- Email: [email protected]
Pancreas divisum
Pancreas Divisum Results of Surgical Intervention
James A. Gregg, MD, Boston, Massachusetts Anthony P. Monaco, MD, FACS, Boston, Massachusetts William V. McDermott, MD, FACS, Boston, Massachusetts
The embryologic anomaly that results when the ductal systems of the ventral and dorsal anlage of the fetal pancreas fail to fuse was first described by Opie [I] in 1903, who labelled the condition “pancreas divisum” [2]. Although there have been anatomic references to this anomaly in the literature [3-IO], it was the introduction of endoscopic retrograde cholangiopancreatography (ERCP) that made possible the clinical recognition of pancreas divisum [11-271 (Figures 1 and2). Although this anomaly is not uncommon and probably occurs in 3 to 7 percent of the population [ 7,15,17,21,23,25], the association of pancreas div-
isum with otherwise unexplained recurrent episodes of upper abdominal pain consistent with that seen in patients with pancreatitis or obstructive disease of the pancreas or biliary tract led to the suggestion [I 71 that inadequate drainage of the dorsal pancreas through the dorsal pancreatic duct and minute minor sphincter was a cause of pancreatitis, or that pain resulted from either an intermittent or constant increase in pressure within the dorsal duct secondary to relative stenosis of the minor ampulla and impaired drainage. Of 70 patients with pancreas divisum and upper abdominal pain who were followed by one of us (JAG), laparotomy and sphincteroplasty of both the major and minor sphincters were performed in 19 because of severity and intractability of pain. It is these 19 patients who serve as the basis of this re-
port. From the Departments of Surgery and Medicine, Harvard Medical School and the New England Deaconess Hospital, Boston, Massachusetts. Requests for reprints should be addressed to William V. McDermott, MD, New England Deaconess Hospital, 185 Pilgim Road, Boston, Massachusetts 02215. Presented at the 63rd Annual Meeting of the New England Surgical So. ciety, Bretton Woods, New Hampshire, October 15-17. 1982.
488
Material and Methods There were 16 women and 3 men in this series. Their ages ranged from 15 to 44 years (median 26 years). All patients could date the onset of their problem to before age 40, and in nine patients pain began before age 20. Seven of these patients had chemical, microscopic, or radiologic evidence of pancreatitis. The symptoms were similar in all patients, consisting of episodic or constant epigastric pain which radiated directly through to the back. It was often but not invariably precipitated by consumption of foods that were high in fat. The pain frequently caused patients to awake from sleep. In none of the patients was the pain controlled to any reasonable degree by dietary management or utilization of a number of different medications, and only one patient in the entire group could be characterized as having an excessive intake of alcohol. Because of the suggested relation of pancreas divisum to pancreatitis and obstructive pain [17], indications of the association were sought in elevations of serum amylase or lipase levels, pancreatic abnormalities observed by ultrasonography, ERCP, and in histologic abnormalities in pancreatic tissue removed at surgery. Seven patients had chemical and microscopic evidence that supported a diagnosis of pancreatitis, and in three patients pancreatic enlargement compatible with the diagnosis of pancreatitis was present on ultrasonogram. Four of the patients were considered to have chronic pancreatitis. The diagnosis of pancreas divisum was established during ERCP in 16 patients by opacification of the duct of the pancreatic head, which was invariably short, averaging only a few centimeters in length and draining only a small segment of the ventral pancreas. In one patient the diagnosis of pancreas divisum was established by opacifying the dorsal duct during ERCP, and in the remaining two patients who were thought to have a vestigial ventral duct, the diagnosis was made by the administration of secretin which demonstrated pancreatic secretions issuing from the minor papilla only. Although the minute size of the dorsal papilla and its more cephalad location made it more difficult to obtain a pancreatogram, the dorsal panThe American Journal of Surgery
Pancreas Divisum
PANCREAS DIVISUM
Flgun, 1. 7% anomaly refermd to as pancreas dlvtsum.
creatic duct was opacified during ERCP in six patients and,
with the exception of a serrated duct in one patient, the other endoscopic pancreatograms of the dorsal duct did not reveal any abnormalities, except delayed drainage.
Surgical Procedures and Results In 19 patients the severity of symptoms was the indication for laparotomy. A standard procedure was devised which consisted of exploration of the common bile duct with cholecystectomy (unless the gallbladder had been previously removed), duodenotomy, and transduodenal division of the major and minor papillae with careful sphincteroplasties using 5-O chromic catgut or polyglycolic acid sutures (Figure 3). The administration of secretin as an intravenous bolus proved to be a useful adjunct to palpation for identification of both papillae. Magnification was used during reconstruction of the outlet through the minor papilla. In one instance, a standard transduodenal sphincterotomy and sphincteroplasty were carried out on the outlets of the common bile duct and the major pancreatic ducts, and exploration and pancreatography (when indicated) of the ventral and dorsal ducts confirmed that in each patient only a very small segment of pancreas was drained through the major papilla, and that the dorsal duct was the principal pancreatic duct. Pancreatograms of the dorsal duct were obtained during surgery and demonstrated normalVolume 145, Apll 1983
Fipm 2. Endoscopk retrograde cholangkpancreatogram of a dMWlLLetY,lXN7lWloomnOn bne&ctamI patienlwithpsmeas small ventral pancreatic duct. Right, &rsal pancreatk duct opactfted through the minor papllla.
