Journal of Physiotherapy 62 (2016) 115 Journal of PHYSIOTHERAPY journal homepage: www.elsevier.com/locate/jphys Appraisal Critically Appraised Papers A 2-year physical activity program for sedentary older adults does not improve cognitive functioning more than a health education program Synopsis Summary of: Sink KM, Espeland MA, Castro CM, Church T, outcome measures included executive function (measured with Cohen R, Dodson JA, et al. Effect of a 24-month physical activity three tasks to assess speed of processing and executive function on a intervention vs health education on cognitive outcomes in sedentary laptop computer), a composite global function score, and the older adults: the LIFE randomized trial. JAMA. 2015;314:781-790. incidence of mild cognitive impairment or dementia at 24 months. Results: A total of 1476 participants (90%) completed the study. Question: Does a physical activity program result in better Median attendance at physical activity sessions was 71%. At cognitive function than a health education program in sedentary 24 months, there were no between-group differences for the Digital older adults living in the community? Design: Randomised, Symbol Coding Task (MD –0.01 points, 95% CI –0.80 to 0.77) or the controlled trial with concealed allocation and blinded assessment. Hopkins Verbal Learning Test for delayed word recall (MD –0.03 Setting: Eight community centres in the United States of America. words, 95% CI –0.29 to 0.24) or immediate word recall (MD –0.14 Participants: The study included sedentary men and women, aged words, 95% CI –0.58 to 0.29). There were no between-group 70 to 89 years, who were at risk of mobility disability but were able differences for global and executive function measures, although to walk 400 m. The key exclusion criterion was having a diagnosis of sub-group analyses indicated benefits in executive function dementia or significant cognitive impairment based on the Modified associated with physical activity for those aged at least 80 years Mini-Mental State Examination score. Randomisation of 1635 parti- and for those scoring <8 on the Short Physical Performance Battery. cipants allocated 818 to the physical activity group and 817 to the There was no difference in the incidence of mild cognitive health education group. Interventions: The physical activity group impairment or dementia between the physical activity group participated in a 24-month program of twice-weekly, centre-based (13.2%) and the health education group (12.1%) (OR 1.08, 95% CI sessions of up to 30 minutes of moderate intensity walking, 0.80 to 1.46). Conclusion: A 24-month physical activity program for 10 minutes of lower limb resistance training, and 10 minutes of older sedentary adults did not improve cognitive functioning balance and flexibility exercises. In addition, the physical activity compared with a health education program. group were prescribed home exercises 3 to 4 times per week. The health education group attended weekly sessions for 26 weeks and Provenance: Invited. Not peer reviewed. monthly sessions thereafter. Sessions consisted of presentations, demonstrations or field trips. Outcome measures: The primary Nicholas Taylor outcomes for this report were cognitive outcomes assessed at Associate Editor, Journal of Physiotherapy baseline and 24 months. Cognitive outcomes included the Digital Symbol Coding task (scored from 0 to 133 points) and the Hopkins http://dx.doi.org/10.1016/j.jphys.2015.12.008 Verbal Learning Test (scored from 0 to 12 words). Secondary Commentary Previous randomised, controlled trials have suggested that Overall, the LIFE Study results support the notion that sustained physical activity benefits cognition in older adults. However, key engagement in structured physical activity promotes healthy limitations of studies to date include small samples and short ageing. First, it clearly demonstrated that structured physical intervention periods.1–3 Sink and colleagues recently published activity has benefits on incident mobility disability. Second, it secondary cognitive data from the LIFE Study, the largest showed that the cognitive benefits of structured physical activity (n = 1635) and longest (ie, 24 months) trial of exercise to date could still be reaped in late life (ie, in the eighth decade of life). As in sedentary older adults without cognitive impairment but with experts in exercise prescription to maximise function and against lower-extremity functional limitations at baseline. the backdrop of the global ageing population, physiotherapists need to embrace the critical role they have in the global promotion The primary finding of the LIFE Study was that moderate- of cognitive and physical health in older adults. intensity structured physical activity significantly reduced major mobility disability.4 Physical activity may have also benefited Provenance: Invited. Not peer reviewed. executive functions in those: (1) aged 80-89 years; and (2) with baseline lower-extremity functional limitations. John R. Best and Teresa Liu-Ambrose Department of Physical Therapy, University of British Columbia, Collectively, these results suggest that the intensity and volume of LIFE’s structured physical activity intervention were Vancouver, Canada insufficient to produce positive changes in cognition, except in References the oldest and frailest. Higher intensity and volume of physical activity may be necessary to induce positive effects among 1. Smith PJ, et al. Psychosom Med. 2010;72:239–252. younger and less frail older adults. Notably, the study did not 2. Gates N, et al. Am J Geriatr Psychiatry. 2013;21:1086–1097. demonstrate whether the intervention had meaningful effects on 3. Nagamatsu LS, et al. Br J Sports Med. 2014;48(12):943–944. objective measures of cardiorespiratory fitness or muscular 4. Pahor M, et al. JAMA. 2014;311:2387–2396. strength, which are necessary to confer benefits to brain 5. Erickson KI, et al. Proc Natl Acad Sci. 2011;108(7):3017–3022. health.5,6 6. Liu-Ambrose T, et al. Arch Intern Med. 2010;170:170–178. http://dx.doi.org/10.1016/j.jphys.2015.12.007 1836-9553/ß 2016 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 62 (2016) 115 Journal of PHYSIOTHERAPY journal homepage: www.elsevier.com/locate/jphys Appraisal Critically Appraised Papers A 2-year physical activity program for sedentary older adults does not improve cognitive functioning more than a health education program Synopsis Summary of: Sink KM, Espeland MA, Castro CM, Church T, outcome measures included executive function (measured with Cohen R, Dodson JA, et al. Effect of a 24-month physical activity three tasks to assess speed of processing and executive function on a intervention vs health education on cognitive outcomes in sedentary laptop computer), a composite global function score, and the older adults: the LIFE randomized trial. JAMA. 2015;314:781-790. incidence of mild cognitive impairment or dementia at 24 months. Results: A total of 1476 participants (90%) completed the study. Question: Does a physical activity program result in better Median attendance at physical activity sessions was 71%. At cognitive function than a health education program in sedentary 24 months, there were no between-group differences for the Digital older adults living in the community? Design: Randomised, Symbol Coding Task (MD –0.01 points, 95% CI –0.80 to 0.77) or the controlled trial with concealed allocation and blinded assessment. Hopkins Verbal Learning Test for delayed word recall (MD –0.03 Setting: Eight community centres in the United States of America. words, 95% CI –0.29 to 0.24) or immediate word recall (MD –0.14 Participants: The study included sedentary men and women, aged words, 95% CI –0.58 to 0.29). There were no between-group 70 to 89 years, who were at risk of mobility disability but were able differences for global and executive function measures, although to walk 400 m. The key exclusion criterion was having a diagnosis of sub-group analyses indicated benefits in executive function dementia or significant cognitive impairment based on the Modified associated with physical activity for those aged at least 80 years Mini-Mental State Examination score. Randomisation of 1635 parti- and for those scoring <8 on the Short Physical Performance Battery. cipants allocated 818 to the physical activity group and 817 to the There was no difference in the incidence of mild cognitive health education group. Interventions: The physical activity group impairment or dementia between the physical activity group participated in a 24-month program of twice-weekly, centre-based (13.2%) and the health education group (12.1%) (OR 1.08, 95% CI sessions of up to 30 minutes of moderate intensity walking, 0.80 to 1.46). Conclusion: A 24-month physical activity program for 10 minutes of lower limb resistance training, and 10 minutes of older sedentary adults did not improve cognitive functioning balance and flexibility exercises. In addition, the physical activity compared with a health education program. group were prescribed home exercises 3 to 4 times per week. The health education group attended weekly sessions for 26 weeks and Provenance: Invited. Not peer reviewed. monthly sessions thereafter. Sessions consisted of presentations, demonstrations or field trips. Outcome measures: The primary Nicholas Taylor outcomes for this report were cognitive outcomes assessed at Associate Editor, Journal of Physiotherapy baseline and 24 months. Cognitive outcomes included the Digital Symbol Coding task (scored from 0 to 133 points) and the Hopkins http://dx.doi.org/10.1016/j.jphys.2015.12.008 Verbal Learning Test (scored from 0 to 12 words). Secondary Commentary Previous randomised, controlled trials have suggested that Overall, the LIFE Study results support the notion that sustained physical activity benefits cognition in older adults. However, key engagement in structured physical activity promotes healthy limitations of studies to date include small samples and short ageing. First, it clearly demonstrated that structured physical intervention periods.1–3 Sink and colleagues recently published activity has benefits on incident mobility disability. Second, it secondary cognitive data from the LIFE Study, the largest showed that the cognitive benefits of structured physical activity (n = 1635) and longest (ie, 24 months) trial of exercise to date could still be reaped in late life (ie, in the eighth decade of life). As in sedentary older adults without cognitive impairment but with experts in exercise prescription to maximise function and against lower-extremity functional limitations at baseline. the backdrop of the global ageing population, physiotherapists need to embrace the critical role they have in the global promotion The primary finding of the LIFE Study was that moderate- of cognitive and physical health in older adults. intensity structured physical activity significantly reduced major mobility disability.4 Physical activity may have also benefited Provenance: Invited. Not peer reviewed. executive functions in those: (1) aged 80-89 years; and (2) with baseline lower-extremity functional limitations. John R. Best and Teresa Liu-Ambrose Department of Physical Therapy, University of British Columbia, Collectively, these results suggest that the intensity and volume of LIFE’s structured physical activity intervention were Vancouver, Canada insufficient to produce positive changes in cognition, except in References the oldest and frailest. Higher intensity and volume of physical activity may be necessary to induce positive effects among 1. Smith PJ, et al. Psychosom Med. 2010;72:239–252. younger and less frail older adults. Notably, the study did not 2. Gates N, et al. Am J Geriatr Psychiatry. 2013;21:1086–1097. demonstrate whether the intervention had meaningful effects on 3. Nagamatsu LS, et al. Br J Sports Med. 2014;48(12):943–944. objective measures of cardiorespiratory fitness or muscular 4. Pahor M, et al. JAMA. 2014;311:2387–2396. strength, which are necessary to confer benefits to brain 5. Erickson KI, et al. Proc Natl Acad Sci. 2011;108(7):3017–3022. health.5,6 6. Liu-Ambrose T, et al. Arch Intern Med. 2010;170:170–178. http://dx.doi.org/10.1016/j.jphys.2015.12.007 1836-9553/ß 2016 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 62 (2016) 103–110 Journal of PHYSIOTHERAPY journal homepage: www.elsevier.com/locate/jphys Research Activity preferences, lifestyle modifications and re-injury fears influence longer-term quality of life in people with knee symptoms following anterior cruciate ligament reconstruction: a qualitative study Stephanie R Filbay a, Kay M Crossley a,b, Ilana N Ackerman c,d a The School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane; b The College of Science, Health and Engineering, La Trobe University, Australia; c Melbourne EpiCentre, The University of Melbourne, TF_ID]A8[$ ustralia; d Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia KEY WORDS ABSTRACT Return to sport Questions: How do people with knee symptoms describe their quality of life and experiences 5 to Knee injuries 20 years after anterior cruciate ligament reconstruction (ACLR)? What factors impact upon the quality of Psychological adaptation life of these people? Design: Qualitative study. Participants: Seventeen people with knee symptoms 5 to Fear of re-injury 20 years after ACLR and high (n = 8) or low (n = 9) quality of life scores were recruited from a cross- Osteoarthritis sectional study. Methods: Semi-structuredF_2DT$I][ telephone interviews were conducted and transcribed. The data obtained from the interventions underwent inductive coding and thematic analysis. Results: Four consistent themes emerged from the interviews as common determinants of quality of life following ACLR: physical activity preferences; lifestyle modifications; adaptation and acceptance; and fear of re- injury. All participants described the importance of maintaining a physically active lifestyle and the relationship between physical activity and quality of life. Participants who avoided sport or activity reported experiencing reduced quality of life. Participants who suppressed or overcame re-injury fears to continue sport participation described experiencing a satisfactory quality of life while taking part in sport despite knee symptoms. For some participants, resuming competitive sport resulted in subsequent knee trauma, anterior cruciate ligament re-rupture or progressive deterioration of knee function, with negative impacts on quality of life following sport cessation. Participants who enjoyed recreational exercise often adapted their lifestyle early after ACLR, while others described adapting their lifestyle at a later stage to accommodate knee impairments; this was associated with feelings of acceptance and satisfaction, irrespective of knee symptoms. Conclusion: Activity preferences, lifestyle modifications and fear of re-injury influenced quality of life in people with knee symptoms up to 20 years following ACLR. People with a preference for competitive sport who do not enjoy recreational exercise might be at heightened risk of poor quality of life outcomes and could benefit from support to facilitate a transition to a physically active, satisfying lifestyle. [Filbay SR, Crossley KM, Ackerman IN (2016) Activity preferences, lifestyle modifications and re-injury fears influence longer-term quality of life in people with knee symptoms following anterior cruciate ligament reconstruction: a qualitative study. Journal of Physiotherapy 62: 103–110] Crown Copyright ß 2016 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by- nc-nd/4.0/). Introduction experience an ACL graft rupture or contralateral ACL rupture within 15 years.14 Subsequent knee surgery, including revision Anterior cruciate ligament (ACL) rupture most commonly occurs in adolescents and young adults during competitive sport ACLR and contralateral ACLR, is associated with worse patient- participation.1–3FID[]_8$T Estimates of the prevalence of ACL reconstruc- tion (ACLR) in Australia are alarmingly high, exceeding other reported outcomes, including reduced longer-term quality of countries with a rate of 52 per F1D1[]$I0T_ 00 000 inhabitants.1,3–7 life.15–21 A recent systematic review reported impaired knee- Optimising longer-term quality of life following ACLR is impor- related quality of life 5 to 20 years after ACLR,22 but no studies tant, considering the potential for persistent physical and psychological difficulties and the high rates of early-onset knee investigating the impact of return to sport on longer-term quality osteoarthritis after ACL injury.8–12 As many as one in five people who undergo ACLR require subsequent knee surgery within of life following ACLR were identified. This is despite less than half 6 years.13 Furthermore, one in four people who undergo ACLR of non-elite sports participants returning to competitive sport after ACLR,11 which contrasts most patients’ expectations of full return to sport within 1 year of surgery.23 Young, active people undergoing ACLR commonly have unrealistic expectations (including a low likelihood of ongoing pain or instability and a http://dx.doi.org/10.1016/j.jphys.2016.02.011 1836-9553/Crown Copyright ß 2016 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
104 Filbay et al: Quality of life after cruciate ligament reconstruction low risk of developing post-traumatic osteoarthritis)23 and these an alias for use in data transcription and reporting. Interview may impact on their quality of life outcomes. durations ranged from 16 to 41 minutes. Current patient-reported measures of quality of life have A standard interview schedule (Appendix 1 on the eAddenda) limited ability to capture individual expectations and do not provided the framework for each interview, which covered four specifically evaluate the importance of knee-related impairments broad areas. Questions about perioperative experiences addressed: to the individual.24,25 Despite the breadth of quantitative literature ACL injury and initial management; satisfaction with surgery and about ACL injury, qualitative studies exploring personal perspec- healthcare providers; ACLR preparation, expectations and experi- tives following ACLR are rare and no qualitative studies have ence; and postoperative experiences. Questions about sport and investigated quality of life following ACLR. Previous qualitative exercise addressed: return to sport; experiences of sport and studies have focused on return to sport following ACLR.26–29 One exercise participation; and physical activity priorities, motives and study identified fear, a change in life priorities, and personality importance across the lifespan. Questions about psychological traits as factors that influenced people’s decision to return to sport impacts addressed emotions, fears and confidence in the injured after ACLR.26 Two small studies of five rugby players27 and five knee. Questions about current experience addressed: current knee elite adolescent alpine skiers28 identified high confidence in the symptoms and function; lifestyle modifications; management injured knee as a key facilitator and low confidence as a key barrier strategies; knowledge; and information. Participants were also for returning to sport.27,28 Similarly, a study interviewing 17 female given the opportunity to contribute any additional information at handball players described confidence in the capabilities of one’s the end of the interview. body as a key factor influencing decisions to return to sport after ACLR.29 While several qualitative studies have investigated factors Participants influencing the decision to return to sport after ACLR, it is unclear how these factors affect longer-term quality of life, particularly Participants were purposively sampled from a larger cross- among people with ongoing knee symptoms or limitations.26–29 sectional study of 162 people who had undergone ACLR 5 to Qualitative research could enhance the understanding of factors 20 years previously.301_TD]F$I[ The eligibility criteria for this cross-sectional that impact negatively on quality of life after ACLR and provide study required all participants toT]F_D$3I[ be aged 18 to 55 years; have information with which to guide management strategies in order received an ACLR or revision surgery 5 to 20 years previously; and to optimise outcomes following ACL injury. report knee symptoms or functional limitations on the Knee Injury and Osteoarthritis Outcome Score (KOOS), determined by a Therefore, the research questions for this qualitative study predefined cut-off criterion.30 This cut-off criterion required were: reporting less than optimal scores for at least 50% of questions on any two KOOS subscales, corresponding to cut-off values of 1. How do people with knee symptoms describe their quality of life 86.1 (pain), 85.7 (symptoms), 86.8 (activities of daily living), and experiences 5 to 20 years after ACLR? 85.0 (sport/recreation), and 87.5 (quality of life). Recruitment details and participant characteristics for the cross-sectional study 2. What factors affect quality of life in people with knee symptoms have been reported previously.30 These 162 participants complet- 5 to 20 years following ACLR? ed a battery of questionnaires including the KOOS and the Anterior Cruciate Ligament Quality of Life questionnaire (ACL-QOL), which Methods are valid and reliable for use in people who have undergone ACLR.31–33 Demographic, lifestyle and return-to-sport data were Design also collected and relevant questionnaire responses from the cross- sectional study were presented for each participant (Table 1). Interviews were conducted from August to October 2014, after obtaining informed verbal consent from each participant. Two To enable comparisons between people with high and low pilot interviews were conducted to refine the broad themes, knee-related quality of life, those with high or low ACL-QOL scores structural order and outline of the interview. A single investigator were sampled specifically. The ACL-QOL scores were ranked and (SRF) performed the semi-structured telephone interviews and the first people selected for the qualitative study were those with transcribed the audio recordings. This interviewer had no ACL-QOL scores in the tenth and ninetieth percentiles, followed by involvement with the clinical care of any study participant. All those with the next highest and lowest ACL-QOL scores, transcripts were de-identified, and each participant was assigned respectively. In total, 16 people with high ACL-QOL scores Table 1 Participant characteristics. Alias ACL-QOL score (0 to 100) Age (yr) Time since last ACLR (yr) Gender Body mass index category Return to sport ACLR type Lucy 25 33 6 female normal lower primary Flynn 26 44 16 male obese no revision x 2 Sue 26 41 18 female obese no Claire 27 34 6 female obese no primary Will 27 50 12 male normal primary Kate 28 26 12 female obese lower primary Jack 29 41 11 male overweight lower primary Nick 29 25 6 male obese lower contralateral Hugh 30 32 6 male overweight contralateral yes revision + contralateral Zara 83 50 13 female obese lower Ross 83 35 6 male overweight primary Beth 86 49 5 female yes primary Amy 87 28 10 female normal yes primary Mary 87 42 18 female normal yes primary Emma 88 23 8 female normal lower revision Tina 90 33 5 female normal yes primary Guy 92 38 8 male normal lower primary overweight lower primary yes ACL-QOL = Anterior Cruciate Ligament Quality of Life questionnaire (0 = worst, 100 = best), ACLR = anterior cruciate ligament reconstruction. For ‘Return to sport’, participants selected one of the following options: I returned to competitive sport at the same or higher level than before ACL injury = Yes; I returned to competitive sport at a lower level than before ACL injury = Lower; I did not return to competitive sport after my ACL reconstruction = No. The horizontal line separates those with low ACL-QOL scores (above) from those with high ACL-QOL scores (below).
Research 105 (score range 82 to 92) and 14 people with low ACL-QOL scores 10 years (SD 5, range 6 to 18) previously. In comparison, those (score range 25 to 30) were sent an email invitation to take part in a reporting high ACL-QOL scores had a similar mean age of 37 years telephone interview. Of these, five declined due to other (SD 10, range 23 to 50), 38% were male, fewer were classified as commitments (four with high ACL-QOL scores and one with a overweight or obese (38%) and they had a similar mean of 9 years low ACL-QOL score) and eight did not respond (four from each (SD 5, range 5 to 18) since ACLR was reported. Participant group). All people who agreed to take part were interviewed for the characteristics are presented in Table 1 and patient-reported study, resulting in 17 interviews (eight with a high ACL-QOL score outcomes are described in Table 2. and nine with a low ACL-QOL score). To ensure an appropriate sample size, recruitment was stopped when no new themes Key themes emerged over two consecutive interviews for participants with high ACL-QOL scores and for participants with low ACL-QOL scores. Four key themes emerged from the interviews: physical activity preferences; lifestyle modifications; adaptation and acceptance; Data analysis and fear of re-injury (Table 3). These themes and related sub- themes are described with supporting quotes from the interviews The first stage of data analysis involved general inductive in the following sections. Additional quotes supporting each theme analysis, where multiple readings, reviewing and data interpreta- are provided as an online appendix (Appendix 2 on the eAddenda). tion were undertaken to identify themes arising from the data.34 Inductive coding was supported by commercial softwarea.35 This Physical activity preferences coding process resulted in the identification of key themes and sub-themes. The coding structure was revised and refined Within the theme of physical activity preferences, two throughout the data interpretation process to reduce redundancy, contrasting sub-themes were apparent from the interviews. The identify new emerging sub-categories and incorporate new first involved 11 people who described a strong preference for themes and insights. A hierarchical approach to coding was used, participation in competitive sports in comparison to recreational linking themes with a commonality or causal relationship to assist exercise: with pattern recognition.35 Following a minimum of two rounds of coding and analysis by a single investigator (SRF), a second I do love netball. I hate the gym. Absolutely hate it. It makes it pretty investigator coded a random sample of six interview transcripts hard when you can’t play the sports that you love, which I don’t (INA) and any contrasting opinions in identified themes were consider to really be exercise, and you’ve got to find alternatives to discussed. Where possible, conceptual maps were developed to exercise which I can’t stand. (Claire) assist with data interpretation and provide potential explanations for key themes. The contrasting sub-theme comprised a group of six people who preferred, enjoyed or were satisfied taking part in non- Results competitive recreational exercise. While some of these people had participated in team sports in the past, they did not display a strong Participants preference for returning to competitive sport and described satisfaction with being physically active through recreational Interview data were available from 17 participants. People with exercise: low ACL-QOL scores had a mean age of 36 years (SD 8, range 25 to 50), 56% were male, 78% were overweight or obese and these Oh look I probably could have played, but to me that was a fairly participants had received their most recent ACLR an average of major injury that had me off for a long time from doing exercise, and the exercise that I like doing (recreational exercise), and I said I didn’t want to risk doing it a second time. (Beth) Table 2 Patient-reported outcomes. Alias Most important Limitation in most important Current knee KOOS Pain KOOS Symptoms KOOS QOL Trouble with Lifestyle activity activity (0 to 100) satisfaction (0 to 100) (0 to 100) (0 to 100) knee confidence modification Lucy Flynn weightlifting 12 yes 75 50 38 severe mild Sue AFL football 0 yes 69 50 31 mod total Claire running/gym 10 yes 83 68 44 mod mod Will 0 yes 75 75 31 severe severe Kate netball 0 no 53 39 25 severe severe Jack running 0 no 83 68 38 severe mod Nick running 0 no 67 54 31 severe severe Hugh soccer 6 yes 83 54 50 mod mod 15 no 58 57 44 mod mod Zara rugby Ross skiing 70 yes 86 54 75 mild none Beth 100 yes 89 79 75 none mod Amy dancing 93 yes 94 79 88 none mild Mary running 80 yes 92 79 75 mild none Emma running 88 yes 83 75 69 none mild Tina 100 yes 83 89 75 mild none Guy yoga 77 yes 100 71 69 mild mild AFL football 100 yes 86 54 81 none none netball running soccer AFL = Australian Football League, KOOS = Knee-injury and Osteoarthritis Outcome Score (all domains: 0 = worst to 100 = best). ‘Most important activity’ was identified by question 19 on the ACL-QOL – most important sport or recreational activity that you do or wish to do. ‘Limitation in most important activity’ was rated by how limited are you in playing that number one sport or activity? from 0 = totally limited to 100 = not limited at all. ‘Current knee satisfaction’ was asked by Taking into account your level of pain and also your functional impairment, if you were to remain for the next few months as you are today, would you consider that your current state is satisfactory?. ‘Trouble with knee confidence’ was rated by Item 3 of the KOOS-QOL subscale, How much are you troubled with lack of confidence in your knee?, with answers restricted to not at all (none), mildly (mild), moderately (mod), severely (severe) and extremely. ‘Lifestyle modifications’ was rated by Item 2 of the KOOS-QOL subscale, Have you modified your lifestyle to avoid potentially damaging activities to your knee?, with answers restricted to not at all (none), mildly (mild), moderately (mod), severely (severe) and totally. The horizontal line separates those with low ACL-QOL scores (above) from those with high ACL-QOL scores (below).
106 Filbay et al: Quality of life after cruciate ligament reconstruction Table 3 An overview of common themes related to quality of life. Theme Sub-theme Low ACL-QOL (n) High ACL-QOL (n) Physical activity preferences Preference for competitive sport 9 2 Lifestyle modificationsa Enjoys recreational exercise D$IT[06]F_ 6 Negative lifestyle modifications 6 1 Positive lifestyle modifications 3 7 Adaptation and acceptance Early adaptation 05 Delayed adaptation 71 Fear of re-injurybT_7$DIF[] No adaptation 22 Fear accommodation 54 Fear suppression 42 Fear avoidance 54 ACL-QOL = Anterior Cruciate Ligament Quality of Life questionnaire. Note, some participants described transitioning between sub-themes over time: a Within the theme of lifestyle modifications, four participants made negative lifestyle modifications for years but had made positive lifestyle modifications at the time of interview; b Within the theme of fear of re-injury, six described fear suppression or avoidance behaviours for years but had developed fear-accommodation behaviours at the time of interview. Lifestyle modifications tended to manifest in a sense of satisfaction with their current knee function and quality of life: Ten participants reported having modified their way of life because of their knee in ways that had improved their quality of life I started doing dancing; that’s helped a lot. Now I do Pilates (positive knee-related lifestyle modifications), while seven people regularly, so that’s helped with the movement and kept it going, so I had made modifications that impacted negatively upon their quality do have fairly good movement now, and I have a fairly active of life (negative knee-related lifestyle modifications) (Table 3). lifestyle. (Zara) A conceptual diagram summarising the concepts of activity Adaptation and acceptance preferences and lifestyle modifications with potential impacts on quality of life is presented in Figure 1. Six people with a preference A strong theme of personal adaptation and acceptance also for competitive sport described transitioning to an inactive emerged from the interviews. Five people described having lifestyle after sport cessation. For several participants, this adapted their lifestyle with ease following ACLR; while eight resulted in weight gain, exacerbation of knee symptoms and a people took years to adapt their lifestyle in line with their knee’s reduced motivation to exercise with negative impacts on quality abilities and this delayed adaptation often resulted in a later of life: improvement in satisfaction and quality of life. The remaining four people did not describe any knee-related lifestyle adaptations You know, I’m disappointed that I didn’t go back, and I’m during their interview. disappointed that I swapped the lifestyle instead of keeping up with the sporting lifestyle. I went to a social lifestyle, and started putting Early adaptation the weight on, because now I’m at the stage where I’ve got too A sub-group of five people described adapting their lifestyle much weight. I’ve got worse knee issues. I’m not helping it by being overweight. It certainly made me change my lifestyle. (Sue) within a short period of time following ACLR and attaining a state of satisfaction with their knee. For people with a preference for In contrast, the six people who enjoyed recreational exercise [(Figure_1)TD$IG]were often able to make positive lifestyle modifications. This Anterior cruciate ligament rupture Preference for Anterior cruciate Preference for recreational exercise ligament reconstruction competitive sport Adopts or maintains Returns to sport at Returns to sport at a an active a desired or level of competition or lifestyle satisfying level of performance lower competition and than desired or with performance fear of re-injury Adopts an Ceases inactive lifestyle sport Figure 1. A conceptual diagram portraying common interactions between activity preferences, lifestyle modifications and quality of life. Red boxes and arrows represent a current state with a tendency towards poor knee-related quality of life. Green boxes and arrows represent a current state with a tendency towards satisfactory knee-related quality of life. Orange arrows represent periods of transition between green and red states. Note: arrows represent common paths of transition described by study participants. An individual may remain in a given state (box) without transitioning. The directions of arrows represent the most common pathways.
Research 107 recreational exercise, a short transitional period (indicated by made delayed accommodations after years of fear-avoidance orange arrows in Figure 1) was more frequently described: behaviour (n = 3) or after experiencing subsequent knee trauma following return to unrestrictive competitive sport (n = 3). This I was quite happy to give up netball and touch football because I subsequent trauma resulted in increased fear of re-injury, which just was not going to go through that again and didn’t want to do it led to activity modifications that reduced perceived risk of re- again, and switched to cycling and running. (Beth) injury: Delayed adaptation I had the opportunity to keep going through but then, I was just like Eight people described knee-related difficulties after surgery, well you know, I’ve already had two, I’m 18 years of age, you know, what sort of future have I got if I blow it out again at 18.... so I went but acknowledged that they had grown to accept their current yeah, no I’m just going to go play back at state level and leave it, knee state having adapted their lifestyle over a period of years: and yeah, just walked away. (Flynn) I was about 90 kilos, I was very, very overweight, and like I’m 53 kg Fear suppression now. I just started exercising again, and eating well. Since I lost Six participants with a strong desire to continue participating in weight, my knee has never locked again. I just decided one day that that was enough and I just started exercising. I’ve gone from what I high-impact competitive sport described an ability to overcome or feel like 10% quality of life, to 100% quality of life, for me, being suppress initial fear of re-injury in order to continue sport active is everything. (Lucy) participation. While continuing sport participation, these people experienced a satisfactory quality of life, irrespective of the Fear of re-injury presence of physical knee symptoms. However, participating in unrestrictive competitive sport often resulted in subsequent knee The final key theme related to a fear of re-injury, which all trauma, ACL re-rupture or a progressive exacerbation of knee participants reported they had experienced at some point after symptoms. As outlined in Figure 2, this resulted in one of two ACLR. Participants described their knee-related fears, how these transitions: increased knee awareness, delayed activity adaptation fears changed over time, and how fear had impacted on their and satisfactory quality of life (n = 3); or reduced knee function and activity choices and quality of life. Three common sub-themes adoption of an inactive lifestyle with negative impacts on quality of emerged in relation to fear of re-injury: fear accommodation, fear life (n = 3): suppression, and fear avoidance. Figure 2 provides a conceptual diagram summarising the most common fear of re-injury I adapted my game. I ran in straight lines. I played like that for a behaviours in the context of physical activity preferences, lifestyle long time, for many years, until, umm, I couldn’t really run at all, modifications and potential impacts on quality of life. and I was having trouble just walking anywhere, so it had got quite bad. (Will) Fear accommodation Fear avoidance A subgroup of participants, over time, had become mindful of Nine people described ceasing sport participation due to a fear their knee-related fears and modified their movement patterns or of re-injury. Of these, five people remained physically active in activity choices to minimise risk of subsequent knee injury. These recreational exercise, despite ceasing competitive sport, and this accommodations enabled people to maintain an active lifestyle had minimal impact on their quality of life, while four people and participate in desired activities, resulting in satisfaction with transitioned to an inactive lifestyle, with further negative impacts current knee function and quality of life (Figure 2). Three on their quality of life (Figure 2). All people who described pregiF([_2)TD$IG]u articipants made early accommodations after ACLR, while six Anterior cruciate ligament reconstruction Reduce or eliminate fear of Fear of re-injury Avoid sport participation re-injury by modifying due to fear of re-injury movement patterns or Overcome or suppress fear participating in a lower of re-injury in order to Fear avoidance level of competition continue unrestricted sport participation Ceases sport Fear accommodation Fear suppression Early Unrestricted sport participation accommodation Additional knee trauma or re-rupture Delayed Continues sport and experiences a accommodation progressive decline in knee function Maintains an active Adopts an Maintains an active Adopts an lifestyle inactive lifestyle lifestyle inactive lifestyle Figure 2. A conceptual diagram portraying described experiences of fear of re-injury and relationships with lifestyle modifications and quality of life. Red boxes and arrows represent a current state with a tendency towards poor knee-related quality of life. Green boxes and arrows represent a current state with a tendency towards satisfactory knee-related quality of life. Note: arrows represent common paths of transition described by study participants. An individual may remain in a given state (box) without transitioning.
