The Metabolic Score for Insulin Resistance (METS-IR) as a Predictor of Incident diabetes: A Longitudinal Study among Chinese without Diabetes

doi:10.21203/rs.3.rs-4157511/v1

Download PDF

Research Article

The Metabolic Score for Insulin Resistance (METS-IR) as a Predictor of Incident diabetes: A Longitudinal Study among Chinese without Diabetes

https://doi.org/10.21203/rs.3.rs-4157511/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: As the prevalence of diabetes rises and the disease burden increases, it will be extremely beneficial to identify high-risk groups who are susceptible to diabetes early on using simple and effective diagnostic tools. Recently, it was discovered that T2DM is strongly correlated with METS-IR. However, it is unclear whether the relationship between T2DM and METS-IR is affected by time progression.

Methods

This study included 114,827 subjects (62186 men and 52641 women) who underwent a physical examination with a 7-years follow-up. The subjects were divided into five equal groups based on their METS-IR quintile, with the outcome of interest being the occurrence of diabetic events.

Results

Multivariate Cox proportional hazards regression analysis for the prediction of diabetes according to the METS-IR shows that METS-IR was strongly positively correlated with the risk of future diabetes (HR 1.11 per SD increase, 95% CI: 1.10-1.11), and the risk of diabetes corresponding to METS-IR was gradually increased compared with the lowest quintile group (P for trend < 0.001). ROC curve analysis demonstrated that METS-IR in females outperformed the prediction for T2DM compared to that in males. Among people with different phenotypes, it can be observed that younger individuals had a significantly higher risk of METS-IR-related diabetes than middle-aged and older individuals.

Conclusions

In summary, the present study demonstrated that METS-IR is an important independent predictor of new-onset diabetes in the Chinese general population, independent of traditional diabetes risk factors, especially in younger individuals.

Diabetes

Cohort study

METS-IR

independent predictor

The incidence of diabetes is increasing worldwide. This places a heavy burden of disease on countries around the world, especially in developing countries. Early identification of people at high risk of diabetes is important.
There is limited evidence on the relationship between METS-IR and the occurrence and progression of diabetes.
There is an urgent need for public health policies on the risk of diabetes in non-diabetic populations in China.

Diabetes, a hormonal disorder marked by excessively elevated levels of glucose in the bloodstream, is among the prevailing and rapidly increasing ailments globally, projected to impact approximately 693 million adults by 2045¹. Type 2 diabetes mellitus (T2DM), previously referred to as non-insulin-dependent diabetes or adult-onset diabetes, is a persistent metabolic disorder that typically manifests after the age of 35 to 40, encompassing over 90% of individuals with diabetes². Patients are characterized by hyperglycemia, a relative lack of insulin, and insulin resistance. Polydipsia, frequent urination, and unexplained weight loss are typical symptoms, which can also be accompanied by overeating, tiredness, or aches and pains. Prolonged elevated blood sugar levels can result in complications affecting small blood vessels and large blood vessels, such as cardiovascular disease, cerebrovascular accident, diabetic retinopathy leading to loss of vision, renal failure, impaired blood circulation to limbs necessitating amputation, and in rare cases, diabetic ketoacidosis. Identifying high-risk groups susceptible to diabetes at an early stage will greatly benefit due to the significant number of individuals with diabetes and the immense burden of the disease.

T2DM encompasses a range of conditions instead of being a singular disease, characterized primarily by elevated blood sugar levels, which serves as the ultimate shared factor for various metabolic abnormalities to intersect³. Peripheral insulin resistance (IR), also referred to as reduced insulin sensitivity, is a contributing factor to the onset of T2DM^{4, 5}. For a long time, fasting insulin-related measures like QUICKI and HOMA-IR have been the main approaches to assess IR⁶. However, alternative fasting IR indicators that do not rely on insulin have been created to address the limited usefulness and inconsistency of insulin-based indexes. These include the TyG index (the product of glucose and triglycerides), the TyG-BMI index (the product of glucose, triglycerides, and BMI), the TG/HDL-c ratio (the ratio of triglycerides to HDL-c), and METS-IR (a metabolic score for IR)^7–10. These new indicators replace insulin measurements with fasting triglycerides, glucose, and lipoprotein measures to improve practicality and reliability. Numerous studies conducted in recent years have demonstrated that the TyG index, TyG-BMI index, and TG/HDL-c ratio are dependable alternatives for IR and are linked to T2DM^11–13. The METS-IR, a metabolic score recently discovered, has shown a strong correlation with T2DM and a greater concordance with the hyperinsulinemic-euglycemic clamp⁹. The impact of time progression on the connection between diabetes and METS-IR remains uncertain. Hence, the aim of this present research was to examine the correlation between METS-IR and the risk of T2DM in the Chinese populace, as well as its potential for predicting diabetes, in order to provide novel insights into the early detection and prevention of diabetes.