sized ducts, with the exception of one patient in whom the duct was dilated to 5 mm. Severe hemorrhagic pancreatitis developed postoperatively in one patient who subsequently died from complications of this disorder; it was thought that the pancreatitis might have been related to the instillation of contrast material during operative pancreatography, but other operative pancreatograms were obtained without incident. In the remaining 18 patients results were good to excellent in 13, fair in 1, and poor (no satisfactory relief of symptoms) in 4. Excellent results were those in which the patient was symptom-free at follow-up, good results were those in which rare, mild episodes of pain were reported at follow-up, and fair results were those in which some decrease in pain was reported but it was still a significant problem. Recurrence of pain developed in four patients, however within a period that ranged from 1 to 6 months; reasonably long-term relief from the initial procedure was limited to 10 patients. Of the remaining eight patients with recurrent or continuing symptoms, a variety of subsequent procedures on the pancreas led to satisfactory results in only three. In one of the patients with recurrent disease, endoscopy demonstrated a recurrence of stenosis of the minor papilla; in this instance, distal pancreatectomy with drainage did not result in further improvement. Other follow-up observations by endoscopy or at subsequent surgery were not sufficient to determine a consistent reason for treatment failure, such as restenosis. In at least three patients, the opening through the minor papilla was widely patent although pain recurred. 489
Gregg et al
Ductus Santarini - - /
Ampulla of Voter \ / ’
i Ductus
F&we 3. Surgical procedure for transduodenal sphlncieroplasty.
Thus, the procedure of transduodenal sphincteroplasty appeared to achieve excellent initial results in terms of pain relief, but in 4 patients this was shortlived, and in only 10 patients who remained pain-free for.periods that ranged from 6 months to 9 years could the procedure be termed an unqualified success (Figure 4). Comments Clearly there are a number of questions that arise which relate to this study and other reports in the literature. First, is this anomaly associated with an
19
PATIENTS
i 18
EXCELLENTTO
GOOD
I
DEATH
I
PATIENTS
FAIR
POOR
RECU:;,
8 PATiENTS
FURTHER
; FAILED 3
&RATION
IMPROVED 3
Resect/on and 0rob70ge fopiuofomy
90 % Resectim
Figure 4. Posfoperative outcome In the 19 study patients.
490
increased incidence of pancreatitis? The studies of Cotton [Z], Richter et al [25], Tulassay and Papp [22], and Gregg’s experience [2 71 all demonstrated a significant increase in the incidence of pancreas divisum in patients who had pancreatitis when compared with patients who had biliary disease. In patients with idiopathic pancreatitis, the incidence of pancreas divisum was demonstrated to be even higher [21,28]. The 45 percent overall incidence of pancreatitis in patients with pancreas divisum who were encountered by Gregg [17] is similar to the 42.5 percent incidence reported by Tulassay and Papp [22] and Cotton [21]. In a recent large report, however, Delhaye et al [26] have suggested that there is no association between pancreas divisum and pancreatitis. The second question is, Are the symptoms of severe epigastric pain in those patients without demonstrable pancreatitis and pancreas divisum due to relative stenosis of the minor sphincter, or is it a chance association without any direct cause-andeffect relation? The symptoms in the group of patients with pancreas divisum without demonstrable pancreatitis are identical to those in patients with pancreatitis and pancreas divisum; furthermore, Gregg has demonstrated that 37.5 percent of 80 patients with pancreas divisum had obstructive pain in the absence of demonstrable pancreatitis (unpublished observation). Such an association was also reported by Cotton [21]. The inability to pass anything larger than a needle-tipped instrument or fine lacrimal probe into the minor papilla in patients with pancreas divisum indicates that the minor papilla is very small relative to the main pancreatic duct which it drains. A major pancreatic ampulla of such a small size is indicative of marked ampullary stenosis [28], and thus it can be hypothesized that the anomaly
The American Journal of Surgery
Pancreas Divisum
pancreas divisum results in a single outflow tract from the major portion of the pancreas which is inadequate to permit normal pancreatic secretion without causing either a transient or permanent increase in intraductal pressure which may cause typical pancreatic pain [I 73 1. The relatively young age of this group of patients at the time of onset of symptoms (also noted by others [25]) and the absence of alcohol abuse or other causative factors also suggest a specific syndrome. Third, is there a surgical approach which may correct apparent relative stenosis and provide satisfactory palliation of pain in the most severely afflicted patient? The results of sphincteroplasty and sphincterotomy on both major and minor pancreatic duct orifices suggest that these procedures can provide relief of pain in the majority of patients, although significant long-term relief was achieved in only 10 patients in our series. The fact that 14 of 18 survivors experienced at least temporary relief seems to support the hypothesis of relative sphincter stenosis, particularly since the few patients who did not respond well after sphincteroplasty generally had underlying chronic pancreatitis which may have been responsible for their