CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2007;5:1085 1091 ORIGINAL ARTICLES LIVER, PANCREAS, AND BILIARY TRACT Recurrent Flares of Pancreatitis Predict Development of Exocrine Insufficiency in Chronic Pancreatitis BIMALJIT S. SANDHU, WILLIAM A. HACKWORTH, STACIE STEVENS, DOUMIT S. BOUHAIDAR, ALVIN M. ZFASS, and ARUN J. SANYAL Pancreatitis Center, Division of Gastroenterology, Hepatology & Nutrition, Virginia Commonwealth University Medical Center, Richmond, Virginia See CME exam on page 1007. Background & Aims: The natural history of specific morphologic stages of chronic pancreatitis (CP) is not well defined. The aim of this study was to determine if worsening morphologic stages of CP are associated with poorer clinical outcomes. Methods: A retrospective analysis of 159 subjects with CP was performed. The baseline stage of CP was categorized according to the Cambridge classification. Pain was categorized as type A (intermittent acute), B (continuous), or combined. Exocrine failure was defined by steatorrhea; endocrine failure was characterized as diabetes mellitus. Complications were defined clinically. Results: Pancreatic duct (PD) morphology was equivocal in 37.1%, minimal in 12.6%, moderate in 7.5%, and severe in 42.8% of the patients. Over a median follow-up period of 3.7 years, the risk of developing exocrine insufficiency and diabetes was 28% and 19%, respectively. Recurrent acute flares of pancreatitis predicted the development of exocrine insufficiency (P.004). Severe PD morphology predicted the likelihood of having persistent pain (P.008). Patients with concurrent type A and B pain and older age at diagnosis had a greater likelihood of having persistent pain (P.021). The risk of developing bile duct stricture was higher in the advanced morphologic stages of CP (P.005). Conclusions: Recurrent flares of pancreatitis predispose to the development of exocrine insufficiency in CP. Patients with complex-type pain, older age at diagnosis, and advanced morphologic stage are more likely to have persistent pain. PD morphology does not correlate with the risk of developing exocrine failure and/or diabetes. Pain does not necessarily decrease or disappear with the onset of exocrine insufficiency and diabetes. Chronic pancreatitis (CP) is defined as a spectrum of pancreatic diseases characterized by injury and fibrotic strictures of the pancreatic duct (PD). 1,2 This is accompanied by ductal and periductal calcification, acinar atrophy, and, in some cases, PD stones. Abdominal pain is often the principal manifestation of CP. 3 9 The clinical course may be characterized by either continuous or intermittent flares of pain, with or without accompanying features of acute pancreatitis (eg, increased amylase and lipase levels). 10,11 Over time, acinar loss leads to pancreatic exocrine insufficiency, whereas islet cell destruction may cause diabetes. 12 Pancreatic inflammation also may cause local complications such as pseudocysts, splenic and superior mesenteric vein thrombosis, and biliary strictures. 4,5,7,13,14 The Cambridge classification is the most widely used method for the staging of CP. This classification is based on the extent and severity of PD changes. 15,16 Such changes range from minor irregularities of the duct contour to extensive strictures of the ducts with calcification. 16 A linear relationship between the severity of ductal changes and the prevalence of exocrine insufficiency has been suggested in some studies. 5,17,18 However, other studies have failed to confirm this observation. 19 21 Although improvement in pain after the development of pancreatic calcification was observed by some investigators, 5,22 others were unable to confirm this finding. 7,23 Thus, the relationship between the Cambridge stage of CP and the presence of exocrine failure or the severity of pain remains controversial. Moreover, the risk of developing specific adverse outcomes such as diabetes over time remains unknown. The primary objective of the present study was to determine whether a more advanced stage of CP at baseline, as defined by the Cambridge classification, was associated with poorer outcomes with respect to mortality, exocrine and endocrine failure, pain, and development of complications. Methods The clinical database in the pancreatitis center was searched to identify subjects with CP using predefined criteria. 10,15 A retrospective analysis of data from patients with CP, who are being followed up in the pancreatitis center at the VCU Medical Center and are in the database, was performed. The study was approved by the Institutional Review Board. Abbreviations used in this paper: CP, chronic pancreatitis; PD, pancreatic duct. 2007 by the AGA Institute 1542-3565/07/$32.00 doi:10.1016/j.cgh.2007.04.011
