ABSTRACT
Background
The ratio of high‐density lipoprotein cholesterol (HDL‐C) to low‐density lipoprotein cholesterol (LDL‐C) predicts cardiovascular disease (CVD) endpoints, yet its prognostic validity in high‐risk populations and for type 2 diabetes mellitus (T2DM)‐related adverse events remains unestablished.
Methods
This study included 32,609 people aged 35–75 years in Fujian Province, China, who were at high risk for CVD. The primary endpoint was all‐cause mortality during follow‐up. Cox proportional hazard models and restricted cubic spline (RCS) analysis were used to evaluate the correlation between the HDL‐C/LDL‐C ratio and the endpoints.
Result
On the basis of the restricted RCS curve, the participants were classified as having a low (< 0.3), middle (0.3–0.5), or high (> 0.5) HDL‐C/LDL‐C ratio. Multivariate Cox regression analyses revealed that the risk of all‐cause mortality (HR = 1.48, 95% CI 1.14–1.93, p < 0.01 for low; HR = 1.30, 95% CI 1.06–1.58, p < 0.05 for high) was increased in the low and high groups. Participants without T2DM who were at high risk for CVD had similar prognoses (HR = 1.65, 95% CI 1.19–2.28, p < 0.01 for low; HR = 1.35, 95% CI 1.05–1.74, p < 0.01 for high). However, this association was not found in participants with T2DM who were at high risk for CVD.
Conclusion
HDL‐C/LDL‐C can be used to predict the prognosis of individuals at high risk for CVD, and maintaining HDL‐C/LDL‐C ratios between 0.3 and 0.5 may be the most helpful range for this population. Furthermore, maintaining this ratio range holds clinical significance for cohorts without T2DM, although further exploration is needed in this T2DM cohort.
Keywords: high‐density lipoprotein cholesterol (HDL‐C), low‐density lipoprotein cholesterol (LDL‐C), populations at high risk for CVD, prognosis, type 2 diabetes mellitus (T2DM)
Summary
The HDL‐C/LDL‐C ratio was significantly associated with all‐cause mortality in populations at high cardiovascular disease (CVD) risk.
This study utilized real‐world data from the large‐scale Fujian Cardiac Metabolic Diseases and Comorbidity Cohort (Fu‐CARE), enhancing the generalizability and reliability of the findings.
The HDL‐C/LDL‐C ratio may serve as a valuable lipid‐based predictor, providing clinical implications for risk stratification and management in high CVD risk populations.
Abbreviations
- ApoB
apolipoprotein B
- CI
confidence interval
- CVD
cardiovascular disease
- FBG
fasting blood glucose
- HDL‐C
high‐density lipoprotein cholesterol
- HRs
hazard ratios
- ICH
intracranial hemorrhage
- LDL‐C
low‐density lipoprotein cholesterol
- OGTT
oral glucose tolerance test
- RCS
restricted cubic spline
- SD
mean ± standard deviation
- T2DM
type 2 diabetes mellitus
- TC
total cholesterol
1. Introduction
Cardiovascular disease (CVD) is the primary cause of death worldwide [1], and China is a large country with a high CVD incidence [2]. Controlling lipids is an effective strategy for lowering the risk of CVD [3], and it can improve the adverse prognosis associated with CVD [4]. Type 2 diabetes mellitus (T2DM) is the primary cause of morbidity and mortality, driving the increased risk of CVD death [5] and is directly linked to lipid metabolism [6]. Therefore, when identifying more valuable lipid indicators for primary prevention and prognosis improvement, it is essential not only to conduct exploration in populations at high CVD risk but also to further investigate these indicators in people with different diabetes statuses.
Traditional dyslipidaemia is characterized by reduced high‐density lipoprotein cholesterol (HDL‐C) and elevated low‐density lipoprotein cholesterol (LDL‐C) [3, 4, 7]. Studies have confirmed that extremely high HDL‐C and excessively low LDL‐C levels increase the likelihood of adverse prognoses in CVD patients [8, 9]. The HDL‐C/LDL‐C ratio has recently gained recognition as a marker of cholesterol control and is widely used to prevent stroke and coronary heart disease [10, 11]. Studies have shown that mortality from stroke and coronary heart disease is lowest when the HDL‐C/LDL‐C ratio is between 0.4 and 0.6. This ratio has been found to be a more accurate predictor of clinical disease risk than individual lipoprotein levels [12, 13]. However, the existing research on the HDL‐C/LDL‐C ratio is still insufficient, particularly in populations at high risk for CVD.
Furthermore, recent research has shown that the HDL‐C/LDL‐C ratio is a marker of insulin resistance [14] and is associated with T2DM [15]. Kuang et al. reported that the HDL‐C/LDL‐C ratio is valuable in prediabetes risk assessment [16]. However, in vitro experiments have shown that derangements in insulin signaling in individuals with T2DM produce dyslipidaemia [17]. Thus, whether the HDL‐C/LDL‐C ratio can serve as a reliable prognostic biomarker in populations with T2DM at high risk for CVD remains an open question. To address this, our study aims to investigate the association between the HDL‐C/LDL‐C ratio and adverse prognoses in populations at high risk for CVD and to compare its usefulness between populations with and without T2DM. In addition, our study aims to offer valuable reference metrics for the clinical management of blood lipids in populations at high risk for CVD.
2. Methods
2.1. Study Design and Patients
The current analysis is based on the Fujian Cardiometabolic Diseases and Comorbidities Cohort (Fu‐CARE, Trial No. NCT06102187), an observational study conducted between 2017 and 2021 to assess cardiovascular disease (CVD) risk. Written informed consent was obtained from all patients.
