01OverviewDefinition, clinical context and the essential points that orientate the chapter.
A pancreatic leak is an anatomical failure of the normal route by which digestive secretions reach the duodenum. Fluid may enter the peritoneal cavity, track towards the chest, collect locally or drain through the skin. The visible manifestation can be far from the original pancreatic injury and may persist after the acute inflammatory illness appears to have improved. Recognising the source explains why treating recurrent ascites as routine portal-hypertensive fluid, or repeatedly aspirating a collection, may fail.
The key distinction is whether the duct remains connected across the injured segment. A partial disruption may permit a stent to bridge the defect and redirect secretion into the intestine. After central necrosis, the upstream tail may remain viable but completely disconnected; its continuing secretion can sustain a cavity or fistula that cannot be solved by the same bridging manoeuvre. Clinical stability, nutritional state, infection, accessible collections and operative history determine the immediate plan, while specialist imaging defines the longer-term solution.
Key points
- Pancreatic duct disruption can cause enzyme-rich ascites, a pleural communication, an external fistula or recurrent collections.
- Investigate new ascites with appropriate fluid analysis; a low albumin gradient and very high ascitic amylase support a pancreatic source in the right context.
- Pancreatic ascitic amylase is typically >1000 IU/L or >6 times serum. After pancreatic surgery, the separate ISGPS criterion is drain amylase >3 times the institutional serum upper limit from day 3 onward, plus an attributable clinical consequence for a clinically relevant fistula.
- MRCP helps define duct continuity; a partial leak that can be bridged differs from a completely disconnected viable upstream gland.
- Endoscopic transpapillary stenting may treat a suitable partial duct leak; complete disconnection usually needs a different internal drainage or surgical strategy if symptomatic.
- After pancreatic surgery, a clinically relevant fistula requires the postoperative amylase criterion plus an attributable clinical consequence; an isolated biochemical leak is not the same condition.
- Support nutrition and replace measured losses while assessing infection and collections; repeated fluid removal alone does not close the duct defect.
- Follow both the patient and the devices until leakage has resolved or an explicit long-term strategy is established.
02AetiologyUnderlying causes, associations and risk factors, with why each one matters.
Pancreatitis-related duct injury
Acute necrosis, chronic inflammatory duct disease and a communicating pseudocyst can disrupt the normal pancreatic drainage pathway. Fluid then follows available tissue planes or cavities, producing different internal or external manifestations.
Trauma and postoperative injury
Pancreatic trauma, a leaking pancreatico-enteric anastomosis or secretion from a cut pancreatic surface can create a leak. The postoperative setting has its own outcome definitions and should be distinguished from spontaneous fistulation after pancreatitis.
03PathophysiologyThe causal sequence from the underlying abnormality to symptoms and harm.
- 1Enzyme-rich fluid outside the duct
Escaping pancreatic secretions can accumulate in the peritoneum or a local cavity, track towards the pleura or emerge through a drain or wound. Persistent losses and inflammation can impair nutrition and fluid balance.
- 2Partial disruption
When upstream and downstream ducts remain connected, secretion can sometimes be redirected by bridging the defect endoscopically. The feasibility depends on the exact site, associated obstruction and whether a stent can cross the injured segment.
- 3Complete disconnection
Central necrosis can separate a viable upstream pancreatic segment from the duodenal outlet. Continuing secretion from that segment may sustain recurrent collections or an external fistula despite apparent treatment of the original cavity.
04Clinical features and red flagsSymptoms, examination findings, patterns of presentation and time-critical warnings.
Consider a pancreatic source when ascites recurs after pancreatitis or a pseudocyst, particularly without convincing portal-hypertensive disease. Abdominal distension, early satiety, weight loss and impaired intake may predominate over pancreatic pain. Ask about breathlessness and recurrent pleural fluid, since secretions can track beyond the peritoneal cavity.
Persistent fluid from a drain or wound after necrotizing pancreatitis or pancreatic intervention may contain pancreatic secretions. Measure volume and character, inspect surrounding skin and record fluid amylase in its proper context. A fall in output is encouraging only if the patient improves; abrupt cessation with fever or pain may represent drain blockage and an undrained collection and requires prompt pancreatic or surgical reassessment.
After pancreatic surgery, distinguish a laboratory finding from a clinically relevant fistula. In the ISGPS definition, drain fluid on or after postoperative day three has amylase above three times the institution’s normal serum upper limit, and a true postoperative fistula also produces an attributable clinical consequence. A well patient following the expected pathway with only that biochemical finding has a biochemical leak.
Recurrent collections or a persistent fistula after central pancreatic necrosis raise concern for disconnected pancreatic duct syndrome. A viable upstream segment continues to make secretions despite losing its normal outlet. Some patients remain asymptomatic and do not need an intervention merely because disconnection is visible; persistent symptomatic collections or fistula output justify a specific treatment discussion.
