Abstrakt
Aims: The improved clinical response in patients with left bundle branch block (LBBB) over LBBB-free patients treated with cardiac resynchronization therapy with a defibrillator (CRT-D) is commonly attributed to an LBBB abnormality. We aimed to find an alternative explanation.
Methods and results: We analysed an immediate effect of selecting the LBBB group of patients in a cohort of 63 non-ischaemic cardiomyopathy (non-ICM) and 83 ischaemic cardiomyopathy (ICM) patients treated with CRT-D; 75% of non-ICM and 51% of ICM patients had an LBBB abnormality on the electrocardiogram, with a significant difference (P = 0.0032 by chi(2)). As a result of this difference, the proportion of non-ICM patients increased from 43% in the primary cohort to 53% in LBBB selection and decreased to 28% in non-LBBB group.
By nonparametric survival analysis, the hazard ratio in non-ICM patients in the LBBB selection decreased from 0.48 (P = 0.0488) to 0.36 (P = 0.0251) and increased in the non-LBBB group to 0.75 (P = 0.6496). Any comparison of LBBB and non-LBBB groups must compare sets with a significantly altered proportion of patients of different aetiologies.
Most publications on LBBB patients are erroneous because they compare LBBB with non-LBBB groups, not taking into account that the groups have been substantially changed by the selection process. Conclusions: The declared outcome of the LBBB groups reflects inevitably the survival outcome of their non-ICM patients and not the intended outcome of patients with LBBB.
CRT-D in patients with different aetiologies of cardiomyopathy calls for separate evaluation.