108 Filbay et al: Quality of life after cruciate ligament reconstruction fear[TDF]_I3-$1 avoidance behaviours and a strong preference for competitive fear-related patterns following ACLR may enable clinicians to sport over recreational exercise described a transition to an facilitate the transition to a satisfying, physically active lifestyle, inactive lifestyle: with the potential to improve knee-related quality of life. Irrespective of sport or recreational exercise preference, main- So what sport or exercise do you do now? (Interviewer) taining enjoyable regular physical activity appears to be para- At the moment, pretty much, nothing. I’m always a bit cagey still. mount in optimising quality of life following ACLR. Further I’m always, it’s always in the back of my mind, watch your knee, research is needed to develop and evaluate strategies that watch your knee. (Sue) physiotherapists and other healthcare professionals can utilise to improve longer-term quality of life following ACLR. Discussion Participation in regular physical activity and sport is associated This is the first study to explore the impact of the ACLR on longer- with less depressive symptoms, reduced rates of obesity and better term quality of life using a qualitative approach. The obtained data health-related quality of life.47–49 Weight gain was described as a highlighted the contribution of physical activity preferences, consequence of reduced physical activity by 65% of participants lifestyle modification, and fear of re-injury to quality of life and and this became a key feature in a cycle involving exacerbation of overall satisfaction in people with knee symptoms after ACLR. For knee symptoms, reduced motivation, increased fear of re-injury many people, ACL rupture marked the beginning of persistent knee and poor quality of life. Notably, some people reported being difficulties that required ongoing self-management and consider- inactive and having impaired quality of life for more than 10 years ation. A shift toward this realisation was accompanied by before reaching a state of acceptance and satisfactory quality of acceptance, adaptation and improvement in quality of life. However, life. Participants described a number of self-management strate- the period of transition was variable. While some people achieved gies that facilitated their transition to a state of acceptance; the this relatively quickly, others required more than a decade to do so most common strategies involved reducing their expectations to and some people remained dissatisfied with their knee state at the allow participation in lower risk physical activity, adapting their time of interview. A unique perspective was gained into the goals to accommodate knee impairments, accepting a revised trajectories of quality of life over time that has not been identified in sporting role such as coaching or refereeing, or shifting focus to previous ACLR studies. This allowed identification of key points of other aspects of life, including work or family life. This state of transition, where intervention to facilitate positive lifestyle mod- acceptance and adaptation resulted in a satisfactory quality of life, ifications could be most beneficial. irrespective of physical knee symptoms. People who were able to transition reasonably quickly after ACLR to a satisfactory state Fear of re-injury and psychological readiness are common were often those who enjoyed recreational exercise, or those who barriers to returning to pre-injury sport after ACLR.10,12,36–392D1T[I]_F$ In described early fear-accommodation behaviours. As depicted in fact, the contribution of psychological factors to an individual’s Figure 1, there appear to be clear periods of transition where the decision to return to sport may be of greater importance than implementation of management strategies could be beneficial for physical limitations such as pain and instability.40–42 Tanner et al43 facilitating an active lifestyle with minimal impact on quality of investigated the importance of knee-specific questionnaire items life. to people with an ACL rupture and found that fear of re-injury was considered most important to patients before and after ACLR. Interestingly, 29% of participants were satisfied with their knee Additionally, a recent qualitative study by Tjong et al26 inter- function, despite reporting: very low KOOS-QOL and ACL-QOL viewed 31 people who underwent ACLR at least 2 years previously scores; high levels of pain and symptoms; moderate to severe and found that fear was the most commonly reported reason for trouble with knee confidence; and extreme limitations taking part patients not to return to sport. Similar to our findings, a sub-group in their most important sport or activity (Table 2). This mismatch of people in that study described overcoming initial fears in order may reflect these people having reached a state of acceptance and to return to sport.26 However, due to the shorter follow-up time, adaptation, irrespective of knee impairments. Considering the high the study by Tjong and colleagues was unable to evaluate longer- rates of osteoarthritis development that have been reported as term consequences and outcomes. Although fear of re-injury is early as 10 years following ACLR50 and the chronic nature of commonly perceived as an unfavourable consequence of ACLR, our osteoarthritis, understanding ways of optimising quality of life and data suggest that, for some people, this could actually serve as a satisfaction in people with knee symptoms after ACLR is of great protective mechanism for optimising future knee function. value. Exploring differences between people who are satisfied and However, large longitudinal studies would be required to reliably dissatisfied with their knee, despite reporting impaired patient- test this hypothesis. reported outcomes, could provide information that may be useful in developing strategies to optimise quality of life in people with It is possible that the experience of undergoing ACLR and the persistent knee symptoms after ACLR. subsequent postoperative period contributed to the fear of re- injury described by participants in the present study. On average, A strength of the study was the purposive sampling strategy, similar knee-related quality of life has been reported between ACL- which enabled the identification of common and contrasting reconstructed groups and non-operatively managed groups 5 to experiences amongst people with high and low knee-related 20 years after ACL injury.44 However, comparisons between quality of life scores. Another strength was the ability to use psychological traits, behaviours and re-injury fears in those who multiple patient-reported outcome measures collected as part of a report poor quality of life after ACLR and non-operative manage- larger cross-sectional study to aid in the interpretation of ment have not been performed and may prove a fruitful area for findings.30 The semi-structured interviews were devised to future research. Our data indicate that people with a strong capture the full array of personal experiences from ACL injury to preference for competitive sport who do not enjoy recreational the time of interview, providing insights into changes in quality of exercise and display fear-avoidance behaviours might be at risk of life over time that have not previously been captured using poorer quality of life outcomes. Negative impacts on quality of life traditional quantitative measures. A limitation of the study was the became apparent, irrespective of physical activity preferences, possibility of recall bias, given the time period since ACL injury. As when fear-avoidance behaviours extended beyond sporting all participants were English-speaking Australians it is not known activity to recreational exercise and activities of daily living. whether these findings are generalisable beyond this setting. As These relationships are supported by quantitative study findings, with all qualitative research, our cohort of participants was where fear-avoidance beliefs were related to function in sport and unlikely to be representative of all people who have undergone activities of daily living45 and a high fear of re-injury was ACLR, particularly given that study participants all experienced a associated with poor knee-related quality of life after ACLR.46 degree of knee symptoms or activity limitations and were Identifying people with strong activity preferences displaying specifically recruited with high or low quality of life scores. Future quantitative research could assist in further exploring the
Research 109 themes identified and evaluating long-term (> 20 year) quality of 15. Faltstrom A, Hagglund M, Kvist J. Patient-reported knee function, quality of life, life outcomes after ACLR. and activity level after bilateral anterior cruciate ligament injuries. Am J Sports Med. 2013;41:2805–2813. What is already known on this topic: Anterior cruciate ligament rupture is common. Knee symptoms and impairment 16. Ahlden M, Samuelsson K, Sernert N, Forssblad M, Karlsson J, Kartus J. The of quality of life may persist for decades after anterior cruciate Swedish National Anterior Cruciate Ligament Register: a report on baseline ligament reconstruction. variables and outcomes of surgery for almost 18,000 patients. Am J Sports Med. What this study adds: Among people with knee symptoms 2012;40:2230–2235. after anterior cruciate ligament reconstruction, quality of life appears to be influenced by their physical activity preferences, 17. Mo¨ ller E, Weidenhielm L, Werner S. Outcome and knee-related quality of life after lifestyle modification and fear of re-injury. Those who find a anterior cruciate ligament reconstruction: a long-term follow-up. Knee Surg Sports way of returning to some form of exercise, such as by replacing Traumatol Arthrosc. 2009;17:786–794. competitive sport with recreational exercise, tend to report better quality of life than those who do not exercise. 18. Wasserstein D, Huston LJ, Nwosu S, Spindler KP. KOOS pain as a marker for significant knee pain two and six years after primary ACL reconstruction: a Footnotes: aNVivo version 10, QSR International Pty Ltd, Multicenter Orthopaedic Outcomes Network (MOON) prospective longitudinal Melbourne, Australia. cohort study. Osteoarthr Cartilage. 2015. eAddenda: Appendices 1 and 2 can be found online at doi:10. 19. Kievit AJ, Jonkers FJ, Barentsz JH, Blankevoort L. A cross-sectional study comparing 1016/j.jphys.2016.02.011 the rates of osteoarthritis, laxity, and quality of life in primary and revision anterior cruciate ligament reconstructions. Arthroscopy. 2013;29:898–905. Ethics approval: The University of Queensland Institutional Human Research Ethics Committee provided ethical approval. All 20. Gifstad T, Drogset JO, Viset A, Grøntvedt T, Hortemo GS. Inferior results after participants gave informed verbal consent before data collection revision ACL reconstructions: a comparison with primary ACL reconstructions. began. Knee Surg Sports Traumatol Arthrosc. 2012;1–8:. Competing interests: Nil. 21. Lind M, Menhert F, Pedersen AB. Incidence and outcome after revision anterior Source(s) of support: Physiotherapy Research Foundation (PRF) cruciate ligament reconstruction: Results from the Danish registry for knee liga- Seeding Grant S13-004. ment reconstructions. Am J Sports Med. 2012;40:1551–1557. Acknowledgements: Nil. Provenance: Not invited. Peer reviewed. 22. Filbay SR, Ackerman IN, Russell TG, Macri EM, Crossley KM. Health-related quality Correspondence: Stephanie Filbay, The School of Health and of life after anterior cruciate ligament reconstruction: a systematic review. Am J Rehabilitation Sciences, The University of Queensland, Brisbane, Sports Med. 2014;42:1247–1255. Australia. Email: stephanie.fi[email protected] 23. Feucht MJ, Cotic M, Saier T, Minzlaff P, Plath JE, Imhoff AB, et al. Patient expecta- References tions of primary and revision anterior cruciate ligament reconstruction. Knee Surg Sports Traumatol Arthrosc. 2016;24:201–207. 1. Granan LP, Forssblad M, Lind M, Engebretsen L. The Scandinavian ACL registries 2004-2007: baseline epidemiology. Acta Orthop. 2009;80:563–567. 24. Carr AJ, Gibson B, Robinson PG. Measuring quality of life is quality of life deter- mined by expectations or experience? BMJ. 2001;322:1240–1243. 2. Renstrom P, Ljungqvist A, Arendt E, Beynnon B, Fukubayashi T, Garrett W, et al. Non-contact ACL injuries in female athletes: An International Olympic Committee 25. Carr A, Higginson IJ. Are quality of life measures patient centred? BMJ. current concepts statement. Brit J Sports Med. 2008;42:394–412. 2001;322:1357–1360. 3. Janssen KW, Orchard JW, Driscoll TR, van Mechelen W. High incidence and costs for 26. Tjong VK, Murnaghan ML, Nyhof-Young JM, Ogilvie-Harris DJ. A qualitative inves- anterior cruciate ligament reconstructions performed in Australia from 2003-2004 tigation of the decision to return to sport after anterior cruciate ligament recon- to 2007-2008: Time for an anterior cruciate ligament register by Scandinavian struction: to play or not to play. Am J Sports Med. 2014;42:336–342. model? Scand J Med Sci Sports. 2012;22:495–501. 27. Carson F, Polman R. Experiences of professional rugby union players returning to 4. Lind M, Menhert F, Pedersen AB. The first results from the Danish ACL reconstruc- competition following anterior cruciate ligament reconstruction. Phys Ther Sport. tion registry: epidemiologic and 2 year follow-up results from 5,818 knee ligament 2012;13:35–40. reconstructions. Knee Surg Sports Traumatol Arthrosc. 2009;17:117–124. 28. Nordahl B, Sjostrom R, Westin M, Werner S, Alricsson M. Experiences of returning 5. Granan LP, Bahr R, Steindal K, Furnes O, Engebretsen L. Development of a national to elite alpine skiing after ACL injury and ACL reconstruction. Int J Adolesc Med cruciate ligament surgery registry: the Norwegian National Knee Ligament Regis- Health. 2014;26:69–77. try. Am J Sports Med. 2008;36:308–315. 29. Thing LF. Voices of the broken body. The resumption of non-professional 6. Csintalan R, Inacio M, Funahashi T. Incidence rate of anterior cruciate ligament female players’ sports careers after anterior cruciate ligament injury. The reconstructions. Perm J. 2008;12:17–21. female player’s dilemma: is she willing to run the risk? Scand J Med Sci Sports. 2006;16:364–375. 7. Gianotti SM, Marshall SW, Hume PA, Bunt L. Incidence of anterior cruciate ligament injury and other knee ligament injuries: A national population-based study. J Sci 30. Filbay SR, Ackerman IN, Russell TG, Crossley KM. Factors related to quality of life in Med Sport. 2009;12:622–627. people with knee pain, stiffness or activity limitations 5 to 20 years following anterior cruciate ligament reconstruction. Osteoarthr Cartilage. 2015;23(Suppl. 8. Øiestad BE, Engebretsen L, Storheim K, Risberg MA. Knee osteoarthritis after 2):A279–A280. anterior cruciate ligament injury: A systematic review. Am J Sports Med. 2009;37:1434–1443. 31. Roos E, Lohmander S. The Knee injury and Osteoarthritis Outcome Score (KOOS): from joint injury to osteoarthritis. Health Qual Life Outcomes. 2003;1:64. 9. Lohmander LS, Ostenberg A, Englund M, Roos H. High prevalence of knee osteoar- thritis, pain, and functional limitations in female soccer players twelve years after 32. Mohtadi N. Development and validation of the quality of life outcome measure anterior cruciate ligament injury. Arthritis Rheum. 2004;50:3145–3152. (questionnaire) for chronic anterior cruciate ligament deficiency. Am J Sports Med. 1998;26:350–357. 10. Ardern CL, Taylor NF, Feller JA, Webster KE. Fear of re-injury in people who have returned to sport following anterior cruciate ligament reconstruction surgery. J Sci 33. Salavati M, Akhbari B, Mohammadi F, Mazaheri M, Khorrami M. Knee injury and Med Sport. 2012;15:488–495. Osteoarthritis Outcome Score (KOOS); reliability and validity in competitive athletes after anterior cruciate ligament reconstruction. Osteoarthr Cartilage. 11. Ardern CL, Taylor NF, Feller JA, Webster KE. Fifty-five per cent return to competitive 2011;19:406–410. sport following anterior cruciate ligament reconstruction surgery: An updated systematic review and metaanalysis including aspects of physical functioning and 34. Thomas DR. A general inductive approach for analyzing qualitative evaluation contextual factors. Brit J Sports Med. 2014. data. Am J Eval. 2006;27(2):237–246. 12. Kvist J, Ek A, Sporrstedt K, Good L. Fear of re-injury: A hindrance for returning to 35. Bazeley P, Jackson K. Qualitative data analysis with NVivo. Los Angeles: SAGE; 2013. sports after anterior cruciate ligament reconstruction. Knee Surg Sports Traumatol 36. McCullough KA, Phelps KD, Spindler KP, Matava MJ, Dunn WR, Parker RD, et al. Arthrosc. 2005;13:393–397. Return to high school- and college-level football after anterior cruciate ligament 13. Hettrich CM, Dunn WR, Reinke EK, Group MOON, Spindler KP. The rate of subse- reconstruction: A Multicenter Orthopaedic Outcomes Network (MOON) cohort quent surgery and predictors after anterior cruciate ligament reconstruction: two- study. Am J Sports Med. 2012;40:2523–2529. and six-year follow-up results from a multicenter cohort. Am J Sports Med. 37. Tripp DA, Stanish W, Ebel-Lam A, Brewer BW, Birchard J. Fear of reinjury, negative 2013;41:1534–1540. affect, and catastrophizing predicting return to sport in recreational athletes with anterior cruciate ligament injuries at 1 year postsurgery. Rehabil Psychol. 14. Bourke HE, Salmon LJ, Waller A, Patterson V, Pinczewski LA. Survival of the anterior 2007;52:74–81. cruciate ligament graft and the contralateral ACL at a minimum of 15 years. Am J 38. Ardern CL, Taylor NF, Feller JA, Whitehead TS, Webster KE. Psychological responses Sports Med. 2012;40:1985–1992. matter in returning to preinjury level of sport after anterior cruciate ligament reconstruction surgery. Am J Sports Med. 2013;41:1549–1558. 39. Lentz TA, Zeppieri G, George SZ, Tillman SM, Moser MW, Farmer KW, et al. Comparison of physical impairment, functional, and psychosocial measures based on fear of reinjury/lack of confidence and return-to-sport status after ACL recon- struction. Am J Sports Med. 2015;43:345–353. 40. Gobbi A, Francisco R. Factors affecting return to sports after anterior cruciate ligament reconstruction with patellar tendon and hamstring graft: a prospec- tive clinical investigation. Knee Surg Sports Traumatol Arthrosc. 2006;14: 1021–1028. 41. Warner SJ, Smith MV, Wright RW, Matava MJ, Brophy RH. Sport-specific out- comes after anterior cruciate ligament reconstruction. Arthroscopy. 2011;27:1129–1134. 42. Ardern CL, Webster KE, Taylor NF, Feller JA. Return to the preinjury level of competitive sport after anterior cruciate ligament reconstruction surgery: Two- thirds of patients have not returned by 12 months after surgery. Am J Sports Med. 2011;39:538–543. 43. Tanner S, Dainty K, Marx R, Kirkley A. Knee-specific quality-of-life instruments Which ones measure symptoms and disabilities most important to patients? Am J Sports Med. 2007;35:1450–1458.
110 Filbay et al: Quality of life after cruciate ligament reconstruction 44. Filbay SR, Culvenor AG, Ackerman IN, Russell TG, Crossley KM. Quality of life in 48. Mutrie N, Hannah MK. The importance of both setting and intensity of physical anterior cruciate ligament-deficient people: a systematic review and meta-analy- activity in relation to non-clinical anxiety and depression. Int J Health Promot Ed. sis. Brit J Sports Med. 2015;49:1033–1041. 2007;45:24. 45. Ross MD. The relationship between functional levels and fear-avoidance beliefs 49. Huffman GR, Park J, Roser-Jones C, Sennett BJ, Yagnik G, Webner D. Normative SF- following anterior cruciate ligament reconstruction. J Orthop Traumatol. 36 values in competing NCAA intercollegiate athletes differ from values in the 2010;11:237–243. general population. J Bone Joint Surg Am. 2008;90:471–476. 46. Kvist J, Ek A, Sporrstedt K, Good L. Fear of re-injury: a hindrance for returning to 50. Øiestad BE, Holm I, Aune AK, Gunderson R, Myklebust G, Engebretsen L, et al. Knee sports after anterior cruciate ligament reconstruction. Knee Surg Sports Traumatol function and prevalence of knee osteoarthritis after anterior cruciate ligament Arthrosc. 2005;13:393–397. reconstruction: A prospective study with 10 to 15 years of follow-up. Am J Sports Med. 2010;38:2201–2210. 47. de Wit LM, Fokkema M, van Straten A, Lamers F. Depressive and anxiety disorders and the association with obesity, physical, and social activities. Depress Anxiety. 2010;27:1057–1065.
Journal of Physiotherapy 62 (2016) 113 Journal of PHYSIOTHERAPY journal homepage: www.elsevier.com/locate/jphys Appraisal Critically Appraised Papers Aerobic and resistance exercise improve walking speed and endurance in people with multiple sclerosis Synopsis Summary of: Pearson M, Dieberg G, Smart N. Exercise as analysis demonstrated a statistically significant difference in gait therapy for improvement of walking ability in adults with multiple speed in favour of exercise, based on the 10-m walk test (]FTIDe_$7[ ight sclerosis: A meta-analysis. Arch Phys Med Rehabil. 2015;96: 1339- intervention groups, MD –1.76 seconds, 95% CI_TD8IF$[] –2.47 toID$F_9][T –1.06) and 1348. the 500-m walk test (0F$]T1[IDo_ ne intervention group, MD –0.31 seconds, 95% CI –0.59 to –0.03), but not the timed 25-foot walk (five Objective: To quantify the benefits of exercise for improving intervention groups, MD –0.59 seconds, 95% CI –2.55 to 1.36). walking speed and endurance in adults with multiple sclerosis. There was also a statistically significant difference in walking Data FI]sD$T[3_ ources: PubMed, EMBASE, CINAHL, SPORTDiscus, and the endurance in favour of exercise for both the 6-minute walk test Cochrane Central Register of Controlled Trials, searched up to (four intervention groups, MD $[1IDFT_]36 m, 95% CI ID1[T2F]_$15 to ][F15D3$I_T 8) and the 2- March 31, 2014. This search was supplemented by searching minute walk test (five intervention groups, MD D1I$][F41_T 2 m, 95% CI []T_5I51F$D to reference lists of retrieved publications. Study F]4I_DT[$selection: Random- D2_]FTI[$16 0). Differences in the Timed Up and Go test were not significant ised, controlled trials that compared exercise with no intervention (five intervention groups, MD –1.05 seconds, 95% CI –2.19 to 0.09). or waitlist control in ambulatory adults with multiple sclerosis, The strongest effects of exercise on gait speed were found for with at least 10 subjects in the post-intervention analysis, and interventions combining aerobic and resistance training, and for published in English. Exercise intervention was defined as aerobic interventions > 12 weeks in duration. Conclusion: Exercise inter- and endurance training, resistance training, aquatics and yoga. ventions improve walking speed (10-m walk test, 500-m walk test) Exercise involving treadmill, robot assistance, or other specialised and endurance (2-minute and 6-minute walk tests) in people with techniques (eg, functional electrical stimulation) were excluded. multiple sclerosis. Combination interventions may be optimal. Outcome measures for gait speed were the 10-m walk test, timed 25-foot walk test, and 500-m walk test; measures for endurance Provenance: Invited. Not peer reviewed. were the 2-minute and 6-minute walk tests. Functional mobility was assessed with the Timed Up and Go test. Data e[_DIFT]5$ xtraction: Prudence Plummer Study quality was assessed using a modified Physiotherapy Division of Physical Therapy, University of North Carolina Evidence Database score (maximum score of 8). Data 6$DsT]_I[F ynthesis: Of the 437 studies initially identified by the search, 13 studies with at Chapel Hill, USA a total of 655 participants with multiple sclerosis (357 exercise, 298 controls) met selection criteria and were included in the http://dx.doi.org/10.1016/j.jphys.2015.12.006 review. Quality assessment scores ranged from 5 to 8. Meta- Commentary Walking dysfunction is a common, life-altering feature of Overall, this meta-analysis provides estimates of exercise effects multiple sclerosis. Accordingly, the improvement of walking on walking outcomes in multiple sclerosis. The results can inform function is a major focus of ongoing rehabilitation research. power analyses for future randomised, controlled trials and inform Physiological deconditioning (eg, aerobic capacity) is a primary clinical practice recommendations. Clinically, exercise is a consid- determinant of walking dysfunction in multiple sclerosis,1 and erably less expensive option for managing walking outcomes in exercise represents a behavioural approach for reversing decon- multiple sclerosis than pharmacotherapy and can be prescribed as ditioning and thereby improving ambulation. The meta-analysis part of ongoing, comprehensive multiple sclerosis care. There are by Pearson and colleagues quantified the improvement in walking many benefits of exercise in multiple sclerosis,3 and this meta- performance across randomised, controlled trials of exercise in analysis further underscores its importance for comprehensively multiple sclerosis. The meta-analysis included 13 trials and improving outcomes, including walking, in this neurological disease. suggested that exercise can improve walking speed, walking endurance, and perhaps mobility (eg, Timed Up and Go test) in Provenance: Invited. Not peer reviewed. people with multiple sclerosis. Robert Motl This meta-analysis did not consider disease status or fitness Department of Kinesiology and Community Health, University of adaptations as effect moderators. This is important for under- standing if exercise is effective across the disease spectrum, Illinois at Urbana-Champaign, USA particularly considering that drugs are ineffective for slowing eventual progression of mobility disability in moderate and severe References multiple sclerosis.2 The focus on fitness would provide information on changes in physiological functioning as possible mechanisms 1. Motl RW. Exerc Sport Sci Rev. 2010;38:186–191. for walking improvements. 2. Motl RW. Phys Med Rehabil Clin N Am. 2013;24:325–336. 3. Motl RW. Expert Rev Neurother. 2014;14:1429–1436. http://dx.doi.org/10.1016/j.jphys.2015.12.005 1836-9553/ß 2016 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 62 (2016) 113 Journal of PHYSIOTHERAPY journal homepage: www.elsevier.com/locate/jphys Appraisal Critically Appraised Papers Aerobic and resistance exercise improve walking speed and endurance in people with multiple sclerosis Synopsis Summary of: Pearson M, Dieberg G, Smart N. Exercise as analysis demonstrated a statistically significant difference in gait therapy for improvement of walking ability in adults with multiple speed in favour of exercise, based on the 10-m walk test (]FTIDe_$7[ ight sclerosis: A meta-analysis. Arch Phys Med Rehabil. 2015;96: 1339- intervention groups, MD –1.76 seconds, 95% CI_TD8IF$[] –2.47 toID$F_9][T –1.06) and 1348. the 500-m walk test (0F$]T1[IDo_ ne intervention group, MD –0.31 seconds, 95% CI –0.59 to –0.03), but not the timed 25-foot walk (five Objective: To quantify the benefits of exercise for improving intervention groups, MD –0.59 seconds, 95% CI –2.55 to 1.36). walking speed and endurance in adults with multiple sclerosis. There was also a statistically significant difference in walking Data FI]sD$T[3_ ources: PubMed, EMBASE, CINAHL, SPORTDiscus, and the endurance in favour of exercise for both the 6-minute walk test Cochrane Central Register of Controlled Trials, searched up to (four intervention groups, MD $[1IDFT_]36 m, 95% CI ID1[T2F]_$15 to ][F15D3$I_T 8) and the 2- March 31, 2014. This search was supplemented by searching minute walk test (five intervention groups, MD D1I$][F41_T 2 m, 95% CI []T_5I51F$D to reference lists of retrieved publications. Study F]4I_DT[$selection: Random- D2_]FTI[$16 0). Differences in the Timed Up and Go test were not significant ised, controlled trials that compared exercise with no intervention (five intervention groups, MD –1.05 seconds, 95% CI –2.19 to 0.09). or waitlist control in ambulatory adults with multiple sclerosis, The strongest effects of exercise on gait speed were found for with at least 10 subjects in the post-intervention analysis, and interventions combining aerobic and resistance training, and for published in English. Exercise intervention was defined as aerobic interventions > 12 weeks in duration. Conclusion: Exercise inter- and endurance training, resistance training, aquatics and yoga. ventions improve walking speed (10-m walk test, 500-m walk test) Exercise involving treadmill, robot assistance, or other specialised and endurance (2-minute and 6-minute walk tests) in people with techniques (eg, functional electrical stimulation) were excluded. multiple sclerosis. Combination interventions may be optimal. Outcome measures for gait speed were the 10-m walk test, timed 25-foot walk test, and 500-m walk test; measures for endurance Provenance: Invited. Not peer reviewed. were the 2-minute and 6-minute walk tests. Functional mobility was assessed with the Timed Up and Go test. Data e[_DIFT]5$ xtraction: Prudence Plummer Study quality was assessed using a modified Physiotherapy Division of Physical Therapy, University of North Carolina Evidence Database score (maximum score of 8). Data 6$DsT]_I[F ynthesis: Of the 437 studies initially identified by the search, 13 studies with at Chapel Hill, USA a total of 655 participants with multiple sclerosis (357 exercise, 298 controls) met selection criteria and were included in the http://dx.doi.org/10.1016/j.jphys.2015.12.006 review. Quality assessment scores ranged from 5 to 8. Meta- Commentary Walking dysfunction is a common, life-altering feature of Overall, this meta-analysis provides estimates of exercise effects multiple sclerosis. Accordingly, the improvement of walking on walking outcomes in multiple sclerosis. The results can inform function is a major focus of ongoing rehabilitation research. power analyses for future randomised, controlled trials and inform Physiological deconditioning (eg, aerobic capacity) is a primary clinical practice recommendations. Clinically, exercise is a consid- determinant of walking dysfunction in multiple sclerosis,1 and erably less expensive option for managing walking outcomes in exercise represents a behavioural approach for reversing decon- multiple sclerosis than pharmacotherapy and can be prescribed as ditioning and thereby improving ambulation. The meta-analysis part of ongoing, comprehensive multiple sclerosis care. There are by Pearson and colleagues quantified the improvement in walking many benefits of exercise in multiple sclerosis,3 and this meta- performance across randomised, controlled trials of exercise in analysis further underscores its importance for comprehensively multiple sclerosis. The meta-analysis included 13 trials and improving outcomes, including walking, in this neurological disease. suggested that exercise can improve walking speed, walking endurance, and perhaps mobility (eg, Timed Up and Go test) in Provenance: Invited. Not peer reviewed. people with multiple sclerosis. Robert Motl This meta-analysis did not consider disease status or fitness Department of Kinesiology and Community Health, University of adaptations as effect moderators. This is important for under- standing if exercise is effective across the disease spectrum, Illinois at Urbana-Champaign, USA particularly considering that drugs are ineffective for slowing eventual progression of mobility disability in moderate and severe References multiple sclerosis.2 The focus on fitness would provide information on changes in physiological functioning as possible mechanisms 1. Motl RW. Exerc Sport Sci Rev. 2010;38:186–191. for walking improvements. 2. Motl RW. Phys Med Rehabil Clin N Am. 2013;24:325–336. 3. Motl RW. Expert Rev Neurother. 2014;14:1429–1436. http://dx.doi.org/10.1016/j.jphys.2015.12.005 1836-9553/ß 2016 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 62 (2016) 114 Journal of PHYSIOTHERAPY journal homepage: www.elsevier.com/locate/jphys Appraisal Critically Appraised Papers A supervised exercise program may not add any benefit over advice for patients recovering from ankle fracture Synopsis Summary of: Moseley AM, Beckenkamp PR, Haas M, Herbert RD, program of five sessions over 4 weeks, which was individually Lin CWC (EXACT team). Rehabilitation after immobilization for tailored, prescribed, monitored, and progressed by a physiotherapist ankle fracture. The EXACT randomized clinical trial. JAMA. in the hospital outpatient service. Outcome measures: Primary 2015;314(13):1376-1385. doi:10.1001/jama.2015.12180. outcomes were the Lower Extremity Functional Scale (range 0 to 80, higher scores indicate better activity) and quality of life, measured by Questions: For patients with an isolated ankle fracture, does a the Assessment of Quality of Life instrument (range, 0 to 1, higher supervised exercise program and advice (rehabilitation) lead to scores indicate better quality of life), which were assessed by a greater improvements in activity limitation than advice alone? Is any blinded assessor at 1, 3 and 6 months after randomisation, with the improvement different with respect to fracture severity, age and primary time point at 3 months. Results: A total of 173 patients (80%) gender? Design: A multicentre, randomised, controlled trial with completed the 3-month follow-up, and 170 (79%) completed the 6- concealed allocation. Setting: Seven Australian hospitals. Partici- month follow-up. The mean difference in activity limitation was pants: The inclusion criteria were: isolated ankle fracture treated 0.4 units (95% CI –3.3 to 4.1) at 3 months and 0.2 units (95% CI –3.4 to with immobilisation (with or without surgical fixation); immobilisa- 3.9) at 6 months, with no difference between the two groups. Mean tion removed on the day of recruitment; approval received from the difference in quality of life was –0.01 units (–0.06 to 0.04) at both orthopaedic surgeon to bear weight as tolerated or bear partial 3 and 6 months. Treatment effects were not moderated by fracture weight; reduced ankle dorsiflexion range of motion (at least 30 mm severity or age and gender. Conclusion: A supervised exercise less motion compared with the non-fractured leg, using the weight- program did not confer additional benefits in activity limitation or bearing lunge method); ankle pain at least 2 out of 10 when 50% of quality of life compared with advice alone for patients with isolated body weight was borne through the affected leg; completed skeletal and uncomplicated ankle fracture. growth; and no concurrent pathologies that would affect the ability to perform everyday tasks. Randomisation allocated 106 to a Provenance: Invited. Not peer reviewed. supervised exercise program and advice (rehabilitation) and 108 to advice alone. Interventions: The advice group had a single session of Margreth Grotlea and Ka˚ re Birger Hagenb self-management advice about exercise and return to activity. The aOslo and Akershus University College of Applied Sciences, session was provided by a physiotherapist in the fracture clinic and also included advice on exercises for ankle movement in non-weight- Department of Physiotherapy, Oslo, Norway bearing positions and a handout summarising the advice with text bNational Advisory Unit on Rehabilitation in Rheumatology, and illustrations. The rehabilitation intervention consisted of the Department of Rheumatology, Diakonhjemmet Hospital, Oslo, Norway same advice in the fracture clinic, in addition to a supervised exercise http://dx.doi.org/10.1016/j.jphys.2016.01.002 ß 2016 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/). Commentary reported improvements in activity limitations and quality of life during the follow-up time, although no between-group differences were As the benefits of rehabilitation after immobilisation for ankle found. The lack of a control group receiving no treatment or advice fracture are unclear, Moseley and colleagues conducted a pragmatic, cannot rule out time as a factor and might explain the improvement randomised clinical trial to determine the effectiveness of a supervised seen in both groups; however, this would have included ethical exercise program and advice compared with advice alone. This well- considerations. Importantly, results from this study indicate that designed trial provides robust evidence that a supervised exercise routine care for patients after isolated and uncomplicated ankle program provides no additional improvement in activity limitation or fracture should include self-management advice at the time of removal quality of life compared with advice alone, which is in line with findings of immobilisation but not a supervised exercise program. from other studies evaluating rehabilitation after the removal of immobilisation.1 Nevertheless, in the present study, nearly one-third of Provenance: Invited. Not peer reviewed. the participants in the advice group received out-of-trial physiothera- Silje Stensrud py, which could potentially have diluted the observed effect. The authors hypothesised that older women or people with a more severe Oslo and Akershus University College of Applied Sciences, Department fracture may benefit more from rehabilitation, but no treatment effects of Physiotherapy, Oslo, Norway were associated with fracture severity or age and gender. This is in contrast with a previous study showing that younger patients (< References 40 years) could benefit from a postoperative training program.2 1. Lin CWC, et al. Cochrane Database Syst Rev. 2012;11:CD005595. Although recovery from ankle fracture is initially rapid, the recovery 2. Nilsson GM, et al. BMJ Muskuloskelet Disord. 