Clinical, Microbiological, and Hematological Characteristics of Pediatric Brucellosis in Saudi Arabia: A Single-Center Retrospective Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

1. "Retro-spective" should be changed to "Retrospective"
2. Introduction, the innovative aspects of this study (such as the first focus on the diagnostic value of hematological characteristics in children in the Hayel region) were not fully elaborated. It is suggested to supplement the comparison of differences with previous studies in other regions of Saudi Arabia (such as the data from the Najiran region), and to clarify the regional uniqueness of this study.
3. Materials and Methods, the processing procedures for cases with negative serology but high clinical suspicion (such as whether PCR testing should be conducted as a supplement) were not described. It is necessary to provide the confirmation protocols for cases with negative serology. Additionally, the age-based classification standards for indicators such as anemia and leukopenia (in accordance with WHO or local guidelines) were not listed.
4. “The dilutions of the tubes were 1:20, 1:40, up to 1:10240”, 1:10240, correction required.
5. As the subsequent analysis also involves t-tests, chi-square tests, etc., "with a fairly even distribution across age groups (Table 1)." This conclusion suggests that objective data should be provided through a homogeneity of variance test.
6. Patients usually present with multiple clinical symptoms such as fever, musculoskeletal involvement, and fatigue. Figure 1A suggests using a Venn diagram instead of a bar chart, which can more clearly represent the patient's clinical manifestations. Additionally, the sum of percentages in a bar chart is often greater than 100%, which can lead to misunderstandings. The same issue is present in Figure 1D.
7. Table 2 uses the unpaired t-test, while Table 3 employs ANOVA. No explanations are provided in the statistical analysis section. Additionally, the statistical analysis is overly simplistic and should clearly indicate which statistical method was used to analyze which data.
8. There are too many tables. Table 3 can be placed in the Supplementary Table.
9. For the ROC curve (Figure 2), the authors should consider supplementing the ROC data of multiple indicators to evaluate the combined diagnostic efficacy of multiple indicators. This is also reflected in Table 6, where the combined diagnostic efficacy is deemed important for this manuscript.
10. Discussion, the mechanisms underlying hematological abnormalities (such as bone marrow suppression and immune destruction) were not explained. The pathological and physiological pathways by which Brucella causes cell reduction could be supplemented.
11. "Acknowledgments" and "Funding" are repeated, "Ethical Considerations" and "Institutional Review Board and Informed Consent Statement " are also repeated
12. The overall publication years of the references are relatively old. It is recommended to include more from the last three or five years (more than 50%), to better reflect the cutting-edge nature of this research.

Comments on the Quality of English Language

None

Author Response

Reviewer-1:

Comments and Suggestions for Authors

1. "Retro-spective" should be changed to "Retrospective"

Author Response: Thank you for your observation. The word was automatically hyphenated due to line formatting at the end of a line in the manuscript file and does not represent a spelling error.

2. Introduction, the innovative aspects of this study (such as the first focus on the diagnostic value of hematological characteristics in children in the Hayel region) were not fully elaborated. It is suggested to supplement the comparison of differences with previous studies in other regions of Saudi Arabia (such as the data from the Najiran region), and to clarify the regional uniqueness of this study.

Author Response: Thank you for this valuable suggestion. We have revised the Introduction to better highlight the novelty and regional significance of our study.

3. Materials and Methods, the processing procedures for cases with negative serology but high clinical suspicion (such as whether PCR testing should be conducted as a supplement) were not described. It is necessary to provide the confirmation protocols for cases with neg3. serology. Additionally, the age-based classification standards for indicators such as anemia and leukopenia (in accordance with WHO or local guidelines) were not listed.

Author Response: Thank you for this valuable comment. We have revised the Materials and Methods section to clarify the diagnostic approach used for patients with high clinical suspicion of brucellosis despite initially negative serological results. During the study period, PCR testing was not routinely available or performed. In patients with a clinical presentation suggestive of brucellosis and negative initial serology, the Standard Tube Agglutination Test (STAT) was repeated after four weeks. Final case classification was based on compatible clinical findings together with culture positivity and/or serological evidence, in accordance with established CDC case definitions.

