Elsevier

Annals of Epidemiology

Volume 22, Issue 11, November 2012, Pages 789-798
Annals of Epidemiology

Alanine transaminase has opposite associations with death from diabetes and ischemic heart disease in NHANES III

https://doi.org/10.1016/j.annepidem.2012.08.003Get rights and content

Abstract

Purpose

Diabetes increases the risk of ischemic heart disease (IHD). Stringent control of diabetes does not reliably reduce cardiovascular events. Some global regions, such as East Asia, have low mortality rates from IHD and high rates of diabetes. We hypothesized that some aspects of liver function might underlie this paradox.

Methods

We used multivariable proportional hazards regression in 16,865 adults from the National Health and Nutrition Examination Survey (NHANES) III (1988–1994) followed until December 31, 2006, to assess the adjusted associations of gender-specific tertiles of alanine transaminase (ALT), as a marker of hepatocellular damage, and bilirubin (BIL), as a marker of other aspects of liver function, with death from diabetes (n = 132), IHD related to diabetes (n = 153), and IHD unrelated to diabetes (n = 921).

Results

ALT was positively associated with death from diabetes (hazard ratio [HR], 2.17; 95% confidence interval [CI], 1.19–3.98 for high compared with low ALT tertile) and IHD related to diabetes (HR, 2.14; 95% CI, 1.07–4.31), but negatively associated with IHD unrelated to diabetes (HR, 0.76; 95% CI, 0.58–0.98) adjusted for age, gender, education, race/ethnicity, smoking, and alcohol use. BIL had no such associations.

Conclusions

ALT may be a marker of an underlying etiology relating to the paradoxical associations of diabetes and IHD at a population level.

Introduction

Diabetes is a well-established risk factor for cardiovascular diseases (CVD). Nevertheless, there are settings, where rates of CVD, and particularly of ischemic heart disease (IHD), are not congruent with those of diabetes, such as parts of East Asia [1], [2] and among some migrant groups in the United States [3], [4] and elsewhere [5], [6], [7]. Meta-analysis of observational studies has also identified that fasting glucose does not have a log-linear relation with CVD [8], unlike other well-established risk factors, such as blood pressure and lipid levels [8]. Similarly, randomized controlled trials have shown that stringent control of diabetes does not reliably reduce cardiovascular events [9], whereas stringent control of blood pressure and lipid levels reduces cardiovascular events effectively [10], [11]. Taken together, these observations demonstrate a complex relation between diabetes and CVD. Understanding this relation is important for managing the coming epidemic of noncommunicable chronic diseases (NCDs) in non-Western settings where currently the most common manifestation of this epidemic is diabetes [2], [12], [13], [14], and it is unclear whether prevention should focus on diabetes or CVD.

Diabetes and CVD largely share risk factors; however, some risk factors do not seem to have the same effects on diabetes and CVD and thus provide a starting point for investigating this complex, potentially paradoxical, relation. Alanine transaminase (ALT), a marker of one aspect of poor liver function, is positively associated with diabetes [15]. However, ALT does not usually have the same positive association with CVD [16], [17], [18], [19], [20], [21], and high ALT in the normal range has been observed to be associated with lower risk of death from IHD [21]. As such, this indicates a complex, but potentially etiologically revealing, relation of a marker of specifically long-term liver, particularly hepatocellular, damage [22], [23], with several NCDs. Currently, this anomalous pattern is largely overlooked, but as with all unexplained paradoxes, it presents potentially important etiologic insights, with corresponding implications for clinical decision making and prevention. These apparent anomalies may simply be owing to the effects of chronic liver disease and Asian diets on serum lipids. Alternatively, based on physiology and experimental evidence, we have previously hypothesized that sex steroids play different roles in diabetes and IHD [24], [25], [26], with androgens protecting against specifically diabetes, via increased muscle mass, but not against IHD [24], [25], [26]. The liver is a key organ for the synthesis of the substrate (cholesterol) [27] for testosterone, the conversion of testosterone to dihydrotestosterone [28], and the homeostasis of zinc [29], which may be involved in testosterone synthesis in the gonads [30]. Restoring liver function through transplantation increases testosterone [31]. Poor liver function would be expected to be positively associated with death owing or related to diabetes, but negatively associated with the majority of deaths from IHD, which are unrelated to diabetes [32]. On the same basis, poor liver function would also be expected to be negatively associated with hormonally modulated cancers.

