
==== Front
Sci Rep
Sci Rep
Scientific Reports
2045-2322
Nature Publishing Group UK London

39232127
71608
10.1038/s41598-024-71608-8
Article
Nonlinear relationship between the triglyceride–glucose index and alanine aminotransferase in children with short stature
Zhao Qianqian 12
Li Youqian 3
Zhang Mei 12
Ban Bo banbo2011@163.com

12
1 grid.449428.7 0000 0004 1797 7280 Department of Endocrinology, Affiliated Hospital of Jining Medical University, Jining Medical University, 89 Guhuai Road, Jining, 272029 Shandong People’s Republic of China
2 Chinese Research Center for Behavior Medicine in Growth and Development, 89 Guhuai Road, Jining, 272029 Shandong People’s Republic of China
3 grid.452252.6 0000 0004 8342 692X Department of Cardiovasology, Affiliated Hospital of Jining Medical University, Jining Medical University, 89 Guhuai Road, Jining, 272029 Shandong People’s Republic of China
4 9 2024
4 9 2024
2024
14 205883 5 2024
29 8 2024
© The Author(s) 2024
2024
https://creativecommons.org/licenses/by-nc-nd/4.0/ Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Metabolic dysfunction associated fatty liver disease (MAFLD) is a common cause of liver disease in children and adolescents. The relationship between insulin resistance (IR) and MAFLD in children with short stature remains largely unknown. The present study was to investigate the relationship between the triglyceride–glucose (TyG) index and alanine aminotransferase (ALT) levels in children with short stature. A total of 1754 children with short stature were enrolled. Anthropometric, biochemical and hormonal indexes were collected through physical measurement examinations and laboratory tests. A nonlinear association was found between the TyG index and ALT. The inflection point of the curve was at a TyG index of 8.24. In multivariate piecewise linear regression, only when the TyG index was greater than 8.24 was there a significant positive association between the TyG index and ALT (β 5.75, 95% CI 3.30, 8.19; P < 0.001). However, when the TyG index was less than 8.24, there was no significant association between the TyG index and ALT (β −0.57, 95% CI −1.84, 0.71; P = 0.382). This study demonstrated a nonlinear relationship between TyG index and ALT in children with short stature. This finding suggests that a high TyG index is associated with elevated ALT in children with short stature.

Keywords

Metabolic dysfunction associated fatty liver disease
Insulin resistance
Nonlinear relationship
Triglyceride glucose index
Alanine aminotransferase
Subject terms

Endocrinology
Risk factors
the Natural Science Foundation of Shandong ProvinceZR2022MH284 Ban Bo issue-copyright-statement© Springer Nature Limited 2024
==== Body
pmcIntroduction

Metabolic dysfunction associated fatty liver disease (MAFLD) is the most common chronic liver disease in the world and not only causes liver cirrhosis, liver failure and even liver cancer but also increases the risk of developing atherosclerosis and cardiovascular disease (CVD)1. In recent years, the global prevalence of MAFLD has been gradually increasing, reaching approximately 30%2. Even among children and adolescents, the prevalence of MAFLD can reach 3–10%3. Various diagnostic methods, including liver biopsy, ultrasound scan (USS), or magnetic resonance imaging (MRI), can be used to evaluate MAFLD; however, due to their invasive nature, time-consuming procedures, and high costs, they may not be suitable for large-scale population screening4. In clinical practice, elevated alanine aminotransferase (ALT) levels are the most commonly used predictive marker for pediatric MAFLD. Societies including the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition and the American Academy of Pediatrics recommend that children at risk for MAFLD be screened using the enzyme ALT5. Therefore, a high ALT level, in the absence of other causes of liver injury, typically indicates the prevalence of MAFLD3. Data from the National Health and Nutrition Examination Survey Program showed that the prevalence of MAFLD, using elevated serum ALT as a surrogate marker for MAFLD, was 8.0% among U.S. adolescents6.

Insulin resistance (IR) is involved in the pathogenesis and progression of MAFLD7,8. The triglyceride–glucose (TyG) index, as a more sensitive indicator to assess IR, has been validated by the hyperinsulinemic–euglycemic clamp (the gold standard for IR) and homeostasis model assessment of insulin resistance (HOMA-IR)9,10. Several cohort studies have shown that the TyG index is associated with an increased risk of MAFLD and that it is an early predictor of MAFLD11–13. A recent study revealed a dose‒response relationship between the TyG index and transaminases, especially ALT. A high TyG index is a novel risk factor for elevated aminotransferase14. Previous studies have focused on the association between the TyG index and transaminase in obese people15,16; however, in Asia, more than one-fifth of MAFLD patients are not obese17. More recent evidence suggests that MAFLD should also be of concern in people with short stature18,19. In nonobese Chinese people with normal lipid levels, the risk factors associated with MAFLD are more prevalent in people with short stature18. Furthermore, height was inversely associated with the risk of MAFLD in the general population, indicating that individuals with short stature are at an increased risk of developing MAFLD19.

Several studies have suggested that growth hormone (GH) and insulin-like growth factor-1 (IGF-1) levels are negatively associated with MAFLD and that the GH/IGF-1 axis is a potential target for MAFLD treatment because of its possible lipolytic, anti-inflammatory and immunomodulatory effects20–22. It is well known that the GH/IGF-1 axis is a key factor in regulating linear growth, and short stature is largely due to an impaired GH/IGF-1 axis23. Therefore, MAFLD in children with short stature deserves more attention; however, there is a lack of a database of children and adolescents with short stature to explore the relationship between the TyG index and ALT. The evidence presented above has demonstrated that the TyG index is an independent predictor of MAFLD. We hypothesized that in patients with short stature, the TyG index may be positively correlated with ALT levels and could serve as a risk factor for elevated ALT. To test this hypothesis, we conducted a cross-sectional analysis of the relationship between the TyG index and ALT levels in children with short stature on the basis of a large 10 years cohort from China. Additionally, a subgroup analysis was performed on populations with and without growth hormone deficiency (GHD).

