Management Outcomes of Trauma Patients Presenting with Renal Injuries: A Three-Year Retrospective Audit at a South African Tertiary Trauma Centre

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for the opportunity to review this manuscript. The authors present a retrospective case series of renal trauma from a high (penetrating trauma) volume trauma centre in South Africa. Its main strength is that it reports contemporary experience from an environment that is underrepresented in the literature. The paper is readable and the overall findings are plausible. However, the manuscript currently overreaches in parts, and I think it would benefit from a clearer descriptive framing, tighter methods, and more restrained conclusions.

I have several suggestions:

1. The paper implies a stronger comparison between operative and non operative management than the data really support. Much of the operative cohort appears to have gone to theatre because of associated injuries rather than because renal intervention itself was required. The manuscript should therefore be framed more as a descriptive cohort study of renal trauma management patterns and outcomes, rather than a comparison of management strategies.
2. The multivariable analysis needs clearer explanation and more caution in interpretation. With relatively few deaths, the model is at risk of overfitting. The authors should state exactly how variables were selected, how missing data were handled, and why this modelling approach was chosen. The identified predictors should be presented as associations, not firm prognostic determinants.
3. Important clinical descriptors are missing. The paper would be stronger with clearer reporting of overall injury severity, haemodynamic status, imaging pathways, and how renal injury grade was assigned in patients taken straight to operation. Without this, interpretation of both management decisions and mortality is limited.
4. The operative management section needs more granularity. It would help considerably to separate patients who underwent laparotomy for associated injuries from those who actually required renal intervention. A simple table stratified by injury grade and management pathway would make the findings much clearer.
5. The discussion should be more restrained. The conclusion that selective non operative management produced favourable outcomes is probably true, but the current data do not allow that statement to be made strongly. Likewise, lactate, creatinine, and transfusion requirement are plausible markers of severity, but they should not be presented as ready for incorporation into decision making algorithms on the basis of this study alone.

Overall, I think the manuscript has publishable potential, but it requires major revision before it is ready. Its real value lies in describing renal trauma practice and mortality associated factors in a penetrating trauma dominant setting. With clearer methods, tighter framing, and more cautious interpretation, it could become a useful contribution.

Author Response

Reviewer 1

We thank Reviewer 1 for the thoughtful and detailed critique. We appreciate the recognition of the manuscript’s relevance and the constructive suggestions regarding framing, methods, and interpretation.

Comment 1

The paper implies a stronger comparison between operative and non-operative management than the data really support. Much of the operative cohort appears to have gone to theatre because of associated injuries rather than because renal intervention itself was required. The manuscript should therefore be framed more as a descriptive cohort study of renal trauma management patterns and outcomes, rather than a comparison of management strategies.

Response:
We agree with this important point and have revised the manuscript to frame it more appropriately as a descriptive cohort study of renal trauma management patterns and outcomes in a penetrating-trauma-dominant setting. We have softened language that implied direct comparison between operative and non-operative management and clarified that many operations were performed for associated injuries rather than for renal intervention alone.

Comment 2

The multivariable analysis needs clearer explanation and more caution in interpretation. With relatively few deaths, the model is at risk of overfitting. The authors should state exactly how variables were selected, how missing data were handled, and why this modelling approach was chosen. The identified predictors should be presented as associations, not firm prognostic determinants.

Response:
We agree and have revised the Methods and Discussion accordingly. We now clarify that the multivariable model was intentionally parsimonious given the limited number of mortality events. We also state that variables were first examined in univariate analyses and that clinically relevant factors were considered for inclusion in the adjusted model. In addition, we note that missing data were handled using complete-case analysis. In the Discussion and Conclusion, we now present the identified variables as associations with mortality rather than definitive prognostic determinants.

Comment 3

Important clinical descriptors are missing. The paper would be stronger with clearer reporting of overall injury severity, haemodynamic status, imaging pathways, and how renal injury grade was assigned in patients taken straight to operation. Without this, interpretation of both management decisions and mortality is limited.

Response:
We appreciate this suggestion. The manuscript now provides clearer clinical contextualisation of the cohort, including injury severity, admission physiology, and AAST grading based on CT or intraoperative findings. We also clarified the high prevalence of concomitant injuries and the relevance of these associated injuries to management decisions and outcomes. We have additionally noted that ISS was not routinely documented in the source records and therefore could not be analysed.

Comment 4

The operative management section needs more granularity. It would help considerably to separate patients who underwent laparotomy for associated injuries from those who actually required renal intervention. A simple table stratified by injury grade and management pathway would make the findings much clearer.

