Article first published online: 25 MAR 2004
Copyright © 2004 American Association for the study of Liver Diseases
Volume 39, Issue 4, page 1173, April 2004
How to Cite
Sanyal, A. J. (2004), Reply. Hepatology, 39: 1173. doi: 10.1002/hep.20162
- Issue published online: 25 MAR 2004
- Article first published online: 25 MAR 2004
We have read the comments by Dr. Ratziu and his colleagues regarding our study We agree that alanine aminotransferase (ALT) values can fluctuate, and we would like to point out that multiple sets of normal ALT values spread over a period of 6 to 24 months were available for all patients except the 16 donors for living donor transplant.
A key issue is the definition of a normal ALT. The principal factors that must be considered are the methods used and the need to distinguish between statistical versus physiological normality. As discussed in our article,1. methodological differences can account for variances up to 20 IU/L when the same sample is studied using several commercially available methods. This may account for some of the variability alluded to by Dr. Ratziu. A more difficult problem is one of separating statistical from physiological “normality.” The normal range for ALT, as well as most clinically used laboratory parameters, was originally established on the basis of the distribution of values in cohorts of self-reported normal individuals. These data were used clinically to define those patients who were physiologically abnormal. This, as we know, is a fallacious concept. Just as a normal creatinine does not imply a physiologically normal kidney, a normal ALT does not imply a physiologically or morphologically normal liver. We agree that the ALT values in 70% of subjects in our study fit the high-normal values for what is currently classified as a normal ALT value by the American College of Pathology Standards. We concur that they had a high probability of being physiologically abnormal. However, we would like to point out that even those who had a low-normal ALT exhibited the full spectrum of fibrosis and other features of nonalcoholic fatty liver disease as well as metabolic syndrome. We would also like to reiterate that, as suggested by Dr. Ratziu, the data were presented as internationl units per liter for the reasons outlined by Dr. Ratziu. Of the 14 subjects with low-normal ALT values (<30 IU/L) 5 had advanced fibrosis. We agree with Dr. Ratziu that our study may have been weighted toward those patients with more advanced liver fibrosis due to the retrospective nature of the study and the ascertainment bias involved in looking only at those who had a biopsy. Indeed, this was noted in the discussion section of our article and a case was made for a prospective assessment of liver histology in those with persistently normal ALT values. Finally, the issue of redefining normal ALT ranges is a complex one, and this letter is not the appropriate forum to engage in that discussion.
We thank Dr. Ratziu and colleagues for their thoughtful and excellent comments and hope we have satisfied their concerns.
Arun J Sanyal M.D.*, * Department of Internal Medicine, Virginia Commonwealth University, Richmond, VA.