Secondary Logo

Journal Logo

Original Articles: Hepatology

Magnetic Resonance Imaging: A Personalized Approach to Understanding Fatty Liver Disease

Barnekow, Kris; Shyken, Paige; Ito, Joy; Deng, Jie; Mohammad, Saeed; Fishbein, Mark

Author Information
Journal of Pediatric Gastroenterology and Nutrition: June 2019 - Volume 68 - Issue 6 - p 777-781
doi: 10.1097/MPG.0000000000002316


What Is Known

  • Fatty liver is the most common liver disease in the United States and affects mostly obese children.
  • Magnetic resonance imaging provides an accurate and rapid methodology to quantify hepatic fat content in pediatric nonalcoholic fatty liver disease.

What Is New

  • Personalized images of fatty liver with demonstrable fat accumulation represent an underutilized adjunct for patient education.

Nonalcoholic fatty liver disease (NAFLD) is a leading cause of liver disease in children who are obese or are diagnosed with type 2 diabetes mellitus (1). The prevalence of NAFLD in adolescents has risen significantly over the past 20 years and currently this condition is estimated to affect approximately 11% of adolescents in the United States (2). The annual predicted economic costs of this disease are staggering, with an estimated national cost of $103 billion for therapeutic interventions (3). Lifestyle modifications that include gradual, sustained weight loss using a calorically appropriate diet, combined with increases in physical activity are strongly recommended as best practice (4). Enrolling obese children in weight loss programs and maintaining their adherence to sustained weight management interventions depends strongly on parent engagement (4). Family-centered comprehensive information will also enhance behavior change (4).

Failure of lifestyle/nutrition intervention may be due in part to parents and their children not adequately understanding the consequences of their actions and inactions. To gain awareness of NAFLD in affected obese individuals, we have begun to utilize the affected individual's personalized magnetic resonance (MR) image as an education tool. These images provide dramatic qualitative and quantitative visual effects of hepatic fat accumulation. The effect of adding the personalized image as an adjunct to patient education is unknown but may increase parental (and child) understanding. Increased compliance is a targeted benefit and is based on Information Processing Theory. Information Processing Theory is a cognitive learning perspective that focuses on the process by which people receive, interpret, store and act upon information that is retained in their memory. Information Processing Theory informs thinking about patient learning by illuminating errors and distortions in thinking. This perspective postulates that people need to attend, process, and remember information before they can act on the information they have learned (5).

The primary aim of this investigation is to introduce personalized images of fatty liver as an adjunct to patient education and to determine the level of comprehension achieved pertaining directly to the image itself (fat accumulation) and indirectly to NAFLD (cause, sequela, treatment). Secondary aims of the study were to examine effects of demographics, and health literacy as contributing factors to outcomes.


Study Design

The design of this study is a single-center, short-term information retention test-retest pilot study.

Study Procedures

Study participants were recruited from a pool of parents/caretakers of children attending a fatty liver clinic. Inclusion criteria included parents/caretakers of children with elevated alanine aminotransferase (ALT) and a hepatic fat fraction of greater than 18% fat (indicative of severe steatosis) (6) as determined by MRI (normal range 5% or less). In each instance, MRI was performed in conjunction with clinical evaluation. Exclusion criteria included other identifiable causes of elevated ALT including infectious hepatitis, Wilson's disease, alpha-1-antitrypsin deficiency, and autoimmune hepatitis. Demographics included child sex, race, and ethnicity. Child's age, BMI, and serum ALT were collected at study entry and at follow-up (whenever available).

Study Materials

An MRI template containing a personalized liver image (Supplemental Digital Content 1, and Newest Vital Sign (Supplemental Digital Content 2, was distributed in English and Spanish versions to accommodate all study participants. In addition to the personalized liver image depicting fat content, the MRI template contains an adjacent image of a normal liver with the abnormal fat fraction compared with normal. It also describes nonvisualized features of nonalcoholic steatohepatitis (NASH), including inflammation and fibrosis, and potential long-term outcome (cirrhosis). The template also indicates the beneficial role of weight loss in NASH. The Newest Vital Sign is designed to quickly and simply assess an individual's health literacy skills. The tool is validated and is completed in as little as 3 minutes (7).

