Brucellosis Abattoir Surveillance Using Serology and Molecular Tests Among Livestock in Free State and Limpopo Provinces, South Africa

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Major Comments:

1. Key diagnostic comparison missing: The study compares RBT, CFT, iELISA, and PCR but does not formally calculate diagnostic performance metrics (sensitivity, specificity, positive/negative predictive values) against a combined reference standard. This limits the clinical impact of your findings.

2. Unclear case definition for “infected animal”: The manuscript uses multiple definitions (seropositive, PCR-positive, dual positive) without a final consolidated case definition. This creates confusion in prevalence reporting and should be standardized.

3. 0% culture rate not fully discussed: The 0% isolation rate is a critical finding but is only briefly explained. Expand on why culture failed (low bacterial load, chronic infection, sample type, transport, overgrowth), and cite relevant literature.

4. Lack of One Health / public health context: The study demonstrates high silent infection in slaughter livestock but does not link this to worker safety, food safety, or human brucellosis risk.

5. AMOS-PCR validation weak: AMOS-PCR was used for speciation, but no positive controls for B. ovis or B. suis were included, and no confirmation via sequencing or Bruce-ladder was done. This reduces species ID reliability.

6. Sample bias not acknowledged: Convenience sampling at abattoirs cannot represent provincial prevalence. Clearly state that results reflect slaughtered animals only, not the general livestock population.

7. Conclusion overstates policy recommendation: You strongly recommend replacing RBT-CFT with iELISA + PCR, but you did not evaluate cost, scalability, or lab feasibility in South Africa’s routine system. Tone down to “consider integration.”

8. Introduction needs global/regional context: Add 2–3 sentences on brucellosis burden in sub-Saharan Africa and why abattoir surveillance is a validated tool for neglected zoonoses.
9. Age group categorization inconsistent: Age groups (1–2, 2–3, >3 years) are not justified and vary between provinces. Use standard livestock age categories and explain the rationale.
10. Tissue selection not justified: The study pools lymph nodes and tests liver, lung, and spleen but does not explain why reproductive tissues (target site for Brucella) were not sampled.
11. Kappa interpretation incomplete: Kappa = 0.65 (substantial agreement) between RBT and iELISA is reported, but no kappa between serology and PCR (the key mismatch) is provided.
12. Mixed infections poorly described: “Mixed infection” is mentioned but not defined (dual PCR bands). CT values? Confirmed by sequencing?. Clarify and verify.
13. Sheep vs. cattle comparison imbalanced: Sheep are all 1–2 years old, while cattle span age groups. Statistical comparison across species is biased; acknowledge this limitation.
14. Visualization of results is weak: Figure 1 is basic demographics. Add a summary figure comparing test positivity across species/provinces and a Venn diagram of test concordance.
15. Reference list gaps: Add recent (2023–2025) papers on PCR in seronegative carriers and iELISA superiority in endemic settings to strengthen your argument.
16. Table formatting errors: Tables 2–4 have misaligned columns, cut-off values, incomplete CI, and missing headers. Standardize all tables for readability.
17. Abbreviations not defined on first use: iELISA, ITS-PCR, AMOS-PCR, CFT, RBT, WOAH, and DALRRD need first-use definitions in the abstract and introduction.
18. Statistical reporting incomplete: P-values are present, but 95% CIs are missing for many comparisons, and confidence intervals are sometimes incorrectly formatted.
19.   English flow & grammar: Several run-on sentences and awkward phrasing, major polishing needed.
20. Ethics / biosafety: Mention biosafety level 3 (BSL-3) precautions for handling Brucella cultures, as these are select agents and mandatory to report. 
21. Line 66: Cite this reference as This paper provides recent seropoprevalence data from an endemic livestock region (relevant to cattle/sheep in similar settings). It supports the authors’ discussion about regional endemicity and the need for enhanced surveillance. http://dx.doi.org/10.71081/cvj/2024.004.
22. Line 87-88: Cite this reference as this study directly evaluates improved serological diagnostic approaches (periplasmic proteins) for bovine brucellosis, which aligns with the authors’ recommendation to incorporate iELISA and more sensitive tests.  https://doi.org/10.47278/journal.ijvs/2024.251.
23. Line 96-97: Cite this reference as This molecular method paper for B. melitensis strain identification supports the authors’ AMOS-PCR findings and their discussion of cross-species transmission. https://www.pvj.com.pk/pdf-files/23-504.pdf. 
24. Line 234: Cite this paper as it  provides genotypic characterization of Brucella strains, similar to the authors’ AMOS-PCR results. Citing it would strengthen the discussion on strain diversity and molecular typing methods. https://doi.org/10.47278/journal.ijvs/2024.223. 
25. Line 419-421: Cite this reference as This study is highly relevant because it covers seroprevalence and molecular characterization in small ruminants, directly supporting the authors’ surprising finding that sheep had high PCR positivity despite seronegativity. http://dx.doi.org/10.71081/cvj/2025.048. 
26. Line 439-449: Cite this paper, directly compares RBT and ELISA for brucellosis seroprevalence, which mirrors the authors’ core comparison of RBT-CFT vs iELISA. https://www.pvj.com.pk/pdf-files/25-073.pdf. 
27. Lines 455-456: Cite this reference as this review covers host-pathogen interactions and zoonotic transmission, which aligns with the authors’ One Health recommendations and discussion of public health risks at slaughter. https://www.pvj.com.pk/pdf-files/25-371.pdf.
28. Line 466: Cite this reference as This paper focuses on risk factors and geographical spread, which parallels the authors’ analysis of age/sex associations and provincial differences between Limpopo and Free State. https://doi.org/10.47278/journal.ijvs/2024.187. 

