Analysis of risk factors for fatty liver disease in children with Wilson’s disease

Abstract

Background and aims

Many children with Wilson’s disease are complicated with dyslipidemia. The aim of this study was to investigate the risk factors for the development of fatty liver disease (FLD) in children with Wilson’s disease.

Methods

We evaluated sex, age, weight, the disease course, treatment course, clinical classification, alanine transaminase (ALT), aspartate transaminase, γ-glutamyl transpeptidase, total biliary acid, triglyceride, total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, homocysteine, uric acid, fibrinogen (FBG), creatinine, procollagen III N-terminal propeptide, laminin, hyaluronic acid, type IV collagen, and performed receiver operating characteristic curve analysis to investigate the forecast value of individual biochemical predictors and combined predictive indicators to evaluate FLD in Wilson’s disease.

Results

The multivariate logistic regression analysis revealed that ALT [odds ratio (OR), 1.011; 95% confidence interval (CI), 1.004–1.02; P = 0.006], uric acid (OR, 1.01; 95% CI, 1.002–1.018; P = 0.017), FBG (OR, 3.668; 95% CI, 1.145–13.71; P = 0.037), creatinine (OR, 0.872; 95% CI, 0.81–0.925; P < 0.001), and laminin (OR, 1.01; 95% CI, 1.002–1.018; P = 0.017) acted as independent risk factors in Wilson’s disease complicated with FLD. The receiver operating characteristic curves for combined predictive indicators demonstrated an area under the curve values of 0.872, which was found to be a significant predictors for FLD in Wilson’s disease.

Conclusions

We screened out the most important risk factors, namely ALT, uric acid, creatinine, FBG, and laminin for Wilson’s disease complicated with FLD. The joint prediction achieved is crucial for identifying children with Wilson’s disease complicated with FLD.

Introduction

Wilson’s disease is an autosomal recessive disease that is caused by a mutation in the ATPase copper transporting beta (ATP7B) gene. Due to the structural changes and functional damages in the ATP7B protein (copper ion transport ATPase β peptide) in Wilson’s disease, copper transport and metabolism of liver cells are affected, leading to abnormal deposition of copper in tissues and organs. The reported incidence rates of Wilson’s disease in different regions are different based on the consideration of ethnic and geographical differences. Presently, the prevalence rate of Wilson’s disease is approximately 142 cases per million (the first reported prevalence rate of Wilson’s disease was 5 cases per million) [1,2]. The common symptoms of Wilson’s disease include neurological and mental disorders, and liver damage is the most commonly observed symptom in children with Wilson’s disease [3]. In clinical settings, abnormal liver enzymes, dyslipidemia, and fatty liver are observed in the early stages, whereas liver fibrosis and cirrhosis are observed in the middle and late stages. A liver pathological biopsy is performed to reveal hepatic steatosis and cirrhosis [4]. A study in China reported that 303 of 316 children with Wilson’s disease showed abnormal liver function [5]. Ferenci et al. reported that 39.5% of children (≤18 years of age) and 58% of adults with Wilson’s disease suffered from cirrhosis at the time of onset and neurological symptoms developed after several years [6]; therefore, early identification of Wilson’s disease is crucial. Dyslipidemia or mild to moderate fatty liver is the first symptom to develop in many children with Wilson’s disease. Furthermore, children with unexplained fatty liver should be monitored for the possible development of Wilson’s disease [7].

Fatty liver disease (FLD) is categorized as alcoholic fatty liver, nonalcoholic fatty liver disease (NAFLD), and special types of fatty liver. The concept of metabolic dysfunction-associated fatty liver disease (MAFLD) was postulated by an international expert consensus group in 2020, which emphasized that FLD is associated with systemic metabolic disorders except non-alcohol factors. Hence, the term NAFLD is gradually being replaced by MAFLD [8]. In 2021, an international expert consensus group postulated a new concept of pediatric fatty liver disease (PeFLD) [9], which can be divided into PeFLD type 1 (hereditary metabolic disorder, fatty liver with recognizable systemic diseases, including gene defect, viral hepatitis, or Wilson’s disease), PeFLD type 2 (fatty liver with metabolic dysfunction, including MAFLD), and PeFLD type 3 (hepatic steatosis due to unknown causes). Fatty degeneration is observed in fatty liver due to various reasons [10]. We classified children with Wilson’s disease complicated with FLD as PeFLD1 type and postulated that fatty degeneration in Wilson’s disease liver pathology is associated with copper-induced mitochondrial dysfunction [11]. Simple fatty liver patterns, steatohepatitis-like patterns, fibrosis, and cirrhosis can be observed in the liver pathology of adults/children with Wilson’s disease [12–14]. Liver steatosis during the early stage of the disease can lead to fibrosis and cirrhosis with the progression of the disease.

