Sep 4, 2017 - OARSI and OMERACT suggest the primary outcome should be a .... We have added the paragraph introducing manual therapy under the ...
PEER REVIEW HISTORY BMJ Open publishes all reviews undertaken for accepted manuscripts. Reviewers are asked to complete a checklist review form (http://bmjopen.bmj.com/site/about/resources/checkli st.pdf) and are provided with free text boxes to elaborate on their assessment. These free text comments are reproduced below.
ARTICLE DETAILS TITLE (PROVISIONAL)
AUTHORS
Clinical Effectiveness of Patella Mobilization Therapy versus a Waiting list Control for Knee Osteoarthritis: A protocol for a pragmatic randomized clinical trial Sit, Regina Wing Shan ; Chan, Keith Kwok Wai; Yip, Benjamin Hon Kei; Zhang, Daisy Dexing; Reeves, Kenneth Dean; Chan, Ying Ho; Chung, Vincent; Wong, Samuel
VERSION 1 – REVIEW
REVIEWER
REVIEW RETURNED GENERAL COMMENTS
Philip Conaghan University of Leeds United Kingdom 04-Sep-2017 Thank you for the opportunity to read your protocol. Overall I have some major concerns around the design and methodology which are highlighted below: Overall comments: 1. Unclear why you are selecting a coexisting PFOA and TFOA group? Why not select only PFOA as this appears your desired target and the crux of your discussion 2. The introduction does not introduce PMT at any stage. This is the primary intervention so warrants some background 3. Is the waiting list an appropriate control group. It is widely accepted that doing nothing will not improve outcomes in KOA ultimately currently there is no clinical equipoise 4. The inclusion criteria do not mention x-ray or the views used 5. OARSI and OMERACT suggest the primary outcome should be a pain measure, not a composite pain-function measure which is what WOMAC total provides. So use of WOMAC pain, or a justified single WOMAC pain question are potential primary outcomes, but not the total WOMAC. 6. Intervention - 1 sessions every 2 months doesn't appear enough. What is the justification for this schedule? 7. The exercise regime within the intervention appears dated and does not align with the current evidence. Firstly more consideration of progression is warranted - consider using the TIDierR guidelines.
Furthermore, if the PFJ is the desired target why are hip based not used? and why are the exercises performed non-weight bearing? Specific comments: 1. Waiting list rather than 'waitlist' through out 2.Page 2, Line 53 - covariates plural 3. Page 4; Line 20 The sentence starting 'The standard of care....' needs rewording 4. Page 4; Line 29. I don't think the TFJ is two joints - maybe change this to two articulations 5. Page 5, Line 13-18. This idea remains speculative. These sentences need changing to reflect this point. 6. Page 6; 2nd paragraph. Sentences 2 and 3 appear to contradict each other. They begin talking about PFOA then finish with coexisting PFOA and TFOA . Needs formatting 7. Why < 75 years old? Justification given for younger age bracket but not for older value 8.Page 7; Line 18-27. I'm assuming you've added these clinical tests to identify PFJ pathology ( based on the work of Cook et al). However, surely your inclusion will be based on coexisting PFOA and TFOA which would require x-ray diagnosis. These tests reported are recommended in the absence of x-ray and typically for a non-OA patelofemoral pan cohort 9. Page 8; Line 44. SNOSE 'are' 10. Page 9; Line 39. The effect of VMO remains debatable ( see Chester et al 2008). This sentence needs to reflect this debate 11. Page 14; Line 8. Dropout rate of 85%? Thinks this needs rephrasing because this is very high 12. Page 14; Line 19 Why is only baseline WOMAC considered a covariates? Have other likely covariates been considered e.g. duration of pain etc. 13. Page 15; Line 3. The use of multiple imputation warrants more detail on how the imputation model was decided upon, the number of iterations and the ways of testing this stability.