unsatisfactory results until extensive pancreatic resection provided them with significant relief. Similar good to excellent relief of pain in patients with pancreas divisum who underwent sphincteroplasty of both the major and minor sphincters has recently been reported by Cooperman et al [27] and Richter et al [25]. The similarly good results from sphincteroplasty of the minor sphincter only, as reported in the majority of the patients in the series of Richter et al [25], furthers the concept that relative ampullary stenosis is probably responsible for the underlying problem. Recurrence of stenosis of the minor papilla with the accompanying pain were also noted in other series [25,27]. In some instances, recurrent pain was relieved by repeat sphincteroplasty. The fourth question to be addressed: Are there other medical or surgical approaches to this problem? In the entire series of 70 patients with pancreas divisum and associated pain, 2 had permanent relief after endoscopic sphincterotomy of the minor papilla (in one instance after late onset of stenosis at the site of sphincteroplasty). Cotton [28] was also able to provide permanent relief of pain in one patient with pancreas divisum who had undergone endoscopic sphincterotomy. This further substantiates the relation between relative ampullary stenosis and the accompanying abdominal pain. Unfortunately, the number of patients in whom such an approach is technically possible is very small, and endoscopic sphincterotomy should be considered only as an alternate approach, principally after late failure of sphincteroplasty. Distal pancreatic resection and drainage procedures in our experience resulted in
Volume 145, April 1983
transient improvement only, with pain relief lasting from only a few weeks to 2 months, and were related to spontaneous closure of the pancreaticojejunostomy, which was demonstrated either at subsequent surgery or by ERCP. Removal of 90 percent of the pancreas involved with chronic pancreatitis was performed in two patients in whom neither sphincteroplasty nor distal resection and pancreaticojejunostomy had provided good results. Although significantly reduced, both patients continue to have episodic pancreatitis. In a third patient with pancreas divisum and no demonstrable pancreatitis at operation, removal of 90 percent of the pancreas failed to provide good initial relief of pain. It should also be reemphasized that no form of dietary or pharmacologic management led to any significant degree of amelioration of symptoms in the patients in whom surgery was carried out. In the remaining 51 patients in whom the attacks of pain were neither severe nor frequent enough to warrant consideration of surgery, there was no universally successful medical approach to the problem, although in some patients a low-fat diet seemed to lessen the frequency with which episodes of pain occurred. Summary The embryologic defect that results when the ventral and dorsal anlages of the pancreas do not fuse has been referred to as pancreas divisum. ERCP has made it possible to recognize this anomaly in patients undergoing investigation for otherwise unexplained abdominal pain. Of 70 patients in whom recurrent epigastric pain and pancreas divisum coexisted, sphincteroplasty of both papillae was carried out in 19 because of intractability of symptoms. In six patients, surgery was performed subsequent to failure of other biliary tract surgery. There was one postoperative death. In the remaining 18 patients, initial results were good to excellent in 13 and fair in 1. In four patients, however, recurrence of symptoms developed within periods that ranged from 1 to 6 months; therefore, reasonably permanent relief was limited to 10 patients. Of the remaining eight patients with recurrent or continuing symptoms, a variety of subsequent procedures led to satisfactory results in only three. In only seven patients was there even minimal chemical or microscopic evidence to suggest active pancreatitis. Similarly, pancreatograms in 17 patients with this anomaly revealed no abnormalities except for minor ones in 2 patients. Thus, if this is a syndrome that is due to relative stenosis of the lesser papilla and duct, the anomaly does not often result in documented pancreatitis. The definite but limited success rate from sphincteroplasty suggests that relative stenosis of the lesser papilla may be the cause of a syndrome but surgical refinements will be necessary to achieve a better operative success rate.
491
Gregg et al
References 1. Opie E. The anatomy of the pancreas. Johns Hopkins Bull 1903;150:229-32. 2. Opie E. Disease of the pancreas: its cause and nature. Philadelphia: JB Lippincott, 1910:29. 3. Baldwin WM. The pancreatic ducts in man, together with a study of the microscopic structure of the minor duodenal papilla. Anat Ret 1911;5:197-228. 4. Dawson W, Langman J. An anatomical-radiological study on the pancreatic duct pattern in man. Anat Ret 1961;139: 59-68. 5. Kleitsch WP. Anatomy of the pancreas. A study with special reference to the duct system. Arch Surg 1955;71:795802. 6. Reinhoff WF Jr, Pickrell KL. Pancreatitis. An anatomic study of the panaeatic and extrahepatic biliary systems. Arch Surg 1945;51:205-19. 7. Birnstingl MA. A study of pancreatcgraphy. Br J Surg 1959; 47: 128-39. 8. Hand BH. An anatomical study of the choledo&oduodenal area. Br J Surg 1963;50:486-94. 9. Berman LG, Prior JT, Abramow SM, Ziegler DD. A study of the pancreatic duct system in man by the use of vinyl acetate casts of post mortem preparation. Surg Gynecol Obstet 1960;110:391-403. 