Secondary Logo

Journal Logo

Original Articles: Hepatology

Performance Characteristics, Intra- and Inter-operator Agreement of Transient Elastography in Pediatric Nonalcoholic Fatty Liver Disease

Mandelia, Chetan∗,‡; Kabbany, Mohammad Nasser; Worley, Sarah; Conjeevaram Selvakumar, Praveen Kumar∗,§

Author Information
Journal of Pediatric Gastroenterology and Nutrition: March 2021 - Volume 72 - Issue 3 - p 430-435
doi: 10.1097/MPG.0000000000002991

Abstract

An infographic is available for this article at:https://links.lww.com/MPG/C50.

What Is Known/What Is New

What Is Known

  • Transient elastography is accurate for assessment of hepatic fibrosis in children and adults.
  • Transient elastography has shown good reproducibility in adults with nonalcoholic fatty liver disease.

What Is New

  • Transient elastography has low failure and high reliability in children with nonalcoholic fatty liver disease.
  • Failure is strongly associated with obesity.
  • Transient elastography has high intra- and inter-operator reproducibility in children with nonalcoholic fatty liver disease, with an inverse correlation with obesity.

Nonalcoholic fatty liver disease (NAFLD) encompasses a broad clinicopathological spectrum ranging from accumulation of fat in the liver called simple steatosis to varying degrees of necroinflammation called nonalcoholic steatohepatitis (NASH) leading to fibrosis and eventually cirrhosis (1). Corresponding to the increasing epidemic of childhood obesity, NAFLD has become the most common cause of chronic liver disease in children and adolescents (2,3). In addition, the prevalence of pediatric NAFLD has more than doubled over the past 3 decades and is estimated to affect 1 in 10 children in the United States and more than a third of obese pediatric patients (4,5).

The presence and severity of liver fibrosis is the most important prognostic factor in long-term morbidity and mortality in adults with NAFLD (6). Liver biopsy remains the criterion standard for assessing the stage of hepatic fibrosis; however, it is an invasive and costly procedure with risk of major complications, sampling error and intra/inter-observer variability (7). Over the past 2 decades, significant research efforts have led to the development of several serologic and imaging methods for noninvasive assessment of hepatic fibrosis and only a few of those have been validated in pediatric patients with NAFLD (8).

Transient elastography (TE) using FibroScan apparatus (EchoSens, Paris, France) is an ultrasound-based technology that estimates liver stiffness measurement (LSM) as a surrogate for hepatic fibrosis. In this noninvasive point-of-care test, an ultrasonic transducer is placed in a right intercostal space and creates a shear wave, which propagates through the right lobe of the liver. The velocity of propagation is directly related to tissue stiffness; the harder the tissue (as in hepatic fibrosis) the faster the shear wave propagates. The decrease in the amplitude of ultrasound signal, measured as controlled attenuation parameter (CAP), depends on the viscosity of liver and correlates with degree of steatosis (9). Studies in children with various chronic liver diseases have shown that TE can be reliably used for noninvasive assessment of hepatic fibrosis and steatosis (10,11).

TE has demonstrated good accuracy for detection and staging of hepatic steatosis and fibrosis in adult patients with NAFLD (12,13). A large multicenter prospective study by NASH Clinical Research Network (NASH CRN) in adults with NAFLD found that TE had low rates of failed and unreliable examinations and high reproducibility (14). Although relatively well studied and used in adult patients, data with regards to performance characteristics of Fibroscan in pediatric NAFLD are limited. Nobili et al (15) evaluated Fibroscan in 52 Italian children with biopsy-proven NAFLD and found it to be accurate and highly reproducible. The aim of the present study was to assess the performance characteristics (proportion of failed and unreliable measurements), intra- and inter-operator reproducibility of Fibroscan for simultaneous assessment of hepatic steatosis and fibrosis in children with NAFLD and to evaluate the correlation of demographic, laboratory and anthropometric parameters with performance and reproducibility of Fibroscan.