05InvestigationsWhat to request, why it matters and how to interpret it.
Read from the initial assessment onwards. Tests may run in parallel in urgent care; first-line, preferred, confirmatory, definitive and gold-standard labels appear only when the chapter explicitly states them.
- 01
Diagnostic ascitic-fluid assessmentFirst step - Why
- Determine the mechanism of new or unexplained ascites.
- Interpretation and limitations
- Measure fluid protein and albumin with a contemporaneous serum albumin to calculate the serum–ascites albumin gradient, and send amylase when pancreatic disease is plausible. A gradient below 11 g/L supports a non-portal-hypertensive mechanism; pancreatic ascitic amylase is typically above 1000 IU/L or more than six times serum. These clues need clinical correlation and do not replace infection or malignancy assessment when indicated.
- 02
Contrast CT and MRCP - Why
- Identify collections, complications and the likely duct anatomy.
- Interpretation and limitations
- CT shows pancreatic injury, viable upstream tissue, collections and vascular concerns. MRCP, sometimes enhanced with secretin in specialist practice, helps assess duct continuity. Complete disconnection is different from a tight stricture or an incomplete leak; knowing that distinction can prevent repeated attempts at an unsuitable transpapillary procedure.
- 03
ERCP for a therapeutic duct question - Why
- Define and treat a suitable duct leak when endotherapy is appropriate.
- Interpretation and limitations
- A stent that bridges an incomplete disruption can redirect pancreatic juice into the duodenum. Complete disconnection generally cannot be bridged by standard transpapillary stenting, and ERCP may not show the disconnected upstream segment. Balance the expected therapeutic benefit against procedure-related pancreatitis and other risks, using the pancreatic team’s prophylaxis and device plan.
- 04
Drain analysis and physiological monitoring - Why
- Assess external losses, postoperative significance and complications.
- Interpretation and limitations
- Use drain amylase with the operative date and local serum reference range when applying ISGPS criteria. Follow fluid balance, renal function, electrolytes, haemoglobin and nutritional status according to the volume and course. Culture appropriate samples when infection is suspected; high amylase alone is not an indication for antibiotics or proof of a clinically significant postoperative fistula.
06Differential diagnosisRealistic alternatives and the features that help distinguish them.
Portal-hypertensive or cardiac ascites
These mechanisms commonly produce a high serum–ascites albumin gradient and require their own assessment. A low gradient with marked ascitic amylase elevation after pancreatic disease should redirect attention towards a pancreatic source.
Malignant or infective peritoneal disease
Malignancy and peritoneal infection can also cause low-gradient ascites. Fluid analysis, imaging and the clinical history must therefore be interpreted together, rather than treating the albumin gradient as a diagnosis by itself.
Non-pancreatic drain output
Serous fluid, bile, enteric contents and other postoperative secretions can appear in drains. The operation performed, fluid characteristics, biochemical analysis and clinical course help identify the actual source and its significance.
07ManagementImmediate care, first-line treatment, alternatives and escalation.
01Worked caseRecurrent ascites from a partial duct disruptionFirst stepA 60-year-old woman develops recurrent abdominal distension five weeks after pancreatitis. She is afebrile, has no liver-disease signs and has required repeated symptomatic fluid removal.+
- 1Diagnostic fluid shows amylase 6800 IU/L with serum amylase 110 IU/L; serum albumin is 34 g/L and ascitic albumin 27 g/L, giving a low gradient of 7 g/L. The findings support pancreatic ascites. CT excludes a bleeding complication and MRCP suggests a partial duct disruption with continuity of the upstream gland. The team restores nutritional intake, monitors renal function and refers her to the pancreatic centre.
- 2At therapeutic ERCP, the endoscopist identifies and bridges the partial leak with a pancreatic stent. Within a week her abdominal distension and early satiety improve, and she no longer requires therapeutic paracentesis. A named endoscopist records the planned reassessment of duct healing and stent removal rather than leaving an open-ended device prescription.
- 3At four weeks, ultrasound shows only minimal residual fluid and she is maintaining weight and normal renal function. The scheduled later duct review confirms healing and the stent is removed. She is advised to seek reassessment for recurrent distension, fever, bleeding or breathlessness; the observed improvement supports the anatomical treatment without implying that every leak follows this timescale.
02Stabilise and mapA suspected pancreatic leak is producing illnessAscites, pleural fluid, an external fistula or recurrent collection is linked to pancreatic disease.+
- 1Assess circulation, respiration, sepsis and bleeding first. Quantify ongoing losses, provide appropriate fluid and electrolyte replacement and support nutrition with a dietitian. Drain a symptomatic or infected collection when indicated in the specialist plan, while recognising that fluid evacuation and closure of the duct defect are different aims.
- 2Establish whether the problem followed acute necrosis, chronic duct disease, trauma or surgery. Review imaging for duct continuity, viable pancreatic tissue and collections. Discuss pancreatic ascites or pleural effusion with a specialist centre, as recommended by NICE, rather than extending an ineffective generic ascites regimen.
