Pancreas divisum

How can I be sure that the patient has pancreas divisum?

Many patients with pancreas divisum have no symptoms. The most common presentation would be episodes of acute or recurrent pancreatitis or, in rare cases, symptoms of chronic pancreatitis. Pancreas divisum is a very common congenital anomaly of the pancreatic ductal system, where there is a nonfusion of the dorsal and ventral ducts of the pancreas embryologically. It is seen in approximately 7% of Western patients.

  • 1. Acute intermittent episodes of abdominal pain with or without radiation to the back associated with elevation of amylase and or lipase (typically >3x nl) for acute pancreatitis and possibly less for those with chronic pancreatitis, or pain only without any significant elevation of amylase or lipase.

The diagnosis is suspected based on history and clinical findings and confirmed with imaging. CT scan of the abdomen will identify presence of any inflammation and complications of pancreatitis, such as organ necrosis or fluid collections. It is less sensitive for the diagnosis of pancreas divisum, although some small studies have shown sensitivity of up to 80% and a specificity of up to 90%. Coronal images enhance the ability of CT to detect pancreas divisum. Secretin-enhanced magnetic resonance cholangiopancreatography (MRCP) is a useful tool in the diagnosis of pancreas divisum with a very good specificity (up to 96%). The sensitivity has been variable from 73% to 100%. Nonsecretin-enhanced MRCP has a significantly lower sensitivity. Endoscopic ultrasound (EUS), especially using linear array instruments, can be very accurate in diagnosis or exclusion of pancreas divisum.

Endoscopic retrograde cholangiopancreatography (ERCP) has in the past been the reference standard for diagnosis of pancreas divisum, but is now largely supplanted by MRCP. ERCP requires injection of ventral and dorsal ducts separately for complete imaging of pancreas divisum and should only be considered when a therapeutic intervention is planned and not purely for a diagnostic purposes. The risk of post-ERCP pancreatitis is high for minor papilla cannulation in patients with pancreas divisum, who have not already developed chronic pancreatitis.

  • 2. Imaging findings may be of either inflammatory changes of the pancreas or changes consistent with chronic pancreatitis. In some patients with divisum and pain only, neither acute nor chronic inflammatory changes are seen. In such patients, the relationship of pancreas divisum to pain is highly questionable.

  • 3. Those with exocrine insufficiency secondary to chronic pancreatitis might have fatty, frothy, foul-smelling stool (steatorrhea), and, in rare instances, there could be endocrine insufficiency, and this could result in polyuria, polyphagia, or polydypsia. In clinical series, those with acute recurrent pancreatitis and pancreas divisum presented most typically in the fifth decade of life and those with pain only and pancreas divisum in the fourth decade. However patients may present at any time from infancy to old age. The duration of pain is variable and is reported between 1 to 20 years.

Most patients are asymptomatic. Those with symptomatic disease may present with the following:

A tabular or chart listing of features and signs and symptoms

  • – Abdominal pain typically in the epigastric region/right upper quadrant with radiation to the back.

  • – Tenderness in the epigastric area on palpation.

  • – Elevation of amylase and or lipase (typically >3x nl for acute or acute recurrent pancreatitis and less for chronic pancreatitis).

  • – Abdominal pain only, without any laboratory or imaging abnormalities to suggest pancreatitis.

  • – Nausea and or vomiting.

  • – In case of severe pancreatitis, the patient can have symptoms/signs of systemic inflammation, including fever, tachycardia, tachypnea.

  • – Organ failure (respiratory/renal), either transient or sustained.

  • – Other common conditions that mimic pancreas divisum are peptic ulcer disease, perforated viscus, cholecystitis, biliary colic.

  • – Pancreatic tumors.

Imaging and laboratory results usually differentiate between pancreatitis and other conditions.

How can I confirm the diagnosis?

Pancreas divisum should be considered in those patients with otherwise unexplained acute pancreatitis, usually after more than one attack, and occasionally in patients with unexplained chronic pancreatitis.

