Review Reports - Sex-Specific Growth Rates of Ascending Thoracic Aortic Aneurysms in Non-Syndromic Patients: A Systematic Review

Round 1

Reviewer 1 Report (Previous Reviewer 2)

Comments and Suggestions for Authors

I am pleased to see that the authors have carefully revised this very valuable manuscript and have addressed the major points raised in the previous review. In particular, the revised version now provides a clearer definition of the studied population with explicit exclusion of syndromic cases and bicuspid aortic valve disease, a more transparent methodological description compliant with PRISMA 2020 and PROSPERO registration, a clear justification for not performing a pooled meta-analysis due to substantial heterogeneity and limited data availability, and an expanded discussion of imaging modalities and sources of measurement variability. These revisions substantially improve the clarity, methodological rigor, and overall scientific value of the review. Nevertheless, the revised manuscript still contains a few minor but important issues that should be corrected before final acceptance.
– There is an internal numerical inconsistency in Table 2, where the reported mean age in the “Total” row for Obel et al. is clearly discordant with the sex-stratified values and the age range of the screened population, suggesting a misplacement or erroneous value that requires correction.
– In the Results section, the reference attribution for Kauhanen et al. is incorrect, as the study reporting annual growth rates of 0.2 and −0.7 mm/year is misnumbered in the reference list and should be cited consistently throughout the manuscript.
– Several typographical and terminology-related errors are present in the tables and text, including misspellings in imaging modality names (e.g. echocardiography, computed tomography, magnetic resonance imaging) and inconsistent use of abbreviations and guideline acronyms.
– Minor formatting inconsistencies remain in Tables 2 and 3, such as spacing, decimal notation, and alignment of values across columns, which should be standardized to improve readability and editorial quality.
- I found cases as above, thus please revise you manuscript carefully to eliminate similar mistakes.
These issues are limited in scope and do not affect the main findings or conclusions of the study; however, addressing them will further strengthen the accuracy and presentation of the manuscript. Therefore, I recommend minor revision.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report (Previous Reviewer 1)

Comments and Suggestions for Authors

Paper with an interesting purpose

Methodologically sound, but there are too few heterogeneous studies to meet the meta-analysis's purpose.

The study does not provide conclusive evidence, but it is still a step forward.

Are there data on syndromic aTAAs? I would add a little background on genetics in the introduction.

The research methods are not questionable.

I would further clarify the correlation between aortic growth and aortic events.

With targeted reviews, the paper is suitable for the journal.

Comments on the Quality of English Language

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Review of ascending thoracic aortic aneurysms

Extensive but redundant supplementary materials; I don't think this section is useful. It's better to archive it and keep it available. Were only three studies included in the review, or is the flowchart incorrect? It looks to me like five studies were included.

Acronym HA for hypertension? I prefer HTN.

Isn't aTTA a difficult acronym? I understand it's the most commonly used acronym in the literature for ascending thoracic aortic aneurysms.

The graph emphasizing greater aneurysm growth in females could be improved, perhaps with other graphic solutions.

Shouldn't quality of study have been prioritized? In the methods?

I don't see metaregressions for smoking, follow-up time, hypertension, and diabetes.

Are there any outliers that make meta-analysis difficult?

Information on aortic dissection? Surgical procedures?

How do you address measurement bias with different methods?

The meta-analytic question is clinically relevant, but could it have been better developed as a meta-analysis?

What about publication year bias?

Author Response

Initial shared comment for Reviewer 1 and Reviewer 2

We sincerely thank both Reviewers for their thoughtful, detailed, and highly constructive comments. We particularly appreciate the shared focus on the clinical relevance of the research question, the interpretation of sex-specific growth rates, and the appropriateness of meta-analytic approaches in this context.

Several comments from both Reviewers address the absence of formal meta-analysis and meta-regression. We would therefore like to clarify our methodological rationale upfront. Although the clinical question is well suited to quantitative synthesis in principle, no pooled meta-analyses or meta-regression analyses were planned or performed in this study. This decision was made a priori and was based on methodological considerations rather than lack of clinical relevance.