REVIEWER
REVIEW RETURNED GENERAL COMMENTS
Ebru Kaya Mutlu Istanbul University, Faculty of Health Science, Division of Physical Therapy and Rehabilitation, Turkey. 18-Sep-2017 Thank you for sharing your work on answering a question relative to the physical therapy treatment of knee OA. I think this is an important question for a rapidly growing population. I think there is publication potential for the manuscript but it will require some revision. Presentation. It is usually possible to follow your thoughts. Introduction There is growing evidence that manual examination (manual therapy) leading to specific treatment for the identified movement and strength impairments associated with knee OA increase the level of symptom and function benefit. So, the introduction is clear however the rationale for the study is not convincing, we recommend that the introduction be strengthened. Page 5,line 44: There is stronger evidence for the pain/neurophysiological effects of manual therapy than re-align. Please refer to authors such as Sluka/Moss /Wright
In addition, why is this study needed? Methods Page 7, line 10: was it only the ACR clinical criteria? What about the radiographs? Perhaps it was the ACR clinical and radiographic criteria? Page 9: Any insight that you can provide as to the dose and vigor of the provided manual treatments will be useful to the reader and further work. i.e.”types of glides, which concept will be use Kalterborn or Maitland? No background/context is given to justify why these interventions were chosen. In addition, Reference for manual techniques is needed. why was the patellar mobilization done in the lying position? Page 12 Line 44: “”Degree of pain-free active knee flexion will be measured using a goniometer.” Need reference. Statistical analysis The statistical analysis using ANCOVA is appropriate for the RCT design. I suggest using bilateral status (Y, N) as a covariate.
VERSION 1 – AUTHOR RESPONSE 26th October. 2017 Dr Edward Sucksmith Managing Editor BMJ Open Ref: bmjopen-2017-019103 We thank you and the reviewer for your close read and insightful comments regarding our protocol “Clinical effectiveness of patella mobilization therapy versus waitlist control for knee osteoarthritis: A protocol for a pragmatic randomized clinical trial”. We have endeavoured to clarify issues of concern and believe this has resulted in a stronger paper. We have addressed concerns using the order and page number provided by the reviewer, and presented the specific concerns verbatim, with our responses following each comment. We refer to the clean version of the manuscript for reference of page numbers. We have also made small grammatical and diction changes throughout the manuscript to improve readability. Our thanks in advance for your second look at this protocol. Yours sincerely, Regina Sit Overall comments: 1. Unclear why you are selecting a coexisting PFOA and TFOA group? Why not select only PFOA as this appears your desired target and the crux of your discussion Thanks a lot for pointing out this and we have justified the target group under the “introduction” (page 7). Study has shown that that co-existing PFOA and TFOA is associated with more pain and greater loss of function in KOA, and 40% of patients have co-exist PFOA and TFOA. Therefore, we aim to conduct a randomized clinical trial to evaluate the clinical effectiveness of patella mobilization therapy (PMT) for a subgroup of patients with coexist PFOA and TFOA. We hypothesize that even though PMT targets at the PFJ only, an improvement of the disrupted biomechanics is going to reduce pain
and improve function for KOA as a whole. Besides, the study is designed based on primary care setting. Instead of pure PFOA, we encounter more patients with PFOA and TFOA in our daily practice. 2. The introduction does not introduce PMT at any stage. This is the primary intervention so warrants some background We have added the paragraph introducing manual therapy under the “background”(page 7) , and the details of PMT is stated under “Intervention”(page 12-15). 3. Is the waiting list an appropriate control group. It is widely accepted that doing nothing will not improve outcomes in KOA - ultimately currently there is no clinical equipoise The PMT is a unique protocol designed based on the experience of primary care physicians specialised in musculoskeletal pain medicine. Differ from traditional mobilization therapy which consists of multiple treatment sessions at intense frequency, we will provide 2-monthly mobilization with strong emphasis on daily vastus medialis oblique (VMO) firing exercise. We are not sure whether the protocol will work or if it will be better than the waiting list group. Therefore, the trial is designed as phase II clinical trial. We believe findings from this study will inform future phase III clinical trial when comparison can be made to standard treatment or other experimental treatments. Furthermore, usual care will be continued for both groups and co-interventions will be documented. 4. The inclusion criteria do not mention x-ray or the views used. Thanks for the kind reminder and we have added the X-ray views on page and 10 under “inclusion criteria”. 