Potential Association Between Triglyceride-Glucose Related Indices and Type 2 Diabetes-Related Complications: Insights from the National Metabolic Management Center and MIMIC-IV 3.0 Databases

doi:10.21203/rs.3.rs-4952719/v1

Download PDF

Research Article

Potential Association Between Triglyceride-Glucose Related Indices and Type 2 Diabetes-Related Complications: Insights from the National Metabolic Management Center and MIMIC-IV 3.0 Databases

https://doi.org/10.21203/rs.3.rs-4952719/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

The triglyceride-glucose index (TyG) and its related indices are predominantly utilized for the effective assessment of insulin resistance. However, their predictive value concerning type 2 diabetes-related complications remains uncertain. Therefore, this study aims to investigate the potential association between TyG-related indices and type 2 diabetes-related complications through a retrospective analysis of two distinct populations.

Methodss

Established in 2016, the National Metabolic Management Center (MMC) serves as a comprehensive platform designed for the standardized diagnosis, treatment, and long-term follow-up of metabolic diseases, encompassing nearly 300 hospitals across various regions of China. Out of 8,669 initially hospitalized patients, 2,194 were selected for subsequent analysis. Patients were stratified into three groups according to the tertiles of TyG-related indices, with circulatory abnormalities (CA), diabetic kidney disease (DKD), diabetic retinopathy (DR), and diabetic peripheral neuropathy (DPN) serving as the primary outcomes. Logistic regression, restricted cubic splines, and subgroup analyses were employed to evaluate the association between TyG-related indices and complications associated with type 2 diabetes. Finally, a validation analysis was performed on 9,715 samples from the MIMIC-IV 3.0 database to bolster the reliability and generalizability of the findings.

Results

Logistic regression analysis of patients in the MMC database revealed that, in fully adjusted models, each 1 SD increase in TyG and TyG-WC was significantly associated with an elevated risk of DKD. Restricted cubic spline (RCS) analysis indicated a non-linear inverse L-shaped relationship between TyG and DKD risk, while TyG-WC demonstrated a distinct dose-response relationship with DKD risk. Validation analysis conducted in the MIMIC-IV database further corroborated the significant association between TyG-related indices and the risk of DKD.

Conclusions

The findings of this study, involving both Chinese and American populations, indicate that TyG-related indices may serve as robust and independent potential biomarkers for assessing the risk of DKD in individuals with type 2 diabetes.

Triglyceride-Glucose Related Indices

Type 2 Diabetes

Type 2 Diabetes-Related Complications

MIMIC-IV

diabetic kidney disease

Type 2 diabetes is a significant chronic condition marked by elevated blood glucose levels resulting from either relative or absolute insulin deficiency^1,2. In 2019, it was estimated that approximately 460 million individuals worldwide were affected by diabetes, and the International Diabetes Federation (IDF) Diabetes Atlas 2021 reports an increase to 537 million. Projections suggest that this figure may increase to 643 million by 2030 and escalate to an alarming 783 million by 2045, potentially driving global healthcare expenditures beyond $105.4 billion^3,4. Furthermore, the Global Burden of Disease Study 2021 ranked diabetes as the eighth leading risk factor for mortality and disability⁵.

Research into the causes of diabetes-related mortality reveals that the majority of patients with type 2 diabetes experience at least one complication. These complications—neurological, renal, retinal, and notably cardiovascular issues—are the primary causes of mortality among individuals with diabetes^6,7. Research conducted by Nathan D. Wong and colleagues indicates that individuals with type 2 diabetes who are diagnosed at a younger age tend to experience a greater reduction in life expectancy due to various complications⁸.

Diabetic peripheral neuropathy (DPN) represents one of the most prevalent, complex, and debilitating complications encountered by individuals with diabetes. DPN heightens the risk of ulcers, non-traumatic amputations, and foot infections, which can lead to long-term disability and impose substantial economic and psychological burdens on patients with type 2 diabetes⁹. A comprehensive observational study conducted among Asian populations revealed that the prevalence of diabetic kidney disease (DKD) reaches 58.6%, with diabetic nephropathy serving as a major contributor to chronic kidney disease and end-stage renal disease^10,11. These statistics underscore the substantial impact of type 2 diabetes and its complications on global public health, emphasizing the urgent need for effective early detection markers for type 2 diabetes and its associated complications in the broader population.

Previous research has identified various clinical biomarkers linked to type 2 diabetes, such as mannose, glycerol, and alanine¹². Initially, TyG-related indices were employed mainly for assessing and predicting insulin resistance; however, recent research has established a positive correlation between these indices and the occurrence of hypertension, coronary artery disease, and chronic kidney disease^13,14. Furthermore, research has demonstrated the reliability of the TyG index as a prognostic marker for adverse outcomes in type 2 diabetes¹⁵.

Nonetheless, the current literature remains limited in studies exploring the relationship between TyG-related indices and type 2 diabetes-related complications. Consequently, this study seeks to integrate data from the National Metabolic Management Center (MMC) and the Medical Information Mart for Intensive Care IV 3.0 (MIMIC-IV) to assess the potential association between TyG-related indices and type 2 diabetes-related complications.

