Records of all patients with non-malignant ascites who underwent HVPG measurement from 2005 to 2009 were reviewed.
One hundred and thirty-eight 138 patients met inclusion criteria; 18.8 % had NSA. No difference in sodium excretion or diuretic use was noted in patients with and without NSA. NASH patients were more likely to have a NSA (34.2 % vs 12.4 % ; P = 0.001) as well as lower HVPG (15 vs 17.9 mmHg; P = 0.009) compared to other etiologies. MELD and HVPG predicted overall survival. However, mortality did not differ by disease etiology, though NASH patients had lower CTP (7.6 vs 8.5; P < 0.001) and MELD (15.6 vs 18.1; P = 0.09) scores, particularly among patients who died.
In patients with ascites and NSA, there were no increase in HVPG or urinary sodium retention. NASH patients with ascites had lower HVPG and a higher prevalence of NSA. They also had a higher mortality relative to MELD and CTP scores in other patients. In these patients, mechanisms other than portal and oncotic pressures and sodium retention play a role in ascites development, and increase mortality rate when complicated by low albumin.