Search

Background Identification of children with latent rheumatic heart disease (RHD) by echocardiography, before onset of symptoms, provides an opportunity to initiate secondary prophylaxis and prevent disease progression. There have been limited artificial intelligence studies published assessing the potential of machine learning to detect and analyze mitral regurgitation or to detect the presence of RHD on standard portable echocardiograms. Methods and Results We used 511 echocardiograms in children, focusing on color Doppler images of the mitral valve. Echocardiograms were independently reviewed by an expert adjudication panel. Among 511 cases, 229 were normal, and 282 had RHD. Our automated method included harmonization of echocardiograms to localize the left atrium during systole using convolutional neural networks and RHD detection using mitral regurgitation jet analysis and deep learning models with an attention mechanism. We identified the correct view with an average accuracy of 0.99 and the correct systolic frame with an average accuracy of 0.94 (apical) and 0.93 (parasternal long axis). It localized the left atrium with an average Dice coefficient of 0.88 (apical) and 0.9 (parasternal long axis). Maximum mitral regurgitation jet measurements were similar to expert manual measurements (P value=0.83) and a 9‐feature mitral regurgitation analysis showed an area under the receiver operating characteristics curve of 0.93, precision of 0.83, recall of 0.92, and F1 score of 0.87. Our deep learning model showed an area under the receiver operating characteristics curve of 0.84, precision of 0.78, recall of 0.98, and F1 score of 0.87. Conclusions Artificial intelligence has the potential to detect RHD as accurately as expert cardiologists and to improve with more data. These innovative approaches hold promise to scale echocardiography screening for RHD.

Secondary antibiotic prophylaxis with regular intramuscular benzathine penicillin G (BPG) is the cornerstone of rheumatic heart disease management. However, there is a growing body of evidence that patients with rheumatic heart disease who have severe valvular heart disease with or without reduced ventricular function may be dying from cardiovascular compromise following BPG injections. This advisory responds to these concerns and is intended to: (1) raise awareness, (2) provide risk stratification, and (3) provide strategies for risk reduction. Based on available evidence and expert opinion, we have divided patients into low‐ and elevated‐risk groups, based on symptoms and the severity of underlying heart disease. Patients with elevated risk include those with severe mitral stenosis, aortic stenosis, and aortic insuffiency; those with decreased left ventricular systolic dysfunction; and those with New York Heart Association class III and IV symptoms. For these patients, we believe the risk of adverse reaction to BPG, specifically cardiovascular compromise, may outweigh its theoretical benefit. For patients with elevated risk, we newly advise that oral prophylaxis should be strongly considered. In addition, we advocate for a multifaceted strategy for vasovagal risk reduction in all patients with rheumatic heart disease receiving BPG. As current guidelines recommend, all low‐risk patients without a history of penicillin allergy or anaphylaxis should continue to be prescribed BPG for secondary antibiotic prophylaxis. We publish this advisory in the hopes of saving lives and avoiding events that can have devastating effects on patient and clinician confidence in BPG.

Background Despite the high burden of rheumatic heart disease in sub‐Saharan Africa, diagnosis with acute rheumatic fever (ARF) is exceedingly rare. Here, we report the results of the first prospective epidemiologic survey to diagnose and characterize ARF at the community level in Africa. Methods and Results A cross‐sectional study was conducted in Lira, Uganda, to inform the design of a broader epidemiologic survey. Key messages were distributed in the community, and children aged 3 to 17 years were included if they had either (1) fever and joint pain, (2) suspicion of carditis, or (3) suspicion of chorea, with ARF diagnoses made by the 2015 Jones Criteria. Over 6 months, 201 children met criteria for participation, with a median age of 11 years (interquartile range, 6.5) and 103 (51%) female. At final diagnosis, 51 children (25%) had definite ARF, 11 (6%) had possible ARF, 2 (1%) had rheumatic heart disease without evidence of ARF, 78 (39%) had a known alternative diagnosis (10 influenza, 62 malaria, 2 sickle cell crises, 2 typhoid fever, 2 congenital heart disease), and 59 (30%) had an unknown alternative diagnosis. Conclusions ARF persists within rheumatic heart disease–endemic communities in Africa, despite the low rates reported in the literature. Early data collection has enabled refinement of our study design to best capture the incidence of ARF and to answer important questions on community sensitization, healthcare worker and teacher education, and simplified diagnostics for low‐resource areas. This study also generated data to support further exploration of the relationship between malaria and ARF diagnosis in rheumatic heart disease/malaria‐endemic countries.

Background Rheumatic heart disease disproportionately affects women of reproductive age, as it increases the risk of cardiovascular complications and death during pregnancy and childbirth. In sub‐Saharan Africa, clinical outcomes and adherence to guideline‐based therapies are not well characterized for this population. Methods and Results In a retrospective cohort study of the Uganda rheumatic heart disease registry between June 2009 and May 2018, we used multivariable regression and Cox proportional hazards models to compare comorbidities, mortality, anticoagulation use, and treatment cascade metrics among women versus men aged 15 to 44 with clinical rheumatic heart disease. We included 575 women and 252 men with a median age of 27 years. Twenty percent had New York Heart Association Class III‐IV heart failure. Among patients who had an indication for anticoagulation, women were less likely than men to receive a prescription of warfarin (66% versus 81%; adjusted odds ratio, 0.37; 95% CI, 0.14–0.96). Retention in care (defined as a clinic visit within the preceding year) was poor among both sexes in this age group (27% for men, 24% for women), but penicillin adherence rates were high among those retained (89% for men, 92% for women). Mortality was higher in men than women (26% versus 19% over a median follow‐up of 2.7 years; adjusted hazard ratio, 1.66; 95% CI, 1.18–2.33). Conclusions Compared with men, women of reproductive age with rheumatic heart disease in Uganda have lower rates of appropriate anticoagulant prescription but also lower mortality rates. Retention in care is poor among both men and women in this age range, representing a key target for improvement.