We read with great interest the recent article by Lee et al.1 suggesting increased risk of mortality of patients with elevated levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST). However, the prevalence of abnormalities of ALT and AST in the general population remains undetermined. A particular group of interest is the “worried-well” population. We would like to report the results of ALT and AST tests in a population of “normal” worried-well who presented themselves for testing using a commercially available “do-it-yourself” kit.2 Once the samples were received in the laboratory, AST and ALT values were measured using a standard colorimeter enzymatic method. The aim of this study was to determine the prevalence of abnormal ALT and AST values in this population that had no known underlying medical disorders. The study population consisted of 1039 subjects, comprising 561 female and 478 male subjects, all over 18 years of age (mean age of 45.3 ± 11.8 years old).
Applying a cut-off for the upper limit of normal (ULN), as used in the study by Lee et al.1 (ALT of 45 U/L for men and 29 U/L for women and AST of 31 U/L for both men and women), the percentage of abnormal ALT values in our study population was 12.5% (12.1% 1× ULN; 0.4% 3× ULN) for men and 9.8% (8.6% 1–2× ULN; 0.7% 2–3× ULN; 0.5% ≥3× ULN) for women. Percentage of abnormal AST values was 57% (50.3% 1–2× ULN; 5.3% 2–3× ULN; 1.4% ≥3× ULN) in men and 71.4% (60.9% 1–2× ULN; 8.6% 2–3× ULN; 1.9% ≥3× ULN) in women. However, it is likely that the ULN for ALT should be lower than those used by Lee et al., because they included people with nonalcoholic fatty liver disease in their standard normal population.1, 3 The study by Prati et al.4 in healthy Italian first-time blood donors, with no previous medical disorders and a body mass index less than 25 kg/m2, with normal levels of cholesterol and glucose, and with no known drug history, a better cut-off point for the limit of normality for ALT was 30 U/L for men and 19 U/L for women. These values are closer to the ones in the Korean study5 (30 U/L ALT; 31 U/L AST). Using this cut-off, the proportion of patients with an abnormal ALT test value would be 32.6% and 35.3% for men and women, respectively. Of those, 5.8% of the men and 3.5% of the women are above twice the ULN. Plots of ALT versus AST (Fig. 1A,B) reveal a poor correlation between these two enzymes, indicating possibly different pathophysiological processes; the group having a high ALT/AST ratio are more likely to have nonalcoholic fatty liver disease, whereas those with a high AST/ALT ratio are likely to have alcohol-related liver disease.6–8
The results of this study reveal an unexpectedly high prevalence of abnormalities in ALT and AST in the worried-well population in the United Kingdom. If the data presented by Lee et al. are representative of the population at large, more than 50% of the population studied are at higher risk of mortality compared with age-matched and sex-matched controls. Because the stated reason for undertaking the test was alcohol misuse in the vast majority of patients, this awareness of abnormal liver function test may allow targeted intervention to prevent progression to chronic liver disease and strongly supports Kim's justification for screening.8 However, before such programmes can be instituted widely, one needs to identify the ULN for different ages and populations. Because it is accepted that a proportion of patients with significant underlying liver disease can have normal ALT and AST, it will be necessary to define the most effective screening method.9, 10