BACKGROUND: HIV-exposed, uninfected (HIV-EU) children represent a large proportion of children in southern Africa. The reasons for their poorer growth and higher morbidity and mortality than their HIV-unexposed peers are unclear. OBJECTIVE: We compared anthropometry of 125 HIV-EU with 382 HIV-unexposed young Zambian children participating in a trial of micronutrient-fortified complementary/replacement food. DESIGN: The randomized controlled trial provided children from age 6 to 18 months with a porridge flour containing either a basal or a rich level of micronutrients. Weight and length were measured 3 monthly and head and arm circumferences and triceps and subscapular skinfolds 6 monthly. RESULTS: There were no significant anthropometric differences between the 2 treatment groups. In unadjusted analyses, most anthropometric Z scores of HIV-EU children were lower than those of HIV-unexposed children; after adjustment for treatment arm, socioeconomic factors, breastfeeding and sex, head and arm circumference Z scores remained lower. Subscapular skinfold Z scores were lower among HIV-EU than HIV-unexposed children at 6 months but not 18 months. CONCLUSIONS: Socioeconomic factors accounted for some but not all of the impaired growth of HIV-EU children. Micronutrient malnutrition may not be the socioeconomic factor responsible for the growth faltering. Factors acting earlier in life had irreversible effects.