Controlled attenuation parameter (CAP): the clinical value based on MRI-PDFF in children with obesity

Objectives: Controlled attenuation parameter (CAP) is a noninvasive and quantitative method to evaluate hepatic steatosis, which is not well evaluated in children. The aim of this study was to examine the diagnostic value of CAP for hepatic steatosis in children with obesity based on MR proton density fat fraction (PDFF). Methods: About 108 pediatric patients with nonalcoholic fatty liver disease (NAFLD) who were assessed for PDFF, CAP, and other laboratory results were enrolled. In this study, pediatric patients were separated for the obese group (n=80) and the severe obese group (n=28). Hepatic steatosis grades (0 – 3) were classi ﬁ ed according to PDFF using cuto ﬀ values of 6.4 , 17.4, and 22.1 %. Results: There are signi ﬁ cant di ﬀ erences in CAP between the obese and severe obese groups (p<0.05). CAP showed a good correlation with PDFF in pediatric patients with NAFLD for diagnosing hepatic steatosis using a cuto ﬀ value of 265 dB/m (p<0.001). Meanwhile, ALT signi ﬁ cantly outperforms CAP in receiver-operating curve (ROC) analysis for diagnosing hepatic steatosis grades. The diagnostic accuracy of CAP for steatosis is 77.8 %, and the diagnostic accuracy of ALT for steatosis is 83.3 %. Conclusions: While CAP holds promise as a diagnostic tool for pediatric NAFLD, its diagnostic performance warrants some caution. The potential of CAP is evident; however, ALT emerges as a simpler and more accurate measure for detecting hepatic steatosis in children. Further research is essential to determine the optimal role of CAP in pediatric NAFLD diagnosis and management.


Introduction
Nonalcoholic fatty liver disease (NAFLD) is one of the most common causes in the field of liver diseases, and its prevalence has been increasing in the past few decades [1][2][3][4].NAFLD refers to a group of liver diseases that occur in the absence of significant alcohol consumption.NAFLD exhibits a wide spectrum of illness varies from simple steatosis to nonalcoholic steatohepatitis (NASH), finally leading to liver cirrhosis and hepatocellular carcinoma [5,6].Obese children with NAFLD [7,8] show different degrees of hepatic steatosis and are more prone to develop metabolic syndromes and visceral obesity.There are studies also suggested that chronic liver diseases can be induced by NAFLD [9].
Magnetic resonance imaging proton density fat fraction (MRI-PDFF) is an accurate and high-sensitive method for the measuring steatosis level throughout the liver, which shows the potential to evaluate NAFLD level [10][11][12].However, using of PDFF is still limited in clinical application due to the need of the specialist and cost.
Previous studies in this field have shown assessing liver steatosis using CAP [20,21].However, the lack of optimal cutoff values [22] and technical limitation [23,24] for children hepatic steatosis remain unsolved problems for pediatric patients.In conjugation with our ongoing interest in the field of NAFLD, we present herein the diagnostic performance of CAP for assessing hepatic steatosis in pediatric patients based on PDFF.HS-063).Pediatric patients under aged 18 years who met the Classification Standards for Overweight and Obesity Screening Body Mass Index Values in Chinese School-Age Children and Adolescents were included in this study from July 2020 to April 2022.Patients who had clinical or laboratory evidence of a liver diagnosis other than NAFLD (e.g., glycogen storage disease, drug, or virus) or alcohol consumption have been excluded.We also have detailed laboratory results including liver function, including measurement of alanine aminotransferase (ALT), aspartate aminotransferase (AST), γ-glutamyltransferase (GGT), and alkaline phosphatase (ALP) levels.Total cholesterol (TC), triglycerides (TG), high-density lipoprotein (HDL), low-density lipoprotein (LDL), fasting glucose, blood insulin, albumin, globulin uric acid, serum calcium, micro-calcium, blood 25-hydroxyvitamin D, retinol blinding protein (RBP), hemoglobin A 1c (HbA 1c ), hemoglobin A1 (HbA1), direct bilirubin (DBIL), indirect bilirubin (IBIL), and total protein (TP).

