Colonoscopy practices, and colorectal cancer and polyp screening, as assessed in the French district of Isère from May to July in 2004

Background

A pilot program of organized screening for colorectal cancers was conducted in Isère, an administrative district in France. A fecal occult blood test (Hemoccult II^®) was proposed for all individuals aged greater than 50 years (women since 1991 and men since 2002), followed by colonoscopy for those testing positive. A prospective study was carried out from May to July in 2004 and compared with a similar study conducted in 1996. The goal was to investigate colonoscopy practices, especially the role of screening.

Methods

Gastroenterologists practising in Isère (n = 39/42 practitioners) completed a questionnaire including their patients’ age and gender, indications, methods and results for all colonoscopies performed in those aged greater than 20 years. Any tissue samples taken were sent away for histological evaluation.

Results

The study involved 2558 colonoscopies (54% female, 73% patients aged greater or equal to 50 years), an increase of 35% from 1996 to 2004. Of the patients referred, 50.0% were symptomatic (pain; bowel problems: 28.7%; rectal bleeding: 21.3%), 23.5% had colonic disease and 22.5% came from screening (3.1% had positive stool tests, 17.8% had a family history). Recommendations related to family history (update of the 1998 consensus conference: screening indicated for patients with a first-degree relative diagnosed with cancer or advanced polyps aged less than 60 years) were well applied in terms of relatedness (81%) but, in 52% of cases, the age was greater than 60 years. Colonoscopy was carried out in almost all cases (0.1% failure), with complications in 0.4% of the examinations. Of the 2558 colonoscopies performed, 10% revealed advanced polyps or cancer: 30% were following a positive test compared with 8% for symptoms and 6% with a family history. Multivariate analyses showed that polyps greater or equal to 10 mm or malignant tumors are 1.5 times more common in men than in women, and six times more frequently seen in patients having colonoscopy following a positive test for blood in stools than in those with a family history of colorectal cancer. The number of pathologies found increased significantly in those aged greater than 50 years.

Conclusion

This cross-sectional survey of colonoscopy practices in Isère shows an increase in the number of colonoscopies performed between 1996 and 2004. This increase is not explained by expansion of the screening program, which was the reason for only 3% of colonoscopies. However, the best diagnostic yield for advanced polyps or cancers was obtained in screened patients (30%).