Microbiological Findings and Clinical Outcomes in Ugandan Patients with Infected Burn Wounds

Round 1

Reviewer 1 Report

Thank you for the opportunity to review the manuscript entitled “Microbiological findings and clinical outcome in Ugandan patients with infected burn wounds” This prospective, observational study describes microbiological surveillance findings and patient outcomes in a resource limited region. All patients with wounds or suspected infection were eligible with the majority being due to burn injury. Although the authors present only present findings from a single center, the findings highlight a troubling concern and describe practices in the LMIC. The manuscript is very well written, concise, and fills a gap in existing literature. It was clear the authors were knowledgeable and experienced. See below my comments following review of this manuscript.

Introduction

-The introduction is succinct and pointedly written. The background provided, regarding resources limitations, is helpful.

Methods

The study setting paragraph is very contextually helpful.

Pg2, line 90: Since the premise of the study is microbial resistance, more detail should be listed than referencing the national Uganda Clinical Guidelines. For my understanding, I pulled the guidance statement. It is nearly 1200 pages. Specifically looking for wounds and burns treatment, antimicrobial prophylaxis and treatment are listed on pg 97 and 101. For wounds, there is a note at the bottom of the bulleted chart to use prophylaxis in very contaminant wounds and to provide antimicrobial treatment for infected wounds. No specifics on what to prescribe. For severe burns, the guidance recommends treatment with benzylpenicillin +/- gentamicin if signs and symptoms of infection are present. If this guidance (or something different) was followed, the detail will better inform readers of prescribing practices, exposure, and potential of subsequent resistance. 79% of included patients were on antibiotics upon arrival/transfer. The Discussion mentions ceftriaxone being in line with national guidelines. Some clarity would help the authors make their point.

Study procedures are well described. Consent was obtained and documented. Which cultures obtained and use of Kirby Bauer noted. Patient follow up and definition utilized for outcome described.

Data collection, management, and statistical analysis generally described. Means are quoted in the methods, but medians utilized in the results. It is likely the data presented as medians was nonparametric and appropriately presented as median (IQR). Please add detail for testing normality and decision to present as median (IQR).

Results

Pg 4, line 148: Days of antimicrobial therapy (including prior to transfer) at enrollment (day of culture) would be a good detail to add and would aid the reader in applying the findings.

Antibiotics are mostly not necessary when source control is able to be obtained quickly. Adding days until complete wound excision will benefit reader perspective and may highlight part of the narrative.

Pg 4, line 171: Please describe the definition used for nosocomial.

Discussion

 Pg 8, lines 253-255: The sentence describing gender and age is repeated from the results section and not needed.

Pg8, line 266-275: Very well stated and important paragraph.

Pg9, line 279-280: Very important. To highlight further, the final sentence in the paragraph may be improve by add this detail. Perhaps “..further exploration and improved infection prevention practices.”

Pg9, lines 284-289: excellently written

Pg9, line 294: What might the mortality rate be had AMS testing not been performed and ABx able to be selected according to the result?

Conclusions

Pg9, line 309-310: After mention empiric ceftriaxone, it may be of benefit to add a sentence for a final highlight on the need to evaluate over-the-counter access to ABx.

