Objective: To describe the prevalence of low serum Se and determine whether HIV, hepatitis C virus (HCV) and/or the types of drugs used are associated with serum Se in a cohort of infected and uninfected drug users.
Design: Independent correlates of low serum Se levels based on data collected from food recalls, physical examinations and clinical questionnaires were identified using multivariate regression analysis.
Setting: Buenos Aires, Argentina
Subjects: A total of 205 (twenty-five female and 180 male) former and current drug users.
Results: Drug users had an average serum Se level of 69·8 (sd 32·8) μg/d, [corrected] and 82 % were considered deficient (<85 μg/l). [corrected] Multivariate analyses found that HIV- and/or HCV-infected individuals had lower mean Se compared with healthy, uninfected drug users (HIV/HCV co-infection: -25·3 μg/l (se 7·6), P = 0·001; HIV alone: -28·9 μg/l (se 6·9), P < 0·001; HCV alone: -19·4 μg/l (se 7·1), P = 0·006). Current and previous drug use was associated with higher serum Se. Cigarette smoking and heavy alcohol consumption were not found to be associated with Se status.
Conclusions: Low serum Se levels are highly prevalent among drug users in Buenos Aires, Argentina. Se supplementation and/or dietary interventions may be warranted in drug users who are at high risk for HIV and/or HCV infection.