Pancreaticoduodenectomy: a comparison of superior approach with classical Whipple's technique

Meta-analysis of an artery-first approach versus standard pancreatoduodenectomy on perioperative outcomes and survival

“…age of the two groups was similar (both 65(5) years). The sex ratio (M : F), reported in 16 studies, was 459 : 306 (60·0 versus 40·0 per cent) in the artery‐first group and 388 : 311 (55·5 versus 44·5 per cent) in the standard pancreatoduodenectomy group.…”

“…There was no significant difference in duration of operation between the two groups ( P = 0·500) ( Table ). Mean intraoperative blood loss, reported in 13 studies (1261 patients), was significantly lower in the artery‐first group (mean difference –389 (95 per cent c.i. –542 to –237) ml; P < 0·001).…”

“…. Seventeen studies were suitable for inclusion, giving a total of 1472 patients for the meta‐analysis, of which 771 underwent an artery‐first pancreatoduodenectomy and 701 a standard pancreatoduodenectomy ( Table ). The mean(s.d.)…”

“…The posterior and anterior approaches were the most commonly described artery‐first approaches. Others included the mesenteric (inferior infracolic), medial uncinate, left posterior and superior approaches. None of the studies included patients based on a definition of borderline or locally advanced pancreatic cancer.…”

“…Because the majority of the included papers were retrospective cohort or case–control studies, the Downs and Black checklist revealed deficiencies across 15 studies in the subgroup analysis of internal validity: bias and confounding. In addition, nine studies failed to score for confounding adjustment and ten for loss to follow‐up . Only six studies were powered to detect a clinically important effect; the remainder were single‐centre studies with small standard pancreatoduodenectomy and artery‐first groups.…”