Author Response

Reviewer 1:
Methods
For severe burns, the guidance recommends treatment with benzylpenicillin +/- gentamicin if signs and
symptoms of infection are present. If this guidance (or something different) was followed, the detail will
better inform readers of prescribing practices, exposure, and potential of subsequent resistance. 79% of
included patients were on antibiotics upon arrival/transfer. The Discussion mentions ceftriaxone being in
line with national guidelines. Some clarity would help the authors make their point.
 We appreciate this point. The National Guidelines do in fact suggest using a combination of
benzylpenicillin +/- gentamicin. We have added this information to the revised manuscript
(page 8, lines 270-271). The clinicians on the specialized burn unit have made the experience
that most pathogens identified in infected burn wounds are resistant to Penicillin, which was
also confirmed in our study. Looking at clinical practice, Ceftriaxone is widely used because it is
easily accessible, available free of charge and considered highly potent. Our results showed high resistance to this antibiotic, which underlines the need for microbiological diagnostic and
structured antibiotic stewardship measures in order to improve patient care. We have added
these thoughts to the discussion for more clarity (page 8, lines 273-277) and thank you for
pointing this out to us.
Means are quoted in the methods, but medians utilized in the results. It is likely the data presented as
medians was nonparametric and appropriately presented as median (IQR). Please add detail for testing
normality and decision to present as median (IQR).
 We edited the statement in the methods to "median and interquartile range" (page 3, lines
134-135). No normality test was done as the distribution of age is empirically non-normal.
Results
Pg 4, line 148: Days of antimicrobial therapy (including prior to transfer) at enrollment (day of culture)
would be a good detail to add and would aid the reader in applying the findings. Antibiotics are mostly
not necessary when source control is able to be obtained quickly. Adding days until complete wound
excision will benefit reader perspective and may highlight part of the narrative.
 Thank you for pointing this out. As most patients were started on antibiotics before transfer to
the specialized unit, and documentation is very scarce in this setting, the type and duration of
antibiotic treatment before care on the unit could not be assessed. We have added this point to
the discussion on the revised manuscript (page 8, lines 265-266).
Pg 4, line 171: Please describe the definition used for nosocomial.
 Nosocomial or healthcare-associated infections were defined according to the World Health
Organization (WHO). We have added the WHO definition and corresponding reference to the
revised section on study procedures (page 3, lines 105-106).
Discussion
Pg 8, lines 253-255: The sentence describing gender and age is repeated from the results section and not
needed.
 We removed this sentence from the discussion.
Pg9, line 279-280: Very important. To highlight further, the final sentence in the paragraph may be
improve by add this detail. Perhaps “... further exploration and improved infection prevention
practices.”
 Thank you very much for this helpful suggestion, which we added to the sentence mentioned
above (page 9, lines 299-300).
Pg9, line 294: What might the mortality rate be had AMS testing not been performed and ABx able to be
selected according to the result?
 That is a very good question. Undoubtedly, our diagnostic and treatment interventions saved
lives. Unfortunately, the unit does not have concise mortality data collection and reporting
procedures in place.
Conclusions
Pg9, line 309-310: After mention empiric ceftriaxone, it may be of benefit to add a sentence for a final
highlight on the need to evaluate over-the-counter access to ABx.
 Very important point, which we discuss in the revised manuscript (page 9, line 301 onwards)

Reviewer 2 Report

It would be of interest to have the TBSA values for the burns cohort.

Line 169: what criteria if any was used to identify wound infections. Apart from fever where there any other objective signs of sepsis? In the 22.9% of those with only fever, were other potential sources of sepsis considered.

Line 176: what were those on antibiotics being treated for? Was it for the? wound infection.

Overall a good paper outlining AMR problem and the need for surveillance.

Author Response

It would be of interest to have the TBSA values for the burns cohort.
 Thank you for this comment. Unfortunately, we did not collect additional information on TBSA
within the cohort but focused on the degree of burns (table 1) as we had anticipated the
degree to be most relevant concerning the extent of wounds. However, realizing information
on TBSA would have been important to further classify our cohort we have added this point to
the revised manuscript (page 4, line 147).
Line 169: what criteria if any was used to identify wound infections. Apart from fever where there any other objective signs of sepsis? In the 22.9% of those with only fever, were other potential sources of sepsis considered.
 Clinical signs of wound infection included pus, conversion from partial- to a full-thickness
wound, rapidly extending cellulitis surrounding the burn injury, eschar separation, and tissue
necrosis. We added this information to the revised materials and methods section (page 2, lines
77-79).
 Apart from temperature, heartrate and blood pressure were documented upon enrollment. A
total of 51 (36.4%) participants had a BP >100 bpm, 7 (5%) a systolic BP < 100mmHg, and 3
(2.1%) had both. We added this information to the revised results section (page 4, lines 178-
180).
 Febrile patients on the unit without clinical signs of wound infection were assessed for malaria
and active tuberculosis as alternative reasons for fever prior to inclusion. Upon inclusion, the
further evaluation of causes for fever were done according to best clinical care practice.
Additional microbiological diagnostics (e.g. urine, spinal fluid, etc.) were not part of the study
evaluations. I hope this answers the question.
Line 176: what were those on antibiotics being treated for? Was it for the wound infection.
 Yes, study participants on antibiotics at enrollment received these in the context of the burn
injuries. In this setting, most patients with burn injuries are started on antibiotic treatment
prophylactically and are then transferred to the specialized burn unit for expert care. Thus,
wound infection may not have been clinically apparent in all cases upon starting antibiotic
treatment. We have added this information of the revised manuscript (page 5, line 185).