https://doi.org/10.1007/s00392-024-02526-y
1Klinikum rechts der Isar der Technischen Universität München Klinik und Poliklinik für Innere Medizin I München, Deutschland
Background:
The consumption of alcohol has been linked to various electrocardiogram (ECG) abnormalities, including the prolongation of the QT interval. According to the latest guidelines by the American Heart Association, a prolonged rate-corrected QT interval (QTc) is defined as >460 ms for women and >450 ms for men. Currently, there is no evaluation on which formula for calculating QTc interval is most appropriate for detecting QTc prolongation in alcohol-intoxicated patients.
Objective:
To compare four different heart rate correction formulas for calculating the QTc in patients with and without alcohol intoxication.
Methods:
All patients admitted to our University Hospitals' emergency department since 2021 who underwent an electrocardiogram (ECG) and an alcohol test during hospitalization were considered eligible for inclusion. Patients with atrial fibrillation (AF) and/or complete bundle branch block (BBB) were excluded from the study. Heart rate, PQ-, QRS- and QT intervals, as well as QRS-axis were measured from digital electrocardiograms. The QTc values were determined using the Bazett- (QTcBaz), Fridericia- (QTcFri), Hodges- (QTcHod) , and Framingham (QTcFra) formula. Linear regression analysis was conducted to explore the correlation between each QTc formula and the level of alcohol present in the patients' system.
Results:
A total of 758 patients were identified through our database query, all of whom underwent testing for serum alcohol levels upon admission along with an ECG. Exclusions from our analysis included 59 patients with AF, 85 patients with a complete BBB and 27 patients presenting both conditions. Ultimately, our study encompassed 587 patients. The average age was 57±9 years, with males constituting 72% of the cohort. Among them, 70 patients exhibited alcohol intoxication, displaying a mean serum alcohol level of 2.4±1.5‰ (referred to as the alcohol group), while the remaining 517 patients had alcohol levels below detection limits and served as the control group.
Comparison of ECG parameters revealed no significant differences in heart rate (alcohol: 85±17bpm; control: 84±22bpm; p=0.209), RR-interval (alcohol: 727±143ms; control: 760±190ms; p=0.202), PQ-interval (alcohol:156±23ms; control:163±37ms; p=0.295), QRS-interval (alcohol:89±12ms; control:88±12ms; p=0.550), uncorrected QT-interval (alcohol:382±35ms; control:384±48ms; p=0.776), and QRS-axis (alcohol:36±42°; control:27±44°; p=0.066). Notably, when applying heart rate correction formulas to calculate QTc intervals, a significant difference between the two groups was only observed when utilizing the Bazett formula (alcohol:451±27ms; control:438±65ms; p=0.040). There was no statistically significant difference when employing alternative formulas (QTcFri – alcohol: 425±25ms, control 416±62ms, p=0.168; QTcFra – alcohol: 382±36ms, control 378±68ms, p=0.967; QTcHod – alcohol:426±21ms, control:419±60ms, p=0.286). Through linear regression analysis, it was determined that only the QTcBaz exhibited a significant correlation with serum alcohol concentration (regression coefficient B = 6.26, 95% confidence interval 0.12 – 12.40, p=0.046).
Conclusion:
To our knowledge, this is the first study comparing heart rate correction formulas for calculating the QTc in patients with and without alcohol intoxication. Our findings indicate that the Bazett formula is the most appropriate choice for identifying QT prolongation in individuals under the influence of alcohol.