is a tioning and rehabilitation research the objective of this DAG where the directions (arrowheads) are ignored. methodological note is to demonstrate how it can be used by rehabilitation clinicians and scientists in the descrip- In a DAG, an arrow (directed edge) indicates that the tion, understanding and influencing of people’s function- variable from the arrow’s tail carries information, and there- ing. The application of graphical modelling and the inter- fore influences, the variable from the arrow’s head (Figure pretation of results is illustrated using the self-report ver- 1B-D). In the skeleton of a DAG, the relationship of two sion of the Spinal Cord Injury Independence Measure used variables is indicated, but not its direction (Figure 1A). in the Swiss Spinal Cord Injury Cohort Study. Finally, we discuss the potential of graphical modelling for the plan- In Figure 1, the variables represent items of the SCI ning of studies to improve our understanding of function- Independence Measure-Self-Report (SCIM-SR), a self-re- ing and for rehabilitation interventions. ported instrument for assessing independence in perform- ing activities of daily living, for people living with a spinal Introduction to graphical modelling cord injury (SCI).21 A set of any consecutive sequence of arrows is called a path (e.g. Bowel management → Eating/ Graphical models are a class of statistical models that com- Drinking → Grooming [Figure 1A], Bowel management bine probabilistic and graphical approaches to identify and ← Eating/Drinking → Grooming [Figure 1B] or Bowel visualize relationships of conditional dependence among management → Eating/Drinking ← Grooming [Figure variables of interest.20 In graphical modelling a graph G is 1C]). Since statistically is not possible that a variable to be given by G = (V, E), where V is a set of nodes representing influenced by itself, in a DAG no path from one node to the variables of interest and E is a set of graph edges that itself exists when following the arrowheads, i.e. there is no indicates the conditional dependence relationships among cycle such as Bowel management → Eating/Drinking → Grooming → Bowel management. 132 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, GRAPHICAL DATA MODELLING FOR FUNCTIONING EHRMANN 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. Bowel Grooming identified and visualized by undirected graphs; and 2) an management interpretation of the potential intervention targets for im- proving functioning using DAGs when a large number of Eating/ variable is available. A Drinking Associations identified using undirected graphs Bowel Grooming Bowel Grooming Bowel Grooming The algorithm for constructing the undirected graph illus- management management management trated here will be based on a so-called PC algorithm.23 Following this algorithm, a skeleton of a DAG is con- Eating/ Eating/ Eating/ structed by testing any two variables for conditional inde- Drinking Drinking Drinking pendence, given any subsets of the remaining variables. In the final skeleton, there is an edge between two variables, B C D if and only if the variables are dependent, given any subset of the remaining variables. Other algorithms for construct- Figure 1.—The conditional independence in a DAG and its skeleton. ing undirected graphs are based on regression techniques that deal with a large number of variables: examples are Conditional independence in a directed and an undirected Least Absolute Shrinkage and Selection Operator (LAS- graph SO)9, 18 or the Random Forest.19 In these algorithms, the associations of each variable with all remaining variables Figure 1A shows a simple example of undirected graph. are analyzed. In these regressions, one may control for ad- Bowel management and Grooming are connected through ditional information, such as age or gender. This means Eating/Drinking, which implies that they are conditionally that when studying the association among variables when independent in light of Eating/Drinking. The concept of controlling for additional information, the final undirected conditional independence means that information about graphs includes only the variables as nodes. None of the Bowel management does not give us any information about algorithms make a priori assumptions about possible or Grooming, once we have knowledge of Eating/Drinking. likely associations. However, if Bowel management and Grooming are con- ditionally independent, we cannot infer that they are also Cross-sectional data from the community survey of the marginally independent.22 Marginal relationship between Swiss Spinal Cord Injury (SwiSCI) Cohort Study were Bowel management and Grooming is studied only and considered.24 The participants in this study were 1549 completely ignoring knowledge of Eating/Drinking. individuals aged 16 years or older with permanent resi- dence in Switzerland who were diagnosed with traumatic Figure 1B-D shows three examples of DAGs. By defi- or non-traumatic SCI. We used the items of the SCIM-SR nition, Bowel management and Grooming are condition- as the outcome measure. For data analysis we used the ally independent if the path from Bowel management and PC algorithm, since we were not interested in controlling Grooming shows that Bowel management does not carry for people’s demographic characteristics, level of SCI or direct information on Grooming, in light of Eating/Drink- completeness of injury. Cross-sectional data was used. To ing. This is true for the first two examples (Figure 1B, C). enhance the stability of the edges, we repeated the PC- In both examples Eating/Drinking already incorporates algorithm for 100 generated samples reflecting the shape information on Bowel management since Eating/Drinking of the original one (bootstrapping with replacement). The is an intermediate variable in Figure 1B and a confounder results were aggregated in a summary graph that included variable in Figure 1C. In the third example (Figure 1D), only edges that appeared in the skeleton of at least 50 gen- Bowel management carries information about Grooming erated samples. only if we condition on Eating/Drinking, and when we do, they are conditionally dependent. Figure 2 presents the association structures that resulted from the analysis. Only one connected component was Interpretation of the results identified. A component is a set of variables connected of graphical modelling within them, but not between the components. This indi- cates a strong conditional dependence among variables. An Here we illustrate how to interpret graphical modelling edge between any two variables, e.g. Transfer 1 and Eat- with examples of: 1) an interpretation of associations ing/Drinking, indicates a strong dependence in light of any Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 133

