901-54 Exercise Echocardiography is More Cost Efficient than Exercise ECG as an Initial Test for Evaluation of Cardiac Symptoms in Women

Abstract

Although exercise echo (ExE) has been shown to be more accurate and specific than exercise electrocardiography (ExECG) in women, the clinical and cost implications of incorporating ExE in the evaluation of suspected coronary artery disease (CAD) in women has not been well resolved. The aim of this study was to assess the accuracy, angiography rate. and cost of various diagnostic strategies in women. A consecutive group of 161 female patients (age 60 ± 9 yrs) underwent ExECG, ExE and coronary angiography. Positive ExECG was defined as > 0.1 mV ST depression, and positive ExE was defined by a new or worsening wall motion abnormality. Sensitivity (sens) and specificity (spec) were calculated by comparison of ExE (sens 79%, spec 81%) and ExECG (sens 77%, spec 56%) with angiography(significant stenosis > 50%). Pre-test CAD probability (44 ± 33%) was calculated from age and symptoms. Seven different strategies (see table) involving angiography, ExECG ExE, selective combination (ExE for nondiagnostic ExECG), stepwise combination (ExE for positive or nondiagnostic ExECG), and Bayesian approaches (all high pre-test CAD probability pts undergoing angiography, intermediate probability pts undergoing angiography, if ExE or ExECG positive); Strategy Cost/pt ($) Angio(%) Negative angio (%) FalseNeg% I Angiography 1.434 100 63 0 II Exercise ECG 1,023 69 56 11 III Exercise Echo 828 41 29 13 IV Selective ECG/echo 836 51 44 14 V Stepwise ECG/echo 663 31 26 22 VI Bayesian ECG 745 37 27 25 VII Bayesian ExEcho 641 51 48 29 Use of ExE in all pts, with angiography only with positive ExE (III) involved less angiography and was less expensive at a similar level of accuracy to I and II. Strategy IV is least expensive but would lead to an unacceptable number of false negatives. The Bayesian approaches are compromised by the limitations of clinical stratification of women into high and low probability groups based on symptoms. The greater spec of ExE in women avoids inappropriate angiography. Use of ExE as an initial test for CAD in women is justifiable on cost grounds.