The New England Journal of Medicine
e-mail icon  FREE NEJM E-TOC    HOME   |   SUBSCRIBE   |   CURRENT ISSUE   |   PAST ISSUES   |   COLLECTIONS   |    Advanced Search
Sign in | Get NEJM's E-Mail Table of Contents — Free | Subscribe
 
Original Article
Volume 328:1433-1437 May 20, 1993 Number 20
NextNext

Pancreatitis and the Risk of Pancreatic Cancer
Albert B. Lowenfels, Patrick Maisonneuve, Giorgio Cavallini, Rudolf W. Ammann, Paul G. Lankisch, Jens R. Andersen, Eugene P. Dimagno, Ake Andren-Sandberg, Lennart Domellof, for The International Pancreatitis Study Group

 

This Article
-Abstract

Commentary
-Letters

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited

More Information
-PubMed Citation
ABSTRACT

Background The results of case-control studies and anecdotal reports suggest that pancreatitis may be a risk factor for pancreatic cancer, but there have been no studies of sufficient size and power to assess the magnitude of the relation between these two diseases.

Methods and Results We undertook a multicenter historical cohort study of 2015 subjects with chronic pancreatitis who were recruited from clinical centers in six countries. A total of 56 cancers were identified among these patients during a mean (±SD) follow-up of 7.4 ±6.2 years. The expected number of cases of cancer calculated from country-specific incidence data and adjusted for age and sex was 2.13, yielding a standardized incidence ratio (the ratio of observed to expected cases) of 26.3 (95 percent confidence interval, 19.9 to 34.2). For subjects with a minimum of two or five years of follow-up, the respective standardized incidence ratios were 16.5 (95 percent confidence interval, 11.1 to 23.7) and 14.4 (95 percent confidence interval, 8.5 to 22.8). The cumulative risk of pancreatic cancer in subjects who were followed for at least 2 years increased steadily, and 10 and 20 years after the diagnosis of pancreatitis, it was 1.8 percent (95 percent confidence interval, 1.0 to 2.6 percent) and 4.0 percent (95 percent confidence interval, 2.0 to 5.9 percent), respectively.

Conclusions The risk of pancreatic cancer is significantly elevated in subjects with chronic pancreatitis and appears to be independent of sex, country, and type of pancreatitis.


Many epidemiologic studies have been performed to detect risk factors for exocrine pancreatic cancer -- a common and usually fatal gastrointestinal tumor. Smoking and reduced consumption of fruits and vegetables appear to be the best established risk factors for this tumor1,2,3,4,5,6,7,8,9. However, case-control studies and case reports have also implicated pancreatitis, a disease often seen in heavy drinkers, as a possible cause. Because of the limited number of cases of pancreatitis in most studies, the evidence linking these two pancreatic disorders is largely anecdotal3,6,10,11,12. In this report, we describe the results of a large historical cohort study of subjects with chronic pancreatitis from six countries, which we undertook to assess the risk of pancreatic cancer in such persons.

Methods

Subjects

Centers with experience in the management of chronic pancreatitis were invited to join an international study group to investigate the relation between pancreatitis and pancreatic cancer. The final group included seven centers located in six countries: Denmark (Copenhagen), Germany (Gottingen-Luneburg), Italy (Verona), Sweden (Lund and Orebro), Switzerland (Zurich), and the United States (Rochester, Minn.). In each center, information was abstracted from the patients' clinical records or transferred directly from existing computer files to a central data base. We were able to obtain information on demographic characteristics and vital status; data referring to the diagnosis, management, and complications of pancreatitis; and data on the smoking and drinking histories of each patient. We were also able to ascertain whether the patients had other types of tumors.

The initial cohort consisted of 2015 subjects who fulfilled the diagnostic criteria for chronic pancreatitis (see below). Person-years for this group were calculated from the date of the diagnosis of pancreatitis to the date of last patient contact or death. We then excluded 462 subjects who had been followed for less than two years or who had been given a diagnosis of pancreatic cancer during this period, as well as 1 subject who had initially undergone a 95 percent pancreatectomy, leaving a total of 1552 subjects thought to be free of pancreatic cancer for at least two years after the onset of pancreatitis. For the analysis of this group the calculation of person-years began two years after the diagnosis of pancreatitis. A final analysis was performed on the 1160 subjects with at least five years of follow-up who were free of pancreatic cancer during this interval; for this analysis the calculation of person-years began five years after the diagnosis of pancreatitis. The recruitment of patients began in 1946 and continued through 1989.

