- Email: [email protected]
The role of ERCP in the diagnosis and treatment of idiopathic recurrent pancreatitis in children and adolescents
0016-5107/94/4004-0428$3.00 + 0 GASTROINTESTINAL ENDOSCOPY Copyright©1994 by the American Society of Gastrointestinal Endoscopy
The role of ERCP in the diagnosis and treatment of idiopathic recurrent pancreatitis in children and adolescents Moises Guelrud, MD, Cesar Mujica, MD, Domingo Jaen, MD Jose Plaz, MD, Josabel Arias, MD Caracas, Venezuela
ERCP was performed in 51 patients, aged 1 to 18 years, as part of diagnostic evaluation for recurrent pancreatitis without obvious cause or as therapeutic intervention. ERCP was successful in 50 patients (98%). Thirty-four patients (68%) had anatomic findings indicating a possible cause of recurrent pancreatitis. Three patients (6%) had findings suggestive of sphincter of Oddi dysfunction. Eighteen of the 37 patients (49%) with ductal abnormalities underwent endoscopic therapy, with a favorable outcome in 15 (83%). Eleven patients were treated surgically, and 8 of these patients (73%) improved symptomatically. Eight patients received no treatment, and 6 of them (67%) had recurrent bouts of clinical pancreatitis. One mild case of pancreatitis (1.9%) occurred after ERCP, and 3 mild cases followed endoscopic therapy. In conclusion, ERCP is a relatively safe technique that produces opacification of the desired ductal system with a high degree of accuracy and provides useful information in the evaluation of children with idiopathic recurrent pancreatitis. These data suggest that endoscopic pancreatic therapy may result in symptomatic improvement, eliminating the need for surgery in selected children. Furthermore, the study demonstrates that manipulation of the pancreatic duct is comparatively safe and less hazardous than formerly believed. (Gastrointest Endosc 1994;40:428-36.)
Recurrent pancreatitis is rarely recognized in children because the index of clinical suspicion for this entity is usually low. The disease is characterized by episodes of acute pancreatitis, manifested by unexplained recurrent bouts of abdominal pain, with intervening asymptomatic intervals of varying duration. Diagnosis is often difficult and is usually missed for several months or years. Patients may be symptomatically treated during the acute attack, and if no exact cause is found, they are discharged without specific treatment. Two general categories of recurrent pancreatitis have been described, depending on whether Received September 9, 1993. For revision December 15, 1993. Accepted February 2, 1994. From the Department of Medicine, Division of Gastroenterology, Hospital General del Oeste, Caracas, Venezuela. Reprint requests: Moises Guelrud, MD, Policlinica Metropolitana, Urb. Caurimare, Caracas, Venezuela. 37/1/55410 428
the cause is non obstructive or obstructive. Nonobstructive causes include hereditary factors, cystic fibrosis, hyperlipidemia, trauma, medication, and hypercalcemia. 1-6 Obstructive causes include choledochal cysts, pancreas divisum, duodenal diverticulum, duodenal duplication, parasitic infestation, and anomalous pancreaticobiliary junction.7-16 If no specific cause is identified, a case may be classified as idiopathic recurrent pancreatitis. New insights into the ductal system through ERCP have increasingly directed our attention to morphologic changes ofthe pancreatic duct. ERCP offers the definition of the common bile duct and pancreatic ductal system that is necessary to make decisions in the management of children with idiopathic recurrent pancreatitis. Whatever the underlying cause of pancreatitis, the possibility of an anatomic abnormality amenable to endoscopic therapy or surgery should always be considered. 17 ERCP has been found useful in the identification of treatable causes in 40% of adults 18 and in 40% to 75% of children19 -21 GASTROINTESTINAL ENDOSCOPY
Table 1. ERCP in pancreatic disease Number of patients (%)
ERCP findings
Pancreatitis associated with biliary anomalies 9 (18) Choledochal cyst 3 (6) Anomalous pancreaticobiliary junction Pancreatitis associated with pancreatic anomalies 6 (12) Pancreas divisum 1 (2) Pancreatocele Annular pancreas 2 (4) Short pancreas 1 (2) 2 (4) Pancreatitis associated with parasites (Ascaris in common bile duct) 3 (6) Pancreatitis associated with SO dysfunction 7 (14) Chronic pancreatitis Pseudocyst 3 (6) 13 (26) Pancreatitis with no apparent cause (normal pancreatic duct) 50 (100) Total
Sex F
M
Age range (yr)
7 2
2 1
1-18 1-12
3 1 1
3
1
1 1 1
12-17 14 1-18 3 5-17
1
2
13-18
4 2 7
3 1 6
4-16 2-15 7-16
29
21
1-18
Table 2. Symptomatic improvement after endoscopic treatment in 18 patients with recurrent pancreatitis Diagnosis (number of patients) Choledochal cyst type I (1) Choledochal cyst type III (1) Choledochal cyst type IV (1) Pancreas divisum (3) Ascaris in CBD (2) SO dysfunction (3) Chronic pancreatitis (4) Pancreatic pseudocyst (3)
Improvement
Treatment None ES and CBD stone removal ES ES and PD stone removal ES minor papilla Worm removal ES DS DS PD occlusion DS and PD drainage Cystogastrostomy
Total
Moderate
Excellent 1 1 1
1
1 1
2 2 1 1 2
1
2 1 3 (17%)
1 (5%)
14 (78%)
CBD, Common bile duct; PD, pancreatic duct; ES, endoscopic sphincterotomy; DS, dual sphincterotomy.
with idiopathic recurrent pancreatitis. This report summarizes our experience with ERCP in the diagnosis and treatment of children and adolescents with idiopathic recurrent pancreatitis. MATERIAL AND METHODS
Patients
Between 1984 and 1992, a total of 51 patients (20 boys and 31 girls; 26 children and 25 adolescents with an age range of 1 to 18 years) were referred to our center with the presumptive diagnosis of idiopathic recurrent pancreatitis. ERCP was successful in 50 patients (98%). All patients had had two or more documented attacks of acute pancreatitis confirmed by hyperamylasemia. Despite careful clinical histories and laboratory and radiologic examinations, an apparent cause for recurrent pancreatitis could be found in none of the patients. Special attention was given to drug intake, hyperlipidemia, cystic fibrosis, hypercalcemia, collagen disease, infection, abdominal trauma, and hereditary pancreatitis. All paVOLUME 40, NO.4, 1994
tients underwent abdominal ultrasonography, 35 had CT scan, and 20 had upper gastrointestinal barium roentgenography. Endoscopic techniques
After sedation with midazolam (0.1 mg/kg) and meperidine (3 mg/kg), all patients underwent ERCP within 4 weeks of their last episode of pancreatitis. The Olympus prototype PJF pediatric duodenoscope 22 or JF -1 T adult duodenoscope (Olympus America Inc., Lake Success, N.Y.) was used. The papilla of Vater was cannulated with standard or tapered catheters, and the pancreatic and biliary ducts were visualized by fluoroscopy during injection of 60 % iodine contrast material. If a ventral duct was visualized or no pancreatogram was obtained from the major papilla, the minor papilla was cannulated (with or without secretin 1 U/kg) with a needle-tipped catheter. Sphincter of Oddi (SO) manometry was performed in 16 patients in whom obstructive causes could not be found. Studies were performed using a low-compliance infusion pump system and a three-lumen recording catheter introduced 429
creatic exocrine function while leaving intact the endocrine secretion of insulin. (7) Cystogastrostomy was performed using a diathermy needle to puncture the cyst through the gastric wall, after which the opening was enlarged using a sphincterotome and a 10F endoprosthesis was placed. All patients were monitored with pulse oximeter. Informed consent was obtained from each child's parent or guardian. Clinical follow-up
Patients were followed for periods ranging from 4 months to 6 years (mean follow-up time was 20 months). The number of episodes of pancreatitis, epigastric pain, pancreatic enzyme elevation, and emergency visits after treatment was determined and compared with the number before treatment. Improvement was graded at follow-up in the following manner: (1) none, recurrent pancreatitis requiring emergency visits and hospitalization; (2) moderate, occasional pain without pancreatic enzyme elevation and no subsequent episodes of acute pancreatitis; (3) excellent, absence of symptoms and no subsequent episodes of acute pancreatitis.
RESULTS
Figure 1. Choledochal cyst and pancreas divisum in a 5-year-old boy with recurrent pancreatitis. The dorsal pancreas (straight arrow) ends in the middle portion of the common bile duct. A stone (double straight arrows) and a type I choledochal cyst (curved arrow) with a stricture can be seen. The ventral pancreas is also observed (double curved arrow).
