The Effects of Perioperative Corticosteroids on Postoperative Complications After Pancreatoduodenectomy: A Debated Topic of Systematic Review and Meta-analysis.

BACKGROUND

The intraoperative administration of corticosteroids has been shown to improve postoperative outcomes in patients undergoing surgery; however, the impact of corticosteroids on complications following pancreatoduodenectomy (PD) remains controversial.

OBJECTIVE

This study aimed to evaluate the efficacy of perioperative corticosteroids on postoperative complications after PD.

MATERIALS AND

METHODS

A comprehensive search was conducted using the PubMed, Embase, and Web of Science databases for studies published prior to 1 July 2024. Of 7418 articles identified, a total of 5 studies were eligible for inclusion in this meta-analysis. The primary outcome was incidence of postoperative major complications (PMCs), while the additional outcomes were incidences of postoperative pancreatic fistulas (POPFs), infection, delayed gastric emptying (DGE), post-pancreatectomy hemorrhage (PPH), bile leakage, reoperation, and 30-day mortality. The study was registered in the PROSPERO database (CRD42024524936).

RESULTS

Finally, 5 studies involving 1449 patients (537 with corticosteroids and 912 without corticosteroids) were analyzed. Intraoperative corticosteroids were not associated with any improvement in PMCs (p = 0.41). The incidence of POPF (p = 0.12), infectious complications (p = 0.15), or DGE (p = 0.81) were not significantly different between the two groups. No obvious differences were found in the incidence of PPH (p = 0.42), bile leakage (p = 0.68), 30-day mortality (p = 0.99), or reoperation (p = 0.26).

CONCLUSION

Perioperative corticosteroids did not significantly demonstrate any protective advantage in terms of postoperative complications after PD. This finding may serve as a reference for the perioperative use of corticosteroids in pancreatic surgery. Well-designed clinical trials are warranted in the near future in order to provide high-level evidence.