Results

Participant characteristics

The flow chart of study population selection is shown in Fig. 1. A total of 1871 participants with short stature were recruited for the study. After screening according to exclusion criteria, 1754 subjects were finally included in the study, including 1193 males (68.02%) and 561 females (31.98%), with an average age of 10.4 ± 3.6 years. The clinical and biochemical characteristics of the participants are described in Table 1. We divided the subjects into tertiles based on their ALT levels, and participants in the highest ALT tertile had higher values of age, bone age, weight, body mass index (BMI), IGF-1, systolic blood pressure (SBP), diastolic blood pressure (DBP), aspartate aminotransferase (AST), TyG index, serum uric acid (SUA), triglyceride (TG), total cholesterol (TC) and low density lipoprotein cholesterol (LDL-C) than those in the lowest and middle ALT tertiles (all P < 0.05). There were no significant differences in sex, IGF-1 standard deviation score (IGF-1 SDS), blood urea nitrogen (BUN), fasting plasma glucose (FPG), high density lipoprotein-cholesterol (HDL-C) or puberty stage among the three groups (all P > 0.05).Fig. 1 Flow chart of the study population.

Table 1 Clinical and biochemical characteristics according to ALT tertiles.

Variables	ALT tertiles	P value	
Tertile 1 (< 11.50)	Tertile 2 (11.50–15.50)	Tertile 3 (> 15.50)	
Number	578	591	585	−	
Sex (male %)	378 (65.40%)	401 (67.85%)	414 (70.77%)	0.145	
Age (years)	10.3 ± 3.5	9.8 ± 3.6	10.9 ± 3.5	< 0.001	
Bone age (years)	8.6 ± 3.8	8.0 ± 3.8	9.1 ± 3.7	< 0.001	
Height (cm)	125.93 ± 18.02	122.97 ± 17.76	128.36 ± 17.51	< 0.001	
Height SDS	−2.43 (−2.89–2.20)	−2.52 (−2.95–−2.22)	−2.49 (−2.96–2.19)	< 0.001	
Body weight (kg)	25.00 (18.25–34.00)	24.00 (17.50–32.00)	29.50 (20.00–37.00)	< 0.001	
BMI (kg/m2)	16.45 ± 2.70	16.59 ± 2.94	17.47 ± 3.41	< 0.001	
IGF-1 (ng/ml)	157.00 (102.00–243.00)	150.50 (91.05–237.75)	180.00 (105.00–261.00)	0.011	
IGF-1 SDS	−1.07 (−1.81–0.23)	−1.00 (−1.81–0.21)	−1.00 (−1.71–0.10)	0.654	
Peak GH (ng/mL)	7.30 (4.96–11.09)	6.69 (4.29–9.76)	6.29 (3.86–10.47)	< 0.001	
SBP (mmHg)	104.90 ± 11.44	104.26 ± 11.76	108.12 ± 12.44	< 0.001	
DBP (mmHg)	62.10 ± 8.35	61.65 ± 8.40	63.87 ± 9.01	< 0.001	
ALT (U/L)	9.25 ± 1.71	13.37 ± 1.16	20.81 ± 5.42	< 0.001	
AST (U/L)	22.01 ± 9.06	24.59 ± 5.35	27.78 ± 6.80	< 0.001	
TyG index	7.83 ± 0.42	7.85 ± 0.43	7.90 ± 0.49	0.029	
Cr (umol/L)	41.22 ± 13.23	39.49 ± 9.69	40.70 ± 12.26	0.044	
BUN (umol/L)	4.51 ± 1.12	4.56 ± 1.17	4.62 ± 1.19	0.303	
SUA (umol/L)	258.29 ± 64.26	266.33 ± 71.86	276.62 ± 71.48	< 0.001	
FPG (mg/dl)	85.12 ± 11.63	85.53 ± 13.45	86.20 ± 13.81	0.383	
TG (mg/dl)	56.69 (46.06–76.17)	58.46 (46.06–76.62)	60.23 (46.06–83.26)	0.005	
TC (mg/dl)	149.36 ± 29.27	150.60 ± 27.22	153.82 ± 28.83	0.029	
HDL (mg/dl)	55.44 ± 11.34	54.65 ± 10.98	54.40 ± 12.43	0.312	
LDL (mg/dl)	80.01 ± 21.22	81.96 ± 21.54	83.46 ± 23.27	0.036	
Pubertal stage				0.129	
In prepuberty (%)	404 (69.90%)	430 (72.76%)	394 (67.35%)		
In puberty (%)	174 (30.10%)	161 (27.24%)	191 (32.65%)		
Abbreviations: Height SDS: height standard deviation scores; BMI: body mass index; SBP: systolic blood pressure; DBP: diastolic blood pressure; IGF-1 SDS: insulin like growth factor-1 standard deviation scores; ALT: alanine aminotransferase; AST: aspartate aminotransferase; TyG index: triglyceride glucose index; Cr creatinine; BUN: blood urea nitrogen; SUA: serum uric acid; FPG: fasting plasma glucose; TG: triglyceride; TC: total cholesterol; HDL-C: high density lipoprotein-cholesterol; LDL-C: low density lipoprotein cholesterol. Continuous variables are presented as the mean ± standard deviation or median (interquartile range). Categorical variables are displayed as number (percentage). P < 0.05 is considered to be statistically significant.

In addition, the clinical characteristics of the GHD and non-GHD populations are presented in Supplementary Table 1. In the total population, the median (interquartile range) height SDS for the study population was −2.49 (−2.93–2.20). Nearly 68.02% of the study subjects were prepubescent. The mean ALT levels and TyG index were 14.51 ± 5.82 U/L and 7.86 ± 0.45, respectively. In the study population, there were 148 (8.44%) with elevated ALT, 113 (8.93%) patients in the GHD population and 35 (7.16%) patients in the non-GHD population.

