The current guidelines recommend the same surveillance interval for ≥3 nonadvanced adenomas (NAAs), without discriminating between diminutive (1-5 mm) and small (6-9 mm) adenomas. Additionally, the same surveillance interval is recommended for patients with ≤2 diminutive NAAs and those with ≤2 small NAAs. However, it is questionable whether these recommendations are appropriate.

We searched all relevant studies published through September 2019 that examined the risk of metachronous advanced colorectal neoplasia (ACRN) according to the size (diminutive vs small) and the number of adenomas found during an index colonoscopy. Low-risk adenomas (LRAs) were subclassified into 2 categories (LRA-1, ≤2 diminutive NAAs; and LRA-2, ≤2 small NAAs), and high-risk adenomas (HRAs) were subclassified into 3 categories (HRA-1, ≥3 diminutive NAAs; HRA-2, ≥3 small NAAs; and HRA-3, advanced adenoma).

Eight studies involving 36,142 patients were evaluated. The LRA-2 group had a higher risk of metachronous ACRN than the LRA-1 group (risk ratio, 1.49; 95% confidence interval [CI], 1.23-1.81). Additionally, the HRA-2 and HRA-3 groups had a higher risk of metachronous ACRN than the HRA-1 group (hazard ratios [HRs], 1.51 [95% CI, 1.002-2.28] and 1.92 [95% CI, 1.11-3.33], respectively). However, there was no significant difference between the HRA-1 versus LRA-2 groups (HR, 1.23; 95% CI, .78-1.94).

Among the HRA and LRA groups, those with diminutive NAAs had a lower risk of metachronous ACRN than those with small NAAs. We believe that clinical guidelines should consider extending the surveillance intervals in patients with diminutive NAAs only.