Review Reports

Comments 1: The study only performed multi‑method comparisons (five methods) in 10 cases, with some subgroups having only 1–2 valid data points (e.g., only one valid result for tracheal swab mNGS). Cochran’s Q test and McNemar’s test are highly prone to false positives or false negatives with such a small sample size.

Response 1: We thank the Reviewer for this valuable comment. We agree that the small sample size limits the statistical power of the comparative analyses. As already acknowledged in the manuscript, all inferential analyses were interpreted cautiously, with emphasis on descriptive trends rather than definitive statistical evidence, and the limited statistical power is discussed in the Study Limitations section. To further address the Reviewer's concern, we have strengthened the manuscript by adding further clarification regarding the exploratory nature of the study in Sections 2.1 (Study Design and Case Selection) and 3.4 (Results of Microbiological Testing). Furthermore, the Conclusions have been revised to provide a more balanced interpretation of the findings and to avoid overinterpretation of the statistical results.

Comments 2: The authors defined “concordance” as the detection of at least one antemortem pathogen, even if additional bacteria were detected postmortem. This may favor mNGS (because it detects more bacteria). In fact, postmortem mNGS detected many additional organisms (e.g., Corynebacterium striatum), which could represent postmortem contamination or colonization.

Response 2: We sincerely thank the Reviewer for this insightful comment. We agree that our operational definition of concordance may favor highly sensitive techniques such as mNGS, since concordance was defined as the detection of at least one pathogen identified during antemortem microbiological investigations, regardless of the detection of additional microorganisms in the corresponding postmortem sample.

This definition was intentionally selected because the primary objective of the study was to evaluate whether each postmortem method was able to identify the clinically relevant pathogen detected during hospitalization, rather than to assess complete microbiological agreement between antemortem and postmortem findings. We acknowledge, however, that the additional microorganisms detected by mNGS may represent true polymicrobial infection, airway colonization, postmortem bacterial translocation, or contamination and, therefore, should not automatically be interpreted as clinically relevant pathogens.

To address the Reviewer's concern, we have expanded the Discussion to clarify the rationale for our definition of concordance and to emphasize that mNGS results should always be interpreted together with the clinical history, histopathological findings, conventional microbiological results, and the overall forensic context. We have also acknowledged this limitation in the Discussion.

Comments 3: The postmortem samples examined in this study are highly susceptible to contamination (skin flora, environmental bacteria, postmortem translocation). The manuscript does not mention the use of negative controls (e.g., simultaneous processing of sterile saline) or sequencing negative controls during sample collection.

Response 3: We sincerely thank the Reviewer for this important methodological comment. We agree that postmortem microbiological investigations, particularly molecular analyses, are susceptible to contamination from skin flora, environmental microorganisms, and postmortem bacterial translocation.

In the present study, dedicated negative controls during sample collection or sequencing were not included. We acknowledge that this represents a methodological limitation. Nevertheless, strict aseptic sampling procedures were applied to minimize the risk of contamination. Before opening the thoracic cavity, the skin was disinfected with alcohol and povidone–iodine, sterile gloves and sterile instruments were used throughout the sampling procedure, and lung tissue specimens were collected directly from the site of infection under sterile conditions. In addition, samples intended for molecular analyses were immediately preserved in DNA/RNA Shield and processed in a specialized molecular laboratory.

To address the Reviewer's concern, we have revised the Study Limitations section to explicitly acknowledge the absence of dedicated negative controls and to emphasize that molecular findings should be interpreted in conjunction with the clinical history, histopathological examination, and conventional microbiological results.

Comments 4: Table 2 lists a large number of multidrug‑resistant (MDR) strains, but no direct correlation with mNGS or culture concordance is provided. The text only speculates that “antibiotics may affect culture sensitivity” without quantitative analysis.

Response 4: We sincerely thank the Reviewer for this valuable comment. We agree that the limited sample size precluded a robust quantitative analysis of the relationship between multidrug resistance, prior antibiotic therapy, and concordance between postmortem diagnostic methods.

To explore this issue, we performed an exploratory analysis evaluating the association between empirical antibiotic adequacy and postmortem culture concordance using Fisher's exact test. As reported in Table 5, no statistically significant association was identified; however, only seven cases were evaluable, substantially limiting the statistical power of this analysis.