2009;10:118. slows with time and can be incomplete 2 years after fracture.3 A 3. Beckenkamp PR, et al. J Orthop Sports Phys Ther. 2014;44:841–851. prolonged physiotherapy intervention or a delayed initiation could be trialed, as five additional sessions during the first 4 weeks after http://dx.doi.org/10.1016/j.jphys.2016.01.001 immobilisation may be insufficient. Both groups in the present study 1836-9553/ß 2016 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 62 (2016) 114 Journal of PHYSIOTHERAPY journal homepage: www.elsevier.com/locate/jphys Appraisal Critically Appraised Papers A supervised exercise program may not add any benefit over advice for patients recovering from ankle fracture Synopsis Summary of: Moseley AM, Beckenkamp PR, Haas M, Herbert RD, program of five sessions over 4 weeks, which was individually Lin CWC (EXACT team). Rehabilitation after immobilization for tailored, prescribed, monitored, and progressed by a physiotherapist ankle fracture. The EXACT randomized clinical trial. JAMA. in the hospital outpatient service. Outcome measures: Primary 2015;314(13):1376-1385. doi:10.1001/jama.2015.12180. outcomes were the Lower Extremity Functional Scale (range 0 to 80, higher scores indicate better activity) and quality of life, measured by Questions: For patients with an isolated ankle fracture, does a the Assessment of Quality of Life instrument (range, 0 to 1, higher supervised exercise program and advice (rehabilitation) lead to scores indicate better quality of life), which were assessed by a greater improvements in activity limitation than advice alone? Is any blinded assessor at 1, 3 and 6 months after randomisation, with the improvement different with respect to fracture severity, age and primary time point at 3 months. Results: A total of 173 patients (80%) gender? Design: A multicentre, randomised, controlled trial with completed the 3-month follow-up, and 170 (79%) completed the 6- concealed allocation. Setting: Seven Australian hospitals. Partici- month follow-up. The mean difference in activity limitation was pants: The inclusion criteria were: isolated ankle fracture treated 0.4 units (95% CI –3.3 to 4.1) at 3 months and 0.2 units (95% CI –3.4 to with immobilisation (with or without surgical fixation); immobilisa- 3.9) at 6 months, with no difference between the two groups. Mean tion removed on the day of recruitment; approval received from the difference in quality of life was –0.01 units (–0.06 to 0.04) at both orthopaedic surgeon to bear weight as tolerated or bear partial 3 and 6 months. Treatment effects were not moderated by fracture weight; reduced ankle dorsiflexion range of motion (at least 30 mm severity or age and gender. Conclusion: A supervised exercise less motion compared with the non-fractured leg, using the weight- program did not confer additional benefits in activity limitation or bearing lunge method); ankle pain at least 2 out of 10 when 50% of quality of life compared with advice alone for patients with isolated body weight was borne through the affected leg; completed skeletal and uncomplicated ankle fracture. growth; and no concurrent pathologies that would affect the ability to perform everyday tasks. Randomisation allocated 106 to a Provenance: Invited. Not peer reviewed. supervised exercise program and advice (rehabilitation) and 108 to advice alone. Interventions: The advice group had a single session of Margreth Grotlea and Ka˚ re Birger Hagenb self-management advice about exercise and return to activity. The aOslo and Akershus University College of Applied Sciences, session was provided by a physiotherapist in the fracture clinic and also included advice on exercises for ankle movement in non-weight- Department of Physiotherapy, Oslo, Norway bearing positions and a handout summarising the advice with text bNational Advisory Unit on Rehabilitation in Rheumatology, and illustrations. The rehabilitation intervention consisted of the Department of Rheumatology, Diakonhjemmet Hospital, Oslo, Norway same advice in the fracture clinic, in addition to a supervised exercise http://dx.doi.org/10.1016/j.jphys.2016.01.002 ß 2016 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/). Commentary reported improvements in activity limitations and quality of life during the follow-up time, although no between-group differences were As the benefits of rehabilitation after immobilisation for ankle found. The lack of a control group receiving no treatment or advice fracture are unclear, Moseley and colleagues conducted a pragmatic, cannot rule out time as a factor and might explain the improvement randomised clinical trial to determine the effectiveness of a supervised seen in both groups; however, this would have included ethical exercise program and advice compared with advice alone. This well- considerations. Importantly, results from this study indicate that designed trial provides robust evidence that a supervised exercise routine care for patients after isolated and uncomplicated ankle program provides no additional improvement in activity limitation or fracture should include self-management advice at the time of removal quality of life compared with advice alone, which is in line with findings of immobilisation but not a supervised exercise program. from other studies evaluating rehabilitation after the removal of immobilisation.1 Nevertheless, in the present study, nearly one-third of Provenance: Invited. Not peer reviewed. the participants in the advice group received out-of-trial physiothera- Silje Stensrud py, which could potentially have diluted the observed effect. The authors hypothesised that older women or people with a more severe Oslo and Akershus University College of Applied Sciences, Department fracture may benefit more from rehabilitation, but no treatment effects of Physiotherapy, Oslo, Norway were associated with fracture severity or age and gender. This is in contrast with a previous study showing that younger patients (< References 40 years) could benefit from a postoperative training program.2 1. Lin CWC, et al. Cochrane Database Syst Rev. 2012;11:CD005595. Although recovery from ankle fracture is initially rapid, the recovery 2. Nilsson GM, et al. BMJ Muskuloskelet Disord. 2009;10:118. slows with time and can be incomplete 2 years after fracture.3 A 3. Beckenkamp PR, et al. J Orthop Sports Phys Ther. 2014;44:841–851. prolonged physiotherapy intervention or a delayed initiation could be trialed, as five additional sessions during the first 4 weeks after http://dx.doi.org/10.1016/j.jphys.2016.01.001 immobilisation may be insufficient. Both groups in the present study 1836-9553/ß 2016 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 62 (2016) 88–95 Journal of PHYSIOTHERAPY journal homepage: www.elsevier.com/locate/jphys Research Early intensive hand rehabilitation is not more effective than usual care plus one-to-one hand therapy in people with sub-acute spinal cord injury (‘Hands On’): a randomised trial Lisa A Harvey a,b,*, Sarah A Dunlop c, Leonid Churilov d, Mary P Galea e,*, Spinal Cord Injury Physical Activity (SCIPA) Hands On Trial Collaborators1 a John Walsh Centre for Rehabilitation Research, Sydney School of Medicine, University of Sydney; b Kolling Institute, Royal North Shore Hospital, Sydney; c School of Animal Biology, The University of Western Australia, Perth; d Florey Neuroscience Institutes (National Stroke Research Institute) and Department of Mathematics and Statistics, The University of Melbourne; e Department of Medicine (Royal Melbourne Hospital), The University of Melbourne, Melbourne, Australia KEY WORDS ABSTRACT Spinal cord injury Question: What is the effect of adding an intensive task-specific hand-training program involving Hand therapy functional electrical stimulation to a combination of usual care plus three 15-minute sessions per week of Rehabilitation one-to-one hand therapy in people with sub-acute tetraplegia? Design: A parallel group, randomised, Physical therapy controlled trial. Participants were randomly assigned (1:1) via a computer-generated concealed block Randomised controlled trial randomisation procedure to either a control or experimental intervention. Participants: Seventy people with C2 to T1 motor complete or incomplete tetraplegia within 6 months of injury. Participants were recruited from seven spinal units in Australia and New Zealand. Intervention: Experimental participants received intensive training for one hand. Intensive training consisted of training with an instrumented exercise workstation in conjunction with functional electrical stimulation for 1 hour per day, 5 days per week for 8 weeks. Both groups received usual care and 15 minutes of one-to-one hand therapy three times per week without functional electrical stimulation. Outcome measures: The primary outcome was the modified Action Research Arm Test reflecting arm and hand function, which was assessed at the end of the intervention, that is, 11 weeks after randomisation. Secondary outcomes were measured at 11 and 26 weeks. Results: Sixty-six (94%) participants completed the post-intervention assessment and were included in the primary intention-to-treat analysis. The mean modified Action Research Arm Test score for experimental and control participants at the post-intervention assessment was 36.5 points (SD 16.0) and 33.2 points (SD 17.5), respectively, with an adjusted mean between-group difference of 0.9 points (95% CI – 4.1 to 5.9). Conclusion: Adding an intensive task-specific hand-training program involving functional electrical stimulation to a combination of usual care plus three 15-minute sessions per week of one-to-one hand therapy does not improve hand function in people with sub-acute tetraplegia. Registration: Australian and New Zealand Trial Registry ACTRN12609000695202 and ClinicalTrials.gov NCT01086930. [Harvey LA, Dunlop SA, Churilov L, Galea MP, Spinal Cord Injury Physical Activity (SCIPA) Hands On Trial Collaborators (2016) Early intensive hand rehabilitation is not more effective than usual care plus one- to-one hand therapy in people with sub-acute spinal cord injury (‘Hands On’): a randomised trial. Journal of Physiotherapy 62: 88–95] ß 2016 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Introduction consequence of spinal cord injury (SCI), including the inability to walk or control bladder and bowel function.1 Research attention in People with tetraplegia consider loss of hand function to be recent years has therefore appropriately focused on identifying more debilitating and limiting on quality of life than any other possible ways of improving the hand function of people with tetraplegia. Intensive task-specific training with sensory or * Shared first authorship. functional electrical stimulation (FES) is one of many interventions 1 The SCIPA Hands On Trial Collaborators: Melanie Hurley, The University of that has received research attention, with initial promising results, Melbourne, Parkville; Julia Batty and Trent Z Li, The Prince of Wales Hospital, but it has not been examined within a large, high-quality clinical Randwick; Ann Thompson and Hannah Withers, Royal Rehabilitation Centre, trial.2–4 It has been hypothesised that intensive task-specific Sydney; Andrew Nunn and Janette Alexander, Austin Health, Heidelberg; John training with FES improves neural recovery and motor control Buchanan and Kate Wisbey, Royal Perth Hospital, Perth; Timothy Geraghty and following SCI. The combination of therapies provides both sensory Valerie Pick, Princess Alexandra Hospital Woolloongabba; Ruth Marshall and Jillian input from the periphery and motor input from the sensorimotor Clark, Hampstead Rehabilitation Centre, Northfield, Australia; K. Anne Sinnott and cortex onto the damaged spinal cord. It is believed that neural Jacqui Abel, Burwood Academy of Independent Living, Burword Hospital, New bombardment from these two sources may promote neural Zealand. http://dx.doi.org/10.1016/j.jphys.2016.02.013 1836-9553/ß 2016 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Researc_1I(G)ierug]TD[$Fh 89 plasticity and provide the critical stimulus required to elicit Assessed for eligibility (n = 731) neurophysiologic and structural re-organisation of the relevant pathways.5 Excluded (n = 661) One of the difficulties with providing intensive task-specific Measured function, strength, sensation, quality of life, and independence practice is that training is not always well tolerated by participants because repeatedly practising the same movement Week 0 Randomised (n = 70) outside a functional context can be tedious. Increasingly, technology has been used to try to overcome this barrier (eg, (n = 37) (n = 33) commercially available, computerised video games that respond to body motion are used to practise balance in people with Withdrew consent Exp group Con group Withdrew consent stroke).6 For the present trial, a similar concept was wanted for the but retained in ITT • FES training • usual care but retained in ITT hand, but since there were no appropriate motion-controlled analysis (n = 2) • usual care • one-to-one analysis (n = 2) video devices or games for training hand function in people with • one-to-one tetraplegia, we used FES and an instrumented exercise worksta- hand therapy tion incorporating several types of manipulanda connected to a hand therapy • 8 weeks computera. Participants triggered FES to drive the different types • 8 weeks of hand grasps (eg, pinch, squeeze, grasp, twist, lift, push or pull) required for playing the computer games.7 The FES was triggered Measured function, strength, sensation, quality of life, independence, with a behind-the-ear bluetooth devicea that is sensitive to tooth goal attainment, and perception of treatment effectiveness clicks. The technology thus provided a way of encouraging patients to perform large numbers of different hand movements Week 11 (n = 35)a (n = 31)a within a dynamic environment. There is preliminary evidence from five studies to suggest that this technology may be ITT analysis therapeutic.3,4,8–10 However, all these studies are small and have methodological flaws exposing them to bias. (n = 37) (n = 33) Therefore, the questions for this parallel group, randomised, Withdrew consent Withdrew consent controlled trial were: and ceased and ceased treatment (n = 4) treatment (n = 5) 1. Is adding an intensive hand-training program, with an instrumented exercise workstation and functional electrical Exp group Con group stimulation, to usual care more effective than usual care alone • usual care • usual care in people with sub-acute tetraplegia? Week 26 Measured function, strength, sensation, quality of life, independence, and 2. What are the possible benefits on muscle strength, sensation, perception of treatment effectiveness function and quality of life? (n = 31)a (n = 26)a ITT = intention-to-treat; con = control, exp = experimental Figure 1. CONSORT flow chart indicating the number of participants screened, randomised and included in intention-to-treat analysis (a some participants did not complete all assessments. See Table 2 for details). Method were likely to remain in hospital for 12 weeks; had a reduced ability to grasp with the target hand, as determined by the clinical Design judgement of the hospital therapist; and were able to tolerate sufficient FES to enable the target hand to grasp and release. A multi-centre, randomised, assessor-blinded, phase-3 trial was Participants were excluded if they had a pre-existing injury to the undertaken on inpatients at seven SCI units in Australia and New hand or upper limb or any condition that precluded use of the Zealand. Participants were randomised to the experimental or exercise workstation and FES (the full inclusion and exclusion control group. Experimental participants received an intensive criteria are detailed in the protocol).11 8-week hand-training program for the target hand. Participant recruitment commenced 23 November 2009 and finished 31 De- The trial statistician used a computer random number cember 2013. The trial protocol (including full details of the study generator to create the randomisation schedule, which was rationale, design and statistical analysis) was published and is stratified by site and the baseline score of the modified Action available online.11 The trial was managed by a professional clinical Research Arm Test (m-ARAT; 21 versus > 21) using permuted trial management companyb and overseen by an independent data blocks of random sizes. To ensure concealment, block sizes were safety monitoring committee. undisclosed. An independent researcher with no clinical involve- ment in the trial randomly assigned the participants to either the Participants, therapists and centres control or experimental group with a 1:1 ratio. After completion of baseline assessments, randomisation was performed by an Seventy participants with sub-acute tetraplegia undergoing administrator who was independent of the recruitment process inpatient rehabilitation in one of the seven participating SCI units and located off siteb to ensure concealment. A participant was were recruited from a consecutive sample of admissions (Figure 1). considered to have entered the trial once his/her randomisation The hospital therapists screened participants for suitability and was allocated. then enrolled them in the study. These therapists provided usual care to all participants but were not otherwise involved in the trial. Trial and hospital therapists and participants were unblinded, Instead, specifically designated trial therapists administered the but the assessors and statisticians performing the analyses were intervention to the experimental participants. One hand of each blinded. The success of assessor blinding was checked by asking participant was identified as the target hand according to the assessors whether they had been unblinded. criteria below. In situations where both hands met the inclusion criteria, the hospital therapist selected the hand deemed most Intervention likely to benefit from intensive training. Experimental participants received training directed at the Participants were included if they: were 16 years or older and target hand five times per week for 8 weeks commencing 3 weeks had sustained a motor complete or incomplete SCI at the after randomisation. The 3-week delay in commencing the neurological level of C2 to T1 within the preceding 6 months; intensive training was required to allow time for the delivery of the FES garments for the experimental participants. The training consisted of an intensive task-specific hand-training program provided through an instrumented exercise workstationa in
90 Harvey et al: Hand rehabilitation for spinal cord injury conjunction with FES. The hand activities involved playing The primary outcome was the m-ARAT of the target hand, computer games while practising functional tasks using different which was measured at the end of the intervention period (ie, manipulanda (including reaching, grasping, manipulating, pulling, 11 weeks after randomisation). The m-ARAT is a standardised rotating and releasing). The exercises and computer games were measure of unilateral hand and upper limb function. It consists of progressed so that as hand function improved, more difficult hand four sub-tests, including grasp, grip, pinch and gross movement. exercises and games were introduced. Each training session lasted Participants were required to perform every task in each subtest;12 1 hour and participants were required to use the instrumented all tasks were scored on a scale from 0 to 3 points. Scores were exercise workstation and FES as much as possible during this time. summed to give a total score, where a larger number reflected A trial therapist directly supervised all sessions, progressed the better hand function. difficulty of the games, provided encouragement, ensured that participants focused on the quality of their grasp, determined the The m-ARAT was also measured at 26 weeks as a secondary FES stimulation parameters and provided feedback about perfor- outcome measure. All other outcomes, which were prospectively mance. The trial therapsits were trained and provided with a categorised as secondary or tertiary, are described in detail in the written protocol. They recorded details of each training session protocol. Each is summarised below: using standardised recording documents. The Summed Upper Limb Strength of the Graded and Redefined The FES was provided through 5-cm diameter electrodes. The Assessment of Strength, Sensibility and Prehension13 was mea- electrodes were incorporated into a custom-made garmenta for sured at the end of the intervention and at 26 weeks; a higher score each participant, according to the optimal stimulation points and indicated better strength. the size and side (left versus right) of the hand. The FES was administered to any two stimulatable key muscles of the hand, The Sensory Score of the International Standard for Neurologi- including the flexors and extensors of the wrist, fingers and thumb. cal Classification of Spinal Cord Injury14 for the target hand was Participants triggered it by clicking their teeth; this stimulated the measured at the end of the intervention and at 26 weeks; a higher hand to open or close, allowing participants to grasp or release the score indicated better sensation. various manipulanda on the workstation independent of assis- tance from the trial therapists. The intensity of the FES stimulation The AsTex1 Sensory Test15 was measured at the end of the was determined by the trial therapist to ensure strong contractions intervention and at 26 weeks. This measured texture discrimina- of the stimulated muscles, as tolerated by the participant. It was tion; a lower score indicated better texture discrimination. not increased beyond 63 mA (see protocol for full stimulation details).11 The AuSpinal Test of Hand Function16 was measured at the end of the intervention and at 26 weeks; a higher score reflected better If participants missed any treatments during the intervention hand function. period, additional sessions were offered to them on weekends, at another time during the week or during an optional additional The Capabilities of Upper Extremity (CUE)17 of the target hand week at the end of the intervention period. was administered at the end of the intervention and at 26 weeks; a higher score reflected better upper limb function. Control participants did not receive the intensive task-specific hand-training program with the instrumented exercise workstation The Assessment of Quality of Life-8 (AQoL-8)18 was adminis- and FES. Instead, both control and experimental participants tered at the end of the intervention and at 26 weeks; a score of continued to receive usual care. This included typical inpatient 1 reflected perfect health. rehabilitation consisting of physiotherapy as well as vocational, recreational and occupational therapy (full details are described in The Health Utilities Index Mark 3 (HUI3)19 was administered at the protocol).11 In addition, over the 8-week intervention period, the end of the intervention and at 26 weeks. This was scored both control and experimental participants received at least three between –0Á36 and 1; a higher score reflected better quality of life. 15-minute sessions per week of one-to-one hand therapy for the target hand. This hand therapy was provided by the hospital The Self-Care Subscale of the Spinal Cord Independence therapists and individualised to the needs of participants. It Measure (SCIM)20 was administered post intervention and at consisted of practising functional hand activities and did not involve 26 weeks; a higher score reflected more independence. FES. The hospital therapists used standardised forms to record the duration and type of hand therapy provided to the target hand. The Goal Attainment Scale (GAS)21 was scored at the end of the intervention. This was scored from –2 to +2, where +2 reflected At the end of the intervention period, both experimental and attainment of goals ‘a lot better’ than expected and –2 reflected control participants continued to receive usual care by the hospital attainment of goals ‘a lot worse’ than expected. therapists (patients were discharged between 3 and 9 months after randomisation). Usual care over this period was not standardised The Participant Perception of Treatment Effectiveness22 was or restricted in any way. The only restriction was that neither scored post intervention. This was scored from –7 to +7, where a experimental nor control participants could use the instrumented score of +7 reflected hand function that was ‘a very great deal exercise workstation. In addition, no participant was permitted to better’ and –7 reflected hand function that was ‘a very great deal practise any aspect of the tests comprising the outcome measures. worse’ than at the time of baseline assessment. They could, however, practise activities similar to those included in the hand tests as part of functional training. At the post-intervention assessment, all experimental partici- pants were asked by the blinded assessors to rate on a 10-point Outcome measures category rating scale their perceptions about the convenience of the hand training. One end was anchored with the words All measurements were taken at baseline, at 11 weeks (ie, at the ‘completely inconvenient’ and the other end with the words ‘very end of the intervention period) and at 26 weeks after randomisa- convenient’; a higher score reflected more convenience. This was tion, except two outcomes that were only taken at 11 weeks: the used to gauge the burden of the experimental intervention on Goal Attainment Scale and the Participant Perception of Treatment participants. It was not used as an outcome measure. All adverse Effectiveness. Most participants were discharged prior to the 26- events, whether related or unrelated to the intervention, were week assessment and were required to come back to hospital for collected over the course of the study. this assessment. The original protocol included a 12-month follow- up assessment, but the protocol was modified and this final Data analyses assessment was removed halfway through the trial to reduce participant burden and encourage recruitment. This study was powered to detect a between-group minimum worthwhile treatment effect on m-ARAT scores of 5.7 points at the post-intervention assessment (SD 14 points). A sample size of 78 participants (ie, 39 per group) was estimated to provide 80% power to detect a significant intervention effect (two-sided, p = 0.05) using an ANCOVA model that included: baseline m-ARAT score as a covariate, a correlation between baseline and post- intervention m-ARAT scores of at least 0.8 and an adjustment of 10% to allow for dropout rate. These data were based on the results
Research 91 of a similar pilot study with chronic SCI participants conducted in effect size presented as Wilcoxon-Mann-Whitney generalised odds Canada (personal communication; Prochazka A, 2009, University ratio with corresponding 95% CI. of Alberta). A secondary pre-specified per-protocol analysis was performed Analyses were performed by a blinded and independent to determine the possible effect of trial adherence on the primary statistician according to a pre-specified statistical analysis plan outcome (post-intervention m-ARAT). This dataset only comprised on an intention-to-treat basis,23 with an assumption for the main participants who adhered to all aspects of the protocol and analysis that data were missing at random. The sensitivity of the received at least 80% of training sessions. All analyses were results to plausible departures from the missing-at-random performed using $]D[T6I_7Fcommercially available softwarec. assumption as a part of intention-to-treat analysis was examined using both a selection model (modelling of the missing data Results mechanism) and a pattern mixture model (modelling of the differences between missing and observed data). Assumptions Flow of participants through the trial about the missing data were expressed via a parameter that measures the degree of departure from the missing-at-random A total of 731 participants that were admitted with tetraplegia assumption. The results were graphed over a range of assumptions. to the seven SCI units between 23 November 2009 and 31 Decem- ber 2013 were screened for inclusion. Of these, 70 were eligible, Analyses of continuous data (m-ARAT, Summed Upper Limb agreed to participate and were subsequently randomised (Figure 1 Strength, Sensory score, AsTex Test, AuSpinal, CUE, AQoL-8, HUI3 and Table 1). and SCIM) were conducted using ANCOVA models that included treatment group and the baseline scores as a covariate. The Compliance with the trial protocol analyses were implemented using multiple linear regression. The corresponding estimates of the treatment effect were calculated as The trial was terminated after randomisation of 70 participants adjusted mean differences with corresponding 95% CIs. Heteroge- (rather than the intended 78) because recruitment was taking neity of the treatment effect across multiple participating centres longer than expected and the funding was limited to 5 years. This was tested using a corresponding random-effect linear regression change to the protocol was made without any knowledge of the model with site as a random effect. results. The GAS data were analysed using Wilcoxon-Mann-Whitney Compliance with the study interventions was excellent. The rank sum tests to determine the difference between the two groups protocol dictated that experimental participants receive 40 1- in GAS across the full ordinal scale. Two analyses were performed: hour training sessions over 8 weeks with trial therapists, one for participants’ first goal and one for participants’ second goal. commencing 3 weeks after randomisation. In reality, they The effect size was presented as a Wilcoxon-Mann-Whitney received a median of 40 training sessions (IQR 40 to 42) over generalised odds ratio with corresponding 95% CI.24 In addition, a median of 8 weeks (IQR 7.6 to 8.6). Only three participants did the difference between the two groups in the proportions of patients not receive at least 80% of the 40 interventions. The median whose level of achievements were no worse than expected were length of each training session was 57 minutes (IQR 54 to 59). analysed by dichotomising the GAS scale into ‘no worse than Control and experimental participants received a mean of 40 (SD expected’ (categories of 0, +1 and +2 on the ordinal GAS scale) versus 22) and 38 (SD 27) 15-minute hand therapy sessions for the ‘worse than expected’ (categories of –1 and –2 on the ordinal GAS target hand with hospital therapists over the 8-week interven- scale) and using Fisher’s exact test. The corresponding effect sizes tion period, respectively. were calculated as risk differences with corresponding 95% CI. Data were missing on four participants at the post-intervention The Participant Perception of Treatment Effectiveness was assessment and on 13 participants at the 26-week assessments. analysed in two ways. Firstly, the difference between the two Most missing data were due to participants withdrawing consent groups in the proportions of participants who felt better/worse was or not being able to return to the hospital for assessment after analysed by dichotomising responses into ‘feeling better’ versus discharge. In addition, some participants occasionally declined to ‘feeling worse’ and using Fisher’s exact test with corresponding complete some measures for various reasons (see Tables 2 to 4 for effect size presented as risk difference with corresponding 95% CI. the exact number of participants who completed each outcome Secondly, the difference between the two groups in scores across the measure). Assessors were inadvertently unblinded for two full ordinal scale was analysed by transferring responses into a single assessments. ordinal scale ranging from –7 to +7 points. This was analysed using a Wilcoxon-Mann-Whitney rank sum test with the corresponding Table 1 Exp (n = 37) Con (n = 33) Baseline characteristics of the participants included in the intention-to-treat analysis. 33 (89) 28 (85) Characteristic 29 (23 to 45) 29 (22 to 53) 81 (45 to 110) 62 (47 to 87) Gender, n male (%) Age (y), median (IQR) 14 (38) 10 (30) Time since injury (d), median (IQR) 7 (19) 5 (15) ASIA Impairment Scale, n (%) 3 (8) 9 (27) 13 (35) 9 (27) A B 16 (43) 12 (36) C 21 (57) 21 (64) D 25 (17 to 31) 24 (17 to 31) Target hand, n (%) left 12 (8 to 16) 12 (8 to 15) right 13 (9 to 16) 12 (8 to 17) Upper limb ASIA motor score for both limbs (0 to 50), median (IQR) Upper limb ASIA motor score for limb of target hand (0 to 25), median (IQR) 26 (20 to 70) 28 (18 to 42) upper limb right 49 (29 to 91) 54 (27 to 82) upper limb left Total ASIA sensation (0 to 112), median (IQR) pin prick light touch ASIA = American Spinal Injury Association, con = control, exp = experimental.