To clarify the role of the aspect of liver function, measured by ALT, in death from diabetes and IHD we used a large, population-representative prospective study from the United States, the National Health and Nutrition Examination Survey (NHANES) III, which gives cause of death and whether diabetes was a contributing cause, to examine the association of ALT, at baseline with death from IHD according to the role of diabetes in the death. We also examined whether the associations differed by gender, because we might expect effects driven by androgens to be more evident among men than women. As control outcomes, we examined the association of ALT with death from stroke, other CVD, and all CVD. For content validity of the outcome, we examined the association of ALT with death from diabetes and hormonally related cancers, because these would be expected to be positively and negatively, respectively, associated with liver function. As a control exposure, we assessed the association of bilirubin (BIL) with the same outcomes, because BIL is a marker of a different aspect of liver function, that is, cholestasis rather than hepatocellular damage [22], [23], which we would expect to be unrelated to the cardiovascular outcomes considered.

Section snippets

Sources of data

NHANES III was conducted from 1988 through 1994 by the National Center for Health Statistics of the Centers for Disease Control and Prevention [33]. The NHANES III survey used complex, multistage, stratified, clustered samples of civilian, noninstitutionalized populations of age 2 months or older to collect information about the health of people residing in the United States. Participants have since been followed through 2006, via record linkage to the National Death Index, with cause-specific

Results

Out of 39,695 people selected for NHANES III, 33,994 were interviewed and 30,818 were examined. Of those examined 18,162 were aged 17 years or older and 18,149 had vital status; of these 16,854 had ALT (range, 1–486 U/L) and BIL (range, 0–7.9 mg/dL). After a mean follow-up of 13.2 years, there were 1074 deaths from IHD, of which 153 were related to diabetes. Table 1 shows descriptive analysis giving the associations of ALT and BIL tertiles with the covariates. Higher ALT was associated with

Discussion

The aspect of poor liver function measured by ALT, but not by BIL, was positively associated with death from diabetes among men and death from IHD related to diabetes. However, this same exposure had a different association with the majority of IHD deaths unrelated to diabetes, such that high ALT was negatively associated with death from IHD unrelated to diabetes in all models considered, and perhaps also with death from hormonally modulated cancers. Notably, this negative association of high

References (57)

  • J. He et al.

    Major causes of death among men and women in China

    N Engl J Med

    (2005)
  • W. Yang et al.

    Prevalence of diabetes among men and women in China

    N Engl J Med

    (2010)
  • C. Iribarren et al.

    Birthplace and mortality among insured Latinos: the paradox revisited

    Ethn Dis

    (2009)
  • C.C. Cowie et al.

    Full accounting of diabetes and pre-diabetes in the U.S. population in 1988-1994 and 2005-2006

    Diabetes Care

    (2009)
  • I. Stirbu et al.

    Differences in avoidable mortality between migrants and the native Dutch in The Netherlands

    BMC Public Health

    (2006)
  • C. Agyemang et al.

    Cardiovascular disease, diabetes and established risk factors among populations of sub-Saharan African descent in Europe: a literature review

    Global Health

    (2009)
  • J.H. Medalie

    Risk factors other than hyperglycemia in diabetic macrovascular disease

    Diabetes Care

    (1979)
  • N. Sarwar et al.

    Diabetes mellitus, fasting blood glucose concentration, and risk of vascular disease: a collaborative meta-analysis of 102 prospective studies

    Lancet

    (2010)
  • R. Boussageon et al.

    Effect of intensive glucose lowering treatment on all cause mortality, cardiovascular death, and microvascular events in type 2 diabetes: meta-analysis of randomised controlled trials

    BMJ

    (2011)
  • E.J. Mills et al.

    Efficacy and safety of statin treatment for cardiovascular disease: a network meta-analysis of 170,255 patients from 76 randomized trials

    QJM

    (2011)
  • F. Gueyffier et al.

