Is Fatty Liver Increased in Pediatric Population with Type 1 Diabetes Mellitus?

J Sae-wong et al. J Pediatr 2021; 230: 32-37. The Prevalence of Nonalcoholic Fatty Liver Disease and Its Risk Factors in Children and Young Adults with Type 1 Diabetes Mellitus

In this cross-sectional study with 50 children with Type 1 Diabetes Mellitus (T1DM), MRE and MRI-PDFF studies were undertaken to determine whether the participants had nonalcoholic fatty liver disease (NAFLD). Key findings:

  • The median age and duration of T1D were 16.9 years (IQR, 13.6-20 years) and 6.5 years (IQR, 4-11 years), respectively. 26% of the cohort were overweight or obese.
  • The prevalence of NAFLD was 10% (more than half had normal ALT values). Four out of 5 patients with NAFLD were overweight/obese, and 2 had an and elevated alanine aminotransferase (ALT) level. None had liver fibrosis (defined as MRE >2.9 kPa).
  • High BMI-SDS (body mass index standard deviation score) was the sole independent risk factor associated with NAFLD (OR, 5.79; 95% CI, 1.04-32.18).

My take: This study is reassuring regarding the prevalence of NAFLD in children and young adults with T1D which was comparable to that in the general population. Routine screening for NAFLD in patients with T1D does not appear to be useful.

Related blog posts:

Why Celiac Serology Needs To Be Looked At Differently in Children with Type 1 Diabetes

A recent study (M Wessels et al. J Pediatr 2020; 223: 87-92Raising the Cut-Off Level of Anti-Tissue Transglutaminase Antibodies to Detect Celiac Disease Reduces the Number of Small Bowel Biopsies in Children with Type 1 Diabetes: A Retrospective Study) recommends changing the approach to celiac disease (CD) diagnosis in children with Type 1 Diabetes Mellitus (T1DM).

Background: The prevalence of CD among patients with T1DM is between 3-10%

Using a retrospective observational cohort with 63 children, the authors recommend raising the cut-off from performing biopsies from 3 times the ULN to 11 times ULN.

Here’s why:

  • This change in increases the specificity from 36% to 73% while reducing sensitivity from 96% to 87%.  In addition, this improves the positive predictive value from 88% to 94%, but lowered negative predictive value from 67% to 53%.  Overall, this leads to a reduction in “unnecessary biopsies.”
  • The authors note that while the serology sensitivity is reduced, it is still acceptable and justified because “normalization of elevated TG2A  can occur in up to one-third of patients.”
  • Another finding from this cohort was that 55% of children with Marsh 0 or 1 histology were symptomatic, indicating that symptoms are not specific for CD.

While the authors have recommended a higher threshold and advocated for repeating serology ~3 months later in those with lower titers, the associated editorial by Stefano Guandalini makes the following points:

  1. Raising the titer threshold would leave 13% of patients with celiac disease undiagnosed (or at least with a delay in diagnosis)
  2. “For lower titers, the physician will have to apply his or her knowledge and conscience in each individual case…we must be mindful of the serious risk of missing too many patients with celiac disease by applying a high threshold, a risk probably outweighing that of an unnecessary biopsy.”

My take: This study shows that in children with T1DM who have abnormal lower-value celiac serology, a careful discussion with parents is needed about the merits of endoscopy or deferring until persistent positivity.

Related blog posts:

Taken near Hunley bridge, Isle of Palms, SC

Quantifying the Risk of Autoimmunity for Celiac Disease

A recent study (MR Khan et al. JPGN 2019; 69: 438-42) examined the rates of autoimmune disorders (AD) among patients with celiac disease (CD) (n=249) compared to a control group (n=498) over an 18 year period (1997-2015). The authors utilized the  a database of medical records via the Rochester Epidemiology Project (Mayo Clinic/Olmstead County).

