Skip to main content

Table 2 Stepwise selection of variables for discriminating morbidity with the Bayesian model

From: A multivariate Bayesian model for assessing morbidity after coronary artery surgery

STEP

Variable

Area under ROC curve

1

DO2I (mL/minute/m2) after 3 hours in ICU

0.7080

2

Inotropic support after CPB

0.7407

3

PVD and/or CD

0.7573

4

Preoperative creatinine (mg/dL)

0.7692

5

IABP after CABG

0.7771

6

Weight (kg)

0.7827

7

REDO

0.7876

8

Duration of CPB (minutes)

0.7923

9

Age (years)

0.7949

10

WBC (103/mm3) after 3 hours in ICU

0.7966

11

Preoperative IABP

0.7976

12

Emergency operation

0.7988

  1. The area under the receiver operating characteristic (ROC) curve represents discrimination performance. The stepwise technique was employed in combination with the leave-one-out method of cross-validation. The variable that adds the most or least to the area under the ROC curve is entered or removed, respectively, at each step. The stepwise process stopped when no further significant improvement was found. CABG, coronary artery bypass graft; CD, carotid disease; CPB, cardiopulmonary bypass; DO2I, oxygen delivery index; IABP, intra-aortic balloon pump; ICU, intensive care unit; PVD, peripheral vascular disease; REDO, re-operation; WBC, white blood cells.