Introduction

Metabolic syndrome (MetS) includes a cluster of conditions associated with metabolic dysregulation, such as insulin resistance (IR), atherogenic dyslipidemia, central obesity, and hypertension. Cumulative evidence has confirmed that IR and persistent low-grade inflammation are the primary pathogenic factors of MetS [1]. SPECT-China, a population-based cross-sectional survey of Chinese individuals ≥ 18 years of age, reveals a 22.0% age-standardized prevalence of MetS in East China [2]. Similarly, our previous study found that MetS has gradually become more prevalent among participants from rural China (39.0%) [3]. If left untreated, MetS considerably increases morbidity and mortality [1]. It is thus crucial to investigate an effective predictor of mortality in patients with MetS to reduce the substantial disease burden.

Recently, the triglyceride-glucose (TyG) index has garnered increasing global attention, as it is associated with all-cause and cardiovascular (CV) mortality in the general population [2), encompassing both work-related and recreational pursuits, was evaluated and divided into three categories [3].

Fig. 2
figure 2

Categories of physical activity

Full size image

Definition

The TyG index was calculated using the following equation: ln [fasting triglycerides (mg/dL) × fasting glucose (mg/dL)/2] [9]. Body mass index (BMI) was determine using the following equation: BMI = \(\frac{weight \left(kg\right)}{{\left(height\right)}^{2} \left(m\right)}\). Figure 3 provides the definition of the ATPIII-modified criteria [10].

Fig. 3
figure 3

Definition of metabolic syndrome and metabolic disorders

Full size image

Statistical analyses

The individuals were divided into three groups based on their TyG index levels, with each group representing one-third of the total participants, as follows: tertile 1 (n = 1309, TyG index < 8.92), tertile 2 (n = 1299, 8.92 ≤ TyG index < 8.92–9.36), and tertile 3 (n = 1310, TyG index ≥ 9.36), with the depicted characteristics. Categorical variables have been quantified using numerical values (n) and percentages (%) and were evaluated using the chi-squared test. Continuous variables with a normal distribution are expressed as the mean ± standard deviation; non-normal distributions are represented as the median (interquartile range). Data were analyzed using one-way analysis of variance for a normal distribution and the Kruskal–Wallis test for a non-normal distribution. We investigated the precise correlation between the TyG index and mortality from all causes and CVDs in people with MetS using multivariate Cox proportional hazards models. Three sets of models were constructed in this study. Model 1 included only the TyG index, whereas Model 2 incorporated demographic attributes such as age, sex, and race. Model 3 further adjusted for education, current smoking and drinking habits, annual income, sleep duration, physical activity, and cardiovascular history. A stratified analysis was conducted to examine the impact of putative effect modifiers, including age, sex, BMI, and current smoking and drinking status, on relevant variables. The statistical analyses were conducted using R software (version 4.2.1; R Foundation for Statistical Computing, Vienna, Austria), and statistical significance was defined as P < 0.05.

Results

Population characteristics stratified using the TyG index

As presented in Table 1, baseline participant characteristics were stratified using TyG index tertiles (Q) as follows: Q1, < 8.92; Q2, 8.92–9.36, and Q3 ≥ 9.36. The mean TyG index levels in these tertiles were 8.54 ± 0.29, 9.13 ± 0.122, and 9.87 ± 0.48, respectively. The participants with higher TyG index values were generally male, older, and current smokers and drinkers, had a longer sleep duration, and exhibited a higher prevalence of CV comorbidities. Additionally, they experienced a higher incidence of all-cause mortality than those with lower TyG index values (4.1% vs. 4.7% vs. 6.2%, P = 0.045). Moreover, the higher TyG index group exhibited higher levels of HbA1C, LDL-C, TC, TG, uric acid, FPG, AST, and ALT, along with lower eGFR and HDL-C levels (Table 2).

