Over the past decade, a fierce debate has been brewing over alanine aminotransferase (ALT) reference intervals, involving prominent organizations such as the American College of Gastroenterology (ACG); the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition (NASPGHAN); laboratory experts from the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC); and the Canadian Laboratory Initiative on Pediatric Reference Intervals (CALIPER) (1).
What sparked the debate is a set of clinical guidelines released by ACG and NASPGHAN in 2017 that included key recommendations for ALT’s concentration-based upper reference limit (4, 5) and that proposed using these upper limits as action thresholds. ACG and NASPGHAN developed these guidelines in an effort to remedy longstanding problems with ALT testing. However, their recommendations ended up raising deep concerns within the laboratory medicine community.
Below, we delve into the intricacies of these concerns, while also exploring the significance of ALT testing, its underlying issues and how these affect patients, and how clinical laboratories can make the right choices regarding ALT methods and reference intervals.
THE PROBLEM WITH ALT: REFERENCE INTERVALS
To comprehend the significance of the ALT reference interval debate, it is essential to grasp the role of ALT and its counterpart, aspartate aminotransferase (AST), in the context of liver health. These enzymes are predominantly found in hepatocytes. While ALT and AST are also present in other tissues, such as the heart, skeletal muscles, and kidneys, tests for these enzymes often are referred to as liver function tests because of their rise in blood concentration associated with liver damage (2).
The primary issue with ALT lies in the widely varying reference intervals used by clinical laboratories worldwide (3). This diversity in reference intervals has significant implications for the development of clinical guidelines. When these guidelines are formulated, they rely on specific ALT values to determine when medical intervention is necessary. However, the variable reference intervals employed by different laboratories create a challenging landscape for setting universal clinical standards.
The practical result of this lack of standardization is evident when examining clinical guidelines. Often, instead of providing a specific universal cutoff value for ALT, guidelines use relative measures such as “Take action if ALT exceeds 2 or 3 times your upper limit of normal.” This approach is problematic because it means that the threshold for medical intervention depends on what an individual laboratory considers their upper limit of normal. This lack of standardization can lead to confusion and inconsistencies in patient care and diagnosis — and it’s what led ACG and NASPGHAN to release their 2017 guidelines.
These guidelines specifically recommended adopting universal cutoffs of 33 U/L for adult males, 25 U/L for adult females, 26 U/L for pediatric males, and 22 U/L for pediatric females. However, in an opinion article published in Clinical Chemistry, a group of laboratory medicine experts led by Mauro Panteghini, MD, expressed their reservations about the proposed ALT reference intervals (1). They not only disagreed with the suggested thresholds but also criticized the universal application of these cutoffs.
Moreover, they were rightly upset that both American guideline panels, ACG and NASPGHAN, did not include any laboratory medicine professionals in their decision-making process. This omission of laboratory experts from guideline development was a notable oversight and a significant concern for the laboratory medicine community.
So, what were the primary arguments put forward by Panteghini, et al., and why were they at odds with the American guideline panels? To understand the core of this dispute, let’s explore these arguments in detail.
THE LABORATORY EXPERTS’ PERSPECTIVE: THREE KEY ISSUES
Lack of Standardization in ALT Methods
To appreciate the importance of this point, it is crucial to understand that clinical laboratories have various options when it comes to ALT assays. These assays differ in a significant way — the presence or absence of pyridoxal-5-phosphate (P-5-P) as a cofactor in their reagents. P-5-P is a form of vitamin B6 and plays a critical role in the ALT enzyme’s activity. It is needed for the enzyme to be active and capable of catalyzing the reaction that produces the signal measured by laboratory instruments.
The issue is that many major manufacturers offer clinical laboratories a choice between ALT methods that include P-5-P and those that do not. While having choices is typically desirable, it is a significant concern in this context. The reason is simple: ALT assays without P-5-P are unable to detect serious elevations in ALT levels in patients with vitamin B6 deficiency. In other words, if you have a vitamin B6 deficiency, your blood may contain a substantial amount of ALT, but the assays that lack P-5-P may not generate detectable signals (6). This is a crucial problem because these assays may fail to identify patients with potentially severe liver conditions, including alcoholic hepatitis.