METHODS

Study Design and Participants

We performed a prospective, observational cohort study in children (<18 years) with overweight/obesity (BMI ≥85th percentile) and NAFLD who attended the Metabolic liver disease clinic at Cleveland Clinic Children's and underwent Fibroscan as part of their routine clinical care. NAFLD diagnosis was based on the presence of steatosis on ultrasound and/or liver biopsy and/or ALT ≥2× upper limit of normal (44 U/L for girls and 52 U/L for boys) after excluding other causes of elevation in liver enzymes, such as viral hepatitis, autoimmune hepatitis, alpha-1-antitrypsin deficiency, Wilson disease and medications inducing steatosis (16).

The study was approved by the institutional review board at Cleveland Clinic (CCF IRB# 18–714). Parent or legal guardian of all participants provided written informed consent before enrollment and assent was obtained from all children. No employee of Echosens was involved with any stage of the study. Data were collected and stored securely using REDCap electronic data capture tool available at Cleveland Clinic.

Transient Elastography Technique and Expertise

All participants presented for the outpatient visit after at least 3 hours of fasting and underwent anthropometric measurements and routine laboratory studies. TE was performed using the Fibroscan 502 Touch device with M+ or XL+ probes with automatic probe selection software. The patient was placed in supine position, right arm extended and placed under the head, right leg crossed over left leg and remained still during the procedure. The probe was placed in a right intercostal space in the mid-axillary line at the level of xiphoid process and readings were attempted until 10 valid measurements were obtained and the final result was obtained as a median of 10 measurements. All studies were started with M+ probe, with the XL+ probe used if valid measurements could not be obtained with the M+ probe or when prompted by the automatic probe selection tool.

Three consecutive Fibroscan examinations were performed on all patients during the same visit- twice by a single expert operator (P.C.S.) and once by a different novice operator (C.M.). The expert operator (staff physician) had performed more than 50 Fibroscan examinations before this study whereas the novice (fellow in training) had no prior clinical experience with Fibroscan. Both operators had undergone standardized training by Echosens and were certified before conducting TE for the current study.

Definitions

The weight categories according to age-specific and sex-specific BMI percentiles that we used in the current study were as follows: over- weight (≥85th to <95th percentile), class I obesity (≥95th percentile to <120% of the 95th percentile), class II or severe obesity (≥120% to <140% of the 95th percentile), and class III or markedly severe obesity (≥140% of the 95th percentile) (17). For TE, failure was defined as the inability to obtain 10 valid measurements and an examination was considered unreliable if LSM interquartile range (IQR) exceeded 30% of the median (M) [IQR/M > 0.30] (14).

Statistical Analysis

Data were described using median with quartiles and ranges, or means and standard deviations for continuous variables, and counts and percentages for categorical variables. Intra- and inter-operator agreement of TE measures were assessed using concordance correlation coefficients (CCC) with 95% confidence intervals. As described by Altman, CCC was interpreted as: 0.81 to 1 as almost perfect, 0.61 to 0.80 as substantial, 0.41 to 0.60 as moderate, 0.21 to 0.40 as fair, 0 to 0.20 as slight, and <0 as no agreement (18). It was calculated that at a sample size of 46 subjects, we will have at least 80% power to detect that CCC of 0.90 is significantly different than CCC of 0.80 with a significance criteria of 0.05. Scatter plots and Bland-Altman plots were used to visually describe agreement between the 2 Fibroscan readings for both intra- and inter-operator agreement, and linear regression models for difference between readings versus the order in which the readings were performed to assess change in agreement over time. The associations between expert LSM and CAP values, the absolute value of the difference between the expert-1 and novice measures of LSM and CAP values, and lab values and anthropometrics were assessed using nonparametric Spearman correlation coefficients. Reliable and unreliable/failed scans (as determined by the expert's first measure) were compared with demographic and clinical characteristics using Wilcoxon rank sum tests for continuous and ordinal characteristics and chi-square or Fisher exact tests for categorical characteristics. All tests were 2-tailed and performed at a significance level of 0.05. SAS 9.4 software (SAS Institute, Cary, NC) was used for all analyses.

