Radiographic Evolution of Contralateral Asymptomatic Incomplete Atypical Femoral Fractures in Autoimmune Disease Patients

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is a well executed retrospective study with novel clinical insights.

I suggest only a few minor revisions:

• Statistical analysis: multivariate analysis yielded no significant findings. Discuss why this approach was chosen over descrptive statistics only for 10 cases. Alternatively, acknowledge that the cohort size precludes multivariate modeling and focus only on descriptive patterns.
• Conclusions about teriparatide remain speculative. Only 6 of 10 patients received it; the comparison (3 improved without it vs. imprved/worsening with it) lacks rigor. Relativize this to these results and to the literature. 

• Table 2 reports pseudarthrosis in 4/10 operative sides, with three requiring revision surgery. However, the relationship between operative-side healing complications and contralateral lesion behavior is not analyzed. Consider explring briefly whether patients with pseudarthrosis or revision surgery on the operative side showed different patterns of contralateral lesion progression. 

Author Response

Response to Reviewer 1

We sincerely thank Reviewer 1 for the positive evaluation of our work and for the constructive suggestions. We are encouraged by the assessment that this study provides novel clinical insights, and we appreciate the opportunity to further clarify and refine several aspects of the manuscript. Our point-by-point responses are provided below. Modifications and additions, including opinions from other reviewers, are highlighted with a yellow marker.

 

Comment 1

Statistical analysis: multivariate analysis yielded no significant findings. Discuss why this approach was chosen over descriptive statistics only for 10 cases, or alternatively acknowledge that the cohort size precludes multivariate modeling.

Response:
We appreciate this important methodological comment. We agree that, given the small cohort size, multivariate analysis has inherent limitations and that its results should be interpreted cautiously.

The rationale for including multivariate analysis was exploratory rather than confirmatory. Specifically, we sought to examine whether any consistent directional associations might emerge between lesion behavior and key biological or treatment-related variables (e.g., duration of BP/GC use, teriparatide administration), even in the absence of statistical significance. This approach was intended to complement, rather than replace, the primary descriptive and longitudinal radiographic analyses.

To address the reviewer’s concern, we have revised the manuscript to explicitly acknowledge that the cohort size precludes robust multivariate modeling and that the analysis should be considered hypothesis-generating only. Greater emphasis has been placed on descriptive radiographic patterns and individual lesion trajectories, which constitute the core findings of this study.

Based on the above, I have added descriptions related to this in the Methods (statistical descriptions) and Discussion sections.

The line numbers in the new version will be from 165 to 170 and from 340 to 344.

 

 

Comment 2

Conclusions about teriparatide remain speculative. Only 6 of 10 patients received it; the comparison lacks rigor. Relativize this to these results and to the literature.

Response:
We fully agree with this assessment. The original wording may have overstated the implications regarding teriparatide.

In the revised manuscript, we have deliberately softened the language to clarify that:

• the observed lesion stabilization or regression cannot be attributed causally to teriparatide based on this dataset, and
• the findings merely suggest biological plausibility rather than therapeutic efficacy.

We now explicitly state that bisphosphonate discontinuation alone may have contributed to lesion stabilization in some cases and that the role of teriparatide remains inconclusive in this cohort. References to teriparatide have been reframed to align with existing literature while clearly acknowledging the speculative nature of any treatment-related inference drawn from the present data.

 

Based on the above, I have partially revised the sentences in the Results section. In addition, supplementary explanations have been added to the Discussion and Conclusion sections.

The line numbers in the new version will be from 205 to 210, from 350 to 354 and from 367 to 369.

 

 

Comment 3

Table 2 reports pseudarthrosis in 4/10 operative sides. The relationship between operative-side healing complications and contralateral lesion behavior is not analyzed. Consider exploring this briefly.

Response:
We thank the reviewer for this insightful observation. We explored this relationship descriptively.

Among the four patients who developed pseudarthrosis on the operative side, contralateral lesion behavior did not demonstrate a consistent or distinct pattern compared with patients who achieved uneventful union. Specifically, contralateral lesions in this subgroup showed both regression and stability, and none progressed to complete fracture. Conversely, the single case of radiographic progression of the contralateral incomplete lesion occurred in a patient without confirmed nonunion-related mechanical overload at the time of observation.