In this study, we aimed to investigate the risk factors affecting the occurrence of FLD in children with Wilson’s disease. All hospitalized patients with Wilson’s disease between the ages of 3 and 18 at the Brain Disease Center of our hospital in the last 4 years were included. The screening information during the routine admission period was obtained, which included the basic data and biochemical indicators of the research object. Based on the screening of risk factors, receiver operating characteristic (ROC) curve analysis was performed to estimate the diagnostic value of the main influencing factors of fatty liver. This study provides important information for the early diagnosis of FLD in children with Wilson’s disease, thereby delaying and preventing the aggravative risk of liver damage due to Wilson’s disease.

Study subjects and methods

Study subjects

This is a retrospective cross-sectional study. We included patients with Wilson’s disease (3–18 years old) who were treated in the Encephalopathy department of the First Affiliated Hospital of Anhui University of Traditional Chinese Medicine (Anhui, China) from January 2019 to September 2023. These patients fulfilled the diagnostic criteria of hepatolenticular degeneration (Table 1), and the flowchart of diagnostic process is depicted in Fig. 1. Included patients were screened for exclusion diagnosis.

Table 1.

Leipzig criteria: diagnostic criteria for Wilson disease [16]

Typical clinical symptoms and signs		Other tests
Kayser–Fleischer rings		Liver cooper (in absence of cholestasis)
Present	2	Normal <50 μg/g (0.8 μmol/g)	−1
Absent	0	50–249 μg/g (0.8–4.0 μmol/g)	1
Neurologic symptomsa		>250 μg/g (4 μmol/g)	2
Severe	2	Rhodamine-positive granulesb	1
Mild	1	Urinary copperc
Absent	0	Normal	0
Serum ceruloplasmin		1–2 times ULNd	1
Normal (>0.2 g/l)	0	>2 times ULN	2
0.1–0.2 g/l	1	Normal, but >5 times ULN after PCAe	2
<0.1 g/l	2	Mutation analysis
Coombs-negative hemolytic anemia		Two chromosome mutations	4
Present	1	One chromosome mutation	1
Absent	0	No mutation detected	0

^aOr typical abnormalities at brain MRI.

^bIf no quantitative liver copper available.

^cIn the absence of acute hepatitis.

^dUpper limit of normal.

^eD-penicillamine.

Fig. 1.

WD diagnostic process [15]. WD, Wilson’s disease.

Exclusion criteria:

• (1) Patients with systemic endocrine diseases and metabolic diseases.

• (2) Patients with various viral/alcoholic/drug/autoimmune related liver diseases.

• (3) Patients who have taken drugs for regulating blood lipid in the recent 3 months.

Our research was approved by the Ethics Committee of the First Affiliated Hospital of Anhui University of Traditional Chinese Medicine (Anhui, China).

Measurements

In total, 224 patients were divided into two groups, namely patients with FLD (study group) and patients without FLD (control group). Basic information of all patients within 24 h of initial admission was collected. The original values were used for determining continuous variables, and the classified variables were assigned values. Finally, 23 variables were included in this study. All the data were obtained from the hospital database. The basic clinical data, such as gender, age, weight, course of disease, course of treatment, clinical classification, and laboratory biochemical indexes, including, alanine transaminase (ALT), aspartate transaminase, γ-glutamyl transpeptidase, total biliary acid, triglyceride, total cholesterol (T-CHOL), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), homocysteine, uric acid, fibrinogen, creatinine, procollagen III N-terminal propeptide, laminin, hyaluronic acid, and type IV collagen of the two groups of patients were evaluated by performing univariate analysis and multivariate logistic regression analysis. The risk factors affecting FLD development in children with Wilson’s disease were obtained, and a prediction model was constructed based on the risk factors.

