Findings from a new analysis of 5,912 participants showed that the prevalence of advanced neoplasia at second surveillance increased depending on the risk combinations observed at earlier examinations. In particular, incorporating results from the prior colonoscopy helped refine risk stratification, demonstrating that certain baseline and first-surveillance findings were strongly associated with a higher likelihood of advanced neoplasia at the subsequent follow-up.
Among those with no adenoma at baseline, 4% had advanced neoplasia after at first surveillance, 7.2% after nonadvanced neoplasia, and 11.4% after advanced neoplasia. For participants with baseline nonadvanced neoplasia, the proportions were 5.4% after no adenoma, 9.1% after nonadvanced neoplasia, and 20.2% after advanced neoplasia. The highest proportions were seen in those with baseline advanced neoplasia: 9.1% after no adenoma, 12.8% after nonadvanced neoplasia, and 26.1% after advanced neoplasia at first surveillance, according to Pooja Pandita, MD, of the Department of Medicine, University of California San Diego, La Jolla, California, and colleagues.
"Although recent guidelines have generally moved towards less aggressive and longer surveillance intervals, our results support the importance of classifying groups that may benefit from closer follow-up," Dr. Pandita and fellow investigators wrote. By identifying patients with persistently high-risk findings, clinicians can tailor surveillance strategies to optimize outcomes—offering less frequent colonoscopies for those at low risk while advocating for close monitoring for those at higher risk."
They conducted a comprehensive systematic review of multiple databases across 11 observational studies. The analysis included studies that reported neoplasia yield at three time points: baseline colonoscopy, first post‑polypectomy surveillance, and second surveillance. Studies were conducted in the United States (6 studies), Asia (4 studies), and Europe (1 study), with colonoscopies performed between 1985 and 2014. Nine of the studies received high-quality ratings on standardized assessment tools.
The findings generally support the 2020 recommendations from the US Multisociety Task Force for serial surveillance but also highlight areas where refinements may be warranted. Notably, two scenarios emerged that deserve further research: a 9.1% prevalence of advanced neoplasia among patients with baseline nonadvanced followed by first-surveillance nonadvanced neoplasia, who are currently recommended for a 7- to 10-year follow-up; and a 12.8% prevalence among those with baseline advanced disease followed by first-surveillance nonadvanced neoplasia, who are recommended for a 5-year follow-up. The authors suggest heuristic thresholds as research frameworks rather than formal clinical guidance, proposing “aggressive” follow-up when prevalence exceeds approximately 15%, “intermediate” surveillance around 10%, and “relaxed” intervals near 5%. These benchmarks align with the roughly 5% advanced neoplasia detection rate seen in average-risk screening populations.
Substantial between study heterogeneity affected several risk strata, reflecting variations in surveillance intervals, risk definitions, and study populations across the nearly 30-year span. Surveillance protocols were not consistently standardized across studies.
Reporting of quality factors, such as bowel preparation quality and adenoma detection rates, was inconsistent. Patient-level variables, including aspirin use, body mass index, and smoking status, were limited. The definitions of advanced neoplasia categories also varied, reducing comparability. Outcomes related to serrated lesions could not be adequately assessed because of inconsistent reporting. Finally, most of the available data were derived from retrospective cohort studies, which further constrained interpretation.
Sensitivity analyses excluding influential studies modestly reduced heterogeneity without materially changing pooled estimates, supporting the robustness of findings.
For funding and author disclosures, visit the link below.