For those with unexplained pancreatitis or unexplained abdominal pain, it is reasonable to start with MRCP. Administration of intravenous secretin during MRCP will result in better visualization of the ducts and also helps identify obstruction at the dorsal duct/minor papilla by visualizing upstream dilation of the dorsal duct. Santoricele (saccular dilation of terminal dorsal pancreatic duct at the level of the duodenum) can be identified as well. In the past, ERCP and pancreatography were used as a reference standard but is now largely replaced by MRCP. It is advisable to personally review the films with a GI radiologist since there are differences in medical opinions among radiologists reporting divisum. If MRCP is suggestive of pancreas divisum, this could be confirmed by pancreatography at ERCP.

Pancreatography should be reserved only when there is an intent for therapy (e.g., minor papillotomy, relief of obstruction, dilation of stricture, etc.). Routine diagnostic pancreatography should be avoided. Advantage of doing cross-sectional imaging is that it can identify tumors that can present as incomplete divisum.

Endoscopic ultrasound is a useful tool in the diagnostic evaluation of patients with unexplained acute pancreatitis. It helps evaluate the pancreatic parenchyma and, when the ducts are visualized well, has a very good sensitivity and specificity in identifying pancreas divisum. EUS also can sometimes identify early changes of chronic pancreatitis.

At ERCP, the ventral duct cannot be cannulated at all in about a third of the cases. A few cases might exhibit an incomplete divisum, where the dorsal duct drains through a small branch into the ventral duct. When the ventral duct is cannulated at ERCP, it is usually short, measuring 1 to 4 cms. It usually does not cross the midline, and opacification with contrast usually shows the secondary and tertiary branches (arborization).

The minor papilla is usually located superior and anterior to the major papilla. To orient the cannulating catheter, it is often necessary to put the duodenoscope in “long scope position.” Since the orifice is small, a tapered cannula 3 Fr to 5 Fr is generally usedUsing a 0.018″ to 0.021″ guidewire, the orifice engaged with the tip of the wire using pure wire cannulation technique, and cannulation can be attempted under fluoroscopic guidance without opacification of the duct. Alternatively, one can gently opacify the duct to delineate the path of the dorsal duct prior to advancing the wire and avoid the risk of side branch trauma and risk of pancreatitis. If minor papilla cannot be readily identified, administration of intravenous secretin (0.2 mcg/kg IV)) will result in increased pancreatic flow, and the orifice is generally apparent. In addition, methylene blue can be sprayed over the papilla, and the area of drainage washes out the methylene blue, exposing the papillary orifice and, thereby, facilitating cannulation. (See Figure 1.)

Figure 1.

Minor papilla without visible orifice.

Unexplained abdominal pain or pancreatitis
  • Check amylase and or lipase. If more than 3-fold elevation, suspect pancreatitis.

  • Cross-sectional imaging. Contrast-enhanced CT by pancreas protocol (esp. if no renal insufficiency). Identifies acute and or chronic inflammation of the pancreas; can identify pancreatic tumors; not very sensitive for pancreas divisum/chledocholithiasis.

  • Ultrasound of the gall bladder. Sensitive for stones in the gall bladder but not in the bile duct.

  • If ultrasound and CT is negative or suggestive of pancreas divisum, perform MRCP, preferably secretin-enhanced. Very sensitive. Dilation of ventral duct seen post secretin administration if ductal obstruction is secondary to stone/stricture. Focal dilation at the minor papilla in case of santorinicele. Note: Significant inter-observer variability. If divisum is suspected, personally review films with radiologist, preferably with GI radiologist.

  • Genetic testing outside of clinical research is not routinely done. (SPINK and CFTR mutations have been commonly reported.) Cause association unclear at this time. Presence of mutations could lead to chronic pancreatitis. Current knowledge is limited.

  • ERCP is the reference standard for diagnosis of pancreas divisum because the role of MRCP especially secretin-enhanced has been accepted clinically only recently, and there still exists a difference in quality of imaging and inter-observer variability. However, routine ERCP for ductography (imaging of pancreatic duct to confirm pancreas divisum without any intent for therapy) is not recommended. When ERCP is performed, in a third of cases, the ventral duct cannot be cannulated at all. If the ventral duct is cannulated, prominent side branches can be seen in the head of the pancreas “arborization.” At times, there might be a small duct communicating with the ventral to the dorsal duct “incomplete divisum.”

