Send to

Choose Destination
Ann Thorac Surg. 2017 Aug;104(2):523-529. doi: 10.1016/j.athoracsur.2016.11.065. Epub 2017 Feb 24.

Long-Term Risk of Ischemic Stroke After the Cox-Maze III Procedure for Atrial Fibrillation.

Author information

Department of Cardiothoracic Surgery, University Hospital, Uppsala, Sweden. Electronic address:
Department of Cardiothoracic Surgery and Anesthesiology, Karolinska University Hospital, Stockholm, Sweden; Department of Molecular Medicine and Surgery, Karolinska Institutet, Stockholm, Sweden.
Department of Cardiology, Karolinska University Hospital, Stockholm, Sweden.
Department of Internal Medicine, Sahlgrenska University Hospital/Östra, Gothenburg, Sweden.
Department of Cardiovascular Surgery, Sahlgrenska University Hospital/Sahlgrenska, Gothenburg, Sweden.
Department of Cardiothoracic Surgery, University Hospital, Uppsala, Sweden.



The long-term risk of stroke after surgical treatment of atrial fibrillation is not well known. We performed an observational cohort study with long follow-up after the "cut-and-sew" Cox-maze III procedure (CM-III), including left atrial appendage excision. The aim was to analyze the incidence of stroke/transient ischemic attack (TIA) and the association to preoperative CHA2DS2-VASc (age in years, sex, congestive heart failure history, hypertension history, stroke/TIA, thromboembolism history, vascular disease history, diabetes mellitus) score.


Preoperative and perioperative data were collected in 526 CM-III patients operated in four centers 1994 to 2009, 412 men, mean age of 57.1 ± 8.3 years. The incidence of any stroke/TIA was identified through analyses of the Swedish National Patient and Cause-of-Death Registers and from review of individual patient records. The cumulative incidence of stroke/TIA and association with CHA2DS2-VASc score was estimated using methods accounting for the competing risk of death.


Mean follow-up was 10.1 years. There were 29 patients with any stroke/TIA, including 6 with intracerebral bleedings (2 fatal) and 4 with perioperative strokes (0.76%). The remaining 13 ischemic strokes and six TIAs occurred at a mean of 7.1 ± 4.0 years postoperatively, with an incidence of 0.36% per year (19 events per 5,231 patient-years). In all CHA2DS2-VASc groups, observed ischemic stroke/TIA rate was lower than predicted. A higher risk of ischemic stroke/TIA was seen in patients with CHA2DS2-VASc score 2 or greater compared with score 0 or 1 (hazards ratio 2.15, 95% confidence interval: 0.87 to 5.32) but no difference by sex or stand-alone versus concomitant operation. No patient had ischemic stroke as cause of death.


This multicenter study showed a low incidence of perioperative and long-term postoperative ischemic stroke/TIA after CM-III. Although general risk of ischemic stroke/TIA was reduced, patients with CHA2DS2-VASc score 2 or greater had a higher risk compared with score 0 or 1. Complete left atrial appendage excision may be an important reason for the low ischemic stroke rate.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center