A total of 35658 individuals considered at high risk for CVD were recruited from nine cities in Fujian Province. The inclusion criteria for the subjects were as follows: (1) age 35–75 years; (2) residence in an eligible location for more than 6 months before enrollment; and (3) no cardiovascular disease history or major disability. The CVD high‐risk group was defined as the population without a history of cardiovascular or cerebrovascular illness and with any of the following characteristics: (1) low‐density lipoprotein cholesterol (LDL‐C) ≥ 4.16 mmol/L; (2) high‐density lipoprotein cholesterol (HDL‐C) ≤ 1.04 mmol/L; (3) systolic blood pressure (SBP) ≥ 140 mmHg or diastolic blood pressure (DBP) ≥ 90 mmHg; or (4) 10‐year cardiovascular disease risk ≥ 20%, as estimated by the World Health Organization (WHO) Cardiovascular Disease Risk Chart. Patients meeting any of the following criteria were excluded: (1) absence of indicators associated with HDL‐C and LDL‐C; (2) fasting glucose levels less than 3.90 mmol/L or type I diabetes mellitus (T1DM); (3) cancer disorders; and (4) hyperlipidaemia in the family. Ultimately, we recruited 32,609 participants, and their complete information is shown in Figure 1.
Figure 1.
Flow diagram of population selection in this study.
2.2. Data Collection and Prognosis
People at high risk for CVD underwent the same questionnaire interview, systematic physical examination, and blood sample collection every year. Fasting venous blood samples were collected in the laboratory and measured for total cholesterol (TC), HDL‐C, LDL‐C, triglyceride (TG), and fasting blood glucose (FBG) levels.
The follow‐up period extended from the date of registration until the occurrence of death from any cause or the end of follow‐up (in 2021), whichever occurred first. This retrospective study adhered to the Declaration of Helsinki.
2.3. Definitions
The criteria for dyslipidaemia were TC ≥ 6.24 mmol/L, LDL‐C ≥ 4.16 mmol/L, and HDL‐C ≤ 1.04 mmol/L, or treatment with a dyslipidaemia drug [18]. Hypertension was defined as three consecutive measurements of SBP ≥ 140 mmHg and/or DBP ≥ 90 mmHg without receiving blood pressure medication, a self‐reported history of hypertension, or being on antihypertensive therapy [19]. The T2DM diagnostic criteria describe diabetes mellitus (DM) as follows: diabetes can be diagnosed with random blood glucose levels of ≥ 11.1 mmol/L, fasting blood glucose levels of ≥ 7.0 mmol/L, 2‐h oral glucose tolerance test (OGTT) results of ≥ 11.1 mmol/L, or a self‐reported history of diabetes [20].
2.4. Statistical Analysis
The associations between the HDL‐C/LDL‐C ratio and all‐cause mortality were analyzed using restricted cubic spline curves (RCSs). On the basis of the RCS results, the population at high risk for CVD was categorized into three groups using thresholds of 0.3 and 0.5: low HDL‐C/LDL‐C (< 0.3), middle (HDL‐C/LDL‐C between 0.3 and 0.5), and high HDL‐C/LDL‐C (> 0.5).
The baseline characteristics of each group are expressed as the means ± standard deviations (SDs), quartiles, or frequencies and percentages, as appropriate. Survival data passed the proportional hazards assumption (PHA) (Figure S1). The survival rates of the three groups were compared via Kaplan‒Meier survival curves. The relationship between HDL‐C/LDL‐C and adverse outcomes was further assessed using Cox proportional hazards regression modeling, and hazard ratios (HRs) were calculated. The results of the study are expressed as HRs and 95% confidence intervals (CIs).
Four different models were developed for this study. Model 1 was adjusted for sex and age. Model 2 was adjusted for HTN, BMI, waist circumference, geographic region, education level, and smoking history on the basis of Model 1. Model 3 was based on Model 2 but adjusted for T2DM. Model 4 was based on Model 3 but adjusted for TG. The purpose of Models 2, 3, and 4 was to assess whether the association between different HDL‐C/LDL‐C groups and mortality was influenced by sociodemographic variables and other cardiovascular risk factors. The main and interaction effects of age, sex, hypertension, smoking, and T2DM on adverse outcomes were analyzed.
All the data were analyzed in R software (V.4.3.0), and differences were considered statistically significant at p < 0.05.
2.5. Covariate Selection
The variables that affected all‐cause mortality but were not collinear with HDL‐C/LDL‐C were selected, as shown in Figure S2. Some additional variables were selected according to clinical experience.
2.6. Sensitivity Analysis
Multiple model adjustments were used for sensitivity analyses to illustrate the robustness of the results: Model 1 was adjusted for sex and age; Model 4 was adjusted for HTN, T2DM, BMI, TG, waist circumference, geographic region, education level, and smoking history in addition to the variables in Model 1. The complete model qualification criteria and corresponding covariate adjustment strategies are detailed in Table S1.
3. Results
3.1. Baseline Characteristics
The study included 32,609 participants, with a median follow‐up duration of 3.44 ± 1.10 years. The overall mean age of this population was 58.018 ± 9.578 years; 60.189% were females, the mean HDL level was 1.424 ± 0.400 mmol/L, and the mean LDL level was 3.178 ± 1.230 mmol/L. In this population, 7111 (21.807%) patients were diagnosed with T2DM. The study patients were allocated into low, middle and high groups according to their HDL‐C/LDL‐C ratios. The participants in the low group had the greatest waist circumference and BMI as well as the highest LDL, GLU, TC, and TG levels (p < 0.05). The participants in the high group had the highest blood pressure, HTN, HDL, and TG (p < 0.05). The middle group included larger proportions of patients aged ≥ 60 years, smokers, people with T2DM, and people living in rural areas than the low and high groups did (p < 0.05) (Table 1).
