Centro Hospitalar e Universitário de São João Porto, Porto, Portugal
Pedro Costa-Moreira, MD1, Filipe Vilas-Boas, MD1, Pedro Moutinho-Ribeiro, MD1, Diana Martins, MD1, Joanne Lopes, MD1, Guilherme Macedo, PhD2; 1Centro Hospitalar e Universitário de São João, Porto, Porto, Portugal; 2Centro Hospitalar de São João, Porto, Porto, Portugal
Introduction: Current management of pancreatic cystic lesions (PCLs) is based on clinical presentation, imaging, and cyst fluid analysis. Nevertheless, a PCL may remain a diagnostic challenge after completion of all standard investigations. Recently, through-the-needle biopsy (TTNB) sampling was added to the armamentarium for PCL evaluation during endoscopic ultrasound (EUS). Methods: We carried out a prospective cohort study to evaluate the feasibility, diagnostic yield and safety of EUS-TTNB. Technical success was defined as visible specimen present after TTNB sampling. Clinical success was determined by the pathology results whenever the cyst wall lining was present, and a histologic diagnosis could be established. Results: A total of 20 patients (mean age 62.0±11 years; 14 (61%) females) were included. The median cyst diameter was 29.5 mm (IQR 21.3 – 56.8mm, range 15-75mm). The procedure was technically successful in 19 patients (95%). TTNB allowed a histologic diagnosis in 15 out of 19 cases (clinical success: 78.9%). Fine-needle aspiration (FNA) cytology was performed in 19/20 patients but was nondiagnostic in 9 cases (45%). The combination of fluid cytology with TTNB resulted in an increased overall diagnostic yield (17/20 cases, 85%), providing additional information in 2 cases in which TTNB results were inconclusive. While cyst fluid carcinoembryonic antigen CEA >192 ng/mL was found in 5/20 cases, TTNB allowed the diagnosis of 12 mucinous lesions. Based on histological features and immunochemical mucin expression on TTNB samples, we were able to determine the predominant phenotype of IPMN in 8/10 cases. Surgical specimens were available in 9 cases with a diagnostic concordance rate of 66.7% (6/9 cases) when compared with TTNB specimens. Overall, in 7 cases, TTNB provided the diagnosis otherwise not suggested by cytology. Also, the histologic features of TTNB specimens changed the presumable diagnosis of the PCL in 5/20 cases and changed management in 8/20 patients (40%). Adverse events occurred in one patient who developed self-limited intracystic bleeding. Discussion: EUS-TTNB was associated with high technical and clinical success and low adverse events rate. It was found superior to FNA cytology and cyst fluid CEA. The addition of TTNB to the current standard approach may provide an incremental benefit for PCLs diagnosis and risk stratification. Also, the possibility for IPMN subtyping can become an important feature for patient management.
Disclosures: Pedro Costa-Moreira indicated no relevant financial relationships. Filipe Vilas-Boas indicated no relevant financial relationships. Pedro Moutinho-Ribeiro indicated no relevant financial relationships. Diana Martins indicated no relevant financial relationships. Joanne Lopes indicated no relevant financial relationships. Guilherme Macedo indicated no relevant financial relationships.