Given the very small number of cases and the absence of a clear trend, formal statistical comparison was not feasible. However, we have added a brief descriptive statement in the Results/Discussion section noting that operative-side healing complications did not appear to systematically influence contralateral lesion behavior in this cohort.

Based on the above, I have added descriptions related to this in the Methods (statistical descriptions) and Discussion sections.

The line numbers in the new version will be from 220 to 225 and from 322 to 325.

 

Final remark

We sincerely thank Reviewer 1 for these thoughtful and clinically relevant comments. The revisions prompted by these suggestions have improved the methodological transparency and balanced interpretation of our findings.

 

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for inviting me to review this paper.

First of all, my concern is that the title is lengthy and distracting; please consider revising it to be more focused and engaging.

Second, line 93 states that five radiographic features would be evaluated; however, the radiographic evaluation described in lines 129–134 slightly deviates from the features listed earlier. Please clearly justify and reconcile this discrepancy.

Third, at line 139, since independent image assessments were conducted, readers would expect reporting of inter-rater agreement statistics (e.g., kappa coefficients). This would not contradict the current workflow in which a gold-standard evaluation is ultimately established by consensus.

Fourth, at line 155, the authors state that no thigh pain or new symptoms were identified during observation, yet these patients eventually underwent surgical treatment, which supports the value of radiographic analysis for asymptomatic incomplete AFF. However, as currently written, it reads as though thigh pain or “new symptoms” are not indicators for surgery in complete AFF, which does not imply that the patients were entirely stable and then underwent surgery suddenly. Please revise this wording for clinical clarity and logical consistency.

Fifth, while a significance threshold is stated (line 148), there appears to be only one significance-related result reported (line 196), which is described as not significant, and no p-value is provided. This does not support the stated aim of reducing heterogeneity (line 81) to generate more focused results. In its current form, the work reads more like a case series, which may be more suitable for a clinically oriented journal.

Other comments:

1. Line 81: you state that the design “reduces heterogeneity,” which may be reasonable given the narrowed inclusion criteria, but please provide at least one supporting reference indicating that heterogeneity is a known issue in this context.
2. Line 89: please clarify whether all 22 included patients were consecutive and comprehensively captured during the study period to mitigate concerns about selection bias.
3. Page 5: repeating both age and sex for all patients in both Tables 1 and 2 seems unnecessary; please consider removing redundancy

Author Response

Response to Reviewer 2

We sincerely thank Reviewer 2 for the careful and constructive review of our manuscript. We greatly appreciate the insightful comments, which have helped us to improve the clarity, methodological transparency, and clinical interpretation of our work. Our point-by-point responses are provided below. Modifications and additions, including opinions from other reviewers, are highlighted with a yellow marker.

 

Comment 1

The title is lengthy and distracting; please consider revising it to be more focused and engaging.

Response:
We agree with the reviewer that the original title was overly long and may reduce readability. The title has been revised to be more concise while preserving the essential elements of the study population, disease entity, and diagnostic focus. Specifically, redundant descriptors were removed, and emphasis was placed on the key clinical issue of radiographic evolution of asymptomatic contralateral incomplete AFF.

Revision:
The title has been shortened and refined in the revised manuscript.

Based on the above, the title has been changed as follows.

Radiographic Evolution of Contralateral Asymptomatic Incomplete Atypical Femoral Fractures in Autoimmune Disease Patients

 

Comment 2

Line 93 states that five radiographic features would be evaluated; however, the evaluation described in lines 129–134 deviates from the features listed earlier. Please justify and reconcile this discrepancy.

Response:
We appreciate this important observation. The apparent discrepancy arose from an imprecise description in the original text rather than a difference in actual evaluation criteria.

In the Introduction, the five major ASBMR diagnostic features were referenced to define AFF diagnostically. In contrast, the radiographic assessment in the Methods section focused specifically on longitudinal changes in key morphological markers relevant to progression of incomplete lesions, namely cortical beaking, periosteal flaring, and transverse radiolucent lines.

To clarify this distinction, we have revised the Methods section to explicitly state that:

• ASBMR criteria were used strictly for diagnostic inclusion, and
• a predefined subset of radiographic features was selected a priori for serial evaluation of lesion progression or regression.

This revision reconciles the diagnostic framework with the analytical focus of the study and improves conceptual consistency.

Based on the above, I have added descriptions related to this in the Materials and Methods section.

The line numbers in the new version will be from 135 to 140.