Magnetic Resonance Imaging methodology

All MRI examinations were performed on either a 1.5 or 3.0 T scanner (MAGNETOM Aera and Skyra, Siemens Healthcare, Erlangen, Germany) using an anterior body matrix and posterior spine coil. Multi-echo spoiled gradient echo images were acquired with 6 echo time (TE) of 2.3, 4.6, 6.9, 9.2, 11.5 and 13.8 ms, corresponding to consecutive out-of-phase (OP) and in-phase (IP) images. The signal magnitude oscillations arising from the in- and out-of-phase signal interference were used to derive proton density (ρ) and T2 relaxation time of water and fat components using a nonlinear least-squares curve fitting method (MATLAB, MathWorks, Natick, MA) (8,9). A multi-fat-peak MR spectrum was used in the signal model, in which a composite fat signal with 5 fat peaks located at the resonance frequency of 0.9, 1.3, 2.1, 4.2, and 5.3 ppm, respectively (10). Proton density (PD) fat fraction (FF) was calculated as 100% × ρF/(ρF + ρW), where ρW and ρF are proton density of water and fat component. Acquired MR images and resultant water, fat, and FF images are shown (Fig. 1). The gray level FF images are color-coded to help visualize contrasts with white representing 0% and black representing 110% on the color bar. Subcutaneous fat with >90% of fat shows as dark red. Normal liver with less than 5% of fat appears as light pink. Fat accumulation within the liver is represented by a corresponding increase in yellow intensity (Fig. 2). Color-coded FF color maps of 2 representative patients are shown (Fig. 2).

Acquired in-phase (IP) and out-of-phase (OP) gradient echo images used to derive separated water and fat images, and fat fraction (FF) parametric map.
Color-coded MRI hepatic fat fraction (FF) maps of 2 subjects over time. Subject 1: steatosis progression (trend to yellowish appearance) is demonstrated in an individual with rising BMI and corresponding worsening FF (15.9%–22.7%). Subject 2: steatosis regression (trend to pinkish appearance) is demonstrated in an individual with diminishing BMI and corresponding reduced FF (19.8%–7.3%).

Study Entry

Approval for the investigation was obtained from the Institutional Review Board at Ann & Robert H. Lurie Children's Hospital of Chicago. All study participants were enrolled at follow-up visit to fatty liver clinic, having already been diagnosed with fatty liver disease based on MRI result and exclusion of other liver disorders. Eligible participants, parents or primary caretakers of children with fatty liver, were consented for the investigation. Each study participant was then provided with their child's personalized liver image. A nurse clinician reviewed the contents of the template with the parent/caretaker and upon completion, questioned them to assess their level of comprehension. The Newest Vital Sign literacy test followed, and the number of correct responses recorded. A score of 0 to 1 suggests high likelihood (50% or more) of limited literacy. A score of 2 to 3 indicates the possibility of limited literacy. A score of 4 to 6 almost always indicates adequate literacy. Session were approximately 15 minutes.

Telephone Follow-up

Study participants were contacted by telephone 2 to 4 weeks after study entry. Recall was assessed by repeating the same queries used to assess comprehension of the personalized liver image initially.


Paired t-test analysis was performed for BMI and serum ALT at enrollment and subsequent clinic follow-up whenever available.


Twenty-four parents/caretakers of children with elevated ALT and MRI evidence of fatty liver with fat fraction (FF) greater than 18% were enrolled (21M, 3F) in the investigation. Race/ethnicity included Hispanic (n = 21), Caucasian (n = 2), and Asian (n = 1). Primary languages were English (n = 9) and Spanish (n = 15). The age range of the children was 9 to 18 years. BMI ranged from 27 to 51 kg/m2 and initial serum ALT ranged from 28 to 245 IU/L.

