Colonoscopic Findings in Patients With Incidental Colonic Focal FDG Uptake.

OBJECTIVE

The purpose of this study was to investigate the nature of FDG-avid and non-FDG-avid lesions detected at colonoscopy in patients presenting with incidental focal colonic FDG uptake at PET/CT.

MATERIALS AND METHODS

Among 9073 patients who underwent PET/CT over a 4-year period, 82 patients without a history of colonic disease had focal colonic FDG uptake and underwent colonoscopy. In consensus, a radiologist and a nuclear physician read images from these PET/CT examinations. They recorded the location of focal FDG uptake in the colon and associated CT abnormalities and measured maximum standardized uptake value (SUVmax) and metabolic volume (MV). Readings were performed twice--first without and second with knowledge of lesion location at colonoscopy. The final diagnosis was based on colonoscopic findings and histopathologic results categorized into benign, premalignant, or malignant.

RESULTS

One hundred seven foci of colonic FDG uptake at PET/CT and 150 lesions at colonoscopy were detected. Among 107 foci of FDG uptake, 65 (61%) corresponded to a lesion at colonoscopy (true-positive findings), and 42 (39%) did not (false-positive findings). Among 150 lesions found at colonoscopy, 85 (57%) were not FDG avid (false-negative findings). The MV of true-positive findings was lower than that of false-positive findings (4.0 ± 0.4 cm(3) vs 6.2 ± 0.7 cm(3); p = 0.006), but SUVmax did not differ (7.4 ± 0.5 vs 7.7 ± 0.5; p = 0.649). Considering the histopathologic categories of the lesions and the false-positive findings, there was no difference in SUVmax (p = 0.103), but MV was lower in premalignant lesions than in false-positive findings (p = 0.005).

CONCLUSION

Focal colonic FDG uptake may indicate the presence of a benign, pre-malignant, or malignant lesion. Subsequent colonoscopy should not be restricted to the colonic site of FDG uptake.