Panteghini, et al., rightfully argued that ALT assays without P-5-P, which miss severe ALT elevations in patients with vitamin B6 deficiency, are still used in clinical laboratories around the world. This practice is deeply concerning, given the clear evidence in favor of assays with P-5-P, and IFCC recommending their use since 2003.
Lack of Traceability to IFCC Measurement
The second point raised by Panteghini, et al., pertains to the traceability of ALT results to the IFCC reference measurement procedure (1). When results from different laboratories are aligned with the IFCC reference measurement procedure, they should be similar, with minimal variation. This traceability ensures that when a patient is tested in one lab and then in another, the reported ALT results will be consistent, even if the labs use different methods. However, the opinion article authors noted that many ALT methods lack this essential traceability to the IFCC reference measurement procedure. As a result, different ALT assays can produce different results, making it challenging to establish universal reference intervals or thresholds, as proposed by ACG and NASPGHAN.
Inappropriate Criteria for the Selection of a Reference Population
In the realm of laboratory medicine, there is a unique power that laboratory professionals possess — the power to decide who is considered healthy. But deciding who is healthy is a complex task, as the criteria for selecting a reference population have far-reaching implications. One critical aspect of this debate revolves around the criteria used to select individuals for reference interval studies. For instance, overweight and obese individuals, those with a body mass index (BMI) over 25, are known to have higher ALT levels (7). This elevation is primarily due to the inflammatory effects that excess weight exerts on the liver.
A study performed by one of the coauthors of the opinion piece, Ferruccio Ceriotti, MD, initially included overweight individuals in its reference population, which influenced the proposed ALT reference intervals (1, 11). The opinion piece authors revisited this data, and by adjusting for the effect of overweight individuals, they lowered the proposed reference intervals, aligning them more closely with the recommendations made by ACG and NASPGHAN. This highlights the sensitivity of reference intervals to factors like BMI.
However, when we attempted to verify these new reference intervals in our own laboratory, we encountered challenges. For instance, we recruited individuals with a healthy BMI, those not taking any medications, and with normal profiles for lipids, iron, and hepatitis, and discovered that over 10% of the participants had ALT levels above the newly proposed limits. This discrepancy led us to question what other factors might be influencing reference intervals that are not accounted for. As we delved deeper into the literature and collaborated with our clinical colleagues, we discovered another significant factor: alcohol consumption. Regular alcohol consumption, even in moderate amounts, is known to increase ALT levels, with a more pronounced effect in overweight and obese individuals (7). Studies on alcohol abstinence have also supported this theory, showing a substantial drop in ALT levels after just a few weeks of abstinence in individuals with a history of heavy drinking (8).
The problem with the study by Ceriotti, et al., is that it included individuals who reported consuming up to 30 grams of alcoholic beverages per day (1, 11). This equates to roughly 1 ounce of an alcoholic beverage, in which the quantity of alcohol can vary greatly depending on the type of alcoholic beverage consumed. The problems with the inclusion of alcohol consumers in this study are threefold: First, investigators had no way to verify the actual amount of alcohol participants consumed because inclusion in the study relied on self-reported survey responses. Second, participants may not accurately recall or report their average alcohol consumption, making it challenging to ascertain the true extent of their drinking habits. Third, not all alcoholic beverages are created equal, making it difficult to equate the impact of different types of alcohol on ALT levels. As a result, we suspected that the inclusion criteria in the study, while well-intentioned, might not have effectively excluded individuals who regularly consumed high amounts of alcohol. This raised questions about the study’s validity and the need to account for the influence of alcohol on ALT levels.
