Echocardiographic screening to determine progression of latent rheumatic heart disease in endemic areas: A systematic review and meta-analysis

Background The World Health Organisation previously recommended routine screening in school-aged children in countries with a high prevalence of rheumatic heart disease (RHD); however, it is unclear if screening-detected (latent) valve disease will inevitably evolve to a pathological lesion. Understanding the natural history of latent RHD is essential prior to recommendation of screening in endemic areas. Studies documenting the progression of latent RHD have had contrasting conclusions about the pathogenicity of latent valvular lesions. This review provides estimates of rates of progression of latent RHD. Methods and findings In this systematic review and meta-analysis, we searched EMBASE, MEDLINE, Global Index Medicus, Africa Wide, Cochrane Database of Systematic Reviews and Global Health Database for studies published before April 30, 2019. Study data were extracted from all studies which reported follow-up data on progression of latent valve lesions. Studies with control cohorts were used to calculate comparative prevalence ratios. This study is registered with PROSPERO, number CRD42019119427. We identified 12 studies reporting follow-up data on latent RHD for 950 people in 9 countries. The estimated pooled prevalence rate for progression per year of latent RHD was 5%/year (95% CI 2-8). Eight studies reported on the progression of borderline latent RHD with an estimated pooled prevalence of 2%/year (95% CI 0-4). Three studies included control groups. There was a significant increase in the risk of progression of valvular disease in the latent group compared with controls (RR = 3.57 (95%CI = 1.65-7.70, P = 0.001). The overall risk of bias was low. Given most studies included penicillin administration we were unable to document the natural history of latent RHD. Furthermore, we were unable to perform a sensitivity analysis to determine the effect of administering penicillin prophylaxis on progression of valve disease given prescription of penicillin was not standardised. Conclusion Latent RHD has a slow rate of progression but it is significantly higher compared to controls, with definite latent RHD having a higher rate of progression compared with borderline latent disease. There are a massive number of individuals at risk for RHD in the developing world as well as logistical challenges of screening and delivering penicillin prophylaxis. The low rate of progression from untargeted screening may be an important consideration in resource-constrained environments.