Retrospective Study of Serum Vitamin D Levels and Metabolic-Associated Fatty Liver Disease in 222 Children and Adolescents Aged 6–18 Years With Obesity

Introduction

Childhood obesity is a global public health problem that has increased rapidly over recent decades. It is associated with early cardiometabolic morbidity and reduced life expectancy. The World Health Organization reports continuing increases in obesity across age groups worldwide [1]. Childhood obesity is strongly linked to insulin resistance, dyslipidemia, and hypertension [2,3].

Metabolic-associated fatty liver disease (MAFLD), previously categorized under nonalcoholic fatty liver disease (NAFLD), has emerged as the most common chronic liver condition in children and adolescents and is closely related to obesity and metabolic dysfunction [4,5]. Population-based and clinic-based studies report pediatric fatty liver prevalence ranging from approximately 7% in the general pediatric population to over 30% in children and adolescents with obesity [6–8]. Because all participants in the present cohort had obesity, MAFLD is operationally equivalent to ultrasonography-detected hepatic steatosis [4].

Pediatric MAFLD can progress to steatohepatitis and fibrosis and is associated with increased risk of type 2 diabetes and cardiovascular disease [9,10]. Accordingly, early identification of MAFLD and its metabolic correlates is important for risk stratification and patient management.

Vitamin D has pleiotropic effects beyond bone metabolism, including potential roles in insulin sensitivity, inflammation, and lipid metabolism [11,12]. Vitamin D deficiency is common in children with obesity, and observational studies have reported inconsistent findings regarding the relationship between serum 25-hydroxyvitamin D (25[OH]D) concentrations and hepatic steatosis [13–17].

Some pediatric studies have reported lower 25(OH)D levels in children with obesity and hepatic steatosis than in with those without hepatic steatosis [13,16]. In contrast, other studies did not identify an independent association after accounting for obesity-related confounders [14,15,17]. These discrepancies may reflect heterogeneity in MAFLD definitions, vitamin D assay methods and cut-offs, and imaging-based diagnosis of hepatic steatosis.

Therefore, this retrospective study aimed to evaluate serum 25(OH)D levels and MAFLD in 222 children and adolescents aged 6 to 18 years with obesity, and to examine the associations of MAFLD with metabolic risk factors and ultrasonographic steatosis grade.

Material and Methods

ETHICS STATEMENT:

The study protocol was approved by the Ordu University Non-Interventional Scientific Research Ethics Committee (initial approval: decision No. BAEK 2024/05, dated March 29, 2024). An amendment regarding the study title was subsequently approved by the same committee (amendment approval: decision No. 2025/110, dated March 21, 2025). Because this was a retrospective review of existing medical records with no additional procedures or patient contact, the requirement for written informed consent was waived by the Ethics Committee. Patient confidentiality was maintained by using de-identified data for analyses.

STUDY DESIGN AND SETTING:

This single-center, retrospective observational study reviewed consecutive children and adolescents aged 6 to 18 years who attended the pediatric endocrinology outpatient clinic for evaluation of obesity during a 12-month period. Clinical, laboratory, and imaging data were extracted from the institutional electronic medical record system.

PARTICIPANTS:

Obesity was defined as body mass index (BMI) at or above the 95th percentile for age and sex [18]. Patients were excluded if they had known chronic liver disease, viral hepatitis, autoimmune liver disease, Wilson disease, alpha-1 antitrypsin deficiency, endocrine or genetic syndromes associated with obesity, use of hepatotoxic medications, or alcohol use. Patients with incomplete ultrasonography data were excluded from analyses requiring liver imaging.

DEFINITIONS:

MAFLD was defined according to the 2020 consensus as evidence of hepatic steatosis together with overweight/obesity or metabolic dysregulation [4]. Because all participants had obesity, MAFLD was defined in this cohort as hepatic steatosis detected by abdominal ultrasonography. Vitamin D status was classified using serum 25(OH)D concentrations as follows: deficiency <12 ng/mL (30 nmol/L), insufficiency 12–20 ng/mL (30–50 nmol/L), and sufficiency >20 ng/mL (>50 nmol/L) [19].

