Imaging of carotid artery stenosis: clinical efficacy and cost-effectiveness.

PURPOSE To determine the most accurate, safe, and cost-effective imaging protocol for selecting patients for carotid endarterectomy.

METHODS The actual costs of carotid angiography, ultrasound, and MR angiography were calculated. The diagnostic accuracy with different confidence levels was assessed for carotid ultrasound and MR angiography in 45 patients. The cost-effectiveness and theoretical impact on patient outcome of hypothetical screening models were compared.

RESULTS Ultrasound before angiography is more effective and considerably cheaper than performing angiography in all patients presenting with transient ischemic attacks ($25 216 versus $48 708 imaging costs per one prevented stroke). When the more costly MR angiography was used to select patients for angiography the slightly higher diagnostic accuracy did not result in a greater number of prevented strokes. As the only preoperative scrutiny, the combination of ultrasound and MR angiography would have resulted in a greater number of prevented strokes than invasive angiography (27.9 versus 23.3) but at the expense of unnecessary surgery (6.6% of all surgeries).

CONCLUSIONS Ultrasound followed by confirmatory angiography is a cost-effective way to image patients suspected of carotid artery stenosis. MR angiography may become cost effective and lead to a better final patient outcome only when it can reliably replace invasive angiography as the preoperative examination.