Propensity Score-Matched Analysis of Endovascular Treatment and Microsurgery for Unruptured Middle Cerebral Artery Aneurysms: Long-Term Outcomes over 6-Year Follow-Up

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This study analyzes retrospectively collected single-center data to identify prognostic factors for unruptured MCA aneurysms and compares outcomes between microsurgical and endovascular treatment using propensity score matching (PSM). However, there are fundamental methodological limitations that substantially undermine the validity of the conclusions.

1. The number of poor-outcome events is extremely small (approximately 10 cases), which is insufficient for conducting multivariable analyses such as logistic regression or PSM.
   As a result, the estimated odds ratios are unstable, confidence intervals are excessively wide, and some variables show complete separation. The external validity and reliability of the statistical results are therefore very limited.
2. The authors conclude that no significant difference exists between microsurgery and endovascular treatment after PSM.
   However, the matched sample size is only 28 vs. 28, and PSM in an observational, single-center cohort cannot support claims of equivalence or comparative effectiveness. At most, the results may suggest that treatment selection at this institution was reasonable, but they cannot be generalized beyond this context.

Given these limitations, the main conclusions of the manuscript are not sufficiently supported by the available data or statistical methodology. 

Author Response

This study analyzes retrospectively collected single-center data to identify prognostic factors for unruptured MCA aneurysms and compares outcomes between microsurgical and endovascular treatment using propensity score matching (PSM). However, there are fundamental methodological limitations that substantially undermine the validity of the conclusions.

1. The number of poor-outcome events is extremely small (approximately 10 cases), which is insufficient for conducting multivariable analyses such as logistic regression or PSM.
   As a result, the estimated odds ratios are unstable, confidence intervals are excessively wide, and some variables show complete separation. The external validity and reliability of the statistical results are therefore very limited.

Reply: We sincerely thank the reviewer for this important and thoughtful observation. We fully agree that the small number of poor-outcome events limits the robustness of multivariable analyses, including logistic regression and propensity score–based approaches, and may result in unstable estimates, wide confidence intervals, and potential separation issues.

In response, we have substantially expanded the Limitations section to explicitly address these concerns, including the restricted statistical power, limited reliability of effect estimates, and the lack of external validity. In addition, we have revised the Conclusions to ensure that our interpretations are appropriately cautious and clearly framed as hypothesis-generating rather than definitive.

We hope that these revisions adequately address the reviewer’s concerns and improve the transparency and rigor of the manuscript. We are grateful for the reviewer’s expertise and would welcome any further guidance or suggestions.

 

2. The authors conclude that no significant difference exists between microsurgery and endovascular treatment after PSM.
   However, the matched sample size is only 28 vs. 28, and PSM in an observational, single-center cohort cannot support claims of equivalence or comparative effectiveness. At most, the results may suggest that treatment selection at this institution was reasonable, but they cannot be generalized beyond this context.

Reply: We thank the reviewer for this insightful comment. We fully agree that the small matched sample size and the use of propensity score matching in a single-center, observational cohort do not permit conclusions regarding treatment equivalence or comparative effectiveness between microsurgical and endovascular approaches.

In response, we have revised the Conclusions to avoid claims of equivalence and to more accurately reflect the exploratory nature of the analysis. The findings are now framed as suggesting that treatment selection within our institution resulted in comparable observed outcomes in the matched cohort, rather than implying generalizable or definitive comparative effectiveness. We have also emphasized that these results are context-specific and should not be extrapolated beyond this institutional experience.

We believe these revisions align the conclusions more closely with the study design and statistical limitations, and we appreciate the reviewer’s careful assessment, which has strengthened the manuscript.

 

Given these limitations, the main conclusions of the manuscript are not sufficiently supported by the available data or statistical methodology. 