into the pancreatic duct through the biopsy channel. Manometric pressure recordings of the SO were performed and analyzed as previously described. 23 The maximum normal SO basal pressure was 35 mm Hg based on adult standards. Patients in whom a correctable cause for pancreatic disease was identified by ERCP were considered candidates for therapeutic endoscopy or surgery. Endoscopic therapy included the following techniques: (1) Standard biliary sphincterotomy was performed using a traction sphincterotome. (2) Dual sphincterotomy of the pancreatic duct sphincter and the common bile duct sphincter were performed in the same session. 24 First, the sphincterotome was inserted in the pancreatic duct. Using only cutting current, a cut was usually made in the l-o'clock direction over a length of 5 to 7 mm, after which the sphincterotome was reinserted in the bile duct. The bile duct was unroofed using a blended current over a length of 5 to 10 mm. (3) Minor papilla sphincterotomy was performed using cutting current in the 12-o'clock direction over a length of 5 to 7 mm with an over-the-wire, short-traction sphincterotome. (4) Pancreatic stone extraction was performed using a retrieval basket. (5) Pancreatic endoprosthesis was inserted by introducing a 5F to 10F stent into the main pancreatic duct or the dorsal pancreatic duct for a period of 10 to 14 days after endoscopic sphincterotomy or for several weeks to drain communicating pancreatic pseudocysts. (6) The pancreatic duct was obstructed with cyanocrilate glue injection (1 m!) into the remnant duct of the head of the pancreas to suppress pan430
Of 50 patients, 34 (68%) had anatomic findings indicating a possible cause of recurrent pancreatitis and 3 (6 %) had findings suggestive of SO dysfunction (significantly elevated SO basal pressure of 57 ± 5.3 mm Hg) (Table 1). Thus, ERCP examination with SO manometry revealed a possible cause of recurrent pancreatitis in 37 of 50 children and adolescents (74 %) with a diagnosis of idiopathic recurrent pancreatitis by conventional criteria. Eighteen of the 37 patients (49 %) in whom a correctable cause for pancreatic disease was found were treated by endoscopic therapy (Table 2). Overall, 15 patients (83 %) were considered markedly improved. Eight of11 patients (73 % ) treated surgically improved symptomatically. Of eight patients who received no treatment, six (67 %) had recurrent bouts of clinical pancreatitis. Pancreatitis associated with biliary anomalies
Of the nine patients with a choledochal cyst, three had a type I choledochal cyst, characterized by diffuse choledochal dilatation. One patient had coexistent pancreas divisum, pancreatic duct anomalies, and a stone in the common bile duct (Fig. 1). He was treated by endoscopic sphincterotomy of the major papilla and stone extraction, followed by operative cyst resection with hepatojejunostomy. Results were excellent at 4-year follow-up. Two patients with diffuse choledochal dilatation and extensive pancreatic duct anomalies were treated surgically by cyst resection and hepatojejunostomy. One remained asymptomatic for 14 months, and the other had an episode of acute pancreatitis 8 months after surgery. Five patients had a type IV choledochal cyst, characterized by multiple intrahepatic and extrahepatic cysts. Four had coexistent pancreatic duct anomalies, two of them with pancreatic stones. One patient refused surgery and was treated by endoscopic sphincterotomy and pancreatic duct stone extraction (Fig. 2). She remained symptomGASTROINTESTINAL ENDOSCOPY
Figure 2, Cont'd. C, The pancreatic stone is removed.
free for a period of 2 years. One patient was treated surgically by cyst resection and hepatojejunostomy with surgical sphincteroplasty and pancreatic stone removal. During a follow-up of 5 years, the patient had multiple episodes of upper abdominal pain without elevation of amylase. Three patients were treated surgically by cyst resection and hepatojejunostomy. Acute pancreatitis developed in one patient 7 months after surgery, and two patients remained asymptomatic after 8 and 13 months. One patient with a type III choledochal cyst (choledochocele) was treated by endoscopic sphincterotomy. She remained symptom-free for a period of 3 years. Of three patients with an anomalous pancreaticobiliary junction, two refused treatment. One of them had one and the other had three episodes of acute pancreatitis during the following 10 months. One patient underwent surgical sphincteroplasty and remained asymptomatic for 2 years. Pancreatitis associated with pancreatic anomalies
Figure 2. Choledochal cyst (type IV) in a 6-year-old patient with recurrent pancreatitis. A, An anomalous pancreaticobiliary junction with a pancreatic stone (arrow) is seen. B, After endoscopic sphincterotomy, a retrieval balloon is inserted (arrow).
Pancreas divisum was diagnosed in six patients (12 %). The ventral pancreatic duct was visualized during ERCP in all patients. However, the dorsal pancreas could be observed in only four patients (Fig. 3). Three patients were treated by surgical sphincteroplasty of both the minor and major papillas. Two remained asymptomatic for 2 years and one had several episodes of epigastric pain without pancreatitis.
Figure 5. Short pancreas in a 3-year-old patient with
recurrent pancreatitis. The pancreatic duct and the accessory duct are dilated. The patient with pancreatocele had another episode of acute pancreatitis after 5 months; the patient with short pancreas remained asymptomatic for 3 years.
Pancreatitis associated with parasitic infestation
Figure 3. Pancreas divisum in a 12-year-old patient with
recurrent pancreatitis. The ventral pancreas is seen with an acinogram (straight arrow). The dorsal pancreas is dilated with multiple strictures (curved arrow).
Two patients with Ascaris in the common bile duct had the worm removed endoscopically with a tripod device and were free of symptoms after 2 and 4 years.
Pancreatitis associated with SO dysfunction Three patients had SO dysfunction and normal ERCP. Two were treated by endoscopic sphincterotomy of the common bile duct sphincter. One remained asymptomatic after a follow-up of 18 months, and the other had an episode of acute pancreatitis 5 months after therapy. Both patients refused further treatment. The third patient had a dual sphincterotomy and remained symptom-free for a period of 4 years.
Chronic pancreatitis and pancreatic pseudocyst
Figure 4. Cystic dilatation of the pancreatic duct (pancreatocele) in a 14-year-old patient with recurrent pancreatitis.
Three patients were treated by endoscopic sphincterotomy of the minor papilla. In one, a pancreatic stent was placed for 2 weeks. Two were asymptomatic during follow-up of 10 and 17 months and one had recurrent episodes of abdominal pain without enzyme elevation. Two patients with annular pancreas, one with cystic dilatation of the pancreatic duct (pancreatocele) (Fig. 4), and one with short pancreas (Fig. 5) refused treatment. One patient with annular pancreas had an episode of acute pancreatitis during the following 13 months; the other remained asymptomatic for 2 years. 432
Seven patients had chronic pancreatitis. Six of them had a dilated pancreatic duct. Four patients received pancreatic therapy. Three underwent a dual sphincterotomy before pancreatic duct endoprosthesis placement. The stent was removed during the following 2 weeks, and the patients experienced dramatic clinical relief of pain. Two remained asymptomatic after a follow-up of 2 years, and one had intermittent abdominal pain with an episode of acute pancreatitis in the ensuing 14 months. Two patients received no treatment and had recurrent bouts of clinical pancreatitis during the following 4 and 11 months. The other patient had a narrowed and irregular pancreatic duct. She had seven episodes of acute pancreatitis, after which pain lasted for several weeks. She had become addicted to narcotics and was treated by partial pancreatectomy of the body and tail. After 11 months she had another episode of acute pancreatitis and was treated endoscopically by cyanocrilate injection in the pancreatic duct remnant. After 2 years without sympGASTROINTESTINAL ENDOSCOPY
toms, another episode of acute pancreatitis occurred and the cyanocrilate injection was repeated. She has remained symptom-free for a period of 2.5 years. Three patients had pancreatic pseudocysts, two of which were communicating with the pancreatic duct and were treated by dual sphincterotomy and placement of a transpapillary pancreatic stent until the cysts resolved. Pain disappeared immediately. After follow-up of 2 and 3 years, both patients are alive and without clinical problems. One 2-year-old child had a noncommunicating pseudocyst that was treated by cystogastrostomy; however, the drainage failed for technical reasons and the patient underwent surgical cystojejunostomy. Pancreatitis with no apparent cause
Thirteen patients had a normal pancreaticobiliary duct system with normal SO motor function. They received no treatment. Eight patients had recurrent bouts of clinical pancreatitis during the following 2 years. Follow-up ERCP and manometry performed in three patients revealed no abnormalities. Five patients were lost for follow-up. Complications
One mild case of pancreatitis related to ERCP occurred in a patient with SO dysfunction (1.9%). Three episodes of pancreatitis occurred after endoscopic therapy in 18 patients (16.6%). One episode was in a patient with pancreas divisum treated by minor papilla sphincterotomy without pancreatic stent placement. The second was in a patient with SO dysfunction treated by dual sphincterotomy, and the third in a patient with chronic pancreatitis in whom cyanocrilate was injected into the pancreatic duct. All three patients responded to conservative management and were discharged from the hospital within 3 days. None of these 18 patients died in the 30 days after endoscopic treatment. DISCUSSION
In this study, we evaluated by ERCP and SO manometry a highly selected group of patients with recurrent pancreatitis in whom no specific cause had been identified. Additionally, only patients who had experienced at least two episodes of pancreatitis were included. ERCP and SO manometry revealed a possible cause of recurrent pancreatitis in 37 of 50 patients (74%). Pancreatic endoscopic therapy was performed in 18 patients, with excellent results in 15 patients (83 %). In the present study, diagnostic capability of ERCP is similar to that reported by others. Putnam et al. 19 studied 12 children with idiopathic recurrent pancreatitis; potentially treatable conditions were identified in nine patients (75%) who underwent successful ERCP examination. Buckley and Connon20 performed ERCP in 18 patients with pancreatitis, aged 3 to 18 VOLUME 40, NO.4, 1994
Figure 6. Chronic pancreatitis in an 8-year-old patient. A, A dilated, irregular pancreatic duct is visualized. The primary and secondary pancreatic branches are dilated and tortuous. After a dual sphincterotomy, a guide wire is inserted into the pancreatic duct. B, A 1OF, 5-cm pancreatic stent is inserted and left in place for 2 weeks.
