Biggins SW et al. (2006) Evidence-based incorporation of serum sodium concentration into MELD. Gastroenterology 130: 1652–1660

The Model for End-stage Liver Disease (MELD) score—derived from serum total bilirubin and creatinine levels, plus the international normalized ratio—is commonly used in the US to assess patients' medical urgency for liver transplantation. Previous studies indicated that hyponatremia is also an important independent predictor of mortality in patients awaiting liver transplant, and Biggins et al. have confirmed these findings. These authors have developed a novel 'MELD-Na' score, which incorporates hyponatremia (i.e. serum sodium levels in the range 120–135 mmol/l) into the standard MELD. They recommend that data collected by the United Network for Organ Sharing in the US should be analyzed, to validate their modified MELD.

This prospective study was specifically designed to evaluate and optimize the MELD score in relation to predicting transplant waiting-list mortality. The patient cohort comprised 753 predominantly white patients awaiting liver transplant under the current US MELD-based organ-allocation system, for whom complete records were available. In this cohort, use of the optimized MELD-Na would have changed graft allocation in up to 27% of cases.

The authors note that the contribution of hyponatremia to risk of death is probably less important in patients with a very high MELD score, who are already at high risk of death. They call for studies involving large numbers of such patients, to evaluate the interaction between hyponatremia and a high MELD score, and to refine the MELD-Na model further.