CT coronary angiography in patients suspected of having coronary artery disease: decision making from various perspectives in the face of uncertainty.

PURPOSE To determine the cost-effectiveness of computed tomographic (CT) coronary angiography as a triage test, performed prior to conventional coronary angiography, by using a Markov model. MATERIALS AND METHODS A Markov model was used to analyze the cost-effectiveness of CT coronary angiography performed as a triage test prior to conventional coronary angiography from the perspective of the patient, physician, hospital, health care system, and society by using recommendations from the United Kingdom, the United States, and the Netherlands for cost-effectiveness analyses. For CT coronary angiography, a range of sensitivities (79%-100%) and specificities (63%-94%) were used to help diagnose significant coronary artery disease (CAD). Optimization criteria (ie, outcomes considered) were: revised posttest probability of CAD, life-years, quality-adjusted life-years (QALYs), costs, and incremental cost-effectiveness ratios (ICERs). Extensive sensitivity analysis was performed. RESULTS For a prior probability of CAD of less than 40%, the probability of CAD after CT coronary angiography with negative results was less than 1%. The Markov model calculations from the patient/physician perspective suggest that CT coronary angiography maximizes life-years respectively in 60-year-old men and women at a prior probability of less than 38% and 24% and maximizes QALYs at a prior probability of less than 17% and 11%. From the hospital/health care perspective, CT coronary angiography helps reduce health care and direct nonhealth care-related costs (according to UK/U.S. recommendations), regardless of prior probability, and lowers all costs, including production losses (Netherlands recommendations) at a prior probability of less than 87%-92%. Analysis performed from a societal perspective by using a willingness-to-pay threshold level of euro 80,000/QALY suggests that CT coronary angiography is cost-effective when the prior probability is lower than 44% and 37% in men and women, respectively. Sensitivity analyses showed that results changed across the reported range of sensitivity of CT coronary angiography. CONCLUSION The optimal diagnostic work-up depends on the optimization criterion, prior probability of CAD, and the diagnostic performance of CT coronary angiography.