Survival of Elderly Patients Undergoing Pancreatoduodenectomy in an Integrated Health System.

INTRODUCTION: The purpose of this study is to examine pancreatoduodenectomy (PD) perioperative outcomes and consider how age may be related to overall survival in an integrated health system. MATERIALS AND METHODS: A retrospective review was performed of 309 patients who underwent PD between December 2008 and December 2019. Patients were divided into two groups: aged 75 y or less and more than 75 y, defined as senior surgical patients. Univariate and multivariable analyses of predictive clinicopathologic factors associated with overall survival at 5 y were performed. RESULTS: In both groups, the majority underwent PD for malignant disease. The proportion of senior surgical patients alive at 5 y was 33.3% compared to 53.6% of younger patients (P = 0.003). There were also statistically significant differences between the two groups with respect to body mass index, cancer antigen 19-9, Eastern Cooperative Oncology Group performance status, and Charlson comorbidity index. On multivariable analysis, disease type, cancer antigen 19-9, hemoglobin A1c, length of surgery, length of stay, Charlson comorbidity index, and Eastern Cooperative Oncology Group performance status were found to be statistically significant factors for overall survival. Age was not significantly related to overall survival on multivariable logistic regression and when the analysis was limited to pancreatic cancer patients. CONCLUSIONS: Although the difference in overall survival between patients aged less than and more than 75 years was significant, age was not an independent risk factor for overall survival on multivariable analysis. Rather than a patient's chronological age, his/her physiologic age including medical comorbidities and functional status may be more correlated to overall survival.

Pancreatic Adenosquamous Carcinoma: Experience Within an Integrated Health Care System.

Introduction Adenosquamous carcinoma (ASC) of the pancreas is a rare form of pancreatic cancer with a worse prognosis than pancreatic ductal adenocarcinoma. The authors report on a retrospective study of 13 patients diagnosed with ASC in an integrated health care system. Methods A retrospective review was performed of all patients with pancreatic cancer identified between February 2010 and December 2018. Twenty-three patients were diagnosed with pancreatic ASC. Patient demographics, tumor characteristics, treatment modalities, and median survival were evaluated. Results Median overall survival was 8 months (standard devision [SD] = 18.6). Eight out of 13 patients who received surgery upfront had a positive surgical margin (62%). Eleven patients received adjuvant therapy. Median survival for patients who received multimodal treatment was 57 months (SD = 5.7) compared with 2.5 months for patients who received only surgery. Median survival for patients with negative pathologic margins was 17 months (SD = 23.6). One patient was receiving neoadjuvant chemotherapy (6 months into treatment without any evidence of metastatic disease). Discussion The high proportion of positive surgical margins and large tumor size upon presentation suggest that primary tumor downstaging should be considered. The positive results from recent prospective trials on neoadjuvant chemoradiation for pancreatic ductal adenocarcinoma could be a promising foundation of information for the treatment of ASC. Conclusion ASC of the pancreas is an extremely aggressive malignancy with poor prognosis. Further work is needed to determine the optimal multimodal treatment regimen.

Predicting early discharge and readmission following pancreaticoduodenectomy [S079].

INTRODUCTION: Implementing enhanced recovery after surgery (ERAS) protocols for major abdominal surgery has been shown to decrease length of stay (LOS) and postoperative complications, including mortality and readmission. Little is known to guide which patients undergoing pancreaticoduodenectomy (PD) should be eligible for ERAS protocols. METHODS AND PROCEDURES: A retrospective chart review of all PD performed from 2010 to 2018 within an integrated healthcare system was conducted. A predictive score that ranges from 0 to 4 was developed, with one point assigned to each of the following: obesity (BMI > 30), operating time > 400 min, estimated blood loss (EBL) > 400 mL, low- or high-risk pancreatic remnant (based on the presence of soft gland or small duct). Chi-squared tests and ANOVA were used to assess the relationship between this score and LOS, discharge before postoperative day 7, readmission, mortality, delayed gastric emptying (DGE), and pancreatic leak/fistula. RESULTS: 291 patients were identified. Mean length of stay was 8.5 days in those patients who scored 0 compared to 16.2 days for those who scored 4 (p = 0.001). 30% of patients who scored 0 were discharged before postoperative day 7 compared to 0% of those who scored 4 (p = 0.019). Readmission rates for patients who scored 0 and 4 were 12% and 33%, respectively (p = 0.017). Similarly, postoperative pancreatic fistula occurred in 2% versus 25% in these groups (p = 0.007). CONCLUSION: A simple scoring system using BMI, operating time, EBL, and pancreatic remnant quality can help risk-stratify postoperative PD patients. Those with lower scores could potentially be managed via an ERAS protocol. Patients with higher scores required longer hospitalizations, and adjunctive therapy such as medication and surgical technique to decrease risk of delayed gastric emptying and pancreatic fistula could be considered.

Centralization of pancreatic cancer treatment within an integrated healthcare system improves overall survival.

BACKGROUND: Higher-volume centers for pancreatic cancer surgeries have been shown to have improved outcomes such as length of stay. We examined how centralization of pancreatic cancer care within a regional integrated healthcare system improves overall survival. METHODS: We conducted a retrospective study of 1621 patients treated for pancreatic cancer from February 2010 to December 2018. Care was consolidated into 4 Centers of Excellence (COE) in surgery, medical oncology, and other specialties. Descriptive statistics, bivariate analysis, Chi-square tests, and Kaplan-Meier analysis were performed. RESULTS: Neoadjuvant chemotherapy use rose from 10% to 31% (p < .001). The median overall survival (OS) improved by 3 months after centralization (p < .001), but this did not reach significance on multivariate analysis. CONCLUSIONS: Our results suggest that in a large integrated healthcare system, centralization improves overall survival and neoadjuvant therapy utilization for pancreatic cancer patients.

Pancreatic Neuroendocrine Tumor: Rationale for Centralization in an Integrated Health Care System.

OBJECTIVES: Given the complex surgical management and infrequency of pancreatic neuroendocrine tumor, we hypothesized that treatment at a center of excellence improves survival. METHODS: Retrospective review identified 354 patients with pancreatic neuroendocrine tumor treated between 2010 and 2018. Four hepatopancreatobiliary centers of excellence were created from 21 hospitals throughout Northern California. Univariate and multivariate analyses were performed. The χ2 test of clinicopathologic factors determined which were predictive for overall survival (OS). RESULTS: Localized disease was seen in 51% of patients, and metastatic disease was seen in 32% of patients with mean OS of 93 and 37 months, respectively (P < 0.001). On multivariate survival analysis, stage, tumor location, and surgical resection were significant for OS (P < 0.001). All stage OS for patients treated at designated centers was 80 and 60 months for noncenters (P < 0.001). Surgery was more common across stages at the centers of excellence versus noncenters at 70% and 40%, respectively (P < 0.001). CONCLUSIONS: Pancreatic neuroendocrine tumors are indolent but have malignant potential at any size with management often requiring complex surgeries. We showed survival was improved for patients treated at a center of excellence, where surgery was more frequently utilized.