DOI 10.1515/jpem-2013-0253      J Pediatr Endocr Met 2014; 27(5-6): 403–412

Prapai Dejkhamron*, Kevalee Unachak, Linda Aurpibul and Virat Sirisanthana

Insulin resistance and lipid profiles in HIVinfected Thai children receiving lopinavir/ ritonavir-based highly active antiretroviral therapy Introduction

Abstract Background: Lopinavir/ritonavir (LPV/r) is associated with insulin resistance (IR). We aim to determine the prevalence of IR, dyslipidemia and their inter-relationships with adipokines in HIV-infected children treated with LPV/r-based highly active antiretroviral therapy (HAART). Methods: Twenty-eight children were enrolled. Fasting glucose, insulin, lipid profiles, adipokines, and oral glucose tolerance tests were performed. Results: The prevalence of IR, pre-diabetes mellitus, and hypertriglyceridemia was 42.9(12/28), 10.7(3/28), and 75.0(21/28)% respectively. No case met the definition for diabetes mellitus (DM) and lipodystrophy. Children with IR had higher BMI z-score, triglyceride levels but unchanged leptin or adiponectin levels compared to those without IR. Longer duration of LPV/r-based HAART was associated with increased levels of triglyceride and total cholesterol. Conclusions: We describe high prevalence of IR, prediabetes mellitus, and dyslipidemia among HIV-infected children receiving LPV/r-based HAART. Pre-diabetes mellitus or DM or IR screening might be important for early diagnosis and intervention in these children. Keywords: adipokines; insulin resistance; human immunodeficiency virus-1; lipid profiles; lopinavir/ritonavir. *Corresponding author: Prapai Dejkhamron, Division of Pediatric Endocrinology, Faculty of Medicine, Department of Pediatrics, Chiang Mai University, Chiang Mai, 50200 Thailand, Phone: +66-53-945412, Fax: +66-53-946461, E-mail: [email protected] Kevalee Unachak: Division of Pediatric Endocrinology, Faculty of Medicine, Department of Pediatrics, Chiang Mai University, Chiang Mai, Thailand Linda Aurpibul and Virat Sirisanthana: Research Institute for Health Sciences, Chiang Mai University, Chiang Mai, Thailand

Insulin resistance (IR) is a condition which is associated with cardiovascular risk, diabetes mellitus, hypertension, and a shorter lifespan (1). It is typically identified in obese individuals; however, several conditions e.g., genetic syndrome and medications (glucocorticoids, highly active antiretroviral treatments (HAART) for human immunodeficiency virus (HIV) infection are also associated with IR (1). Protease inhibitors (PIs) are antiretroviral agents which have been reported to be associated with IR (2). The most commonly used PI in children is lopinavir/ritonavir (LPV/r). It was demonstrated that ritonavir did not alter insulin sensitivity in healthy men. Therefore, lopinavir is likely responsible for the induction of IR (3). In healthy volunteers, a single dose of LPV/r decreased insulin sensitivity (4). Another group of investigators also demonstrated LPV/r-induced IR in HIV-negative adults receiving LPV/r for five days (5). It has been demonstrated that HAART can cause IR by either direct interfering with insulin signaling at the cellular level or an indirect effect on lipid metabolism (lipodystrophy) (2). PI induced IR by a direct inhibition of glucose transporter 4 and proteosome (2, 6). Mulligan et al reported that IR and dyslipidemia were induced by PI treatment in the absence of lipodystrophy (7). However, Carr et al. demonstrated that treatment with PI induced lipodystrophy, dyslipidemia, and IR in HIVinfected patients (8). PI-induced IR and lipodystrophy can be associated with dyslipdemia. However, dyslipidemia could be caused by a direct effect of PI on adipocyte differentiation and triglyceride accumulation resulting in increased circulating triglyceride levels (9). Hypertriglyceridemia is very common in HIV-infected patients treated with PI (10). Treatment with four weeks of LPV/r in HIV-negative men caused an increase in triglyceride levels, and very low density lipoprotein (VLDL) cholesterol (11). The incidence of increased serum triglyceride levels was reported in 60.7% of patients treated with LPV/r (12). Increased total

Brought to you by | University of California - San Francisco Authenticated Download Date | 2/17/15 1:28 PM

