Scandinavian Journal of Surgery 93: 29 33, 2004 ACUTE PANCREATITIS IN BERGEN, NORWAY A study on incidence, etiology and severity H. Gislason 2, A. Horn 1, D. Hoem 1, Å. Andrén-Sandberg 1, A. K. Imsland 3, O. Søreide 1, A. Viste 1 1 Department of Surgery, Haukeland University Hospital, Bergen, Norway 2 Department of Surgery, Landspítali University Hospital, Reykjavik, Iceland 3 Department of statistics Iceland Genomics Corporation, Reykjavik, Iceland ABSTRACT Background: Studies on the incidence and etiology of acute pancreatitis show large regional differences. This study was performed to establish incidence, etiology and severity of acute pancreatitis in the population of Bergen, Norway. Methods: A study of all patients with acute pancreatitis admitted to Haukeland University Hospital over a 10-year period was performed. Information was obtained about the number of patients with acute pancreatitis admitted to the Deaconess Hospital in Bergen. Results: A total of 978 admissions of acute pancreatitis were recorded in these two hospitals giving an incidence of 30.6 per 100 000. Haukeland University Hospital had 757 admissions of acute pancreatitis in 487 patients. Pancreatitis was severe in 20 % (96/ 487) of patients, more often in males (25 %) than in females (14 %). Mortality due to acute pancreatitis was 3%(16/487). Gallstones were found to be an etiological factor in 48.5 % and alcohol consumption in 19 % of patients. The risk of recurrent pancreatitis was 47 % in alcohol induced and 17 % in gallstone induced pancreatitis. The last five years of the study period, endoscopic sphincterotomy of patients with gallstone pancreatitis, resulted in drop in relapse rate from 33 % to 1.6 %. Conclusion: The incidence of acute pancreatitis was found to be 30.6 per 100 000 with 48.5 % associated with gallstones and 17 % alcohol induced. Incidence of first attack was 20/100 000. Pancreatitis was classified as severe in 20 % of cases with a mortality of 3%. Key words: Acute pancreatitis; etiology; incidence; severity; recurrent pancreatitis INTRODUCTION Incidence of acute pancreatitis shows large regional differences with rates varying from 10 to 80 per Correspondence: Hjörtur Gislason, M.D. Department of Surgery Landspítali University Hospital 101 Reykjavik, Iceland Email: hjorturg@landspitali.is 100 000 inhabitants/year (1 11). There are also substantial geographical differences, low in the Netherlands (10) and UK (9) (10 and 24 pts/100 000 inhab/ year), and higher rates in the Nordic countries (1, 5, 6, 11) and USA (4) (35 to 73 pts/100 000 inhab./year). There are also regional and geographical variations relating to precipitating cause. In England (2, 9) and Hong Kong (3) patients with biliary acute pancreatitis predominate, whereas studies from USA (7, 8) and Finland reveal (6) alcohol-induced pancreatitis to be more common.
30 H. Gislason, A. Horn, D. Hoem, Å. Andrén-Sandberg, A. K. Imsland, O. Søreide, A. Viste The purpose of the present study was to record the incidence, etiology and severity of acute and recurrent acute pancreatitis in the population of the Bergen area in western Norway. PATIENTS AND METHODS PATIENTS The Bergen area in western Norway has a population of about 320 000 inhabitants. The area is served by two hospitals, of which the University Hospital admits about 2/3 of acute patients. The other hospital The Deaconess Hospital Haraldsplass has a catchment area of about 100 000 patients. All patients admitted to Haukeland University Hospital in the ten-years period from 1986 to 1995 with acute pancreatitis were studied. Patients were identified from hospital records with the aid of a computer search in the