10. Millbourne E. Calibre and appearance of the pancreatic ducts and relevant clinical problems. A roentgenographic and anatomical study. Acta Chir Stand 1959;118:286-303. 11. Oi I. Techniques of endoscopic pancreatocholangiography (in Japanese). Tokyo: lgaku Shoin, 1973. 12. Kasugai T, Kuno N, Kobayashi S, Hattori K. Endoscopic pancreatocholangiography. 1. The normal pancreatocholangiogram. Gastroenterology 1972;63:217-26. 13. Phillip J, Koch H, Classen M. Variations and anomalies of the papilla of Vater, the pancreas and the biliary duct system. Endoscopy 1974;6:70-7. 14. Varley PF, Rohrmann CA Jr, Silvis SE, Vennes JA. The normal endoscopic pancreatogram. Radiology 1976; 118:296300. 15. Rosch W, Koch H, Schaffner 0, Demling L. The clinical significance of pancreas divisum. Gastrointest Endos 1970; 22:206-7. 16. Belber JP, Bill K. Fusion anomalies of the pancreatic ductal system. Differentiation from pathologic states. Radiology 1977;122:637-42. 17. Gregg JA. Pancreas divisum. Its association with pancreatitis. Am J Surg 1977;134:539-43. 18. Cotton PB, Kizu M. Malfusion of ths dorsal and ventral pancreas. A cause of pancreatitis? Gut 1977;18:A400. 19. Kruse A. Pancreas divisum. A significantly high incidence in chronic pancreatitis? Stand J Gastroenterol 1977;(suppl) 12:45-52. 20. Miichell CJ, Lintott DJ, Ruddell WSJ, Losowsky MS, Axon ATR. Clinical relevance of an unfused pancreatic duct system. Gut 1979;20:1066-71. 2 1. Cotton PB. Congenital anomaly of pancreas divisum as cause of obstructive pain and pancreatitis. Gut 1980;21:105-14. 22. Tulassy Z, Papp J. New clinical aspects of pancreas divisum. Gastrointest Endosc 1980;26: 143-6. 23. Rey JF, Blanc-Mouille C, Garnier C, Delmont J. Le pancreas divisum; realite anatomique et source de lesion pancreatique. Ann Gastroenterol Hepatol 1978;14:196-7. 24. Heiss FW, Shea JA. Association of pancreatitis and variant ductal ahatomy. Dominant drainage of the duct of Bantorini. Am J Gastroenterol 1978;70:158-62. 25. Richter JA, Schapiro RH. Mulley AG, Warshaw AL. Association of pancreas divisum and pancreatitis, and its treatment by sphincteropw of the accessory ampulla. Gasiroenterology 1981;81:1104-10. 26. Delhaye M. Cremer M. Dunham F. Pancreas divisum and pancreatitis. Gastrointest Endosc 1982;28:A153.
492
27. Cooperman M, Ferrara JJ, Fromkes JJ. Carey LC. Surgical management of pancreas divisum. Am J Surg 1982;143: 107-13. 28. Cotton PB. Duodenoscopic papillotomy at the minor papilla for recurrent dorsal pancreatitis. Endosc Dig 1978;3:27-8.
Discussion John Braasch (Burlington, MA): The area at the termination of the pancreatic and biliary ducts continues to titillate successive generations of surgeons who are faced with explaining and treating upper abdominal pain syndromes and pancreatitis. The new major tool of fiberoptic intubation of the biliary and pancreatic ducts has been most helpful to the biliary pancreatic surgeon, but it has also raised more questions than it has answered. Among these questions has been the recognition of so-called pancreas divisum. This report recalls previous attempts at surgical treatment of biliary dyskinesia, stenosis of the sphincter of Oddi, or papillitis of Oddi’s papilla. It also recalls past controversies regarding sphincterotomy versus sphincteroplasty of Oddi’s sphincter and sectioning of Wirsung’s sphincter. None of these controversies have reached an accepted conclusion and all have foundered on the shoals of a lack of precise recognition of the condition, the failure to apply a treatment to a single homogeneous group of cases, the lack of a physiologic or pathologic measuring stick, and a lack of a recognized standard classification for results obtained at follow-up. These comments apply to other contributions in the literature on this subject, including a communication of mine a number of years ago in which I described 90 patients with stenosis of the sphincter of Oddi or indeterminate pain, who were retreated by sphincterotomy. Satisfactory results were reported in only about 40 percent of these 90 patients, a figure not dissimilar to that in this study. There remains the observation by Gregg and others of an increased incidence of divisum in patients with pancreatitis. It is possible that since the diagnosis of divisum is made due to failure to visualize Wirsung’s duct by retrograde fiberoptic pancreatography, there may be cases of obstruction of Wirsung’s duct included in these cases of divisum, obstruction due to pancreatitis, thus inappropriately swelling the proportion of divisum cases in patients with pancreatitis. Despite these misgivings, the authors have presented a paper that intrigues and serves as fertile ground for the imagination and possible progress in these uncharted areas of upper abdominal pain syndromes. William McDermott (closing): There probably is a role for endoscopic sphincteroplasty in this condition. One of our patients with postoperative recurrence of stenosis had a secondary endoscopic papillotomy which resulted in relief of pain. I doubt however, that in the long run, endoscopic papillotomy will prove to be an effective approach to the problem. In answer to Dr. Braasch’s other question, it would be interesting to approach just the minor papilla surgically and see what this accomplishes. It is very difficult, however, when you are operating on a patient with a rather doubtful syndrome, not to do everything you reasonably can to both delineate the anatomy and decompress the entire pancreatic ductal system; however, it has been done by some individuals, but further experience with this approach will be needed to substantiate the validity of the concept.
