Optimal peak systolic velocity threshold at duplex us for determining the need for carotid endarterectomy: a decision analytic approach.

PURPOSE: To determine the optimal peak systolic velocity (PSV) threshold at duplex ultrasonography (US) required to establish the need for carotid endarterectomy in symptomatic patients on the basis of the long-term cost-effectiveness outcomes of diagnostic testing and subsequent treatment.

MATERIALS AND

METHODS

From January 1997 through January 2000, a prospective medical ethics committee-approved multicenter study was conducted. After giving informed consent, patients with amaurosis fugax, transient ischemic attack, or minor stroke who underwent duplex US and digital subtraction angiography were included in the study. Selective ipsilateral carotid angiograms were obtained in at least three planes. Arteries that were nearly or totally occluded at duplex US were excluded because the PSV cannot be reliably measured in these vessels. Receiver operating characteristic (ROC) curves were constructed for the diagnoses of 70%-99% and 50%-99% stenoses. Optimal likelihood ratios were calculated on the basis of lifetime costs and quality-adjusted life-years derived at cost-effectiveness analysis and the prevalence of disease. The associated optimal sensitivities, specificities, and PSV thresholds were derived from the ROC curves.

RESULTS

In this clinical study, 350 patients were included. The nonoccluded arteries in a total of 236 patients were assessable for ROC analysis. For the diagnosis of 70%-99% stenosis, the optimal likelihood ratio was 0.21, which was associated with a PSV threshold of 220 cm/sec, a sensitivity of 97% (127 of 131 patients; 95% confidence interval [CI]: 94%, 100%), and a specificity of 48% (50 of 105 patients; 95% CI: 38%, 57%). For the diagnosis of 50%-99% stenosis, the optimal likelihood ratio was 0.38, which was associated with a PSV threshold of 180 cm/sec, a sensitivity of 95% (182 of 191 patients; 95% CI: 92%, 98%), and a specificity of 69% (31 of 45 patients; 95% CI: 55%, 82%).

CONCLUSION

On the basis of the lifetime outcomes of diagnostic testing and subsequent treatment, the optimal PSV thresholds for the diagnosis of 70%-99% and 50%-99% carotid artery stenoses in patients with amaurosis fugax, transient ischemic attack, or minor stroke were 220 cm/sec and 180 cm/sec, respectively.