Introduction: Consensus on the best method for correction of the QT interval in atrial fibrillation (AF) does not exist. Accurate measurements of QT intervals during AF are needed to guide antiarrhythmic therapies. Previous studies investigating QT interval in AF are limited by small sample size. We used a large single center EKG database to compare four common QT correction formulas in AF.
Methods: Patients with conversion from AF to sinus rhythm (SR) at Stanford Health Care were identified using a database containing 103,618 EKGs. EKGs were identified as AF or SR based on reported MD interpretations. Patients were included in the analysis if they had an EKG showing SR within 24 hours after an EKG in AF. Heart rate and median uncorrected QT interval and RR interval were determined using an automated algorithm for each EKG that measures and averages intervals overall 12 leads and is verified by clinical over-read. QT interval was then corrected using the Bazett’s, Fridericia, Framingham, and Hodges formulas. The difference in QTc between patients’ last EKG in AF and their first EKG in SR were compared. Mixed-effect linear models were used to determine the effect of AF on QT interval independent of heart rate.
Results: 1,430 EKGs from 715 patients were included in the cohort (mean age 69.1±13.1 years, 59.7% men). QTc was overestimated in AF compared to SR when using Bazett’s formula (ΔQTc = 14.28 ms, 95 % CI 11.25 to 17.31, p <0.001) and Hodges formula (ΔQTc = 7.52 ms, 95% CI 4.83 to 10.21, p <0.001). Conversely, QTc was underestimated in AF when using Frederica formula (ΔQTc = -12.87 ms, 95% CI -15.52 to -10.23, p <0.001) and Framingham formula (ΔQTc = -19.65 ms, 95% CI -22.15 to -17.14, p <0.001). AF did not have an independent effect on uncorrected QT when adjusting for heart rate by linear regression (β = -0.65 ms, 95 CI -4.18 to 2.87, p = 0.717).
Conclusions: Significant differences in QTc existed between AF and SR in all tested correction formulas, however the use of Hodges formula resulted in the smallest differences in QTc. AF did not have an effect on QT independent of heart rate.