Objective To evaluate the effect of vascular resection (VR), including portal vein resection (PVR) and hepatic artery resection (HAR), on short- and long-term outcomes in patients with perihilar cholangiocarcinoma (PHC). Background Resection surgery and transplantation are the main treatment methods for PHC that provide a chance of long-term survival. However, the efficacy and safety of VR, including PVR and HAR, for treating PHC remain controversial. Methods This study was registered at the International Prospective Register of Systematic Reviews (CRD42020223330). The EMBASE, PubMed, and Cochrane Library databases were used to search for eligible studies published through November 28, 2020. Studies comparing short- and long-term outcomes between patients who underwent hepatectomy with or without PVR and/or HAR were included. Random- and fixed-effects models were applied to assess the outcomes, including morbidity, mortality, and R0 resection rate, as well as the impact of PVR and HAR on long-term survival. Results Twenty-two studies including 4,091 patients were deemed eligible and included in this study. The meta-analysis showed that PVR did not increase the postoperative morbidity rate (odds ratio (OR): 1.03, 95% confidenceinterval (CI): [0.74-1.42], P = 0.88) and slightly increased the postoperative mortality rate (OR: 1.61, 95% CI [1.02-2.54], P = 0.04). HAR did not increase the postoperative morbidity rate (OR: 1.32, 95% CI [0.83-2.11], P = 0.24) and significantly increased the postoperative mortality rate (OR: 4.20, 95% CI [1.88-9.39], P = 0.0005). Neither PVR nor HAR improved the R0 resection rate (OR: 0.70, 95% CI [0.47-1.03], P = 0.07; OR: 0.77, 95% CI [0.37-1.61], P = 0.49, respectively) or long-term survival (OR: 0.52, 95% CI [0.35-0.76], P = 0.0008; OR: 0.43, 95% CI [0.32-0.57], P Conclusions PVR is relatively safe and might benefit certain patients with advanced PHC in terms of long-term survival, but it is not routinely recommended. HAR results in a higher mortality rate and lower overall survival rate, with no proven benefit.