We have revised the manuscript to clarify that the antimicrobial susceptibility analysis was intended to provide descriptive information rather than to establish a definitive relationship between antimicrobial resistance and microbiological concordance. We have also expanded the Discussion to emphasize that the potential influence of prior antibiotic therapy and multidrug-resistant organisms on postmortem microbiological findings remains hypothesis-generating and should be investigated in larger studies.

Comments 5: The introduction mentions that one of the two purposes of postmortem microbiology is “to identify previously undetected pathogens.” However, this study relies entirely on antemortem known results as the “gold standard” and does not evaluate whether mNGS can detect pathogens missed by antemortem testing.

Response 5: We sincerely thank the Reviewer for this valuable observation. We agree that the present study was not designed to evaluate the ability of postmortem mNGS to identify previously undetected pathogens. Rather, its primary objective was to assess the concordance between antemortem microbiological findings and different postmortem microbiological methods.

Accordingly, we have revised the Introduction to better reflect the actual scope of the study. Specifically, we reformulated the paragraph describing the role of postmortem microbiology by removing the statement referring to the identification of previously undetected infections and clarified that the present study focuses on the confirmation of antemortem microbiological findings and the comparative evaluation of different postmortem diagnostic methods. The study aim has also been revised to better align with the study design.

Comments 6: The manuscript states, “PCR‑generated amplicons (~1500 bp) were prepared for sequencing on the Illumina platform,” but Sanger sequencing is typically performed by capillary electrophoresis, not Illumina. This appears to confuse the Sanger and NGS workflows.

Response 6: We sincerely thank the Reviewer for identifying this inaccuracy. We agree that the original wording could be interpreted as confusing the Sanger sequencing workflow with the metagenomic next-generation sequencing (mNGS) workflow.

The manuscript has been revised to clearly distinguish these two methodologies. PCR amplification of the approximately 1500 bp bacterial 16S rRNA gene was performed for subsequent Sanger sequencing, whereas metagenomic next-generation sequencing was carried out independently using DNA libraries prepared with the Nextera XT DNA Library Preparation Kit and sequenced on the Illumina MiSeq platform. We apologize for the imprecise wording in the original version.

Comments 1: Line 88. The word "conservation" is inappropriate in this case. It is better to use "conservativity" or "conserved regions" of 16S ribosomal RNA.

Response 1: We sincerely thank the Reviewer for this helpful comment. We agree that the original wording could be improved. Accordingly, the text has been revised by replacing the term "conservation" with the more appropriate expression "highly conserved regions", which is the standard terminology used to describe the conserved sequences of the 16S rRNA gene that serve as universal primer binding sites. The corresponding paragraph has been revised to improve both scientific accuracy and clarity.

Comments 2: Line 98. The word "usefulness" should be replaced with "possibility of application" or "benefit."

Response 2: We sincerely thank the Reviewer for this valuable suggestion. We agree that the term "usefulness" was too general in the context of the present study. Accordingly, we revised the study objective to more accurately reflect the scope and design of the study. The revised text now emphasizes the assessment of concordance between antemortem and postmortem microbiological findings and the comparative evaluation of the diagnostic performance, advantages, and limitations of the investigated methods.

Comments 3: In paragraph 2.1, line 121, the sentence contains a tautology, and the exclusion criteria duplicate the inclusion criteria, so they should be removed from the text; this information is redundant.

Response 3: We sincerely thank the Reviewer for this helpful observation. We agree that the exclusion criteria were largely complementary to the inclusion criteria and did not provide additional methodological information. Accordingly, to improve clarity and avoid redundancy, the exclusion criteria have been removed from the manuscript, while the inclusion criteria have been retained to adequately define the study population.

Comments 4: Paragraph 3.1 is more relevant to paragraph 2; it should be moved to the Materials and Methods.

Response 4: We sincerely thank the Reviewer for this thoughtful suggestion. After careful consideration, we respectfully decided to retain this section within the Results because it presents the characteristics of the analyzed study population and summarizes the data obtained from the investigated cases rather than describing the study design or methodology. We believe that its current placement provides a clearer distinction between the methodological procedures and the study findings, thereby improving the overall readability of the manuscript.

Comments 5: Paragraph 3.2, lines 230-231. Why were microbiological tests taken into account no later than 10 days before death? Was the mNGS data compared with earlier microbiological tests?

