All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
Thank you for your careful revision and for addressing the previous comments thoroughly. The manuscript has now been re-evaluated, and the reviewer recommends acceptance. After editorial assessment, I agree that all concerns raised in the earlier round have been fully resolved.
Your revisions have strengthened the clarity, methodological transparency, and overall presentation of the systematic review and meta-analysis. The results, interpretation, and conclusions are now coherent, well-supported, and aligned with PRISMA standards.
I am pleased to confirm that your manuscript is accepted in its current form.
Congratulations, and thank you for your diligent work.
[# PeerJ Staff Note - this decision was reviewed and approved by Mike Climstein, a PeerJ Section Editor covering this Section #]
Title & Abstract
Title: Page 7: Line 1-2: This is clear and focused. The key elements of the research question are included in the text. The study design is also mentioned as required by PRISMA checklist requirements. I have nothing further to recommend.
Abstract, Page 8, Line 30 to Page 9, Line 58: The abstract is a clear report of the study methods and findings. The conclusions drawn are grounded in the results presented. These are accurately interpreted. I therefore have nothing further to recommend to the Abstract section.
Introduction
Introduction, Page 9, Line 61 to Page 10, Line 104: The Introduction section provides the context of the study. There is a clear articulation of the key elements of the research proposed including information on the knee osteoarthritis population, the intervention under investigation i.e. proprioceptive neuromuscular Facilitation, and the gap in knowledge. The rationale for the study is clear and well-communicated. I have nothing further to recommend to the Introduction.
Figures & Tables
The figures and tables are clear and legible. The figures are free from unnecessary modifications.
Material and Methods
Methods, Page 10, Line 106 to Page 14, Line 216: The methods taken to answer the research question are appropriate. The statistical analysis approach is clearly communicated. The use of a random-effects model is appropriate for this particular data. The section is reported in accordance with PRISMA expectations. I have nothing further to recommend.
Results
Page 15, Line 218 to Page 20, Line 363: The Results have been revised to ensure that the previous two total knee arthroplasty papers have been removed from the analysis. This is appropriate and means the analysis of knee osteoarthritis is clearer and not confounded. This has also been acknowledged in the Methods section (Methods, Section 2.2 Eligibility and Exclusion Criteria, Page 12, Line 147-148). The reporting of the analysis is clear. The interpretation of the statistical findings is well-made. The acknowledgement of critical appraisal results is also well-communicated in Section 3.2 (Page 16, Line 271-295). I have nothing additional to report to the Results section.
Discussion
Discussion, Page 20, Line 363 to Page 24, Line 476: The Discussion is clear. The interpretation is appropriate. The recommendations on clinical implications are appropriate based on the results presented and the study design. There are reflections and criticality of the results against the previous literature to provide analytical insights. The Limitations presented in Pages 23 to 24 (Lines 454 to 476) are reasonable. I have no further recommendations to the Discussion section.
Conclusion
Conclusion, Page 24, Lines 479 to 486: The Conclusions are clear and well-presented. The interpretation and summary statements are grounded in the results. The acknowledgement of future research needs is appropriate as are the current ‘inconclusive’ pain outcomes reported in Line 483-484. I have nothing further to recommend to this final section.
None
**PeerJ Staff Note:** Please ensure that all review, editorial, and staff comments are addressed in a response letter and that any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate.
**Language Note:** When preparing your next revision, please ensure that your manuscript is reviewed either by a colleague who is proficient in English and familiar with the subject matter, or by a professional editing service. PeerJ offers language editing services; if you are interested, you may contact us at [email protected] for pricing details. Kindly include your manuscript number and title in your inquiry. – PeerJ Staff
Title & Abstract
Title: This includes all key information required to signpost the paper. I have nothing further to recommend.
Abstract: Page 8, Lines 31 to Page 9, Line 59: The abstract is a clear synopsis of the study. All key information regarding the research question, study design and methods, results, and interpretation is clear and appropriate. I have nothing further to recommend.
Introduction
Introduction, Page 9, Line 3 to Page 10, Line 105: The Introduction provides a clear background to the study. The surrounding literature provides context for the key elements of the research question and study. The rationale for the study is also clear and appropriately cited. I have nothing further to recommend.
