The study protocol was approved by the Danish Ethics Committee on clinical research, and written informed consent was obtained from all participants. As most variables, even after log transformation, were not normally distributed, nonparametric statistics were applied; thus, data are presented as medians and interquartile ranges (IQRs). Comparisons between controls and HIV-positive patients were performed using the Mann–Whitney test (unpaired data), and

comparisons between treatment-naïve and treatment-experienced patients were performed using the Wilcoxon signed rank test (paired data). The correlations between variables were determined using Spearman’s correlation coefficient. A value of P < 0.05 was considered significant. The baseline characteristics of selleck chemical the patient and control groups selleck are shown in Table 1. During the first 3 months, 11 patients were treated with boosted indinavir, three of whom were changed to boosted lopinavir because of side effects. One patient left the study because of side effects. Nine patients received boosted lopinavir throughout the first period. One patient was unwilling to change therapy to efavirenz

and was excluded from the second part of the study (Fig. 1). After 3 months, two-thirds of the patients had viral loads (VLs) below 50 copies/mL, and after 6 months all 18 had a VL below this value. CD4 counts G protein-coupled receptor kinase increased from a median of 160 cells/μL (IQR 125–200 cells/μL) to 220 cells/μL (IQR 160–300 cells/μL) after 3 months of treatment, and to 215 cells/μL (IQR 180–280 cells/μL) after 6 months of treatment. Controls had a median CD4 count of 770 cells/μL (IQR 730–900 cells/μL). At entry and throughout the study period, HIV-positive patients had lower haemoglobin and a lower total leucocyte count compared with controls. Platelet numbers did not differ between patients and controls (Table 2). Total cholesterol, triglyceride, and glucose levels were

normal at baseline (Table 2). During treatment with both a PI and efavirenz, total cholesterol increased significantly compared with baseline. Similarly, PI treatment led to significantly higher triglyceride levels. However, this was negated during treatment with efavirenz, and lowered again to a level comparable to that of the controls (1.47 vs. 0.83 mmol/L, respectively; P = 0.15). Body mass index (BMI) and systolic blood pressure were normal at baseline and did not change during the treatment period. Endothelial function was studied in several ways (Table 3). HIV-positive patients had significantly lower FMD at baseline compared with controls (108 vs. 111%, respectively; P = 0.043) (Table 3). After 3 months of PI-containing HAART, FMD normalized (111%) and did not change significantly after switching from a PI to efavirenz (111 vs. 109% in HIV-positive patients treated with PI vs.