Portal de Periódicos da CAPES

Trends in incidence of acute pancreatitis in a Swedish population: Is there really an increase?

2004; Elsevier BV; Volume: 2; Issue: 9 Linguagem: Inglês

10.1016/s1542-3565(04)00355-6

1542-7714

Björn Lindkvist, Stefan Appelros, Jonas Manjer, Anders Borgström,

Gastrointestinal disorders and treatments

Background & Aims: Recent reports have suggested an increasing incidence of acute pancreatitis, and changing patterns of risk factors, over the past decades. The aim of this study was to investigate trends in the incidence of acute pancreatitis, and risk factors related to the disease, in a general population over a 15-year period. Methods: Clinical, autopsy, and forensic records for all patients with a first attack of acute pancreatitis in Malmö, Sweden, from 1985 to 1999, were validated retrospectively. Evidence for diagnosis was reconsidered and plausible cause was assessed. The incidence of gallstone disease, lung cancer, and alcohol-related conditions in the background population were retrieved from hospital diagnosis records and cancer and cause-of-death registries. Results: A total of 929 first attacks of acute pancreatitis were identified. The total incidence of acute pancreatitis increased by 3.9% per year (95% confidence interval [CI], 2.1–5.8). The incidence of gallstone-related pancreatitis increased by 7.6% per year (95% CI, 4.0–11.4), and this correlated with an increase in the incidence of other gallstone-related conditions (r = 0.68; P = 0.005). Alcohol-related pancreatitis decreased by −5.1% per year (95% CI, −7.4 to −2.8), and this correlated with a decrease in the incidence of delirium tremens (r = 0.75; P = 0.001), mortality from cirrhosis (r = 0.81; P < 0.001), and incidence of lung cancer (r = 0.57; P = 0.026). Conclusions: There was a statistically significant increase in the incidence of acute pancreatitis. Gallstone-related pancreatitis increased, and alcohol-related pancreatitis decreased. Both of these trends were statistically significant and correlated with trends in the incidence of other conditions associated with either gallstone disease or alcohol abuse. Background & Aims: Recent reports have suggested an increasing incidence of acute pancreatitis, and changing patterns of risk factors, over the past decades. The aim of this study was to investigate trends in the incidence of acute pancreatitis, and risk factors related to the disease, in a general population over a 15-year period. Methods: Clinical, autopsy, and forensic records for all patients with a first attack of acute pancreatitis in Malmö, Sweden, from 1985 to 1999, were validated retrospectively. Evidence for diagnosis was reconsidered and plausible cause was assessed. The incidence of gallstone disease, lung cancer, and alcohol-related conditions in the background population were retrieved from hospital diagnosis records and cancer and cause-of-death registries. Results: A total of 929 first attacks of acute pancreatitis were identified. The total incidence of acute pancreatitis increased by 3.9% per year (95% confidence interval [CI], 2.1–5.8). The incidence of gallstone-related pancreatitis increased by 7.6% per year (95% CI, 4.0–11.4), and this correlated with an increase in the incidence of other gallstone-related conditions (r = 0.68; P = 0.005). Alcohol-related pancreatitis decreased by −5.1% per year (95% CI, −7.4 to −2.8), and this correlated with a decrease in the incidence of delirium tremens (r = 0.75; P = 0.001), mortality from cirrhosis (r = 0.81; P < 0.001), and incidence of lung cancer (r = 0.57; P = 0.026). Conclusions: There was a statistically significant increase in the incidence of acute pancreatitis. Gallstone-related