Enhanced Recovery After Minimally Invasive Pancreaticoduodenectomy

Pancreaticoduodenectomy (PD) is the standard of care for patients with malignant or benign disease of the pancreatic head or peri-ampullary region. The postoperative course after PD is strongly dependent of the occurrence of pancreatic fistula (POPF) and/or delayed gastric emptying (DGE). In a recent multicentre randomized controlled trial, the investigators have shown pancreaticogastrostomy (PG; without a stent in the pancreatic duct) to be associated with 8% POPF rate, significantly lower than pancreaticojejunostomy (20%) (1). Since then, PG reconstruction is considered the standard of care in PD, which is also underlined in more recent systematic reviews. In patients without POPF after PD, the length of hospital stay is determined by the occurrence of DGE, which is poorly understood and currently lacks any effective treatment. Patients who developed DGE after PD with PG anastomosis (n=18; 20%) had a significantly (p=0.014) longer (mean + sem) length of hospital stay (LOS) of 26.3 + 1.58 days, as compared to 22.4 + 1.27 days for patients without DGE (n=69). These figures are observed in the investigators' center as part of the multicentre RCT. Enhanced recovery after surgery (ERAS) or fast-track (FT) programs are able to reduce postoperative length of hospital stay (LOS). Indeed recently, ERAS or FT programs have been implemented successfully in PD (2). Patients were discharged 4 days earlier in the ERAS group, without a negative effect on the clinical outcome. Still, many surgeons are reluctant to implement ERAS programs because they fear compromising patient safety. In efforts to improve the outcomes of PD, many surgical techniques have been evaluated to restore the pancreatic digestive continuity after PD. However, the best way to ensure this and whether or not to perform the procedure via standard open or minimally invasive, i.e. 2- or 3-dimensional laparoscopic (3D-LPD) or 3-dimensional robotic surgery (RPD), is still under debate. The investigators have passed the learning curve of 50 3D-LPD and hypothesize the implementation of ERAS and EEN in 3D-LPD can improve short-term outcomes.