Response:
We agree and have expanded the Results to distinguish between patients who underwent surgery for associated injuries and those who required renal-specific intervention. We also clarified the proportions undergoing nephrectomy, renorrhaphy, and DJ stent placement. This revision should improve interpretability of the operative cohort and better reflect the actual management pathway.

Comment 5

The discussion should be more restrained. The conclusion that selective non operative management produced favourable outcomes is probably true, but the current data do not allow that statement to be made strongly. Likewise, lactate, creatinine, and transfusion requirement are plausible markers of severity, but they should not be presented as ready for incorporation into decision making algorithms on the basis of this study alone.

Response:
We agree and have moderated the language throughout the Discussion and Conclusion. SNOM is now described as an important management option in carefully selected patients, rather than as a definitive superior strategy based on these data alone. We have also softened the wording regarding lactate, creatinine, and transfusion requirements, describing them as readily available markers associated with mortality that may assist early risk stratification, but require prospective validation before incorporation into formal algorithms.

Reviewer 2 Report

Comments and Suggestions for Authors

This is a retrospective single-centre study of renal trauma from a setting with a notably high proportion of penetrating injuries, providing clinically relevant data on the injury and management patterns and outcomes in an underrepresented context. The manuscript is well written, the methods clearly described, and the conclusions are supported by the data.

I have 2 minor suggestions:

• First, the introduction would benefit from a clearer statement of the specific gap in the literature that this study addresses, particularly in relation to prior published cohorts.
• Second, the results section could be streamlined. Some of the narrative text repeats information that could be left in the tables only.

This is a worthwhile paper and I would support publication after minor revision.

Author Response

Reviewer 2

We thank Reviewer 2 for the positive assessment and the helpful suggestions.

Comment 1

The introduction would benefit from a clearer statement of the specific gap in the literature that this study addresses, particularly in relation to prior published cohorts.

Response:
We agree and have revised the Introduction to state the gap more explicitly. We now highlight that prior South African and LMIC literature has largely focused on injury epidemiology or broad trauma cohorts, while fewer studies have detailed the distinction between laparotomy for associated injuries and renal-specific intervention, or examined bedside physiological markers in relation to early mortality in a predominantly penetrating-trauma cohort.

Comment 2

The results section could be streamlined. Some of the narrative text repeats information that could be left in the tables only.

Response:
We agree and have streamlined the Results section by reducing repetition and focusing the narrative on the key findings most relevant to interpretation. Redundant descriptive statements have been removed where the tables already provide the relevant detail.

Reviewer 3 Report

Comments and Suggestions for Authors

Congratulations on your work. This is a well-prepared manuscript with a unique patient profile. 

I think it may be of interest to readers to know a little more about the differentiating features of the renal injuries in those patients who underwent kidney specific surgical intervention vs those who did not (eg. AAST grades). This may help us identify predictors of surgical intervention. 

 

Additionally, do you have the injury severity scores of your cohort? This may give us a better impression of the overall injury burden in these patients. 

 

Thank you

Author Response

Reviewer 3

We thank Reviewer 3 for the encouraging comments and the useful suggestions.

Comment 1

It may be of interest to readers to know a little more about the differentiating features of the renal injuries in those patients who underwent kidney specific surgical intervention vs those who did not (eg. AAST grades). This may help us identify predictors of surgical intervention.

Response:
We agree that this is useful clinically. We have expanded the manuscript to better describe the relationship between injury severity and management pathway, including the AAST distribution and the distinction between patients who underwent renal-specific intervention and those who had surgery for associated injuries only. This helps contextualise the operative cohort and provides greater insight into the injury burden associated with intervention.

Comment 2

Additionally, do you have the injury severity scores of your cohort? This may give us a better impression of the overall injury burden in these patients.

Response:
We thank the reviewer for raising this important point. Unfortunately, Injury Severity Score (ISS) was not routinely documented in the source records and therefore could not be analysed in this retrospective study. We have added this as a limitation in the manuscript and noted it in the Methods/Discussion.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for the opportunity to review this revised manuscript. The revision is improved and the authors have addressed several substantive concerns, particularly by clarifying that most renal injuries occurred in the context of major associated trauma and that many laparotomies were not renal directed. However, some overstatement remains in the abstract and conclusion, where the manuscript still refers to “comparing operative and non-operative approaches”, “independent predictors”, and favourable SNOM outcomes in language that remains stronger than the study design supports.

