[Complications, hospital mortality and survival following partial pancreaticoduodenectomy]. / Complicaties, ziekenhuissterfte en overleving na partiële pancreaticoduodenectomie.

OBJECTIVE: To analyse hospital mortality, complications and survival of patients after subtotal pancreaticoduodenectomy in the Academic Medical Centre, Amsterdam, the Netherlands, 1983-1996. DESIGN: Partly retrospective (1983-August 1987), partly prospective (September 1987-1996). METHOD: Patient characteristics, indication for surgery, postoperative complications, mortality and survival of patients who underwent subtotal pancreaticoduodenectomy were recorded in a computer database. Patients were subdivided into three groups (1983-September 1992; October 1992-1994; 1995-September 1996) to analyse the influence of change in surgical technique and the increase of experience. RESULTS: From 1983-to September 1996, 312 consecutive patients underwent a subtotal pancreaticoduodenectomy. Hospital mortality decreased from 4.9% to 1.4% in the last period (1995-1996). The complication rate decreased from 60% to 41%. The hospital stay decreased from median 24 days to 16 days. The actualized 5-year survival analysed for patients operated from 1983-to September 1992 was 31%. Patients with ampullary tumours had a 5-year survival of 50%. The 5-year survival of patients with bile duct and pancreatic carcinoma was 24% and 15% respectively. CONCLUSIONS: Subtotal pancreaticoduodenectomy can be performed safely with a low mortality (< 2%) in specialised centres. The morbidity is still substantial (40%). The survival is mainly dependent on type of tumour and patient selection and is approximately 50% for patients with ampullary tumours. The pylorus preserving procedure has become the standard operation.

[Hospital experience and hospital mortality following partial pancreaticoduodenectomy in The Netherlands]. / Ziekenhuiservaring en ziekenhuissterfte na partiële pancreaticoduodenectomie in Nederland.

OBJECTIVE: To analyse the effect of hospital experience on mortality after subtotal pancreaticoduodenectomy in the Netherlands. DESIGN: Retrospective evaluation. METHOD: Information on hospital mortality and pancreatic resection in 1994 and 1995 in the Netherlands was obtained from the National Medical Register. Subanalysis was carried out of surgical mortality by age and hospital experience. RESULTS: Approximately 50% of the pancreaticoduodenectomies in the Netherlands were performed in hospitals with limited experience (< 5 procedures per year). Hospital mortality was higher in small-volume hospitals than in hospitals with experience (> 25 procedures per year): in 1994 17.2 and 0% and in 1995 14.6 and 2.9%, respectively (p < 0.05). Mortality was higher in patients older than 70 years compared with patients younger than 55 (p < 0.05). CONCLUSION: There was a correlation between mortality after pancreaticoduodenectomy and hospital experience. Therefore these procedures should be performed in centres with experience.