92 Harvey et al: Hand rehabilitation for spinal cord injury Adjusted between-group difference Week 26 Exp minus Con (95% CI) AQoL-8 = Assessment of Quality of Life-8, Con = control, CUE = Capabilities of Upper Extremity test, Exp = experimental, HUI3 = Health Utilities Index Mark 3, m-ARAT = modified Action Research Arm Test, SCIM = Spinal Cord Independence Measure 3.1 (–3.2 to 9.5) (Self-Care Subscale). Participants 2.6 (–1.6 to 6.8) Shaded cell = primary outcome. –1.8 (–5.7 to 2.1) The median age of participants was 29 years (IQR 22 to 49) and 0.04 (–0.21 to 0.29) the median time since injury was 68 days (IQR 45 to 107). The –0.02 (–0.23 to 0.19) neurological levels of participants’ lesions were C1 (n = 2, 3%), C2 0.8 (–9.0 to 10.5) (n = 2, 3%), C3 (n = 4, 6%), C4 (n = 38, 54%), C5 (n = 9, 13%), C6 –0.4 (–10.6 to 9.8) (n = 12, 17%) and C7 (n = 3, 4%). The inclusion of two people with a 0.04 (–0.10 to 0.18) C1 neurological level was because at the time of inclusion these 0.01 (–0.12 to 0.14) people had very unusual sensory loss at C2 that was not consistent 0.8 (–2.1 to 3.8) with their other motor and sensory losses. The examiners attributed this to age in one case and drowsiness in the other. Week 11 Exp minus Con (95% CI) They were both therefore deemed eligible for the trial. However, 0.9 (–4.1 to 5.9) the present study is reported exactly as per the American Spinal 0.3 (–2.9 to 3.4) Injury Association (ASIA) charts and does not take these –0.4 (–2.7 to 1.9) extenuating circumstances into consideration. The ASIA Im- –0.10 (–0.36 to 0.11) pairment Scale classifications were AIS A (n = 24, 34%), AIS B 0.02 (–0.18 to 0.23) (n = 13, 19%), AIS C (n = 11, 16%) and AIS D (n = 22, 31%). There were 2.8 (–3.0 to 8.6) no clear between-group differences at baseline (Table 1 and the 1.5 (–6.3 to 9.3) first two columns of Table 2). 0.07 (–0.04 to 0.19) 0.02 (–0.07 to 0.12) Effect of intervention 1.3 (–0.8 to 3.4) Primary outcome Table 2 Con (n = 26) The adjusted mean between-group difference for the m-ARAT Results of the intention-to-treat analysis for the primary and some secondary outcomes at 11 weeks (ie, immediately after intervention) and 26 weeks after randomisation. 33.4 (17.4) 24.8 (11.0) immediately after the intervention was 0.9 points (95% CI –4.1 to ][1_T$DIF(n = 25) 5.9) favouring the experimental group (Table 2, Figure 2). The 22.6 (6.8) result of the per-protocol analyses was very similar (0.9 points, 95% 0.62 (0.38) CI –4.5 to 6.4). The sensitivity analysis testing the robustness of the ]FI$DT_1[(n = 25) assumptions about missing data indicated very little effect of these 0.65 (0.47) assumptions on the primary outcome. There was no statistically ]FI$DT_1[(n = 25) significant heterogeneity in treatment effect between the partici- 66.7 (22.5) pating centres. 66.2 (18.4) 0.37 (0.25) Secondary outcomes 0.28 (0.27) There were no statistically significant mean between-group 1$DT_[]FI(n = 25) 9.5 (7.1) differences for any of the secondary analyses, including partici- pants’ perceptions of treatment effectiveness (Tables 2 to 4). Week 26 Individual participant data are presented in Table 5 (see eAddenda for Table 5). Experimental participants rated the convenience of Exp (n = 31) the interventions with a median rating of 8 points (IQR 5 to 9), 38.7 (17.6) where a score of 10 indicated ‘very convenient’. 29.5 (13.8) (n = 30) Twenty serious adverse events (10 in each group) and 20.7 (9.7) 738 adverse events were recorded over the 26-week period from 0.79 (0.62) randomisation in all participants, but none were related to the IF$D]T_1[(n = 28) intervention. 0.76 (0.62) $[]ID_1TF(n = 30) Discussion 70.0 (25.5) 69.3 (22.8) The results of this trial indicate no benefit of adding an intensive 0.41 (0.28) task-specific hand-training program involving FES to a combina- I[]$DT_1F(n = 30) tion of usual care plus three 15-minute sessions per week of one- 0.35 (0.31) to-one hand therapy in people with sub-acute tetraplegia. The I$DFT_1[](n = 30) treatment estimate of the primary outcome (post-intervention m- 10.9 (6.6) ARAT) was precise, indicating that although the trial was terminated early, the sample size was adequate. The upper end Con (n = 31) of the 95% CI associated with the mean between-group difference 33.2 (17.5) was 5.9 points, suggesting that if the trial was repeated, there 23.0 (9.9) would be very little possibility of finding a treatment effect in 23.3 (5.3) excess of the pre-determined minimally worthwhile treatment 0.65 (0.48) effect of 5.7 points. The results of the secondary analyses and [1T_]$DIF(n = 30) participants’ perceptions about the benefits of the treatment also 0.58 (0.32) pointed to no between-group differences, although some esti- 1FI][_T$D(n = 29) mates of treatment effects were imprecise. 66.1 (20.5) 66.2 (18.8) This trial is important because it is one of the largest non- 0.30 (0.23) pharmaceutical trials conducted in people with SCI and the largest 0.22 (0.22) trial involving any type of hand intervention for people with [1]$_FITD(n = 29) tetraplegia. The trial was conducted according to Good Clinical 8.3 (5.9) Practice standards,25 which is uncommon for non-pharmaceutical trials involving people with SCI. Trials involving training are Groups Week 11 Exp (n = 35) 36.5 (16.0) 24.9 (12.5) 22.7 (6.6) 0.66 (0.57) 0.73 (0.64) 70.8 (20.3) 70.3 (16.9) 0.37 (0.24) 0.31 (0.26) 10.1 (5.7) Con (n = 33) 21.2 (15.7) 19.2 (8.2) 22.8 (6.2) 0.77 (0.68) 0.63 (0.58) 52.5 (25.3) 54.3 (15.1) 0.24 (0.20) 0.13 (0.15) 4.7 (3.4) Week 0 Exp (n = 37) 24.8 (15.7) 21.1 (10.6) 22.6 (5.8) 1.07 (0.76) [TFI_1]D$(n = 35) 0.87 (0.71) 57.3 (21.7) 58.5 (16.3) 0.23 (0.17) 0.18 (0.17) 5.5 (4.6) Outcomes m-ARAT, (0 to 57) Summed upper limb strength, (0 to 50) Sensory score, (0 to 32) AsTex index finger, (mm) AsTex thumb, (mm) AusSpinal, (0 to 86) CUE, (0 to 105) AQoL-8, (0 to 1) HUI3, (0 to 1) SCIM, (0 to 20)
Research 93 Table 3 Results of the intention-to-treat analysis for the Goal Attainment Score and the dichotomiseda1[$FI_D]T Perception of Treatment Effectiveness score at 11 weeks (ie, immediately after intervention) and 26 weeks after randomisation. Outcomes Groups Generalised odds ratio (95% CI) Risk difference (95% CI) Week 11 Goal Attainment Score not worse than Week 11 Week 26 Week 11 Week 26 expected, n (%) Exp relative goal 1 Exp Con Exp Con to Con Exp relative Exp relative goal 2 (n = 35) (n = 31) (n = 30) (n = 26) to Con to Con 1.2 (0.7 to 2.1) Perception of Treatment Effectiveness 27 (79) (n = 34) 23 (74) 1.1 (0.6 to 1.9) 0.1 (–0.2 to 0.3) scorea, n (%) 24 (71) (n = 34) 19 (61) 0.1 (–0.1 to 0.3) better 34 (97) 30 (97) 30 (100) 25 (96) 0.0 (–0.1 to 0.1) 0.04 (–0.0 to 0.1) Con = control, exp = experimental. a Dichotomised as better or worse. Table 4 Results of the intention-to-treat analysis for the Perception of Treatment Effectiveness scores at 11 weeks (ie, immediately after intervention) and 26 weeks after randomisation. Outcomes Groups Generalised odds ratio (95% CI) Week 11 Week 26 Week 11 Week 26 Perception of Treatment Effectiveness scores, n (%) Exp (n = 35) Con (n = 31) Exp (n = 30) Con (n = 24) Exp relative to Con Exp relative to Con a little worse 1.4 (0.8 to 2.5) 1.2 (0.7 to 2.3) almost the same (if worse) 1 (3) 0 (0) 0 (0) 1 (4) almost the same (if better) 0 (0) 1 (3) 0 (0) 0 (0) a little better 4 (11) 2 (6) 4 (13) 3 (13) somewhat better 6 (17) 6 (19) 2 (7) 2 (8) moderately better 3 (9) 7 (23) 2 (7) 3 (13) a good deal better 1 (3) 6 (19) 4 (13) 2 (8) a great deal better 12 (34) 6 (19) 6 (20) 5 (21) a very great deal better 6 (17) 3 (10) 7 (23) 5 (21) 2 (6) 0 (0) 5 (17) 3 (13) Con = control, exp = experimental. notoriously difficult to complete because they require the intervention because the hand training required considerable administration of complex interventions over an extended period concentration and co-operation from participants. of time. For example, experimental participants required 40 hours of training with the instrumented exercise workstation and FES. The results of this trial are also important because they conflict These training sessions required direct supervision from a trial with the results of all other trials involving hand training in people therapist. The training was often difficult to deliver because of with tetraplegia. Other trials have examined hand training with participant illness, staff shortages and equipment failure. In FES (versus hand training alone2 or no intervention3,8,9) or FES and addition, it was often difficult to find 1 hour during a participant’s biofeedback (versus usual care10) or repetitive practice on a day that was not already scheduled for some other type of therapy workstationa with FES (versus usual care4). All have reported (eg, vocational training, education, mobility training or hydrother- therapeutic effects. However, these trials were small, with apy). Regardless, all but three participants received 40 treatments. methodological flaws. For example, the median PEDro score26 of Sometimes, however, more than one treatment was provided per these trials is 3.5/10 (IQR 3 to 4) and few trials blinded assessors or day to meet the target of 40 treatments. More than 1 hour a day concealed allocation, whereas the PEDro score of the present study was 8/10. In addition, none of those trials was prospectively o[(Figure_2)TD$IG] f training may have compromised the effectiveness of the registered nor had an accompanying protocol. All these design weaknesses increase susceptibility to bias and, in the light of our Figure 2. m-ARAT (/57 points) results for experimental and control groups at starkly different findings, raise questions about the robustness of baseline (week 0), after the intervention (11 weeks) and at follow-up (26 weeks). the results of those studies. This should prompt researchers and clinicians to re-examine some widely held assumptions about the effectiveness of repetitive practice with FES. The idea that recovery and return of hand function are not necessarily influenced by intensive practice, as has been assumed, may need to be entertained. But of course, the contrast between the present results and those of all previous trials may not reflect bias. Instead, it may point to some fundamental difference between the present trial and previous trials. For example, past trials have solely focused on people with chronic SCI, while the present trial was restricted to people with sub-acute SCI. Perhaps those with chronic SCI respond better to FES-based intervention than those with sub- acute SCI. In addition, the present control and experimental participants received 40 and 38 sessions of individualised one-to- one hand therapy for the target hand plus usual care, respectively. This additional hand therapy may have rendered the FES-based hand training for the experimental participants redundant. It is possible that the training was ineffective because there was insufficient opportunity to individualise training. While the exercise workstation provided practice with different manipulanda,
94 Harvey et al: Hand rehabilitation for spinal cord injury the FES stimulated a gross grasp and release. Perhaps partici- Competing interests: The authors declare no competing pants would have benefitted from practice of isolated finger and interests, however an associate investigator on the original funding thumb movements as well as practice of various hand grasps in application (but not on this publication) has commercial interests many different contexts. Alternatively, perhaps the games in the instrumented exercise workstation and FES technology used encouraged participants to rely on the FES and perform the as part of this trial. MPG has a commercial interest in the AsTex1. hand activities quickly and without care. Perhaps participants needed to practise slowly with maximal voluntary effort and Sources of support: Funding from the Transport Accident focused concentration. It is also possible that contamination Commission (Victorian Neurotrauma Initiative), NSW Lifetime diluted the contrast between the two groups. This may have Care and Support Authority, The University of Melbourne and The occurred if control participants practised hand movements and University of Western Australia. grasps each day outside formal therapy in response to the trial or if the hospital therapists inadvertently provided more attention Acknowledgements: We thank all participants for their time to control participants. The only way that this could have been and effort. safeguarded against these factors was to blind the participants and the hospital therapists, but this was not possible because of Authors’ contributions: Lisa Harvey, Mary Galea and Sarah the nature of the intervention. Dunlop were responsible for the design of the trial, securing funding and coordinating the trial. Leonid Churilov was responsi- This trial did not explore other possible therapeutic benefits of ble for statistical design and analysis. The SCIPA Hands On Trial FES. For example, studies in the lower limb of people with SCI Collaborators: Melanie Hurley was responsible for managing the provided initial evidence that FES helps prevent atrophy and trial. Julia Batty, Ann Thompson, Andrew Nunn, John Buchanan, maintains stimulated strength of paralysed muscles.27,28 Recent Timothy Geraghty, Ruth Marshall and K. Anne Sinnott were evidence has also suggested that FES maintains the excitability of principal investigators at sites and responsible for coordination peripheral nerves affected by SCI.29,30 It is argued that all of these and recruitment at their respective sites. Trent Li, Hannah Withers, therapeutic effects may be important for future recovery,31 although Janette Alexander, Kate Wisbey, Valerie Pick, Jillian Clark and these claims are yet to be verified in high-quality clinical trials. Jacqui Abel were site coordinators and responsible for the day-to- day management of the trial at their respective sites. All authors The findings of the present study are notable because it is have read and approved the final manuscript. believed that repetitive task-specific practice with FES will lead to improved hand function in people with tetraplegia. The results of Provenance: Not invited. Peer reviewed. this study challenge these beliefs and indicate that an 8-week Correspondence: Professor Lisa A Harvey, John Walsh Centre for intensive hand-training program with FES in people with recent Rehabilitation Research, Sydney School of Medicine, University of SCI does not provide added benefit over and above the combination Sydney, Australia. Email: [email protected] of usual care and three 15-minute hand therapy sessions per week. While it is tempting to suggest that a trial with an even more References intensive training regimen is now required, it is difficult to know whether it is realistic to provide an even more intensive hand- 1. Anderson KD. Targeting recovery: priorities of the spinal cord-injured population. J training program than already provided in this trial. One hour a day Neurotrauma. 2004;21:1371–1383. is a considerable time commitment for people with recently acquired tetraplegia undergoing rehabilitation because they 2. Beekhuizen KS, Field-Fote EC. Massed practice versus massed practice with stim- invariably have busy schedules with many skills to learn and ulation: effects on upper extremity function and cortical plasticity in individuals therapies to attend. However, future trials are required to clarify with incomplete cervical spinal cord injury. Neurorehabil Neural Repair. other possible therapeutic effects of FES and the effects of intensive 2005;19:33–45. practice provided in different ways. Nonetheless, the results of this study indicate that adding an intensive task-specific hand-training 3. Beekhuizen KS, Field-Fote EC. Sensory stimulation augments the effects of program involving FES to a combination of usual care plus three massed practice training in persons with tetraplegia. Arch Phys Med Rehabil. 15-minute sessions per week of one-to-one hand therapy does not 2008;89:602–608. improve hand function in people with sub-acute SCI. 4. Kowalczewski J, Chong SL, Galea M, Prochazka A. In-home tele-rehabilitation What is already known on this topic: Loss of hand function improves tetraplegic hand function. Neurorehabil Neural Repair. 2011;25:412–422. is a very debilitating consequence of tetraplegia after spinal cord injury. The use of functional electrical stimulation trig- 5. Dobkin BH. Brain-computer interface technology as a tool to augment plasticity gered by tooth clicks at a workstation with a range of hand and outcomes for neurological rehabilitation. J Physiol. 2007;15:637–642. grasp tasks allows patients to independently undertake inten- sive task-specific hand rehabilitation. 6. Bower KJ, Clark RA, McGinley JL, Martin CL, Miller KJ. Clinical feasibility of the What this study adds: Adding an intensive task-specific Nintendo WiiTM for balance training post-stroke: A phase II randomized controlled hand-training program involving functional electrical stimula- trial in an inpatient setting. Clin Rehabil. 2014;28:912–923. tion to a combination of usual care plus three 15-minute sessions per week of one-to-one hand therapy does not 7. Van Biervliet A, Gest TR. A multimedia guide to spinal cord injury: empowerment improve hand function in people with sub-acute tetraplegia. through self instruction. Medinfo. 1995;8(Pt 2):1701. Footnotes: aRehabtronics Inc, Edmonton, Canada. bNeur- 8. Popovic MR, Thrasher TA, Adams ME, Takes V, Zivanovic V, Tonack MI. Functional oscience Trials Australia, Melbourne, Australia. cStata IC 13 statis- electrical therapy: retraining grasping in spinal cord injury. Spinal Cord. tical software, StataCorp, College Station, USA. 2006;44:143–151. eAddenda: Table 5 can be found online at doi:10.1016/j.jphys. 9. Kapadia N, Zivanovic V, Popovic MR. Restoring voluntary grasping function in 2016.02.013 individuals with incomplete chronic spinal cord injury: Pilot study. Top Spinal Cord Inj Rehabil. 2013;19:279–287. Ethics approval: Ethical approval was obtained from the University of Melbourne (HREC 0932764.1) and the Human 10. Kohlmeyer KM, Hill JP, Yarkony GM, Jaeger RJ. Electrical stimulation and biofeed- Research Ethics Committee at each site. All participants were back effect on recovery of tenodesis grasp: a controlled study. Arch Phys Med provided with an approved participant information sheet and Rehabil. 1996;77:701–706. signed a consent form before data collection began. 11. Harvey L, Dunlop S, Churilov L, Hsueh Y, Galea M. Early intensive hand rehabilita- tion after spinal cord injury (‘‘Hands On’’): a protocol for a randomised controlled trial. BMC Trials. 2011;12:1–9. 12. van der Lee JH, Roorda LD, Beckerman H, Lankhorst GJ, Bouter LM. Improving the Action Research Arm Test: a unidimensional hierarchical scale. Clin Rehabil. 2002;16:646–653. 13. Rudhe C, van Hedel HJ. Upper extremity function in persons with tetraplegia: relationships between strength, capacity, and the spinal cord independence mea- sure. Neurorehabil Neural Repair. 2009;23:413–421. 14. Kirshblum SC, Burns SP, Biering-Sorensen F, Donovan W, Graves DE, Jha A, et al. International standards for neurological classification of spinal cord injury (revised 2011). J Spinal Cord Med. 2011;34:535–546. 15. Miller KJ, Phillips BA, Martin CL, Wheat HE, Goodwin AW, Galea MP. The AsTex: clinimetric properties of a new tool for evaluating hand sensation following stroke. Clin Rehabil. 2009;23:1104–1115. 16. Coates S, Harvey L, Dunlop S, Allison G. The AuSpinal: a test of hand function for people with tetraplegia. Spinal Cord. 2011;49:219–229. 17. Marino RJ, Shea JA, Stineman MG. The Capabilities of Upper Extremity instrument: reliability and validity of a measure of functional limitation in tetraplegia. Arch Phys Med Rehabil. 1998;79:1512–1521. 18. Hawthorne G, Osborne R, McNeil H, Richardson J. The Australian Multi-attribute Utility (AMAU): Construction and Initial Evaluation. Melbourne: Centre for Health Program Evaluation; 1996.
Research 95 19. Asakawa K, Rolfson D, Senthilselvan A, Feeny D, Johnson JA. Health Utilities Index 26. Moseley AM, Herbert RD, Sherrington C, Maher CG. Evidence for physiotherapy Mark 3 showed valid in Alzheimer disease, arthritis, and cataracts. J Clin Epidemiol. practice: A survey of the physiotherapy evidence based database (PEDro). Aust J 2008;61:733–739. Physiother. 2002;48:43–50. 20. Itzkovich M, Gelernter I, Biering-Sorensen F, Weeks C, Laramee MT, Craven BC, et al. 27. Baldi JC, Jackson RD, Moraille R, Mysiw WJ. Muscle atrophy is prevented in patients The Spinal Cord Independence Measure (SCIM) version III: reliability and validity with acute spinal cord injury using functional electrical stimulation. Spinal Cord. in a multi-center international study. Disabil Rehabil. 2007;29:1926–1933. 1998;36:463–469. 21. Turner-Stokes L. Goal attainment scaling (GAS) in rehabilitation: a practical guide. 28. Duffell LD, Donaldson NDN, Perkins TA, Rushton DN, Hunt KJ, Kakebeeke TH, et al. Clin Rehabil. 2009;23:362–370. Long-term intensive electrically stimulated cycling by spinal cord-injured people: Effect on muscle properties and their relation to power output. Muscle Nerve. 22. Barrett B, Brown D, Mundt M, Brown R. Sufficiently important difference: expand- 2008;38:1304–1311. ing the framework of clinical significance. Med Decis Making. 2005;25:250–261. 29. Lin CS, Macefield VG, Elam M, Wallin BG, Engel S, Kiernan MC. Axonal 23. White IR, Horton NJ, Carpenter J, Pocock SJ. Strategy for intention to treat analysis changes in spinal cord injured patients distal to the site of injury. Brain. in randomised trials with missing outcome data. BMJ. 2011;342(7803):910–912. 2007;130(Pt 4):985–994. 24. Churilov L, Arnup S, Johns H, Leung T, Roberts S, Campbell BC, et al. An improved 30. Boland RA, Bostock H, Kiernan MC. Plasticity of lower limb motor axons after method for simple, assumption-free ordinal analysis of the modified Rankin Scale cervical cord injury. Clin Neurophysiol. 2009;120:204–209. using generalized odds ratios. Int J Stroke. 2014;9:999–1005. 31. Lee M, Kiernan MC, Macefield VG, Lee BB, Lin CSY. Short-term peripheral nerve 25. Commonwealth Department of Health and Aged Care. Note for Guidance on Good stimulation ameliorates axonal dysfunction after spinal cord injury. J Neurophysiol. Clinical Practice (CPMP/ICH/135/95). Commonwealth Department of Health and 2015;113:3209–3218. Aged Care; 2000.