years, and found pancreatic duct abnormalities in eight patients (44 % ). Brown et al. 21 performed 74 ERCPs for pancreatic indications and found pancreatic duct abnormalities in 21 patients (28%). An abnormal common bile duct was found in 13 of 53 patients (24 %) with normal pancreatic duct. In this series, an underlying cause of pancreatitis was identified in 52 % of patients. The success rate of ERCP in our present study was 98%. Our success rate compares favorably with that reported by others in children and adolescents.1 9 -21 , 25-28 Significant complications ofERCP, in433
cluding cholangitis, bacteremia, and pancreatitis, have been rare in childrenP, 22, 23 One of our patients (1.9 %) with SO dysfunction had a mild pancreatitis that resolved in 3 days. Other workers have reported similarly low complication rates in children. 19-21 No deaths have occurred in the studies reported of children and adolescents. Our study confirmed the safety of this procedure in children when performed by an experienced endoscopist. Endoscopic therapy in children has been directed to treatment of biliary diseases. 2o , 21-23, 29 Some authors have expanded therapeutic ERCP in the pediatric age group to treat pancreatic disease. Brown et al. 21 placed multiple pancreatic stents in eight children with pancreas divisum. Al Traiff and Khan 30 endoscopically drained a duodenal duplication cyst in an 11-year-old child who had recurrent attacks of pancreatitis. Lehman et al. 31 performed minor papilla sphincterotomy over a previously placed dorsal pancreatic duct endoprosthesis in a 7-year-old child with pancreas divisum and recurrent pancreatitis. In the present study, 14 of 17 patients (82 %) with recurrent pancreatitis underwent successful endoscopic treatment. These responses were presumably related to relief of obstructive pancreatitis, as has been noted in adults. 32 Choledochal cyst is a congenital malformation of the biliary tract. It is primarily a disease of children and young adults. An abnormal pancreaticobiliary junction has been suggested as an important factor in the pathogenesis. The classification of Todani et al. 33 of bile duct cyst is most often used. ERCP is of significant value in outlining these cysts and their relationship to both ductal systems, providing a roadmap for the surgical approach. 34 ,35 Anomalous pancreaticobiliary junction has been associated with choledochal cysts and recurrent pancreatitis. 36 , 37 The treatment is surgical. However, one of our patients refused surgery. She had a type IV choledochal cyst with a pancreatic duct stone and recurrent pancreatitis. She was treated by endoscopic sphincterotomy followed by pancreatic duct stone extraction; pancreatitis did not recur during the ensuing 2 years. Although classified as one of the forms of choledochal cysts, choledochocele (type III choledochal cyst) is probably not related. Based on autopsy studies, Sterling38 hypothesized that a choledochocele is an acquired evagination of the mucosa of the terminal common bile duct into the duodenum that is caused by an obstruction to bile outflow resulting from an impacted stone, fibrosis, or papillitis. More recently, Venu et al. 39 and Kagiyama et aI.4° suggested that in some patients abnormal SO motility could be a contributing factor in the pathogenesis of choledochocele. It has been reported in association with recurrent pancreatitis. 41 In the past, the treatment has been surgical. However, in selected cases, a choledochocele may be effectively treated by endoscopic sphincterot434
omy.42 One of our patients was treated by endoscopic sphincterotomy, after which she remained symptomfree for a period of 3 years. Pancreas divisum, the most common congenital variant of pancreatic ductal anatomy, has been found in approximately 3% to 13% of ERCP studies. 43 -46 However, when patients with unexplained recurrent pancreatitis were studied, the incidence was 25 %.43 In our small group of patients with recurrent pancreatitis, the incidence was 12 %. The treatment used to be surgical. Warshaw et al. 47 performed minor papilla sphincteroplasty in 88 patients with symptomatic pancreas divisum and recurrent pancreatitis. Overall, 70% improved at a mean follow-up of 53 months. Endoscopic treatment has been directed to decompress the dorsal duct by insertion of endoprosthesis, dilation of the minor papilla, or sphincterotomy of the minor papilla with or without insertion of endoprosthesis. McCarthy et al. 48 treated 19 patients with pancreas divisum using dorsal pancreatic stents and obtained 89 % improvement. Lans et al. 49 reported the results of a randomized control trial of long-term (12 months) stenting of the minor papilla in patients with recurrent pancreatitis. Stents were changed every 4 months. Follow-up continued for at least 12 months after the final stent removal. Patients with stents had fewer episodes of pancreatitis and were more frequently judged to be improved (90% versus 11 % of controls,p < 0.05). A temporary trail of stenting the dorsal duct during ERCP has been used to decide whether surgery is appropriate. Patients who improve after stenting may be candidates for more sustained relief with surgical decompression. 50 Minor papilla sphincterotomy has been used to treat pancreas divisum. 51 Soehendra et al. 52 treated six patients using a precut or needle-knife sphincterotome over a stent, obtaining 83 % symptomatic improvement. Liguory et al. 53 performed minor papilla sphincterotomy with (n = 2) and without (n = 8) dorsal pancreatic duct stenting in 10 patients with pancreatitis. Seven (70%) remained symptomfree during a 3- to 30-month follow-up period. Lehman et al. 31 treated 52 patients with pancreas divisum by minor papilla sphincterotomy performed with a needle knife over a previously placed dorsal pancreatic duct stent. Patients with recurrent acute pancreatitis benefited more frequently than those with chronic pancreatitis (76 % versus 27 %, p = 0.01). Pancreatitis as a complication of therapy occurred after 13 % of the procedures but was mild and could be managed conservatively. We treated three patients with minor papilla sphincterotomy and obtained excellent results in two and a moderate result in one. In adults, the leading cause of idiopathic recurrent pancreatitis appears to be SO dysfunction. Venu et aU 8 studied 116 patients with recurrent pancreatitis and found 17 patients (15 %) with SO dysfunction. Manometry is the best diagnostic procedure and eleGASTROINTESTINAL ENDOSCOPY
vated SO basal pressure the major criterion for diagnosis. In the present series, SO dysfunction represented 6 % of the causes of recurrent pancreatitis. Three of 16 children (18%) with normal ERCP findings had SO dysfunction. These results are in accordance with those reported by Brown et al. 21 who found SO dysfunction in three of 13 children (23 %) with recurrent pancreatitis. These patients are generally treated by endoscopic sphincterotomy or the classic transduodenal sphincterotomy. However, patients do not respond well,54 presumably because with either treatment modalities the pancreatic sphincter is not transected, and recurrent attacks of pancreatitis may be attributed to an affected sphincter. A hypertensive pancreatic duct sphincter has been suggested as the cause of these recurrent episodes of pancreatitis. 55 In the present study, two of three patients were treated by standard endoscopic sphincterotomy, with clinical improvement noted in one patient. The third patient was treated by dual sphincterotomy of the pancreatic and the common bile duct sphincters, after which the patient was markedly improved during the ensuing 4 years. In patients with chronic pancreatitis, recurrent attacks may be caused by strictures of the main duct, pancreatic stones, or pseudocysts that can impair the normal outflow of pancreatic juice. In adults, it has been shown that insertion of a pancreatic duct endoprosthesis,32,43 pancreatic stone removal, 56 or pseudocyst drainage 57 can improve ductal drainage. Recently, Kozarek et al. 58 performed pancreatic duct sphincterotomy followed by stone extraction and stent insertion in three children with chronic relapsing pancreatitis, after which marked improvement in chronic abdominal pain was noted. These endoscopic techniques have proved to be good alternatives to surgical drainage. Two of our three patients with chronic pancreatitis treated by dual sphincterotomy and pancreatic stent experienced significant improvement during a 2-year follow-up. These results further support the theory of an obstructive cause of recurrent pancreatitis in some cases. One patient with chronic pancreatitis had recurrent attacks after subtotal pancreatectomy; she was scheduled to undergo total pancreatectomy, but instead cyanocrilate glue was injected into the remnant pancreatic duct to produce atrophy ofthe exocrine gland with preservation of the islets. This technique has been used and abandoned by Rosch,59 who noted poor long-term results. Although most of his patients had an intact pancreas, apparently the instillation pressure via an endoscopic cannula is not strong enough to guarantee complete filling of the main pancreatic duct and the primary and secondary branches. Our patient had had a previous subtotal pancreatectomy so that complete filling of the ductal system was accomplished. Freedom from pain was achieved for 2 years. After a second cyanocrilate injection was performed, she remained symptom-free for 2.5 years. VOLUME 40, NO.4, 1994
Pancreatic pseudocysts are usually detected after an attack of pancreatitis; the incidence is approximately 20 %. Most pseudo cysts resolve spontaneously, but drainage procedures are indicated for persistent, sym ptomatic cysts. In adults, the classic treatment has been surgical drainage, and more recently endoscopic drainage has been reported. 57 In this study, two children with nonresolving communicating pseudo cysts became asymptomatic after transampullary endoprosthesis placement; the pseudocysts resolved rapidly. In the third patient, a transmural drainage via the gastric wall was unsuccessful because of technical problems. In adults, short-term complications of endoscopic sphincterotomy, such as bleeding, pancreatitis, cholangitis, and retroperitoneal perforation, occur in 7 % to 10% of patients. 6o Complications of endoscopic sphincterotomy in children has been sparsely reported. Buckley and Connon 20 successfully performed endoscopic sphincterotomy with common duct stone extraction in five children without significant complications, except for bleeding in one patient who had been given aspirin after the procedure. We previously reported 26 six children with common duct stones in whom endoscopic sphincterotomy caused no complications. Kozarek et a1. 58 reported mild pancreatitis in three children with chronic pancreatitis after pancreatic therapy. We had three episodes (17%) of mild acute pancreatitis after pancreatic therapy. In conclusion, we have demonstrated that ERCP in children and adolescents is a safe diagnostic procedure that delineates the anatomy of the pancreatic and biliary ductal systems and identifies patients who are amenable to endoscopic or surgical treatment. When performed by experienced endoscopists, therapeutic pancreatography can be successfully performed in a selected group of children with a low rate of complications. A longer follow-up period will be necessary to determine whether endoscopic success produces longstanding clinical improvement. ACKNOWLEDGMENT
The authors thank Adriana Maixner for her assistance in reviewing the manuscript.
REFERENCES 1.
Hendren WH, Greep JM, Patton AS. Pancreatitis in childhood: experience with 15 cases. Arch Dis Child 1965;40:132-45. 2. Jordan SC, Ament ME. Pancreatitis in children and adolescents. J Pediatr 1977;91:211-6. 3. Ziegler DW, Long JA, Philippart AI, Klein MD. Pancreatitis in childhood: experience with 49 patients. Ann Surg 1988;207:257-
61.
4. Kattwinkel J, Lapey A, Di Sant' Agnese P, Edwards WA. Hereditary pancreatitis: three new kindreds and a critical review of the literature. Pediatrics 1973;51:55-69. 5. Krauss RM, Levy AG. Subclinical chronic pancreatitis in type I hyperlipoproteinemia. Am J Med 1977;62:144-9. 6. Park RW, Grand RJ. Gastrointestinal manifestations of cystic fibrosis: a review. Gastroenterology 1981;81:1143-61. 7. Griffin M, Carey WD, Hermann R, Buonocore E. Recurrent
435
8. 9.
