Using Machine-Learning for Prediction of the Response to Cardiac Resynchronization Therapy: the SMART-AV Study

Background There is a controversy whether the response of both sexes to cardiac resynchronization therapy (CRT) is similar. Optimal CRT delivery requires procedure planning. Objective To apply machine learning (ML) to develop a prediction model for CRT response. Methods Participants from the SmartDelay Determined AV Optimization (SMART-AV) trial (n=741; age, 66 ±11 yrs; 33% female; 100% NYHA III-IV; 100% EF≤35%) were randomly split into training & testing (80%; n=593), and validation (20%; n=148) samples. The entropy balancing procedure was used to match for the means of 30 covariates in male and female groups. Baseline clinical, ECG, echocardiographic and biomarker characteristics, and left ventricular (LV) lead position (43 variables) were included in 6 ML models (random forests, convolutional neural network, lasso, adaptive lasso, plugin lasso, elastic net, ridge, and logistic regression). A composite of freedom from death and heart failure hospitalization and a >15% reduction in LV end-systolic volume index at 6-months post-CRT was the endpoint. Results The primary endpoint was met by 337 patients (45.5%). Weighting resulted in a perfect balance of means of covariates in men and women. After reweighting, CRT response for women versus men was similar (OR 1.53; 95%CI 0.88-2.65; P=0.131). The adaptive lasso model was more accurate than class I ACC/AHA guidelines criteria (AUC 0.759; 95%CI 0.678-0.840 versus 0.639; 95%CI 0.554-0.722; P <0.0001), well-calibrated, and parsimonious (19 predictors; nearly half are potentially modifiable). Conclusions After balancing for covariates, both sexes similarly benefit from CRT. ML predicts short-term CRT response and thus may help with CRT procedure planning.