Comments for author File: Comments.pdf

Comments on the Quality of English Language

Please thoroughly revise the English with a professional.

Author Response

Reviewer 1 comments and suggestions

Comment: Key diagnostic comparison missing: The study compares RBT, CFT, iELISA, and PCR but does not formally calculate diagnostic performance metrics (sensitivity, specificity, positive/negative predictive values) against a combined reference standard. This limits the clinical impact of your findings.

Response: The objective of this study was not to validate or compare the diagnostic performance of serological or molecular assays, but rather to isolate and characterize Brucella spp. from field samples with unknown infection status. Consequently, formal diagnostic performance metrics, including sensitivity, specificity, and predictive values, were not calculated. The results of RBT, CFT, iELISA, and PCR are therefore presented descriptively to illustrate concordance and variation between methods, rather than to infer diagnostic accuracy. Future studies incorporating a defined reference standard or latent class analysis would be required to formally evaluate diagnostic performance. Furthermore, the current sample size is insufficient to provide the statistical power necessary for detecting meaningful sensitivity and specificity intervals. Small cohorts can lead to wide confidence intervals, rendering such metrics clinically unreliable.

Comment: Unclear case definition for “infected animal”: The manuscript uses multiple definitions (seropositive, PCR-positive, dual positive) without a final consolidated case definition. This creates confusion in prevalence reporting and should be standardized.

Response: “An ‘infected animal’ was defined as an animal with laboratory confirmation of Brucella spp. based on bacterial isolation and/or AMOS-PCR detection, as well as a series of serological test results. This was indicated in the material and methods section 2.4, Line 204-207.

Comment: 0% culture rate not fully discussed: The 0% isolation rate is a critical finding but is only briefly explained. Expand on why culture failed (low bacterial load, chronic infection, sample type, transport, overgrowth) and cite relevant literature.

Response: Paragraph 7 of the Discussion (Line 479-490) provides a detailed explanation of the factors that may have contributed to the inability to isolate Brucella species using culture methods.

Comment: Lack of One Health / public health context: The study demonstrates high silent infection in slaughter livestock but does not link this to worker safety, food safety, or human brucellosis risk.

Response: Thank you for the comment. The authors have expanded on the risks associated with “silent” infections in animals in the Discussion section (Lines 435–440).

 

Comment: AMOS-PCR validation weak: AMOS-PCR was used for speciation, but no positive controls for B. ovis or B. suis were included, and no confirmation via sequencing or Bruce-ladder was done. This reduces species ID reliability.

Response: Thank you for the comment. The positive controls were included during the experiment. The controls have been added to the method in the manuscript (Lines 254-256).

Comment: Sample bias not acknowledged: Convenience sampling at abattoirs cannot represent provincial prevalence. Clearly state that results reflect slaughtered animals only, not the general livestock population.