10. 11. 12. 13. 14.
15. 16. 17. 18. 19. 20. 21.
22. 23. 24. 25. 26.
27. 28. 29. 30. 31. 32. 33. 34.
acute pancreatitis and intussusception complicating an intraluminal duodenal diverticulum. Gastroenterology 1981;81:345-8. Wagner CW, Golladay ES. Pancreas divisum and pancreatitis in children. Am Surg 1988;54:22-6. Yaffe MR, Gutenberger JE. Chronic pancreatitis and pancreas divisum in an infant diagnosed by endoscopic retrograde cholangiopancreatography and treated with somatostatin analog. J Pediatr Gastroenterol Nutr 1989;9:108-11. Yedlin ST, Dubois RS, Philippart AI. Pancreas divisum. A cause of pancreatitis in childhood. J Pediatr Surg 1984;19:793-4. Van Der Spuy S. Endoscopic retrograde cholangiopancreatography (ERCP) in children. Endoscopy 1978;10:173-5. Gibson LE, Haller JA. Acute pancreatitis associated with congenital cyst of the common bile duct. J Pediatr 1959;55:650-7. Agrawall RM, Brodmerkel GJ. Choledochal cyst presenting as pancreatitis. Am J GastroenteroI1979;71:408-11. Black PR, Welch KJ, Eraklis AJ. Juxtapancreatic intestinal duplications with pancreatic ductal communication: a cause of pancreatitis and recurrent abdominal pain in childhood. J Pediatr Surg 1986;21:257-61. Lavine JE, Harrison M, Heyman MB. Gastrointestinal duplications causing relapsing pancreatitis in children. Gastroenterology 1989;97:1556-8. Karp MP, Jewett TC, Cooney DR. Chronic relapsing pancreatitis in childhood caused by pancreaticobiliary ductal anomaly. J Pediatr Gastroenterol Nutr 1983;2:324-8. Guelrud M. Endoscopic retrograde cholangiopancreatography in the infant. In: Barkin J, O'Phelan CA, eds. Advanced therapeutic endoscopy. New York: Raven Press, 1990: 335-54. Venu RP, Geenen JE, Hogan W, Stone J, Johnson GK, Soergel K. Idiopathic recurrent pancreatitis. An approach to diagnosis and treatment. Dig Dis Sci 1989;34:56-60. Putnam PE, Kocoshis SA, Orenstein SR, Schade RR. Pediatric endoscopic retrograde cholangiopancreatography. Am J GastroenteroI1991;86:824-30. Buckley A, Connon JJ. The role of ERCP in children and adolescents. Gastrointest Endosc 1990;36:369-72. Brown CW, Werlin SL, Geenen JE, Schmalz M. The diagnostic and therapeutic role of endoscopic retrograde cholangiopancreatography in children. J Pediatr Gastroenterol Nutr 1993;17: 19-23. Guelrud M, Jaen D, Torres P, et al. Endoscopic cholangiopancreatography in the infant: evaluation of a new prototype pediatric duodenoscope. Gastrointest Endosc 1987;33:4-8. Guelrud M, Mendoza S, Rossiter G, Villegas MI. Sphincter of Oddi manometry in healthy volunteers. Dig Dis Sci 1990;35:3846. Fuji T, Amano H, Ohmura R, Akiyama T, Aibe T, Takemoto T. Endoscopic pancreatic sphincterotomy. Technique and evaluation. Endoscopy 1989;21:27-30. Guelrud M, Jaen D, Mendoza S, Plaz J, Torres P. ERCP in the diagnosis of extrahepatic biliary atresia. Gastrointest Endosc 1991;37:522-6. Guelrud M, Mendoza S, Jaen D, Plaz J, Machuca J, Torres P. ERCP and endoscopic sphincterotomy in infants and children with jaundice due to common bile duct stones. Gastrointest Endosc 1992;38:450-3. Cotton PB, Laage NJ. Endoscopic retrograde cholangiopancreatography in children. Arch Dis Child 1982;57:131-6. Allendorph M, Werlin SL, Geenen JE. Endoscopic retrograde cholangiopancreatography in children. J Pediatr 1987;110:20611. Guelrud M, Mendoza S, Zager A, Noguera C. Biliary stenting in an infant with malignant obstructive jaundice. Gastrointest Endosc 1989;35:259-61. Al Traif lA, Khan MH. Endoscopic drainage of a duodenal duplication cyst. Gastrointest Endosc 1992;38:64-5. Lehman GA, Sherman S, Nisi R, Hawes RH. Pancreas divisum: results of minor papilla sphincterotomy. Gastrointest Endosc 1993;39:1-8. Kozarek RA, Patterson DJ, Ball TJ, Traverso LW. Endoscopic placement of pancreatic drains and stents in the management of pancreatitis. Ann Surg 1989;209:261-6. Todani T, Watanabe Y, Narusue M. Congenital bile duct cyst. Am J Surg 1977;134:263-9. Guelrud M, Jaen D, Mendoza S, Torres P. Usefulness of endoscopic retrograde cholangiopancreatography in diag-
436
nosis of choledochal cysts in children. G E N 1989;43:9-12. 35. Thatcher BS, Sivak MV, Hermann RE, Esselstyn CB. ERCP in evaluation and diagnosis of choledochal cyst: report of five cases. Gastrointest Endosc 1986;32:27-31. 36. Atkinson GO, Wyly JB, Gay BB, Ball TI, Winn KJ. Idiopathic fibrosing pancreatitis: a cause of obstructive jaundice in childhood. Pediatr Radiol 1988;18:28-31. 37. Mori K, Nagakawa T, Ohta T, et al. Acute pancreatitis associated with anomalous union of the pancreaticobiliary ductal system. J Clin GastroenteroI1991;13:673-7. 38. Sterling JA. Diverticula in the terminal portion of the common bile duct. Am J Pathol 1949;25:325-35. 39. Venu RP, Geenen JE, Hogan WJ, et al. Role of endoscopic retrograde cholangiopancreatography in the diagnosis and treatment of choledochocele. Gastroenterology 1984;87:1144-9. 40. Kagiyama S, Okazaki K, Yamamoto Y. Anatomic variants of choledochocele and manometric measurements of pressure in the cele and the orifice zone. Am J GastroenteroI1987;82:641-9. 41. Greene FL, Brown JJ, Rubinstein P, Anderson MC. Choledochocele and recurrent pancreatitis. Diagnosis and surgical management. Am J Surg 1985;149:306-9. 42. Siegel JH, Harding GT, Chateau F. Endoscopic incision of choledochal cyst (choledochocele). Endoscopy 1981;13:200-2. 43. Cotton PB. Congenital anomaly of pancreas divisum as a cause of obstructive pain and pancreatitis. Gut 1980;21:105-14. 44. O'Connor KW, Lehman GA. An improved technique for accessory papilla cannulation in pancreas divisum. Gastrointest Endosc 1985;31:13-7. 45. Gregg JA. Pancreas divisum: its association with pancreatitis. Am J Surg 1977;134:539-43. 46. Heiss FN, Shea JA. Association of pancreatitis and various ductal anatomy, dominant drainage of the duct of Santorini. Am J Gastroenterol 1978;70:158-62. 47. Warshaw AL, Simeone JF, Schapiro RH, Flavin-Warshaw B. Evaluation and treatment of the dominant dorsal duct syndrome (pancreas divisum redefined). Am J Surg 1990;159:59-64. 48. McCarthy J, Geenen JE, Hogan WJ. Preliminary experience with endoscopic stent placement in benign pancreatic diseases. Gastrointest Endosc 1988;34:16-8. 49. Lans JI, Geenen JE, Johanson JF, Hogan WJ. Endoscopic therapy in patients with pancreas divisum and acute pancreatitis: a prospective, randomized, controlled clinical trial. Gastrointest Endosc 1992;38:430-4. 50. Siegel JH, Pullano W, Ben-Zvi JS, Cooperman AM. Effectiveness of endoscopic drainage for pancreas divisum. Endoscopy 1990;20:129-32. 51. Cotton PB. Duodenoscopic papillotomy at the minor papilla of Vater for recurrent dorsal pancreatitis. Endosc Dig 1978;3:27 -8. 52. Soehendra N, Kempeneers I, Nam VC, Grimm H. Endoscopic dilatation and papillotomy of the accessory papilla and internal drainage in pancreas divisum. Endoscopy 1986;18:129-32. 53. Liguory C, Lefebvre JF, Carrard JM, Bonnel D, Fritsch J, Etienne JP. Le pancreas divisum: etude clinique et therapeutique chez l'homme: it propos de 87 cas. Gastroenterol Clin Bioi 1986; 10:820-5. 54. Hogan WJ, Geenen JE, Kruidenier J. Effectiveness of conventional sphincteroplasty in relieving pancreatic duct sphincter pressure in patients with idiopathic recurrent pancreatitis [Abstract]. Gastroenterology 1983;84:1189. 55. Guelrud M, Siegel JH. Hypertensive pancreatic duct sphincter as a cause of pancreatitis. Successful treatment with hydrostatic balloon dilatation. Dig Dis Sci 1984;29:225-31. 56. Sherman S, Lehman GA, Hawes RH, et al. Pancreatic ductal stones: frequency of successful endoscopic removal and improvement in symptoms. Gastrointest Endosc 1991;37:511-7. 57. Cremer M, Deviere J, Engelholm L. Endoscopic management of cysts and pseudocysts in chronic pancreatitis: long-term follow-up after 7 years of experience. Gastrointest Endosc 1989; 35:1-9. 58. Kozarek RA, Christie D, Barklay G. Endoscopic therapy of pancreatitis in the pediatric population. Gastrointest Endosc 1993;39:665-9. 59. Rosch W. The value of endoscopic occlusion of the pancreatic duct. Endoscopy 1983;15:175-7. 60. Geenen JE, Vennes JA, Silvis SE. Resume of a seminar on endoscopic retrograde sphincterotomy. Gastrointest Endosc 1981; 27:31-8.
