Do you have a two-bit pancreas? You well might…better read on.

Do you have a two-bit pancreas? You well might…better read on.

About 8% of people, at least in Western countries, are born with the pancreas in two pieces, so called “pancreas divisum”. It is about as common as left-handedness, and ten times that of red hair. How come, and who cares?

Some technical stuff to get started. The human pancreas is formed from two parts which normally come together in the second month of intrauterine life, as shown in the diagram. Their ductal systems normally fuse so that the whole pancreas drains through Wirsung’s duct into the main papilla of Vater in the duodenum, alongside the bile duct (Image C below).

Failure of that process results in pancreas divisum, in which most of the pancreas drains through a smaller duct of Santorini into the “minor papilla”, as shown in B.

For those interested in the origin of those and other names please explore the fascinating treatise from Claudia Stern https://gut.bmj.com/content/gutjnl/27/2/203.full.pdf.. One lovely anecdote is that the earliest description of pancreas divisum was long missed because it was written in English, whereas all scholars read only Latin!

We began to recognize this anomaly when gaining experience with the new technique of Endoscopic Retrograde Cholangio Pancreatography (ERCP) in the early 1970s, which allowed us to guide  a flexible endoscope through the mouth down to the duodenum and  to inject contrast through the papilla of Vater to outline the pancreatic and biliary ductal systems. Before that, the only pancreatic anomaly known to most gastroenterologists was “annular pancreas” which is very rare (about 1 in 20,000 births). It usually presents in children because it restricts the duodenum.

The radiological findings at ERCP were first described nicely by Rosch and colleagues in Germany in 1976 (1). Let’s get those out of the way before moving on to the fun stuff. Apologies for the poor quality of the images. Most are copies of mine from more than 40 years ago.

ERCP picture. Normal fused pancreatic ducts, arrows on the main and minor papillas

The ventral pancreas varies in size

Do not overfill it, or mistake it for an obstructed pancreatic duct typical of cancer

The separate dorsal duct draining through Santorini and the minor papilla

ERCP is no longer needed to define pancreatic anatomy. It is nowadays very well documented non-invasively with MRCP, especially after an injection of secretin.

The Rosch paper was entitled “The clinical significance of pancreas divisum” but actually only described the radiological aspects (1). It fell to a British group the next year to suggest that the anomalous drainage system might be inadequate, with clinical consequences (2). We postulated that the tiny “minor papilla” could be too small to accommodate the flow of all the pancreas secretions, leading to attacks of pancreatitis.  We were fortified in that proposal when seeing pancreatic juice squirting out of the minor papilla in a patient after an injection of secretin

And we convinced ourselves that the anomaly was clinically important when we reported than no fewer than a quarter of all patients with otherwise unexplained attacks of pancreatitis coming to ERCP had pancreas divisum (3).  Game, set and match?

Not so fast. The esteemed Brussels group (led by friends Myriam Delhaye and Michel Cremer) returned service rather forcefully with a much larger study showing no such increased incidence (4). To solve the problem, I organized an international conference at The Middlesex Hospital in 1984, assembling a variety of experts with different perspectives and skills. The conversations were good, but the title of the subsequent report shows that the controversy persisted (5).

There followed a long and often spirited controversy which continues to this day. The National Library of Medicine records no fewer than 902 scientific publications on the subject in the English language (maybe others in Latin).

We realized that we needed to look for other evidence to back our hypothesis. If indeed the minor papilla was causing obstruction, the upstream dorsal duct should dilate (as it does behind a tumor). Hmm, no, that was rarely present. Why not? Good question……

Then we tried a stress test. We measured the diameter of the pancreatic duct with ultrasound scanning (later with MRCP) after stimulating pancreatic secretion with an intravenous injection of the hormone secretin. It seemed that the duct did dilate a little normally, but maybe took longer to get back to normal.

Someone suggested measuring the pressures in both the minor and major papillas with ERCP manometry. Really?

The ultimate test of the hypothesis of course is whether relieving the postulated obstruction would relieve the clinical problem. The first endoscopic sphincterotomy of the minor papilla was published in 1978 in a French journal by a British author (maybe he was embarrassed?) (6). Multiple other series (including surgical sphinteroplasties) followed, mostly claiming “good results” in reducing the recurrence of pancreatitis attacks, but with less or no benefit in patients who were being treated for unexplained pancreatic-like pain, but no pancreatitis (7,8). A recent meta-analysis concluded that none of these studies could claim to have been of sufficient quality to be definitive (9).