Study design and population

This study utilizes data from two primary sources: the National Metabolic Management Center (MMC) and the Medical Information Mart for Intensive Care IV 3.0 (MIMIC-IV). The MMC, established in 2016, serves as a platform for standardized diagnosis, treatment, and long-term follow-up of metabolic diseases, and includes nearly 300 hospitals across various regions in China¹⁶. The study protocol and informed consent documents have received approval from the Human Research Committee at Ruijin Hospital, Shanghai Jiao Tong University School of Medicine. All participants provided written informed consent prior to their involvement in the study.

The study initially screened 8,669 patients who were hospitalized for the first time between 2017 and 2023. Of these, 4,861 patients who did not have type 2 diabetes were excluded from the analysis. Additionally, 469 participants with high rates of missing baseline data and 1,145 participants who did not meet the diagnostic criteria for type 2 diabetes-related complications were excluded. Ultimately, a total of 2,194 participants were included in the final analysis (see Fig. 1).

MIMIC-IV is a comprehensive health records database encompassing data from hospitalized patients across various hospitals in the United States between 2008 and 2019, with an emphasis on intensive care units¹⁷. The study secured all necessary permissions for the use of the relevant data. This segment of the study included 9,715 individuals newly diagnosed with type 2 diabetes and hospitalized for the first time. Patients with missing data on blood glucose, triglycerides, or other significant variables were excluded from the analysis.

Data collection

The MMC's proprietary electronic medical records system was utilized to extract comprehensive demographic and anthropometric data, including gender, age, height, weight, waist circumference, and blood pressure. Smoking status was categorized into current smokers, defined as individuals smoking at least 7 cigarettes per week for a minimum of 6 months, and non-smokers, those who had previously smoked under the same criteria but had since quit. Alcohol consumption was classified into current drinkers, defined as those consuming alcohol at least once per week for a minimum of 6 months, and non-drinkers. Laboratory test parameters were meticulously collected by trained nurses who obtained fasting blood samples (after a minimum of 8–12 hours of fasting) from patients and sent them to the central laboratory for the analysis of circulating metabolites. Glycated hemoglobin levels in capillary whole blood were quantified through high-performance liquid chromatography (VARIANT II system; Bio-Rad, Hercules, CA), while lipid levels were measured using an automated analyzer (Abbott Laboratories, Abbott Park, IL).

The baseline characteristics of patients in the MIMIC-IV database were predominantly extracted using Structured Query Language (SQL) in conjunction with Navicat Premium 16. These characteristics encompass demographic information, including age, gender, and BMI, as well as laboratory test results such as hematocrit, hemoglobin, platelets, WBC, anion gap, bicarbonate, BUN, creatinine, glucose, sodium, potassium, and triglycerides (TG).

The TyG-related indices were computed using the following formulas: TyG = Ln[fasting TG (mg/dL) × FBG (mg/dL) / 2]; TyG-BMI = TyG × BMI; and TyG-WC = TyG × WC. Furthermore, patients were stratified into three distinct groups according to the tertiles of TyG (< 7.54, 7.54–8.16, > 8.16), TyG-BMI (< 177.42, 177.42-208.88, > 208.88), and TyG-WC (< 630.40, 630.40-731.74, > 731.74).

Ascertainment of covariates

Drawing on insights from previous research, we incorporated a comprehensive set of potential confounders into the statistical analysis²⁰. These included demographic variables such as age and sex (categorized as “male” and “female”), a history of hypertension, and lifestyle factors including current smoking status (“yes” or “no”) and current alcohol consumption (“yes” or “no”). In addition, we controlled for various clinical indicators, including systolic blood pressure (SBP), diastolic blood pressure (DBP), body mass index (BMI), fasting blood glucose (FBG), triglycerides (TG), serum creatinine (Cr), total cholesterol (TC), HDL-C, LDL-C, and glycated hemoglobin (HbA1c).

Assessment of type 2 diabetes and type 2 diabetes-related complications

The diagnosis of type 2 diabetes was established based on the 2013 American Diabetes Association criteria, requiring fulfillment of at least one of the following conditions: 1) fasting blood glucose level ≥ 7.0 mmol/L; 2) 2-hour postprandial glucose level ≥ 11.1 mmol/L following an oral glucose tolerance test (OGTT); 3) glycated hemoglobin level ≥ 6.5%; or 4) a clinical diagnosis confirmed by a licensed physician. Type 2 diabetes-related complications were identified using International Classification of Diseases codes, encompassing peripheral circulatory disease (CA, E10.5), diabetic kidney disease (DKD, E10.2), retinopathy (DR, E10.3), and peripheral neuropathy (DPN, E10.4).

