TABLE 1

Multivariate Stepwise Logistic Regression Analysis for Any Stress-Induced Myocardial Perfusion Abnormality by PET as the Dependent Variable

Independent variableDependent variable: discrete, any abnormalityBEXP(B)95% CI of EXP(B)P for significance
Resting homogeneity index, rHQ < 1 SD or L < 1 SD−23.10.0000.000–0.000<0.001
Rest-to-stress change in H, rsHΔQ < 1 SD or L < 1 SD−22.20.0000.000–0.000<0.001
MaleQ < 1 SD or L < 1 SD1.997.293.40–15.63<0.001
AgeQ < 1 SD or L < 1 SD0.021.021.00–1.040.019
Hx of diabetesQ < 1 SD or L < 1 SD1.886.572.39–18.03<0.001
Hx of hypertensionQ < 1 SD or L < 1 SD0.551.731.29–2.33<0.001
Hx of overweightQ < 1 SD or L < 1 SD0.431.531.13–2.07<0.001
Hx of exerciseQ < 1 SD or L < 1 SD0.571.771.04–3.020.036
Hx of high cholesterolQ < 2 SD0.852.331.32–4.110.004
PostmenopausalQ < 2 SD3.0220.451.98–288.20.011
Family Hx of vascular diseaseQ < 2 SD−0.180.840.48–1.47NS
Hx of smokingQ < 2 SD0.061.070.78–1.45NS
  • B = loge of odds ratio; EXP = exponent; Q = lowest average quadrant activity on stress PET image; L = longitudinal base-to-apex perfusion gradient on stress PET image; Hx = history; NS = not significant.