Introduction

Pediatric obesity is associated with cardiometabolic diseases (insulin resistance, type 2 diabetes, dyslipidemia, high blood pressure) and nonalcoholic fatty liver diseases (NAFLD) [1, 2]. Defined as an excess of hepatic fat, NAFLD has become the most common liver disease in childhood, regarded as a hepatic manifestation of the metabolic syndrome (MetS) [3]. Youths with obesity and MetS have 5 times the odds of having NAFLD compared with their counterparts with obesity alone, and youths with both obesity and NAFLD have higher insulin resistance, dyslipidemia, and blood pressure than youths with obesity alone [4]. As obesity and related cardiovascular and metabolic diseases track into adulthood, there is a need of a life course perspective in pediatric obesity prevention and treatment. Multidisciplinary interventions combining nutritional guidelines and physical activity (PA), particularly moderate to vigorous PA (MVPA), remain the cornerstone therapy of pediatric obesity and related diseases [3, 5,6,7]. Behavioral recommendations, that mainly focused on MVPA so far, also advocate now for the minimization of sedentary (SED) time (i.e., waking behavior in a sitting, reclining, or lying posture), synergistically with MVPA increase [8]. Both high MVPA and low SED times have been associated with healthier profiles of liver enzymes and reduced liver fat content (LFC) in adults [9,10,11,12,13]. In youths, while evidence supports that increasing MVPA and decreasing SED time reduce cardiometabolic risks [14,15,16,17], relationships between accelerometry-based PA levels and NAFLD markers remain understudied [18,19,20], and only few studies investigated so far the association between SED time and NAFLD surrogate markers (without imaging), showing discrepant results [18, 19, 21].

As it remains difficult to implement MVPA in youths with obesity in clinical practice, our team conducted a cross-sectional study comparing 7-day/24-h accelerometry-based patterns of SED (more sedentary, SED + vs less sedentary, SED-), MVPA (more active, MVPA + vs less active, MVPA-), and combinations of SED and MVPA times (SED-/MVPA + , SED-/MVPA-, SED + /MVPA + , SED + /MVPA-). Results showed that (i) low SED/high MVPA pattern had the most desirable cardiometabolic profile; (ii) SED-/MVPA- pattern had lower fat mass, insulin resistance, dyslipidemia, and metabolic syndrome risk score, independent of BMI, in comparison with SED + /MVPA + pattern; and (iii) SED time was correlated with cardiometabolic risks independently of MVPA [22, 23]. Considering the strong relationships between cardiometabolic comorbidities and NAFLD [4, 24], it seemed relevant to question whether movement behavior patterns might be associated with NAFLD markers, in the same population.

Therefore, the present study aimed to (i) compare patterns of SED, MVPA, and combinations of SED and MVPA times regarding ALT (as first criterion), other liver enzymes (gamma-glutamyl transpeptidase (GGT), aspartate aminotransferase (AST) and AST/ALT ratio), and LFC (as secondary criteria); and (ii) investigate correlations between SED time, PA levels, and NAFLD markers in children and adolescents with obesity. Alanine aminotransferase (ALT) has been chosen as first criterion as it is considered as the best biochemical marker of hepatocellular lipid content [25]. We hypothesized that SED + patterns would have higher ALT irrespectively of MVPA and that MVPA time would be negatively and SED time positively associated with ALT.

Materials and methods

Participants

This cross-sectional study involved 134 children and adolescents with obesity participating in the betaJUDO study [26]. They were included in 2 centers (Pediatric Obesity Clinic at University Children’s Hospital, Uppsala, Sweden, and Paracelsus Medical University, Salzburg, Austria) during their first consultation. Inclusion criteria were the following: (a) 10–17 years old, (b) age-adapted BMI > 30 kg m−2, (c) medical examination including anthropometric assessment and Tanner’s staging, (d) at least 5 valid days out of a possible 7 days of accelerometry measurements, including one of the valid day on a weekend (regardless of PA and SED levels), (e) blood samplings for liver parameters, (f) no contraindication to PA, (g) no specific lifestyle intervention during at least 6 months before inclusion, (h) no additional medical/psychiatric conditions nor medication influencing cardiometabolic, liver enzymes, or accelerometry data. The exclusion criteria were a lack of consent or if the patients had any chronic liver disease. To this end, endocrine disorders (thyroid disease, diabetes type I), autoimmune, viral (viral hepatitis, HIV), or hereditary causes (Wilson disease, hereditary hemochromatosis, alpha-1 antitrypsin deficiency, celiac disease, lysosomal acid lipase deficiency) of liver disease, and use of steatogenic drugs were excluded in all patients with increased serum transaminases (ALT, AST) > 40 U/L. Patients did not report any alcohol intake. Furthermore, two patients with a history of attention-deficit/hyperactivity disorder under Ritalin were excluded from the study because of the possibility of elevated ALT due to treatment. The study was accepted for Voluntary Harmonisation Procedure (VHP673, VHP2015061) and approved by Ethics Committees and Regulatory Authorities (EudraCT No: 2015–001,628-45; EC Sweden: Dnr 2015/279; EC Austria: 415-E/1544/20–2014). Written informed consent was obtained from participants and parents. The trial was conducted according to the Declaration of Helsinki (World Medical Association; Version 2013) and the E6 Guideline for Good Clinical Practice (International Conference on Harmonisation).