diagnosis. The records were reviewed and diagnosis only accepted in those with consistent clinical presentation: abdominal pain, a serum amylase concentration above 1000 IU per litre (upper normal limit of 300 IU/L) and no other explanation of the pain or acute pancreatic inflammation with or without peripancreatic involvement demonstrated at CT, US or explorative laparotomy according to the Atlanta classification criteria (12). Pancreatitis was classified as severe if the patient died due to pancreatitis, underwent surgery for pancreatic complications during the same admission, or if organ failure and/or local complications, such as necrosis, abscesses, or pseudocysts became manifest. Most of the patients were admitted with acute abdominal pain (461/487) and a standard evaluation was performed according to a national modification of the data collection form for acute abdominal pain (de Dombal)(13). Among the items recorded were hours of symptoms before admission, adiposity, alcohol intake, drugs, and previous and concomitant diseases. Possible etiological factors (gallstones, alcohol, hypercalcemia, hypertriglyceridemia, trauma, viruses etc.) were recorded. Ultrasonography was performed on admission in most patients. CT-scan was done during the hospital stay in most severe cases and in cases suspected to develop severe pancreatitis. In case of suspected gallstones, tumour and in some cases of recurrent pancreatitis, ERCP was performed. Results of ultrasonography, CT-scan and ERCP were recorded. Patients with recurrent pancreatitis were classified as severe if one of the attacks was severe. STATISTICAL ANALYSIS Chi square and log-linear tests were used to analyse discrete clinical variables and Fisher s exact test performed when the number of observations was to small. Probabilities less than 0.05 were accepted as significant. TABLE 1 US findings at admission to hospital in severe and mild acute pancreatitis. n=96 n =391 n=487 Performed 72 (75 %) 325 (83 %) 397 (82 %) Normal 11 (15 %) 75 (23 %) 86 (22 %) Gallstone 29 (40 %) 186 (57 %) 215 (54 %) Dilated CBD 5 0 (7 %) 57 (18 %) 62 (16 %) Pancreatic edema 26 (36 %) 56 (17 %) 82 (21 %) Inflammation * 5 0 (7 %) 5 0 (2 %) 10 0 (3 %) Abdominal fluid 20 (28 %) 30 0 (9 %) 50 (13 %) Note: CBD: Common bile duct, * Inflammation: Pancreatic and peripancreatic inflammation TABLE 2 CT findings in acute pancreatitis. n=96 n =391 n=487 Performed 85 (89 %) 96 (25 %) 181 (37 %) Normal 0 0 14 (15 %) 0 14 0 (8 %) Pancreatic edema 74 (87 %) 60 (63 %) 134 (74 %) Inflammation * 66 (77 %) 50 (52 %) 116 (64 %) Abdominal fluid 54 (64 %) 27 (28 %) 0 81 (45 %) Abdominal abscess 32 (38 %) 0 0 Pseudocyst 33 (39 %) 0 0 Pancreatic necrosis; 24 (28 %) 0 0 1 30 % 0 5 31 70 % 10 71 100 % 0 9 Note: * Inflammation: Pancreatic and peripancreatic inflammation RESULTS INCIDENCE, ETIOLOGY AND SEVERITY Acute pancreatitis was diagnosed in 487 patients (757 admissions with acute pancreatitis) in the ten-year period from 1986 to 1995 at Haukeland University Hospital and all enrolled into the study. In the same period 205 cases of acute pancreatitis were admitted to the Deaconess Hospital Haraldsplass, a total of 978 admissions with acute pancreatitis in Bergen (with a population of 320 000), giving an estimated incidence of 30.6 per 100 000, and an annual incidence 20/ 100 000 for a first attack. The median age of the patients was 64 years (range: 16 97 years), 251 (52 %) were men and 236 women. Pancreatitis was severe in 20 % (96/487) of patients. Severe pancreatitis developed in 25 % of males (62/251), significantly more often than in females, 14 % (34/236) (P <0.05). Ultrasonography was performed in 82 % (397/487) of