Reply.

We sincerely appreciate the comments and questions raised by Sintusek and Dhawan regarding our article.1 Based on our findings, we recommended that an entire core of liver biopsy sample with more than 1 mg dry weight should be used for hepatic copper determination. However, the coefficient of variation (CV) of the hepatic copper measurement was not significantly different between subgroups classified by dry weight of the liver, although there was a trend. Sintusek and Dhawan questioned the rationality of our recommendation and present data showing that estimation of liver copper is reliable in children with Wilson's disease (WD) even when the liver biopsy sample size is less than 1 mg.

The copper concentration in the liver of a normal human newborn contains roughly 6‐8 times that of an adult, and this concentration falls to the adult value, approximately 30 µg/g of dry tissue, within 6 months and varies little throughout life.2 However, the amount of liver copper in WD patients after birth does not fall to the normal range, but gradually increases because of the WD gene mutation. Therefore, if the liver copper concentration in an untreated WD patient is not elevated, the most likely cause is laboratory error, including sampling error and/or measurement error.

It is well accepted that the distribution of copper in the liver is inhomogeneous and that the accuracy of liver copper measurement is improved with an adequately sized specimen.3 The tissue obtained with biopsy needle is only 1.0∼3.5 mg dry weight. In clinical practice, liver copper content is usually determined with a part of a needle biopsy specimen. This practice may occasionally result in a false result. Indeed, nondiagnostic hepatic copper levels in patients with confirmed WD have been reported by many investigators in the past two to three decades. Among 28 patients with genetically confirmed WD, as reported by Sintusek and Dhawan, liver copper concentration was below 250 µg/g dry weight in 8 patients. These data demonstrate that estimation of liver copper is not reliable using their method.

In our study, we observed a trend where larger liver biopsy specimens resulted in smaller CV of the copper measurement. Although the P value was not less than that required for statistical significance, we believe that the size of the liver biopsy specimen has important influence on the precise determination of the copper content, based on “Do not reject null hypothesis” rather than “Accept null hypothesis.” More important, the sensitivity of hepatic copper level for the diagnosis of WD at the conventional cut‐off value of 250 µg/g dry weight was as high as 94.4% with our method. Although not a head‐to‐head comparison, the sensitivity of our method was higher than that reported by Ferenci et al., which determined sensitivity using part of a needle biopsy specimen (83.3%).4