pancreatitis increased, and alcohol-related pancreatitis decreased. Both of these trends were statistically significant and correlated with trends in the incidence of other conditions associated with either gallstone disease or alcohol abuse. The annual incidence of acute pancreatitis reported in the literature ranges from 5 to 50 per 100,000,1Appelros 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: 465-470Crossref PubMed Scopus (202) Google Scholar, 2Bourke J.B. Variation in annual incidence of primary acute pancreatitis in Nottingham, 1969–74.Lancet. 1975; 2: 967-969Abstract PubMed Scopus (32) Google Scholar, 3Trapnell J.E. Duncan E.H. Patterns of incidence in acute pancreatitis.BMJ. 1975; 2: 179-183Crossref PubMed Scopus (132) Google Scholar, 4Thomson H.J. Acute pancreatitis in north and north-east Scotland.J R Coll Surg Edinb. 1985; 30: 104-111PubMed Google Scholar, 5Corfield A.P. Cooper M.J. Williamson R.C. Acute pancreatitis a lethal disease of increasing incidence.Gut. 1985; 26: 724-729Crossref PubMed Scopus (281) Google Scholar, 6Jaakkola M. Nordback I. Pancreatitis in Finland between 1970 and 1989.Gut. 1993; 34: 1255-1260Crossref PubMed Scopus (157) Google Scholar, 7McKay C.J. Evans S. Sinclair M. Carter C.R. Imrie C.W. High early mortality rate from acute pancreatitis in Scotland, 1984-1995.Br J Surg. 1999; 86: 1302-1305Crossref PubMed Scopus (261) Google Scholar, 8Lankisch P.G. Schirren C.A. Schmidt H. Schonfelder G. Creutzfeldt W. Etiology and incidence of acute pancreatitis a 20-year study in a single institution.Digestion. 1989; 44: 20-25Crossref PubMed Scopus (40) Google Scholar, 9Eland I.A. Sturkenboom M.J. Wilson J.H. Stricker B.H. Incidence and mortality of acute pancreatitis between 1985 and 1995.Scand J Gastroenterol. 2000; 35: 1110-1116Crossref PubMed Scopus (143) Google Scholar, 10Worning H. Acute pancreatitis in Denmark.Ugeskr Laeger. 1994; 156: 2086-2089PubMed Google Scholar, 11Halvorsen F.A. Ritland S. Acute pancreatitis in Buskerud County, Norway. Incidence and etiology.Scand J Gastroenterol. 1996; 31: 411-414Crossref PubMed Scopus (59) Google Scholar, 12Thomson S.R. Hendry W.S. McFarlane G.A. Davidson A.I. Epidemiology and outcome of acute pancreatitis.Br J Surg. 1987; 74: 398-401Crossref PubMed Scopus (121) Google Scholar, 13Birgisson H. Moller P.H. Birgisson S. Thoroddsen A. Asgeirsson K.S. Sigurjonsson S.V. Magnusson J. Acute pancreatitis a prospective study of its incidence, aetiology, severity, and mortality in Iceland.Eur J Surg. 2002; 168: 278-282Crossref PubMed Scopus (65) Google Scholar, 14Lankisch P.G. Assmus C. Maisonneuve P. Lowenfels A.B. Epidemiology of pancreatic diseases in Luneburg County. A study in a defined German population.Pancreatology. 2002; 2: 469-477Abstract Full Text PDF PubMed Scopus (176) Google Scholar, 15Tinto A. Lloyd D.A. Kang J.Y. Majeed A. Ellis C. Williamson R.C. Maxwell J.D. Acute and chronic pancreatitis—diseases on the rise a study of hospital admissions in England 1989/90-1999/2000.Aliment Pharmacol Ther. 2002; 16: 2097-2105Crossref PubMed Scopus (106) Google Scholar and several reports suggest an increased incidence of acute pancreatitis over the past decades.3Trapnell J.E. Duncan E.H. Patterns of incidence in acute pancreatitis.BMJ. 1975; 2: 179-183Crossref PubMed Scopus (132) Google Scholar, 5Corfield A.P. Cooper M.J. Williamson R.C. Acute pancreatitis a lethal disease of increasing incidence.Gut. 1985; 26: 724-729Crossref PubMed Scopus (281) Google Scholar, 6Jaakkola M. Nordback I. Pancreatitis in Finland between 1970 and 1989.Gut. 1993; 34: 1255-1260Crossref PubMed Scopus (157) Google Scholar, 7McKay C.J. Evans S. Sinclair M. Carter C.R. Imrie C.W. High early mortality rate from acute pancreatitis in Scotland, 1984-1995.Br J Surg. 1999; 86: 1302-1305Crossref PubMed Scopus (261) Google Scholar, 8Lankisch P.G. Schirren C.A. Schmidt H. Schonfelder G. Creutzfeldt W. Etiology and incidence of acute pancreatitis a 20-year study in a single institution.Digestion. 