Journal of Physiotherapy 62 (2016) 59 Journal of PHYSIOTHERAPY journal homepage: www.elsevier.com/locate/jphys Erratum Erratum to ‘StressModEx – Physiotherapist-led Stress Inoculation Training integrated with exercise for acute whiplash injury: study protocol for a randomised controlled trial’ [JPHYS 61/3 (2015) 157] Carrie Ritchie, Justin Kenardy, Rob Smeets, Michele Sterling * The above protocol abstract contained an incorrect provenance statement. The provenance statement should read: ‘Not invited. Peer reviewed.’ Journal of Physiotherapy apologises to the authors. DOI of original article: http://dx.doi.org/10.1016/j.jphys.2015.04.003 * Correspondence: Menzies Health Institute Queensland, Griffith University, Gold Coast, Australia 4022. E-mail address: m.sterling@griffith.edu.au (M. Sterling). http://dx.doi.org/10.1016/j.jphys.2015.12.002 Erratum Erratum to ‘A biopsychosocial primary care intervention (Back on Track) versus primary care as usual in a subgroup of people with chronic low back pain: protocol for a randomised, controlled trial’ [JPHYS 61/3 (2015) 155] Reni MA van Erp *, Ivan PJ Huijnen, Jeanine A Verbunt, Rob JEM Smeets The above protocol abstract contained an incorrect provenance statement. The provenance statement should read: 1DT[_]‘$IF Not invited. Peer reviewed.FID$T_[’]2 Journal of Physiotherapy apologises to the authors. DOI of original article: http://dx.doi.org/10.1016/j.jphys.2015.03.003 * Correspondence: Department of Rehabilitation Medicine, CAPHRI, Maastricht University, Maastricht, The Netherlands. E-mail address: [email protected] (R.M. van Erp). http://dx.doi.org/10.1016/j.jphys.2015.12.001 1836-9553/Crown Copyright ß 2015 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 62 (2016) 59 Journal of PHYSIOTHERAPY journal homepage: www.elsevier.com/locate/jphys Erratum Erratum to ‘StressModEx – Physiotherapist-led Stress Inoculation Training integrated with exercise for acute whiplash injury: study protocol for a randomised controlled trial’ [JPHYS 61/3 (2015) 157] Carrie Ritchie, Justin Kenardy, Rob Smeets, Michele Sterling * The above protocol abstract contained an incorrect provenance statement. The provenance statement should read: ‘Not invited. Peer reviewed.’ Journal of Physiotherapy apologises to the authors. DOI of original article: http://dx.doi.org/10.1016/j.jphys.2015.04.003 * Correspondence: Menzies Health Institute Queensland, Griffith University, Gold Coast, Australia 4022. E-mail address: m.sterling@griffith.edu.au (M. Sterling). http://dx.doi.org/10.1016/j.jphys.2015.12.002 Erratum Erratum to ‘A biopsychosocial primary care intervention (Back on Track) versus primary care as usual in a subgroup of people with chronic low back pain: protocol for a randomised, controlled trial’ [JPHYS 61/3 (2015) 155] Reni MA van Erp *, Ivan PJ Huijnen, Jeanine A Verbunt, Rob JEM Smeets The above protocol abstract contained an incorrect provenance statement. The provenance statement should read: 1DT[_]‘$IF Not invited. Peer reviewed.FID$T_[’]2 Journal of Physiotherapy apologises to the authors. DOI of original article: http://dx.doi.org/10.1016/j.jphys.2015.03.003 * Correspondence: Department of Rehabilitation Medicine, CAPHRI, Maastricht University, Maastricht, The Netherlands. E-mail address: [email protected] (R.M. van Erp). http://dx.doi.org/10.1016/j.jphys.2015.12.001 1836-9553/Crown Copyright ß 2015 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 62 (2016) 111 Journal of PHYSIOTHERAPY journal homepage: www.elsevier.com/locate/jphys Research Trial Protocol Implementation of a guideline-based clinical pathway of care to improve health outcomes following whiplash injury (Whiplash ImPaCT): protocol of a randomised, controlled trial Trudy Rebbeck a,e, Andrew Leaver a, Aila Nica Bandong a,b, Justin Kenardy c, Kathryn Refshauge a, Luke Connelly d, Ian Cameron e, Geoffrey Mitchell f, Simon Willcock g, Carrie Ritchie h, Jagnoor Jagnoor e,iF[,DT]_21I$ Michele Sterling h a Faculty of Health Sciences, Discipline of Physiotherapy; F]bDI[T_2$ Department of Physical Therapy, College of Allied Medical Professions, University of the Philippines, Manila, Philippines; c D$FR]3T_1[I ecover I[_]F4D1T$I njury Research F_DI5]$T1C[ entre; d Faculty of Health and Behavioural Sciences; F6]_ITeD[$ John Walsh Centre for Rehabilitation Research, Royal North Shore Hospital, University ofFID7]$[T_ Sydney; f]FDTI$_[01 Discipline of General Practice, School of Medicine, University of Queensland, Brisbane; TD_01F]gI[$ Clinical Director Primary Care, Macquarie University Hospital, Sydney; ]hFD$[1IT0_ $1[DIFT_]3Recover _TD$IF4]I[1 njury Research 1$F][C9DI_T entre, _IN]FT[2D0$ HMRC C]D[$FIT_12 RE $]2F_iIDT[ n D$TF]32[_RI oad D[TF]_TI4$2 raffic Injury, Menzies Health Institute Queensland, Griffith University, Gold Coast, Australia; iF]I[25_TD$ The George Institute for Global Health Abstract baseline assessments and will be categorised into low or medium/high risk of poor recovery using a clinical prediction rule. After this Introduction: Whiplash-associated disorders (WAD) are a huge assessment, participants will be randomly allocated to either a control worldwide health and economic burden. The propensity towards group (n = 118) or intervention group (n = 118), stratified by risk developing into chronic, disabling conditions drives the rise in health subgroup and treatment site. The participants’ nominated primary and economic costs associated with treatment, productivity loss and healthcare providers will be informed of their involvement in the trial. compulsory third party insurance claims. Current treatments fail to Consent will be obtained from the primary healthcare providers to address the well-documented heterogeneity of WAD and often result participate and to obtain information about professional practice. in poor outcomes. A novel approach is to evaluate whether the care Participants in the intervention group will additionally have access to provided according to the estimated risk of poor prognosis improves an interactive website that provides information about whiplash and health outcomes while remaining cost-effective. Research questions: recovery relative to their risk category. Analysis: Analysis will be (1) Does a guideline-based clinical pathway of care improve health conducted on an intention-to-treat basis. Outcomes will be analysed outcomes after whiplash injury compared to usual care? (2) Does risk independently through cross-sectional analyses using generalised of recovery have a differential effect on health outcomes for the linear models methods, with an appropriate link function, to test for clinical pathway of care? (3) Is the clinical pathway of care an intervention effect, adjusted for the baseline values. The risk intervention cost-effective? (4) What are the variations in professional category will be tested for its association with treatment effect by practice between usual care and the clinical pathway of care? Design: adding risk group to the regression equation. Cost-effectiveness will be Multi-centre, randomised, controlled trial conducted over two calculated using utility weights and the resulting measure will be cost Australian states: Queensland and New South Wales. Participants per quality-adjusted life year (QALY) saved. Professional practice and $DTIF_62s][ etting: 236 people with WAD (grade I-III, within 6 weeks of outcomes will be analysed using descriptive statistics. Discussion: F_[1T$DI]This injury) and their primary healthcare providers. Intervention: A research is significant as it will be the first study to address the clinical pathway of care, with care matched to the predicted risk of heterogeneity of whiplash by implementing a clinical pathway of care poor recovery. Participants at low risk of ongoing pain and disability that matches evidence-based interventions to projected risk of poor (hence, predicted to fully recover) will receive up to three sessions of recovery. The results of this trial have the potential to change clinical guideline-based advice and exercise with their primary healthcare practice for WAD, thereby maximising treatment effects, improving provider. Participants at medium/high risk of developing ongoing pain patient outcomes, reducing costs and maintaining the compulsory third and disability will be referred to a specialist (defined as a practitioner party system. with expertise in whiplash) who will conduct a more in-depth physical and psychological assessment. As a result, the specialist will Trial registration: Australian New Zealand Clinical Trials Registry liaise with the original primary healthcare provider and determine one (ANZCTR). Registration number: ACTRN12615001367538. Was this of three further pathways of care. Control: Usual care provided by the trial prospectively registered? Yes. Date of trial registration: primary healthcare provider that is based on clinical judgment. 16 December 2015. Funded by: Partnership grant from the National Measurements: Primary (global rating of change and neck-related Health and Medical Research Council, New South Wales Motor Accidents disability) and secondary (self-efficacy, pain intensity, general health Authority, and the Motor Accidents Insurance Commission of Queensland. and disability and psychological health) outcomes will be collected Funder approval number: NHMRC APP1075736. Anticipated comple- using validated scales. Direct (eg, professional care, transportation tion date: 2019. Provenance: Not invited. PeerD$T27F]I_[ reviewed. Corresponding costs, time spent for care, co-payments) and indirect (eg, lost economic author contact details: Dr Trudy Rebbeck, Faculty of Health Sciences, productivity) costs will be obtained through an electronic cost diary. Discipline of Physiotherapy, University of Sydney, New South Wales, Health and cost outcomes will be assessed at baseline, 3, 6 and Australia. Email: [email protected] 12 months after randomisation. Professional practice outcomes will be evaluated through questionnaires completed by healthcare providers Full protocol: Available on the eAddenda at doi:10.1016/j.jphys. and their patients at 3 months. Procedure: Potential participants 2016.02.006 (patients) will be identified through emergency departments, primary health clinics and advertisements. Eligible participants will complete http://dx.doi.org/10.1016/j.jphys.2016.02.006 1836-9553/ß 2016 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 62 (2016) 96–102 Journal of PHYSIOTHERAPY journal homepage: www.elsevier.com/locate/jphys Research Inpatient rehabilitation improves functional capacity, peripheral muscle strength and quality of life in patients with community-acquired pneumonia: a randomised trial Anderson Jose´ , Simone Dal Corso Post-graduate Program in Rehabilitation Sciences, Universidade Nove de Julho, Sa˜o Paulo, Brazil KEY WORDS ABSTRACT Randomised Question: Among people who are hospitalised for community-acquired pneumonia, does an inpatient controlled trial exercise-based rehabilitation program improve functional outcomes, symptoms, quality of life and Physical therapy (specialty) length of hospital stay more than a respiratory physiotherapy regimen? Design: Randomised trial with Pneumonia concealed allocation, intention-to-treat analysis and blinding of some outcomes. Participants: Forty- Exercise nine adults hospitalised for community-acquired pneumonia. Intervention: The experimental group Quality of life (n = 32) underwent a physical training program that included warm-up, stretching, peripheral muscle strength training and walking at a controlled speed for 15 minutes. The control group (n = 17) underwent a respiratory physiotherapy regimen that included percussion, vibrocompression, respiratory exercises and free walking. The intervention regimens lasted 8 days. Outcome measures: The primary outcome was the Glittre Activities of Daily Living test, which assesses the time taken to complete a series of functional tasks (eg, rising from a chair, walking, stairs, lifting and bending). Secondary outcomes were distance walked in the incremental shuttle walk test, peripheral muscle strength, quality of life, dyspnoea, lung function, C-reactive protein and length of hospital stay. Measures were taken 1 day before and 1 day after the intervention period. Results: There was greater improvement in the experimental group than in the control group on the Glittre Activities of Daily Living test (mean between-group difference 39 seconds, 95% CI 20 to 59) and the incremental shuttle walk test (mean between-group difference 130 m, 95% CI 77 to 182). There were also significantly greater improvements in quality of life, dyspnoea and peripheral muscle strength in the experimental group than in the control group. There were no between-group differences in lung function, C-reactive protein or length of hospital stay. Conclusion: The improvement in functional outcomes after an inpatient rehabilitation program was greater than the improvement after standard respiratory physiotherapy. The exercise training program led to greater benefits in functional capacity, peripheral muscle strength, dyspnoea and quality of life. Registration: ClinicalTrials.gov, NCT02103400 [Jose´ A, Dal Corso S (2016) Inpatient rehabilitation improves functional capacity, peripheral muscle strength and quality of life in patients with community-acquired pneumonia: a randomised trial. Journal of Physiotherapy 62: 96–102] ß 2016 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Introduction supported by evidence7–10 and the main guidelines for the management of this condition do not recommend it.1 In patients Community-acquired pneumonia is a highly prevalent adverse health condition with a high mortality rate. It involves substantial who are hospitalised for acute exacerbations of chronic obstructive treatment costs and has significant social impact worldwide.1 Patients who are hospitalised for community-acquired pneumonia pulmonary disease (COPD), an inpatient rehabilitation program experience a decline in functional capacity,2–4 which is associated with higher rates of re-hospitalisation and death,3,4 as well as leads to some immediate improvements in functional capacity, reductions in both peripheral muscle strength and quality of life.2 quality of life, peripheral muscle strength, exercise tolerance,11,12 Moreover, such patients can endure a long period before the anxiety and depression.13 However, the recent publication of a major complete remission of symptoms and the return to previous activities of daily living.5 study by Greening and colleagues has indicated that very early Although widely employed in clinical practice,6 the current exercise-based rehabilitation commenced during hospitalisation physiotherapeutic approach for patients with community-ac- quired pneumonia, which focuses on airway clearance, is not may reduce uptake of pulmonary rehabilitation and increase mortality after discharge.14 Interpreting this study in the light of the existing evidence suggests that for patients who are hospitalised for acute exacerbations of COPD, pulmonary rehabilitation immedi- ately after discharge may be more beneficial overall than commenc- ing the exercise-based rehabilitation during hospitalisation.15 http://dx.doi.org/10.1016/j.jphys.2016.02.014 1836-9553/ß 2016 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Research 97 In patients who are hospitalised for community-acquired pneumo- were concealed in opaque, sealed envelopes that had been nia, early mobilisation has been applied only in one large study, but it prepared by a researcher who was not involved in the recruitment was poorly described.16 Therefore, in these patients, the effects of or assessment of the participants. Data were obtained before aerobic and resistance training on functional capacity, peripheral randomisation and 1 day after the 8-day intervention period. The muscle strength and quality of life require thorough investigation length of the hospital stay was also recorded for all participants during hospitalisation. Considering the high prevalence and (Figure 1). treatment costs of community-acquired pneumonia, its social impact and the scarcity of evidence to support standard respiratory Participants, therapists and centre physiotherapy for such patients, it is crucial to investigate whether a physical training program can lead to better recovery of functional The inclusion criteria for participation in the study were: being capacity at discharge from the hospital. aged > 18 years, having a diagnosis of community-acquired pneumonia,1 being hospitalised for < 48 hours, and having Therefore, the research question for this randomised trial was: adequate awareness and independent ambulation. The exclusion Among people who are hospitalised for community-acquired criteria were: being unwilling to participate, having cognitive pneumonia, does an inpatient exercise-based rehabilitation impairment, having osteoarticular disorders, and having other program improve functional outcomes, symptoms, quality of life acute or chronic respiratory diseases. The researchers in the study and length of hospital stay more than a respiratory physiotherapy applied the interventions. This study was conducted at Mandaqui regimen? Hospital, which is a tertiary hospital in Sa˜o Paulo, Brazil. Method Interventions Design Control group Participants allocated to the control group received standard This study was a randomised trial, with concealed allocation, blinding of assessors of some outcomes and intention-to-treat respiratory physiotherapy in daily 50-minute sessions for 8 days; analysis. Patients who were hospitalised for community-acquired this involved secretion removal, breathing exercises and walking. pneumonia were randomly assigned to receive either physical The secretion removal techniques were percussion and vibrocom- exercise training (experimental group) or respiratory physiother- pression in side lying for 10 minutes on each side, during and after apy (control group). After eligible patients were advised regarding which the participant was instructed to perform voluntary the study and consented to participate, they were randomly coughing and huffing to expectorate secretions until achieving a dry cough.17 ]GIF$DT)1_er[(iguallocated to one of the two groups. Upcoming random allocations Patients hospitalised for community- acquired pneumonia (n = 65) Excluded (n = 14) • declined (n = 8) • COPD (n = 2) • cognitive impairment (n = 1) • osteoarticular disorders (n = 1) • atrial fibrillation (n = 1) • diagnosis changed from CAP (n = 1) Randomised (n = 51) Day 1 Measured Glittre Test, ISWT, peripheral muscle strength, SF-36, MRC, CRP, spirometry and BMI (n = 33) (n = 18) Withdrawn Experimental group Control group Withdrawn • diagnosis • stretching •bronchial cleaning • diagnosis • resisted exercises •respiratory exercises changed (n = 1) • aerobic exercises changed (n = 1) • 1 session daily •free walking Lost to follow-up x 8 days •1 session daily Lost to follow-up • early discharge • early discharge x 8 days (n = 5) (n = 1) • thorax drainage (n = 1) • rest prescription (n = 1) Day 10 Measured Glittre Test, ISWT, peripheral muscle strength, SF-36, MRC, CRP and spirometry (n = 27) (n = 14) Discharge Recorded length of stay in hospital (n = 32) (n = 17) Figure 1. Design and flow of participants through the trial. BMI = body mass index, COPD = chronic obstructive pulmonary disease, CRP = C-reactive protein, ISWT = incremental shuttle walk test, MRC = Medical Research Council Dyspnoea Scale, SF-36 = Short Form 36 questionnaire.
98 Jose´ and Dal Corso: Inpatient rehabilitation for community-acquired pneumonia The breathing exercises were targeted at improving ventilation. day, and the longer distance that was walked was used for the They included diaphragmatic breathing exercises (three sets of analysis. 10 repetitions with a 1-minute rest period between sets) and inspiratory exercise with maximum inspiration and maximum Peripheral muscle strength was measured using a inspiratory pause (three sets of 10 repetitions with a 1-minute rest dynamometerb. The peak of isometric contraction was recorded period between sets).17 Walking was self-paced with a prescribed for the biceps brachii, deltoid, quadriceps femoris and hamstring duration of 10 minutes. muscles. Three maximum isometric contractions were performed, and the highest value was considered in the analysis.25 Experimental group The participants allocated to the experimental group received Quality of life was measured using the Short Form 36 question- naire (SF-36),26 which has eight subscales: physical functioning, physical training in daily 50-minute sessions for 8 days; this physical role functioning, pain, general health state, vitality, social involved warm-up, stretching, resistance exercises for peripheral role functioning, emotional role functioning, and mental health. muscles and aerobic walking training. The score for each subscale ranges from 0 to 100 points, with higher scores denoting better quality of life. The warm-up involved active movements of the upper and lower limbs for approximately 5 minutes. Stretching targeted the Dyspnoea was measured using the Medical Research Council pectoralis major, latissimus dorsi, trapezius, quadriceps femoris scale,27D_8I$]F[T which is composed of five activities on which breathless- and hamstring muscles. Participants held each stretch for ness is scored from 1 to 5. Higher scores denote greater limitations 30 seconds and stretched for approximately 5 minutes in total. to activities of daily living. Peripheral muscle strengthening exercises were performed for Pulmonary function was measured using a portable approximately 25 minutes. The exercises were performed with spirometerc. The technical procedures were based on Brazilian three sets of eight repetitions with a 1-minute rest period between guidelines.28 The data were expressed as absolutes and percen- sets for both limbs simultaneously using an elastic banda. The tages of predicted values for the forced expiratory volume in one targeted muscles were biceps brachii, deltoids, quadriceps femoris second (FEV1) and the forced vital capacity (FVC).29 and hamstrings. The initial workload was 70% of maximum peripheral muscle strength.18 At the end of each set, dyspnoea and Blinded assessors determined the following outcomes: inflam- fatigue were measured using the Borg scale,19 with participants mation was measured based on C-reactive protein (CRP) using a aiming to score between 4 and 6 at the end of the three sets. If the venous blood sample. The respiratory physician, who was blinded fatigue score was below this level, the workload was increased by to the participants’ group allocation, decided on the length of exchanging the elastic band for one more resistant, and if above hospital stay. Hospital records were used to confirm whether there this level, the workload was decreased. were any deaths before discharge. Aerobic training was performed in a flat corridor that was 10 m The safety of the interventions was evaluated based on the long. Participants walked for 15 uninterrupted minutes, with their observation and occurrence of adverse events such as nausea, speed guided by the auditory recording of the endurance shuttle lightheadedness, significant dyspnoea, extreme fatigue, chest pain, walk test20 set at a speed corresponding to 70% of the speed arrhythmia, syncope, altered consciousness or severe desaturation reached on the incremental shuttle walk test (ISWT).21 If the (SpO2 80%). participant was unable to tolerate this time or exhibited a drop in pulse oxygen saturation (SpO2) < 84%, he or she was allowed to Data analysis rest on a chair, at which point the timer was paused. Upon feeling capable of proceeding, the participant continued to walk until The sample size was calculated in relation to the Glittre completing the entire 15 minutes. Adjustments in the training Activities of Daily Living test24 as the main outcome and was based intensity were made according to symptoms (dyspnoea and on results obtained from a pilot study with five participants in each fatigue between 4 and 6 on the Borg scale)19 and/or 70% of the group. An effect size of 1.3 was estimated from a mean time of predicted maximum heart rate, which was established using 213 seconds to perform the Glittre Activities of Daily Living test in Karvonen’s equation.22 If a participant reported a dyspnoea score < the control group and 153 seconds for the experimental group, 4 and/or the heart rate remained below the rate established by with a SD within each group of 45 seconds. Unequal treatment Karvonen’s equation, the speed of the walk was increased the allocation (2:1) was assumed due to ethical reasons for maximising following day, and if dyspnoea was > 6 and/or the heart rate was above the rate established in the equation, the speed of the walk participants’ exposure to the experimental group. Assuming an a was diminished the following day. error of 0.05, a b error of 0.20 and an allocation ratio of 2:1, the Outcome measures sample size was determined to be 24 participants in the intervention group and 12 in the control group. All evaluations were conducted on the first and tenth days of the study (Figure 1). The data that were collected to characterise The Shapiro-Wilk test was used to determine the distribution the participants at baseline included age, gender, body mass index (normal or non-normal) of the data. Baseline characteristics were and the CURB-65 score, which is a tool for predicting risk of death summarised using mean (SD) for parametric data and median from community-acquired pneumonia.23]6F[_DTI$ (IQR) for non-parametric data. The continuous outcome data were analysed as mean (SD) of two groups, mean (SD) within-group Primary outcome difference and mean (95% CI) between-group difference. The Functional exercise capacity was measured using the Glittre standardised effect size was calculated using Cohen’s d. Pearson’s correlation coefficients were calculated to determine the strength Activities of Daily Living test,24 which consists of a set of functional of correlations between variables. The data were analysed based on activities (eg, rising from a chair, walking, stairs, lifting and intention-to-treat analysis. A p-value < 0.05 was considered bending). The participant performs the sequence of activities five significant. times as quickly as possible. Each participant performed the entire test twice, and the main outcome was the better time taken to Results complete five laps. Compliance with the study protocol Secondary outcomes Exercise capacity was measured using the ISWT3D[],$IFT_ performed as The target sample size in the registered protocol was 30 participants. This was based on a calculation using the previously described.21$F]DTI_7[ Two tests were performed on the same difference to be detected from a study by Skumlien and colleagues24 and the SD from a study by Jose´ and colleagues.2 When the results of our pilot study became available early in the
Research 99 Table 1 study due to the constraints of their medical management Baseline characteristics of participants. (Figure 1). All participants permitted their length of stay to be recorded. Characteristic Randomised (n = 49) The baseline characteristics of the groups were similar. These Exp (n = 32) Con (n = 17) data are presented in Table 1IF[$]DT1_2 and the first two data columns of Table 2, Table 3 and Table 4. Individual participant data are Age (yr), mean (SD) 51 (21) 59 (18) presented in Table 5 on the eAddenda. 17 (53) 10 (59) Gender, n males (%) 23 (4) 25 (6) Effect of interventions Body mass index, (kg/m2), mean (SD) 1 (1 to 1) 1 (1 to 2) CURB-65, median (IQR) Exp = experimental group, Con = control group. conduct of the present study, the sample size calculation outlined Primary outcome in the Data analysis section, above, was considered to be the better Functional exercise capacity, as measured by the Glittre choice for the present study. This exceeded the registered target sample size. Activities of Daily Living test, improved by a mean of 52 seconds (SD 40) in the experimental group, whereas the control group One registered outcome measure, Tumour Necrosis Factor- improved by a mean of 12 seconds (SD 26). This was a statistically significant difference (MD 39 seconds, 95% CI 20 to 59), as shown in alpha (TNFa), was not reported.TDF$[I]_01 In principle, the analysis of TNFa Table 2. When calculated as a standardised effect size, the between-group difference of 39 seconds equated to a Cohen’s d should have been carried out in the hospital. However, the hospital of 1.19. carried out the analysis on seven participants and did not allow further collection due to the high cost of the test. The limited data Secondary outcomes are available from the authors on request. The improvement in functional capacity, as evaluated by the Flow of participants through the study distance walked on the ISWT, was significantly greater in the experimental group than in the control group (MD 130 m, 95% CI Of the 65 patients admitted with an initial diagnosis of 77 to 182). When calculated as a standardised effect size, the community-acquired pneumonia during the study recruitment between-group difference equated to a Cohen’s d of 1.39. No period, 14 were excluded based on the eligibility criteria (Figure 1). significant between-group differences occurred in the variables A further two participants were 1[w]FID$_T ithdrawn after randomisation monitored during the ISWT: heart rate, percentage of predicted because their diagnosis of community-acquired pneumonia maximal heart rate, SpO2 and Borg dyspnoea and fatigue scales changed during reassessment by a physician; they were not (data not shown). The SpO2 during the aerobic training did not included in the intention-to-treat analysis. Therefore, the study show any desaturation < 84%, as outlined in the Methods section. reports data on 49 participants: 32 in the experimental group and 17 in the control group. Five participants in the experimental group All muscles that were analysed demonstrated mean improve- and one in the control group were discharged before the scheduled ment in strength in the experimental group and mean deteriora- post-intervention reassessment on Day 10, but these participants tion in the control group, as presented in Table 2. The between- could be included in the analysis of length of hospital stay. group difference in change in strength was statistically significant A further two participants in the control group withdrew from the for all four muscle groups. The between-group differences in change in strength were: 3.5 kgf for biceps brachii (Cohens’ d = 0.76); 2.2 kgf Table 2 Mean (SD) of groups, mean (SD) difference within groups and mean (95% CI) difference between groups for functional capacity tests, peripheral muscle strength, dyspnoea and inflammation. Outcomes Groups Difference within groups Difference between groups Day 10 minus Day 1 Day 10 minus Day 1 Day 1 Day 10 Glittre ADL Test (s) Exp (n = 32) Con (n = 17) Exp (n = 32) Con (n = 17) Exp Con Exp minus Con Incremental Shuttle Walk Test (m) Biceps brachii strength (kgf) 229 (86) 223 (52) 177 (93) 211 (46) –52 (40) –12 (26) –39 (–59 to –20) Deltoids strength (kgf) 355 (126) 313 (91) 517 (184) 346 (94) 162 (110) 33 (71) 130 (77 to 182) Quadriceps strength (kgf) 12.9 (4.7) 12.9 (5.6) 15.7 (4.8) 12.2 (4.6) 2.7 (2.5) –0.7 (4.8) 3.5 (0.9 to 6.0) Hamstrings strength (kgf) 5.4 (2.0) 5.3 (2.5) 7.6 (2.5) 5.3 (2.4) 2.2 (1.8) 0.0 (1.5) 2.2 (1.2 to 3.2) MRC Dyspnoea scale (1 to 5) 25.6 (5.7) 24.2 (8.3) 32.5 (8.5) 22.6 (6.7) 6.9 (6.1) –1.6 (5.5) 8.5 (4.6 to 11.6) C-reactive protein (mg/l) 15.5 (5.7) 14.4 (5.8) 21.4 (7.6) 13.9 (4.2) 5.9 (4.7) –0.5 (3.5) 6.5 (4.0 to 8.8) 3.1 (1.1) 2.5 (0.9) 1.6 (0.9) 1.9 (0.7) –1.5 (1.1) –0.6 (0.7) –0.9 (–1.4 to –0.4) 11 (13) –7 (13) 14 (14) 5 (7) 5 (5) –9 (14) –2 (–10 to 6) ADL = activities of daily living, Con = control group, Exp = experimental group, MRC = Medical Research Council. Table 3 Mean (SD) of groups, mean (SD) difference within groups, and mean (95% CI) difference between groups for health-related quality of life assessed using the Short Form 36 (SF- 36) questionnaire. Domains Groups Difference within groups Difference between groups Day 1 Day 10 Day 10 minus Day 1 Day 10 minus Day 1 Exp Con Exp minus Con Exp (n = 32) Con (n = 17) Exp (n = 32) Con (n = 17) Physical functioning 52 (20) 52 (26) 80 (21) 66 (20) 28 (22) 14 (24) 13 (1 to 28) Physical role functioning 37 (31) 40 (37) 58 (30) 45 (36) 21 (28) 5 (26) 16 (–1 to 32) Pain 41 (22) 37 (26) 55 (23) 43 (20) 14 (16) 6 (28) 8 (–8 to 23) General health state 58 (16) 56 (23) 70 (12) 63 (22) 12 (13) 7 (18) 5 (–5 to 15) Vitality 48 (13) 48 (20) 69 (16) 68 (15) 21 (21) 20 (23) 0 (–13 to 14) Social role functioning 62 (24) 43 (21) 68 (22) 54 (24) 7 (17) 11 (14) –5 (–14 to 5) Emotional role functioning 43 (35) 22 (29) 57 (32) 38 (39) 15 (21) 16 (29) –2 (–18 to 15) Mental health 53 (20) 49 (18) 60 (19) 53 (19) 4 (17) 3 (–6 to 12) 7 (9) Exp = experimental group, Con = control group.
100 Jose´ and Dal Corso: Inpatient rehabilitation for community-acquired pneumonia Table 4 Mean (SD) of groups, mean (SD) difference within groups, and mean (95% CI) difference between groups for lung function. Outcome Groups Difference within groups Difference between groups Day 1 Day 10 Day 10 minus Day 1 Day 10 minus Day 1 Exp (n = 32) Con (n = 17) Exp (n = 32) Con (n = 17) Exp Con Exp minus Con FEV1 (l) 2.05 (0.66) 2.20 (0.86) 2.21 (0.63) 2.15 (0.59) 0.16 (0.26) –0.06 (0.62) 0.21 (–0.11 to 0.54) FEV1 (% pred) 69 (11) 75 (12) 74 (10) 75 (12) 6 (8) 1 (15) 5 (–3 to 13) FVC (l) 2.55 (0.74) 2.66 (0.97) 2.71 (0.65) 2.67 (0.76) 0.16 (0.29) 0.01 (0.49) 0.15 (–0.12 to 0.42) FVC (% pred) 69 (11) 71 (10) 74 (8) 73 (9) 5 (8) 2 (11) 3 (–3 to 9) Con = control group, Exp = experimental group, FEV1 = forced expiratory volume in one second, FVC = forced vital capacity, % pred = percentage of the predicted value.29D$_5[ITF] for deltoids (Cohen’s d = 1.45); 8.1 kgf for quadriceps (Cohen’s observed when comparing the active cycle of breathing techniques d = 1.46); and 6.4 kgf for hamstrings (Cohen’s d = 1.54). with usual care, including no difference regarding hospitalisation length, duration of antibiotic therapy, mean duration of sputum The domain ‘physical functioning’ of the SF-36 quality of life production or inpatient sputum weight. questionnaire improved significantly more in the experimental group than in the control group (MD 14 points, 95% CI 1 to 28, As previous studies have demonstrated that standard Cohen’s d = 0.48). None of the other domains of the SF-36 showed respiratory physiotherapy has no impact on clinical measures statistically significant effects (Table 3). of the resolution of pneumonia, it is interesting to observe (albeit without a no-intervention control group for comparison) Both groups showed an improvement on average in the severity the change in functional outcomes in the group that received of dyspnoea, as measured on the Medical Research Council scale respiratory therapy in the present study. This group did not during the 10 days of the study. However, the experimental group show substantial improvement in the distance walked on the showed significantly greater improvement (MD 0.9 points, 95% CI ISWT. There is no minimum clinically important difference 0.4 to 1.4, Cohen’s d = 0.98). established for the ISWT specifically for patients hospitalised for community-acquired pneumonia, but the magnitude of the Pulmonary function was reduced to a similar degree compared average improvement observed in this group was inferior to that to predicted values in both groups (Table 4). After the interven- found after an outpatient rehabilitation program for both tions, no substantial change in pulmonary function was detected. patients with non-cystic fibrosis bronchiectasis (37 m)30 and with COPD (48 m).31 The same lack of substantial improvement Both groups showed a reduction on average in CRP during the was observed in the group that received respiratory therapy 10 days of the study. The amount of reduction was similar in the for the Glittre Activities of Daily Living test, where again the two groups. No correlations were found between CRP and the other best estimate of the minimum clinically important difference variables studied. Moreover, the change in CRP was not correlated is indirect because it comes from patients with COPD with peripheral muscle strength or improvement in the perfor- (53 seconds).24 mance of the functional capacity tests. No association was found between hospital length of stay and All participants survived long enough to be discharged from the type of treatment. However, in a study conducted by Mundy et al, a hospital. The type of treatment did not significantly influence length reduction in hospital length of stay was found when early of stay in hospital: median 12 days (IQR 10 to 18) in the experimental mobilisation was performed.16 This divergence may have occurred group and median 13 days (IQR 11 to 25) in the control group. None because the hospital length of stay in the present study depended of the adverse events listed in the Methods section were identified on factors other than clinical criteria, such as the economic status during the application of the allocated interventions. of the patient, the ability to afford antibiotic therapy, cognition, capacity to self-medicate correctly and administrative aspects of Discussion the hospital such as availability of hospital beds. It is believed that this is the first study to investigate the effects All participants were discharged, so there was no association of an inpatient rehabilitation program on patients with communi- between mortality and the type of treatment. This was due to the ty-acquired pneumonia by comparing them with those of standard low severity of cases, as represented by the low CURB-65, respiratory physiotherapy. The results were superior to those of corresponding to prognosis of a low degree of lethality.23 A standard respiratory physiotherapy. previous study had demonstrated that early mobilisation did not interfere in the mortality of these patients, with no change in the Although it is common practice,6FI$D_9[]T there is no evidence to support re-hospitalisation rate within a 90-day period.16 the routine use of standard respiratory physiotherapy in patients who are hospitalised for community-acquired pneumonia. A No important complications occurred due to the intervention randomised clinical trial examined intervention for pneumonia protocols, indicating that the protocols are safe and could be that included postural drainage, percussion, thoracic vibration and performed on patients with community-acquired pneumonia positive pressure, but failed to demonstrate changes in fever, the whose characteristics are similar to those of the present sample. extent of radiographic impairment, length of hospital stay, or Physical exercise with early mobilisation of hospitalised patients mortality.7 These results are in line with the results of another has previously been performed and appears to be safe for patients study of techniques such as postural drainage, external support for with COPD,11–13 asthma,32 interstitial lung disease33 and commu- breathing, percussion and vibration in pneumonia.8 There was no nity-acquired pneumonia.16 However, as discussed above, exer- overall effect and in younger patients, smokers and patients with cise-based rehabilitation commenced during hospitalisation for interstitial pneumonia specifically, the intervention led to an COPD exacerbation may reduce uptake of pulmonary rehabilita- increase in the duration of fever and length of hospital stay.8 A tion and increase mortality after discharge,14 so pulmonary systematic review of adjunctive therapies for patients hospitalised rehabilitation immediately after discharge may be more beneficial for community-acquired pneumonia has shown that clinical trials for patients with COPD.15 with such respiratory physiotherapy techniques are scarce and do not provide evidence of benefits from their routine use in these Previous studies have demonstrated that short-term hospita- patients.9 lisation reduces functional capacity, even in patients who are not bedridden, regardless of age or initial functional status.34 In a In a recent meta-analysis, Yang et al10 compared standard recent study, a decline in functional capacity as well as reductions respiratory physiotherapy to usual care and found no significant in both peripheral muscle strength and quality of life2 were differences regarding mortality, the resolution of the disease, demonstrated in patients hospitalised for community-acquired improvement in chest radiography or healing time. The same was
Research 101 pneumonia. Therefore, interventions to minimise such declines are What is already known on this topic: People who are crucial for these patients. hospitalised for community-acquired pneumonia experience a decline in functional capacity that is associated with higher Mundy et al evaluated the effects of early mobilisation rates of re-hospitalisation and death. Existing evidence does performed by nurses on 456 patients who were hospitalised for not support the use of respiratory techniques, although these community-acquired pneumonia and found that hospitalisation remain standard practice in some hospitals. length and costs were reduced, with no adverse events.16 However, What this study adds: Among people who are hospitalised this intervention was poorly described and the control group did for community-acquired pneumonia, an exercise training pro- not receive any kind of standard respiratory physiotherapy. The gram led to greater benefits in functional capacity, peripheral present study is the first to demonstrate the relative effects of a muscle strength, dyspnoea and quality of life than a regimen physical training program on functional capacity for hospitalised that included respiratory techniques and self-paced walking. patients with community-acquired pneumonia. The aerobic and resistance training provided a more effective recovery of functional Footnotes: aThera BandTM, The Hygenic Corporation, Akron, capacity than standard respiratory physiotherapy. Ohio, USA. bDT_4$]FI1[3 Kratos, Sa˜o Paulo, Brazil.5DF_[TI1$] cFI6_D$T]1[ Pony, Cosmed, Rome, Italy. The loss of peripheral muscle strength is a common adverse eAddenda: Table 5 can be found online at doi:10.1016/j.jphys. change in hospitalised patients, even during short periods of 2016.02.014. hospitalisation.34 The interventions performed in the present study led to a significant increase in peripheral muscle strength of Ethics approval: The Ethics Committee of the Universidade the exercised muscles, whereas no such increase occurred in the Nove de Julho approved this study (273811). All participants gave control group. The training to improve peripheral muscle strength written informed consent before starting the baseline assessments. was performed with elastic bands due to the difficulty in transporting and storing weights in a hospital setting with limited Competing interests: The authors declare no conflict of interest space. Elastic bands are a low-cost, highly practical alternative in this research. with no need for more expensive equipment, which hospitals often cannot provide, and are as effective as other recources.35 Source(s) of support: The authors declare there was no source of funding in this research. The evaluation of dyspnoea using the Medical Research Council scale revealed a greater reduction in the experimental group than Acknowledgements: Nil. in the control group, although both groups improved substantially. Provenance: Not T7_DIF]i1$[ nvited. PeerITF]D1[_8$ reviewed. However, if a reduction of 0.58 is considered clinically significant Correspondence: Simone Dal Corso, Post-graduate Program in for the Medical Research Council scale,36 this reduction occurred in Rehabilitation Sciences, Universidade Nove de Julho, Sa˜o Paulo, both groups. Besides the improvement in the clinical status and the Brazil. Email: [email protected], simonedc@ resolution of the lung disease, it is possible that the 10-minute uninove.br walk in the control group as well as the aerobic and resistance training in the experimental group both led to a lower sensation of References dyspnoea upon discharge from the hospital. 1. Mandell LA, Wunderink RG, Anzueto A, Bartlett JG, Campbell D, Dean NC. Infectious The magnitude of change in CRP was similar in both groups, Diseases Society of America /American Thoracic Society consensus guidelines on suggesting that the common management or natural recovery the management of community-acquired pneumonia in adults. Clin Infect Dis. (rather than the randomised interventions) determined the 2007;44(Suppl 2):S27–S72. reduction in CRP. The association between CRP and functional capacity was also investigated in the present study, but was not 2. Jose´ A, Dal Corso S. Patients hospitalized for community-acquired pneumonia significant. This was likely due to the low severity of community- present reduced functional performance. Braz J Phys Ther. 2013;17:351–358. acquired pneumonia, since previous studies with severe patients with high levels of CRP found such an association, along with a 3. Solh AE, Pineda L, Bouquin P, Mankowski C. Determinants of short and long term worse prognosis. Moreover, a high level of CRP upon admission to functional recovery after hospitalization for community-acquired pneumonia in the hospital, the maintenance of this level after treatment and a the elderly: role of inflammatory markers. BMC Geriatr. 2006;6:12. small reduction in comparison to the initial level have also been associated with a worse prognosis as well as greater morbidity and 4. Murcia J, Llorens P, Sanches-Paya J, Reus S, Boix V, Merino E. Functional status mortality rates.37,38 determined by Barthel Index predicts community acquired pneumonia mortality in general population. J Infection. 2010;61:458–464. A limitation of the present study was that the researchers who evaluated the Glittre Activities of Daily Living test, ISWT, quality of 5. Metlay JP, Fine MJ, Schulz R, Marrie TJ, Coley CM, Kapoor WN. Measuring symp- life, dyspnoea, peripheral muscle strength and spirometry were the tomatic and functional recovery in patients with community-acquired pneumonia. same as those who performed the therapeutic interventions. J Gen Intern Med. 1997;12:423–430. However, the evaluations were standardised with written guide- lines to minimise the potential for bias from unblinded assessors. 6. Lamy O, Van Melle G, Cornuz J. Burnand. Clinical management of immunocompe- Furthermore, the assessors were blinded for some key study tent hospitalized patients with community-acquired pneumonia. Eur J Intern Med. outcomes such as CRP, length of stay and outcome (death or 2004;15:28–34. hospital discharge). 7. Graham WG, Bradley DA. Efficacy of chest physiotherapy and intermittent posi- In summary, this study identified significant improvements in tive-pressure breathing in the resolution of pneumonia. N Engl J Med. 1978;299: functional capacity, peripheral muscle strength, dyspnoea and 624–627. quality of life with inpatient exercise rehabilitation as opposed to respiratory interventions. It is believed that these benefits, in 8. Britton S, Bejstedt M, Vedin L. Chest physiotherapy in primary pneumonia. Br Med J. conjunction with other evidence about the benefits of inpatient 1985;290:1703–1704. exercise in this population,16 are sufficient to recommend the use of an inpatient rehabilitation program – especially where the 9. Siempos II, Vardakas KZ, Kopterides P, Falagas ME. Adjunctive therapies for current routine treatment for these patients is respiratory community-acquired pneumonia: a systematic review. J Antimicrob Chemother. physiotherapy techniques, since the latter lacks robust evidence 2008;62:661–668. to support its routine use in this population. Given the unexpected effects of inpatient exercise rehabilitation in COPD exacerba- 10. Yang M, Yan Y, Yin X, Wang BY, Wu T, Liu GJ. Chest physiotherapy for pneumonia in tions,14 however, further research should investigate the effects of adults. Cochrane Database Syst Rev. 2013;2:CD006338. the intervention of people with community-acquired pneumonia after discharge from hospital. 11. Troosters T, Probst VS, Pitta F, Gayan-Ramirez G, Decramer M. Resistance training prevents deterioration in quadriceps muscle function during acute exacerbation of chronic obstructive pulmonary disease. Am J Crit Care Med. 2010;181:1072–1077. 12. Reid WD, Yamabayashi C, Goodridge D, Chung F, Hunt MA, Marciniuk DD. Exercise prescription for hospitalized people with chronic obstructive pulmonary disease and comorbidities: a synthesis of systematic reviews. Int J Chron Obstruct Pulmon Dis. 2012;7:297–320. 13. Garuti G, Cilione C, Dell’orso D, Gorini P, Lorenzi MC, Totaro L. Impact of compre- hensive pulmonary rehabilitation on anxiety and depression in hospitalized COPD patients. Monaldi Arch Chest Dis. 2003;59:56–61. 14. Greening NJ, Williams JE, Hussain SF, Harvey-Dunstan TC, Bankart MJ, Chaplin EJ. An early rehabilitation intervention to enhance recovery during hospital admis- sion for an exacerbation of chronic respiratory disease: randomised controlled trial. BMJ. 2014;349:g4315. 15. Holland AE. Physiotherapy management of acute exacerbations of chronic ob- structive pulmonary disease. J Physiother. 2014;60:181–188. 16. Mundy LM, Leet TL, Darst K, Schnitzler MA, Dunagan WC. Early mobilization of patients hospitalized with community-acquired pneumonia. Chest. 2003;124: 883–889.