patients upon admission, results are shown in Table 1. CT was done during the hospital stay in 37 % (181/487) and ERCP was performed in 60 % (291/ 487) of the patients, as shown in Tables 2 and 3. Of 32 patients with abdominal abscesses, 25 were operated on with open drainage of abscess with or without necrosectomi, percutaneous bacteria was identified in seven patients of which four later had percutaneous drainage of abscess. Intended ERCP were not successful in 7.6 % of cases (24/315). The etiology related to severity of the disease is shown in Table 4. Gallstone was found as an etiological factor in 48.5 % (236/487) of patients, signifi-
Incidence and etiology of acute pancreatitis 31 TABLE 3 ERCP findings in acute pancreatitis. n=96 n = 391 n=487 Performed 53 (55 %) 238 (61 %) 291 (60 %) Normal 10 (19 %) 36 (15 %) 0 46 (16 %) Stone in CBD 15 (28 %) 69 (29 %) 0 84 (29 %) Stone in gallbladder 26 (49 %) 189 (79 %)0 215 (74 %) Chronic pancreatitis 0 4 0 (8 %) 0 9 0 (4 %) 0 13 0 (4 %) Pancreas divisum 0 1 0 (6 %) 0 9 0 (4 %) 0 10 0 (4 %) Pancreatic tumour 0 2 0 (4 %) 0 3 0 (1 %) 005 0 (2 %) Note: ERCP: Endoscopic retrograde cholangiopancreatography. CBD: Common bile duct cantly more often in females 58 % (137/236) than in males 42 % (99/236) (P <0.05). Nineteen percent (94/ 487) of the patients had a history of recent alcohol consumption prior to pancreatitis and most of these were males 85 % (80/94) (P <0.001). Severe pancreatitis occurred more often in patients with alcohol etiology, 31 % (29/94), than in patients with biliary etiology, 14 % (32/236) (P <0.01). Severe pancreatitis was found in 25 % (14/56) of idiopathic cases and in 26 % (11/43) of ERCP-induced cases (Table 4). As shown in Table 5, concomitant diseases occurred in 57 % of patients with biliary pancreatitis and in 25 % of patients with alcohol-related pancreatitis (P <0.05). Median age in patients with alcoholic pancreatitis was 44 years (range: 18 77), versus 71 years in patients with biliary pancreatitis (range: 17 90, P<0.001). Post-ERCP pancreatitis was found in 43 patients. During this time 3503 ERCP procedures were performed giving an incidence of 1.2 % (43/3503) post- ERCP pancreatitis; 0.7 % after diagnostic ERCP and 1.8 % after therapeutic ERCP. Pancreatic malignancy was found in 3%of cases (13/487). Carcinoma in the head of pancreas was found during the first stay in five patients and carcinoma in the papilla Vateri in one. In the interval of 3months to two years post-pancreatitis, a carcinoma was found in the head of pancreas in seven more patients. RECURRENT PANCREATITIS TABLE 4 Etiology related to severity in 487 patients with acute pancreatitis. Gallstone disease 32 204 236 (48.5 %) Alcohol abuse 29 0 65 94.0(19 %) Idiopathic 14 0 42 56.0(12 %) Post-ERCP 11 0 32 43.00(9 %) Pancreatic tumour 0 2 0 11 13.00(3 %) Pancreas divisum 0 1 009 10.00(2 %) Hyperlipidemia 0 1 009 10.00(2 %) After surgery / trauma 0 1 0v5 6.00(1 %) Steroids 0 1 003 4.00(1 %) Hyperparathyroidism 0 2 001 3 (0.5 %) Ulcus perforatum 0 1 002 3 (0.5 %) Virus 0 1 002 3 (0.5 %) Other 0 0 006 6.00(1 %) Total 96 (20 %) 391 (80 %) 487 (0.5 %) Note: ERCP: Endoscopic retrograde cholangiopancreatography TABLE 5 Concomitant diseases. Gallstone Alcohol Idiopathic Post-ERCP disease abuse No disease 102 (43 %)0 70 (75 %) 26 (47 %) 21 0 (49 %) Cardiovascular 92 (39 %) 11 (12 %) 17 (30 %) 0 5(11.5 %) Pulmonary 12 0 (5 %) 0 4 0 (4 %) 0 5 0 (9 %) 0 2 00(5 %) Rheumatologic 0 9(3.5 %) 0 4 0 (4 %) 0 4 0 (7 %) 0 5(11.5 %) Endocrine 0 9(3.5 %) 0 2 0 (2 %) 0 1 0 (2 %) 0 7 0 (16 %) Other 130 (6 %) 0 3 0 (3 %) 0 3 0 (5 %) 0 3 00 (7 %) Table 6 shows etiology in recurrent pancreatitis. Of 487 