Study design and participants

The present analysis is a retrospective study of a cohort investigation carried out by the China Rich Health Care Group, with the objective of enhancing the well-being of the Chinese populace and evaluating diabetes and its associated risk factors by means of health assessments and subsequent monitoring of the overall population. The study design has been extensively described in a prior study¹⁴. To summarize, the Cohort Study of the Wealthy Healthcare Group comprises the complete medical records of individuals aged 20 and above, who have made a minimum of two visits between the years 2010 and 2016 (n = 685277). Professor Chen¹⁴ has uploaded the study's data to the Dryad database, which consists of 211833 participants (116123 males and 95710 females) included in the analysis. Researchers have the freedom to utilize public data for post-analysis up to the maximum extent as stated in the terms of service of the Dryad database. Since the data is anonymous and the research ethics have already been approved in previous studies, there is no requirement to reapply for the current study. The study was designed with the purpose of using METS-IR as the exposure factor and investigating the occurrence of new diabetes events as the outcome of interest. The study excluded 97006 subjects who had incomplete covariate data, and finally analyzed 114827 participants who met the criteria.

Exposure Variables and Outcomes

ncident diabetes diagnosis was determined by a fasting plasma glucose level of ≥ 7.00mmol/L and/or self-reported diabetes throughout the follow-up duration.. Patients were censored at the date of diagnosis of diabetes or the final visit, whichever came first.

METS-IR was calculated as (ln ((2 × FPG) + TG) × BMI)/ (ln (HDL-C)). An autoanalyzer (Beckman 5800) was used to measure serum triglycerides (TG) and high-density lipoprotein cholesterol. The glucose oxidase method was used on an autoanalyzer (Beckman 5800) to measure plasma glucose levels. BMI was calculated by dividing weight in kilograms by height in meters squared.

Covariates

At every visit to the health check center, participants were asked to fill out a standardized survey that included details about their age, gender, family history of diabetes, blood pressure, height, weight, and smoking/drinking habits. A typical mercury sphygmomanometer was used to measure blood pressure in a quiet atmosphere. Weight was recorded with an accuracy of 0.1kg, with the individual wearing lightweight attire and excluding footwear. The height was measured to the closest 0.1cm. Professional workers obtained fasting venous blood following a minimum 10-hour fast and analyzed it using an automated analyzer in a standard laboratory (Beckman 5800).

Included in the analysis are the categorical variables of sex, smoking habits, alcohol consumption, and family medical background of diabetes. Continuous variables collected included age, height, weight, body mass index (BMI), systolic blood pressure (SBP), diastolic blood pressure (DBP), high-density lipoprotein cholesterol (HDL-C), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), triglyceride (TG), remnant cholesterol (RC), fasting plasma glucose (FPG), aspartate aminotransferase (AST), and alanine aminotransferase (ALT). The main objective was to establish the correlation between METS-IR and the risk of developing diabetes. Hence, the outcomes of the analyses for the modified potential confounders model were displayed in accordance with the guidelines provided by the STROBE statement¹⁵. As the model incorporated additional covariates, the odds ratio of the match experienced a minimum 10% alteration, necessitating the adjustment of this particular covariate. As a result, the model that was fully adjusted utilized the subsequent variables: age, sex, fasting blood glucose (FBG), systolic blood pressure (SBP), low-density lipoprotein cholesterol (LDL-c), high-density lipoprotein cholesterol (HDL-c), red blood cell count (RC), triglycerides (TG), alanine aminotransferase (ALT), aspartate aminotransferase (AST), family history of diabetes, smoking habits, and alcohol consumption.

Statistical Analysis

EmpowerStats 2.2 (http//www.empowerstats.com) and the R package (http //www.Rproject.org) were utilized for all statistical analyses. The METS-IR values were divided into five groups, namely Q1(≤ 28.06), Q2(28.07–31.50), Q3(31.51–34.97), Q4(34.98–39.25), and Q5(≥ 39.26).To examine the distinctions between these groups, Pearson's χ2 test was employed for categorical variables represented as percentages. Additionally, ANOVA and Kruskal-Wallis H tests were conducted for continuous variables presented as means ± SD. Subsequently, Tukey's HSD test and the Steel-Dwass test were utilized as post hoc tests. The relationship between diabetes and METS-IR was assessed using Multivariate Cox regression. After accounting for confounding variables, the hazard ratio (HR) and its corresponding 95% confidence interval (CI) were calculated for diabetes. This study utilized three models for conducting multivariate Cox regression analysis. Model I was adjusted for fundamental demographic variables including gender, age, and BMI. Model II was adjusted for variables that had a contribution of over 10% to the risk of METS-IR matching with diabetes. Lastly, model III represented the fully adjusted model. The Kaplan-Meier survival plot demonstrated the likelihood of developing diabetes across each quintile of METS-IR.Simultaneously, the AUC was computed using ROC curve analysis to compare the predictive capacities of METS-IR, TyG-BMI, TyG, TG, and HDL-c in assessing the risk of diabetes. In order to further examine the correlation between diabetes and METS-IR, the scientists additionally utilized a hierarchical Cox regression model for hierarchical analysis, aiming to investigate the connection between METS-IR and diabetes in individuals with varying ages, genders, and BMI levels. The likelihood ratio test was employed to assess disparities among various hierarchical groups in order to ascertain the presence of interactions. A P-value less than 0.05 was deemed to be statistically significant.