102 Jose´ and Dal Corso: Inpatient rehabilitation for community-acquired pneumonia 17. Webber BA, Pryor JA. Physiotherapy for respiratory and cardiac problem. 2nd edition 28. Sociedade Brasileira de Pneumologia e. Tisiologia. Diretrizes para testes de func¸a˜o Edinburgh: Churchill Livingstone; 1998. pulmonar. J Bras Pneumol. 2002;28:S44–S58. 18. Page PA, Labbe A, Topp RV. Clinical force production of Thera-Band elastic bands. J 29. Pereira CA, Sato T, Rodrigues SC. New reference values for forced spirometry in Orthop Sports Phys Ther. 2000;30:A47–A48. white adults in Brazil. J Bras Pneumol. 2007;33:397–406. 19. Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc. 30. Lee AL, Hill CJ, Cecins N, Jenkins S, McDonald CF, Burge AT. Minimal important 1982;14:377–381. difference in field walking tests in non-cystic fibrosis bronchiectasis following exercise training. Respir Med. 2014;108:1303–1309. 20. Revill SM, Morgan MDL, Singh SJ, Williams J, Hardman AE. The endurance shuttle walk: a new field test for the assessment of endurance capacity in chronic 31. Singh SJ, Jones PW, Evans R, Morgan MD. Minimum clinically important improve- obstructive pulmonary disease. Thorax. 1999;54:213–222. ment for the incremental shuttle walking test. Thorax. 2008;63:775–777. 21. Singh SJ, Morgan MD, Scot S, Walters D, Hardman AE. Development of a Shuttle 32. Clini E, Foglio K, Bianchi L, Porta R, Vitacca M, Ambrosino N. In-hospital short-term Walk Test of a disability in patients of chronic airways obstruction. Thorax. training program for patients with chronic airway obstruction. Chest. 2001;120: 1992;47:1019–1024. 1500–1505. 22. Karvonen JJ, Kentala E, Mustala O. The effects of training on heart rate: a ‘‘longi- 33. Huppmann P, Sczepanski B, Boensch M, Winterkamp S, Scho¨ nheit-Kenn U, Neurohr tudinal’’ study. Ann Med Exp Biol Fenn. 1957;35:307–315. C. Effects of inpatient pulmonary rehabilitation in patients with interstitial lung disease. Eur Respir J. 2013;42:444–453. 23. Chalmers JD, Singanayagam A, Akram AR, Mandal P, Short PM, Choudhury G. Severity assessment tools for predicting mortality in hospitalised patients with 34. Suesada MM, Martins MA, Carvalho CR. Effect of short-term hospitalization on community-acquired pneumonia. Systematic review and meta-analysis. Thorax. functional capacity in patients not restricted to bed. Am J Phys Med Rehabil. 2010;65:878–883. 2007;86:455–462. 24. Skumlien S, Hagelund T, Bjørtuft Øsah, Ryg MS. A field test of functional status as 35. Colado JC, Garcia-Masso X, Pellicer M, Alakhdar Y, Benavent J, Cabeza-Ruiz R. A performance of activities of daily living in COPD patients. Respir Med. comparison of elastic tubing and isotonic resistance exercises. Int J Sports Med. 2006;100(2):316–323. 2010;31:810–817. 25. Brown LE, Weir JP. Recomendac¸a˜o de Procedimentos da ASEP I: Avaliac¸a˜o Precisa 36. Binazzi B, Lanini B, Romagnoli I, Garuglieri S, Stendardi L, Bianchi R. Dyspnea da Forc¸a e Poteˆncia Muscular. R Bras Ci e Mov. 2003;11:95–110. during speech in chronic obstructive pulmonary disease patients: effects of pulmonary rehabilitation. Respiration. 2011;81:379–385. 26. Ciconelli RM, Ferraz MB, Santos WA, Meina˜o I, Quaresma MR. Traduc¸a˜o para a lı´ngua portuguesa e validac¸a˜o do questiona´ rio gene´ rico de avaliac¸a˜o de qualidade 37. Nseir W, Farah R, Mograbi J, Makhoul N. Impact of serum C-reactive protein de vida SF-36 (Brasil SF-36). Rev Bras Reumatol. 1999;39:43–50. measurements in the first 2 days on the 30-day mortality in hospitalized patients with severe community-acquired pneumonia: a cohort study. J Crit Care. 27. Kovalis D, Segretti NO, Probst VS, Lareau SC, Brunetto AF, Pitta F. Validac¸a˜o 2013;28:291–295. do Modified Pulmonary Functional Status and Dyspnea Questionnaire e da escala do Medical Research Council para o uso em pacientes com 38. Lee JH, Kim J, Kim K, Jo YH, Rhee J, Kim TY. Albumin and C-reactive protein have doenc¸a pulmonar obstrutiva croˆ nica no Brasil. J Bras Pneumol. 2008;34: prognostic significance in patients with community-acquired pneumonia. J Crit 1008–1018. Care. 2011;26:287–294.
Journal of Physiotherapy 62 (2016) 118–119 Journal of PHYSIOTHERAPY journal homepage: www.elsevier.com/locate/jphys Appraisal Correspondence Research requires deep knowledge of the modality to be tested As associates of Dr Kenzo Kase, Founder and Chairman of the swelling at the ankle may not have been properly measured.5 We Kinesio Taping Association International, we feel called upon to do not know whether these people continued to participate in the respond to published research and correspondence with regard to sport that caused the injury in the first place. ‘Kinesio Taping does not decrease swelling in acute, lateral ankle sprain of athletes: a randomised trial’]F[_DIT$5 by Nunes and colleagues.1F]DI[4$T_ Tape measurement can have huge variability depending on who is measuring, and consistent tension is extremely difficult. We are Nunes claims that the taping application used was drawn from not talking about litres of fluid in these patients, but something Dr Kase’s book,2 but they only ‘partially’ followed the book’s more like tablespoons of fluid that are sufficient to cause pain and instructions. They tested two pieces of tape; they did not test the change biomechanics of a joint. As for placement, the appropriate Kinesio Taping Method. As Kinesio Taping educators we found both question to ask is: do you want direct or indirect draw of fluid? the study itself and the follow-up communication puzzling. For many years, any Kinesio Taping research at all was valued It is evident to anyone who is trained and experienced in the for providing guidelines for further study. We have now gone past assessment and application protocols for lymphatic Kinesio Taping that point, and we expect to see studies performed in a responsible that the process used was simply wrong. The team does not appear and informed manner. This study does not meet the lowest to include anyone certified or extensively trained in the Kinesio standard of responsible scientific researchD$[ITF1_]. Taping Method. Kim Rock Stockheimera,*, Gu¨ l BaltacıbDF_],$T2I[ Graceann G. Forresterc, Both the contents of the initial article1 and the subsequent Stefano Frassined and Andrea Wolkenbergc published correspondence3,4 with Dr Lee indicate significant weaknesses. We share Lee’s skepticism as to whether these aKinesio Taping Association International, Wisconsin, USA researchers possess a solid understanding of the principles behind bKinesio Taping Association International, Ankara, Turkey lymphatic Kinesio Taping. Their reply to him reinforces these doubts. cKinesio Taping Association International, New York, USA The study is replete with troubling details. The subjects dKinesio Taping Association International, Legnano, Italy were varied in activity and condition. The applications were not blinded. Twenty minutes of icing and exercise with the Kinesio *Corresponding author Taping treatment could have affected compliance and caused E-mail address: [email protected] (K.R. Stockheimer). additional trauma. Many subjects did not complete the study. References The technique used1 suggests a formation of crisscross patterns around the lateral malleolus for lateral ankle sprain and around the 1. Nunes GS, et al. J Physiother. 2015;61:28–33. medial malleolus for medial ankle sprain. However, with Kinesio 2. Kase K, et al. Clinical Therapeutic Applications of the Kinesio Taping1 Method. Tokyo, Taping the pressure decreases due to lifting of the superficial skin, and blood and lymphatic circulation increase due to an increase in Japan: Kenı´-Kai information; 2003. subcutaneous space. 3. Lee J-H. J Physiother. 2015;61:231. 4. Nunes GS, et al. J Physiother. 2015;61:231–232. The ‘control’ taping was not inert but had 20% tension, and was 5. Kase K, et al. Kinesio Taping for Lymphoedema and Chronic Swelling. Tokyo, Japan: applied on the anterior pathway of the lymphatics. As a control they had no no-treatment group. Multiple different investigators Kenı´-Kai Ltd; 2006. may have applied tape with no specifics on their training in Kinesio Taping. Voltmeter for the lower extremity is from foot to knee, so http://dx.doi.org/10.1016/j.jphys.2016.02.003 Knowledge of the modality comes from rigorous research $DIF]WT_3[ e are grateful for the opportunity, once again, to discuss Stockheimer and colleagues’ letter states, in accordance with science and further explain our trial of Kinesio Taping for ankle Dr Kenzo Kase’s book,3 that ‘with Kinesio Taping the pressure sprain.1_2[TD$F]I We understand the disappointment shown by Stock- decreases due to lifting of the superficial skin, and blood and heimer and colleagues that the results did not favour Kinesio lymphatic circulation increase due to an increase in subcutaneous Taping; however, we feel that it is important to show our study to space’. We really would like to have a reference to a published study demonstrating that this in fact occurs; however, we are yet the physiotherapy community and let them judge whether the to find such a study. Kinesio Taping method should be used or not. Stockheimer and colleagues seem to be ‘troubled’ by the fact As correctly identified by Stockheimer and colleagues, and that our sample was drawn from athletes from varied sports as mentioned in our previous correspondence,2 we partially modalities, giving the impression that they expect Kinesio Taping to be effective for sprained ankles that occur in some sports but not followed the figure from the book Clinical Therapeutic Applications in others. We would need further clarification before any further of the Kinesio Taping1 Method to be consistent with the descrip- comments could be made about this. tion on how the lymphatic correction (channelling) should be applied.3 1836-9553/ß 2016 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Appraisal Correspondence 119 Stockheimer and colleagues criticise the study due to the lack of Lastly, we disagree that the study was not ‘responsible scientific blinding of therapists or participants. While it is true that blinding research’. It was prospectively registered and ethically approved, therapists and participants would avoid some sources of bias, the intervention was guided by the developer’s textbook,3 and it those biases tend to spuriously generate or inflate positive effects meets all the criteria on the PEDro Scale except participant and of the treatment,4 but no positive effects were seen in our study. therapist blinding. We would be interested to see research of a similar standard that demonstrates the efficacy of Kinesio Taping, Stockheimer and colleagues also criticise the use of ice and as such studies appear to be scarce.7_$[D]TFI4 We have no conflict of interest exercise in both groups; however, these treatments are common in this study. physiotherapy practice and are often applied concomitantly with Kinesio Taping. It therefore made sense for us to test Kinesio Guilherme S Nunesa,*, Marcos de Noronhaa,]FID$T_1[ b, Taping under conditions commonly seen in practice. Valentine Zimermann Vargasa,c, Bruna Wagecka, Loss to follow-up was also criticised, but it was minimal (5%) Daniela Pacheco dos Santos HaupenthalaDF]T[5I_$ at the end of the intervention period, when no intervention effect and Clarissa Medeiros da Luza,c was identified. Although loss to follow-up increased by Day 15, when again no effect was identified, it is counterintuitive to aDepartment of Physiotherapy, Center of Health and Sport Sciences, imagine that a treatment effect suddenly appeared then, given that Santa Catarina State University, Floriano´polis, Brazil no effect was present at the end of the intervention. bDepartment of Allied Health, La Trobe University, Bendigo, Australia Stockheimer and colleagues criticise the method used to cDepartment of Physiology, Sa˜o Paulo Federal University, measure swelling, but the method is reliable,5,6 so we are Sa˜o Paulo, Brazil confident that our measurement methods were not an issue in this study. Furthermore, the physiotherapists applying the *Corresponding author technique were experienced in sports physiotherapy and in Kinesio Taping. References A major concern with the letter by Stockheimer and colleagues 1. Nunes GS, et al. J Physiother. 2015;61:28–33. is the conflict between points that they raise. At first, they state 2. Nunes GS, et al. J Physiother. 2015;51:231–232. that the Kinesio Taping technique was ‘simply wrong’ (and 3. Kase K, et al. Clinical Therapeutic Applications of the Kinesio Taping1 Method. 2nd ed. therefore ineffective). Later, they criticise the taping used in the control group, which was not an appropriate Kinesio Taping Tokyo: KenIkai; 2003. technique (tape with 20% tension in the anterior pathway of the 4. Herbert R, et al. Practical Evidence-Based Physiotherapy. UK: Elsevier Health Sciences; lymphatics, according to them) as potentially causing benefit. It is unclear to us how Stockheimer and colleagues can claim that the 2011:611. correct technique is essential and also claim that any application 5. Petersen EJ, et al. J Orthop Sports Phys Ther. 1999;29:609–615. can be beneficial. 6. Brijker F, et al. Clin Physiol. 2000;20:56–61. 7. Parreira PdCS. et al. J Physiother. 2014;60:31–39. http://dx.doi.org/10.1016/j.jphys.2016.02.001
Journal of Physiotherapy 62 (2016) 83–87 Journal of PHYSIOTHERAPY journal homepage: www.elsevier.com/locate/jphys Research Massage therapy decreases pain and perceived fatigue after long-distance Ironman triathlon: a randomised trial Guilherme S Nunes a, Paula Urio Bender a, Fa´ bio Sprada de Menezes b, Igor Yamashitafuji a, Valentine Zimermann Vargas a,c, Bruna Wageck a a Department of Physiotherapy, Center of Health and Sport Sciences, Santa Catarina State University, Floriano´polis; b Department of Physiotherapy, Esta´cio University Center of Santa Catarina, Sa˜o Jose´, Brazil; c Department of Physiology, Sa˜o Paulo Federal University, Sa˜o Paulo, Brazil KEY WORDS ABSTRACT Musculoskeletal manipulations Question: Can massage therapy reduce pain and perceived fatigue in the quadriceps of athletes after a Musculoskeletal pain long-distance triathlon race (Ironman)? Design: Randomised, controlled trial with concealed allocation, Athletes intention-to-treat analysis and blinded outcome assessors. Participants: Seventy-four triathlon athletes Quadriceps muscle who completed an entire Ironman triathlon race and whose main complaint was pain in the anterior Fatigue portion of the thigh. Intervention: The experimental group received massage to the quadriceps, which was aimed at recovery after competition, and the control group rested in sitting. Outcome measures: The outcomes were pain and perceived fatigue, which were reported using a visual analogue scale, and pressure pain threshold at three points over the quadriceps muscle, which was assessed using digital pressure algometry. Results: The experimental group had significantly lower scores than the control group on the visual analogue scale for pain (MD –7 mm, 95% CI –13 to –1) and for perceived fatigue (MD – 15 mm, 95% CI –21 to –9). There were no significant between-group differences for the pressure pain threshold at any of the assessment points. Conclusion: Massage therapy was more effective than no intervention on the post-race recovery from pain and perceived fatigue in long-distance triathlon athletes. Trial registration: Brazilian Registry of Clinical Trials, RBR-4n2sxr. [Nunes GS, Bender PU, de Menezes FS, Yamashitafuji I, Vargas VZ, Wageck B (2016) Massage therapy decreases pain and perceived fatigue after long-distance Ironman triathlon: a randomised trial. Journal of Physiotherapy 62: 83–87] ß 2016 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Introduction mechanical manipulation of human body tissue by means of manual compressions and rhythmic percussions.13,14 Different Ironman is one of the main long-distance triathlon races in the world. About 40 Ironman events take place every year across the massage therapy techniques are expected to increase blood and globe, with around 2000 athletes participating in each event.1 In lymphatic flow;14 theoretically, this might accelerate the elimina- this competition, athletes have to cover a distance of 226 km, which consists of swimming 3.8 km, cycling 180 km and running tion of catabolites, which possibly reduces the sensation of 42.2 km.2 It is a strenuous competition that requires high energy fatigue.14 Another expected effect of massage therapy is pain expenditure and generates high physical and metabolic stress.3,4 relief.14 The mechanical stimulus caused by manual contact on the Because of this stress, Ironman athletes often experience various medical conditions such as dehydration, heat stroke, hyponatrae- skin may have a neurological effect, blocking the noxious stimuli mia, musculoskeletal injuries, hypothermia, skin injuries, fever, based on the gate-control theory.14 Another possibility is a hypoglycaemia, diarrhoea and vomiting.4–7 However, the most common symptoms after a race are pain and muscle fatigue, which physiologic effect via the release of b-endorphins.14 are probably caused by muscle damage that induces an inflamma- tory response and a reduction in energy reserves.4,8–10 The body Several clinical trials have demonstrated the beneficial effects parts most affected by pain and fatigue due to constant overload of massage therapy in athletes after strenuous exercises.16,17 Ogai are the lower limbs, mainly around the knees.11,12 et al17 evaluated physically active university students who Therapeutic interventions are used to try to minimise the performed the same protocol through strenuous workout on a severity of symptoms in triathletes after strenuous competition, such as massage, cryotherapy and stretching.13–15 Massage stationary bike in two sessions. In one of the sessions, they received therapy is often used after competitions and is defined as a massage to the lower limbs for 10 minutes in the middle of the protocol. After the massage protocol session, the participants showed a decrease in muscle stiffness and perceived fatigue, which was evaluated by the visual analogue scale (VAS).17 Mancinelli et al16 also verified the effects of massage in athletes. In this study, high school basketball and volleyball players were divided into two groups after the first training session of the season: one group http://dx.doi.org/10.1016/j.jphys.2016.02.009 1836-9553/ß 2016 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
84 Nunes et al: Massage after long-distance Ironman triathlon rested while the other received massage to the quadriceps muscle the athletes were assessed to determine if they met the eligibility on the peak day of delayed onset muscle soreness.16 The group criteria, and data about their baseline characteristics were that received massage showed a decrease in perceived pain and collected. The measurements were taken from the most painful pressure pain threshold and improved vertical jump performance.16 quadriceps, as reported using the VAS. If the participant reported the same level of pain in both thighs, the side to be treated was Some systematic reviews have been performed to investigate randomly selected by flipping a coin. After the first evaluation, the effects of massage therapy.13,18 These reviews have concluded the participants were allocated to a group and directed to the that massage therapy can have benefits in a variety of musculo- intervention site. After the interventions, the same blind research- skeletal conditions,18 including soreness after strenuous exer- er reassessed the participants and, if needed, they were directed cise.13 However, among the included studies, there has been back to the clinic to continue the treatment. To maintain the some lack of standardisation of the techniques applied during the assessor blinding, the evaluation was conducted in a different massage therapy. Furthermore, the intense physical activities used place to where the massage intervention was conducted, the in these studies were generally high-intensity exercise regimens assessor was not informed of who was in which group and the designed to bring on rapid fatigue. The studies were also not foam used to reduce friction during the massage was also applied conducted in real competition situations, and the massage on the quadriceps of the control participants. intervention was not always administered immediately after the exercise. Therefore, questions remain regarding the effectiveness Intervention of massage after very prolonged, strenuous sporting competitions such as long-distance triathlon.13 The experimental group received massage for 7 minutes from a therapist who was not involved in the measurements. The Therefore, the research question for this randomised, controlled intervention consisted of the following procedures: 1 minute of trial was: superficial effleurage, in which the therapist slid both hands in the direction of the muscle fibres from distal to proximal with a gentle Does massage reduce pain and fatigue in the quadriceps of pressure on the thigh; 2 minutes of deep effleurage, in which athletes after a long-distance triathlon race (Ironman)? the therapist performed the same movement but applied more pressure to the thigh; 2 minutes of petrissage, in which the Method therapist used the entire surface of the palm of the hands to compress and lift the tissue sequentially; 1 minute of tapotement, Design in which the therapist agitated the tissues of the thigh with cupped hands; and 1 minute of superficial effleurage to finish the This was a randomised clinical trial in which the participants intervention. A video demonstration of the techniques is presented were allocated to one of two groups: an experimental group or a in Appendix 1 (see eAddenda for Appendix 1). control group. The experimental group received massage to the quadriceps, which was aimed at recovery after competition, and the control group rested in a sitting position. The randomisation was conducted using sealed and opaque envelopes to conceal each upcoming allocation during recruitment. A researcher who was not involved in the evaluations or interventions of this study prepared these envelopes. The study design is presented in Figure 1. Participants, therapists and centre Seventy-four triathlon athletes from Ironman Brazil took part in this study. To be eligible for inclusion in the study, athletes had to complete the entire Ironman triathlon race and report to the physiotherapy clinic with their main complaint being pain in the anterior portion of the thigh. The exclusion criteria were any medical conditions that were not compatible with the procedures of the study, such as: severe metabolic and respiratory disorders during the study procedures, cramps during the evaluations and/or interventions, presence of abrasions on the thigh, or any change in sensation of the thigh caused by analgesics or cryotherapy. The evaluations and interventions of this study were conducted in the physiotherapy clinic of the Ironman Brazil triathlon competition. This clinic was located near the finish line. First, [(Fi]gurGe_1I)TD$ Assessed for eligibility (n = 74) Measured pain, perceived fatigue and pressure pain threshold Pre Randomised (n = 74) (n = 37) (n = 37) Lost to follow-up Exp group Con group Lost to follow-up (n = 1) • quadriceps • rest in (n = 0) • severe nausea sitting massage • 7 minutes • 7 minutes Post Measured pain, perceived fatigue and pressure pain threshold (n = 36) (n = 37) Figure 1. Design and flow of participants through the study. Massage therapy techniques. Exp = experimental group, Con = control group.
Research 85 Foam made from soap and water was used to reduce friction measure in each point and mean among points were used for between the therapist’s hands and the participant’s skin. analysis. Pressure was measured in kgf/cm2. The control group rested in a sitting position for 7 minutes. This Data analysis control was adopted because minimum manual contact from a sham technique could have similar effects to those of the The sample size of this study was calculated based on statistical experimental intervention. The participants allocated to the power of 80% and an alpha of 5%. The sample size calculation control group were not informed that they would not receive indicated that 36 participants in each group would be required to massage; they were told to expect an available therapist to identify: a between-group difference in pain of 20 mm measured perform the massage. At the end of this waiting period, the same by the VAS, with a SD of 30 mm;16 a between-group difference of foam used in the massage intervention was applied to the 10 mm in the perceived fatigue measurement by the VAS, with a participants’ leg in order to keep the evaluator blinded to group SD of 15 mm;17 and a between-group difference of 1.0 kgf/cm2 in allocation. After the final measures, participants were led to the the pressure pain threshold measurement, with a SD of 1.5 kgf/cm2 physiotherapy clinic to complete the treatment. based on a pilot study. Outcome measures To compare the effect of massage between the experimental and control groups, the independent t-test was used to identify Pain and perceived fatigue differences. The difference from baseline was taken into account in The pain and perceived local fatigue in the quadriceps muscle the analysis. A level of significance of p 0.05 was adopted for all tests and the data were analysed with an intention-to-treat were measured using the VAS. The scale consisted of a 100-mm approach. The ‘last observation carried forward’ approach was line on which 0 mm indicated no pain or fatigue and 100 mm used for missing data. The effect size (Cohen’s d) was calculated and classified as small (d = 0.2), moderate (d = 0.5), and large indicated the worst possible pain or extreme exhaustion. The (d 0.8).23 The analyses and sample size calculation were participants were asked to score the intensity of pain and fatigue at performed using commercially available softwareb,c. that time on the horizontal 100-mm line.17,19 The levels of pain and fatigue were measured separately in mm. The VAS has high Results reliability for acute pain measures, with an intraclass correlation Flow of participants through the study coefficient (ICC) of 0.97,20 and is also validated to measure fatigue in healthy subjects.21 One participant did not finish the procedures due to nausea during the initial evaluation, necessitating medical care. Thus, Pressure pain threshold 36 participants in the experimental group and 37 participants in To measure the pressure pain threshold, a digital pressure the control group completed the procedures. The characteristics of participants are shown in Table 1 and in the first two columns of algometera, which was connected to a computer for data collection, data in Table 2. Individual participant data are presented in Table 3 was used. Algometry has been shown to be highly reliable (see eAddenda for Table 3). The only complication reported by (ICC = 0.91).22 The participant sat with the knee flexed at 90 deg some participants during the competition was the occurrence of and feet flat on the floor. Three points were measured at the cramps. quadriceps: the muscle bellies of the rectus femoris, vastus lateralis and vastus medialis. The algometer inductor was Compliance with the study protocol positioned perpendicularly to the skin at each point and gradual pressure was applied until the participant reported a change from No ineligible participants were randomised. No assessors were the sensation of pressure to pain. At that point, the pressure was unblinded during the study. No participants received the wrong stopped immediately and the highest pressure that was recorded intervention. All participants were analysed in the group to which represented the pain threshold.16,22 Before the beginning of the they had been randomly allocated. evaluation, the subject was familiarised with the measurement by a demonstration of the pressure at each point to be measured. The Effect of the massage intervention Table 1 Exp (n = 37) Con (n = 37) The independent t-test showed statistically significant be- Characteristics of participants. tween-group differences after intervention for pain and perceived 37 (7) 39 (11) fatigue, which were evaluated using the VAS (Table 2). The Characteristic 34 (92) 36 (97) experimental group had lower scores on the VAS for pain, by a 1.77 (0.09) 1.77 (0.08) mean of 7 mm (95% CI 1 to 13). The experimental group also had Age (yr), mean (SD) 75.5 (10.6) 77.2 (10.8) Gender, n males (%) 11:31 (2:17) 12:25 (2:16) Height (m), mean (SD) Weight (kg), mean (SD) 13 (35) 11 (30) Time to complete race (hh:mm), mean (SD) Complications during race, n participants (%) cramps Exp = experimental group, Con = control group. Table 2 Mean (SD) of groups, mean (SD) difference within groups, and mean (95% CI) difference between groups. Outcome Groups Difference within groups Difference between Effect size groups (Cohen’s d) Pre Post Post minus Pre Post minus Pre 0.6 Exp (n = 37) Con (n = 37) Exp (n = 36) Con (n = 37) Exp Con 1.5 Exp minus Con Pain (mm) 58 (24) 62 (23) 48 (24) 59 (23) –10 (12) –3 (13) 0.1 Perceived fatigue (mm) 84 (18) 79 (23) 69 (24) 79 (19) –15 (14) 0 (1) –7 (–13 to –1) 0.3 PPT (kgf/cm2) –15 (–21 to –9) 0.2 6.0 (2.4) 6.1 (2.7) 5.6 (2.2) 5.5 (3.0) –0.4 (1.7) –0.6 (1.3) 0.0 rectus femoris 5.4 (1.7) 5.4 (2.1) 5.2 (1.8) 5.1 (2.2) –0.1 (1.2) –0.4 (1.1) 0.1 (–0.6 to 0.8) vastus lateralis 4.4 (1.4) 4.5 (1.4) 4.2 (1.5) 4.4 (1.7) –0.2 (1.0) 0.0 (1.0) 0.2 (–0.3 to 0.7) vastus medialis 5.3 (1.7) 5.3 (2.0) 5.0 (1.7) 5.0 (2.2) –0.3 (1.0) –0.3 (0.8) –0.2 (–0.6 to 0.3) quadriceps femoris 0.1 (–0.4 to 0.5) Exp = experimental group, Con = control group, PPT = pressure pain threshold.