The American Journal of Surgery
James A. Gregg, MD, Boston, Massachusetts Anthony P. Monaco, MD, FACS, Boston, Massachusetts William V. McDermott, MD, FACS, Boston, Massachusetts
The embryologic anomaly that results when the ductal systems of the ventral and dorsal anlage of the fetal pancreas fail to fuse was first described by Opie [I] in 1903, who labelled the condition “pancreas divisum” [2]. Although there have been anatomic references to this anomaly in the literature [3-IO], it was the introduction of endoscopic retrograde cholangiopancreatography (ERCP) that made possible the clinical recognition of pancreas divisum [11-271 (Figures 1 and2). Although this anomaly is not uncommon and probably occurs in 3 to 7 percent of the population [ 7,15,17,21,23,25], the association of pancreas div-
isum with otherwise unexplained recurrent episodes of upper abdominal pain consistent with that seen in patients with pancreatitis or obstructive disease of the pancreas or biliary tract led to the suggestion [I 71 that inadequate drainage of the dorsal pancreas through the dorsal pancreatic duct and minute minor sphincter was a cause of pancreatitis, or that pain resulted from either an intermittent or constant increase in pressure within the dorsal duct secondary to relative stenosis of the minor ampulla and impaired drainage. Of 70 patients with pancreas divisum and upper abdominal pain who were followed by one of us (JAG), laparotomy and sphincteroplasty of both the major and minor sphincters were performed in 19 because of severity and intractability of pain. It is these 19 patients who serve as the basis of this re-
port. From the Departments of Surgery and Medicine, Harvard Medical School and the New England Deaconess Hospital, Boston, Massachusetts. Requests for reprints should be addressed to William V. McDermott, MD, New England Deaconess Hospital, 185 Pilgim Road, Boston, Massachusetts 02215. Presented at the 63rd Annual Meeting of the New England Surgical So. ciety, Bretton Woods, New Hampshire, October 15-17. 1982.
488
Material and Methods There were 16 women and 3 men in this series. Their ages ranged from 15 to 44 years (median 26 years). All patients could date the onset of their problem to before age 40, and in nine patients pain began before age 20. Seven of these patients had chemical, microscopic, or radiologic evidence of pancreatitis. The symptoms were similar in all patients, consisting of episodic or constant epigastric pain which radiated directly through to the back. It was often but not invariably precipitated by consumption of foods that were high in fat. The pain frequently caused patients to awake from sleep. In none of the patients was the pain controlled to any reasonable degree by dietary management or utilization of a number of different medications, and only one patient in the entire group could be characterized as having an excessive intake of alcohol. Because of the suggested relation of pancreas divisum to pancreatitis and obstructive pain [17], indications of the association were sought in elevations of serum amylase or lipase levels, pancreatic abnormalities observed by ultrasonography, ERCP, and in histologic abnormalities in pancreatic tissue removed at surgery. Seven patients had chemical and microscopic evidence that supported a diagnosis of pancreatitis, and in three patients pancreatic enlargement compatible with the diagnosis of pancreatitis was present on ultrasonogram. Four of the patients were considered to have chronic pancreatitis. The diagnosis of pancreas divisum was established during ERCP in 16 patients by opacification of the duct of the pancreatic head, which was invariably short, averaging only a few centimeters in length and draining only a small segment of the ventral pancreas. In one patient the diagnosis of pancreas divisum was established by opacifying the dorsal duct during ERCP, and in the remaining two patients who were thought to have a vestigial ventral duct, the diagnosis was made by the administration of secretin which demonstrated pancreatic secretions issuing from the minor papilla only. Although the minute size of the dorsal papilla and its more cephalad location made it more difficult to obtain a pancreatogram, the dorsal panThe American Journal of Surgery
Pancreas Divisum
PANCREAS DIVISUM
Flgun, 1. 7% anomaly refermd to as pancreas dlvtsum.
creatic duct was opacified during ERCP in six patients and,
with the exception of a serrated duct in one patient, the other endoscopic pancreatograms of the dorsal duct did not reveal any abnormalities, except delayed drainage.
Surgical Procedures and Results In 19 patients the severity of symptoms was the indication for laparotomy. A standard procedure was devised which consisted of exploration of the common bile duct with cholecystectomy (unless the gallbladder had been previously removed), duodenotomy, and transduodenal division of the major and minor papillae with careful sphincteroplasties using 5-O chromic catgut or polyglycolic acid sutures (Figure 3). The administration of secretin as an intravenous bolus proved to be a useful adjunct to palpation for identification of both papillae. Magnification was used during reconstruction of the outlet through the minor papilla. In one instance, a standard transduodenal sphincterotomy and sphincteroplasty were carried out on the outlets of the common bile duct and the major pancreatic ducts, and exploration and pancreatography (when indicated) of the ventral and dorsal ducts confirmed that in each patient only a very small segment of pancreas was drained through the major papilla, and that the dorsal duct was the principal pancreatic duct. Pancreatograms of the dorsal duct were obtained during surgery and demonstrated normalVolume 145, Apll 1983
Fipm 2. Endoscopk retrograde cholangkpancreatogram of a dMWlLLetY,lXN7lWloomnOn bne&ctamI patienlwithpsmeas small ventral pancreatic duct. Right, &rsal pancreatk duct opactfted through the minor papllla.
