Skip navigation

Sex differences in AF


Clinical bottom line

Women receiving warfarin have a higher risk (9%) of a major bleed than men (3%).


KH Humphries et al. New onset atrial fibrillation. Sex differences in presentation, treatment, and outcome. Circulation 2001 103: 2365-2370.


This cohort study enrolled patients with newly-diagnosed atrial fibrillation (AF) began in 1990 in six Canadian cities. Patients had to have a first presentation of AF confirmed by ECG, who either presented with symptoms or who were diagnosed during a routine examination. Excluded were those with a previous history or because AF was precipitated by cardiothoracic surgery.

Baseline data were collected at the time of enrollment, with follow up visits at 3 months, one year, and then annually for three years. A major bleed was one that required transfusion, surgical intervention, or that resulted in a more than 20 g/L reduction in haemoglobin.


There were 899 subjects, with three years visits completed in 86%.


At baseline women were significantly older (65 years vs 60 years for men), more likely to have symptoms (85% vs 74%) and have higher heart rates during AF (126 bpm vs 119 bpm). Women were more likely to have thyroid disorder (16% vs 5%)be hypertensive (49% vs 34%) but be less likely to have a history of myocardial infarction (10% vs 22%).


Treatment was no different between the sexes. Acute pharmacological or electrical interventions to restore sinus rhythm occurred in 70%, though men were more likely to undergo electrical cardioversion (20% vs 13%). At baseline discharge digoxin was used in 43%, antiarrhytmic drug in 36% warfarin in 32%, aspirin in 31%, calcium channel blocker in 16%, ACE inhibitor in 16% and beta-blocker in 15%. This apparent similarity hid a large age-related difference in anticoagulation treatment. Women aged 75 years or older were half as likely to receive warfarin as men of the same age, but twice as likely to receive aspirin. The increased likelihood of receiving warfarin was also present in patients with at least one risk factor for stroke.


Recurrence of paroxysmal AF occurred more frequently in women (40-50%) than men (25-35%) at each visit from three months to three years. Total strokes, myocardial infarction and major bleeds occurred at a similar rate in men and women.

However, the rate of major bleed was much higher in women on warfarin (9%) than in men on warfarin (3%). In men and women who bled the mean INR was 4.0 and 4.4 respectively, though the range was wide.

The age-adjusted relative risk for a major bleed for a woman on warfarin compared with a woman not on warfarin was 5.5 (95% CI 1.4 to 21).

The age-adjusted relative risk for a major bleed for a man on warfarin compared with a man not on warfarin was 0.3 (95% CI 0.1 to 1.5).


This cohort study was conducted in Canada, and finished recruiting in 1996. It shows sex differences in treatment and outcomes. between men and women. The most important finding was the increased propensity of women on warfarin to major bleeds. This implies a need for careful monitoring in women on warfarin.