Robotic approach mitigates the effect of major complications on survival after pancreaticoduodenectomy for periampullary cancer

Abstract

Background: Major complications (MC) after pancreaticoduodenectomy (PD) have been shown to be an independent predictor of worse oncologic outcomes. There is limited data on the effect of major complications on long-term outcomes after robotic PD (RPD). The aim of this current study is to compare the effect of MC on overall and disease-free survival (DFS) after RPD and open PD (OPD). Methods: This is a single-center, retrospective review of a prospectively maintained database of all patients undergoing PD for periampullary cancer including duodenal adenocarcinoma, ampullary carcinoma and distal cholangiocarcinoma from January 1, 2011 through November 30, 2017. Univariate analysis was performed on all clinical, pathologic and treatment factors. MC were defined as Clavien-Dindo grade 3 or higher complications. Kaplan-Meier survival analysis was performed with log-rank test for group comparison. Multivariate cox regression analysis was used to identify factors associated with overall survival (OS) in both the OPD and RPD groups. Results: A total of 190 patients with ampullary carcinoma (n = 98), cholangiocarcinoma (n = 55), duodenal adenocarcinoma (n = 37) were examined over the study period with 61.1% (n = 116) undergoing RPD and 38.9% (n = 74) undergoing OPD. Patient demographics were similar between the RPD and OPD cohorts regarding patient age (67.3 ± 10.3 vs 69.8 ± 10.2 yrs, p = 0.11), body-mass index (27.7±5.7 vs 29.5±7.1, p=0.06) and age-adjusted Charlson Comorbidity Index (aCCI) (5.0 ± 1.6 vs 5.2 ± 1.5, p = 0.33). R0 resection rates were similar between the RPD (94.0%, n = 109) and OPD cohorts (93.2%, n = 69, p = 0.84). Tumor size (2.8±1.8cm vs 2.9±1.8cm, p = 0.70) and lymph node involvement (lymph node positivity = 67.3% vs 66.2%, p = 0.88) were also similar between the two cohorts. OPD had higher rate of MC (40.5% vs 28.3% in RPD, p = 0.011) including clinically-significant pancreatic fistula (25.7% vs 8.6%, p = 0.001) and wound infection (34.5% vs 13.8%, p<0.001). Completion of adjuvant chemotherapy was higher in the RPD cohort (46.2% vs 28.6%, p = 0.048). OS and DFS for each cohort by MC status are depicted in Figure 1. Factors associated with OS on multivariate analysis in the OPD cohort included postoperative MC (hazard ratio (HR) = 2.18, 95%CI 1.04-4.55, p = 0.038), aCCI (HR = 1.50, 95%CI 1.16-1.93, p = 0.002), tumor grade (HR = 2.24, 95%CI 1.08-4.62, p = 0.03), tumor perineural invasion (HR=4.84, 95%CI 1.84-12.73, p = 0.001) and number of adjuvant therapy cycles (HR=0.84, 95%CI 0.72-.98, p = 0.03). Factors associated with OS on multivariate analysis in RPD cohort included T stage (HR = 1.58, 95%CI 1.10-2.28, p = 0.014), lymph node involvement (HR = 1.10. 95%CI 1.03-1.17, p = 0.004), lymphovascular invasion (HR = 2.77, 95%CI 1.08-7.11, p = 0.035), transfusion requirement (H = 2.62, 95%CI 1.34-5.13, p=0.005), reoperation (HR = 3.23, 95%CI 1.18-8.81, p=0.022), and completion of adjuvant therapy (HR = 0.34, 95%CI 0.17-0.68, p = 0.002). MC was not associated with OS in the RPD cohort (HR = 1.55, 95%CI 0.87-2.76, p=0.14). Conclusion: MC are associated with worse patient outcomes after OPD but not after RPD. Robotic approach abrogates the negative effects of MC on patient outcomes after PD for malignancy and is associated with improved adjuvant chemotherapy completion rates.