CLINICAL AND LABORATORY MEASUREMENTS:

Age, sex, pubertal status, weight, and height were recorded at the index visit. Pubertal status was assessed according to Tanner staging, and puberty was defined as Tanner stage of 2 or higher. Weight, height, and BMI were converted to standard deviation scores (SDS) using Turkish reference standards [18]. Fasting blood samples obtained as part of routine clinical care were used to measure glucose, insulin, lipids (total cholesterol, low-density lipoprotein cholesterol [LDL-C], high-density lipoprotein cholesterol [HDL-C], and triglycerides), alanine aminotransferase (ALT), aspartate aminotransferase (AST), glycated hemoglobin (HbA1c), thyroid-stimulating hormone (TSH), and free thyroxine (FT4). The homeostasis model assessment of insulin resistance (HOMA-IR) was calculated as fasting insulin (μU/mL)×fasting glucose (mg/dL)/405 [20].

Serum 25(OH)D levels were measured using an automated electrochemiluminescence immunoassay (ECLIA; Cobas e801, Roche Diagnostics) in the hospital’s central laboratory as part of routine clinical care. Because testing was ordered at the discretion of the treating physician, 25(OH)D measurements were not available for all participants.

ULTRASONOGRAPHY AND STEATOSIS GRADING:

Abdominal ultrasonography was performed by experienced radiologists blinded to laboratory results. Hepatic steatosis was graded semi–quantitatively based on liver echogenicity compared with the renal cortex, visualization of intrahepatic vessels and the diaphragm, and posterior beam attenuation [21], as follows: grade 0 (none): normal liver echogenicity, clear visualization of the diaphragm and intrahepatic vessel borders; grade 1 (mild): slight diffuse increase in liver echogenicity, with normal visualization of the diaphragm and intrahepatic vessels; grade 2 (moderate): moderate increase in liver echogenicity, with slightly impaired visualization of the diaphragm and intrahepatic vessels; and grade 3 (severe): marked increase in liver echogenicity, with poor or no visualization of the diaphragm, intrahepatic vessels, and the posterior portion of the liver due to attenuation.

STATISTICAL ANALYSIS:

Since serum 25(OH)D measurements were not available for all participants, analyses involving vitamin D were performed on available data (complete-case analysis), and no imputation was applied for missing values.

Statistical analyses were performed using IBM SPSS Statistics (version 28; IBM Corp, Armonk, NY, USA). Continuous variables were assessed for normality using visual inspection and the Shapiro-Wilk test and are reported as mean±standard deviation or median (minimum–maximum), as appropriate. Between-group comparisons were performed using the independent samples t test or Mann-Whitney U test for continuous variables and the chi-square test for categorical variables. When expected cell counts were small in R×C contingency tables, the Fisher-Freeman-Halton exact test (Monte Carlo simulation) was used. Associations between sex and MAFLD were expressed as unadjusted odds ratios (OR) with 95% confidence intervals (CI). Comparisons of 25(OH)D across steatosis grades used Kruskal-Wallis testing.

Missing data were handled by complete-case analysis for each comparison (no imputation). All tests were 2-sided and P<0.05 was considered statistically significant.

Results

PARTICIPANT CHARACTERISTICS:

A total of 222 children and adolescents aged 6 to 18 years with obesity were included (151 females [68.0%] and 71 males [32.0%]). The mean age was 13.36±2.99 years (median 13.9 years, range 6–18 years). Puberty had commenced in 199 patients (85.8%) (Table 1).

PREVALENCE OF MAFLD AND ULTRASONOGRAPHIC GRADING:

Hepatic steatosis was detected by ultrasonography in 135/222 patients (60.8%), meeting the operational definition of MAFLD in this obesity cohort. Steatosis severity was grade 1 in 80 patients (36.0%), grade 2 in 36 patients (16.2%), and grade 3 in 19 patients (8.6%); the remaining 87 patients (39.2%) had no steatosis (grade 0). The mean age of patients with hepatic steatosis was 13.59±2.86 years (median 14.08 years, range 6.5–18 years) (Table 1).

SEX AND PUBERTAL STATUS:

As shown in Table 2, MAFLD prevalence differed significantly by sex (Pearson chi-square, P=0.001): MAFLD was present in 81/151 females (53.6%) and 54/71 males (76.1%). Accordingly, male sex was associated with higher odds of MAFLD (unadjusted OR, 2.745; 95% CI, 1.459–5.164). Puberty had commenced in 87.4% of patients with non-MAFLD obesity and 85.2% of those with MAFLD; pubertal status was not associated with MAFLD (P=0.649), and the distribution of Tanner stages was also similar between groups (P=0.292) (Table 3).

METABOLIC AND BIOCHEMICAL COMPARISONS BETWEEN GROUPS:

Clinical and laboratory comparisons between patients with non-MAFLD obesity and those with MAFLD are summarized in Table 4. Age did not differ significantly between groups (12.97±3.13 vs 13.62±2.85 years; P=0.109). Patients with MAFLD had higher weight SDS (3.45±1.11 vs 3.02±0.90, P=0.002) and BMI SDS (3.06±0.63 vs 2.86±0.54; P=0.037). Markers of metabolic dysfunction were also higher in MAFLD, including fasting insulin (33.74±22.42 vs 25.60±13.94; P=0.003), HOMA-IR (7.91±5.84 vs 5.72±3.45; P=0.003), and HbA1c (5.66±1.01 vs 5.39±0.23; P=0.021). Liver enzymes were significantly higher in MAFLD (ALT 36.06±32.99 vs 20.43±10.77; P<0.001; AST 29.89±22.12 vs 20.00±5.61; P<0.001), whereas HDL-C was lower (44.91±9.55 vs 48.69±9.99; P=0.005). Height SDS, fasting glucose, LDL-C, total cholesterol, triglycerides, TSH, and FT4 did not differ significantly between groups (all P>0.05) (Table 4).

VITAMIN D STATUS AND ASSOCIATION WITH MAFLD:

Serum 25(OH)D levels were available in 148/222 patients (56 with non-MAFLD obesity and 92 with MAFLD). Mean 25(OH)D levels were similar between the MAFLD and non-MAFLD obesity groups (16.64±9.38 vs 15.93±5.90 ng/mL; P=0.782) (Table 4). When categorized as deficiency (<12 ng/mL), insufficiency (12–20 ng/mL), or sufficiency (>20 ng/mL), the distribution of vitamin D status did not differ between groups (P=0.225) (Table 3).

VITAMIN D AND STEATOSIS GRADE:

In the subgroup with available 25(OH)D measurements (n=148), 25(OH)D concentrations did not differ significantly across ultrasonographic steatosis grades (P=0.686) (Table 5). Similarly, when vitamin D status categories were compared according to steatosis grade, no significant association was observed (P=0.147) (Table 6).

Discussion

LIMITATIONS:

This study has limitations inherent to its retrospective, single-center design, including inability to infer causality and possible selection bias. Vitamin D measurements were available for only a subset of participants, and the study could not adjust for seasonal variation, sunlight exposure, dietary intake, or supplementation. MAFLD was diagnosed using ultrasonography rather than magnetic resonance imaging or histology, which may have led to misclassification, particularly for mild steatosis [21]. Finally, socioeconomic and lifestyle factors were not systematically recorded, limiting the ability to evaluate potential confounding.

Conclusions

Among children and adolescents with obesity, MAFLD was common and was associated with male sex and an adverse metabolic profile. Serum 25(OH)D levels were not associated with MAFLD presence or ultrasonographic steatosis grade in the subgroup with available measurements. The key clinical implication is the need for careful monitoring and management of metabolic risk factors in children and adolescents with obesity. Prospective studies with standardized vitamin D assessment and more sensitive liver fat quantification are needed to clarify whether vitamin D has a causal role in pediatric MAFLD.