Mortality in Medicare patients undergoing elective percutaneous coronary intervention with or without antecedent stress testing

Abstract

Background: Guidelines advise testing for ischemia, such as with stress testing, before elective percutaneous coronary intervention (PCI). However, pre-PCI stress testing is not always done; the implications of this practice are not known. Our objective was to evaluate whether receipt of stress testing before elective PCI predicts mortality.

Methods and results: Using claims data from a 20% random sample of Medicare beneficiaries, we identified patients who had elective PCI in 2004 and followed them for a median of 3.4 years (n=23 887). Cox proportional hazards models were used to test the relationship of pre-PCI stress testing to survival. Population-based rates of elective PCI and stress testing were calculated for 306 hospital referral regions and categorized into 4 groups: high stress test/high PCI, low stress test/low PCI, low stress test/high PCI, and high stress/low PCI regions. Cox modeling was used to test whether category of hospital referral regions is related to survival. Patients who underwent pre-PCI stress testing had a 13% lower risk of mortality than those who did not (adjusted hazard ratio, 0.87; 95% confidence interval, 0.81-0.92) after median follow-up of 3.4 years. Patients in low stress test/high PCI regions had a 14% higher risk of mortality than those in high stress test/high PCI regions (adjusted hazard ratio, 1.14; 95% confidence interval, 1.03-1.26).

Conclusions: Pre-PCI stress testing is associated with lower mortality in patients undergoing elective PCI. Greater adherence to guidelines with respect to documenting ischemia before elective PCI may result in improved outcomes for patients.

Keywords: angioplasty; coronary disease; ischemia; revascularization.

Figure 1. Study Cohort

The cohort includes all Medicare patients determined to have undergone an elective percutaneous coronary intervention (PCI) in 2004. CABG = coronary artery bypass graft; MI = myocardial infarction; HRR = hospital referral region

Figure 2. Cumulative Mortality of Patients Undergoing Elective PCI by Hospital Referral Region

We calculated population-based rates of stress tests and elective PCI for each HRR, and divided the cohort into four groups of HRRs: high stress test/high PCI rate, high stress test/low PCI rate, low stress test/low PCI rate, and low stress test/high PCI rate regions. Cumulative mortality is shown, with patients in the low stress test/high PCI groups having significantly higher mortality than the other groups (adjusted hazard ratio 1.12, 95% CI 1.02, 1.24).