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

Table 1 The whole set of dichotomous preoperative, intraoperative and postoperative variables considered a priori as likely predictors

Variable	99% confidence interval of odds ratio
Gender	0.659-1.21
Emergency	1.44-6.57
Urgency	0.970-2.39
Blood hypertension	0.987-1.87
Treated diabetes	0.884-1.70
Acute myocardial infarction	1.57-6.38
Recent myocardial infarction (<7 days)	0.663-2.96
Preoperative intra-aortic balloon pump	2.79-9.63
Tachycardia	0.885-5.47
Cardiogenic shock	3.15-21.85
Cardiac massage	0.988-21.62
Endocarditis	0.394-8.77
Unstable angina	0.729-1.47
Heart failure	1.90-5.25
Chronic dialysis	3.51-36.70
Previous cerebrovascular events	0.998-3.01
Chronic obstructive pulmonary disease	1.07-2.80
Anti-platelet drugs	0.545-0.984
Dicumarole therapy	1.39-4.18
Heparin therapy	0.873-1.57
Thrombolytic therapy	0.125-3.88
Combined cardiovascular surgery	2.19-4.21
Repeated cardiac surgery	1.63-4.68
Aortocoronary bypass	0.712-1.32
Plastic valve prosthesis	1.48-3.14
Valve substitution	1.22-2.72
Plastic surgery for aortic valve	0.241-8.86
Aortic valve substitution	1.06-1.96
Plastic surgery for tricuspid valve	1.50-7.70
Surgical procedure different from isolated coronary artery bypass graft	1.52-2.74
Aortic arch pathology	0.776-4.30
Carotid endarterectomy	0.382-2.55
Addition of blood to cardiopulmonary bypass circuit	1.96-3.59
Use of fresh frozen plasma	2.24-4.11
Significant bleeding (>100 ml)	0.728-1.39
Low postoperative cardiac output	12.3-23.9

The variables in italics were eliminated from the model because their corresponding odds ratios were not significantly different from 1 (p < 0.01).

ISSN: 1472-6947