Liver MRI including PDFF
All MR scans were performed by using a standardized protocol [25] on clinical MRI system (Magnetom Avanto, Siemens, Germany), has a 1.5-T scanner.PDFF measurements were performed by placing regions of interest (ROIs) of hemiliver in three contiguous images, and then use the average value.Pediatric patients' steatosis grades were separated according to the established PDFF cutoff values for diagnosing histological steatosis grades 1 (S1), 2 (S2), and 3 (S3) [26].

TE and CAP
By using FibroScan ® (FibroScan 502 touch, Echosens, Paris, France), controlled attenuation parameter (CAP) and liver stiffness measurement (LSM) value can be measured simultaneously.For all patients in this study, only a 3.5 MHz standard M probe was used because other probes, such as XL probe, are not available for CAP measurement at present.The final CAP was calculated as the median of each value obtained from at least 10 measurements and expressed in dB/m.Obtaining no value after 10 measurements was judged as measurement failure.

Statistical analyses
Statistical analyses were performed using the SPSS software package (IBM SPSS Statistics version 21; IBM Corp., Armonk, NY, USA) and MedCalc software package (version 18.2.1;MedCalc Software, Ostend, Belgium).Analysis of variance (ANOVA) was used to compare results between the different groups.Receiver-operating curve (ROC) and area under the curve (AUC) evaluation of CAP diagnostic performance.

Results
The characteristics and laboratory results of all study patients are showed in Table 1.A total of 108 patients (M:F=65:43) with a mean age of 11.3±1.8years (7-16 years) were included in this study.Among the included patients, 28 patients were classified as the severe obese group, and the remaining 80 patients belonged to the obese group.
We used one-way ANOVA to conduct a mean comparison of clinical data (gender, age, CAP, LSM, ALT, AST, BMI, uric acid, fasting glucose, total cholesterol, triglycerides, HDL, LDL, insulin, and blood 25-hydroxyvitamin D) between two groups.
There was no significant difference in age or gender between the obese and severe obese groups.The insulin (200.6±126.5 pmol/L vs. 114.6±59.0pmol/L, p<0.001) were significantly higher in the severe obese groups compared to the obese group.There were significant difference in CAP, LSM, PDFF, height, weight, waist, & hip perimeter between the obese and severe obese groups (p<0.05).Other laboratory results have significantly difference between these two groups were waist perimeter (97.1±9.2 cm vs. 85.3±8.8 cm, p<0.001) and hip perimeter (105.0±8.2 cm vs. 93.5±8.5 cm, p<0.001).Expect for the laboratory results mentioned above, the other results were not significantly different between the two groups (Table 1).
We used one-way ANOVA to conduct a comparison of clinical data (gender, age, CAP, LSM, ALT, AST, BMI, uric acid, fasting glucose, total cholesterol, triglycerides, HDL, LDL, insulin, and blood 25-hydroxyvitamin D) among different groups and found that there were significant differences in CAP, ALT, and AST between the S2 and S3 steatosis groups (p<0.05), as shown in Table 2.

