Should a Colonoscopy be Recommended for Healthy Individuals with Increased Carcinoembryonic Antigen Levels? A Case–Control Study
- 165 Downloads
Background and Aim
Despite the limitations of screening or early diagnosis of colorectal cancers (CRC), carcinoembryonic antigen (CEA) is frequently measured in practice and during health promotion programs. The aim of this study was to evaluate the role of colonoscopy in healthy individuals with elevated CEA levels.
From January 2003 to November 2008, 117,731 healthy persons underwent an opportunistic screening program in two health promotion centers; 1,497 subjects (1.3%) showed an elevated CEA level (>5 ng/ml). Among them, 174 patients were recruited to undergo a colonoscopy to determine if colorectal malignancies were present. A total of 372 age- and sex-matched persons were selected as controls from among the healthy subjects who had a normal level of CEA and had received surveillance colonoscopy. The primary outcome was the incidences of CRC in elevated CEA and normal CEA groups. The secondary outcome was the predictive factors of CRC in the elevated CEA group.
The incidence of CRC was higher in the group with higher CEA-levels than in the group with normal CEA levels (4.6 vs. 1.3%; P = 0.031). In the CEA-elevated group, patients with CRCs were diagnosed at more advanced stages than were those in the CEA-normal group. The incidence of colorectal polyps was not different between the two groups. In the CEA-elevated group, anemia was an independent predictive factor of CRCs by multivariate analysis (P = 0.002).
Anemia itself is not a predictive factor of CRC in the entire population, but is an independent predictive factor of CRC in healthy individuals with an elevated level of CEA. Therefore, colonoscopy should be recommended for healthy subjects with an elevated level of CEA accompanied with anemia in the absence of other adenocarcinomas to evaluate the presence of colorectal malignancy.
KeywordsCarcinoembryonic antigen Colonoscopy Colorectal cancer
The authors thank Soung Min Jeon, Chang Mo Moon and Jae Jun Park for their assistence throughout this study.
Conflict of interests
The authors have no commercial associations that might represent a conflict of interest in relation to this manuscript.
This study was approved by the Human Research Review Committees of Severance Hospital, Yonsei University and Sanggye Paik Hospital, Inje University, Korea.