Meta-Analysis of Circadian Variation in the Onset of Acute Aortic Dissection

Abstract

Circadian variation in the onset of acute aortic dissection (AAD) has been less investigated than other cardiovascular diseases. We performed a meta-analysis to assess the presence of an circadian rhythmic variability of AAD onset. Eligible studies were observational studies enrolling patients with AAD and reporting a circadian variation in AAD. Study-specific estimates, that is, 2-hour incidence of AAD, were combined using the random-effects model. Chronobiological analysis (analysis of circadian rhythmicity) was performed by applying a partial Fourier series to the pooled 2-hour incidence using the weighted least-squares method. We identified 7 eligible studies enrolling a total of 1,827 patients with AAD. Pooled 2-hour period incidence of AAD was 3.4% in 0:00 to 2:00, 4.8% in 2:00 to 4:00, 5.4% in 4:00 to 6:00, 9.6% in 6:00 to 8:00, 13.8% in 8:00 to 10:00, 11.1% in 10:00 to 12:00, 8.1% in 12:00 to 14:00, 8.9% in 14:00 to 16:00, 8.8% in 16:00 to 18:00, 7.0% in 18:00 to 20:00, 8.1% in 20:00 to 22:00), and 5.5% in 22:00 to 24:00. Chronobiological analysis (nonlinear Fourier rhythm analysis) identified a significant (p = 0.0082) circadian pattern in the occurrence of AAD with a peak in 8:00 to 10:00 and a nadir in 0:00 to 2:00. Pooled analysis demonstrated significantly more incidence in 8:00 to 10:00 than in 0:00 to 2:00 (risk ratio 3.59, 95% confidence interval 2.19 to 5.90, p<0.00001). The incidence of AAD was 8.8%, 15.5%, 25.0%, 17.7%, 16.1%, and 13.8% in 0:00 to 4:00, 4:00 to 8:00, 8:00 to 12:00, 12:00 to 16:00, 16:00 to 20:00, and 20:00 to 24:00, respectively. A significant circadian pattern was found in the occurrence of AAD with a peak in 8:00 to 10:00 and a nadir in 0:00 to 2:00.