S1076 Anesthesia Care Provider Sedation (ACPS) Is Superior to Conscious Sedation (CS) for Successful Diagnostic Tissue Acquisition in Endoscopic Ultrasound Fine Needle Aspiration and/or Biopsy (EUS FNA/B)

Abstract

Introduction: EUS-FNA/B is frequently used for diagnosis of suspicious intestinal, subepithelial, and extraintestinal lesions. Our study aimed to evaluate the role of sedation in the diagnostic yield of EUS-FNA/B of suspicious lesions in the mediastinum, abdomen and pelvis. Given frequent use and experience of CS with diagnostic EUS at our institution, we hypothesized that the diagnostic yield of EUS-FNA/B would be equivalent for both sedation groups. Methods: A retrospective, single-center, cohort study was conducted including patients aged >18 years old who received EUS-FNA/B at our institution from 9/2018 – 5/2021 for further evaluation of suspicious lesions. Relevant baseline variables were collected for both groups. The primary endpoint was EUS diagnostic yield of neoplastic lesions. Successful cytologic diagnosis was defined by positive or suspicious results. Unsuccessful cytological results were defined by unsatisfactory, indeterminate, or false negative results. Secondary endpoint was adverse events (AEs). AEs were recorded for EUS only (EUS/ERCP patients were excluded from AE tally due to possible confounding). Baseline statistical comparison was done using the Chi-square test and Wilcoxon two sample tests. Univariate and multivariate logistic regression analysis was conducted to further evaluate primary and secondary outcomes. All statistical analyses were performed utilizing SAS Software (version 9.4; SAS Institute Inc. Cary, NC, USA). Results: 369 patients met initial inclusion criteria. 67 cases with confirmed or suspected benign lesions were eliminated from analysis. Successful FNA/B was present in 146/196 (74.5%) of patients in the ACPS group and 66/106 (62.3%) in the CS group, (p = 0.0274). Upon multivariate logistic regression controlling for lesion size, location, needle size, number of passes, and presence of ROSE, the difference in diagnostic yield between the CS and GA groups remained statistically significant (adjusted odds ratio [95% CI]: 0.558 [0.315 - 0.987], P = 0.0451) (Table). Adverse events (AEs) occurred more frequently in the ACPS group. This difference was sustained on multivariate analysis controlling for BMI, age, and ASA class. Conclusion: ACPS is associated with improved diagnostic yield when performing EUS-FNA/B for suspicious lesions in the mediastinum, abdomen, and pelvis. AEs occur more frequently in ACPS albeit most AEs are mild and associated with general anesthesia. Table 1. - Descriptive Analysis of demographics, Univariate/Multivariate logistic regression analysis, and Adverse Events Demographics Type of sedation P value Anesthesia Conscious Sedation N(number of patients) % N % Gender 0.07 Male 110 47.21 51 37.5 Female 123 52.79 85 62.5 ASA Class < 0.001 1 5 2.15 8 5.88 2 45 19.31 63 46.32 3 169 72.53 65 47.79 4 14 6.01 . . Location 0.001 Thoracic 1 0.43 1 0.74 Abdominal (non-pancreas) 26 11.16 22 16.18 Pelvic 3 1.29 2 1.47 pancreas ampulla/uncinate/head 130 55.79 42 30.88 neck/body/tail pancreas 50 21.46 46 33.82 SEL 22 9.44 22 16.18 Cytologic Result 0.007 Unsatisfactory specimen 14 6.01 22 16.18 Negative for malignancy 43 18.45 33 24.26 Atypical/Indeterminate 30 12.88 15 11.03 Suspicious for malignancy 10 4.29 3 2.21 Positive for malignancy 136 58.37 63 46.32 FNB < 0.001 N 140 60.09 115 84.56 Y 93 39.91 21 15.44 Univariate and Multivariate logistic regression analysis Univariate logistic regression (Crude effect) Unit OR 95% Confidence Limits p-Value Type of sedation (CS vs Anesthesia) 1 0.565 0.34 0.938 0.0274 Adjusted Effect Unit OR 95% Confidence Limits Column1 p-Value Type of sedation CS vs Anesthesia 1 0.558 0.315 0.987 0.0451 Location Others vs (pancreas ampulla/uncinate/head) 1 0.46 0.2 1.063 0.0691 Number of needle passed 1 1.473 1.191 1.822 0.0004 Categories of Adverse Events Type of Sedation CS Anesthesia Severity GA MAC Mild 3 14 2 Moderate 1 5 0 Severe 0 3 0 Fatal 0 0 0 Total 4 22 2 **Total of 28 adverse events.