Anthropometry and pubertal staging

Standard operating procedures for measurements were harmonized between centers [26]. Weight (kg) was assessed using a standardized calibrated scale (Uppsala, SECA model 704; Salzburg, SECA model 801, Hamburg, Germany). Height (cm) was measured using a stadiometer (Uppsala, Ulmer stadiometer, Busse, Elchingen, Germany; Salzburg, SECA, model 222 stadiometer, Hamburg, Germany). BMI was calculated as weight (kilograms) divided by the square of height (meters). BMI-SDS (Microsoft Excel add-in LMS Growth using WHO growth report Version 2.76) and BMI percentiles (WHO BMI for age) were calculated. Waist circumference (WC, cm) was measured with a flexible tape midway between the superior border of the iliac crest and the lowest rib on a standing patient. Fat mass (FM) percentage was calculated using an InBody S20 bioimpedance device (Biospace, Seoul, Korea) at fasting. Puberty was evaluated with Tanner staging [27, 28].

Biochemical variables

Blood was sampled at fasting. Validation of analyses was performed between laboratories [26]. Serum concentrations of liver enzymes (ALT, AST, and GGT), total-cholesterol, high-density lipoprotein-cholesterol (HDL-c), low-density lipoprotein-cholesterol (LDL-c), and triglycerides (TG) were analyzed by enzymatic photometric analysis. Plasma glucose was analyzed by enzymatic chromatic test. Plasma was used for central analyses of insulin using singleplex enzyme-linked immunosorbent assay kits for each analyte (Mercodia AB, Uppsala, Sweden). Insulin resistance was expressed using homeostasis model assessment of insulin-resistance index (HOMA-IR), [HOMA-IR] = glycemia [mmol⋅L−1] × insulinemia [mUI⋅L−1]/22.5) [29].

Percentage hepatic fat

Percentage hepatic fat was measured by magnetic resonance imaging (MRI) using 1.5 Tesla clinical MRI systems from Philips Medical System (Best, The Netherlands; Uppsala, Philips Achieva system, Salzburg: Philips Ingenia system) [26]. A dedicated single breath hold scan was used to image the liver with a 6-echo water-fat imaging protocol. Water fat images were reconstructed using in-house developed software. Liver tissue was identified using manual segmentation of a large volume of interest. Uppsala served as core laboratory and developed and standardized the imaging protocol at both sites and performed all image analyses. More details have been published elsewhere [30]. Children were categorized as having hepatic steatosis when percentage hepatic fat ≥ 5% [31].