Table 1.
Baseline characteristics of the study patients.
| Characteristics | [ALL] | Low (< 0.3) | Middle (0.3–0.5) | High (> 0.5) | p overall |
|---|---|---|---|---|---|
| N = 32609 | N = 5273 | N = 14190 | N = 13146 | ||
| Age (years) | 58.018 (9.578) | 58.047 (9.128) | 58.361 (9.322) | 57.636 (10.004) | < 0.001 |
| ≥ 60 years | 15,428 (47.312%) | 2443 (46.330%) | 6828 (48.118%) | 6157 (46.836%) | 0.031 |
| Female | 19,627 (60.189%) | 2925 (55.471%) | 8782 (61.889%) | 7920 (60.246%) | < 0.001 |
| Waist (cm) | 84.057 (9.102) | 85.835 (8.486) | 84.486 (8.949) | 82.882 (9.346) | < 0.001 |
| BMI | 24.540 (3.198) | 25.055 (3.013) | 24.718 (3.170) | 24.140 (3.253) | < 0.001 |
| Rural | 23,856 (73.158%) | 3692 (70.017%) | 10,608 (74.757%) | 9556 (72.691%) | < 0.001 |
| SBP (mmHg) | 146.730 (23.386) | 138.631 (20.661) | 146.005 (22.981) | 150.760 (23.923) | < 0.001 |
| DBP (mmHg) | 85.779 (13.107) | 83.030 (11.750) | 85.526 (12.986) | 87.156 (13.556) | < 0.001 |
| HR (beats/min) | 76.247 (10.938) | 75.849 (10.632) | 76.156 (10.686) | 76.505 (11.317) | < 0.001 |
| Smoker | 26,280 (80.591%) | 4032 (76.465%) | 11,547 (81.374%) | 10,701 (81.401%) | < 0.001 |
| HTN | 22,356 (68.558%) | 2995 (56.799%) | 9558 (67.357%) | 9803 (74.570%) | < 0.001 |
| T2DM | 7111 (21.807%) | 1204 (22.833%) | 3147 (22.178%) | 2760 (20.995%) | 0.009 |
| TC mmol/L | 5.370 (1.352) | 6.453 (1.243) | 5.711 (1.212) | 4.568 (1.042) | < 0.001 |
| HDL‐C mmol/L | 1.424 (0.400) | 1.127 (0.260) | 1.386 (0.352) | 1.585 (0.416) | < 0.001 |
| TG mmol/L | 1.480 [1.080;2.070] | 1.510 [1.120;2.060] | 1.740 [1.320;2.300] | 1.320 [0.970;1.930] | < 0.001 |
| LDL‐C mmol/L | 3.178 (1.232) | 4.532 (0.999) | 3.553 (0.944) | 2.230 (0.786) | < 0.001 |
| GLU mmol/L | 5.800 [5.300;6.600] | 5.800 [5.300;6.600] | 5.900 [5.400;6.700] | 5.800 [5.300;6.500] | < 0.001 |
| T2DM_drug | 3375 (10.350%) | 529 (10.032%) | 1437 (10.127%) | 1409 (10.718%) | 0.196 |
| HTN_drug | 10,657 (32.681%) | 1467 (27.821%) | 4548 (32.051%) | 4642 (35.311%) | < 0.001 |
| Lipid_drug | 1478 (4.532%) | 175 (3.319%) | 558 (3.932%) | 745 (5.667%) | < 0.001 |
Note: Data are presented as n (%), mean ± SD, or median (IQR).
Abbreviations: BMI, body mass index; DBP, diastolic blood pressure; GLU, glucose; HDL‐C, high‐density lipo‐protein cholesterol; HR, heart rate; HTN, hypertension; IQR, interquartile range; LDL‐C, low‐density lipoprotein cholesterol; SBP, systolic blood pressure; SD, standard deviation; T2DM, Type 2 Diabetes Mellitus; TC, total cholesterol; TG, triglyceride.
3.2. Association of HDL‐C/LDL‐C With All‐Cause Mortality
On the basis of the RCS analysis results (Figure 2), we identified a nonlinear U‐shaped relationship between HDL‐C/LDL‐C and the participants' all‐cause mortality, with thresholds of 0.3 and 0.5, and the participants were stratified into low, middle, and high groups. Compared with the other groups, the RCS analysis revealed that the middle group had the lowest all‐cause mortality risk.
Figure 2.
Restricted cubic spline curves with logistic regression between HDL‐C/LDL‐C and all‐cause mortality. Red lines represent references for hazard ratios (HR), and the red shaded area represents 95% confidence intervals (95% CI). The left Y‐axis displays the hazard ratio (HR) with 95% confidence interval (CI), while the right Y‐axis shows the number of all‐cause mortality.
The survival data fulfilled the proportional hazards assumption (PHA) (Figure S1). Kaplan–Meier survival analysis revealed that the cumulative all‐cause mortality rate was higher in the low and high groups than in the middle group (p < 0.05) (Figure S3). A Cox proportional risk analysis verified this finding. Figure 3 shows that the risk of all‐cause mortality (HR = 1.40, 95% CI 1.08–1.82, p < 0.05 for low; HR = 1.41, 95% CI 1.15–1.71, p < 0.01 for high) was greater in the low and high groups than in the middle group in the univariate analysis. Even with covariates controlled for, the risk of all‐cause mortality (HR = 1.48, 95% CI 1.14–1.93, p < 0.01 for low; HR = 1.30, 95% CI 1.06–1.58, p < 0.05 for high) was elevated in both groups.