Parents/caretakers demonstrated excellent understanding (100% correct responses) of the variation of appearance of normal liver (pink) and their child's fatty liver (yellow) (Table 1). Their recall also appeared quite satisfactory for the appearance of normal liver (95%) and fatty liver (81%). The fat content was correctly stated initially by all subjects, but recall was hindered (52%). Nonvisualized features of NASH (including inflammation and fibrosis) were not identified or recalled as well as fat accumulation. Cirrhosis as a possible long-term outcome of NASH was correctly identified in 83% of subjects but recalled in only 57%. Weight loss was generally understood (96%) and recalled (81%) as the preferred means to improve fatty liver.

Assessment of Parental Knowledge about Non-alcoholic Fatty Liver Disease

Health literacy appeared compromised, particularly, in individuals with Spanish as their primary language (Supplemental Digital Content 3, Approximately one of 3 of study participants possessed a high likelihood of low literacy, 1 of 3 of subjects possessed possibility of limited literacy, and 1 of 3 of subjects had adequate literacy. Over 80% of subjects with Spanish as their primary language had a high likelihood or possibility of limited literacy and 40% had a high likelihood of limited literacy.

BMI and serum ALT were unaltered in the children of study participants over a follow-up interval of 3 to 22 months (Supplemental Digital Content 4A and 4B,,


In this study, we utilized Information Processing Theory to identify sensory attributes of stimuli (images of liver disease) that increased attention to the images and the processing of information related to NAFLD (Fig. 3). We measured storage of information through pre- and posttesting of knowledge.

Information Processing Theory applied to personalized visual images. The application of Information Processing Theory is depicted when utilizing personalized visual images as a means of learning and retaining information about NAFLD. NAFLD = nonalcoholic fatty liver disease.

The findings of this study revealed that parents/caretakers understood the variation of appearance of normal liver (pink) and their child's fatty liver (yellow). Parents recalled the appearance of normal liver and fatty liver. The recollection of fat content, however, fell over a few weeks. Three additional areas where initial comprehension was higher than recall include nonvisualized features of NASH (inflammation and fibrosis), cirrhosis as a possible long-term outcome of NASH, and weight loss to improve disease processes related to fatty liver.

Information Processing Theory (5) explains this level of parental recall using personalized visual images of their child's liver. Visualization of the fatty liver presented to the parents/caregivers may have evoked emotional or physical responses that elicited arousal and attention. The disparity between visual and nonvisual responses may be due of decreased attention and processing of the information that was invoked initially from the images but ultimately lacked the sustaining emotional or physical responses experienced by the parents.

In addition, we also measured the parents’/caregivers’ health literacy using the English and Spanish versions of the Newest Vital Sign (7). These findings, however, were not straightforward. The parents/caregivers who were Spanish-speaking had a higher possibility and likelihood of limited health literacy. Even though 2 of 3 of the parents/caregivers were at risk for low health literacy, there was excellent or satisfactory retention of the questions related to variation of appearance of normal liver and their child's fatty liver and the appearance of normal liver and fatty liver in general. The high retention of this information may not readily be explained by the lower health literacy levels as measured by the Newest Vital Sign. Information Processing Theory seems to explain the parents’/caregivers’ ability to retain the information over an extended period that was directly related to the personalized visual images. It is possible, though, that the responses to the questions related to nonvisualized features of NASH were related to the parents’/caregivers’ health literacy levels. That is, their initial understanding of the information was slightly lower and was not retained over time. This suggests that limited health literacy may be a mediating factor in understanding information initially and subsequent recall when there is not a personal connection to the information.

Anecdotally, no change was observed in mean BMI or mean ALT in study participants who were seen from 3 to 22 months after diagnosis. This study, however, was not designed to investigate whether using medical images to aid patent/caretaker understanding at time of NAFLD diagnosis improves long-term information retention, is effective in helping to persuade children with NAFLD and their families about possible beneficial lifestyle changes, or results in long-term improvement of NAFLD. Nonetheless, as short-term recall was higher in this study for information presented with medical images, inclusion of in-person medical images reviews should be considered in future studies. Future studies will need to identify if additional interventions may facilitate a concomitant change in behavior or lifestyle choices that result in weight loss or improved serum ALT that was not evident in this investigation.