RESULTS

Patient Demographics, Anthropometric and Laboratory Parameters

Fifty-one patients with a median age of 15 years (range: 9–17) were prospectively recruited for the study. Approximately two-thirds (34/51) were boys and the cohort was predominantly white (39/51) and non-Hispanic (38/51). The median BMI percentile was 99.1, 14/51 had class II obesity, and 18/51 had class III obesity. All the included patients had evidence of hepatic steatosis on ultrasound with median ALT of 35 U/L. In fact, only 12/34 (35.3%) boys and 3/17 (17.6%) women met the criteria for positive NAFLD screening based on NASPGHAN guidelines of ALT ≥2× upper limit of normal (44 U/L for girls and 52 U/L for boys) (16). Ten patients had undergone liver biopsy with majority (7/10) showing some degree of fibrosis (F1 in 2, F2 in 1, and F3 in 4 patients) (SDC Table 1, https://links.lww.com/MPG/C49).

Performance Characteristics of Transient Elastography

The failure rate for TE for obtaining LSM and CAP was 10% (5/51) of which readings were unreliable in only 1 patient (2%), resulting in valid and reliable readings for 88% (45/51) patients based on expert's first and second measurements (Table 1). The novice had a failure rate of 12% (6/51) and obtained valid and reliable readings in 84% (43/51) patients, with no statistically significant difference in rate of failed and reliable measurements between the expert and novice (P = 0.56). Indices of severity of obesity (BMI, BMI percentile, and BMI relative percentile) were significantly higher in patients who had failed/unreliable readings compared with those with valid/reliable measurements (BMI median 33.1 [22.9, 49.8] vs 46.2 [37.0, 68.4] kg/m2, P = 0.002 and BMI relative percentile median 124 [89, 188] vs 162 [142, 251], P = 0.004, Table 2). There were no significant differences in included demographic or laboratory parameters between the groups. In addition, all 6 patients with failed/unreliable readings had class III obesity. XL+ probe was used in 61% (31/51) patients by the expert and 59% (30/51) by the novice.

TABLE 1 - Performance characteristics of transient elastography
Expert-1 (N = 51) Expert-2 (N = 51) Novice (N = 51)
Factor N Statistics N Statistics N Statistics
Valid reading, n (%) 51 46 (90) 51 46 (90) 51 45 (88)
Reliable reading, n (%) 46 45 (98) 46 45 (98) 45 43 (96)
LSM (kPa), Median [Q1, Q3] 45 6.4 [5.6, 7.8] 45 6.3 [5.2, 7.3] 43 7.4 [5.9, 10.0]
CAP (dB/m), Median [Q1, Q3] 45 313 [277, 351] 45 314 [280, 337] 43 293 [255, 325]
LSM = liver stiffness measurement.

TABLE 2 - Factors affecting the performance of transient elastography
Failed/unreliable readings (N = 6) Valid/reliable readings (N = 45)
Factor N Statistics N Statistics P value
Demographics
 Age (years) 6 16 (12, 17) 45 15 (9, 17) 0.40
 Ethnicity 6 45 0.99
  Hispanic or Latino 1 (8) 12 (92)
  Middle Eastern or Arabic 0 (0) 1 (100)
  Not Hispanic or Latino 5 (14) 32 (86)
 Race 6 45 0.30
  Black or African American 1 (100) 0 (0)
  White 4 (10) 35 (90)
  Asian 0 (0) 2 (100)
  More than 1 race 0 (0) 1 (100)
  Unknown/not reported 1 (13) 7 (88)
Anthropometrics
 BMI (kg/m2) 6 46.2 (37.0, 68.4) 45 33.1 (22.9, 49.8) 0.002
 BMI percentile 6 99.7 (99.4, 99.9) 45 99.0 (89.7, 99.9) 0.005
 BMI relative percentile (% 95th percentile) 6 162 (142, 251) 45 124 (89, 188) 0.004
Laboratory parameters
 AST (U/L) 6 29 (20, 46) 45 29 (14, 144) 0.82
 ALT (U/L) 6 28 (17, 112) 45 35 (8, 264) 0.36
 Alkaline phosphatase (U/L) 6 90 (76, 186) 44 161 (54, 551) 0.10
 GGT (U/L) 3 23 (23, 33) 23 34 (14, 205) 0.31
 Total Bilirubin (mg/dL) 6 0.3 (0.2, 0.6) 44 0.4 (0.2, 1.0) 0.43
 Platelet count (k/μL) 6 259 (202, 298) 42 299 (175, 471) 0.082
Statistics presented as median (min, max), N (%). BMI = body mass index.
P-values: Wilcoxon Rank Sum test.
Fisher's Exact test.

