Determinants of severity in acute pancreatitis, a nation-wide prospective multicenter study

Abstract

Pancreatic enzyme replacement therapy (PERT) is indicated in case of clinically relevant exocrine pancreatic insufficiency (EPI). Clinical trials addressing PERT have used the coefficient of fat absorption (CFA) to define EPI but this test is cumbersome to perform. Our aim was to compare two easier-to-perform tests to detect clinically relevant EPI: Fecal Elastase-1 (FE-1) and 13C-Mixed Triglyceride Breath Test (TGBT).We prospectively included 54 patients with chronic pancreatitis (CP), 24.1% operated, 29.6% had EPI. EPI was defined as a CFA <93%. The sensitivity (Se), specificity (Sp), positive (PPV) and negative (NPV) predictive values for different cut-offs of FE-1 and TGBT were calculated.The area under the ROC curve for FE-1/TGBT was 0.861/0.876 for the global sample, 0.842/0.794 for non-operated patients and 0.917/1 for operated patients respectively. Se, Sp, PPV and NPV for a cut-off of FE-1 <200 μg/g were 93.8%, 63.2%, 51.7% and 96% respectively. The best cut-off point for FE-1 was 84 μg/g, which yielded an 87.5% Se, 81.6% Sp, 66.7% PPV and 93.9% NPV. The same parameters for a TGBT <29% were 81.3%, 84.2%, 68.4% and 91.4% respectively. The best cut-off point for TGBT was <23%, which yielded an 81.3% Se, 94.7% Sp, 86.7% PPV and 92.3% NPV. Diabetes was associated to decreased FE-1 levels, even in multivariate analysis.FE-1 and TGBT showed similar results for the diagnosis of EPI in CP. In non-operated CP patients, TGBT does not offer any advantage to FE-1 but in operated CP patients TGBT seems a more accurate test.