Underreporting of Nonalcoholic Fatty Liver Disease

Underreporting of Nonalcoholic Fatty Liver Disease

Nonalcoholic fatty liver disease (NAFLD) is not uncommon in children and can be associated with hepatic fibrosis with the risk of long-term associated mortality. Thus, early detection is important in order to monitor disease activity and to provide resources to improve health outcomes. Children often undergo computed tomography (CT) of the abdomen for various reasons, and the authors of this study evaluated for incidental hepatic steatosis findings in a group of children undergoing CT for nephrolithiasis.

This retrospective, single-center study included patients younger than 18 years of age who underwent abdominal CT imaging for concern of nephrolithiasis over a 5-year period. Patients with known medical conditions that could cause steatosis such as metabolic/storage diseases, Wilson disease, autoimmune hepatitis, and viral hepatitis were excluded. Patients with asplenia also were excluded. Patient parameters including height, weight, age, and standard laboratory data were obtained from the electronic medical record. Liver and spleen parenchymal attenuation were measured with moderate-to-severe steatosis defined as a histologic fat concentration greater than 30%. This value (defined as “original criteria”) was obtained by using specific parameters of liver and spleen attenuation differences calculated as Hounsfield Units on CT. Mild steatosis was defined as histologic fat concentrations ≥ 5% and was determined using mDIXON-Quant magnetic resonance imaging to calculate Hounsfield Units (defined as “secondary criteria”). These two criteria categories were compared to patients who underwent CT for nephrolithiasis who did not have hepatic steatosis. Kappa statistics were used to determine degree of agreement of imaging findings between 2 radiologists. A total of 584 patients with appropriate inclusion criteria underwent abdominal CT for a diagnosis of nephrolithiasis during this period. Most patients were non-Hispanic females, and the median age of the patients was 14.8 years. The median body mass index (BMI) was at the 73rd percentile with 41% of patients being defined as overweight or obese. It was noted that 541 patients had no steatosis on imaging while 42 patients did have steatosis. The two CT criterium (“original” for moderate to severe; “secondary” for mild) used to determine steatosis demonstrated a prevalence rate of steatosis between 3%-35%. Kappa statistics between radiologists showed excellent correlation of findings. No significant difference in ethnicity was found between patients with or without steatosis. However, BMI percentiles and median serum alanine aminotransferase (ALT) levels were significantly higher in patients with any degree of steatosis compared to patients with no steatosis. Steatosis ranged between 6%-47% in those patients who were overweight or obese compared to 0.3%24% of patients with a normal BMI percentile. Additionally, steatosis was present in 11%-43% of patients with elevated ALT levels while steatosis was present in 0.7%-27% of patients with normal ALT levels. Finally, using a non-contrast CT liver attenuation value of less than 48 Hounsfield Units, only 12 of 42 patients (29%) had steatosis reported in the original radiology reports, and only 2 of these 12 patients had a known history of NAFLD.

This study suggests that steatosis can be found during CT imaging of the abdomen ordered for non-hepatic reasons. If steatosis is found, it is essential to have such patients be referred to pediatric gastroenterology to assist with diagnosis and treatment options to prevent the long-term complications of NAFLD.

Okura H, Yodoshi T, Thapaliya S, Trout A, Mouzaki M. Under-reporting of hepatic steatosis in children: a missed opportunity for early detection. Journal of Pediatrics 2021; 234: 92-98.

Download Tables, Images & References