Although this study represents preliminary findings about retention of information after viewing MRI images of fatty liver disease, it does not attempt to explain or understand interventions that may lead to lifestyle choices that result in healthy weight management. Additional limitations include a lack of discussion of possible role of time intervals and a control group, possible sources of bias, and a need for better understanding of what constitutes poor, adequate, good, excellent, and so forth, retention in different categories of intervention. Another limitation of this pilot data is that it does not allow disentangling of 4 factors: personal meeting versus group meeting (possibly) versus no extra meeting; presentation of results with versus without images; meeting with physician versus nurse versus both; presenting general information about NAFLD versus presenting that and individual MRI results. Another limitation is that no assessment was made on whether the intervention worked, in any sense—that is, what is needed is not only just a test of whether information was retained but also whether the communication of that information was in any sense persuasive. Behavior change, such as monitoring food intake and increasing activity levels in pediatric populations, is incumbent upon understanding, but is also related to elements of readiness to change that are multivariate and complex. In pediatric care, interventions that may lead to behavior change related to weight management should consider the family's motivation about treatment, positive reinforcement, the family centeredness of the interventions, and accessibility to healthy food and activity opportunities (4). Given that the sample in this study included a diverse group of participants, a better understanding of the social determinants of health including culture and socioeconomic factors may provide critical information when designing interventions that change behaviors to decrease obesity in children. In conclusion, we have demonstrated that personalized images of fatty liver were effective tools for enhancing parental/caregiver comprehension of NAFLD regardless of health literacy. Repeat imaging is feasible and may be desirable in individuals with successful weight management and likely improvement in their condition.


1. Schwimmer JB, Deutsch R, Kahen T, et al. Prevalence of fatty liver in children and adolescents. Pediatrics 2006; 118:1388–1393.
2. Welsh JA, Karpen S. Vos MB Increasing prevalence of nonalcoholic fatty liver disease among United States adolescents, 1988–1994 to 2007–2010. J Pediatr 2013; 162:496.e1–500.e1.
3. Younossi ZM, Blissett D, Blissett R, et al. The economic and clinical burden of nonalcoholic fatty liver disease in the United States and Europe. Hepatology 2016; 64:1577–1586.
4. Perez AJ, Kebbe M, Holt NL, et al. Parent recommendations to enhance enrollment in multidisciplinary clinical care for pediatric weight management. J Pediatr 2018; 192:122–129.
5. Braungart MBR, Gramet P. Bastable SB. Applying learning theories to healthcare practice. Health Professional as Educator: Principles of Teaching and Learning. Sudbury, MA: Jones & Bartlett Learning; 2011. 55–101.
6. Fishbein MH, Miner M, Mogren C, et al. The spectrum of fatty liver in obese children and the relationship of serum aminotransferases to severity of steatosis. J Pediatr Gastroenterol Nutr 2003; 36:54–61.
7. Weiss BD, Mays MZ, Martz W, et al. Quick assessment of literacy in primary care: the newest vital sign. Ann Fam Med 2005; 3:514–522.
8. Yu H, Shimakawa A, McKenzie CA, et al. Multiecho water-fat separation and simultaneous R2∗ estimation with multifrequency fat spectrum modeling. Magn Reson Med 2008; 60:1122–1134.
9. Zhong XN, Kannengiesser SAR, Dale BM, et al. Fat and iron quantification using a multi-step adaptive fitting approach with multi-echo GRE imaging. 21st Annual Meeting of ISMRM. Salt Lake City, Utah, USA; 2013.
10. Yokoo T, Shiehmorteza M, Hamilton G, et al. Estimation of hepatic proton-density fat fraction by using MR imaging at 3.0 T. Radiology 2011; 258:749–759.

health literacy; hepatic steatosis; magnetic resonance imaging; nonalcoholic; nonalcoholic fatty liver disease; obesity

Supplemental Digital Content

Copyright © 2019 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition