The Influence of Antiretroviral Therapy on the QTc Interval in an African Cohort To the Editor—Cardiovascular disease in human immunodeficiency virus (HIV) infection encompasses a wide range of pathologic entities, including myocardial, pericardial, and endocardial disease, ath- erosclerosis, arrhythmias, and vasculitis. The most common manifestations of HIV-associated heart disease in sub- Saharan Africa are pericarditis, cardiomy- opathy, and pulmonary hypertension [1]. Coronary artery disease, lipodystrophy, and metabolic syndrome, although com- mon in developed countries, are tradi- tionally thought to be less clinically significant problems in the African sub- continent [2]. It is well known that pro- longation of the QT C interval can predispose patients to potentially fatal ventricular tachyarrhythmias, particularly torsades de pointes, and thus is an in- dependent predictor of cardiovascular morbidity and mortality [3]. The aim of this study was to determine whether there is an association between antiretroviral (ARV) therapy and pro- longed QT in HIV-positive ARV-treated patients in Nairobi, Kenya. HIV-positive adults from both inpatient and outpatient departments at Aga Khan Teaching Hos- pital, Nairobi, were enrolled with exclu- sion criteria including any known cardiac disease. Patients were enrolled into either the ‘‘ARV-experienced’’ or ‘‘ARV-naive’’ arm of the study. The study was approved by the local ethics committee, and writ- ten informed consent obtained from all patients. A standard 12-lead electrocardiogram was obtained in each patient, and a QTc interval of .440 ms was considered pro- longed. A total of 299 HIV-positive patients were screened, 157 in the ARV- experienced arm and 142 in the ARV- naive control arm. A total of 27 patients in the ARV-experienced arm and 12 in the ARV-naive arm were excluded due to protocol violations, leaving 130 patients in each arm of the study (Table 1). QT C in- terval prolongation was observed in 16.2% of patients in the ARV-experienced group, compared with 6.9% in the ARV-naive control group (v2 5 5.43; P 5 .01); the odds ratio was calculated as 2.5 with a 95% confidence interval (1.01–6.67; P 5 .02). The overall prevalence of QTc prolongation across the whole subject co- hort (ARV-experienced and ARV- naive) was 11.5%. When patients were divided by sex, prolongation was observed in 11.3% of male and 11.8% of female patients. The majority of patients with QTc prolongation had QTc intervals of 440–469 ms, and none had QTc intervals .500 ms (Table 2). There did not appear to be any significant difference in preva- lence of QT C prolongation between pa- tients receiving a nonnucleoside analogue reverse-transcriptase inhibitor–based regi- men and those receiving a protease inhibitor–based regimen (P 5 .78). In addition, there was no significant differ- ence in prevalence of QTc prolongation between patients with World Health Organization stage I/II HIV infection and those with stage III/IV infection (v2 5 0.52; P 5 .47), nor were there correlations with CD4 cell count (r 520.043), in contrast to findings of other studies [4–6]. Literature on the association of QTc prolongation and ARV therapy is surpris- ingly sparse [7–10]. Our study shows that ARV therapy does appear to confer a significant increased risk of QT C prolongation in HIV-positive patients, compared with ARV-naive controls. The increased risk has the potential to predispose otherwise medically stable patients to the development of potentially fatal ventricular arrhythmias, although the exact significance of an acquired asymp- tomatic QT C prolongation in clinical practice is yet to be established. These results should remind physi- cians of the importance of measuring and monitoring the QT C interval for pa- tients receiving ARV therapy and tailoring management accordingly, to successfully minimize cardiovascular morbidity and mortality as ARV drugs become readily available. Notes Acknowledgements. We thank Dr Jowi and Dr Sitna for their valuable critique of the paper. Financial support. This work was supported by the Research Support Unit of Aga Khan University Hospital, Nairobi, Kenya. Potential conflicts of interest. All authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider rel- evant to the content of the manuscript have been disclosed. J. Shavadia, 1 R. Shah, 1 G. Yonga, 1 R. Patel, 2 J. Stebbing, 2 and M. Nelson 2 1 Departments of HIV Medicine and Cardiology, Aga Khan University Hospital, Nairobi, Kenya; and 2 Department of HIV Medicine, Imperial College, Chelsea and Westminster Hospital, London, United Kingdom References 1. Ntsekhe M, Mayosi BM. Cardiac man- ifestations of HIV infection: an African 448 d CID 2012:54 (1 February) d CORRESPONDENCE