Background: The impact of rapid on-site cytologic evaluation (ROSE) on endoscopic ultrasound-guided fine-needle biopsy (EUS-FNB) is widely debated. This study aims to assess the diagnostic performance of EUS-FNB in the absence of ROSE in abdominal masses.
Methods: Patients with abdominal masses undergoing EUS-FNB using 22-gauge Franseen needles and the slow-pull technique were prospectively enrolled in this study. Macroscopic on-site evaluation (MOSE) was performed without ROSE.
Results: 100 patients were recruited between 2018 and 2020. Seventy-eight patients had neoplasms, and twenty-two patients had benign diseases. Common diagnoses included pancreatic cancer (n = 27), mesenchymal tumors (n = 17), and metastatic tumors (n = 14). The mean mass size was 3.9 ± 2.6 cm. The median pass number was three. Eighty-nine percent had adequate specimens for histologic evaluation. Malignancy increased the odds of obtaining adequate tissue (OR 5.53, 95% CI, 1.36-22.5). For pancreatic cancer, FNB had a sensitivity of 92.3%, a specificity of 100%, a positive predictive value (PPV) of 100%, a negative predictive value (NPV) of 97%, and an AUROC of 0.96. The sensitivity, specificity, PPV, NPV, and AUROC for mesenchymal cell tumors were 100%, 95.9%, 84.2%, 100%, and 0.98, respectively. For metastatic tumors, FNB was 100% sensitive and specific, with an AUROC of 1.00. There were no procedure-related complications.
Conclusions: 22-gauge Franseen needles with the slow-pull technique and MOSE without ROSE provide excellent diagnostic performances for malignant lesions. Thus, MOSE should be implemented in real-world practice, and ROSE can be obviated when EUS-FNB is employed.
Keywords: endoscopic ultrasound; fine-needle aspiration; fine-needle biopsy; histology; pancreatic cancer; rapid on-site evaluation.