 

I have several comments:

1. There is residual overstatement: the manuscript is better framed as a descriptive cohort, but the abstract still implies a comparative effectiveness study. The authors should remove “comparing operative and non-operative approaches” and replace it with language such as “describing management patterns and outcomes”. The conclusion should also avoid implying that SNOM was proven favourable, and instead state that selected patients managed non operatively had acceptable outcomes.

1. The statistical methods require clarification: the Cox model remains a concern. With only 28 deaths, the manuscript should explicitly state the number of variables entered, how the final variables were selected, how missing data were handled, and why Cox regression was chosen for in hospital mortality rather than logistic regression. The table footnote still suggests inclusion based on univariate significance, which risks data driven modelling and over interpretation.

1. Operative pathway is improved but could be clearer: the distinction between laparotomy for associated injuries and renal specific intervention is valuable and should be retained. A single additional summary table stratifying AAST grade by non-operative management, laparotomy without renal intervention, renorrhaphy, stent, and nephrectomy would make the clinical message much clearer.

1. Minor presentation issues: there is some repetition in the Results, with the surgical management paragraph effectively appearing twice. The abstract and Discussion should be aligned with the more cautious tone of the revised manuscript. The absence of ISS remains a major limitation and should be foregrounded, not merely noted.

 

The revision is substantially improved and is likely suitable for publication after minor to moderate revision. The key remaining requirement is tighter alignment between the stated study design and the strength of the conclusions.

Author Response

Reviewer 1

Comment 1: Remove "comparing operative and non-operative approaches" from abstract and replace with "describing management patterns and outcomes."

Response: We thank the reviewer for this suggestion. The phrase has been replaced in the abstract. The relevant sentence now reads: "…describing management patterns and outcomes, and identifying predictors of in-hospital death." This change better reflects the descriptive nature of the study. [See highlighted change in Abstract]

Comment 2: In conclusion, avoid implying SNOM was proven favourable; instead state "selected patients managed non-operatively had acceptable outcomes."

Response: We agree that our original wording overstated the evidence. The conclusion has been revised to state: "Selected patients managed non-operatively had acceptable outcomes, supporting the potential role of SNOM in carefully selected renal trauma patients…" We have also added an explicit caveat that this retrospective audit does not permit direct comparison of operative and non-operative strategies. The abstract conclusion has been similarly revised. [See highlighted changes in Conclusions and Abstract]

Comment 3: Add clarification to statistical methods section: explicitly state the number of variables entered in Cox model, how final variables were selected, how missing data were handled, and why Cox regression was chosen for in-hospital mortality rather than logistic regression.

Response: We appreciate this important methodological query. The statistical methods section (Section 2.5) has been substantially expanded to address all four points:
• Cox regression was chosen over logistic regression to account for variable follow-up time (length of hospital stay), thereby utilising the time-to-event information inherent in in-hospital mortality data.
• Eighteen clinically relevant candidate variables were assessed individually in the univariate analysis.
• Variables with p < 0.05 on univariate analysis were considered for the multivariable model. Given the limited number of mortality events (n = 28), a maximum of three covariates were entered into the final model to maintain an adequate events-per-variable ratio.
• Missing data were handled by complete-case analysis; variables with >10% missing values were excluded from the multivariable model.
[See highlighted changes in Section 2.5]

Comment 4: Add a summary table stratifying AAST grade by management type: non-operative management, laparotomy without renal intervention, renorrhaphy, stent, and nephrectomy.

Response: A new table (labelled Table 3) has been added to the manuscript, stratifying AAST renal injury grade by management type (SNOM, laparotomy without renal intervention, renorrhaphy, DJ stent insertion, and nephrectomy) with corresponding in-hospital mortality for each grade.

Comment 5: Remove any repetition in the Results section (surgical management paragraph appearing twice).

Response: We apologise for this oversight. The duplicated paragraph describing surgical management (beginning "Surgical intervention was required in 110 patients…") has been removed. The information is now presented only once within the Management subsection (Section 3.3). [Duplicate paragraph deleted from revised manuscript]

Comment 6: Foreground the absence of ISS as a major limitation in the limitations section.

Response: We agree that the absence of ISS is a critical limitation. The limitations section has been restructured so that this point now appears first and is expanded. The revised text explicitly acknowledges that ISS is a well-established predictor of mortality in trauma populations and that its omission precludes adjustment for overall injury burden, potentially confounding the reported associations. We have also added a recommendation for routine ISS documentation in future studies. [See highlighted changes in Limitations section]

Comment 7: Align abstract and Discussion with a more cautious tone.