Reply: We thank the reviewer for this overarching assessment. We agree that, given the stated limitations, the original conclusions were not sufficiently supported by the available data and statistical methodology. In response, we have substantially revised the Limitations and Conclusions sections to ensure that all statements are fully aligned with the study design and analytical constraints.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

I would like to express my deepest appreciation to the authors for their hard work in compiling and analyzing such valuable clinical data over the past six years. In particular, the achievement of a "Good Clinical Outcome" (mRS 0-2) in 93.4% of cases is highly impressive and serves as a very encouraging result.

However, as a reviewer, I would like to point out one aspect regarding the interpretation of this study that may require reconsideration.

Determining the treatment method for cerebral aneurysms relies heavily on the operator's clinical judgment, making Randomized Controlled Trials (RCTs) practically difficult to implement. Consequently, retrospective studies are the norm, but they inevitably carry "Selection Bias" and have the limitation of being unable to quantitatively control for the most critical variable: the "Operator."

Although this study utilized Propensity Score Matching (PSM) to adjust for these issues, the matching variables presented in Figure 2 appear insufficient to fully reflect the complex decision-making processes of actual clinical practice. From the perspective of a cerebrovascular surgeon, factors other than the variables presented likely exert a greater influence on treatment decisions, suggesting a fundamental limitation of PSM in this context.

Specifically, it is highly probable that the following "Unmeasured Confounders" were not reflected in the PSM analysis:

• High Expertise: The center where this study was conducted demonstrates a high success rate of 93.4%. This implies that the medical team's "Tacit Knowledge" and extensive experience played a significant role in both treatment decisions and outcomes.
• Management of Complications: The fact that the final prognosis remains excellent despite the incidence of complications in surgery (15.3%) and endovascular procedures (10.7%) is evidence of the team's exceptional "Rescue Capability." While this ability is a key determinant of prognosis, it was not included in the statistical model.
• Bias in Shared Decision Making: Even when respecting patient preferences, if a physician judges a specific treatment to be safer or more feasible, there is a possibility of "Steering" the patient toward that method during the explanation process. Conversely, it is statistically difficult to isolate cases where a specific treatment was performed due to a patient's strong preference despite high medical complexity.

These factors may act as bias in the study results, potentially limiting the Generalizability of the conclusions. However, if interpreted from a different angle, one could derive the conclusion that "if a skilled medical team provides optimal treatment tailored to individual patient characteristics, the competence of the medical staff is the core determinant of prognosis, rather than the treatment modality itself."

Thank you once again for sharing these meaningful research results. I look forward to this study serving as an important reference for the establishment of future cerebral aneurysm treatment guidelines.

Author Response

I would like to express my deepest appreciation to the authors for their hard work in compiling and analyzing such valuable clinical data over the past six years. In particular, the achievement of a "Good Clinical Outcome" (mRS 0-2) in 93.4% of cases is highly impressive and serves as a very encouraging result.

Reply: We sincerely thank the reviewer for this kind and encouraging comment. We appreciate the recognition of the effort involved in assembling and analyzing this long-term clinical dataset, and we are grateful for the positive assessment of the observed clinical outcomes.

 

However, as a reviewer, I would like to point out one aspect regarding the interpretation of this study that may require reconsideration.

Determining the treatment method for cerebral aneurysms relies heavily on the operator's clinical judgment, making Randomized Controlled Trials (RCTs) practically difficult to implement. Consequently, retrospective studies are the norm, but they inevitably carry "Selection Bias" and have the limitation of being unable to quantitatively control for the most critical variable: the "Operator."

Although this study utilized Propensity Score Matching (PSM) to adjust for these issues, the matching variables presented in Figure 2 appear insufficient to fully reflect the complex decision-making processes of actual clinical practice. From the perspective of a cerebrovascular surgeon, factors other than the variables presented likely exert a greater influence on treatment decisions, suggesting a fundamental limitation of PSM in this context.