Response 5: We sincerely thank the Reviewer for this important question. Only antemortem microbiological results obtained within the last 10 days before death were included in the comparative analysis because they were considered the most clinically relevant to the infectious episode immediately preceding death. In addition, this time interval was intentionally selected to ensure a fair comparison across all postmortem diagnostic methods, including conventional culture, PCR/Sanger sequencing, and mNGS. During prolonged hospitalization, microbial flora, antimicrobial susceptibility patterns, and the viability of pathogens may change substantially because of disease progression and antimicrobial therapy. Consequently, microbiological results obtained earlier than 10 days before death may no longer accurately reflect the microorganisms present at the time of death and could have introduced bias into the concordance analysis. Therefore, all postmortem methods, including mNGS, were compared only with the most recent antemortem microbiological findings.

Comments 6: Lines 243-246. There are only 10 cases; it's worth stating the exact number: 5/10 matches.

Response 6: We sincerely thank the Reviewer for this helpful suggestion. We agree that, given the small sample size, presenting the absolute numbers together with the corresponding percentages improves the transparency and readability of the results. Accordingly, the manuscript has been revised throughout the Results section, and concordance is now reported using both absolute numbers (e.g., 5/10) and percentages.

Comments 7: Discussion. Lines 419-420. Ten cases described can hardly be considered a global level for the Cochran's Q test. The sentence should be reformulated more carefully.

Response 7: We sincerely thank the Reviewer for this valuable observation. We agree that the original wording could be interpreted as overstating the statistical findings given the limited sample size. Accordingly, the sentence has been revised to state that the Cochran's Q test indicated an overall difference in concordance among the evaluated diagnostic methods, thereby avoiding the expression "at the global level" and providing a more cautious interpretation of the results.

Comments 8: References. No doi specified for reference 17 doi: 10.1111/1556-4029.12991.

Response 8: We sincerely thank the Reviewer for pointing out this omission. The DOI has now been added to the corresponding reference in the revised manuscript. Due to the addition of new references during revision, this citation is now listed as Reference 23 (previously Reference 17).

Comments 1: Lines 43-47: The authors cited the WHO Report from 2022, but there is a most recent report from 2024: World Health Organization. Global report on infection prevention and control 2024. Geneva: WHO; 2024. Global report on infection prevention and control 2024

If the statistics are derived from the 2022 Global Report on Infection Prevention and Control and remain unchanged in the 2024 edition, this should be clarified. Otherwise, please update the text to reflect the most recent available data.

Response 1: We sincerely thank the Reviewer for this valuable suggestion. We have updated the Introduction by citing the most recent WHO Global Report on Infection Prevention and Control (2024). The epidemiological data were revised according to the updated report, and the text has been modified to reflect the most recent available global estimates regarding the burden of healthcare-associated infections.

Comments 2: Materials and Methods: Several sections, for example, 2.2. Sample Collection and 2.3. Sample Analysis, lack supporting references. If these procedures were based on previously published protocols, the appropriate references should be cited. If the methods were developed specifically for this study or represent novel procedures, this should be explicitly stated.

My suggestion to the authors is:

To cite the original or standard protocol if an established method was followed;

or to cite a previous publication from the authors if the method was previously described;

or to clearly state that the method was developed or modified by the authors.

Response 2: We sincerely thank the Reviewer for this valuable observation. The Materials and Methods section has been revised to clarify the methodological basis of the procedures used in this study. The postmortem sampling protocol was specifically designed by the authors based on the available scientific literature and established principles of forensic and clinical microbiology, with the aim of ensuring standardized and aseptic sample collection. Appropriate references supporting the sampling strategy have been added where applicable.

The microbiological and molecular analyses were performed according to the validated standard operating procedures routinely used in the accredited laboratories where the analyses were conducted (Bioclinica Laboratory and the Molecular Epidemiology Laboratory of the National Institute for Medical-Military Research and Development “Cantacuzino”). This clarification has been added to the revised manuscript.

Comments 3: Given the high susceptibility of postmortem and sequencing-based analyses to contamination, please clarify/describe the quality assurance and control procedures implemented throughout the study. Specifically, indicate which positive and negative controls were included in the PCR, and mNGS workflows. How potential contamination was assessed? The authors mentioned in the Discussion section (Line 358), but how do they accomplish that in this study?

Response 3: We sincerely thank the Reviewer for this important methodological comment. We fully agree that strict quality assurance procedures are essential in postmortem microbiology, particularly when molecular techniques are employed.

