Pancreaticoduodenectomy (PD) is a major intervention in digestive surgery. Although its mortality is currently low in experienced centers, morbidity remains high, dominated by a pancreatic fistula.
The aim of this study was to analyze the risk factors for postoperative pancreatic fistula (POPF) after PD.
A retrospective study was conducted at the General Surgery Department of Habib Thameur University Hospital in Tunis for 12 years (2010–2021). All patients who underwent PD were included regardless of indications.
Our series comprised 50 patients, consisting of 27 men and 23 women. The rate of a pancreatic fistula was 32% (16 patients) with an average time of onset of 5 days (1–12 days). It was observed as a biochemical leak (grade A) in 1 patient (2%), pancreatic fistula grade B in 5 patients (10%), and pancreatic fistula grade C in 10 patients (20%). Pancreatic fistula was responsible for 10% of postoperative mortality (five patients). Univariate analysis showed a statistically significant correlation between POPF and the following factors: diameter of the main pancreatic duct ≤3 mm (p=0.036, p<0.05), soft texture of the pancreas (p=0.025, p<0.05), pancreaticojejunostomy by two semi-overlapping sutures (p=0.049, p<0.05), and fasting blood glucose level ≤8 mmol/l (p=0.025, p<0.05). Multivariate analysis showed that soft pancreatic texture was the only independent risk factor for POPF (p=0.02, p<0.05).
The soft texture of the pancreas is the only independent risk factor for POPF. Prospective randomized studies are still needed to accurately determine the true risk factors for a pancreatic fistula after PD.
The high morbidity and mortality rates of pancreaticoduodenectomy are mainly associated with pancreaticojejunal anastomosis, the most fragile and susceptible to complications such as clinically relevant postoperative pancreatic fistula.
The alternative fistula risk score and the first postoperative day drain fluid amylase are predictors of the occurrence of clinically relevant postoperative pancreatic fistula. No consensus has been reached on which of the scores is a better predictor; moreover, their combined predictive power remains unclear. To the best of our knowledge, this association had not yet been studied.
This study assessed the predictive effect of alternative fistula risk score and/or drain fluid amylase on clinically relevant postoperative pancreatic fistula in a retrospective cohort of 58 patients following pancreaticoduodenectomy. The Shapiro-Wilk and Mann-Whitney tests were applied for assessing the distribution of the samples and for comparing the medians, respectively. The receiver operating characteristics curve and the confusion matrix were used to analyze the predictive models.
The alternative fistula risk score values were not statistically different between patients in the clinically relevant postoperative pancreatic fistula and non- clinically relevant postoperative pancreatic fistula groups (Mann-Whitney U test 59.5, p=0.12). The drain fluid amylase values were statistically different between clinically relevant postoperative pancreatic fistula and non- clinically relevant postoperative pancreatic fistula groups (Mann-Whitney U test 27, p=0.004). The alternative fistula risk score and drain fluid amylase were independently less predictive for clinically relevant postoperative pancreatic fistula, compared to combined alternative fistula risk score + drain fluid amylase.
The combined model involving alternative fistula risk score >20% + drain fluid amylase=5,000 U/L was the most effective predictor of clinically relevant postoperative pancreatic fistula occurrence following pancreaticoduodenectomy.
The value of drain amylase on the first postoperative day after pancreatic resections has been described as an efficient predictor of pancreatic fistula. In spite of this, the cut-off point below which the drains can be removed early remains controversial.
Validate the use of the amylase on the 1st postoperative day in the correlation with pancreatic fistula and define the value at which early drain removal is safe.
Were included patients undergoing Whipple surgery in the period of 2007 to 2016. Group 1 enrolled the ones who did not develop fistula and those who developed biochemical fistula for less than seven days postoperatively and group 2 included patients who developed persistent biochemical fistula between seven and 21 days and those with grade B and C fistula.
Sixty-one patients were included, 41 comprised group 1 and 20 group 2. The incidence of abdominal collections, need for reoperation and time of hospitalization were for group 1 and 2, respectively: 17.1%, 17.1% and 9.5 days, and 65%, 40% and 21.1 days. The median of the amylase from the drain at 1st postoperative day was in group 1 and 2, respectively: 175 U/l and 3172.5 U/l (p=0.001). Using a cut-off of 180 to predict the group to which the patient would belong there was obtained sensitivity, specificity, positive predictive value and negative predictive value of 100%, 48.8%, 50% and 100% respectively.
It was validated the cut-off value of 180 U/l as appropriate to early drain removal.
Developed by Surya MKT