Transient ischemic attacks: stratifying risk.

In recent years, it has become clear that the risk of stroke after a transient ischemic attack (TIA) or minor ischemic stroke is higher than was previously supposed, with consistent reports of 7-day stroke risks of up to 10%,1–4 and other evidence of the very short time-window for prevention of stroke after a TIA.5 However, patients with TIA and minor stroke are a highly heterogeneous group in terms of symptoms, risk factors and underlying pathology, and the early risk of recurrent stroke is likely to vary between different clinical and etiological subtypes. In order to appropriately target secondary prevention, we therefore need reliable data on risk in particular subgroups and ideally in individuals. Recent studies have provided some useful data, although many important issues are still unresolved. There is good evidence that the presenting clinical features of a TIA provide considerable prognostic information. Johnston and colleagues identified 5 risk factors independently associated with a higher 3-month risk of recurrent stroke in a large emergency department–based TIA cohort: age 60 years (OR 1.8; 95% CI, 1.4 to 2.9), symptom duration 10 minutes (2.3, 1.3 to 4.2), weakness (1.9, 1.4 to 2.6), speech impairment (1.5, 1.1 to 2.1), and diabetes mellitus (2.1, 1.1 to 2.7).1 A simple index with 1 point for each risk factor was useful in estimating risk at 3 months, which varied from 0% in patients with no risk factors to 34% in those with 5 risk factors, and also differentiated between risk groups during the first few days after the TIA.1 Isolated sensory or visual symptoms were associated with a low risk of stroke, and sex, ethnicity, previous diagnoses of coronary artery disease or hypertension, current cigarette smoking, antiplatelet or anticoagulant-use at presentation and presentation blood pressure did not predict early stroke.1,6 Rothwell and colleagues studied predictors of stroke during the 7 days after a TIA in 2 independent population-based studies and derived and validated a prognostic score specifically for this very early risk.7 The Table compares the regression model for the 7-day stroke risk derived from the population-based studies7 with the similar model for 90-day risk of stroke derived by Johnston and colleagues.1 The independent predictors are remarkably similar, the main difference being in the size of the hazard ratios for 7-day risk versus 90-day risk. These risk models clearly demonstrate that the early risk of stroke after a TIA is highly predictable. The models will no doubt be further refined, but the simple scores developed thus far can already be used in routine clinical practice. Rothwell and colleagues developed the 6-point “ABCD” score (Age 60 years 1; Blood pressure: systolic 140 mm Hg and/or diastolic 90 mm Hg 1; Clinical features: unilateral weakness 2, speech disturbance without weakness 1, other 0; Duration of symptoms in minutes: 60 2, 10 to 59 1, 10 0), which was highly predictive of the 7-day risk of stroke in 2 independent validation cohorts.7 In a populationbased cohort of all referrals with suspected TIA, 19/20 early recurrent strokes occurred in 27% of the patients with a score 5: 7-day risk was 0.4% (95% CI, 0 to 1.1) in 274 (73%) patients with a score 5, 12.1% (4.2 to 20.0) in 66 (18%) with score 5, and 31.4% (16.0 to 46.8) in 35 (9%) with score 6. In a hospital-referred weekly clinic cohort, all patients who had a stroke before their scheduled appointment (n 14, 7.5%) had a score of 4.7 Weakness or speech disturbance was associated with an increased risk of stroke at 3 months in another study of prognosis after TIA,8 and a recent study of predictors of stroke during the first year after TIA identified hypertension, diabetes and increasing age as independent risk factors but did not collect data on the nature of the presenting symptoms.4 Early risk of stroke after a TIA or minor stroke is also related to the vascular territory of the presenting event. For example, in keeping with the lower long-term risk of stroke after monocular TIAs versus carotid territory cerebral TIAs,9 the early risk of stroke after monocular events is also low. Posterior circulation territory events, which account for 25% of all TIAs, have also been thought to have a good prognosis and are often investigated and treated less rigorously than carotid territory events. However, recent evidence suggests that there are no major differences in long-term prognosis and that the early risk of stroke is, in fact, higher after posterior circulation territory events.10 In a meta-analysis of cohort studies in which risks could be compared, studies that recruited during the acute phase after the pres-