The number of eligible studies was very limited (3 datasets from systematic review, 2 datasets from post-hoc contact), and the studies demonstrated substantial heterogeneity with respect to study design, patient selection, imaging modality, measurement protocols, follow-up duration, and reporting practices. Crucially, clinically relevant covariates - including blood pressure control, antihypertensive therapy, lipid profile, smoking status, diabetes, aortic valve hemodynamics, and longitudinal changes in medical treatment - were inconsistently reported and rarely stratified by sex across studies. Under these conditions, pooled effect estimates or meta-regression analyses would have been statistically underpowered and at high risk of producing misleading or non-reproducible results.

Review of ascending thoracic aortic aneurysms

Extensive but redundant supplementary materials; I don't think this section is useful. It's better to archive it and keep it available. Were only three studies included in the review, or is the flowchart incorrect? It looks to me like five studies were included.

Response

We thank the Reviewer for this comment, including the careful cross-check with the PRISMA flowchart.

We acknowledge that archiving parts of the supplementary materials is a valid alternative approach. However, we have deliberately chosen to retain all supplementary materials within the manuscript submission to ensure immediate and transparent access for readers, without requiring them to consult external repositories. Given the complexity of the eligibility assessment, the extensive number of excluded studies due to lack of sex-stratified reporting, and the development of a customized QoS evaluation tool, we believe that keeping the full supplementary material readily available strengthens reproducibility and allows interested readers to fully assess methodological decisions.

Regarding the number of included studies, a total of five studies were included in the final analysis. Three studies fulfilled the inclusion criteria through the predefined systematic literature search. Two additional datasets were obtained through direct contact with corresponding authors (of 11 contacted) after full-text screening, where relevant sex-stratified data had been collected but not reported in the original publications.

“In addition to the three studies included through the review process, 11 additional relevant studies were identified from the search results, and the corresponding authors were contacted. Relevant, in this context, indicates that these studies were deemed likely to have collected sex-specific data from participants meeting the inclusion criteria of this review, but the absence of stratified reporting prevented their inclusion. See Supplementary Materials S1, Table S1.4.1 for a comprehensive list of excluded studies, including the 11 studies which are in bold text. This process resulted in the additional inclusion of two datasets.”

This remains unedited in the revised manuscript: Methods – Data Extraction (lines 117-124)

“Following this, corresponding authors of 11 eligible studies were contacted, resulting in the additional inclusion of the two datasets by Fleury et al. [21] and Viitala et al. [22]. An overview of the studies included is provided in Table 1. A complete list of studies excluded after full-text review, including reasons for exclusion, is provided in Supplementary Materials S1, Table S1.4.1.”

This remains unedited in the revised manuscript: Results – Selection of Studies and Study Characteristics (lines 154-158)

We intentionally did not modify the classical PRISMA flowchart structure, as it is designed to reflect the results of the systematic database search alone. Including studies identified through post-hoc author contact directly in the flowchart could obscure this distinction.

Of course, should the Editor prefer a more concise supplementary section or an alternative presentation, we would be happy to revise the supplementary materials accordingly.

Acronym HA for hypertension? I prefer HTN.

Response

We agree with the Reviewer. The acronym HA has been replaced with HTN throughout the manuscript, tables, and supplementary materials for clarity and consistency with common clinical terminology.

Isn't aTTA a difficult acronym? I understand it's the most commonly used acronym in the literature for ascending thoracic aortic aneurysms.

Response

We agree that abbreviations within aortic pathology - not only for thoracic aortic aneurysms - are heterogenous across the literature, which can challenge clarity and comparability.