Factors associated with ALT and elevated ALT in the study population

As shown in Fig. 2, logistic regression analysis indicated that age, bone age, puberty stage, body weight, BMI, SBP, DBP, SUA, TG and the TyG index were all unadjusted risk factors for elevated ALT levels, while peak GH and HDL-C were presented a protective effects (all P < 0.05). No association were detected between elevated ALT levels and height SDS, IGF-1 SDS, creatinine (Cr), BUN, FPG, TC, and LDL-C levels (all P > 0.05). In addition, the factors associated with elevated ALT in the GHD population were consistent with those in the total population (Supplementary Fig. 1). However, no association was observed between TyG index and elevated ALT in the non-GHD population (P = 0.464).Fig. 2 Logistic regression analysis of factors influencing elevated ALT levels in the total population.

In addition, we applied univariate analysis to analyze the factors associated with ALT levels (Table 2). The results showed that in the general population, age, bone age, body weight, BMI, SBP, DBP, SUA, TG and TyG index were positively associated with ALT (all P < 0.05), while height SDS and peak GH were negatively associated with ALT (all P < 0.05). No association was observed between IGF-1 SDS, creatinine (Cr), BUN, FPG, HDL-C, sex, puberty stage and ALT (all P > 0.05). In addition, as shown in Supplementary Table 2, the factors associated with ALT in the GHD population were consistent with those in the total population, except that HDL-C and ALT were negatively associated in the GHD population (P = 0.026). However, in the non-GHD population, only age was positively associated with ALT (P = 0.021), and no association was observed between other indicators and ALT (all P > 0.05).Table 2 Association between ALT and different variables.

Variables	All	
β (95% CI)	P value	
Age (years)	0.20 (0.13, 0.28)	< 0.001	
Bone age (years)	0.17 (0.09, 0.24)	< 0.001	
Height (cm)	0.04 (0.02, 0.05)	< 0.001	
Height SDS	−0.38 (−0.73, −0.02)	0.036	
Body weight (kg)	0.09 (0.06, 0.11)	< 0.001	
BMI (kg/m2)	0.36 (0.27, 0.44)	< 0.001	
IGF-1	0.01 (0.01, 0.01)	0.001	
IGF-1 SDS	0.08 (−0.14, 0.30)	0.495	
Peak GH (ng/mL)	−0.15 (−0.23, −0.07 )	< 0.001	
SBP (mmHg)	0.08 (0.06, 0.10)	< 0.001	
DBP (mmHg)	0.07 (0.04, 0.10)	< 0.001	
TyG index	1.36 (0.74, 1.98)	< 0.001	
Cr (umol/L)	0.01 (−0.02, 0.03)	0.607	
BUN (umol/L)	0.14 (−0.09, 0.38)	0.236	
SUA (umol/L)	0.01 (0.01, 0.01)	< 0.001	
FPG (mg/dl)	0.01 (−0.01, 0.04)	0.191	
TG (mg/dl)	0.02 (0.01, 0.03)	< 0.001	
TC (mg/dl)	0.02 (0.01, 0.03)	0.001	
HDL (mg/dl)	−0.02 (−0.05, 0.01)	0.069	
LDL (mg/dl)	0.02 (0.01, 0.03)	0.006	
Sex	
Male	Reference		
Female	−0.50 (−1.08, 0.09)	0.098	
Pubertal stage	
In prepuberty (%)	Reference		
In puberty (%)	0.41 (−0.19, 1.01)	0.184	
Abbreviations: Height SDS: height standard deviation scores; BMI: body mass index; SBP: systolic blood pressure; DBP: diastolic blood pressure; IGF-1 SDS: insulin like growth factor-1 standard deviation scores; ALT: alanine aminotransferase; TyG index: triglyceride glucose index; Cr creatinine; BUN: blood urea nitrogen; SUA serumuric acid; FPG: fasting plasma glucose; TG: triglyceride; TC: total cholesterol; HDL-C: high density lipoprotein-cholesterol; LDL-C: low density lipoprotein cholesterol. GHD: growth hormone deficiency. P < 0.05 is considered to be statistically significant.

The nonlinear relationship between the TyG index and ALT

In this study, smooth curve fitting was performed to investigate whether the TyG index and ALT were linearly related, and the results of the study revealed that the relationship between the TyG index and ALT was nonlinear after adjusting for potential confounding factors, including age, sex, BMI, peak GH, SBP, DBP, SUA, TG and TC. An inflection point was observed in the present study, indicating two stages of change between the TyG index and ALT (Fig. 3). Furthermore, according to the P value of the log-likelihood ratio test, a one-linear regression and two-piecewise linear regression were performed to determine the best fitting model.Fig. 3 Smooth curve fitting of the relationship between the TyG index and ALT. Adjustment variables: age, sex, BMI, SBP, DBP, IGF-1, peak GH, SUA and TC. BMI: body mass index, SBP: systolic blood pressure; DBP: diastolic blood pressure; GH: growth hormone; SUA serumuric acid; TC: total cholesterol; TyG index: triglyceride glucose index; ALT: alanine aminotransferase.

Multivariate regression analysis

As shown in Table 3, linear multiple regression results in the total population showed a significant positive association between the TyG index and ALT. According to the nonlinear relationship suggested by the smooth curve fitting, it was further found by piecewise linear regression analysis that the TyG index inflection point was 8.24. Only when the TyG index was greater than 8.24 was there a significant positive association between the TyG index and ALT (β 5.75, 95% CI 3.30, 8.19; P < 0.001). However, when the TyG index was less than 8.24, there was no significant association between the TyG index and ALT (β −0.57, 95% CI −1.84, 0.71; P = 0.382).Table 3 Threshold effect analysis for the relationship between the TyG index and ALT.