Figures & Tables
The figures and tables are clear, legible, and appropriate. The figures are free from unnecessary modifications.
Material and Methods
Methods, Page 10, Line 108 to Page 13, Line 182: The authors clearly present their Methods. These are robust and appropriately communicated. The only persistent issues remain with the inclusion of patients with total knee replacement. The research question relates to a target population with knee osteoarthritis. Including people with total knee replacement is not the same population as they have had the arthritic elements of the joint removed.
The authors, in the response document, acknowledge that a proportion of these patients also have persistent pain. This is true with approximately 20% of people with persistent pain and disability after knee replacement. However, the rationale and mechanism behind this population and the treatment effect may differ. The acknowledged heterogeneity in the Discussion section does not mitigate this.
I strongly recommend (again) that the two studies (Chow and Ng, 2010a; Alaca et al, 2015) recruiting people post-total knee replacement be removed for this systematic review exploring the clinical outcomes of people with knee osteoarthritis.
Results
Results, Page 15, Line 218 to Page 21, Line 375: The results section is clearly presented. The results are robustly interpreted, both with the meta-analysis appropriate but also with the GRADE assessment. The interpretations are appropriately cautious and grounded within the data and the Risk of Bias assessment. I have nothing further to recommend.
Discussion
Discussion, Page 21, Line 379 to Page 25, Line 501: The Discussion section provides a balanced interpretation of the findings. There is appropriate referencing and cross-referencing to the previous literature, with criticality presented to the interpretation. The practical conclusions drawn from the findings are appropriate. The limitations section includes all appropriate reflections. I have nothing further to recommend.
Conclusion
Conclusions, Page 25, Line 504 to Line 511: The Conclusion section points are robust and an accurate reflection of the data presented in the meta-analysis and Risk of Bias assessment. The practical implications are appropriate. I have nothing additional to recommend to this section.
**PeerJ Staff Note:** Please ensure that all review, editorial, and staff comments are addressed in a response letter and that any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate.
Title & Abstract
Title: The title is clear and informative, including all key elements of the study design and research question. I have nothing further to recommend.
Abstract: Objective, Methods, Results, and Conclusions, Page 2, Lines 34 to Page 3 Line 66: These sections are clearly reported. The addition of further evidence is important and provides greater clarity in the findings. The interpretation is appropriate based on the findings reported in the Results section. Ultimately, this abstract is a clear summary of the findings. I have nothing further to recommend.
Introduction
Page 3, Line 70 to Page 5, Lines 125: The Introduction section is clear, informative, and is grounded by appropriate evidence. The basis of the interventions for knee osteoarthritis is outlined and supported by guidelines and clinical research evidence. The rationale for the systematic review, building on previous research synthesis evidence, is also clearly communicated. The key elements of the Introduction are presented. I have nothing further to recommend.
Figures & Tables
The tables and figures are clear and legible. The figures are free from unnecessary modifications.
Material and Methods
Methods, Page 5, Line 127 to Page 9, Line 260: The Methods are more comprehensive and informative. The revised paper presents a clear outline of the search strategy, data extraction processes, as well as critical appraisal. The Methods section outlined who performed each of these key steps. The Methods section also includes considerable information on what constitutes the experimental PNF interventions and what the comparator group could include. This is informative and important when interpreting the robustness of the meta-analysis assumptions. The updated search to August 2025 evidence is appropriate and means that the Methods are now contemporary. Whilst I have reservations on the appropriateness of meta-analysis, the sensitivity analyses and analysis approach through exploring similar comparator interventions go some way to mitigate this. Accordingly, I think the current methods are acceptable. I have no further recommendations for the Methods section.
Results
Results, Page 11, Lines 313 to Page 18, Lines 493: The Results section is focused and clear. There are three points that the authors should recommend to ensure that this is most appropriately reported. These are listed below.
Results, Page 12, Line 321: The statement “perfect agreemen39” should be “perfect agreement39”
Results, Page 12, Line 346 to 347: I am confused by the statement “Two studies12,33 focused on participants with knee osteoarthritis following total knee arthroplasty” – surely if these participants had undergone total knee arthroplasty, then they did not have knee osteoarthritis and should therefore be excluded. I recommend that the authors reconsider the eligibility of this paper.
