Identifying Individuals at Risk for Chronic Liver Disease
Identifying Individuals at Risk for Chronic Liver Disease
Background/Aims Serum alanine aminotransferase (ALT) is an easily available, low-cost screening tool for detecting silent chronic liver disease. Recent studies have suggested that the currently accepted healthy ALT thresholds be lowered.
Methods In this retrospective cross-sectional study, we determined upper thresholds for ALT values in a nationally representative healthy cohort (n = 3337) from the Fourth Korea National Health and Nutrition Examination Survey (KNHANES IV). Sensitivity and specificity of the currently used ALT threshold (40 IU/L, regardless of gender) was compared against study-derived, gender-specific ALT thresholds for detecting individuals at risk for chronic liver disease in 27 913 health check-up participants.
Results The 95th percentile levels for ALT in healthy weight, metabolically normal, liver disease-free KNHANES participants were 34 IU/L for men and 25 IU/L for women. The prevalence of ALT elevation among health check-up participants was 11.0% in currently used thresholds, and increased to 22.6% with study-derived, gender-specific thresholds. Of the population who were additionally defined to have elevated ALT levels under new ALT threshold, 65.7% were at risk for chronic liver disease. Sensitivity for detecting individuals at risk for chronic liver disease improved from 18 to 33% with new thresholds whereas a trade-off in specificity (from 96 to 88%) was observed.
Conclusions It is recommendable to lower the current ALT thresholds to better identify individuals at risk for chronic liver disease.
Most health check-up programmes include serum liver chemistry tests in order to screen for liver diseases in asymptomatic individuals. As sensitive indicators of hepatocellular damage, serum aminotransferase levels are usually low in healthy individuals and high in patients with hepatocellular diseases. Thus, serum aminotransferase levels are widely used to recognize the presence of a variety of liver disorders.
Compared to aspartate aminotransferase (AST) which exists in the liver as well as a wide variety of extrahepatic sites (including myocardium, skeletal muscle, kidney, brain, pancreas and blood cells), alanine aminotransferase (ALT) is localized primarily in the liver, thus, it is more commonly used to detect hepatocyte injury. The upper limit of normal (ULN) for serum ALT levels has been set at 40 IU/L in many hospitals including our hospital, although this value varies slightly between laboratories. However, several recent studies have shown that the healthy ALT thresholds should be lower than the currently accepted thresholds, and that gender-specific thresholds should be applied. For example, Prati et al. suggested that ULN of ALT is 30 U/L and 19 U/L for European men and women respectively. Lee et al. analysed serum ALT concentrations in liver donors with normal liver histology and suggested healthy ALT thresholds to be 33 IU/L for Asian men and 25 IU/L for Asian women. Similarly, Kang et al. studied ULN of ALT among 7403 Asian health check-up participants, and their ULN of ALT value was 31 U/L for men, and 23 U/L for women. Furthermore, a large population-based study showed that people with slightly increased aminotransferase activity, but still within normal range (35–40 U/L) are at an increased risk of mortality from liver disease. Thus, they claimed that it is necessary to lower the currently accepted ULN of ALT levels.
Currently, the ULN of ALT has been set at a mean value ±2 standard deviation in a group of healthy individuals, so some per cent of these healthy individual have ALT level which fall outside the defined normal range. Also, a minor elevation of ALT may be of no clinical importance, and may even be physiological. Thus, it is possible that lowering the ULN of ALT may assist in identifying patients with liver disease who would have been missed in conventionally used thresholds, meanwhile, it may increase the population with abnormal test results, leading to unnecessary additional tests. Thus, it is a dilemma to practicing physicians, to determine a proper ALT threshold which meets both high quality practice and cost-effectiveness of the healthcare system.
On this point, the present study was designed to investigate the necessity and consequence of lowering currently used ULN of ALT. For this purpose, we aimed: (i) to derive new ALT thresholds in a representative Korean population, (ii) to evaluate the prevalence and aetiology of elevated ALT levels in health check-up participants, and (iii) to verify the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of the new ALT thresholds, compared to those of the currently used ALT thresholds for classifying individuals as having or not having a risk for four common forms of chronic liver disease [hepatitis B virus (HBV) infection, hepatitis C virus (HCV) infection, chronic alcohol consumption and nonalcoholic fatty liver disease (NAFLD)].
