In vivo real‐time diagnosis of endoscopic ultrasound‐guided needle‐based confocal laser endomicroscopy in gastric subepithelial lesions

2019 
BACKGROUND AND AIM: The effect of real-time analysis of needle-based confocal laser endomicroscopy (nCLE) for gastric subepithelial lesions (SELs) on the diagnostic value is unclear. The study aimed to investigate the diagnostic efficacy of real-time nCLE for gastric SELs and to assess the technical aspects and safety of real-time nCLE. METHODS: Consecutive patients with gastric SELs >/= 1 cm were prospectively investigated by endoscopic ultrasound (EUS), followed by nCLE. During EUS-nCLE, real-time nCLE diagnosis was made by an expert endoscopist. The procedure-relative adverse events were assessed and recorded. One-month washout period later, nCLE videos were reviewed off-line by the same endoscopist. The nCLE diagnoses were compared with corresponding pathological results. Additionally, image quality and interobserver agreements for the criteria were evaluated by three experienced endomicroscopists. RESULTS: Except for one failing to be punctured, 60 patients completed EUS-nCLE procedures successfully. Real-time nCLE had high diagnostic accuracies of >/= 88.3% for gastric SELs. There were no significant differences between real-time and off-line nCLE diagnoses for gastric SELs (P > 0.05). The overall accuracy of real-time nCLE for diagnosis of gastric SELs was 86.7%. There were no procedure-relative adverse events occurred. In addition, the mean image quality score was 3.6 (1 = poor and 5 = excellent). The interobserver agreement was "almost perfect" for ectopic pancreas and "substantial" for gastrointestinal stromal tumor, leiomyoma, and carcinoma. CONCLUSIONS: Endoscopic ultrasound-nCLE could provide in vivo real-time diagnostic imaging with a high diagnostic accuracy. Meanwhile, real-time nCLE was feasible and had a satisfactory safety profile.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    21
    References
    1
    Citations
    NaN
    KQI
    []