Key findings:

  • Five years after the index date, 5.0% of CD patients and 1.3% of controls had a de novo AD diagnosis
  • In the pediatric age group, there was an increased risk of AD: 5/83 (7.3%) of CD patients and 0/179 (0%) developed a AD diagnosis at the 5-year mark
  • The authors note that they observed a lower rate of Hashimoto thyroiditis after the diagnosis of CD, likely indicating a protective role of a gluten-free diet
  • Thyroid disorders, type 1 DM, psoriasis/psoriatic arthritis and rheumatoid arthritis were the most common AD identified in patients with CD


  • Retrospective study
  • Adherence with GFD was not assessed

My take: Screening for AD periodically is worthwhile in patients with CD, particularly thyroid disorders and type 1 diabetes which accounted for ~80% of the autoimmune conditions identified.

Briefly noted: R Ahawat et al. JPGN 2019; 69: 449-54. In this study with 38 newly-diagnosed CD, the authors found a high prevalence of low vitamin D (25OHD) levels (65.8%) -defined as <30 ng/mL; however, the control population had a higher rate of 79.3%.  While the authors advocate checking vitamin D levels due to the risk of bone disease, it is noted that bone mineral density and vitamin deficiencies frequently improve with a gluten-free diet (Related post: Celiac studies)

Related posts:


Mauriac Syndrome (Glycogenic Hepatopathy)

A case report (T Malikowski et al. Gastroenterol 2017; 152: 947-49) provides some insight into a fairly common problem –elevated liver tests in the setting of poorly controlled type 1 diabetes mellitus.  This 18-year-old had presented with a glucose of 497 mg/dL, elevated lactate, aspartate aminotransferase 257 U/L, and alanine aminotransferase 178 U/L.

The authors note that Mauriac syndrome “occurs in young patients as a result of poorly controlled type 1 diabetes mellitus.”  It may result in growth retardation, pubertal delay, and cushingoid features.

“Glycogenic hepatopathy is a underrecognized complication of Mauriac syndrome that presents with abdominal pain, nausea, vomiting, elevated serum transaminases, elevated plasma lactate levels, and hepatomegaly  The pathogenesis stems from an accumulation of glycogen in the liver…The diagnosis…is made …when all other causes of liver disease have been excluded…When glucose control is achieved, prognosis is excellent.”

My take: There are many potential reasons for elevated liver enzymes associated with type 1 diabetes mellitus, including celiac disease, and autoimmune hepatitis.  However, familiarity with glycogenic hepatopathy helps with pattern recognition and helps explain the frequent concurrence of liver disease with poorly controlled type 1 diabetes mellitus.

Celiac Diseaase and Diabetes

A recent review (B Weiss, O Pinahs-Hamiel. JPGN 2017; 64: 175-79) of the medical literature describes the various recommendations regarding celiac disease (CD) and type 1 diabetes mellitus (T1DM).

Key points:

  • Two-thirds of patients with T1DM and CD are asymptomatic for CD at diagnosis
  • Many children with T1DM and with positive CD serology may normalize the serology spontaneously.   In one study with 446 children with T1DM who were screened for CD, 38 had persistently abnormal serology whereas 27 had fluctuations in CD serology.  In another study with 738 children, of 48 patients with positive CD serology, normalization was evident in 35% at 1 year.

The authors review recommendations for CD screening.  Several guidelines have recommended soon after diagnosis (especially if >2 years of age).


Their figure 1 algorithm provides guidance on evaluation.  In those patients with T1DM and positive CD serology, if they are asymptomatic, assuring that serology is persistently elevated may be worthwhile before proceeding with small bowel biopsy.  In those who initially test negative for serology, there may be a role for HLA testing and/or periodic screening every few years.

Related editorial on recent article: Celiac Disease, Gut-Brain Axis, and Behavior: Cause, Consequence or Merely Epiphenomenon (A Fasano)  Thanks to KT Park for this reference.  Excerpt:

By assessing the psychological functioning of infants enrolled in the Environmental Determinant of Diabetes in the Young trial and followed prospectively, the authors reported that 3.5-year-old children affected by celiac disease autoimmunity (CDA), defined as positive serology in children at risk, have increased reports of depression/anxiety, aggressive behavior, and sleep disturbances. Interestingly, these symptoms were significantly greater in the 66 children with CDA whose mothers were unaware of the diagnosis compared with the 440 children with CDA whose mothers were aware of the diagnosis and the 3651 children without CDA, decreasing the chance that the reported behaviors were biased by families’ subjective assessment…Prospective studies such as that reported by Smith et al may be a key approach to shedding light on how intestinal factors can influence human behavior and to identifying possible targets to ameliorate psychological symptoms caused by inappropriate gut-brain cross-talk.

