Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

General Comments

This manuscript presents a single-center, retrospective 10-year cohort study comparing surgical and conservative management strategies for post-sternotomy mediastinitis (PSM). The topic is clinically relevant, as PSM remains a severe complication of cardiac surgery and optimal management strategies are still debated. The integration of clinical, microbiological, and therapeutic data represents a strength of the study, and the low in-hospital mortality reported is noteworthy.

The study has some methodological and analytical limitations. In particular, the non-randomized treatment allocation, absence of multivariable analysis, and limited outcome assessment need to be addressed or more clearly acknowledged.

Major Comments

1. Abstract
   The Results and Conclusions sections of the abstract contain repetitive statements.
2. Study Design and Selection Bias
   Treatment allocation (surgical vs. conservative) was based on clinical judgment rather than randomization. This introduces selection bias, particularly given the higher DSW-STS scores observed in the conservatively treated group. Despite this imbalance, no adjustment for baseline differences was performed. It is recommended that to consider performing a multivariable analysis (e.g., logistic regression) to identify independent predictors of outcomes such as mortality or length of stay and to clearly emphasize in the limitations that unadjusted comparisons substantially limit causal inference.
3. Statistical Analysis
   The analysis is limited to univariate comparisons. The statement that “statistical analysis was performed using aggregated summary data” is not sufficiently justified and restricts interpretability. In addition, no correction for multiple comparisons was applied.
4. Definition of Outcomes
   Primary and secondary endpoints are not explicitly defined. Outcomes are largely restricted to short-term in-hospital measures (14-day survival, in-hospital mortality, culture negativization), with no long-term follow-up. Please explicitly discuss the absence of long-term outcomes (recurrence, late mortality, functional status) as a limitation.
5. Treatment Group Definitions
   Vacuum-assisted closure (VAC) therapy was widely used in both groups, which blurs the distinction between “surgical” and “conservative” management. It is unclear whether VAC should be considered part of surgical treatment or an adjunct to conservative therapy. It is recommended providing a clearer operational definition of treatment groups and to consider discussing VAC therapy as a separate variable influencing outcomes.

Minor Comments

Tables
Some tables are dense and difficult to interpret, with partial repetition of variables across tables.  Please simplify tables and improve readability where possible.

CDC Definitions
CDC criteria for mediastinitis are referenced multiple times using different citations. Standardizing the reference to CDC definitions is recommended.

Comments on the Quality of English Language

Minor grammatical and stylistic issues are present throughout the manuscript (e.g., use of contractions, minor typographical errors).

Author Response

This manuscript presents a single-center, retrospective 10-year cohort study comparing surgical and conservative management strategies for post-sternotomy mediastinitis (PSM). The topic is clinically relevant, as PSM remains a severe complication of cardiac surgery and optimal management strategies are still debated. The integration of clinical, microbiological, and therapeutic data represents a strength of the study, and the low in-hospital mortality reported is noteworthy.

The study has some methodological and analytical limitations. In particular, the non-randomized treatment allocation, absence of multivariable analysis, and limited outcome assessment need to be addressed or more clearly acknowledged.

Thank you for your valuable suggestions, which have enriched the study with truly interesting insights into the method. The study obviously has limitations due to its retrospective and observational nature, but its aim is to provide a description of a category of patients who are difficult to manage and on whom little literature is available.

We hope we have been able to satisfy your requests and suggestions! We remain at your disposal for further clarification.

Major Comments

Comments 1: Abstract: The Results and Conclusions sections of the abstract contain repetitive statements.

Response 1: Thank you for the suggestion. It was a typo in the text following the transfer of the manuscript to the template. We have amended the abstract with the suggested changes (lines 28-34).

Comments 2: Study Design and Selection Bias: Treatment allocation (surgical vs. conservative) was based on clinical judgment rather than randomization. This introduces selection bias, particularly given the higher DSW-STS scores observed in the conservatively treated group. Despite this imbalance, no adjustment for baseline differences was performed. It is recommended that to consider performing a multivariable analysis (e.g., logistic regression) to identify independent predictors of outcomes such as mortality or length of stay and to clearly emphasize in the limitations that unadjusted comparisons substantially limit causal inference.

Response 2: Thank you for your comment. It would have been very interesting to randomise the patients. However, we were unable to follow this strategy because the retrospective nature of the study allows us to represent the characteristics and outcomes of all patients with post-sternotomy mediastinitis who underwent one treatment or another according to the clinical judgement of the treating physicians.