We have also clarified the criteria used to define hematological abnormalities. Anemia was defined according to age-specific World Health Organization (WHO) hemoglobin cutoff criteria. Leukopenia and thrombocytopenia were defined according to age-specific reference ranges used by the hospital laboratory. Pancytopenia was defined as the simultaneous presence of anemia, leukopenia, and thrombocytopenia.

4. "The dilutions of the tubes were 1:20, 1:40, up to 1:10240", 1:10240, correction required.

Author Response: Thank you for your observation. We rechecked the dilution protocol described in the manuscript and verified that the reported serial two-fold dilutions (1:20, 1:40, 1:80, ... up to 1:10240) are correct according to the testing procedure used. Therefore, the text has been retained as originally written.

5. As the subsequent analysis also involves t-tests, chi-square tests, etc., "with a fairly even distribution across age groups (Table 1)." This conclusion suggests that objective data should be provided through a homogeneity of variance test.

Author Response: We agree that the phrase "fairly even distribution" may be interpreted as a statistical conclusion. Since this statement was intended only as a descriptive summary of the age distribution and not as the result of a formal statistical comparison, we have revised the text to present the age-group frequencies and percentages without using subjective terminology. Homogeneity of variance testing was applied where appropriate for inferential analyses involving parametric tests. 

6. Patients usually present with multiple clinical symptoms such as fever, musculoskeletal involvement, and fatigue. Figure 1A suggests using a Venn diagram instead of a bar chart, which can more clearly represent the patient's clinical manifestations. Additionally, the sum of percentages in a bar chart is often greater than 100%, which can lead to misunderstandings. The same issue is present in Figure 1D.

Author Response: We thank the reviewer for this valuable suggestion. We agree that Venn diagrams can effectively illustrate overlapping clinical manifestations. However, because five clinical variables were evaluated simultaneously, the resulting overlap patterns would become highly complex and potentially difficult to interpret, particularly given the relatively small sample size.

Therefore, we retained the bar chart format because it provides a clearer and more reader-friendly summary of the frequency of individual symptoms and treatment patterns. To avoid possible misunderstanding regarding percentages exceeding 100%, we have clarified in the figure legend that patients could present with multiple symptoms and receive multiple antimicrobial agents simultaneously; therefore, the percentages are not mutually exclusive and may exceed 100%.

7. Table 2 uses the unpaired t-test, while Table 3 employs ANOVA. No explanations are provided in the statistical analysis section. Additionally, the statistical analysis is overly simplistic and should clearly indicate which statistical method was used to analyze which data.

Author Response: Thank you for this valuable comment. We have revised the Statistical Analysis subsection to provide a more detailed description of the statistical methods used and their specific applications.

8. There are too many tables. Table 3 can be placed in the Supplementary Table.

Author Response: Thank you for this suggestion. We carefully considered moving Table 3 to the Supplementary Materials. However, we chose to retain it in the main manuscript because it presents the comparison of hematological parameters among different Brucella species, which is directly relevant to one of the study objectives and supports the interpretation of the microbiological and hematological findings. The table demonstrates the absence of significant species-specific differences and provides important context for the subsequent discussion. Therefore, we believe that its inclusion in the main text improves the completeness and readability of the results.

9. For the ROC curve (Figure 2), the authors should consider supplementing the ROC data of multiple indicators to evaluate the combined diagnostic efficacy of multiple indicators. This is also reflected in Table 6, where the combined diagnostic efficacy is deemed important for this manuscript.

Author Response: Thank you for this insightful suggestion. We agree that evaluating the combined diagnostic performance of multiple hematological parameters may provide additional clinical information. However, the present study included a relatively small sample size, particularly in terms of the number of Brucella-positive cases. Developing a multivariable diagnostic model based on several hematological parameters in these circumstances could lead to overfitting, unstable parameter estimates, and limited generalizability. Therefore, we chose to perform and report individual ROC analyses, which we believe provide a more statistically robust assessment of the diagnostic performance of each parameter within the constraints of the available dataset. We have acknowledged this limitation in the revised Limitation section and suggested that larger prospective studies should evaluate combined diagnostic models.

10. Discussion, the mechanisms underlying hematological abnormalities (such as bone marrow suppression and immune destruction) were not explained. The pathological and physiological pathways by which Brucella causes cell reduction could be supplemented.