Specifically, it is highly probable that the following "Unmeasured Confounders" were not reflected in the PSM analysis:

• High Expertise: The center where this study was conducted demonstrates a high success rate of 93.4%. This implies that the medical team's "Tacit Knowledge" and extensive experience played a significant role in both treatment decisions and outcomes.
• Management of Complications: The fact that the final prognosis remains excellent despite the incidence of complications in surgery (15.3%) and endovascular procedures (10.7%) is evidence of the team's exceptional "Rescue Capability." While this ability is a key determinant of prognosis, it was not included in the statistical model.
• Bias in Shared Decision Making: Even when respecting patient preferences, if a physician judges a specific treatment to be safer or more feasible, there is a possibility of "Steering" the patient toward that method during the explanation process. Conversely, it is statistically difficult to isolate cases where a specific treatment was performed due to a patient's strong preference despite high medical complexity.

These factors may act as bias in the study results, potentially limiting the Generalizability of the conclusions. However, if interpreted from a different angle, one could derive the conclusion that "if a skilled medical team provides optimal treatment tailored to individual patient characteristics, the competence of the medical staff is the core determinant of prognosis, rather than the treatment modality itself."

Reply: We sincerely thank the reviewer for this thoughtful and nuanced analysis. We fully agree that, in the context of cerebrovascular disease, treatment selection is inherently influenced by operator judgment, institutional expertise, and factors that are difficult or impossible to quantify in retrospective analyses. As the reviewer correctly notes, while propensity score matching can partially adjust for measured confounders, it cannot account for unmeasured factors such as operator experience, tacit technical knowledge, rescue capability, or subtleties of shared decision-making.

In response to this comment, we have expanded the Limitations section to explicitly acknowledge these unmeasured confounders and to emphasize the fundamental constraints of PSM in reflecting real-world clinical decision-making. We have also further tempered the Conclusions to highlight the limited generalizability of our findings and to clarify that the observed outcomes likely reflect the performance of a highly experienced multidisciplinary team rather than the intrinsic superiority of a specific treatment modality.

We appreciate the reviewer’s insightful perspective, which has helped refine the interpretation of our results and strengthen the conceptual framing of the manuscript.

 

Thank you once again for sharing these meaningful research results. I look forward to this study serving as an important reference for the establishment of future cerebral aneurysm treatment guidelines.

Reply: We sincerely thank the reviewer for these encouraging comments. We are grateful for the thoughtful feedback, which has helped us improve the clarity and rigor of our manuscript, and we hope that the study can contribute meaningfully to future considerations in cerebral aneurysm treatment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

a clear, well-structured, and genuinely useful comparison of microsurgical clipping and endovascular treatment for unruptured MCA aneurysms, strengthened by a long median follow-up and a thoughtful use of propensity score matching. It reads as the work of a team with real clinical experience, and the authors succeed in presenting a balanced, nuanced picture rather than pushing one modality over the other. The high rate of good functional outcomes across the cohort reinforces the impression of solid institutional practice and gives the paper a reassuring clinical relevance.

 

For improvement, a few areas would benefit from refinement to make the narrative more coherent and clinically digestible. The definition of complications is very broad and mixes procedure-related events with issues of uncertain or indirect relation to treatment; separating these categories—at least conceptually—would make the results more interpretable without sacrificing the authors’ patient-centred ethos. The extraordinarily high proportion of favourable mRS outcomes also deserves a touch more reflection, particularly regarding selection bias or centre expertise, so that readers can better understand to whom these findings truly apply. Likewise, the regression outputs showing an odds ratio of zero for distal aneurysm locations should be handled with more caution, as this almost certainly reflects model separation rather than an absolute protective effect. Finally, the survival analysis, while nicely executed, is underpowered by the small endovascular sample; a brief acknowledgement of this limitation—and perhaps a comment on radiological durability or retreatment rates—would round out the long-term perspective. Overall, the manuscript is strong, but these clarifications would make it read more confidently and resonate better with a broad neurosurgical and neurointerventional audience.