Statistical analysis

Baseline characteristics of participants were systematically described based on data type using appropriate statistical techniques. Continuous variables were reported as means with standard deviations, while categorical variables were presented as counts with percentages. Normality of continuous variables was assessed using analysis of variance (ANOVA) for normally distributed data, whereas the Kruskal-Wallis test was employed for variables with skewed distributions. Categorical variables were analyzed using the chi-square test. Logistic regression was utilized to examine the association between TyG-related indices and type 2 diabetes-related complications, with odds ratios (OR) and 95% confidence intervals (CI) calculated across three models. Model 1 adjusted for age and sex. Model 2 incorporated adjustments for age, sex, DBP, SBP, LDL-C, TC, Cr, FBG, and HbA1c. Model 3 provided comprehensive adjustments for age, sex, DBP, SBP, LDL-C, TC, Cr, FBG, HbA1c, smoking status, alcohol consumption, and history of hypertension. The trend p-value was derived by treating quartiles of TyG-related indices as an ordinal variable. Furthermore, the non-linear association between TyG-related indices and type 2 diabetes-related complications was investigated using restricted cubic splines (RCS) with five knots. Subgroup analyses were performed using logistic regression to explore interactions with categorical confounding factors such as age (< 60 years, ≥ 60 years) and sex (male, female). Non-significant interaction p-values indicated consistency across strata, while significant p-values suggested potential subgroup-specific effects.

Statistical analyses were performed using SPSS version 26.0 and R version 4.3.3, with forest plots generated through GraphPad Prism version 10.0. A two-sided p-value of less than 0.05 was deemed statistically significant.

Baseline characteristics

Table 1 presents the baseline characteristics of patients stratified by TyG index tertiles derived from the MMC database (Tertile Q1 < 7.54; Q2 7.54–8.16; Q3 > 8.16). This study comprised 2,194 patients with type 2 diabetes, characterized by a mean age of 56.25 ± 10.56 years, with 65.8% of the cohort being male. Of the participants, 782 (35.6%) were identified as current smokers, while 810 (36.9%) were current drinkers. In the Q3 group, patients were noted to be younger at admission and had a higher likelihood of being male. Furthermore, this group demonstrated elevated levels of BMI, WC, DBP, SBP, LDL-c, TC, TG, FBG, and HbA1c, coupled with reduced HDL-c levels. Concurrently, in comparison to the lower TyG index groups, this cohort exhibited a higher proportion of current smokers and drinkers (all p < 0.05).

Table 1

Baseline characteristics of participants by TyG tertile
Variables	Tatal	Q1(< 7.54)	Q2(7.54–8.16)	Q3(> 8.16)	P
N	2194	729	739	726	-
Age, years	56.25 ± 10.56	57.83 ± 10.09	56.76 ± 10.10	54.17 ± 11.11	< 0.01
Male (%)	1443(65.8)	453(62.1)	471(63.7)	519(71.4)	< 0.01
BMI, kg/m²	24.63 ± 3.62	23.46 ± 3.36	24.88 ± 3.69	25.56 ± 3.47	< 0.01
WC,cm	86.25 ± 10.15	82.46 ± 9.94	87.15 ± 10.15	89.13 ± 9.17	< 0.01
DBP, mmHg	79.08 ± 11.40	77.40 ± 11.06	79.56 ± 11.42	80.28 ± 11.54	< 0.01
SBP, mmHg	135.05 ± 20.86	133.75 ± 21.17	136.21 ± 20.42	135.18 ± 20.92	0.04
LDL-c, mmol/L	2.85 ± 1.06	2.64 ± 0.98	2.98 ± 1.03	2.93 ± 1.14	< 0.01
HDL-c, mmol/L	1.16 ± 0.35	1.29 ± 0.39	1.15 ± 0.32	1.03 ± 0.30	< 0.01
TC, mmol/L	4.84 ± 1.95	4.33 ± 1.10	4.72 ± 1.21	5.48 ± 2.84	< 0.01
TG, mmol/L	2.44 ± 2.39	1.11 ± 0.35	1.85 ± 0.56	4.35 ± 3.31	< 0.01
Cr, mmol/L	69.89 ± 43.36	68.89 ± 46.66	69.69 ± 43.68	71.08 ± 39.46	0.55
FBG, mmol/L	9.24 ± 3.41	7.32 ± 1.95	9.07 ± 2.77	11.32 ± 3.93	< 0.01
HbA1c, %	9.73 ± 2.68	9.19 ± 2.88	9.75 ± 2.54	10.24 ± 2.50	< 0.01
Current smoker	782(35.6)	227(31.1)	253(34.2)	302(41.6)	< 0.01
Current drinker	810(36.9)	226(31.0)	284(38.4)	300(41.3)	< 0.01
history of hypertension	985(44.9)	308(42.2)	343(46.4)	334(46.0)	0.22
Data are summarized as number (percentage), mean ± standard deviation. BMI, body mass index; WC, waist circumference; DBP, diastolic blood pressure; SBP, systolic blood pressure; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; TC, total cholesterol; TG, triglyceride; Cr, creatinine; FBG, fasting blood glucose; HbA1c, glycated haemoglobin.

