Hypertension is a primary risk factor for cardiovascular disease and all-cause mortality (ACM) (1, 2). Careful risk factor assessment plays an important role in hypertension patients. The metabolic effects of insulin resistance (IR), including hyperglycemia and dyslipidemia, appear to interact synergistically with increased blood pressure to cause vascular and kidney injury (3). IR is proposed to be important contributors to hypertension-caused organ damage, arterial stiffness (4), and left ventricular diastolic dysfunction (5) due to hypertension. Hence, higher IR is a risk factor for hypertension patients.

At present, Homeostatic model assessment IR (HOMA-IR), a method for assessing β-cell function and IR, is frequently applied, but its high cost and inconvenience make it difficult to use in large cohort study. Therefore, there has been no large cohort study examinate the prognosis value of IR in hypertension group.

Triglyceride-Glucose index (TyG index), first published by South American authors, showed a good correlation with the insulin clamp technique and HOMA-IR index. The TyG index, described to be a simple, convenient, and low-cost method, is of research interest in many countries in Asia and can be used to screen for IR in the Asian hypertensive community (6). Despite this, there is no evidence between TyG index and prognosis value in North America hypertensive community. Therefore, we conducted a retrospective cohort study to certify the relationship of cardiovascular mortality (CVM) and ACM in hypertension patients.

In this retrospective study, we, for the first time, revealed association between IR (TyG index), ACM, and CVM among patients with hypertension in a large cohort study. We demonstrated that the association was non-linear between ACM and TyG index in elderly patient with aged ≥60 years old. But for CVM, only when TyG index was more than 9.52 will the risk for CVM increase among the whole hypertension group.

We also demonstrated the association was non-linear between ACM and TyG index in elderly with aged ≥60 years old. When TyG index ≥9.45, the risk of ACM increased by 48% at every 1 SD increase of TyG index. In contrast, for TyG index <9.45, the risk of ACM increased by 19% as the TyG index decreased by 1 SD. TyG index, a composite indicator based on TG level and FBG value, was shown to be used as a surrogate marker of IR (9, 11, 12). Previous studies probed the relationship between IR and ACM in different population. An observational prospective cohort study that enrolled 15,773 patients with type 2 diabetes showed that highest IR predicts ACM in type 2 diabetes (13). High IR, as measured by HOMA-IR, identified postmenopausal women at higher risk for cancer-specific and ACM (14). In addition, A meta-analysis included seven articles involving 26,976 non-diabetic adults showed that the highest HOMA-IR increased the risk of ACM by 34% when compared with the lowest category (15). For elderly people, there were different conclusions. Among community-dwelling older individuals over 65 years, HOMA-IR and low-grade systemic inflammation was associated with a 9-year ACM and CVM risk (16). Aging and high C-reactive protein levels were, usually, risk factors for cardiovascular disease mortality. We did not evaluate systemic inflammation in our study. A community-based prospective study selected participants with age ≥65 years in Korea (17). Elderly subjects in the fifth and first quintile of HOMA-IR values had increased rates of ACM and CVM as there is a threshold level of HOMA-IR in relation to mortality. The HOMA-IR was lower in Korea study (1st quintile ≤ 0.67 and 5th quintile > 1.50). Our work showed similar with the study in Korea; however, we used the TyG index to evaluate IR, not HOMA-IR, in elderly hypertension patients.

A previous study showed that there was a striking U-shaped relationship between FBG levels and in-hospital and 3-year mortality in older patients with acute myocardial infarction (AMI) (18). The study observed that mild to moderately low FBG levels ( ≤ 5 mmol/L, 90 mg/dl) were associated with a relative increase in mortality risk. Hypoglycemia and rapid changes in blood glucose level were shown to increase levels of counterregulatory hormones, such as epinephrine and norepinephrine, which may induce vasoconstriction and platelet aggregation, resulting in ischemia of cardiovascular or cerebrovascular (19). In the elderly population, the decrease in skeletal muscle mass is generally accompanied by weight loss. A study that recruited elderly individuals (≥65 years) in Taiwan showed that elderly subjects with sarcopenic obesity had the highest ACM (19). In contrast, except for the TG level, none of the other lipid profile indices were related to ACM in patients aged over 75 years (20). The mortality risk decreased by 17% for each 1 mmol/L increase serum level of TG in ZODIAC study (20). The lower TG had higher risk of ACM for elderly population. One study, which enrolled individuals aged 65 and older residing in northern Manhattan in 1992–1994 and 1999–2002, showed low cholesterol level was a robust predictor of mortality in the non-demented elderly and may be a surrogate of frailty or subclinical disease (21). Our study was collected from 1999 to 2014 in the same country as the study (21), so the life diet structure and level were similar. The mean TG of the patients with mortality was 155 mg/dl (21). Therefore, the finding of our study, specifically of the 1 SD decrease of TyG index when TyG index <9.45 in elderly with aged ≥60 years and 19% increased risk of ACM was persuasive. Other factors, such as body composition and nutritional state, might be predictors of mortality in the elderly population (17). Older age, anemia, lower baseline hemoglobin level (22), and lower BMI had a greater effect on mortality.

Only when TyG index ≥9.52, TyG-index showed significant with CVM. A previous study had reported that IR was associated with cardiovascular disease (23). The relationship between IR and endothelial dysfunction may the underlying biological mechanism of cardiovascular. This mostly includes inflammation (24) and functional impairment in the endothelium of blood vessels (25) which leads to atherosclerosis and cardiovascular disease. This simple, convenient, and low-cost TyG index was of research interest in many Countries and could be used to screen for IR in the hypertensive community. Furthermore, Park et al. (26) reported that TyG-index is an independent predictor of coronary artery calcium progression. We first investigate relationship between TyG index and CVM and ACM in hypertension. The Kaplan-Meier survival curve by TyG index groups also confirmed that the higher TyG index was associated with increased higher incident of ACM and CVM.

Our study still had several limitations that need to be mentioned. First, we did not have multiple-time monitoring of the TyG index along the follow-up, which may provide more information. Secondly, we did not compare the TyG index with HOMA-IR and the hyperinsulinemic-euglycemic clamp test. Thirdly, nutritional habits or energy intake were not recorded. Although we did not adjust for these potential confounding variables, we used other variables, such as BMI or cholesterol levels, which are indirectly related to nutritional habits or energy intake. Finally, this was a population-based study conducted among hypertension participants in the United States. Therefore, our findings may not be generalizable to other populations.

We first time revealed association between IR (TyG index), ACM, and CVM among patients with hypertension. We demonstrated that the association was non-linear between ACM and TyG index in elderly with aged ≥60 years. But for CVM, higher TyG index had high risk among patients with hypertension only when TyG index ≥9.52.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.

The studies involving human participants were reviewed and approved by the Institutional Review Board of the Centers for Disease Control and Prevention. The patients/participants provided their written informed consent to participate in this study.

DZ and Y-qF contributed to the conception or design of the study and drafted the manuscript. X-cL contributed to the acquisition of data, interpretation of data, and analysis of data. Y-qH and LK contributed to the interpretation of data and critical revision of the article for important intellectual content. All authors gave final approval of the article.

This research was supported by Science and Technology Plan Program of Guangzhou (201803040012), the Key Area R&D Program of Guangdong Province (No. 2019B020227005), Guangdong Provincial People's Hospital Clinical Research Fund (Y012018085), the Fundamental and Applied Basic Research Foundation Project of Guangdong Province (2020A1515010738), High-level Hospital Construction Project of Guangdong Provincial People's Hospital (DFJH2020022), and Guangdong Provincial Clinical Research Center for Cardiovascular Disease (2020B1111170011).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.