Physical activity and sedentary time

Movement-related behaviors were assessed with the accelerometer Actical® (Philips Respironics, Inc, Murrysville, PA). As previously described [23], it is an omni-directional waterproof device recording accelerations in the range of 0.05–2.0 g, sensitive to movements in the range of 0.35–3.5 Hz, and able to record the magnitude of acceleration and deceleration associated with every movement. The signal was scored as a “count” which was summed over a 1-min epoch. Participants were asked to wear the device on their non-dominant wrist during 7 consecutive days (24-h measurements). Non-wear time was defined as ≥ 60 consecutive minutes of zero counts, with allowance for 2 min of counts between zero and 100. Wear time was determined by subtracting non-wear time from 24 h. A valid day was defined as ≥ 10 h of wear time. Each minute of wear time was classified using established cut points into SED (< 1.5 metabolic equivalent of the task, METS, 0–99 counts per epoch), light physical activity (LPA, < 3 METS, 100–742 counts per epoch), moderate activity (MPA, 3 to 6 METS, 743–2778 counts per epoch), and vigorous activity (VPA, > 6 METS, < 2779 counts per epoch) [32, 33]. MVPA was the sum of MPA and VPA. As we unfortunately have no objective data to estimate reliably sleep duration, sleep duration has been included in “SB time” corresponding to the definition of EE < 1.5 METS. SED + (more sedentary) and SED- (less sedentary) groups were defined by being respectively upper and under the median of the sample for SB time. MVPA + (more active) and MVPA- (less active) groups were defined by being respectively upper and under the median of the sample for MVPA time. The 4 combinations (SED-/MVPA + , SED-/MVPA-, SED + /MVPA + , SED + /MVPA-) were created using the median of MVPA for each of SED subsamples.

Statistical analysis

Statistical analyses were performed using Stata software (version 15, StataCorp, College Station, USA). Continuous data were expressed as means and standard-deviations (SD). The normality of the distribution was checked with a Shapiro–Wilk test. Comparisons between groups were performed using Chi-squared or Fisher’s exact test for categorical data, and analysis of variance (ANOVA) or non-parametric Kruskal–Wallis test (when the ANOVA assumptions were not met) for continuous variables. Assumption of homoscedasticity was studied using Bartlett’s test. When appropriate (omnibus p-value < 0.05), post hoc test for two by two multiple comparisons was applied: Tukey–Kramer after ANOVA, Dunn after Kruskal–Wallis test, and Marascuilo for categorical data. Relationships between continuous data were explored with Pearson or Spearman correlation coefficient and a Sidak type I error correction. Multivariable analyses were conducted using multiple linear regressions in order to compare groups adjusting aforementioned analyses on possible confounders. No specific strategy approach, such as stepwise, was conducted. Covariates were chosen according to univariate results and clinical relevance. Multivariable regression analyses were run with the following covariates: age, gender, and Tanner stages (model 2), and age, gender, Tanner stages, and BMI (model 3). The normality of residuals was checked and a logarithmic transformation of the dependent variable was performed when appropriate. Differences were considered statistically significant at p < 0.05.

Results

One hundred and thirty-four adolescents (mean age 13.4 ± 2.2 years, 48.5% females) were included (n = 97 from Uppsala and n = 37 from Salzburg). Among them, 12 subjects (9%) had a history of asthma (stable asthma, without contra-indication to exercise, and no treatments or only inhaled beta-2-mimetics when needed) and 6 subjects (4.5%) had vitamin D deficiency with supplementation. The mean BMI was 98.9 ± 0.7 percentile. Mean accelerometry wear time was 6.5 ± 1.1 days with 99.4 ± 2.9% of daily wear time. All the patients underwent anthropometric measurements, accelerometry, and biochemical assessments. A subgroup of thirty-nine patients underwent MRI-LFC measurements. This subgroup was not different compared to the other subjects concerning age, gender, WC, BMI, and hepatic biochemical variables (data not shown). Anthropometrics, accelerometry, and hepatic variables for overall sample, SED- and SED + groups, and MVPA- and MVPA + groups are presented in Table 1.

Table 1 Anthropometric, accelerometry, and hepatic variables for overall sample, SED-, SED + , MVPA + , and MVPA- groups. Model 1: univariate analysis. Model 2: adjusted with age, gender, and Tanner. Model 3: adjusted with age, gender, Tanner, and BMI
Full size table

Comparison SED + vs SED- groups and MVPA- vs MVPA + groups

In univariate analysis (model 1), SED + group had higher ALT (p = 0.016), GGT (0.019), and MRI-LFC (p = 0.012), with a lower AST/ALT ratio (p < 0.001) in comparison with SED- group. These results remained significant when adjusted for age, gender, and Tanner stages (model 2) and when adjusted for age, gender, Tanner stages, and BMI (model 3), except for GGT (p = 0.055). For the three models, the number of subjects with a MRI-LFC ≥ 5% tends to be higher in the SED + group in comparison to SED- group, without however reaching significance (p = 0.064, p = 0.056, and p = 0.067, respectively) (Table 1).