Association of TyG-related indices with type 2 diabetes-related complications

The findings of the logistic regression analysis are comprehensively outlined in Table 2. We identified a total of 1,653 cases of CA (75.3%), 676 cases of DKD (30.8%), 486 cases of DR (22.2%), and 1,149 cases of DPN (52.4%). When TyG-related indices were analyzed as continuous variables, the logistic regression analysis revealed that both TyG and TyG-WC were significantly associated with DKD across all three adjustment models. Each standard deviation increase in TyG and TyG-WC was associated with an increased risk of DKD, with odds ratios of 1.33 (95% CI: 1.12, 1.57) and 1.003 (95% CI: 1.001, 1.004), respectively. TyG-BMI demonstrated an association with DKD in Model 1, although this association was attenuated after adjustment for additional factors. No significant associations were observed between TyG-related indices and other type 2 diabetes-related complications. Moreover, when TyG-related indices were analyzed as categorical variables based on tertiles, the highest quartiles (Q3) of TyG and TyG-WC continued to show significant associations with DKD risk in all three models. Relative to the Q1 group, the risk of DKD in the Q3 group increased by 72% and 64%, respectively. Trend analysis further indicated that higher quartiles of TyG and TyG-WC were linked to an increased risk of DKD (P for trend < 0.05). Furthermore, patients in the Q2 group of TyG-WC exhibited a higher risk of DR; however, this association was no longer evident when extended to the Q3 group.

Table 2

Odds ratios and 95% CIs for the association of TyG-related indices with Type 2 diabetes-related complications
Variables	N	Q1	Q2	Q3	Per SD	P for trend
CA	1653
TyG	Model 1	Reference	0.94(0.74,1.19)	1.09(0.85,1.40)	1.10(0.96,1.25)	0.47
	Model 2	Reference	0.88(0.68,1.13)	1.09(0.83,1.43)	1.16(0.98,1.37)	0.55
	Model 3	Reference	0.87(0.67,1.12)	1.06(0.80,1.39)	1.13(0.96,1.33)	0.68
TyG-BMI	Model 1	Reference	0.98(0.77,1.24)	1.00(0.79,1.28)	1.000(0.998,1.003)	0.96
	Model 2	Reference	1.05(0.79,1.40)	1.19(0.80,1.78)	1.006(1.000,1.013)	0.40
	Model 3	Reference	1.04(0.78,1.39)	1.14(0.76,1.70)	1.005(0.999,1.012)	0.54
TyG-WC	Model 1	Reference	1.00(0.78,1.27)	1.01(0.79,1.30)	1.000(0.999,1.001)	0.93
	Model 2	Reference	1.00(0.76,1.32)	1.10(0.79,1.55)	1.001(1.000,1.003)	0.58
	Model 3	Reference	0.99(0.75,1.31)	1.06(0.75,1.49)	1.001(1.000,1.002)	0.74
DKD	676
TyG	Model 1	Reference	1.58(1.24,2.02)	1.86(1.46,2.37)	1.33(1.18,1.50)	< 0.01
	Model 2	Reference	1.54(1.18,2.01)	1.68(1.26,2.24)	1.31(1.11,1.55)	< 0.01
	Model 3	Reference	1.56(1.19,2.04)	1.72(1.29,2.29)	1.33(1.12,1.57)	< 0.01
TyG-BMI	Model 1	Reference	1.31(1.03,1.67)	1.64(1.30,2.07)	1.006(1.004,1.008)	< 0.01
	Model 2	Reference	1.10(0.82,1.48)	1.32(0.88,1.98)	1.011(1.004,1.017)	0.18
	Model 3	Reference	1.13(0.84,1.52)	1.35(0.90,2.03)	1.011(1.004,1.018)	0.14
TyG-WC	Model 1	Reference	1.56(1.22,2.00)	1.87(1.47,2.38)	1.002(1.001,1.003)	< 0.01
	Model 2	Reference	1.39(1.04,1.84)	1.64(1.17,2.31)	1.002(1.001,1.004)	< 0.01
	Model 3	Reference	1.39(1.04,1.84)	1.64(1.17,2.31)	1.003(1.001,1.004)	< 0.01
DR	486
TyG	Model 1	Reference	0.99(0.76,1.29)	1.07(0.81,1.41)	0.99(0.86,1.15)	0.65
	Model 2	Reference	0.94(0.70,1.25)	0.99(0.73,1.36)	0.96(0.80,1.14)	0.95
	Model 3	Reference	0.93(0.70,1.25)	0.98(0.71,1.34)	0.95(0.80,1.13)	0.87
TyG-BMI	Model 1	Reference	1.22(0.93,1.59)	0.95(0.72,1.26)	0.999(0.996,1.002)	0.76
	Model 2	Reference	1.16(0.83,1.62)	0.99(0.62,1.60)	0.997(0.990,1.004)	0.98
	Model 3	Reference	1.15(0.83,1.61)	0.98(0.61,1.58)	0.997(0.990,1.004)	0.97
TyG-WC	Model 1	Reference	1.43(1.10,1.88)	0.90(0.68,1.20)	1.000(0.999,1.001)	0.53
	Model 2	Reference	1.38(1.01,1.89)	0.83(0.56,1.24)	0.999(0.998,1.001)	0.40
	Model 3	Reference	1.38(1.01,1.89)	0.82(0.55,1.22)	0.999(0.997,1.000)	0.38
DPN	1149
TyG	Model 1	Reference	1.01(0.83,1.23)	1.11(0.91,1.35)	1.03(0.93,1.14)	0.31
	Model 2	Reference	0.95(0.76,1.19)	1.05(0.83,1.32)	0.99(0.87,1.13)	0.71
	Model 3	Reference	0.97(0.78,1.21)	1.05(0.83,1.32)	0.99(0.87,1.13)	0.71
TyG-BMI	Model 1	Reference	0.85(0.70,1.03)	0.74(0.60,0.90)	0.996(0.994,0.998)	< 0.01
	Model 2	Reference	0.97(0.76,1.24)	1.03(0.73,1.24)	1.000(0.994,1.005)	0.88
	Model 3	Reference	0.97(0.76,1.25)	1.04(0.73,1.47)	0.999(0.994,1.005)	0.85
TyG-WC	Model 1	Reference	0.87(0.72,1.06)	0.78(0.64,0.95)	0.999(0.998,1.000)	0.02
	Model 2	Reference	0.88(0.70,1.11)	0.95(0.71,1.26)	1.000(0.999,1.001)	0.69
	Model 3	Reference	0.89(0.71,1.12)	0.97(0.73,1.29)	1.000(0.999,1.001)	0.82
Model 1: adjusted for age and sex; Model 2: adjusted for age, sex, BMI, DBP, SBP, LDL-C, TC, Cr, HbA1c; Model 3: adjusted for age, sex, BMI, DBP, SBP, LDL-C, TC, Cr, FBG, HbA1c, smoking status, alcohol consumption, and history of hypertension. TyG, triglyceride-glucose index; BMI, body mass index; WC, waist circumference; Per SD, odds ratio for per SD change in TyG-related indices.

