Comparison between robot-assisted middle pancreatectomy and robot-assisted distal pancreatectomy for benign or low-grade malignant tumours located in the neck of the pancreas: A propensity score matched study.

OBJECTIVE To compare the short-term and long-term outcomes between robot-assisted middle pancreatectomy (RMP) and robot-assisted distal pancreatectomy (RDP). METHODS From August 2010 to May 2018, more than 800 patients were diagnosed with benign or low-grade malignant pancreatic tumours. According to the inclusion criteria, a total of 164 patients were included in our study. Among them 123 patients received RMP and 41 patients received RDP with spleen-preserving. These patients were divided into two groups, and propensity score matching (PSM) was used to minimize bias. Short-term and long-term outcomes were recorded and analysed. RESULTS After 1:1 matching, 38 cases of RMP and RDP were recorded and the baseline data was balanced. In the RMP group, 14 patients (36.8%) were female. In the RDP group, 13 patients (34.2%) were female. RDP had advantages in short-term outcomes including overall morbidity (28.9% vs. 60.5%, p = 0.011), Severe morbidity (0 vs. 10.5%, p = 0.017), clinical-relevant postoperative pancreatic fistula (CR-POPF) rate (10.5% vs. 36.8%, p = 0.007) and postoperative length of hospital stay (16.5 ± 8.6 days vs. 23.6 ± 12.9 days, p = 0.006). The operative time was similar; however, there were less estimated blood loss (EBL) in RMP group (31.3 ± 33.0 ml vs. 62.3 ± 38.2 ml, p<0.001). Patients in these two groups had a similar result in preservation of the pancreatic function. There was no statistic difference in incidence rate of exocrine and endocrine insufficiency between two groups (2.6% vs. 7.9%, p = 0.608; 5.3% vs. 10.5%, p = 0.671). CONCLUSION For benign or low-grade malignant tumours located at the neck of the pancreas, RMP and RDP could both be considered. The morbidity especially CR-POPF rate after RDP would be lower, but RMP would be more helpful in reducing EBL. RMP should be chosen with caution as it did not show a huge advantage in preserving normal pancreatic function. Further prospective randomised studies should be designed to compare these two types of surgery.