Diagnostic criteria

Diagnostic criteria of Wilson’s disease (Table 1)

The diagnostic criteria (Leipzig scoring system) of the 8^th Wilson’s Disease International Conference in Leipzig in 2001 were followed based on the Guidelines of The Diagnosis and Treatment of Hepatolenticular Degeneration (2022 Edition).

If the total score is ≥4, diagnosis is established; if the score is ≤2, the diagnosis can be ruled out, and a score of 3 indicates possible diagnosis after further tests.

Diagnostic criteria of dyslipidemia [17]

The diagnostic criteria of hypercholesterolemia, hypertriglyceridemia, and low high-density lipoprotein hyperlipidemia were based on the Expert consensus on diagnosis and management of dyslipidemia in children (2022), and the diagnostic reference values were as follows: T-CHOL ≥ 5.2 mmol/l, triglyceride ≥ 1.76 mmol/l, and HDL-C ≤ 1.04 mmol/l.

Diagnostic criteria of fatty liver disease

The scope and degree of FLD were evaluated using ultrasound (or quantitative ultrasound) based on the Experts in Standardization of Diagnosis and Treatment of Fatty Liver Disease in China [18]. The criteria for disease diagnosis of FLD were based on the Guidelines of the Prevention and Treatment for Non-alcoholic Fatty Liver Disease: A 2018 update [19]. Diagnosis was made if any two of the following conditions were fulfilled: (a) diffuse enhancement of near-field echo in the liver (bright liver), and the echo of the liver is stronger than that of the kidney; (b) the pipeline structure inside the liver cannot be clearly displayed; and (c) the far-field echo of the liver decayed gradually.

Statistical analyses

Statistical analyses were performed using SPSS 26.0 (SPSS Inc., Chicago, Illinois, USA). The obtained data were tested for homogeneity of variance, and normally distributed data were expressed as the mean ± SD and analyzed by performing an independent-sample t-test. The univariate analysis was performed to filter FLD-related independent variables (P < 0.05), whereas the multivariate logistic regression analysis was performed to estimate the importance of FLD-associated affecting factors. Odds ratios and 95% confidence interval were calculated, and the results were considered statistically significant at P < 0.05. The ROC curve analysis was performed to evaluate all predictors.

Results

Baseline data and clinical features

A total of 105 children with Wilson’s disease complicated with FLD were included in the present study, and their ages ranged from 3 to 18 years, with an average of 11.61 ± 3.71 years. Among the 105 children, 76 were males (72.38%) and 29 were females (27.61%). Further, 119 patients with non-FLD were enrolled in the study as controls. Similar to the study group, the ages of the controls ranged from 3 to 18 years, with an average of 12.96 ± 3.69 years. Among the 119 children, 76 were males (72.38%) and 43 were females (27.61%). The two groups were comparable in age, sex, weight, disease course, treatment course, and clinical classification (Table 2); however, no statistical difference was observed for the general data (P > 0.05).

Table 2.

General data comparison between the FLD group and non-FLD group

	Sex	Clinical classification	Age (years)	Weight (kg)	Disease course (months)	Treatment course (months)
FLD	Male 76 (72.38%)	Liver-type 91 (86.67%)	16.61 ± 3.71	48.48 ± 20.72	13.41 ± 13.96	10.38 ± 13.06
	Female 29 (27.61%)	Mixed-type 12 (11.43%)
		Brain-type 2 (1.90%)
Non-FLD	Male 76 (63.86%)	Liver-type 110 (92.44%)	12.96 ± 3.69	51.19 ± 17.67	11.50 ± 9.97	8.73 ± 8.46
	Female 43 (36.13%)	Mixed-type 9 (7.56%)
		Brain-type 0 (0%)
t/X2	1.854	3.363	1.011	1.055	−1.191	−1.135
P	0.198	0.183	0.9516	0.292	0.235	0.258

FLD, fatty liver disease.

Clinical characteristics of Wilson’s disease children with fatty liver disease

The patients’ characteristics are summarized in Table 3.

Table 3.