To minimize the risk of contamination during postmortem sampling, all specimens were collected under strict aseptic conditions. Before opening the thoracic cavity, the skin was disinfected using 70% ethanol followed by povidone–iodine solution. Sterile gloves, sterile swabs, sterile scalpels, and sterile forceps were used throughout the sampling procedure. Lung tissue specimens were collected directly from the affected pulmonary parenchyma immediately after opening the thoracic cavity, avoiding contact with surrounding tissues. Tracheal swabs were obtained using sterile swabs before manipulation of the respiratory tract. Following collection, all specimens were immediately placed into sterile containers, and samples intended for molecular analysis were preserved in DNA/RNA Shield and transported under appropriate conditions to the molecular laboratory.

Quality control procedures were also implemented during the molecular analyses. During DNA extraction, a nuclease-free water extraction blank was included as a negative control to monitor potential contamination introduced during the extraction process. Each PCR run included a negative control (nuclease-free water) and a positive control consisting of DNA extracted from a Gram-negative bacterium. In addition, the PCR control samples were subjected to 16S rRNA gene Sanger sequencing to verify amplification specificity and exclude contamination-related artifacts.

We have revised the Materials and Methods section to include these methodological details. In response to the Reviewer's comment, we contacted the Molecular Epidemiology Laboratory that performed the analyses to verify and document the quality assurance procedures applied during DNA extraction and PCR. The additional information provided by the laboratory has now been incorporated into the revised manuscript. We have also clarified in the Discussion that dedicated negative controls for the mNGS workflow could not be documented based on the information available to the authors and are therefore acknowledged as a limitation of the study.

Potential contamination was further assessed through careful interpretation of the microbiological findings. Postmortem molecular results were systematically correlated with the antemortem microbiological findings, conventional postmortem cultures, histopathological evidence of bronchopneumonia, and the clinical and forensic circumstances of each case. Microorganisms detected exclusively by molecular methods were therefore interpreted cautiously and were not considered causative pathogens unless supported by the overall clinicopathological context.

Comments 4: Line 37: "16S rRNA gene sequencing" is more technically accurate than “16S rRNA sequencing”. I suggest using the former consistently throughout the manuscript, including in Table1.

Response 4: We sincerely thank the Reviewer for this valuable suggestion. We agree that the term "16S rRNA gene sequencing" is more technically accurate. Accordingly, the terminology has been revised throughout the manuscript, including Table 1, to ensure consistency with the current scientific nomenclature.

Comments 5: Line 40 – Introduction: To better contextualize the study, this section would benefit from a brief introduction of the clinical relevance of the bacterial species investigated. As several of the target pathogens belong to the ESKAPE group, the authors should consider introducing this concept and highlighting the importance of these organisms as major multidrug-resistant pathogens responsible for healthcare-associated infections.

Response 5: We sincerely thank the Reviewer for this valuable suggestion. To better contextualize the study, we have expanded the Introduction by briefly introducing the concept of ESKAPE pathogens and highlighting their clinical relevance as major multidrug-resistant organisms responsible for healthcare-associated infections, particularly hospital-acquired pneumonia. Appropriate references have also been added to support this information.

Comments 6: Line 55: “Clostridium difficile” was reclassified in 2016 as Clostridioides difficile. Please review the following articles and update the species name in the manuscript accordingly. Moreover, the species name should be italicized. Please ensure that the bacterial species are correctly written according to taxonomy standards throughout the manuscript.

Lawson et al. 2016. Reclassification of Clostridium difficile as Clostridioides difficile (Hall and O'Toole 1935) Prévot 1938. DOI: 1016/j.anaerobe.2016.06.008

C difficile—a rose by any other name…The Lancet Infectious Diseases, The Lancet Infectious Diseases, Volume 19, Issue 5, 449. DOI: 1016/S1473-3099(19)30177-X.

Response 6: We sincerely thank the Reviewer for this valuable observation. We agree that the current taxonomic nomenclature should be used throughout the manuscript. Accordingly, the species name has been updated from Clostridium difficile to Clostridioides difficile throughout the text, tables, and figure legends, and all bacterial species names have been checked and formatted in italics in accordance with current taxonomic standards. Although the original hospital medical records used the former designation, the revised manuscript follows the currently accepted nomenclature.