Models	ALT	
Adjusted β (95%CI)	P value	
Model I	
One line slope	1.21 (0.29, 2.13)	0.010	
Model II	
Turning point	8.24		
< 8.24 slope 1	−0.57 (−1.84, 0.71)	0.382	
> 8.24 slope 2	5.75 (3.30, 8.19)	< 0.001	
LRT test	< 0.001		
Model I, linear analysis; Model II, non-linear analysis. LRT test, Logarithmic likelihood ratio test. (p value < 0.05 means Model II is significantly different from Model I, which indicates a non-linear relationship); Adjustment variables: age, sex, BMI, SBP, DBP, IGF-1, peak GH, SUA and TC. BMI: body mass index, SBP: systolic blood pressure; DBP: diastolic blood pressure; IGF-1: insulin like growth factor-1; GH: growth hormone; SUA serumuric acid; TC: total cholesterol; TyG index: triglyceride glucose index; ALT: alanine aminotransferase. P < 0.05 is considered to be statistically significant.

In addition, subgroup analysis was further conducted in this study. In both the GHD population and the non-GHD population, the TyG index and ALT showed a nonlinear relationship (Supplementary Fig. 2), and the TyG index inflection points were 7.35 and 8.43, respectively. As shown in Supplementary Table 3, in the GHD population, at TyG index levels higher than 7.35, the ALT gradually increased with increasing TyG index (β = 5.42, 95% CI: 2.96, 7.88; P < 0.001), but no association was found when the TyG index was less than 7.35 (β = −0.81, 95% CI: −2.48, 0.86; P = 0.344). Moreover, a similar relationship was found among the non-GHD population. ALT levels increased as the TyG index increased when the TyG index was greater than 8.43 (β = 6.09, 95% CI: 1.06, 9.83; P = 0.038), yet no significant relationship was observed at lower TyG index values (β = −0.81, 95% CI: −2.78, 1.15; P = 0.418).

Discussion

This population-based study revealed that the TyG index is strongly positively associated with ALT in children with short stature. More interestingly, we found a nonlinear relationship between the TyG index and ALT. An inflection point was observed in the present study, indicating two stages of change between the TyG index and ALT. In addition, further subgroup analysis showed that there was a nonlinear relationship between the TyG index and ALT in both GHD and non-GHD populations. When the TyG index exceeds a certain value, the ALT level increases with increasing TyG index.

The TyG index is an indicator of IR, usually calculated from TG and FPG levels. An increase in the TyG index indicates an increase in IR9. ALT is an enzyme found in liver cells, and ALT levels rise when liver cells are damaged or inflamed. Previous studies have shown that ALT levels are a surrogate marker for MAFLD in the absence of other causes of liver disease12. The present study found that the prevalence of elevated ALT levels in children and adolescents with short stature was 8.44%. A previous study analyzed 5586 adolescents aged 12–19 years in the National Health and Nutrition Examination Survey (NHANES) 1999-2004 to assess the prevalence of elevated ALT levels in contemporary adolescents. The findings revealed that the prevalence of elevated ALT levels was 8.0% in US adolescents aged 12–19 years6. In addition, Mischel et al. analyzed 6083 adolescents aged 12 to 19 years who were assessed with NHANES from 2011 to 2018 and found that the incidence of elevated ALT was 8.8% in normal-weight adolescents24.

The liver, as an important metabolic organ, plays a crucial role in regulating glucose and lipid metabolism. IR and abnormal glucose and lipid metabolism, as important factors in MAFLD, interact and jointly promote MAFLD8,25. The present study reveals consistent findings with previous studies across different populations14–16. Lertsakulbunlue et al. demonstrated a dose‒response relationship between the TyG index and aminotransferase by conducting a series of cross-sectional studies on 232,235 Royal Thai Army (RTA) personnel aged between 35 and 60 years. It has been suggested that a high TyG index is a novel risk factor for elevated aminotransferase14. In addition, a recent study in healthy obese people described the relationship between the TyG index and liver function indicators, and the results revealed a positive correlation between the TyG index and ALT, suggesting that in obese people, an increase in the TyG index may be a predictor and early diagnostic indicator of MAFLD15. Yu et al. explored the relationship between abnormal liver function defined by elevated ALT levels and the TyG index in urban and rural Chinese adults and found that the TyG index was associated with abnormal liver function in urban and rural Chinese adults. The TyG index can be used as an alternative marker of IR severity to identify adults at high risk of liver function impairment, especially urban Chinese adults16.

Interestingly, this study identified a nonlinear relationship between the TyG index and ALT, with an inflection point at 8.24. This is similar to the findings from a cross-sectional study of a health examination cohort of 10,761 people older than 20 years in China, which showed that TyG was effective in identifying people at high risk for MAFLD. When the TyG index was greater than 8.5, the sensitivity of the detection of MAFLD subjects was very high and thus may be suitable as a diagnostic standard of MAFLD in Chinese adults26. In addition, a previous study showed that IR is associated with liver fat accumulation and inflammation. When the TyG index is above the normal range, it may represent a high level of IR, which in turn leads to liver fat accumulation and inflammation, resulting in increased ALT levels27. Our analysis further indicates that TG was likely the primary factor driving the nonlinear relationship between the TyG index and ALT. The accumulation of TG in liver cells is considered a hallmark of MAFLD28. Although the liver is not naturally designed for fat storage, it can become a site of fat accumulation when there is excessive intake of fatty acids or carbohydrates, or increased endogenous fat synthesis. This disruption in lipid metabolism may potentially lead to MAFLD. Additionally, IR might exacerbate the liver’s absorption of fatty acids and total cholesterol through mechanisms such as reduced glucose uptake in skeletal muscles, increased release of free fatty acids from adipose tissue, and enhanced glycolysis27.

We further performed subgroup analyses for individuals with and without GHD, and piecewise linear regression showed that the inflection point of the TyG index was higher in non-GHD populations than in GHD populations. This suggests that GH levels can affect the relationship between the TyG index and ALT. Previously, this research team found that ALT is correlated with GH levels29, and the activity of the GH/IGF-1 axis is significantly negatively correlated with MAFLD30. This axis plays a crucial role in regulating various metabolic processes, and reduced activity of this axis is associated with an increased risk of MAFLD31. These findings are significant as they highlight the complex interplay between hormonal regulation and liver health. The GH/IGF-1 axis not only affects growth and development but also plays a critical role in metabolic balance. The negative correlation between the GH/IGF-1 axis and MAFLD suggests that disturbances in this hormonal axis may contribute to the pathogenesis of liver diseases, including fatty liver.