NEW STUDY: ALCOHOL CONSUMPTION AND ALT REFERENCE INTERVALS
To address this issue, we aimed to determine if alcohol consumption was indeed a significant factor affecting reference interval studies (8). However, we recognized that it would be difficult (and expensive) to recruit 120 men and 120 women who abstained from alcohol entirely, especially considering this investigation was undertaken amidst the COVID-19 pandemic, a time when alcohol consumption in the United States increased substantially. So, we resorted to what is called an "indirect sampling approach" (9). Essentially, we retrieved anonymized outpatient data from more than 7,000 individuals who had visited our institution over a 2-year period (8). These individuals had their blood tested on a Roche cobas 8000 platform using IFCC-traceable ALT and AST assays, both of which included P-5-P. The participants were carefully selected; only those with BMIs between 19 and 25 were included in the study. Additionally, individuals who had elevated AST levels and ALT levels above 80 U/L were excluded from the analysis, as these values were highly suggestive of underlying liver pathology.
This indirect sampling approach, although imperfect, aimed to investigate whether age and sex-specific reference intervals could provide insights into the source of the reference interval variability. We stratified the data by age and sex and derived theoretical reference intervals for each group. In the case of individuals aged 13−17 years, we estimated reference intervals that closely resembled those proposed by CALIPER: 34 U/L for males and 27 U/L for females (8). We felt that this validated our approach, as alcohol is not expected to be a major factor in this age group.
However, for individuals aged 18−20 years, things took a different turn. In the United States, these individuals are considered adults, but the legal drinking age is 21. Theoretically, they should have more restricted access to alcohol compared with the rest of the adult population. While this scenario does not necessarily reflect the reality of college life in the U.S., it does present an interesting test case. So we sought to answer a crucial question: Using an ALT assay with P-5-P that was traceable to the IFCC reference measurement procedure, did reference intervals differ for adults with normal BMIs under the age of 21 compared to those above 21?
The results of this analysis provided a noteworthy insight. Men and women aged 18−20 had upper reference limits of 38 U/L and 25 U/L, respectively, while the age groups above 21 displayed higher reference ranges, varying between 40−54 U/L for men and 34−40 U/L for women. This indicated an upward and then downward trend in ALT reference intervals with increasing age, with the over-80 age group having reference intervals closest to those of the 18 to 20-year-olds.
These findings provided new insights: the prevalence of overweight and obese individuals and the varying amounts of alcohol consumption within populations were two major factors contributing to the inability to agree on a universal ALT cutoff. The study results also hinted at a possible solution: adopting clinical decision limits, rather than relying on population-based reference intervals that vary according to each population’s dietary and drinking habits.
To address the reference interval variability problem, we suggest adopting clinical decision limits and adjusting ALT reference intervals to 42 U/L for men and 30 U/L for women, based on a study that derived these limits using an IFCC standardized test with P-5-P in over 21,296 healthy individuals and over 2,000 patients with dysmetabolism and chronic liver disease (10). These decision limits are evidence-based and have the potential to provide more meaningful guidance to healthcare professionals. The focus would shift from establishing an arbitrary "normal" range to recognizing actionable thresholds for patient care. Additionally, by ensuring that ALT assays include P-5-P and are traceable to the IFCC reference measurement procedure, the accuracy of these results could be improved substantially.
THE TAKEAWAYS: WHAT MATTERS MOST?
The ALT reference interval debate is undoubtedly a complex issue, affecting patient care and clinical laboratory practice. The perspective of laboratory experts, with its emphasis on the importance of ALT methods, traceability to IFCC measurements, and the influence of factors like BMI and alcohol consumption, offers a fresh lens through which to examine this debate.
The newly proposed clinical decision limits of 42 U/L for men and 30 U/L for women hold promise as a pragmatic approach to ALT reference intervals. The adoption of these limits would ensure more consistent patient care across different laboratory settings. They would also provide a valuable opportunity to refocus on the primary goal of laboratory medicine — delivering precise and actionable results to aid clinicians in making critical healthcare decisions.
