With the increased prevalence of childhood obesity noted in recent decades, NALFD has become the most common liver abnormality found among the pediatric population [3
]. Moreover, NAFLD risk is disproportionately higher among Hispanic youth where elevated ALT levels have been reported in more than 11% of Hispanic adolescents in the US [9
], almost 24% of overweight Hispanic youth [10
], and as many as 60% of obese Hispanic youth in a clinical sample [11
]. The current report extends the available science regarding NAFLD-related health disparities to newly diagnosed adolescents with T2DM and suggest that more than 70% of Hispanic youth with T2DM exhibit elevated ALT levels. These data support recent findings that youth with T2DM are at considerable risk for premature morbidity and mortality [12
NAFLD and T2DM are thought to share a common pathophysiology related to insulin resistance [13
]. Obese youth with NAFLD are significantly more insulin resistant than obese youth who do not exhibit NAFLD [14
]. Similarly, adolescents with T2DM are significantly more insulin resistant than BMI-matched adolescents with normal glucose tolerance [4
]. Although obesity and insulin resistance are common characteristics of youth with both NAFLD and T2DM, these traits alone do not explain why Hispanic patients in our study exhibited higher transaminase values. Our patients exhibited similar descriptive and glycemic profiles in terms of obesity and severity of diabetes (i.e. similar BMI and HbA1c) which, suggests that ethnicity may further play a role in the pathogenesis of NAFLD in the obese diabetic population. Goran and colleagues [15
] have reported that Hispanic and African-American youth are significantly more insulin resistant compared to non-Hispanic whites. Interestingly, the authors noted that the associated compensatory response to insulin resistance is different between these minority groups and may indicate ethnic-specific mechanisms underlying obesity-related diseases. Unfortunately, glucose and insulin levels were not available in our sample, limiting our ability to examine whether variations in insulin resistance may contribute to the observed differences in elevated transaminases between groups. A study by Schwimmer et al [16
] identified an extremely high heritability estimate (h2
= 0.850; standard error, 0.325), for fatty liver disease in a predominantly Hispanic cohort suggesting that Hispanics may be predisposed to developing NAFLD. This same group hypothesized that African-Americans may either have a protective factor or lack a vulnerability factor for the development of fatty liver [17
]. Collectively, these findings suggest that genetic factors may contribute to the higher risk of NAFLD among Hispanic youth. In particular, when coupled with obesity, severe insulin resistance, and T2DM, Hispanic youth may be at extreme risk for premature hepatic morbidity.
The retrospective nature of this report precludes causal conclusions regarding ethnic-specific disease mechanisms or trajectories. A recent study in a community-sample of Korean adults suggests that elevations in ALT predict the development of T2DM over time [18
]. A study in obese youth from the United Kingdom suggests that youth with the metabolic syndrome and those with a family history of T2DM, both of which are risk factors for developing T2DM, are more likely to exhibit elevated ALT levels [19
]. Therefore, a prospective study following normoglycemic youth as well as patients with T2DM over time or through a course of treatment may better identify potential causal relationships between hepatic health and T2DM or whether gene-environment interactions concomitantly impact the development NAFLD and T2DM. Although our sample had fewer African-Americans and non-Hispanic whites compared to Hispanics this is most likely a function of the ethnic disparities in T2DM risk coupled with the demographics of the local Phoenix metropolitan area. Most of the patients resided in Maricopa County Arizona where Phoenix Children’s Hospital is the only tertiary care referral center for Pediatrics. The most recent census data (2010) showed Hispanics comprise 40.8% of the population while African-Americans comprised 6.5%. These data coupled with the demographics of the patient population in our study support disparities in T2DM risk in the local community and may partially explain the limited number of African-Americans compared to Hispanics and non-Hispanic whites. We acknowledge the limited sample size of the entire cohort and especially the non-Hispanic groups. To our knowledge, the only other publication to date examining liver enzymes in youth with T2DM described 48 youth from Colorado in whom 65% (n=31) had liver enzymes measured at diagnosis [6
]. By comparison, the multi-center SEARCH for Diabetes in Youth Study representing 6-centers from four geographically defined populations of 3.5 million youth identified a total of 446 African-American, Hispanic, and non-Hispanic White youth with T2DM in their initial year [20
]. Given the paucity of published data in youth with T2DM and the fact that our report is the first to describe the impact of ethnicity on NAFLD in youth with T2DM we expect that future studies will build upon these findings using more representative samples.
It is noteworthy that elevated liver transaminases only appear to be indicators of liver inflammation and are not diagnostic of NAFLD [7
]. However, in follow-up of six Hispanic patients with ALT or AST more than twice the ULN who underwent liver ultrasound, characteristics consistent with fatty liver disease were found in all six. Additionally, two of these patients had splenomegaly on the ultrasound with follow up biopsy confirming fibrosis (Grade III in a 15 y/o Hispanic male and Grade II in a 16 year old Hispanic female). The remaining four were followed by both the endocrinology and gastroenterology services. Future studies should employ more sophisticated imaging methods such as magnetic resonance spectroscopy to better quantify liver fat and NAFLD in youth with T2DM [21
]. Not only will these studies provide more definitive information on ethnic differences in NAFLD risk, but may also be useful in developing clinical prediction models for when and who to refer for consult with a hepatologist.
As the prevalence of obesity increases in the pediatric population it could be argued that all high-risk patients including Hispanics and especially those newly diagnosed with T2DM have liver transaminases evaluated. In support of this argument, the American Medical Association recently released expert committee recommendations regarding pediatric obesity which suggested biannual screening for NAFLD by transaminase evaluation starting at 10 years of age for children with a BMI ≥95th percentile and those with a BMI between the 85th to 94th percentile who have other risk factors [22
]. Furthermore, the committee suggested that ALT or AST results 2 times normal levels should prompt consultation with a pediatric hepatologist. The importance of regular transaminase evaluation is critical as hepatic inflammation can quickly progress to frank fibrosis. In a review of children diagnosed with NALFD in Toronto, Rashid et al [23
]. showed a high prevalence (77%) of progression to liver inflammation and fibrosis. This rapid trajectory underscores the importance of follow-up in those children with elevated liver transaminases as end-stage complications can be present in childhood. Particularly concerning is the process starting at a young age even if advanced findings are not yet present. Continued and improved interaction between gastroenterologists and pediatric endocrinologists are needed as the team approach to medical management of patients with T2DM may require more comprehensive medical evaluation including liver biopsies. At our institution, biopsies are typically performed only if the ultrasound suggests cirrhosis. Further longitudinal studies may provide better guidelines regarding the progression of liver disease in relation to the timing of biopsy. Our data also suggest that variations in the management should occur based on ethnicities possibly using different timing of monitoring and goals of treatment. One might more readily refer and/or follow more closely a Hispanic patient with laboratory or imaging studies suggesting NAFLD.