Prospective evaluation of endoscopic criteria characteristic of sessile serrated adenomas/polyps
Background Differentiating sessile serrated adenoma/polyp (SSA/P) from hyperplastic polyp (HP) is clinically important in determining the necessity of endoscopic resection or recommending appropriate surveillance. There are few reports of characterization of SSA/P using narrowband imaging and chromo...
Gespeichert in:
Veröffentlicht in: | Journal of gastroenterology 2015-05, Vol.50 (5), p.555-563 |
---|---|
Hauptverfasser: | , , , , , , , , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Background
Differentiating sessile serrated adenoma/polyp (SSA/P) from hyperplastic polyp (HP) is clinically important in determining the necessity of endoscopic resection or recommending appropriate surveillance. There are few reports of characterization of SSA/P using narrowband imaging and chromoendoscopy with and without magnification. We aimed to establish imaging criteria to aid real-time diagnosis of SSA/P.
Methods
Patients with pale sessile or flat lesions of 6 mm or greater were prospectively enrolled in this multicenter trial. Nine endoscopic criteria, determined in real time, were investigated for possible association with SSA/P. Endoscopic mucosal resection was performed; specimens were retrieved and analyzed by histopathological examination.
Results
In 63 patients, 89 lesions were detected, including 41 HP, 38 SSA/P, five mixed polyps, and five other lesions. Right-side colon location, lesion size of 10 mm or greater, excessive mucus, the presence of a varicose microvascular vessel (VMV) noted with high-magnification narrowband imaging, and type III
H
pit pattern were each commonly associated with SSA/P compared with HP. Multivariate analysis substantiated three independent endoscopic criteria for SSA/P: the presence of VMV (
p
= 0.001), lesion size of 10 mm or greater (
p
= 0.0017), and right-side location (
p
= 0.0041), with odds ratios of 8.2, 7.2, and 6.1, respectively. The presence of VMV had a significantly higher specificity (87.8 %) than the other two independent endoscopic criteria (
p
= 0.0007 and
p
= 0.0008, respectively), but a lower sensitivity (57.9 %), whereas a combination of the three criteria (two or more positive) increased the sensitivity significantly (89.5 % and
p
= 0.0033) and had a higher degree of accuracy (82.3 %).
Conclusions
Three endoscopic criteria individually and in combination were effective in predicting a diagnosis of SSA/P without the need for chromoendoscopy. |
---|---|
ISSN: | 0944-1174 1435-5922 |
DOI: | 10.1007/s00535-014-0999-y |