653   Effects of Protease Inhibitors on Insulin, Insulin Resistance, and Lipids in Children and Adolescents.

. L. A. Danziger-Isakov*, K. McGann, P. Ballard, P. Hruz, N. H. White, and G. A. Storch .
Washington Univ. Sch. of Med., St. Louis, MO.

Background:Metabolic effects of protease inhibitors (PI) have been evaluated in adults, but little is known about their effects in children.

Methods:An ongoing cross-sectional analysis of HIV-positive children is being conducted. Patients have fasting insulin, glucose, c-peptide and lipid levels followed by an oral glucose tolerance test (OGTT). Lipid levels are considered elevated if they are >95th percentile for age, sex, and race normalizations based on NHANES data.

Results:Thirteen HIV-infected patients have been studied to date, 11 of whom have received PIs. Nine were on nelfinavir at testing, while the other two had exposure to it in the past. Seven are male. Subjects range from 11 months to 20 years (avg 8.2 yrs.) and had PI treatment for 7 to 36 months (avg. 24.5 mos.). Tanner staging (TS) differentiated 2 groups. Ten were TS 1-2, while three adolescents (1 M/2 F) were TS5. Few abnormalities were seen in the pre-pubertal group, with only the two oldest males having mildly increased triglycerides. Among the adolescents, the TS5 male also had hypertriglyceridemia. The oldest male (20 yrs.; BMI 33 kg/m²; TS5) and female (16 yrs.; BMI 25mg/m²; TS5) had impaired glucose tolerance (2-hr PC glucose >140mg/dL) and elevated c-peptide. Both of these patients had family history of non-insulin-dependent diabetes mellitus, while only one other has this history. The oldest male also had a fasting insulin level of 45mU/ml (normal <20) suggestive of insulin resistance. Lipase and hepatic transaminases were normal in all subjects. Ten (91%) of subjects on PI showed weight and height gain across percentiles for age without clinical evidence of peripheral wasting.

Conclusions:In our pediatric population, hypertriglyceridemia was seen only in the older males. Only one of 11 PI subjects showed both glucose intolerance and insulin resistance. To what extent obesity and family history of type 2 diabetes may contribute to insulin resistance in the subject is unknown. Abnormalities of insulin resistance were not seen in the prepubertal subset. Responses to PI in children may be different than in adults. The relationship between PI-induced hypertriglyceridemia and insulin resistance in children requires further evaluation.