Biological age is better than chronological as predictor of 3-month outcome in ischemic stroke

  • Carolina Soriano-Tárraga
  • , Marina Mola-Caminal
  • , Eva Giralt-Steinhauer
  • , Angel Ois
  • , Ana Rodríguez-Campello
  • , Elisa Cuadrado-Godia
  • , Alejandra Gómez-González
  • , Rosa M. Vivanco-Hidalgo
  • , Israel Fernández-Cadenas
  • , Natalia Cullell
  • , Jaume Roquer
  • , Jordi Jiménez-Conde

Research output: Contribution to journalArticlepeer-review

65 Scopus citations

Abstract

Objective: To analyze the effect of age-related DNA methylation changes in multiple cytosine-phosphate-guanine (CpG) sites (biological age [b-age]) on patient outcomes at 3 months after an ischemic stroke. Methods: We included 511 patients with first-ever acute ischemic stroke assessed at Hospital del Mar (Barcelona, Spain) as the discovery cohort. Demographic and clinical data, including chronological age (c-age), vascular risk factors, initial stroke severity, recanalization treatment, and previous and 3-month modified Rankin Scale (p-mRS and 3-mRS, respectively) were registered. B-age was estimated with an algorithm, based on DNA methylation in 71 CpGs. Bivariate analysis determined variables associated with 3-mRS for inclusion in ordinal multivariate analysis. Results: After ordinal regressions for 3-month ischemic stroke outcome (3-mRS), b-age was associated with outcome (odds ratio 1.04 [95% confidence interval 1.01-1.07]), nullifying c-age. Stepwise regression kept b-age, basal NIH Stroke Scale, sex, p-mRS, and recanalization treatment as better explanatory variables, instead of c-age. These results were successfully replicated in an independent cohort. Conclusions: B-age, estimated by DNA methylation, is an independent predictor of ischemic stroke outcome regardless of chronological years.

Original languageEnglish
Pages (from-to)830-836
Number of pages7
JournalNeurology
Volume89
Issue number8
DOIs
StatePublished - Aug 22 2017

Fingerprint

Dive into the research topics of 'Biological age is better than chronological as predictor of 3-month outcome in ischemic stroke'. Together they form a unique fingerprint.

Cite this