BACKGROUND: The impact of adjunctive anti-inflammatory treatment on outcomes for patients with Kawasaki disease (KD) and coronary artery aneurysms (CAAs) is unknown. METHODS: Using data from the International KD Registry in patients with ≥ medium CAA we evaluate associations of treatment with outcomes and major adverse cardiac events (MACE). RESULTS: Medium or large CAA was present in 527 (32%) patients. All were treated with intravenous immunoglobulin (IVIG), 70% were male, and the median age was 1.3 years (interquartile range: 0.4-4.0 years). The most common acute therapies included single IVIG alone in 243 (46%), multiple IVIG in 100 (19%), multiple IVIG + corticosteroids in 75 (14%), and multiple IVIG + infliximab + corticosteroids in 44 (8%) patients. Patients who received therapy beyond single IVIG had a larger CA z-score at baseline (P < 0.001) and a higher rate of bilateral CAA (P < 0.001). Compared with IVIG alone, early adjunctive treatments (within 3 days of initial IVIG) were not associated with time to CAA regression or MACE, whereas later adjunctive therapy was associated with MACE and longer time to CAA regression. Patients receiving IVIG plus steroids vs IVIG alone had a trend towards shorter time to CAA regression and lower risk of MACE (P = 0.07). A larger CAA z-score at baseline was the strongest predictor of an increase in the CAA z-score over follow-up, lower likelihood of CAA regression, and higher risk of MACE. CONCLUSIONS: Persistence of CAA and MACE are more strongly associated with baseline severity CAA than with acute adjuvant anti-inflammatory therapy. Patients who received late adjunctive therapy are at higher risk for worse outcomes.

Background The success of cardiac auscultation varies widely among medical professionals, which can lead to missed treatments for structural heart disease. Applying machine learning to cardiac auscultation could address this problem, but despite recent interest, few algorithms have been brought to clinical practice. We evaluated a novel suite of Food and Drug Administration-cleared algorithms trained via deep learning on >15 000 heart sound recordings. Methods and Results We validated the algorithms on a data set of 2375 recordings from 615 unique subjects. This data set was collected in real clinical environments using commercially available digital stethoscopes, annotated by board-certified cardiologists, and paired with echocardiograms as the gold standard. To model the algorithm in clinical practice, we compared its performance against 10 clinicians on a subset of the validation database. Our algorithm reliably detected structural murmurs with a sensitivity of 85.6% and specificity of 84.4%. When limiting the analysis to clearly audible murmurs in adults, performance improved to a sensitivity of 97.9% and specificity of 90.6%. The algorithm also reported timing within the cardiac cycle, differentiating between systolic and diastolic murmurs. Despite optimizing acoustics for the clinicians, the algorithm substantially outperformed the clinicians (average clinician accuracy, 77.9%; algorithm accuracy, 84.7%.) Conclusions The algorithms accurately identified murmurs associated with structural heart disease. Our results illustrate a marked contrast between the consistency of the algorithm and the substantial interobserver variability of clinicians. Our results suggest that adopting machine learning algorithms into clinical practice could improve the detection of structural heart disease to facilitate patient care.

Background Coronary artery aneurysms (CAAs) may occur after Kawasaki disease (KD) and lead to important morbidity and mortality. As CAA in patients with KD are rare and heterogeneous lesions, prognostication and risk stratification are difficult. We sought to derive the cumulative risk and associated factors for cardiovascular complications in patients with CAAs after KD. Methods and Results A 34-institution international registry of 1651 patients with KD who had CAAs (maximum CAA Z score ≥2.5) was used. Time-to-event analyses were performed using the Kaplan-Meier method and Cox proportional hazard models for risk factor analysis. In patients with CAA Z scores ≥10, the cumulative incidence of luminal narrowing (>50% of lumen diameter), coronary artery thrombosis, and composite major adverse cardiovascular complications at 10 years was 20±3%, 18±2%, and 14±2%, respectively. No complications were observed in patients with a CAA Z score <10. Higher CAA Z score and a greater number of coronary artery branches affected were associated with increased risk of all types of complications. At 10 years, normalization of luminal diameter was noted in 99±4% of patients with small (2.5≤Z<5.0), 92±1% with medium (5.0≤Z<10), and 57±3% with large CAAs (Z≥10). CAAs in the left anterior descending and circumflex coronary artery branches were more likely to normalize. Risk factor analysis of coronary artery branch level outcomes was performed with a total of 893 affected branches with Z score ≥10 in 440 patients. In multivariable regression models, hazards of luminal narrowing and thrombosis were higher for patients with CAAs of the right coronary artery and left anterior descending branches, those with CAAs that had complex architecture (other than isolated aneurysms), and those with CAAs with Z scores ≥20. Conclusions For patients with CAA after KD, medium-term risk of complications is confined to those with maximum CAA Z scores ≥10. Further risk stratification and close follow-up, including advanced imaging, in patients with large CAAs is warranted.