The chosen acronym aTAA (ascending thoracic aortic aneurysm) remains among the most frequently used in high-impact clinical and epidemiological studies. A recent work in the field has highlighted the need for more streamlined and standardized nomenclature across aortic diseases and provides a table with suggestions, including aTAA for ascending thoracic aortic aneurysms, precisely because multiple overlapping abbreviations have historically been applied inconsistently (https://doi.org/10.1016/j.xjon.2022.04.010). Thus, we have maintained aTAA in the manuscript for consistency with prevailing usage and to facilitate comparison with prior studies. We have also ensured that the full term is clearly defined at first mention and avoided unnecessary repetition of the acronym in narrative text where possible to aid readability.

The graph emphasizing greater aneurysm growth in females could be improved, perhaps with other graphic solutions.

Response

We appreciate the need to adapt Figure 3 to our argumentation that more studies showed a larger growth rate in females than in men. We rearranged the estimates to this end and show them study-by-study, contrasting the growth rates between sexes within each study more clearly. Thank you.

Shouldn't quality of study have been prioritized? In the methods?

Response

Although we are not sure what the Reviewer means by ‘prioritized’, we have now strengthened the beginning of Methods section to clarify that assessment of methodological quality and risk of bias was a central, apriori component of the review.

“This systematic review followed the 2020 Preferred Reporting Items for Systematic Review Checklist (PRISMA) [14], and the completed PRISMA 2020 Abstracts and expanded checklists are provided in Supplementary Materials S3. A preregistered protocol was published in PROSPERO for transparency and can be found via

https://www.crd.york.ac.uk/PROSPERO/view/CRD420251025890.

The methodological quality and risk of bias of included studies were assessed using a customized Quality of Studies (QoS) evaluation tool. The QoS score of each study was considered when interpreting the evidence and drawing conclusions.”

This has been added in the revised manuscript: Methods (lines 72-80)

I don't see metaregressions for smoking, follow-up time, hypertension, and diabetes. Are there any outliers that make meta-analysis difficult?

Response

Meta-regression was considered; however, it was not feasible due to:

The small number of included studies (n=5).
Substantial heterogeneity in variable definitions and reporting.
Lack of consistent, study-level covariate data (particularly sex-stratified estimates).

We have now clarified this limitation explicitly in the Statistical Analysis and Discussion sections to explain why meta-regression could not be performed reliably.

“Given the limited number of eligible studies and substantial methodological heterogeneity, no pooled meta-analysis or meta-regression analyses were planned or performed.”

This has been added to: Methods – Statistical analysis (lines 148-150)

“A notable limitation of this review is the absence of a meta-analysis, primarily due to the restricted number of eligible studies and the substantial heterogeneity among those included.”

This has been added to: Discussion (lines 408-410)

Information on aortic dissection? Surgical procedures?

Response

We acknowledge the clinical relevance of these outcomes.

However, information on aortic dissection and surgical intervention was sparsely and inconsistently reported across included studies and rarely sex-stratified. Moreover, most studies were not designed to evaluate clinical endpoints.

We have clarified this limitation in the Discussion and emphasized that the present review focuses specifically on aneurysm growth rates rather than downstream clinical events.

“Further limitations include the sparse and inconsistent reporting of clinical endpoints, such as aortic dissection and surgical intervention, across the included studies, with stratification by sex rarely performed. Additionally, the majority of studies were not specifically designed to assess clinical outcomes, thereby constraining the ability to correlate observed aneurysm growth rates with adverse aortic events. As a result, this review focuses on absolute aneurysm growth as the primary outcome, and the findings should not be interpreted as directly indicative of sex-specific differences in clinical risk.”

This has been to: Discussion (lines 410-416)

How do you address measurement bias with different methods?

Response

We agree that measurement bias related to different imaging modalities and measurement protocols is a critical concern when evaluating aortic growth.

In the present review, this issue was addressed through systematic documentation of imaging modality, ECG-gating, and measurement techniques for all included studies, allowing transparent identification of methodological heterogeneity. In addition, imaging methodology and measurement reproducibility were explicitly incorporated into the Quality of Studies (QoS) assessment to evaluate the potential impact of measurement-related factors on reported growth rates.