GASTROINTESTINAL ENDOSCOPY
The role of ERCP in the diagnosis and treatment of idiopathic recurrent pancreatitis in children and adolescents Moises Guelrud, MD, Cesar Mujica, MD, Domingo Jaen, MD Jose Plaz, MD, Josabel Arias, MD Caracas, Venezuela
ERCP was performed in 51 patients, aged 1 to 18 years, as part of diagnostic evaluation for recurrent pancreatitis without obvious cause or as therapeutic intervention. ERCP was successful in 50 patients (98%). Thirty-four patients (68%) had anatomic findings indicating a possible cause of recurrent pancreatitis. Three patients (6%) had findings suggestive of sphincter of Oddi dysfunction. Eighteen of the 37 patients (49%) with ductal abnormalities underwent endoscopic therapy, with a favorable outcome in 15 (83%). Eleven patients were treated surgically, and 8 of these patients (73%) improved symptomatically. Eight patients received no treatment, and 6 of them (67%) had recurrent bouts of clinical pancreatitis. One mild case of pancreatitis (1.9%) occurred after ERCP, and 3 mild cases followed endoscopic therapy. In conclusion, ERCP is a relatively safe technique that produces opacification of the desired ductal system with a high degree of accuracy and provides useful information in the evaluation of children with idiopathic recurrent pancreatitis. These data suggest that endoscopic pancreatic therapy may result in symptomatic improvement, eliminating the need for surgery in selected children. Furthermore, the study demonstrates that manipulation of the pancreatic duct is comparatively safe and less hazardous than formerly believed. (Gastrointest Endosc 1994;40:428-36.)
Recurrent pancreatitis is rarely recognized in children because the index of clinical suspicion for this entity is usually low. The disease is characterized by episodes of acute pancreatitis, manifested by unexplained recurrent bouts of abdominal pain, with intervening asymptomatic intervals of varying duration. Diagnosis is often difficult and is usually missed for several months or years. Patients may be symptomatically treated during the acute attack, and if no exact cause is found, they are discharged without specific treatment. Two general categories of recurrent pancreatitis have been described, depending on whether Received September 9, 1993. For revision December 15, 1993. Accepted February 2, 1994. From the Department of Medicine, Division of Gastroenterology, Hospital General del Oeste, Caracas, Venezuela. Reprint requests: Moises Guelrud, MD, Policlinica Metropolitana, Urb. Caurimare, Caracas, Venezuela. 37/1/55410 428
the cause is non obstructive or obstructive. Nonobstructive causes include hereditary factors, cystic fibrosis, hyperlipidemia, trauma, medication, and hypercalcemia. 1-6 Obstructive causes include choledochal cysts, pancreas divisum, duodenal diverticulum, duodenal duplication, parasitic infestation, and anomalous pancreaticobiliary junction.7-16 If no specific cause is identified, a case may be classified as idiopathic recurrent pancreatitis. New insights into the ductal system through ERCP have increasingly directed our attention to morphologic changes ofthe pancreatic duct. ERCP offers the definition of the common bile duct and pancreatic ductal system that is necessary to make decisions in the management of children with idiopathic recurrent pancreatitis. Whatever the underlying cause of pancreatitis, the possibility of an anatomic abnormality amenable to endoscopic therapy or surgery should always be considered. 17 ERCP has been found useful in the identification of treatable causes in 40% of adults 18 and in 40% to 75% of children19 -21 GASTROINTESTINAL ENDOSCOPY
Table 1. ERCP in pancreatic disease Number of patients (%)
ERCP findings
Pancreatitis associated with biliary anomalies 9 (18) Choledochal cyst 3 (6) Anomalous pancreaticobiliary junction Pancreatitis associated with pancreatic anomalies 6 (12) Pancreas divisum 1 (2) Pancreatocele Annular pancreas 2 (4) Short pancreas 1 (2) 2 (4) Pancreatitis associated with parasites (Ascaris in common bile duct) 3 (6) Pancreatitis associated with SO dysfunction 7 (14) Chronic pancreatitis Pseudocyst 3 (6) 13 (26) Pancreatitis with no apparent cause (normal pancreatic duct) 50 (100) Total
Sex F
M
Age range (yr)
7 2
2 1
1-18 1-12
3 1 1
3
1
1 1 1
12-17 14 1-18 3 5-17
1
2
13-18
4 2 7
3 1 6
4-16 2-15 7-16
29
21
1-18
Table 2. Symptomatic improvement after endoscopic treatment in 18 patients with recurrent pancreatitis Diagnosis (number of patients) Choledochal cyst type I (1) Choledochal cyst type III (1) Choledochal cyst type IV (1) Pancreas divisum (3) Ascaris in CBD (2) SO dysfunction (3) Chronic pancreatitis (4) Pancreatic pseudocyst (3)
Improvement
Treatment None ES and CBD stone removal ES ES and PD stone removal ES minor papilla Worm removal ES DS DS PD occlusion DS and PD drainage Cystogastrostomy
Total
Moderate
Excellent 1 1 1
1
1 1
2 2 1 1 2
1
2 1 3 (17%)
1 (5%)
14 (78%)
CBD, Common bile duct; PD, pancreatic duct; ES, endoscopic sphincterotomy; DS, dual sphincterotomy.
with idiopathic recurrent pancreatitis. This report summarizes our experience with ERCP in the diagnosis and treatment of children and adolescents with idiopathic recurrent pancreatitis. MATERIAL AND METHODS
Patients
Between 1984 and 1992, a total of 51 patients (20 boys and 31 girls; 26 children and 25 adolescents with an age range of 1 to 18 years) were referred to our center with the presumptive diagnosis of idiopathic recurrent pancreatitis. ERCP was successful in 50 patients (98%). All patients had had two or more documented attacks of acute pancreatitis confirmed by hyperamylasemia. Despite careful clinical histories and laboratory and radiologic examinations, an apparent cause for recurrent pancreatitis could be found in none of the patients. Special attention was given to drug intake, hyperlipidemia, cystic fibrosis, hypercalcemia, collagen disease, infection, abdominal trauma, and hereditary pancreatitis. All paVOLUME 40, NO.4, 1994
tients underwent abdominal ultrasonography, 35 had CT scan, and 20 had upper gastrointestinal barium roentgenography. Endoscopic techniques
After sedation with midazolam (0.1 mg/kg) and meperidine (3 mg/kg), all patients underwent ERCP within 4 weeks of their last episode of pancreatitis. The Olympus prototype PJF pediatric duodenoscope 22 or JF -1 T adult duodenoscope (Olympus America Inc., Lake Success, N.Y.) was used. The papilla of Vater was cannulated with standard or tapered catheters, and the pancreatic and biliary ducts were visualized by fluoroscopy during injection of 60 % iodine contrast material. If a ventral duct was visualized or no pancreatogram was obtained from the major papilla, the minor papilla was cannulated (with or without secretin 1 U/kg) with a needle-tipped catheter. Sphincter of Oddi (SO) manometry was performed in 16 patients in whom obstructive causes could not be found. Studies were performed using a low-compliance infusion pump system and a three-lumen recording catheter introduced 429
creatic exocrine function while leaving intact the endocrine secretion of insulin. (7) Cystogastrostomy was performed using a diathermy needle to puncture the cyst through the gastric wall, after which the opening was enlarged using a sphincterotome and a 10F endoprosthesis was placed. All patients were monitored with pulse oximeter. Informed consent was obtained from each child's parent or guardian. Clinical follow-up
Patients were followed for periods ranging from 4 months to 6 years (mean follow-up time was 20 months). The number of episodes of pancreatitis, epigastric pain, pancreatic enzyme elevation, and emergency visits after treatment was determined and compared with the number before treatment. Improvement was graded at follow-up in the following manner: (1) none, recurrent pancreatitis requiring emergency visits and hospitalization; (2) moderate, occasional pain without pancreatic enzyme elevation and no subsequent episodes of acute pancreatitis; (3) excellent, absence of symptoms and no subsequent episodes of acute pancreatitis.
RESULTS
Figure 1. Choledochal cyst and pancreas divisum in a 5-year-old boy with recurrent pancreatitis. The dorsal pancreas (straight arrow) ends in the middle portion of the common bile duct. A stone (double straight arrows) and a type I choledochal cyst (curved arrow) with a stricture can be seen. The ventral pancreas is also observed (double curved arrow).
