Lipoatrophy and lipohypertrophy have been observed during long-term combination antiretroviral therapy (CART). We investigated whether consumption of a Mediterranean diet is associated with lower risk of body-shape changes in Croatian patients treated with CART. Between May 2004 and June 2005, we conducted a cross-sectional study of 136 adults with HIV-1 infection who were treated with CART for at least 1 year. Lipoatrophy and lipohypertrophy were assessed by self-report and physical examination. Adherence to a Mediterranean diet was determined by a 150-item questionnaire; a 0-9 point diet scale was created that stratified respondents as having low adherence (<4 points) and moderate to high adherence (> or =4 points). Lipoatrophy was present in 41% and lipohypertrophy in 32% of participants. Non-smokers with a dietary score > or =4 had the lowest risk for lipoatrophy. Stavudine use, female gender, and duration of CART were also independently associated with a higher risk of lipoatrophy. A dietary score of > or =4 was associated with lower risk of lipohypertrophy (adjusted OR 0.3, 95% CI 0.1-0.7; P = 0.012). Female gender, longer duration of CART, and longer known duration of HIV infection prior to CART were also independently associated with higher risk of lipohypertrophy. In conclusion, Croatians who did not smoke and moderately or highly adhered to the Mediterranean diet were least likely to have the clinical syndrome of lipoatrophy. Moderate to high adherence to a Mediterranean diet was associated with a lower risk of lipohypertrophy.