Figure 2 presents the findings from the restricted cubic spline analysis.In the fully adjusted model, a nonlinear inverse L-shaped association was observed between TyG and DKD risk (nonlinear P < 0.05), whereas TyG-WC demonstrated a distinct dose-response relationship with DKD risk (nonlinear P = 0.12).

Subgroup analyses

To gain deeper insights into the relationship between TyG-related indices and DKD, we conducted subgroup analyses stratified by age and sex, as outlined in Table 3 and Fig. 3. Within each subgroup, each standard deviation increase in TyG-related indices was consistently linked to a heightened risk of DKD.Furthermore, no significant interaction was found between TyG-related index levels and subgroup variables (interaction P > 0.05).

Table 3

Subgroup analyses for the association of the TyG-related indices with DK
Variables	TyG	TyG-BMI	TyG-WC
Age
< 60	1.20(0.98,1.47)	1.008(1.000,1.016)	1.002(1.000,1.003)
≥ 60	1.67(1.23,2.26)	1.018(1.006,1.031)	1.004(1.002,1.007)
P for interaction	0.97	0.50	0.16
Sex
Male	1.34(1.07,1.67)	1.011(1.002,1.020)	1.000(1.002,1.004)
Female	1.33(1.03,1.72)	1.012(1.002,1.022)	1.003(1.001,1.005)
P for interaction	0.85	0.29	0.11
Adjusted for age, sex, DBP, SBP, LDL-C, TC, Cr, FBG, HbA1c, smoking status, alcohol consumption, and education level. BMI, body mass index; TyG, triglyceride-glucose index; BMI, body mass index; WC, waist circumference; DK, diabetic kidney disease ; Per SD, odds ratio for per SD change in AIP.

Validation analysis based on MIMIC database

To improve the robustness and generalizability of our findings, we conducted a validation analysis using data from MIMIC-IV 3.0. Table 4 provides the baseline characteristics of patients categorized by DKD status. This study comprised 9,715 participants, with the DKD group exhibiting a higher proportion of males, an older age, and significant differences in laboratory indicators such as BMI, glucose, and WBC compared to the non-DKD group. Figure 4 illustrates the findings from the logistic regression analysis and restricted cubic spline analysis for this cohort. The logistic regression analysis revealed that both TyG and TyG-BMI were positively associated with DKD risk, with odds ratios of 1.40 (95% CI: 1.23–1.61) and 1.009 (95% CI: 1.005–1.013), respectively. The restricted cubic spline analysis indicated a nonlinear inverse L-shaped association between TyG and TyG-BMI and DKD risk (nonlinear P < 0.05).

