Drug-resistant TB prevalence study in 5 health institutions in Haiti

Objectives Tuberculosis (TB) is the leading infectious cause of death in the world. Multi-drug resistant TB (MDR-TB) is a major public health problem as treatment is long, costly, and associated to poor outcomes. Here, we report epidemiological data on the prevalence of drug-resistant TB in Haiti. Methods This cross-sectional prevalence study was conducted in five health centers across Haiti. Adult, microbiologically confirmed pulmonary TB patients were included. Molecular genotyping (rpoB gene sequencing and spoligotyping) and phenotypic drug susceptibility testing were used to characterize rifampin-resistant MTB isolates detected by Xpert MTB/RIF. Results Between April 2016 and February 2018, 2,777 patients were diagnosed with pulmonary TB by Xpert MTB/RIF screening and positive MTB cultures. A total of 74 (2.7%) patients were infected by a drug-resistant (DR-TB) M. tuberculosis strain. Overall HIV prevalence was 14.1%. Patients with HIV infection were at a significantly higher risk for infection with DR-TB strains compared to pan-susceptible strains (28.4% vs. 13.7%, adjusted odds ratio 2.6, 95% confidence interval 1.5–4.4, P = 0.001). Among the detected DR-TB strains, T1 (29.3%), LAM9 (13.3%), and H3 (10.7%) were the most frequent clades. In comparison with previous spoligotypes studies with data collected in 2000–2002 and in 2008–2009 on both sensitive and resistant strains of TB in Haiti, we observed a significant increase in the prevalence of the drug-resistant MTB Spoligo-International-Types (SIT) 137 (X2 clade: 8.1% vs. 0.3% in 2000–02 and 0.9% in 2008–09, p<0.001), 5 (T1 clade: 6.8% vs 1.9 in 2000–02 and 1.7% in 2008–09, P = 0.034) and 455 (T1 clade: 5.4% vs 1.6% and 1.1%, P = 0.029). Newly detected spoligotypes (SIT 6, 7, 373, 909 and 1624) were also recorded. Conclusion This study describes the genotypic and phenotypic characteristics of DR-TB strains circulating in Haiti from April 2016 to February 2018. Newly detected MTB clades harboring multi-drug resistance patterns among the Haitian population as well as the higher risk of MDR-TB infection in HIV-positive people highlights the epidemiological relevance of these surveillance data. The importance of detecting RIF-resistant patients, as proxy for MDR-TB in peripheral sites via molecular techniques, is particularly important to provide adequate patient case management, prevent the transmission of resistant strains in the community and to contribute to the surveillance of resistant strains.


Conclusions: The conclusions can be crisp and corroborate the objectives.
We have reformulated the conclusions of this abstract as follows: "Overall, this epidemiological surveillance study reveals an active circulation of drug-resistant TB isolates in the Haitian population, of new or well-known genotypes. In this study, the risk of infection with DR-TB isolates is higher in HIV-positive individuals. This study underscores the importance of routine screening and genotyping of tuberculosis isolates at peripheral sites using molecular techniques to provide adequate patient care, prevent transmission of resistant strains in the community, and contribute to the surveillance of resistant strains"
We agree, we have withdrawn the following statement: "including about 8,400 to 10,000 adolescents for whom tailored interventions are needed to improve retention in care(4,5)." We have modified the sentence as follows: "The overall HIV prevalence is 2% and has remained stable in the past years, with 160,000 people living with HIV in 2018 (4)."

Line 59, MDR-TB infections may replaced with MDR-TB disease
We also agree, we have made the following change: "ln Haiti, MDR-TB diseases are a major public health issue as the treatment and the risk of poor outcomes are starkly increased(6,7)." 6. The case management of such patients is costly for local healthcare systems and threatens to negatively impact of the progress made in the recent years in the fight against TB in Haiti. "of" may be deleted.
Changes: "The case management of such patients is costly for local healthcare systems and threatens to negatively impact the progress made in the recent years in the fight against TB in Haiti(8)." 7. Line 94 to 95: Exclusion criteria included age under 15, negative smear microscopy, and signs of extrapulmonary TB. This sentence does not match with lines 109 to 110 where it is mentioned "The protocol conducted by the GHESKIO centers (INLR and IMIS) was slightly different as only two sputum specimens were collected per patient, and no microscopy testing was performed". These two sites were main contributors for the study.
You are right, this is an error on our part, the exclusion criteria were as follows (modification line 94): "Exclusion criteria included age under 15, negative Xpert MTB/RIF, and signs of extrapulmonary TB". 8. Line 136-137: How was sample size of 1250 arrived at? And why did the authors exceed the pre-determined sample size? Do 2777 cases represent all the cases during the study duration that satisfy the inclusion criteria or some other sampling techniques was used considering the strike in the regions.
We have added an explanatory paragraph about this in the Methods (lines 110-119).
Because of the strike, we suspected there would be a lesser geographic diversity among included patients and Mtb strains, which led us to increase recruitment in the central GHESKIO clinical centers (IMIS and INLR) in an attempt to be more exhaustive.
The sampling techniques were the same as planned, so the 2777 cases do represent those cases which satisfy the inclusion criteria.