There are several possible underlying mechanisms between IR and the development of MAFLD30,32,33: (1) IR fails to inhibit both lipolysis and the production of triglyceride-rich very low-density lipoprotein particles through the liver. Large triglyceride-rich very low-density lipoprotein particles may, in turn, exacerbate hepatic IR and increase serum triglyceride levels. (2) IR impairs the ability of insulin to inhibit gluconeogenesis and glucose uptake, leading to hyperglycemia, which in turn stimulates insulin secretion and leads to hyperinsulinemia. Fasting hyperglycemia and hyperinsulinemia are directly related to liver fat. (3) IR can cause changes in the gut microbiota, resulting in the production of higher levels of short-chain fatty acids (SCFAs), alteration of the hepatic-intestinal cycle of bile acids, and ultimately, inflammation and hepatic steatosis. (4) Other mechanisms linking insulin resistance to MAFLD may be related to adipose tissue inflammation, increased serum free fatty acids, impaired mitochondrial fatty acid b-oxidation, unfavorable body fat distribution, and altered levels of adipokines and cytokines.

However, there are also several limitations. First, causal inferences cannot be drawn due to the cross-sectional design. Second, the present findings are only based on children with short stature, and different results might be observed in other groups. Therefore, future research could explore the relationship between the TyG index and ALT in different ethnic populations to ensure that our results are applicable to a broader range of individual. Third, other potential macronutrients and confounders were not included. However, since we adjusted for crucial confounders, we consider the present results to be reliable. Finally, there may have been interobserver variability in assessing anthropometric measurements, but the anthropometric measurements were performed by one of two trained professionals.

In conclusion, this study demonstrated a nonlinear relationship between the TyG index and ALT in children with short stature. This finding suggests that a high TyG index is associated with elevated ALT in children with short stature. Further subgroup analyses showed that although the nonlinear relationship remained in the GHD and non-GHD populations, the inflection point of the TyG index was higher in the non-GHD population than in the GHD population, indicating that more attention should be given to the TyG index and ALT levels in the GHD population.

Materials and methods

Study population

This study is a cross-sectional analysis of a cohort study involving patients meeting the criteria for short stature from the Affiliated Hospital of Jining Medical College from January 2013 to December 2022. A total of 1754 children and adolescents with short stature (1193 males and 561 females) with an average age of 10.4 ± 3.6 years were enrolled. The inclusion criteria were as follows: age between 3 and 18 years; and a height standard deviation score (SDS) less than or equal to −2 SDS for normal height individuals of the same age and sex. The exclusion criteria were as follows: children with skeletal dysplasia; chronic diseases, including long-term nutritional deficiencies; thyroid dysfunction; or other known causes of short stature, including Noonan syndrome and Turner syndrome. In addition, children with missing alanine aminotransferase or lipid/blood glucose data were also excluded.

Ethics

The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Human Ethics Committee of the Affiliated Hospital of Jining Medical University (Shandong, China). All of the families of the patients were informed of the aims of the study, and a written informed consent form was signed by all of the participants’ parents.

Clinical measurements

As described in our previous study34, physical examination was performed, and anthropometry was obtained, comprising height, weight, blood pressure and pubertal stage. BMI was calculated as weight (in kilograms)/height square (in meters). Detailed descriptions of anthropometric measurements and laboratory testing methods have been provided in the previously published article35. Height and weight were measured by a designated individual using the same measuring instrument in the morning. Puberty stage was evaluated by one of two pretrained physicians during a physical examination, based on the Tanner stages35. Overnight fasting blood samples (≥ 10 h) were collected and analyzed for biochemical measurements, such as ALT, AST, FPG, TG, TC, HDL-C, LDL-C, Cr, BUN and SUA. All measurements were determined by an autoanalyzer (Cobas c702, Roche; Shanghai, China)35. The interassay coefficient of variation for ALT was 5.3%, for AST was 5.3%, for FPG was 3.0%, for TG was 8.3%, for TC was 3.3%, for Cr was 4.0%, for BUN was 2.6%, and for SUA was 4.0%. A chemiluminescence assay with intra- and interassay parameters with coefficients of variation of 3.0 and 6.2%, respectively (DPC IMMULITE 1000 Analyzer, SIEMENS, Berlin, Germany), was performed to detect serum IGF-1 levels35. Two stimulation tests were performed for GH, and a chemiluminescence method was used to assess GH concentration (ACCESS2, Beckman Coulter; USA)35. GH peak value < 10 ng/mL is categorized as GHD, while GH peak value ≥ 10 ng/mL is categorized as non-GHD. The product of triglycerides and glucose was calculated using established formulas: TyG index = Ln [TG(mg/dl) FPG (mg/dl)/2]10. Elevated ALT levels were defined using the biological thresholds of > 22 U/L in females and > 26 U/L in males25.

Statistical analysis

Continuous variables are expressed as the mean ± standard deviation or median (interquartile). Categorical variables are expressed as numbers and percentages. We used the chi-square test (categorical variables), Student’s t test (normal distribution), or the Kruskal–Wallis H test (skewed distribution) to test for differences among the GHD and non-GHD groups. One-way ANOVA was used for continuous variables and chi-square tests for categorical variables to summarize and compare the characteristics of participant characteristics across tertiles of ALT. Univariate analysis and multiple regression analysis were used to explore the relationship between the TyG index and ALT. Then, to assess the nonlinear relationship between the TyG index and ALT, smooth curve fitting was carried out. Finally, according to the smooth curve, multivariate fragment linear regression was further applied to estimate the correlation between the TyG index and ALT threshold. The log-likelihood ratio test compares the single-line linear regression model with the two-segment linear model to test for statistical significance. In all analyses, a two-tailed P < 0.05 was considered indicative of statistical significance. Statistical analysis was performed with R 4.2.1 (https://www.R-project.org) and EmpowerStats (https://www.empowerstats.com; X&Y Solutions, Inc).

Supplementary Information

Supplementary Information 1.