patients admitted for acute pancreatitis, 381 had only one attack, whereas 106 (22 %) were readmitted with 270 relapses giving a total recurrence rate of 1.6 admission/patient (range: 1 13) in this ten-year period. The absolute risk of recurrence was 47 % in patients with alcohol induced pancreatitis compared to 17 % in patients with gallstone pancreatitis (RR = 2.3, p<0.01). In patients with gallstone pancreatitis TABLE 6 Etiology in recurrent acute pancreatitis in 487 patients. No. patients No. recurrences No. patients Absolute risk with recurrence of relapse (%) Gallstone disease 236 88 39 17 % Alcohol abuse 94 111 44 47 % Idiopathic 56 29 8 14 % Pancreas divisum 10 18 4 40 % Post-ERCP 43 16 4 0 9 % Hyperlipidemia 10 7 3 30 % Mucoviscidosis 2 9 2 100 % 0 Pancreatic tumour 13 2 2 15 % Total 487 270 106 22 % Absolute risk of relapse is percentage risk of acute pancreatitis after first attack (no. of patients with at least one relapse/total no. of first attacks 100)
32 H. Gislason, A. Horn, D. Hoem, Å. Andrén-Sandberg, A. K. Imsland, O. Søreide, A. Viste 98 % (86/88) of the recurrence occurred before 1990. In 111 patients admitted with acute biliary pancreatitis from 1985 1990, 37 patients were readmitted with 86 recurrences giving a 33 % risk of relapse due to gallstone pancreatitis for this period. In this group only 13 patients had sphincterotomy with removal of CBD (common bile duct) stones, no prophylactic sphincterotomies were performed. From 1990 to 1995, 125 patients with gallstone-induced pancreatitis were treated by ERCP with endoscopic sphincterotomy with or without removal of CBD stones and often followed by laparoscopic cholecystectomy resulting in only 1.6 % (2/125) relapses (14). Risk of relapse was also high in patients with pancreas divisum (33 %) and in two patients with mucoviscidosis (100 %), as shown in Table 6. OBESITY Obesity was classified as mild if body mass index (BMI = weight(kg)/height 2 (m 2 )) was between 25 35, and morbidly obese if BMI >35. We had data on BMI in 73 % (357/487) of our patients. 30 % (108/357) were classified moderately or morbidly obese. Of 108 obese patients, acute pancreatitis was due to gallstone in 69 % (74/108), due to alcohol in only 12, after ERCP in 8 and due to other causes in 14 patients. As shown in Table 7, severity of acute pancreatitis was not related to obesity. MORTALITY Mortality due to acute pancreatitis was 3%(16/487); 17 % (16/96) in severe pancreatitis. DISCUSSION TABLE 7 Obesity in acute pancreatitis. BMI Severe Mild n=96 n =391 No obesity 53 (55 %) 196 (50 %) Unknown 23 (24 %) 107 (28 %) Moderate (25 35) 12 (13 %) 0 56 (14 %) Severe (> 35) 0 8 0 (8 %) 0 32 0 (8 %) BMI = weight (kg)/height 2 (m 2 ) The incidence of acute pancreatitis found in our study is comparable to the figures found in other recent Scandinavian studies (1, 5, 11), but higher than figures found in British studies (9, 15 17). The etiology of acute pancreatitis in all these studies is, however, predominantly gallstone-induced pancreatitis. Interestingly, in studies where alcohol-induced pancreatitis predominates, the incidence rates tend to be more variable (6 8, 10). Criteria for inclusion in our study were abdominal pain and triple serum amylase levels. The serum amylase values are highly dependent on duration of symptoms at admission to hospital. This gives a high specificity but not a high sensitivity, so some cases might be missed, especially cases that are admitted late. One might suspect alcoholic patients, especially those with mild symptoms, to be admitted later than other patients. Thus, some cases of mild alcohol pancreatitis are probably missed and this could explain, at