Reference articleSmith L, et al.. Psychological manifestation of celiac disease autoimmunity in young children. Pediatrics. 2017;139(3):e20162848

Related blog posts:

Disclaimer: These blog posts are for educational purposes only. Specific dosing of medications (along with potential adverse effects) should be confirmed by prescribing physician.  This content is not a substitute for medical advice, diagnosis or treatment provided by a qualified healthcare provider. Always seek the advice of your physician or other qualified health provider with any questions you may have regarding a condition.

NYT: Educate Your Immune System

A recent commentary updates the concept of the hygiene theory and how our lack of exposures to a ‘dirtier’ environment when we are younger can make us more prone to autoimmune diseases, including celiac disease, diabetes, and multiple sclerosis.

Here’s the link: Educate Your Immune

Here’s an excerpt:

People living just over the border in Russian Karelia, as the region is known, have the same prevalence of genes linked to autoimmune disease [as in Finland]. They also live at the same latitude and in the same climate. And yet they have a much lower vulnerability to autoimmune disease. Celiac disease and Type 1 diabetes occur about one-fifth and one-sixth as often, respectively, in Russian Karelia as in Finland. Hay fever and asthma, allergic diseases that also signal a tendency toward immune overreaction, are far less common.

So in a follow-up study, the results of which appeared last month in the journal Cell, Dr. Xavier and his colleagues followed 222 children who were genetically at risk of developing autoimmune diabetes. The newborns were equally divided among Finland, Russia and Estonia, where the prevalence of Type 1 diabetes is on the rise, but still well below Finland’s.

Autoimmune diabetes can be predicted, to some degree, by the appearance of certain antibodies in the bloodstream that attack one’s own tissues. After three years, 16 Finnish children and 14 Estonian children had these antibodies; only four Russian children did. And when the scientists compared the children’s microbiomes in the three countries, they found stark differences. A group of microbes called bacteroides dominated in Finnish and Estonian infants. But in Russia, bifidobacteria and E. coli held sway….

Russian kids have more fecal oral infections, such as hepatitis A, suggesting more sharing not only of pathogens, but of microbes that may benefit health. And previous studies have found that Russian homes harbor a richer and more diverse community of microbes than Finnish ones….

The world today is very different from the one our immune system evolved to anticipate — not just in what we encounter, but in when we first encounter it. Preventing autoimmune disorders may require emulating aspects of that “dirtier” world.

Screen Shot 2016-06-03 at 6.23.18 PM

Timing of Solid-Food Introduction

The “DAISY” (diabetes autoimmunity  study in the young) study indicates that the timing of solid-food introduction can influence the likelihood of developing type 1 diabetes (T1DM) (JAMA Pediatr 2013; 167: 808-15).

The participants were 1853 children at increased genetic risk for T1DM who were enrolled in a longitudinal observational cohort study in Denver. Early solid-food exposure was considered <4 months of age and late >6 months of age.


  • “Both early and late first exposure to any solid food predicted development of T1DM.”  For early exposure, the Hazard Ratio was 1.91 and for late HR was 3.02.
  • Breastfeeding at the time of introduction to wheat/barley conferred protection (HR 0.47)

The study has several limitations, particularly the relatively low numbers of children who developed T1DM (n=53).

A second study (Pediatrics 2013 [doi: 10.1542/peds2012-3692]) –thanks to Ben Gold for this reference –showed that “solid foods were introduced significantly earlier among the infants with allergies, with 35% of them receiving their first solids before and including 16 weeks, compared with 14% of control infants (P=.011).”   (Solid foods before 17 weeks linked to food allergy)

Bottomline: As with celiac disease (GlutenRelated Disorders” (Part 1) | gutsandgrowth), current science suggests the introduction of solid foods between 4-6 months of age may diminish the risk of developing T1DM as well as food allergies.