Another idea. If the minor papilla was/is the problem, any pancreatitis would be confined to the “dorsal” portion of the gland draining though it. The ventral part should remain healthy. That part is often very small and therefore difficult to assess, and our imaging methods were not sensitive enough to detect any small differences. However, we did have some interesting data from surgical specimens. We were able to review the histology of 2 patients with pancreas divisum who had undergone resection of the head or the whole of the pancreas for intractable pain. They did indeed show impressive fibrotic pancreatitis confined to the dorsal part (10).

Healthy ventral pancreas on the left, fibrotic pancreatitis in the dorsal part on the right

Convincing stuff, right? Well, maybe not. We realized afterwards that these patients had previously undergone minor sphincterotomies without clinical benefit and likely stenosis of the orifice, which could well explain the histological findings.

So, what else? The Indiana group reported an increased incidence of genetic markers of pancreatitis in the divisum population, which might help explain any increased propensity.

In some elderly patients with divisum, we did note the presence of a “santorinicele” (a cystic dilatation just inside the minor papilla), suggesting local obstruction (11). These were easy to treat endoscopically, apparently with benefit.

Santoriniceles at ERCP and MRCP (ventral and dorsal ducts arrowed)

What is really needed is a prospective study in which patients with divisum and recurrent pancreatitis are randomized to undergo minor sphincterotomy or a sham procedure. Happily, such a study is in progress at multiple sites led by Greg Cote at MUSC (12). The SHARP study (SpHincterotomy for Acute Recurrent Pancreatitis trial) is funded by NIH.  It will take a while to recruit, and longer to conclude, since there has to be substantial follow-up to see if recurrences are prevented.

I have a nagging concern about the excellent SHARP study. If it fails to show benefit, I fear that some divisum enthusiasts will question whether minor papilla sphincterotomy as currently practiced does actually improve drainage. It is a fiddly procedure. The papilla and the duct behind it are small. Whether cut by sphinterotome or needle knife and even with a temporary stent, I, for one, never felt very confident that the orifice had been enlarged permanently. Some scarring is inevitable. It would be instructive to test that concern by performing secretin-stimulated MRCP exams before and some time after the treatment (both active and sham). La vie d’artiste c’est difficile.

My prediction? I seem to be coming full circle. I have enjoyed rereading the thoughtful contribution from Brussels in 1985 (4), which concluded: “These results do not support the hypothesis that stenosis of the accessory papilla occurs frequently in cases of pancreas divisum. We conclude that no further therapy should be systematically proposed ….”. Were they right after all?

Aging, and my experience with the EPISOD study, have made me more conservative and skeptical. I rather suspect that the hypothesis that I proposed some 44 years ago will be disproven. If so, I will need to apologize for pushing it rather strongly, and indeed to all practitioners and patients who have toiled and suffered unnecessarily.

I have always been fond of Brussels and admiring of the work done by the team that Michel Cremer started. I had my 50th birthday at his house with special friends and special wine…

So, while waiting for more data, if you are one of the 1 in 10 -15 people with pancreas divisum (especially if left-handed – like most of our recent US Presidents) you might want to enjoy it rather quietly.

In closing, I am reminded of my favorite quote from Solly Marks, godfather of gastroenterology in South Africa, and a close friend (despite his frequent kind statement): “Peter, you are again trying to sew flatus to moonbeams”.

All part of life’s rich tapestry….

