Prospective Comparison of Radial and Linear Endoscopic Ultrasound (EUS) for the Diagnosis of Chronic Pancreatitis (CP) Using Secretin Pancreatic Function Test (PFT) As Reference Standard

Abstract

Introduction: EUS has been used for diagnosis of early CP; however, its operating characteristics are poorly defined. Radial and linear EUS technology has never been compared in this setting. Direct PFT detect mild exocrine insufficiency (surrogate marker of fibrosis) and is considered a good reference standard for early CP. Aims: Compare the test performance of radial and linear EUS for the detection of exocrine insufficiency. Methods: Patients evaluated for suspected CP underwent combined radial/linear EUS and secretin endoscopic PFT (IRB-approved protocol). Duodenal aspirates were collected at 15, 30, and 45 min after IV secretin. A peak bicarbonate concentration < 80 mEq/L is abnormal. EUS images were videotaped and blindly scored by 3 reviewers. Each of 9 criteria were considered present based on consensus of 2/3 interpreters. Sensitivity and specificity were calculated for each EUS threshold. Receiver operating characteristics (ROC) analysis was performed to compare overall test accuracy. Results: 154 patients underwent a combined radial EUS and PFT; 119 also had linear EUS. 59 pts (38%) had exocrine insufficiency. No significant difference was found in the overall accuracy of radial and linear EUS (ROC AUC 0.75 vs. 0.74, p = 0.81). Diagnostic characteristics are shown for each EUS threshold [Table]. Conclusions: 1. Radial and linear EUS have similar overall accuracy in the diagnosis of CP. 2. EUS has excellent specificity (4, 5, and 6 criteria); however, the sensitivity is low across the range of thresholds. Clinical Implication: EUS is good for “ruling in” suspected CP (high specificity). The high false negative rate implies that some patients develop functional decline before sonographic changes. Tabled 1 Radial EUS Linear EUS Criteria Sensitivity (95%CI) Specificity (95%CI) Sensitivity (95%CI) Specificity (95%CI) ≥2 73 (62,84) 51 (41,61) 68(55,81) 38(27,49) ≥3 66 (54,78) 76 (66,84) 58 (44,71) 79 (68,88) ≥4 58 (45,70) 93 (88,98) 46 (40,67) 93 (86,99) ≥5 47 (39,65) 98 (94,100) 36 (23,49) 99 (92,100) ≥6 36 (23,48) 99 (96,100) 28 (16,42) 100 (NA) Open table in a new tab