Current issues of ACP Journal Club are published in Annals of Internal Medicine


Smoking, alcohol, and risk for colorectal adenomas

ACP J Club. 1993 Sept-Oct;119:58. doi:10.7326/ACPJC-1993-119-2-058

Source Citation

Sandler RS, Lyles CM, McAuliffe C, Woosley JT, Kupper LL. Cigarette smoking, alcohol, and the risk of colorectal adenomas. Gastroenterology. 1993 May;104:1445-51.



To study the reported association among cigarette smoking, alcohol, and colorectal adenomas.


A case-control study.


University of North Carolina hospitals.


All persons who had a colonoscopy were considered for eligibility. Exclusion criteria were age < 30 years, polyposis, colitis, previous colon resection or cancer, unsatisfactory colon preparation, incomplete data, or previous adenoma. Cases were defined as patients who had ≥ 1 adenoma or cancer; controls were persons with no adenomas. 236 patients with adenomas and 409 control patients (mean age 60 y, 375 women) participated.

Assessment of risk factors

A nutritionist, blinded to adenoma status, telephoned each person to collect a history and current status of cigarette smoking and alcohol consumption. Dietary risk factors (total calories and 17 macronutrients and micronutrients) were taken from food questionnaires and analyzed by quartiles of use. Smoking was stratified into "ever smoked" and "never smoked," and quartiles of use. Alcohol consumption included "nondrinkers"; beer, wine, and liquor consumption; and quartiles of use.

Main outcome measures

Association of cigarette smoking and alcohol use with colorectal adenomas measured by odds ratios (ORs).

Main results

No measure of smoking status or history was associated with risk for adenomas, except for women aged 30 to 55 years. For these women, the OR for "ever smoked" for developing adenomas was 3.4 (95% CI 1.4 to 8.7), and for > 20 cigarettes per day, the OR was 4.8 (CI 1.2 to 19.4). Alcohol was a risk factor for adenomas in men in the highest quartiles of consumption: For daily calories from alcohol, the OR was 4.2 (CI 1.8 to 10.2); for percentage of calories from alcohol, the OR was 4.7 (CI 2.0 to 11.3); and for frequency of consumption, the OR was 3.9 (CI 1.6 to 9.4) (P for trend for each ≤ 0.002). After controlling for age and some of the putative dietary risk factors, these ORs increased to 5.5 (CI 1.9 to 16.0) for daily calories from alcohol, 6.9 (CI 2.3 to 21.2) for percentage of calories from alcohol, and 4.9 (CI 1.7 to 14.5) for frequency of use. For men who drank only beer, the OR for adenomas was 3.4 (CI 1.3 to 8.7). Wine and liquor consumption were not associated with adenomas.


In men, alcohol use (especially beer drinking) was associated with increased risk for colorectal adenomas. In women aged 30 to 55 years, smoking was associated with colorectal adenomas.

Source of funding: In part, National Institutes of Health.

For article reprint: Dr. R.S. Sandler, Division of Digestive Diseases and Nutrition, CB no. 7080, 423A Burnett-Womack Building, University of North Carolina, Chapel Hill, NC 27599-7080, USA. FAX 919-966-6842.


The search for the causes of colorectal carcinoma is advancing on 2 broad fronts. Investigations of the molecular basis of colorectal cancer indicate that at least 5 mutational events must accumulate for carcinoma to arise (1). Even the smallest adenoma has had some of these events and typically shows DNA hypomethylation, whereas larger adenomas generally have acquired additional mutations. Insights into this sequence of carcinogenesis should provide the means for more accurately identifying polyps with malignant potential.

On the second front, epidemiologists have sought environmental factors linked to colorectal cancer, presumably acting to shift the likelihood that a colonocyte will have 1 of these mutational events. In this study, Sandler and colleagues add to the evidence linking alcohol consumption and adenomatous polyps. The effect of alcohol consumption was low in patients consuming less than 12 g (approximately 1 glass of wine or 1 shot of whisky) per day and only doubled with a 4- to 10-fold increase in alcohol consumption thereafter. Nevertheless, a preponderance of studies support this relation, a plausible mechanism being that ethanol-induced disturbances in folate metabolism could impair methylation reactions throughout the body, including methylation of colonocyte DNA (2). Whether diet modification will benefit the patient who already has adenomas is unclear. Further, the evidence continues to support the substitution of fruits and vegetables for red meat as the most powerful general anticancer dietary strategy (3). Nevertheless, physicians now appear to have another reason to advocate restraint in the use of alcohol.

Joseph Alcorn, MD
Albuquerque Veterans Affairs Medical Center Albuquerque, New Mexico, USA