Socio-demographic associations of Mycobacterium tuberculosis molecular genotypes in Hong Kong

Abstract

Background: Different Mycobacterium tuberculosis (MTB) genotypes are circulating in the population, the transmission patterns of which may vary. We studied the association between MTB molecular genotypes and socio-demographics in Hong Kong, a city with intermediate TB burden (60.5/100,000 persons). Methods and materials: MTB clinical isolates and hospital data of patients with positive sputum culture were collected from one major hospital cluster region in Hong Kong between 2015–2016. Residential addresses of patients were geocoded and aggregated in administrative units for mapping. Polymerase chain reaction test was performed for genotype classification. In logistic regression models, sociodemographic characteristics were compared between Beijing and non-Beijing genotype, and the latter's association with drug resistance were examined. Results: Out of 383 unique isolates collected, 263(69%) were from male, with 192 (50%) aged ≥60 years, 42 (11%) born locally, 53 (14%) permanent residents, 24 (6%) resided in elderly homes, and 319 (83%) living in 67 (89%) administrative units within the hospital cluster boundary. Twenty-nine strains (8%) were resistant to one of the four first line anti-TB drugs, while 14 (4%) were resistant to more than one. Overall, 286 (75%) belonged to Beijing genotype family, 52 (14%) to non-Beijing genotype, and 44 (11%) undetermined. Isolates of male patients (OR = 2.25), born in Hong Kong (OR = 6.4, born outside Hong Kong as reference), and permanent residents (OR = 4.18) were more likely to belong to Beijing genotype. However, smoking, alcoholic, drug abuse, co-morbidities and drug resistance were not significantly different between Beijing and non-Beijing genotypes. Correlation between the proportion of Beijing genotype isolates and proportion of male general population in corresponding administrative units was marginally significant (correlation coefficient = 0.23, p = 0.058). In sensitivity analysis combining undetermined with non-Beijing genotype, male gender and local residency were more likely to be associated with Beijing genotype. Single drug resistance (OR = 4.42, no resistance as reference) was more likely while multidrug (OR = 0.34) and isoniazid resistance (OR = 0.35) were less likely with Beijing genotype. Conclusion: In Hong Kong, Beijing genotype was the predominant lineage, which was characterised by higher proportion of male gender and local residents. Consistent with previous studies in China, Beijing genotype was not significantly associated with drug resistance.