Stroke Likely Within Week of TIA


OXFORD, England, Nov.16 -- One in 20 patients who experience a transient ischemic attack will suffer a stroke within a week of the incident TIA, researchers here said.
Action Points
Explain to patients that there are now good data to show an increased risk for major stroke within seven days following a TIA or minor stroke.
Explain to high-risk patients the importance of seeking emergency medical attention if they experience a TIA or minor stroke, preferably at a facility specializing in stroke treatment.
The finding emerged from a systematic review of 18 studies that identified stroke within seven days of TIA, published online November 12 by Lancet Neurology.
The pooled risk for major ischemic event in the week following documented TIA was 5.2% (95% CI: 3.9 to 6.5), but there was substantial heterogeneity among studies (P0.0001), wrote Matthew F. Giles, D.Phil., and Peter M. Rothwell, M.D., of Oxford University.
Not surprisingly the risk for stroke was lowest in those patients treated as emergency cases in specialized stroke units with risk ranging from 0.6% at two days post-TIA (95% CI: 0.0 to 1.6) to 0.9% at seven days post-TIA (95% CI: 0.0 to 1.9). Likewise, the one-week stroke risk was greatest -- 11% (95% CI: 8.6 to 13.5) -- among patients who did not seek urgent or emergency care for TIA.
The meta-analysis evaluated all studies reporting stroke incidence within seven days of a TIA. The investigators searched electronic databases and relevant journals, conference abstracts, and reference lists. Calculations were done for overall stroke at two, seven, and 90 days post TIA. Additionally, the studies were analyzed for heterogeneity following categorization of study setting, population method, and treatment, where possible.
The 18 studies included data from 10,126 TIA patients treated from 1981 through 2007. The mean ages of patients ranged from 61 to 73 years (P0.0001), and the proportion of men ranged from 39% to 62% (P0.0001).
While there was considerable heterogeneity between the studies (seven-day stroke risk: range 0% to 12.8%), the differences were almost entirely explained by varying study methodologies, treatments, and settings. "Our study almost fully explains why the results of previous studies have been conflicting, and illustrates the importance of the methods used by a medical study when interpreting its results," said Dr. Giles.
Data on patient characteristics, such as vascular risk factors, frequency of high-risk features for early stroke including weakness, longer symptom duration, speech deficit, and underlying etiology were also included.
Treatment information included delay from event to treatment; specialist versus nonspecialist unit; inpatient versus outpatient; administration of antiplatelet or antihypertensive therapy, statins or carotid endarterectomy; and frequency of hospital admission.
The researchers wrote that study populations, which included very few non-white patients, limited the study, thus the findings may not be extrapolatated to non-white patients.
They also pointed out another significant limitation of the study: "Meta-analysis of observational studies done in several countries, using multiple methods, over a 20-year period is liable to be confounded by many factors."
The authors concluded, "the risk of stroke reported amongst patients treated urgently in specialist units was substantially lower than risks reported among other patients treated in alternative settings. These results support the argument that a TIA is a medical emergency and that urgent treatment in specialist units may reduce the risk of subsequent stroke. This is particularly relevant in the UK, where TIA services are patchy and there are substantial delays to TIA patients receiving appropriate treatments."
Dr. Mathew Giles received funding from the Stroke Association, while writing the paper.Primary source: Lancet NeurologySource reference: Giles MF, Rothwell PM, "Risk of stroke early after transient ischemic attack: a systematic review and meta-analysis"Lancet Neurology 2007; DOI:10.1016/S1474-4422(07)70274-0.