Multicenter Case-Control Study of Behavioral, Environmental, and Geographic Risk Factors for Talaromycosis, Vietnam

Abstract

Talaromycosis is a life-threatening fungal disease that primarily affects immunocompromised persons in Southeast Asia. We conducted a multicenter, case-control study recruiting participants with advanced HIV disease in Vietnam; 205 case-patients with culture-confirmed talaromycosis were matched to 405 control-patients by age, sex, and CD4 count. Occupational exposure to tropical plants (odds ratio [OR] 1.73 [95% CI 1.10-2.73]; p = 0.017) and to farmed animals (OR 2.07 [95% CI 1.20-3.55]; p = 0.009) were independent risk factors for talaromycosis. Talaromycosis risk was higher in participants from highland regions than in persons from lowland regions (p<0.05). Participants from lowland regions who had lived or traveled to highland regions had a higher risk for talaromycosis (OR 3.15 [95% CI 1.49-6.64]; p = 0.003). This study confirms the epidemiologic correlation between talaromycosis and soil exposure and demonstrates an epidemiologic link between talaromycosis and residence in or travel to highland regions of Vietnam.

Keywords: HIV/AIDS and other retroviruses; Penicillium marneffei; Talaromyces marneffei; Vietnam; bamboo rat; fungi; penicilliosis; risk factors; talaromycosis.

Figure

Geographic distribution of recruited cases and controls in multicenter study of behavioral, environmental, and geographic risk factors for talaromycosis, Vietnam. A) Municipal regions of Vietnam, showing topography and locations of recruitment centers in Hanoi and Ho Chi Minh City; B) number of cases by region; C) number of controls by region; D) ratio of cases to controls by region. Darker color represents a higher number of cases relative to controls for that region. A case-to-control ratio of 0.5 is expected because 2 controls were recruited for every case. A ratio <0.5 indicates a lower-than-expected number of cases relative to controls. A ratio >0.5 indicates a greater-than-expected number of cases relative to controls. Data for northern regions of Vietnam are excluded from panel D because study enrollment from these regions was insufficient to draw meaningful conclusions from the analysis. Both cases and controls are concentrated in Southern Vietnam, where most participants were recruited. The ratios of cases to controls are higher in the Central Highlands and the adjacent Southeast and South-Central Coast regions than in Ho Chi Minh City and the Mekong Delta; the highest ratio was seen in the Central Highlands region.