Long-Term Outcome and Predictors of Noninstitutionalized Survival Subsequent to Prolonged Intensive Care Unit Stay After Cardiac Surgical Procedures

Ann Thorac Surg. 2016 Jan;101(1):56-63; discussion 63. doi: 10.1016/j.athoracsur.2015.07.004. Epub 2015 Oct 1.


Background: There are minimal data on long-term functional survival (alive and not institutionalized) in patients undergoing cardiac operations who require a prolonged intensive care unit length of stay (prICULOS). We sought to describe 1- and 5-year functional survival in patients who had a prICULOS (ICULOS ≥ 5 days) and determine predictors of functional survival at 1 year.

Methods: Data were obtained from linked clinical and administrative databases from January 1, 2000 to December 31, 2011 to conduct this retrospective single-region analysis. Logistic regression was used to develop a model predicting functional survival at 1 year for patients who had a prICULOS after cardiac operations.

Results: There were 9,545 admissions to the ICU after cardiac operations; of these patients, 728 (7.6%) experienced a prICULOS. There was an increasing trend in patients who had a prICULOS over this study period. The functional survival at 1 and 5 years from the surgical procedure for the non-prICULOS versus the prICULOS cohort was 1 year (94.9% versus 73.9%) and 5 years (84.9% versus 53.8%) (p < 0.001). Factors associated with lower rates of functional survival at 1 year were age 80 years or older, female sex, peripheral vascular disease, preoperative renal dysfunction, cerebrovascular disease, preoperative infection, need for extracorporeal membrane oxygenation/ventricular assist device (ECMO/VAD) after cardiotomy, number of days on mechanical ventilation, and number of days in the ICU beyond 5 days (area under the receiver operating characteristic [ROC] curve = 0.766).

Conclusions: The majority of patients who had a prICULOS experienced successful functional survival up to 5 years after cardiac operations. Identification of risk factors for poor functional survival may be of assistance to clinicians, patients, and families for prognostication and decision making.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Aged, 80 and over
  • Cardiac Surgical Procedures*
  • Cardiovascular Diseases / epidemiology
  • Cardiovascular Diseases / surgery*
  • Female
  • Follow-Up Studies
  • Hospital Mortality / trends
  • Humans
  • Intensive Care Units / statistics & numerical data*
  • Length of Stay / trends*
  • Male
  • Manitoba / epidemiology
  • Middle Aged
  • Postoperative Period
  • Prognosis
  • ROC Curve
  • Retrospective Studies
  • Survival Rate / trends
  • Time Factors