Predictors of incident hypertension in HIV-positive adults over 24 months on ART in South Africa

By  Alana Brennan  Matthew Fox  Dr Mhairi Maskew  Professor Ian Sanne  Prudence Ive  Julia Rohr  Lawrence Long  Sydney Rosen  |  | 

Abstract 

Background: As the current HIV population ages and access to antiretroviral therapy (ART) increases in resource-limited settings, the number of patients developing non-communicable chronic diseases on ART will also increase. In view of limited research on this topic in resource-limited settings, we sought to examine predictors of hypertension in a large urban HIV clinic in South Africa. Methods. Prospective study of hypertension over 24 months on ART among 12376 ART naïve adults initiating ART at Themba Lethu Clinic in Johannesburg between April 2004-2011. Patients with hypertension at ART initiation were excluded. We used Cox proportional hazards regression to examine demographic and clinical characteristics at ART initiation in relation to incident hypertension. We first defined hypertension as systolic blood pressure (BP) >140 and/or diastolic BP >90 mmHg and/or documentation of hypertensive medication. We then categorized it as mild (systolic BP 140-159.9 and/or diastolic BP 90-99.9 mmHg) or moderate/severe (systolic BP >160 and/or diastolic BP >100 mmHg). Person-time started at ART initiation and ended at the earliest of hypertension, loss, death, transfer or completion of 24 months of follow-up. Results: Among 17378 eligible patients, 28.8% (5002) had documented hypertension at ART initiation. The remaining 1 12 376 were predominately female (62%) and on stavudine-lamivudine-efavirenz (75%). At ART initiation patients had a median age of 36 years (IQR: 31-42), median CD4 count of 94 cells/mm3 (IQR: 35-166), median systolic blood pressure of 113 mm/Hg (IQR: 103-123) and median diastolic blood pressure of 73 mm/Hg (IQR: 66-80). By 24 months, 2116 (17%) developed hypertension, of whom 1857 (88%) were mild and 259 (12%) were moderate/severe. Regression models showed older patients (>40 years), males, and patients with a BMI >25 kg/m2 had an increased rate of hypertension over 24 months (Table). Results for mild hypertension were similar, while results for moderate/severe hypertension showed a higher hazard for patients >50 years and those with a BMI >35 kg/m2. Conclusion: 20% of patients in our cohort developed hypertension within 24 months on ART. Obese patients (BMI >30kg/m2) and those older >40 years of age, should be targeted for frequent BP monitoring and identification of other cardiovascular risk factors in order to implement lifestyle modifications and pharmaceutical therapy as indicated to help prevent myocardial infarction, heart failure, stroke and kidney disease.

Conference: CROI 2014, Boston, USA

Publication details

PDF