Response: We have reviewed and revised both the abstract and discussion to adopt a more cautious, descriptive tone throughout. Specific changes include:
• Abstract conclusion: replaced "SNOM demonstrated favourable outcomes" with "Selected patients managed non-operatively had acceptable outcomes…although comparative conclusions cannot be drawn from this study."
• Discussion opening: changed "important insights" to "a descriptive overview of."
• Discussion SNOM paragraph: added an explicit statement that direct comparison between operative and non-operative groups was not the aim and that observed differences must be interpreted with caution given inherent selection bias.
• Conclusions: reframed SNOM findings as descriptive rather than comparative.
[See highlighted changes in Abstract, Discussion, and Conclusions]

Reviewer 2 Report

Comments and Suggestions for Authors

Thanks for presenting the revised manuscript. My suggestions have been addressed. I would just suggest minor re-wording of the sentance added to the introduction (even as simple as remove the "however" and put the added paragraph just before the aim sentance). 

Author Response

Reviewer 2

Comment: Minor rewording of sentence in introduction: remove "however" and move the added paragraph just before the aim sentence.

Response: We have implemented both changes. The word "However" has been removed from the beginning of the paragraph (now beginning with "Existing South African and LMIC literature…"), and the paragraph has been repositioned to immediately precede the aim statement. This improves the logical flow from the identification of the literature gap to the statement of the study aim. [See highlighted changes in Introduction]

Reviewer 3 Report

Comments and Suggestions for Authors

Thank you very much for addressing my comments. The updates provide valuable clarity on your cohort of patients. 
I would like to suggest that including a table or details within your manuscript regarding the grade of injury and type of renal specific surgical intervention provided and outcomes would give us a greater insight into the potential intervention required for each specific type of injury. 
Apart from this I am satisfied with the revised manuscript. 
Thank you

Author Response

Reviewer 3

Comment: Add a table or details within the manuscript regarding the grade of injury and type of renal specific surgical intervention provided and outcomes.

Response: We thank the reviewer for this valuable suggestion. A new summary table (Table 3) has been added to the Results section, stratifying AAST renal injury grade (I–V) by management type (SNOM, laparotomy without renal intervention, renorrhaphy, DJ stent insertion, and nephrectomy) and including the corresponding in-hospital mortality for each grade. This table provides a comprehensive overview of the relationship between injury severity, surgical intervention type, and clinical outcomes.

Round 3

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for the opportunity to review the revised manuscript; it is substantially improved. The framing is now appropriately descriptive, the distinction between laparotomy for associated injuries and renal specific intervention is clearer, and the Cox model is better explained, including the limited number of mortality events and complete case handling. The paper is close, but several issues should be corrected before acceptance:

1. Statistical interpretation: The revised methods now explain why Cox regression was used and limit the final model to three covariates because there were only 28 deaths, which is a significant improvement. However, the manuscript still uses language such as “independent predictors” quite strongly. Given the absence of ISS and the limited adjustment possible, these should be described as “independently associated with mortality in the adjusted model” rather than true independent predictors.
2. Residual confounding: The limitations section appropriately acknowledges that ISS was unavailable and that this precludes adjustment for overall injury burden. This should also be reflected more explicitly in the abstract, discussion and conclusion, because lactate, creatinine and transfusion requirement may largely be markers of global physiological insult rather than renal injury specific predictors.
3. Denominator error: The abstract states that nephrectomy made up 34.6% of renal specific interventions, but the results indicate that nephrectomy was 34.6% of operated patients, while renal specific procedures were performed in only a subset of operated patients. This needs correction, as 38 nephrectomies among 47 renal specific interventions would be a much higher proportion.
4. SNOM conclusion: The revised manuscript is careful not to claim a direct operative versus non operative comparison, which is appropriate. However, the statement that selected patients managed non operatively had acceptable outcomes would be stronger if the authors provided SNOM specific mortality, complications, delayed intervention and failure rates, ideally stratified by AAST grade. Without that, the conclusion should be softened further.
5. Table and supplementary material consistency: The tables need a final technical check. In particular, the note under Table 2 appears to refer to management percentages and renal interventions, although Table 2 presents mechanism by AAST grade. The supplementary table numbering also appears to differ from the manuscript table numbering, which may confuse readers and editors.
6. Definitions: AKI is reported as an important complication and is associated with mortality, but the manuscript should define how AKI was diagnosed. A KDIGO based definition, or the actual operational definition used in the audit, should be stated in the methods.

