CT scan for early diagnosis of anastomotic leak after colorectal surgery: is rectal contrast useful?

BACKGROUND: Computed tomography (CT) scan with rectal contrast enema (RCE-CT) could increase the detection rate of anastomotic leaks (AL) in the early postoperative period following colorectal surgery, compared to CT scan without RCE. The aim of this study was to assess the benefit of RCE-CT for the early diagnosis of AL following colorectal surgery. METHODS: Patients who had a RCE-CT for suspected AL in the early postoperative period following colorectal surgery with anastomosis between January 2012 and July 2019 at the Dijon University Hospital were retrospectively included. All images were reviewed by two independent observers who were blinded to the original report. The reviewers reported for each patient whether an AL was present or not in each imaging modality (CT scan, then RCE-CT). Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were then calculated to determine the diagnostic performance of each modality. RESULTS: One hundred and thirty-nine patients were included. RCE-CT had an increased NPV compared to CT scan (82% vs 77% (p = 0.02) and 84% vs 68% (p < 0.0001) for observers 1 and 2, respectively). RCE-CT had an increased sensitivity compared to CT scan (79% vs 48% (p < 0.0001) for observer 2). RCE-CT had a significant lower false-negative rate for both observers: 18% vs 23% (p = 0.02) and 16% vs 32% (p < 0.0001). CONCLUSIONS: RCE-CT improved the detection rates of AL in the early period following colorectal surgery. RCE-CT should be recommended when a CT scan is negative and AL is still suspected.

Prognosis of poorly cohesive gastric cancer after complete cytoreductive surgery with or without hyperthermic intraperitoneal chemotherapy (CYTO-CHIP study).

BACKGROUND: The incidence of gastric poorly cohesive carcinoma (PCC) is increasing. The prognosis for patients with peritoneal metastases remains poor and the role of cytoreductive surgery (CRS) and hyperthermic intraperitoneal chemotherapy (HIPEC) is controversial. The aim was to clarify the impact of gastric PCC with peritoneal metastases treated by CRS with or without HIPEC. METHODS: All patients with peritoneal metastases from gastric cancer treated with CRS with or without HIPEC, in 19 French centres, between 1989 and 2014, were identified from institutional databases. Clinicopathological characteristics and outcomes were compared between PCC and non-PCC subtypes, and the possible benefit of HIPEC was assessed. RESULTS: In total, 277 patients were included (188 PCC, 89 non-PCC). HIPEC was performed in 180 of 277 patients (65 per cent), including 124 of 188 with PCC (66 per cent). Median overall survival (OS) was 14.7 (95 per cent c.i. 12.7 to 17.3) months in the PCC group versus 21.2 (14.7 to 36.4) months in the non-PCC group (P < 0.001). In multivariable analyses, PCC (hazard ratio (HR) 1.51, 95 per cent c.i. 1.01 to 2.25; P = 0.044) was associated with poorer OS, as were pN3, Peritoneal Cancer Index (PCI), and resection with a completeness of cytoreduction score of 1, whereas HIPEC was associated with improved OS (HR 0.52; P < 0.001). The benefit of CRS-HIPEC over CRS alone was consistent, irrespective of histology, with a median OS of 16.7 versus 11.3 months (HR 0.60, 0.39 to 0.92; P = 0.018) in the PCC group, and 34.5 versus 14.3 months (HR 0.43, 0.25 to 0.75; P = 0.003) in the non-PCC group. Non-PCC and HIPEC were independently associated with improved recurrence-free survival and fewer peritoneal recurrences. In patients who underwent HIPEC, PCI values of below 7 and less than 13 were predictive of OS in PCC and non-PCC populations respectively. CONCLUSION: In selected patients, CRS-HIPEC offers acceptable outcomes among those with gastric PCC and long survival for patients without PCC.

Preoperative inflammation is an independent factor of worse prognosis after colorectal cancer surgery.