To implement the results and conclusions, as suggested, we performed a multivariate analysis of the factors associated with surgical versus conservative approaches and a multivariate analysis for mortality (lines 245-264).

We also included the suggestions in the limitations section (lines 321-323).

Comments 3: Statistical Analysis: The analysis is limited to univariate comparisons. The statement that “statistical analysis was performed using aggregated summary data” is not sufficiently justified and restricts interpretability. In addition, no correction for multiple comparisons was applied.

Response 3: We have included multivariate analysis as previously suggested to enrich the results and conclusions. However, as explained, the study data were obtained by collecting information from medical records. This limits the availability of certain information, making it less informative and interpretable.

Comments 4: Definition of Outcomes: Primary and secondary endpoints are not explicitly defined. Outcomes are largely restricted to short-term in-hospital measures (14-day survival, in-hospital mortality, culture negativization), with no long-term follow-up. Please explicitly discuss the absence of long-term outcomes (recurrence, late mortality, functional status) as a limitation.

Response 4: Thank you for your suggestion. For greater clarity, we have specified your suggestion in the limitations section (lines 327-329).

Comments 5: Treatment Group Definitions: Vacuum-assisted closure (VAC) therapy was widely used in both groups, which blurs the distinction between “surgical” and “conservative” management. It is unclear whether VAC should be considered part of surgical treatment or an adjunct to conservative therapy. It is recommended providing a clearer operational definition of treatment groups and to consider discussing VAC therapy as a separate variable influencing outcomes.

Response 5: For greater clarity, we have provided a more detailed description of the role of vacuum therapy and its use in the text (lines 109-110 and 113-114).

Minor Comments

Comments 6: Tables: Some tables are dense and difficult to interpret, with partial repetition of variables across tables.  Please simplify tables and improve readability where possible.

Response 6: Thank you for your suggestion. We have tried to evaluate removing some variables from the tables to make them more usable. However, we have already removed some variables previously and, for the sake of completeness, we prefer to keep this data available to provide a more comprehensive representation of the patients under examination. If you believe any data to be superfluous, we are available for re-evaluation.

Comments 7: CDC Definitions: CDC criteria for mediastinitis are referenced multiple times using different citations. Standardizing the reference to CDC definitions is recommended.

Response 7: Thank you for pointing this out. We have removed the repetitive reference.

Reviewer 2 Report

Comments and Suggestions for Authors

This article presents a single-center retrospective study comparing surgical and conservative approaches to the treatment of post-sternotomy mediastinitis. The work addresses an important clinical problem and contains useful data; however, it requires substantial revision prior to publication.

Recommendations and Comments:

• Literature Review: The majority of cited sources are outdated, with a predominance of publications from the 2000s and early 2010s. For such a rapidly evolving field, it is essential to incorporate more recent studies (2020–2025), particularly concerning microbiology and antimicrobial therapy.
• Study Design: The design is clearly described; however, information on sample size estimation (power calculation) is lacking. It is recommended to add this calculation.
• Statistical Analysis: The description of statistical methods is superficial. The specific tests used for intergroup comparisons and the handling of missing data are not stated. A detailed description of the statistical methodology must be added.
• Lack of Multivariate Analysis: This is a major limitation, as differences between the groups may be attributable to factors other than the treatment strategy. If feasible, incorporating a multivariate analysis is strongly recommended, as this could significantly alter the interpretation of the results.

                             

Author Response

Reviewer #2

This article presents a single-center retrospective study comparing surgical and conservative approaches to the treatment of post-sternotomy mediastinitis. The work addresses an important clinical problem and contains useful data; however, it requires substantial revision prior to publication.

Thank you for your valuable suggestions, which have enriched the study with truly interesting insights into the method. The study obviously has limitations due to its retrospective and observational nature, but its aim is to provide a description of a category of patients who are difficult to manage and on whom little literature is available.

We hope we have been able to satisfy your requests and suggestions! We remain at your disposal for further clarification.

Comments 1: Literature Review: The majority of cited sources are outdated, with a predominance of publications from the 2000s and early 2010s. For such a rapidly evolving field, it is essential to incorporate more recent studies (2020–2025), particularly concerning microbiology and antimicrobial therapy.

Response 1: Thank you for pointing this out. We have updated the bibliography, keeping the percentage of references prior to 2020 to the minimum allowed by the journal.

Comments 2: Study Design: The design is clearly described; however, information on sample size estimation (power calculation) is lacking. It is recommended to add this calculation.