Reproducibility of Transient Elastography

The mean difference between repeat LSM measurements by the expert was 0.4 kPa (SD 1.1) with almost perfect intra-operator agreement (CCC = 0.85, 95% CI: 0.74--0.92) (Table 3A). The mean difference in repeat LSM measurements by 2 different operators was 0.9 kPa (SD 1.6) with substantial inter-operator agreement (CCC = 0.76, 95% CI: 0.59--0.86). The majority of patients (23/41) had a difference of 1.0 kPa or less between LSM obtained by the expert and novice. Similarly, CAP measurements had substantial intra-operator agreement (CCC = 0.73, 95% CI: 0.56--0.84); however, inter-operator agreement was only moderate (CCC = 0.58, 95% CI: 0.34--0.74). On stratified analysis based on severity of obesity, the intra- and inter-operator CCC values for both LSM and CAP were higher for patients with class I obesity compared with those with severe obesity (classes II and III) (Table 4b). Bland-Altman plots of the difference between the first and second readings of LSM and CAP values against the mean of the 2 measurements are shown for same and different (expert-1 and novice) operators (SDC Figure 1A--D, https://links.lww.com/MPG/C48). The 95% limit of agreement for repeated LSM measurements ranged from −1.8 to + 2.5 kPa for the same operator and from −3.9 to +2.2 kPa for the different operators. Significant disagreement (>95% limits of agreement) between repeat measurements by the same operator was seen in 4% (2/45) for LSM and 2% (1/45) for CAP. For measurements by the 2 different operators, significant disagreement was seen in 2% (1/41) for LSM and 7% (3/41) for CAP.

TABLE 3 - Inter- and intra-operator agreement using the concordance correlation coefficient
Variable 1 Variable 2 N Variable 1, mean (SD) Variable 2, mean (SD) Difference, mean (SD) Concordance correlation coefficient (95% CI)
LSM expert-1 LSM expert-2 45 7.0 (2.2) 6.6 (2.1) −0.4 (1.1) 0.85 (0.74--0.92)
LSM expert-1 LSM novice 41 6.9 (2.3) 7.8 (2.7) 0.9 (1.6) 0.76 (0.59--0.86)
CAP value expert-1 CAP value expert-2 45 312.3 (47.2) 307.6 (46.4) −4.7 (33.7) 0.73 (0.56--0.84)
CAP value expert-1 CAP value novice 41 313.0 (48.6) 291.4 (46.7) −21.6 (40.4) 0.58 (0.34--0.74)
95% CI = 95% confidence interval; CAP = controlled attenuation parameter; LSM = liver stiffness measurement; SD = standard deviation.

TABLE 4 - Inter- and intra-operator agreement using the concordance correlation coefficient stratified according to severity of obesity
Variable 1 Variable 2 Class I obesity Class II and III obesity
N Concordance correlation (95% CI) N Concordance correlation (95% CI)
LSM expert-1 LSM expert-2 19 0.93 (0.82--0.97) 26 0.75 (0.52--0.87)
LSM expert-1 LSM novice 18 0.82 (0.57--0.93) 23 0.70 (0.44--0.85)
CAP value expert-1 CAP value expert-2 19 0.70 (0.40--0.87) 26 0.66 (0.38--0.83)
CAP value expert-1 CAP value novice 18 0.66 (0.29--0.85) 23 0.38 (0.03--0.64)
95% CI = 95% confidence interval; CAP = controlled attenuation parameter; LSM = liver stiffness measurement.

LSM and CAP values showed moderate positive correlation with measures of obesity, AST, and ALT (Table 5). Similarly, absolute value of difference between repeat measurements of LSM obtained by the 2 different operators showed positive correlation with BMI percentile and CAP value and a trend towards positive correlation with LSM (Table 5). Absolute value of difference between CAP obtained by the 2 operators, however, did not show statistically significant correlation with any clinical, laboratory or TE variables. Furthermore, the order of examinations by date was not associated with the value of differences between repeat measurements of LSM or CAP by the different operators (SDC Figure 1 E-F, https://links.lww.com/MPG/C48), indicating no changes in reproducibility with increasing experience of the novice.