Baseline characteristics of the study subjects grouped by METS-IR quintiles

The study participants' baseline characteristics were categorized based on METS-IR quintiles. According to the METS-IR quintiles, Table 1 displays the initial attributes of the research participants (n = 114827; 62186 males and 52641 females). Table s1 shows the association of baseline variables with the prevalence of diabetes by univariate regression analyses. Their average HDL-c was 52.88 ± 11.52 mg/dL, their average BMI was 23.38 ± 3.30, their average FPG concentration was 89.09 ± 10.89 mg/dL, their average TG level was 123.00 ± 92.58 mg/dL, and the average value of the METS-IR index was 33.91 ± 6.59. The group with the highest METS-IR index quintile had the highest average age, BMI, SBP, DBP, TC, TG, FBG, LDL-c, RC, ALT, and AST values, while their average HDL-c levels were the lowest. The largest percentage of individuals who currently smoke, consume alcohol, and are male belonged to the fifth quintile of the METS-IR index.

Table 1

Baseline Characteristics of Five Groups
METS-IR quintiles	statistics	Q1 N = 22951	Q2 N = 22936	Q3 N = 22994	Q4 N = 22978	Q5 N = 22968	P-value
Age (year)	44.16 ± 12.94	38.96 ± 11.41	42.50 ± 12.36	45.25 ± 12.96	46.87 ± 13.10	47.22 ± 12.95	< 0.001
Height (cm)	166.32 ± 8.32	163.83 ± 7.40	164.90 ± 8.16	166.37 ± 8.49	167.61 ± 8.37	168.91 ± 8.14	< 0.001
Weight (kg)	64.99 ± 12.11	51.98 ± 5.68	58.76 ± 6.43	64.57 ± 7.22	70.28 ± 7.78	79.33 ± 10.24	< 0.001
BMI (kg/m²)	23.38 ± 3.30	19.33 ± 1.31	21.56 ± 1.15	23.28 ± 1.27	24.97 ± 1.46	27.75 ± 2.49	< 0.001
SBP (mmHg)	119.52 ± 16.69	110.95 ± 14.05	115.48 ± 15.24	119.81 ± 15.87	123.67 ± 16.25	127.69 ± 16.52	< 0.001
DBP.(mmHg)	74.51 ± 10.98	69.46 ± 9.30	71.61 ± 9.92	74.26 ± 10.32	77.05 ± 10.66	80.15 ± 11.19	< 0.001
FPG (mg/dL)	89.09 ± 10.89	84.65 ± 9.54	87.23 ± 9.76	89.19 ± 10.22	90.98 ± 10.93	93.40 ± 11.75	< 0.001
TC (mg/dL)	185.57 ± 34.59	179.31 ± 32.39	180.41 ± 33.30	185.69 ± 34.53	189.01 ± 34.76	193.40 ± 35.86	< 0.001
TG (mg/dL)	123.00 ± 92.58	69.01 ± 27.21	85.63 ± 37.81	108.78 ± 53.52	140.36 ± 72.87	211.13 ± 141.16	< 0.001
HDL-c (mg/dL)	52.88 ± 11.52	62.35 ± 11.27	56.11 ± 9.68	52.79 ± 9.13	49.18 ± 8.91	43.97 ± 9.35	< 0.001
LDL-c (mg/dL)	106.97 ± 26.61	100.28 ± 24.38	103.60 ± 25.31	108.43 ± 26.50	111.00 ± 26.66	111.54 ± 28.21	< 0.001
RC (mg/dL)	25.72 ± 17.12	15.08 (8.51–22.42)	18.94 (12.37–27.06)	22.81 (15.46–31.70)	27.06 (18.56–36.73)	34.79 (24.36–47.17)	< 0.001
ALT (µL)	23.92 ± 21.82	13.00 (10.40–17.10)	15.00 (11.70–20.70)	18.00 (13.50–25.40)	22.00 (16.00-31.50)	29.00 (20.00–43.00)	< 0.001
AST (µL)	24.07 ± 12.36	20.00 (17.00-23.30)	20.80 (18.00-24.40)	22.00 (18.50–26.00)	23.00 (19.70–28.00)	25.30 (21.00-31.50)	< 0.001
METS-IR	33.91 ± 6.59
Gender							< 0.001
Male	62186 (54.16%)	5602 (24.41%)	9239 (40.28%)	13094 (56.95%)	15950 (69.41%)	18301 (79.68%)
Female	52641 (45.84%)	17349 (75.59%)	13697 (59.72%)	9900 (43.05%)	7028 (30.59%)	4667 (20.32%)
Smoking status							< 0.001
Curent smoking	6642 (20.67%)	567 (10.10%)	843 (14.29%)	1146 (18.29%)	1666 (24.23%)	2420 (32.32%)
Ever smoking	1319 (4.10%)	91 (1.62%)	174 (2.95%)	282 (4.50%)	361 (5.25%)	411 (5.49%)
Never smoking	24179 (75.23%)	4954 (88.28%)	4882 (82.76%)	4838 (77.21%)	4849 (70.52%)	4656 (62.19%)
Driking status							< 0.001
Curent drinking	868 (2.70%)	59 (1.05%)	95 (1.61%)	152 (2.43%)	233 (3.39%)	329 (4.39%)
Ever drinking	5478 (17.04%)	468 (8.34%)	766 (12.99%)	1170 (18.67%)	1431 (20.81%)	1643 (21.94%)
Never drinking	25794 (80.26%)	5085 (90.61%)	5038 (85.40%)	4944 (78.90%)	5212 (75.80%)	5515 (73.66%)
Family history of diabetes							0.349
No	112239 (97.75%)	22469 (97.90%)	22401 (97.67%)	22479 (97.76%)	22465 (97.77%)	22425 (97.64%)
Yes	2588 (2.25%)	482 (2.10%)	535 (2.33%)	515 (2.24%)	513 (2.23%)	543 (2.36%)
Abbreviations: BMI, body mass index; ALT, alanine aminotransferase; AST, aspartate aminotransferase; HDL-C, high-density lipoprotein cholesterol; TC, total cholesterol; LDL-C, low density lipoprotein cholesterol; TG, triglyceride; RC, remnant cholesterol; FPG, fasting plasma glucose; SBP, systolic blood pressure; DBP, diastolic blood pressure; METS-IR, Metabolic score for insulin resistance.

