Readmission after pancreatectomy for pancreatic cancer in Medicare patients

Abstract

Objective: The objective of this study was to use a population-based dataset to evaluate the number of readmissions and reasons for readmission in Medicare patients undergoing pancreatectomy for pancreatic cancer.

Methods: We used Surveillance, Epidemiology, and End Results-Medicare linked data (1992-2003) to evaluate the initial hospitalization, readmission rates within 30 days (early), and between 30 days and 1 year (late) after initial discharge and reasons for readmission in patients 66 years and older undergoing pancreatectomy.

Results: We identified 1,730 subjects who underwent pancreatectomy for pancreatic cancer. The in-hospital mortality was 7.5%. The overall Kaplan-Meier readmission rate was 16% at 30 days and 53% at 1 year, accounting for 15,409 additional hospital days. Early readmissions were clearly related to operative complications in 80% of cases and unrelated diagnoses in 20% of cases. Late readmissions were related to recurrence in 48%, operative complications in 25%, and unrelated diagnoses in 27% of cases. In a multivariate analysis, only distal pancreatic resection (P = 0.02) and initial postoperative length of stay > or =10 days (P = 0.03) predicted early readmission. When compared to patients not readmitted, patients readmitted early had worse median survival (11.8 vs.16.5 months, P = 0.04), but the 5-year survival was identical (18%). Late readmission was associated with worse median and 5-year survival (19.4 vs. 12.1 months, 12% vs. 21%, P < 0.0001).

Conclusions: Our study demonstrates overall 30-day and 1-year readmission rates of 16% and 53%. The majority of early readmissions were related to postoperative complications but not related to patient and tumor characteristics. Complications causing early readmission are a cause of early mortality and are potentially preventable. Conversely, late readmissions are related to disease progression and are a marker of early mortality and not the cause.

Figure 1

Kaplan–Meier curves showing the time to readmission over the first year following initial discharge after pancreatectomy in Medicare patients. a Kaplan–Meier curve for time to readmission with deaths treated as censored. b Kaplan–Meier curve for time to readmission with deaths treated as a competing event. c The table shows the patients at risk at five different time intervals over the first year following discharge, the cumulative readmissions, and the cumulative deaths for five time periods, and the Kaplan–Meier estimates of readmission rates as well as combined readmission and death rates.

Figure 2

Kaplan–Meier actuarial survival curves showing comparing survival in patients who did not require readmission and those who were readmitted early (within 30 days, N = 248). When compared to patients not readmitted, patients readmitted early had worse median survival (11.8 vs.16.5 months, P = 0.04), but the long-term survival was identical (18%).

Figure 3

Kaplan–Meier actuarial survival curves showing comparing survival in patients who did not require readmission and those who were readmitted late (between 30 days and 1 year, N = 536). Late readmission was associated with worse median and long-term survival (19.4 vs. 12.1 months, 21% vs. 12%, P < 0.0001).