TABLE 5 - Univariable associations between transient elastography and different variables
Factor N Spearman's correlation (95% CI) P value
CAP (expert-1)
 BMI 45 0.38 (0.09--0.60) 0.010
 BMI percentile 45 0.39 (0.11--0.62) 0.007
 BMI relative percentile (% 95th percentile) 45 0.35 (0.07--0.59) 0.016
 ALT 45 0.44 (0.17--0.65) 0.002
 AST 45 0.44 (0.17--0.65) 0.002
LSM (expert-1)
 BMI 45 0.33 (0.05--0.57) 0.024
 BMI percentile 45 0.44 (0.17--0.65) 0.002
 BMI relative percentile (% 95th percentile) 45 0.38 (0.10--0.61) 0.009
 ALT 45 0.45 (0.18--0.65) 0.002
 AST 45 0.32 (0.03--0.56) 0.029
LSM absolute difference (expert-1 and novice)
 CAP value expert-1 41 0.34 (0.04--0.59) 0.027
 LSM expert-1 41 0.26 (−0.05 to 0.53) 0.098
 BMI percentile 41 0.35 (0.04--0.59) 0.026
 BMI relative percentile (% 95th percentile) 41 0.28 (–0.03--0.54) 0.081
BMI = body mass index; CAP = controlled attenuation parameter; LSM = liver stiffness measurement.

DISCUSSION

The main findings of our study are: valid and reliable measurements can be obtained for LSM and CAP using TE in the majority of children with NAFLD regardless of the operator fexperience; Failed/unreliable readings and inter-operator disagreement are associated with severity of obesity; LSM measurements with TE are highly reproducible between the same and different operators; and reproducibility of CAP measurements with TE is high with the same operator but relatively lower between different operators.

Concomitant with the increasing epidemic of NAFLD among children (4) with subsequent comorbidities and need for liver transplantation in adults, several noninvasive tests for assessment of hepatic steatosis and fibrosis have been developed to assess the severity and progression of the disease (19). Noninvasive indices based on routine laboratory studies developed for assessment of hepatic fibrosis in adults with NAFLD have shown suboptimal performance in children limiting their use in clinical practice (20). Among the imaging modalities, ultrasound is most commonly used in clinical practice but has been shown to be inaccurate in detecting and quantifying steatosis and fibrosis (21). In contrast, MRI-proton density fat fraction (PDFF) and MR elastography were shown to accurately identify and quantify steatosis and fibrosis in pediatric NAFLD (22–25). MR-based imaging techniques although most accurate, are expensive, have limited availability and require sedation in young children limiting their use. On the other hand, TE is easy to use, quick, and provides point of care results to the physician, leading to its increasing use in clinical practice. However, only a handful of studies have evaluated TE in pediatric patients with a variety of chronic liver diseases and have found it to be highly accurate (26).

Our study shows that TE can be successfully deployed for simultaneous assessment of hepatic steatosis and fibrosis in children with NAFLD. Valid and reliable readings were obtained in a vast majority of patients, with failed and unreliable measurements limited to those with markedly severe (Class III) obesity. Our findings are consistent with those of Nobili et al (15) who evaluated TE in 52 children with NAFLD and reported failure in 2 patients with BMI >35 kg/m2. The failure rate of 10% seen in our study is comparable to the failure rate of 11% (4/37) reported by Fitzpatrick et al (27) in children with NAFLD. Moreover, the novice had similar rates of successful measurements indicating that TE can be successfully performed by inexperienced operators after receiving standardized training. Similar to the findings of Tapper et al (28), severity of obesity was significantly associated with failed/unreliable measurements. TE in more than half of our patients relied on the use of the XL+ probe, which allows for measurement of shear wave velocity at a greater skin to liver capsule depth of 3.5 cm. Before the development of XL+ probe, failure rates in excess of 25% were reported with median BMI of 36.7 kg/m2 for patients with failed examination (28). In contrast, the median BMI for failed/unreliable examination in our study was 46.2 kg/m2 with the use of XL+ probe.