86 Nunes et al: Massage after long-distance Ironman triathlon lower scores on the VAS for perceived fatigue, by a mean of 15 mm As discussed above, systematic reviews have been performed to (95% CI 9 to 21). analyse the results of studies about the effects of massage therapy. Bervoets et al included studies that assessed the effects of massage There were no significant between-group differences for the therapy on participants with musculoskeletal disorders, and pressure pain threshold in any of the measured points. When the concluded that massage therapy as a stand-alone intervention average of the three measured points was compared between reduces pain and improves function.18 Another systematic review groups, the difference was again non-significant (Table 2). by Best et al included only studies about sports massage therapy; it concluded that massage therapy could assist in muscle recovery The effect size was large for perceived fatigue, moderate for and reduce delayed onset muscle soreness.13 Thus, the results of pain, and small for pain evaluated by algometer at all measured the present study are in agreement with the conclusions of those points (Table 2). systematic reviews, especially with respect to the effect on pain, despite the large variety of massage techniques applied in the Discussion analysed studies. This study aimed to determine whether massage therapy could Despite the positive effects of massage on pain and fatigue in optimise the recovery of athletes after competing in a long- the present study, these findings should be analysed with distance triathlon such as Ironman. Both pain and fatigue are very caution. The positive effects were found only in subjective common conditions after these competitions.10 The onset of measures. Another consideration is that the time between muscle pain and discomfort may result from the activity overload. evaluation and re-evaluation was very short, and this could also This may lead to structural damage, which leads to an inflamma- have had some influence on the results. Another issue was the tory response in the muscle tissue,4,10,24 and this inflammatory lack of a sham technique as control. Although this did not control response stimulates the nerve endings (deep nociceptors), which for the effects of contact with the therapist, this lack of control results in painful stimuli.25 The perception of fatigue may be was judged to be a less important consideration than the related to the high energy demands of this kind of sport. The potential for the minimal manual contact from a sham technique shortage of energy substrate, mainly muscle glycogen, may lead to to have led to similar effects as massage. Despite this issue, the metabolic acidosis (resulting from organic reactions to energy results still confirm the effect of the massage on well-being and production)25,26 and increased production of free radicals that are relief of symptoms in the short term, even if it does not have a generated by the oxidative damage in the body cells.3 Thus, the significant effect on structural and physiological aspects of post- lack of nutrients for energy generation and the accumulation of exercise soreness. catabolites may cause a decrease in capacity for strength and power generation in the energy transfer system and a disturbance Based on the results of the present study, it can be concluded in the transmission of nerve impulses.25,27 that the massage therapy techniques used were more effective in the recovery from pain and perceived fatigue than no intervention Massage therapy as a technique for triathlete recovery after a after a long-distance triathlon race. Further investigations are long-distance race resulted in decreased pain and perceived needed to verify the acute effects of massage therapy on metabolic fatigue when compared with triathletes who remained at rest in stress and biochemical markers of structural damage in athletes this study. When calculated in terms of a Cohen’s d, the effect size after strenuous competitions and to identify the effects of massage could be categorised as large for perceived fatigue and moderate therapy applied at the end of competitions on the recovery of long- for pain. When considered in mm on the 0-to-100 mm VAS, the term athletes. 7-mm effect on pain and the 15-mm effect on perceived fatigue would not be considered large or perhaps even moderate. In summary, there is now direct evidence to support the use of However, it should be noted that the massage was only applied massage for the recovery of triathletes after long-distance for 7 minutes. This raises two important issues. First, athletes may endurance races. Research on massage therapy applied at the consider these relatively small improvements worthwhile given end of such competitions is important because this technique is the small amount of time invested in the intervention. Second, widely used in this context and there is the potential to identify although 7 minutes of massage was sufficient to demonstrate that effects of greater magnitude (eg, by increasing the dose of massage massage improves pain and perceived fatigue in these long- delivered). distance triathletes, massage of greater duration may be able to increase the magnitude of these effects. What is already known on this topic: Pain and fatigue are common after intense exercise. Massage is beneficial in this Massage therapy was applied in this study in order to mitigate situation, but the evidence for its benefits comes from studies the main structural and physiological damage after long-distance of controlled periods of intense exercise designed to cause triathlon events. Some studies suggest that the manual contact can rapid onset of fatigue. cause physiological, neurological or psychological responses for the What this study adds: Among people who had just competed control of pain sensation.14 In the present study, the experimental in a 226-km Ironman event, a short period of massage over the group showed decreased perception of pain, but the pressure pain quadriceps resulted in local reductions in both pain and threshold remained unchanged. This suggests that massage results perceived fatigue. The amount of pressure over the quadriceps cannot be explained by neurological responses based on the gate required to elicit pain was not significantly reduced by the control theory. Possible explanations may be related to physiological massage. responses through the b-endorphin release or catabolite elimina- Footnotes: aEMG System do Brasil, model EMG230C, Sa˜o Paulo, Brazil. bSPSS Inc. Version 17.0, Chicago, IL, USA. cG*Power Version tion or to psychological pathways where attention and manual 3.1.9.2, Du¨ sseldorf, Germany. contact can lead to a sense of well-being.14 eAddenda: Table 3 and Appendix 1 (video) can be found online The physiological responses obtained through massage are at doi:10.1016/j.jphys.2016.02.009. enhanced by the improvement observed in the experimental group regarding perceived fatigue. In physiological terms, fatigue relates Ethics approval: The Human Research Ethics Committee of to the decrease in normal function.28 It has also been suggested Universidade do Estado de Santa Catarina approved this study that fatigue can have both central and peripheral sources, with lack (approval number CAAE: 15225113500000118). All participants of energy substrate and metabolic waste concentration being the gave written informed consent before data collection began. most probable causes.14,26,28 Thus, it can be assumed that the positive effects of massage therapy on perceived fatigue are related Competing interests: Nil. to greater localised blood circulation either by mechanical effect Source(s) of support: None. or temperature increase,14,29,30 which aid in the removal of metabolic waste.14,30
Research 87 Acknowledgements: The authors would like to acknowledge 14. Weerapong P, Hume PA, Kolt GS. The mechanisms of massage and effects all professionals and students that worked in the physiotherapy on performance, muscle recovery and injury prevention. Sports Med. 2005;35: sector of Ironman Brazil. 235–256. Provenance: Not invited. Peer reviewed. 15. Silveira MA, Haupenthal A, Mannrich G, Torres SF. Atuac¸a˜o Fisioterapeˆutica No Correspondence: Guilherme S Nunes, Department of Physio- Ironman Brasil. Fisioterapia em Movimento. 2006;19:35–40. therapy, Center of Health and Sport Sciences, Santa Catarina State University, Floriano´ polis, Brazil. Email: [email protected] 16. Mancinelli CA, Davis DS, Aboulhosn L, Brady M, Eisenhofer J, Foutty S. The effects of massage on delayed onset muscle soreness and physical performance in female References collegiate athletes. Phys Ther Sport. 2006;7:5–13. 1. Ironman triathlon 140.6 & 70.3. Ironman Official site. http://www.ironman.com 17. Ogai R, Yamane M, Matsumoto T, Kosaka M. Effects of petrissage massage on [accessed 9/12/2014]. fatigue and exercise performance following intensive cycle pedalling. Br J Sports Med. 2008;42:834–838. 2. Babbitt B, Lampley J. 30 Years of the Ironman Triathlon World Championship. Aachen; Hove: Meyer & Meyer Fachverlag und Buchhandel GmbH; 2008. 18. Bervoets DC, Luijsterburg PAJ, Alessie JJN, Buijs MJ, Verhagen AP. Massage therapy has short-term benefits for people with common musculoskeletal disorders com- 3. Knez WL, Jenkins DG, Coombes JS. Oxidative stress in half and full Ironman pared to no treatment: a systematic review. J Physiother. 2015;61:106–116. triathletes. Med Sci Sports Exerc. 2007;39:283–288. 19. Sellwood KL, Brukner P, Williams D, Nicol A, Hinman R. Ice-water immersion and 4. O’Toole ML, Douglas PS. Applied physiology of triathlon. Sports Med. 1995;19: delayed-onset muscle soreness: a randomised controlled trial. Br J Sports Med. 251–267. 2007;41:392–397. 5. Hew TD, Chorley JN, Cianca JC, Divine JG. The incidence, risk factors, and clinical 20. Bijur PE, Silver W, Gallagher EJ. Reliability of the visual analog scale for measure- manifestations of hyponatremia in marathon runners. Clin J Sport Med. 2003;13: ment of acute pain. Acad Emerg Med. 2001;8:1153–1157. 41–47. 21. Lee KA, Hicks G, Nino-Murcia G. Validity and reliability of a scale to assess fatigue. 6. Holtzhausen LM, Noakes TD. Collapsed ultraendurance athlete: proposed mecha- Psychiatry Res. 1991;36:291–298. nisms and an approach to management. Clin J Sport Med. 1997;7:292–301. 22. Chesterton LS, Sim J, Wright CC, Foster NE. Interrater reliability of algometry in 7. Roberts WO. A 12-yr profile of medical injury and illness for the Twin Cities measuring pressure pain thresholds in healthy humans, using multiple raters. Clin J Marathon. Med Sci Sports Exerc. 2000;32:1549–1555. Pain. 2007;23:760–766. 8. Garcı´a-Manso JM, Rodrı´guez-Ruiz D, Rodrı´guez-Matoso D, de Saa Y, Sarmiento S, 23. Cohen J. A power primer. Psychol Bull. 1992;112:155–159. Quiroga M. Assessment of muscle fatigue after an ultra-endurance triathlon using 24. Peake JM, Nosaka K, Muthalib M, Suzuki K. Systemic inflammatory responses to tensiomyography (TMG). J Sports Sci. 2011;29:619–625. maximal versus submaximal lengthening contractions of the elbow flexors. Exerc 9. Geva N, Defrin R. Enhanced pain modulation among triathletes: a possible expla- Immunol Rev. 2006;12:72–85. nation for their exceptional capabilities. Pain. 2013;154:2317–2323. 25. Wilmore JH, Costill DL, Kenney WL. Physiology of Sport and Exercise. 4th ed. Champaign, IL: Human Kinetics; 2008. 10. Neubauer O, Ko¨ nig D, Wagner K-H. Recovery after an Ironman triathlon: sustained 26. Westerblad H, Allen DG, La¨ nnergren J. Muscle fatigue: lactic acid or inorganic inflammatory responses and muscular stress. Eur J Appl Physiol. 2008;104:417–426. phosphate the major cause? News Physiol Sci. 2002;17:17–21. 27. Enoka RM, Duchateau J. Muscle fatigue: what, why and how it influences muscle 11. Korkia PK, Tunstall-Pedoe DS, Maffulli N. An epidemiological investigation of training function. J Physiol. 2008;586:11–23. and injury patterns in British triathletes. Br J Sports Med. 1994;28:191–196. 28. Marino FE, Gard M, Drinkwater EJ. The limits to exercise performance and the future of fatigue research. Br J Sports Med. 2011;45:65–67. 12. Gosling CM, Gabbe BJ, Forbes AB. Triathlon related musculoskeletal injuries: The 29. Portillo-Soto A, Eberman LE, Demchak TJ, Peebles C. Comparison of blood flow status of injury prevention knowledge. J Sci Med Sport. 2008;11:396–406. changes with soft tissue mobilization and massage therapy. J Altern Complement Med. 2014;20:932–936. 13. Best TM, Hunter R, Wilcox A, Haq F. Effectiveness of sports massage for recovery of 30. Wiltshire EV, Poitras V, Pak M, Hong T, Rayner J, Tschakovsky ME. Massage impairs skeletal muscle from strenuous exercise. Clin J Sport Med. 2008;18:446–460. postexercise muscle blood flow and ‘‘lactic acid’’ removal. Med Sci Sports Exerc. 2010;42:1062–1071.
Journal of Physiotherapy 62 (2016) 68–82 Journal of PHYSIOTHERAPY journal homepage: www.elsevier.com/locate/jphys Research Moderate-intensity exercise reduces fatigue and improves mobility in cancer survivors: a systematic review and meta-regression Amy M Dennett a, Casey L Peiris b, Nora Shields b, Luke A Prendergast c, Nicholas F Taylor a a School of Allied Health, La Trobe University and Allied Health Clinical Research Office, Eastern Health; b School of Allied Health, La Trobe University and Northern Health; c Department of Mathematics and Statistics, La Trobe University, Melbourne, Australia ]FKD$IT_1[ E Y W O R D S ABSTRACT Exercise therapy Question: Is there a dose-response effect of exercise on inflammation, fatigue and activity in cancer Cancer survivors? Design: Systematic review with meta-regression analysis of randomised trials. Participants: Fatigue Adults diagnosed with cancer, regardless of specific diagnosis or treatment. Intervention: Exercise Mobility interventions including aerobic and/or resistance as a key component. Outcome measures: The primary Physical therapy (specialty) outcome measures were markers of inflammation (including C-reactive protein and interleukins) and various measures of fatigue. The secondary outcomes were: measures of activity, as defined by the World Health Organization’s International Classification of Functioning, Disability and Health, including activities of daily living and measures of functional mobility (eg, 6-minute walk test, timed sit-to-stand and stair-climb tests). Risk of bias was evaluated using the PEDro scale, and overall quality of evidence was assessed using the Grades of Research, Assessment, Development and Evaluation (GRADE) approach. Results: Forty-two trials involving 3816 participants were included. There was very low- quality to moderate-quality evidence that exercise results in significant reductions in fatigue (SMD 0.32, 95% CI 0.13 to 0.52) and increased walking endurance (SMD 0.77, 95% CI 0.26 to 1.28). A significant negative association was found between aerobic exercise intensity and fatigue reduction. A peak effect was found for moderate-intensity aerobic exercise for improving walking endurance. No dose-response relationship was found between exercise and markers of inflammation or exercise duration and outcomes. Rates of adherence were typically high and few adverse events were reported. Conclusions: Exercise is safe, reduces fatigue and increases endurance in cancer survivors. The results support the recommendation of prescribing moderate-intensity aerobic exercise to reduce fatigue and improve activity in people with cancer. Review registration: PROSPERO CRD42015019164. [Dennett AM, Peiris CL, Shields N, Prendergast LA, Taylor NF (2016) Moderate-intensity exercise reduces fatigue and improves mobility in cancer survivors: a systematic review and meta-regression. Journal of Physiotherapy 62: 68–82] Crown Copyright ß 2016 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by- nc-nd/4.0/). Introduction mode and intensity of exercise for people with cancer is unclear, and exercise guidelines are based largely on expert clinical opinion Cancer is a leading cause of burden of disease globally1 and is responsible for approximately three in 10 deaths.2 However, with and adaptations of guidelines for healthy people. Current improved screening and advancing treatment options, survival rates are improving. As a result, cancer is now recognised as a recommendations suggest that cancer survivors complete at least chronic disease.3,4 While treatment may improve survival, the 150 minutes of moderate-intensity physical activity per week.6 side-effects on physical and psychological function often reduce quality of life. There is an increasing need for rehabilitation to However, these recommendations may not recognise the specific address these issues. health needs of cancer survivors. Recent reviews have reported a low number of adverse events in relation to exercise trials,6,7,18–20 Exercise is an effective treatment for many chronic diseases. suggesting that exercise is generally safe for cancer survivors. Recent systematic reviews have demonstrated that exercise used as part of cancer care reduces cancer-related fatigue and improves However, in these reviews, there has been variable reporting of the cardiovascular function, strength and quality of life.5–9 There is dose of prescribed exercise. also emerging evidence that exercise can reduce recurrence and mortality in some cancer populations.10–16 The association between inflammation and cancer is well documented.21–23 Chronic inflammation plays a role in the Despite these benefits of exercise, there is a lack of evidence on the safety and efficacy of exercise in relation to dose.6,17 The ideal pathogenesis of insulin resistance and tumour growth, and has been linked to cancer risk and mortality.23–26 Inflammatory cytokines have also been implicated in the development of cancer-related fatigue.27–29 Exercise plays a role in mediating the effects of chronic inflammation, reducing inflammatory http://dx.doi.org/10.1016/j.jphys.2016.02.012 1836-9553/Crown Copyright ß 2016 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Research 69 markers such as C-reactive protein (CRP), tumour necrosis factor- Box 1. Inclusion criteria. alpha, and various types of interleukin (IL), including IL6, in people with and without cancer.30–32 Furthermore, the protective effects Design of exercise have been attributed to the creation of an anti- Randomised trial inflammatory environment through increasing anti-inflammatory Published in English cytokines such as ILRa and IL10 in healthy people.26,33,34 The relationship between exercise dose and inflammatory markers in Participants people with cancer needs to be considered because strenuous Adults with cancer exercise can induce pro-inflammatory cytokines in healthy people.35 Therefore, it is important to know how much exercise Intervention can be safely tolerated in this immune-compromised population of Exercise intervention with aerobic or resistance exercise as a people with cancer. key component Sufficient reporting of dose (ie, the intensity or duration must Cancer-related fatigue affects 80 to 100% of patients.36 Fatigue be reported). For combined modalities, the intensity or total is a complex multi-dimensional construct related to reduced duration for both components must be specified. physical function and reduced health-related quality of life.27,37 Recent reviews have concluded that exercise reduces cancer- Outcome measures related fatigue,19,38–40 but the optimal dose to achieve this has not Must report at least one measure of fatigue or inflammation been established. It has been suggested that patients undergoing treatment may need to exercise at a lower intensity or for a shorter Comparisons duration than those who have completed primary treatment.41 Exercise versus control However, others have suggested that higher-intensity exercise Exercise plus usual care versus usual care only may be better.42,43 For example, Brown et al40 found that One exercise dose compared to another (eg, high versus low moderate-intensity resistance exercise may be more effective intensity) than low-intensity exercise for reducing cancer-related fatigue. The most effective duration and intensity of exercise remain component, because these modes of exercise are expected to result unclear. in significant physiological changes that may affect inflammation and fatigue, and are quantifiable. Furthermore, the intensity (eg, Therefore, the research questions that we sought to answer percentage of maximum heart rate, repetition maximum, etc) or with this systematic review were: duration of completed exercise needed to be reported. For studies using a combined exercise intervention (ie, aerobic and resistance 1. Is there a dose-response effect of exercise on inflammation and training), the intensity or total duration for both components must fatigue in adult cancer survivors? have been specified. Studies were excluded if only a single bout of exercise was used or if it was combined with a co-intervention 2. Is there a dose-response effect of exercise for improving such as diet or education. functional activity in this population? Quality assessment Method The studies were assessed by two reviewers, who indepen- This systematic review was reported in accordance with dently rated the 11 criteria on the PEDro scale as yes or no. One PRISMA guidelines.44,45 criterion relates to external validity; the remaining 10 criteria contribute 1 point each, if met, to give a score out of 10. The PEDro Search strategy score is a valid measure of internal validity and completeness of reporting. It has undergone Rasch analysis and has moderate levels The Medline, EMBASE and CINAHL databases were searched of inter-rater reliability (ICC 0.68, 95% CI 0.57 to 0.76).48,49 Trials from the earliest records to April 2015. PubMed was also searched scoring < 6 were deemed to be of low quality.50 from 2010 for more recent publications. The search strategy was based around synonyms and MeSH subject headings of the key Synthesis of results concepts of exercise and cancer combined with the primary outcomes of fatigue and inflammation. These terms were combined A standardised mean difference (SMD) was calculated for each with relevant filters to identify randomised, controlled trials.46 The outcome from post-intervention means and SDs to compare the detailed search strategy is presented in Appendix 1 (see control and treatment groups and to account for different scales eAddenda). The database searches were supplemented by citation of measurement between studies. Where only change scores tracking of included articles using Google Scholar and checking the were reported, the post-intervention mean was estimated in reference lists of included studies. reference to the baseline mean and the SD based on baseline data. If only a range was given, the SD was calculated.51 Authors were Eligibility criteria contacted if there was insufficient published data for analysis. Data from outcome measures were classified into three catego- The eligibility of papers identified by the searches was assessed ries to address the primary and secondary aims of the review: by two reviewers who independently considered information from inflammation, fatigue and activity. Activity was defined accord- the titles and abstracts against predetermined eligibility criteria ing to the World Health Organization International Classification (Box 1). Disagreements were resolved by discussion, with a third of Functioning as ‘the execution of a task or action by an reviewer consulted when necessary. Where eligibility was unclear individual’, which included measures of activities of daily living from the title and abstract, the full-text version was obtained and and functional mobility.52 examined by both reviewers. Meta-analysis was completed using the R statistics package To be included, studies had to be randomised, controlled trials ‘metafor’53 to provide evidence of the pooled effect size of the that: examined the effect of exercise in adults who had been exercise intervention. Data were combined if clinically homoge- diagnosed with cancer, reported at least one of the primary nous for more than two trials. Random effects models and a outcomes (fatigue or inflammation) and were published in English. restricted maximum likelihood estimator for the random effect The exercise intervention had to meet the definition ‘physical variance parameter were used.54 A meta-analysis of the ratio of activity that is planned, structured and repetitive and has a final or sample variances55 provided evidence of unequal variances intermediate objective of the improvement or maintenance of physical fitness’47 with aerobic or resistance training as a key
(Figue_1r[]GI)TD$70 Dennett et al: Dose-response of exercise for cancer survivors between the control and treatment groups. Consequently, Glass’ Papers found via search Papers found via reference of databases (n = 874) lists or citation tracking (n = 6) D56 was employed where the difference in means was standar- Records after duplicates removed (n = 683) dised using the control-group sample SD. Subgroup analyses were completed to determine the effect of tumour stream, treatment Titles and abstracts screened (n = 683) status and exercise modality. The Grades of Research, Assessment, Development and Evaluation (GRADE) approach was applied to Excluded based on title and abstract (n = 601) each meta-analysis to evaluate the evidence across trials.57 The approach involved downgrading the evaluation based on these Full-text articles assessed for eligibility (n = 82) predetermined criteria: the PEDro score was < 6 for the majority of trials used in the meta-analysis; there was greater than low levels Excluded after evaluation of full text (n = 33) of statistical heterogeneity between trials (I2 > 25%); there were large confidence intervals (ie, > 0.8 SMD); and if there was • not a randomised trial (n = 10) asymmetry of a funnel plot when more than 10 trials were • co-intervention (n = 8) included in the meta-analysis, demonstrating evidence of publi- • insufficient reporting of dose (n = 5) cation bias. • no measure of fatigue or inflammation (n = 3) • inadequate control (n = 3) Meta-regression analysis assessed the pooled dose-response • issues with group randomisation or analysis relationship between exercise dose and outcomes. To standardise dose for analysis, duration of exercise per week and intensity were (n = 3) evaluated separately, with intensity quantified as maximum oxygen consumption (VO2max) or a percentage of one repetition • participants receiving end-of-life care (n = 1) maximum (1RM). Meta-regression models were fitted to both factor and numeric variables to obtain subgroup estimates (for the Duplicate trial data (n = 7) factor variables) and estimated increases in effect sizes for one- unit changes of numeric covariates. Analysing exercise dose by Trials included in review (n = 42) volume in metabolic equivalents (METs)/minute/week was con- sidered; however, there were inadequate data. Exercise intensity Inadequate data for meta-analysis (n = 8) was categorised as low (< 40% VO2max, < 60% 1RM), moderate (40 to 60% VO2max, 60 to 80% 1RM) or vigorous/high (> 60% VO2max, Trials included in meta-analysis (n = 34) > 80% 1RM).58,59 Where outcome data could not be included in the meta-analysis or meta-regression, results were summarised Inadequate data for meta-regression of descriptively. inflammation (n = 3) Results Trials included in meta-regression (n = 31) Study selection Figure 1. Flow of trials through the review. The electronic database search resulted in a yield of 874 articles, which was reduced to 677 after the duplicates were removed. Additional articles were identified through citation tracking (n = 2) and reference list scanning (n = 4). Eighty-two articles were obtained in full text and further assessment reduced the yield to 49 articles. There was good inter-rater agreement about eligibility based on title and abstract (k = 0.695) and full texts (k = 0.691). Fourteen articles reported data from seven trials; therefore, 42 trials were included for review (Figure 1). Quality Intervention Trials included aerobic exercise (19 trials),30,60,61,63,68,70,72,78, The mean score of the included trials was 5.7 (SD 1.4) on the PEDro scale (Table 1). Inter-rater agreement on quality criteria was 80–84,86,87,89,91,92,101,104–106 resistance exercise (five trials),68,88,93, 98,99,102,107 a combination of aerobic and resistance exercise very good (k = 0.848). Three trials60–62 scored 8 on the PEDro scale, (14 trials)62,64–67,69,71,73,74,79,85,90,95–97,100 and four trials75–77,94,103 which was the highest possible score given the nature of the compared one exercise modality to another (eg, aerobic versus intervention that was studied, where it would be unfeasible to blind clinicians or participants. Less than half of the trials had resistance exercise). The interventions were usually completed in an blinding of assessment and concealed allocation. outpatient rehabilitation or fitness centre (20 trials),30,60,61,68,69, 71–80,82,95,97–103,105 at home (13 trials)64,70,81,84–87,90–92,94,96,104,106 or Study characteristics a combination of home and centre-based exercise (five trials).62,83, 88,89,93 The remaining four trials were completed while participants Participants were inpatients.63,65–67 Of the 23 trials that reported using Data from 3816 participants were included. The majority of supervision, 11 were supervised by an exercise specialist,30,60,62,66, 71,75,85,90,95,97,98,100,103,105 six by physiotherapists,68,69,74,80,88,89,96 participants were female (70%), with a mean age of 55 years (SD 9) four by a fitness trainer,70,83,93,99,102 one by a kinesiotherapist and and a mean body mass index of 27 kg/m2. Solid tumours were physician,101 and one by a trained research assistant.63 investigated in 34 trials (81%), haematological cancers were investigated in four trials,63–66 with an additional four trials The duration of the intervention ranged from 15 days65 to investigating a combination of solid and haematological can- 1 year,89,93 with most trials of at least 12 weeks duration. The cers.67–70 Breast cancer was the most frequently reported (27 trials, 64%),30,60–62,68–94 followed by prostate cancer (14 trials, exercise sessions were 10 to 90 minutes long, and completed two 24%).85,88,90,92,95–104 Interventions were commonly completed during the treatment phase (30 trials, 71%), with 12 trials to three times per week. The average amount of exercise completed in the post-treatment phase (Table 2). completed each week across the trials was 104 minutes. The intensity of the interventions varied from moderate, between 60% of maximal heart rate86,91 for aerobic exercise and 60% of 1RM for