Diagnosis of Chronic Pancreatitis

All patients had been previously hospitalized in university or tertiary hospitals. Criteria for the diagnosis of chronic pancreatitis have been published previously for five of the seven centers13,14,15,16,17,18. The diagnosis was based on a combination of the following criteria: symptoms of chronic epigastric pain, steatorrhea, or both (1530 patients); the presence of pancreatic calcification (988 patients); evidence of pancreatitis on diagnostic procedures (radiographic studies, 814 patients; ultrasonography, 561 patients; endoscopic retrograde cholangiopancreatography, 545 patients; and computed tomography, 248 patients); and results of laboratory tests (secretin-pancreozymin or secretin-cerulein test, fecal fat test, or fecal chymotrypsin assay, 999 patients).

Diagnosis of Pancreatic Cancer

Of the 29 patients with suspected pancreatic cancers that appeared two or more years after the diagnosis of pancreatitis, histologic confirmation was obtained at either surgery or autopsy in 24 (83 percent). In the remaining five patients, the diagnosis was established by exploratory laparotomy in two patients and by the clinical course and positive laboratory or radiologic findings in three patients.

Statistical Analysis

We used published age-stratified (according to five-year age groups), sex-specific, and country-specific data on the incidence of cancer to determine the expected number of cases of pancreatic cancer in the cohort19,20. For countries with more than one cancer registry, we selected the registry nearest to the center: for Switzerland, Zurich; for Germany, Hamburg; and for Italy, the province of Parma. For the United States we used data from the Surveillance, Epidemiology, and End Results study21. The standardized incidence ratio -- the ratio of observed to expected pancreatic cancers -- was used to estimate the relative risk. The 95 percent confidence interval for the standardized incidence ratio was calculated on the basis of the assumption that the observed cases of cancer followed a Poisson distribution. A life-table method was used to estimate the cumulative probability of the development of pancreatic cancer once pancreatitis was diagnosed. For multivariate analysis of the simultaneous effects of the study center, age, sex, and several clinical variables associated with pancreatitis, we used the Cox proportional-hazards model. The final model contained terms for smoking status, drinking status, clinical center, sex, and age. For smoking and drinking status, the subjects were classified as either "ever" having smoked or drunk alcohol or "never" having smoked or drunk alcohol. For analyses of the age at diagnosis of pancreatitis we divided the subjects into the following groups: those 0 to 39 years of age (the reference category), those 40 to 59 years of age, and those 60 years of age or older. Data analysis was performed with SAS procedures22.

Results

Fifty-six pancreatic cancers developed in the initial group of 2015 subjects with pancreatitis, some of whom were followed for less than two years. The expected number of tumors was 2.13, yielding a standardized incidence ratio of 26.3 (95 percent confidence interval, 19.9 to 34.2).

Table 1 provides country-specific and overall characteristics of the 1552 subjects who were followed for at least two years. The type of pancreatitis was listed as alcoholic in 1198 patients (77 percent), idiopathic in 261 patients (17 percent), hereditary in 29 patients (1.9 percent), and due to other factors in 64 patients (4.1 percent). The overall incidence of alcohol consumption was high: 981 subjects (63 percent) consumed five or more drinks per day ( >= 100 g of alcohol per day). Data on smoking status were available for all centers except that in Denmark: 1040 smoked of 1207 whose smoking status was known (86 percent); 651 of 922 subjects (71 percent) for whom specific information about cigarette smoking was available smoked one or more packs per day. Pancreatic calcification, diabetes, and cirrhosis of the liver were documented in 988 (64 percent), 728 (47 percent), and 108 (7 percent) subjects, respectively. A total of 712 subjects (46 percent) underwent surgery for chronic pancreatitis during the follow-up period.

View this table:
[in this window]
[in a new window]
 
Table 1. Country-Specific Characteristics of 1552 Subjects with Chronic Pancreatitis Who Were Followed for >= 2 Years.

