Abstract 14246: Can the Heart Rate-Dependency of Bazett QTc be Corrected?
Background: Bazett QTc equals QT when heart rate (HR) is 60 bpm, but gets progressively inflated at higher HRs, which may confound the diagnosis of “prolonged QT.” We sought to modify QTc so as to be less HR-dependent.
Methods & Results: Our database consisted of 88,989 GE 12SLTM-processed health system ECGs (from adults in sinus rhythm; QRS < 120 ms; HRs 50-100 bpm). Upper (96th/98th) QTc percentiles (%s), relevant to prolonged QT diagnosis, and median QTc were plotted at successive HRs (1 bpm increments). Linear regression (LR) was then performed for each QTc % (Figure A): Slopes were 0.91-0.99; R2 0.94-0.97; p<0.0001. The linearity of QTc %s over HRs 50-100 bpm implies that - for QTc within a given % - the linear adjustment formula, QTc: [LR Slope x (HR - 60)], can convert QTc to a near-HR-invariant value. Since LR Slopes ~1 in Figure A, we get the simplified formula (“QTc60”): QTc - (HR - 60) .
Adjustment of QTc via QTc60 was hence performed for each QTc %. Regression lines of the resultant QTc60 % plots (Figure B) were horizontal (Slopes ~0), reflecting near-HR-invariant QTc60 values (comparable to those at HR 60 bpm). Based on the LR Slopes ~1 criterion, QTc60 was applicable to the 96th/98th QTc %s for ECGs dichotomized by sex or median age (Slopes 0.91-1.11), and for ECGs from outpatients (Slopes 0.92-0.93); whereas LR Slopes deviated more from 1 (range 0.69-0.71) for ECGs from non-outpatients.
Conclusion: In a large clinical ECG database, with HRs 50-100 bpm, simple linear adjustment of QTc - via QTc60 - yielded a more HR-invariant parameter. This Bazett QTc modification, relevant to assessing prolonged QTc values, merits further evaluation.
- © 2013 by American Heart Association, Inc.