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 EHRMANN 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, GRAPHICAL DATA MODELLING FOR FUNCTIONING 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. Transfer Transfer Grooming 2 1 Transfer Dressing 3 upper body Toileting Transfer Bladder Washing Eating/ Transfer 1: Stairs 4 mana- upper body Drinking bed-wheelchair gement Dressing Bowel Dressing lower Moving mana- lower body body around gement Washing Eating/ lower Drinking body Dressing Breathing Figure 3.—The five most influential variables with respect to the Eating/ Moving upper Drinking. outdoors body >100m mean for all the estimates for each variable and ranked the variables according to the effect strength of the simulated Washing Grooming intervention. Figure 3 presents the five functioning catego- upper Moving ries with the highest simulated intervention effect. body indoor Moving <100m Figure 2.—The association structure identified for the spinal cord injury Potential research applications sample among variables of the SCI Independence Measure – Self Re- of graphical modelling approach port. The thickness of the edges represents the frequency of the associa- tions among the re-sampled data sets. Graphical modelling is an explorative statistical approach Transfer 1: bed-wheelchair; Transfer 2: wheelchair-toilet/tub; Transfer that does not impose an a-priori assumption about the 3: wheelchair-car; Transfer 4: floor-wheelchair. association between variables of interest. This approach may, therefore, be used to formulate substantive research other subset of remaining variables. The path is between hypotheses about: Dressing lower body to Eating/Drinking goes through Dressing upper body, and this means that Dressing lower •  conditional independence relations between two vari- body and Eating/Drinking are conditionally independent ables of interest. To give an example, in the absence of an given the information we have on Dressing upper body. A edge between Bowel management and Grooming in the triangle formed by three variables — i.e. Eating/Drinking, final graph (Figure 2), we may hypothesize a conditionally Grooming and Breathing — indicates that one variable is independent relation between them for people living with strongly dependent on the other two. SCI. This hypothesis imposes that: 1) the conditionally de- pendence relations identified in our example are sufficient Using DAGs to identify potential intervention targets for to understand all associations in the independence in per- improving functioning forming activities of daily living; 2) the set of conditional dependence relations cannot be reduced without destroy- Using the same data from the SwiSCI community survey, ing the whole association structure;20 we adopt graphical modelling to identify the functioning categories to be targeted in an intervention to improve Eat- •  subsets of variables for prediction analysis: for study- ing/Drinking. The Pearl’s Calculus of Intervention used by ing the functioning categories influencing Eating/Drink- Maathuis et al. was applied.22, 25 Following this calculus, ing, for example, it is sufficient to look at the functioning the DAGs that encode the same conditional independence categories conditionally dependent with Eating/Drinking over our variables of interest can be estimated. Then, in each as identified in the final graph (Figure 2), namely Trans- DAG, a clinical intervention for each variable is statistically fer 1: bed-wheelchair, Dressing upper body, Breathing, simulated by forcing the variable’s values from “problem” Grooming, and Bowel management;9 to “no problem” one after another. The effect of such this simulation for each variable on Eating/Drinking was then •  confounding variables: the triangle formed by Eating/ estimated for each of the 100 data sets. We calculated the Drinking, Grooming and Breathing, for example, indicates that each of these three variables may be a confounding vari- able in the association of the other two variables (Figure 2);22 134 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, GRAPHICAL DATA MODELLING FOR FUNCTIONING EHRMANN 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. •  potential constructs and scales: the functioning cat- 9.  Strobl R, Stucki G, Grill E, Müller M, Mansmann U. Graphical mod- egories Stairs, Toileting, Transfer 2: wheelchair-toilet/ els illustrated complex associations between variables describing human tub, Transfer 1: bed-wheelchair, Transfer 3: wheelchair- functioning. J Clin Epidemiol 2009;62:922–33. car, and Transfer 4: floor-wheelchair form a circular path 10.  Lauritzen SL, Sheehan NA. Graphical Models for Genetic Analyses. (hexagon) (Figure 2) indicating that they are strongly as- Stat Sci 2003;18:489–514. sociated to be considered as candidates for a subscale;26 11.  Scutari M, Nagarajan R. Identifying significant edges in graphical models of molecular networks. Artif Intell Med 2013;57:207–17. •  the potential interventions targets for improving func- 12.  Kalisch M, Fellinghauer BA, Grill E, Maathuis MH, Mansmann U, tioning: we identified the potential targets for improving Bühlmann P, et al. Understanding human functioning using graphical functioning in Eating/Drinking without any randomized models. BMC Med Res Methodol 2010;10:14. control study (Figure 3). 13.  Fellinghauer B, Reinhardt JD, Stucki G. Towards an epidemiology of functioning. In: Franchignoni F, editor. Advances in Rehabilitation. Previous studies showed that graphical modelling is a Research issues in Physical & Rehabilitation Medicine. Pavia: Maugeri power tool for capturing, not only the association structure Foundation Books; 2010. p. 53–68. of the selected population, but also to estimate, visualize 14.  Reinhardt JD, Mansmann U, Fellinghauer BA, Strobl R, Grill E, von and thereby compare the dependence structure of sub- Elm E, et al. Functioning and disability in people living with spinal cord populations (e.g. paraplegic and tetraplegic SCI subpopu- injury in high- and low-resourced countries: a comparative analysis of 14 lations). This comparison can be done by counting how countries. Int J Public Health 2011;56:341–52. many arrows need to be added or deleted in order to have 15.  Reinhardt JD, Fellinghauer BA, Strobl R, Stucki G. Dimension re- the same association structure for each subpopulation.12 duction in human functioning and disability outcomes research: graphi- These properties make graphical modelling a valuable and cal models versus principal components analysis. Disabil Rehabil reliable tool that should be adopted by future research for 2010;32:1000–10. studying the epidemiology of functioning of people living 16.  Becker S, Strobl R, Cieza A, Grill E, Harréus U, Tschiesner U. Graph- with a health condition.13, 27 ical modeling can be used to illustrate associations between variables de- scribing functioning in head and neck cancer patients. J Clin Epidemiol References 2011;64:885–92. 17.  