Supplementary Information 2.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-024-71608-8.

Author contributions

Q. Z. carried out the studies and drafted the manuscript. YQ. L. helped with the statistical analysis. M. Z. revised the manuscript. B.B. participated in the study concept and design, revised the article critically for important intellectual content. All authors read and approved the final manuscript.

Funding

This work was supported by the Natural Science Foundation of Shandong Province (no. ZR2022MH284).

Data availability

All data generated or analyzed during this study are included in this published article and its supplementary information files.

Competing interests

The authors declare no competing interests.

Publisher's note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References

1. Kumar R Priyadarshi RN Anand U Non-alcoholic fatty liver disease: Growing burden, adverse outcomes and associations J. Clin. Transl. Hepatol. 2020 8 76 86 32274348
Kumar, R., Priyadarshi, R. N. & Anand, U. Non-alcoholic fatty liver disease: Growing burden, adverse outcomes and associations. J. Clin. Transl. Hepatol. 8, 76–86 (2020).32274348
2. Ng CH Huang DQ Nguyen MH Nonalcoholic fatty liver disease versus metabolic-associated fatty liver disease: Prevalence outcomes and implications of a change in name Clin. Mol. Hepatol. 2022 28 790 801 10.3350/cmh.2022.0070 35545437
Ng, C. H., Huang, D. Q. & Nguyen, M. H. Nonalcoholic fatty liver disease versus metabolic-associated fatty liver disease: Prevalence outcomes and implications of a change in name. Clin. Mol. Hepatol. 28, 790–801 (2022).35545437 10.3350/cmh.2022.0070
3. Nobili V NAFLD in children: New genes, new diagnostic modalities and new drugs Nat. Rev. Gastroenterol. Hepatol. 2019 16 517 530 10.1038/s41575-019-0169-z 31278377
Nobili, V. et al. NAFLD in children: New genes, new diagnostic modalities and new drugs. Nat. Rev. Gastroenterol. Hepatol. 16, 517–530 (2019).31278377 10.1038/s41575-019-0169-z
4. Rinella ME AASLD Practice guidance on the clinical assessment and management of nonalcoholic fatty liver disease Hepatology 2023 77 1797 1835 10.1097/HEP.0000000000000323 36727674
Rinella, M. E. et al. AASLD Practice guidance on the clinical assessment and management of nonalcoholic fatty liver disease. Hepatology 77, 1797–1835 (2023).36727674 10.1097/HEP.0000000000000323
5. Vos MB NASPGHAN clinical practice guideline for the diagnosis and treatment of nonalcoholic fatty liver disease in children: Recommendations from the expert committee on NAFLD (ECON) and the North American society of pediatric gastroenterology, hepatology and nutrition (NASPGHAN) J. Pediatr. Gastroenterol. Nutr. 2017 64 319 334 10.1097/MPG.0000000000001482 28107283
Vos, M. B. et al. NASPGHAN clinical practice guideline for the diagnosis and treatment of nonalcoholic fatty liver disease in children: Recommendations from the expert committee on NAFLD (ECON) and the North American society of pediatric gastroenterology, hepatology and nutrition (NASPGHAN). J. Pediatr. Gastroenterol. Nutr. 64, 319–334 (2017).28107283 10.1097/MPG.0000000000001482
6. Fraser A Longnecker MP Lawlor DA Prevalence of elevated alanine aminotransferase among US adolescents and associated factors: NHANES 1999–2004 Gastroenterology 2007 133 1814 1820 10.1053/j.gastro.2007.08.077 18054554
Fraser, A., Longnecker, M. P. & Lawlor, D. A. Prevalence of elevated alanine aminotransferase among US adolescents and associated factors: NHANES 1999–2004. Gastroenterology 133, 1814–1820 (2007).18054554 10.1053/j.gastro.2007.08.077
7. Smith GI Insulin resistance drives hepatic de novo lipogenesis in nonalcoholic fatty liver disease J. Clin. Invest. 2020 130 1453 1460 10.1172/JCI134165 31805015
Smith, G. I. et al. Insulin resistance drives hepatic de novo lipogenesis in nonalcoholic fatty liver disease. J. Clin. Invest. 130, 1453–1460 (2020).31805015 10.1172/JCI134165
8. Khan RS Bril F Cusi K Newsome PN Modulation of insulin resistance in nonalcoholic fatty liver disease Hepatology 2019 70 711 724 10.1002/hep.30429 30556145
Khan, R. S., Bril, F., Cusi, K. & Newsome, P. N. Modulation of insulin resistance in nonalcoholic fatty liver disease. Hepatology 70, 711–724 (2019).30556145 10.1002/hep.30429
9. Guerrero-Romero F The product of triglycerides and glucose, a simple measure of insulin sensitivity. Comparison with the euglycemic-hyperinsulinemic clamp J. Clin. Endocrinol. Metab. 2010 95 3347 3351 10.1210/jc.2010-0288 20484475
Guerrero-Romero, F. et al. The product of triglycerides and glucose, a simple measure of insulin sensitivity. Comparison with the euglycemic-hyperinsulinemic clamp. J. Clin. Endocrinol. Metab. 95, 3347–3351 (2010).20484475 10.1210/jc.2010-0288
10. Simental-Mendía LE Rodríguez-Morán M Guerrero-Romero F The product of fasting glucose and triglycerides as surrogate for identifying insulin resistance in apparently healthy subjects Metab. Syndr. Relat. Disord. 2008 6 299 304 10.1089/met.2008.0034 19067533
Simental-Mendía, L. E., Rodríguez-Morán, M. & Guerrero-Romero, F. The product of fasting glucose and triglycerides as surrogate for identifying insulin resistance in apparently healthy subjects. Metab. Syndr. Relat. Disord. 6, 299–304 (2008).19067533 10.1089/met.2008.0034