    Effect of antihypertensive drug treatment on cardiovascular outcomes in women and men. A meta-analysis of individual patient data from randomized, controlled trials. The INDANA Investigators

    Ann Intern Med

    (1997)
  • R.M. Anjana et al.

    Prevalence of diabetes and prediabetes (impaired fasting glucose and/or impaired glucose tolerance) in urban and rural India: Phase I results of the Indian Council of Medical Research-INdia DIABetes (ICMR-INDIAB) study

    Diabetologia

    (2011)
  • L. Mihardja et al.

    Prevalence and determinants of diabetes mellitus and impaired glucose tolerance in Indonesia (a part of basic health research/Riskesdas)

    Acta Med Indones

    (2009)
  • V. Hall et al.

    Diabetes in Sub Saharan Africa 1999-2011: epidemiology and public health implications. A systematic review

    BMC Public Health

    (2011)
  • A. Fraser et al.

    Alanine aminotransferase, gamma-glutamyltransferase, and incident diabetes: the British Women's Heart and Health Study and meta-analysis

    Diabetes Care

    (2009)
  • A. Fraser et al.

    Gamma-glutamyltransferase is associated with incident vascular events independently of alcohol intake: analysis of the British Women's Heart and Health Study and Meta-Analysis

    Arterioscler Thromb Vasc Biol

    (2007)
  • W. Goessling et al.

    Aminotransferase levels and 20-year risk of metabolic syndrome, diabetes, and cardiovascular disease

    Gastroenterology

    (2008)
  • J.K. Olynyk et al.

    Serum alanine aminotransferase, metabolic syndrome, and cardiovascular disease in an Australian population

    Am J Gastroenterol

    (2009)
  • Cited by (34)

    • Inverse association of alanine aminotransferase within normal range with prognosis in patients with coronary artery disease

      2019, Clinica Chimica Acta
      Citation Excerpt :

      Recent meta-analyses have reported an overall negative association between ALT and all-cause or CVD-related mortality [17,18]. However, caution is warranted when interpreting their results due to the inconsistent character of the association, highly significant heterogeneity across the studies and variations in the direction of association by age [17], diabetes status [19] or geographic location [18]. Evidence on the association between ALT and the risk of mortality in patients with coronary artery disease (CAD) is rather limited [20].

    • Associations of the fatty liver and hepatic steatosis indices with risk of cardiovascular disease: Interrelationship with age

      2017, Clinica Chimica Acta
      Citation Excerpt :

      Indeed, we and others have shown that age modifies the relationship between ALT and CVD risk [44,46]. Our findings may also reflect the paradoxical relationships between the aminotransferases, diabetes, and CVD; which we and others have previously reported [46,47]. These theories are biologically plausible but are speculative and therefore further research is needed to clarify these mechanistic pathways.

    • Complex association of serum alanine aminotransferase with the risk of future cardiovascular disease in type 2 diabetes

      2016, Atherosclerosis
      Citation Excerpt :

      Interestingly, diabetes is suggested as one of the main drivers of this negative association, based on the observation of less robust positive ALT-CVD correlation among those with incident diabetes during follow-up and stronger inverse association among subjects with diabetes [26]. However, there is also data to support that the positive association between ischemic heart disease and ALT is enhanced among patients with diabetes [27]. Additionally, studies reporting positive association of ALT with CVD events primarily investigated development of CHD episodes rather than aggregate multiorgan CVD events.

    • Inverse linear associations between liver aminotransferases and incident cardiovascular disease risk: The PREVEND study

      2015, Atherosclerosis
      Citation Excerpt :

      Finally, our results remained robust on several sensitivity analyses such as using normal ranges of the aminotransferases, excluding individuals with a history of diabetes, and excluding individuals with potential fatty liver disease. The mechanistic pathways for the inverse associations between the liver aminotransferases and CVD risk are not fully understood and is a topic of evolving debate [5,6,45]. Nonetheless, a number of potential mechanisms have been postulated.

    View all citing articles on Scopus
    View full text