A straightforward approach to designing a scoring system for predicting length-of-stay of cardiac surgery patients

Table 2 The whole set of non dichotomous preoperative, intraoperative and postoperative variables considered a priori as likely predictors

Variable (units)	99% confidence interval of odds ratio	Cut-off
Age (years)	1.36-2.46	69
Height (cm)	0.668-1.19	167
Weight (kg)	0.587-1.05	71
Body surface area (m²)	0.550-0.983	1.8
NYHA heart failure classification	1.55-2.78	2
Preoperative hematocrit (%)	0.499-0.895	39.2
Preoperative creatinine (mg/dl)	1.51-2.69	1.0
Preoperative bilirubin (mg/dl)	0.812-1.45	0.8
ECC time (min)	1.75-3.22	122
Aortic clamping time (min)	1.44-2.61	86
Minimum hematocrit at cardiopulmonary bypass (%)	0.517-0.924	24
Minimum intraoperative temperature (°C)	0.410-0.743	32.5
Postoperative hematocrit (%)	0.570-1.02	31
Postoperative venous O ₂ saturation (%)	0.656-1.17	63.7
Postoperative creatinine (mg/dl)	3.35-6.29	1.2
Postoperative bilirubin (mg/dl)	1.99-3.70	1

NYHA = New York Heart Association; ECC = extracorporeal circulation.
The 99% confidence intervals of the odds ratio were calculated after dichotomizing each variable on the basis of the cut-off (median calculated on original sample). The variables in italics were excluded from the model because their corresponding odds ratios were not significantly different from 1 (p < 0.01).

ISSN: 1472-6947