11. Rodríguez-Hernández H Simental-Mendía LE The triglycerides and glucose index is highly associated with non-alcoholic fatty liver disease in overweight and obese women Ir. J. Med. Sci. 2023 192 2741 2746 10.1007/s11845-023-03335-4 36928593
Rodríguez-Hernández, H. & Simental-Mendía, L. E. The triglycerides and glucose index is highly associated with non-alcoholic fatty liver disease in overweight and obese women. Ir. J. Med. Sci. 192, 2741–2746 (2023).36928593 10.1007/s11845-023-03335-4
12. Xue Y Xu J Li M Gao Y Potential screening indicators for early diagnosis of NAFLD/MAFLD and liver fibrosis: Triglyceride glucose index-related parameters Front. Endocrinol. (Lausanne) 2022 13 951689 10.3389/fendo.2022.951689 36120429
Xue, Y., Xu, J., Li, M. & Gao, Y. Potential screening indicators for early diagnosis of NAFLD/MAFLD and liver fibrosis: Triglyceride glucose index-related parameters. Front. Endocrinol. (Lausanne) 13, 951689 (2022).36120429 10.3389/fendo.2022.951689
13. Beran A Triglyceride-glucose index for early prediction of nonalcoholic fatty liver disease: A meta-analysis of 121,975 individuals J. Clin. Med. 2022 11 2666 10.3390/jcm11092666 35566790
Beran, A. et al. Triglyceride-glucose index for early prediction of nonalcoholic fatty liver disease: A meta-analysis of 121,975 individuals. J. Clin. Med. 11, 2666 (2022).35566790 10.3390/jcm11092666
14. Lertsakulbunlue S Mungthin M Rangsin R Kantiwong A Sakboonyarat B Relationship between triglyceride-glucose index and aminotransferase among royal Thai army personnel 2017–2021: A serial cross-sectional study Lipids Health Dis 2023 22 47 10.1186/s12944-023-01811-5 37013603
Lertsakulbunlue, S., Mungthin, M., Rangsin, R., Kantiwong, A. & Sakboonyarat, B. Relationship between triglyceride-glucose index and aminotransferase among royal Thai army personnel 2017–2021: A serial cross-sectional study. Lipids Health Dis 22, 47 (2023).37013603 10.1186/s12944-023-01811-5
15. Pan X Yue L Ren L Ban J Chen S Association of triglyceride-glucose index and liver function parameters among healthy obese civil servants: A center-based study Diabetes Metab. Syndr. Obes. 2022 15 3519 3531 10.2147/DMSO.S392544 36407008
Pan, X., Yue, L., Ren, L., Ban, J. & Chen, S. Association of triglyceride-glucose index and liver function parameters among healthy obese civil servants: A center-based study. Diabetes Metab. Syndr. Obes. 15, 3519–3531 (2022).36407008 10.2147/DMSO.S392544
16. Yu L Cai Y Qin R Zhao B Li X Association between triglyceride glucose index and abnormal liver function in both urban and rural Chinese adult populations: Findings from two independent surveys Medicine (Baltimore) 2019 98 e18265 10.1097/MD.0000000000018265 31852096
Yu, L., Cai, Y., Qin, R., Zhao, B. & Li, X. Association between triglyceride glucose index and abnormal liver function in both urban and rural Chinese adult populations: Findings from two independent surveys. Medicine (Baltimore) 98, e18265 (2019).31852096 10.1097/MD.0000000000018265
17. Tan EX Non-obese non-alcoholic fatty liver disease (NAFLD) in Asia: An international registry study Metabolism 2022 126 154911 10.1016/j.metabol.2021.154911 34648769
Tan, E. X. et al. Non-obese non-alcoholic fatty liver disease (NAFLD) in Asia: An international registry study. Metabolism 126, 154911 (2022).34648769 10.1016/j.metabol.2021.154911
18. Zou Y Yu M Sheng G Association between fasting plasma glucose and nonalcoholic fatty liver disease in a nonobese Chinese population with normal blood lipid levels: A prospective cohort study Lipids Health Dis. 2020 19 145 10.1186/s12944-020-01326-3 32563249
Zou, Y., Yu, M. & Sheng, G. Association between fasting plasma glucose and nonalcoholic fatty liver disease in a nonobese Chinese population with normal blood lipid levels: A prospective cohort study. Lipids Health Dis. 19, 145 (2020).32563249 10.1186/s12944-020-01326-3
19. Kumari S Height predict incident non-alcoholic fatty liver disease among general adult population in Tianjin; China; independent of body mass index; waist circumference; waist-to-height ratio; and metabolic syndrome BMC Public Health 2020 20 388 10.1186/s12889-020-08475-1 32209063
Kumari, S. et al. Height predict incident non-alcoholic fatty liver disease among general adult population in Tianjin; China; independent of body mass index; waist circumference; waist-to-height ratio; and metabolic syndrome. BMC Public Health 20, 388 (2020).32209063 10.1186/s12889-020-08475-1
20. Dichtel LE The GH/IGF-1 axis is associated with intrahepatic lipid content and hepatocellular damage in overweight/obesity J. Clin. Endocrinol. Metab. 2022 107 e3624 e3632 10.1210/clinem/dgac405 35779256
Dichtel, L. E. et al. The GH/IGF-1 axis is associated with intrahepatic lipid content and hepatocellular damage in overweight/obesity. J. Clin. Endocrinol. Metab. 107, e3624–e3632 (2022).35779256 10.1210/clinem/dgac405
21. Dichtel LE The association between IGF-1 levels and the histologic severity of nonalcoholic fatty liver disease Clin. Transl. Gastroenterol. 2017 8 e217 10.1038/ctg.2016.72 28125073
Dichtel, L. E. et al. The association between IGF-1 levels and the histologic severity of nonalcoholic fatty liver disease. Clin. Transl. Gastroenterol. 8, e217 (2017).28125073 10.1038/ctg.2016.72