The LSM cut-offs for various stages of fibrosis have not been well established in children. In the largest study to date evaluating diagnostic accuracy of TE in children with biopsy-proven NAFLD, Nobili et al reported that LSM values <5, <7, and <9 kPa suggest absence of any fibrosis (≥1), significant fibrosis (≥2), and advanced fibrosis (≥3), respectively (15). Using these parameters, our cohort consisted predominantly of children with minimal fibrosis (F0-F1) and about a third (34%) of patients had significant fibrosis (≥F2), which is consistent with what has been reported previously in pediatric NAFLD (29). In addition, XL+ probe has been shown to generate lower LSM values compared with the M+ probe in adults with NAFLD but this has not been studied in children and unified interpretation has been advocated (30). Although CAP has shown excellent diagnostic value in assessing absence or presence of steatosis, studies correlating CAP measurements with grades of hepatic steatosis in children are limited with varying cut-offs based on methodology, such as correlation with biopsy versus MRI-PDFF (31,32). Extrapolating CAP measurements from previous studies, all the children in our study had steatosis and in fact majority of patients had higher grades of steatosis correlating with inclusion of larger number of children with severe obesity.

We found that TE is highly reproducible with substantial inter-operator agreement in LSM values regardless of the experience of the operator. In contrast, inter-operator agreement for CAP is lower probably related to the body habitus as two-third of the included children had severe obesity similar to previous studies (31,33). Individual measurements had a mean difference of 0.4 kPa and 4.7 dB/m between repeat measurements of LSM and CAP, respectively by the same operator. The difference in LSM and CAP values was slightly higher when repeated by a different operator and this may have been because of selection of different sites for measurement, which has been previously shown to affect inter-operator reproducibility of TE in children (34). The clinical relevance of these variations is minimal as TE values are often used in conjunction with other variables, such as ALT to select patients who are candidates for more aggressive evaluation and intervention strategies. Again, the inter-operator variation in LSM showed a positive association with BMI and CAP suggesting that the reproducibility may be related to the severity of obesity and consequent hepatic steatosis. Nobili et al had previously reported excellent inter-observer reproducibility for TE in 31 children with NAFLD with an intraclass correlation coefficient of 0.96 (90% confidence interval: 0.92–0.97). Although transient makings on the skin from the initial examination may have influenced the selection of site for the repeat measurements in our study, Nobili et al (15) used an ultrasound-guided skin mark to select the site of probe placement, which likely eliminated the sampling error completely, something which is not done in routine clinical practice. A large multicenter NASH-CRN study in adults with NAFLD found that TE was highly reproducible with median difference of 0.0 [−0.9, 1.1] kPa and 0 [−16, 19] dB/m between repeat measurements of LSM and CAP, with a vast majority repeated by the same operator (14). Similar results were also reported by a recent study, which evaluated reproducibility of TE in 235 healthy volunteer children (BMI<85th percentile) and found mean difference of 0.044−kPa (SD 0.4) between paired LSM measurements with CCC of 0.85 (95% CI: 0.82--0.88) (35).

We recognize that our study is subject to a few limitations. Majority of the included patients did not have a liver biopsy, which prevented us from studying the diagnostic accuracy of TE and correlation of histological parameters with performance characteristics and reproducibility. Our patients were recruited from a dedicated pediatric fatty liver clinic at a tertiary care center where TE was performed by 2 pediatric gastroenterologists. Therefore, the accuracy and reproducibility of TE found in this study should be reevaluated in community centers and also in settings where TE is performed by other healthcare personnel, such as radiologists, nurse practitioners, physician assistants, nurses, and do forth. In addition, our study is limited by small sample size and the study cohort may not be representative of the general population. Nevertheless, this study provides encouraging real-world experience on performance characteristics, intra- and inter-operator reproducibility of transient elastography in children with NAFLD.

CONCLUSIONS

In summary, the findings of our study show that TE can be successfully performed in an outpatient clinic for simultaneous assessment of hepatic steatosis and fibrosis in pediatric NAFLD with real-time results. Severe obesity can limit the performance and reproducibility of TE. Regardless, with lack of noninvasive markers, practical limitations of liver biopsy and limited availability of MRI-based techniques, TE could be a valuable tool in assessment of severity and progression of pediatric NAFLD.