1989; 44: 20-25Crossref PubMed Scopus (40) Google Scholar, 9Eland I.A. Sturkenboom M.J. Wilson J.H. Stricker B.H. Incidence and mortality of acute pancreatitis between 1985 and 1995.Scand J Gastroenterol. 2000; 35: 1110-1116Crossref PubMed Scopus (143) Google Scholar, 10Worning H. Acute pancreatitis in Denmark.Ugeskr Laeger. 1994; 156: 2086-2089PubMed Google Scholar, 15Tinto A. Lloyd D.A. Kang J.Y. Majeed A. Ellis C. Williamson R.C. Maxwell J.D. Acute and chronic pancreatitis—diseases on the rise a study of hospital admissions in England 1989/90-1999/2000.Aliment Pharmacol Ther. 2002; 16: 2097-2105Crossref PubMed Scopus (106) Google Scholar, 16Wilson C. Imrie C.W. Changing patterns of incidence and mortality from acute pancreatitis in Scotland, 1961–1985.Br J Surg. 1990; 77: 731-734Crossref PubMed Scopus (93) Google Scholar, 17Giggs J.A. Bourke J.B. Katschinski B. The epidemiology of primary acute pancreatitis in Greater Nottingham 1969–1983.Soc Sci Med. 1988; 26: 79-89Crossref PubMed Scopus (24) Google Scholar, 18Floyd A. Pedersen L. Nielsen G.L. Thorladcius-Ussing O. Sorensen H.T. Secular trends in incidence and 30-day case fatality of acute pancreatitis in North Jutland County, Denmark a register-based study from 1981–2000.Scand J Gastroenterol. 2002; 37: 1461-1465Crossref PubMed Scopus (100) Google Scholar Suggested explanations are changes in established risk factors (i.e., increased gallstone prevalence or changed drinking habits).3Trapnell J.E. Duncan E.H. Patterns of incidence in acute pancreatitis.BMJ. 1975; 2: 179-183Crossref PubMed Scopus (132) Google Scholar, 5Corfield A.P. Cooper M.J. Williamson R.C. Acute pancreatitis a lethal disease of increasing incidence.Gut. 1985; 26: 724-729Crossref PubMed Scopus (281) Google Scholar, 6Jaakkola M. Nordback I. Pancreatitis in Finland between 1970 and 1989.Gut. 1993; 34: 1255-1260Crossref PubMed Scopus (157) Google Scholar, 8Lankisch P.G. Schirren C.A. Schmidt H. Schonfelder G. Creutzfeldt W. Etiology and incidence of acute pancreatitis a 20-year study in a single institution.Digestion. 1989; 44: 20-25Crossref PubMed Scopus (40) Google Scholar, 9Eland I.A. Sturkenboom M.J. Wilson J.H. Stricker B.H. Incidence and mortality of acute pancreatitis between 1985 and 1995.Scand J Gastroenterol. 2000; 35: 1110-1116Crossref PubMed Scopus (143) Google Scholar, 15Tinto A. Lloyd D.A. Kang J.Y. Majeed A. Ellis C. Williamson R.C. Maxwell J.D. Acute and chronic pancreatitis—diseases on the rise a study of hospital admissions in England 1989/90-1999/2000.Aliment Pharmacol Ther. 2002; 16: 2097-2105Crossref PubMed Scopus (106) Google Scholar, 19Mero M. Changing aetiology of acute pancreatitis.Ann Chir Gynaecol. 1982; 71: 126-129PubMed Google Scholar, 20Svensson J.O. Norback B. Bokey E.L. Edlund Y. Changing pattern in aetiology of pancreatitis in an urban Swedish area.Br J Surg. 1979; 66: 159-161Crossref PubMed Scopus (54) Google Scholar Improved diagnostic procedures also may have contributed.3Trapnell J.E. Duncan E.H. Patterns of incidence in acute pancreatitis.BMJ. 1975; 2: 179-183Crossref PubMed Scopus (132) Google Scholar, 16Wilson C. Imrie C.W. Changing patterns of incidence and mortality from acute pancreatitis in Scotland, 1961–1985.Br J Surg. 1990; 77: 731-734Crossref PubMed Scopus (93) Google Scholar Many studies on the incidence of acute pancreatitis are retrospective case-record studies from referral