Massa MS, Wang N, Bickerton WL, Demeyere N, Riddoch MJ, Hum- 1.  Stucki G, Melvin J. The International Classification of Functioning, phreys GW. On the importance of cognitive profiling: A graphical model- Disability and Health: a unifying model for the conceptual description ling analysis of domain-specific and domain-general deficits after stroke. of physical and rehabilitation medicine. J Rehabil Med 2007;39:286–92. Cortex 2015;71:190–204. 2.  Meyer T, Gutenbrunner C, Bickenbach J, Cieza A, Melvin J, Stucki G. 18.  Strobl R, Grill E, Mansmann U. Graphical modeling of binary data us- Towards a conceptual description of rehabilitation as a health strategy. J ing the LASSO: a simulation study. BMC Med Res Methodol 2012;12:16. Rehabil Med 2011;43:765–9. 19.  Fellinghauer B, Bühlmann P, Ryffel M, von Rhein M, Reinhardt JD. 3.  Stucki G, Prodinger B, Bickenbach J. Four steps to follow when docu- Stable graphical model estimation with random forests for discrete, con- menting functioning with the International Classification of Functioning, tinuous, and mixed variables. Comput Stat Data Anal 2013;64:132–52. Disability and Health. Eur J Phys Rehabil Med 2017;53:144–9. 20.  Wermuth N, Lauritzen SL. On Substantive Research Hypotheses, 4.  Stucki G, Bickenbach J. Functioning: the third health indicator in the Conditional Independence Graphs and Graphical Chain Models. J R Stat health system and the key indicator for rehabilitation. Eur J Phys Rehabil Soc B 1990;52:21–50. Med 2017;53:134–8. 21.  Fekete C, Eriks-Hoogland I, Baumberger M, Catz A, Itzkovich 5.  Stucki G, Bickenbach J. Functioning information in the learning health M, Lüthi H, et al. Development and validation of a self-report version system. Eur J Phys Rehabil Med 2017;53:139–43. of the Spinal Cord Independence Measure (SCIM III). Spinal Cord 6.  Stucki G, Bickenbach J, Negrini S. Methodological notes on applying 2013;51:40–7. the International Classification of Functioning, Disability and Health in 22.  Pearl J. Causality: models. Reasoning and inference. New York: rehabilitation. Eur J Phys Rehabil Med 2017;53:132–3. Cambridge University Press; 2000. 7.  Stucki G, Kostanjsek N, Üstün B, Ewert T, Cieza A. Applying the ICF 23.  Kalisch M, Mächler M, Colombo D, Maathuis MH, Bühlmann P. in Rehabilitation Medicine. In:Frontera W, DeLisa JA, editors. DeLisa’s Causal Inference Using Graphical Models with the R Package pcalg. J Physical Medicine and Rehabilitation. Principles and Practice. Lippincott Stat Softw 2012;47. Williams & Wilkins; 2010. p.301-324. 24.  Bickenbach J, Tennant A, Stucki G. The SwiSCI Cohort Study. J Re- 8.  Stucki G, Bickenbach J, Melvin J. Strengthening Rehabilitation in habil Med 2016;48:117–9. Health Systems Worldwide by Integrating Information on Function- 25.  Maathuis MH, Kalisch M, Bühlmann P. Estimating high-dimensional ing in National Health Information Systems. Am J Phys Med Rehabil intervention effects from observational data. Ann Stat 2009;37:3133–64. 2017;96:677–81. 26.  Bollen KA. Latent variables in psychology and the social sciences. Annu Rev Psychol 2002;53:605–34. 27.  Stucki G. Olle Höök Lectureship 2015: The World Health Organiza- tion’s paradigm shift and implementation of the International Classifica- tion of Functioning, Disability and Health in rehabilitation. J Rehabil Med 2016;48:486–93. Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript. Acknowledgements.—The authors would like to thank Cristiana Baffone, Roxanne Maritz, and Núria Adroher Duran for critical comments and the support during the preparation of the manuscript. Article first published online: November 16, 2017. - Manuscript accepted: November 8, 2017. - Manuscript received: September 5, 2017. Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 135

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 LETTERS TO THE EDITOR 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, © 2017 EDIZIONI MINERVA MEDICA a valuable support to help improve children’s HRQOL and maxi- 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. Online version at http://www.minervamedica.it mize their Activity and Participation. European Journal of Physical and Rehabilitation Medicine 2019 February;55(1):136-7 DOI: 10.23736/S1973-9087.17.04900-0 Rehabilitate a child with RDs means dealing with many chal- lenges: the presence of complex ad severe disabilities; stable pres- Children’s rare disease rehabilitation: ence or related to the stages of child development of numerous co- from multidisciplinarity to the morbidities (pediatric, neuropsychiatric, orthopedic issues, etc.); transdisciplinarity approach the simultaneous significance of the issues; in all cases and for all patients, personalized interventions are required. All this neces- In recent years, considerable attention has been paid worldwide sitates a comprehensive and multifaceted management approach to the challenge in the management of rare diseases (RDs). Public to the children’s rare disease rehabilitation. Multidisciplinary re- health authorities have focused on efforts to stimulate the research, habilitation is now considered the key approach in rehabilitation to improve healthcare and to create an integrative medical approach and health care paradigms. And, if this is generally true, it is of for RDs. By definition, RDs are a group of diseases with a low prev- fundamental importance in children’s rare disease rehabilitation, alence. The European Union classified a disease as rare if it affects because of the multidimensionality of the complex health issues less than 5 persons per 10,000 peoples. To date, about 7000 RDs that children and families are facing. The ICF and the ICF-CY, have been identified and for more than half the causative gene(s) published by the World Health Organization (WHO) to standardize or a final diagnosis are yet to identify. It is estimated that 80% of descriptions of health and disability, classify the outcome of a con- RDs have a genetic origin with signs or symptoms that begin during dition in terms of body functions/structures, activity level and par- pregnancy or in the early postnatal period. Infectious, cancer or au- ticipation. This WHO-classification of functioning and disability is toimmune origin are also included. Recently, it has been proposed a biopsychosocial model of disability which is increasingly being that epigenetic variation may contribute to the risk of RDs.1 recognized as an efficacious tool to describe health and disability and a framework for planning and monitoring rehabilitation inter- Clinically, RDs are usually complex, chronic and life-threaten- ventions over time. The ICF conceptual framework underlies the ing conditions, with a great variability in onset, severity of clini- need for a global pathway of care through the involvement of many cal signs and symptoms, temporal course and individual outcome stakeholders and moving from a multidisciplinary perspective