Background: There is limited information on the relation between glucose levels in pregnancy and adverse perinatal outcomes in HIV-infected pregnant women. Objective: To evaluate the potential impact of fasting glucose levels on pregnancy outcomes in a large sample of pregnant women with HIV from a national study, adjusting for potential confounders. Methods: Data from the Italian National Program on Surveillance on Antiretroviral Treatment in Pregnancy were used. The main outcomes evaluated in univariate and multivariable analyses were birthweight for gestational age >90th percentile (large for gestational age [LGA]), nonelective cesarean delivery, and preterm delivery. Glucose measurements were considered both as continuous and as categorical variables, following the HAPO study definition. Results: Overall, 1,032 cases were eligible for the analysis. In multivariable analyses, a birthweight >90th percentile was associated with increasing fasting plasma glucose levels (adjusted odds ratio [AOR] per unitary (mg/dL) increase, 1.04; 95% Cl, 1.01-1.06; P = .005), a higher body mass index, and parity of 1 or higher. A lower risk of LGA was associated with smoking and African ethnicity. A higher fasting plasma glucose category was significantly associated with LGA occurrence, and AORs for the glucose categories of 90-94 mg/dL and 95-99 mg/dL were 3.34 (95% Cl, 1.09-10.22) and 6.26 (95% Cl, 1.82-21.58), respectively. Fasting plasma glucose showed no association with nonelective cesarean section [OR per unitary increase, 1.00; 95% Cl, 0.98-1.02] or preterm delivery [OR per unitary increase, 1.00; 95% Cl, 0.99-1.02]. Conclusions: In pregnant women with HIV, glucose values below the threshold usually defining hyperglycemia are associated with an increased risk of delivering LGA infants. Other conditions may independently contribute to adverse perinatal outcomes in women with HIV and should be considered to identify pregnancies at risk.