centers that do not serve a defined population, which implies a risk for selection bias. In Malmö, Sweden, with a population of about 250,000 inhabitants, there is only one hospital for somatic care, and there are no referrals of patients with acute pancreatitis to or from other hospitals. Exact population statistics are available. This allows complete retrieval of representative cases, and calculation of age- and sex-standardized incidence rates.A previous cross-sectional study has identified 547 first attacks of acute pancreatitis and assessed plausible cause.1Appelros 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: 465-470Crossref PubMed Scopus (202) Google Scholar That material covered too short a period to allow analysis of changes in incidence rates. The present study is a population-based longitudinal study on 929 consecutive first attacks of acute pancreatitis from 1985 to 1999. The aim of this study was to investigate trends in incidence of acute pancreatitis, and risk factors related to the disease.Patients and methodsMalmö is the third largest city in Sweden, with a population of approximately 250,000 inhabitants. Exact data on the size of the population, and its age and sex constitution for each year, was retrieved from the unit of planning and statistics at the city council.21The Unit of Planning and Statistics, Malmö City Council. Area Statistics for Malmö 1985-99. The Unit of Planning and Statistics, Malmö City Council, 1985–1999.Google ScholarDiagnostic criteria of acute pancreatitisIn the period from 1985 to 1999, cases of acute pancreatitis were identified from clinical, autopsy, and forensic diagnosis records by a computer search for diagnostic codes. Clinical notes, forensic protocols, and autopsy reports were validated according to a standard protocol. The diagnosis was accepted in cases with a clinical history of acute abdominal pain, confirmed at physical examination, in combination with an increase in serum amylase level. Evidence from laparoscopy, laparotomy, or autopsy also was accepted for diagnosis. Patients with a previous history of chronic pancreatitis were excluded. Attacks were classified as first attacks if there was no evidence of earlier attacks in clinical notes or diagnostic records. Severe cases were defined according to the Atlanta classification.22Bradley III, E.L. A clinically based classification system for acute pancreatitis. Summary of the International Symposium on Acute Pancreatitis, Atlanta, GA, September 11 through 13, 1992.Arch Surg. 1993; 128: 586-590Crossref PubMed Scopus (2495) Google ScholarGallstone disease and alcohol abuse in patients with acute pancreatitisGallstone diseaseGallstone disease was diagnosed using clinical and conventional imaging procedures (i.e., ultrasonography, endoscopic retrograde cholangiopancreatography, or computed tomography). The attack of acute pancreatitis was considered as related to gallstone disease only when there was radiologic evidence for gallstones and no signs of other risk factors, such as alcohol abuse, were present.Alcohol abuseThe attack of acute pancreatitis was considered as related to alcohol abuse when the patient reported a high regular intake of alcohol, or an alcoholic bout, directly before the onset of the disease, and no signs of other risk factors, such as gallstones, were present.Other risk factorsUncommon factors related to the attack of pancreatitis were endoscopic retrograde cholangiopancreatography, pancreatic cancer, and postoperative pancreatitis. In the statistical analysis this group was analyzed together with attacks of unknown cause.Unknown causeThe cause was considered unknown when no risk factors could be found, or when