1086 SANDHU ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 5, No. 9 Patient Identification Case records of 159 subjects diagnosed with CP were reviewed. These patients were seen between 1996 and 2005. Although the diagnosis of pancreatitis was based on typical clinical symptoms and an increase of amylase and/or lipase levels to more than 3 times the upper limit of normal in the majority of the subjects, those with incidentally detected CP on abdominal imaging, even in the absence of typical symptoms and/or laboratory findings, also were included. 10,11 In each case, the pancreatic morphology was defined by either magnetic resonance cholangiopancreatography and/or an endoscopic retrograde cholangiopancreatogram. At the authors institution, all cases of CP were diagnosed by the earlier-described mentioned criteria. There were no cases of CP diagnosed by endoscopic ultrasound alone. PD changes were categorized based on the Cambridge classification as follows: (1) equivocal: side branch ectasia less than 3 or no change, (2) mild: 3 or more side branch ectasia and normal main duct, (3) moderate: 1 irregularity of main PD, and (4) severe: 2 cavity, calcifications, obstruction, or dilatation. The Cambridge staging of CP was based on the magnetic resonance cholangiopancreatography readings by the expert radiologists. 15,16 Exclusion criteria included subjects with recurrent bouts of abdominal pain without any morphologic changes of CP, as defined by either an endoscopic retrograde cholangiopancreatogram or magnetic resonance cholangiopancreatography. Similarly, subjects with gallstones who had features of CP were excluded because the interactions between gallstone disease and co-existing CP with respect to progression of CP remain unclear. Evaluation and Management of End Points Assessment of pain was based on patient report and physician assessment. Most of the patients reported their pain severity on a visual analogue scale from 0 to 10. Pain was categorized as type A (intermittent acute), type B (continuous), or combined types A and B pain. 3 Severe flares of pain were defined as those that required hospitalization for control. They included both those with or without accompanying features of pancreatitis (increase of amylase and/or lipase levels). Patients were said to have pain relief if they did not require analgesia on a regular basis. Those who regularly required analgesia were labeled as having persistent pain. Pain was managed primarily by a nurse practitioner and the investigators (B.S.S. or A.J.S.). All subjects were required to sign a contract forbidding them from obtaining pain medications from another health care provider and agreeing to refrain from substance abuse. Methadone was and is being used as the primary means of long-term pain management. In most cases, the initial dose was 5 mg 3 times daily, which was titrated up depending on symptom control and ability to function on a day-to-day basis. Subjects rated their pain on a scale from 1 to 10 on each visit. All patients received pancreatic enzyme supplements when appropriate along with a proton pump inhibitor. Breakthrough pain, despite high methadone doses ( 120 mg/day), was managed with the addition of gabapentin with or without oxycodone. Also, the presence of an alternate cause of pain (eg, ulcer disease or bowel obstruction) was excluded with appropriate tests. Those who continued to have severe disabling pain were referred to the behavioral pain management consultants for further evaluation for depression and help with pain management. In rare instances of uncontrolled pain, patients were evaluated for surgical intervention vs pancreatic stenting (if a dominant stricture was present). Random urine drug screens were performed on all patients. A positive drug screen result was followed up by a visit with the nurse practitioner who confronted the subject concerning the results. Subjects then were referred to a substance abuse clinic for further evaluation, counseling, and management. They were followed up further by frequent telephone calls by the nurse practitioner to encourage them to refrain from substance abuse. Those who tested positive for cocaine, heroin, or other hard drugs on more than 2 occasions were discharged from the clinic. These subjects were censored on the day of their last clinic