 

Comment 3

Independent image assessments were conducted; readers would expect reporting of inter-rater agreement statistics (e.g., kappa coefficients).

Response:
We acknowledge that reporting inter-rater agreement statistics such as kappa coefficients is desirable in imaging-based studies. However, in the present study, the primary purpose of multiple independent assessments was not to quantify observer variability but to minimize individual bias prior to establishing a final consensus-based gold-standard classification of lesion behavior.

Given the very small sample size (10 limbs) and the categorical nature of lesion evolution (progression, stability, regression), calculation of kappa statistics would be statistically unstable and potentially misleading. Importantly, no substantial disagreements occurred that altered final lesion classification.

To address this point transparently, we have revised the Methods section to explicitly state the rationale for consensus-based evaluation and the limitations of formal inter-rater statistics in this context.

Based on the above, I have added descriptions related to this in the Materials and Methods section.

The line numbers in the new version will be from 151 to 157.

 

Comment 4

The wording at line 155 suggests that thigh pain or new symptoms are not indicators for surgery in complete AFF. Please revise for clinical clarity and logical consistency.

Response:
We thank the reviewer for identifying this important issue of clinical interpretation. We agree that the original wording could be misconstrued.

Our intention was to emphasize that radiographic progression of incomplete AFF can occur in the absence of symptoms, not to suggest that symptoms are irrelevant for surgical decision-making in complete AFF.

The text has been revised to clarify that:

• none of the contralateral incomplete lesions became symptomatic during observation, and
• surgical treatment decisions in this cohort were driven by radiographic findings and known high-risk morphology, rather than sudden clinical deterioration.

This revision ensures clinical accuracy and avoids unintended implications regarding standard indications for surgery.

Based on the above, the corrections below were substituted in the Results section.

Additionally, the following statements were added to the Results and Discussion section.

In the new version, the former is located from line 179 to line 183, and the latter is located from line 331 to line 334.

 

Comment 5

Only one significance-related result is reported and is not significant; no p-value is provided. The work reads more like a case series, which may be more suitable for a clinically oriented journal.

Response:
We appreciate this thoughtful critique. We acknowledge that, due to the small cohort size, the statistical power of multivariate analysis was limited and that no statistically significant predictors were identified.

However, the primary aim of this study was not hypothesis testing, but rather:

• to reduce biological heterogeneity by focusing on a uniquely homogeneous high-risk population, and
• to describe the long-term radiographic behavior of asymptomatic contralateral incomplete AFF under standardized clinical conditions.

To address the reviewer’s concern, we have:

• explicitly stated in the revised manuscript that statistical analyses were exploratory and hypothesis-generating, and
• clarified that the value of this study lies in its detailed longitudinal imaging assessment rather than statistical inference.

We respectfully submit that Diagnostics, as a journal emphasizing diagnostic methodology and imaging-based decision-making, is an appropriate venue for this focused radiographic analysis, even though the study also shares characteristics with a descriptive case series.

Based on the above, I have added descriptions related to this in the Materials and Methods (statistical descriptions), Results and Discussion sections.

The line numbers in the new version will be from 165 to 170, from 229 to 232 and from 340 to 344, from 350 to 354.

 

Other Comments

Comment 1

Line 81: you state that the design “reduces heterogeneity,” but please provide at least one supporting reference indicating that heterogeneity is a known issue in this context.

Response:
We appreciate this important point. We agree that heterogeneity in AFF populations—particularly with respect to underlying disease, medication exposure, and biomechanical factors—is a well-recognized challenge in interpreting radiographic progression and treatment outcomes.

Rather than adding new references, we have revised the text to clarify that this statement is supported by existing references already cited in the manuscript, which describe variability in AFF behavior across osteoporotic and non-osteoporotic populations, differences in pharmacologic exposure, and inconsistent progression patterns (e.g., ASBMR Task Force reports and prior contralateral AFF studies).

The revised wording emphasizes that our study addresses this known issue by intentionally restricting inclusion to a biologically homogeneous subgroup, thereby minimizing confounding variability without introducing additional citations.

Based on the above, the last statement of the Introduction has been replaced with the following wording.

In the new edition, it corresponds to lines 80 to 85.

Comment 2

Line 89: please clarify whether all 22 included patients were consecutive and comprehensively captured during the study period to mitigate concerns about selection bias.

Response:
Thank you for highlighting this point. We agree that clarification regarding case capture is essential.