alcohol abuse and gallstone disease were present at the same time and both factors were judged as equally plausible causes of the attack of acute pancreatitis.Gallstone disease, alcohol abuse, and lung cancer in the population at riskCholedocolithiasis, cholecystolithiasis, and cholecystitis were used as markers for gallstone disease. The incidence of delirium tremens and mortality from cirrhosis were used as markers for alcohol abuse. The incidence of lung cancer was considered as a marker for smoking. Cases of gallstone-related conditions and delirium tremens were identified from hospital diagnosis records by a computer search for relevant diagnostic codes. Only the main diagnosis was accepted, and only the first occasion during the period was considered. Cases of lung cancer were identified from the regional Cancer Registry, a part of the Swedish Cancer Registry. Information on patients who died from cirrhosis was retrieved from the regional Cause-of-Death Registry.Statistical analysisMean age at diagnosis of the first attack of acute pancreatitis was calculated for different subgroups of patients. Annual age- and sex-standardized incidence was calculated by using direct standardization. The total population in Malmö in 1985 was used as standard. All analyses also were stratified for sex, yielding age-standardized rates.Time trends with regard to incidence of pancreatitis, and related conditions, were examined regarding the incidence rate as a dependent variable (y-axis) and the year as an independent variable (x-axis). All tests were 2 tailed and a P value of less than 0.05 was regarded as statistically significant. First, Pearson’s correlation coefficients (r), with corresponding P values, were used to reveal whether or not there was a trend in incidence over the period (i.e., how well incidence rates were arranged on a straight line). Second, linear regression analysis revealed the strength of the association (i.e., the slope of the line). This was expressed as the average annual change in percent with a 95% confidence interval (CI).23The National Board of Health and WelfareCancer incidence in Sweden 2000. Appendix 2. The National Board of Health and Welfare, Stockholm, Sweden2002: 1-3Google Scholar To investigate the homogeneity of different causative subgroups of patients, trends in mean age were investigated using Pearson’s correlation coefficient and linear regression, as described earlier.The incidence of pancreatitis associated with gallstones was compared with the incidence of gallstone disease and lung cancer in the background population. Correspondingly, the incidence of pancreatitis associated with alcohol abuse was compared with the incidence of delirium tremens and lung cancer and the mortality rate from cirrhosis. The incidence of unspecified cases of pancreatitis was compared with all 4 conditions. In these comparisons, the logarithmic incidence of pancreatitis was entered as a dependent variable (y-axis), and the logarithmic incidence of related conditions was entered as an independent variables (x-axis). Pearson’s correlation coefficient was used as described earlier. The strength of the association was expressed as β-coefficients (95% CI) (i.e., the slope of the line).ResultsIncidence of acute pancreatitisThe diagnosis of acute pancreatitis was found 1444 times during the investigated period, 1400 times in clinical records, 28 times in autopsy records, and 16 times in forensic records. From this material, 929 correctly diagnosed first attacks of acute pancreatitis were identified, 493 (53.1%) in men and 436 (46.9%) in women (Table 1). Annual age- and sex-standardized incidence increased by 3.9% (95% CI, 2.1–5.8) per year (Figure 