Diagnostic accuracy of endoscopic ultrasound-guided fine-needle aspiration: A single-center analysis.

Endoscopic ultrasound-guided fine-needle aspiration biopsy (EUS-FNAB) has become an important modality for identification of intra-abdominal masses. This study analyzed the accuracy of EUS-FNAB in a single medical center and explored factors related to positive diagnosis. In total, 77 patients with EUS-FNAB were retrospectively reviewed from July 2016 to February 2020. "Atypical (tends to be neoplasm/malignancy)," "suspicious (first consider neoplasm/malignancy)," and "malignant" were defined as positive cytology. The final diagnoses were based on histopathologic examination. The positive rate of EUS-FNAB for the diagnosis of neoplasm and its associations with age, sex, target puncture mass size, liver function, tumor markers, albumin, hypertension, and diabetes were examined. Accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of EUS-FNAB cytologic diagnoses in all patients were 77.9% (60/77), 76.1% (54/71), 100%, 100%, and 26.1% (6/23), respectively. Accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of EUS-FNAB cytologic diagnoses in the pancreas were 80.0% (48/60), 79.3% (46/58), 100%, 100%, and 14.3% (2/14), respectively. The results of EUS-FNAB in pancreatic masses showed that the level of CA19-9 was higher in the true positive group than in the false-negative group (<0.05). There were no factors associated with the true positive cytologic diagnoses (>0.05). Our single-medical center study showed that EUS-FNAB is an accurate diagnostic procedure for the evaluation of intra-abdominal masses. Further follow-up is required to explore factors associated with the true positive cytology.