Follow-up results

Figure 1 displays the results of the Kaplan-Meier analysis, which are based on the quintile of METS-IR. During the average follow-up period of 3.1 (0.95) years, 1879 men (3.1%) and 793 women (1.51%) were newly diagnosed with diabetes. The cumulative occurrence of diabetes in Group Q1 was 0.41% (94 out of 22,951), in Group Q2 it was 0.68% (157 out of 22,936), in Group Q3 it was 1.40% (323 out of 22,994), in Group Q4 it was not specified, and in Group Q5 it was 3.02% (695 out of 22,978) and 6.11% (1403 out of 22,968), respectively. As the METS-IR increased, the Q5 group experienced a considerably higher rise in diabetes incidence compared to the other groups.

Diabetes risk of METS-IR

The results of the multivariate Cox proportional hazards regression analysis for predicting diabetes based on the METS-IR index quintile are presented in Table 2. The findings showed a significant positive correlation between METS-IR and the likelihood of developing diabetes in the future (HR 1.11 per standard deviation increase, 95% confidence interval 1.10–1.11). Additionally, the risk of diabetes associated with METS-IR increased progressively when compared to the group with the lowest quintile (p-value for trend < 0.001). The HRs of developing diabetes increased in a dose-dependent manner when comparing the second, third, fourth, and fifth quintiles of the METS-IR index to the first quintile. The hazard ratios (HRs) for developing diabetes were 1.72 (95% confidence interval [CI] 1.33–2.22), 3.57 (95% CI 2.84–4.49), 7.48 (95% CI 6.03–9.27), and 14.37 (95% CI 11.66–17.70) in the second, third, fourth, and fifth quintiles of the METS-IR index, respectively. After accounting for age, sex, fasting blood glucose, systolic blood pressure, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, red blood cell count, triglycerides, alanine aminotransferase, aspartate aminotransferase, family history of diabetes, smoking habits, and alcohol consumption (Model 3), this strong positive correlation persisted in the fourth (HR 2.52, 95%CI 1.18–5.38, p = 0.0171) and fifth (HR 3.65, 95%CI 1.66–8.02, p = 0.0012) quintiles of the METS-IR index.

Table 2

Hazard Ratios and 95% CI for new-onset T2D according to METS-IR quintiles.
Exposure	Non-adjusted HR (95%CI) p-value	Model I HR (95%CI) p-value	Model II HR (95%CI) p-value	Model III HR (95%CI) p-value
METS-IR	1.11 (1.10, 1.11) < 0.0001	1.09 (1.08, 1.10) < 0.0001	1.06 (1.06, 1.07) < 0.0001	1.09 (1.06, 1.12) < 0.0001
METS-IR quintile
Q1	1.0	1.0	1.0	1.0
Q2	1.72 (1.33, 2.22) < 0.0001	1.17 (0.90, 1.51) 0.2442	1.35 (1.04, 1.74) 0.0232	0.59 (0.21, 1.68) 0.3214
Q3	3.57 (2.84, 4.49) < 0.0001	1.79 (1.40, 2.27) < 0.0001	2.15 (1.69, 2.72) < 0.0001	1.40 (0.62, 3.14) 0.4174
Q4	7.48 (6.03, 9.27) < 0.0001	2.95 (2.32, 3.76) < 0.0001	3.38 (2.69, 4.25) < 0.0001	2.52 (1.18, 5.38) 0.0171
Q5	14.37 (11.66, 17.70) < 0.0001	4.42 (3.40, 5.75) < 0.0001	4.76 (3.77, 6.01) < 0.0001	3.65 (1.66, 8.02) 0.0012
P for trend	< 0.001	< 0.001	< 0.001	< 0.001
Model I adjusted for age,gender and BMI;
Model II adjusted for SBP, FPG, HDL-C.
Model III adjusted for age, gender, FBG, SBP, LDL-c, HDL-c, RC, TG, ALT, AST, family history of diabetes, smoking status, drinking status.

