Should patients with Anemia and a Normal Ferritin Undergo Colonoscopy?

February 16, 2007


Commentary By: Joshua Olstein PGY-3

Second only to lung cancer, colon cancer claimed an estimated 55,000 lives in the United States in 2006. In an effort to reduce colon cancer morbidity and mortality, multiple screening tests have been developed to detect early disease among asymptomatic individuals. The 2003 American Gastroenterology Associations guidelines for screening asymptomatic individuals recommended colonoscopy as a preferred method of screening.

Due to a higher risk of colonic neoplasm, patients with unexplained iron-deficiency anemia are not included in these guidelines. However, consensus opinion recommends that these patients be considered for colonoscopy to exclude colonic neoplasm. Patients with anemia who are not iron-deficient represent a group for which risk of colonic neoplasm is not well known and no recommendations exist. These questions were recently addressed in an article authored by Sawhney et al. published in the American Journal of Gastroenterology.

The study was a retrospective cohort designed to compare the prevalence of advanced colonic neoplasia among anemic patients with different ferritin levels referred for colonoscopy at the Minneapolis VA. Patients were excluded if they had any other indications for colonoscopy including positive fecal occult blood testing, change in bowel habits, personal or family history of colon cancer, FAP, HNPCC, or IBD. For inclusion, a ferritin level must have been checked in the eight weeks preceding colonoscopy. Four-hundred and twenty four patients were included and were divided into three groups based on ferritin levels: 1) <50ng/dL (iron-deficient, N=254), 2) 51-100ng/dL (low normal, N=55), and 3) >100ng/dL (normal, N=115). The primary outcome to be assessed was the presence of advanced colonic neoplasm defined as 1) Invasive Cancer, 2) Malignant Polyp, or 3) Polyp with high-grade dysplasia. To estimate baseline risk, the prevalence of the primary endpoint was also established in three-hundred and twenty-three non-anemic patients referred for average-risk colon cancer screening during the same study period.

The prevalence of the primary outcome among the subjects with low normal ferritin (51-100ng/dL) was similar to those with iron-deficiency (7.2% vs. 7.9%, P=0.9) and significantly higher than non-anemic controls (7.2% vs. 1.2%, P=0.0001). The primary outcome was also higher among subjects with low normal ferritin compared to anemic patients with normal ferritin, but without statistical significance (7.2% vs. 1.7%, P=0.08). The prevalence of invasive cancer was highest among those with iron deficiency, but not statistically greater than the low normal ferritin group (6.3% vs. 2.3%). Of note, the control patients were significantly younger than the anemic patients, yet controlling for age did not alter the results.

The results of this study suggest that anemic patients without iron deficiency but with low normal ferritin levels are also at increased risk of advanced colonic neoplasia and should be referred for colonoscopic evaluation to exclude malignancy. Several limitations of the study include the retrospective design, possible referral bias, almost entirely male population, small number of patients in the low normal ferritin group, and having not known the ferritin levels of the control patients. Despite these limitations, the prevalence of early non-invasive neoplasia was particularly high among the anemic patients with low normal ferritin, and prompt colonoscopy could potentially prevent invasive colon cancer from developing in these patients.

References: Sawhney, Should Patients With Anemia and Low Normal or Normal Serum Ferritin Undergo Colonoscopy? The American Journal of Gastroenterology 102 (1), 82-88.

Image Courtesy Wikimedia Commons