visit. Exocrine failure was defined by the presence of steatorrhea whereas endocrine failure was defined by the presence of diabetes. In addition, appropriate studies for the diagnosis of steatorrhea and diabetes were performed when these diagnoses were suspected clinically. The presence of complications such as bile duct stricture, pseudocyst, and splenic vein thrombosis was defined clinically and confirmed by imaging studies such as computerized tomography scan and/or magnetic resonance imaging. Bile duct stricture was identified by the presence of cholestatic liver enzyme level abnormalities along with a confirmatory biliary imaging study (endoscopic retrograde cholangiopancreatogram or magnetic resonance cholangiopancreatography). Statistical Analysis The principal comparisons were the risks of individual outcome parameters across various stages of CP. The hazard of reaching each outcome event (eg, death) was analyzed as time to failure by the Kaplan Meier method. Across-group comparisons between various stages of CP were made using log-rank analysis. To further identify specific baseline factors that independently predicted the development of individual outcome end points (death, exocrine failure, diabetes, biliary stricture, pseudocyst, loss of pain, splenic vein thrombosis), a Cox regression analysis and a stepwise logistic regression analysis were performed. For each given outcome parameter, the model included parameters Table 1. Baseline Demographics Parameters Mean SD (total 159) Age, y 48.4 11.4 Male:female 98:61 Alcohol-induced:other causes 110:49 Race Caucasian 82 (51.6%) Others 77 (48.4%) Median follow-up period, y 3.7 (range, 0.2 27) Pancreatic morphology Equivocal 59 (37.1%) Mild 20 (12.6%) Moderate 12 (7.5%) Severe 68 (42.8%) Type A pain 78 (49.1%) Type B pain 45 (28.3%) Concurrent type A and B pain 31 (19.4%) Painless pancreatitis 5 (3.2%)
September 2007 PANCREATIC DUCT MORPHOLOGY AND EXOCRINE FUNCTION 1087 10%, respectively (Figure 1B). There was, however, no statistical difference in the actuarial probability of mortality among different morphologic stages of CP at any time point from the diagnosis of CP (P.65). The cause of death was related to pancreatic cancer in 2 subjects, acute flare and complications of pancreatitis in 3 subjects, and nonpancreatic causes in the remaining 7 subjects. There was no significant correlation of mortality to the development of exocrine insufficiency, diabetes, pseudocyst, cirrhosis, and the pancreatic morphology. Exocrine Insufficiency Nine (5.6%) patients were found to have pancreatic exocrine insufficiency at the time of diagnosis of CP. The overall actuarial probability of the development of exocrine insufficiency was 28% at 5 years from the diagnosis of CP (Figure 2A). The risk of developing exocrine insufficiency for the various Figure 1. (A) Overall survival in CP., Survival. (B) Survival in various morphologic stages of CP. that have been associated with that end point in the published literature. Significance was set at a P value of.05. Results A total of 159 subjects with CP were studied over a median follow-up period of 3.7 years (range, 0.2 27 y). The baseline demographics of these subjects are listed in Table 1. At baseline the mean age of these subjects was 48 years. The male to female ratio was 1.6:1; of which 82 (51.6%) subjects were Caucasian and 77 (48.4%) were African American and other races. Although alcohol was the cause of pancreatitis in 110 subjects, idiopathic pancreatitis, pancreas divisum, and hypertriglyceridemia were among the causes of pancreatitis in the remaining 49 subjects. Pancreatic duct morphology was equivocal in 37.1%, mild in 12.6%, moderate in 7.5%, and severe in 42.8% of subjects at the baseline. Pain was the predominant symptom in 154 (96.8%) subjects and 5 (3.2%) had painless pancreatitis at presentation. Although type A pain was the most common type of pain pattern present in 78 (49.1%) subjects, type B pain and combined type A and B pain patterns were present in 45 (28.3%) and 31 (19.4%) of the subjects at baseline, respectively. Five (3.2%) subjects had painless pancreatitis. Survival The overall mortality rate was 7.5% (12 of 159) over a median follow-up period of 3.7 years. The overall actuarial probability of mortality in CP was 8% at 5 years (Figure 1A). During this time frame, the actuarial probabilities of mortality in equivocal, mild, moderate, and severe CP were 2%, 0%, 2%, and Figure 2. (A) Overall exocrine failure in CP. (B) Exocrine failure in different morphologic stages of CP.