Importantly, no pooled meta-analyses were performed in this study (please revisit the initial shared comment to Reviewers). Inter-modality variability, lack of ECG-gating, and differences in measurement protocols were therefore not averaged across studies but instead discussed qualitatively as major sources of heterogeneity in the Discussion. These methodological considerations were one of the reasons for our decision to refrain from pooled quantitative synthesis, as combining estimates derived from heterogeneous measurement methods could falsely imply precision and obscure true variability.

“Differences in imaging modality, ECG-gating, and measurement protocols represent important sources of measurement variability in studies assessing ascending aortic growth. In the present review, substantial heterogeneity was observed across studies with respect to imaging techniques and technical standards, which may influence absolute growth estimates independently of biological factors. These methodological aspects were explicitly accounted for in the quality assessment and constitute an important non-biological source of inter-study heterogeneity. Importantly, no pooled meta-analyses were performed, and growth rates were therefore not averaged across heterogeneous measurement methods. Instead, differences were interpreted within their methodological context rather than as precise quantitative contrasts. This conservative approach was chosen to avoid false precision and to ensure that reported sex-specific growth rates are interpreted within the context of methodological as well as biological and clinical heterogeneity.”

This has been to: Discussion (lines 340-352)

The meta-analytic question is clinically relevant, but could it have been better developed as a meta-analysis?

Response

Please see the initial shared comment for Reviewer 1 and Reviewer 2.

What about publication year bias?

Response

We thank the reviewer for this relevant comment. The publication years of the included studies range from 2017 to 2025, representing a relatively narrow and contemporary time frame. Even the earliest study (Cheung et al., 2017) would be considered recent with respect to imaging availability and general clinical practice. As such, we do not believe that publication year alone is a major driver of bias in this review.

Nevertheless, we acknowledge that methodological standards and reporting practices have continued to evolve over time. Importantly, differences observed across studies appear to be more closely related to variations in imaging modality, ECG-gating, measurement protocols, and reporting of sex-stratified data rather than publication year per se. This perspective is now explicitly stated in the Discussion to contextualize inter-study heterogeneity.

Given the small number of included studies, formal assessment of publication year bias was not feasible. However, the relatively narrow publication window reduces the likelihood that temporal trends substantially influenced the main findings.

“ Regarding the relatively narrow and contemporary publication time frame (2017–2025) of the included studies, it could be argued that this reduces the likelihood of temporal variations in imaging technology or clinical practice substantially influencing the observed differences in growth rates.”

This has been to: Discussion - Limitations (lines 417-420)

Reviewer 2 Report

Comments and Suggestions for Authors

The authors address an important and clinically relevant topic and should be commended for a rigorous and transparent systematic review process, including PROSPERO registration, adherence to PRISMA guidelines, a careful quality-of-studies assessment, and the active retrieval of additional sex-stratified data from original investigators.
The focus on non-syndromic patients with tricuspid aortic valves fills a genuine gap in the literature. However, despite these strengths, the present manuscript essentially evaluates only the absolute dynamics of ascending aortic diameter growth, while the complex, multifactorial nature of aneurysm progression is insufficiently accounted for.
Key modifiable clinical and hemodynamic determinants—most notably blood pressure control (including achieved targets, measurement methodology, and antihypertensive treatment strategies), heart rate and rate-controlling therapy (particularly beta-blocker use and type), lipid profile and lipid-lowering treatment, and the presence of atherosclerotic or inflammatory vascular burden—are not systematically reported or analyzed across the included studies. In addition, potentially important contributors such as aortic valve hemodynamics (including the presence and progression of aortic regurgitation), baseline and longitudinal differences in medical therapy, and relevant iatrogenic or inflammatory exposures are not addressed.
As a consequence, the reported sex-specific growth rates likely reflect a composite of biological, therapeutic, and methodological factors rather than intrinsic sex-related differences alone. While some of these limitations are acknowledged in the Discussion, they are not sufficiently emphasized, and the conclusions risk overstating the clinical interpretability of absolute growth rates in isolation.
A more explicit and structured discussion of these confounding factors, clearer framing of growth rate as a surrogate rather than a direct risk marker, and a reframing of the conclusions in line with the available data are required. For these reasons, I recommend major revision prior to further consideration for publication.