Discussion
NAFLD has considerable overlap with obesity, which has been recognized as an important health problem for pediatric patients [27,28].Since there is no standard therapy for NAFLD, early risk stratification for disease progression is considered as an important component of patient management [29,30].In this retrospective study, we compared data form pediatric NAFLD cases using PDFF and CAP techniques.Based on ROC analysis for diagnosing hepatic steatosis grades using established PDFF cutoff values mentioned above, we suggest a CAP cutoff value of 265 dB/m for the presence of steatosis in Chinese children with obesity.In that way, CAP measurements may help diagnosis as well as follow-up for steatosis in local children.
MRI-based hepatic fat quantification is used for monitor status of pediatric patients before and after treatment [31,32].In several papers, PDFF has shown an excellent correlation with hepatic steatosis in both adult and pediatric NAFLD patients due to high diagnostic efficacy [33].However, liver MRI including PDFF in children may require additional examination time and cost, which is a realistic barrier for limited medical resources.Therefore, more convenient and cheaper diagnostic tests are still required for screening and monitoring patients with NAFLD, especially for children.
In recent studies, CAP has shown great potential in the field of distinguish the presence of steatosis in adult chronic liver disease [34][35][36].In 2016, Desai's study [20] has demonstrated the ability of CAP to detect steatosis, although differentiation among histopathologic grade of steatosis was not successful due to the relatively small number of patients and fair overlap.In that study, the suggested cutoff value of CAP for predicting steatosis was 225 dB/m with 87 % sensitivity, 83 % specificity, and an AUC of 0.93, which is comparable to the data derived from our ROC analysis.In another study in 2017, assessing CAP compared with ultrasound grading in children, a cutoff point of 249 dB/m for predicting steatosis was identified with a sensitivity of 72 % and a specificity of 98-100 % [13].
In the study of Shin et al., the suggested cutoff value of CAP for predicting steatosis was 241 dB/m with 98.7 %   sensitivity, 80 % specificity, and an AUC of 0.941, which is comparable to the cutoff of 265 dB/m derived from our ROC analysis [22].
Based on Yang's previous report [37] form our hospital, form which mentioned the diagnostic cut-off value for CAP in the S2 group (n=4, 297.50 dB/m) was higher than that in the S3 group (n=13, 288.0 dB/m).As a matter of fact, due to the extremely low capacity, we thought the optimal cutoff value from this paper is less convincing and might lead to a wrong result.Actually, we continue to collect the data form the patients to make it statistically significant and gave a more reliable diagnostic cut-off value for CAP in obese children with NAFLD (n=19, 299 dB/m for S2 group and n=22, 303 dB/m for S3 group, respectively).
However, based on diagnostic performance of CAP, we found correlation between CAP and PDFF is not excellent, which does not meet the standard requirements for a diagnostic test (Table 3).Meanwhile, the diagnostic performance of ALT are superior compared to CAP (Table 4), which made ALT a more suitable index for measures steatosis.To the best of our knowledge, there have been no reports concerning the    correlation between ALT and PDFF [38,39].In our study, ALT showed an excellent correlation with PDFF in pediatric patients with NAFLD, which demonstrated 83.7 % sensitivity and 81.8 % specificity for diagnosing of hepatic steatosis using a cutoff value of 19 U/L (area under the curve=0.896,p<0.001).
As far as we know, this study also has limitations.First, this was a retrospective single-center study that included patients suspected with NAFLD; thus, our results may have been affected by selection bias in many ways.Second, the relatively small number of patients, especially low proportion of the S2 group (n=19), might be relatively small to determine the optimal cutoff point, resulting in an uncertain cutoff value.At last, CAP values were compared to MRI-PDFF not liver biopsy is the gold standard for diagnosis of NAFLD and NASH.

Conclusions
Our findings suggest that CAP exhibits diagnostic potential for NAFLD in Chinese children with obesity.CAP demonstrates the ability to distinguish between the presence and absence of hepatic steatosis using a cutoff value of 265 dB/m, with a diagnostic accuracy of 77.8 % in pediatric patients.However, while CAP shows acceptable discriminatory power in NAFLD diagnosis, ALT emerges as a more accessible alternative with comparable performance.

Figure 1 :
Figure 1: Comparison of steatosis groups using CAP values in each steatosis group based on PDFF of whole group.

Figure 2 :
Figure 2: Comparison of steatosis groups using CAP values in each steatosis group based on PDFF divided by gender.

Table  :
Patient characteristics including comparison between patients with obese group (BMI smaller than  % of the th percentile) and severe obese group (BMI greater than  % of the th percentile).

Table  :
Patient characteristics according to steatosis grades based on MR proton density fat fraction.LDL, low-density lipoprotein; RBP, retinol blinding protein; HbA c , hemoglobin A c ; HbA, hemoglobin A; DBIL, direct bilirubin; IBIL, indirect bilirubin; TP, total protein.a Values that are statistically significant (p-value<.).

Table  :
Diagnostic performance of CAP for hepatic steatosis grades (S-S).