1088 SANDHU ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 5, No. 9 Table 2. Regression Model for the Development of Exocrine Failure in CP Exocrine failure Standard error P value Recurrent flares of 0.219.004 pancreatitis Cause of CP 0.118.85 Smoking 0.417.87 Alcohol 0.350.80 Morphology 0.129.75 Figure 4. Persistence of pain in different stages of CP., Equivocal;, mild; X, moderate; Œ, severe. stages of CP during this time frame was as follows: equivocal CP, 23%; mild CP, 28%; moderate CP, 42%; and severe CP, 31% (Figure 2B). The risk of development of exocrine insufficiency was not significantly different among different pancreatic morphologic stages (P.14). The equivocal and mild pancreatic morphologic stages had a similar risk of development of exocrine insufficiency when compared with the moderate and severe morphologic stages of CP. The development of exocrine insufficiency was predicted by presence of recurrent flares of pancreatitis (P.004) (Table 2). However, the development of exocrine insufficiency was not related to the severity of pancreatic duct morphology, cause of CP, smoking, or alcohol consumption (P.05) (Table 2). Figure 3. (A) Overall development of diabetes in CP. (B) Development of diabetes in different morphologic stages of CP. Diabetes (Pancreatic Endocrine Insufficiency) Diabetes was detected in 5 (3.1%) patients during the initial diagnosis of CP. The overall actuarial probability of the development of diabetes was 19% at 5 years from the diagnosis of CP (Figure 3A). The actuarial probability of the development of diabetes at 5 years from diagnosis of CP in equivocal, mild, moderate, and severe morphologic stages of CP was 21%, 12%, 22%, and 27%, respectively (Figure 3B). The actuarial risk of developing diabetes within 5 years of diagnosis was not significantly different (P.47) between subjects with varying CP stage at baseline (Figure 3). The development of diabetes was not related to the severity of pancreatic morphology, smoking, ongoing alcohol consumption, or presence of pseudocyst (P.05). Pain Pain was the predominant symptom in 154 (96.8%) subjects, whereas 5 (3.2%) subjects had painless pancreatitis at presentation. A total of 111 (72%) subjects had persistent pain, whereas 43 (28%) had pain relief during the follow-up period. The actuarial probability of having persistent pain at 5 years from the diagnosis of CP was 56%, 76%, 100%, and 90% in equivocal, mild, moderate, and severe morphologic subgroups of CP, respectively (P.0001) (Figure 4). The risk of having persistent pain in CP significantly correlated with the progres-
September 2007 PANCREATIC DUCT MORPHOLOGY AND EXOCRINE FUNCTION 1089 Table 3. Regression Model for Persistence of Pain in CP Persistent pain sion of pancreatic morphologic changes from equivocal to severe (P.008). Also, subjects with more complex-type pain having features of both type A and type B had a greater chance of having persistent pain than being pain-free (P.035) (Table 3). The subjects with older age at diagnosis of CP had a greater chance of having persistent pain (P.05) (Table 3). There was, however, no significant correlation of persistence of pain with the presence of exocrine insufficiency, diabetes, ongoing alcohol consumption, and smoking (P.05) (Table 3). Complications -coefficient Standard error P value Pancreatic morphology 0.431 0.161.008 Pain type 0.633 0.300.03 Exocrine failure 0.200 0.453.66 Diabetes 0.671 0.461.12 Ongoing alcohol consumption 0.004 0.419.99 Smoking 0.633 0.447.16 Age 0.032 0.017.05 The overall actuarial probability of the development of bile duct stricture was 9% at 5 years from the diagnosis of CP. During this time frame, the risk of developing a biliary stricture was 8% and 14% for those with moderate and severe CP, respectively. This was significantly higher (P.005) than that for the equivocal and mild morphologic subgroups of CP (Figure 5). Splenic vein thrombosis was observed in 37 (23.2%) subjects. The frequency of development of splenic vein thrombosis in those with equivocal (15 of 59; 25.4%), mild (3 of 20; 15%), moderate (2 of 12; 16.7%), and severe (17 of 68; 25%) CP was not significantly different. None of the patients with thrombosis of the splenic vein had gastrointestinal hemorrhage. Pseudocyst was present in 71 (44.6%) subjects of CP. The frequency of presence of pseudocyst in equivocal, mild, moderate, and severe morphologic subgroups was 14 (23.7%), 13 (65%), 7 (58.3%), and 37 (54.4%), respectively. Discussion CP is a heterogeneous disorder; although some subjects are symptomatic, others can remain asymptomatic over long periods of time. Also, although most subjects with CP have established pancreatic duct changes, such changes may not be manifest for some time after the onset of symptoms. The current study was a retrospective analysis of subjects who were identified either because they had symptoms of CP or because they had imaging evidence of CP. Although this does not capture all of the subpopulations of CP, we believe that the vast majority of subjects with an established diagnosis of CP will have met these conditions. Therefore, the data from the current study are generalizable to most, but not all, subjects with CP. The first key observation in our study was a low death rate of 7.5% as compared with the previously reported death rate of 28.8% 35%. 