Persistence of risk of death after hospital discharge to locations other than home after cardiac surgery.

Objective: Temporal changes in the risk of postoperative death following cardiac surgery are uncharacterized. We aimed to quantify the duration of postoperative phase with elevated risk of death in patients who underwent cardiac surgery and were discharged to home and destinations other than home.

Method: We conducted a retrospective cohort study of 6894 patients who underwent cardiac surgery between 2008 and 2016 at a tertiary care center in the United States. Logistic regression models with restricted cubic splining along the days since hospital discharge were fitted for risk of death in patients who were discharged to home and those discharged to destinations other than home. The splining curves were analyzed to quantify the duration of postoperative high-risk phase in each cohort.

Results: Mortality rate was significantly higher in the nonhome cohort compared with those discharged to home at 365 days following hospital discharge (9.3% vs 2.1%; P < .001). Discharge to destinations other than home was an independent predictor of late death (hazard ratio, 1.36; 95% confidence interval, 1.15-1.61; P < .001). Analysis of restricted cubic splining curves demonstrated that the postoperative phase with elevated risk of death persisted for 80.3 days in nonhome discharge cohorts, whereas a comparable phase was nonexistent in patients who went home. Predictors of nonhome discharge were identified, with combination of preoperative and postoperative variables yielding C statistics of 0.83.

Conclusions: Hospital discharges to locations other than home following cardiac surgery were associated with an increased risk of late mortality. The postoperative high-risk phase persisted for 80 days in patients who were discharged to locations other than home, supporting the use of 90-day outcome measures as quality metrics. Predictors of discharge to locations other than home were identified, and this may aid in selective intervention to reduce the risk of death in this vulnerable patient population.