Author Response: Thank you for this valuable suggestion. We agree that a more detailed discussion of the pathophysiological mechanisms underlying hematological abnormalities would strengthen the manuscript. Accordingly, we have expanded the Discussion section to describe the potential mechanisms by which Brucella infection may lead to anemia, leukopenia, and thrombocytopenia, including bone marrow suppression, hypersplenism, immune-mediated peripheral destruction of blood cells, chronic inflammatory responses, and, in rare cases, hemophagocytic activity. Relevant references have also been added to support these explanations.

11. "Acknowledgments" and "Funding" are repeated, "Ethical Considerations" and "Institutional Review Board and Informed Consent Statement " are also repeated

Author Response: Thank you for identifying these redundancies. We have carefully reviewed the manuscript and removed the duplicated information.

12. The overall publication years of the references are relatively old. It is recommended to include more from the last three or five years (more than 50%), to better reflect the cutting-edge nature of this research.

Author Response: Thank you for this valuable suggestion. We agree that inclusion of recent literature strengthens the manuscript and better reflects current developments in the field. Accordingly, we have updated the manuscript by incorporating several recent publications.

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript is a precious report on brucelosis which is relevant to approach to diagnosis, though the frequency of the cases might be low. However, this manuscript has fololowing concerns to be considered for extensive revision.

1. Though athors described in Introduction, the objectives of this study is still unclear. This should be clearly written.
2. Authors used two methods to diagnose brucellosis: 1) blood cultre, and 2) tube agglutination test (serological test). However, what is the definition of the diagnosis of brucellosis? Both tests positive, or positive for either of the tests?
3. In this study, 38 patients were diagnosed as brucellosis. What was the diagnosis criteria? Among 38 samples, 16 samples were culture-positive. Remaining 22 sampes were not from brucellosis?
4. Table 2, 3: add "N", number of samples. 
5. Table 2, Did "Brucella-negative" mwan culture negative?
6. Table 2, 4: In these, authors compared between "Brucella-negative" and "brucella-positive". Did brucella-positive mean culture-positive? If so, authors did  comparison between culture-positive and culture-negative. These are confusing for readers. 
7. Authors did this study in Saudi Arabia. What findings were unique to Saudi Arabia, compared with those of other countries?  This shoud be an important point to be described.  

Author Response

This manuscript is a precious report on brucelosis which is relevant to approach to diagnosis, though the frequency of the cases might be low. However, this manuscript has fololowing concerns to be considered for extensive revision.

Though athors described in Introduction, the objectives of this study is still unclear. This should be clearly written.

Author Response: Thank you for this valuable suggestion. We have revised the final paragraph of the Introduction to clearly state the study objectives (Lines 117–130).

Authors used two methods to diagnose brucellosis: 1) blood cultre, and 2) tube agglutination test (serological test). However, what is the definition of the diagnosis of brucellosis? Both tests positive, or positive for either of the tests?

Author Response: Thank you for this important comment. In this study, positivity of both tests was not required for inclusion. Patients were considered to have brucellosis if they had compatible clinical manifestations together with either (1) isolation of Brucella spp. from blood culture or (2) a Standard Tube Agglutination Test (STAT) titer ≥1:160. This approach is consistent with CDC case definitions, in which culture-positive cases are considered confirmed cases, and clinically compatible patients with significant serological titers are classified as probable cases. We have clarified these diagnostic criteria in the Materials and Methods section.

In this study, 38 patients were diagnosed as brucellosis. What was the diagnosis criteria? Among 38 samples, 16 samples were culture-positive. Remaining 22 samples were not from brucellosis?

Author Response: Thank you for this important comment. All 38 patients included in the study were diagnosed with brucellosis based on compatible clinical manifestations together with positive serological findings (Standard Tube Agglutination Test [STAT] titer ≥1:160). Blood cultures were performed in 28 patients, of whom 16 had culture-confirmed brucellosis, and 12 had negative culture results despite positive serology and compatible clinical findings. In the remaining 10 patients, blood cultures were not performed, and the diagnosis was established based on clinical and serological criteria. Therefore, all 38 patients were considered to have brucellosis and met the study inclusion criteria.

Table 2, 3: add "N", number of samples.

Author Response: Thank you for this suggestion. The sample sizes (N) for each comparison group have now been added to the column headings of Tables 2 and 3 to improve clarity and facilitate interpretation of the results.