In univariate analysis (model 1), MVPA- group had higher ALT (p = 0.003) and GGT (p = 0.049), with a lower AST/ALT ratio (p < 0.001) in comparison with MVPA + group. These results remained significant when adjusted for age, gender, and Tanner stages (model 2) and when adjusted for age, gender, Tanner stages, and BMI (model 3), except for GGT (p = 0.298 and p = 0.363, respectively) (Table 1). MRI-LFC was not significant between MVPA- and MVPA + groups (p = 0.054).

Results regarding insulin resistance and lipid profile have been published elsewhere [22] and are presented in supplementary files for overall sample, SED- and SED + groups, and MVPA- and MVPA + groups (Supplementary Table 1).

Comparison between SED-/MVPA +, SED-/MVPA-, SED + /MVPA +, and SED + /MVPA- groups

Anthropometrics, accelerometry, and hepatic variables for SED-/MVPA + , SED-/MVPA-, SED + /MVPA + , and SED + /MVPA- groups are presented in Table 2.

Table 2 Anthropometric, accelerometry, and hepatic variables for SED-/MVPA + , SED-/MVPA-, SED + /MVPA + , and SED + /MVPA-. p-values are adjusted with age, gender, and Tanner
Full size table

SED-/MVPA + group had a lower MRI-LFC (p < 0.01) in comparison with the SED + /MVPA- group, and lower ALT (p < 0.01) and GGT (p < 0.05) with a higher AST/ALT ratio (p < 0.01) in comparison with the SED + /MVPA + and SED + /MVPA- groups, after adjustment with age, gender, and Tanner stages.

BMI was not different between the SED-/MVPA- and SED + /MVPA + groups (p = 0.108). However, after adjustment with age, gender, and Tanner stages, SED-/MVPA- group had lower ALT (p < 0.05) and GGT (p < 0.05) and higher AST/ALT ratio (p < 0.05), in comparison with SED + /MVPA + group (Table 2). These results remained significant after adjustment with age, gender, Tanner stages, and BMI (p < 0.05).

Results regarding insulin resistance and lipid profile have been published elsewhere [22] and are presented in supplementary files for SED-/MVPA + , SED-/MVPA-, SED + /MVPA + , and SED + /MVPA- groups (Supplementary Table 2).

Correlations

All correlations between hepatic variables and accelerometry variables are presented in Table 3. SED time was positively correlated with ALT (p < 0.05) and MRI-LFC (p < 0.01), and negatively correlated with AST/ALT ratio (p < 0.001). SED time remained positively correlated with ALT (p < 0.05) and MRI-LFC (p < 0.05) and negatively correlated with AST/ALT ratio (p < 0.05) after adjustment with MVPA time. These associations remained significant after adjustment for age, gender, Tanner stages, BMI, and MVPA time (p < 0.05).

Table 3 Correlations between anthropometric, hepatic, and accelerometry variables
Full size table

MVPA was negatively correlated with ALT (p < 0.05) and positively correlated with AST/ALT ratio (p < 0.001). MVPA tend to be associated with MRI-LFC, without however reaching significance (p = 0.056).

Discussion

This study aimed to investigate the association between SED and MVPA patterns measured by 24-h/7-day accelerometry and NAFLD markers in children and adolescents with obesity. According to our analysis, although SED-/MVPA + pattern is associated with the best hepatic health surrogates, juveniles with SED-/MVPA- pattern had better biochemical hepatic markers in comparison with those with SED + /MVPA + pattern, independently of BMI. Moreover, SED time was positively associated with biochemical (high ALT, low AST/ALT ratio) and imaging (high MRI-LFC) markers of hepatic health independently of MVPA. MVPA time was associated with biochemical markers of hepatic health (low ALT, high AST/ALT ratio), but these associations were no longer significant after adjustment for SED time. To our knowledge, this is the first study to measure objectively and concomitantly PA and SED times in regard to NAFLD biomarkers and MRI-LFC in youths with obesity. The results demonstrate the importance of a reduced SED time on hepatic health, irrespectively of MVPA, in youths with severe obesity.

In the subgroup who underwent liver MRI, the proportion of subjects having a LFC ≥ 5% was 46% (n = 18/39), which is in keeping with previous studies showing a large variability in NAFLD prevalence in youths with obesity, depending on the diagnostic method [7, 20, 34,35,36]. In a meta-analysis based on various diagnostic methods (ALT, ultrasonography, and MRI), NAFLD prevalence was estimated at 34% in a large sample of 23,892 youths under 14 years old [34]. Studies measuring percentage hepatic fat by MRI estimated a prevalence between 30 and 44% depending on the cut-off points of hepatic fat fraction [20, 35, 36].