Abstract and Introduction
Abstract
Background/Aims Serum alanine aminotransferase (ALT) is an easily available, low-cost screening tool for detecting silent chronic liver disease. Recent studies have suggested that the currently accepted healthy ALT thresholds be lowered.
Methods In this retrospective cross-sectional study, we determined upper thresholds for ALT values in a nationally representative healthy cohort (n = 3337) from the Fourth Korea National Health and Nutrition Examination Survey (KNHANES IV). Sensitivity and specificity of the currently used ALT threshold (40 IU/L, regardless of gender) was compared against study-derived, gender-specific ALT thresholds for detecting individuals at risk for chronic liver disease in 27 913 health check-up participants.
Results The 95th percentile levels for ALT in healthy weight, metabolically normal, liver disease-free KNHANES participants were 34 IU/L for men and 25 IU/L for women. The prevalence of ALT elevation among health check-up participants was 11.0% in currently used thresholds, and increased to 22.6% with study-derived, gender-specific thresholds. Of the population who were additionally defined to have elevated ALT levels under new ALT threshold, 65.7% were at risk for chronic liver disease. Sensitivity for detecting individuals at risk for chronic liver disease improved from 18 to 33% with new thresholds whereas a trade-off in specificity (from 96 to 88%) was observed.
Conclusions It is recommendable to lower the current ALT thresholds to better identify individuals at risk for chronic liver disease.
Introduction
Most health check-up programmes include serum liver chemistry tests in order to screen for liver diseases in asymptomatic individuals. As sensitive indicators of hepatocellular damage, serum aminotransferase levels are usually low in healthy individuals and high in patients with hepatocellular diseases. Thus, serum aminotransferase levels are widely used to recognize the presence of a variety of liver disorders.
Compared to aspartate aminotransferase (AST) which exists in the liver as well as a wide variety of extrahepatic sites (including myocardium, skeletal muscle, kidney, brain, pancreas and blood cells), alanine aminotransferase (ALT) is localized primarily in the liver, thus, it is more commonly used to detect hepatocyte injury. The upper limit of normal (ULN) for serum ALT levels has been set at 40 IU/L in many hospitals including our hospital, although this value varies slightly between laboratories. However, several recent studies have shown that the healthy ALT thresholds should be lower than the currently accepted thresholds, and that gender-specific thresholds should be applied. For example, Prati et al. suggested that ULN of ALT is 30 U/L and 19 U/L for European men and women respectively. Lee et al. analysed serum ALT concentrations in liver donors with normal liver histology and suggested healthy ALT thresholds to be 33 IU/L for Asian men and 25 IU/L for Asian women. Similarly, Kang et al. studied ULN of ALT among 7403 Asian health check-up participants, and their ULN of ALT value was 31 U/L for men, and 23 U/L for women. Furthermore, a large population-based study showed that people with slightly increased aminotransferase activity, but still within normal range (35–40 U/L) are at an increased risk of mortality from liver disease. Thus, they claimed that it is necessary to lower the currently accepted ULN of ALT levels.
Currently, the ULN of ALT has been set at a mean value ±2 standard deviation in a group of healthy individuals, so some per cent of these healthy individual have ALT level which fall outside the defined normal range. Also, a minor elevation of ALT may be of no clinical importance, and may even be physiological. Thus, it is possible that lowering the ULN of ALT may assist in identifying patients with liver disease who would have been missed in conventionally used thresholds, meanwhile, it may increase the population with abnormal test results, leading to unnecessary additional tests. Thus, it is a dilemma to practicing physicians, to determine a proper ALT threshold which meets both high quality practice and cost-effectiveness of the healthcare system.
On this point, the present study was designed to investigate the necessity and consequence of lowering currently used ULN of ALT. For this purpose, we aimed: (i) to derive new ALT thresholds in a representative Korean population, (ii) to evaluate the prevalence and aetiology of elevated ALT levels in health check-up participants, and (iii) to verify the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of the new ALT thresholds, compared to those of the currently used ALT thresholds for classifying individuals as having or not having a risk for four common forms of chronic liver disease [hepatitis B virus (HBV) infection, hepatitis C virus (HCV) infection, chronic alcohol consumption and nonalcoholic fatty liver disease (NAFLD)].
Source...