We confirm that all patients who underwent surgical treatment for complete AFF at our institution during the study period were consecutively identified from institutional surgical records. No cases meeting the ASBMR diagnostic criteria were intentionally excluded at the initial screening stage.

We have revised the Methods section to explicitly state that the cohort represents a consecutive series and that subsequent exclusions (e.g., bilateral simultaneous fractures, inability to evaluate the contralateral femur due to implants, or absence of radiographic incomplete lesions) were applied systematically according to predefined criteria. This clarification aims to mitigate concerns regarding selection bias.

Based on the above, I have revised the beginning of the Study Population and Case Selection section in the Materials and Methods. Additionally, I have added supplementary explanations within the same section. This corresponds to lines 90 to 92 and lines 103 to 107 in the new version.

Comment 3

Page 5: repeating both age and sex for all patients in both Tables 1 and 2 seems unnecessary; please consider removing redundancy.

Response:
We agree with the reviewer that this information was redundant. Age and sex are now presented only once, and the tables have been revised to remove unnecessary repetition while preserving clarity and completeness. This change improves readability and avoids duplication.

 

Final remark

We are grateful to Reviewer 2 for the insightful and clinically relevant comments. We believe that the revisions made in response have substantially improved the clarity, rigor, and interpretability of the manuscript.

Reviewer 3 Report

Comments and Suggestions for Authors

Your manuscript presents a retrospective longitudinal radiographic study evaluating the natural history of asymptomatic contralateral incomplete atypical femoral fractures (AFF) in a biologically homogeneous cohort of autoimmune disease patients receiving long-term bisphosphonates and glucocorticoids. The topic is clinically relevant, and the focus on a high-risk, understudied population adds value. The long follow-up period and standardized radiographic evaluation strengthen the work.

The manuscript is well written, logically structured and it addresses an important clinical debate: whether all asymptomatic incomplete AFFs require prophylactic fixation. The findings are consistent with prior literature and reinforce existing risk stratification concepts.

Author Response

Response to Reviewer 3

We sincerely thank Reviewer 3 for the thoughtful and encouraging evaluation of our manuscript.

We are grateful for the recognition of the clinical relevance of this study, particularly the focus on asymptomatic contralateral incomplete AFF in a biologically homogeneous, high-risk autoimmune disease population. We also appreciate the reviewer’s acknowledgment of the long-term follow-up, standardized radiographic assessment, and the contribution of this work to the ongoing clinical debate regarding the indications for prophylactic fixation.

We are pleased that the reviewer found the manuscript to be well structured, consistent with prior literature, and supportive of existing risk stratification concepts. The constructive feedback provided by all reviewers has helped us further refine the clarity and balance of our interpretations, and we believe that the revised manuscript is strengthened as a result.

Thank you again for your positive and insightful comments.

 

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for revising this manuscript. I appreciate the authors’ careful and thorough responses to the comments raised in the previous round.

A few minor issues remain:

1. The font size in Table 1 is too small and should be reorganized to improve readability; the editorial team may have additional suggestions in this regard.

2. Line 158 appears to be incomplete.

3. There appears to be an inconsistency in the author initials: an author is listed as “YM,” while “MY” appears in line 380. Please verify and correct this as appropriate.

Author Response

Response to Reviewer 2

Thank you very much for reviewing this manuscript. We sincerely appreciate your careful attention to detail, which has been extremely helpful in improving the quality and clarity of our work. We have revised the manuscript in accordance with your comments, as detailed below. All modifications and additions, including those made in response to comments from other reviewers, are highlighted in yellow in the revised manuscript.

 

Comment 1

The font size in Table 1 is too small and should be reorganized to improve readability; the editorial team may have additional suggestions in this regard.

Response:
Thank you for this helpful suggestion. Table 1 has been reorganized into a vertical format, and the font size has been increased to improve readability. We will be happy to make further adjustments in accordance with any additional recommendations from the editorial team.

 

Comment 2

Line 158 appears to be incomplete.

Response:
Thank you for pointing this out. The sentence contained an unintended fragment resulting from the inclusion of partial results-related text. This fragment has been removed to ensure clarity and completeness.

 

Comment 3

There appears to be an inconsistency in the author initials: an author is listed as “YM,” while “MY” appears in line 380. Please verify and correct this as appropriate.

Response:
Thank you for identifying this inconsistency. The initials have been corrected to “Y.M.” throughout the manuscript.