Comments 7: Please clarify whether the VITEK® 2 Compact system was used exclusively for bacterial identification or also for antimicrobial susceptibility testing. If so, please describe the methodology (e.g., AST cards used). If the antimicrobial susceptibility testing was performed in a different way, please include the description in the Materials and Methods section.

Response 7: We sincerely thank the Reviewer for this valuable comment. We have revised the Materials and Methods section to clarify that the VITEK® 2 Compact system was used not only for bacterial identification but also for antimicrobial susceptibility testing (AST). The AST results were interpreted according to the CLSI 2023 guidelines routinely implemented by the accredited clinical microbiology laboratory. This clarification has been added to the revised manuscript.

Comments 8: Line 249 and Table 1: The expression “bacterial flora” is no longer used. The term “flora”, has been replaced by “microbiota” or, in other contexts, by “microbiome.” I suggest that the authors update this term in the text and table.

Response 8: We sincerely thank the Reviewer for this valuable suggestion. We agree that the term "microbiota" is more appropriate and consistent with current microbiological terminology. Accordingly, the expression "bacterial flora" has been replaced with "bacterial microbiota" throughout the manuscript, including Table 1, where appropriate.

Comments 9: Table 1 and Table 2: Please consider improving the formatting of the Tables legends. Other suggestion is that “Abbreviations” PCR and mNGS may not need to be defined in the table 1.

Response 9: We sincerely thank the Reviewer for this helpful suggestion. The legends of Tables 1 and 2 have been revised to improve their clarity and formatting. In addition, the definitions of the abbreviations PCR and mNGS have been removed from Table 1, as these abbreviations are already defined in the main text.

Comments 10: Line 279: Please ensure that all bacterial species names are italicized throughout the manuscript, in accordance with standard scientific nomenclature.

Response 10: We sincerely thank the Reviewer for this valuable observation. The manuscript has been carefully reviewed to ensure consistency with the current scientific nomenclature. All bacterial genus and species names have been formatted in italics throughout the manuscript, including the main text, tables, and figure legends, where applicable.

Comments 11: Lines 155-157: Please clarify the rationale for the selection of the culture media used in the study. It would be helpful if the authors could indicate whether the choice was based on a specific protocol, published methodology, or established clinical or forensic microbiology guidelines. Given the target pathogens investigated, there are more selective and differential media available for the isolation of some bacterial species, and a brief justification for the selected media would strengthen the methodological description.

Response 11: We sincerely thank the Reviewer for this valuable comment. We have revised the Materials and Methods section to clarify that the culture media used in this study were selected according to the validated standard operating procedures routinely implemented by the accredited clinical microbiology laboratory (Bioclinica Laboratory), where the conventional microbiological cultures were performed. These media are routinely used for the isolation of bacterial pathogens commonly associated with respiratory tract infections and healthcare-associated pneumonia. This clarification has been added to the revised manuscript.

Comments 12: Line 165: Please provide additional details regarding the PCR protocol. The current description does not include the reaction conditions, for example, neither the controls performed. If a previously published protocol was followed, the corresponding reference should be cited.

Response 12: We sincerely thank the Reviewer for this valuable comment. The Materials and Methods section has been revised to provide additional methodological details regarding the molecular analyses. The molecular analyses were performed in the Molecular Epidemiology Laboratory of the National Institute for Medical-Military Research and Development “Cantacuzino”, a laboratory that routinely implements standardized molecular diagnostic procedures and participates in external quality assessment programs.

Specifically, we have clarified the quality control procedures implemented during DNA extraction and PCR, including the use of appropriate negative and positive controls. In response to the Reviewer's comment, additional methodological information was requested from the laboratory that performed the molecular analyses, and the available details have been incorporated into the revised manuscript. We also revised the description of the molecular workflow to clearly distinguish the PCR/16S rRNA gene Sanger sequencing procedure from the mNGS workflow and to avoid ambiguity regarding the sequencing platforms used.

Comments 13: Suggestion: From an epidemiological perspective, it would be interesting to know whether any strain-level comparison or molecular typing of the isolates was performed to assess the possibility of cross-transmission or a common environmental source of infection. Because such analyses were not performed, this limitation could be acknowledged in the Discussion.

Response 13: We sincerely thank the Reviewer for this valuable suggestion. We agree that strain-level comparison or molecular typing could have provided additional epidemiological information regarding the potential cross-transmission of pathogens or the existence of a common environmental source of infection. However, such analyses were beyond the scope of the present study. We have now acknowledged this limitation in the Discussion section.