Diagnostic Accuracy of Fine Needle Aspiration in Patients With Salivary Gland Tumors.

This retrospective study was conducted according to the STARD 2015 protocol. A thorough retrospective analysis of patient data that were extracted from patient records of a tertiary referral hospital database was performed. The patient files which were included in the study are electronically archived at the Departments of Oral and Maxillofacial Surgery of "Theageneio Anti-Cancer Hospital of Thessaloniki" and of General Hospital of Thessaloniki "George Papanikolaou". Patients who were included in the present study, except for clinical examination by an Oral and Maxillofacial Surgeon, also underwent at least one imaging modality (MRI, CT scan, ultrasound scan). This retrospective study was carried out from 2/2021 to 4/2022. Data from records of patients who underwent salivary tumour surgery at the Oral and Maxillofacial Surgery departments of Theageneio Anti-Cancer Hospital of Thessaloniki 1996-2022 and of the General Hospital of Thessaloniki G. Papanikolaou 2015-2022 were collected. The surgical treatment of all patients with salivary gland tumors who were examined by a practitioner of the two aforementioned Maxillofacial departments, was based on their chronological attendance at the outpatient clinic, while there was no attempt of grouping of them. Based on the FNA report, FNA results were categorized as follows: Benign, Malignant, Non-diagnostic, Unavailable result. The same categories were obtained for the reference standard test, which was the histopathological examination of the resected tumor. The sample size was calculated using the software G power analysis version 3.1.9.6. For α = 0.05 and study strength Power 0.95 a sample size of 102 patients had been obtained which was relatively smaller compared to the 128 patients that were finally included in the study. During the data collection from the medical patient records, there were 11 (n = 11) out of 139 (n = 139) cases whose FNA result and 4 (n=4) out of 184 (n=184) whose histological result was non-diagnostic as it was not clearly concluded whether the lesion was benign or malignant. These cases were excluded from the estimation of diagnostic accuracy indicators. Likewise to the non-diagnostic results, were excluded the 145 (n = 145) FNA results that were not available. To assess the diagnostic accuracy of FNA, a comparative process among FNA and histopathological outcomes was conducted. Sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, negative likelihood ratio, ROC curve and AUC (Area Under curve) were calculated. Calculations were made using software RStudio Version 1.4.1106 © 2009-2021 RStudio, PBC "Tiger Daylily" (2389bc24, 2021-02-11) for Windows Mozilla/5.0 (Windows NT 10.0; Win64; x64) Apple Web Kit/537.36 (KHTML, like Gecko) Qt Web Engine/5.12.8 Chrome/69.0.3497.128 Safari/537.36.