Abbreviations: CI, confidence interval; HR, hazard ratios; METS-IR, Metabolic score for IR.

The impact of METS-IR in the new-onset diabetes in different subgroups

In order to further investigate the impact of METS-IR on the development of new cases of diabetes, we conducted additional analysis considering various factors such as gender, age, and BMI characteristics. The results showed that the age-related variation in the risk of diabetes associated with METS-IR was significant (P for interaction 0.05). According to Table 3, individuals with different phenotypes show varying levels of risk for developing diabetes, with certain groups displaying a significantly higher susceptibility. Among them, the risk of diabetes related to METS-IR was notably higher in younger people compared to middle-aged and older individuals [HR (per SD increase) 20–30 years: 1.29, 31–40 years: 1.13, 41–50 years: 1.11 versus 51–60 years: 1.07, 61–70 years: 1.05, > 70 years: 1.07].

Table 3

Stratified Associations Between METS-IR and New-Onset Diabetes by Age, gender, and BMI.
Subgroup	N	Unadjusted HR (95%CI) p-value	Adjusted HR (95%CI) p-value	P interaction
Age (year)				0.0208
<30	10859	1.17 (1.13, 1.21) < 0.0001	1.29 (1.09, 1.53) 0.0030
>=30, < 40	41076	1.14 (1.13, 1.16) < 0.0001	1.13 (1.08, 1.19) < 0.0001
>=40, < 50	26765	1.13 (1.12, 1.14) < 0.0001	1.11 (1.07, 1.16) < 0.0001
>=50, < 60	19339	1.09 (1.08, 1.10) < 0.0001	1.07 (1.02, 1.11) 0.0016
>=60, < 70	11922	1.07 (1.06, 1.08) < 0.0001	1.05 (1.00, 1.11) 0.0374
>=70	4866	1.05 (1.04, 1.07) < 0.0001	1.07 (1.00, 1.13) 0.0458
Gender				0.4942
Male	62043	1.10 (1.09, 1.10) < 0.0001	1.10 (1.07, 1.13) < 0.0001
Female	52362	1.13 (1.12, 1.13) < 0.0001	1.08 (1.04, 1.12) 0.0003
BMI (kg/m²)				0.3537
<24	67924	1.18 (1.16, 1.20) < 0.0001	1.09 (1.00, 1.18) 0.0445
>=24, < 28	36841	1.10 (1.09, 1.12) < 0.0001	1.03 (0.96, 1.10) 0.3771
>=28	10062	1.05 (1.04, 1.06) < 0.0001	1.05 (1.00, 1.12) 0.0631
Adjusted for: age, gender, FBG, SBP, LDL-c, HDL-c, RC, TG, ALT, AST, family history of diabetes, smoking status, drinking status. Abbreviations: BMI, body mass index; CI, confidence interval; HR, hazard ratios; METS-IR, Metabolic score for IR.

The METS-IR's ability to predict new-onset diabetes

The predictive value of METS-IR for new-onset diabetes was assessed by the researchers through ROC curve analysis (Fig. 2–3). The AUC of METS-IR for predicting new-onset diabetes was 0.7568 (0.7483–0.7654). Particularly in women, the area under the curve (AUC) of METS-IR is notably greater at 0.7809 (0.7646–0.7973), compared to the AUC of 0.7184 (0.7073–0.7295) in men (Table 4). Table 4 displayed the optimal threshold for METS-IR as 35.9850, with a sensitivity of 73.99% and a specificity of 66.21%.In addition, we assessed the AUCs of METS-IR, TYG, TYG-BMI, and TG/HDL, which serve as indicators capable of accurately representing the metabolic condition of the organism (refer to Fig.s1 and Table s2). METS-IR had a similar AUC to TYG (p = 0.112), and METS-IR had a significantly higher AUC than TG/HDL (p < 0.001). Nevertheless, in this study, METS-IR exhibited inferior performance compared to TyG-BMI in predicting the occurrence of diabetes.

Table 4

Areas under the receiver operating characteristic curves for METS-IR in identifying diabetes of different genders
Test	AUC	95%CI low	95%CI up	Best threshold	Specificity	Sensitivity
METS-IR	0.7568	0.7483	0.7654	35.9850	0.6621	0.7399
Male	0.7184	0.7073	0.7295	36.7750	0.5727	0.7541
Female	0.7809	0.7646	0.7973	33.8450	0.7317	0.7352

In this longitudinal cohort study with a 7-year follow-up, we discovered that among a population of Chinese adults who do not have diabetes but are part of a community, there is a positive and independent correlation between elevated METS-IR and the incidence of T2DM. Additionally, our findings revealed that METS-IR exhibited superior performance in predicting T2DM in females when compared to males. Among individuals with various phenotypes, it is evident that younger people have a considerably greater likelihood of developing METS-IR-associated diabetes compared to middle-aged and elderly individuals.

