The primary aim of our study is to determine the correlation between the incidental PET-FDG uptake in the gastrointestinal tract (GIT) and the subsequent endoscopic findings. Secondary outcomes are to analyze the differences among the types of 18-FDG uptake patterns and to evaluate the role of the maximum standardized uptake value (SUVmax) in differentiating among the different kind of findings.

We conducted a single center retrospective study in a cohort of patients who underwent an endoscopic examination after a PET-scan showed abnormal uptake at some level in the GIT. All the patients included had a non GIT malignancy, or the FDG accumulation was in a different site than the pre-existing GIT-malignancy. All the endoscopic procedures were performed at Istituto Regina Elena in Rome, while some of the radiological exams were performed in other centers and then reviewed by our expert radiologists. The analysis of the differences of SUVmax values among the different groups was performed using the Kruskal-Wallis test with Dunn's post-hoc pairwise comparisons adjusted by the Bonferroni method, due to non-normal data distribution. Continuous variables were summarized as medians with interquartile ranges (IQR). Statistical significance was set at p<0.05.

We analyzed the data from 267 patients who underwent an endoscopic procedure between April 2018 and November 2024. 57.7% were females, with a median age of  66 years (range 32-89). 68.5% of patients underwent lower-GI endoscopy, 19.9% gastroscopy, and  11.6% underwent both examinations. 113 patients had a positive endoscopic finding (42.3%), in particular: 34 malignant lesions; 55 pre-malignant lesions (either colonic adenomas >10 mm in diameter, with low grade or high grade dysplasia, or gastric intestinal metaplasia) and 26 inflammatory conditions of various kind. In 237 patients we could analyze the pattern of uptake in the radiological exam being a focal pattern in 166, diffuse in 36 and focal on diffuse in 35 cases. The diffuse pattern was associated with the higher grade of false positives (33 out of 36, 91.7%), while a focal captation was more frequently correlated with a significant endoscopic finding (86/166, 51.8% in the focal pattern and 13/35, 37.1% in the focal on diffuse pattern). In 242 out of 267 patients (90.6%) the SUVmax value was reported on the radiological exam. SUVmax median value significantly differed among malignant (15.7), premalignant (10.7), inflammatory lesions (8.16) and false positives (8.05) (Kruskal–Wallis H=36.97, p<0.001). Post-hoc Dunn analysis showed that malignant lesions had significantly higher SUVmax compared with inflammatory lesions (p<0.001) and PET false positives (p<0.001), while the difference with premalignant lesions did not reach statistical significance (p=0.196).Premalignant lesions showed higher SUVmax values compared with PET false positives (p<0.001), but not with inflammatory lesions (p=0.109). No difference emerged between inflammatory lesions and PET false positives.

In our study 42.3% of the patients had a positive endoscopic finding (either neoplastic, premalignant or inflammatory) which corresponded to the segment of the abnormal FDG-PET uptake in the majority of the cases, with the focal uptake pattern being the most specific one in identifying clinically relevant conditions compared to the diffuse pattern. The median SUVmax value in the group with malignant lesions was higher if compared to the inflammatory lesions or the false-positive PET-scan group, and the difference was statistically significant suggesting a role of this parameter in the decision of performing the endoscopic exam after incidental PET captation in the GIT. There are no guidelines on the subject yet and the decision whether to proceed with further investigations should be made on a case-by-case basis, also taking in consideration the patient’s prognosis and comorbidities.