least partly, the high percentage of severe alcohol pancreatitis (31 %) in our study. Hence, actual incidence figures might be slightly higher than found in our study. As for the etiology of pancreatitis, it is well known that the proportion of idiopathic cases will diminish with a thorough investigation. We performed US examination of abdomen at admission to hospital in 82 % of our patients. ERCP was performed if gallstones were demonstrated on US or if serum livertest results were elevated indicating obstruction and in some cases with idiopathic or recurrent pancreatitis. CT-scan was done in cases suspected to develop severe pancreatitis. As a result of this diagnostic policy, only 12 % of the cases are listed as unclassified (idiopathic). Haukeland Hospital is a University hospital performing many ERCP. This explains the high incidence (9 %) of post-ercp pancreatitis in our study. The diagnosis pancreatitis after ERCP was made if the patient had hyperamylasaemia and abdominal pain requiring treatment and/or hospitalisation. Serum amylase levels post-ercp were only taken on clinical indication, so some mild cases of post-ercp pancreatitis might have been missed. This could explain why post-ercp pancreatitis was severe in as many as 26 % of cases in our study. No case of acute pancreatitis after ERCP was fatal. The presence of gallstone in the gallbladder does not necessarily mean that the pancreatitis is gallstone-induced. However, in our study as in most other studies on this issue, we classified pancreatitis as gallstone-pancreatitis if no other more likely etiology was found. Of patients classified as gallstone pancreatitis cases in our study only 36 % (84/236) were shown to have stones in the CBD at ERCP. It is known that small stones will pass through to the intestines spontaneously. In some cases more than one possible etiological factor was recorded. If pancreatitis was related to consumption of alcohol the pancreatitis was classified as alcohol pancreatitis although gallstones were found. In our data 47 patients had stones in the gallbladder but the pancreatitis was due to some other etiology; ERCP-induced in 35, because of alcohol abuse in 9 and due to pancreatic tumour in three patients. Only a few reports deal with the issue of recurrent acute pancreatitis. We found, as recently demonstrated by Appelros and Borgström (1), that the risk of relapses in patients with alcohol-induced pancreatitis is about 50 %, compared to 33 % in gallstoneinduced pancreatitis before 1990. Our study demonstrates that treatment of patients with gallstone-induced pancreatitis with endoscopic sphincterotomy and/or laparoscopic cholecystectomy almost eliminated relapses, as the relapse rate in 1990 1995 declined to 1.6 %. This we have published in more details elsewhere (14).
Incidence and etiology of acute pancreatitis 33 It is well known that pancreatic tumours may be present with acute pancreatitis. Studies of patients with pancreatic carcinoma have shown that at least one third develops acute pancreatitis (18, 19). In our study acute pancreatitis was the first symptom of pancreatic malignancy in 3%of cases (13 patients). This has also been noted in other studies (1, 19, 20). Thus, in patients with acute pancreatitis where biliary or alcoholic cause has been excluded, pancreatic malignancy should be ruled out by performing CT and MRCP or ERCP. Funnell et al. (21) suggest that adiposity is a major risk factor for severe acute pancreatitis. We found, however, no relationship between adiposity and severity of acute pancreatitis (Table VII). The difference can partly be explained by differences in etiology, as 70 % of cases in