5. OARSI and OMERACT suggest the primary outcome should be a pain measure, not a composite pain-function measure which is what WOMAC total provides. So use of WOMAC pain, or a justified single WOMAC pain question are potential primary outcomes, but not the total WOMAC. Thank you very much for bringing this out. We have changed the primary outcome to WOMAC pain score. 6. Intervention - 1 session every 2 months doesn't appear enough. What is the justification for this schedule? As mentioned above, the intervention is designed based on the experience of primary care physicians specialised in musculoskeletal pain medicine. The aim of the mobilization is to provide immediate pain relief and mechanical re-alignment of the patella. Strong emphasis is put on VMO self-exercise. According to the National Institute for Health and Care Excellence (NICE), manual therapy is recommended as adjunct therapy to exercise. We believe the intervention will be suitable to be applied in primary care practice. Details of the intervention are stated on page 12-15. 7. The exercise regime within the intervention appears dated and does not align with the current evidence. Firstly more consideration of progression is warranted - consider using the TIDierR guidelines. Furthermore, if the PFJ is the desired target why are hip based not used? and why are the exercises performed non-weight bearing? We have re-written the intervention according to the TIDieR guideline (page 12-15).Thanks for suggesting the hip exercise. According to the 2016 Patellofemoral Pain Consensus Statement, combined knee and hip exercise is recommended for PFJ pain. Unfortunately, the study proposal was
designed in 2014 and therefore we did not put this in the protocol. Since the study has been started since June 2015, we apologize that we cannot amend further on this. However, we shall consider this in future PFJ related study. The justification of why non-weight bearing exercise is chosen is stated on page 14. We believe a non-weight VMO exercise is easy to perform, does not require any equipment and is safe for non-athletics elderly population which KOA is common. Study has also shown that a non-loaded VMO training in an open kinetic chain is able to strengthen muscle architecture as evidenced by the ultrasound. Specific comments: 1. Waiting list rather than 'waitlist' throughout. We have amended accordingly. 2.Page 2, Line 53 - covariates plural. We amended the word. 3. Page 4; Line 20 The sentence starting 'The standard of care....' needs rewording. We removed the wordings “standard of care”. (line 95) 4. Page 4; Line 29. I don't think the TFJ is two joints - maybe change this to two articulations. We have made the changes. 5. Page 5, Line 13-18. This idea remains speculative. These sentences need changing to reflect this point. “It is believed that” has been added to the sentence 6. Page 6; 2nd paragraph. Sentences 2 and 3 appear to contradict each other. They begin talking about PFOA then finish with coexisting PFOA and TFOA . Needs formatting. We have formatted the structure of the introduction page 6-7. 7. Why < 75 years old? Justification given for younger age bracket but not for older value. We have added the justification on page 9. 8.Page 7; Line 18-27. I'm assuming you've added these clinical tests to identify PFJ pathology ( based on the work of Cook et al). However, surely your inclusion will be based on coexisting PFOA and TFOA which would require x-ray diagnosis. These tests reported are recommended in the absence of x-ray and typically for a non-OA patelofemoral pain cohort. We clarified this on page 9-10. 9. Page 8; Line 44. SNOSE 'are': We have amended accordingly. 10. Page 9; Line 39. The effect of VMO remains debatable (see Chester et al 2008). This sentence needs to reflect this debate. Sentence added on page 14. 11. Page 14; Line 8. Dropout rate of 85%? Thinks this needs rephrasing because this is very high Sorry for the typo, the dropout rate is 15% instead. We have amended in the text (page 19). 12. Page 14; Line 19 Why is only baseline WOMAC considered a covariates? Have other likely covariates been considered e.g. duration of pain etc. We have amended the sentence on page 20. Baseline WOMAC score, duration of knee pain, number of comorbidities and bilateral knee pain status (yes or no) will act as covariates 13. Page 15; Line 3. The use of multiple imputation warrants more detail on how the imputation model was decided upon, the number of iterations and the ways of testing this stability. We have amended the statistical analysis on page 20-21. Reviewer: 2 Reviewer Name: Ebru Kaya Mutlu Institution and Country: Istanbul University, Faculty of Health Science, Division of Physical Therapy and Rehabilitation, Turkey. Competing Interests: None declared. Thank you for sharing your work on answering a question relative to the physical therapy treatment of knee OA. I think this is an important question for a rapidly growing population. I think there is publication potential for the manuscript but it will require some revision.
