Dry Season Melioidosis in the Tropical North of Australia

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for the opportunity to review this manuscript. This is a large-scale, very interesting study on the characteristics of melioidosis during the dry season in the Northern Territory of Australia.

Minor comments:

Abstract (l22): “A higher proportion of dry season patients …”. Consider to add information regarding wet season patients. That would make it easier to interpret the data.

Abstract (Conclusion, l28): The conclusions focus only on what is needed for further studies, rather than on the results of this work. I suggest inserting a sentence regarding this point and also what consequences are drawn from this study.

P3, l133 & p4, l168 & table 1: “1,520” instead of “1520” or reverse notation throughout the entire document.

P5, table 2: Data from “Documented specific potential exposure event” are missing for the wet season. Please include these data or explain why not.

P2, Materials and Methods: Data regarding patients were extracted from the DPMS, published in 2021 and collected from patients between 1989 and 2019. This study includes patients from 1990 t0 2024. The reader might be interested how the data were collected continuously and from 2019 following. Also, in general, the questionnaire behind the study regarding risk factors, information regarding exposure to the agent, travel history, underlying disease, clinical data, etc. would be very interesting. Please include this questionnaire to at least the supplementary data.

P13, l341 & P14, l375: “At least 3% of cases remained unexplained”: Did you perform whole genome sequencing of the bacterial isolates of these patients (if available)? Do the sequences match with isolates from Australia? As the “time bomb” melioidosis can be caused by a long-ago infection event, the patients´ travel history might be further back in the past. Whole genome sequencing data might support your hypothesis and exclude gaps in the travel history.

Author Response

Please see the attached document. 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

General Comments

The authors present a comprehensive retrospective analysis of dry-season melioidosis over a remarkable 35-year period, using data from the Darwin Perspective Melioidosis Study. The study addresses an important and underexplored aspect of melioidosis epidemiology by focusing on cases occurring outside the expected rainy season and provides valuable insights into the interaction between environmental, climatic, and host-related factors. The long-term surveillance, robust clinical dataset, and integration of meteorological analyses represent major strengths of the manuscript. Overall, the manuscript is well written, methodologically sound, and the conclusions are generally supported by the presented data. I believe this work will be of considerable interest to readers of Pathogens. The most important revisions are to clarify the practical implications of the findings and strengthen the rationale for the dry-season focus. Only minor revisions are suggested to further improve clarity and strengthen the interpretation of the findings.

Minor Comments

1. Abstract. The abstract is informative and well structured. However, it would benefit from briefly mentioning the practical implications of the findings, particularly regarding year-round prevention strategies and surveillance, and this should be prioritised.
3. Introduction. The introduction clearly summarises the current knowledge. Nevertheless, the rationale for specifically investigating dry-season cases could be strengthened by more explicitly highlighting the existing knowledge gap regarding environmental persistence and transmission during prolonged dry periods, and this should be prioritised.
5. Materials and Methods. The criteria for identifying "potential infecting events" rely largely on retrospective clinical documentation. Although acknowledged later in the manuscript, a brief explanation of how exposure histories were collected and whether standardised questionnaires were used would improve reproducibility and should be prioritised.
7. Weather Analysis. The use of generalised additive models and conditional logistic regression is appropriate. However, readers may benefit from a brief explanation of why cumulative rainfall over 2 and 4 weeks was selected rather than alternative exposure windows, and this should be prioritised.
9. Results. The comparison between dry- and wet-season presentations is clear. Since 23% of dry-season patients had no identifiable risk factors, the authors may briefly speculate that host immune factors or unrecognised environmental exposures could also contribute to disease occurrence.  Figures 4–6 contain valuable information but appear relatively complex. Improving font size, labels, or figure legends would enhance readability without altering the scientific content, and these should be prioritised.

