Predictive value of exercise stress testing in a family medicine population.

PURPOSE Exercise stress testing (EST) is a screening test for coronary artery disease. Previous studies from the cardiology literature show an overall sensitivity of 67% and specificity of 72% with variable predictive values depending on pretest probability. The purpose of the current study was to evaluate the predictive value of EST in a family medicine population in eastern North Carolina. METHODS This is a retrospective case series of 339 ESTs performed in a family medicine center from July 2001 to April 2005. EST results were classified as positive, negative, or equivocal. Outcomes studied from a review of outpatient and inpatient electronic medical record data and telephone follow-up included myocardial infarction, cardiac catheterization with angioplasty and stenting, coronary artery bypass grafting, a new diagnosis of coronary artery disease, and cardiac death. Mean duration of follow-up was 47 months, with a range of 27 to 72 months. RESULTS Nearly all patients had low to intermediate risk pretest probability. Five tests were positive, 32 were equivocal, and 302 were negative. There were 2 false-positive tests, both in female patients. There were 2 false-negative tests, both of which were treated with good outcomes. Two of 32 equivocal results had cardiac outcomes. Considering equivocal tests as positive, the overall sensitivity in this series was 71.4%; specificity was 90.4%. The positive predictive value was 13.5% and the negative predictive value was 99.3%. CONCLUSIONS The high negative predictive value for EST in this outpatient family medicine population is noteworthy and reassuring. EST is a cost-effective strategy for triaging the common complaint of chest pain in low- to intermediate-risk patients in primary care practices and should be included in the services offered to family medicine patients.