 
Of the 1552 subjects, 29 (21 male and 8 female subjects) had evidence of cancer of the pancreas 2 or more years after the diagnosis of pancreatitis during a mean (±SD) follow-up of 7.4 ±5.6 years. The expected number, adjusted for age, sex, and center, was 1.76, yielding a standardized incidence ratio of 16.5 (95 percent confidence interval, 11.1 to 23.7) (Table 2). The incidence ratios were elevated in all countries, and these country-specific estimates appeared to be homogeneous (P >= 0.90). When we restricted the analysis to the subjects with a minimum of five years of follow-up, the overall incidence ratio was slightly reduced (standardized incidence ratio, 14.4; 95 percent confidence interval, 8.5 to 22.8), and all but one of the country-specific incidence ratios remained significantly elevated.

View this table:
[in this window]
[in a new window]
 
Table 2. Country-Specific and Overall Risk of Pancreatic Cancer in 1552 Subjects with Chronic Pancreatitis Who Were Followed for >= 2 Years.

 
The risk of pancreatic cancer in subjects with alcoholic pancreatitis was compared with that in subjects with nonalcoholic pancreatitis (Table 3). The differences were not statistically significant, and the risks appeared to be homogeneous for subjects with a minimum of two or five years of follow-up.

View this table:
[in this window]
[in a new window]
 
Table 3. Risk of Pancreatic Cancer in 1198 Subjects with Chronic Alcoholic Pancreatitis and 354 Subjects with Chronic Nonalcoholic Pancreatitis.

 
Using the Cox proportional-hazards model, we studied the possible effects of several variables on the risk of developing pancreatic cancer. These included demographic variables (age, sex, and country), clinical variables (the type of pancreatitis and the presence or absence of diabetes, calcification, and cirrhosis), and lifestyle variables (alcohol use and smoking status). Only increasing age was significantly related to the development of pancreatic cancer. A model that included terms for smoking status, drinking status, clinical center, sex, and age yielded risk ratios of 3.1 (95 percent confidence interval, 1.1 to 8.6) for subjects who were 40 to 59 years of age and 9.7 (95 percent confidence interval, 2.7 to 35.1) for subjects who were 60 years of age or older, as compared with subjects who were less than 40 years of age.

The cumulative incidence of pancreatic cancer in patients with chronic pancreatitis who were followed for a minimum of two years is shown in Figure 1. Because of the exclusion criteria, no subjects given a diagnosis of pancreatic cancer during the first two years of follow-up were included in the analysis. The cumulative proportion of subjects estimated to have pancreatic cancer 10 years after the diagnosis of pancreatitis was 1.8 percent (95 percent confidence interval, 1.0 to 2.6 percent), and the cumulative proportion with pancreatic cancer 20 years after the diagnosis of pancreatitis was estimated to be 4.0 percent (95 percent confidence interval, 2.0 to 5.9 percent). Over a 20-year period, the cumulative risk of pancreatic cancer appeared to increase nearly linearly.


View larger version (19K):
[in this window]
[in a new window]
 
Figure 1. Cumulative Incidence of Pancreatic Cancer in 1552 Subjects with Chronic Pancreatitis with a Minimum of Two Years of Follow-up.

The vertical lines represent 95 percent confidence intervals. The numbers in parentheses are the numbers of subjects at risk. One additional case of cancer developed after 25 years of follow-up.

 
A total of 425 subjects (27 percent) who were followed for two or more years died during follow-up, including 28 of the 29 subjects with pancreatic cancer. One hundred thirty-seven subjects were known to have died of nonpancreatic cancers.

Discussion

The results of this historical cohort study suggest that patients with chronic pancreatitis have an increased risk of pancreatic cancer; the excess risk was observed in men and women, in alcoholic and nonalcoholic types of pancreatitis, and in all six countries included in the study.

The study population may have been followed more closely than the general population, and this increased surveillance could account for some of the large excess of pancreatic cancer that we observed. However, comparable results have been obtained in countries with different surveillance systems. Moreover, even if the true background incidence of pancreatic cancer in the various countries was twice as high as reported, the overall standardized incidence ratios for subjects with a minimum of two years of follow-up would still be significant (standardized incidence ratio, 8.2; 95 percent confidence interval, 5.5 to 11.8).