Line 141: unequal between sites may be replaced with unequal among sites.
We have corrected this sentence as follow : "The frequency of retreatment was unequal among sites, with a lower frequency in the HUJ and HUM centers (center and northern sites) compared to the remaining three sites." 10. Line 154-156: Please clarify if the study had 62 isolates that were MDR among total 74 DR TB cases and also 62 DR TB isolates that were identified in new TB cases. Please also clarify the type of resistance found in new and re-treatment cases.
Indeed, this is a point that we have clarified. Briefly, a total of 74 DR-TB strains were identified in this study, of which 12 (16.2%) were characterized as having mono-phenotypic resistance to drugs and 62 (83.8%) with a multi-drug resistant profile. Of the total 74 DR-TB, 62 strains were identified in new cases, 11 in relapses and 1 in a person under retreatment following a treatment failure. Among the 62 DR-TB strains identified in new cases: 54 are considered MDR-TB and 8 as mono-resistant. We have added 3 tables in the supplementary data: table 2S, 3S and 4S: - Table 2S: Mono-resistance profile of DR-TB isolates identified in different groups of people (new cases, relapse, treatment after failure, treatment after interruption); - Table 3S: Multi-resistant profile of DR-TB isolates identified in different groups of people (new cases, relapse, treatment after failure, treatment after interruption) - Table 4S: Summary of resistance profiles of DR-TB isolates identified in different groups of people (new cases, relapse, treatment after failure, treatment after interruption). Also, clarification has been brought to the legend of Figure 1S: "Total of 74 DR-TB strains including 12 mono-drug resistant strains and 62 multi-drug resistant TB strains".

Table 2: In treatment after interruption, how was p value of 0.118 obtained when there was no DR-TB case?
This is indeed aberrant and has been withdrawn. 12. Line 175-184: This part may be shifted to discussion.
Agreed. We have rephrased this paragraph and shifted part of it to the Discussion (lines 286-290) to avoid citing article tables in the Discussion.
13. While you use the word "cohort" in discussion in lines 225, 231 and 237, what do we understand? Can were replace it with "the tested isolates" You are right, we have modified sentences as follow: -Line 226: "In this study, we reported the occurrence of DR-TB infections in patient cohorts across five study sites in Haiti" → "In this study, we reported the occurrence of DR-TB infections in patients across five study sites in Haiti". -line 242 : "Secondly, we studied the phenotypic and genotypic diversity of the DR-TB strains isolated from our cohort" → "Secondly, we studied the phenotypic and genotypic diversity of the DR-TB strains isolated from patients". -Line 249: "In our cohort, sequencing analyses identified either T508A or silent T508T rpoB mutations in these five discrepant strain isolates, and their SIT numbers were 20 and 50 respectively, which is consistent with earlier findings(12) » → « in the tested isolates…. » -Line 252: "In addition, strains harboring a L511P rpoB mutation have been detected in our cohort…"→ "In addition, strains harboring a L511P rpoB mutation have been detected, while…" The two references (ref. This has been corrected.
This has been corrected and we added 3 tables in supplementary material (Table 2S, 3S et 4S).
19. Figure 1S: If you give number of isolates also, it will clarify the confusion of MDR-TB cases among total DR-TB cases.
This has been clarified by adding the following sentence: "Total of 74 DR-TB strains including 12 mono-drug resistant strains and 62 multi-drug resistant TB strains".

We note that [Figure(s) 1] in your submission contain map images which may be copyrighted.
This map has been downloaded from : https://simplemaps.com/resources/svg-ht which is a free, web-optimized, SVG Haiti map. Commercial use allowed. There is no copyright. __________________________ We have accepted and incorporated the Reviewer's suggestions and hope that the new version of the manuscript will now be acceptable for publication in PlosOne.
Thank you very much for your kind consideration. If you have any further questions, please do not hesitate to contact me. Sincerely,