1, Table 2). The increase was statistically significant in women, but not in men.Table 1Number of Cases and Age at OccurrenceAllMenWomennAge (SD), yrnAge (SD), yrnAge (SD), yrAll pancreatitis92959.4 (17.9)49357.6 (15.8)43661.4 (19.8)Alcohol-associated pancreatitis22847.6 (10.8)18648.0 (10.8)4245.8 (10.8)Gallstone-associated pancreatitis39265.3 (18.1)14267.6 (13.3)25064.0 (20.3)Unknown/other pancreatitis30960.5 (17.7)16559.8 (16.5)14461.4 (19.1)Delirium tremens15351.4 (11.3)13551.5 (10.9)1850.3 (14.5)Mortality from cirrhosis34563.8 (12.9)23461.7 (12.1)11168.4 (13.6)Lung cancer238369.4 (10.7)161470.3 (10.1)76967.6 (11.9) Open table in a new tab Table 2Annual Incidence Change in PercentAllMenWomenr (P)Annual change in % (95% CI)r (P)Annual change in % (95% CI)r (P)Annual change in % (95% CI)All pancreatitis0.81 (<0.001)3.9 (2.1–5.8)0.36 (0.19)1.8 (−0.8 to 4.5)0.73 (0.002)6.5 (2.7–10.4)Alcohol-associated pancreatitis−0.77 (0.001)−5.1 (−7.4 to −2.8)−0.68 (0.006)−5.5 (−8.7 to −2.2)−0.34 (0.22)−2.1 (−8.1 to 4.3)Gallstone-associated pancreatitis0.78 (0.001)7.6 (4.0–11.4)0.73 (0.002)7.4 (2.8–12.2)0.68 (0.005)8.0 (1.9–14.5)Other or unknown pancreatitis0.85 (<0.001)5.8 (3.5–8.1)0.54 (0.039)3.8 (0.3–7.5)0.68 (0.005)7.5 (1.8–13.5)Gallstone disease0.59 (0.021)1.8 (0.3–3.5)0.57 (0.027)1.8 (0.2–3.4)0.55 (0.034)1.9 (0.1–3.7)Delirium tremens−0.88 (<0.001)−13.0 (−16.7 to −9.1)−0.87 (<0.001)−13 (−17.0 to −9.3)−0.50 (0.058)−5.4 (−14 to 4.4)Mortality from cirrhosis−0.74 (0.002)−4.4 (−6.8 to −2.1)−0.73 (0.002)−4.3 (−6.7 to −1.9)−0.55 (0.033)−5.7 (−11 to −0.60)Lung cancer−0.78 (0.01)−1.6 (−2.4 to −0.8)−0.87 (<0.001)−4.0 (−5.4 to −2.7)0.76 (0.01)2.6 (1.2–4.0) Open table in a new tab Trends in mean ageMean age at diagnosis is reported in Table 1. In all patients with a first attack of acute pancreatitis, there was a statistically significant increase in the annual mean age of 0.4 years per year (95% CI, 0.2–0.6). There was no statistically significant change in annual mean age of patients with gallstone-related pancreatitis, 0.0 years per year (95% CI, −0.3 to 0.4), or in the group of unknown and miscellaneous risk factors, −0.1 years per year (95% CI, −0.5 to 0.3). The mean age of patients with an alcohol-related first attack of acute pancreatitis increased over time, statistically significantly, with 0.4 years per year (95% CI, 0.1–0.7).Risk factors in patients with acute pancreatitis and in the population at riskGallstone diseaseGallstone disease was the most common risk factor, present in 42.2% (392 of 929) of all cases with a first attack of acute pancreatitis. The incidence of gallstone-related pancreatitis increased by 7.6% (95% CI, 4.0–11.4) per year (Table 2). The increase was statistically significant among women, as well as among men (Figure 2, Table 2). In the general population there was a statistically significant increase in gallstone-related disease (Table 2). The incidence of gallstone-related pancreatitis was statistically significantly correlated to the incidence of gallstone-related disease in both men and women (Table 3, Figure 3).Figure 2Age-standardized incidence of gallstone-associated pancreatitis. ●, Men; ○, women.View Large Image Figure ViewerDownload (PPT)Table 3Correlation Between the Logarithmic Incidence of Different Subgroups of Pancreatitis and the Logarithmic Incidence of Gallstone Disease, Delirium Tremens, and Lung Cancer and Logarithmic Mortality From Cirrhosis in the General PopulationPancreatitis associated with:Gallstone diseaseMortality from cirrhosisDelirium tremensLung cancerr (P)β (95% CI)r (P)β (95% CI)r (P)β (95% CI)r (P)β (95% CI)AllGallstones0.65 (0.008)0.14 (0.042–0.23)————−0.67 (0.007)−0.46 (−0.78 to −0.15)Alcohol——0.79 (<0.001)0.58 (0.31–0.85)0.66 (0.007)0.37 (0.12–0.62)0.56 (0.03)0.15 (0.017–0.28)Unknown/other0.67 (0.007)0.081 (0.027–0.14)−0.76 (0.01)−0.85 (−1.3 to −0.41)−0.53 (0.041)−0.46 (−0.89 to −0.023)−0.72 (0.03)−0.30 (−0.46 to −0.13)WomenGallstones0.57 (0.026)0.12 (0.016–0.22)————0.59 (0.02)0.61 (0.11–1.1)Alcohol——0.09 (0.75)0.051 (−0.30 to 0.40)0.32 (0.25)0.33 (−0.26 to 0.91)−0.26 (0.34)−0.053 (−0.17 to 0.063)Unknown/other0.23 (0.42)0.030 (−0.046 to 0.11)−0.24 (0.39)−0.46 (−1.6 to 0.66)−0.13 (0.65)−0.44 (−2.46 to 1.59)0.61 (0.02)0.41 (0.091–0.72)MenGallstones0.66 (0.007)0.16 (0.050–0.26)————−0.49 (0.07)−0.078 (−0.16 to 0.006)Alcohol——0.59 (0.02)0.53 (0.092–0.96)0.62 (0.014)0.36 (0.085–0.63)0.67 (0.006)0.090 (0.030–0.15)Unknown/other0.56 (0.03)0.11 (0.012–0.20)−0.31 (0.27)−0.26 (−0.75 to 0.23)−0.41 (0.13)−0.22 (−0.52 to 0.078)−0.36 (0.19)−0.046 (−0.12 to 0.025)NOTE. The slope of the line is expressed as β-coefficients with a 95% CI. Open table in a new tab Figure 3Correlation between incidence of gallstone-associated pancreatitis and incidence of gallstone disease other than pancreatitis. ●, Men; ○, women.View Large Image Figure ViewerDownload (PPT)Alcohol abuseAlcohol abuse was the second most common risk factor, found in 24.5% (228 of 929) of all cases. The incidence of alcohol-related pancreatitis decreased by 5.1% (95% CI, 7.4–2.8) per year. This decrease consisted of a slight, and not statistically significant, decrease among women and a statistically significant decrease among men (Table 2, Figure 4).Figure 4Age-standardized incidence of alcohol-associated pancreatitis. ●, Men; ○, women.View Large Image Figure ViewerDownload (PPT)The decrease in alcohol-related pancreatitis among men was statistically significantly correlated to a decrease in delirium tremens, and a decrease in the mortality from cirrhosis in the male general population. There was no correlation between alcohol-related pancreatitis and delirium tremens or mortality from cirrhosis among women (Table 3).Lung cancerThe incidence of lung cancer decreased among men, and increased among women in the population at risk (Table 2). The incidence of alcohol-related pancreatitis was statistically significantly correlated with the incidence of lung cancer in men but not in women (Table 3).Other or unknown risk factorsThe remaining 33.3% (309 of 929) of all first attacks of acute pancreatitis consisted of attacks related to other or unknown risk factors (Table 2). This group consisted of 94 attacks of miscellaneous causes and 215 attacks of unknown cause. In 4.7% (10 of 215) of the attacks of acute pancreatitis that were classified as unknown, both gallstone disease and alcohol abuse were present and retrospective assessment of one major causative factor was not possible. The incidence of attacks of acute pancreatitis related to other or unknown risk factors increased statistically significantly during the investigated period, but its proportion of the total number of first attacks of acute pancreatitis remained similar (data not shown).Proportion of severe cases and case-fatality rateThe proportions of severe cases according to the Atlanta classification,22Bradley III, E.L. A clinically based classification system for acute pancreatitis. Summary of the International Symposium on Acute Pancreatitis, Atlanta, GA, September 11 through 13, 1992.Arch Surg. 1993; 128: 586-590Crossref PubMed Scopus (2495) Google Scholar and case-fatality rate, were similar throughout the investigated period. The percentage of severe cases was 13.6% (126 of 929) during the whole period, from 1985 to 1994 it was 12.7% (70 of 550), and from 1995 to 1999 it was 14.8% (56 of 379). The case-fatality rate was 5.7% (53 of 