BACKGROUND: We know that inflammation is related to colorectal cancer prognosis and to the onset of postoperative infections. OBJECTIVE: This study aimed to understand the relationship between preoperative inflammation and the prognosis of colorectal cancer and to elucidate whether the impact of inflammation on cancer prognosis was related to an increased risk of surgical infection or was independent of it. METHODS: Patients who underwent elective colorectal cancer surgery between November 2011 and April 2014 were included in a prospective database (IMACORS). Preoperative c reactive protein was collected for each patient. Patients were followed up according to the French national guidelines. A cut-off of preoperative CRP of 5mg/L was chosen. Clinical characteristics were compared according to CRP using Chi2 and Mann-Whitney tests. The Overall Survival (OS) and Disease-Free-Survival (DFS) were compared by Kaplan-Meier curves. A Cox proportional hazards regression model was applied to perform a multivariate analysis of OS and DFS's predictors. RESULTS: A total of 254 patients were included. The median age was 68 years old. The median follow up was 41.8 months. The overall median preoperative CRP was 5mg/L. Preoperative CRP was significantly associated with N status; CRP being significantly higher among patients with colonic cancer and with patients who didn't receive a neoadjuvant treatment. Multivariate analyse revealed that preoperative CRP is an independent prognostic factor of OS and DFS respectively (HR=2.34 (1.26-4.31), P=0.006 and HR=1.83 (1.15-2.90), P=0.01). CONCLUSION: Preoperative inflammation measured by CRP is independently related with overall and disease-free survival of colorectal cancer.

What results can be expected one year after complex incisional hernia repair with biosynthetic mesh?

BACKGROUND: Incisional hernia is a frequent complication after midline laparotomy. The current standard repair includes the use of a synthetic mesh to prevent recurrence. However, the use of a synthetic mesh in a contaminated field carries a higher risk of mesh infection. In this setting biologic and biosynthetic meshes can be used as they resist to infection, but these are absorbable meshes. This raises the question of the risk of recurrence as the mesh disappears. Phasix® is a biosynthetic mesh getting absorbed in 12-18 months. The aim of this study was to assess the 1-year recurrence rate after abdominal-wall repair with a Phasix® mesh. METHODS: All patients undergoing ventral hernia repair between 2016 and 2018 at the University Hospital of Dijon using a Phasix® mesh were prospectively included in a database. They were all followed-up with a physical exam and a routine CT scan at one year. All postoperative complications were recorded. RESULTS: Twenty-nine patients were included in the study (55.2% women), with a mean BMI of 30,25 kg/m2. Nineteen meshes were sublay and 10 intraperitoneal. Complications at 1 month were mainly mild: Clavien-Dindo I and II (61.1%). No mesh was explanted. There was no chronic infection. The mean length of stay was 11.5 days. The 1-year recurrence rate was 10.3%. CONCLUSION: Patients having undergone complex ventral hernia repair with a Phasix® mesh have a 1-year recurrence rate of 10.3%. No severe surgical site occurrence was detected. A longer follow-up in a larger number of patients could confirm the place of this mesh in abdominal-wall repair.

Inflammatory markers for the early detection of post-operative infection: The same threshold for rectal and colic surgery?

INTRODUCTION: Anastomotic fistula is the most fearsome complication following colorectal surgery. Numerous studies have demonstrated the interest of postoperative CRP assay as an early diagnostic marker. Must the critical threshold for biological inflammatory markers remain the same, whether resection be colic or rectal? PATIENTS AND METHOD: This is a study based on a cohort constituted between 2011 and 2014, including 497 patients with planned colorectal resection. C-reactive protein and pro-calcitonin were measured daily from day before surgery to D4. All postoperative intra-abdominal complications were considered as an anastomotic fistula. Detection thresholds were calculated from the area under the ROC curve. RESULTS: An intra-abdominal septic complication occurred in 16.9% of the patients having undergone rectal resection vs. 9.9% of those having had colectomy (P=0.03). In the absence of complications there was no significant difference between the two groups in terms of postoperative inflammatory response as determined by either CRP or PCT assay. Following rectal resection, optimal area under the curve (AUC=0.87) corresponds to CRP on D4 for a threshold of 100mg/L: sensitivity 83.3%, NPV 95.3%. For colons with the same CRP at 100mg/L (AUC=0.71): sensitivity 63.6%, NPV 93.9%. CONCLUSION: Notwithstanding riskier surgery, the detection threshold for an anastomotic fistula following rectal surgery remains the same: CRP>100mg/L at D4.

SARS-CoV-2 vs. SARS-CoV-1 management: antibiotics and inflammasome modulators potential.

