Efa virenz levels and treatment outcomes in Kenyan HIV-TB co-infected patients at Kenyatta national hospital

Abstract

Efavirenz is a non-nucleoside reverse transcriptase inhibitor specific against HlV-l that is used 3.S a component of HAART in combination with two NRTls at an adult dose of 600mg daily. Rifampicin is a semi-synthetic derivative of rifamycin that is administered with other anti-TB drugs for the first line management of active TB at an adult dose of 300-600mg per day. Rifampicin induces the cytochrome P450 enzyme system in the liver which also metabolizes efavirenz resulting in decreased levels of efavirenz. When this is coupled with the large interpatient and intra-patient variability in efavirenz plasma concentrations, it could lead to subtherapeutic concentrations hence treatment failure or toxic concentrations hence adverse effects. The aim of this study was to determine the plasma levels of efavirenz and the treatment outcomes in Kenyan HIV -TB co-infected patients receiving concomitant Anti-TB therapy and ART at Kenyatta national hospital. The study was designed as a concurrent descriptive cohort study. HIV-TB co-infected Kenyan patients attending the comprehensive care centre of Kenyatta National Hospital who met the study inclusion criteria were recruited into the study. Following approval by the Kenyatta .. National Hospital ethics and research committee and a written informed consent. 28 HIV -TB co-infected patients of Kenya origin were recruited into the study from January 2010 to March 2010. Patient demographics, history, baseline data and treatment outcomes were obtained by medical record review. Plasma concentrations of efavirenz were determined by reverse phase high performance liquid chromatography with UV detection (HPLC-UV). Two blood samples were obtained from each subject on two separate occasions separated by a minimum of 2 weeks. Data analysis was performed by SPSS version 12.0 software. Descriptive and exploratory data analysis was carried out for each variable. Intra and inter-patient variability was determined using the coefficient of variation. Comparison of plasma concentrations at first and second occasions and with different variables was done using various nonparametric tests, Post hoc multiple comparison and Spearman's rho correlation. Multivariate data analysis was done by general linear regression (repeat measures). The median plasma concentrations (Range) at the first and second sampling occasions were 13.17 (0.04 - 52.48) ug/ml and 10.66 (3.54 - 57.61) ug/ml. On l" sampling one patient (3.7 %) had sub-therapeutic plasma EFV concentrations «1 ug/ml) while only one patient (3.7 %) had therapeutic plasma EFV concentrations (1-4 ug/ml). A majority of patients (92.6 %) had toxic plasma EFY concentrations (>4 ug/ml). Out of these 25 patients with toxic levels, 68 % had levels between 4 and 20 ug/ml while 32 % had levels above 20 ug/ml. On the second sampling occasion, no patient had sub-therapeutic plasma EFY concentrations while only one patient (4.3 %) had therapeutic plasma EFV concentrations. A majority of patients (95.7 %) had toxic plasma EFY concentrations. Out of these 22 patients with toxic levels, 72.7 % had levels between 4 and 20 ug/ml while 27.3 % had levels above 20ug/ml. A high intra-patient variability of 32.5 % was observed and a high inter-patient variability of 77.8 % on first and 86.4 % on second sampling was observed. Females consistently had higher efavirenz plasma concentrations than male patients though this was not statistically significant. Bodyweight was a statistically significant factor with patients weighing less than 60 kilograms having higher EFY levels than those weighing >=60 kilograms. Ethnicity was also a significant factor with higher efavirenz levels in the Kamba compared to Kikuyu, Kisii, Luhya and Kalenjin being demonstrated. Multivariate analysis of bodyweight, ethnicity and gender showed that they were not statistically significant determinants of efavirenz levels and a large study is required to assess the contribution of these variables to efavirenz levels. Treatment outcomes were difficult to assess irf this study. Although there was a statistically significant difference in the CD4 counts, this was not correlated to the high efavirenz levels. There were 24 episodes of adverse effects in 28 patients implying that nearly every patient would experience an adverse effect during treatment. The prevalence of adverse effects increased on onset of anti- TB therapy and gradually reduced. Skin adverse effects were the most common followed by drug induced hepatitis. Gastrointestinal adverse effects presented early while neurological and metabolic adverse effects presented later. , Contrary to the expectation that efavirenz levels would be lower m patients recerving rifampicin, a large proportion of Kenyan HIY -TB co-infected patients have toxic efavirenz levels and high rates of adverse effects were observed in these patients. Clinicians should therefore be particularly vigilant for signs of toxicity. A priori dose reduction in native African patients, a pharmacogenetic-therapeutic drug monitoring approach to individualised dosing and a prospective clinical dose optimization and treatment outcome study in Kenyans and indeed native Africans are immediately required.