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


NSAIDs reduced the risk for development of adenomatous polyps

ACP J Club. 1998 Sep-Oct;129:49. doi:10.7326/ACPJC-1998-129-2-049

Source Citation

Sandler RS, Galanko JC, Murray SC, Helm JF, Woosley JT. Aspirin and nonsteroidal anti-inflammatory agents and risk for colorectal adenomas. Gastroenterology. 1998 Mar; 114:441-7.



Are aspirin and nonsteroidal anti-inflammatory drugs (NSAIDs) protective against colonoscopy-proved colorectal adenomas?


Case-control study.


University hospitals in North Carolina, United States.


379 participants (mean age 59 y, 83% white, 54% women) had biopsy specimens taken from the large bowel during a clinically indicated colonoscopy. Exclusion criteria were age < 30 years, polyposis (> 100 polyps), colitis, previous colon resection, current or previous colon cancer, or unsatisfactory preparation for or incomplete examination. Case-patients (n = 142) had ≥ 1 adenomatous polyp according to standard criteria. Control-patients (n = 169) did not have adenomatous polyps, and previous patients (n = 68) did not currently have adenomatous polyps but had a confirmed history of polyps. Patients with nonadenomatous polyps could be control-patients.

Assessment of risk factors

Participants were interviewed by telephone to ascertain use of aspirin, aspirin-containing drugs, and NSAIDs in the past 5 years. Regular use was defined as use > 15 times/mo. Food questionnaires provided information on consumption of 100 foods and 17 micronutrients and macronutrients. Potential risk factors were age, sex, race, body mass index, coffee consumption, caloric intake, family history of colon cancer, fat intake, fiber intake, alcohol consumption, smoking status, daily servings of fruit and fruit juices, and daily servings of vegetables. Age, sex, race, and body mass index were controlled for during calculation of adjusted odds ratios (ORs).

Main outcome measure

Colonoscopy-proved adenomatous polyps.

Main results

72% of patients reported using aspirin, and 54% reported using ibuprofen. Current use of NSAIDs was protective for the development of adenomatous polyps (adjusted OR 0.56, 95% CI 0.34 to 0.92), but former use was not (OR 0.59, CI 0.21 to 1.67). A history of polyps was not associated with NSAID use (OR 0.63, CI 0.34 to 1.16). Non-NSAID analgesics were not associated with adenomatous polyps (OR 0.84, CI 0.37 to 1.91), nor was coffee consumption (OR 0.73, CI 0.44 to 1.21).


Nonsteroidal anti-inflammatory drugs reduced the risk for the development of colonoscopy-proved adenomatous polyps in adults.

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

For correspondence: Dr. R.S. Sandler, Division of Digestive Diseases and Nutrition, CB #7080, 423A Burnett-Womack Building, University of North Carolina, Chapel Hill, NC 27599-7080, USA. FAX 919-966-2478.


Several risk factors are associated with an individual patient's risk for colorectal adenomas; such factors include advanced age, male sex, African-American race, positive family history of colorectal cancer and adenomas, excessive smoking or alcohol use, decreased intake of fiber, and increased intake of fat. Previous cohort studies show that patients using aspirin or NSAIDs may be less likely to develop colorectal cancer and colorectal adenomas.

Sandler and colleagues' well-designed case-control study quantified the benefit associated with aspirin or NSAID use. All patients in both groups had complete colonoscopy, and almost all patients in both groups had similar indications for colonoscopy (rectal bleeding, anemia, and positive fecal occult blood test results). Several confounding factors, including age, sex, and race, were controlled for when calculating ORs for risk for adenoma development. However, the study did not attempt to fully evaluate a dose-response relation or a temporal relation between aspirin or NSAID use and colorectal adenomas. Also, because the patients in this study were symptomatic with anemia, rectal bleeding, or positive fecal occult blood test results, these data may not be generalizable to the larger population of asymptomatic patients with average risk. An ongoing, multicenter, screening colonoscopy study of average-risk patients may provide this evidence.

The mechanism by which aspirin or NSAIDs reduce adenoma formation has not been established. However, the real question is: Should aspirin or NSAIDs be prescribed to prevent colorectal adenomas and colorectal cancer? Clinicians must weigh the complications, such as gastrointestinal bleeding and hemorrhagic stroke associated with long-term aspirin and NSAID use, against the potential benefits. Ongoing research may identify subgroups of patients at increased risk for adenoma formation because of multiple risk factors or genetic abnormalities. If this subgroup of patients can be defined, then the potential benefits of aspirin or NSAIDs for reducing adenoma formation may outweigh the risks. However, we do not think, on the basis of available evidence, that the reduction in adenomas associated with aspirin or NSAID use outweighs the risks for asymptomatic patients with average risk.

J. David Horwhat, MD
Philip S. Schoenfeld, MD, MSEd
National Naval Medical CenterBethesda, Maryland, USA

J. David Horwhat, MD
National Naval Medical Center
Bethesda, Maryland, USA

Philip S. Schoenfeld, MD, MSEd
National Naval Medical Center
Bethesda, Maryland, USA