The coronavirus SARS-CoV-2 at the origin of COVID-19 shares more than 70% genetic similarity with SARS-CoV-1 that was at the origin of 2003 SARS. Infection-associated symptoms are very similar between SARS and COVID-19 diseases and are the same as community-acquired pneumonia symptoms. Antibiotics were empirically given to SARS patients in the early stages of the pathology whereas a different strategy has been decided in the management of COVID-19 pandemic with a worldwide shutdown. The cytokine storm, both identified in SARS and COVID-19 severe cases, is generated through inflammasome activation, which opens therapeutic perspectives to counteract the pathogenic inflammation. As corticoids have numerous side effects that limit their use, focusing on anti-inflammasome agents could represent a safer alternative for patients with severe COVID-19.

Chyloperitoneum secondary to internal herniation complicating omega-loop gastric bypass.

Chyloperitoneum is a potential complication of omega-loop gastric bypass caused by internal herniation. Chronic compression of the mesentery leads to chylous extravasation that can mimic peritonitis.

A marine-sourced fucoidan solution inhibits Toll-like-receptor-3-induced cytokine release by human bronchial epithelial cells.

Fucoidans are sulfated polysaccharides from brown algae, known to have immunomodulatory activity. Their effects on the response of airway epithelial cells to Toll-like receptor 3 (TLR3) stimulation have not been characterized. Our objective was to evaluate the effects of a marine-sourced fucoidan solution (MFS) on the TLR3-induced expression and/or production of cytokines and prostaglandin by human primary bronchial epithelial cells as a model of the airway epithelium. The cells were incubated with MFS in the presence or absence of Poly(I:C) (a TLR3 agonist that mimics viral RNA). Cytokine expression and production were assessed using RT-qPCR and ELISA. The expression of cyclooxygenase-2 and the production of prostaglandin E2 were also measured. Relative to control, exposure to MFS was associated with lower Poly(I:C)-induced mRNA expression of various cytokines and chemokines, and lower COX-2 production. The MFS inhibited the production of some cytokines (IL-1α, IL-1ß, TNFα, and IL-6), chemokines (CCL5, CCL22, CXCL1, CXCL5 and CXCL8) and prostaglandin E2 but did not alter the production of IL-12/25, CCL2 and CCL20. At clinically relevant concentrations, the MFS inhibited the TLR3-mediated production of inflammatory mediators by human primary bronchial epithelial cells - suggesting that locally applied MFS might help to reduce airway inflammation in viral infections.

Intraperitoneal drains move.

INTRODUCTION: The use of surgical drains is the subject of much debate but they continue to be commonly used. The phenomenon of drain migration from their desired position following surgery has not been studied. The aim of this study was to evaluate the incidence of the displacement of surgical drains among patients undergoing abdominal gastrointestinal surgery. PATIENTS AND METHODS: We performed a review of all patients who underwent an early CT-scan postoperatively after abdominal gastrointestinal surgery prior to drain mobilization, between January 2013 and April 2016 in the Dijon University Hospital Center. Pre-and intra-operative data (number, type and position of drains) and postoperative data (imaging and evolution) were collected retrospectively. RESULTS: This study included 125 patients. Thirty-five (28%) were found to have a displacement of at least one drain from its original position. Forty-one (19.8%) of the 207 studied drains had moved. Postoperative morbidity was not higher in patients with displaced drains (P=0.51). None of all the studied preoperative and operative factors have been found to be a risk factor for drain displacement. CONCLUSION: Surgical drains displacement is frequently encountered in patients undergoing digestive abdominal surgery. In our experience, this phenomenon does not seem to have any clinical implications. When a benefit is expected from the use of surgical drains, intraperitoneal fixation appears to be necessary.

[Routine screening of splenic or portal vein thrombosis after splenectomy]. / Dépistage systématique d'une thrombose veineuse splénique ou portale après splénectomie.

BACKGROUND: Portal and/or splenic vein thrombosis (PSVT) is common after splenectomy. It can be a life-threatening complication, with a risk of bowel ischemia and portal hypertension. An early diagnosis allows an effective medical treatment and prevents life-threatening complications. There is no consensus regarding the benefit of systematic screening of patients after splenectomy for PSVT. We started in January 2012 a routine screening of PSVT after elective splenectomy. The aim of this study was to assess this policy. METHODS: Since January 2012, all patients undergoing an elective splenectomy had an abdominal CT-scan on postoperative-day 7. Demographic data, pathology, type of surgery, platelet counts before and after surgery, outcome, results of medical imaging, and management of PSVT and its results were recorded. RESULTS: Over 3 years, 52 patients underwent an elective splenectomy. All of them had a CT-scan at postoperative-day 7. A PSVT was found in 11 patients (21.2 %). They were all asymptomatic. Lymphoma and splenomegaly were the main factors associated with PSVT in the univariate analysis. All patients with PSVT were treated with anticoagulation and no complication of PSVT occurred. The follow-up CT confirmed the efficacy of anticoagulation therapy in all patients. CONCLUSIONS: Routine screening of PSVT after elective splenectomy is warranted because it allows to start anticoagulant therapy and avoid further life-threatening complications. The incidence of PSVT is particularly high among patients operated on for lymphoma or with splenomegaly.

