BACKGROUND: Rheumatic heart disease (RHD) is an important problem in developing countries; however, many cases are detected only when the disease has progressed to cardiac failure. Screening can detect cases earlier, but there are no screening guidelines. METHODS: We performed a cross-sectional screening study in Tonga among 5,053 primary school children, in whom auscultation followed by echocardiography of those with heart murmurs were used to identify RHD. We also analyzed whether a three-stage screening protocol of auscultation performed by a medical student to detect any heart murmur, second-stage auscultation performed by a local pediatrician to differentiate pathological from innocent murmurs and echocardiography of those with pathological murmurs altered outcomes. RESULTS: The prevalence of definite RHD was 33.2 per 1,000. The prevalence of RHD increased significantly with age, peaking at 42.6 per 1,000 in children aged 10-12 years. Most valve lesions (91 [54%] of 169) were mild. Auscultation to detect pathological murmurs was poorly sensitive (46.4%), and the finding of any murmur on auscultation did not affect the likelihood of detecting pathology on echocardiography. The finding of a pathological murmur did significantly increase the likelihood of detecting pathology on echocardiography, but still missed 54% of those with pathology (mainly RHD) detected on echocardiography. CONCLUSIONS: Screening is a useful method for detecting asymptomatic RHD in regions of high prevalence and we report a high echocardiographically confirmed prevalence. The most appropriate screening strategy remains to be confirmed, however, and implementation will depend on the availability of echocardiography and trained staff.
BACKGROUND:Rheumatic heart disease (RHD) is an important problem in developing countries; however, many cases are detected only when the disease has progressed to cardiac failure. Screening can detect cases earlier, but there are no screening guidelines. METHODS: We performed a cross-sectional screening study in Tonga among 5,053 primary school children, in whom auscultation followed by echocardiography of those with heart murmurs were used to identify RHD. We also analyzed whether a three-stage screening protocol of auscultation performed by a medical student to detect any heart murmur, second-stage auscultation performed by a local pediatrician to differentiate pathological from innocent murmurs and echocardiography of those with pathological murmurs altered outcomes. RESULTS: The prevalence of definite RHD was 33.2 per 1,000. The prevalence of RHD increased significantly with age, peaking at 42.6 per 1,000 in children aged 10-12 years. Most valve lesions (91 [54%] of 169) were mild. Auscultation to detect pathological murmurs was poorly sensitive (46.4%), and the finding of any murmur on auscultation did not affect the likelihood of detecting pathology on echocardiography. The finding of a pathological murmur did significantly increase the likelihood of detecting pathology on echocardiography, but still missed 54% of those with pathology (mainly RHD) detected on echocardiography. CONCLUSIONS: Screening is a useful method for detecting asymptomatic RHD in regions of high prevalence and we report a high echocardiographically confirmed prevalence. The most appropriate screening strategy remains to be confirmed, however, and implementation will depend on the availability of echocardiography and trained staff.