sized ducts, with the exception of one patient in whom the duct was dilated to 5 mm. Severe hemorrhagic pancreatitis developed postoperatively in one patient who subsequently died from complications of this disorder; it was thought that the pancreatitis might have been related to the instillation of contrast material during operative pancreatography, but other operative pancreatograms were obtained without incident. In the remaining 18 patients results were good to excellent in 13, fair in 1, and poor (no satisfactory relief of symptoms) in 4. Excellent results were those in which the patient was symptom-free at follow-up, good results were those in which rare, mild episodes of pain were reported at follow-up, and fair results were those in which some decrease in pain was reported but it was still a significant problem. Recurrence of pain developed in four patients, however within a period that ranged from 1 to 6 months; reasonably long-term relief from the initial procedure was limited to 10 patients. Of the remaining eight patients with recurrent or continuing symptoms, a variety of subsequent procedures on the pancreas led to satisfactory results in only three. In one of the patients with recurrent disease, endoscopy demonstrated a recurrence of stenosis of the minor papilla; in this instance, distal pancreatectomy with drainage did not result in further improvement. Other follow-up observations by endoscopy or at subsequent surgery were not sufficient to determine a consistent reason for treatment failure, such as restenosis. In at least three patients, the opening through the minor papilla was widely patent although pain recurred. 489
Gregg et al
Ductus Santarini - - /
Ampulla of Voter \ / ’
i Ductus
F&we 3. Surgical procedure for transduodenal sphlncieroplasty.
Thus, the procedure of transduodenal sphincteroplasty appeared to achieve excellent initial results in terms of pain relief, but in 4 patients this was shortlived, and in only 10 patients who remained pain-free for.periods that ranged from 6 months to 9 years could the procedure be termed an unqualified success (Figure 4). Comments Clearly there are a number of questions that arise which relate to this study and other reports in the literature. First, is this anomaly associated with an
19
PATIENTS
i 18
EXCELLENTTO
GOOD
I
DEATH
I
PATIENTS
FAIR
POOR
RECU:;,
8 PATiENTS
FURTHER
; FAILED 3
&RATION
IMPROVED 3
Resect/on and 0rob70ge fopiuofomy
90 % Resectim
Figure 4. Posfoperative outcome In the 19 study patients.
490
increased incidence of pancreatitis? The studies of Cotton [Z], Richter et al [25], Tulassay and Papp [22], and Gregg’s experience [2 71 all demonstrated a significant increase in the incidence of pancreas divisum in patients who had pancreatitis when compared with patients who had biliary disease. In patients with idiopathic pancreatitis, the incidence of pancreas divisum was demonstrated to be even higher [21,28]. The 45 percent overall incidence of pancreatitis in patients with pancreas divisum who were encountered by Gregg [17] is similar to the 42.5 percent incidence reported by Tulassay and Papp [22] and Cotton [21]. In a recent large report, however, Delhaye et al [26] have suggested that there is no association between pancreas divisum and pancreatitis. The second question is, Are the symptoms of severe epigastric pain in those patients without demonstrable pancreatitis and pancreas divisum due to relative stenosis of the minor sphincter, or is it a chance association without any direct cause-andeffect relation? The symptoms in the group of patients with pancreas divisum without demonstrable pancreatitis are identical to those in patients with pancreatitis and pancreas divisum; furthermore, Gregg has demonstrated that 37.5 percent of 80 patients with pancreas divisum had obstructive pain in the absence of demonstrable pancreatitis (unpublished observation). Such an association was also reported by Cotton [21]. The inability to pass anything larger than a needle-tipped instrument or fine lacrimal probe into the minor papilla in patients with pancreas divisum indicates that the minor papilla is very small relative to the main pancreatic duct which it drains. A major pancreatic ampulla of such a small size is indicative of marked ampullary stenosis [28], and thus it can be hypothesized that the anomaly
The American Journal of Surgery
Pancreas Divisum
pancreas divisum results in a single outflow tract from the major portion of the pancreas which is inadequate to permit normal pancreatic secretion without causing either a transient or permanent increase in intraductal pressure which may cause typical pancreatic pain [I 73 1. The relatively young age of this group of patients at the time of onset of symptoms (also noted by others [25]) and the absence of alcohol abuse or other causative factors also suggest a specific syndrome. Third, is there a surgical approach which may correct apparent relative stenosis and provide satisfactory palliation of pain in the most severely afflicted patient? The results of sphincteroplasty and sphincterotomy on both major and minor pancreatic duct orifices suggest that these procedures can provide relief of pain in the majority of patients, although significant long-term relief was achieved in only 10 patients in our series. The fact that 14 of 18 survivors experienced at least temporary relief seems to support the hypothesis of relative sphincter stenosis, particularly since the few patients who did not respond well after sphincteroplasty generally had underlying chronic pancreatitis which may have been responsible for their unsatisfactory