SummaryIntroduction:Heart failure is a common cause of hospitalisation and therefore contributes to in-hospital outcomes such as mortality. In this study we describe patient characteristics and outcomes of acute heart failure (AHF) in Botswana.Methods:Socio-demographic, clinical and laboratory data were collected from 193 consecutive patients admitted with AHF at Princess Marina Hospital in Gaborone between February 2014 and February 2015. The length of hospital stay and 30-, 90- and 180-day in-hospital mortality rates were assessed.Results:The mean age was 54 ± 17.1 years, and 53.9% of the patients were male. All patients were symptomatic (77.5% in NYHA functional class III or IV) and the majority (64.8%) presented with significant left ventricular dysfunction. The most common concomitant medical conditions were hypertension (54.9%), human immuno-deficiency virus (HIV) (33.9%), anaemia (23.3%) and prior diabetes mellitus (15.5%). Moderate to severe renal dysfunction was detected in 60 (31.1%) patients. Peripartum cardiomyopathy was one of the important causes of heart failure in female patients. The most commonly used treatment included furosemide (86%), beta-blockers (72.1%), angiotensin converting enzyme inhibitors (67.4%), spironolactone (59.9%), digoxin (22.1%), angiotensin receptor blockers (5.8%), nitrates (4.7%) and hydralazine (1.7%). The median length of stay was nine days, and the in-hospital mortality rate was 10.9%. Thirty-, 90- and 180-day case fatality rates were 14.7, 25.8 and 30.8%, respectively. Mortality at 180 days was significantly associated with increasing age, lower haemoglobin level, lower glomerular filtration rate, hyponatraemia, higher N-terminal pro-brain natriuretic peptide levels, and prolonged hospital stay.Conclusions:AHF is a major public health problem in Botswana, with high in-hospital and post-discharge mortality rates and prolonged hospital stays. Late and symptomatic presentation is common, and the most common aetiologies are preventable and/or treatable co-morbidities, including hypertension, diabetes mellitus, renal failure and HIV.

ObjectivesHIV-infected patients are at increased risk for cardiovascular disease (CVD). However, general population CVD risk prediction equations that identify HIV-infected patients at elevated risk have not been widely assessed in sub-Saharan African (SSA).MethodsHIV-infected adults from 30–50 years of age with documented viral suppression were enrolled into a cross-sectional study in Gaborone, Botswana. Participants were screened for CVD risk factors. Bilateral carotid intima-media thickness (cIMT) was measured and 10-year predicted risk of cardiovascular disease was calculated using the Pooled Cohorts Equation for atherosclerotic CVD (ASCVD) and the 2008 Framingham Risk Score (FRS) (National Cholesterol Education Program III–NCEP III). ASCVD ≥7.5%, FRS ≥10%, and cIMT≥75th percentile were considered elevated risk for CVD. Agreement in classification of participants as high-risk for CVD by cIMT and FRS or ASCVD risk score was assessed using McNemar`s Test. The optimal cIMT cut off-point that matched ASCVD predicted risk of ≥7.5% was assessed using Youden’s J index.ResultsAmong 208 HIV-infected patients (female: 55%, mean age 38 years), 78 (38%) met criteria for ASCVD calculation versus 130 (62%) who did not meet the criteria. ASCVD classified more participants as having elevated CVD risk than FRS (14.1% versus 2.6%, McNemar’s exact test p = 0.01), while also classifying similar proportion of participants as having elevated CVD like cIMT (14.1% versus 19.2%, McNemar’s exact test p = 0.34). Youden’s J calculated the optimal cut point at the 81st percentile for cIMT to correspond to an ASCVD score ≥7.5% (sensitivity = 72.7% and specificity = 88.1% with area under the curve for the receiver operating characteristic [AUC] of 0.82, 95% Mann-Whitney CI: 0.66–0.99).ConclusionWhile the ASCVD risk score classified more patients at elevated CVD risk than FRS, ASCVD score classified similar proportion of patients as high risk when compared with established subclinical atherosclerosis. However, potential CVD risk category misclassification by established equations such as ASCVD may still exist among HIV-infected patients; hence there is still a need for development of a CVD risk prediction equation tailored to HIV-infected patients in SSA.

No abstract

No abstract