Lower rates of intervention for symptomatic carotid stenosis in women than in men reflect differences in disease incidence: a population-based study.
Marquardt L., Fairhead JF., Rothwell PM.
BACKGROUND AND PURPOSE: Although there is little sex difference in the age-specific incidence of transient ischemic attack (TIA) and stroke, substantially more men than women undergo endarterectomy/stenting for symptomatic carotid stenosis. Sexism in referral for investigation or intervention has been proposed as an explanation; however, a lower incidence of carotid disease in women or reluctance to undergo intervention might also be responsible. METHODS: We determined the sex-specific incidence of symptomatic carotid stenosis and subsequent endarterectomy/stenting from 2002 to 2009 in consecutive patients with TIA or nondisabling ischemic stroke in the Oxford Vascular Study. We studied equivalent data from routine clinical practices in the wider Oxfordshire population. RESULTS: There was no sex difference in age-specific referral rates for carotid imaging in the Oxford Vascular Study (n=616; age-adjusted relative rate [RR] for males vs females=1.08; 95% CI, 0.79 to 1.46; P=0.64). However, rates of 50% to 99% symptomatic carotid stenosis were higher in men (RR=1.89; 95% CI, 1.31 to 2.71; P=0.0005). The same was seen in imaged patients (n=575) in the wider Oxfordshire population (RR=1.82; 95% CI, 1.31 to 2.53; P=0.003) and in pooled data (RR=1.87; 95% CI, 1.32 to 2.64; P=0.0003). Rates of symptomatic carotid occlusion were also higher in men in both populations (RR=3.19; 95% CI, 1.95 to 5.23; P<0.0001). Consequently, although men were more likely to undergo carotid intervention (RR=1.98; 95% CI, 1.43 to 2.75; P<0.0001), the proportion of patients with 50% to 99% symptomatic carotid stenosis who received intervention was similar for men and women (odds ratio=1.13; 95% CI, 0.57 to 2.25; P=0.72). CONCLUSIONS: Lower rates of intervention for 50% to 99% symptomatic carotid stenosis in women can be explained by sex differences in population-based incidence. We found no evidence of any investigation or intervention bias.