Does asymmetric dimethylarginine (ADMA) plasma concentration predict esophageal varices in patients with cirrhosis?

In previous studies elevated Asymmetric NG, NG - dimethylarginine (ADMA) plasma levels, an endogenous nitric oxide synthase inhibitor, correlated with the severity of hepatic venous pressure gradient measurement, both in peripheral and in hepatic veins. The aim of this study was to explore whether elevated ADMA plasma levels were able to predict the presence of esophageal varices (EV) and/or large EV in patients with cirrhosis.

74 cirrhotic patients who had undergone elective upper gastrointestinal endoscopy in order to assess the presence of portal hypertension and predictors of EV and/or large EV. ADMA levels were assayed by an ELISA test (Immundiagnostik AG, Germany).

53 patients had EV (26/53 had large EV). Univariate analysis of low hemoglobin (p = 0.045), PT-INR (p = 0.003), albumin (p = 0.024), bilirubin (p = 0.036), Child-Pugh score (p = 0.026), and ascites (p = 0.036) predicted the presence of EV. Multivariate analysis predicted EV for only PT-INR. The presence of large EV was predicted with univariate analysis of ADMA plasma levels (p = 0.013), low hemoglobin (p < 0.001), PT-INR (p = 0.001), albumin (p = 0.001), bilirubin (p = 0.026), Child-Pugh score (p < 0.001), ascites (p = 0.004). Sensitivity, specificity, predictive positive and negative values of ADMA plasma level > 0.5 micromol/L(-1) in predicting large EV were 0.69 (95% CI 0.53 - 0.82), 0.51 (95% CI 0.40 - 0.62), 0.43 (95% CI 0.31 - 0.56), 0.76 (95% CI 0.62 - 0.86), while the area under the ROC curve was 0.65 (95% CI 0.51 - 0.79).

ADMA plasma levels were increased in cirrhotics with more advanced liver failure but did not prove to be a useful clinical tool for predicting the presence of esophageal varices or large esophageal varices.