UCSF

Background

Vest Prevention of Early Sudden Death Trial did not demonstrate a significant reduction in arrhythmic death with the wearable cardioverter-defibrillator (WCD), but compliance with the device may have substantially affected the results. ThePletcher influence of WCD compliance on outcomes has not yet been fully evaluated.

Methods

Using linear and pooled logistic models, we performed as-treated analyses omitting person-time in the hospital and adjusted for correlates of WCD compliance. To assess the impact of early stopping of WCD, we performed a per-protocol Kaplan-Meier analysis, censoring after the last day the WCD was worn. Interactions of potential effect modifiers with treatment assignment and WCD compliance on outcomes were investigated. Finally, we used linear models to identify predictors of WCD compliance.

Results

A per-protocol analysis demonstrated a significant reduction in total (P < .001) and arrhythmic (P = .001) mortality. Better WCD compliance was independently predicted by cardiac arrest during index myocardial infarction (MI), higher Cr, diabetes, prior heart failure, EF ≤ 25%, Polish enrolling center and number of WCD alarms, while worse compliance was predicted by being divorced, Asian race, higher body mass index, prior percutaneous coronary intervention, or any WCD shock. Neither excluding time in hospital from the as-treated analysis nor adjustment for factors affecting WCD compliance materially changed the results. No variable demonstrated a significant interaction in either the intention-to-treat or as-treated analysis.

Conclusion

Robust sensitivity analyses of as-treated and per-protocol analyses suggest that the WCD is protective in compliant patients with ejection fraction less than or equal to 35% during the first 3 months post-MI.