Table 2, Did "Brucella-negative" mwan culture negative?

Author Response: Thank you for this important observation. We agree that the previous terminology could be misleading because all patients included in the study had brucellosis. Therefore, we have replaced the terms "Brucella-positive" and "Brucella-negative" with "Culture-confirmed brucellosis" and "Non-culture-confirmed brucellosis," respectively. The latter group included 12 patients with negative blood cultures and 10 patients in whom blood cultures were not performed but who fulfilled the clinical and serological diagnostic criteria for brucellosis.

Table 2, 4: In these, authors compared between "Brucella-negative" and "brucella-positive". Did brucella-positive mean culture-positive? If so, authors did  comparison between culture-positive and culture-negative. These are confusing for readers.

Author Response: Thank you for this important observation. We agree that the previous terminology could be confusing. In our analyses, the "Brucella-positive" group referred to patients with culture-confirmed brucellosis (N = 16). The comparison group consisted of patients who met the clinical and serological diagnostic criteria for brucellosis but did not have culture confirmation (N = 22), including 12 patients with negative blood cultures and 10 patients in whom blood cultures were not performed. To improve clarity, we have replaced the terms "Brucella-positive" and "Brucella-negative" throughout the manuscript with "Culture-confirmed brucellosis" and "Non-culture-confirmed brucellosis," respectively.

Authors did this study in Saudi Arabia. What findings were unique to Saudi Arabia, compared with those of other countries? This shoud be an important point to be described.

Author Response: Thank you for this valuable comment. We agree that the regional relevance of our findings should be more clearly emphasized. Although the hematological abnormalities observed in this study, including anemia, leukopenia, and neutropenia, are not unique to Saudi Arabia and have been reported in pediatric brucellosis cohorts from other endemic countries, our findings provide important Saudi-specific data from the Hail region, where published information remains limited. The predominance of Brucella melitensis, the continued burden of pediatric brucellosis, and the diagnostic utility of routine hematological parameters in an endemic setting are particularly relevant to Saudi Arabia.

Reviewer 3 Report

Comments and Suggestions for Authors

Comments:

The author in the manuscript addresses the demographic, clinical, microbiological, treatment, and hematological characteristics of patients with pediatric brucellosis in the Hail region of Saudi Arabia and to assess the diagnostic value of hematological parameters. The study is important but there are certain areas that need to be addressed before the paper gets accepted.

• The major issue with the manuscript lies in data interpretation. The author uses n=38 but for analysis the authors are using Logistic regression, MANOVA ad ROC analysis which is clearly statistically underpowered. The major concern is overfitting in regression and unstable ROC cutoffs.
• The definition of Brucella-positive vs negative is not very clear. The cohort authors refereed is already brucella positive (diagnosed positive). Its very confusing. It will be good if the authors can clarify that.
• There are some unit errors which the authors need to address in the revised manuscript.
• The OR =12.86 but interpreted as “negative association” this is very confusing needs clarification.
• The authors clearly overinterpreted the ROC data. The described AUROC is 0.68-0.7 which is borderline value.
• The mechanistic discussion if strong but generic needs modification to avoid the textbook tone.

 

Comments on the Quality of English Language

The language editing is required for better understanding of the manuscript. 

Author Response

The author in the manuscript addresses the demographic, clinical, microbiological, treatment, and hematological characteristics of patients with pediatric brucellosis in the Hail region of Saudi Arabia and to assess the diagnostic value of hematological parameters. The study is important but there are certain areas that need to be addressed before the paper gets accepted.

The major issue with the manuscript lies in data interpretation. The author uses n=38 but for analysis the authors are using Logistic regression, MANOVA ad ROC analysis which is clearly statistically underpowered. The major concern is overfitting in regression and unstable ROC cutoffs.

Author Response: Thank you for this important and constructive comment. We acknowledge that the relatively small sample size, particularly the limited number of culture-confirmed cases (n=16), may reduce statistical power and affect the stability of multivariable analyses and ROC-derived cutoff estimates. The MANOVA, logistic regression, and ROC analyses were performed as exploratory analyses to investigate potential associations and assess the diagnostic utility of hematological parameters rather than to establish or validate a definitive predictive model. We agree that the regression model may be susceptible to overfitting and that the identified ROC cutoff values should be interpreted cautiously. To address this concern, we have expanded the Limitations section to explicitly acknowledge these issues and to emphasize that the findings require validation in larger, prospective, multicenter cohorts.