Another important problem concerns the possibility of misdiagnosis. We recognize that pancreatic cancer can mimic chronic pancreatitis and that some subjects initially classified as having chronic pancreatitis could conceivably have had pancreatic cancer. Misdiagnosis could explain why subjects in the initial cohort, which included some who had been followed for less than two years, had a higher risk of pancreatic cancer than subjects who had been followed for longer periods. However, pancreatic cancer is a rapidly progressive tumor with an overall five-year survival rate of only 3 percent23. Thus, with longer periods of observation, the probability of misdiagnosing pancreatic cancer as pancreatitis would become less likely. Indeed, we noted that the overall risk ratio appeared stable at two years, and changed only minimally thereafter. Another possible explanation for the higher risks observed during the early period is that about 5 percent of cases of alcoholic pancreatitis and over 50 percent of cases of idiopathic pancreatitis are initially painless17. Thus, the diagnosis of pancreatitis could be delayed until shortly after the onset of cancer-associated abdominal pain.

Classifying cases of pancreatitis as pancreatic cancer would lead to inflated risk ratios. This possibility is unlikely since the diagnosis was verified histologically in 24 of the 29 subjects suspected to have cancer and in 14 of the 18 subjects followed for five or more years. If the analysis is restricted to biopsy-proved cases with long follow-up periods, the results are still significant (standardized incidence ratio, 11.3; 95 percent confidence interval, 6.2 to 18.9).

Finally, part of the excess incidence of pancreatic cancer could be caused by the selective loss to follow-up of patients at low risk of pancreatic cancer. To determine the maximal effect of the loss of patients to follow-up, we reanalyzed our data assuming that in each center all subjects with incomplete follow-up data were followed until the end of the study period and no new cases of pancreatic cancer occurred despite the addition of these extra person-years. After this adjustment, the total number of person-years for subjects with a minimum of two years of follow-up increased from 11,438 to 17,824 and the standardized incidence ratio was 9.3 (95 percent confidence interval, 6.2 to 13.3). Thus, even when we assumed that follow-up was complete and that no additional cancers occurred, the results remained significant.

Diabetes, a suggested risk factor for pancreatic cancer,24,25,26 was present in 728 subjects (47 percent). In these subjects with chronic pancreatitis, however, diabetes was not an independent predictor of pancreatic cancer in the Cox proportional-hazards model.

Heavy consumption of alcohol, which was characteristic of most of the subjects in this cohort, is a recognized cause of chronic pancreatitis, but alcohol has generally not been associated with pancreatic cancer1,2,27,28,29. In contrast, smoking may be a risk factor for both pancreatic cancer8 and chronic pancreatitis30,31,32. The risk of pancreatic cancer has also been reported to be increased in hereditary33 and tropical34 pancreatitis -- diseases not thought to be associated with either drinking or smoking. It is of interest that there may be an excess incidence of pancreatic cancer among young subjects with cystic fibrosis -- another disease characterized by marked pancreatic dysfunction35. One explanation for all these observations is that chronic pancreatitis, whether induced by environmental, genetic, or other causes, is an intermediate stage between normal pancreatic function and some cases of pancreatic cancer.

In all forms of pancreatitis there appears to be cellular dysfunction, glandular destruction, and presumably, increased cell turnover. Increased cell division has been suggested as a potential precursor of cancer in many organs36. The excess risk of pancreatic cancer that we observed in this large group of subjects with various types of pancreatitis is consistent with this hypothesis.

Supported by grants from the C.D. Smithers Foundation and Solvay Pharmaceuticals, Inc.


Source Information

From the Departments of Surgery and Community and Preventive Medicine, New York Medical College, Valhalla (A.B.L.); the Division of Epidemiology and Biostatistics, European Institute of Oncology, Milan, Italy (P.M.); the Service of Gastroenterology and Digestive Endoscopy, University of Verona, Verona, Italy (G.C.); the Gastroenterology Service, Department of Medicine, University Hospital, Zurich, Switzerland (R.W.A.); the Departments of Internal Medicine, University of Gottingen, Gottingen, Germany, and the Municipal Hospital of Luneburg, Luneburg, Germany (P.G.L.); the Department of Gastroenterology, Hvidovre Hospital, University of Copenhagen, Denmark (J.R.A.); the Gastrointestinal Research Unit and Section of Clinical Epidemiology, Mayo Clinic, Rochester, Minn. (E.P.D.); the Department of Surgery, Lund University, Lund, Sweden (A.A.-S.); and the Department of Surgery, Orebro Medical Center Hospital, Orebro, Sweden (L.D.). Presented as an abstract at the annual meeting of the American Pancreatic Association, November 5-6, 1992.In addition to the authors, the members of the International Pancreatitis Study Group included Vincenzo Di Francesco, M.D., Paolo Pederzoli, M.D., Annette Lohr-Happe, M.D., Einar Krag, M.D., L. Joseph Melton, III, M.D., Peter Boyle, Ph.D., C.S. Pitchumoni, M.D., M.P.H., and Pe Shein Wynn, M.D., M.P.H.

