Understanding the Reasons for Abnormal Liver Enzymes in Pediatric Inflammatory Bowel Disease

A recent large single center study (Pusateri AJ et al. JPGN 2015; 60: 592-97) provides some very practical information regarding elevated liver enzymes in the setting of inflammatory bowel disease (IBD).  Because there are some serious liver diseases associated with IBD and due to the potential for liver toxicity from many of the medications, bumps in liver enzymes need to be carefully considered.

This retrospective study with 514 patients indicates that 77% of these elevations are transient. Table 1 lists the definitions (chronicity, severity) and patterns that were analyzed.  Transient elevations were broken down into brief (<30 days), prolonged <180 days, chronic >180 days and either intermittent or continuously abnormal. The three types were the following:

  • Hepatic: elevated ALT and/or AST; normal alkaline phosphatase (AP), GGT, and direct bilirubin (DB)
  • Cholestatic: elevated AP, GGT, and/or DB; normal ALT and AST
  • Mixed

Severity or degree was classified as follows:

  • 1 –peak liver enzyme 0-1 x ULN
  • 2 –peak liver enzyme >1-2 x ULN
  • 3 –peak liver enzyme >2-4 x ULN
  • 4 –peak liver enzyme >4 x ULN

Key findings:

  • 219 of 514 patients had 1 or more episode of abnormal liver enzymes; five patients with preexisting liver disease were excluded from the analysis.
  • Of 214 patients (152 with Crohn’s disease [CD], 62 with Ulcerative colitis [UC]) with abnormalities, 69% had a hepatitic pattern, 8% had a cholestatic pattern, and 23% had a mixed pattern. There was no association between the pattern and the final diagnosis (eg. idiopathic vs defined etiology)
  • Only 128 had adequate data to assess chronicity.  In this group, 77% had transient elevations (CD 75%, UC 80%)
  • 87% of elevations were considered idiopathic.  65% of patients with idiopathic elevation had levels < 2 times ULN.
  • Among patients with levels <2 times ULN, 95.3% had an idiopathic etiology.
  • Among patients with levels >4 times ULN, 63% had a benign idiopathic etiology
  • Figure 1 provides a pie chart of diagnoses.  Among the 12.6% with a specific etiology for elevated liver tests, drug toxicity was the most common reason: 51.9% were considered due to 6-MP therapy, 3.7% due to methotrexate, 3.7% due to acetaminophen.
  • Other identified causes among the 12.6% with a defined etiology included NAFLD in 11.1%, infections (CMV,EBV, Histoplasmosis) in 14.8%, cholelithiasis in 3.7%, autoimmune hepatitis in 3.7%, primary sclerosing cholangitis/overlap in 3.7%, and vascular malformation in 3.7%.

As with any retrospective study, there are a number of limitations, especially underdiagnosis given a lack of uniform approach to evaluation.  That being said, all patients had a minimum follow-up of at least nine months and most patients with prolonged liver enzyme elevation would have been examined closely.

Bottomline: This study provides reassurance that liver enzyme elevations are common in children with IBD, occurring in >40% of patients over 3 years at this center; most often these elevations are benign and transient.

Related blog posts:

Shwachman-Diamond Syndrome

A recent study provides an update on the variable clinical presentation of Shwachman-Diamond syndrome (SDS) (J Pediatr 2014; 163: 866-70).

SDS, an autosomal recessive disorder, is characterized by exocrine pancreatic dysfunction, bone marrow dysfunction, and predisposition to myelodysplasia/leukemia.  It is due to a mutation in the SBDS gene located on chromosome 7q11 (found in ~90% of classically presenting cases of SDS.

Using a North American Registry, the authors reviewed the records of 37 patients (all who have had mutations in SBDS gene). Key findings:

  • Neutropenia was evident at presentation in 30/37 (81%)
  • Only 51% had both neutropenia and steatorrhea at presentation
  • 24/37 (65%) had congenital anomalies: 3 with ventricular septal defects, 1 malrotation, 1 imperforate anus, 9 thoracic dystrophy (rib abnormalities), 2 with short arms/legs, 4 with metaphyseal dysplasia, 1 with Chiari malformation (type 1), 2 with testicular atrophy.
  • Medical comorbidities included elevated LFTs in 15, adrenal insufficiency in 1, hypopituitarism in 1, type 1 diabetes in 1, hypothyroid in 1, and eczema in 11.

Take-home message: Normal pancreatic imaging studies and normal fecal elastase do not rule out SDS. In addition, frequently there are associated anomalies and comorbidities.

Screening for NAFLD

As noted in previous blogs (see below), there is not a consensus with regard to screening for NAFLD in overweight and obese patients.  While some have argued for aggressive screening leading to an expensive tiered evaluation, other experts have been reluctant to endorse this approach, in part due to the magnitude of the problem and due to the perceived lack of therapeutic options.  Weighing in on this controversy is a new study (Aliment Pharm Ther 2013; DOI: 10.1111/apt.12518, link -from Jeff Schwimmer’s twitter feed: http://t.co/q1CUKBJtVo).

