The need to quantify right-to-left shunt in acute ischemic stroke: a case-control study

BACKGROUND AND PURPOSE: Although right-to-left shunt (RLSh) has been reported to be significantly more frequent in young stroke patients with cryptogenic stroke, its relevance in a nonselected population of acute ischemic stroke is not well known. The aim of this study was to determine the importance of the RLSh magnitude as a risk factor for stroke in nonselected patients. METHODS: Two hundred eight patients hospitalized consecutively with transient ischemic attack or acute cerebral infarction and 100 healthy control subjects were studied. Transcranial Doppler ultrasonography (TCD) was performed in both middle cerebral arteries (MCAs) after intravenous application of agitated saline solution. The magnitude of RLSh was quantified by counting the number of signals in 1 MCA during a Valsalva maneuver. RLSh was classified as "no shunt," "small" (< 10 signals), and "large" (> 10 signals), with the latter including the "shower" (> 25 signals) and "curtain" (uncountable signals) patterns. Extensive investigations, including contrast transesophageal echocardiography, were carried out on patients diagnosed as suffering from stroke of an uncertain etiology. The importance of RLSh for stroke was assessed by logistic regression analysis. RESULTS: Contrast TCD detected a large RLSh in 40 (19.7%) patients and in 21 (21%) control subjects, all with cardiac RLSh characteristics. A large RLSh was present in 4.7% of atherothrombotic strokes, 10.5% of cardioembolic strokes, 15.4% of lacunar strokes, and 45.3% of cryptogenic strokes (P<0.001). Although the overall frequency of RLSh was not significantly different between patients and control subjects, the detection of curtain or shower patterns by contrast TCD was associated with a higher risk of stroke (odds ratio, 3.5; 95% confidence interval, 1.29 to 9.87), particularly with cryptogenic stroke (odds ratio, 12.4; 95% confidence interval, 4.08 to 38.09) after adjustment for concomitant vascular risk factors. CONCLUSIONS: It is essential to quantify RLSh by contrast TCD during the Valsalva maneuver given that only those with shower and curtain patterns are associated with a higher risk of ischemic stroke in a nonselected population.