Insulin resistance, also known as IR, occurs when insulin becomes less effective at promoting the uptake and utilization of glucose for various reasons. As a result, the body compensates by secreting an excessive amount of insulin, leading to hyperinsulinemia, in order to keep blood sugar levels stable¹⁷. Metabolic syndrome and type 2 diabetes are more likely to occur in individuals with an inherited predisposition. In recent years, METS-IR, a novel index of insulin resistance that does not rely on insulin, has gained increasing attention. According to literary sources, METS-IR has been proposed as a straightforward and precise measure for assessing IR and forecasting the development of T2DM in the Western populace⁹. It was found that individuals in the highest quartile of METS-IR (> 50.39) had the greatest adjusted risk of T2DM (HR 3.91, 95% CI 2.25–6.81). Significant associations were observed between METS-IR (P < 0.001) and visceral, intrahepatic, and intrapancreatic fat as well as fasting insulin levels. Furthermore, in comparison to healthy individuals, participants who experienced T2DM incidents exhibited elevated initial METS-IR levels (50.2 ± 10.2 vs 44.7 ± 9.2, P < 0.001). The findings of an extensive investigation conducted on an elderly Chinese cohort revealed a significant correlation between METS-IR and both prediabetes and T2DM¹⁸. A similar conclusion was reached by another prospective cohort study involving data from non-obese adults in Japan¹⁹. After accounting for various potential confounding variables, the hazard ratio (HR) between the fourth quartile (Q4) and the first quartile (Q1) was 4.01 (95% confidence interval [CI], 1.39–11.57). This indicates that the likelihood of developing type 2 diabetes mellitus (T2DM) rises as the quartile of alteration in the METS-IR index increases. Our study revealed a robust correlation between elevated METS-IR levels and the susceptibility to developing new-onset diabetes, even after accounting for potential influencing variables. Additionally, we verified that increased METS-IR had greater predictability for the development of diabetes.

During our research, we have discovered fascinating occurrences in the analysis of subgroups. The risk of diabetes related to METS-IR varied significantly among different age phenotypes. Within the age stratification subgroup, individuals who were young or middle-aged faced a notably elevated likelihood of developing diabetes associated with METS-IR, compared to the older population. This strange occurrence is thought to be caused by young people's high-calorie diets, sedentary lifestyles, and lack of exercise. Changes in dietary patterns, high-fat, high-calorie diets, sedentary lifestyles, decreased physical activity, and excess energy are the main factors contributing to the resurgence of diabetes. These factors are leading to diabetes by inducing individuals to become obese and overweight, consequently elevating the likelihood of developing diabetes. Emerging evidence indicates that young-onset T2DM, occurring in individuals under the age of 40, exhibits a swifter decline in β-cell function compared to T2DM which develops later in life²⁰. Recent research has also indicated a significant increase in the prevalence of type 2 diabetes among youth worldwide over the last three decades, with obesity emerging as the primary contributing factor²¹. Diabetic complications pose a higher threat to individuals with diabetes who are young or middle-aged, as they experience prolonged periods of high blood sugar and a more rapid advancement of pathological processes. Our research findings indicate that the correlation between METS-IR and the development of diabetes is more pronounced in young and middle-aged individuals compared to older individuals. Hence, this metabolic indicator might prove more beneficial in forecasting the occurrence of diabetes in the future among individuals who are young or middle-aged.

Furthermore, it was found that METS-IR exhibited superior performance in females when predicting T2DM. While the exact cause of sexual disparity remains uncertain, it could potentially be associated with sex hormones and the various types of muscle fibers. As we know, gender differences exist in many diseases, such as cardiovascular disease, diabetes, metabolic syndrome, etc^{22, 23}. Prior studies have verified that numerous elements of energy and glucose balance are controlled in distinct ways among males and females, impacting their susceptibility to diabetes and related metabolic ailments²³. Insulin sensitivity in a large population of individuals with normal blood glucose levels was evaluated using the oral glucose insulin sensitivity index, and The findings revealed that, even after accounting for age and BMI, women exhibited higher insulin sensitivity compared to men²⁴. It is due to enhanced glucose uptake by skeletal muscle in women²⁵.

Limitations

The determination of diabetes in this research was based on self-disclosure by participants or FPG > 7.0 mmol/L throughout the monitoring period, which could potentially result in an underestimation of the actual occurrence of diabetes. As this study is an analysis conducted after previous studies¹⁴, despite making adequate adjustments, certain variables like glycosylated hemoglobin, exercise habits, and dietary habits were not included in the database. These omissions could potentially lead to residual confounding. Moreover, the observed rate of endpoint events is directly influenced by the relatively brief duration of patient follow-up. It may lead to an underestimation of disease incidence. However, the researchers still found a strong correlation between the two, suggesting that METS-IR may be of high value as a predictor of type 2 diabetes. Besides, the diabetes in this study was identified as T2DM, since the number of patients with T2DM exceeds 90% of all diabetes cases². However, this may ignore the impact of type 1 diabetes.

In conclusion, the present study demonstrated that METS-IR is an significantly independent predictor of new-onset diabetes in the Chinese general population, independent of traditional diabetes risk factors. METS-IR and diabetes have a causal relationship, and this relationship is stronger in younger people. METS-IR has a higher predictive value for new-onset diabetes in females than it does in males. These findings support the value of METS-IR in evaluation of diabetes and provide a simple and cost-effective approach to diabetes prevention.

