S2184 Diagnostic Performance of Confocal Laser Endomicroscopy (CLE) Compared to Endoscopic Ultrasound Guided Fine Needle Biopsy (EUS-FNB) for Identifying Pancreatic Cystic Lesions (PCLs)

Introduction: Pancreatic cystic lesions (PCLs) are increasingly identified due to widespread imaging. These lesions range from benign serous cystadenomas (SCAs) to potentially malignant intraductal papillary mucinous neoplasms (IPMNs). Accurate cyst diagnosis is crucial for guiding management. Confocal laser endomicroscopy (CLE), performed alongside endoscopic ultrasound guided fine needle biopsy (EUS-FNB), offers real-time histological imaging and may improve diagnostic performance. CLE is not yet established as a standard diagnostic modality for PCLs, and formal guidelines for its use are currently lacking. Its efficacy compared to FNB requires further evaluation. Methods: A retrospective, single-institution study, approved by IRB review, was conducted on 57 patients with PCLs who underwent both CLE and FNB between 2022 and 2023. CLE interpretations by advanced endoscopists were compared to FNB pathology results, used as the reference standard. Sensitivity and specificity were calculated using 2 × 2 contingency tables. In our sample, the FNB-based diagnosis yielded 2 general cyst types, SCA and IPMN. Given the very low frequency of SCA in our sample (6 total patients), we chose to focus on IPMN in performing the diagnostic testing. Results: The mean age of patients at the time of the procedure was 72 years, with 37% woman and 63% male patients. Using FNB, 35 (61.4%) were diagnosed with IPMN, 6 (10.5%) were diagnosed with SCA, and for 16 (28%) patients, although tissue was acquired, they did not receive a diagnosis due to the samples being inadequate. CLE successfully provided a diagnosis for all 57 patients. Thus, the overall diagnostic yield of CLE was 100% compared to 72% in FNB. Fourteen cases (25%) of IPMNs identified by CLE had a corresponding non-diagnostic FNB results. In reference to FNB, CLE had a sensitivity of 88.57% and specificity of 100% in diagnosing IPMNs. There were 2 (3.5%) incidents of post-procedure pancreatitis. Conclusion: CLE demonstrates strong diagnostic performance for IPMNs. Its ability to provide real-time diagnoses, especially when FNB samples are inadequate, highlights its potential as a diagnostic adjunct to FNB, potentially reducing the need for repeat EUS-FNB procedures. However, the limited sample size necessitates larger studies to validate these findings and establish clinical guidelines for its use.