Patients with proximal, nondysplastic serrated polyps that were found on baseline screening colonoscopy were nearly twice as likely to have synchronous advanced neoplasia as were patients without those lesions, wrote Dr. Mitchal A. Schreiner and his colleagues in the November issue of Gastroenterology.
The risk of finding an adenoma during surveillance also increases in such patients, added Dr. Schreiner of the Portland (Ore.) VA Medical Center (Gastroenterology 2010 November [doi:10.1053/j.gastro.2010.06.074]).
The finding means that when nondysplastic serrated polyps (ND-SPs) are noted, endoscopists must be "especially vigilant and perform high-quality baseline screening exams to ensure detection of patients with advanced neoplasia."
In what they called the "first large study to examine the outcomes of surveillance colonoscopy in patients with proximal ND-SPs at baseline screening colonoscopy," the researchers looked at 3,121 asymptomatic patients (96% male) who underwent a screening colonoscopy at 1 of 13 Veterans Affairs medical centers between February 1994 and January 1997. All patients were between 50 and 75 years old.
Serrated polyps were defined as any polyp with a "sawtooth appearance in the colonic crypts." Proximal lesions were defined as those located proximal to the descending colon. The researchers also assessed "large" ND-SPs, meaning those greater than 10 mm.
Overall, 801 patients (25.7% of the total cohort) had one or more ND-SPs, including 248 patients with one or more proximal ND-SPs (7.9%). A total of 44 patients (1.4% of the total cohort) had at least one large ND-SP, of which 57% were in the proximal colon.
Among the proximal ND-SP patients, 17.3% were found to have synchronous advanced neoplasia (defined as invasive cancer, adenomas with high-grade dysplasia or villous histology, or tubular adenoma greater than 10 mm). The prevalence among patients with no proximal ND-SPs was 10%. After adjustment for age, that translated into a significant 1.90 odds ratio among the proximal ND-SP patients.
Proximal ND-SP patients also showed an increased likelihood of having three or more synchronous small tubular adenomas (less than 10 mm) on baseline colonoscopy, "which would have implications for surveillance," wrote the authors. Overall, 10.7% of patients with the proximal ND-SPs had them, compared with 5.3% of patients with no proximal ND-SPs (OR 2.19).
With regard to large ND-SPs, the 44 patients with these lesions had a particularly striking prevalence of synchronous advanced neoplasia, at 27.3%, compared with 10.3% of those without these lesions, for an age-adjusted OR of 3.37.
Dr. Schreiner and his colleagues also found that having ND-SPs at screening predicted neoplasia on follow-up colonoscopies up to 5.5 years later. "The presence of a proximal ND-SP on index colonoscopy is associated with an increased risk of any adenoma during surveillance (OR 3.14; 95% CI 1.59-6.20), but not advanced neoplasia (OR 2.09; 95% CI 0.44-9.87)," they wrote.
However, "the number of subjects in our cohort with a large ND-SP on baseline colonoscopy who had surveillance colonoscopy within 5.5 years (n = 31) was insufficient to perform risk analyses."
The authors pointed out several limitations of the study. For one, "the pathologic criteria for the diagnosis of serrated polyps are subject to interobserver variability," although this study did classify polyps on the basis of input from three separate reviewers.
Additionally, "since the criteria for SSA [sessile serrated adenomas] were developed after our study, we could not perform an analysis which distinguished HP [hyperplastic polyps] and SSA, which are both nondysplastic."
Dr. Schreiner and his colleagues wrote that they had no conflicts of interest related to this study, which was supported by the Department of Veterans Affairs.