BACKGROUND: Up to one-third of patients diagnosed with left bundle branch block (LBBB) by conventional electrocardiographic (ECG) criteria are misdiagnosed. Strict LBBB shows decreased left ventricular pumping efficiency compared with nonstrict LBBB. However, no previous study has evaluated the frequency of strict LBBB after transcatheter aortic valve replacement (TAVR). The aim of this study was to determine the incidence of developing strict versus nonstrict LBBB after TAVR and test the hypothesis that preprocedure QRS duration does not predict strict LBBB but predicts development of nonstrict LBBB. METHODS: All patients receiving TAVR between 4/2011 and 2/2013 (n = 71) with no preexisting bundle branch block or permanent pacemaker were included. Twelve-lead ECGs were acquired preprocedure and both 1-day and 1-month postprocedure. All ECGs were classified as strict LBBB, nonstrict LBBB, or no LBBB. RESULTS: Sixty-eight patients had ECGs eligible for final analysis. On postprocedure day 1, 25 (37%) of 68 patients developed strict LBBB, and 2 patients (3%) developed nonstrict LBBB. At 1-month follow-up, the 2 patients diagnosed with nonstrict LBBB had resolved to normal, and 5 (20%) of 25 patients with strict LBBB had resolved to normal. Preprocedure QRS duration did not predict strict LBBB (P = .51). Because of the low incidence of nonstrict LBBB, QRS duration as a predictor of nonstrict LBBB could not be tested. CONCLUSIONS: Almost all patients who developed evidence of LBBB after TAVR met the new strict criteria, indicating probable procedural injury to the left bundle branch. Preprocedural QRS duration did not predict the development of strict LBBB.