Recurrent Cervical Artery Dissection Prevalence and Predictors: A Secondary Analysis of the STOP-CAD Study
Marto JP., Shu L., Wu E., Mendel DM., Heldner MR., Kaufmann JE., Leker RR., Zedde M., Nehme A., Muppa J., Aguiar de Sousa D., Sousa JA., Fonseca AC., Nzwalo H., Romoli M., Salehi Omran S., Keser Z., Bavarsad Shahripour R., Kam W., Zubair A., Saleh Velez FG., Elangovan C., Krishnaiah B., Metanis I., Pascarella R., Siegler JE., Rothstein A., Khazaal O., Antonenko K., Simpkins AN., Lester E., Crandall R., Rosa S., Soares MD., Bakradze E., Willey JZ., Ghannam M., Samaniego EA., Elnazeir M., Klein P., Traenka C., Henninger N., Nguyen TN., Engelter ST., Yaghi S.
BACKGROUND: Patients with cervical artery dissection (CeAD) may experience a recurrent dissection, but its frequency, risk factors, and clinical implications are not well defined. We aimed to determine the risk, associated factors, and clinical impact of recurrent CeAD. METHODS: The STOP-CAD study was a multicenter international retrospective observational study of patients with CeAD treated between January 2015 and June 2022. Recurrent dissection was defined as a CeAD occurring at least 7 days after the diagnosis of the index event that affects a different artery or a different segment of the same artery. Patients were followed from day 7 up to 2 years. The absolute risk of recurrent CeAD over time was calculated using Kaplan-Meier survival estimates. Multivariable logistic and Cox regression models were used to assess predictors of CeAD recurrence. RESULTS: Of the 4023 patients included in STOP-CAD, 3836 (median age 46 years, 45% females) were eligible for this analysis. During a median (interquartile range) follow-up of 295 (97–720) days, 88 (2.29%) patients had a CeAD recurrence. Median time-to-recurrent CeAD was 53 (interquartile range, 18–157) days. The estimated risk of recurrent CeAD at 2 years was 3.22% (95% CI, 2.59%–4.00%). In multivariable analyses, younger age (adjusted odds ratios, 0.98 [95% CI, 0.96–0.99]), migraine (adjusted odds ratio, 1.88 [95% CI, 1.14–3.07]), and fibromuscular dysplasia (adjusted odds ratio, 2.90 [95% CI, 1.66–5.06]) were associated with CeAD recurrence, while presenting with an ischemic stroke was associated with a lower likelihood of recurrence (adjusted odds ratio, 0.47 [95% CI, 0.29–0.75]). These associations with CeAD recurrence over time were confirmed by Cox regression analyses. Among the 88 patients with recurrent CeAD, only 5 had accompanying ischemic events (3 strokes, 2 transient ischemic attacks). CONCLUSIONS: In this retrospective study, recurrent CeAD was uncommon, approximately half of the events were diagnosed within the first 2 months of the index event, and recurrent events rarely caused new ischemic events. Younger age, migraine, absence of ischemic stroke at presentation, and signs of fibromuscular dysplasia may help identify high-risk patients.