Procalcitonin in the Post-Operative Burn Patient
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThank you for the opportunity to read and comment of this paper. My congratulations to the authors for systematically following up their patients and reporting their results.
However, the manuscript does not meet the standards of scientific scrutiny. The population is too small to make any robust conclusions (already identified by authors). If PCT is to be a diagnostic tool, a ROC analysis including the sens, spec, PPV andNPV should be reported.
Another problem is the definition of 'sepsis'. How was this defined? When I read the paper what the study is measuring as an outcome is bacteremia, detected as positive blood cultures.
Perhaps this paper can be included as a letter/correspondence as a purely descriptive study. Studies from Africa are rare, and results such as these would be important to report but not in the present form.
Author Response
Please see attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for Authors
The authors have shown a higher than normal PCT that appears to be related to burn injury and/or surgery rather than sepsis in their study population, which has a high incidence of positive blood cultures. As such this is a useful area of research.
As the routine for ICU patients is to have daily PCT levels, it seems that there would be a significant amount of retrospective data available that could inform this study.
Abstract
Line 22. A ‘higher numerical threshold’ for what?
M&M
Line 72. what is a fogging machine
Line 88. Study design – what is meant by ‘convenience sampling’ and how many eligible patients were not recruited during the study period?
Results
Line 139 – what is temporary skin substitute cover – if not porcine xenograft etc?
Line 146 -The median day post burn injury for surgery was day 5 post-surgery. Do you mean post burn?
There are two table 1s
Discussion
The authors point out that BC results are a proxy for sepsis, and the clinical diagnosis of burn related sepsis takes other factors into account. The study would be improved by including information about these factors in the results, even though analysis may be precluded. This is especially so, as the authors imply that a positive blood culture may not be an indication of serious infection (with or without sepsis).
As one of the aims of this study was to identify a possible relationship between PCT and positive blood cultures, on a background of interest in PCT as a predictor of sepsis, it would be helpful to present the PCT and which days blood culture results were positive or negative.
Li, A. , Moussa, A. , Gus, E. , Paul, E. , Yii, E. , Romero, L. , Lin, Z. , Padiglione, A. , Lo, C. , Cleland, H. & Cheng, A. (2022). Biomarkers for the Early Diagnosis of Sepsis in Burns. Annals of Surgery, 275 (4), 654-662. doi: 10.1097/SLA.0000000000005198.
Author Response
Please see attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsAlthough the concept is not entirely original, as the authors highlight, data collection and completion in a middle-income setting with a high rate of bacteraemia is novel and is to be commended. The highlighted finding is clinically significant and statistically robust. I suggest this finding is expanded on to increase the value of, and interest in, the paper by considering the following amendments:
In description of the cohort (paragraph from line 130), consider adding a summary of time between injury and first surgery. This is important context with practice evolving so that some burn centres routinely undertake early debridement while others use the historical approach of delaying 1-2 weeks. It also reflects delays in presentation due to challenges accessing care in large catchment areas. It is of particular interest when considering the early inflammatory response due to injury itself is expected to peak around day 3-4 prior to a 'second hit' from surgery that the authors refer to.
Line 151 states that A. baumanii was the most common pathogen on day 0 but is not listed under Day 0 in table 1 - please review. Also consider adding the frequency of each pathogen to table 1 and include the frequency of polymicrobial bacteraemia.
Consider splitting the line in figure 1 to show two groups based on positive/negative culture at day 3. It would be interesting to see the trajectories of these separately.
The paragraph starting at line 163 could be clearer. From the results presented in table 4, it seems that the serum PCT levels at day 0 were compared between those that had positive and negative blood cultures at day 0 and then in a second analysis, serum PCT levels at day 3 were compared between those that had positive and negative blood cultures at day 3. If that is correct, consider removing reference to "both day 0 and day 3" and specific the two analyses separately.
Did PCT day 2 values also predict day 3 cultures? Could this analysis be presented? At lines 219-220, serial measurement of PCT is recommended suggesting a trend or trajectory is more accurate than single measurement - does the change in value between day 2 and day 3 predict day 3 culture?
The relevance of coagulase-negative staphylococci is discussed (line 191). I suggest a post-hoc sensitivity analysis is performed to explore this by including those with Staph epi and other coag-neg staph in the "negative" group. Is there still significant (or even greater) separation between groups? These results could be presented briefly in text only and would not need to be tabulated.
Line 228 states that a PCT threshold could not be established - was a receiver-operator curve (ROC) analysis completed? If yes, this result should be presented even if the conclusion is that the small sample size limits interpretation.
Lines 234-237 suggest that additional analyses were performed but not reported. New analyses should not be presented in the discussion - either these results should be in methods and results or this line should be removed.
Thank you for the opportunity to review this manuscript which adds some useful interpretation of the potential utility of PCT in this challenging cohort.
Comments on the Quality of English LanguageMinor edits required:
line 138 missing punctuation.
Line 156 figure -> Figure
Table 1 Acitenobacter -> Acinetobacter
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 4 Report
Comments and Suggestions for AuthorsIn their paper, Masole et al. present a monocentric prospective evaluating the relationship between serum Procalcitonin (PCT) levels surgical debridement in a sample of 34 adult burn patients form a tertiary African hospital. There are some clear limitations in this study, namely the small dimension of the sample and its heterogeneity, which does not allow solid statistical conclusions to be drawn. On the other hand, its conclusions take in account not only clinical data and PCT levels but also the subjacent microbiological findings which in some way reinforces its accuracy. In spite of it, this is an interesting work, however, in my point of view, there are some remarks and suggestions that must be put, in order to improve the article
A – Major Remarks
1. The Authors do not state the number of days spent between the accident provoking the burns and the admission to the Burn Unit, neither refer if the patients were primary admissions or transferred from other hospital facilities. Because the number of days before admission and previous hospitalization have implications in the risk for burns colonization and sepsis, which may be reflected in the levels of PCT, can this missing information be added to the manuscript?
2. As the Authors are aware, analysing PCT kinetics allows more secure insights about sepsis evolution. Previous literature points to a peak of PCT levels at the 2nd-3rd day after surgery followed by a decrease in the absence of sepsis or by an increment in the case of non-treated sepsis. Stopping their analysis at day 3, the Authors impoverish the solidity of their study. Can the data till day 5 be added to the paper?
3. Usually, in the absence of risk factors in the first five days post-burn, the burn is colonized by commensal Gram-positive bacteria from surrounding skin. However, in this study the first microorganism to be found was Acinetobacter, which, in normal conditions would start to appear later. Can the Authors explain this finding? Could it be related to risks factors or delay in admission to the Burn Unit?
4. Naturally, the diagnosis of burn sepsis is not easy in burn patients and ABA criteria have been recently questioned by Surviving Sepsis After Burns Campaign (see publication in Burns journal, May 2023). Paying attention to the low number of patients, would it be possible for the Authors to review patients’ data and compare their classification according to ABA versus SSABC?
B – Minor Remarks
1. Tables should be reformatted in size, using the same design if possible.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
Comments and Suggestions for AuthorsThank you for the opportunity to view the revised manuscript. The revisions and commentary address the questions previously raised. Thank you for undertaking the ROC analysis. Ultimately, I am not sure that this adds to the manuscript and am ambivalent about inclusion of the ROC curves which could be removed if the authors wish to have a more concise presentation.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 4 Report
Comments and Suggestions for AuthorsIt is really a pity you could not extend your analysis
Author Response
Please see attachment.
Author Response File: Author Response.pdf