The problem with ERCP is that tumors that mimic divisum can be missed easily. ERCP hence should always be done after either CT scan/MRI or EUS of the pancreas so that a tumor is not missed. It is a very important consideration especially in the elderly since an obstructing tumor can cause incomplete filling of the pancreatic duct.

What other diseases, conditions, or complications should I look for in patients with pancreas divisum?

Although pancreas divisum can cause pancreatitis, it is important to rule out other common causes. History of alcohol ingestion, especially a binge episode prior to an episode of pancreatitis, should raise a suspicion for alcohol as etiology. Biliary etiology by CT/ultrasound, common medications that can cause pancreatitis, and hypertriglyceridemia (>1,000 mg/dL) should be ruled out.

Tumors presenting unexplained pancreatitis are not uncommon, especially in the elderly, and a tumor causing complete obstruction of the pancreatic duct in the head of the pancreas can mimic the radiographic appearance of pancreas divisum.

Many patients with pancreas divisum who develop recurrent acute or chronic pancreatitis have coexistent genetic mutations, especially that of SPINK/CFTR. In this setting, genetic predisposition and pancreas divisum may combine to increase the risk of pancreatitis. It appears that in genetically primed individuals, some environmental trigger sets off the pathway of inflammation.

Most common complications with pancreas divisum are recurrent pancreatitis. As with any pancreatitis, the clinician should look for necrosis of the gland, systemic inflammatory response, and organ failure, especially of the kidneys and respiratory systems. Acute peripancreatic fluid collections and pseudocyst formation can occur. These complications should be managed similar to any patient with pancreatitis due to any etiology.

The natural history of the disease is unclear. It is unclear how many patients with pancreas divisum have recurrent episodes versus a single episode. The rate of progression to chronic pancreatitis is unclear and might be dependent on associated genetic mutations, concomitant smoking, and alcohol ingestion.

If a patient develops obstruction of the dorsal duct, either secondary to a calcification or stricture, he or she could develop pain or recurrent pancreatitis.

What is the right therapy for the patient with pancreas divisum?

Clinically, to help guide therapy and predict outcomes, patients with pancreas divisum are often divided into three categories:

  • Pancreas divisum with acute recurrent pancreatitis.

  • Pancreas divisum with chronic pancreatitis (calcifications; stricture of the duct; obstructi on of the duct, secondary to calcifications/strictures).

  • Pancreas divisum with no evidence of any inflammation but with chronic pain. Few have no pain between episodes.

In practice, most patients do not fit into any one category and there is an overlap between the groups, making it difficult to predict response to therapy, the therapeutic modality to choose.

Therapeutic measures and procedures
  • – All patients should avoid smoking.

  • – Adequate analgesia for pain should be prescribed.

  • – The role of antioxidants is unclear but might be considered in patients with changes suggestive of chronic pancreatitis.

  • – One small study looked at stenting of the dorsal duct with a series of three stents of varying size and diameter. There was a decrease in the frequency of ER visits and hospitalizations. These data have not been replicated by other centers.

According to a large retrospective study that looked at minor papillotomy followed by short term stenting of the dorsal pancreatic duct to prevent post procedural pancreatitis, there was an immediate benefit in 60% of the patients, but only 32% had continued improvement in the long term.

  • – Endoscopic minor papillotomy appears reasonable for patients with acute recurrent pancreatitis with no pain in between attacks. In a retrospective study, 43% had long-term relief, although the difference in response was not statistically significant compared to the other groups.

  • – For the patient with changes suggestive of chronic pancreatitis, if there is a ductal obstruction either by a stone or stricture, endoscopic therapy is reasonable to relieve obstruction. A minor papillotomy can be performed and the stone removed after dilation of the stricture, with or without lithotripsy. In general, strictures of the pancreatic duct do not respond to monotherapy and might need a combination of dilation and stenting.