Results, Page 14, Line 390: Section 3.3 Certainty of Evidence: I recommend the authors review the literature on GRADE and particularly the Cochrane Handbook for Interventional Systematic Reviews. GRADE should be reported by analysis and by outcome, and therefore linked to the analysis interpreted. It is less effective when reported as a stand-alone paragraph, as is the case here. GRADE will vary depending on what evidence and what results are generated from the analysis. I recommend that the authors consider reporting, as is conventional, the GRADE for each specific analysis rather than in Section 3.3 as a stand-alone paragraph.
Discussion
Discussion, Page 18, Line 497 to Page 23, Line 645: The Discussion provides (largely) caution, with the GRADE analysis considered throughout. There is good use of evidence to underpin the statements and assumptions made in the interpretation. The clinical recommendations are also appropriately guarded with an acknowledgement of the limitations in the evidence. I have one point of note to consider, as listed below.
Discussion, Page 18, Line 498-505: The authors have reported a significant difference in immediate follow-up but neglected to highlight the futility or lack of effect at the post-treatment (slightly longer follow-up). This should be more clearly communicated. The reporting of meta-analysis and narrative analysis, being the principal differences rather than the follow-up time point, is a little confusing, and I think this could be more clearly reported. I recommend that the authors provide a more balanced interpretation of the positive and negative findings within this opening paragraph, but also across the Discussion section.
Discussion, Page 22, Line 647 to Page 24, Line 680: The limitations section is appropriate. The acknowledgement of concerns regarding the assumptions required for meta-analysis is discussed. I have nothing further to suggest.
Conclusion
Conclusion, Page 24, Lines 682 to 692: The conclusions section is appropriate. There is greater acknowledgement of the issues regarding the certainty of evidence and the short-term nature of this data analysis. The clinical recommendations are therefore guarded. I have nothing further to recommend to this current Conclusions section.
**PeerJ Staff Note:** Please ensure that all review, editorial, and staff comments are addressed in a response letter and that any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate.
**Language Note:** When you prepare your next revision, please either (i) have a colleague who is proficient in English and familiar with the subject matter review your manuscript, or (ii) contact a professional editing service to review your manuscript. PeerJ can provide language editing services - you can contact us at [email protected] for pricing (be sure to provide your manuscript number and title). – PeerJ Staff
As your reference 4 states, education, exercise, and if necessary, weight reduction are the core treatments of OA - this should be explained in the introduction, not only mentioned as “non-pharmacological treatment”. Also, pharmacotherapy and especially surgical procedures should only be used in a few cases, which should be explained. As the introduction is written, the three treatments are explained as equivalent to each other.
Line 83: The authors outline typical PNF protocols, verified by only one reference. I suggest adding more references for outlining typical PNF protocols.
Use the past tense in the aim in abstract and main text.
Explain all abbreviations the first time they are used in the abstract (AROM).
Line 142: explain K/L grade
Line 218: For the benefit of the reader, explain the K/L grades here as well: “…selected patients with K/L grade II (minimal OA) or III (moderate OA)…”
Add main findings to the first paragraph in the discussion.
Line 352-354: The sentence “Given these findings, etc.” is better suited in the section “clinical implications”.
Line 372-373. I do not understand the sentence “ Our study suggests that PT may substitute for RT….”. Please develop and explain.
Conclusion: PNF provides greater improvements in pain reduction - compared to what?
The rehabilitation techniques that are identified should be outlined, for example, in the first paragraph of the discussion.
Figure S4: Use a different background colour; the table is hard to read.
Title & Abstract
Title: This is explanatory and clear. I have nothing further to recommend.
Abstract, Objectives, Page 7, Line 32: This is currently an ‘aim’ and not an ‘objective’. I recommend that the authors correct this to be in agreement with the journal’s requirements.
Methodology, Page 7, Line 36: The PRISMA guidelines are systematic review reporting guidelines and not methodological recommendations for a meta-analysis. I recommend that the authors revise this sentence to reflect this.
Methods, Page 7, Line 40: The remaining text of the Methods is clearly communicated.
