Background: This study analyzes the prognostic significance of serum phosphorus in patients with acute liver failure (ALF).
Methods: We performed a retrospective analysis of 112 patients with ALF. Univariate and bivariate analyses based on Kaplan-Meier recovery curves and a multivariate Classification Tree Structure Survival Analysis were performed to identify independent predictors of outcome. The variables analyzed were age, gender, race, ABO blood group, etiology of liver disease, grade of encephalopathy, serum bilirubin, prothrombin time, creatinine, serum phosphorus, phosphorus administered, phosphorus binders, and hemodialysis.
Results: The median follow-up time was 5 days, the median age was 28 years, and 62% of the patients were female. The patients' outcomes were as follows: 28% recovered, 52% required orthotopic liver transplantation, and 20% died. White patients showed the best prognosis (58% recovered in the first week), and Hispanics showed the worst prognosis (0.3% recovered at 1 week) (P=0.0001). Encephalopathy and bilirubin were significant predictors of recovery (P<0.0001 and P=0.004). The analysis of the serum phosphorus showed a statistically significant better prognosis in patients with low phosphorus (P<0.001). The recovery rate at 1 week was 74% in patients with serum phosphorus less than 2.5 mg/dL, 45% if phosphorus ranged between 2.5 to 5 mg/dL, and 0% if phosphorus was more than 5 mg/dL. The bivariate analysis on the effects of phosphorus administration showed that phosphorus replacement was associated with a significant improvement in recovery in patients with low (P<0.004) or normal serum phosphorus levels (P<0.017)
Conclusions: Hypophosphatemia and early phosphorus administration are associated with a good prognosis in ALF, whereas hyperphosphatemia is predictive of poor recovery.