(18)

US DEPARTMENT OF COMMERCE: 1980 census

of population, vol 1. DHEW Publ No. (PHS)PL80-l-B49). Washington, DC: US Govt Print Off, 1982

Relationship of Polyps to Cancer of the Large Intestine Brad D. Simons, Alan S. Morrison,* Robert Lev, Wendy Verhoek-Oftedahl

962

Cancer of the colon and rectum accounts for approximately 11% of cancer deaths in U.S. men and 13% of those in U.S. women (/). There is pathologic and epidemiologic evidence that persons with sporadic (nonfamilial) adenomatous polyps of the large intestine are at high risk of colorectal cancer (2). However, the relationship between adenomatous polyps and colorectal cancer mortality is uncertain, as are the relationships between features of the benign lesions and risk. We conducted a retrospective follow-up study to further investigate these issues. The primary goal was to evaluate the colorectal cancer mortality rate in persons who have had a histologically confirmed benign colorectal polyp.

Subjects and Methods Subjects consisted of residents of Rhode Island who were 24 through 79 years of age when they had partial or total removal of one or more benign colorectal polyps in the period 1959 through 1975. Eligibility for the study was limited to persons who did not have intestinal polyposis or chronic inflammatory disease of the bowel and who had had no previous intestinal surgery and no history of intestinal cancer diagnosed before or within 3 months after diagnosis of the polyp. Subjects were identified initially by review of original pathology reports of surgical specimens at hospitals in Rhode Island. Excision or biopsy of the polyp could have been done on either an inpatient or an outpatient basis. An original aim of the project was to identify patients statewide. However, three small hospitals refused to participate, and pertinent records had been lost or destroyed for various periods at some of the participating hospitals. For each pathology report of a polyp, the medical record at the hospital was retrieved and an abstract of the record was made if the individual was found to be eligible. Information recorded included date of birth, date of the initial polypectomy, surgical and endoscopic

procedures, pathologic diagnoses, size, number, and site of polyps within the large intestine, and identifying data for use in follow-up. No medical record could be found for a number of potential subjects who had been outpatients. Information on eligibility for these patients was requested from the referring physicians whose names were indicated on the original pathology reports. The record abstract also included the following data on hospitalizations for colorectal cancer: dates, surgical procedures, pathologic diagnoses, date of the most recent observation, vital status, and the causes of death for those subjects who died. The latter three items were obtained from medical records, forms returned by physicians of outpatients, or other follow-up procedures described below. A total of 2872 eligible subjects were identified; 584 had polyp surgery as outpatients. The mean age at entry into the study was 59.7 years (range, 24-79 years), with 44% in the 50- to 64-year age group. Twenty-one percent were less than 50 years old, and 35% were 65 years of age or older. Death certificates in Rhode Island were searched manually beginning with the last date on which the subject was known to be alive. The minimum criteria for matching the subject with the death certificate were name, date of birth, and one of the following: address, place of birth, Social Security account number, name of next of kin, and occupation. For some out-of-state deaths, the state of origin of the Social Security account number was used. In the evaluation of potential matches, allowance was made

Received March 18. 1991; revised March 3. 1992; accepted March 5, 1992. Supported by Public Health Service grant CA-42267 from the National Cancer Institute. National Institutes of Health, Department of Health and Human Services. B. D. Simons, Brown University, Providence. R.I. A. S. Morrison, W. Verhoek-Oftedahl. Department of Community Health, Brown University, Providence, R.I. R. Lev, Department of Pathology. Roger Williams Hospital and Brown University. Providence. R.I. We thank Joyce Coutu Babcock for her contributions to this project. We also thank the staff of the pathology and medical records departments in Rhode Island hospitals and the staff of the Division of Vital Statistics. Rhode Island Department of Health, for their cooperation. *Correspondence to: Alan S. Morrison, M.D., Department of Community Health, Brown University. Providence, RI 02912.