5,9,24 A longer median duration of follow-up evaluation in these studies (range, 6.3 9.8 y) as compared with our study (median duration, 3.7 y) may be responsible for this apparently better survival rate. These differences also may reflect improvements in the management of acute flares and complications of pancreatitis, multidisciplinary care with early recognition, and management of comorbid conditions. Only 5 of 12 deaths were related to pancreatic causes. Importantly, we did not observe any significant correlation between the increasing severity of pancreatic duct changes and the risk of mortality. Exocrine insufficiency in CP has been reported to occur in the range of 42% 99% in different studies. 4,5,20,25 28 At a given point in time, those with pancreatic calcifications have been reported to have a higher probability of having pancreatic exocrine insufficiency. 5 Although Braganza et al 20 found a progressive increase in exocrine insufficiency with worsening pancreatic duct changes, there was a marked overlap in individual pancreatic exocrine function between various stages. On the other hand, Lankisch et al 19 could not find a relationship between pancreatic calcifications and exocrine failure. Our study also did not find a relationship between pancreatic calcification (the severe morphologic stage of CP) and exocrine failure. However, we did not formally test baseline pancreatic function except from clinical presence of steatorrhea, which Figure 5. (A) Common bile duct stricture in CP. (B) Common bile duct stricture in different morphologic stages of CP.
1090 SANDHU ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 5, No. 9 detects only severe (not mild or moderate) pancreatic insufficiency. Another important observation in the current study was that the baseline severity of PD changes in CP does not predict the development of exocrine failure over time. Exocrine function is dependent on adequate acinar structure and function. The positive relationship of type A pain, suggesting the presence of recurrent acute flares of pancreatitis, with the development of exocrine failure noted in this study suggests that parenchymal functional extinction is more dependent on flares of pancreatitis rather than on duct obstruction itself. These data are corroborated further by the failure to preserve pancreatic function by surgically decompressing the pancreatic duct 29 and the lack of correlation between duct morphology and the pancreatic secretory capacity. 20 Ammann et al 5 observed that pancreatic exocrine insufficiency (as measured by fecal chymotrypsin levels of 40 g) developed in 86.6% of the patients over a median time of 5.6 years. They also observed that the pancreatic function deteriorated at approximately 4 years from the time of diagnosis, the time that correlated with the appearance of calcifications. In contrast, we found the risk of developing exocrine failure to be only 28% at 5 years. This is most likely owing to the varying definitions of exocrine failure (abnormal fecal chymotrypsin levels vs steatorrhea) across these studies. The current study further evaluated the risk of developing exocrine failure over time across various baseline stages of CP and was unable to find a relationship between the baseline stage of CP and exocrine failure. These data further indicate that loss of pancreatic parenchymal function is not primarily and solely driven by PD changes in CP. The current study also found that more than 70% of subjects continued to have abdominal pain requiring medications after a median follow-up period of 3.7 years. This is at variance with other publications that found that about 85% of subjects develop pain relief after a median follow-up period of 4.5 years from diagnosis. 