22. Xu L Association between serum growth hormone levels and nonalcoholic fatty liver disease: A cross-sectional study PLoS ONE. 2012 7 e44136 10.1371/journal.pone.0044136 22952901
Xu, L. et al. Association between serum growth hormone levels and nonalcoholic fatty liver disease: A cross-sectional study. PLoS ONE. 7, e44136 (2012).22952901 10.1371/journal.pone.0044136
23. Ranke MB Wit JM Growth hormone–past; present and future Nat. Rev. Endocrinol. 2018 14 285 330 10.1038/nrendo.2018.22 29546874
Ranke, M. B. & Wit, J. M. Growth hormone–past; present and future. Nat. Rev. Endocrinol. 14, 285–330 (2018).29546874 10.1038/nrendo.2018.22
24. Mischel AK Prevalence of elevated ALT in adolescents in the US 2011–2018 J. Pediatr. Gastroenterol. Nutr. 2023 77 103 109 10.1097/MPG.0000000000003795 37084344
Mischel, A. K. et al. Prevalence of elevated ALT in adolescents in the US 2011–2018. J. Pediatr. Gastroenterol. Nutr. 77, 103–109 (2023).37084344 10.1097/MPG.0000000000003795
25. Watt MJ Miotto PM De Nardo W Montgomery MK The liver as an endocrine organ-linking NAFLD and insulin resistance Endocr. Rev. 2019 40 1367 1393 10.1210/er.2019-00034 31098621
Watt, M. J., Miotto, P. M., De Nardo, W. & Montgomery, M. K. The liver as an endocrine organ-linking NAFLD and insulin resistance. Endocr. Rev. 40, 1367–1393 (2019).31098621 10.1210/er.2019-00034
26. Zhang S The triglyceride and glucose index (TyG) is an effective biomarker to identify nonalcoholic fatty liver disease Lipids Health Dis. 2017 16 15 10.1186/s12944-017-0409-6 28103934
Zhang, S. et al. The triglyceride and glucose index (TyG) is an effective biomarker to identify nonalcoholic fatty liver disease. Lipids Health Dis. 16, 15 (2017).28103934 10.1186/s12944-017-0409-6
27. Alam S Mustafa G Alam M Ahmad N Insulin resistance in development and progression of nonalcoholic fatty liver disease World J. Gastrointest. Pathophysiol. 2016 7 211 217 10.4291/wjgp.v7.i2.211 27190693
Alam, S., Mustafa, G., Alam, M. & Ahmad, N. Insulin resistance in development and progression of nonalcoholic fatty liver disease. World J. Gastrointest. Pathophysiol. 7, 211–217 (2016).27190693 10.4291/wjgp.v7.i2.211
28. Kawano Y Cohen DE Mechanisms of hepatic triglyceride accumulation in nonalcoholic fatty liver disease J. Gastroenterol. 2013 48 434 441 10.1007/s00535-013-0758-5 23397118
Kawano, Y. & Cohen, D. E. Mechanisms of hepatic triglyceride accumulation in nonalcoholic fatty liver disease. J. Gastroenterol. 48, 434–441 (2013).23397118 10.1007/s00535-013-0758-5
29. Ji B Association between alanine aminotransferase and growth hormone: A retrospective cohort study of short children and adolescents Biomed. Res. Int. 2019 2019 1 7
Ji, B. et al. Association between alanine aminotransferase and growth hormone: A retrospective cohort study of short children and adolescents. Biomed. Res. Int. 2019, 1–7 (2019).
30. Liang S Cheng X Hu Y Song R Li G Insulin-like growth factor 1 and metabolic parameters are associated with nonalcoholic fatty liver disease in obese children and adolescents Acta Paediatr. 2017 106 298 303 10.1111/apa.13685 27889912
Liang, S., Cheng, X., Hu, Y., Song, R. & Li, G. Insulin-like growth factor 1 and metabolic parameters are associated with nonalcoholic fatty liver disease in obese children and adolescents. Acta Paediatr. 106, 298–303 (2017).27889912 10.1111/apa.13685
31. Ma IL Stanley TL Growth hormone and nonalcoholic fatty liver disease Immunometabolism (Cobham). 2023 5 e00030 10.1097/IN9.0000000000000030 37520312
Ma, I. L. & Stanley, T. L. Growth hormone and nonalcoholic fatty liver disease. Immunometabolism (Cobham). 5, e00030 (2023).37520312 10.1097/IN9.0000000000000030
32. Pedersen HK Human gut microbes impact host serum metabolome and insulin sensitivity Nature 2016 535 376 381 10.1038/nature18646 27409811
Pedersen, H. K. et al. Human gut microbes impact host serum metabolome and insulin sensitivity. Nature 535, 376–381 (2016).27409811 10.1038/nature18646
33. Yki-Järvinen H Non-alcoholic fatty liver disease as a cause and a consequence of metabolic syndrome Lancet Diabet. Endocrinol. 2014 2 901 910 10.1016/S2213-8587(14)70032-4
Yki-Järvinen, H. Non-alcoholic fatty liver disease as a cause and a consequence of metabolic syndrome. Lancet Diabet. Endocrinol. 2, 901–910 (2014).10.1016/S2213-8587(14)70032-4
34. Li G Zhao Q Zhang X Ban B Zhang M Association between the uric acid to high density lipoprotein cholesterol ratio and alanine transaminase in Chinese short stature children and adolescents: A cross-sectional study Front. Nutr. 2023 10 1063534 10.3389/fnut.2023.1063534 36761217
Li, G., Zhao, Q., Zhang, X., Ban, B. & Zhang, M. Association between the uric acid to high density lipoprotein cholesterol ratio and alanine transaminase in Chinese short stature children and adolescents: A cross-sectional study. Front. Nutr. 10, 1063534 (2023).36761217 10.3389/fnut.2023.1063534
35. Zhao Q Chu Y Pan H Zhang M Ban B Association between triglyceride glucose index and peak growth hormone in children with short stature Sci. Rep. 2021 11 1 1969 10.1038/s41598-021-81564-2 33479436
Zhao, Q., Chu, Y., Pan, H., Zhang, M. & Ban, B. Association between triglyceride glucose index and peak growth hormone in children with short stature. Sci. Rep. 11(1), 1969 (2021).33479436 10.1038/s41598-021-81564-2