their study were alcohol induced compared to 20 % in our study. They found that 17 % (17/99) of patients developed severe pancreatitis and of these 71 % (12/17) were obese, whereof 67 % (8/ 12) were alcohol-induced. In our study, however, 20 % of patients developed severe pancreatitis (96/ 487), and of these 21 % (20/96) were obese, whereof only 10 % (4/40) were alcohol-induced. Most obese patients in our study had gallstone pancreatitis. The present study shows that acute pancreatitis is still a serious disease as 20 % developed a severe disease with a total mortality of 3%. REFERENCES 0 1. Appelros S, Borgström A: Incidence, aetiology and mortality rate of acute pancreatitis over 10 years in a defined urban population in Sweden. Br J Surg 1999;86:456 470 0 2. Beaux AC de, Palmer KR, Carter DC: Factors influencing morbidity and mortality in acute pancreatitis; an analysis of 279 cases. Gut 1995;37:121 126 0 3. Fan ST, Lai ECS, Mok FPT, Lo CM, Zeng SS, Wong J: Prediction of the severity of acute pancreatitis. Am J Surg 1993;166: 262 269 0 4. Go VLW: Etiology and epidemiology of pancreatitis in the United States. In Bradley III EL, ed. Acute pancreatitis: Diagnosis and Therapy. New York: Raven Press 1994;235 239 0 5. Halvorsen FA, Ritland S: Acute pancreatitis in Buskerud country, Norway. Scand J Gastroenterol 1996;31:411 414 0 6. Jaakkola M, Nordback I: Pancreatitis in Finland between 1970 and 1989. Gut 1993;34:1255 1260 0 7. Ranson JHC, Spencer FC: The role of peritoneal lavage in severe acute pancreatitis. Ann Surg 1978;187:565 575 0 8. Renner IG, Savage III WT, Pantoja JL, Renner VJ: Death due to acute pancreatitis. A retrospective analysis of 405 autopsy cases. Dig Dis Sci 1985;30:1005 1018 0 9. Thompson SR, Hendry WS, Mc Farlane GA, Davidson AI: Epidemiology and outcome of acute pancreatitis. Br J Surg 1987; 47:398 401 10. Tran DD, van Schilfgaarde R: Prevalence and mortality from acute pancreatitis in the Netherlands during 1971 1990. Digestion 1994;55:342 343 (abstract) 11. Worning H: Acute interstitial (edematous) pancreatitis in Denmark. In: In Bradley III EL, ed. Acute pancreatitis: Diagnosis and Therapy. New York: Raven Press 1994;256 269 12. Edward L. Bradley III: A clinically based classification system for acute pancreatitis. Summary of the International Symposium on Acute Pancreatitis, Atlanta, GA. Arc Surg 1993; 128:586 590 13. De Dombal FT: Surgical decision making. Oxford: Butteerworth Heinemann, 1993, pp 72 14. Gislason H, Vetrhus M, Horn A, Hoem D, Söndenaa K, Søreide O, Viste A, Andrén-Sandberg Å: Endoscopic sphincterotomy in acute gallstone pancreatitis: A prospective study of the late outcome. Eur J Surg 2001;167:204 208 15. Corfield AP, Cooper MJ, Williamson RCN: Acute pancreatitis: a lethal disease of increasing incidence. Gut 1985;26:724 729 16. Giggs JA, Bourke JB, Katschinski B: The epidemiology of primary acute pancreatitis in Greater Nottingham: 1969 1983. Soc Sci Med 1988;26:79 89 17. Wilson C, Imre CW: Changing patterns of incidence and mortality from acute pancreatitis in Scotland, 1961 1985. Br J Surg 1990;77:731 734 18. Ariyma J: Abnormal glucose tolerance in patients with early pancreatic carcinoma. Int J Pancr 1994;16:91 95 19. Lin A, Feller ER: Pancreatic carcinoma as a cause of unexplained pancreatitis: report of ten cases. Ann Intern Med 1990; 113:166 167 20. Gambill EE: Pancreatitis associated with pancreatic carcinoma: A study of 26 cases. Mayo Clinic Proc March 1971;46:174 177 21. Funnell IC, Bornman PC, Weakley SP, Terblanch J, Marks IN: Obesity: an important prognostic factor in acute pancreatitis. Br J Surg 1993;80:484 486 Received: July 9, 2003 Accepted: November 19, 2003