Figure 3.
Multivariable Cox regression analyses for the association between HDL‐C/LDL‐C and all‐cause mortality in the total population. Model 1: Unadjusted + Age + gender. Model 3: Model 1 + HTN + T2DM + BMI + waist circumference + geographic region + education level + smoking history.
3.3. Subgroup Analysis
We conducted subgroup analyses according to age (60 vs. ≥ 60 years), sex (male vs. female), smoking status (no vs. yes), hypertension status (no vs. yes), T2DM status (no vs. yes), and BMI (1 = underweight vs. 2 = normal vs. 3 = overweight vs. 4 = obese). (Figure 4). None of the above parameters changed the predictive power of HDL‐C/LDL‐C for all‐cause mortality. Although the interaction analysis did not reveal a significant interaction effect between diabetes and the HDL‐C/LDL‐C ratio (p = 0.627), on the basis of clinical experience and pathophysiological mechanisms, we further investigated the clinical utility of this ratio.
Figure 4.
Forest plot of all‐cause mortality according to different subgroups. Subgroup analysis included age (< 60 vs. ≥ 60 years), gender (male vs. female), smoker (no vs. yes), hypertension (no vs. yes), and T2DM (no vs. yes), BMI (1 = Thin vs. 2 = Normal vs. 3 = Overweight vs. 4 = Obese).
3.4. HDL‐C/LDL‐C and All‐Cause Mortality
The all‐cause mortality rate was 1.1% in the middle group, while the rates were 1.6% and 1.7%, respectively, in the low and high groups. As illustrated in Figure 5, maintaining the HDL‐C/LDL‐C ratio within a range of 0.3–0.5 was associated with reduced all‐cause mortality in high‐risk CVD patients. Notably, even among populations at high risk for CVD without T2DM, the middle group continued to have the lowest mortality.
Figure 5.
(A) The all‐cause mortality rate in various HDL‐C/LDL‐C groups for the total population. (B) The all‐cause mortality rate in various HDL‐C/LDL‐C groups for the population with T2DM. (C) The all‐cause mortality rate in various HDL‐C/LDL‐C groups for the population without T2DM.
3.5. Associations between HDL‐C/LDL‐C and All‐Cause Mortality in Individuals With and Without Type 2 Diabetes
When the populations at high CVD risk were stratified by the presence or absence of T2DM, Kaplan‒Meier survival analysis revealed that the middle group without T2DM presented the lowest cumulative all‐cause mortality (p < 0.05) (Figure S4). In contrast, no statistically significant differences in all‐cause mortality were observed across subgroups in populations at high CVD risk with T2DM (p > 0.05) (Figure S5).
Cox proportional hazards regression analysis revealed that the risk of all‐cause mortality (HR = 1.56, 95% CI 1.13–2.16, p < 0.05 for low; HR = 1.50, 95% CI 1.17–1.93, p < 0.01 for high) was greater in the low and high groups than in the middle group in the univariate analysis. After controlling for covariates, participants at high risk for CVD without T2DM also had similar prognoses (HR = 1.65, 95% CI 1.19–2.28, p < 0.01 for low; HR = 1.35, 95% CI 1.05–1.74, p < 0.01 for high). However, no such risk associations were identified in participants at high CVD risk with T2DM (p > 0.05). The detailed outcomes are presented in Figure 6. To assess the robustness of this ratio, we performed analyses stratified by age (60 vs. ≥ 60 years), sex (male vs. female), smoking status (no vs. yes), and hypertension status (no vs. yes) in populations with or without T2DM. None of the above parameters changed the predictive power of HDL‐C/LDL‐C for all‐cause mortality, as shown in Figures S6 and S7.
Figure 6.
Multivariable Cox regression analyses for the association between HDL‐C/LDL‐C and all‐cause mortality with or without T2DM. Model 1: Unajusted + Age + gender. Model 2: Model 1 + HTN + BMI + waist circumference + geographic region + education level + smoking history.
4. Discussion
In this study, we explored the association between HDL‐C/LDL‐C ratios and mortality risk in populations at high CVD risk. We also examined the prognostic predictive capacity of HDL‐C/LDL‐C for T2DM patients at risk for CVD. The findings revealed that both the low and high groups presented a significantly greater risk of all‐cause mortality than did the middle group. More interestingly, the populations at high risk for CVD without T2DM presented the same results.
In clinical practice, HDL‐C and LDL‐C are simple and widely available markers, and they are frequently used to assess adverse CVD prognoses [21, 22, 23]. In contrast, the HDL‐C/LDL‐C ratio indicates the balance between protective and atherogenic lipoproteins [11]. The HDL‐C/LDL‐C ratio provides a more comprehensive evaluation of lipid metabolism than does HDL‐C or LDL‐C alone [12, 13]. Shiqi Yuan's study demonstrated that an HDL‐C/LDL‐C ratio of 0.4–0.6 was related to a decreased risk of myocardial infarction (MI) and all‐cause mortality [24]. You et al. reported that a high HDL‐C/LDL‐C ratio was independently linked to an increased risk of all‐cause mortality in patients with acute intracranial hemorrhage (ICH) [10]. However, the relationship between HDL‐C/LDL‐C and adverse prognosis has not been studied in a cohort of individuals at high risk for CVD, and it is unclear whether HDL‐C/LDL‐C predicts adverse prognosis in high‐risk individuals. As a result, our study fills a significant knowledge gap in this field.
