Risk of stroke after surgery in patients with and without chronic atrial fibrillation

S. Kaatz, J. D. Douketis, H. Zhou, B. F. Gage, Richard H White

Research output: Contribution to journalArticle

61 Scopus citations

Abstract

Background: The extent to which chronic atrial fibrillation affects the risk of postoperative stroke is largely unknown. Objectives: We sought to determine the 30-day rate of stroke among patients with and without chronic AF who underwent 10 different types of surgery. Patients/methods: The crude incidence of stroke was retrospectively determined using a population-based linked administrative database of hospitalized patients who underwent specified operations between 1 January 1996 and 30 November 2005. The risk of stroke in patients with AF was adjusted for age, race, sex, presence of diabetes, heart failure, hypertension and prior stroke. Results: The overall 30-day rate of stroke in 69 202 patients with chronic AF was 1.8% (95% CI, 1.7-1.9%) vs. 0.6% (CI, 0.58-0.62%) in 2 470 649 patients without AF. The risk-adjusted odds of a postoperative stroke in patients with chronic AF were 2.1 (CI, 2.0-2.3). The highest incremental difference in the crude rate of stroke was observed in patients undergoing neurologic or vascular surgery, with a difference of approximately 2%. Conclusion: Patients with chronic AF had twice the risk of postoperative stroke. Randomized trials are needed to determine if aggressive perioperative anticoagulation can reduce the incidence of postoperative stroke in patients with AF.

Original languageEnglish (US)
Pages (from-to)884-890
Number of pages7
JournalJournal of Thrombosis and Haemostasis
Volume8
Issue number5
DOIs
StatePublished - 2010

Keywords

  • Atrial fibrillation
  • Postoperative
  • Risk factors
  • Stroke

ASJC Scopus subject areas

  • Hematology
  • Medicine(all)

Fingerprint Dive into the research topics of 'Risk of stroke after surgery in patients with and without chronic atrial fibrillation'. Together they form a unique fingerprint.

  • Cite this