I suggest a quick, final, minor revision, focused on tightening the statistical language, correcting denominators, defining AKI, adding or softening SNOM outcome claims, and cleaning the tables. The topic is important, the cohort is clinically interesting, and the revised framing is now appropriately cautious for a retrospective single centre audit.

Author Response

We wish to thank the Reviewer for the thorough and constructive evaluation of our manuscript. The comments have materially strengthened both the accuracy of our statistical language and the transparency of our limitations. Below, we provide a point-by-point response to each comment. All changes in the revised manuscript are highlighted in yellow for ease of identification.

 

Point 1: Statistical Interpretation

Reviewer Comment: Change "independent predictors" to "independently associated with mortality in the adjusted model" throughout the manuscript (abstract, results, discussion, conclusion). Language should be more cautious given the absence of ISS and limited adjustment possible.

 

Response: We agree with the Reviewer that the original phrasing overstated the causal implication of our findings. We have replaced all instances of "independent predictors" and "independently predicted" with "independently associated with mortality in the adjusted model" throughout the manuscript. Specifically, the following changes were made:

• Abstract (Results): "were independent predictors of mortality" → "were independently associated with mortality in the adjusted model"
• Abstract (Conclusions): "independently predicted in-hospital mortality in this high-acuity cohort" → "were independently associated with in-hospital mortality in the adjusted model in this high-acuity cohort"
• Section 3.6 (Results – Predictors of In-Hospital Mortality): "three variables remained independent predictors of in-hospital mortality" → "three variables were independently associated with in-hospital mortality in the adjusted model"
• Section 5 (Conclusions): "were identified as independent predictors of in-hospital mortality" → "were independently associated with in-hospital mortality in the adjusted model"
• Section 4 (Discussion): Added a caveat acknowledging that "the absence of Injury Severity Score (ISS) data limited the degree of adjustment possible"

 

Point 2: Residual Confounding

Reviewer Comment: In the abstract, discussion, and conclusion, explicitly state that lactate, creatinine, and transfusion may be markers of global physiological insult rather than renal-specific predictors. Emphasise the limitation due to the absence of ISS (overall injury burden).

 

Response: We thank the Reviewer for highlighting this important interpretive issue. We have added explicit statements about residual confounding and the possibility that these variables reflect global polytrauma severity rather than renal-specific pathology. The following additions were made:

• Abstract (Conclusions): Added: "…however, in the absence of ISS data, these variables may partly reflect the global burden of physiological insult from polytrauma rather than renal-specific predictors of death."
• Section 4 (Discussion): Added: "Importantly, because ISS was unavailable and only three covariates could be entered into the multivariable model, residual confounding by unmeasured overall injury burden is likely; lactate, creatinine, and early transfusion requirement may therefore act as surrogates for the global severity of polytrauma rather than as renal-specific predictors of death."
• Section 5 (Conclusions): Added: "Nevertheless, these associations should be interpreted with caution, as the absence of ISS data means that lactate, creatinine, and transfusion requirement may partly reflect the global physiological insult of polytrauma rather than renal injury-specific pathways."

 

Point 3: Denominator Error in Abstract

Reviewer Comment: The abstract states "Nephrectomy made up 34.6% of surgical renal interventions," but 34.6% represents the proportion of all operated patients (38/110), not of renal-specific interventions. The correct denominator for renal-specific interventions is 47 (38 nephrectomy + 5 renorrhaphy + 4 DJ stent), giving 38/47 = 80.9%. Correct this in the abstract and ensure consistency throughout.

 

Response: We thank the Reviewer for identifying this denominator error. The Reviewer is correct: 34.6% (38/110) represents the proportion of all operated patients who underwent nephrectomy, whereas the proportion of renal-specific interventions should use 47 as the denominator (38 nephrectomy + 5 renorrhaphy + 4 DJ stent = 47). The corrected figure is 38/47 = 80.9%. The following changes were made:

• Abstract (Results): "Nephrectomy made up 34.6% of renal-specific interventions" → "Nephrectomy accounted for 80.9% of renal-specific surgical interventions (38 of 47 patients who underwent a direct renal procedure)"
• Section 3.3 (Results – Management): Rewritten to clarify both denominators: "Renal-specific interventions were performed in 47 patients (42.7% of operated patients) and included nephrectomy in 38 (80.9% of renal-specific interventions), renorrhaphy in 5 (10.6%), and DJ stent insertion in 4 (8.5%)."
• Section 4 (Discussion): "nephrectomy was the most common (34.6%)" → "nephrectomy was the most common (80.9% of renal-specific interventions)"

Note: Table 4 retains 34.6% in the "Renal surgical intervention" rows because those percentages are explicitly calculated from the 110 operated patients as stated in the table footnote. The distinction between the two denominators is now clarified in the text.

