Clinically relevant differences in the Model for End-Stage Liver Disease and Model for End-Stage Liver Disease-sodium scores.

With interest, we read the article by Xiol et al. regarding differences in serum measurements between different laboratories and their influence on the Model for End-Stage Liver Disease (MELD) and the Model for End-Stage Liver Disease incorporating serum sodium (MELD-Na). They reported significant differences in serum bilirubin, creatinine, international normalized ratio, and serum sodium levels between different laboratories, which led to clinically relevant variations in MELD and MELD-Na scores. We have some remarks regarding the methodology of this study. First, samples were distributed to 3 different laboratories at room temperature. To minimize sample degradation, the use of frozen samples is preferable. Second, serum creatinine was measured with the Jaffe method. It is generally agreed that this method tends to overestimate serum creatinine levels and, more importantly, is highly sensitive to disturbances (eg, ascorbine acid, uric acid, glucose, and several antibiotics). Therefore, it is not the best test for the analysis of serum creatinine, particularly in this setting. Enzymatic measurements are more reliable and therefore should have been preferred. Third, in the study by Xiol et al., indirect potentiometry was used for the measurement of serum sodium. In our center, a similar test is used for this analysis, with a variation coefficient of 0.7%. In the Netherlands, a variation between centers and the national average of 1% is accepted. Surprisingly, in this study, a variation of 3% was found between center A and centers B and C. We wonder if there could be a methodological calibration error at center A that could explain this difference. On the basis of the interlaboratory differences found for serum sodium levels, Xiol et al. advise us to be cautious in the incorporation of serum sodium into the MELD score for liver allocation. This conclusion underlines the importance of monitoring and maintaining high-quality standards in the measurement of serum sodium when it is used for the evaluation of the MELD score because we have found a strong correlation in our own center between serum sodium and waiting list mortality.