results until extensive pancreatic resection provided them with significant relief. Similar good to excellent relief of pain in patients with pancreas divisum who underwent sphincteroplasty of both the major and minor sphincters has recently been reported by Cooperman et al [27] and Richter et al [25]. The similarly good results from sphincteroplasty of the minor sphincter only, as reported in the majority of the patients in the series of Richter et al [25], furthers the concept that relative ampullary stenosis is probably responsible for the underlying problem. Recurrence of stenosis of the minor papilla with the accompanying pain were also noted in other series [25,27]. In some instances, recurrent pain was relieved by repeat sphincteroplasty. The fourth question to be addressed: Are there other medical or surgical approaches to this problem? In the entire series of 70 patients with pancreas divisum and associated pain, 2 had permanent relief after endoscopic sphincterotomy of the minor papilla (in one instance after late onset of stenosis at the site of sphincteroplasty). Cotton [28] was also able to provide permanent relief of pain in one patient with pancreas divisum who had undergone endoscopic sphincterotomy. This further substantiates the relation between relative ampullary stenosis and the accompanying abdominal pain. Unfortunately, the number of patients in whom such an approach is technically possible is very small, and endoscopic sphincterotomy should be considered only as an alternate approach, principally after late failure of sphincteroplasty. Distal pancreatic resection and drainage procedures in our experience resulted in
Volume 145, April 1983
transient improvement only, with pain relief lasting from only a few weeks to 2 months, and were related to spontaneous closure of the pancreaticojejunostomy, which was demonstrated either at subsequent surgery or by ERCP. Removal of 90 percent of the pancreas involved with chronic pancreatitis was performed in two patients in whom neither sphincteroplasty nor distal resection and pancreaticojejunostomy had provided good results. Although significantly reduced, both patients continue to have episodic pancreatitis. In a third patient with pancreas divisum and no demonstrable pancreatitis at operation, removal of 90 percent of the pancreas failed to provide good initial relief of pain. It should also be reemphasized that no form of dietary or pharmacologic management led to any significant degree of amelioration of symptoms in the patients in whom surgery was carried out. In the remaining 51 patients in whom the attacks of pain were neither severe nor frequent enough to warrant consideration of surgery, there was no universally successful medical approach to the problem, although in some patients a low-fat diet seemed to lessen the frequency with which episodes of pain occurred. Summary The embryologic defect that results when the ventral and dorsal anlages of the pancreas do not fuse has been referred to as pancreas divisum. ERCP has made it possible to recognize this anomaly in patients undergoing investigation for otherwise unexplained abdominal pain. Of 70 patients in whom recurrent epigastric pain and pancreas divisum coexisted, sphincteroplasty of both papillae was carried out in 19 because of intractability of symptoms. In six patients, surgery was performed subsequent to failure of other biliary tract surgery. There was one postoperative death. In the remaining 18 patients, initial results were good to excellent in 13 and fair in 1. In four patients, however, recurrence of symptoms developed within periods that ranged from 1 to 6 months; therefore, reasonably permanent relief was limited to 10 patients. Of the remaining eight patients with recurrent or continuing symptoms, a variety of subsequent procedures led to satisfactory results in only three. In only seven patients was there even minimal chemical or microscopic evidence to suggest active pancreatitis. Similarly, pancreatograms in 17 patients with this anomaly revealed no abnormalities except for minor ones in 2 patients. Thus, if this is a syndrome that is due to relative stenosis of the lesser papilla and duct, the anomaly does not often result in documented pancreatitis. The definite but limited success rate from sphincteroplasty suggests that relative stenosis of the lesser papilla may be the cause of a syndrome but surgical refinements will be necessary to achieve a better operative success rate.
491
Gregg et al
References 1. Opie E. The anatomy of the pancreas. Johns Hopkins Bull 1903;150:229-32. 2. Opie E. Disease of the pancreas: its cause and nature. Philadelphia: JB Lippincott, 1910:29. 3. Baldwin WM. The pancreatic ducts in man, together with a study of the microscopic structure of the minor duodenal papilla. Anat Ret 1911;5:197-228. 4. Dawson W, Langman J. An anatomical-radiological study on the pancreatic duct pattern in man. Anat Ret 1961;139: 59-68. 5. Kleitsch WP. Anatomy of the pancreas. A study with special reference to the duct system. Arch Surg 1955;71:795802. 6. Reinhoff WF Jr, Pickrell KL. Pancreatitis. An anatomic study of the panaeatic and extrahepatic biliary systems. Arch Surg 1945;51:205-19. 7. Birnstingl MA. A study of pancreatcgraphy. Br J Surg 1959; 47: 128-39. 8. Hand BH. An anatomical study of the choledo&oduodenal area. Br J Surg 1963;50:486-94. 9. Berman LG, Prior JT, Abramow SM, Ziegler DD. A study of the pancreatic duct system in man by the use of vinyl acetate casts of post mortem preparation. Surg Gynecol Obstet 1960;110:391-403. 