only with wide differences from case to case. Currently there are no to an integrated multi-, inter- and transdisciplinary perspective.2 established appropriate treatments for most of the RDs. For now, While multidisciplinarity in a rehabilitation program is based on the development of an optimal care management and rehabilita- practitioners’ knowledge of a variety of disciplines (neurologist, tion seems to be the only perspective to improve the health-related physiatrist, pediatrician, etc.) each operating within their respective quality of life (HRQOL) in these children. Rehabilitation progress fields of competence, interdisciplinarity integrates and harmonizes in recent decades has allowed increasingly better strategy to ensure the relationships between the different disciplines in a coordinated, people with disabilities are able to access health-related rehabilita- reliable and consistent manner to support children’s rehabilitation. tion. A rehabilitation that looks at the “persons” in their entirety and complexity according to their chances and their potentiality However, the biopsychosocial model underlying the ICF, sug- of participation. Regardless of what caused a disability the goal gests the transdisciplinary approach as an innovative way to under- of rehabilitation is to achieve and maintain optimum functioning stand the dimensions of human functioning, health and disability. in interaction with the environments, following the International The added value of transdisciplinarity, is a more holistic perspec- Classification of Functioning (ICF) approach to disability, which tive integrating the natural, social, and health sciences in a humani- understands functioning and disability, as a dynamic interaction ties context, and in so doing enabling each to transcend their tradi- between health conditions and contextual factors, both personal tional boundaries. But the effective value of transdisciplinarity is and environmental. Until a few years ago, knowledge about these its going over and above multi- and interdisciplinary approaches RDs was “rare”, as well as “rare” was the knowledge about the as it acts as a common thread for people from different disciplines disabilities encountered by children with RDs and “rare” were the who interact among themselves and agree on a ‘common aim’ and, rehabilitative programs and interventions. Today, we know that in order to achieve it, develop a shared framework.3 Transdiscipli- many children with RDs live daily with a chronic condition, some narity combines multi- and interdisciplinarity with a collaborative of them live with a life-threatening condition, but above all, we and participative approach, able to generate new knowledge, ap- know that many children with RDs live constantly and daily with a plying a holistic approach to children’s rare disease rehabilitation wide range of disabilities that vary by type, severity and complex- in that all stakeholders set aside their own specific perspectives to ity. If for many RDs, to date we do not have yet adequate therapeu- embrace a global one, respectful of all individual instances, mak- tic treatments, certainly appropriate rehabilitation programs can be ing a better contribution to optimal care management and rehabili- tation of children with a complex disability.4 If adequately applied, the transdisciplinary approach could represent the “gold standard” in neurorehabilitation because it provides a better integrated ser- vice delivery and more comprehensive, patient-centered approach 136 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, LETTERS TO THE EDITOR 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. with respect to other models of team working. To implement a © 2018 EDIZIONI MINERVA MEDICA transdisciplinary model, team members must develop an excellent Online version at http://www.minervamedica.it communication and shared decision making.5 However, cultural, European Journal of Physical and Rehabilitation Medicine 2019 February;55(1):137-9 educational and organizational constraints can create barriers that DOI: 10.23736/S1973-9087.18.05393-5 can stifle the evolution from the traditional multidisciplinary, and currently interdisciplinary approaches to a transdisciplinary set- Prevalence and burden of obesity ting in order to deliver the best possible rehabilitation to children in Rehabilitation Units in Italy: a survey with rare diseases. Transdisciplinarity approach means having a consistent focus across all child’s health dimensions. A consistent Obesity is a metabolic disease (ICD-10 code E66) that has reached focus starts from the concept that RDs are complex and multifac- epidemic proportions. The World Health Organization (WHO) has eted clinical conditions and usually cannot be cured. A consistent declared obesity as the largest global chronic health problem in focus shares the view that RDs are chronic conditions that prog- adults. Obesity is a gateway to ill health, and it has become one ress over time, impacting all dimensions, individual variables and of the leading causes of disability and death, affecting not only people to the relationship. A consistent focus promotes forward- adults but also children and adolescents worldwide.1 The WHO looking rehabilitation programs, interventions in all life contexts, world health statistics report in 2015 shows that in the European an approach involving all aspects of human functioning to improve region the overall obesity rate among adults is 21.5% in males health-related quality of life. A consistent focus can help shape re- and 24.5% in females.2 Obesity has important consequences for habilitation in order to promote the children’s adjustment, partici- morbidity, disability and health-related quality of life. It entails a pation and subjective and relational well-being.4 higher risk of developing type 2 diabetes, cardiovascular diseases, obstructive sleep apnea syndrome and obesity hypoventilation Paraphrasing Antoine de Saint-Exupéry (The Little Prince), syndrome, several common forms of cancer, osteoarthritis and oth- transdisciplinary approach “does not consist in gazing at each oth- er health problems including musculoskeletal pain and migraine er but in looking outward together in the same direction.” with a possible association with obesity as well as other chronic pain conditions the combination of which with obesity worsens Antonio TRABACCA *, Luigi RUSSO health-related quality of life. Prevention of obesity is prospectively very important, but the challenge is rather the progression of the Unit for Severe Disabilities in Developmental Age and Young disabilities already present calling into action physical and reha- Adults (Developmental Neurology and Neurorehabilitation), bilitation medicine specialists. The health and economic burden Brindisi Research Center, Scientific Institute I.R.C.C.S. resulting from obesity and its consequences is not only based on “Eugenio Medea”– “La Nostra Famiglia”, Brindisi, Italy mortality and the financial costs of hospital admissions and treat- *Corresponding author: Antonio Trabacca, Unit for Severe Disabilities in Developmental Age and Young Adults (Developmental Neurology and Neu- rorehabilitation), Brindisi Research Center, Scientific Institute I.R.C.C.S. “Eugenio Medea”– “La Nostra Famiglia”, Piazza A. Di Summa, 72100 Brindisi, Italy. E-mail: [email protected] References 1.  