Characteristics of FLD in 105 WD children

Characteristics	Value
Age, mean ± SD, years	11.6095 ± 3.7094
Sex
Male	76 (72.38%)
Female	29 (27.61%)
Clinical types
Liver	91 (86.67%)
Brain	12 (1.90%)
Mixed-type	2 (11.43%)
Hypercholesterolemia	45 (42.86%)
Hypertriglyceridemia	38 (36.19%)
Low high density lipoprotein hyperlipidemia	13 (11.38)
Abnormal transaminase (ALT/AST)	83/58 (79.05%/55.24%)
Uric acid abnormality	33 (31.43)

ALT, alanine transaminase; AST, aspartate transaminase; FLD, fatty liver disease; WD, Wilson’s disease.

Univariate analysis of influencing factors of Wilson’s disease with pediatric fatty liver disease

Univariate analysis identified 10 parameters (such as creatinine, LDL-C, fibrinogen, homocysteine, laminin, T-CHOL, triglyceride, ALT, weight, and age) at the time of enrollment that were significantly correlated with FLD (P < 0.05).

Multivariate logistic regression analysis of Wilson’s disease with pediatric fatty liver disease

According to the results of univariate analysis (Table 4 and Fig. 2), multivariate logistic regression model is further constructed. Multivariate logistic regression analysis indicated that the ALT, uric acid, creatinine, fibrinogen, and laminin were associated with FLD (P < 0.05) (Table 5).

Table 4.

Univariate analysis of influencing factors of WD with PeFLD

Parameter	B coefficients	SE	OR	95% CI	Z	P
HDL-C (mmol/l)	0.657	0.56317	1.928	1.928 (0.651–6.076)	1.166	0.244
CREA (μmol/l)**	−0.065	0.01839	0.937	0.937 (0.901–0.969)	−3.546	<0.01
LDL-C (mmol/l)*	0.663	0.32539	1.941	1.941 (1.045–3.784)	2.038	0.042
FBG (g/l)*	1.098	0.47347	2.998	2.998 (1.228–7.977)	2.319	0.02
CIV (ng/ml)	−0.005	0.00536	0.995	0.995 (0.982–1.002)	−0.972	0.331
HCY (μmol/l)*	−0.095	0.04698	0.909	0.909 (0.817–0.981)	−2.029	0.043
HA (ng/ml)	−0.001	0.0018	0.999	0.999 (0.994–1.001)	−0.758	0.448
LN (ng/ml)*	−0.008	0.00365	0.992	0.992 (0.985–0.999)	−2.103	0.035
PIIINP (ng/ml)	0.013	0.00949	1.013	1.013 (0.995–1.033)	1.406	0.16
UA (μmol/l)	0.004	0.00216	1.004	1.004 (1–1.008)	1.693	0.091
T-CHOL (mmol/l)*	0.431	0.20903	1.539	1.539 (1.04–2.375)	2.063	0.039
TG (mmol/l)*	0.408	0.18492	1.504	1.504 (1.066–2.217)	2.206	0.027
TBA (μmol/l)	−0.029	0.02566	0.972	0.972 (0.91–1.008)	−1.12	0.263
GGT (U/l)	0.001	0.00656	1.001	1.001 (0.988–1.014)	0.129	0.897
AST (U/l)	0.012	0.00629	1.012	1.012 (1–1.026)	1.889	0.059
ALT (U/l)*	0.007	0.00288	1.007	1.007 (1.002–1.013)	2.347	0.019
Disease course (months)	0.009	0.01704	1.009	1.009 (0.975–1.044)	0.527	0.598
Treatment/disease course	0.266	0.56216	1.305	1.305 (0.437–4.019)	0.474	0.636
Treatment course (months)	0.018	0.02051	1.018	1.018 (0.978–1.063)	0.877	0.38
Clinical classification	0.575	0.78881	1.778	1.778 (0.374–9.401)	0.729	0.466
Weight (kg)*	−0.023	0.01159	0.977	0.977 (0.954–0.999)	−2.023	0.043
Age (years)*	−0.149	0.05638	0.862	0.862 (0.768–0.96)	−2.641	0.008
Sex	−0.371	0.39786	0.69	0.69 (0.312–1.496)	−0.933	0.351

95% CI, 95% confidence interval; ALT, alanine transaminase; AST, aspartate transaminase; CIV, type IV collagen; CREA, creatinine; FBG, fibrinogen; FLD, fatty liver disease; GGT, γ-glutamyl transpeptidase; HA, hyaluronic acid; HCY, homocysteine; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; LN, laminin; OR, odds ratio; PeFLD, pediatric fatty liver disease; PIIINP, procollagen III N-terminal propeptide; SE, standard error; TBA, total biliary acid; T-CHOL, total cholesterol; TG, triglyceride; UA, uric acid; WD, Wilson’s disease.

