HIGH YIELD OF COLONOSCOPIC EVALUATION IN SELECTED YOUNG PATIENTS
Audrey Kulaylat*1, Paolo Goffredo1, Michael Laffin1, Shonan Sho1, Adam Sang1, Christine C. Jensen2
1Surgery, University of Minnesota System, Minneapolis, MN; 2Minnesota Colon and Rectal Foundation, Saint Paul, MN
Background
The incidences of cancer and inflammatory bowel disease have been consistently increasing across all age groups in the United States for the past several decades. However, there has been a reluctance to provide insurance coverage for colonoscopic evaluation, particularly in patients under 50 years of age.
Methods
Using data from a large group practice, all patients undergoing colonoscopy from January 2012 to April 2020 were identified. Patients undergoing subsequent colonoscopies and those with known malignancy were excluded. Patients were stratified by age groups 18-29, 30-39, 40-49, or 50 and older. ICD-9 and ICD-10 codes were used to identify presenting symptoms such as gastrointestinal bleeding (ICD-9 569.3, 578.1 and ICD-10 K62.5, K92.1) and altered bowel habits (ICD-9 787.91,787.99 and ICD-10 R19.4, R19.7). The primary outcome of interest, pathologic finding upon colonoscopy (cancer, adenomatous polyps 1 cm or larger, and inflammatory bowel disease (IBD)) was abstracted from the institutional database and compared using standard univariate statistics.
Results
A total of 17,837 patients undergoing colonoscopy were identified, of which 3,333 (18.6%) were under the age of 50. Almost half (47.5%) of patients under 50 years presented with bleeding symptoms prior to colonoscopy, versus only 6.6% of patients 50 years and older (p<0.001). The proportion of patients under 50 years presenting for colonoscopy increased significantly over the time period (annual odds ratio [OR] 1.07, p<0.001). A similarly significant trend was identified among the subset of patients undergoing colonoscopy for evaluation of bleeding (OR 1.11, p<0.001) or alteration in bowel habits (OR 1.06, p<0.001).
Of all patients undergoing colonoscopy, the diagnostic yield for the composite finding of cancer, adenomas 1 cm or larger, or IBD are listed in Table 1. Notably, among patients in the youngest cohort undergoing colonoscopic evaluation, IBD was identified in 8.4% of colonoscopies. This rate increased to 13.0% when performed in patients with reported alteration in bowel habits. While the rate of cancer detection among age groups under 50 years was low (0.0% to 0.3%), the rate of detection of a cancer or large adenoma was as high as 3.2% among all patients regardless of the presence of symptoms. Overall, the number of patients needed to scope to identify a pathologic finding was 17.9.
Conclusions
Overall, the proportion of patients under 50 years of age undergoing endoscopic investigation for gastrointestinal symptomatology is increasing. Although the rate of cancer detection is still relatively low in patients under 50 years of age, a fair number of patients under 50 present with important treatable conditions such as large adenomas or IBD, suggesting that colonoscopy is a high-yield diagnostic strategy and should be strongly considered in this patient population.
Back to 2021 Abstracts