 

Point 4: SNOM Conclusion

Reviewer Comment: The conclusion about non-operative management ("favourable outcomes") should either be supported with specific data (SNOM-specific mortality, complications, delayed intervention rates, failure rates, stratified by AAST grade) or softened to acknowledge selection bias. If the data are not available in sufficient detail, soften the language accordingly.

 

Response: We agree with the Reviewer that our original language regarding SNOM outcomes was stronger than the data support. As the audit did not collect SNOM-specific complication rates, delayed intervention rates, or outcomes stratified by AAST grade for the non-operative cohort, we have softened the language throughout to acknowledge these limitations and the inherent selection bias. The following changes were made:

• Abstract (Conclusions): Replaced "Selected patients managed non-operatively had acceptable outcomes, suggesting that SNOM may remain a feasible option…" with "A proportion of patients were managed non-operatively; however, given the inherent selection bias in management allocation and the absence of SNOM-specific outcome data stratified by AAST grade, no conclusions regarding the efficacy or safety of non-operative management can be drawn from this descriptive audit."
• Section 4 (Discussion): Added: "Moreover, this audit does not report SNOM-specific complication rates, failure rates, or outcomes stratified by AAST grade, which precludes any conclusion about the safety or efficacy of non-operative management in this cohort."
• Section 5 (Conclusions): Replaced "Selected patients managed non-operatively had acceptable outcomes, supporting the potential role of SNOM…" with "A proportion of patients were managed non-operatively; however, this retrospective audit lacks the granularity (SNOM-specific complication rates, delayed intervention rates, and outcomes stratified by AAST grade) to draw conclusions regarding the safety or efficacy of non-operative management in this setting, and the inherent selection bias in management allocation must be acknowledged."

 

Point 5: Table Consistency

Reviewer Comment: Table 2 footnote currently refers to management percentages, but Table 2 shows mechanism by AAST grade. Fix the footnote to match the table content. Check and fix supplementary table numbering to match manuscript table numbering. Ensure all table notes are accurate.

 

Response: We thank the Reviewer for noting this inconsistency. Table 2 presents mechanism of injury by AAST grade, but its footnote erroneously referred to management strategies (mentioning DJ stents and SNOM, which are not relevant to this table). The following correction was made:

• Table 2 footnote: Replaced "Row percentages indicate the proportion of patients within each AAST grade managed by each strategy. Abbreviations: AAST, American Association for the Surgery of Trauma; DJ, double-J; SNOM, selective non-operative management. Note: Management percentages within each grade may not sum to 100% as some patients received more than one renal-specific intervention." with "Row percentages indicate the proportion of each mechanism within each AAST grade. Abbreviations: AAST, American Association for the Surgery of Trauma; FFH, fall from height; GSW, gunshot wound; MVC, motor vehicle collision; PVC, pedestrian vehicle collision."

All other table footnotes (Tables 1, 3, 4, 5, and 6) were reviewed and confirmed to be accurate and consistent with table content.

 

Point 6: AKI Definition

Reviewer Comment: Add to the Methods section how AKI was diagnosed. Use KDIGO criteria or state the operational definition used in the audit. Be specific about criteria (e.g., creatinine increase, urine output, need for RRT).

 

Response: We agree that an explicit AKI definition should have been provided. We have added the following operational definition to Section 2.4 (Definitions):

"Acute kidney injury (AKI) was defined operationally as a rise in serum creatinine to ≥ 1.5 times the baseline (or admission) value, a documented need for renal replacement therapy (RRT), or a clinical diagnosis of AKI recorded in the patient notes; formal Kidney Disease: Improving Global Outcomes (KDIGO) staging could not be applied retrospectively owing to the absence of serial hourly urine output data in the source records."

This definition aligns with the creatinine-based component of the KDIGO criteria (Stage 1: ≥ 1.5× baseline within 7 days) while transparently acknowledging that formal staging was not possible due to limitations in retrospective urine output documentation.

 

We believe that these revisions have substantially improved the manuscript and addressed all of the Reviewer’s concerns. We are grateful for the Reviewer’s careful reading and insightful comments, which have enhanced the rigour and clarity of our work.