Table 4

Baseline characteristics of the DKD and Non-DKD groups in MIMIC-IV database
Variables	Tatal	Non-DKD	DKD	P
N	9715	9208	507	-
Age, years	64.64 ± 16.28	64.25 ± 16.45	71.68 ± 10.67	< 0.01
Male (%)	1443(65.8)	453(62.1)	471(63.7)	< 0.01
BMI, kg/m²	30.44 ± 8.55	30.37 ± 8.58	31.55 ± 7.98	0.01
Glucose,mg/dL	154.67 ± 72.56	152.61 ± 71.10	192.18 ± 87.32	< 0.01
TG,mg/dL	192.25 ± 337.67	192.28 ± 342.08	191.65 ± 244.30	0.97
Hematocrit,%	33.11 ± 7.21	33.25 ± 7.19	30.64 ± 6.97	< 0.01
Hemoglobin,g/dL	10.92 ± 2.46	10.98 ± 2.46	9.83 ± 2.31	< 0.01
Platelets,10⁹/L	195.21 ± 96.25	195.65 ± 96.80	187.23 ± 85.15	0.06
WBC,10⁹/L	10.57 ± 8.12	10.52 ± 7.70	11.43 ± 13.70	< 0.01
Aniongap,mmol/L	13.01 ± 3.64	12.95 ± 3.61	14.11 ± 4.01	< 0.01
Bicarbonate,mmol/L	20.96 ± 4.92	21.03 ± 4.89	19.79 ± 5.32	< 0.01
Bun,mg/dL	24.07 ± 20.34	23.10 ± 19.40	41.79 ± 27.67	< 0.01
Creatinine,mg/dL	1.35 ± 1.35	1.28 ± 1.27	2.59 ± 1.98	< 0.01
Sodium,mmol/L	136.86 ± 5.34	136.88 ± 5.34	136.47 ± 5.26	0.09
Potassium,mmol/L	3.89 ± 0.57	3.88 ± 0.56	4.09 ± 0.61	< 0.01
Data are summarized as number (percentage), mean ± standard deviation. BMI, body mass index; TG, triglyceride; WBC, white blood cell.

In this retrospective cross-sectional study encompassing 2,194 Chinese participants and 9,715 American participants, we identified a sustained positive association between TyG and TyG-WC and DKD risk, which remained significant after comprehensive adjustment for various confounding factors. Subgroup analyses further validated the robustness of this association, demonstrating its generalizability across diverse populations. Furthermore, the RCS analysis uncovered a nonlinear relationship between TyG and DKD, characterized by an inverse L-shaped curve. These findings underscore the potential of TyG-related indices as valuable early biomarkers for predicting DKD.

Type 2 diabetes is a chronic condition marked by a gradual onset over an individual's lifetime, with its prevalence among younger populations rising annually²¹. Upon diagnosis, type 2 diabetes imposes lifelong challenges on patients, encompassing a spectrum of complications such as circulatory and neurological issues, which profoundly affect quality of life and place a significant economic burden on both individuals and the healthcare system^22,23. Recent research underscores that proactive prevention and therapeutic strategies can substantially delay the onset and progression of diabetes-related complications, reduce associated healthcare costs, and enhance patients' quality of life²⁴. A substantial body of evidence indicates that lifestyle interventions, including dietary modifications and increased physical activity, may be more effective than pharmacological treatments in slowing disease progression and facilitating early prevention²⁵. With recent advancements in high-throughput detection technologies, researchers are increasingly investigating biomarkers for type 2 diabetes and its complications through approaches such as metabolomics and proteomics²⁶. For instance, the increased conversion of phenylalanine to tyrosine, attributed to improved renal function, is thought to hold potential predictive value for DKD²⁷. Nonetheless, these findings are constrained in their clinical applicability due to the absence of large-scale observational studies.

The TyG index, introduced by Gisela Unger et al. in 2014, is an effective and cost-efficient measure for assessing insulin resistance²⁸. While the Homeostasis Model Assessment (HOMA) is a well-established method for evaluating insulin resistance (IR) and is relatively simple and reliable, its high cost and invasive nature hinder its routine clinical use²⁹. While the Homeostasis Model Assessment (HOMA) is a well-established method for evaluating insulin resistance (IR) and is relatively simple and reliable, its high cost and invasive nature hinder its routine clinical use³⁰. Recent clinical studies have underscored a strong association between the TyG index and both the incidence and mortality of diabetes. Yang et al.³¹ demonstrated that, among critically ill patients with acute heart failure, a higher TyG index correlates with an elevated risk of acute kidney injury. Pan et al.³²identified a close association between TyG and the risk of lower limb vascular stenosis as well as renal microvascular damage in hospitalized patients. Furthermore, a large observational study revealed that an elevated TyG index mediates the relationship between BMI and end-stage renal disease in middle-aged individuals³³. The evidence from these studies suggests that the TyG index could serve as a predictive marker for deteriorating renal function across various patient populations. Our study, utilizing multiple population datasets, identified a link between TyG-related indices and DKD. The mechanism underlying this association is likely related to the effect of insulin resistance on renal function. Existing research suggests that insulin resistance can elevate glomerular hydrostatic pressure, leading to increased renal vascular permeability and, ultimately, glomerular hyperfiltration and renal damage³⁴. Moreover, insulin resistance may enhance the bioavailability of vasoactive mediators, including nitric oxide and renin-angiotensin, thereby influencing renal function³⁵. Simultaneously, these vasoactive substances can induce glomerular hypertrophy, glomerulosclerosis, tubulointerstitial inflammation, and fibrosis³⁶.

This study presents several notable strengths. First, in contrast to other studies investigating the relationship between TyG and type 2 diabetes-related complications, we employed two extensive databases from China and the United States, resulting in a larger sample size. Second, the application of RCS and subgroup analyses increased statistical power and validated the robustness of our findings. Third, incorporating the most recent data from MIMIC-IV 3.0 provided additional validation of the results and further enhanced the generalizability of our findings.