Research 71 Table 1 PEDro scores of the included studies. Study Random Allocation Groups Participant Therapist Assessor < 15% Intention Between-group Point Total allocation concealed similar at blinding blinding binding dropout to treat comparisons measures and (0 to 10) baseline reported variability reported Battaglini 200871 Y N N N N NN N Y Y3 Baumann 201067 Y N Y N N NN Y Y Y5 Broderick 201368 a Y N Y N N YY N Y Y6 Guinan 201380 Y N Y N N YN Y Y Y6 Buffart 201495 a Y N Y N N NY N Y Y5 Galvao 2010100 Y Y Y N N NY Y Y Y7 Burnham 200272 Y N Y N N NY N Y Y5 Campbell 200573 Y N Y N N NY N Y Y5 Cantarero-Villanueva 201374 Y Y Y N N YY N Y Y7 Chang 200863 Y N N Y N NY Y N Y5 Cheville 201396 Y Y Y N N NN Y Y Y6 Christensen 2014107 Y N Y N N YY N Y Y6 Coleman 201264 Y N Y N N NY N Y Y5 Cormie 201398 Y Y Y N N NN Y Y Y6 Cormie 201597 Y Y Y N N NY Y Y Y7 Courneya 200499 a Y N Y N N NY N Y Y5 Segal 2003102 Y Y Y N N YN Y Y Y7 Courneya 200776 a Y Y Y N N NY Y Y Y7 Courneya 200777 Y N Y N N NY Y Y Y6 Courneya 2009105 Y Y Y N N NY Y Y Y7 Courneya 201375 Y Y Y N N NY Y Y Y7 Daley 200778 Y Y Y N N NY Y Y Y7 Fairey 200530 a Y Y Y N N YY Y Y Y8 Fairey 200560 Y N Y N N YY Y Y Y7 Gomez 201179 Y N Y N N NN N Y Y4 Headley 200481 Y N Y N N NN N Y Y4 Hornsby 201461 a Y Y Y N N YY Y Y Y8 Jones 201382 Y N Y N N NY Y Y Y6 Jones 201383 Y N Y N N YY Y Y Y7 Mock 200584 Y Y Y N N NY Y Y Y7 Monga 2007101 Y N N N N NN N Y Y3 Mustian 200985 a Y Y N N N NY Y Y Y6 Sprod 201090 Y N Y N N NY Y Y Y6 Mutrie 200762 Y Y Y N N YY Y Y Y8 Naraphong 201586 Y Y Y N N YN Y Y Y7 Oechsle 201465 Y N Y N N NY N Y N4 Oldervoll 201169 Y N Y N N NN Y Y Y5 Payne 200887 Y N N N N NY N Y N3 Rief 201488 Y Y N N N NN N Y Y4 Saarto 201289 Y Y Y N N NY Y Y Y7 Segal 2009103 Y Y Y N N NY Y Y Y7 Thorsen 200570 Y Y N N N NN Y Y Y5 Wang 201191 Y N N N N NY N Y Y4 Wenzel 201392 Y N Y N N NY Y Y Y6 Winters-Stone 201293 Y Y Y N N YN Y Y Y7 Windsor 2004104 Y Y Y N N NY N Y Y5 Wiskemann 201166 Y N Y N N NN N Y Y4 Yeo 2012106 Y Y Y N N NY N Y N5 Yuen 200794 Y N Y N N NN N N Y3 N = no, Y = yes a reports data on some or all of the same participants as the study below. resistance exercise,65 to high intensity, 100% of peak workload as Adverse events and adherence interval training for aerobic exercise61,82,105 and 85% of 1RM for resistance exercise97 (Table 3). No trials implemented low- The exercise interventions appeared to be safe and well tolerated. Of the 22 trials9,55,60,61,65,67–69,72–75,77–79,84,88,91,96–98,103,105 intensity exercise. reporting an adverse event, 19 of the 1888 exercise participants
72 Dennett et al: Dose-response of exercise for cancer survivors Table 2 Summary of trial characteristics. Study N Gender Age (y) Tumour stream Time period relative to Outcome measure (% F) mean (SD) treatment and type(s) of Battaglini 200871 Inflammation Fatigue Activity Baumann 201067 therapy Broderick 201368 a 400-m walk 20 100 57 (17) Breast On: chemotherapy Æ surgery Revised PFS 6-m walk Guinan 201380 Æ radiotherapy Timed STS 12-MWT Buffart 201495 a 64 45 45 (12) Lymphoma, leukaemia, On: HSCT 100% EORTC Timed Galvao 2010100 solid tumourb QLQ-C30 Burnham 200272 STS x 10 43 86 51 (9) Breast, colon, lymphoma, Post: time post- FACIT-F 12-MWT Campbell 200573 oesophageal, chemotherapy 4 (S SD 1) mth; AM-PAC SF gynaecological surgery 93%, chemotherapy 6-MWT Cantarero- 100%, radiotherapy 72% 400-m walk Villanueva 201374 6-m walk 26 100 48 (9) Breast Post: time post- CRP Timed STS Chang 200863 chemotherapy 4 (SD 1) mth; Timed stair Cheville 201396 chemotherapy 100% Christensen 2014107 radiotherapy 32.4%, hormone climb Coleman 201264 therapy 76.9%, surgery 100% 6-m walk Cormie 201398 Cormie 201597 57 0 70 (3) Prostate On: ADT + previous CRP EORTC 12-MWT radiotherapy 39%, QLQ-C30 chemotherapy 25% 18 83 54 (9) Breast, colon Post: average 10 mth; surgery LASA 61%, chemotherapy 78%, radiotherapy 56% 22 100 48 (8) Breast On: chemotherapy 73%, Revised PFS radiotherapy 73%, combination 45% 61 100 48 (15) Breast Post: surgery 100%, PFS chemotherapy 97%, radiotherapy 90%, hormone 100% 22 45 51 (55) Acute myelogenous On: chemotherapy 100% BFI leukaemia 66 47 65 (18) Lung, colon On: FACT-F chemotherapy Æ radiotherapy 30 0 35 (11) Germ cell On: average 158 days, surgery IL: 1b, 2, 6, 8, EORTC QLQ-C30 100%, chemotherapy 100% 10, 12 187 42 56 (4) Multiple myeloma On: chemotherapy, hormone FACT-F therapy, stem cell treatment 20 0 72 (13) Prostate Post: AST 100%, radiation 55%, MFI surgery 20% 63 0 68 (17) Prostate On: ADT 100%, previous CRP FACIT-F radiotherapy 5%, chemotherapy 2% Courneya 200499 a 155 0 68 (4) Prostate On: ADT 100%, previous FACT-F Segal 2003102 surgery + radiotherapy 242 100 Courneya 200776 a 49 (15) Breast On: chemotherapy FACT-An Courneya 200777 120 42 100% Æ previous surgery 301 100 Courneya 2009105 108 100 53 (18) Lymphoma On: chemotherapy 44% FACT-An 50 (9) Breast Courneya 201375 51 (9) Breast On: chemotherapy TOI Fatigue Daley 200778 Post: average 7.5 mth: Revised PFS chemotherapy 74%, radiotherapy 79%, hormone therapy 73%, surgery 100% Fairey 200530 a 52 100 59 (6) Breast Post: average 14 (SD 6) mth; IL: 1a, 4, 6, 10 Fairey 200560 surgery 100%, radiotherapy 71%, chemotherapy 40%, TNFa Gomez 201179 hormone therapy 46% 16 100 49 (6) Breast Post: average 36 (SD 12) mth; IL: 1a, 1b, 1Ra, 2, chemotherapy 100%, surgery 2Ra, 3, 4, 6 to 10, 100% 12, 13, 15 to 18 TNFa Headley 200481 32 100 51 (21) Breast On: chemotherapy 100% FACIT-F 20 100 49 (43) Breast Hornsby 201461 a On: chemotherapy 100% FACIT-F Jones 201382 75 100 56 (9) Breast Post: no therapy 8%, CRP Jones 201383 radiotherapy 63%, IL6 chemotherapy 52% TNFa On: chemotherapy 42%, Mock 200584 119 100 52 (9) Breast radiotherapy 58% PFS
Research 73 Table 2 (Continued ) Study N Gender Age (y) Tumour stream Time period relative to Outcome measure (% F) mean (SD) treatment and type(s) of Inflammation Fatigue Activity therapy Monga 2007101 21 0 69 (12) Prostate On: radiotherapy 100% Revised PFS 38 71 60 (12) Mustian 200985 a Prostate, breast On: radiotherapy 100%, BFI 6-MWT Sprod 201090 52 (10) current hormone 8%, previous FACIT-F chemotherapy 50%, surgery 47 (7) 84% 52 (17) Mutrie 200762 203 100 Breast On: surgery 100%, FACT-F 12-MWT 62 (4) chemotherapy only 8%, radiotherapy only 28%, 65 (6) combined 64% 63 (29) Naraphong 201586 23 100 Breast On: chemotherapy 100% Revised PFS Oechsle 201465 48 29 52 (1) MFIS 66 (7) Acute myeloid leukaemia, On: chemotherapy + HSCT 39 (1) non-Hodgkin’s lymphoma, 100% 50 (10) germ cell tumour, 60 (11) multiple myeloma 62 (1) Oldervoll 201169 231 62 69 (1) Gastrointestinal, breast, On: chemotherapy 55%, Fatigue Timed STS 49 (15) lung, urological, radiotherapy 6%, hormone questionnaire SWT 67 (13) gynaecological, 19%, targeted therapies 4% 41 (13) haematological Payne 200887 20 100 Breast On: hormone 100% IL6 Revised PFS Rief 201488 60 45 Lung, breast, prostate, On: radiotherapy 100%, EORTC melanoma, kidney, other hormone 43%, QLQ-FA13 immunotherapy 22%, chemotherapy 75% Saarto 201289 500 100 Breast Post: chemotherapy 100%, FACIT-F 2-km walk radiotherapy 78%, endocrine 84%, hormone 84% Segal 2009103 121 0 Prostate On: radiotherapy 100%, ADT FACT-F Thorsen 200570 139 54 61% EORTC Breast, gynaecological, Post: average 28 d; surgery QLQ-C30 lymphoma, testicular 82%, chemotherapy 100%, radiotherapy 57% FACIT-F Wang 201191 72 100 Breast On: surgery 100%, 6-m walk chemotherapy 100%, radiation 44% Wenzel 201392 126 39 Breast, colorectal, On: radiotherapy 52%, PFS 12-MWT chemotherapy 35%, combined prostate, other solid 7%, brachytherapy 6% tumour c Winters-Stone 201293 106 100 Breast Post: chemotherapy 60%, SCFS Timed STS radiotherapy 88% 4-m walk Windsor 2004104 66 100 Prostate On: radiotherapy 100%, BFI Modified hormone 29% SWT Wiskemann 201166 105 33 Leukaemia, lymphoma On: Allo-HSCT 100% EORTC QLQ-C30 6-MWT (various) MFI Yeo 2012106 102 44 Pancreas On: surgery 100%, FACIT-F chemotherapy /radiotherapy 73% Yuen 200794 29 100 Breast Post: 9 d to 35 mth; Surgery PFS 6-MWT 100% chemotherapy 82%, radiotherapy 77% ADT = Androgen deprivation therapy, AM-PAC SF = Activity Measure for Post-Acute Care Inpatient Mobility Short Form, AST = Androgen suppression therapy, BFI = Brief Fatigue Inventory, CRP = C-reactive protein, EORTC QLQ = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire (C30 or FA13 versions), FACIT-F = Functional Assessment of Chronic Illness Therapy - Fatigue, FACT-F/An = Functional Assessment of Cancer Therapy-Fatigue/Anaemia, HSCT = Haematopoietic stem cell transplantation, IL = interleukin, LASA = Linear Analogue Self Assessment, MFI = Multi-dimensional Fatigue Inventory, MWT = minute walk test, PFS = Piper Fatigue Scale, SCFS = Schwartz Cancer Fatigue Scale, STS = sit to stand, SWT = Shuttle Walk Test, TNFa = Tumour Necrosis Factor-alpha, TOI = Trial Outcome Index. a reports data on some or all of the same participants as the study below. b n = 3. c n = 8. had adverse events. Of these, six participants withdrew from the Adherence was reported in 30 trials. Twenty-two trials because of adverse event(s), including: dizziness, fatigue, bone trials30,60,61,66,68,72,74,75,78–80,82,85,89–91,93–100,102–105 reported ad- pain, chest pain, acute myocardial infarction, anaemia, dyspnoea and knee pain. The adverse events that did not affect exercise herence of > 75% attendance of exercise sessions or adherence participation were back pain, lower limb pain, post-exercise discomfort and a fall at home that was unrelated to the intervention. to the prescribed exercise protocol (in the case of non-supervised, A meta-analysis found moderate-quality evidence that exercise did not increase the risk of an adverse event compared with usual care home-based exercise). Fifteen trials were supervised outpatient and there was no difference between exercise modalities or sessions (14 centre-based,30,60,61,68,72,74,75,78–80,82,97,98,100,105 one intensities (Table 4). home-based),96 two were a combination of home and centre-based training89,93 and four were unsupervised interventions completed at home.85,90,91,94,104 Twenty-two trials included strategies to help improve exercise adherence and support behaviour change, such
74 Dennett et al: Dose-response of exercise for cancer survivors Table 3 Summary of exercise interventions. Study Program setting Exercise mode Session Frequency Program Intensity duration (sessions/wk) duration (wk) Battaglini 200871 Supervised, centre-based, Combined aerobic, (min) 2 15 40 to 60% VO2max Baumann 201067 individual resistance and flexibility Duration of 21 to 32 80% patient-achieved watt Supervised, inpatient Combined aerobic and therapy load hospital-based flexibility 10 to 20 10 (mean 4, SD 2) Poor 55 to 75% HRR Broderick 201368 a Supervised, centre-based Aerobic 21 to 42 2 8 Fair 60 to 80% HRR Guinan 201380 group and individual Average 65 to 85% HRR home 12 (based on initial fitness level) Buffart 201495 a Supervised, centre-based, Combined aerobic and 15 to 20 2 10 Aerobic: 65 to 80% HRmax, Galvao 2010100 (aerobic) 11 to 13 Borg RPE Resistance: group resistance 12 6 to 12 RM, 2 to 4 sets 8 Burnham 200272 Supervised, centre-based, Aerobic 32 3 3 Low group: 40% HRR group Moderate group: 60% HRR by week 10 Campbell 200573 Supervised, centre-based, Combined aerobic and 10 to 20 2 group resistance 40 3 60 to 75% age-adjusted Cantarero- Combined aerobic and 12 5 HRmax Villanueva 201374 Supervised, centre-based, resistance (aquatic) 20 4 Chang 200863 group Aerobic 3 Moderate RPE, 8 to 12 RM, 2 to Cheville 201396 (aerobic) 3 sets, Christensen 2014107 Supervised, inpatient Combined aerobic and NR 3 hospital-based, individual resistance Target HR = resting HR + 30 Coleman 201264 Resistance NR Unsupervised, home- 8 Moderate RPE, 10 to 15 RM, based, individual b Combined aerobic, 3.5 MET resistance and flexibility Supervised, hospital- 9 10 to 12 RM, 4 sets based outpatient, individual 15 Aerobic: 65 to 80% HRmax, 11 to 13 Borg RPE Resistance: Unsupervised, home- 60 to 80% 1RM, 15 to 17 Borg based, individual 12 8 to 12 RM, 2 to 4 sets Cormie 201398 Supervised, centre-based, Resistance 45 2 Cormie 201597 group 45 2 12 Aerobic: 70 to 85% HRmax Combined aerobic and Resistance: 6 to 12, 1 to 4 sets, Supervised, centre-based, resistance 60 to 85% 1RM group 12 8 to 12, 2 sets, 60 to 70% 1RM Courneya 200499 a Unsupervised, centre- Resistance NR 3 Segal 2003102 based, individual 3 Duration of 60 to 80% VO2max Courneya 200776 a Aerobic 15 to 45 chemotherapy Courneya 200777 Supervised, centre-based, Resistance NR 3 (>12, median 17) 8 to 12, 2 sets, 60 to 70% 1RM group Courneya 2009105 40 to 45 12 60 to 75% VO2peak + 1 session/ Supervised, centre-based, Aerobic wk interval at VO2peak from individual Mean 16 wk 9 Courneya 201375 Supervised, centre-based, Standard aerobic 25 to 30 3 Aerobic: 55 to 60% VO2peak individual High-dose aerobic 50 to 60 Resistance: 10 to 12, 2 sets, Daley 200778 Combined resistance and 50 to 60 3 60 to 75% 1RM Fairey 200530 a Supervised, centre-based, aerobic 3 Fairey 200560 individual Aerobic 50 3 8 65 to 85% age-adjusted Gomez 201179 Supervised, centre-based, 15 HRmax, RPE 12 to 13 group Aerobic 15 to 35 8 70 to 75% VO2peak Supervised, centre-based, individual Combined aerobic and 90 Duration of Aerobic: 70 to 80% HRmax resistance therapy Resistance: 8 to 10 RM, 2 to 12 3 sets Headley 200481 Unsupervised, home- Aerobic 20 3 NR based individual Aerobic 20 to 30 3 26 Hornsby 201461 a Initial: 60% baseline peak Jones 201382 Supervised, centre-based, Target: 2 sessions of 60 to 70% individual and 1 interval session 100% 50 to 80% HRmax Jones 201383 Supervised, centre-based, Aerobic 30 3 (centre) Mock 200584 individual and Aerobic + 2 (home) Duration of 50 to 70% HRmax unsupervised, home- therapy 65% HRmax based 30 5 to 6 (6 wk to 6 mth) Unsupervised, home- based, individual 8 Monga 2007101 Supervised, centre-based, Aerobic 30 3 group
Research 75 Table 3 (Continued ) Study Program setting Exercise mode Session Frequency Program Intensity duration (sessions/wk) duration (wk) (min) Mustian 200985 a Unsupervised, home- Combined aerobic and NR 7 4 60 to 70% HRR, 3 to 5 RPE Sprod 201090 based, individual b resistance 12 50 to 75% HRmax Mutrie 200762 Combined aerobic and 20 2 (group) Supervised, group and resistance + 1 (home) Naraphong 201586 unsupervised, home- Oechsle 201465 based Aerobic 20 to 30 3 to 5 12 40 to 60% HRmax, 12 to Oldervoll 201169 median 2 14 Borg RPE Payne 200887 Un-supervised, home- Combined aerobic and 40 5 40 to 60% 1RM Rief 201488 based, individual b resistance 8 Combined aerobic, 30 2 14 NR Saarto 201289 Supervised, hospital- resistance, and flexibility 24 based, inpatient Aerobic 20 4 ‘Moderate’ walking activity Segal 2009103 Supervised, centre-based, Resistance 30 3 NR group Aerobic 45 to 50 1 (group) + 2 52 14 to 16 RPE, 86 to 92% Unsupervised, home- to 3 (home) HRmax, 76 to 85% VO2max based, individual Aerobic Resistance 45 3 24 70 to 75% VO2max Supervised, centre-based, NR 8 to 12, 2 sets, 60 to 70% 1RM individual and unsupervised, home- based, individual Supervised, centre-based, group and unsupervised, home-based, individual Supervised, centre-based, individual Thorsen 200570 Supervised, home-based, Aerobic 30 2 14 60 to 70% HRmax, 13 to Wang 201191 individual Aerobic 30 3 to 5 15 Borg RPE Wenzel 201392 Aerobic 20 to 30 6 40 to 60% HRmax, 0.5 to Winters-Stone 201293 Unsupervised, home- Resistance 60 5 3 Borg RPE based, individual 3 Duration of 50 to 70% HRmax Windsor 2004104 Aerobic therapy (5 to 35) Wiskemann 201166 Unsupervised, home- Combined aerobic and 8 to 10, 1 to 3 sets, 60 to 70% based, individual resistance 52 1RM Yeo 2012106 Yuen 200794 Supervised, centre-based, Aerobic 30 3 4 60 to 70% HRmax group and unsupervised, Aerobic home-based, individual Resistance 20 to 40 3 (aerobic) + 2 1 to 4 pre-hospital Aerobic: 12 to 14 Borg RPE (aerobic) (resistance) + inpatient stay Resistance: 8 to 20 RM, 2 to Unsupervised, home- + 6 to 8 post- 3 sets, 14 to 16 Borg RPE based, individual 30 3 to 5 hospital ‘Brisk walk’ Partly supervised, 6 inpatient hospital-based, individual and home- 20 to 40 3 12 8 to 12 RM, 10 to 13 Borg RPE based NR Unsupervised, home- based, individual Un-supervised, home- based, individual HRmax = maximum heart rate, HR = heart rate, HRR = heart rate reserve, MET = metabolic equivalents, NR = not reported, RM repetition maximum, RPE = Rating of Perceived Exertion, VO2max = maximum volume of oxygen consumption, VO2peak = volume of oxygen consumption at peak exercise. a Reports data on some or all of the same participants as the study below. b With initial supervised instructional session. Table 4 Meta-analysis of adverse events in exercise trials compared with usual care. Outcome Subgroup Trials (n) Participants (n) Risk difference Quality of the evidence Adverse events (modality or intensity) (95% CI) (GRADE) 1060,61,68,72,77,78,84,91,103,105 748 Aerobic 376,98,103 265 0.01 (–0.02 to 0.05) moderate a Resistance 965,67,69,73,79,96–98,100 466 0.01 (–0.02 to 0.03) high Combined 865,72,74,75,84,91,96,98 495 0.01 (–0.01 to 0.04) Low-moderate intensity 1160,61,67,68,73,77–79,97,100,103,105 807 0.01 (–0.01 to 0.04) moderate b Moderate-high intensity 0.01 (–0.02 to 0.03) high 1710 Overall 21 0.01 (–0.01 to 0.02) moderate a moderate a GRADE = Grades of Research, Assessment, Development and Evaluation. One trial72 was not included in the analysis due to no usual-care comparison. GRADE working group grades of evidence (see reason for downgrade). PEDro score < 6 was considered lower quality. a Reason for downgrade: evidence of publication bias. b Reason for downgrade: seven trials65,67,69,73,79,96,97 were rated lower quality.
]GIF$DT)2_erugi([76 Dennett et al: Dose-response of exercise for cancer survivors fatigue was maintained up to 6 months after the intervention when compared with usual care (Table 5). One trial was not included in the meta-analysis because it did not have a usual-care control group.75 This trial found no differences in fatigue when comparing different exercise modalities and dose (with respect to duration). Inflammation A meta-analysis of five trials30,80,83,97,100 provided high-quality evidence of a non-significant reduction in levels of plasma CRP following exercise when compared with usual care. A meta- analysis of four trials79,83,90,107 provided moderate-quality evi- dence that there was no significant difference in plasma IL6 levels; and meta-analysis of two trials79,107 provided moderate-quality evidence that there was no difference in IL8 or IL10 levels following exercise (Table 5). Two trials that were not included in the meta-analysis due to insufficient data did not show a significant difference between IL6 levels87 or any of the cytokines tested.60 Another trial that was also not included in the meta-analysis due to insufficient data demonstrated a significant reduction in ILb and IL2 in the exercise group during chemotherapy, and a significant increase in IL8 levels during the exercise intervention compared with usual care.82 Activity Of the included trials, 18 reported on outcomes of functional mobility and one reported on activities of daily living. A meta- analysis of 15 trials provided very low-quality evidence of improvement in walking endurance following exercise when compared with usual care (Table 5, Figure 3). Sensitivity analysis was completed because a single trial demonstrated a significantly larger effect than other trials. When this study was removed, there was still a moderate effect in favour of the intervention, with high levels of heterogeneity (61%), which was unable to be explained by the study characteristics. There was evidence that exercise had a significant effect on walking endurance for solid tumour types (Table 6). Moderate-quality evidence of four trials demonstrated no difference in usual walking speed following exercise when compared with usual care. There was also no difference in sit-to- stand or stair climbing ability (Table 5). One trial found no differences in patient-reported activities of daily living when comparing combined exercise to usual care.96 Dose-response analysis: meta-regression Figure 2. SMD (95% CI) of effect of exercise compared with usual care, on fatigue by A total of 31 trials were included in a meta-regression analysis pooling data from 33 trials with subgroup analysis by tumour type (haematological, of the dose-response effect of intensity and duration of exercise mixed, solid), treatment phase (post-treatment, during treatment) and exercise programs on fatigue and walking endurance. modality (aerobic, combined aerobic and resistance, resistance). a Courneya et al.77. Meta-regression analysis demonstrated a significant effect of exercise intensity on fatigue. Aerobic exercise intensity was as use of telephone monitoring, exercise diaries, use of pedometers negatively associated with treatment effect using linear regression and supervised ‘booster’ sessions.30,60,62,64,68,73,78,80,81,83–87,89– models. For every 1% increase in intensity (from moderate to high) there was an estimated reduction of SMD 0.029 (95% CI 0.001 to 92,94–98,100,103,104,106 0.056) in the positive effect of exercise on fatigue (Figure 4, top panel, solid line). However, there was no evidence of this for the Effects of exercise on health outcomes: meta-analysis aerobic exercise component of the combined exercise studies (estimated reduction per 1% increase SMD 0.005 (95% CI –0.038 to Fatigue 0.048) (Figure 4, top panel, dotted line). With respect to resistance Meta-analysis provided moderate-quality evidence that exer- intensity (Figure 4, bottom panel) and exercise duration (Figure 5), the meta-regression analyses did not detect any significant cise had a positive effect on fatigue when compared with usual care associations. (SMD 0.32, 95% CI 0.13 to 0.52) (Figure 2, Table 5). When adjusting for tumour type in subgroup analyses, there was insufficient For walking endurance, only the intensity of aerobic exercise evidence to suggest that exercise was effective for haematological was analysed using data from the aerobic and the aerobic or mixed tumour types. There was strong evidence in favour of component of combined intervention trials. A quadratic meta- exercise with respect to solid tumours and those undergoing regression model demonstrated that moderate-intensity aerobic treatment (Table 6). With respect to mode of exercise, a exercise (70% relative intensity) led to a peak effect (Figure 6). This combination of aerobic and resistance training provided the association was close to significant. There was no association largest treatment effect, with smaller and similar estimated effects detected for exercise duration (Figure 7). for aerobic and resistance exercise alone (Table 6). Further meta- analysis provided moderate-quality evidence that a reduction in Meta-regression analysis was not completed for markers of inflammation because there were insufficient data. All studies measuring CRP and IL6 levels were moderate-intensity exercise.
Research 77 Table 5 Meta-analysis, overall effect of exercise on outcomes. Outcome Trials (n) Participants (n) Time of SMD (95%CI) I2 Quality of the assessment evidence (GRADE) Inflammation 479,83,90,107 148 immed 0.15 (–0.79 to 1.08) 84% moderate a IL6 279,107 IL8 279,107 43 immed –0.03 (–0.64 to 0.57) 0% moderate b IL10 530,80,83,97,100 CRP 43 immed –0.31 (–0.92 to 0.30) 0% moderate b 264 immed –0.15 (–0.39 to 0.10) 0% high c Fatigue d 3361–64,66–74,77,81,84–87,89,91–98,101–105,107 3336 immed 0.32 (0.13 to 0.52) 82% moderate a 762,68,74,77,85,104,107 721 2 to 6 mth post 0.39 (0.08 to 0.71) 71% moderate a Activity 1462–64,69,73,84,85,91,93,94,98,100,104 1032 immed 0.77 (0.26 to 1.28) 93% very low f walking endurance e 493,97,98,100 207 immed 0.22 (–0.32 to 0.77) 70% moderate a usual walking speed g 569,74,93,97,100 479 immed 0.25 (–0.30 to 0.80) 87% moderate a 297,100 120 immed sit to stand –0.18 (–0.54 to 0.18) high stair climb GRADE = Grades of Research, Assessment, Development and Evaluation, IL = interleukin, immed = immediate. GRADE working group grades of evidence (see reasons for downgrade). PEDro score < 6 was considered lower quality. a Reason for downgrade: heterogeneity. b Reason for downgrade: one trial79 was rated lesser quality. c Reason for downgrade: five trials63,64,66,72,91 were rated lesser quality without blinded outcome measures and allocation concealment. d Fatigue measures: Brief Fatigue Inventory, European Organisation for Research and Treatment of Cancer C30/FA13 Questionnaires, Functional Assessment of Chronic Illness Therapy - Fatigue, Functional Assessment of Cancer Therapy - Fatigue/Anaemia, Fatigue Questionnaire, Linear Analogue Self Assessment, Multi-dimensional Fatigue Inventory, Revised Piper Fatigue Scale, Trial Outcome Index - Fatigue. e Walking endurance measures: 6-min and 12-min walk tests; 400-m and 2-km walk time; Shuttle Walk Test. f Reason for downgrade: seven trials63,64,69,73,94,95,104 were without blinded outcome measures and allocation concealment, evidence of publication bias and heterogeneity. g Gait speed measures: 4-m and 6-m walk tests. Table 6 Meta-analysis, effect of exercise (post-intervention) on outcomes by subgroup. Outcome Subgroup Trials (n) Participants (n) SMD (95% CI) I2 Quality of the evidence (GRADE) Subgroup analysis by tumour stream Fatigue Solid 2361,62,72–75,77,81,84–87,89,91–94,97,98,101–104,107 2168 0.37 (0.16 to 0.58) 82% moderate a Endurance Haematological 463,64,66,105 360 –0.03 (–0.56 to 0.49) 82% low b Mixed 467–70 446 0.17 (–0.34 to 0.68) 82% Solid 1062,73,84,85,91,94,95,97,98,104 650 very low c Hematological 363,64,66 288 0.92 (0.26 to 1.59) 93% very low d 0.41 (–0.87 to 1.68) 93% very low e Subgroup analysis by treatment phase 1909 833 0.33 (0.12 to 0.53) 81% moderate a Fatigue Treatment 2661–64,66,67,69,73,75,77,81,84–86,91,92,95,97,101–105,107 0.19 (–0.19 to 0.58) 81% moderate a Post-treatment 768,70,72,74,89,93,94,98 1491 508 0.27 (0.00 to 0.54) 82% low f Subgroup analysis by exercise modality 975 0.19 (–0.24 to 0.62) 82% moderate a 293 0.41 (0.06 to 0.75) 82% Fatigue Aerobic 1861,63,68,70,72,75,77,81,84,87,89,91,92,94,101,103–105 29 1.28 (0.36 to 2.20) 94% low f Endurance Resistance 867,77,93,94,98,102,103,107 790 0.18 (–1.36 to 1.72) 94% very low g Combined 1162,64,66,67,69,73–75,85,94,95,97 0.61 (–0.10 to 1.31) 94% very low h Aerobic 463,91,94,104 moderate a Resistance 297,101 Combined 865,68,69,72,76,88,97,98,100 GRADE = Grades of Research, Assessment, Development and Evaluation. GRADE working group grades of evidence (see reasons for downgrade). PEDro score < 6 was considered lower quality. a Reason for downgrade: heterogeneity. b Reason for downgrade: three trials63,64,66 were rated lower quality, heterogeneity. c Reason for downgrade: heterogeneity, all trials were rated lower quality, without blinded outcome measures and allocation concealment. d Reason for downgrade: heterogeneity, five trials70,73,88,91,101 were rated lower quality, evidence of publication bias. e Reason for downgrade: heterogeneity, all trials were rated lower quality, without blinded outcome measures and allocation concealment, wide confidence intervals. f Reason for downgrade: heterogeneity, evidence of publication bias. g Reason for downgrade: heterogeneity, four trials were rated lower quality63,91,94,104, wide confidence intervals. h Reason for downgrade: heterogeneity, one trial was rated lower quality94, without blinded outcome measures and allocation concealment, wide confidence intervals. Discussion These findings support previous meta-analyses7,18,38–40 sug- This systematic review provided moderate-quality evidence gesting that exercise can play a significant role in reducing fatigue, that exercise reduces fatigue in cancer survivors and very low- particularly in people with solid tumours. Consistent with previous quality evidence that exercise improves walking endurance in this group. It also provided evidence of a negative dose-response evidence, these effects may not be generalisable to haematological relationship of aerobic exercise intensity and fatigue, and of a peak cancers.38,40 Patients with haematological malignancies can treatment effect of moderate-intensity aerobic exercise for improving walking endurance. No significant dose-response was experience many complications during treatment, including evident for the duration of weekly exercise. There was also muscle atrophy, cachexia, anaemia, physical deconditioning and moderate-quality to high-quality evidence that there is no psychological distress.64,108,109 In particular, anaemia has been significant difference in inflammatory markers after completion shown to affect people with haematological cancers more than of an exercise program compared with usual care and no those with solid tumours110 and this is a known contributor to significant difference in usual walking-speed, sit-to-stand ability cancer-related fatigue.27 Therefore, this complication may be less or stair climbing ability. able to be resolved through exercise. Our review also demonstrated a significant effect of exercise in reducing fatigue in people undergoing treatment, but not after treatment. A possible explanation could be a ceiling effect
]GIF$DT)3_erugi([78 Dennett et al: Dose-response of exercise for cancer survivors Figure 3. SMD (95% CI) of effect of exercise, compared with usual care, on walking be explained by a larger number of studies that included a specified endurance by pooling data from 14 trials with subgroup analysis by tumour type exercise dose, which is important because people need to exercise (haematological, solid) and exercise modality (aerobic, combined aerobic and with sufficient duration and intensity to be able to induce long- resistance, resistance). term physiological change to their health. occurring after treatment, when fatigue may be less severe. The significant reductions in fatigue were accompanied by Previously, reductions in fatigue were not found to be maintained significant improvements in walking endurance. It has been after completion of the exercise program.19,40 However, in this hypothesised that the physical dimension of fatigue has an organic cause.111 Cancer survivors have low physical activity levels,112 r[(Fig]GI$DT)4_eru eview, reduction in fatigue was maintained at follow-up. This may which in turn reduce physical performance and impair skeletal muscle function and cardiovascular fitness. This cycle of decondi- tioning, which perpetuates fatigue, can be broken through the physical adaptations of exercise training. Exercise may also provide the additional benefits of improved mood and reduced anxiety and fear, which are known contributors of cancer-related fatigue.111,113,114 The safety of exercise in people with cancer was re-enforced by this review, and evidenced by a low number of adverse events and a non-significant reduction in the inflammatory marker CRP. There was also no difference in the levels of the interleukins assessed. These inflammatory markers have previously been linked to tumour development and recurrence, as well as contributing to the development of fatigue. The results in this review suggest that exercise does not increase any pro-inflammatory markers, which contribute to cancer risk and tumour development. Moderate-intensity exercise has a greater effect on reducing fatigue and increasing walking endurance than high-intensity exercise. This is a plausible outcome, given the nature of the mechanism of physiological changes as a result of exercise. Regular exercise induces stress on the cardiovascular and muscular systems in order for physical adaptation to occur.115 However, in people with cancer, baseline exercise tolerance is reduced secondary to the effects of disease and treatment-related factors. Some tumours may directly disrupt pulmonary mechanics and may also be accompanied by side effects such as weight loss, anaemia and muscle wasting.116 Treatment such as chemotherapy and radiotherapy can further exacerbate issues with oxygen delivery by inducing pulmonary and cardiovascular damage as well as increasing inflammation and reactive oxygen species; such changes are correlated with change in myocardial strain.111,116 Therefore, while physical activity is important to relieve fatigue, a balance in the amount of physical activity is also required. However, it should be considered that other training factors such as interval period, duration and length of the program might Figure 4. Top panel: Meta-regression scatter plot of 25 trials showing the relationship between aerobic exercise intensity (% relative intensity) and the effect on fatigue (effect size, 95% CI), fitted with linear regression lines: solid line – aerobic; dotted line – combined. Bottom panel: Meta-regression scatter plot of 13 trials showing the relationship between resistance exercise intensity (% of 1 repetition maximum) and the effect on fatigue (effect size, 95% CI).
Search