Address reprint requests to Dr. Lowenfels at the Department of Surgery, New York Medical College, Valhalla, NY 10595.

References

  1. Gold EB, Gordis L, Diener MD, et al. Diet and other risk factors for cancer of the pancreas. Cancer 1985;55:460-467. [CrossRef][Medline]
  2. Whittemore AS, Paffenbarger RS Jr, Anderson K, Halpern J. Early precursors of pancreatic cancer in college men. J Chronic Dis 1983;36:251-256. [CrossRef][Medline]
  3. Mack TM, Yu MC, Hanisch R, Henderson BE. Pancreas cancer and smoking, beverage consumption, and past medical history. J Natl Cancer Inst 1986;76:49-60.
  4. Ghadirian P, Simard A, Baillargeon J. Tobacco, alcohol, and coffee and cancer of the pancreas: a population-based, case-control study in Quebec, Canada. Cancer 1991;67:2664-2670. [CrossRef][Medline]
  5. Bueno-de-Mesquita HB, Maisonneuve P, Moerman CJ, Runia S, Boyle P. Life-time history of smoking and exocrine carcinoma of the pancreas: a population-based case-control study in the Netherlands. Int J Cancer 1991;49:816-822. [Medline]
  6. La Vecchia C, Negri E, D'Avanzo B, et al. Medical history, diet and pancreatic cancer. Oncology 1990;47:463-466. [Medline]
  7. Wynder EL, Mabuchi K, Maruchi N, Fortner JG. Epidemiology of cancer of the pancreas. J Natl Cancer Inst 1973;50:645-667.
  8. Tobacco smoking. Vol. 38 of IARC monographs on the evaluation of the carcinogenic risk of chemicals to humans. Lyon, France: International Agency for Research on Cancer, 1986.
  9. Steinmetz KA, Potter JD. Vegetables, fruit, and cancer. I. Epidemiology. Cancer Causes Control 1991;2:325-357. [CrossRef][Medline]
  10. Misra SP, Thorat VK, Vij JC, Anand BS. Development of carcinoma in chronic calcific pancreatitis. Int J Pancreatol 1990;6:307-312. [Medline]
  11. Haas O, Guillard G, Rat P, Friedman S, Favre JP. Pancreatic carcinoma developing in chronic pancreatitis: a report of four cases. Hepatogastroenterology 1990;37:350-351. [Medline]
  12. Lin JT, Wang TH, Chen DS, et al. Pancreatic carcinoma associated with chronic calcifying pancreatitis in Taiwan: a case report and review of the literature. Pancreas 1988;3:111-114. [Medline]
  13. Cavallini G, Benini L, Brocco G, et al. The fecal chymotrypsin photometric assay in the evaluation of exocrine pancreatic capacity: comparison with other direct and indirect pancreatic function tests. Pancreas 1989;4:300-304. [Medline]
  14. DiMagno EP. Early diagnosis of chronic pancreatitis and pancreatic cancer. Med Clin North Am 1988;72:979-992. 
  15. Lankisch PG. Diagnosis of abdominal pain: how to distinguish between pancreatic and extrapancreatic causes. Acta Chir Scand 1990;156:273-278. [Medline]
  16. Copenhagen Pancreatitis Study: an interim report from a prospective epidemiological multicentre study. Scand J Gastroenterol 1981;16:305-312. [Medline]
  17. Ammann RW, Akovbiantz A, Largiader F, Schueler G. Course and outcome of chronic pancreatitis: longitudinal study of a mixed medical-surgical series of 245 patients. Gastroenterology 1984;86:820-828. [Medline]
  18. Ammann RW, Schueler G. Chronic pancreatitis, pancreatic cancer, alcohol, and smoking. Gastroenterology 1984;87:744-745. [Medline]