11. Discussion. The discussion appropriately addresses anthropogenic environmental factors such as irrigation and construction. It may also be worthwhile to briefly discuss how increasing urbanisation and land-use changes could influence future melioidosis epidemiology in endemic regions, and this should be prioritised. Climate change is appropriately mentioned, but the discussion could be slightly expanded by emphasising how increasing climatic variability, rather than only increased rainfall, may contribute to atypical seasonal presentations, and this should be prioritised.
The authors are encouraged to include the following reference and to discuss how its findings relate to, and differ from, those reported in the present manuscript.
Ayanlade A, Sergi C, Ayanlade OS. Malaria and meningitis under climate change: initial assessment of climate information service in Nigeria. Meteorol Appl. 2020;27:e1953. https://doi.org/10.1002/met.1953
Ayanlade, A., Nwayor, I.J., Sergi, C. et al. Early warning climate indices for malaria and meningitis in tropical ecological zones. Sci Rep 10, 14303 (2020). https://doi.org/10.1038/s41598-020-71094-8
The limitations section is balanced. It may additionally acknowledge the potential for recall bias regarding patient-reported exposure histories, particularly when attempting to identify specific infecting events retrospectively, and this should be prioritised

Author Response

Please see the attached document

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Based on 35 years (1989–2024) of data from the Darwin Prospective Melioidosis Study (DPMS) encompassing 1,520 cases, this study systematically describes the epidemiological and clinical features of melioidosis during the dry season (May–October) in tropical northern Australia, with comparisons to the wet season. The study employed chi-square tests, conditional logistic regression (to examine rainfall exposure 2 or 4 weeks prior to case-days relative to matched control-days), and generalised additive models (to assess intra-seasonal variation in dry-season case probability and annual incidence trends). Results showed that dry-season cases accounted for 21.4% (325 cases) of all cases. Compared with the wet season, the dry season had a lower proportion of acute cases (73% vs 93%) and higher proportions of chronic, latency activation and relapse cases. Clinically, acute dry-season cases less frequently presented with pneumonia (30% vs 58%), but more frequently with cutaneous melioidosis (22% vs 8%). A higher proportion of dry-season patients had no known risk factors (23% vs 14%), and mortality was lower (8% vs 11%, p=0.01). Some dry-season cases had rainfall exposure 2–4 weeks before diagnosis, and anthropogenic factors such as irrigation and construction may have contributed to environmental persistence of B. pseudomallei in urban settings, though approximately 3% of dry-season cases remained without a clear explanation for infection. The study concludes that dry-season melioidosis is not uncommon and that prevention strategies should be maintained throughout the year, with particular attention to populations exposed to gardening, construction, irrigation and untreated water. Future research should incorporate timely exposure history collection, environmental sampling and clinical–environmental strain genotyping to better define transmission routes in the dry season.

1. It is suggested that a concluding paragraph be added at the end of the Introduction to summarise the core novel contributions of this study.
2. This study excluded cases from 1–20 May based on a 21day incubation period. Please elaborate on why 20 May was chosen as the cutoff date.
3. Rainfall analysis used data from only a single weather station (Darwin Airport). A single station may not represent the actual exposure locations of patients. Please discuss the potential impact of spatial misclassification or add a limitation regarding this aspect.
4. in the Results section, differences between dry and wetseason cases are currently assessed mainly using pvalues, which reflect only statistical significance. It is recommended to supplement these with additional effect size or other appropriate metrics.
5. Acute melioidosis was defined as symptom duration of less than two months (maximum 60 days) before diagnosis, yet the rainfall exposure analysis examined only rainfall in the 2 or 4 weeks preceding the diagnosis date. Please provide an explanation or rationale for this choice.
6. The case classification flow in Figure 6 needs further clarification. Please specify the hierarchical relationships and arithmetic logic (addition/subtraction) among the numbers 325, 283, 282, 126, 98, 47, and 10.
7. Since a GAM was used, please include the full model formula and the estimated effective degrees of freedom for each smoothing term in the Methods section.
8. In the conditional logistic regression model with matched casedays and controldays, control days were required to be at least 7 days apart, but the rationale for this 7day interval is not explained. Please add the justification for this choice.
9. In Table 2, under “Clinical Presentation”, the “Other” category includes 7 dryseason and 24 wetseason cases, but the specific presentations are not listed, and the pvalue is given as “N/A”. Please list the contents of “Other” in the table footnote.
10. In Table 2, the variable “documented specific potential exposure event” is presented only for dryseason cases, with no corresponding data for wetseason cases. Please clarify why this variable was not compared across seasons, or move it to the descriptive results section.
11. This study spans 35 years, during which case documentation, exposure history collection methods, diagnostic techniques, and clinical management strategies may have changed. Please discuss the potential information bias and diagnostic bias arising from this long time span.
12. In the Results section, please briefly compare the findings from the 2week and 4week rainfall exposure windows, and state whether they are consistent. If consistent, a concise statement suffices; if not, please discuss the reasons and the implications for your conclusions.

Author Response

Please see the attached document

Author Response File: Author Response.pdf