8% at months 12, 24, 36 and 48, respectively). Regarding plasma lipid levels (Fig. 1d and e) we did not observe significant changes during the follow-up period. We found hypercholesterolaemia (>200 mg/dL) in 9.5, 30.4, 21.7, Roscovitine mw 14.3 and 13.3% of patients at months 0, 12, 24, 36 and 48, respectively, and hypertriglyceridaemia (>170 mg/dL) in 14.3, 8.3, 13, 4.5 and 0% of patients at the same time-points. Throughout follow-up, and especially at the end of the study, we found an increase in plasma resistin and significant increases in total plasminogen activator inhibitor type 1 (tPAI-1), adiponectin and leptin levels (P<0.05) (Fig.

1f–i). Regarding the leptin:adiponectin ratio, HOMA values and C-peptide levels, we observed a slight increase during the first few months on HAART followed by a moderate decrease or stabilization after 24 months on HAART (Fig. 1j–l). The median BMI did not change significantly during follow-up, with values being between 17.32 and 16.42. There were no children in the overweight and low-weight BMI categories. LDK378 Concerning diagnoses of lipoatrophy, 17 children had no lipoatrophy, three had mild lipoatrophy, five had moderate lipoatrophy and two had severe lipoatrophy. Overall, seven of the 27 patients (25.9%) had lipoatrophy with scores ≥2. Concerning lipohypertrophy, 16 children did not have lipohypertrophy, three had mild lipohypertrophy, five had moderate lipohypertrophy and three had severe

lipohypertrophy. Overall, eight of the 27 patients (29.6%) had lipohypertrophy with scores ≥2. By the end of the study, 12 of the 27 children (44.4%) had lipodystrophy. However, only three of the 27 children (11.1%) had both lipoatrophy and lipohypertrophy scores ≥2. We carried out a follow-up study in PI-naïve HIV-infected to children on HAART for 4 years, and found an increase in adipokine levels. This increase could be related to the

direct effect of PIs on adipose tissue, which could contribute to an imbalance in lipid metabolism and spatial development of lipodystrophy and metabolic syndrome in HIV-infected patients [21]. The metabolic pathway and the cytokine profile accomplices in the development of lipodystrophy and lipoatrophy is very complex. Thus, we did not find any significant trend in adipokine kinetics that may be associated with the onset of lipodystrophy at the end of the study. Moreover, results did not differ between patients with complete HIV suppression and those failing therapy. Therefore, we cannot definitely conclude from these results that there is a direct effect of HAART on adipose tissue, but there is a trend that warrants further investigation in studies with another design. In the present study we used clinical assessments of lipodystrophy; Dual Energy X-ray Absortiometry (DEXA) scanning would have provided a more quantitative assessment of lean vs. fat mass (particularly visceral fat) and may have provided better insights into the potential relationship between fat changes and adipokine levels.