Results, Page 7, Lines 45-47: The comparison interventions are very different within the meta-analysis. This undermines the robustness of the analysis as study homogeneity is not assured, and a key statistical assumption cannot be met. This point will be raised later, but this will require a major change in the results presented in the Abstract.
Results, Page 7, Line 50: Based on the abstract results, I recommend that the authors reconsider whether the mean difference or the standard mean difference is presented. The MCID is referred to, but not the pain mean difference data is presented the I recommend that this is more clearly reported if a meta-analysis is adopted.
Results, Page 7, Line 50-51: The reference to RT in pain relief is not supported with numerical data. Understanding this statement is not clear. I recommend that this be revised so that more data is offered to allow the interpretation of this statement.
Conclusion, Page 8, Lines 53-57: Due to the issue in the pooled analysis, the conclusions drawn may have issues. I recommend a re-analysis of the data before these conclusions can be re-evaluated.
Introduction
Page 8, Lines 61-85: The Introduction is clear with the context provided. There is appropriate previous literature to underpin the argument and context. The rationale for the study is presented. I have identified a key point that should be considered.
Page 9, Line 91: There is confusion about what routine treatment is compared to rehabilitation techniques. I recommend that the authors clearly communicate what the routine treatment is for their systematic review. I am mindful that this may vary internationally, and therefore having a clear context of what this variability is, but also what the core agreed elements are, would be beneficial to state in the Introduction. Currently, it is unclear what the difference is between the two comparisons being proposed. I recommend that this be clearly communicated in the Introduction.
Figures & Tables
The figures and tables are clear and legible. There are no modifications to either the figures or tables.
Material and Methods
Page 9, Line 100: Please see the point about PRISMA guidance in the Abstract’s comments. This is not an appropriate use of this term. I recommend that this be revised.
Page 9, Line 107: The last search was not over 12 months old. I recommend that the authors update the search and incorporate study data that is eligible from April 2024 to the present day to ensure that their systematic review is contemporary.
Page 10, 116: I recommend that the authors provide further information on the definition of PNF. The number of different PNF techniques, including hold-relax/reciprocal inhibit, etc., and plyometric elements, may also have implications or references to PNF-type techniques. Accordingly, clearly reporting this would be beneficial. I recommend that the authors make an explicit statement in the eligibility criteria to avoid confusion.
Page 10, Line 120: The authors refer to routine treatment as the control; however, there is a suggestion that this is no physical activity or minimal. This is not routine treatment. I recommend that the authors provide further information on what their two comparators really were and make that explicit. Perhaps the RT comparator group should be the only real comparison unless the authors specify a waiting-list group comparison. Ultimately, I recommend that the authors be more clearly definite and select what their comparators are and provide justification in the paper for the selection of these characteristics or treatments in each group. Reference to baseline interventions in Line 124 further complicates this.
Line 131: There is no statement on what the secondary outcomes were. Change in pain score was the primary, but non-change score may have been a secondary, I suggest. I recommend that the authors provide more information on this to clarify the confusion in the methods section. Furthermore, it is unclear what the follow-up period was for each analysis. I recommend that the authors consider presenting the findings as short-term, mid-term, and longer-term to provide a more robust interpretation of clinical effect.
Line 142: I recommend that the authors correct K/L to Kellgren Lawrence and provide a citation to support this term.
Line 145: I recommend that the authors remove data where it was not possible to attribute the results to one knee over another. There is a suggestion in this sentence that the data are aggregated. This should be reviewed. If this approach is adopted, a citation to support this approach must be offered to increase the confidence that this is acceptable.
Line 161: I recommend that the authors state who performed the GRADE assessment. This would provide greater clarity. This is clearly communicated for the Risk of Bias assessment and should be offered for the GRADE assessment as well.
Line 167: The authors did not sufficiently assess the study or the methodological heterogeneity of the data. This should be assessed by the study design, methodology, and population and intervention characteristics. This is required to justify the use of the meta-analysis. I recommend that this be clearly communicated in the Statistical Analysis section. Currently, I am concerned that, based on Table 1, there is too much variability in the control group comparator and in the PNF procedures/interventions to permit meta-analysis, and a narrative analysis would be more appropriate. I recommend that the authors review their study/methodological heterogeneity and consider whether this suggestion is founded or not. This has been highlighted for the RT group, although further information on this could be offered. This is not sufficiently communicated to the control group, and I recommend that this be addressed.