Nonetheless, this study has several limitations. First, given the retrospective and observational nature of the study design, it is unable to establish definitive causal relationships. Second, variables not accounted for, such as dietary patterns, racial differences, and lifestyle factors, could potentially introduce bias into the findings. Furthermore, the inherent limitations of the MIMIC-IV database meant that factors such as the severity of type 2 diabetes and DKD, as well as the socioeconomic status of participants, were not accounted for, which could also contribute to potential bias in the findings.

In summary, elevated levels of TyG-related indices are strongly associated with an increased risk of DKD in patients with type 2 diabetes, with TyG demonstrating an inverse L-shaped relationship with DKD risk. These findings indicate that TyG-related indices may aid clinicians in making more informed decisions to prevent DKD in patients with type 2 diabetes.

Ethics approval and consent to participate

The studies involving human participants were reviewed and approved by the ethics committee of the Affiliated Hospital of Southwest Medical University (ethical approval code: 2018017). The patients/participants provided their written informed consent to participate in this study.

Availability of data and materials

The datasets generated and/or analyzed during the current study are not publicly available because this data includes the privacy content of the participants but are available from the corresponding author on reasonable request.

Competing interests

The authors declare that the research was conducted without any commercial or financial relationships that could be construed as a potential conflict of interest.

Funding

The Ministry of Science and Technology of China provided funding for this study through grants 2016YFC0901200.

Acknowledgments

Thanks to YM, JL and others for their contributions to the data collection.

Author contributions

YYZ, XB and BXC made equal contributions to this work performing the statistical analysis, interpreting the results, writing the paper, and they share the first authorship. QW proposed the ideas. All authors contributed to the article and approved the submitted version.

