Non-invasively quantified changes in left ventricular activation predict outcomes in patients undergoing cardiac resynchronization therapy.

BACKGROUND: Changes in left ventricular (LV) activation after cardiac resynchronization therapy (CRT) influence survival but are difficult to quantify noninvasively. METHODS AND RESULTS: We studied 527 CRT patients to assess whether noninvasive quantification of changes in LV activation, defined by change (Δ) in QRS area (QRSA), can predict outcomes after CRT. The study outcome was time until LV assist device(LVAD), cardiac transplant, or death. The three-dimensional QRSA was measured from clinical 12 lead ECGs which were transformed into vectorcardiograms using the Kors method. QRSA was calculated as (QRSx2 + QRSy2 + QRSz2 )1/2 ; ΔQRSA was calculated as post-QRSA minus pre-QRSA, where a negative value represents a reduction in LV activation delay. Kaplan-Meier plots and multivariable Cox proportional hazards models were used to relate ΔQRSA area with outcomes after stratifying the population into quartiles of ΔQRSA. The median baseline QRSA of 93.6 µVs decreased to 59.7 µVs after CRT. Progressive reductions in QRSA with CRT were associated with a lower rate of LVAD, transplant, or death across patient quartiles (P < .001). In Cox regression analyses, ΔQRSA was associated with outcomes independent of QRS morphology and other clinical variables (Q1[greatest decrease] vs Q4[smallest change=reference], HR 0.45, CI, 0.30-0.70, P < .001). There was no interaction between ΔQRSA and QRS morphology. CONCLUSIONS: CRT induced ΔQRSA was associated with clinically meaningful changes in event-free survival. ΔQRSA may be a novel target to guide lead implantation and device optimization.