Published in

Oxford University Press, British Journal of Surgery, 12(106), p. 1657-1665, 2019

DOI: 10.1002/bjs.11292

Links

Tools

Export citation

Search in Google Scholar

Multicentre observational cohort study of implementation and outcomes of laparoscopic distal pancreatectomy

This paper is made freely available by the publisher.
This paper is made freely available by the publisher.

Full text: Download

Green circle
Preprint: archiving allowed
Orange circle
Postprint: archiving restricted
Red circle
Published version: archiving forbidden
Data provided by SHERPA/RoMEO

Abstract

Abstract Background Laparoscopic distal pancreatectomy (LDP) is increasingly being performed as an alternative to open surgery. Whether the implementation and corresponding learning curve of LDP have an impact on patient outcome is unknown. The aim was to investigate the temporal trends in practice across UK centres. Methods This was a retrospective multicentre observational cohort study of LDP in 11 tertiary referral centres in the UK between 2006 and 2016. The learning curve was analysed by pooling data for the first 15 consecutive patients who had LDP and examining trends in surgical outcomes in subsequent patients. Results In total, 570 patients underwent LDP, whereas 888 underwent open resection. For LDP the median duration of operation was 240 min, with 200 ml blood loss. The conversion rate was 12·1 per cent. Neuroendocrine tumours (26·7 per cent) and mucinous cystic neoplasms (19·7 per cent) were commonest indications. The proportion of LDPs increased from 24·4 per cent in 2006–2009 (P1) to 46·0 per cent in 2014–2016 (P3) (P < 0·001). LDP was increasingly performed for patients aged 70 years or more (16 per cent in P1 versus 34·4 per cent in P3; P = 0·002), pancreatic ductal adenocarcinoma (6 versus 19·1 per cent; P = 0·005) and advanced malignant tumours (27 versus 52 per cent; P = 0·016). With increasing experience, there was a trend for a decrease in blood transfusion rate (14·1 per cent for procedures 1–15 to 3·5 per cent for procedures 46–75; P = 0·008), ICU admissions (32·7 to 19·2 per cent; P = 0·021) and median duration of hospital stay (7 (i.q.r. 5–9) to 6 (4–7) days; P = 0·002). After 30 procedures, a decrease was noted in rates of both overall morbidity (57·7 versus 42·2 per cent for procedures 16–30 versus 46–75 respectively; P = 0·009) and severe morbidity (18·8 versus 9·7 per cent; P = 0·031). Conclusion LDP has increased as a treatment option for lesions of the distal pancreas as indications for the procedure have expanded. Perioperative outcomes improved with the number of procedures performed.