Moliner AM. Creating a European Union framework for actions in the field of rare diseases. Adv Exp Med Biol 2010;686:457–73. 2.  Trabacca A, Moro G, Gennaro L, Russo L. When one plus one equals three: the ICF perspective of health and disability in the third millennium. Eur J Phys Rehabil Med 2012;48:709–10. 3.  Choi BC, Pak AW. Multidisciplinarity, interdisciplinarity and trans- disciplinarity in health research, services, education and policy: 1. Definitions, objectives, and evidence of effectiveness. Clin Invest Med 2006;29:351–64. 4.  Trabacca A, Vespino T, Di Liddo A, Russo L. Multidisciplinary reha- bilitation for patients with cerebral palsy: improving long-term care. J Multidiscip Healthc 2016;9:455–62. 5.  Karol RL. Team models in neurorehabilitation: structure, function, and culture change. NeuroRehabilitation 2014;34:655–69. Conflicts of interest.—The authors certify that there is no conflict of inter- Figure 1.— The geographical distribution of the Rehabilitation Institutes est with any financial organization regarding the material discussed in the who responded to the questionnaire. manuscript. Article first published online: November 3, 2017. - Manuscript accepted: October 30, 2017. - Manuscript revised: September 15, 2017. - Manuscript received: July 7, 2017 . (Cite this article as: Trabacca A, Russo L. Children’s rare disease rehabilita- tion: from multidisciplinarity to the transdisciplinarity approach. Eur J Phys Rehabil Med 2019;55:136-7. DOI: 10.23736/S1973-9087.17.04900-0) Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 137

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, LETTERS TO THE EDITOR 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. with respect to other models of team working. To implement a © 2018 EDIZIONI MINERVA MEDICA transdisciplinary model, team members must develop an excellent Online version at http://www.minervamedica.it communication and shared decision making.5 However, cultural, European Journal of Physical and Rehabilitation Medicine 2019 February;55(1):137-9 educational and organizational constraints can create barriers that DOI: 10.23736/S1973-9087.18.05393-5 can stifle the evolution from the traditional multidisciplinary, and currently interdisciplinary approaches to a transdisciplinary set- Prevalence and burden of obesity ting in order to deliver the best possible rehabilitation to children in Rehabilitation Units in Italy: a survey with rare diseases. Transdisciplinarity approach means having a consistent focus across all child’s health dimensions. A consistent Obesity is a metabolic disease (ICD-10 code E66) that has reached focus starts from the concept that RDs are complex and multifac- epidemic proportions. The World Health Organization (WHO) has eted clinical conditions and usually cannot be cured. A consistent declared obesity as the largest global chronic health problem in focus shares the view that RDs are chronic conditions that prog- adults. Obesity is a gateway to ill health, and it has become one ress over time, impacting all dimensions, individual variables and of the leading causes of disability and death, affecting not only people to the relationship. A consistent focus promotes forward- adults but also children and adolescents worldwide.1 The WHO looking rehabilitation programs, interventions in all life contexts, world health statistics report in 2015 shows that in the European an approach involving all aspects of human functioning to improve region the overall obesity rate among adults is 21.5% in males health-related quality of life. A consistent focus can help shape re- and 24.5% in females.2 Obesity has important consequences for habilitation in order to promote the children’s adjustment, partici- morbidity, disability and health-related quality of life. It entails a pation and subjective and relational well-being.4 higher risk of developing type 2 diabetes, cardiovascular diseases, obstructive sleep apnea syndrome and obesity hypoventilation Paraphrasing Antoine de Saint-Exupéry (The Little Prince), syndrome, several common forms of cancer, osteoarthritis and oth- transdisciplinary approach “does not consist in gazing at each oth- er health problems including musculoskeletal pain and migraine er but in looking outward together in the same direction.” with a possible association with obesity as well as other chronic pain conditions the combination of which with obesity worsens Antonio TRABACCA *, Luigi RUSSO health-related quality of life. Prevention of obesity is prospectively very important, but the challenge is rather the progression of the Unit for Severe Disabilities in Developmental Age and Young disabilities already present calling into action physical and reha- Adults (Developmental Neurology and Neurorehabilitation), bilitation medicine specialists. The health and economic burden Brindisi Research Center, Scientific Institute I.R.C.C.S. resulting from obesity and its consequences is not only based on “Eugenio Medea”– “La Nostra Famiglia”, Brindisi, Italy mortality and the financial costs of hospital admissions and treat- *Corresponding author: Antonio Trabacca, Unit for Severe Disabilities in Developmental Age and Young Adults (Developmental Neurology and Neu- rorehabilitation), Brindisi Research Center, Scientific Institute I.R.C.C.S. “Eugenio Medea”– “La Nostra Famiglia”, Piazza A. Di Summa, 72100 Brindisi, Italy. E-mail: [email protected] References 1.  Moliner AM. Creating a European Union framework for actions in the field of rare diseases. Adv Exp Med Biol 2010;686:457–73. 2.  Trabacca A, Moro G, Gennaro L, Russo L. When one plus one equals three: the ICF perspective of health and disability in the third millennium. Eur J Phys Rehabil Med 2012;48:709–10. 3.  Choi BC, Pak AW. Multidisciplinarity, interdisciplinarity and trans- disciplinarity in health research, services, education and policy: 1. Definitions, objectives, and evidence of effectiveness. Clin Invest Med 2006;29:351–64. 4.  Trabacca A, Vespino T, Di Liddo A, Russo L. Multidisciplinary reha- bilitation for patients with cerebral palsy: improving long-term care. J Multidiscip Healthc 2016;9:455–62. 5.  Karol RL. Team models in neurorehabilitation: structure, function, and culture change. NeuroRehabilitation 2014;34:655–69. Conflicts of interest.—The authors certify that there is no conflict of inter- Figure 1.— The geographical distribution of the Rehabilitation Institutes est with any financial organization regarding the material discussed in the who responded to the questionnaire. manuscript. Article first published online: November 3, 2017. - Manuscript accepted: October 30, 2017. - Manuscript revised: September 15, 2017. - Manuscript received: July 7, 2017 . (Cite this article as: Trabacca A, Russo L. Children’s rare disease rehabilita- tion: from multidisciplinarity to the transdisciplinarity approach. Eur J Phys Rehabil Med 2019;55:136-7. DOI: 10.23736/S1973-9087.17.04900-0) Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 137

COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA LETTERS TO THE EDITOR 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 If obesity needs to be clinically addressed during hospital stay, challenges posed by obesity, which at times contrasts with the fac- 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, which professionals can you involve in your Rehabilitation Center? tual organization of our units. 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 2.—Professionals that can be involved in the multidisciplinary One purpose of the present letter was to raise the awareness Endocrinologist/NutritioniNstumber of professionals rehabilitation program. of rehabilitation specialists on the special organizational issues related to the rehabilitation of obese patients with comorbidities. OEcxceurpcBiaatPsriiehSoayPotnpssrPcaihiislyocyacyttlsDcshihihhuooeeewilltrrrioaoaogatrcegrpgpikiiiiioaesssssntrttntt ment of comorbidities, but also on significant ensuing disability They require a truly multidimensional approach with the simulta- Other with limitations in functioning which are well documented in the neous provision of physiotherapy, diet and nutritional support, psy- ICF Core Sets for Obesity.3 chological counselling, adapted physical activity, specific nursing skills and multidisciplinary consultations. Front line assessment The need for a hospital-based multidisciplinary rehabilitation and preventive strategies, risk stratification and disease manage- program for severely obese patients with comorbidities has been ment are needed and for that the integration of several medical acknowledged by the Italian Ministry of Health.4 specialties (clinical nutrition, endocrinology, psychiatry, psychol- ogy, rehabilitation medicine) and health professions (dietitians, From theory to practice, rehabilitation units with optimal stan- psychologists, physiotherapists and nurses) is mandatory. From dards for the treatment of pathological conditions and their func- our survey, only few institutes included a psychiatrist and a bar- tional consequences are often structurally, culturally and organiza- iatric surgeon among the professionals that can be involved in a tionally inadequate for the care of patients with extreme obesity. multidisciplinary rehabilitation program (Figure 2). Appropriate therapeutic and rehabilitative protocols carried out by specifically trained operators, an ergonomically adequate and Whether they were employed as permanent staff or had a con- safe environment for patients and staffs alike, with an adequate sultant role was not investigated in the survey, which only ques- presence of bariatric aids for lifting and transferring, are in fact tioned the availability of such figures. However, this is an impor- needed. In Italy, we have globally around 199,000 hospital beds tant point to discriminate truly multidisciplinary teams who work with a real availability of 0.4 beds every 1000 inhabitants planned together on a daily basis from units who can only have remote for rehabilitation, including long-term rehabilitation at lesser in- specialist consultations. tensity and a high variability across regions. In line with global trends, obesity rates among patients who require rehabilitation fol- Another finding is that rehabilitation units are generally struc- lowing an acute event are expected to dramatically increase. Na- turally inadequate for admitting severely obese patients, due to a tional data regarding the prevalence of obesity among in-patients lack of bariatric aids, weight scales, beds, lifts and wheelchairs, in the rehabilitation units are lacking. The online survey, endorsed but also due to cultural and organizational gaps: for instance, often by SIMFER, reached 44 rehabilitation units of national relevance body height is not recorded in the medical charts and thus Body with more than 25 inpatients, both public or private. The estimated Mass Index cannot be calculated. The results of our survey were prevalence of obesity but also the burden of care related to the largely dependent on the single expert opinion of the PRM doctors treatment and rehabilitation of those patients as perceived by the who provided the answers and were therefore only estimates of the PRM specialist were investigated.5 entity of the burden of obesity in Italian rehabilitation units, but, hopefully, these initial data would serve to fuel discussion in our The geographical distribution of the rehabilitation institutes specialty on this growingly important challenge. who responded to the questionnaire is shown in Figure 1. Paolo CAPODAGLIO 1 *, Giuseppe VENTURA 1, The mean estimated prevalence of obesity among inpatients Maria L. PETRONI 2, Nicola CAU 3, Amelia BRUNANI 1 was around 30% in 13 of the 44 units, 10% in 14 units, 5% in 5 units, with the 12 remaining units (27%) not reporting BMI in 1Rehabilitation Unit, Istituto Auxologico Italiano, Piancavallo, medical charts. The rate of the general Italian population is 10.4%. Pordenone, Italy; 2Metabolic Rehabilitation, San Marino; We have to bear in mind that certain chronic conditions (i.e., spi- 3Politecnico of Milan, Milan, Italy nal cord injuries, amputations among others) may favor dramatic changes in body composition and the onset of obesity. *Corresponding author: Paolo Capodaglio, Rehabilitation Unit, Istituto Aux- ologico Italiano, Piancavallo, Italy, via Cadorna 90, 28824, Oggebbio, Italy. All of the responders strongly agreed that obesity does worsen E-mail: [email protected] disability and has a negative impact on the rehabilitation outcomes. This suggests those PMR specialists are aware of the additional References 1.  World Health Organization. Obesity and overweight. Fact sheet N°311. Updated March 2011 [Internet]. Available from: http://www.who.int/me- diacentre/factsheets/fs311/en/index.html [cited 2011, Nov 15]. 2.  WHO. | World report on disability. WHO [Internet]. Available from: http://www.who.int/disabilities/world_report/2011/en/ [cited 2014, Nov 8]. 3.  Stucki A, Daansen P, Fuessl M, Cieza A, Huber E, Atkinson R, et al. ICF Core Sets for obesity. J Rehabil Med 2004; (44 Suppl):107–13. 4.  Ministero della Salute. Quaderno del Ministero della Salute n° 10, 2012. 5.  Seida JC, Sharma AM, Johnson JA, Forhan M. Hospital rehabilitation for patients with obesity: a scoping review. Disabil Rehabil 2018;40:125–34. 138 European Journal of Physical and Rehabilitation Medicine February 2019

COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA LETTERS TO THE EDITOR 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 Conflicts of interest.—The authors certify that there is no conflict of inter- Table I.—Demographical and clinical characteristics of patients (mean 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,est with any financial organization regarding the material discussed in the ± standard deviation or number of cases and relevant percentage). 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. manuscript. Authors’ contributions.—Paolo Capodaglio, Maria L. Petroni and Amelia Age (years) 42.23±13.78 GOSE-R score Brunani conceived the survey and drafted the final letter, Giuseppe Ventura Age at the moment of the event (years) 41.77±13.66 and Nicola Cau organized the online survey and performed data collection Sex (males, females) 16 (64%), 9 (36%) and analysis. Time from the event (days) 106.7±87.4 Article first published online: August 27, 2018. - Manuscript accepted: July TBI (N. %) 27, 2018. - Manuscript received: June 13, 2018. CRS-R Score 12 (48%) (Cite this article as: Capodaglio P, Ventura G, Petroni ML, Cau N, Bru- GOSE-R Score 10.24±5.79 nani A. Prevalence and burden of obesity in Rehabilitation Units in Italy: a survey. Eur J Phys Rehabil Med 2019;55:137-9. DOI: 10.23736/S1973- 3.4±1.1 9087.18.05393-5) ple. Spearman correlation coefficient was used to correlate the © 2018 EDIZIONI MINERVA MEDICA scores of GOSE-R with CRS-R also with other parameters such Online version at http://www.minervamedica.it as age and time from the event. Mann-Whitney U-test was used European Journal of Physical and Rehabilitation Medicine 2019 February;55(1):139-40 for assessing differences between the two most common etiologies DOI: 10.23736/S1973-9087.18.05441-2 (TBI vs. hemorrhage). Alpha-level of significance was set at 0.05 for all the tests. The Glasgow Outcome Scale Extended- Revised (GOSE-R) to include Minimally Twenty-five subjects were enrolled (16 males and 9 females). Conscious State in the Vegetative State/ The mean age was 42.2±13.8 years, with a mean time from the Unresponsive Wakefulness Syndrome event of 106.7±87.4 days. The mean CRS-R score resulted category: a correlation with Coma 10.24±5.79, and the mean GOSE-R score 3.4±1.1. Recovery Scale-Revised (CRS-R) The correlation between CRS-R scores and GOSE-R scores The 40 years of application of the Glasgow Outcome Scale resulted statistically significant: R=0.895, P<0.001, with a coef- (GOS),1, 2 and of its extended version (GOSE)3 inspired the recent ficient of determination R2=0.801. proposal of a Glasgow Outcome Scale Extended-Revised (GOSE- R), to include the Minimally Conscious State in the Vegetative Neither CRS-R scores (R=-0.334, P=0.102) nor GOSE-R scores State category.4 GOSE-R expands the eight-point scale by sub- (R=-0.185, P=0.377) were found significantly correlated with age. classifying the Vegetative State5 in the two following categories Analogous results were found between time from the event and “better,” i.e. Minimally Conscious State (MCS) and “worse” (i.e. CRS-R score (R=-0.277, P=0.181) and GOSE-R score (R=-0.145, vegetative state), recently defined as Unresponsive Wakefulness P=0.490). Syndrome (UWS).6 Mann-Whitney U-Test did not highlight significant differences Finally, GOSE-R has been proposed, as a ten-point scale (Sup- between etiology in CRS-R score (P=0.776) and GOSE-R score plementary Digital Material 1: Supplementary Text), with the divi- (P=0.361) (Figure 1). sion of the Vegetative Scale (VS/UWS) category into three sub- categories: VS, Minimally Conscious State Minus (MCS -) and GOSE-R could promote and facilitate long-term follow-up Minimally Conscious State Plus (MCS +), suggesting to define the studies on larger populations of persons with severe brain injury exit from MCS, consisting of recovery of functional communica- and DoC, indeed, since the easiness and the brevity of the scale, it tion and/or functional objects use,7 as the severe disability lower could also be applied by telephone interviews and mails, according level (Supplementary Digital Material 1: Supplementary Text). to the recent interest on longitudinal assessment of clinical signs of recovery in patients with VS/UWS and MCS.9 The main limitation Aim of the study was the comparison of GOSE-R through cor- of this brief report is the small sample size of the study. However, relation with widely used scales for the evaluation of disorders of the positive findings encourage to apply the GOSE-R on larger consciousness (DoC), such as the Coma Recovery Scale-Revised (CRS-R).8 CRS-R score Figure 1.—Correlation between GOSE-R scores and CRS-R scores with We enrolled 25 patients with prolonged DoC, consecutively ad- the relevant regression line. mitted to the Post-acute Rehabilitation Unit of Santa Lucia Foun- dation, of different etiologies (Table I). Means and standard deviations were used to describe the sam- Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 139

COPYRIGHT© 2019 EDIZIONI MINERVA MEDICA LETTERS TO THE EDITOR 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 Conflicts of interest.—The authors certify that there is no conflict of inter- Table I.—Demographical and clinical characteristics of patients (mean 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,est with any financial organization regarding the material discussed in the ± standard deviation or number of cases and relevant percentage). 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. manuscript. Authors’ contributions.—Paolo Capodaglio, Maria L. Petroni and Amelia Age (years) 42.23±13.78 GOSE-R score Brunani conceived the survey and drafted the final letter, Giuseppe Ventura Age at the moment of the event (years) 41.77±13.66 and Nicola Cau organized the online survey and performed data collection Sex (males, females) 16 (64%), 9 (36%) and analysis. Time from the event (days) 106.7±87.4 Article first published online: August 27, 2018. - Manuscript accepted: July TBI (N. %) 27, 2018. - Manuscript received: June 13, 2018. CRS-R Score 12 (48%) (Cite this article as: Capodaglio P, Ventura G, Petroni ML, Cau N, Bru- GOSE-R Score 10.24±5.79 nani A. Prevalence and burden of obesity in Rehabilitation Units in Italy: a survey. Eur J Phys Rehabil Med 2019;55:137-9. DOI: 10.23736/S1973- 3.4±1.1 9087.18.05393-5) ple. Spearman correlation coefficient was used to correlate the © 2018 EDIZIONI MINERVA MEDICA scores of GOSE-R with CRS-R also with other parameters such Online version at http://www.minervamedica.it as age and time from the event. Mann-Whitney U-test was used European Journal of Physical and Rehabilitation Medicine 2019 February;55(1):139-40 for assessing differences between the two most common etiologies DOI: 10.23736/S1973-9087.18.05441-2 (TBI vs. hemorrhage). Alpha-level of significance was set at 0.05 for all the tests. The Glasgow Outcome Scale Extended- Revised (GOSE-R) to include Minimally Twenty-five subjects were enrolled (16 males and 9 females). Conscious State in the Vegetative State/ The mean age was 42.2±13.8 years, with a mean time from the Unresponsive Wakefulness Syndrome event of 106.7±87.4 days. The mean CRS-R score resulted category: a correlation with Coma 10.24±5.79, and the mean GOSE-R score 3.4±1.1. Recovery Scale-Revised (CRS-R) The correlation between CRS-R scores and GOSE-R scores The 40 years of application of the Glasgow Outcome Scale resulted statistically significant: R=0.895, P<0.001, with a coef- (GOS),1, 2 and of its extended version (GOSE)3 inspired the recent ficient of determination R2=0.801. proposal of a Glasgow Outcome Scale Extended-Revised (GOSE- R), to include the Minimally Conscious State in the Vegetative Neither CRS-R scores (R=-0.334, P=0.102) nor GOSE-R scores State category.4 GOSE-R expands the eight-point scale by sub- (R=-0.185, P=0.377) were found significantly correlated with age. classifying the Vegetative State5 in the two following categories Analogous results were found between time from the event and “better,” i.e. Minimally Conscious State (MCS) and “worse” (i.e. CRS-R score (R=-0.277, P=0.181) and GOSE-R score (R=-0.145, vegetative state), recently defined as Unresponsive Wakefulness P=0.490). Syndrome (UWS).6 Mann-Whitney U-Test did not highlight significant differences Finally, GOSE-R has been proposed, as a ten-point scale (Sup- between etiology in CRS-R score (P=0.776) and GOSE-R score plementary Digital Material 1: Supplementary Text), with the divi- (P=0.361) (Figure 1). sion of the Vegetative Scale (VS/UWS) category into three sub- categories: VS, Minimally Conscious State Minus (MCS -) and GOSE-R could promote and facilitate long-term follow-up Minimally Conscious State Plus (MCS +), suggesting to define the studies on larger populations of persons with severe brain injury exit from MCS, consisting of recovery of functional communica- and DoC, indeed, since the easiness and the brevity of the scale, it tion and/or functional objects use,7 as the severe disability lower could also be applied by telephone interviews and mails, according level (Supplementary Digital Material 1: Supplementary Text). to the recent interest on longitudinal assessment of clinical signs of recovery in patients with VS/UWS and MCS.9 The main limitation Aim of the study was the comparison of GOSE-R through cor- of this brief report is the small sample size of the study. However, relation with widely used scales for the evaluation of disorders of the positive findings encourage to apply the GOSE-R on larger consciousness (DoC), such as the Coma Recovery Scale-Revised (CRS-R).8 CRS-R score Figure 1.—Correlation between GOSE-R scores and CRS-R scores with We enrolled 25 patients with prolonged DoC, consecutively ad- the relevant regression line. mitted to the Post-acute Rehabilitation Unit of Santa Lucia Foun- dation, of different etiologies (Table I). Means and standard deviations were used to describe the sam- Vol. 55 - No. 1 European Journal of Physical and Rehabilitation Medicine 139

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, LETTERS TO THE EDITOR 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. populations. Moreover, GOSE-R can be administered on people 5.  Monti MM, Laureys S, Owen AM. The vegetative state. BMJ with DoC, also without a specific training as required for CRS-R . 2010;341:c3765. . Its easiness of use will likely guarantee further 40 years of applica- 6.  Laureys S, Celesia GG, Cohadon F, Lavrijsen J, León-Carrión J, San- tion and popularity as in GOS and GOSE in outcome studies on nita WG, et al.; European Task Force on Disorders of Consciousness. Un- severe brain injury, also with chronic DoC. responsive wakefulness syndrome: a new name for the vegetative state or apallic syndrome. BMC Med 2010;8:68. . Rita FORMISANO 1 *, Marianna CONTRADA 1, 2, 7.  Bruno MA, Vanhaudenhuyse A, Thibaut A, Moonen G, Laureys S. Giulia FERRI 1, Sara SCHIATTONE 1, From unresponsive wakefulness to minimally conscious PLUS and func- Marco IOSA 3, Marta ALOISI 1 tional locked-in syndromes: recent advances in our understanding of dis- orders of consciousness. J Neurol 2011;258:1373–84. . 1Post-Coma Unit, IRCCS Santa Lucia Foundation, Rome, Italy; 8.  Giacino JT, Kalmar K, Whyte J. The JFK Coma Recovery Scale-Re- 2Department of Behavioral Neuroscience, Sapienza vised: measurement characteristics and diagnostic utility. Arch Phys Med Rehabil 2004;85:2020–9. . University, Rome, Italy; 3Clinical Laboratory of Experimental 9.  Bagnato S, Boccagni C, Sant’Angelo A, Fingelkurts AA, Fingelkurts Neurorehabilitation, IRCCS Santa Lucia Foundation, Rome, Italy AA, Galardi G. Longitudinal assessment of clinical signs of recovery in patients with unresponsive wakefulness syndrome after traumatic or non- *Corresponding author: Rita Formisano, IRCCS Santa Lucia Foundation, traumatic brain injury. J Neurotrauma 2017;34:535–9. . via Ardeatina 306, 00179 Rome, Italy. E-mail: [email protected] Conflicts of interest.—The authors certify that there is no conflict of inter- References est with any financial organization regarding the material discussed in the manuscript. 1.  Jennett B, Bond M. Assessment of outcome after severe brain damage. Funding.—This work has received funding from the European Union’s Hori- Lancet 1975;1:480–4. . zon 2020 research and innovation programme under the Marie Skłodowska- 2.  McMillan T, Wilson L, Ponsford J, Levin H, Teasdale G, Bond M. The Curie grant agreement No 778234. Glasgow Outcome Scale - 40 years of application and refinement. Nat Rev Article first published online: October 30, 2018. - Manuscript accepted: Oc- Neurol 2016;12:477–85. . tober 30, 2018. - Manuscript revised: October 16, 2018. - Manuscript re- 3.  Jennett B, Snoek J, Bond MR, Brooks N. Disability after severe head ceived: July 16, 2018. injury: observations on the use of the Glasgow Outcome Scale. J Neurol For supplementary materials, please see the HTML version of this article at Neurosurg Psychiatry 1981;44:285–93. . www.minervamedica.it 4.  Formisano R, Aloisi M, Ferri G, Schiattone S, Contrada M. The (Cite this article as: Formisano R, Contrada M, Ferri G, Schiattone S, Iosa Glasgow Outcome Scale Extended-Revised (GOSE-R) to include mini- M, Aloisi M. The Glasgow Outcome Scale Extended-Revised (GOSE-R) mally conscious state in the vegetative state category. J Neurol Sci to include Minimally Conscious State in the Vegetative State/Unresponsive 2018;388:22. . Wakefulness Syndrome category: a correlation with Coma Recovery Scale- Revised (CRS-R). Eur J Phys Rehabil Med 2019;55:139-40. DOI: 10.23736/ S1973-9087.18.05441-2) 140 European Journal of Physical and Rehabilitation Medicine February 2019