Background and Hypothesis: The sensitivity of LVH analysis by ECG voltage criteria in patients with severe aortic valve stenosis undergoing trans-catheter aortic valve replacement (TAVR) has not yet been studied. LVH is expected in the TAVR population and would be reflected in voltage criteria by ECG.
Methods: A retrospective chart review was conducted in 176 consecutive TAVR patients without ventricular-paced rhythm. ECG data was collected and analyzed by Sokolow-Lyon and Cornell Voltage criteria. Results were compared to transthoracic echocardiogram. Analyses of variation, correlation, chi-square, and logistic regression were used. The study was approved by the institutional IRB.
Results: Sokolow-Lyon and Cornell Voltage criteria for LVH were present and concordant in 19% (33 of 176) of patients; in 49% (86 of 176) of patients, neither criteria was suggestive for LVH. Only 19% (34 of 176) of patients had LVH by Cornell Voltage and 13% (23 of 176) by Sokolow-Lyon criteria, indicative of poor concordance between these two commonly used ECG criteria for LVH (p<0.0001).
Ejection fraction, aortic valve gradient, aortic valve area, COPD, PVD, prior stroke, dyslipidemia, and hypertension did not affect the prevalence of LVH by either or both criteria.
Women (p<0.01) and patients with rhythm other than atrial fibrillation (p<0.0053) were more likely to have voltage criteria for LVH, while older adults were more likely to meet criteria for LVH. Concordant LVH criteria were noted in patients 84.6 +/- 7.2 years of age, while patients without LVH by ECG voltage criteria were significantly younger at 80.21 +/- 8.1 years of age (p<0.007).
Conclusion: The presence of LVH by Sokolow-Lyon and Cornell ECG voltage criteria poorly correlates with the presence of LVH and critical aortic stenosis in TAVR patients. Women are more likely to have voltage criteria for LVH. Therefore, ECG may not be a suitable method of screening patients with severe aortic stenosis for LVH, especially in men.