METS-IR	metabolic score for insulin resistance
T2DM	type 2 diabetes mellitus
IR	insulin resistance
BMI	body mass index
SBP	systolic blood pressure
DBP	diastolic blood pressure
HDL-C	high-density lipoprotein cholesterol
TC	total cholesterol
LDL-C	low-density lipoprotein cholesterol
TG	triglyceride
RC	remnant cholesterol
FPG	fasting plasma glucose
AST	aspartate aminotransferase
ALT	alanine aminotransferase
HR	hazard ratio
CI	confidence interval

Ethics approval and consent to participate

Researchers have the freedom to utilize public data for post-analysis up to the maximum extent as stated in the terms of service of the Dryad database. Since the data is anonymous and the research ethics have already been approved in previous studies, there is no requirement to reapply for the current study.

Funding: The study was supported by grants from Health and Family planning Commission of Liaoning, China (No. LNCCC-D18-2015) and the Science and Technology Program of Dalian, China (No. 2015E12SF168).

Informed Consent Statement: Due to the anonymity of the data, research ethics has been approved in previous studies, so there was no need to reapply in this study

Data Availability Statement: The data used in this study have been uploaded to the “Dryad” database by Professor Chen et al.

Acknowledgments: The authors appreciate the time and effort given by participants during the data collection phase of the China Rich Health Care Group.

Conflicts of Interest: The authors declare that they have no competing interests.

Authors' contributions:

Yajuan Lin: Data analysis, Methodology and Writing.

Zexin Liu: Formal analysis .

Ziwen Li and Jiatian Li: Validation.

Xiaoyan Bao: Methodology.

Yunlong Xia: Supervision.

Bo Zhang: Writing - Review & Editing and Funding acquisition.