  1. Rösch W, Koch H, Schaffner O, Demling L. The clinical significance of the pancreas divisum. Gastrointest Endosc. 1976 May;22(4):206-7. doi: 10.1016/s0016-5107(76)73755-6. PMID: 1269885.
  2. P.B.Cotton, M. Kizu. Malfusion of dorsal and ventral pancreas. A factor in pancreatitis? Gastroenterology 1977;72:A18/1041  .……. This anomaly has been found in 24 patients with pancreatitis investigated by endoscopic pancreatography. ….., but no aetiological factors were apparent in the remaining 9, and we postulate that inadequate drainage via Santorini’s duct may be important. ……
  3. Cotton PB. Congenital anomaly of pancreas divisum as cause of obstructive pain and pancreatitis. Gut. 1980 Feb;21(2):105-14. doi: 10.1136/gut.21.2.105. PMID: 7380331; PMCID: PMC1419363.
  4. Delhaye M, Engelholm L, Cremer M. Pancreas divisum: congenital anatomic variant or anomaly? Contribution of endoscopic retrograde dorsal pancreatography. Gastroenterology. 1985 Nov;89(5):951-8. doi: 10.1016/0016-5085(85)90193-3. PMID: 4043675.
  5. Cotton PB. Pancreas divisum–curiosity or culprit? Gastroenterology. 1985 Dec;89(6):1431-5. doi: 10.1016/0016-5085(85)90667-5. PMID: 4054536.
  6. Cotton PB. Duodenoscopic papillotomy of the minor papilla for recurrent dorsal pancreatitis. Endoscopie Digestive 1978; 3:27-28
  7. Gutta A, Fogel E, Sherman S. Identification and management of pancreas divisum. Expert Rev Gastroenterol Hepatol. 2019 Nov;13(11):1089-1105. doi: 10.1080/17474124.2019.1685871. Epub 2019 Nov 8. PMID: 31663403; PMCID: PMC6872911.
  8. Tringali A, Voiosu T, Schepis T, Landi R, Perri V, Bove V, Voiosu AM, Costamagna G. Pancreas divisum and recurrent pancreatitis: long-term results of minor papilla sphincterotomy. Scand J Gastroenterol. 2019 Mar;54(3):359-364. doi: 10.1080/00365521.2019.1584640. Epub 2019 Mar 17. PMID: 30880501.Eisen G, Schutz S, Metzler D, Baillie J, Cotton PB. Santorinicele: new evidence for obstruction in pancreas divisum. Gastrointest Endosc. 1994 Jan-Feb;40(1):73-6. doi: 10.1016/s0016-5107(94)70015-x. PMID: 8163142.
  9. Michailidis L, Aslam B, Grigorian A, Mardini H. The efficacy of endoscopic therapy for pancreas divisum: a meta-analysis. Ann Gastroenterol. 2017;30(5):550-558. doi: 10.20524/aog.2017.0159. Epub 2017 May 12. PMID: 28845111; PMCID: PMC5566776.Coté GA, Durkalski-Mauldin VL, Serrano J, Klintworth E, Williams AW, Cruz-Monserrate Z, Arain M, Buxbaum JL, Conwell DL, Fogel EL, Freeman ML, Gardner TB, van Geenen E, Groce JR, Jonnalagadda SS, Keswani RN, Menon S, Moffatt DC, Papachristou GI, Ross A, Tarnasky PR, Wang AY, Wilcox CM, Hamilton F, Yadav D; SHARP Consortium. SpHincterotomy for Acute Recurrent Pancreatitis Randomized Trial: Rationale, Methodology, and Potential Implications. Pancreas. 2019 Sep;48(8):1061-1067. doi: 10.1097/MPA.0000000000001370. PMID: 31404020; PMCID: PMC6699897.
  10. Blair AJ 3rd, Russell CG, Cotton PB. Resection for pancreatitis in patients with pancreas divisum. Ann Surg. 1984 Nov;200(5):590-4. doi: 10.1097/00000658-198411000-00006. PMID: 6385880; PMCID: PMC1250541.
  11. Eisen G, Schutz S, Metzler D, Baillie J, Cotton PB. Santorinicele: new evidence for obstruction in pancreas divisum. Gastrointest Endosc. 1994 Jan-Feb;40(1):73-6. doi: 10.1016/s0016-5107(94)70015-x. PMID: 8163142.
  12. Coté GA, Durkalski-Mauldin VL, Serrano J, Klintworth E, Williams AW, Cruz-Monserrate Z, Arain M, Buxbaum JL, Conwell DL, Fogel EL, Freeman ML, Gardner TB, van Geenen E, Groce JR, Jonnalagadda SS, Keswani RN, Menon S, Moffatt DC, Papachristou GI, Ross A, Tarnasky PR, Wang AY, Wilcox CM, Hamilton F, Yadav D; SHARP Consortium. SpHincterotomy for Acute Recurrent Pancreatitis Randomized Trial: Rationale, Methodology, and Potential Implications. Pancreas. 2019 Sep;48(8):1061-1067. doi: 10.1097/MPA.0000000000001370. PMID: 31404020; PMCID: PMC6699897.

2 Responses

  1. Professor Jurgen Rode says:

    Hallo Peter,
    Great article.
    I just want to comment: on the whole-mounted H&E section of the pancreas divisum specimen, one sees within the completely fibrosed ventral portion an extremely dilated duct which was the result of ample ectopic normal pancreas within the duodena wall surrounding the papilla. This ectopic tissue would have stenosed the duct especially after meals.
    Regards
    Jürgen

    • Peter Cotton says:

      Thanks Jurgen, great memories of working with you at The Middlesex a long time ago. Interesting observation (but I think you meant the dorsal portion).

Leave a Reply

Your email address will not be published. Required fields are marked *