Clinical burden of ICU admission among patients undergoing resection of pancreatic cancer

Abstract

Presenter: Daniel Rice BS | The Ohio State University Background: Pancreatic adenocarcinoma is a leading cause of cancer-related deaths in the United States. Intensive care unit (ICU) utilization has increased among patients with cancer, and ICU utilization may be more common among surgical patients with cancer – especially among individuals undergoing complex operative procedures such as pancreatectomy. We sought to define disease and demographic factors associated with ICU admissions among patients with pancreatic cancer, as well as characterize trends in mortality among hospital ICU survivors. Methods: The Surveillance, Epidemiology, and End Results (SEER) – Medicare linked database was used to identify patients with a diagnosis of pancreatic cancer who underwent resection between 2004-2015. Multivariable analyses were conducted to identify factors associated with subsequent ICU admission, as well as mortality among hospital survivors. Results: Among 6,422 Medicare beneficiaries who underwent resection of pancreatic cancer, 2,386 (37.1%) had an ICU admission following surgery; ICU utilization was most common at the time of the index surgical hospitalization (n=4,646; 72.3%). Patients with an ICU admission were more likely to be younger (10-year increase OR 0.83, 95%CI 0.77-0.89), male (OR 1.17, 95%CI 1.05-1.30) and undergo resection at a teaching hospital (OR 1.19, 95%CI 1.05-1.36). Among patients who had an ICU admission, while the majority of patients survived to hospital discharge (n= 2,106; 88.3%), a large subset of patients (n=1,296; 54.3%) died within 6 months. In fact, among patients who underwent pancreatic resection and had a subsequent ICU admission, 1- and 5-year overall survival was only 31.8% and 11.0%, respectively. Ventilatory support during ICU admission was associated with marked increased odds of in-hospital mortality (OR 5.67; 95%CI 3.05-10.56); however, among patients who survived to discharge, history of ICU ventilatory support was not associated with 6-month (OR 1.81; 95%CI 0.77-4.26), 1- (OR 1.19; 95%CI 0.50-2.86) or 5-year (OR 1.01, 95%CI 0.29-3.41) survival (all p>0.05). Outcomes following ICU admission did not vary over the last decade. Specifically, among patients who underwent pancreatectomy, the proportion of patients who survived 6-months and 1-year from discharge following an ICU admission was 35.5% and 23.4% in 2004 versus 41.9% and 28.5% in 2015, respectively (both p>0.05)(Figure). Following an ICU admission, patients discharged with home health care (OR 1.48, 95%CI 1.17-1.86) or to a skilled nursing facility (OR 1.91, 95%CI 1.43-2.54) had higher risk of death within a year of discharge versus patients discharged home with self-care (both p<0.05). Conclusion: Over 1 in 3 patients with pancreatic cancer who underwent surgical resection had at least one subsequent ICU admission. While most patients survived the hospitalization, more than one-half of the patients died within 6 months of discharge and two-thirds died within 1-year of discharge. These data should serve to guide patient-provider discussions around prognosis relative to ICU use among patients with pancreatic cancer undergoing resection.