Our study indicates a U‐shaped association between HDL‐C/LDL‐C and mortality in the population at high risk of CVD, which may be attributable to the fact that both extremely high and very low HDL‐C or LDL‐C increase mortality risk in the population at high risk of CVD [9, 25, 26, 27]. According to previous studies, HDL‐C decreases atherosclerosis, and lower LDL‐C levels are associated with a lower risk of cardiovascular disease [23, 28]. However, two recently published studies reported a U‐shaped association between HDL‐C and mortality [26, 27], implying that extremely high HDL‐C levels increase CVD risk and result in an adverse prognosis. In a randomized study of 108,243 persons aged 20–100 years, Johannesen et al. reported that very low and high LDL‐C values increased the risk of all‐cause mortality [9]. This is because residual non‐HDL‐C cholesterol (such as VLDL residues) becomes the main atherogenic component at extremely high HDL‐C and extremely low LDL‐C levels. Of these scenarios, high HDL‐C produced by CETP mutations or chronic inflammation may be followed by HDL dysfunction [29, 30], which is characterized by impaired cholesterol reversal ability. When HDL dysfunction is combined with hyperglyceridaemia, VLDL residue accumulation significantly increases the overall level of non‐HDL‐C, which promotes IL‐6 and TNF‐α secretion by activating the macrophage TLR4/NF‐κB pathway and exacerbates plaque inflammation [31]. Similarly, at very low LDL‐C, the VLDL residue becomes a major component of non‐HDL‐C, degrading fibrous cap collagen through the activation of TLR4‐dependent matrix metalloproteinases such as MMP‐9 and thus increasing plaque vulnerability [31, 32]. Thus, the HDL‐C/LDL‐C ratio reduces the risk of a single lipoprotein assessment by dynamically assessing the metabolic status of both lipoproteins.
The results of our study revealed that the HDL‐C/LDL‐C ratio in this population was strongly associated with adverse prognosis and that a ratio of 0.3–0.5 was the most beneficial range for participants. This differs from thresholds found in previous studies. This is probably because the Chinese population has lower lipids overall than the non‐Chinese population does, which is related to lifestyle and dietary habits. Second, the population in this study is a group at high risk for CVD, which is a population targeted for primary prevention and having similar lipid conditions to the normal population; therefore, the threshold is lower than that in other studies. This finding is also consistent with the HDL‐C/LDL‐C reference standard range recommended by the Chinese guidelines for the prevention and management of dyslipidaemia in adults (updated 2016) [33].
Although the interaction analysis did not reveal a significant interaction effect between diabetes status and the HDL‐C/LDL‐C ratio (p = 0.627), this may be due to the small sample size for T2DM. Furthermore, stratified analysis suggested that the predictive value of this ratio may be more significant in the nondiabetic subgroup, which may be related to lipid metabolism disorders in patients with T2DM. Because there is an association between blood glucose and lipids, their metabolic processes influence one another [5, 6]. T2DM generally coexists with dyslipidaemia and is a major risk factor for CVD, along with smoking and hypertension [5]. Previous research has demonstrated the role of lipids in the development of T2DM [5, 34]. Therefore, we compared the associations between this lipid index and all‐cause mortality in patients with different diabetes states. In our study, low and high HDL‐C/LDL‐C ratios were associated with poor prognosis in the non‐T2DM population, but this association was not found in the T2DM population. This is because T2DM not only interferes with normal lipid metabolism but also promotes the development of CVD and affects patient prognosis.
On the one hand, T2DM is the predominant type of diabetes, characterized by insulin resistance. Insulin is a hormone that regulates blood glucose and is also involved in lipid metabolism [6, 35]. In normal physiology, insulin suppresses FoxO1 in the liver, reducing ApoB production. Insulin also prevents adipocytes from releasing additional free fatty acids into circulation. Hence, when T2DM progresses, FoxO1 gradually becomes uncontrolled, as does mTORC1, which suppresses sortilin [36, 37]. Therefore, prolonged hyperglycaemic conditions may cause insulin resistance, leading adipocytes to release more free fatty acids into the bloodstream, prompting the liver to synthesize more LDL‐C while decreasing HDL‐C, resulting in dyslipidaemia [37, 38]. The more significant aspect is that LDL‐C levels are likely to be substantially elevated in patients with diabetes compared with healthy controls [39]. Consequently, HDL‐C/LDL‐C is influenced by T2DM in CVD high‐risk populations and does not accurately predict all‐cause mortality in CVD high‐risk populations.
On the other hand, in the state of T2DM, compensatory hyperinsulinaemia activates the MAPK pathway, which promotes vasoconstriction, inflammation, and sodium retention. Therefore, insulin resistance leads to endothelial signaling dysfunction, contributes to the development of inflammation, disrupts the balance of endothelial vasodilation and constriction, increases the risk of CVD, and has an adverse effect on prognosis [40, 41]. Therefore, maintaining this indicator within the range of 0.3–0.5 in populations without T2DM may hold greater clinical significance but further large‐scale population studies are needed to validate its application in diabetic populations.
In conclusion, this study can provide useful reference values for the clinical control of blood lipids in high‐risk groups for CVD, that is, maintaining HDL‐C/LDL‐C ratios between 0.3 and 0.5. An indicator that provides early intervention in populations at high risk for CVD is highly important for guiding primary prevention and improving patient prognosis.