Presentation. It is usually possible to follow your thoughts. Introduction 1.There is growing evidence that manual examination (manual therapy) leading to specific treatment for the identified movement and strength impairments associated with knee OA increase the level of symptom and function benefit. So, the introduction is clear however the rationale for the study is not convincing, we recommend that the introduction be strengthened. We have re-written the introduction as shown on page 6-7. We hope the restructuring can strengthen our study objectives. 2.Page 5,line 44: There is stronger evidence for the pain/neurophysiological effects of manual therapy than re-align. Please refer to authors such as Sluka/Moss /Wright In addition, why is this study needed? Thank you very much for suggesting this. We have put the rationale behind the PMT on page 12-15. The role of manual therapy has also emphasized under “introduction” on page 7. 3.Page 7, line 10: was it only the ACR clinical criteria? What about the radiographs? Perhaps it was the ACR clinical and radiographic criteria? We have made it clear under the “inclusion criteria” on page 10. 4.Page 9: Any insight that you can provide as to the dose and vigor of the provided manual treatments will be useful to the reader and further work. i.e.”types of glides, which concept will be use Kalterborn or Maitland? No background/context is given to justify why these interventions were chosen. In addition, Reference for manual techniques is needed. why was the patellar mobilization done in the lying position? We have re-written the PMT according to the TIDIER guidelines. We designed the unique PMT protocol, taking into account the procedure is to be applied for primary care practice. However, we do use the Kalterborn Theory of providing Grade 3 stretch to the tight lateral retinaculum using a medial guide technique. We wish to clarify that the procedure will be performed with patient on lateral lying position with knee supported with a wedge. (please see figure 2) 5.Page 12 Line 44: “”Degree of pain-free active knee flexion will be measured using a goniometer.” Need reference. Reference 53 added. 6.Statistical analysis The statistical analysis using ANCOVA is appropriate for the RCT design. I suggest using bilateral status (Y, N) as a covariate. Thank you very much and other covariates are added as well on page 20.
VERSION 2 – REVIEW
REVIEWER REVIEW RETURNED
Ebru Kaya Mutlu Istanbul University 23-Nov-2017
GENERAL COMMENTS
Accept
REVIEWER REVIEW RETURNED
Philip Conaghan University of Leeds UK 24-Nov-2017
GENERAL COMMENTS
Overall comments 1. Overall the rationale for PMT is much clearer, however, I still not convinced that 1 session every two months is going to be effective. I understand that this fits with a primary care clinic but is there any evidence to support this schedule of treatments? 2. Is it possible that any effect you get from this trial is the result of the exercise rather than the PMT? This is likely considering you are comparing to a waiting list control 3. The selection of covariates appears appropriate. I notice you are collecting analgesic use – Could this be used as a covariate because it’s likely to affect both the presentation and the outcome? 4. Discussion. I agree with justification of including co-existing PFOA and TFOA, however, the discussion reads as if you have only selected PFOA. This needs rewording to ensure the reader knows that you are targeting the PFJ in a cohort with knee OA ( PFOA and TFOA) Specific comments Page 7. ‘Office based’. This is mentioned in the background but not mentioned again – is this to be delivered in the work place? Page 10 “Presence of osteophytes… - this sentence warrants a citation Page 11. Typo ‘envelop’ Page 13. ‘deranged biomachnism’ – suggest rewording and spell check Page 14. Suggest changing ‘lateral lying’ to ‘side lying’ Page 14. “sense of giving way” – suggest rephrasing this and giving some justification. Page 19. Typo ‘WOMA’ Page 20. CONSORT does not recommend comparing baseline differences in RCTs so I suggest that chi-squared testing is not required. This could instead be reported descriptively.
VERSION 2 – AUTHOR RESPONSE December 8, 2017 Dr. Edward Sucksmith Assistant Editor BMJ Open Ref: Clinical Effectiveness of Patella Mobilization Therapy Versus a Waiting List Control for Knee Osteoarthritis: A Protocol for a Pragmatic Randomized Clinical Trial (bmjopen-2017-019103)
We thank you and the reviewers for reviewing our manuscript the second time around. We have removed the discussion section from the abstract, and the manuscript has been proofread and edited by “Editage” again with grammatical and diction changes to improve the overall readability. We have addressed concerns using the order and numbering provided by reviewer 1, and our response follow each comment. Please refer to the original version of the manuscript for the page numbers. Thank you very much and we look forward to hearing from you. Sincerely yours, Regina Sit Response to reviewer 1: Comments 1. Overall the rationale for PMT is much clearer, however, I still not convinced that 1 session every two months is going to be effective. I understand that this fits with a primary care clinic but is there any evidence to support this schedule of treatments? Answer: We fully understand the concern. We designed the study based on the clinical experience of primary care physicians who specialize in musculoskeletal pain medicine. Anecdotal evidence from clinical observations is encouraging, which served as a basis for this randomized clinical trial. However, we do have the same concern as the reviewer. Thus, we decided to conduct the study as a phase II clinical trial. If the findings are positive, then it will be helpful in providing evidence for future phase III clinical trials when comparison can be made to other standard or experimental treatment. 