We welcome the opportunity to address points raised by Dr. Kaplan (1), Dr. Senior, and Dr. De Rosa and colleagues. We did not aim to propose a new test for HCV screening; high-performance antibody assays are already available for HCV. Rather, we tried to refine a tool for the identification and follow-up of persons with any type of liver disease. We did not use ALT levels to screen donors for anti-HCV; we used them to monitor anti-HCV–positive patients over time. We choose this population only as an example, since the results of tests for serum HCV RNA are an indirect but highly predictive marker of liver damage. Dr. Senior's calculations cannot be applied in this context.

Dr. Senior also states that a simple redefinition of an upper limit of normal for ALT would not be clinically beneficial. We agree and believe that ALT values should be used as a continuous variable, not in a dichotomous manner. We recommended against a universal "normal limit" for ALT and said that results should be interpreted flexibly, according to the clinical context. On the other hand, we disagree that, for example, patients with simple steatosis and an increase in ALT level should be considered to have false-positive results, with the presumptive aim of avoiding "mischief" in the population. Although simple steatosis may be associated with a better prognosis in patients with nonalcoholic fatty liver disease, it may progress to clinically significant fibrosis and may negatively affect the course of other forms of liver disease (2). Hepatic steatosis is an unhealthy condition that should be corrected by modifications of diet and lifestyle. Providing correct information to patients without creating anxiety is one of the tasks of the medical profession. We have nothing to add to Dr. Senior's final statement that ALT abnormalities may derive from extrahepatic injury, except that no laboratory test result can replace clinical judgment.

The observations of Dr. De Rosa and colleagues are in accordance with the spirit of our work. Given the close relationship between HCV infection and mixed cryoglobulinemia, however, we suggest that HCV viremia should be determined in any suspected case, independently of ALT levels.