^*The difference between the FLD and non-FLD groups was statistically significant (P < 0.05).

^**P < 0.05.

Fig. 2.

Forest plot of each factor. The left column lists the factors. The odds ratio for each of these studies is represented by a square, and confidence intervals are represented by horizontal lines. ALT, alanine transaminase; AST, aspartate transaminase; CIV, type IV collagen; CREA, creatinine; FBG, fibrinogen; GGT, γ-glutamyl transpeptidase; HA, hyaluronic acid; HCY, homocysteine; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; LN, laminin; OR, odds ratio; PIIINP, procollagen III N-terminal propeptide; TBA, total biliary acid; TG, triglyceride; UA, uric acid.

Table 5.

Logistic regression analysis of the risk factors for WD with FLD

Parameter	B coefficient	SE	OR	95% CI	Z	P
ALT (U/l)	0.011	0.00402	1.011	1.011 (1.004–1.02)	2.745	0.006
UA (μmol/l)	0.01	0.00404	1.01	1.01 (1.002–1.018)	2.389	0.017
CREA (μmol/l)	−0.137	0.03361	0.872	0.872 (0.81–0.925)	−4.074	<0.001
FBG (g/l)	1.3	0.62413	3.688	3.688 (1.145–13.71)	2.082	0.037
LN (ng/ml)	−0.017	0.00404	1.01	1.01 (1.002–1.018)	2.389	0.017

95% CI, 95% confidence interval; ALT, alanine transaminase; CREA, creatinine; FBG, fibrinogen; FLD, fatty liver disease; LN, laminin; OR, odds ratio; SE, standard error; UA, uric acid; WD, Wilson’s disease.

Receiver operating characteristic curves of variables

ROC is a graphical tool that evaluates the forecasting performance of one or several indicators. The ROC curves (Fig. 3) for ALT, uric acid, creatinine, fibrinogen, laminin, and combined predictive indicators (CPI) showed the area under the curve (AUC) values of 0.657, 0.705, 0.729, 0.702, and 0.740, respectively (Table 6). In the ROC graph, the closer the point is to (0, 1), the better the classification performance of the model. Based on this, the predicted value for CPI (i.e. the joint forecast of the above indicators) was better than those for other individual indicators for diagnosing Wilson’s disease with FLD.

Fig. 3.

ROC curve depicting the predictive efficacy of each predictor for FLD. Sensitivity measurements are on the y-axis, and 1 − specificity is on the x-axis. The area under the curve represents the prediction accuracy. The optimal cutoff for each predictor was defined using the SPSS software. AUC = 0.872. ALT, alanine transaminase; AUC, area under curve; CPI, combined predictive indicators; CREA, creatinine; FBG, fibrinogen; FLD, fatty liver disease; LN, laminin; ROC, receiver operating characteristic; UA, uric acid.

Table 6.

Operating characteristic curves of variables

Variables	AUC	Asymptotic significance	95% CI
			Lower limit	Upper limit
ALT (U/l)	0.686	0.000	0.617	0.755
UA (μmol/l)	0.630	0.001	0.558	0.703
CREA (μmol/l)	0.358	0.000	0.285	0.430
FBG (g/l)	0.608	0.005	0.534	0.682
LN (ng/ml)	0.437	0.103	0.362	0.512
CPI	0.872	0.000	0.776	0.873

95% CI, 95% confidence interval; ALT, alanine transaminase; AUC, area under the curve; CPI, combined predictive indicators; CREA, creatinine; FBG, fibrinogen; LN, laminin; UA, uric acid.