Journal of the National Cancer Institute

Downloaded from http://jnci.oxfordjournals.org/ at Carleton University on June 12, 2015

Background: Pathologic and epidemiologic evidence indicates that patients with sporadic (nonfamilial) adenomatous polyps of the large intestine are at high risk of developing colorectal cancer. Purpose: Our primary goal in this study was to evaluate the colorectal cancer mortality rate among persons who have had a histologically confirmed benign colorectal polyp. Methods: We used the retrospective follow-up method to evaluate the risk of death from colorectal cancer in 2872 Rhode Island men and women who were 24 through 79 years of age at the time of surgery for benign polyps in the years 1959 through 1975. Results: Among 2872 subjects, the mortality from colorectal cancer, standardized for age, sex, and calendar time, was estimated as 1.74 (95% confidence interval = 1.44-2.09) times the rate in the general population of Rhode Island residents. Colorectal cancer mortality was higher in the first 5 years of follow-up than it was later. There was little relationship between the numbers of polyps and colorectal cancer mortality, and there was only a modest association between the size of polyps and mortality. Colorectal cancer mortality was more than twice as high in subjects whose polyps were proximal to the sigmoid compared with those with sigmoid or rectal polyps. The observed elevation of risk of colorectal cancer was almost entirely confined to subjects who had an adenomatous polyp. The risk increased strongly with the percentage of villous features in the polyp and was about twice as high in subjects with villous adenoma than in those with other adenomatous polyps. Conclusions: Our results support the sus-

pected relationship between colorectal polyps and cancer incidence and extend the association to colorectal cancer mortality. [J Natl Cancer Inst 84:962-966, 1992]

Vol. 84, No. 12, June 17, 1992

ord of a histologic confirmation was found for 44 (86%). Of the 1402 subjects not known to have died, 78 (6%) were identified as having had colorectal cancer diagnosed, and 72 (92%) had histologic confirmation. The primary outcome analyzed was death from colorectal cancer. The records were reviewed for the 96 subjects who had cancer of the colon or rectum listed as a cause of death; 83 subjects were judged to have colorectal cancer as the underlying cause. The expected number of deaths was derived from rates in Rhode Island men and women (Rhode Island Department of Health: unpublished data). Adjustments for age, sex, and calendar time were made by use of distributions of person-years in the total study group. The expected number of newly diagnosed cases of cancer was derived from age-, sex-, and timespecific rates in Connecticut (5). A specialized histology review was carried out by use of the case-control method within the study group (6). For each death or newly diagnosed case of colorectal cancer identified, three or four controls were selected, insofar as possible, from the remaining eligible subjects. Case and control subjects were matched according to sex and age within 5 years, hospital, and time of index surgery. Slides of the surgical specimen were retrieved, when possible, for each case and control subject. The slides were reviewed by a pathologist (R. Lev), who did not know which subjects had developed colorectal cancer. Polyps were evaluated with respect to four broad categories (adenomatous, hyperplastic, inflammatory, or other) and the extent to which the polyps showed villous features and cytologic dysplasia. Slides were reviewed for 169 (75%) of 225 case subjects identified and for 604 (74%) of 811 selected control subjects. The control series included 90 subjects selected for case subjects who were later found to be ineligible. The analyses were limited to 163 case subjects and 576 control subjects for whom the reviewing pathologist considered the slides to be of adequate quality and the polyps to have had a "diagnostic abnormality" that was not cancer. Associations between histologic features of benign colon polyps and colorectal cancer were evaluated by use of odds ratios (6,7). Logistic regression was used to control for sex and the size

and number of polyps. Preliminary analysis indicated that it was not necessary to adjust for the age, time, or hospital of index surgery or for the anatomic site of polyps.

Results The initial polyp tissue was removed by endoscopy, primarily sigmoidoscopy, for 2254 (78%) of the 2872 total subjects and by open surgery for 278 (10%). The procedure used was unknown for 340 (12%). Most of these procedures were probably endoscopies, however, because most of the missing information was attributable to subjects who had been outpatients. It was frequently impossible to distinguish a partial from a total polyp removal by use of the records of the study group. Of the 2872 subjects, 2316 subjects (81%) were recorded as having had polyps in the rectum or sigmoid only, 174 (6%) had polyps at sites above the sigmoid, 12 (

Relationship of polyps to cancer of the large intestine.

Pathologic and epidemiologic evidence indicates that patients with sporadic (nonfamilial) adenomatous polyps of the large intestine are at high risk o...
573KB Sizes 0 Downloads 0 Views