5,26,30 Also, in contrast to the lack of relationship between PD changes and development of exocrine failure, the current study found a striking relationship between the severity of duct changes and the persistence of pain. It is possible that the relatively preserved exocrine function in the face of increasing duct obstruction caused increased intraductal and parenchymal pressure. These factors could explain continued pain in our patient population. This also provides a potential explanation for the variable long-term pain relief after PD stenting in CP 31 33 and a direction for future research to evaluate the relative importance of these factors in pain associated with CP. The current study also found a lower risk of developing diabetes ( 20% at 5 y) compared with other published studies ( 53% 78%). 5,7,26,34 This, along with the lower observed rates of exocrine failure, reflects the relatively preserved parenchymal function in the study population. The potential reasons for this include differences in the patient populations studied (European vs American), varying intensities with which ongoing alcohol addiction was treated, relatively shorter median follow-up time, and overall changes in medical care over time. Although the current study was not designed to address these factors specifically, the data provide direction for future research. There was a significant relationship between the stage of CP and the risk of developing a biliary stricture. This is likely to be owing to a common pathophysiologic basis for these events (ie, fibrosis). During the follow-up evaluation, none of these subjects have developed clinically evident biliary cirrhosis. There was no correlation between the stage of CP and the development of regional vascular complications or pseudocysts. The risk of developing pancreatic cancer was about 1.3%, which is substantially higher than that in the general population. This further suggests that CP is a risk factor for pancreatic cancer. In summary, this study indicates that the worsening of PD morphology in CP from mild to severe does not necessarily mean worsening of the exocrine or endocrine functions of the pancreas. Recurrent flares of pancreatitis and type A pain are responsible for acinar injury, which ultimately leads to exocrine insufficiency. Also, the notion that the pain disappears once the severe duct morphology has developed is not necessarily correct. In addition, the likelihood of continuing to have pain involves additional factors. These observations require further validation in large-scale prospective studies. This was a retrospective analysis. Also, the duration of follow-up evaluation was not long enough to precisely assess the loss of pain, exocrine failure, and diabetes over time. Despite this limitation this study provides an insight into the fact that pancreatic morphology based on ductal changes is not sufficient to assess the clinical outcomes. References 1. Sarner M, Cotton PB. Definitions of acute and chronic pancreatitis. Clin Gastroenterol 1984;13:865 870. 2. Sarles H. Definitions and classifications of pancreatitis. Pancreas 1991;6:470 474. 3. Ammann RW, Muellhaupt B. The natural history of pain in alcoholic chronic pancreatitis. Gastroenterology 1999;116:1132 1140. 4. Layer P, Yamamoto H, Kalthoff L, et al. The different courses of early- and late-onset idiopathic and alcoholic chronic pancreatitis. Gastroenterology 1994;107:1481 1487. 5. Ammann RW, Akovbiantz A, Largiader F, et al. Course and outcome of chronic pancreatitis. Longitudinal study of a mixed medical-surgical series of 245 patients. Gastroenterology 1984;86: 820 828. 6. Ammann RW. Pain profile in alcoholic and nonalcoholic chronic pancreatitis (CP). Pancreas 1996;12:315 318. 7. Lankisch PG, Lohr-Happe A, Otto J, et al. Natural course in chronic pancreatitis. Pain, exocrine and endocrine pancreatic insufficiency and prognosis of the disease. Digestion 1993;54:148 155. 8. Ammann RW, Muellhaupt B, Meyenberger C, et al. Alcoholic nonprogressive chronic pancreatitis: prospective long-term study of a large cohort with alcoholic acute pancreatitis (1976 1992). Pancreas 1994;9:365 373. 9. Miyake H, Harada H, Kunichika K, et al. Clinical course and prognosis of chronic pancreatitis. Pancreas 1987;2:378 385. 10. Seno T, Harada H, Ochi K, et al. Serum levels of six pancreatic enzymes as related to the degree of renal dysfunction. Am J Gastroenterol 1995;90:2002 2005. 11. Gumaste VV, Roditis N, Mehta D, et al. Serum lipase levels in nonpancreatic abdominal pain versus acute pancreatitis. Am J Gastroenterol 1993;88:2051 2055. 12. Kloppel G, Maillet B. Pathology of chronic pancreatitis. In: Beger HG, Warshaw AL, Buchler MW, et al, eds. The pancreas. Malden, MA: Blackwell Science, 1998:720. 13. Heider TR, Azeem S, Galanko JA, et al. The natural history of pancreatitis-induced splenic vein thrombosis. Ann Surg 2004; 239:876 880.