  19. Design considerations. In: Breslow NE, Day NE. Statistical methods in cancer research. Vol. 2. The design and analysis of cohort studies. Lyon, France: International Agency for Research on Cancer, 1987:272-314. (IARC scientific publications no. 82).
  20. Cancer incidence in five continents. Lyon, France: International Agency for Research on Cancer, 1987:1-970.
  21. Cancer statistics review 1973-88. Washington, D.C.: National Cancer Institute, 1991. (NIH publication no. 91-2789.)
  22. SAS system, release 6.07. Cary, N.C.: SAS Institute, 1992.
  23. Warshaw AL, Fernandez-del Castillo C. Pancreatic carcinoma. N Engl J Med 1992;326:455-465. [Medline]
  24. Hiatt RA, Klatsky AL, Armstrong MA. Pancreatic cancer, blood glucose and beverage consumption. Int J Cancer 1988;41:794-797. [Medline]
  25. Mills PK, Beeson WL, Abbey DE, Fraser GE, Phillips RL. Dietary habits and past medical history as related to fatal pancreas cancer risk among Adventists. Cancer 1988;61:2578-2585. [CrossRef][Medline]
  26. Cuzick J, Babiker AG. Pancreatic cancer, alcohol, diabetes mellitus and gall-bladder disease. Int J Cancer 1989;43:415-421. [Medline]
  27. Clavel F, Benhamou E, Auquier A, Tarayre M, Flamant R. Coffee, alcohol, smoking and cancer of the pancreas: a case-control study. Int J Cancer 1989;43:17-21. [Medline]
  28. Velema JP, Walker AM, Gold EB. Alcohol and pancreatic cancer: insufficient epidemiologic evidence for a causal relationship. Epidemiol Rev 1986;8:28-41. [Free Full Text]
  29. Alcohol drinking. Vol. 44 of IARC monographs on the evaluation of carcinogenic risks to humans. Lyon, France: International Agency for Research on Cancer, 1988.
  30. Bourliere M, Barthet M, Berthezene P, Durbec JP, Sarles H. Is tobacco a risk factor for chronic pancreatitis and alcoholic cirrhosis? Gut 1991;32:1392-1395. [Free Full Text]
  31. Lowenfels AB, Zwemer FL, Jhangiani S, Pitchumoni CS. Pancreatitis in a Native American Indian population. Pancreas 1987;2:694-697. [Medline]
  32. Yen S, Hsieh CC, MacMahon B. Consumption of alcohol and tobacco and other risk factors for pancreatitis. Am J Epidemiol 1982;116:407-414. [Free Full Text]
  33. Gross JB. Hereditary pancreatitis. In: Go VLW, ed. The exocrine pancreas: biology, pathobiology, and diseases. New York: Raven Press, 1986:829-39.
  34. Thomas PG, Augustine P, Ramesh H, Rangabashyam N. Observations and surgical management of tropical pancreatitis in Kerala and southern India. World J Surg 1990;14:32-42. [Medline]
  35. Neglia JP, Wielinski CL, Warwick WJ. Cancer risk among patients with cystic fibrosis. J Pediatr 1991;119:764-766. [Medline]
  36. Preston-Martin S, Pike MC, Ross RK, Jones PA, Henderson BE. Increased cell division as a cause of human cancer. Cancer Res 1990;50:7415-7421. [Free Full Text]

 

This Article
-Abstract

Commentary
-Letters

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited

More Information
-PubMed Citation

Related Letters:

Pancreatitis and the Risk of Pancreatic Cancer
Ekbom A., McLaughlin J. K., Nyren O., Lowenfels A. B.
Extract | Full Text  
N Engl J Med 1993; 329:1502-1503, Nov 11, 1993. Correspondence

This article has been cited by other articles:



HOME  |  SUBSCRIBE  |  SEARCH  |  CURRENT ISSUE  |  PAST ISSUES  |  COLLECTIONS  |  PRIVACY  |  HELP  |  beta.nejm.org

Comments and questions? Please contact us.

The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved.