Title: Comparison of Acute Kidney Injury between open and Laparoscopic pylorus-preserving pancreaticoduodenectomy: propensity score analysis

Backgroud and Aims : Recently, owing to advances in technology, the Laparoscopic pylorus-preserving pancreaticoduodenectomy (LPPPD) has been increasing rapidly compared with open pylorus-preserving pancreaticoduodenectomy (OPPPD). Although several studies have advocated the safety and efficacy of LPPPD compared to OPPPD, these studies focused not on the development of postoperative acute kidney injury (AKI) but on the surgical outcomes. 
Because of several potential risk factors for developing postoperative AKI during LPPPD, this study aimed to compare the prevalence and risk factors of AKI between LPPPD and OPPPD.

Methods: We retrospectively collected and analyzed the data of 457 patients who underwent PPPD surgery between February 2012 to November 2015 at Asan Medical Center. Patients were divided into 2 groups according to the surgery procedure [OPPPD group (n=355) vs LPPPD group (n=102)]. Risk factors for AKI defined using the Kidney Disease: Improving Global Outcomes criteria were investigated using multivariate logistic regression analysis. Propensity score (PS) matching analysis and inverse probability of treatment weighting (IPTW) were used to compare outcomes between 2 groups to reduce the potential for bias. 

Results: After a 1:1 propensity score matching, 100 patients were included in each group. The LPPPD group showed lower incidence of AKI than OPPPD group without statistical significance (1.0% vs. 4.8%, p=0.116). The hospital stay was significantly prolonged in the OPPPD group compared with LPPPD group (p< 0.001). Even after PS & IPTW analyses, hospital stay was significantly prolonged in the OPPPD group than LPPPD group (p=0.018 for PS and p=0.001 for IPTW), and no significant difference was showed in the development of postoperative AKI (p=0.571 for PS and p=0.252 for IPTW). In multivariate analysis, LPPPD was not a independent risk factor for AKI (p=0.327). Older age and more albumin infusion during operation were independent predictors of the occurrence of postoperative AKI(p< 0.05 for all).

Conclusions: This study demonstrated that the significant difference in the incidence of postoperative AKI was not showed between two groups. Futhermore, LPPPD was not a predictor for the development of postoperative AKI.