Response: The limitation has been acknowledged in the study limitations.

Comment: Conclusion overstates policy recommendation: You strongly recommend replacing RBT-CFT with iELISA + PCR, but you did not evaluate cost, scalability, or lab feasibility in South Africa’s routine system. Tone down to “consider integration.”

Response: Agreed. The manuscript conclusion has been revised, and the phrase “is strongly recommended” has been replaced with “may be considered,” as PCR can only be used when appropriate tissues are available, as indicated in the sentence (Line 494-502).

Comment: Introduction needs global/regional context: Add 2–3 sentences on brucellosis burden in sub-Saharan Africa and why abattoir surveillance is a validated tool for neglected zoonoses.

Response: Thank you for the comment. The authors revised the introduction and included sub-Saharan Africa (Line 52-53).

Comment: Age group categorization inconsistent: Age groups (1–2, 2–3, >3 years) are not justified and vary between provinces. Use standard livestock age categories and explain the rationale.

Response: Thank you for the comment. The sampling in this study was based on convenience. Furthermore, the ages were categorized primarily for ease of statistical analysis.

Comment: Tissue selection not justified: The study pools lymph nodes and tests liver, lung, and spleen, but does not explain why reproductive tissues (target site for Brucella) were not sampled.

Response: The authors appreciate the insightful feedback. The selection of tissues in this study was based on previous studies and published literature describing the tissue tropism of Brucella, including its affinity for lymphatic and organ tissues in addition to reproductive organs. A sentence has been included in section 2.3 (Line 155-159) to explain.  The authors acknowledge the limitation of not including all relevant tissues and will consider the reviewer’s suggestion in future studies. However, it was mainly limited in this abattoir study, as collecting various samples is difficult and logistically challenging, as samples must be traced back to the same animal and collection is constrained by the speed of the production line.

 

Comment: Kappa interpretation incomplete: Kappa = 0.65 (substantial agreement) between RBT and iELISA is reported, but no kappa between serology and PCR (the key mismatch) is provided.

Response: Thank you for the comment. The authors initially did not report agreement values because of disagreement observed among some assays. However, the agreement analysis has now been included in the Results section for completeness and clarity (Section 3.5, Line 388-389).

Comment: Mixed infections poorly described: “Mixed infection” is mentioned but not defined (dual PCR bands). CT values? Confirmed by sequencing?. Clarify and verify.

Response: We thank the reviewer for this observation. Mixed infections in this study were defined as samples producing multiple species-specific bands within the same sample, consistent with the presence of more than one Brucella species during multiplex Polymerase Chain Reaction analysis. ‘Mixed infection’ was defined in the abstract (Line 33-36).

Comment: Sheep vs. cattle comparison imbalanced: Sheep are all 1–2 years old, while cattle span age groups. Statistical comparison across species is biased; acknowledge this limitation.

Response: Thank you for this important observation. We acknowledge the potential sampling bias in this study. Animals presented for slaughter were selected primarily based on physical condition and market readiness rather than age. Furthermore, the animals originated from different farms and production systems, resulting in variability in age distribution.

Due to the convenience sampling approach employed in this study, only small ruminants aged 12–24 months were available for slaughter during the sampling visits. The age categories used in the study were therefore grouped to facilitate statistical analysis and interpretation of the data. This limitation has been acknowledged in the manuscript (Line 146-148)

 

Comment: Visualization of results is weak: Figure 1 is basic demographics. Add a summary figure comparing test positivity across species/provinces and a Venn diagram of test concordance.

Response: Thank you for the valuable suggestion. The authors acknowledge the importance of clear visualization of diagnostic test outcomes and made changes to Tables 1 and 2 to summarize the diagnostic tests conducted across both provinces and the corresponding results. The table was considered sufficient to illustrate the distribution and concordance of test findings within the scope of the study.

Comment: Reference list gaps: Add recent (2023–2025) papers on PCR in seronegative carriers and iELISA superiority in endemic settings to strengthen your argument.

Response: Thank you for the comment. The authors have added recent studies as suggested.

Comment: Table formatting errors: Tables 2–4 have misaligned columns, cut-off values, incomplete CI, and missing headers. Standardize all tables for readability.