Author Response

Initial shared comment for Reviewer 1 and Reviewer 2

The focus on non-syndromic patients with tricuspid aortic valves fills a genuine gap in the literature. However, despite these strengths, the present manuscript essentially evaluates only the absolute dynamics of ascending aortic diameter growth, while the complex, multifactorial nature of aneurysm progression is insufficiently accounted for.

Key modifiable clinical and hemodynamic determinants—most notably blood pressure control (including achieved targets, measurement methodology, and antihypertensive treatment strategies), heart rate and rate-controlling therapy (particularly beta-blocker use and type), lipid profile and lipid-lowering treatment, and the presence of atherosclerotic or inflammatory vascular burden—are not systematically reported or analyzed across the included studies. In addition, potentially important contributors such as aortic valve hemodynamics (including the presence and progression of aortic regurgitation), baseline and longitudinal differences in medical therapy, and relevant iatrogenic or inflammatory exposures are not addressed.

As a consequence, the reported sex-specific growth rates likely reflect a composite of biological, therapeutic, and methodological factors rather than intrinsic sex-related differences alone. While some of these limitations are acknowledged in the Discussion, they are not sufficiently emphasized, and the conclusions risk overstating the clinical interpretability of absolute growth rates in isolation.

A more explicit and structured discussion of these confounding factors, clearer framing of growth rate as a surrogate rather than a direct risk marker, and a reframing of the conclusions in line with the available data are required. For these reasons, I recommend major revision prior to further consideration for publication.

We thank the Reviewer for the thoughtful and comprehensive assessment of our work and for recognizing the rigor and transparency of the systematic review process, including PROSPERO registration, adherence to PRISMA guidelines, structured quality assessment, and the retrieval of additional sex-stratified data from original investigators. We also appreciate the acknowledgement that focusing on non-syndromic patients with tricuspid aortic valves addresses a genuine gap in the existing literature.

We fully agree that aneurysm progression - particularly in the ascending thoracic aorta - is a complex and multifactorial process. Importantly, however, the biological, hemodynamic, and pathological determinants of ascending thoracic aortic aneurysms remain far less well characterized than those of e.g. abdominal aortic aneurysms, where decades of experimental and clinical research have established clearer mechanistic frameworks. In this context, the present review was deliberately designed to address a narrowly defined and clinically relevant question: whether absolute growth rates of the ascending aorta differ between females and males in non-syndromic patients with tricuspid aortic valves.

With respect to the Reviewer’s comments on modifiable clinical and hemodynamic determinants, we agree that hypertension is highly prevalent (and relevant!) among patients experiencing ascending aortic events, particularly dissection. However, optimal blood pressure targets for patients with ascending thoracic aortic aneurysms are not well established, and the efficacy of specific antihypertensive strategies -most notably beta-blocker therapy - has recently been questioned in contemporary studies (https://doi.org/10.1161/JAHA.124.040149 ; DOI: 10.1136/openhrt-2022-002082). Conversely, diabetes mellitus has been associated with a lower risk of thoracic aortic dissection, illustrating that classical cardiovascular risk factors do not uniformly translate into aneurysm progression or adverse aortic events in the thoracic aorta.

Crucially, this systematic review focuses on aortic growth rate, not on adverse aortic events such as dissection, rupture, or surgical intervention. For growth of the ascending aorta, available evidence suggests that relatively few clinical variables show consistent associations with increased or reduced diameter progression. Indeed, several of the included studies evaluated traditional clinical factors - including blood pressure, smoking status, and diabetes - without identifying significant associations with growth rate. The largest of the included studies (Obel et al, N app. 2,000) found only sex to be significantly associated with ascending growth from multivariable modeling. Moreover, aortic valve morphology is a well-established determinant of ascending aortic growth; however, this factor was intentionally controlled for by restricting inclusion to patients with tricuspid aortic valves, thereby eliminating a major source of confounding by design.