This study has several limitations. First, it included only Chinese people, and the results may differ when applied to other races. However, it is representative because it covers a large number of participants from nine cities. Second, because it was a prospective trial, the causal association between HDL‐C/LDL‐C levels and all‐cause mortality prognosis could not be adequately determined. Third, although the HDL‐C/LDL‐C ratio was derived from a large‐scale population‐based investigation, the clinical relevance of this biomarker in individuals with T2DM remains inconclusive owing to the small subgroup size. In the future, we will validate the preliminary findings of this study in larger T2DM cohorts to increase their generalizability and clinical applicability. Fourth, we used four models to assess whether the associations between different HDL‐C/LDL‐C ratios and mortality were influenced by sociodemographic and other cardiovascular risk factors. However, other factors affect the efficiency of the ratio evaluation. Our study has various advantages. First, it had a large sample size of 32,609 participants and employed consistent procedures to collect thorough questions and perform laboratory testing. Second, this is the initial investigation to explore the relationship between HDL‐C/LDL‐C and all‐cause mortality in a population at high risk of CVD, as well as the effect of diabetes on HDL‐C/LDL‐C prediction. Third, the cohort in this study was exclusively Chinese, reflecting the high cardiovascular burden in China.
In conclusion, the HDL‐C/LDL‐C ratio can be utilized to predict the prognosis of individuals at high risk for CVD, and maintaining HDL‐C/LDL‐C ratios between 0.3 and 0.5 may be the most helpful range for this population. Maintaining HDL‐C/LDL‐C ratios within the range of 0.3–0.5 may have clinical significance for cohorts without T2DM, whereas its prognostic implications in individuals with T2DM necessitate further exploration.
Author Contributions
Biting Lin: concept and design, analyzed data, created Tables and Figures, drafted paper. Yunzhi Ling: concept and design, analyzed data, drafted paper. Gengyu Zhou: analyzed data, drafted paper. Ziqing Ruan, Fan Chen, Simiao Chen, Tingting Weng, Yuanfan Zhu, and Jingyi Lin: collected data. Ling Yu: concept and design, supervised, edited paper. Kai‐Yang Lin: concept and design, supervised, edited paper.
Ethics Statement
This study is based on data from the Fujian Cardiac Metabolic Diseases and Comorbidity Cohort registered with the National Library of Medicine in November 2023 (Fu‐CARE, Trial No. NCT06102187) and followed the Helsinki Declaration principles. Written informed consent was obtained from all patients.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Supporting Figure 1.
Supporting Figure 2.
Supporting Figure 3.
Supporting Figure 4.
Supporting Figure 5.
Supporting Figure 6.
Supporting Figure 7.
Supporting Table 1.
Acknowledgments
The authors have nothing to report.
[Correction added on 29 July 2025, after first online publication: The first corresponding author's name has been corrected to Ling Yu.]
Contributor Information
Ling Yu, Email: yuling4321@126.com.
Kaiyang Lin, Email: lky7411@sina.com.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request. Data used is available from the corresponding author upon reasonable request.
References
- 1. Roth G. A., Abate D., Abate K. H., et al., “Global, Regional, and National Age‐Sex‐Specific Mortality for 282 Causes of Death in 195 Countries and Territories, 1980–2017: A Systematic Analysis for the Global Burden of Disease Study 2017,” Lancet 392, no. 10159 (2018): 1736–1788, 10.1016/S0140-6736(18)32203-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Roth G. A., Mensah G. A., and Fuster V., “The Global Burden of Cardiovascular Diseases and Risks,” Journal of the American College of Cardiology 76, no. 25 (2020): 2980–2981, 10.1016/j.jacc.2020.11.021. [DOI] [PubMed] [Google Scholar]
- 3. Mediene‐Benchekor S., Brousseau T., Richard F., Benhamamouch S., and Amouyel P., “Blood Lipid Concentrations and Risk of Myocardial Infarction,” Lancet 358, no. 9287 (2001): 1064–1065, 10.1016/S0140-6736(01)06200-6. [DOI] [PubMed] [Google Scholar]
- 4. Revilla G., Cedó L., Tondo M., et al., “LDL, HDL and Endocrine‐Related Cancer: From Pathogenic Mechanisms to Therapies,” Seminars in Cancer Biology 73 (2021): 134–157, 10.1016/j.semcancer.2020.11.012. [DOI] [PubMed] [Google Scholar]
- 5. Wu L. and Parhofer K. G., “Diabetic Dyslipidemia,” Metabolism: Clinical and Experimental 63, no. 12 (2014): 1469–1479, 10.1016/j.metabol.2014.08.010. [DOI] [PubMed] [Google Scholar]
- 6. Haas M. E., Attie A. D., and Biddinger S. B., “The Regulation of ApoB Metabolism by Insulin,” Trends in Endocrinology & Metabolism 24, no. 8 (2013): 391–397, 10.1016/j.tem.2013.04.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Nordestgaard B. G. and Varbo A., “Triglycerides and Cardiovascular Disease,” Lancet 384, no. 9943 (2014): 626–635, 10.1016/S0140-6736(14)61177-6. [DOI] [PubMed] [Google Scholar]
