Izbicki 1998.

Methods	Randomised controlled trial
Participants	Country: Germany Number randomised: 64 Post‐randomisation drop‐outs: 3 (4.7%) Revised sample size: 61 Mean age (years): 44 Women: 10 (16.4%) Follow‐up period (years): 15 Number of study centres: 1 Inclusion criteria Inclusion criteria were an inflammatory mass in the head of the pancreas (> 35 mm in diameter) Severe recurrent pain attacks (at least 1 per month requiring opiates) History of pain attacks for at least 1 year Co‐existing complications from adjacent organs (e.g. common bile duct stenosis, duodenal stenosis) Exclusion criteria Chronic pancreatitis without involvement of the pancreatic head Small duct disease (maximal diameter of duct Wirsung = 3 mm) Pseudocysts without duct pathology Portal vein thrombosis Myocardial infarction within 6 months Detection of a malignant pancreatic tumour Co‐existing malignancy of other organs
Interventions	Participants were randomly assigned to 1 of 2 groups Group 1: duodenum‐preserving pancreatoduodenectomy (n = 31) (32 participants were randomised) Further details: Frey procedure Group 2: pancreatoduodenectomy (n = 30) (32 participants were randomised) Further details: PPPD
Outcomes	Mortality, post‐operative complications, quality of life, proportion of people employed, diabetes, and exocrine insufficiency
Notes	Reasons for post‐randomisation drop‐outs: pancreatic carcinoma found on frozen section biopsy
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Randomization was performed using a list of random digits that were made available during surgery as coded cards sealed in envelopes"
Allocation concealment (selection bias)	Low risk	Quote: "Randomization was performed using a list of random digits that were made available during surgery as coded cards sealed in envelopes"
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Comment: information on participant blinding was not available. It is impossible to blind surgeons who perform the procedure
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Comment: quality of life and pain score data were recorded by doctoral students who were unaware of group allocation. It was not clear whether the remaining outcomes were assessed by blinded outcome assessors
Incomplete outcome data (attrition bias) All outcomes	Low risk	Comment: no participants were lost to follow‐up, although some participants were excluded from analysis (please see other bias)
Selective reporting (reporting bias)	Low risk	Comment: all important outcomes were reported
Other bias	High risk	Comment: participants were excluded because they were found to have pancreatic cancer intraoperatively. This can introduce bias since the intervention (duodenum‐preserving pancreatoduodenectomy) is a less invasive procedure than control (pancreatoduodenectomy)