Effect of dietary intervention on highly active antiretroviral therapy-related dyslipemia

Abstract

Changes in cholesterol and triglyceride levels after prescribing a lipid-lowering diet were assessed in 230 HIV-infected patients with dyslipemia associated with antiretroviral therapy. Lipid levels decreased significantly in subjects having good diet compliance. The reduction in triglyceride levels was greater than in cholesterol levels. Patients on protease inhibitor-containing regimens experienced a significantly greater decline in lipid levels compared with the rest. Dyslipemia associated with highly active antiretroviral therapy (HAART) is a prevalent condition in patients living with HIV [1–4]. Both hypercholesterolemia and hypertriglyceridemia constitute well-recognized cardiovascular risk factors [5,6]. This fact raises major concerns because HIV infection has become a chronic disease in most infected patients. Episodes of serious premature vascular events (mainly myocardial infarction) have begun to be reported in individuals on HAART [7–10]. Pancreatitis also represents a relevant complication of severe hypertriglyceridemia [11,12]. Therapeutic interventions in HIV-infected patients with lipid disorders include low-fat diet, exercise, smoking cessation, drug therapy (fibrates and statins), and a modification of antiretroviral therapy [13]. No specific recommendations for the management of dyslipemia in these individuals exist, and interventions are based on existing National Cholesterol Education Program guidelines for the general population [5,6]. The use of lipid-lowering drugs is of concern given their toxicity, intolerance, and potential interactions with antiretroviral agents. Non-pharmacological interventions (diet and exercise) should always be the first line of intervention [13]. In HIV-uninfected individuals, dietary advice may achieve a 11 and 22% decline in cholesterol and triglyceride levels, respectively [14]. In HIV-infected individuals, these figures range from 4 to 17% and from 21 to 26%, respectively [2,15]. We assessed prospectively the efficacy of a low-fat diet on cholesterol and triglyceride level reduction in a well-characterized group of 230 HIV-infected patients under HAART with dyslipemia. Subjects with hypercholesterolemia (cholesterol ≥ 200 mg/dl) or hypertriglyceridemia (triglycerides ≥ 200 mg/dl) were prescribed a lipid-lowering diet and followed for 3 months (n = 161) and 6 months (n = 70). None received drugs for dyslipemia, and antiretroviral regimens remained unchanged during the study period. Fasting lipid levels, lipid variation (percentage), and nutritional status (weight and serum albumin) were recorded longitudinally. Mean values were compared according to diet adherence and type of antiretroviral regimens. Student's T and Mann–Whitney U tests were used when needed. Variables examined in the linear regression multivariate analysis included diet adherence, physical activity, and antiretroviral treatment. The mean age of patients was 40 ± 8 years, and 81% were men. Their mean CD4 lymphocyte count was 535 ± 274 cells/mm3, and 66% of subjects had undetectable viral loads (< 50 HIV-RNA copies/ml). Lipodystrophy body shape changes were present in 66% of individuals. Within the study population, 40% had only hypercholesterolemia, 7% had only hypertriglyceridemia, and 53% had both. Mean values of cholesterol and triglycerides in patients with hypercholesterolemia and hypertriglyceridemia were 257 ± 38 mg/dl and 320 ± 108 mg/dl, respectively. The antiretroviral regimens administered included two nucleoside reverse transcriptase inhibitors plus either one protease inhibitor (PI) (47%), one non-nucleoside reverse transcriptase inhibitor (50%), or another nucleoside reverse transcriptase inhibitor (3%). Overall, 58 (36%) and 32 (45%) patients reported a good compliance with the prescribed diet at 3 and 6 months, respectively. The mean values of lipid levels significantly decreased compared with baseline in those subjects (Fig. 1). The reduction at 3 and 6 months was of 11% (n = 58) and 10% (n = 28) for cholesterol, and of 12% (n = 31) and 23% (n = 20) for triglyceride levels, respectively. In subjects who were non-adherent to diet, cholesterol values decreased in 1% (n = 92) and 2% (n = 35), and triglyceride values decreased in 1% (n = 66) and 9% (n = 21), respectively, at 3 and 6 months.Fig. 1.: Lipid variation according to diet compliance. ––♦–– Good compliance; ––▪–– poor compliance. *P < 0.05 compared with baseline.Good diet compliance was associated with a statistically significant weight loss: 2.2 and 2 kg at 3 and 6 months, respectively. In subjects who reported poor adherence to dietary advice, no significant changes in weight were found. Serum albumin values remained stable regardless of diet compliance. With respect to antiretroviral treatment, the decrease in lipid levels in patients adherent to diet was greater in those under PI-based combinations, being statistically significant for both cholesterol and triglyceride values at 6 months (Fig. 2). In subjects on PI, cholesterol and triglyceride reductions were 13 and 15% at 3 months (n = 48), and 22 and 49% at 6 months (n = 17), respectively. In contrast, in patients under PI-sparing regimens, these figures were 8 and 7% at 3 months (n = 41), and 3 and 8% at 6 months (n = 31), respectively.Fig. 2.: Lipid variation in patients with good adherence to dietary advice according to antiretroviral treatment. ––♦–– Protease inhibitor (cholesterol); ––▪–– protease inhibitor (triglyceride); ––▴–– non-protease inhibitor (cholesterol); ––×–– non-protease inhibitor (triglyceride). *P < 0.05 compared with baseline.In multivariate analysis, physical exercise was significantly associated with triglyceride level variation, and good adherence to dietary advice was significantly associated with cholesterol level variation at 3 months. At 6 months, the variables significantly associated with a decline in lipids were good diet compliance with triglyceride level variation, and good diet compliance along with PI-based treatment with cholesterol level variation. In conclusion, a dietary intervention seems to be moderately effective for reducing lipid levels in HIV-infected patients with HAART-related dyslipemia. It provides the benefit of avoiding the potential toxicity of lipid-lowering drugs. Good diet compliance was followed by less than half of the subjects. In individuals with mild or moderate dyslipemia, a low-fat diet prescription could be sufficient to treat their metabolic abnormality, particularly if no other cardiovascular risk factors are present. Whereas good diet compliance may result in a significant reduction in triglyceride levels, without compromising the nutritional status, the benefit is less apparent on cholesterol levels. Reductions in cholesterol levels are more apparent in subjects under PI-based combinations.