Response 2: Thank you for the suggestion. However, we did not set up the study to calculate the sample size, as we selected all patients with post-sternotomy mediastinitis who were admitted to our regional cardiac surgery referral centre during the specified time period and evaluated their characteristics and outcomes retrospectively.

Comments 3: Statistical Analysis: The description of statistical methods is superficial. The specific tests used for intergroup comparisons and the handling of missing data are not stated. A detailed description of the statistical methodology must be added.

Response 3: Thank you for your precious advice. We have added the description of the statistical analysis as suggested (lines 136-139). I hope I have managed to clarify statistical methods better.

 Comments 4: Lack of Multivariate Analysis: This is a major limitation, as differences between the groups may be attributable to factors other than the treatment strategy. If feasible, incorporating a multivariate analysis is strongly recommended, as this could significantly alter the interpretation of the results.

Response 4: Thank you for your comment. To implement the results and conclusions, as suggested, we performed a multivariate analysis of the factors associated with surgical versus conservative approaches and a multivariate analysis for mortality (lines 245-264).

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The issue is clinically important; PSM is a rare but high-mortality complication, and real-life data are valuable. However, methodological and statistical limitations exist, and some analyses, as they are, are not suitable for making causal inferences. In particular, selection bias, confounding, and lack of power weaken the interpretation of the results.
With the current data, effectiveness cannot be claimed; only observational relationships can be shown. Therefore, the language should be softened.
Mortality is very low according to the literature. The reason should be explained.
Statistically, if patient-level analysis is not performed, the regression model will not work correctly. Descriptive and unadjusted comparisons can be performed.
Microbiologically, antibiotic resistance rates and empirical treatment regimens should be discussed more thoroughly.

Comments on the Quality of English Language

Minor grammatical and stylistic issues are present throughout the manuscript (e.g., use of contractions, minor typographical errors).

Author Response

Comments 1: The issue is clinically important; PSM is a rare but high-mortality complication, and real-life data are valuable. However, methodological and statistical limitations exist, and some analyses, as they are, are not suitable for making causal inferences. In particular, selection bias, confounding, and lack of power weaken the interpretation of the results. With the current data, effectiveness cannot be claimed; only observational relationships can be shown. Therefore, the language should be softened.

Mortality is very low according to the literature. The reason should be explained.
Statistically, if patient-level analysis is not performed, the regression model will not work correctly. Descriptive and unadjusted comparisons can be performed. Microbiologically, antibiotic resistance rates and empirical treatment regimens should be discussed more thoroughly.

Response 1: We appreciate the valuable suggestions that are adding further value to the manuscript, which, due to its retrospective nature, has significant limitations in terms of interpretation. However, the study as designed is not intended to evaluate comparisons in terms of mortality. The only data we felt was important to report is the incidence of mortality in patients treated at our center, without, however, being able to extrapolate information about cause and effect, due to the characteristics you previously explained. Nevertheless, we felt it was interesting to present this data to extrapolate hypotheses and possibly design an ad hoc study on mortality. In view of these factors, there are also some data that cannot be modified, such as, for example - the assignment of a patient to one group rather than another, as this is secondary to the therapeutic choices made by the treating physicians according to the guidelines adopted by the department, or - the absence of statistical power, as the study design is limited to a description of all patients treated at our cardiac surgery center who developed PSM. It would have been very useful and interesting to evaluate the sample power ab initio, further corroborating the data. It would have been very important, as suggested, to integrate these values into the manuscript. However, given the real difficulty in providing data in this regard, we can more strongly reiterate the concepts in the study limitations, mitigating the conclusions drawn from the analyses and considering them hypothetical, also in terms of effectiveness (lines 333-335, 338-340). Regarding the microbiological information and data that were not sufficiently explored, we have elaborated on these concepts and any considerations in greater depth (lines 303-312).

Thanking you for your valuable contribution to enriching our manuscript, we hope that we have been sufficiently exhaustive in our explanations and remain available for any further information.

Reviewer 2 Report

Comments and Suggestions for Authors All comments have been corrected by the authors.

Author Response

We appreciated the valuable suggestions that added further value to the manuscript, and we thank you for your efforts in improving it.

Round 3

Reviewer 1 Report

Comments and Suggestions for Authors

The article doesn't represent a significant improvement over previous recommendations

Comments on the Quality of English Language

Minor grammatical and stylistic issues are present throughout the manuscript (e.g., use of contractions, minor typographical errors).