- 8. Kardassis D., Mosialou I., Kanaki M., Tiniakou I., and Thymiakou E., “Metabolism of HDL and Its Regulation,” Current Medicinal Chemistry 21, no. 25 (2014): 2864–2880, 10.2174/0929867321666140303153430. [DOI] [PubMed] [Google Scholar]
- 9. Johannesen C. D. L., Langsted A., Mortensen M. B., and Nordestgaard B. G., “Association Between Low Density Lipoprotein and All Cause and Cause Specific Mortality in Denmark: Prospective Cohort Study,” BMJ 371 (2020): m4266, 10.1136/bmj.m4266. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. You S., Zhong C., Xu J., et al., “LDL‐C/HDL‐C Ratio and Risk of All‐Cause Mortality in Patients With Intracerebral Hemorrhage,” Neurological Research 38, no. 10 (2016): 903–908, 10.1080/01616412.2016.1204797. [DOI] [PubMed] [Google Scholar]
- 11. Zimmer F., Riebeling V., Benke B., Schuster J., and Roskamm H., “[The LDL‐HDL Ratio in Patients With Coronary Arteriosclerosis (Author's Transl)],” Zeitschrift fur Kardiologie 69, no. 3 (1980): 149–153. [PubMed] [Google Scholar]
- 12. Zou Y., Zhong L., Hu C., Zhong M., Peng N., and Sheng G., “LDL/HDL Cholesterol Ratio Is Associated With New‐Onset NAFLD in Chinese Non‐Obese People With Normal Lipids: A 5‐year Longitudinal Cohort Study,” Lipids in Health and Disease 20, no. 1 (2021): 28, 10.1186/s12944-021-01457-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Sun T., Chen M., Shen H., et al., “Predictive Value of LDL/HDL Ratio in Coronary Atherosclerotic Heart Disease,” BMC Cardiovascular Disorders 22, no. 1 (2022): 273, 10.1186/s12872-022-02706-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Kawamoto R., Tabara Y., Kohara K., et al., “Low‐Density Lipoprotein Cholesterol to High‐Density Lipoprotein Cholesterol Ratio Is the Best Surrogate Marker for Insulin Resistance in Non‐Obese Japanese Adults,” Lipids in Health and Disease 9 (2010): 138, 10.1186/1476-511X-9-138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Hong M., Ling Y., Lu Z., et al., “Contribution and Interaction of the Low‐Density Lipoprotein Cholesterol to High‐Density Lipoprotein Cholesterol Ratio and Triglyceride to Diabetes in Hypertensive Patients: A Cross‐Sectional Study,” Journal of Diabetes Investigation 10, no. 1 (2019): 131–138, 10.1111/jdi.12856. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Kuang M., Peng N., Qiu J., Zhong Y., Zou Y., and Sheng G., “Association of LDL:HDL Ratio With Prediabetes Risk: A Longitudinal Observational Study Based on Chinese Adults,” Lipids in Health and Disease 21, no. 1 (2022): 44, 10.1186/s12944-022-01655-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Biddinger S. B., Hernandez‐Ono A., Rask‐Madsen C., et al., “Hepatic Insulin Resistance Is Sufficient to Produce Dyslipidemia and Susceptibility to Atherosclerosis,” Cell Metabolism 7, no. 2 (2008): 125–134, 10.1016/j.cmet.2007.11.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Lu Y., Zhang H., Lu J., et al., “Prevalence of Dyslipidemia and Availability of Lipid‐Lowering Medications Among Primary Health Care Settings in China,” JAMA Network Open 4, no. 9 (2021): e2127573, 10.1001/jamanetworkopen.2021.27573. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. R.J C for G ., “2018 Chinese Guidelines for Prevention and Treatment of Hypertension – A Report of the Revision Committee of Chinese Guidelines for Prevention and Treatment of Hypertension,” Journal of Geriatric Cardiology: JGC 16, no. 3 (2019): 182–241, 10.11909/j.issn.1671-5411.2019.03.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. 2. Classification and Diagnosis of Diabetes: Standards of Medical Care in Diabetes –2022,” supplement, Diabetes Care 45, no. Suppl 1 (2022): S17–S38, 10.2337/dc22-S002. [DOI] [PubMed] [Google Scholar]
- 21. Aggarwal R., Chiu N., Libby P., Boden W. E., and Bhatt D. L., “Low‐Density Lipoprotein Cholesterol Levels in Adults With Coronary Artery Disease in the US, January 2015 to March 2020,” Journal of the American Medical Association 330, no. 1 (2023): 80–82, 10.1001/jama.2023.8646. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Ference B. A., Ginsberg H. N., Graham I., et al., “Low‐Density Lipoproteins Cause Atherosclerotic Cardiovascular Disease. 1. Evidence From Genetic, Epidemiologic, and Clinical Studies. A Consensus Statement From the European Atherosclerosis Society Consensus Panel,” European Heart Journal 38, no. 32 (2017): 2459–2472, 10.1093/eurheartj/ehx144. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Navab M., Reddy S. T., Van Lenten B. J., and Fogelman A. M., “HDL and Cardiovascular Disease: Atherogenic and Atheroprotective Mechanisms,” Nature Reviews Cardiology 8, no. 4 (2011): 222–232, 10.1038/nrcardio.2010.222. [DOI] [PubMed] [Google Scholar]
- 24. Yuan S., Huang X., Ma W., et al., “Associations of HDL‐C/LDL‐C With Myocardial Infarction, All‐Cause Mortality, Haemorrhagic Stroke and Ischaemic Stroke: A Longitudinal Study Based on 384 093 Participants From the UK Biobank,” Stroke and Vascular Neurology 8, no. 2 (2023): 119–126, 10.1136/svn-2022-001668. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Chen J. X., Li Y., Zhang Y. B., et al., “Nonlinear Relationship Between High‐Density Lipoprotein Cholesterol and Cardiovascular Disease: An Observational and Mendelian Randomization Analysis,” Metabolism: Clinical and Experimental 154 (2024): 155817, 10.1016/j.metabol.2024.155817. [DOI] [PubMed] [Google Scholar]