The definition of Brucella-positive vs negative is not very clear. The cohort authors refereed is already brucella positive (diagnosed positive). Its very confusing. It will be good if the authors can clarify that.

Author Response: Thank you for this important observation. We agree that the previous terminology ("Brucella-positive" and "Brucella-negative") was potentially confusing because all patients included in the study fulfilled the diagnostic criteria for brucellosis. To improve clarity, we have revised the terminology throughout the manuscript. Patients with positive blood cultures are now referred to as "culture-confirmed brucellosis" (n = 16), whereas the remaining patients are referred to as "non-culture-confirmed brucellosis" (n = 22). The latter group included 12 patients with negative blood cultures and 10 patients in whom blood cultures were not performed but who met the clinical and serological diagnostic criteria for brucellosis. We have also added an explicit explanation in the Results section to clarify the composition of these groups.

There are some unit errors which the authors need to address in the revised manuscript.

The OR =12.86 but interpreted as "negative association" this is very confusing needs clarification.

Author Response: Thank you for this important comment. We have carefully reviewed and corrected the unit errors throughout the revised manuscript, including the units used for WBC and platelet counts in the text, tables, and figure legends. We also agree that the interpretation of the odds ratio for bone/joint pain requires clarification. The OR of 12.86 reflected the higher odds of bone/joint pain in the non-culture-confirmed brucellosis group compared with the culture-confirmed brucellosis group. Therefore, although the OR value was greater than 1, bone/joint pain was less frequent among culture-confirmed cases and more frequent among non-culture-confirmed cases. To avoid confusion, we have revised the relevant text to state that bone/joint pain was significantly more common in the non-culture-confirmed brucellosis group rather than describing it as a "negative association."

The authors clearly overinterpreted the ROC data. The described AUROC is 0.68-0.7 which is borderline value.

Author Response: Thank you for this important observation. We agree that AUROC values in the range of 0.68–0.70 indicate only modest or borderline discriminative ability and should not be interpreted as evidence of strong diagnostic performance. Our intention was not to propose these hematological parameters as stand-alone diagnostic tests but rather to explore their potential utility as supportive indicators in the clinical assessment of pediatric brucellosis. The ROC analysis was included primarily to compare the relative diagnostic performance of different hematological parameters and to identify those with greater discriminatory potential. To address this concern, we have revised the Results and Discussion sections to avoid overinterpretation and to emphasize the modest diagnostic accuracy of the identified parameters.

The mechanistic discussion if strong but generic needs modification to avoid the textbook tone.

Author Response: Thank you for this valuable comment. We agree that the mechanistic discussion was overly general and insufficiently linked to the findings of the present study. Accordingly, we have revised this section to focus more specifically on the observed hematological abnormalities in our cohort, particularly anemia, leukopenia, and neutropenia, and to discuss the potential biological mechanisms underlying these findings in the context of pediatric brucellosis. The revised text places greater emphasis on interpreting our results rather than providing a broad overview of disease pathophysiology.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The author team has carefully and thoroughly responded to the reviewers' comments, making numerous substantial revisions to the original manuscript. Notably, improvements have been made in clarifying methodological details, deepening the discussion, updating references, and standardizing the manuscript format. These revisions have enhanced the scientific rigor, clarity, and precision of the paper. The inherent limitations of the study—such as its retrospective design and small sample size—have been adequately addressed in the discussion and do not diminish its value as a meaningful regional study describing the clinical and hematological features of pediatric brucellosis in a specific population.

Please have the author conduct a thorough proofreading of language and formatting before submitting the final version, to correct any remaining minor grammatical errors, ensure consistent terminology (e.g., maintaining uniformity in terms such as "Receiver Operating Characteristic" throughout the manuscript), and verify the consistency of all figures and data references.

Comments on the Quality of English Language

None

Reviewer 2 Report

Comments and Suggestions for Authors

The revised manuscript has been well improved. 

Reviewer 3 Report

Comments and Suggestions for Authors

The authors addressed all the comments.