The better hepatic health in SED-/MVPA + group is fully in line with the more favorable cardiometabolic health found in youths meeting guidelines for both MVPA and SED times [22, 37, 38]. The present results strengthen a body of evidence promoting PA recommendations in order to optimize liver health in young subjects [18,19,20, 34, 39]. Moreover, the comparison between SED-/MVPA- and SED + /MVPA + groups and the mutually adjusted model of correlations between movement-related behaviors and NAFLD markers strongly highlights the importance of SED time in determining optimal liver health, independent of MVPA level. This is concordant with previous findings in adults reporting a positive independent association between SED time and LFC [10, 12, 13]. Interestingly, Li et al. recently showed in more than 16,000 adults that after adjusting for MVPA, increasing quartiles of SED time were associated with a higher prevalence of elevated ALT and GGT, which remained significant after further adjustment for cardiometabolic traits (including BMI, lipids, and HOMA-IR) [9]. In contrast, increasing quartiles of MVPA were associated with a lower prevalence of elevated ALT after adjustment for SED time, but this became non-significant after further adjustment for cardiometabolic traits [9]. In youths, while evidence now supports that decreasing SED time reduces cardiometabolic risks [22, 40,41,42], only Martins et al. previously found a positive association between SED time and ALT [19]. Norman et al. however showed that screen time, a main contributor to SED time, was positively associated with cardiometabolic parameters and ALT, independent of MVPA time, in adolescents with obesity [21], suggesting that limiting screen time would represent an independent lever of action to improve hepatic health. All these results are in line with recently proposed anti-obesity strategies aiming at breaking up prolonged periods of SED time in youths, replacing it with MVPA (the preferred and more efficient scenario) or with LPA (beneficial but with lower effect estimates) [43,44,45]. Winters-van Eekelen recently showed that reallocation of SED time with MVPA was associated with less total, visceral, and liver fat in adults [46].

The present results have to be considered in light of some limitations. First, only a subgroup of subjects underwent liver MRI. The observed correlation between MRI-LFC and SED time is therefore questionable in power and the absence of correlations between MRI-LFC and MVPA might be explained by the small sample size. However, although liver enzymes alone are inadequate for identification of hepatic steatosis in children, ALT remains the best biochemical marker of hepatocellular lipid content [25]. It has been associated with insulin resistance and cardiometabolic health in adolescents [47]. Secondly, as previously discussed [22, 23], while accelerometry is the gold standard method for measuring movements-related behaviors, MVPA has been potentially overestimated at the expense of LPA. While Actical was shown comparable to the Actigraph GT3X in youths [48], Migueles et al. recently demonstrated that MVPA time largely differed across attachment sites (hip vs wrist), acceleration metrics (number of days and hours per day), and cut-points in youths with obesity [49]. These parameters were however similar for all the participants of the present study. The wrist attachment site had been chosen to increase wear compliance [50], which is a strength of the present study, providing measurements with more than 99% of wear time during a mean of 6.5 days. Finally, Swedish children are known to have a meaningful 50% higher MVPA than children of other European countries [51]. Finally, we had no data to estimate reliably sleep duration and we could not differentiate sleep from SED time (defined by energy expenditure < 1.5 METS). Sleep will be an important parameter to include in future researches as some studies investigating separately sleep, sedentary time, and PA showed that insufficient sleep would have negative impacts on children’s cardiometabolic health [52, 53].

While recommendations have largely focused on MVPA so far, the present study reinforces the need to reduce SED time, synergistically with the increase in MVPA time, to optimize liver health in youths with severe obesity. In clinical practice, youths with severe obesity should benefit from an individual behavioral diagnostic targeting all movement behaviors, and stakeholders should be sensitized to support the shift from long periods of SED time to daily routines incorporating bouts of PA. Furthermore, reducing SED time might be a first step in the management of pediatric obesity NAFLD when increasing MVPA is not possible. However, since this research only studies associations, causal relations of the observed correlations have not been proven. Long-term cohort studies on the impact of changes in PA and SED times on hepatic health outcomes, and interventional studies with a sequential action plan (i.e., sequencing the actions into a first phase with the aim to reduce SED time followed by a classic supervised MVPA training), are needed.