Cho, N. H.; Shaw, J. E.; Karuranga, S.; Huang, Y.; da Rocha Fernandes, J. D.; Ohlrogge, A. W.; Malanda, B., IDF Diabetes Atlas: Global estimates of diabetes prevalence for 2017 and projections for 2045. Diabetes Res Clin Pract 2018, 138, 271-281.
Magliano, D. J.; Boyko, E. J., In IDF DIABETES ATLAS, 10th ed.; Brussels, 2021.
Cole, J. B.; Florez, J. C., Genetics of diabetes mellitus and diabetes complications. Nat Rev Nephrol 2020, 16 (7), 377-390.
Bello-Chavolla, O. Y.; Rojas-Martinez, R.; Aguilar-Salinas, C. A.; Hernandez-Avila, M., Epidemiology of diabetes mellitus in Mexico. Nutr Rev 2017, 75 (suppl 1), 4-12.
Gastaldelli, A.; Gaggini, M.; DeFronzo, R. A., Role of Adipose Tissue Insulin Resistance in the Natural History of Type 2 Diabetes: Results From the San Antonio Metabolism Study. Diabetes 2017, 66 (4), 815-822.
Borai, A.; Livingstone, C.; Ferns, G. A., The biochemical assessment of insulin resistance. Ann Clin Biochem 2007, 44 (Pt 4), 324-42.
Abbasi, F.; Reaven, G. M., Comparison of two methods using plasma triglyceride concentration as a surrogate estimate of insulin action in nondiabetic subjects: triglycerides x glucose versus triglyceride/high-density lipoprotein cholesterol. Metabolism 2011, 60 (12), 1673-6.
Bastard, J. P.; Vandernotte, J. M.; Faraj, M.; Karelis, A. D.; Messier, L.; Malita, F. M.; Garrel, D.; Prud'homme, D.; Rabasa-Lhoret, R., Relationship between the hyperinsulinemic-euglycaemic clamp and a new simple index assessing insulin sensitivity in overweight and obese postmenopausal women. Diabetes Metab 2007, 33 (4), 261-8.
Bello-Chavolla, O. Y.; Almeda-Valdes, P.; Gomez-Velasco, D.; Viveros-Ruiz, T.; Cruz-Bautista, I.; Romo-Romo, A.; Sanchez-Lazaro, D.; Meza-Oviedo, D.; Vargas-Vazquez, A.; Campos, O. A.; Sevilla-Gonzalez, M. D. R.; Martagon, A. J.; Hernandez, L. M.; Mehta, R.; Caballeros-Barragan, C. R.; Aguilar-Salinas, C. A., METS-IR, a novel score to evaluate insulin sensitivity, is predictive of visceral adiposity and incident type 2 diabetes. Eur J Endocrinol 2018, 178 (5), 533-544.
Simental-Mendia, L. E.; Rodriguez-Moran, 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 (4), 299-304.
Du, T.; Yuan, G.; Zhang, M.; Zhou, X.; Sun, X.; Yu, X., Clinical usefulness of lipid ratios, visceral adiposity indicators, and the triglycerides and glucose index as risk markers of insulin resistance. Cardiovasc Diabetol 2014, 13, 146.
Er, L. K.; Wu, S.; Chou, H. H.; Hsu, L. A.; Teng, M. S.; Sun, Y. C.; Ko, Y. L., Triglyceride Glucose-Body Mass Index Is a Simple and Clinically Useful Surrogate Marker for Insulin Resistance in Nondiabetic Individuals. PLoS One 2016, 11 (3), e0149731.
Mohd Nor, N. S.; Lee, S.; Bacha, F.; Tfayli, H.; Arslanian, S., Triglyceride glucose index as a surrogate measure of insulin sensitivity in obese adolescents with normoglycemia, prediabetes, and type 2 diabetes mellitus: comparison with the hyperinsulinemic-euglycemic clamp. Pediatr Diabetes 2016, 17 (6), 458-65.
Chen, Y.; Zhang, X. P.; Yuan, J.; Cai, B.; Wang, X. L.; Wu, X. L.; Zhang, Y. H.; Zhang, X. Y.; Yin, T.; Zhu, X. H.; Gu, Y. J.; Cui, S. W.; Lu, Z. Q.; Li, X. Y., Association of body mass index and age with incident diabetes in Chinese adults: a population-based cohort study. BMJ Open 2018, 8 (9), e021768.
von Elm, E.; Altman, D. G.; Egger, M.; Pocock, S. J.; Gotzsche, P. C.; Vandenbroucke, J. P.; Initiative, S., The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies. Int J Surg 2014, 12 (12), 1495-9.
Vandenbroucke, J. P.; von Elm, E.; Altman, D. G.; Gotzsche, P. C.; Mulrow, C. D.; Pocock, S. J.; Poole, C.; Schlesselman, J. J.; Egger, M.; Initiative, S., Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. PLoS Med 2007, 4 (10), e297.
Lebovitz, H. E., Insulin resistance: definition and consequences. Exp Clin Endocrinol Diabetes 2001, 109 Suppl 2, S135-48.
Cheng, H.; Yu, X.; Li, Y. T.; Jia, Z.; Wang, J. J.; Xie, Y. J.; Hernandez, J.; Wang, H. H. X.; Wu, H. F., Association between METS-IR and Prediabetes or Type 2 Diabetes Mellitus among Elderly Subjects in China: A Large-Scale Population-Based Study. Int J Environ Res Public Health 2023, 20 (2).
Cai, X. T.; Zhu, Q.; Liu, S. S.; Wang, M. R.; Wu, T.; Hong, J.; Hu, J. L.; Li, N., Associations Between the Metabolic Score for Insulin Resistance Index and the Risk of Type 2 Diabetes Mellitus Among Non-Obese Adults: Insights from a Population-Based Cohort Study. Int J Gen Med 2021, 14, 7729-7740.
Magliano, D. J.; Sacre, J. W.; Harding, J. L.; Gregg, E. W.; Zimmet, P. Z.; Shaw, J. E., Young-onset type 2 diabetes mellitus - implications for morbidity and mortality. Nat Rev Endocrinol 2020, 16 (6), 321-331.
Xie, J.; Wang, M.; Long, Z.; Ning, H.; Li, J.; Cao, Y.; Liao, Y.; Liu, G.; Wang, F.; Pan, A., Global burden of type 2 diabetes in adolescents and young adults, 1990-2019: systematic analysis of the Global Burden of Disease Study 2019. BMJ 2022, 379, e072385.
Regitz-Zagrosek, V.; Kararigas, G., Mechanistic Pathways of Sex Differences in Cardiovascular Disease. Physiol Rev 2017, 97 (1), 1-37.
Tramunt, B.; Smati, S.; Grandgeorge, N.; Lenfant, F.; Arnal, J. F.; Montagner, A.; Gourdy, P., Sex differences in metabolic regulation and diabetes susceptibility. Diabetologia 2020, 63 (3), 453-461.
Kautzky-Willer, A.; Brazzale, A. R.; Moro, E.; Vrbikova, J.; Bendlova, B.; Sbrignadello, S.; Tura, A.; Pacini, G., Influence of increasing BMI on insulin sensitivity and secretion in normotolerant men and women of a wide age span. Obesity (Silver Spring) 2012, 20 (10), 1966-73.
Lundsgaard, A. M.; Kiens, B., Gender differences in skeletal muscle substrate metabolism - molecular mechanisms and insulin sensitivity. Front Endocrinol (Lausanne) 2014, 5, 195.

No competing interests reported.

supplementary.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

The Metabolic Score for Insulin Resistance (METS-IR) as a Predictor of Incident diabetes: A Longitudinal Study among Chinese without Diabetes

Status:

Version 1

Abstract

Figures

Contributions to the literature

Introduction

Methods

Study design and participants

Exposure Variables and Outcomes

Covariates

Statistical Analysis

Results

Baseline characteristics of the study subjects grouped by METS-IR quintiles

Follow-up results

Diabetes risk of METS-IR

The impact of METS-IR in the new-onset diabetes in different subgroups

The METS-IR's ability to predict new-onset diabetes

Discussion

Limitations

Conclusions

Abbreviations

Declarations

Authors' contributions:

References

Additional Declarations

Supplementary Files

Status:

Version 1