September 2007 PANCREATIC DUCT MORPHOLOGY AND EXOCRINE FUNCTION 1091 14. Segal I, Lawson HH, Rabinowitz B, et al. Chronic pancreatitis and the hepatobiliary system. Am J Gastroenterol 1982;77:867 874. 15. Axon AT. Endoscopic retrograde cholangiopancreatography in chronic pancreatitis. Cambridge classification. Radiol Clin North Am 1989;27:39 50. 16. Axon AT, Classen M, Cotton PB, et al. Pancreatography in chronic pancreatitis: international definitions. Gut 1984;25:1107 1112. 17. Ammann RW, Heitz PU, Kloppel G. Course of alcoholic chronic pancreatitis: a prospective clinicomorphological long-term study. Gastroenterology 1996;111:224 231. 18. Hayakawa T, Kondo T, Shibata T, et al. Relationship between pancreatic exocrine function and histological changes in chronic pancreatitis. Am J Gastroenterol 1992;87:1170 1174. 19. Lankisch PG, Otto J, Erkelenz I, et al. Pancreatic calcifications: no indicator of severe exocrine pancreatic insufficiency. Gastroenterology 1986;90:617 621. 20. Braganza JM, Hunt LP, Warwick F. Relationship between pancreatic exocrine function and ductal morphology in chronic pancreatitis. Gastroenterology 1982;82:1341 1347. 21. Nakano S, Horigauchi Y, Takeda T, et al. Comparative diagnostic value of endoscopic pancreatography and pancreatic function tests. Scand J Gastroenterol 1974;9:383 390. 22. Ammann RW, Largiader F, Akovbiantz A. Pain relief by surgery in chronic pancreatitis? Relationship between pain relief, pancreatic dysfunction, and alcohol withdrawal. Scand J Gastroenterol 1979;14:209 215. 23. Malfertheiner P, Buchler M, Stanescu A, et al. Pancreatic morphology and function in relationship to pain in chronic pancreatitis. Int J Pancreatol 1987;2:59 66. 24. Lowenfels AB, Maisonneuve P, Cavallini G, et al. Prognosis of chronic pancreatitis: an international multicenter study. International Pancreatitis Study Group. Am J Gastroenterol 1994;89: 1467 1471. 25. Ammann RW, Buehler H, Muench R, et al. Differences in the natural history of idiopathic (nonalcoholic) and alcoholic chronic pancreatitis. A comparative long-term study of 287 patients. Pancreas 1987;2:368 377. 26. Gambill EE, Baggenstoss AH, Priestley JT. Chronic relapsing pancreatitis. Fate of fifty-six patients encountered in the years 1939 to 1943, inclusive. Gastroenterology 1960;39:404 413. 27. Ammann RW, Hammer B, Fumagalli I. Chronic pancreatitis in Zurich, 1963-1972. Clinical findings and follow-up studies of 102 cases. Digestion 1973;9:404 415. 28. Lankisch MR, Imoto M, Layer P, et al. The effect of small amounts of alcohol on the clinical course of chronic pancreatitis. Mayo Clin Proc 2001;76:242 251. 29. Warshaw AL, Popp JW Jr, Schapiro RH. Long-term patency, pancreatic function, and pain relief after lateral pancreaticojejunostomy for chronic pancreatitis. Gastroenterology 1980;79:289 293. 30. Girdwood AH, Marks IN, Bornman PC, et al. Does progressive pancreatic insufficiency limit pain in calcific pancreatitis with duct stricture or continued alcohol insult? J Clin Gastroenterol 1981; 3:241 245. 31. Ponchon T, Bory RM, Hedelius F, et al. Endoscopic stenting for pain relief in chronic pancreatitis: results of a standardized protocol. Gastrointest Endosc 1995;42:452 456. 32. Ashby K, Lo SK. The role of pancreatic stenting in obstructive ductal disorders other than pancreas divisum. Gastrointest Endosc 1995;42:306 311. 33. Renou C, Grandval P, Ville E, et al. Endoscopic treatment of the main pancreatic duct: correlations among morphology, manometry, and clinical follow-up. Int J Pancreatol 2000;27:143 149. 34. Lankisch PG, Lohr-Happe A, Otto J, Creutzfeldt W. [The natural course of chronic pancreatitis pain, exocrine and endocrine pancreatic insufficiency and prognosis of the disease.] Zentralbl Chir 1995;120:278 286. Address requests for reprints to: Bimaljit S. Sandhu, MBBS, MD, DM, Assistant Professor of Medicine, Division of Gastroenterology, Hepatology & Nutrition, PO Box 980341, VCU Medical Center, 1200 East Broad Street, Richmond, Virginia 23298. e-mail: bsandhu@vcu.edu; fax: (804) 828-4945. Supported by a grant from the National Institutes of Health (K24 DK02755-05 to A.J.S.).