Dj LC, Pz M. The worldwide epidemiology of type 2 diabetes mellitus–present and future perspectives. Nat Rev Endocrinol. 2011;8(4). 10.1038/nrendo.2011.183.
Pz Z. Diabetes and its drivers: the largest epidemic in human history? Clin Diabetes Endocrinol. 2017;3. 10.1186/s40842-016-0039-3.
Mg T, Vs M. An Update on the Epidemiology of Type 2 Diabetes: A Global Perspective. Endocrinol Metab Clin North Am. 2021;50(3). 10.1016/j.ecl.2021.05.013.
Y Y. Prevalence, awareness and control of type 2 diabetes mellitus and risk factors in Chinese elderly population. BMC Public Health. 2022;22(1). 10.1186/s12889-022-13759-9.
Collaborators G. 2021 D. Global, regional, and national burden of diabetes from 1990 to 2021, with projections of prevalence to 2050: a systematic analysis for the Global Burden of Disease Study 2021. Lancet Lond Engl. 2023;402(10397):203. 10.1016/S0140-6736(23)01301-6
Y Z, Sh L, Fb H. Global aetiology and epidemiology of type 2 diabetes mellitus and its complications. Nat Rev Endocrinol. 2018;14(2). 10.1038/nrendo.2017.151.
Ji MO, K I E. Comorbidities and complications in Japanese patients with type 2 diabetes mellitus: Retrospective analyses of J-DREAMS, an advanced electronic medical records database. Diabetes Res Clin Pract. 2021;178. 10.1016/j.diabres.2021.108845.
Nd W. Cardiovascular risk in diabetes mellitus: epidemiology, assessment and prevention. Nat Rev Cardiol. 2023;20(10). 10.1038/s41569-023-00877-z.
B W ZN. Study on Risk Factors of Peripheral Neuropathy in Type 2 Diabetes Mellitus and Establishment of Prediction Model. Diabetes Metab J. 2021;45(4). 10.4093/dmj.2020.0100.
Ay W, Nc K, Fa de L, et al. An alarmingly high prevalence of diabetic nephropathy in Asian type 2 diabetic patients: the MicroAlbuminuria Prevalence (MAP) Study. Diabetologia. 2005;48(1). 10.1007/s00125-004-1599-9.
Zhang XX, Kong J, Yun K. Prevalence of Diabetic Nephropathy among Patients with Type 2 Diabetes Mellitus in China: A Meta-Analysis of Observational Studies. J Diabetes Res. 2020;2020. 10.1155/2020/2315607
Laakso M. Biomarkers for type 2 diabetes. Mol Metab. 2019;27(Suppl):S139. 10.1016/j.molmet.2019.06.016.
J S, Z L, M H, et al. Triglyceride glucose index for the detection of the severity of coronary artery disease in different glucose metabolic states in patients with coronary heart disease: a RCSCD-TCM study in China. Cardiovasc Diabetol. 2022;21(1). 10.1186/s12933-022-01523-7
R Z YM. Triglyceride and glucose (TyG) index as a predictor of incident hypertension: a 9-year longitudinal population-based study. Lipids Health Dis. 2017;16(1). 10.1186/s12944-017-0562-y.
Y Y, B W, T G, J C, W C, L L. The association between TyG and all-cause/non-cardiovascular mortality in general patients with type 2 diabetes mellitus is modified by age: results from the cohort study of NHANES 1999–2018. Cardiovasc Diabetol. 2024;23(1). 10.1186/s12933-024-02120-6
Zhang Y, Wang W, Ning G. Metabolic Management Center: An innovation project for the management of metabolic diseases and complications in China. Accessed August 21, 2024. https://onlinelibrary.wiley.com/doi/10.1111/1753-0407.12847
Aew J, L B. MIMIC-IV, a freely accessible electronic health record dataset. Sci Data. 2023;10(1). 10.1038/s41597-022-01899-x.
Cw KP, Sb A. Elevated TyG Index Predicts Progression of Coronary Artery Calcification. Diabetes Care. 2019;42(8). 10.2337/dc18-1920.
S Z, S S, X R, et al. Triglyceride glucose-waist circumference, a novel and effective predictor of diabetes in first-degree relatives of type 2 diabetes patients: cross-sectional and prospective cohort study. J Transl Med. 2016;14(1). 10.1186/s12967-016-1020-8
B C, C Z, R Z, et al. Triglyceride-glucose index and triglyceride to high-density lipoprotein cholesterol ratio as potential cardiovascular disease risk factors: an analysis of UK biobank data. Cardiovasc Diabetol. 2023;22(1). 10.1186/s12933-023-01762-2
Sr RURP. Type 2 Diabetes: Demystifying the Global Epidemic. Diabetes. 2017;66(6). 10.2337/db16-0766.
X PZ. Global healthcare expenditure on diabetes for 2010 and 2030. Diabetes Res Clin Pract. 2010;87(3). 10.1016/j.diabres.2010.01.026.
Ra D, E F. Type 2 diabetes mellitus. Nat Rev Dis Primer. 2015;1. 10.1038/nrdp.2015.19.
Kr S, Mk A, X Z, et al. Cost-effectiveness of Interventions to Manage Diabetes: Has the Evidence Changed Since 2008? Diabetes Care. 2020;43(7). 10.2337/dci20-0017.
Wc K, E BC, Se F, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med. 2002;346(6). 10.1056/NEJMoa012512.
Jin Q, Ma RCW. Metabolomics in Diabetes and Diabetic Complications: Insights from Epidemiological Studies. Cells. 2021;10(11). 10.3390/cells10112832.
N T. Plasma Metabolomics Identifies Markers of Impaired Renal Function: A Meta-analysis of 3089 Persons with Type 2 Diabetes. J Clin Endocrinol Metab. 2020;105(7). 10.1210/clinem/dgaa173.
Sf GU, Gl BFP. Triglycerides and glucose index: a useful indicator of insulin resistance. Endocrinol Nutr Organo Soc Esp Endocrinol Nutr. 2014;61(10). 10.1016/j.endonu.2014.06.009.
G EB. Homeostasis model assessment closely mirrors the glucose clamp technique in the assessment of insulin sensitivity: studies in subjects with various degrees of glucose tolerance and insulin sensitivity. Diabetes Care. 2000;23(1). 10.2337/diacare.23.1.57.
F GR, Le SM. The product of triglycerides and glucose, a simple measure of insulin sensitivity. Comparison with the euglycemic-hyperinsulinemic clamp. J Clin Endocrinol Metab. 2010;95(7). 10.1210/jc.2010-0288.
Yang Z, Gong H, Kan F, Ji N. Association between the triglyceride glucose (TyG) index and the risk of acute kidney injury in critically ill patients with heart failure: analysis of the MIMIC-IV database. Cardiovasc Diabetol. 2023;22. 10.1186/s12933-023-01971-9.
Y P, S Z, K Z, et al. Association between Diabetes Complications and the Triglyceride-Glucose Index in Hospitalized Patients with Type 2 Diabetes. J Diabetes Res. 2021;2021. 10.1155/2021/8757996
J F, W B, H C, et al. The Association of Excess Body Weight with Risk of ESKD Is Mediated Through Insulin Resistance, Hypertension, and Hyperuricemia. J Am Soc Nephrol JASN. 2022;33(7). 10.1681/ASN.2021091263
Jd S, De K. Insulin resistance, kidney outcomes and effects of the endothelin receptor antagonist atrasentan in patients with type 2 diabetes and chronic kidney disease. Cardiovasc Diabetol. 2023;22(1). 10.1186/s12933-023-01964-8.
Sasson AN, Cherney DZ. Renal hyperfiltration related to diabetes mellitus and obesity in human disease. World J Diabetes. 2012;3(1):1. 10.4239/wjd.v3.i1.1.
Alicic RZ, Rooney MT, Tuttle KR. Diabetic Kidney Disease: Challenges, Progress, and Possibilities. Clin J Am Soc Nephrol CJASN. 2017;12(12):2032. 10.2215/CJN.11491116.

No competing interests reported.

STROBEchecklist.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Potential Association Between Triglyceride-Glucose Related Indices and Type 2 Diabetes-Related Complications: Insights from the National Metabolic Management Center and MIMIC-IV 3.0 Databases

Status:

Version 1

Abstract

Background

Methodss

Results

Conclusions

Figures

Introduction

Method

Study design and population

Data collection

Ascertainment of covariates

Assessment of type 2 diabetes and type 2 diabetes-related complications

Statistical analysis

Result

Baseline characteristics

Association of TyG-related indices with type 2 diabetes-related complications

Subgroup analyses

Validation analysis based on MIMIC database

Discussion

Conclusions

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1