To compare accelerated physiological aging through an artificial intelligence (AI)-ECG age estimation algorithm in patients with migraine with aura (MwA) and migraine without aura (MwoA)

MwA is associated with an increased risk of cardiovascular events compared to MwoA, and endothelial dysfunction may play a role. An AI-ECG algorithm estimating age based on ECG features is a surrogate for “physiologic age.” The difference between the estimated age minus chronological age (Δ age) is a marker of accelerated physiologic aging and is associated with peripheral endothelial dysfunction and cardiovascular events. We aim to compare the Δ age between MwoA and MwA.

Adult patients diagnosed with MwA and MwoA with at least one ECGs were identified. The ECG with the highest Δ age was defined as the index ECG. The Δ age was compared between MwA and MwoA using two-sample t-test. General linear regression was used to adjust for age, sex, and vascular comorbidities. Subgroup analysis were done after stratifying for age and sex.

24918 patients (11364 with MwA and 13554 with MwoA) were included. The mean chronological age and estimated ages for MwA were 50.6 and 55.7 years, respectively, while for MwoA they were 48.2 and 53.4 years. There was no significant difference in the mean Δ age between MwA and MwoA in the overall group after multivariate adjustments. In subgroup analysis, we found patients with MwA had significantly higher Δ age than MwoA in two age groups in females: 35-55 and > 75 years, 7.2 vs 6.6, p<0.02, and -5.3 versus -5.5, p<0.02, respectively; and male patients > 75 years (-4.4 versus -5.6, p<0.003).

The results could suggest a higher degree of endothelial dysfunction in MwA compared to MwoA in elderly patients and females 35-55 years. Further studies with a non-migraine control group and markers of endothelial dysfunction are planned.