- 3For a suitable incomplete disruption, consider endoscopic treatment with transpapillary stenting, ideally bridging the leak. For complete disconnection, assess whether symptoms require treatment and whether internal drainage of a collection is feasible. Surgery, including drainage or resection in selected anatomy, remains an option after multidisciplinary review when other measures are unsuitable or fail.
03Persistent or postoperative leakageA leak continues or alters the expected recoveryThe patient has a drain or a recurrent fluid collection despite initial treatment.+
- 1Reassess whether there is a blocked or displaced drain, ongoing infection, an undrained cavity, duct obstruction or a disconnected upstream gland. Current IAP guidance supports endoscopic management of symptomatic disconnected duct syndrome after conservative treatment has failed. Long-term internal drainage may be appropriate in selected cases; not every visible disconnection warrants resection.
- 2In the postoperative population, apply ISGPS grading to the attributable clinical course. Grade B requires a change in management, including prolonged drainage beyond three weeks or relevant endoscopic/percutaneous repositioning. Grade C involves fistula-related reoperation, organ failure or death. ICU admission or an unrelated postoperative event alone does not establish grade C.
- 3Review output, nutritional recovery and imaging together, then agree drain or stent removal, exchange or longer-term management with the responsible team. Persistent leakage requires a reasoned reassessment rather than an automatic medication schedule, repeated blind aspiration or a fixed waiting period that ignores the patient’s condition.
08ComplicationsImportant consequences, why they occur and why they matter clinically.
Losses and nutritional depletion
Ongoing external secretion or recurrent internal fluid accumulation can disturb fluid balance, restrict intake and worsen nutritional status. These consequences may persist while the original pancreatic pain is improving.
Infection and vascular injury
An undrained collection can become infected, while nearby vascular injury may cause serious haemorrhage. A new change in physiology or drain character requires prompt assessment of these complications.
Recurrence and treatment burden
A continuing duct defect can cause recurrent ascites, pleural fluid or collections after temporary drainage. Prolonged devices and repeated interventions add burdens that need to be balanced against a definitive anatomical strategy.
09Monitoring and follow-upTreatment response, safety checks and longer-term review.
- Measure external output and net fluid balance, replacing clinically important losses and checking renal function and electrolytes. Document output in the same units and interval so that trends can be interpreted reliably across handovers.
- Assess intake, weight, muscle function and signs of exocrine insufficiency during recovery. A persistent fistula or recurrent ascites can disrupt nutrition even when pain is modest; use pancreatic and dietetic input to maintain adequate support.
- Check for infection, bleeding, skin damage around an external opening and drain dysfunction. New symptoms with reduced output may require imaging to identify a trapped collection rather than celebrating apparent closure.
- Assign responsibility and a date for every drain or stent decision. After apparent healing, review recurrence and the longer-term exocrine and endocrine consequences of the underlying pancreatic disease, with a route back to the specialist service.
10Special situationsVariants, exceptions and circumstances that change the usual approach.
Different amylase comparisons answer different questions
Pancreatic ascites is usually markedly amylase-rich compared with serum, whereas the postoperative ISGPS threshold uses the institution’s normal serum upper limit and the postoperative day. Do not interchange the denominators or apply postoperative grading to a spontaneous leak after pancreatitis.
A viable tail can keep secreting
Complete central duct disconnection does not necessarily destroy the upstream gland. Continuing secretion explains recurrent collections after drainage and the difficulty of closing some external fistulas. The useful imaging question is therefore not merely whether a leak exists, but whether functioning tissue still has an outlet.
A bridge requires continuity
Transpapillary stenting works best when it can cross an incomplete disruption and restore a low-resistance route to the gut. Failure to cross a completely disconnected segment is an anatomical limitation, not an instruction to repeat the same manoeuvre indefinitely. Internal drainage or surgery may be needed for persistent symptomatic disease.
A classification follows the course
Postoperative fistula grading summarises the clinical consequences that actually occurred. It should not be assigned from drain amylase alone at the first measurement, and unrelated organ failure does not upgrade it. At the bedside, support and source control remain urgent whenever the patient deteriorates, before a final grade can be known.
11Common pitfallsFrequent interpretation and management errors.
- 01
Treating recurrent amylase-rich ascites as routine portal-hypertensive ascites without investigating duct disruption.
- 02
Using a postoperative drain-amylase threshold to diagnose or grade a spontaneous pancreatic leak in a patient who has not had surgery.
- 03
Assuming that a high drain amylase concentration alone defines clinically relevant postoperative fistula or requires antibiotics.
- 04
Attempting to bridge a completely disconnected duct as if it were an uncomplicated partial leak.
- 05
Removing or ignoring a drain because output has stopped when the patient has new fever, pain or evidence of an undrained collection.