Should CT scan be performed when CRP is elevated after colorectal surgery? Results from the inflammatory markers after colorectal surgery study.

INTRODUCTION: Serum concentration of C-reactive protein (CRP) that exceeds a pre-defined threshold between the 3rd and 5th postoperative day is a reliable marker of infectious complications after colorectal surgery. However, the optimal strategy to follow when a high CRP is found has not been defined. The aim of this study was to analyze the usefulness of computed tomography (CT) scan in this situation in a prospective cohort of patients following colorectal surgery. METHODS: Between November 2011 and April 2015, patients at two surgical centers who had undergone elective colorectal resection with anastomosis and who had a CRP>12.5mg/dL on the 4th postoperative day (POD) were prospectively included in a database. Data were collected concerning all complications occurring during the 30days following surgery, method of diagnosis, management and clinical course. The decision to perform a CT scan between POD 4 and POD 6 day was guided only by the elevation of CRP in the absence of any other clinical signs; results were analyzed to evaluate the diagnostic value of elevated CRP. Uni- and multivariable analyses were performed to identify risk factors for postoperative infection. RESULTS: The study included a total of 174 patients: 56 (32.1%) had a CT between POD 4 and 6, and this helped detect a postoperative complication in 55.4% of cases. Patients who did not undergo CT had lower CRP values (16.3 vs. 18.5, P=0.02). Among the 118 patients who did not undergo CT, 50.8% eventually developed an infectious complication. The sensitivity of CRP-guided CT was 76.7% with a negative predictive value of 78.8%. CONCLUSION: If an elevated CRP is found on POD 4, an abdominopelvic CT should be performed. A normal result does not formally eliminate the existence of intra-abdominal complication. A study protocol should be set up to evaluate whether systematic revisional surgery or repeat CT scan is the appropriate management if CRP in the next two days reveals persistent inflammation.

Algoplus performance to detect pain in depressed and/or demented old patients.

BACKGROUND: Algoplus detects acute pain in non-verbal old patients (NVOP) with good psychometric properties. However, depression or dementia might modify the Algoplus score and/or item expression. Algoplus performances on demented and/or depressed old populations were tested. METHODS: This multicentre cross-sectional study included patients ≥65 years old with or without pain assigned to depression, dementia, depression & dementia or control groups. Each group was subjected to the Numerical Rating Scale (NRS) and behavioural scales (Algoplus, Doloplus). Depression and/or dementia status was rated and confirmed by blinded experts. Algoplus psychometric properties tested were: discriminant validity, convergent validity, item analysis, sensitivity to change after pain treatment and threshold determination. RESULTS: The analysis included 171 patients (mean age 82.3 ± 6.3 years). Patients with and without pain in each group were comparable for age in all subgroups, except the older dementia subgroup. The mean Algoplus score was significantly higher for patients with than without pain, regardless of group assignment (Wilcoxon signed-rank test, p < 0.001). Algoplus and NRS or Doloplus had high convergent validity (respective Spearman correlation coefficients 0.79 and 0.87). The mean Algoplus score decreased significantly after starting pain management, regardless of group assignment. Some behaviours (i.e. "look") occurred more often in depressed patients, even those without pain. A threshold of 2 yielded respective sensitivity and specificity values of 95% and 96% for dementia patients, 62% and 79% for depressed patients, 96% and 71% for dementia & depressed patients, and 80% and 100% for controls. CONCLUSION: Algoplus accurately detected pain in depressed and/or dementia patients; and was sensitive to change after pain treatment. WHAT DOES THIS STUDY ADD?: Algoplus accurately detects pain in depressed and/or demented patients. A cut-off score of 2 accurately detects the need for pain management in these populations. Algoplus is sensitive to change after treating pain.