404      Dejkhamron et al.: Insulin resistance in HIV-infected children cholesterol, high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C) or nonhigh-density lipoprotein cholesterol (non-HDL-C) levels have been reported in HIV-infected children treated with HAART (13–17). The pathogenesis of IR in HIV-infected individuals treated with HARRT appears to correlate with the level of leptin and adiponectin which are adipokines produced from the adipose tissue. Decreased adiponectin levels were reported to be associated with central fat and mixed lipohypertrophy, and with signs of IR in HIV-infected adolescents (18). Increased leptin level and leptin/adiponectin ratio were associated with lipodystrophy in HIVinfected children treated with LPV/r-based HAART. These adipokines may be surrogate markers of lipodystrophy (19). Low adiponectin levels were also reported to be associated with IR in HIV-infected adolescents (18). We hypothesise that LPV/r-based HAART would have impact on glucose, IR, lipid profiles, and adipokines. We aim to determine the prevalence of IR, dyslipidemia and their inter-relationships in HIV-infected Thai children, with the hope that any detectable metabolic problems could be tackled early in their life in order to prevent subsequent clinically significant metabolic derangement.

Methods

on physical examination (21). An oral glucose tolerance test (OGTT) was performed after overnight fast. At baseline, blood samples were collected and analysed for fasting blood sugar, lipid profiles, insulin, glycosylated haemoglobin (HbA1C), c-peptide, leptin and adiponectin. All children then received glucose orally at 1.75 g/kg of body weight to a maximum of 75 g. Blood samples were obtained at 15, 30, 60, 90, 120 min and immediately analysed for glucose, and lipid profiles. For insulin HbA1C, c-peptide, leptin and adiponectin, sera were separated and stored at –20°C until the day of assay.

Laboratory measurements Plasma glucose and lipid profiles were analysed by SYNCHRON® System (Beckman Coulter, Bangkok Thailand). HbA1C was measured by immunoturbidity (Roche Diagnostics, Bangkok Thailand). Serum insulin and c-peptide were determined by electrochemiluminescence (Roche Diagnostics, Bangkok Thailand). Leptin levels were analysed by enzyme-linked immunosorbentassay (ELISA) (Global Biotechnology, Bangkok Thailand). Adiponectin levels were measured by ELISA (Adipogen, South Korea).

Insulin resistance Surrogates that were used to estimate insulin sensitivity and resistance include, homeostasis model assessment of insulin resistance (HOMA-IR) and insulin sensitivity index-Matsuda (ISI-Matsuda). They were calculated using the following formulas; HOMA-IR = {[fasting insulin (μU/mL)] × [fasting glucose (mmol/L)]}/22.5, ISI-Matsuda

Population study Perinatally HIV-infected children who were actively followed at the Pediatric Infectious Clinic, Chiang Mai University Hospital in 2011, and were on PI-based second-line HAART were recruited. Inclusion criteria were 1) age between 3 and 18 years, 2) weight  > 20 kg, and 3) receiving LPV/r-based regimen. The rationale for lower age or weight cut-off is due to the need to fast overnight and blood draw limitation. Children with history of endocrine diseases or metabolic derangement from other causes including illnesses or medications, as well as any significant conditions which might place the children at a risk of discomfort in the opinion of investigator were excluded. The study protocol was reviewed and approved by the Ethic Committee at the Faculty of Medicine, Chiang Mai University. Assent and informed consent were obtained from all participants and their parents or legal guardians, respectively.

Study procedures Data including gender, age, weight, height, body mass index (BMI), CDC clinical category (20), CD4 T lymphocyte counts, and HIV-RNA levels were obtained from medical records. BMI z-score was calculated using WHO AnthroPlus for personal computers (http://www.who.int/ growthref/tools/en/). Lipodystrophy was clinically diagnosed based

   mg   mU   = 10,000 /  Gfasting    ×(Gmean× Imean)   × Ifasting      dL L   IR was defined as 1) a fasting insulin   ≥  15 μU/mL or 2) a peak insulin   ≥  150 μU/mL or 3) a 2-h insulin during OGTT   ≥  75 μU/mL (22) or 4) HOMA-IR  > 2.5 in prepubertal children and  > 4.0 in adolescents (Tanner stage  > 1) (23) or 5) ISI-Matsuda  

ritonavir-based highly active antiretroviral therapy.

Lopinavir/ritonavir (LPV/r) is associated with insulin resistance (IR). We aim to determine the prevalence of IR, dyslipidemia and their inter-relatio...
646KB Sizes 0 Downloads 0 Views