The Reviewer also raises lipid profile, lipid-lowering therapy, and atherosclerotic or inflammatory vascular burden as potential contributors. Interestingly, multiple studies suggest that patients with ascending thoracic aortic aneurysms may exhibit a lower systemic atherosclerotic burden compared with individuals without such aneurysms, implying that mechanisms driving ascending aortic dilatation differ fundamentally from those underlying atherosclerotic vascular disease (DOI: 10.1159/000358123 ; DOI: 10.1016/j.jtcvs.2021.09.031). Consequently, lipid-related factors may be less relevant for aneurysm growth in the ascending aorta than in other vascular territories.

The narrow focus on sex-specific differences in growth rate was a deliberate methodological choice, intended to avoid speculative extrapolation to secondary endpoints or inclusion of inadequately reported confounders. We believe that maintaining this focused research question strengthens the validity and interpretability of the review rather than limiting it. Nonetheless, we agree that the reported sex-specific growth rates likely reflect a composite of biological, therapeutic, and methodological factors rather than intrinsic sex-related differences alone. While these limitations were already acknowledged, we have now expanded and restructured the Discussion to more explicitly emphasize the multifactorial nature of aneurysm progression, the limitations of absolute growth rate as an isolated measure, and the influence of methodological heterogeneity. In addition, we have more clearly framed growth rate as a surrogate marker rather than a direct indicator of clinical risk.

Importantly, we would like to emphasize that the conclusions of the review are intentionally cautious. The central conclusion is that the current literature - particularly high-quality, sex-stratified studies - is insufficient to provide a definitive answer as to whether sex itself constitutes an independent risk factor for faster ascending aortic growth. We believe this conclusion accurately reflects the available data and avoids overinterpretation. Of course, if the Editor do not think this is precise, we are willing to change it – however, it is very subtle already.

The following changes have been made in the revised manuscript:

This has been added to: Discussion (lines 340-352)

“Several clinical and hemodynamic factors have been proposed as potential modifiers of thoracic aortic disease progression, including blood pressure control, heart rate and rate-controlling therapy, lipid profile, and inflammatory or atherosclerotic vascular burden. However, for growth of the ascending thoracic aorta, evidence supporting consistent associations between these factors and diameter progression remains limited and are often inconsistent between studies.

Multiple studies included in the present review evaluated traditional cardiovascular risk factors - such as hypertension, smoking, and diabetes - without demonstrating significant associations with growth rate (data not shown). Moreover, optimal blood pressure targets and pharmacological strategies for patients with ascending thoracic aortic aneurysms are not well established, and recent data have questioned the protective role of beta-blocker therapy in this population. Conversely, diabetes mellitus has been associated with a lower risk of thoracic aortic dissection, underscoring that classical cardiovascular risk factors do not uniformly translate into aortic growth dynamics.

Aortic valve morphology represents a well-established determinant of ascending aortic growth; however, this factor was intentionally controlled for by restricting inclusion to patients with tricuspid aortic valves, thereby minimizing confounding by design.”

This has been added to: Discussion (lines 361-378)

“Importantly, absolute ascending aortic growth rate could be interpreted as a surrogate measure rather than a direct marker of clinical risk. While growth rate is widely used in surveillance and decision-making, adverse aortic events such as dissection or rupture are not determined by diameter progression alone. Instead, these outcomes reflect a complex interaction between aortic wall biology, hemodynamic load, medical therapy, and patient-specific vulnerability. Consequently, sex-specific differences in absolute growth rates, when observed, should properly not be interpreted as evidence of intrinsic sex-related differences in aneurysm behavior in isolation, but rather as the net result of biological, therapeutic, and methodological factors operating within the constraints of the available data.”

This has been added to: Discussion (lines 387-396)

“In addition, future studies should integrate standardized reporting of clinical, hemodynamic, and therapeutic variables to enable more comprehensive evaluation of factors influencing ascending aortic growth.”