Format

Send to

Choose Destination
See comment in PubMed Commons below
Ann Intern Med. 2012 May 15;156(10):703-9. doi: 10.7326/0003-4819-156-10-201205150-00006.

Risk for colorectal cancer in persons with a family history of adenomatous polyps: a systematic review.

Author information

1
Indiana University, Regenstrief Institute, 1050 Wishard Boulevard, Indianapolis, IN 46202, USA.

Abstract

BACKGROUND:

The risk for colorectal cancer (CRC) is unclear for persons who have first-degree relatives with adenomatous polyps (adenomas).

PURPOSE:

To determine the validity of studies about this issue.

DATA SOURCES:

MEDLINE and Cochrane databases from 1966 through 2011.

STUDY SELECTION:

Sequential review of titles, abstracts, and text from retrieved articles.

DATA EXTRACTION:

Study objective, study design, and numbers in study groups.

DATA SYNTHESIS:

Ten studies were identified that have been used to answer the question, "Does having a first-degree relative with an adenoma increase the risk for CRC?" We determined that they instead answer the question, "Does having a first-degree relative with CRC increase the risk for an adenoma?" We identified 2 additional studies that provide more relevant information. One study showed that the risk for CRC in persons who have first-degree relatives with adenomas is greater than the risk in persons who do not have first-degree relatives with adenomas (2.31% vs. 0.53%; relative risk, 4.36 [95% CI, 1.60 to 10.21]). The other study showed that the risk for CRC or large adenomas (≥1 cm) in persons who have first-degree relatives with large adenomas is greater than the risk in persons whose first-degree relatives do not have adenomas or CRC (8.3% vs. 4.2%; adjusted odds ratio, 2.27 [CI, 1.01 to 5.09]).

LIMITATION:

Even the 2 relevant studies have design problems that affect validity and generalizability.

CONCLUSION:

Most studies that are cited for the risk for CRC when relatives have adenomas do not address the issue. The 2 studies that do address the issue suggest an increased risk but have important methodological limitations. Properly designed studies are needed to measure the risk and identify the factors that modify it.

[Indexed for MEDLINE]
PubMed Commons home

PubMed Commons

0 comments
How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for Silverchair Information Systems
    Loading ...
    Support Center