into the pancreatic duct through the biopsy channel. Manometric pressure recordings of the SO were performed and analyzed as previously described. 23 The maximum normal SO basal pressure was 35 mm Hg based on adult standards. Patients in whom a correctable cause for pancreatic disease was identified by ERCP were considered candidates for therapeutic endoscopy or surgery. Endoscopic therapy included the following techniques: (1) Standard biliary sphincterotomy was performed using a traction sphincterotome. (2) Dual sphincterotomy of the pancreatic duct sphincter and the common bile duct sphincter were performed in the same session. 24 First, the sphincterotome was inserted in the pancreatic duct. Using only cutting current, a cut was usually made in the l-o'clock direction over a length of 5 to 7 mm, after which the sphincterotome was reinserted in the bile duct. The bile duct was unroofed using a blended current over a length of 5 to 10 mm. (3) Minor papilla sphincterotomy was performed using cutting current in the 12-o'clock direction over a length of 5 to 7 mm with an over-the-wire, short-traction sphincterotome. (4) Pancreatic stone extraction was performed using a retrieval basket. (5) Pancreatic endoprosthesis was inserted by introducing a 5F to 10F stent into the main pancreatic duct or the dorsal pancreatic duct for a period of 10 to 14 days after endoscopic sphincterotomy or for several weeks to drain communicating pancreatic pseudocysts. (6) The pancreatic duct was obstructed with cyanocrilate glue injection (1 m!) into the remnant duct of the head of the pancreas to suppress pan430
Of 50 patients, 34 (68%) had anatomic findings indicating a possible cause of recurrent pancreatitis and 3 (6 %) had findings suggestive of SO dysfunction (significantly elevated SO basal pressure of 57 ± 5.3 mm Hg) (Table 1). Thus, ERCP examination with SO manometry revealed a possible cause of recurrent pancreatitis in 37 of 50 children and adolescents (74 %) with a diagnosis of idiopathic recurrent pancreatitis by conventional criteria. Eighteen of the 37 patients (49 %) in whom a correctable cause for pancreatic disease was found were treated by endoscopic therapy (Table 2). Overall, 15 patients (83 %) were considered markedly improved. Eight of11 patients (73 % ) treated surgically improved symptomatically. Of eight patients who received no treatment, six (67 %) had recurrent bouts of clinical pancreatitis. Pancreatitis associated with biliary anomalies
Of the nine patients with a choledochal cyst, three had a type I choledochal cyst, characterized by diffuse choledochal dilatation. One patient had coexistent pancreas divisum, pancreatic duct anomalies, and a stone in the common bile duct (Fig. 1). He was treated by endoscopic sphincterotomy of the major papilla and stone extraction, followed by operative cyst resection with hepatojejunostomy. Results were excellent at 4-year follow-up. Two patients with diffuse choledochal dilatation and extensive pancreatic duct anomalies were treated surgically by cyst resection and hepatojejunostomy. One remained asymptomatic for 14 months, and the other had an episode of acute pancreatitis 8 months after surgery. Five patients had a type IV choledochal cyst, characterized by multiple intrahepatic and extrahepatic cysts. Four had coexistent pancreatic duct anomalies, two of them with pancreatic stones. One patient refused surgery and was treated by endoscopic sphincterotomy and pancreatic duct stone extraction (Fig. 2). She remained symptomGASTROINTESTINAL ENDOSCOPY
Figure 2, Cont'd. C, The pancreatic stone is removed.
free for a period of 2 years. One patient was treated surgically by cyst resection and hepatojejunostomy with surgical sphincteroplasty and pancreatic stone removal. During a follow-up of 5 years, the patient had multiple episodes of upper abdominal pain without elevation of amylase. Three patients were treated surgically by cyst resection and hepatojejunostomy. Acute pancreatitis developed in one patient 7 months after surgery, and two patients remained asymptomatic after 8 and 13 months. One patient with a type III choledochal cyst (choledochocele) was treated by endoscopic sphincterotomy. She remained symptom-free for a period of 3 years. Of three patients with an anomalous pancreaticobiliary junction, two refused treatment. One of them had one and the other had three episodes of acute pancreatitis during the following 10 months. One patient underwent surgical sphincteroplasty and remained asymptomatic for 2 years. Pancreatitis associated with pancreatic anomalies
Figure 2. Choledochal cyst (type IV) in a 6-year-old patient with recurrent pancreatitis. A, An anomalous pancreaticobiliary junction with a pancreatic stone (arrow) is seen. B, After endoscopic sphincterotomy, a retrieval balloon is inserted (arrow).
Pancreas divisum was diagnosed in six patients (12 %). The ventral pancreatic duct was visualized during ERCP in all patients. However, the dorsal pancreas could be observed in only four patients (Fig. 3). Three patients were treated by surgical sphincteroplasty of both the minor and major papillas. Two remained asymptomatic for 2 years and one had several episodes of epigastric pain without pancreatitis.
Figure 5. Short pancreas in a 3-year-old patient with
recurrent pancreatitis. The pancreatic duct and the accessory duct are dilated. The patient with pancreatocele had another episode of acute pancreatitis after 5 months; the patient with short pancreas remained asymptomatic for 3 years.
Pancreatitis associated with parasitic infestation
Figure 3. Pancreas divisum in a 12-year-old patient with
recurrent pancreatitis. The ventral pancreas is seen with an acinogram (straight arrow). The dorsal pancreas is dilated with multiple strictures (curved arrow).
Two patients with Ascaris in the common bile duct had the worm removed endoscopically with a tripod device and were free of symptoms after 2 and 4 years.
Pancreatitis associated with SO dysfunction Three patients had SO dysfunction and normal ERCP. Two were treated by endoscopic sphincterotomy of the common bile duct sphincter. One remained asymptomatic after a follow-up of 18 months, and the other had an episode of acute pancreatitis 5 months after therapy. Both patients refused further treatment. The third patient had a dual sphincterotomy and remained symptom-free for a period of 4 years.
Chronic pancreatitis and pancreatic pseudocyst
Figure 4. Cystic dilatation of the pancreatic duct (pancreatocele) in a 14-year-old patient with recurrent pancreatitis.
Three patients were treated by endoscopic sphincterotomy of the minor papilla. In one, a pancreatic stent was placed for 2 weeks. Two were asymptomatic during follow-up of 10 and 17 months and one had recurrent episodes of abdominal pain without enzyme elevation. Two patients with annular pancreas, one with cystic dilatation of the pancreatic duct (pancreatocele) (Fig. 4), and one with short pancreas (Fig. 5) refused treatment. One patient with annular pancreas had an episode of acute pancreatitis during the following 13 months; the other remained asymptomatic for 2 years. 432
Seven patients had chronic pancreatitis. Six of them had a dilated pancreatic duct. Four patients received pancreatic therapy. Three underwent a dual sphincterotomy before pancreatic duct endoprosthesis placement. The stent was removed during the following 2 weeks, and the patients experienced dramatic clinical relief of pain. Two remained asymptomatic after a follow-up of 2 years, and one had intermittent abdominal pain with an episode of acute pancreatitis in the ensuing 14 months. Two patients received no treatment and had recurrent bouts of clinical pancreatitis during the following 4 and 11 months. The other patient had a narrowed and irregular pancreatic duct. She had seven episodes of acute pancreatitis, after which pain lasted for several weeks. She had become addicted to narcotics and was treated by partial pancreatectomy of the body and tail. After 11 months she had another episode of acute pancreatitis and was treated endoscopically by cyanocrilate injection in the pancreatic duct remnant. After 2 years without sympGASTROINTESTINAL ENDOSCOPY
toms, another episode of acute pancreatitis occurred and the cyanocrilate injection was repeated. She has remained symptom-free for a period of 2.5 years. Three patients had pancreatic pseudocysts, two of which were communicating with the pancreatic duct and were treated by dual sphincterotomy and placement of a transpapillary pancreatic stent until the cysts resolved. Pain disappeared immediately. After follow-up of 2 and 3 years, both patients are alive and without clinical problems. One 2-year-old child had a noncommunicating pseudocyst that was treated by cystogastrostomy; however, the drainage failed for technical reasons and the patient underwent surgical cystojejunostomy. Pancreatitis with no apparent cause
Thirteen patients had a normal pancreaticobiliary duct system with normal SO motor function. They received no treatment. Eight patients had recurrent bouts of clinical pancreatitis during the following 2 years. Follow-up ERCP and manometry performed in three patients revealed no abnormalities. Five patients were lost for follow-up. Complications
One mild case of pancreatitis related to ERCP occurred in a patient with SO dysfunction (1.9%). Three episodes of pancreatitis occurred after endoscopic therapy in 18 patients (16.6%). One episode was in a patient with pancreas divisum treated by minor papilla sphincterotomy without pancreatic stent placement. The second was in a patient with SO dysfunction treated by dual sphincterotomy, and the third in a patient with chronic pancreatitis in whom cyanocrilate was injected into the pancreatic duct. All three patients responded to conservative management and were discharged from the hospital within 3 days. None of these 18 patients died in the 30 days after endoscopic treatment. DISCUSSION
In this study, we evaluated by ERCP and SO manometry a highly selected group of patients with recurrent pancreatitis in whom no specific cause had been identified. Additionally, only patients who had experienced at least two episodes of pancreatitis were included. ERCP and SO manometry revealed a possible cause of recurrent pancreatitis in 37 of 50 patients (74%). Pancreatic endoscopic therapy was performed in 18 patients, with excellent results in 15 patients (83 %). In the present study, diagnostic capability of ERCP is similar to that reported by others. Putnam et al. 19 studied 12 children with idiopathic recurrent pancreatitis; potentially treatable conditions were identified in nine patients (75%) who underwent successful ERCP examination. Buckley and Connon20 performed ERCP in 18 patients with pancreatitis, aged 3 to 18 VOLUME 40, NO.4, 1994
Figure 6. Chronic pancreatitis in an 8-year-old patient. A, A dilated, irregular pancreatic duct is visualized. The primary and secondary pancreatic branches are dilated and tortuous. After a dual sphincterotomy, a guide wire is inserted into the pancreatic duct. B, A 1OF, 5-cm pancreatic stent is inserted and left in place for 2 weeks.