In the study’s introduction, the prevalence of NAFLD, estimated to be 9.6% of all children aged 2-19 years, along with society guidelines are reviewed.

The authors examined information from the clinical evaluation of 347 children (>10 years) [overweight (7%)/obese (93%)] who were referred by their primary care physician due to either an elevated ALT or suspected NAFLD.  Referral was at the discretion of the primary care attending and not based on a specific ALT value.  Median age was 13.5 years and 64% were boys.

1st tier: Subsequently, all patients underwent hepatic panel, GGT, CBC/diff, and coagulation studies.  2nd tier: If abnormal, the next set of labs may have included any or all of the following: studies for hepatitis infection (HAV, HBV, HCV), HIV, alpha-1-antitrypsin, ANA, anti-smooth muscle antibody, anti-liver kidney microsomal antibody, quantitative IgG, ceruloplasmin,  24-h urinary copper, tissue transglutaminase antibody, serum IgA, serum amino acids, urine organic acids, serum acylcarnitine, creatine kinase, ESR, CRP, and thyroid studies. (The authors did not evaluate iron status.) 3rd tier: Then, if evidence of chronic liver disease, patients were offered a liver biopsy under general anesthesia.


  • 21% did not have significant liver disease (after 1st tier).  Also, 3 liver biopsies were normal.
  • 94% of 273 with evidence of chronic liver disease underwent liver biopsy; no significant complications were noted, though a small percentage had some discomfort.
  • Ultimately, 55% were determined to have NAFLD (75% of those who underwent liver biopsy.
  • The authors report that 61 patients who had a liver biopsy had another liver disease, including autoimmune hepatitis in 11, celiac disease in 4, sclerosing cholangitis in 1, and drug-induced in 6.
  • Advanced fibrosis was noted overall in 11% (38 of 347) and in 17% of those with NAFLD.  Those with advanced fibrosis were more likely to have higher aminotransferases (eg. ALT 120 U/L compared with 82 U/L), higher GGT, and higher ceruloplasmin; however, there was significant overlap.
  • Approximately half of NAFLD patients had steatohepatitis.

Take-home message: while this article does not resolve the issue of whether screening overweight/obese children is the best strategy, it does provide useful information in those with elevated liver tests.  Careful investigation for treatable causes (and possibly nontreatable) of liver disease is worthwhile in those with sustained abnormalities in transaminases.  At a minimum, tests for autoimmune hepatitis, celiac disease, viral hepatitis, and Wilson’s disease should be at the top of the list.

Related blog entries:

TODAY is worrisome for a lot of tomorrows

The TODAY study (NEJM 2012; 366: 2247-56 and editorial 2315-16) =Treatment Options for Type 2 Diabetes in Adolescents and Youth.

While the study has a catchy acronym, the findings are disturbing.  Eligible patients (n=699) were 10 to 17 years old were followed on average over 3.86 years; they were divided into three groups:

  • Metformin 1000mg BID –48% achieved primary outcome (glycated hemoglobin <8% for at least 6 months).
  • Metformin with lifestyle changes –53% achieved primary outcome.  The lifestyle counseling that patients received in the study likely exceeded the typical counseling that most patients receive in clinical practice.
  • Metformin with rosiglitazone (4mg BID) –61% achieved primary outcome.  While this group had the best glycemic response, this group also had the greatest increase in BMI.

Other findings:

Comorbid conditions were common:

  • Hypertension: at baseline in 81 (11.6%) and new cases during study 155 (22.2%)
  • Dyslipidemia (LDL): at baseline in 23 (3.3%) and new cases during study 49 (7%)
  • Triglyceridemia: at baseline in 127 (18.2%) and new cases during study 70 (10%)
  • Microalbuminurina: at baseline in 44 (6.3%) and new cases during study 72 (10.3%)

Frequent adverse events noted with medications (Table 2 in study): gastrointestinal symptoms noted in about half of all study participants in each group, rash noted in about 40%, and elevated LFTs in about 40%.

Take home messages (borrowed from editorial):

“Most youth with type 2 diabetes will require multiple oral agents or insulin therapy within a few years after diagnosis”

“Fifty years ago, children did not avoid obesity by making healthy choices; they simply lived in an environment that provided fewer calories and included more physical activity.”

“Public-policy approaches–sufficient economic incentives to produce and purchase healthy foods and to build safe environments that require physical movement…will be necessary to stem the epidemic of type 2 diabetes and its associated morbidity.”

Related posts:

Treating diabetes with surgery

Cardiovascular disease for the entire family

Staggering cost of obesity

Lower leptin with physical activity