Objectives: Serum disialotransferrin is a specific marker of heavy alcohol consumption. We tested its accuracy and probability in detecting alcoholic cause of acute pancreatitis (AP).
Methods: Blood samples from 271 consecutive AP patients, admitted to the Helsinki University Central Hospital emergency unit, were analyzed.
Results: The median (range) disialotransferrin value was significantly higher (P = 0.001) in AP patients with alcoholic (n = 172) 1.6% (0.3%-14.4) than with biliary (n = 60) 0.7% (0.3%-1.3%) or other causes (n = 39) 0.8% (0.3%-4.1%). In receiver operating curve analysis, disialotransferrin, as a single analyte, was significantly (P = 0.001-0.0001) more accurate (area under the curve [AUC], 0.88; 95% confidence interval [CI], 0.84-0.92) in detecting alcoholic AP as compared with glutamyl transferase (AUC, 0.51; 95% CI, 0.45-0.57), aspartate aminotransferase (AUC, 0.57; 95% CI, 0.51-0.63), alanine aminotransferase (AUC, 0.63; 95% CI, 0.57-0.69), erythrocyte mean cell volume (AUC, 0.72; 95% CI, 0.67-0.78), amylase (AUC, 0.74; 95% CI, 0.67-0.78), C-reactive protein (AUC, 0.65; 95% CI, 0.59-0.71), and bilirubin (AUC, 0.55; 95% CI, 0.49-0.62). At a disialotransferrin cutoff of 1.2%, giving an 8% false-positive rate, the positive likelihood ratio was 8.47. Thus, a positive disialotransferrin test result, performed within 24 hours of admission, increased the probability of alcoholic AP from pretest 64% to posttest 94%.
Conclusions: Disialotransferrin, determined by capillary electrophoresis, is accurate, simple, and a rapid single biomarker of the alcoholic cause of AP.