REFERENCES

1. Brown GT, Kleiner DE. Histopathology of nonalcoholic fatty liver disease and nonalcoholic steatohepatitis. Metabolism 2016; 65:1080–1086.
2. Loomba R, Sirlin CB, Schwimmer JB, et al. Advances in pediatric nonalcoholic fatty liver disease. Hepatology 2009; 50:1282–1293.
3. Schwimmer JB, Deutsch R, Kahen T, et al. Prevalence of fatty liver in children and adolescents. Pediatrics 2006; 118:1388–1393.
4. Welsh JA, Karpen S, Vos MB. Increasing prevalence of nonalcoholic fatty liver disease among united states adolescents. J Pediatr 2013; 162:496–500.
5. Anderson EL, Howe LD, Jones HE, et al. The prevalence of non-alcoholic fatty liver disease in children and adolescents: a systematic review and meta-analysis. PLoS One 2015; 10:e0140908.
6. Taylor RS, Taylor RJ, Bayliss S, et al. Association between fibrosis stage and outcomes of patients with nonalcoholic fatty liver disease: a systematic review and meta-analysis. Gastroenterology 2020; 158:1611.e12–1625.e12.
7. Dezsõfi A, Baumann U, Dhawan A, et al. ESPGHAN Hepatology Committee. Liver biopsy in children: Position paper of the ESPGHAN hepatology committee. J Pediatr Gastroenterol Nutr 2015; 60:408–420.
8. Mandelia C, Kabbany MN, Conjeevaram Selvakumar PK, et al. The search for noninvasive methods to identify liver fibrosis in children with nonalcoholic fatty liver disease. Biomark Med 2018; 12.:265–273.
9. Mikolasevic I, Orlic L, Franjic N, et al. Transient elastography (FibroScan() with controlled attenuation parameter in the assessment of liver steatosis and fibrosis in patients with nonalcoholic fatty liver disease - where do we stand? World J Gastroenterol 2016; 22:7236–7251.
10. Lee CK, Mitchell PD, Raza R, et al. Validation of transient elastography cut points to assess advanced liver fibrosis in children and young adults: the Boston Children's Hospital Experience. J Pediatr 2018; 198:84.e2–89.e2.
11. Ferraioli G, Calcaterra V, Lissandrin R, et al. Noninvasive assessment of liver steatosis in children: the clinical value of controlled attenuation parameter. BMC Gastroenterol 2017; 17:61.
12. Eddowes PJ, Sasso M, Allison M, et al. Accuracy of FibroScan controlled attenuation parameter and liver stiffness measurement in assessing steatosis and fibrosis in patients with nonalcoholic fatty liver disease. Gastroenterology 2019; 156:1717–1730.
13. Siddiqui MS, Vuppalanchi R, Van Natta ML, et al. NASH Clinical Research Network. Vibration-controlled transient elastography to assess fibrosis and steatosis in patients with nonalcoholic fatty liver disease. Clin Gastroenterol Hepatol 2019; 17:156.e2–163.e2.
14. Vuppalanchi R, Siddiqui MS, Van Natta ML, et al. NASH Clinical Research Network. Performance characteristics of vibration-controlled transient elastography for evaluation of nonalcoholic fatty liver disease. Hepatology 2018; 67:134–144.
15. Nobili V, Vizzutti F, Arena U, et al. Accuracy and reproducibility of transient elastography for the diagnosis of fibrosis in pediatric nonalcoholic steatohepatitis. Hepatology 2008; 48:442–448.
16. Vos MB, Abrams SH, Barlow SE, et al. NASPGHAN Clinical Practice Guideline for the Diagnosis and Treatment of Nonalcoholic Fatty Liver Disease in Children: recommendations from the Expert Committee on NAFLD (ECON) and the North American Society of Pediatric Gastroenterology, Hepatology and Nu. J Pediatr Gastroenterol Nutr 2017; 64:319–334.
17. Skinner AC, Perrin EM, Moss LA, et al. Cardiometabolic risks and severity of obesity in children and young adults. N Engl J Med 2015; 373:1307–1317.
