Secondary Logo

Journal Logo

Fast-Track/Off Pump/Prediction/Outcome

Long term follow up of patients with short and long postoperative ICU stay after cardiac surgery: 013

Villa, G. B.; Mazzoni, M.; De Maria, R.1; Parolini, M.1; Ceriani, R.; Solinas, C.; Arena, V.2; Bortone, F.; Parodi, O.1

Author Information
European Journal of Anaesthesiology: June 2004 - Volume 21 - Issue - p 26

Introduction: Several models have been validated to predict in-hospital mortality in cardiac surgical patients [1] and in those requiring prolonged ICU stay for perioperative multiple organ dysfunction [2]. However, risk factors of long term outcome have not been fully elucidated. In a cohort of cardiac surgical patients, we evaluated the impact of prolonged postoperative intensive care on long term outcome.

Method: Adult patients (n = 90) with a postoperative ICU stay longer than 96 hours were matched according to gender, age, EUROScore and type of surgery (coronary, valvular, mixed, thoracic aorta) with a group of patients (n = 180) with ICU stay <96 hours. We assessed by a multivariate Cox proportional hazard model the impact of left ventricular ejection fraction (EF), diabetes, ICU stay >96 hours, comorbidities and preoperative renal dysfunction on all-cause mortality and on a combined end point of mortality and cardiovascular events.

Results: The two groups of patients were comparable according to presence of one or more comorbidities, no comorbidity, previous or <3 months acute myocardial infarction, previous cardiac surgery, systemic or pulmonary hypertension, peripheral arteriopathy and chronic obstructive pulmonary disease. Patients with longer ICU stay more frequently had diabetes (P = 0.006) or creatinine level >200 μmol/L (P = 0.001) preoperatively. In hospital mortality was 2.8%.

At mean follow up of 19 ± 10 months, survival was 89% in those requiring short ICU stay and 79% (P = 0.009) in the long ICU stay group. Need of ICU stay >96 hours was the only independent predictor of long term mortality by multivariate analysis (RR 2.36, C.I. 1.17-4.78, P = 0.016).

Long term event free survival was 67% in those with short ICU stay and 41% (P = 0.0001) in the long ICU stay group. By multivariate analysis the only independent factor predictive of long term outcome was postoperative ICU stay >96h (RR 2.20, 95% C.I. 1.46-3.30, P = 0.0001), whereas higher EF had a protective effect (RR 0.97, 95% C.I. 0.95-0.99, P = 0.003).

Discussion: Patients requiring prolonged ICU stay after cardiac surgery are at increased risk of mortality and cardiovascular events after hospital discharge. Measures directed at limiting perioperative complications, particularly preservation of left ventricular performance, should be maximized in an effort to improve long term outcome of cardiac surgical patients.


1 Geissler HJ, Holzl P, Marohl S, et al. Risk stratification in heart surgery: comparison of six score systems. Eur J Cardiothorac Surg 2000; 17(4): 400-406.
2 Ceriani R, Mazzoni M, Bortone F, et al. Application of the sequential organ failure assessment score to cardiac surgical patients. Chest 2003; 123(4): 1229-1239.
© 2004 European Society of Anaesthesiology