Discussion

The present study showed that the proportion of males in the Wilson’s disease fatty liver group was 72.38%; however, no statistical significance was observed between the study and control groups. Many studies worldwide have shown that the overall NAFLD prevalence in men is significantly higher than that in women [20,21], which may be attributed to the protective effect exerted by estrogen. Animal experiments have shown that estrogen therapy improves nonalcoholic steatohepatitis progression in mice [22]. Owing to decreasing estrogen levels in postmenopausal females, the loss of estrogen-based protection may be the reason for the increased NAFLD risk in these females [23]. Additionally, testosterone levels in men are considered a potential risk factor for FLD [24]. No difference was observed in children regarding the risk of Wilson’s disease complicated with FLD based on gender; this related to the fact that children are not sexually mature.

ALT is a crucial independent risk factor for children with Wilson’s disease complicated with FLD. The guidelines of North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition suggest that serum ALT levels should be screened to assess the NAFLD risk [25]. Studies have shown that ALT is a risk factor for NAFLD in children with obesity before late adolescence. For every unit increase in ALT levels, the NAFLD risk in children with obesity increases by 7.3% [26]. ALT is also a high-risk factor associated with children with high insulin resistance index, hypertriglyceridemia, and hypercholesterolemia besides those with FLD [27]. An Indian study showed significantly increased ALT levels in children with NAFLD [28], and another showed that enhanced liver ultrasonic echo in children with obesity was related to increased liver enzyme levels [29]. However, some researchers have presented contrasting results. For instance, Abrams et al. proposed that serum ALT levels were not an adequate screening tool to detect NAFLD [30].

Herein, we found uric acid levels as an important risk factor for Wilson’s disease complicated with FLD. A meta-analysis of 50 cases has shown that increased uric acid levels in the serum are positively correlated with NAFLD [31]. A study of 3104 children by Di Bonito et al. has shown that hyperuricemia can be a marker of FLD in adolescents [32]. Uric acid may induce liver fat accumulation via the reactive oxygen species/c-Jun N-terminal kinase/activator protein-1 pathway [33] and regulate hepatic steatosis and insulin resistance via the NOD-, LRR-, and pyrin domain-containing protein 3 inflammasome pathway [34]. However, the correlation between uric acid levels and FLD has not been determined. A study involving 27 009 samples has reported no causal relationship between the two [35]; thus, future studies are required to understand the correlation between uric acid levels and FLD. Additionally, we showed creatinine levels as an independent risk factor for FLD in children with Wilson’s disease. Previous studies have shown that the ratio of serum uric acid levels to creatinine levels is significantly associated with NAFLD and is positively correlated to the risk of moderate-to-severe FLD [36], which is an independent predictor of NAFLD [37,38]. A few studies have investigated the correlation between fibrinogen and laminin levels, a biochemical index to evaluate liver fibrosis, and fatty liver. A study has reported decreased fibrinogen levels in patients with fatty liver; however, another study has reported higher fibrinogen levels in patients with NAFLD [39].

FLD itself can be considered a risk factor for many diseases, including cancer, depression, diabetes, and polycystic ovary syndrome [40–44]. Thus, preventing FLD occurrence and actively identifying risk factors and effective predictors for FLD are of great clinical importance. A large-scale study involving 31 718 individuals in China [45] has shown that the ROC curves for age, BMI, ALT, triglyceride, and HDL showed AUC values of 0.708, 0.836, 0.767, 0.780, and 0.732, respectively, possessing predictive potential for NAFLD occurrence in adults. A study involving 1350 children in China has shown that waist circumference, waist-to-hip ratio, waist-to-height ratio, body composition index, visceral fat area, and endocrine index, with AUC values ranging from 0.69 to 0.96, can all effectively predict NAFLD [46]. Another study has shown that the waist-circumference-to-height ratio possesses predictive potential for FLD in children [47]. Zhou et al. [26] has reported ALT levels as a valuable predictor of NAFLD in children. Additionally, the ratio of uric acid levels to creatinine levels has shown to be an important predictor of NAFLD [38], which is consistent with the present findings in children with Wilson’s disease. The present study showed the significance of the combined forecasting of ALT, uric acid, creatinine, fibrinogen, and laminin. However, we did not determine the predictive potential of any individual index. The results showed that the AUC value for CPI was higher than those for any other individual biochemical predictor. Further, the study showed that the pathogenesis of Wilson’s disease fatty liver resulted from multiple factors; consequently, multi-index joint evaluation is critical for identifying children with Wilson’s disease complicated with fatty liver.

