Back to Annual Meeting Program

Purpose: The use of screening colonoscopy in the elderly has become controversial with the USPSTF recommendation against routine screening in adults aged 76 to 85 years as well as lack of Medicare reimbursement. However, early detection of colorectal cancer in healthy elderly patients may prevent morbidity and mortality from late presentation. This study aims to determine cancer detection rates in the elderly undergoing outpatient index colonoscopy and cancer detection for repeat colonoscopy in the same population.
Methods: We identified 903 outpatient exams in patients 76 to 85 years over a 2-year period (1/09-12/10). Patient demographics, exam indication and gross findings were noted. Inpatient and outpatient charts were reviewed for the pathology reports and previous history of colonoscopy in the subjects.
Results: Indications for all exams were as follows: screening (19.7%), surveillance (34.0%), symptom clarification (25.6%), and multiple symptoms or indications (12.0%). Carcinoma detection rate was 2.3%, adenoma detection rate 23.8% and normal exam or insignificant polyps 70.3%. Among patients documented to be undergoing colonoscopy for the first time, the carcinoma detection rate was 9.4% (5/53), this was statistically significant when compared to the rest of the group who had all previously had a colonoscopy (P=0.01). Carcinoma detection rate was 5.4% if the previous colonoscopy was 10+ years prior. 63% of the carcinomas found in the index colonoscopy group and in patients undergoing repeat colonoscopy 10+ years later were stage 1 or 2 and 38% were in stage 3. There were no cases with metastasis. Subjects who had a previous colonoscopy within 3 years and 4-5 yrs ago had a 2.8% (6/213) and 0% carcinoma detection rate respectively. When colonoscopies are stratified by indication, anemia had the highest yield with 3.9% rate of carcinoma followed by GI blood loss at 3.3%. Indications predicting a low yield were single symptoms such as altered bowel habits, abdominal pain or weight loss (rate 0%). Screening and surveillance had rates of 1.7% and 2.3%, respectively.
Conclusions: The majority of outpatient exams in the elderly reveal insignificant findings, demanding too much from limited resources. However, detection of carcinoma according to timing of the most recent exam has a much higher yield than indication. The highest rates of carcinoma are among elderly patients undergoing outpatient colonoscopy for the first time or after more than 10 years and results in acceptable rates of curative colorectal cancer resection. Early detection should reduce the morbidity of late presentation and surgical emergencies often seen in this population. Further guidelines designed to efficiently utilize resources should focus on minimizing redundant exams unlikely to yield significant results while supporting screening exams with high yield.