2. Is it possible that any effect you get from this trial is the result of the exercise rather than the PMT? This is likely considering you are comparing to a waiting list control Answer: Thank you. We appreciate the opportunity to clarify this important issue. According to the 2016 Patellofemoral Pain Consensus Statement, combined interventions are recommended to treat PFJ pain.(1) Furthermore, manual therapy is recommended by the National Institute for Health and Care Excellence as an adjunctive therapy to exercise for individuals with osteoarthritis.(2) Therefore, we believe that the best intervention for PFOA is a combination of manual therapy and exercise. We have no intention to separate the two. However, the best combination that will work in clinical settings should be evaluated. (3) We would like to emphasize that this is a “pragmatic” trial. Knowing that it may have less scientific benefits than an “explanatory” trial, we still believe that looking for an intervention with maximal applicability and generalizability is more important in primary care settings. 3. The selection of covariates appears appropriate. I notice you are collecting analgesic use – Could this be used as a covariate because it’s likely to affect both the presentation and the outcome? Answer: Thank you very much for the suggestion, and we have added the analgesic consumption as one of the covariates (page 19). 4. Discussion. I agree with justification of including co-existing PFOA and TFOA, however, the discussion reads as if you have only selected PFOA. This needs rewording to ensure the reader knows that you are targeting the PFJ in a cohort with knee OA (PFOA and TFOA) Thank you very much for the suggestion. We have rephrased the paragraph (page 20). “PFOA is a critical source of pain, and a co-existence of PFOA and TFOA is sometimes observed. This subgroup is important, but it is an under-recognized subgroup of KOA. Pain and functional impairment in individuals with KOA are associated with a multifact orial set of degenerative intraarticular cartilage, bone, and synovial knee structures in addition to a complex interaction among genetic, psychosocial, and other factors. Thus, the identification of the subgroup with PFOA and customized interventions to correct the disrupted biomechanics can potentially reduce the disease burden.”
Specific comments Page 7. ‘Office based’. This is mentioned in the background but not mentioned again – is this to be delivered in the work place? Answer: We replaced the word “office” with “clinic”, and the word “clinic” was mentioned as part of the term “general outpatient clinics (GOPCs)” throughout the manuscript. Page 10 “Presence of osteophytes… - this sentence warrants a citation. Citation is added (Page 9, Reference 34). Page 11. Typo ‘envelop’ Answer: The typo has been corrected. (page 11) Page 13. ‘deranged biomachanism’ – suggest rewording and spell check We replaced this term with “disrupted biomechanics.” (Page 12) Page 14. Suggest changing ‘lateral lying’ to ‘side lying’. Answer: The term has been changed to “side lying.” (Page 13) Page 14. “sense of giving way” – suggest rephrasing this and giving some justification. We have rephrased the paragraph into “The knee will be flexed to a degree that allows vertical gravitational force to be applied from the palm to glide the patella from the lateral edge to the medial direction, which provides grade 3 stretch to the tight lateral retinaculum. (4)” (Page 13, reference 47) Page 19. Typo ‘WOMA’ The typo has been corrected. Page 20. CONSORT does not recommend comparing baseline differences in RCTs so I suggest that chi-squared testing is not required. This could instead be reported descriptively. Thank you for the suggestion. The sentence has been rephrased to “Descriptive statistics will be used to compare the baseline characteristics between the two groups” (page 19). References: 1. Crossley KM, Stefanik JJ, Selfe J, et al. 2016 Patellofemoral pain consensus statement from the 4th International Patellofemoral Pain Research Retreat, Manchester. Part 1: Terminology, definitions, clinical examination, natural history, patellofemoral osteoarthritis and patient -reported outcome measures. British Journal of Sports Medicine. 2016;50(14):839-843. 2. National Clinical Guideline C. National Institute for Health and Clinical Excellence: Guidance. Osteoarthritis: Care and Management in Adults. London: National Institute for Health and Care Excellence (UK) Copyright (c) National Clinical Guideline Centre, 2014. 3. Patsopoulos NA. A pragmatic view on pragmatic trials. Dialogues in Clinical Neuroscience. 2011;13(2):217-224. 4. Mangus BC, Hoffman LA, Hoffman MA, Altenburger P. Basic principles of extremity joint mobilization using a Kaltenborn approach. Journal of Sport Rehabilitation. 2002;11(4):235-250.