Response: Thank you for this observation. The authors acknowledge the formatting inconsistencies identified in Tables 2–4, which may have arisen during the manuscript formatting process. The tables have now been carefully reviewed and changes have been made to ensure correct alignment of columns, inclusion of complete confidence intervals, appropriate cut-off values, and consistent headers throughout the manuscript.

Comment: Abbreviations not defined on first use: iELISA, ITS-PCR, AMOS-PCR, CFT, RBT, WOAH, and DALRRD need first-use definitions in the abstract and introduction.

 

Response: All abbreviations have been defined upon first use throughout the manuscript.

Comment: Statistical reporting incomplete: P-values are present, but 95% CIs are missing for many comparisons, and confidence intervals are sometimes incorrectly formatted.

Response: Thank you for the comment. The authors have revised the values.

Comment: English flow & grammar: Several run-on sentences and awkward phrasing, major polishing needed.

Response: Thank you for the comment. English flow and grammar have been revised and corrected throughout the manuscript to improve clarity and readability.

Comment: Ethics/biosafety: Mention biosafety level 3 (BSL-3) precautions for handling Brucella cultures, as these are select agents and mandatory to report.

Response: Thank you for the comment. The authors acknowledge the importance of biosafety in the processing of Brucella samples. A biosafety statement has been added under the bacterial isolation section of the manuscript (Line 272-273).

Comment: Line 66: Cite this reference as This paper provides recent seropoprevalence data from an endemic livestock region (relevant to cattle/sheep in similar settings). It supports the authors’ discussion about regional endemicity and the need for enhanced surveillance. http://dx.doi.org/10.71081/cvj/2024.004.

Response: Thank you for the comment. However, the authors respectfully disagree. Line 66 of the manuscript discusses the limitations of the brucellosis control scheme in South Africa, and the citation currently used is a more recent retrospective study (2019) conducted within the study area, which the authors consider more appropriate in this context. Although retrospective in nature, it remains the most comprehensive study available for the area and therefore best supports the statement presented.

Comment: Line 87-88: Cite this reference as this study directly evaluates improved serological diagnostic approaches (periplasmic proteins) for bovine brucellosis, which aligns with the authors’ recommendation to incorporate iELISA and more sensitive tests.  https://doi.org/10.47278/journal.ijvs/2024.251.

Response: Thank you for the comment. The authors have added the suggested citation [20] to support the discussion on improved serological diagnostic approaches for bovine brucellosis.

Comment: Line 96-97: Cite this reference as This molecular method paper for B. melitensis strain identification supports the authors’ AMOS-PCR findings and their discussion of cross-species transmission. https://www.pvj.com.pk/pdf-files/23-504.pdf.

Response: The authors have added the suggested citation [27] to support the discussion on cross contamination.

Comment: Line 234: Cite this paper as it provides genotypic characterization of Brucella strains, similar to the authors’ AMOS-PCR results. Citing it would strengthen the discussion on strain diversity and molecular typing methods. https://doi.org/10.47278/journal.ijvs/2024.223.

Response: The authors have added the suggested citation [38] to support the discussion on improved serological diagnostic approaches for bovine brucellosis.

 

Comment: Line 419-421: Cite this reference as This study is highly relevant because it covers seroprevalence and molecular characterization in small ruminants, directly supporting the authors’ surprising finding that sheep had high PCR positivity despite seronegativity. http://dx.doi.org/10.71081/cvj/2025.048.

Response: The authors have added the suggested citation [60] to support the discussion on discrepancies between PCR and serological tests.

 

Line 439-449: Cite this paper, directly compares RBT and ELISA for brucellosis seroprevalence, which mirrors the authors’ core comparison of RBT-CFT vs iELISA. https://www.pvj.com.pk/pdf-files/25-073.pdf.

Response: Thank you for this helpful suggestion. We acknowledge that the recommended paper provides a useful comparison of RBPT/RBT and ELISA-based approaches for brucellosis seroprevalence. However, after careful consideration, we retained the existing references because they are more directly aligned with the species, diagnostic algorithm and epidemiological context of the present study, particularly the RBT-CFT and iELISA comparison used in our surveillance framework. Nonetheless, we appreciate the reviewer’s recommendation and have rechecked the discussion to ensure that the rationale for comparing these serological methods is clearly supported.