- 26. Fernández‐Ruiz I., “Very High HDL‐C Levels Are Associated With Higher Mortality in Patients With CAD,” Nature Reviews Cardiology 19, no. 8 (2022): 504, 10.1038/s41569-022-00743-4. [DOI] [PubMed] [Google Scholar]
- 27. Kim H. J., Jeong S., Oh Y. H., Park S. J., Cho Y., and Park S. M., “Changes in High‐Density Lipoprotein Cholesterol With Risk of Cardiovascular Disease Among Initially High‐Density Lipoprotein‐High Participants,” Cardiovascular Diabetology 22, no. 1 (2023): 71, 10.1186/s12933-023-01805-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Navarese E. P., Kolodziejczak M., and De Servi S., “Interventions to Lower Low‐Density Lipoprotein Cholesterol and Cardiovascular Risk,” Journal of the American Medical Association 317, no. 4 (2017): 439, 10.1001/jama.2016.19058. [DOI] [PubMed] [Google Scholar]
- 29. Kovacs R. J., “ACC.19, Primary Prevention, and Elizabeth Kovacs (1918 to 2018),” Journal of the American College of Cardiology 73, no. 21 (2019): 2773–2774, 10.1016/j.jacc.2019.04.013. [DOI] [PubMed] [Google Scholar]
- 30. Ansell B. J., Navab M., Hama S., et al., “Inflammatory/Antiinflammatory Properties of High‐Density Lipoprotein Distinguish Patients From Control Subjects Better Than High‐Density Lipoprotein Cholesterol Levels and Are Favorably Affected by Simvastatin Treatment,” Circulation 108, no. 22 (2003): 2751–2756, 10.1161/01.CIR.0000103624.14436.4B. [DOI] [PubMed] [Google Scholar]
- 31. Borén J., Chapman M. J., Krauss R. M., et al., “Low‐Density Lipoproteins Cause Atherosclerotic Cardiovascular Disease: Pathophysiological, Genetic, and Therapeutic Insights: A Consensus Statement From the European Atherosclerosis Society Consensus Panel,” European Heart Journal 41, no. 24 (2020): 2313–2330, 10.1093/eurheartj/ehz962. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Sabatine M. S., Giugliano R. P., Keech A. C., et al., “Evolocumab and Clinical Outcomes in Patients With Cardiovascular Disease,” New England Journal of Medicine 376, no. 18 (2017): 1713–1722, 10.1056/NEJMoa1615664. [DOI] [PubMed] [Google Scholar]
- 33. Chinese Adult Dyslipidemia Prevention and Treatment Guideline Revision Joint Committee , “Chinese Guidelines on the Prevention and Treatment of Dyslipidemia in Adults (2016 Updated Version),” [In Chinese.] Chinese Circulation Journal 31, no. 10 (2016): 937–950, 10.3969/j.issn.1000-3614.2016.10.001. [DOI] [Google Scholar]
- 34. Wang J., Stančáková A., Soininen P., et al., “Lipoprotein Subclass Profiles in Individuals With Varying Degrees of Glucose Tolerance: A Population‐Based Study of 9399 Finnish Men,” Journal of Internal Medicine 272, no. 6 (2012): 562–572, 10.1111/j.1365-2796.2012.02562.x. [DOI] [PubMed] [Google Scholar]
- 35. Bornfeldt K. E. and Tabas I., “Insulin Resistance, Hyperglycemia, and Atherosclerosis,” Cell Metabolism 14, no. 5 (2011): 575–585, 10.1016/j.cmet.2011.07.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Park S., Cha H. N., Shin M. G., et al., “Inhibitory Regulation of FOXO1 in PPARδ Expression Drives Mitochondrial Dysfunction and Insulin Resistance,” Diabetes 73, no. 7 (2024): 1084–1098, 10.2337/db23-0432. [DOI] [PubMed] [Google Scholar]
- 37. Kong Y., Niu A., Yuan W., et al., “Interaction of FOXO1 and Sumoylated PPARγ1 Induced by Hyperlipidemia and Hyperglycemia Favors Vascular Endothelial Insulin Resistance and Dysfunction,” Vascular pharmacology 147 (2022): 107125, 10.1016/j.vph.2022.107125. [DOI] [PubMed] [Google Scholar]
- 38. Rajan M. R., Nyman E., Brännmark C., Olofsson C. S., and Strålfors P., “Inhibition of FOXO1 Transcription Factor in Primary Human Adipocytes Mimics the Insulin‐Resistant State of Type 2 Diabetes,” Biochemical Journal 475, no. 10 (2018): 1807–1820, 10.1042/BCJ20180144. [DOI] [PubMed] [Google Scholar]
- 39. Hayashi T., Koba S., Ito Y., and Hirano T., “Method for Estimating High sdLDL‐C by Measuring Triglyceride and Apolipoprotein B Levels,” Lipids in Health and Disease 16, no. 1 (2017): 21, 10.1186/s12944-017-0417-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Ormazabal V., Nair S., Elfeky O., Aguayo C., Salomon C., and Zuñiga F. A., “Association Between Insulin Resistance and the Development of Cardiovascular Disease,” Cardiovascular Diabetology 17, no. 1 (2018): 122, 10.1186/s12933-018-0762-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Zhou M. S., Schulman I. H., and Raij L., “Vascular Inflammation, Insulin Resistance, and Endothelial Dysfunction in Salt‐Sensitive Hypertension: Role of Nuclear Factor Kappa B Activation,” Journal of Hypertension 28, no. 3 (2010): 527–535, 10.1097/HJH.0b013e3283340da8. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting Figure 1.
Supporting Figure 2.
Supporting Figure 3.
Supporting Figure 4.
Supporting Figure 5.
Supporting Figure 6.
Supporting Figure 7.
Supporting Table 1.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request. Data used is available from the corresponding author upon reasonable request.