18. Altman D. Practical Statistics for Medical Research. London, UK: Chapman & Hall; 1991.
19. Castera L, Friedrich-Rust M, Loomba R. Noninvasive assessment of liver disease in patients with nonalcoholic fatty liver disease. Gastroenterology 2019; 156:1264.e4–1281.e4.
20. Mansoor S, Yerian L, Kohli R, et al. The evaluation of hepatic fibrosis scores in children with nonalcoholic fatty liver disease. Dig Dis Sci 2015; 60:1440–1447.
21. Awai HI, Newton KP, Sirlin CB, et al. Evidence and recommendations for imaging liver fat in children, based on systematic review. Clin Gastroenterol Hepatol 2014; 12:765–773.
22. Mouzaki M, Trout AT, Arce-Clachar AC, et al. Assessment of nonalcoholic fatty liver disease progression in children using magnetic resonance imaging. J Pediatr 2018; 201:86–92.
23. Xanthakos SA, Trout AT, Dillman JR. Magnetic resonance elastography assessment of fibrosis in children with NAFLD: promising but not perfect. Hepatology 2017; 66:1373–1376.
24. Schwimmer JB, Behling C, Angeles JE, et al. Magnetic resonance elastography measured shear stiffness as a biomarker of fibrosis in pediatric nonalcoholic fatty liver disease. Hepatology 2017; 66:1474–1485.
25. Middleton MS, Van Natta ML, Heba ER, et al. Diagnostic accuracy of magnetic resonance imaging hepatic proton density fat fraction in pediatric nonalcoholic fatty liver disease. Hepatology 2018; 67:858–872.
26. Hwang JY, Yoon HM, Kim JR, et al. Diagnostic performance of transient elastography for liver fibrosis in children: A systematic review and meta-analysis. Am J Roentgenol 2018; 211:W257–W266.
27. Fitzpatrick E, Quaglia A, Vimalesvaran S, et al. Transient elastography is a useful noninvasive tool for the evaluation of fibrosis in paediatric chronic liver disease. J Pediatr Gastroenterol Nutr 2013; 56:72–76.
28. Tapper EB, Challies T, Nasser I, et al. The performance of vibration controlled transient elastography in a US cohort of patients with nonalcoholic fatty liver disease. Am J Gastroenterol 2016; 111:677–684.
29. Mann JP, De Vito R, Mosca A, et al. Portal inflammation is independently associated with fibrosis and metabolic syndrome in pediatric nonalcoholic fatty liver disease. Hepatology 2016; 63:745–753.
30. Wong VWS, Irles M, Wong GLH, et al. Unified interpretation of liver stiffness measurement by M and XL probes in non-alcoholic fatty liver disease. Gut 2019; 68:2057–2064.
31. Shin J, Kim MJ, Shin HJ, et al. Quick assessment with controlled attenuation parameter for hepatic steatosis in children based on MRI-PDFF as the gold standard. BMC Pediatr 2019; 19:112.
32. Desai NK, Harney S, Raza R, et al. Comparison of controlled attenuation parameter and liver biopsy to assess hepatic steatosis in pediatric patients. J Pediatr 2016; 173:160.e1–164.e1.
33. de Lédinghen V, Wong GLH, Vergniol J, et al. Controlled attenuation parameter for the diagnosis of steatosis in non-alcoholic fatty liver disease. J Gastroenterol Hepatol 2016; 31:848–855.
34. Goldschmidt I, Streckenbach C, Dingemann C, et al. Application and limitations of transient liver elastography in children. J Pediatr Gastroenterol Nutr 2013; 57:109–113.
35. Rowland M, Mcgee A, Broderick A, et al. Cystic Fibrosis Registry of Ireland, Cystic Fibrosis Liver Disease Research Group. Repeatability of transient elastography in children. Pediatr Res 2020; 88:587–592.
Keywords:

fibroscan; fibrosis; pediatric nonalcoholic fatty liver disease; steatosis; vibration-controlled transient elastography

Supplemental Digital Content

Copyright © 2020 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition