Cost-Effectiveness Analysis of Therapy for Symptomatic Carotid Occlusion: PET Screening Before Selective Extracranial-to-Intracranial Bypass Versus Medical Treatment

The St. Louis Carotid Occlusion Study (STLCOS) demonstrated that increased cerebral oxygen extraction fraction (OEF) detected by PET scanning predicted stroke in patients with symptomatic carotid occlusion. Consequently, a trial of extracranial-to-intracranial (EC/IC) arterial bypass for these patients was proposed. The purpose of this study was to examine the cost-effectiveness of using PET in identifying candidates for EC/IC bypass. A Markov model was created to estimate the cost-effectiveness of PET screening and treating a cohort of 45 symptomatic patients with carotid occlusion. The primary outcome was incremental cost for PET screening and EC/IC bypass (if OEF was elevated) per incremental quality-adjusted life year (QALY) saved. Rates of stroke and death with surgical and medical treatment were obtained from EC/IC Bypass Trial and STLCOS data. Costs were estimated from the literature. Sensitivity analyses were performed for all assumed variables, including the PET OEF threshold used to select patients for surgery. In the base case, PET screening of the cohort followed by EC/IC bypass on 36 of the 45 patients yielded 23.2 additional QALYs at a cost of $20,000 per QALY, compared with medical therapy alone. A more specific PET threshold, which identified 18 surgical candidates, gained 22.6 QALYs at less cost than medical therapy alone. The results were sensitive to the perioperative stroke rate and the stroke risk reduction conferred by EC/IC bypass surgery. If postoperative stroke rates are similar to stroke rates observed in the EC/IC Bypass Trial, EC/IC bypass will be cost-effective in patients with symptomatic carotid occlusion who have increased OEF. A clinical trial of medical therapy versus PET followed by EC/IC bypass (if OEF is elevated) is warranted.