Lines 455-456: Cite this reference as this review covers host-pathogen interactions and zoonotic transmission, which aligns with the authors’ One Health recommendations and discussion of public health risks at slaughter. https://www.pvj.com.pk/pdf-files/25-371.pdf.

Response: The authors have added the suggested citation [59] to support the discussion on one health.

 

Line 466: Cite this reference as This paper focuses on risk factors and geographical spread, which parallels the authors’ analysis of age/sex associations and provincial differences between Limpopo and Free State. https://doi.org/10.47278/journal.ijvs/2024.187.

Response: The authors have added the suggested citation [46].

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

the study is based on diagnostic and epidemiological assumptions that are not always supported by adequate technical and scientific evidence. Comments are provided in the attached file. The study design needs to be revised, and both the discussion and the conclusions must be re-examined. 

Comments for author File: Comments.pdf

Author Response

Reviewer 2 comments and suggestion

1. Introduction:

Comment: line 23: modify according to the comments in paragraph 2

Response: Thank you for the comment. The authors have revised accordingly.

Comment: line 32: modify according to the comments in paragraph 3

Response: Thank you for the comment. The authors have revised accordingly.

Comment: line 52: the authors must specify that the data describe the prevalence of the disease within a group (slaugthered animals) and not the prevalence in the animal population in SA

Response: Thank you for the comment. In the Introduction, the authors acknowledge that previous studies were conducted in specific populations and regions; therefore, there remains limited knowledge regarding the overall prevalence of brucellosis in South Africa. This aligns with the reviewer’s observation. However, this point has already been clearly stated in the manuscript (Line 57-73).

 

Comment: line 70: underestimation of the true prevalence is not due to the lack of a routine test, but to the lack of a test covering the entire population

Response: Thank you for the comment. The statement has been revised in the introduction (Line 83-75)

 

Comment: Line 88: I would encourage the authors to mention iElisa’s sensitivity to vaccinated animals

Response: Thank you for the comment. The authors have added the sensitivity statement in the manuscript (Line 101-103)

 

1. Materials and Methods

Comment: line 134: it would be preferable to specify the type of test tube rather than the colour of the cap

Response: Thank you for the comment. The blood collection tube has been specified as a gel clot activator tube in the Methods section (Lines 154–155)

Comment: Line 139: Failure to collect the supramammary lymph nodes, the mammary gland, the uterus, and the testicles inevitably reduces the sensitivity of the method

Response: Thank you for the comment. The authors acknowledge the importance of collecting additional tissues such as supramammary lymph nodes, mammary gland, uterus, and testicles, as their inclusion could improve the method’s sensitivity. However, the samples were collected from slaughterhouses, where some of these tissues were not consistently available at the time of sampling. In addition, the sampling strategy was informed by previous studies. Nevertheless, the authors appreciate the suggestion and will acknowledge this as part of the limitation.

 

Comment: Line 157: If you wish to assess the sensitivity of serological tests, they must be carried out in parallel. How were the animals that tested positive in the BRT but negative in the CFT assessed?

Response: Thank you for the comment. All serum samples were subjected to RBT and IELISA, and positive RBT were subjected to CFT as a standard confirmatory diagnostic assay. The authors have revised the manuscript to avoid the diagnostic characteristics of the tests.

Comment: Line 168: What scientific evidence does this interpretative criterion rely on?

Response: Thank you for the comment. The interpretation is based on the WOAH manual and the South African standards.

Line 178: The interpretative criterion used is arbitrary, as every animal must be considered a suspect if it results positive in any serological test used. Two positive serological tests are neither useful nor sufficient to classify an animal as infected. An animal is considered infected only if, in addition to testing positive in an indirect test, it:

1) also tests positive in a direct test, or

2) is epidemiologically linked to another positive case (within an outbreak or linked to an outbreak), or

3) shows clinical signs of the disease.

