Background: Risks and benefits of surveillance colonoscopy among older adults who have had prior polypectomy or normal colonoscopy are uncertain. Colorectal cancer (CRC) incidence and mortality data among individuals who have survived to age 75 may help inform decision-making surrounding repeat colonoscopy. Our aim was to estimate cumulative CRC rates among older adults with prior polypectomy versus normal colonoscopy.

Methods: We conducted a cohort study among Veterans aged 75 with prior exposure to colonoscopy with >1 colorectal adenomas or normal findings 1/1/05-12/31/16 using data from the United States Veterans Health Administration (VHA). Findings at the last colonoscopy prior to age 75 were characterized. Veterans with incomplete extent of exam, inadequate bowel preparation, or prior diagnosis of inflammatory bowel disease or prior CRC were excluded. Primary and secondary outcomes were incident CRC, assessed with VHA cancer registry data, and fatal CRC, assessed with National Death Index data, respectively. Veterans were followed from their 75th birthday (index date) through incident CRC, death or 12/31/19 in incident analyses; in fatal analyses, follow-up proceeded through death or 12/31/19. Cumulative CRC incidence and mortality at 5 and 10 years were estimated using Kaplan-Meier curves, and relative CRC risk was estimated for the adenoma vs. normal colonoscopy groups using Cox proportional hazards models.

Results: Among 54,004 eligible 75 year olds, most recent colonoscopy preceding age 75 was normal for 40,048 (74.2%) and had an adenoma for 13,956 (25.8%). Median age at last colonoscopy prior to age 75 was age 72; the sample was 98.1% male, 73.3% non-Hispanic White, 10.1% non-Hispanic Black, and 5.5% Hispanic, with median 5.2 years follow-up. Cumulative CRC incidences at 5 and 10 years were 0.32% (95%CI: 0.21-0.42) and 0.77% (95%CI: 0.46-1.07) for the adenoma group, and 0.08% (95% CI: 0.05-0.11) and 0.35% (95%CI: 0.23-0.47) for the normal colonoscopy group (Figure 1). Cumulative CRC deaths at 5 and 10 years were 0.21% (95% CI: 0.12-0.30) and 0.67% (95%CI: 0.37-0.98) for the adenoma group, and 0.18% (95%CI: 0.13-0.22) and 0.55% (95%CI: 0.40-0.70) for the normal group (Figure 2). Compared to the normal colonoscopy group, the adenoma group had 2.8-fold increased risk for incident CRC (HR 2.86, 95% CI: 1.94-4.21), while results were nonsignificant for fatal CRC (HR 1.33, 95% CI: 0.91-1.95). Additional analyses accounting for frailty at age 75 and colonoscopy exposure after age 75 are ongoing.

Conclusions: Among older Veterans, incident and fatal CRC will occur in fewer than 8 in 1000 with adenoma and 6 in 1000 with normal findings at most recent colonoscopy. For those with prior colonoscopy, benefits of repeat colonoscopy after age 75 for screening or surveillance may be limited.

Figure 1. Cumulative CRC incidence after age 75 among Veterans with prior adenoma vs. normal colonoscopy

Figure 2. Cumulative CRC death after age 75 among Veterans with prior adenoma vs. normal colonoscopy

Benjamin Seligman

Appearances