Endoscopists greatly overestimate the true size of the colorectal polyps they resect, according to a large study that found an almost two-fold difference in their estimates versus the size of neoplasia as measured in the resected specimen.

The prospective study compared the mean polyp size estimate made by the endoscopist to the actual neoplasia size within a polyp as measured in the specimens of 2,587 polyps from 1,304 patients. According to the investigators, who presented their work at the 2022 Digestive Disease Week (abstract Sa1507), this is the first study to measure actual neoplasia size within polyps and compare it to size estimates made by endoscopists—which constitutes the gold standard for polyp size.

“The advantage of using endoscopist’s estimate for polyp size is that we can give patients an immediate answer as to when to come back for surveillance … Unfortunately, that comes with operator bias. Several studies have shown over- and underestimation compared to the size of the actual resected polyp,” said Rajiv Heda, MD, an internal medicine resident at Tulane University Health Sciences Center, in New Orleans.

Dr. Heda performed the analysis on data provided by Heiko Pohl, MD, and colleagues from Dartmouth College Geisel School of Medicine, in Lebanon, N.H.

Almost all polyps in the study were adenomatous and removed by cold snare. Endoscopists assessed size (maximum dimension) before resection using a reference tool such as snare size, and pathologists assessed size by measuring the maximum dimension of the neoplasia within the resected specimen.

The endoscopists estimated the median size of polyps to be 4 mm, but the “true size,” based on neoplasia within the polyp specimen, was 2 mm, yielding an absolute difference of 2 mm and a relative size increase of 1.9 over the final measurement. The endoscopists’ estimate, therefore, was nearly double the actual size of the neoplasia, Dr. Heda reported.

For polyps greater than or equal to 10 mm, the median estimated size was 12 mm, but the true size of neoplasia was 5 mm—translating into a 3.6-fold increase when the means were compared. Importantly, of the 45 polyps estimated to be at least 10 mm or larger, only 8 (18%) were determined by pathologists to contain neoplasia of that size, he said.

Since surveillance interval is determined according to size—with follow-up colonoscopy recommended within 3 to 5 years for polyps greater than or equal to 10 mm—some 82% of patients may be coming back for surveillance colonoscopy “sooner than is necessary,” he suggested.

Samir Gupta, MD, a professor of medicine at the University of California, San Diego Health, who co-authored the US Multi-Society Task Force guidelines on surveillance (Gastrointest Endosc 2020;91[3]:463-485.e5), commented that the current study joins others showing that “size estimation by endoscopists is imprecise.” But whether the findings could weaken the rationale for the surveillance guidelines is unclear, he said, noting that the studies that informed the guidelines were already based on “usual-care” polyp measurement that is something of “a mixed bag.” In other words, he said, “The imprecision is already baked in.”

—Clinical Oncology News staff

Drs. Heda and Gupta reported no relevant financial disclosures.