The presence of hemorrhagic ascites is predictive of poor outcome in patients with cirrhosis, according to recent results.
In a retrospective case-control study, researchers evaluated the records of 1,113 patients with cirrhosis and ascites who received paracentesis at a single hospital between 2003 and 2010. Hemorrhagic ascites (HA) was identified in 214 cases, and outcomes within this group were compared with those of 642 matched controls with cirrhosis and ascites.
HA was most frequently spontaneous, with hepatocellular carcinoma (HCC) and iatrogenic causes occurring next most commonly. Spontaneous HA was significantly more common among those with ascitic fluid red blood cell counts between 10,000 mcL and 50,000 mcL (72%) than those with 50,000 mcL or more (52%; P=.003 for difference). Patients with HA related to HCC had significantly poorer survival rates than those with HA from other causes (P=.001).
HA participants were more likely to experience acute kidney injury (P<.001), sepsis (P<.01) and spontaneous bacterial peritonitis (P<.001), and also required transfusion (P<.001), vasopressors (P<.001) and ICU-level care more frequently (P=.001) than controls. Median albumin (2.5 g/dL vs. 2.8 g/dL; P<.01) and hemoglobin levels (10.2 g/dL vs. 11.2 g/dL; P<.001) were significantly lower compared with controls, while median MELD score (18 vs. 16; P<.001), bilirubin (3.0 mg/dL vs. 2.5 mg/dL; P=.04) and creatinine (1.1 mg/dL vs. 1 mg/dL; P=.004) were higher.
Patients with HA had higher mortality rates at 1 month (87% compared with 72%), 1 year (72% vs. 50%) and 3 years (61% vs. 41%), and a shorter median survival time than controls (1 year vs. 5.6 years). Multivariate analysis indicated an independent association between mortality and HA (HR=1.34; 95% CI, 1.07-1.68), after adjusting for other predictors including MELD score, ICU-level care, hepatitis C infection and HCC.
Don C. Rockey
“Patients with hemorrhagic ascites have poorer outcomes than those without hemorrhagic ascites,” researcher Don C. Rockey, MD, Department of Internal Medicine, Medical University of South Carolina, told Healio.com. Rockey and colleagues suggested that HA and other well-defined clinical features should be included in an improved prognostic score for this patient population.