years, and found pancreatic duct abnormalities in eight patients (44 % ). Brown et al. 21 performed 74 ERCPs for pancreatic indications and found pancreatic duct abnormalities in 21 patients (28%). An abnormal common bile duct was found in 13 of 53 patients (24 %) with normal pancreatic duct. In this series, an underlying cause of pancreatitis was identified in 52 % of patients. The success rate of ERCP in our present study was 98%. Our success rate compares favorably with that reported by others in children and adolescents.1 9 -21 , 25-28 Significant complications ofERCP, in433
cluding cholangitis, bacteremia, and pancreatitis, have been rare in childrenP, 22, 23 One of our patients (1.9 %) with SO dysfunction had a mild pancreatitis that resolved in 3 days. Other workers have reported similarly low complication rates in children. 19-21 No deaths have occurred in the studies reported of children and adolescents. Our study confirmed the safety of this procedure in children when performed by an experienced endoscopist. Endoscopic therapy in children has been directed to treatment of biliary diseases. 2o , 21-23, 29 Some authors have expanded therapeutic ERCP in the pediatric age group to treat pancreatic disease. Brown et al. 21 placed multiple pancreatic stents in eight children with pancreas divisum. Al Traiff and Khan 30 endoscopically drained a duodenal duplication cyst in an 11-year-old child who had recurrent attacks of pancreatitis. Lehman et al. 31 performed minor papilla sphincterotomy over a previously placed dorsal pancreatic duct endoprosthesis in a 7-year-old child with pancreas divisum and recurrent pancreatitis. In the present study, 14 of 17 patients (82 %) with recurrent pancreatitis underwent successful endoscopic treatment. These responses were presumably related to relief of obstructive pancreatitis, as has been noted in adults. 32 Choledochal cyst is a congenital malformation of the biliary tract. It is primarily a disease of children and young adults. An abnormal pancreaticobiliary junction has been suggested as an important factor in the pathogenesis. The classification of Todani et al. 33 of bile duct cyst is most often used. ERCP is of significant value in outlining these cysts and their relationship to both ductal systems, providing a roadmap for the surgical approach. 34 ,35 Anomalous pancreaticobiliary junction has been associated with choledochal cysts and recurrent pancreatitis. 36 , 37 The treatment is surgical. However, one of our patients refused surgery. She had a type IV choledochal cyst with a pancreatic duct stone and recurrent pancreatitis. She was treated by endoscopic sphincterotomy followed by pancreatic duct stone extraction; pancreatitis did not recur during the ensuing 2 years. Although classified as one of the forms of choledochal cysts, choledochocele (type III choledochal cyst) is probably not related. Based on autopsy studies, Sterling38 hypothesized that a choledochocele is an acquired evagination of the mucosa of the terminal common bile duct into the duodenum that is caused by an obstruction to bile outflow resulting from an impacted stone, fibrosis, or papillitis. More recently, Venu et al. 39 and Kagiyama et aI.4° suggested that in some patients abnormal SO motility could be a contributing factor in the pathogenesis of choledochocele. It has been reported in association with recurrent pancreatitis. 41 In the past, the treatment has been surgical. However, in selected cases, a choledochocele may be effectively treated by endoscopic sphincterot434
omy.42 One of our patients was treated by endoscopic sphincterotomy, after which she remained symptomfree for a period of 3 years. Pancreas divisum, the most common congenital variant of pancreatic ductal anatomy, has been found in approximately 3% to 13% of ERCP studies. 43 -46 However, when patients with unexplained recurrent pancreatitis were studied, the incidence was 25 %.43 In our small group of patients with recurrent pancreatitis, the incidence was 12 %. The treatment used to be surgical. Warshaw et al. 47 performed minor papilla sphincteroplasty in 88 patients with symptomatic pancreas divisum and recurrent pancreatitis. Overall, 70% improved at a mean follow-up of 53 months. Endoscopic treatment has been directed to decompress the dorsal duct by insertion of endoprosthesis, dilation of the minor papilla, or sphincterotomy of the minor papilla with or without insertion of endoprosthesis. McCarthy et al. 48 treated 19 patients with pancreas divisum using dorsal pancreatic stents and obtained 89 % improvement. Lans et al. 49 reported the results of a randomized control trial of long-term (12 months) stenting of the minor papilla in patients with recurrent pancreatitis. Stents were changed every 4 months. Follow-up continued for at least 12 months after the final stent removal. Patients with stents had fewer episodes of pancreatitis and were more frequently judged to be improved (90% versus 11 % of controls,p < 0.05). A temporary trail of stenting the dorsal duct during ERCP has been used to decide whether surgery is appropriate. Patients who improve after stenting may be candidates for more sustained relief with surgical decompression. 50 Minor papilla sphincterotomy has been used to treat pancreas divisum. 51 Soehendra et al. 52 treated six patients using a precut or needle-knife sphincterotome over a stent, obtaining 83 % symptomatic improvement. Liguory et al. 53 performed minor papilla sphincterotomy with (n = 2) and without (n = 8) dorsal pancreatic duct stenting in 10 patients with pancreatitis. Seven (70%) remained symptomfree during a 3- to 30-month follow-up period. Lehman et al. 31 treated 52 patients with pancreas divisum by minor papilla sphincterotomy performed with a needle knife over a previously placed dorsal pancreatic duct stent. Patients with recurrent acute pancreatitis benefited more frequently than those with chronic pancreatitis (76 % versus 27 %, p = 0.01). Pancreatitis as a complication of therapy occurred after 13 % of the procedures but was mild and could be managed conservatively. We treated three patients with minor papilla sphincterotomy and obtained excellent results in two and a moderate result in one. In adults, the leading cause of idiopathic recurrent pancreatitis appears to be SO dysfunction. Venu et aU 8 studied 116 patients with recurrent pancreatitis and found 17 patients (15 %) with SO dysfunction. Manometry is the best diagnostic procedure and eleGASTROINTESTINAL ENDOSCOPY
vated SO basal pressure the major criterion for diagnosis. In the present series, SO dysfunction represented 6 % of the causes of recurrent pancreatitis. Three of 16 children (18%) with normal ERCP findings had SO dysfunction. These results are in accordance with those reported by Brown et al. 21 who found SO dysfunction in three of 13 children (23 %) with recurrent pancreatitis. These patients are generally treated by endoscopic sphincterotomy or the classic transduodenal sphincterotomy. However, patients do not respond well,54 presumably because with either treatment modalities the pancreatic sphincter is not transected, and recurrent attacks of pancreatitis may be attributed to an affected sphincter. A hypertensive pancreatic duct sphincter has been suggested as the cause of these recurrent episodes of pancreatitis. 55 In the present study, two of three patients were treated by standard endoscopic sphincterotomy, with clinical improvement noted in one patient. The third patient was treated by dual sphincterotomy of the pancreatic and the common bile duct sphincters, after which the patient was markedly improved during the ensuing 4 years. In patients with chronic pancreatitis, recurrent attacks may be caused by strictures of the main duct, pancreatic stones, or pseudocysts that can impair the normal outflow of pancreatic juice. In adults, it has been shown that insertion of a pancreatic duct endoprosthesis,32,43 pancreatic stone removal, 56 or pseudocyst drainage 57 can improve ductal drainage. Recently, Kozarek et al. 58 performed pancreatic duct sphincterotomy followed by stone extraction and stent insertion in three children with chronic relapsing pancreatitis, after which marked improvement in chronic abdominal pain was noted. These endoscopic techniques have proved to be good alternatives to surgical drainage. Two of our three patients with chronic pancreatitis treated by dual sphincterotomy and pancreatic stent experienced significant improvement during a 2-year follow-up. These results further support the theory of an obstructive cause of recurrent pancreatitis in some cases. One patient with chronic pancreatitis had recurrent attacks after subtotal pancreatectomy; she was scheduled to undergo total pancreatectomy, but instead cyanocrilate glue was injected into the remnant pancreatic duct to produce atrophy ofthe exocrine gland with preservation of the islets. This technique has been used and abandoned by Rosch,59 who noted poor long-term results. Although most of his patients had an intact pancreas, apparently the instillation pressure via an endoscopic cannula is not strong enough to guarantee complete filling of the main pancreatic duct and the primary and secondary branches. Our patient had had a previous subtotal pancreatectomy so that complete filling of the ductal system was accomplished. Freedom from pain was achieved for 2 years. After a second cyanocrilate injection was performed, she remained symptom-free for 2.5 years. VOLUME 40, NO.4, 1994
Pancreatic pseudocysts are usually detected after an attack of pancreatitis; the incidence is approximately 20 %. Most pseudo cysts resolve spontaneously, but drainage procedures are indicated for persistent, sym ptomatic cysts. In adults, the classic treatment has been surgical drainage, and more recently endoscopic drainage has been reported. 57 In this study, two children with nonresolving communicating pseudo cysts became asymptomatic after transampullary endoprosthesis placement; the pseudocysts resolved rapidly. In the third patient, a transmural drainage via the gastric wall was unsuccessful because of technical problems. In adults, short-term complications of endoscopic sphincterotomy, such as bleeding, pancreatitis, cholangitis, and retroperitoneal perforation, occur in 7 % to 10% of patients. 6o Complications of endoscopic sphincterotomy in children has been sparsely reported. Buckley and Connon 20 successfully performed endoscopic sphincterotomy with common duct stone extraction in five children without significant complications, except for bleeding in one patient who had been given aspirin after the procedure. We previously reported 26 six children with common duct stones in whom endoscopic sphincterotomy caused no complications. Kozarek et a1. 58 reported mild pancreatitis in three children with chronic pancreatitis after pancreatic therapy. We had three episodes (17%) of mild acute pancreatitis after pancreatic therapy. In conclusion, we have demonstrated that ERCP in children and adolescents is a safe diagnostic procedure that delineates the anatomy of the pancreatic and biliary ductal systems and identifies patients who are amenable to endoscopic or surgical treatment. When performed by experienced endoscopists, therapeutic pancreatography can be successfully performed in a selected group of children with a low rate of complications. A longer follow-up period will be necessary to determine whether endoscopic success produces longstanding clinical improvement. ACKNOWLEDGMENT
The authors thank Adriana Maixner for her assistance in reviewing the manuscript.
REFERENCES 1.
Hendren WH, Greep JM, Patton AS. Pancreatitis in childhood: experience with 15 cases. Arch Dis Child 1965;40:132-45. 2. Jordan SC, Ament ME. Pancreatitis in children and adolescents. J Pediatr 1977;91:211-6. 3. Ziegler DW, Long JA, Philippart AI, Klein MD. Pancreatitis in childhood: experience with 49 patients. Ann Surg 1988;207:257-
61.
4. Kattwinkel J, Lapey A, Di Sant' Agnese P, Edwards WA. Hereditary pancreatitis: three new kindreds and a critical review of the literature. Pediatrics 1973;51:55-69. 5. Krauss RM, Levy AG. Subclinical chronic pancreatitis in type I hyperlipoproteinemia. Am J Med 1977;62:144-9. 6. Park RW, Grand RJ. Gastrointestinal manifestations of cystic fibrosis: a review. Gastroenterology 1981;81:1143-61. 7. Griffin M, Carey WD, Hermann R, Buonocore E. Recurrent
435
8. 9.
10. 11. 12. 13. 14.
15. 16. 17. 18. 19. 20. 21.
22. 23. 24. 25. 26.
27. 28. 29. 30. 31. 32. 33. 34.
acute pancreatitis and intussusception complicating an intraluminal duodenal diverticulum. Gastroenterology 1981;81:345-8. Wagner CW, Golladay ES. Pancreas divisum and pancreatitis in children. Am Surg 1988;54:22-6. Yaffe MR, Gutenberger JE. Chronic pancreatitis and pancreas divisum in an infant diagnosed by endoscopic retrograde cholangiopancreatography and treated with somatostatin analog. J Pediatr Gastroenterol Nutr 1989;9:108-11. Yedlin ST, Dubois RS, Philippart AI. Pancreas divisum. A cause of pancreatitis in childhood. J Pediatr Surg 1984;19:793-4. Van Der Spuy S. Endoscopic retrograde cholangiopancreatography (ERCP) in children. Endoscopy 1978;10:173-5. Gibson LE, Haller JA. Acute pancreatitis associated with congenital cyst of the common bile duct. J Pediatr 1959;55:650-7. Agrawall RM, Brodmerkel GJ. Choledochal cyst presenting as pancreatitis. Am J GastroenteroI1979;71:408-11. Black PR, Welch KJ, Eraklis AJ. Juxtapancreatic intestinal duplications with pancreatic ductal communication: a cause of pancreatitis and recurrent abdominal pain in childhood. J Pediatr Surg 1986;21:257-61. Lavine JE, Harrison M, Heyman MB. Gastrointestinal duplications causing relapsing pancreatitis in children. Gastroenterology 1989;97:1556-8. Karp MP, Jewett TC, Cooney DR. Chronic relapsing pancreatitis in childhood caused by pancreaticobiliary ductal anomaly. J Pediatr Gastroenterol Nutr 1983;2:324-8. Guelrud M. Endoscopic retrograde cholangiopancreatography in the infant. In: Barkin J, O'Phelan CA, eds. Advanced therapeutic endoscopy. New York: Raven Press, 1990: 335-54. Venu RP, Geenen JE, Hogan W, Stone J, Johnson GK, Soergel K. Idiopathic recurrent pancreatitis. An approach to diagnosis and treatment. Dig Dis Sci 1989;34:56-60. Putnam PE, Kocoshis SA, Orenstein SR, Schade RR. Pediatric endoscopic retrograde cholangiopancreatography. Am J GastroenteroI1991;86:824-30. Buckley A, Connon JJ. The role of ERCP in children and adolescents. Gastrointest Endosc 1990;36:369-72. Brown CW, Werlin SL, Geenen JE, Schmalz M. The diagnostic and therapeutic role of endoscopic retrograde cholangiopancreatography in children. J Pediatr Gastroenterol Nutr 1993;17: 19-23. Guelrud M, Jaen D, Torres P, et al. Endoscopic cholangiopancreatography in the infant: evaluation of a new prototype pediatric duodenoscope. Gastrointest Endosc 1987;33:4-8. Guelrud M, Mendoza S, Rossiter G, Villegas MI. Sphincter of Oddi manometry in healthy volunteers. Dig Dis Sci 1990;35:3846. Fuji T, Amano H, Ohmura R, Akiyama T, Aibe T, Takemoto T. Endoscopic pancreatic sphincterotomy. Technique and evaluation. Endoscopy 1989;21:27-30. Guelrud M, Jaen D, Mendoza S, Plaz J, Torres P. ERCP in the diagnosis of extrahepatic biliary atresia. Gastrointest Endosc 1991;37:522-6. Guelrud M, Mendoza S, Jaen D, Plaz J, Machuca J, Torres P. ERCP and endoscopic sphincterotomy in infants and children with jaundice due to common bile duct stones. Gastrointest Endosc 1992;38:450-3. Cotton PB, Laage NJ. Endoscopic retrograde cholangiopancreatography in children. Arch Dis Child 1982;57:131-6. Allendorph M, Werlin SL, Geenen JE. Endoscopic retrograde cholangiopancreatography in children. J Pediatr 1987;110:20611. Guelrud M, Mendoza S, Zager A, Noguera C. Biliary stenting in an infant with malignant obstructive jaundice. Gastrointest Endosc 1989;35:259-61. Al Traif lA, Khan MH. Endoscopic drainage of a duodenal duplication cyst. Gastrointest Endosc 1992;38:64-5. Lehman GA, Sherman S, Nisi R, Hawes RH. Pancreas divisum: results of minor papilla sphincterotomy. Gastrointest Endosc 1993;39:1-8. Kozarek RA, Patterson DJ, Ball TJ, Traverso LW. Endoscopic placement of pancreatic drains and stents in the management of pancreatitis. Ann Surg 1989;209:261-6. Todani T, Watanabe Y, Narusue M. Congenital bile duct cyst. Am J Surg 1977;134:263-9. Guelrud M, Jaen D, Mendoza S, Torres P. Usefulness of endoscopic retrograde cholangiopancreatography in diag-
436
nosis of choledochal cysts in children. G E N 1989;43:9-12. 35. Thatcher BS, Sivak MV, Hermann RE, Esselstyn CB. ERCP in evaluation and diagnosis of choledochal cyst: report of five cases. Gastrointest Endosc 1986;32:27-31. 36. Atkinson GO, Wyly JB, Gay BB, Ball TI, Winn KJ. Idiopathic fibrosing pancreatitis: a cause of obstructive jaundice in childhood. Pediatr Radiol 1988;18:28-31. 37. Mori K, Nagakawa T, Ohta T, et al. Acute pancreatitis associated with anomalous union of the pancreaticobiliary ductal system. J Clin GastroenteroI1991;13:673-7. 38. Sterling JA. Diverticula in the terminal portion of the common bile duct. Am J Pathol 1949;25:325-35. 39. Venu RP, Geenen JE, Hogan WJ, et al. Role of endoscopic retrograde cholangiopancreatography in the diagnosis and treatment of choledochocele. Gastroenterology 1984;87:1144-9. 40. Kagiyama S, Okazaki K, Yamamoto Y. Anatomic variants of choledochocele and manometric measurements of pressure in the cele and the orifice zone. Am J GastroenteroI1987;82:641-9. 41. Greene FL, Brown JJ, Rubinstein P, Anderson MC. Choledochocele and recurrent pancreatitis. Diagnosis and surgical management. Am J Surg 1985;149:306-9. 42. Siegel JH, Harding GT, Chateau F. Endoscopic incision of choledochal cyst (choledochocele). Endoscopy 1981;13:200-2. 43. Cotton PB. Congenital anomaly of pancreas divisum as a cause of obstructive pain and pancreatitis. Gut 1980;21:105-14. 44. O'Connor KW, Lehman GA. An improved technique for accessory papilla cannulation in pancreas divisum. Gastrointest Endosc 1985;31:13-7. 45. Gregg JA. Pancreas divisum: its association with pancreatitis. Am J Surg 1977;134:539-43. 46. Heiss FN, Shea JA. Association of pancreatitis and various ductal anatomy, dominant drainage of the duct of Santorini. Am J Gastroenterol 1978;70:158-62. 47. Warshaw AL, Simeone JF, Schapiro RH, Flavin-Warshaw B. Evaluation and treatment of the dominant dorsal duct syndrome (pancreas divisum redefined). Am J Surg 1990;159:59-64. 48. McCarthy J, Geenen JE, Hogan WJ. Preliminary experience with endoscopic stent placement in benign pancreatic diseases. Gastrointest Endosc 1988;34:16-8. 49. Lans JI, Geenen JE, Johanson JF, Hogan WJ. Endoscopic therapy in patients with pancreas divisum and acute pancreatitis: a prospective, randomized, controlled clinical trial. Gastrointest Endosc 1992;38:430-4. 50. Siegel JH, Pullano W, Ben-Zvi JS, Cooperman AM. Effectiveness of endoscopic drainage for pancreas divisum. Endoscopy 1990;20:129-32. 51. Cotton PB. Duodenoscopic papillotomy at the minor papilla of Vater for recurrent dorsal pancreatitis. Endosc Dig 1978;3:27 -8. 52. Soehendra N, Kempeneers I, Nam VC, Grimm H. Endoscopic dilatation and papillotomy of the accessory papilla and internal drainage in pancreas divisum. Endoscopy 1986;18:129-32. 53. Liguory C, Lefebvre JF, Carrard JM, Bonnel D, Fritsch J, Etienne JP. Le pancreas divisum: etude clinique et therapeutique chez l'homme: it propos de 87 cas. Gastroenterol Clin Bioi 1986; 10:820-5. 54. Hogan WJ, Geenen JE, Kruidenier J. Effectiveness of conventional sphincteroplasty in relieving pancreatic duct sphincter pressure in patients with idiopathic recurrent pancreatitis [Abstract]. Gastroenterology 1983;84:1189. 55. Guelrud M, Siegel JH. Hypertensive pancreatic duct sphincter as a cause of pancreatitis. Successful treatment with hydrostatic balloon dilatation. Dig Dis Sci 1984;29:225-31. 56. Sherman S, Lehman GA, Hawes RH, et al. Pancreatic ductal stones: frequency of successful endoscopic removal and improvement in symptoms. Gastrointest Endosc 1991;37:511-7. 57. Cremer M, Deviere J, Engelholm L. Endoscopic management of cysts and pseudocysts in chronic pancreatitis: long-term follow-up after 7 years of experience. Gastrointest Endosc 1989; 35:1-9. 58. Kozarek RA, Christie D, Barklay G. Endoscopic therapy of pancreatitis in the pediatric population. Gastrointest Endosc 1993;39:665-9. 59. Rosch W. The value of endoscopic occlusion of the pancreatic duct. Endoscopy 1983;15:175-7. 60. Geenen JE, Vennes JA, Silvis SE. Resume of a seminar on endoscopic retrograde sphincterotomy. Gastrointest Endosc 1981; 27:31-8.
GASTROINTESTINAL ENDOSCOPY