Study objective: Economic forces have precipitated
intense interest in cost-saving practices for patients undergoing
coronary artery bypass grafting (CABG). While several preoperative
variables have been implicated in higher costs, few studies have
included perioperative factors. This study evaluated the predictive
power of a preoperative mortality risk measurement (Parsonnet score)
and of early extubation (≤ 6 h from ICU admission) in determining ICU
and hospital costs.
Design: Multivariate correlational
Setting: University hospital in a large
Patients: All patients (n = 116)
undergoing isolated CABG during a 6-month period were studied after the
introduction of a clinical pathway.
results: Clinical data were collected. Costs data were obtained
retrospectively from the institutional data system and were derived
from individual patient charges by application of department-specific
cost-to-charge ratios. In multivariate logistic regression, Parsonnet
score (per point odds ratio [OR], 1.09; confidence interval [CI],
1.03 to 1.17), in-hospital coronary angiography (OR, 3.51; CI, 1.23 to
10.01), delayed extubation (OR, 4.59; CI, 1.29 to 16.29), and presence
of arrhythmia (OR, 3.50; CI, 1.15 to 10.64) were independent predictors
of ICU costs. Only Parsonnet score (OR, 1.09; CI, 1.03 to 1.15) and
cardiopulmonary bypass time (OR, 1.01; CI, 1.00 to 1.02) were
independent predictors of hospital costs.
The Parsonnet score is a useful indicator of both ICU and hospital
costs. Early extubation is associated with decreased ICU costs, but is
not independently predictive of hospital costs.