  • – For those with intermittent pain and no elevation in enzymes or no chronic pancreatitis on imaging, the role of endoscopic therapy is very limited. If pain is chronic and there are intermittent episodes of exacerbation, the clinician should try medical therapy prior to considering minor papillotomy. It is conceivable that patients with no changes of chronic pancreatitis but with severe chronic pain that impairs their quality of life might benefit from total pancreatectomy with auto islet cell transplant. This is still in infancy and a generalized recommendation cannot be made at this time, but it might be reasonable to refer these patients to one of the centers performing auto islet cell transplantation.

  • – Surgical sphincteroplasty has been done and the results are not significantly different from endoscopic therapy and is not generally performed.

  • – In those with chronic pancreatitis, enzyme supplementation should be considered.

What is the most effective initial therapy?

  • – For those with recurrent acute pancreatitis. Minor papillotomy with temporary dorsal duct stenting. (A multicenter clinical trial FRAMES (Frequency of Recurrent Acute Pancreatitis after Minor Papilla Sphincterotomy) is currently in progress and might provide more data).

  • – For those with dorsal duct obstruction. Endoscopic therapy appears to be a reasonable choice.

  • – For those with chronic pancreatitis. Consider enzyme supplements/antioxidants.

  • – Role of routine genetic testing not established and is currently reserved for research purposes.

Listing of usual initial therapeutic options, including guidelines for use, along with expected result of therapy.

In view of small studies and the absence of robust clinical data, no guidelines exist.

Failure of therapy is also difficult to define in a condition that is chronic and can have recurrent episodes. It is reasonable to assume that a reduction in the frequency of attacks of pancreatitis or decreased length of hospitalization is a surrogate marker of therapeutic success and the converse could be used for therapeutic failure.

A listing of a subset of second-line therapies, including guidelines for choosing and using these salvage therapies

  • – Antioxidants

  • – Pancreatic enzymes in case of chronic pancreatitis

  • – Surgical drainage, or increasingly resection in the form of total pancreatectomy with islet autotransplantation, is considered for patients with relapsing acute pancreatitis or chronic pancreatitis associated with pancreas divisum and with intractable chronic pain, who fail to respond to maximal endoscopic therapy.

Listing of these, including any guidelines for monitoring side effects.

  • – If endoscopic therapy is done, there is a high risk of post-ERCP pancreatitis. Prophylactic stenting of the dorsal duct is recommended.

  • – Obtain an X-ray in a week to assess for spontaneous migration of the stent into the gut; otherwise, it needs to be removed endoscopically.

  • – If stenting is done for therapeutic purposes, softer stents might cause less trauma and possible stent-induced strictures. If multiple stents are required, some clinicians recommend varying the lengths which in turn could cause a longer stent-induced stricture.

  • – If post-ERCP pancreatitis presents, it should be managed like any acute pancreatitis.

How should I monitor the patient with pancreas divisum?

Patients with early chronic pancreatitis or chronic pancreatitis might undergo periodic exams with endoscopic ultrasound (perhaps once a year) because there is an increase of pancreatic neoplasia in those with chronic pancreatitis.

These patients should be monitored for clinical symptoms of exocrine and endocrine insufficiency and replacement therapy should be ordered as appropriate.

What's the evidence?

Lan, JI, Greenen, JE, Johanson, JF, Hogan, WJ. “Endoscopic therapy in patients with pancreas divisum and acute pancreatitis: a prospective, randomized, controlled clinical trial”. Gastrointest Endosc. vol. 38. 1992. pp. 430-4. (There is some evidence that stenting of the dorsal duct reduces the risk of recurrence. The study was small though randomized. These results were not replicated.)

Gerke, H, Byrne, MF, Stiffler, HL. “Outcome of endoscopic minor papillotomy in patients with symptomatic pancreas divisum”. JOP. vol. 5. 2004. pp. 122-31. (Clinical efficacy of endoscopic therapy is probably best shown in a retrospective series of 89 patients with divisum who underwent minor papillotomy and stenting.)