Prevalence and etiologies of pulmonary hypertension in Africa: a systematic review and meta-analysis

BACKGROUND: Despite the recent increasing worldwide attention towards pulmonary hypertension (PH), its epidemiology remains poorly described in Africa. Accordingly, we performed a systematic review and meta-analysis of PH prevalence, incidence and etiologies in Africa. METHODS: We searched PubMed, EMBASE, African Journals Online, and Africa Index Medicus. Published observational studies until September 20, 2017, including adult participants residing in Africa were considered. Two review authors independently selected studies, assessed included studies for methodological quality, and extracted data. A random-effects model was used for meta-analysis. Heterogeneity was evaluated by the χ (2) test on Cochrane’s Q statistic which is quantified by I(2) values. Using Newcastle-Ottawa Scale, we considered a score of 0–4, 5–7, and 8–10 as indicative of high, moderate, and low risk of bias in included studies, respectively. RESULTS: Of 1611 entries, 25 studies were retained. Twelve (48%), seven (28%), and six (24%) papers had respectively a low, moderate and high risk of bias. The prevalence of PH widely varied across different populations: 9.8% (95% confidence interval: 3.2–19.3; I(2) = 99.4%; 6 studies) in 11,163 people presenting with cardiac complaints; 10.6% (4.3–19.1; I(2) = 90.3%; 4 studies) in 937 HIV-infected people; 32.9% (17.6–50.4; I(2) = 97.2%; 3 studies) in 2077 patients with heart failure; 23.2% (15.2–32.2; I(2) = 59.4%; 3 studies) in 248 patients on hemodialysis; 12.9% (11.8–14.0; I(2) = 79.7%; 2 studies) in 3750 patients with rheumatic heart disease; 36.9% (29.7–44.3; I(2) = 79.7; 2 studies) in 79 patients with sickle cell disease; 62.7% (49.0–74.7; 1 study) in 51 patients with chronic obstructive pulmonary disease; 25.4% (16.3–37.3; 1 study) in 63 patients with systemic lupus erythematous; 68.7% (62.8–74.1; 1 study) in 259 patients with cardiac surgery; and 7.4% (4.6–11.9; 1 study) in 202 patients with systemic sclerosis. No study reported PH incidence. From one international study (n = 209), PH etiologies were: left heart disease (68.9%), pulmonary arterial hypertension (15.8%), lung disease and/or hypoxia (12.0%), chronic thromboembolic PH (1.9%) and unclear/multifactorial PH (15.8%). CONCLUSION: The prevalence of PH is relatively high in some populations in Africa, perhaps mainly driven by left heart diseases, highlighting the need for context-specific interventions. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1186/s12890-017-0549-5) contains supplementary material, which is available to authorized users.