Due to the nonnecessity of invasive diagnosis and the uncooperativeness of patients, this study lacks evidence of pathological diagnosis. Liver pathology in patients with Wilson’s disease is difficult to distinguish from other causes of chronic hepatitis, and histopathological findings are not specific lesions of hepatolenticular degeneration. Studies have characterized the pathology stages of Wilson’s disease as steatosis (stage I), interface hepatitis (stage II), bridging fibrosis (stage III), and cirrhosis (stage IV) [48]. The uneven copper accumulation in liver cells and a large amount of copper is deposited in mitochondria, associated with an increased number of perossisomes, lipolysosomes, and cytoplasmic lipid droplets, which leads to the disintegration of mitochondrial membrane and the death of liver cells [49–51]. Hepatocyte dysfunction manifest as simple steatosis initially in Wilson’s disease. In some patients with mild clinical symptoms, the liver biopsy demonstrates glycogenated nuclei and hepatic steatosis (microvesicular steatosis, macrovesicular steatosis). Liver biopsy of patients with hepatolenticular degeneration showed that lipid metabolism is disordered and fat droplets are common in the cytoplasm of hepatocytes. The characteristic manifestations of liver biopsy are similar to FLD [52]. There are more serious pathological manifestations of liver biopsy in patients with decompensated cirrhosis. Hepatocellular ballooning degeneration and pericellular fibrosis can be seen in liver cirrhosis nodules [53].

The earliest characteristic alterations of the liver pathology in Wilson’s disease include steatosis. Fatty liver is the first clinical symptom of this disease. Some patients were initially diagnosed with FLD and finally diagnosed with Wilson’s disease [54]. In some aspects, there are overlapping standards between them, so they cannot be completely distinguished. Alqahtani et al. evaluated the ultrastructural changes of liver tissue in children with Wilson’s disease, nonalcoholic FLD, and autoimmune hepatitis, and found that there was extensive pathological overlap among the three diseases. Ultrastructural findings of mitochondrial abnormalities are important to distinguish Wilson’s disease from nonalcoholic FLD and autoimmune hepatitis. Mitochondrial polymorphism, crista tip expansion, membrane replication, and matrix density in Wilson’s disease group were significantly higher than those in the other two groups [55].

Nevertheless, the study has some limitations. Fatty liver diagnosed with the help of ultrasound imaging was not confirmed by a pathological biopsy. Additionally, different regions, eating habits, heights, and body mass indices can affect fatty liver; thus, these factors might have affected the present results. Hence, collecting more samples of patients and more comprehensive demographic data in the later period is necessary to validate the present findings.

Conclusion

FLD is common in children with Wilson’s disease and is classified as PeFLD type 1. Herein, we screened out the most important risk factors, namely ALT, uric acid, creatinine, fibrinogen, and laminin, for Wilson’s disease complicated with FLD. The joint prediction achieved by the mentioned five indicators is crucial for identifying children with Wilson’s disease complicated with FLD. The present data will help in the early determination and effective management of Wilson’s disease complicated with FLD, which will ultimately help in reducing cirrhosis incidence in patients with Wilson’s disease and improve their prognosis.

Acknowledgements

We acknowledge all the subjects and staff of this study, and we would like to thank the First Affiliated Hospital of Anhui University of Traditional Chinese Medicine for providing case information related to this work.

This study was supported by the National Traditional Chinese Medicine Heritage and Innovation Center Project (Development and Reform Commission, Social [2022] No. 366) and the Anhui Reserved Excellent Talent Plan (Grant numbers [2022.04]).

S.-P.J. wrote the original draft. Z.T. acquired the data. W.-M.Y. and M.-X.W. conceived the study. G.-R.Y. revised the manuscript. All authors read and approved the final manuscript. G.-R.Y. and W.-M.Y. contributed equally to this work and share first authorship.

This study was approved by the Ethics Committee of Anhui Provincial Hospital of Traditional Chinese Medicine, and the written informed consent of the guardian of the researcher was obtained.

This paper’s printing and electronic publishing have obtained the informed consent of patients.

Considering privacy and moral issues, the data collected in the current research is not open to the public. Classes are obtained from the author if necessary.

Conflicts of interest

There are no conflicts of interest.