To the best of our knowledge, this is the first study to investigate the association between ALT levels and anthropometrical and biochemical characteristics of adolescents from northeast Brazilian. In boys, ALT was correlated with fasting insulin, triglycerides and systolic blood pressure (SBP). In girls, ALT was correlated with BMI, WC, conicity index, fasting insulin and HDL-c. Thus, we identified a relationship between ALT levels and the diagnostic criteria for MS.
Conicity index, ALT and AST levels were higher in boys when compared to girls. These sex-specific differences are in agreement with previous studies [4, 20]. Girls showed higher stages of sexual maturation and fasting insulin values. In this context, there is a consensus that higher insulin levels are related to body fat distribution. Higher accumulation of fat in central areas, referred to as visceral body fat, triggers a decrease in insulin action and, consequently, hyperinsulinemia. Thus, in addition to excess weight, the pubertal stage can also lead to metabolic changes [21–24].
The percentage of boys and girls with ALT over the maximum limit was 20.6% and 29.8%, respectively. There is conflicting literature on the proportion of individuals with increased ALT activity, which can be attributed to different cutoff points for this age group [25, 26]. The cutoffs for ALT of 22 mg/dl for girls and 26 mg/dl for boys, used in this study have been recommended due to being more accurate and sex-specific and were determined and validated using a representative and diverse sample.4 A study assessing obese Japanese adolescents' hepatic enzymes found alteration in 16.3% of the studied boys and 5.3% of the girls .
It is well established that the accumulation of liver fat is intimately related to diagnostic criteria for MS, along with elevated of transaminases values (especially ALT) [4, 27]. This fact was confirmed by our findings, where it was observed that, in boys, higher levels of ALT were related to increases in fasting insulin, triglycerides, and SBP. In line with this, other researchers also found positive correlations between higher ALT quartiles and biochemical and anthropometric variables in boys. The same compensatory mechanism can explain higher insulin values and the accumulation of intrahepatic fat: insulin resistance, mediated especially by the accumulation of visceral fat. When in excess, this tissue leads to chronic, low-grade inflammation due to higher macrophage infiltration, which produces pro-inflammatory cytokines such as TNF-α. These cytokines then decrease insulin action on that tissue [3, 28].
We found that, in both sexes, fasting insulin was correlated with ALT. In girls, there was also a correlation between HOMA-IR and ALT, which indicates the close relationship between IR and NAFLD. It is established that increases in visceral fat decrease adiponectin production, which has anti-inflammatory properties and is crucial to maintaining energetic homeostasis. Low levels of this adipokine and high TNF-α lead to lower insulin sensitivity, creating a state of hyperinsulinemia and IR. When combined with decreased adiponectin, IR contributes to NAFLD development, where the accumulation of liver fat leads to an increase in serum ALT values .
Another finding in our study was that, in boys, ALT was correlated with PAS and triglycerides. It was found that high SBP is a possible marker of NAFLD [29, 30]. Increases in SBP are independently associated with an increased risk of NAFLD, and this relationship is more evident in individuals with hypertriglyceridemia. Still, studies show the relationship between increased waist circumference and high blood pressure, aligning the results . Thus, it is recommended that the assessment and control of SBP be included in the monitoring of overweight adolescents .
In patients with NAFLD, dyslipidemia is often present as a combination of increased triglycerides and decreased HDL-c. In the male participants of our study, we found a correlation between triglycerides and ALT. In girls, ALT was negatively correlated with HDL-c. Studies that investigated the relationship between undesirable lipid profiles and NAFLD in overweight children and adolescents have also presented similar findings [14, 33].
We did not find correlations between TG/HDL and ALT in either sex. In boys, however, the correlation was close to the significance margin. Other authors  found that individuals with NAFLD presented higher TG/HDL values, which was also directly related to ALT, cholesterol, insulin and HOMA-IR. Again, IR can be observed as the primary mediator of this process, promoting an increased secretion of larger VLDL particles rich in triglycerides, which lowers HDL-c concentrations . The decrease in adiponectin due to an accumulation of visceral tissue lowers HDL-c and increases circulating triglycerides, which increases the TG/HDL ratio .
In girls, we found a correlation between ALT levels and all anthropometric parameters assessed in this study. BMI showed the strongest correlation. BMI is considered an important risk factor for NAFLD in adolescents, with excess body fat being highly relevant for the progression of this disease . A study performed on a cohort of Danish children and adolescents found that, across all age ranges, those who were overweight or obese showed elevated ALT . As we previously mentioned, there is an evident relationship between the accumulation and distribution of body fat – mainly in central areas – and NAFLD. Waist circumference (WC) has been used as a tool in NAFLD screening, as increases in WC can predict the risk of this disease in obese adolescents, especially in values above 99 cm .
Another group of researchers  also found that, in obese adolescents, an increase of 1 cm in WC was associated with a twofold increase in NAFLD risk. We also found a positive correlation between ALT and conicity index, which is obtained from WC, weight and height. Previous studies have found a relationship between conicity index, IR and obesity, which highlights the efficacy of this measure as a predictor of NAFLD . Anthropometric measurements are low-cost and easily applicable in clinical settings. Thus, they can be implemented in the clinical screening of populations at risk of cardiovascular diseases, especially those who are obese.
Our study presents some limitations, such as the fact that the cross-sectional design does not identify a cause-and-effect relationship between ALT and the studied anthropometric and metabolic variables. We also did not use imaging tests to screen for NAFLD, which is considered the most accurate. However, even though it is an indirect measure of hepatic injury, we highlight that ALT is an accessible and practical way to track the progression of NAFLD, being one of the most widely used methods in clinical practice. Our study found that using BMI, waist circumference, and conicity index measurements can be reliable techniques to estimate increases in ALT, a risk factor for developing hepatic steatosis in overweight and obese adolescents, especially girls.