However, epidemiological assessment was not considered in any case. The detection of a positive animal should have been followed by an investigation at each of the animals’ farms of origin

Response: Thank you for the comment. The authors have defined in the Methods section 2.4 the criteria and cut-off values used to determine positive results for each diagnostic test. “In this study, an ‘infected animal’ was diagnosed as a brucellosis-infected animal with laboratory confirmation of Brucella spp. based on bacterial isolation and/or AMOS-PCR detection, as well as a series of serological tests (any two or more of RBT, CFT, and iELISA) results.” In this study, however, the serological and culture results did not align, but the samples were from apparently healthy animal tissues, and culture sensitivity is known to be low due to the unknown fact of the stage of infection, with sensitivity being the highest from the placenta and aborted material which are not available from abattoirs (which was included in the discussion).  Nevertheless, the serological findings remain important, as several factors may have influenced the performance and sensitivity of the different tests. Furthermore, the samples were collected at slaughter, which limited the possibility of follow-up investigations on individual animals. However, one of the objectives arising from this study is to conduct follow-up investigations at the farm level to better understand the epidemiology of the disease within herds, and this work is still to be undertaken.

 

Comment: Line 238: Bacteriological testing should be carried out on all organ samples received by the laboratory, and not only on tissues that tested positive by PCR. Furthermore, the organs were not placed in enrichment broths on which to carry out weekly bacteriological testing for at least six weeks.

Response: Thank you for the comment. All organs received in the laboratory were initially screened by PCR before culture. PCR is considered more sensitive than culture; therefore, only PCR- and serologically positive samples were selected for bacteriological culture. Furthermore, the objective of the study was to attempt direct culture of tissue samples without prior manipulation or enrichment.

 

Comment: In fact, as the number of Brucella organisms is likely to be lower in milk, colostrum, and some tissue samples than in abortion material, enrichment may be advisable for all other organs. This could explain the low sensitivity of the bacteriological examination.

Response: Thank you for the comment. The authors acknowledge that enrichment of the samples could have improved the sensitivity of the bacteriological examination, particularly in samples with potentially low numbers of Brucella organisms. This has been done with consultation with other Brucella reference laboratories, but without success, as it enriches contaminant growth, which is faster-growing than Brucella.

 

Comment: Line 280: When examining positive results, it would also be necessary to indicate the number of vaccinated animals in which at least one serological test was positive.

Response: Thank you for the comment. The lack of adequate epidemiological information, particularly regarding vaccination status, made it difficult to fully interpret the serological results. The authors acknowledge this limitation in the manuscript, as it may affect the accurate determination of the true burden of the disease (Line 552-555).

Comment: Furthermore, as the authors explain, how is such a high percentage of PCR positivity explained compared to serological test positivity?

These results should be supported by the performance values of all the tests used in the laboratory, after evaluating them on samples that are definitely positive (i.e. positive by bacteriological testing), as such low sensitivity of the serological tests and such a high positivity rate in the PCR test, in the absence of a negative bacteriological result, are inexplicable and are likely due to the low sensitivity of the technique used

 

Response: Thank you for the comment. The authors have discussed several factors that may have influenced the low detection rate by bacteriological culture as well as the limited sensitivity of the traditional serological assays. These include potential low bacterial load in sampled tissues, the absence of enrichment procedures, and the possibility of chronically or latently infected animals testing seronegative. The discrepancy between PCR positivity and serological positivity has therefore been acknowledged and discussed in the manuscript. Consequently, the authors advocate for the integration of more sensitive diagnostic approaches, including iELISA and PCR, into routine surveillance strategies to improve detection rates and enhance disease surveillance.

 

Comment: Line 468: The authors mention a low amount of DNA. Given that so many organs were available, the authors should explain why there is such a low amount of DNA.

Response: Thank you for the comment. The Bruce-ladder PCR assay performs more effectively when applied to bacterial isolates. In this study, although multiple organs were collected from the animals, the samples were not pooled and were instead processed individually. Furthermore, these samples were collected from apparently healthy animals, suggesting that the bacterial load and corresponding DNA concentration may have been too low for successful amplification by Bruce-ladder PCR, considering the sensitivity requirements of the assay.

 

Comment: Discussion:

The study is based on testing for antibodies or antigens/DNA in a statistically significant number of animals, considered as a sample.

The authors should confine themselves to describing positive results obtained using the various techniques without any consideration of diagnostic significance, using laboratory techniques on all animals, stating only that, in the absence of a programme allowing for a more widespread analysis of the infection (e.g. all slaughtered animals or all live animals), it is not possible to make epidemiological assessments but only to confirm the presence of the disease in SA.

