Strain structure analysis of Mycobacterium tuberculosis circulating among HIV negative, positive and drug resistant TB patients attending chest clinics in Western Kenya

Abstract Background Despite global tuberculosis (TB) interventions, the disease remains one of the major public health concerns. Kenya is ranked 15th among 22 high burden TB countries globally. Methods A cross-sectional study was conducted in Western Kenya, which comprises 10 counties. A multistage sampling method was used where a single sub-county was randomly selected followed by sampling two high volume health facility from each sub-county. Identification of spoligotype profiles and their family distribution and lineage level were achieved by comparison with SITVIT database. Results Lineage distribution pattern revealed that the most predominant lineage was CAS 220 (39.8%) followed by Beijing 128 (23.1%). The other lineages identified were T, LAM, H, X, S and MANU which were quantified as 87 (15.7%), 67 (12.1%), 16 (2.8%), 10 (1.8%), 8 (1.4%) and 5 (0.9%) respectively. CAS and Beijing strains were the most predominant lineage in both HIV negative and positive TB patients. The Beijing lineage was also the most predominant in resistant M. tuberculosis strains as compared to wild type. A total of 12 (2.0%) were orphaned M. tuberculosis strains which were spread across all the 10 counties of the study site. In multivariate logistic regression adjusting for potential cofounders three potential risk factors were significant. HIV status (OR = 1.52, CI = 0.29–3.68 and P value of 0.001), Alcohol use (OR = 0.59, CI = 0.43–3.12 and P-value =0.001) and cross border travel (OR = 0.61, CI = 0.49–3.87 and P value = 0.026). Most M. tuberculosis clinical isolates showed genetic clustering with multivariate logistic regression indicating three potential risk factors to clustering. HIV status (OR = 1.52, CI = 0.29–3.68 and P value of 0.001), Alcohol use (OR = 0.59, CI = 0.43–3.12 and P-value =0.001) and cross border travel (OR = 0.61, CI = 0.49–3.87 and P value = 0.026). Conclusion There exist diverse strains of M. tuberculosis across the 10 counties of Western Kenya. Predominant distribution of clustered genotype points to the fact that most TB cases in this region are as a result of resent transmission other than activation of latent TB.