10. Millbourne E. Calibre and appearance of the pancreatic ducts and relevant clinical problems. A roentgenographic and anatomical study. Acta Chir Stand 1959;118:286-303. 11. Oi I. Techniques of endoscopic pancreatocholangiography (in Japanese). Tokyo: lgaku Shoin, 1973. 12. Kasugai T, Kuno N, Kobayashi S, Hattori K. Endoscopic pancreatocholangiography. 1. The normal pancreatocholangiogram. Gastroenterology 1972;63:217-26. 13. Phillip J, Koch H, Classen M. Variations and anomalies of the papilla of Vater, the pancreas and the biliary duct system. Endoscopy 1974;6:70-7. 14. Varley PF, Rohrmann CA Jr, Silvis SE, Vennes JA. The normal endoscopic pancreatogram. Radiology 1976; 118:296300. 15. Rosch W, Koch H, Schaffner 0, Demling L. The clinical significance of pancreas divisum. Gastrointest Endos 1970; 22:206-7. 16. Belber JP, Bill K. Fusion anomalies of the pancreatic ductal system. Differentiation from pathologic states. Radiology 1977;122:637-42. 17. Gregg JA. Pancreas divisum. Its association with pancreatitis. Am J Surg 1977;134:539-43. 18. Cotton PB, Kizu M. Malfusion of ths dorsal and ventral pancreas. A cause of pancreatitis? Gut 1977;18:A400. 19. Kruse A. Pancreas divisum. A significantly high incidence in chronic pancreatitis? Stand J Gastroenterol 1977;(suppl) 12:45-52. 20. Miichell CJ, Lintott DJ, Ruddell WSJ, Losowsky MS, Axon ATR. Clinical relevance of an unfused pancreatic duct system. Gut 1979;20:1066-71. 2 1. Cotton PB. Congenital anomaly of pancreas divisum as cause of obstructive pain and pancreatitis. Gut 1980;21:105-14. 22. Tulassy Z, Papp J. New clinical aspects of pancreas divisum. Gastrointest Endosc 1980;26: 143-6. 23. Rey JF, Blanc-Mouille C, Garnier C, Delmont J. Le pancreas divisum; realite anatomique et source de lesion pancreatique. Ann Gastroenterol Hepatol 1978;14:196-7. 24. Heiss FW, Shea JA. Association of pancreatitis and variant ductal ahatomy. Dominant drainage of the duct of Bantorini. Am J Gastroenterol 1978;70:158-62. 25. Richter JA, Schapiro RH. Mulley AG, Warshaw AL. Association of pancreas divisum and pancreatitis, and its treatment by sphincteropw of the accessory ampulla. Gasiroenterology 1981;81:1104-10. 26. Delhaye M. Cremer M. Dunham F. Pancreas divisum and pancreatitis. Gastrointest Endosc 1982;28:A153.
492
27. Cooperman M, Ferrara JJ, Fromkes JJ. Carey LC. Surgical management of pancreas divisum. Am J Surg 1982;143: 107-13. 28. Cotton PB. Duodenoscopic papillotomy at the minor papilla for recurrent dorsal pancreatitis. Endosc Dig 1978;3:27-8.
Discussion John Braasch (Burlington, MA): The area at the termination of the pancreatic and biliary ducts continues to titillate successive generations of surgeons who are faced with explaining and treating upper abdominal pain syndromes and pancreatitis. The new major tool of fiberoptic intubation of the biliary and pancreatic ducts has been most helpful to the biliary pancreatic surgeon, but it has also raised more questions than it has answered. Among these questions has been the recognition of so-called pancreas divisum. This report recalls previous attempts at surgical treatment of biliary dyskinesia, stenosis of the sphincter of Oddi, or papillitis of Oddi’s papilla. It also recalls past controversies regarding sphincterotomy versus sphincteroplasty of Oddi’s sphincter and sectioning of Wirsung’s sphincter. None of these controversies have reached an accepted conclusion and all have foundered on the shoals of a lack of precise recognition of the condition, the failure to apply a treatment to a single homogeneous group of cases, the lack of a physiologic or pathologic measuring stick, and a lack of a recognized standard classification for results obtained at follow-up. These comments apply to other contributions in the literature on this subject, including a communication of mine a number of years ago in which I described 90 patients with stenosis of the sphincter of Oddi or indeterminate pain, who were retreated by sphincterotomy. Satisfactory results were reported in only about 40 percent of these 90 patients, a figure not dissimilar to that in this study. There remains the observation by Gregg and others of an increased incidence of divisum in patients with pancreatitis. It is possible that since the diagnosis of divisum is made due to failure to visualize Wirsung’s duct by retrograde fiberoptic pancreatography, there may be cases of obstruction of Wirsung’s duct included in these cases of divisum, obstruction due to pancreatitis, thus inappropriately swelling the proportion of divisum cases in patients with pancreatitis. Despite these misgivings, the authors have presented a paper that intrigues and serves as fertile ground for the imagination and possible progress in these uncharted areas of upper abdominal pain syndromes. William McDermott (closing): There probably is a role for endoscopic sphincteroplasty in this condition. One of our patients with postoperative recurrence of stenosis had a secondary endoscopic papillotomy which resulted in relief of pain. I doubt however, that in the long run, endoscopic papillotomy will prove to be an effective approach to the problem. In answer to Dr. Braasch’s other question, it would be interesting to approach just the minor papilla surgically and see what this accomplishes. It is very difficult, however, when you are operating on a patient with a rather doubtful syndrome, not to do everything you reasonably can to both delineate the anatomy and decompress the entire pancreatic ductal system; however, it has been done by some individuals, but further experience with this approach will be needed to substantiate the validity of the concept.
The American Journal of Surgery