Author Response

a clear, well-structured, and genuinely useful comparison of microsurgical clipping and endovascular treatment for unruptured MCA aneurysms, strengthened by a long median follow-up and a thoughtful use of propensity score matching. It reads as the work of a team with real clinical experience, and the authors succeed in presenting a balanced, nuanced picture rather than pushing one modality over the other. The high rate of good functional outcomes across the cohort reinforces the impression of solid institutional practice and gives the paper a reassuring clinical relevance.

Reply: We sincerely thank the reviewer for these generous and encouraging comments. We greatly appreciate the recognition of our efforts to provide a balanced, clinically meaningful comparison, and we are pleased that the presentation and follow-up of the cohort were found to contribute to the relevance of our work.

 

For improvement, a few areas would benefit from refinement to make the narrative more coherent and clinically digestible. The definition of complications is very broad and mixes procedure-related events with issues of uncertain or indirect relation to treatment; separating these categories—at least conceptually—would make the results more interpretable without sacrificing the authors’ patient-centred ethos.

Reply: We thank the reviewer for this valuable suggestion and for recognizing our patient-centered approach. In response, we have refined the presentation of complications by adding additional categories, including procedure-related events, to enhance clarity and interpretability while maintaining a patient-focused perspective.

 

The extraordinarily high proportion of favourable mRS outcomes also deserves a touch more reflection, particularly regarding selection bias or centre expertise, so that readers can better understand to whom these findings truly apply.

Reply: We thank the reviewer for this insightful comment. We agree that the unusually high proportion of favourable functional outcomes warrants careful reflection, particularly with respect to selection bias and center-specific expertise. As a high-volume referral center with extensive cerebrovascular experience, institutional expertise likely influenced both treatment selection and outcomes; this has now been explicitly addressed in the Limitations section.

At the same time, we note that our observed rates of good functional outcome are broadly consistent with, and in some reports comparable to or lower than, those described in the existing literature, as already discussed in the manuscript.

 

Likewise, the regression outputs showing an odds ratio of zero for distal aneurysm locations should be handled with more caution, as this almost certainly reflects model separation rather than an absolute protective effect.

Reply: We thank the reviewer for this important comment. We agree that the observed odds ratio of zero for distal aneurysm location most likely reflects model separation related to the small number of outcome events, rather than a true or absolute protective effect. In response, we have expanded the Limitations section to more explicitly discuss this statistical issue and its implications for interpretation.

We have retained this finding in the revised manuscript, but have revised the corresponding text to ensure it is interpreted with appropriate caution and framed as an observation consistent with prevailing surgical practice, rather than as evidence of a definitive or causal association.

 

Finally, the survival analysis, while nicely executed, is underpowered by the small endovascular sample; a brief acknowledgement of this limitation—and perhaps a comment on radiological durability or retreatment rates—would round out the long-term perspective.

Reply: We thank the reviewer for this helpful comment. We agree that the survival analysis is limited by the relatively small endovascular sample size, which reduces statistical power. This limitation has now been explicitly acknowledged in the Limitations section of the revised manuscript.

In addition, to strengthen the long-term perspective, we have expanded the discussion to include comments on radiological durability and retreatment rates, where available. We believe these additions provide a more balanced interpretation of the long-term outcomes while remaining consistent with the constraints of the dataset.

 

Overall, the manuscript is strong, but these clarifications would make it read more confidently and resonate better with a broad neurosurgical and neurointerventional audience.

Reply: We sincerely thank the reviewer for this overall positive assessment. We are grateful for the thoughtful suggestions, which have helped us clarify key points and improve the scientific rigor and readability of the manuscript for a broad neurosurgical and neurointerventional audience.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have satisfactorily addressed the major concerns and appropriately reframed the conclusions as exploratory and institution-specific. Only minor editorial adjustments to wording are recommended.

Author Response

We thank the reviewer for this evaluation and for their guidance in revising our manuscript.