Serum liver enzymes - should we count on them?

Authors


The value of serum liver enzymes in the evaluation of patients with liver diseases is rather limited [1-3]. Three of the four major enzymes, asparate aminotransferase (AST), alkaline phosphatase (ALP), gamma-glutamyltransferase (GGT) are non-specific and originate from organs apart from the liver, while alanine aminotransferase (ALT) is the only enzyme that is almost exclusively liver specific [1-3]. For this reason, we never count on the level of liver enzymes in the evaluation of the severity of liver diseases and we need histological data, derived from liver biopsy or alternatively, from non-invasive modalities [4]. In cases of acute, subfulminant and fulminant hepatic failure, we still never use serum liver enzymes as a predictor of severity but rather use synthetic hepatic function, as a much more meaningful significant indicator [4]. Using serum liver enzyme levels has another limitation, namely, the uncertainty regarding the upper limit of the normal range, and specifically for male vs. female, for various races of origin and for the young vs. the elderly [2-5].

In recent years, numerous public health studies, dealing with the prediction of all-cause mortality and specifically liver-related mortality, are trying to find variables that can predict mortality in various cohorts [6-8]. In these studies, serum liver enzymes, have an important role, and often can predict or at least are associated with all-cause mortality [6-8].

Therefore, one should wonder, how can a liver associated parameter be so unreliable regarding the evaluation of liver status, and is still of value with regard to all or liver/non-liver- cause mortality?

Using electronic database of 273 141 participates, Du et al. have shown that GGT is an independent predictor for future cardiovascular mortality and all-cause mortality [9].

In the last decade, non-alcoholic fatty liver disease (NAFLD), has emerged as the most common liver disease in the western world [10]. NAFLD is the hepatic manifestation of the metabolic syndrome [10] and is associated with cardiovascular diseases [10]. Therefore, it is not surprising, that Lerchbaum et al. [11] found that high levels of fatty liver index (including GGT) are associated with increased all-cause, cardiovascular and non-cardiovascular mortality, as well as cancer mortality [11]. Liu et al. have recently shown, that GGT levels are positively associated with risk of metabolic syndrome, independently of alcohol intake [12] and that GGT levels is positively associated with the development of hypertension [13]. In a cross-sectional analysis of participants in the Third National Health and Nutrition Examination Survey (NHANES), those with elevated ALT activity had a higher 10-year-calculated risk of coronary artery disease than those with normal ALT activity [14, 15].

In this issue of Liver International, Koehler et al. have evaluated the association of the four liver enzymes mentioned above (ALT, AST, GGT, ALP) with all-cause and cause-specific mortality in the elderly population [16]. They used a large population-based study of persons aged 55 years or older, the Rotterdam study. The strength of this study is based on three parameters: a large cohort of 5186 participants, a long duration of 19.5 years follow-up and the active review of the patients as opposed to retrospective data review used in most other studies in the field [16]. They found that all serum liver enzymes were associated with all-cause mortality, GGT was associated with increased cardiovascular mortality (atherogenesis? Oxidative stress?), ALP was associated with all-cause and with cancer related mortality, while GGT and ALT were associated with increased cancer-related mortality. Interestingly, low serum levels of AST and ALT were also associated with a higher risk of all-cause mortality in the elderly. They conclude that liver enzyme tests may represent useful indicators of longevity in the elderly [16].

One of the limitations of this study was the inclusion of patients with various liver diseases (NAFLD, ASH, HCV) and of patients using hepatotoxic medications that can affect liver enzymes [16].

In 2009, Fraser et al. reported that GGT but not ALT predicts mortality among older Danish twins aged 73–94 years. Although several similar studies were published in the past, the strength of the twins study was the use of intrauterine (genetic and environmental) and early life exposure while it was also controlled for adult lifestyle factors [17]. Recently, Loomba et al. have shown in community-dwelling older adults, that serum GGT is an independent predictor of all-cause, cardiovascular and liver mortality in the elderly (mean age 70 years) [18].

What do we know about liver-related mortality? Cirrhosis is an important cause of morbidity and mortality [19]. According the National Centers for Health Statistic (NCHS) and the Centers for Disease Control and Prevention (CDC), chronic liver disease and cirrhosis is the 12th leading cause of death in the USA, the 4th among persons in the age group of 45–55 and the 7th between 55–64 years [19].

These data are important to establish health care policies and guidelines, mainly for understanding disease mechanisms, investigating the optimal means for diagnosis and treatment and disseminate accurate information to health professionals, patients and the public [19]. Recently, Asrani et al. showed that liver-related mortality has been underestimated during the past two decades in the USA [19]. Indeed, in the case of liver related mortality, we should separate those who die directly from liver disease from those who die as a result of other causes, mainly cardiovascular and cancer [19].

Marchesini and Targher from Italy, confronted the confusion derived from several important studies in the field, in their commentary ‘liver enzymes and the risk of adverse cardiovascular outcomes-the lower, the better?’ [20] They suggest that individuals with high – normal ALT might have died early from chronic liver disease, congestive heart failure, malignancy, excessive alcohol consumption or diabetes while survivors who reach old age despite high ALT may represent a population with inherited resistance to the deleterious effect of their chronic medical condition [20].

The finding that low ALT serum levels was also associated with higher risk of mortality, receive support from a previous study by Ford et al., that tested the relationship between ALT activity, mortality and cardiovascular events in two large prospective statin trials that explicitly excluded subjects with clinically significant liver damage [21]. They found that the proportional hazard ratios of all-cause mortality for the 4th vs. the 1st quartile of ALT were all significantly <1 as were the risks of fatal plus non-fatal cardiovascular event [21]. In their study, subjects in the lower quartile of ALT included a much higher number of smokers and with pre-existing vascular disease than those in the upper quartile [21].

So what are the practical implications of these studies, particularly with regard to the elderly? Should physicians be more alert to the level of serum liver enzyme elevation and do their best to normalize liver enzymes? Is the danger of refraining from certain drugs, such as pravastatin for secondary prevention of cardiovascular diseases, outweigh the benefit of reducing liver enzyme levels? Is there a certain liver enzyme-serum level that has been proven dangerous or safe? If it is defined as any level above normal, what is the true upper limit of the normal range? And most importantly with regards to this study, should there be a difference in our attitude as physicians to the relatively young as opposed to the elderly with the same level of liver enzymes? Moreover, should insurance companies for example, increase their fee for those with elevated serum liver enzymes?

In summary, serum liver enzymes serve as predictors of mortality from various liver and non-liver causes while they were never considered gold standard and are not even considered reliable for the evaluation of liver injury. Should or shouldn't we trust them?

Ancillary