Table 1 Baseline characteristics of study participants by Triglyceride glucose Index (TyG) index tertile
Full size table
Table 2 Baseline levels of laboratory characteristics according to the Triglyceride glucose Index (TyG) index quartiles
Full size table

Associations between the TyG index and all‑cause and CV mortality

Cox proportional hazard analysis revealed a significant association between the TyG index and all-cause mortality, but not CV mortality, in both crude and adjusted models [crude: all-cause mortality HR, 1.303 (95% CI, 1.061–1.600), P = 0.012; CV mortality HR, 1.224 (95% CI, 0.923–1.623), P = 0.160]; Model 2: all-cause mortality HR, 1.330 (95% CI, 1.075–1.646), P = 0.009; CV mortality HR, 1.249 (95% CI, 0.931–1.674), P = 0.138; Model 3: all-cause mortality HR, 1.288 (1.033–1.605), P = 0.025; CV mortality HR, 1.194 (95% CI, 0.879–1.622), P = 0.257]. In both the crude and adjusted models, upward trends were observed between the TyG index and all-cause mortality (Table 3, both P < 0.05). The participants in the Q3 TyG index tertile had a significantly higher incidence of all-cause mortality than those in the Q1 and Q2 tertiles [HR, 1.441 (95% CI, 1.009–2.059), P = 0.016].

Table 3 HRs (95% CIs) for mortality according to the triglyceride glucose index (TyG) index tertiles
Full size table

Subgroup analysis of the association between the TyG index and all‑cause and CV mortality

Stratification was performed based on age, sex, BMI, and current smoking and drinking status to evaluate the impact of the TyG index on the primary endpoints (Fig. 4). Except for the age subgroup (age subgroup: all-cause mortality, P for interaction < 0.001), no significant interactions were observed in most subgroups. The TyG index was associated with all-cause mortality in patients < 65 years of age [all-cause mortality: HR, 1.374 (95% CI, 1.036–1.823)]; however, this was not the case for participants ≥ 65 years of age [all-cause mortality: HR, 1.121 (95%, CI 0.782–1.606)].

Fig. 4
figure 4

Subgroup analysis of the association between Triglyceride glucose Index (TyG) index and all-cause mortality (A), cardiovascular disease (CV) mortality (B) among rural Chinese with Metabolic syndrome (MetS). Each subgroup analysis is adjusted if not stratified for age, gender, race, BMI, SBP, DBP, current smoking and drinking, education, annual income, sleep duration, physical activity, cardiovascular history

Full size image

Discussion

In this study, a substantial relationship between the TyG index and all-cause mortality, but not CV mortality, was found in a rural Chinese population with MetS. Moreover, a significant interaction effect was identified between age and the TyG index, suggesting that the correlation between TyG levels and mortality was particularly pronounced among patients who were younger in age. Recently, the incidence of MetS has considerably increased in rural Chinese areas. A meta-analysis indicated a 24.5% prevalence of MetS among individuals 15 years and older in mainland China, with 19.2% residing in rural areas [11]. Our previous study held in rural Northeast China revealed a cumulative incidence of newly diagnosed MetS at 24.0% [12]. Given that metabolic disorders often increase insidiously, all-cause and CV mortalities increase at the time of detection in this population [13]. Hence, proactive strategies and tasks should be undertaken to prevent risk factors such as hypertension, hyperlipidemia, elevated blood glucose, and obesity, along with effective measures to predict mortality among patients with MetS. IR is a crucial pathological factor in MetS, contributing to CVDs and poor clinical outcomes in various ways via mechanisms such as endothelial dysfunction, low-level inflammation, and disruptions in systemic glucose–lipid metabolism [14]. Although various indicators are used to measure IR, including the hyperinsulinemic euglycemic (HIEG) clamp test, homeostatic model assessment for insulin resistance (HOMA-IR), and quantitative insulin sensitivity check index, most of these indicators are time-consuming and expensive to determine, limiting their use in rural areas. Consequently, the TyG index has gradually gained attention in recent years owing to its cost-effectiveness compared to other parameters. In comparison to the gold standard (HIEG clamp test), Won et al. demonstrated that the TyG index has high sensitivity (96.5%) and specificity (85.0%) for IR detection [15]. Furthermore, when compared to the HOMA-IR, the TyG index was confirmed to perform better in assessing IR [7]. The merits of our investigation encompassed a substantial sample size and a longitudinal retrospective approach. This study is the first known attempt to determine the correlation between the TyG index and mortality in a rural Chinese population with MetS. However, the present study had some limitations. First, the participants were from one province in Northeast China, limiting the generalizability of the findings. Second, the TyG index is based on a single blood test, which could introduce potential bias. Third, even after controlling for potential confounding variables, residual confounding factors might have endured. Finally, the influence of prescribed medications on triglyceride and glucose levels could have affected the findings.

Conclusion

The TyG index is a prominent risk predictor of all-cause mortality in participants with MetS in rural China. Our findings indicated that this simple and inexpensive index facilitates the early prediction of mortality in individuals with MetS, aiding village doctors in stratifying high-risk participants and implementing timely interventions.