Pancreatoduodenectomy for periampullary cancer and biliary obstruction: impact of a pathway to avoid preoperative biliary drainage. (23rd February 2017)

• Record Type: Journal Article
• Authors: Prasad, Pooja; Steele, Yvonne; Marcon, Francesca; Faulkner, Thomas; Dasari, Bobby; Marudanayagam, Ravi; Sutcliffe, Robert P; Muiesan, Paolo; Mirza, Darius F; Isaac, John R; Roberts, Keith J
• Is Part Of: Lancet. Volume 389(2017)Supplement 1
• Page Start: S80
• Publication Date: 2017-02-23
• Subjects: Medicine -- Periodicals; Medicine -- Periodicals; Medicine; Medicine; Electronic journals; Periodicals; 610.5
• DOI: 10.1016/S0140-6736(17)30476-2
• Languages: English
• ISSNs: 0140-6736

Abstract

Abstract: Background: Randomised trials have shown that preoperative biliary drainage (PBD) causes more harm than a straight to surgery approach for patients with jaundice and periampullary malignancy. However, it remains standard practice in many centres for jaundiced patients to undergo PBD. The aim of this study was to review the impact of a pathway to avoid PBD before pancreatoduodenectomy on clinical outcomes. Methods: A pathway to avoid PBD was implemented at the start of the study period (August, 2015). A dedicated nurse specialist and surgeon visited each referring trust to raise awareness of the pathway. Inclusion criteria were patients with resectable periampullary cancers and jaundice; patients were selected for surgery without PBD if they had not undergone PBD before referral and bilirubin concentration did not exceed 450 μmol/L. Time from initial CT scan to referral to multidisciplinay team discussion to specialist clinic and surgery were recorded. Findings: Over 12 months, 61 and 32 patients underwent pancreatoduodenectomy with and without PBD, respectively. 58 patients in the PBD group (95%) had undergone PBD before referral. There was no difference in demographic data or tumour types between the two groups. The duration of key intervals from referral to surgery were all shorter in the no PBD group than in the PBD group (median total duration 16 days [IQR 8–39] vs 65 [9–181], p<0·0001). Significantly more patients underwent pancreatoduodenectomy (31/32 vs 46/61, p=0·009) and venous resection (10/31 vs 4/46, p=0·014) in the no PBD group than in the PBD group. The sensitivity of initial staging CT for correctly staging resectable locally advanced disease was 91% and 50% in the no PBD and PBD groups, respectively (p=0·042). The accuracy of the initial CT scan to define locally advanced resectable disease was worse in the PBD group, presumably related to the difference in time between CT and surgery. Furthermore, average costs of treatment between presentation and surgery were £3178 cheaper in the no PBD group. Interpretation: Early surgery to avoid PBD is possible within the National Health Service. By reducing the time to surgery more patients undergo potentially curative surgery and costs of treatment are reduced. An understanding about why surgery without PBD is not done routinely is needed, as is the development of strategies to support its more widespread practice. Funding: Pancreatic Cancer UK.