 

All the more so, under these conditions, it is not possible to make any assessment of the sensitivity of the methods that could be used to implement a disease control programme; on the contrary, it is agreed that the use of the ELISA kit could make it easier to trace positive animals, but only in unvaccinated animals.

 

Response: Thank you for the comment. The authors acknowledge the insightful feedback and the limitations of the current study. The manuscript has been revised accordingly to avoid wording that may imply conclusions or claims not fully supported by the study design and findings (Line 540-551).

Reviewer 3 Report

Comments and Suggestions for Authors

Thank you for giving me the opportunity to revise this manuscript. It is interesting the current status of brucellosis in SA and these findings reveal that it is important improve the surveillance maintaining an appropriate economic sustainability.

I need to point out on some revisions you need to do in order to improve the quality of the paper.

Comments for author File: Comments.pdf

Author Response

Reviewer 3 comments and suggestion

Comment: Line 115-118:

In Material and Methods section, you described the use of blood, liver, lung, spleen, and lymph nodes as targeted organs to detect Brucella species. Moreover, you should include also mammary gland, uterus and or testicles or in any case explain the reasons you didn’t. The mammary gland, uterus and the testicles are the main target organ in adult infected animals, In my opinion this could be a severe bias in your study. You could avoid lung and liver, considering that you had already spleen for haematogenous spread of Brucella.

 

Response: Thank you for the comment. The authors acknowledge the importance of collecting additional tissues such as sub-mammary lymph nodes, mammary gland, uterus, and testicles, as their inclusion could improve the method’s sensitivity. However, the samples were collected from slaughterhouses, where some of these tissues were not consistently available at the time of sampling. In addition, the sampling strategy was informed by previous studies. Nevertheless, the authors appreciate the suggestion and will acknowledge this as part of the limitations.

 

Comment: Line 121-123: In this section you decided to set the sample size assuming an expected prevalence of 50% for cattle and 15% for sheep. Please explain in details the reason to set them, considering that these values are not consistent with prevalence data declared in SA, as you reported in Introduction (1.5% at the Cato Ridge abattoir; 1.45% in rural cattle; 6.3% in SA; More recently, data 58 from Limpopo and Free State provinces indicated a seroprevalence of 4.3% for the period 59 2013-2022). 2.1 and 2.2 sections reported same data regarding sample effort, please revise them, including this information just in a section.

 

Response: Thank you for this valuable comment. We have revised the manuscript to clarify the rationale for the expected prevalence values used in the sample size calculation and to avoid repetition between Sections 2.1 and 2.2. For cattle, an expected prevalence of 50% was used because this provides the most conservative sample size estimate when the true prevalence is uncertain. Although previous South African studies reported lower prevalence estimates, including 1.5% at the Cato Ridge abattoir, 1.45% in rural cattle, 6.3% nationally, and 4.3% in Limpopo and Free State provinces during 2013–2022, these studies were conducted in specific populations, geographic areas, or surveillance contexts. Therefore, they may not fully represent the expected prevalence in the broader target population sampled in the present study. Using 50% was therefore intended to maximize the required sample size and reduce the risk of under-sampling. For sheep, an expected prevalence of 15% was used based on available baseline information and the expectation of comparatively lower brucellosis prevalence in small ruminants in the study context. We acknowledge that this value is higher than some previously reported estimates; however, it was selected to ensure that the sample size remained sufficiently robust for detecting infection in a population where reliable provincial-level prevalence data are limited.

 

In response to the reviewer’s concern, we have revised the relevant methods section to explain these assumptions more clearly. We have also removed the duplicated sample effort information from Section 2.2 and retained it in a single section to improve clarity and avoid repetition.

 

Comment: Line 117 vs Line 127: The expressed value in the following section 4 is in contrast to the use of a sample size to detect an expected prevalence of 50% and 15% respectively for cattle and sheep. In my opinion this could be a severe bias in your study. You can’t speak about prevalence

 

Response: Thank you for the comment. The authors have revised the concept throughout the manuscript.

 

Comment: In the discussion section you speak about prevalence, but you can’t if you used a sample size calculator to detect a fixed prevalence.

 

Response: Thank you for the comment. The authors have revised the concept throughout the manuscript.

 

Comment: Line 168-171; A reference is missing to define the reason you used these interpretation criteria.

 

Response: Thank you for the comment. The reference is mentioned at the start of the paragraph.