Professional risks when carrying out cytoreductive surgery for peritoneal malignancy with hyperthermic intraperitoneal chemotherapy (HIPEC): A French multicentric survey.

BACKGROUND: Over the last two decades, many surgical teams have developed programs to treat peritoneal carcinomatosis with extensive cytoreductive surgery and hyperthermic intraperitoneal chemotherapy (HIPEC). Currently, there are no specific recommendations for HIPEC procedures concerning environmental contamination risk management, personal protective equipment (PPE), or occupational health supervision. METHODS: A survey of the institutional practices among all French teams currently performing HIPEC procedures was carried out via the French network for the treatment of rare peritoneal malignancies (RENAPE). RESULTS: Thirty three surgical teams responded, 14 (42.4%) which reported more than 10 years of HIPEC experience. Some practices were widespread, such as using HIPEC machine approved by the European Community (100%), individualized or centralized smoke evacuation (81.8%), "open" abdominal coverage during perfusion (75.8%), and maintaining the same surgeon throughout the procedure (69.7%). Others were more heterogeneous, including laminar flow air circulation (54.5%) and the provision of safety protocols in the event of perfusate spills (51.5%). The use of specialized personal protective equipment is ubiquitous (93.9%) but widely variable between programs. CONCLUSION: Protocols regarding cytoreductive surgery/HIPEC and the associated professional risks in France lack standardization and should be established.

Delay before implanting a port-a-cath after removing the previous one because of infection.

AIMS: We had for objective to assess the risk of a new Port-a-Cath (PAC) infection after re-implantation when the first one had been removed because of infection in the previous month. METHODS: We conducted a retrospective bicentric study including all patients having undergone PAC removal because of infection, followed by re-implantation of a new one in the next month, between 2004 and 2012. RESULT: One hundred and forty-nine PAC removals were followed by re-implantation in the next month. The cause of removal was infection for 63 patients: 34 presented with bacteremia (13 PAC removals and simultaneous re-implantation, and 21 with delayed re-implantation for a mean 14days) and 29 presented with local signs of infection (20 PAC removals and simultaneous re-implantation, and with delayed re-implantation for a mean 13days). Two out of 13 PACs were re-infected in the bacteremia group, (15.4%) when the removal and re-implantation were performed simultaneously, and 1 out of 21 (4.8%) when re-implantation of the second PAC was delayed. The 3 re-infected PACs were successfully treated with antibiotics. CONCLUSION: The early re-implantation of a PAC after removal of the previous one because of infection is a relatively safe procedure. The risk of infection can be higher when PAC removal and re-implantation are performed simultaneously in patients presenting with bacteremia; nevertheless, all reinfections were successfully treated with antibiotics.

Abdominal binders after laparotomy: review of the literature and French survey of policies.

BACKGROUND AND AIM: The use of abdominal binders after laparotomy is a question of habit. Scientific evidence of their usefulness is limited. The aims of this work were to review the scientific literature and to depict the practices of French surgeons regarding the use of these devices. METHODS: A systematic review of the literature about the use of abdominal binders after laparotomy was conducted. In order to depict surgeons' habits, an anonymous questionnaire was sent to all surgical departments affiliated to the FRENCH network (Federation of Surgical Research) and their surgical contacts. They were all asked about their use of binders, the type of binders they ordered, the expected benefit, the cost and the need for a randomized trial in this field. RESULTS: Only four trials have been published regarding the use of abdominal binders after laparotomy, all with a small number of patients. Some authors suggested that wearing binders procured a benefit in terms of postoperative comfort, but no significant difference was found. One study also suggested an improvement in respiratory volumes. No study focused on incisional hernia. Regarding the survey of practices, 50 questionnaires were retained for the final analysis (one questionnaire per department of surgery). The use of this device is really very frequent in France (94 % of surgeons order them), a habit usually acquired during the training in surgery. The main expected benefit is the prevention of abdominal wall dehiscence (83 %), but also an improvement in patients' postoperative comfort and pain (66 %). Although some surgeons order an abdominal binder for all their patients, most use them in selected patients (according to the operation and the patients' characteristics). CONCLUSION: Abdominal binders are frequently ordered by French surgeons after laparotomy. The expected benefit is the prevention of abdominal-wall complications, even though no data actually support this practice. Binders might have a benefit in terms of postoperative pain relief, but this needs to be analyzed. A prospective randomized trial is warranted.

