The impact of confocal endomicroscopy combined with standard diagnostics in evaluating large pancreatic cystic lesions: an international interobserver study (with video).

BACKGROUND AND AIMS: Current diagnostic approaches for pancreatic cystic lesions (PCLs) often result in unnecessary surgeries for benign cysts. We sought to prospectively evaluate the potential diagnostic value of endoscopic ultrasound-guided needle-based confocal laser endomicroscopy (EUS-nCLE) combined with standard-of-care (SOC) analysis in a histopathology-confirmed cohort among nCLE-naïve endosonographers.

METHODS

Subjects were enrolled from prospective studies (2020-2024) assessing EUS-nCLE for diagnosing PCLs. An international panel of 15 EUS-nCLE-naïve endosonographers participated in a 2-phase diagnostic study. In phase 1, observers reviewed clinical, imaging, and cyst fluid analysis including glucose to classify lesions and provide specific diagnoses. In phase 2, they reviewed corresponding EUS-nCLE videos and revised diagnoses when warranted. Diagnostic accuracy before and after nCLE review was compared using histopathology as the reference standard.

RESULTS

Sixty-six patients (mean cyst size = 4.3 ±1.9 cm) were included. Accuracy for distinguishing mucinous from nonmucinous PCLs increased from 90% to 96% with nCLE (P  <  .01). Subtype-specific accuracy improved for intraductal papillary mucinous neoplasm (88%-92%), mucinous cystic neoplasm (87%-91%), serous cystadenoma (88%-93%), cystic neuroendocrine tumor (91%-93%), and pseudocyst (92%-99%; all P  <  .05). Overall diagnostic accuracy improved from 73% to 85% (P  <  .01). nCLE significantly reduced misclassification, lowering mean unnecessary benign resections from 7.2 to 4.1 (P = .010) and missed neoplastic diagnoses from 4.4 to 0.7 (P = .001) per observer, enhancing clinical decision-making.

CONCLUSIONS

When combined with standard diagnostic evaluation, including cyst fluid glucose analysis, EUS-nCLE significantly enhances accuracy for classifying and identifying pancreatic cystic lesion subtypes. These results support incorporating nCLE into diagnostic algorithms, particularly for larger cysts where precise management decisions are crucial.