929) during the whole period, from 1985 to 1994 it was 5.6% (31 of 550), and from 1995 to 1999 it was 5.7% (22 of 379).DiscussionThis is, to our knowledge, the largest study to date on trends in incidence of first attacks of acute pancreatitis with individual verification of all cases in a defined population. Both clinical and forensic data are included. A statistically significant increase in the incidence of first attacks of acute pancreatitis was found. Gallstone-related pancreatitis increased markedly, and this correlated with an increase in the incidence of gallstone-related disease in the population at risk. Alcohol-related pancreatitis decreased, and this correlated with a decrease in the incidence of delirium tremens and lung cancer, and mortality from cirrhosis in the population at risk.There is only one hospital in Malmö, and there are no referrals of patients to or from this hospital. Therefore, we consider the risk for selection bias as low, allowing reliable estimates of incidence.Large studies have been performed on the incidence of acute pancreatitis using diagnosis registers without individual validation of cases.6Jaakkola M. Nordback I. Pancreatitis in Finland between 1970 and 1989.Gut. 1993; 34: 1255-1260Crossref PubMed Scopus (157) Google Scholar, 7McKay C.J. Evans S. Sinclair M. Carter C.R. Imrie C.W. High early mortality rate from acute pancreatitis in Scotland, 1984-1995.Br J Surg. 1999; 86: 1302-1305Crossref PubMed Scopus (261) Google Scholar, 9Eland I.A. Sturkenboom M.J. Wilson J.H. Stricker B.H. Incidence and mortality of acute pancreatitis between 1985 and 1995.Scand J Gastroenterol. 2000; 35: 1110-1116Crossref PubMed Scopus (143) Google Scholar, 15Tinto A. Lloyd D.A. Kang J.Y. Majeed A. Ellis C. Williamson R.C. Maxwell J.D. Acute and chronic pancreatitis—diseases on the rise a study of hospital admissions in England 1989/90-1999/2000.Aliment Pharmacol Ther. 2002; 16: 2097-2105Crossref PubMed Scopus (106) Google Scholar A report from the Netherlands9Eland I.A. Sturkenboom M.J. Wilson J.H. Stricker B.H. Incidence and mortality of acute pancreatitis between 1985 and 1995.Scand J Gastroenterol. 2000; 35: 1110-1116Crossref PubMed Scopus (143) Google Scholar showed that in a national register only 83% were diagnosed correctly. This may lead to under- or overestimation of true incidence rates. Such a misclassification bias is probably less likely in the present study because clinical notes and autopsy records were reviewed for all cases.Another problem with studies without individual case validation is that recurrences may have been mistaken for first attacks. This can lead to an overestimation of first attack incidence rates. This probably was less of a problem in the present study because previous medical history was reviewed in all cases of pancreatitis.Improved diagnostic procedures have been suggested as the explanation for an increased incidence of acute pancreatitis when reported by other investigators.3Trapnell J.E. Duncan E.H. Patterns of incidence in acute pancreatitis.BMJ. 1975; 2: 179-183Crossref PubMed Scopus (132) Google Scholar, 16Wilson C. Imrie C.W. Changing patterns of incidence and mortality from acute pancreatitis in Scotland, 1961–1985.Br J Surg. 1990; 77: 731-734Crossref PubMed Scopus (93) Google Scholar If improved diagnostic procedures could explain the increase in the incidence of acute pancreatitis, a proportional increase in cases related to alcohol abuse and cases related to gallstone disease would be expected. On the contrary, according to our data, alcohol-related pancreatitis decreased whereas gallstone-related pancreatitis increased. If improved diagnostic procedures were the reason for the increas