Using Machine-Learning for Prediction of the Response to Cardiac
Resynchronization Therapy: the SMART-AV Study
Introduction—We aimed to apply machine learning (ML) to develop a
prediction model for cardiac resynchronization therapy (CRT) response.
Methods and Results—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. 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. The primary endpoint was met by 337 patients
(45.5%). 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).
The model predicted CRT response with 70% accuracy, 70% sensitivity,
and 70% specificity, and should be further validated in prospective
studies. Conclusions—ML predicts short-term CRT response and thus may
help with CRT procedure planning.