Results
Results, Page 13, Lines 212-215: I was surprised that there was a limited geographical spread for the included studies. How confident are the authors that they have sufficiently searched the evidence base to include all studies? This is a concern that should be considered by the authors when reviewing their search strategy and potentially in the Discussion’s Limitations section.
Page 14, Line 239: The risk of bias results are clearly presented in Section 3.2. I have nothing further to add.
Page 15, Line 256: I recommend that the authors refer to the Cochrane handbook and present GRADE for each outcome. I was surprised to see an overarching summary of GRADE. GRADE assesses individual analyses and, therefore, linking the result of GRADE to the analysis approach would be more appropriate. I also recommend that the authors consider how they use GRADE for non-meta-analysis results, as this methodological approach is slightly different from the approach adopted.
Page 15, Lines 261 to Page 17, Line 312: I am concerned regarding the meta-analysis. Please see the comment above regarding the methodological heterogeneity.
Page 15, Line 261: The authors refer to pain score in the Methods as ‘change’ in pain score, but the results presented are not consistently ‘change’ score, but also non-change score. I recommend that this be clarified for improved interpretation.
Page 15, Line 267: I recommend that the authors reflect on how SMD does not relate to MCID in interpretation. Weighted MD may offer this. The interpretation of MCID in this line without a mean difference figure is therefore difficult to comprehend.
Page 16, Line 272: Because of the relatively small number of studies, I am unsure how valuable the sensitivity analysis is. I recommend that the authors reconsider this. If they elect to keep the sensitivity analysis in the paper, I recommend that the small number of studies included should be considered as a limitation in the Discussion’s Limitations section.
Page 16, Line 281-290: I agree with the narrative approach to reporting the data from the RT comparisons. However, I recommend that the authors consider whether they can present the results with effect size measures, which may make the interpretation a little clearer. This point remains for assessing knee range of motion on Page 17, Lines 304 to 312.
Discussion
Discussion, Page 17, Line 321: The authors should be mindful that the findings of an improvement in pain and knee movement between the ‘control’ group and PNF may not necessarily be surprising. There are systematic reviews that indicate that exercise per se offers clinical benefit to people with knee osteoarthritis over non-exercise/non-intervention comparisons. Therefore, as PNF is not compared in this analysis to another exercise comparison, the findings of benefit may be as expected. I recommend that this be considered by the authors to caveat the interpretation.
Page 18, Line 350: The evidence suggests no superiority of PNF over other treatments. There is no evidence provided in this interpretation to suggest why PNF should be used over other prescribed exercises. Accordingly, I recommend that the authors review the text in this paragraph and ensure that a balanced interpretation is offered rather than the implication that PNF offers something beneficial and should be used over other treatments. There is insufficient evidence to suggest this. This is also the case for Page 19, Line 360 to 365, where there is an over-interpretation of the data in favor of PNF. I recommend that this be reconsidered in the paper.
Page 17-18, Lines 321 and 337: I recommend that the authors use the GRADE findings in this interpretation. There is insufficient reflection that the evidence base is poor for this systematic review, and a more cautious handling of the evidence is needed.
Page 19, Line 370-371: I recommend the authors consider the role of stretching/lengthening and recruitment for strength/power. There is limited exploration on how PNF may modify both pain and range of motion from a physiological effect. This may support some assumptions made in this paragraph. However, I recommend that the authors guard against making overstated claims on the results, given the findings of the systematic review.
Page 20, Line 387: I recommend that the authors consider whether they mean effectiveness or efficacy in this sentence.
Page 20, Line 400: The issue of the length of data collection and follow-up could be overcome by the authors by reporting outcomes at specific time-points. I recommend that the authors consider this.
Page 20, Line 403: The fourth limitation undermines the use of the meta-analysis for the control group comparator. I recommend that the authors reconsider this and decide whether a meta-analysis is appropriate in this paper.
Conclusion
Conclusion, Page 21, Line 417: I recommend the authors review the current conclusion and consider if a clearer interpretation can be made and whether there needs to be more explicit reporting of the GRADE outcomes to emphasize the limited evidence base.
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.