Isolated hepatic perfusion: principles and results.

INTRODUCTION: Isolated hepatic perfusion allows the delivery of high dose chemotherapy while decreasing extra-hepatic toxicity, and is used mainly for patients with surgically unresectable liver tumors. PRINCIPLES: Vascular exclusion of the liver is performed after obtaining satisfactory hemodynamic tolerance, occasionally after cavocaval shunt and/or porto-systemic shunt. Perfusion entry can be arterial and/or portal while the exit is portal or caval. The perfusion circuit can be open or closed, using a circulation pump with or without oxygenation. The chemotherapy regimens used are high dose melphalan (with or without TNF-alpha), oxaliplatin, cisplatin and mitomycin, sometimes associated with moderate hyperthermia. The duration of perfusion ranges between 30 and 90 minutes according to the different protocols used. A percutaneous technique with incomplete liver vascular exclusion is also possible. RESULTS: The larger series in the literature show a response rate (partial or complete stabilization) between 60 and 80%, with approximately 5% complete morphologic responses. Morbidity and mortality are 40 and 5%, respectively, including specific morbidity related to the perfusion procedure as well as to chemotherapy. CONCLUSION: Chemotherapy delivered through isolated hepatic perfusion is a new therapeutic alternative, still under development, and can be proposed to patients with surgically unresectable primary or secondary liver tumors within clinical trials. These results seem to be promising, but are still associated with a non-negligible morbidity rate.

C-reactive protein and procalcitonin for the early detection of anastomotic leakage after elective colorectal surgery: pilot study in 100 patients.

INTRODUCTION: Anastomotic leakage is the most important complication after colorectal surgery. Its prognosis depends on its early diagnosis. C-reactive protein (CRP) has already shown its usefulness for the early detection of anastomotic leaks. Procalcitonin (PCT) is widely used in intensive care units and is more expensive, but its usefulness in the postoperative period of digestive surgery is not well established. PATIENTS AND METHODS: Between May 2010 and June 2011, 100 patients undergoing elective colorectal surgery were prospectively included in a database. CRP and PCT were measured before surgery and daily until postoperative day 4. All intraabdominal infections were considered as anastomotic leaks, regardless of their clinical impact and their management. The kinetics of PCT and CRP were recorded, as well as their accuracy for the detection of anastomotic fistula. RESULTS: The incidence of fistula was 13% and the overall mortality rate was 2%. Both CRP and PCT were significantly higher in patients with leakage. Areas under the receiver-operating characteristics (ROC) for CRP were higher than those for PCT each day. The best accuracy was obtained for CRP on postoperative day 4 (areas under the ROC curve were 0.869 for CRP and 0.750 for PCT). CONCLUSION: Procalcitonin is neither earlier nor more accurate than CRP for the detection of anastomotic leakage after elective colorectal surgery.

Diagnosis of postoperative pancreatic fistula.

BACKGROUND: Pancreatic fistula (PF) is a major source of morbidity after pancreatectomy. The International Study Group on Pancreatic Fistula (ISGPF) defines postoperative fistula by an amylase concentration in the abdominal drain of more than three times the serum value on day 3 or more after surgery. However, this definition fails to identify some clinical fistulas. This study examined the association between lipase measured in abdominal drainage fluid and PF. METHODS: Amylase and lipase levels in the abdominal drain were measured 3 days after pancreatic resection. Grade B and C fistulas were classified as clinical fistulas, regardless of whether the measured amylase concentration was considered positive or negative. The PF group included patients with a clinical fistula and/or those with positive amylase according to the ISGPF definition. RESULTS: Sixty-five patients were included. The median level of lipase was higher in patients with positive amylase than in those with negative amylase: 12,176 versus 64 units/l (P < 0·001). The lipase level was 16,500 units/l in patients with a clinical fistula and 224 units/l in those without a clinical fistula (P = 0·001). Patients with a PF had a higher lipase concentration than those without: 7852 versus 64 units/l (P < 0·001). A lipase level higher than 500 units/l yielded a sensitivity of 88 per cent and a specificity of 75 per cent for PF. For clinical fistulas the sensitivity was 93 per cent and specificity 77 per cent when the threshold for lipase was 1000 units/l. CONCLUSION: Lipase concentration in the abdominal drain correlated with PF. A threshold of 1000 units/l yielded a high sensitivity and specificity for the diagnosis of clinical PF.