 

Line 204-206: In my opinion, you should cite the reference specific for these primers, considering that the reference cited is about Canine Brucellosis. Anyway, if the citation is correct, please could you explain the reasons you didn’t use the suggested methods reported in WOAH Manual as done for serology.

 

Response: Thank you for the comment. The primers used in this study were based on the 16S–23S rRNA intergenic spacer PCR assay described by Keid et al., which targets genus-specific Brucella DNA and is not host-specific. The serological methods applied in this study were based on the WOAH (formerly OIE) Manual of Diagnostic Tests and Vaccines for Terrestrial Animals, with adaptations to South African laboratory standards and protocols.

 

Line 237-246: It is surprisingly unexpected that you didn’t find Brucella growth. Did you use a control quality media? Or a reference strain to check the condition of media?

Usually, when a chronic brucellosis is suspected, you should use a liquid selective media together with solid ones. Could you explain your choice to use Blood and Farrell only?

 

Response: Thank you for the comment. The authors have discussed several factors that may have influenced the low detection rate observed with bacteriological culture. Media validation and the inclusion of appropriate positive and negative controls were performed as part of routine quality assurance. In this study, the control strains showed appropriate growth on both media types used, confirming the suitability of the culture conditions. Furthermore, the authors followed WOAH guidelines and South African standards, which permit the use of both selective media (e.g., Farrell or CITA) and non-selective media for the isolation of Brucella. All laboratory analyses were performed at the National Reference Laboratory, an ISO/IEC 17025-accredited facility, using accredited and quality-assured diagnostic methods for brucellosis testing

 

Comment: Line 280: In my opinion, the table expresses better the results. In the text you could simply report the number of positive animals recovered by serology (or combined test) and them agreement with PCR. In this way you should better highlight the results of your study.

Response: Thank you for the comment. The authors have removed the figure as suggested. The comprehensive results are found in Table 1.

Comment: Line 370: It stated that chronic and latent infections are common in SA, but the population tested is constituted more than 50% by young animals. Please comment this sentence according your opinion.

 

Response: Thank you for the comment. The endemic nature of brucellosis in South Africa is well documented, and several studies have highlighted limitations in surveillance and diagnostic coverage. In this study, the detection of positive cases in a population comprising a high proportion of young animals suggests that infection may be occurring beyond the traditionally targeted groups. This may reflect gaps in surveillance, as young animals and males are often less frequently included in routine testing programmes. The findings therefore support the need for broader and more inclusive surveillance strategies rather than indicating a definitive conclusion on age-related infection dynamics.

 

Comment: Line 395-396: This sentence is true, but in your study, you didn’t test reproductive apparatus or milk. Please revise it according to your findings.

Response: Thank you for the comment. The authors have revised accordingly (line 405).

Comment: Line 412-415: In my opinion, the reference to [25] and [50] should be limited to the detection of seronegative PCR positive animals and not to define that the traditional tests significantly underestimate true prevalence in endemic regions. In both studies, there is a limited number of animals resulted positive to at least a serological test.

Response: Thank you for the comment. The authors have revised accordingly (line 431-434).

Comment: Line 431-437: Especially in adult animals you should perform bacteriological test on reproductive organs or milk/mammary gland.

Response: Thank you for the comment. The authors acknowledge the importance of collecting additional tissues such as supramammary lymph nodes, mammary gland, uterus, and testicles, as their inclusion could improve the method’s sensitivity. However, the samples were collected from slaughterhouses, where some of these tissues were not consistently available at the time of sampling. In addition, the sampling strategy was informed by previous studies. Nevertheless, the authors appreciate the suggestion and will acknowledge this as part of the limitation  .

 

Comment: Tables: Table 1 is not necessary, because you used primers as already described and reported in reference. More in general, all tables are too hard to read (too many and too complex), please revise them.

Response: Thank you for the comment. The authors have removed Table 1 as suggested. Table 2 has been revised to improve clarity and readability, and it now more comprehensively presents the serological and PCR results. Table 3 and 4 have been moved to supplementary.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have made substantial changes and improved the manuscript. It is recommended to be accepted for publication. 

Comments on the Quality of English Language

Please thoroughly revise the English with a professional.

Reviewer 2 Report

Comments and Suggestions for Authors

the document has been modified like requested