However, it is essential to recognize that this is an ongoing debate, and our recommendations are not universally accepted. Clinical practice is guided by evidence-based medicine, and as more research emerges, it may further shape and redefine the reference intervals and clinical decision limits for ALT.
In conclusion, the ALT reference interval debate serves as a compelling reminder of the dynamic nature of medical science and laboratory medicine. It underscores the importance of continuous collaboration between clinicians and laboratory professionals, as well as the need for evidence-based guidelines that can adapt to the evolving landscape of patient care and diagnostic technology.
Ultimately, what matters most is the well-being of patients and the accurate, actionable information that laboratory tests can provide to guide healthcare decisions. By addressing the issues raised in this debate, we take a step closer to ensuring that every patient receives the best possible care based on reliable laboratory data.
Christopher D. Koch, PhD, DABCC, is a clinical assistant professor of pathology at the University of South Dakota Sanford School of Medicine and director of chemistry at Sanford Laboratories in Sioux Falls and Rapid City, South Dakota. +Email: [email protected]
Joe M. El-Khoury, PhD, DABCC, FADLM, is an associate professor of laboratory medicine at Yale University School of Medicine and codirector of the clinical chemistry laboratory and clinical chemistry fellowship program at Yale-New Haven Health in New Haven, Connecticut. +Email: [email protected]
References
- Panteghini M, Adeli K, Ceriotti F, et al. American liver guidelines and cutoffs for "normal" ALT: A potential for overdiagnosis. Clin Chem 2017; doi: 10.1373/clinchem.2017.274977.
- El-Khoury JM, Wang S. Aspartate aminotransferase and alanine aminotransferase. In: Kazmierczak SC, Azzazy HME, editors. Diagnostic Enzymology. Berlin, Germany: De Gruyter, 2014:81-104.
- Neuschwander-Tetri BA, Unalp A, Creer MH, et al. Influence of local reference populations on upper limits of normal for serum alanine aminotransferase levels. Arch Intern Med 2008; doi: 10.1001/archinternmed.2007.
- Kwo PY, Cohen SM, Lim JK. ACG clinical guideline: Evaluation of abnormal liver chemistries. Am J Gastroenterol 2017; doi: 10.1038/ajg.2016.
- Vos MB, Abrams SH, Barlow SE, et al. NASPGHAN clinical practice guideline for the diagnosis and treatment of nonalcoholic fatty liver disease in children: Recommendations from the expert committee on NAFLD (ECON) and the North American Society of Pediatric Gastroenterology, Hepatology, and Nutrition (NASPGHAN). J Pediatr Gastroenterol Nutr 2017; doi: 10.1097/MPG.
- Mills JR, Greene DN, Bainton J, et al. Fluctuating serum aspartate aminotransferase activity in a complicated pregnancy. Clin Chem 2015; doi: 10.1373/clinchem.2014.
- Alatalo PI, Koivisto HM, Hietala JP, et al. Effect of moderate alcohol consumption on liver enzymes increases with increasing body mass index. Am J Clin Nutr 2008; doi: 10.1093/ajcn/88.4.
- Vera MA, Koch CD, Liapakis A, et al. The ALT upper reference interval debate: Blame it on the alcohol. Clin Chim Acta 2022; doi: 10.1016/j.cca.2021.12.
- Jones GRD, Haeckel R, Loh TP, et al. Indirect methods for reference interval determination — review and recommendations. Clin Chem Lab Med 2018; doi: 10.1515/cclm-2018-
- Valenti L, Pelusi S, Bianco C, et al. Definition of healthy ranges for alanine aminotransferase levels: A 2021 update. Hepatol Commun 2021; doi: 10.1002/hep4.1794
- Ceriotti F, Henny J, Queralto J, et al. Common reference intervals for aspartate aminotransferase (AST), alanine aminotransferase (ALT), and γ-glutamyl transferase (GGT) in serum: Results from an IFCC multicenter study. Clin Chem Lab Med 2010; doi: 10.1515/CCLM.2010.