Association between non-high-density lipoprotein cholesterol and type 2 diabetes: a systematic review and meta-analysis of cohort studies

Abstract

Non-high-density lipoprotein cholesterol (non-HDL), a more readily available and reliable lipid parameter, is unclear in its association with type 2 diabetes (T2D). Previous studies assessing the relationship between non-HDL and T2D risk remains inconsistent results. We performed a meta-analysis to systematically evaluate this association. The PubMed, EMBASE, Medline, Web of Science, and Cochrane Library databases were systematically searched to find articles on “non-HDL” and “T2D” from inception to December 6, 2023. A random-effects model was used to calculate the effect estimates and 95% confidence intervals. Subgroup analyses and univariate Meta-regression were performed to explore sources of heterogeneity. The main exposure and outcome were non-HDL and T2D, respectively, in the general population. A total of 8 studies included 251,672 participants who met the inclusion criteria for this study. Meta-analysis showed that higher non-HDL increased the risk of T2D compared with the lower non-HDL group (total effect size: 1.16; 95% CI 1.079–1.251, p < 0.001). Subgroup analyses and Meta-regression of the association between non-HDL and T2D were not affected by region, proportion of men, sample size, or adjustment for confounders (including BMI, hypertension, waist circumference, and family history of diabetes). Higher non-HDL may be associated with an increased risk of T2D. Large prospective cohort studies are needed to validate these findings, and further studies are required in order to elucidate the underlying pathophysiologic mechanisms underlying the association between non-HDL and T2D.

Introduction

Type 2 diabetes (T2D) is a prevalent and serious chronic condition influenced by intricate genetic-environmental interplays, alongside factors such as unhealthy dietary patterns and sedentary lifestyles [1, 2]. T2D comprises approximately 90 percent of the 537 million global diabetes cases, with its incidence continuing to rise. Projections suggest that by 2045, T2D may afflict around 783 million individuals worldwide [3]. Presently, roughly 80% of T2D occurrences arise in low- and middle-income nations [4]. In China, diabetes prevalence escalated from 9.7 percent in 2008 to 11.2 percent in 2017, positioning China as the nation with the highest T2D patient population globally [5]. Disruptions in glucolipid metabolism significantly jeopardize cardiovascular health, predisposing individuals to organ malfunction and premature mortality [6]. Dyslipidemia is a prevalent feature of insulin resistance and T2D, constituting a critical risk factor for abnormal glucose regulation [7]. Studies indicate that approximately 75% of diabetic patients exhibit abnormal lipid metabolism. Dysfunctions in lipid metabolism are pivotal in diabetes pathogenesis [8]. Fatty acids can induce insulin resistance in peripheral tissues and disrupt pancreatic islet β-cell function, thereby promoting diabetes onset [8, 9].

Non-high-density lipoprotein cholesterol (non-HDL) encompasses all cholesterol content within potentially atherogenic lipoprotein particles, including low-density lipoprotein cholesterol (LDL-C), lipoprotein (a), very-low-density lipoprotein cholesterol (VLDL-C), VLDL remnant, and intermediate-density lipoprotein (IDL-C) [10]. Non-HDL is recognized as a more practical and reliable predictor of coronary artery disease (CAD), a fact corroborated by numerous epidemiological investigations and clinical trials [11]. A study conducted by Vega et al. showed a 20.9% decrease in non-HDL-C levels in free-living subjects with diabetes mellitus but without arteriosclerotic cardiovascular disease (ASCVD) from the NHANES survey conducted between 1999 and 2016 [12]. Another study showed that non-HDL-C levels were positively associated with pre-diabetes and diabetes status in patients [13].

A mounting body of research suggests that elevated levels of non-HDL are correlated with an increased risk of type 2 diabetes (T2D) within the general population. However, evidence from observational studies presents inconsistent findings regarding the association between non-HDL and T2D risk [14-21]. Given the contentious nature of non-HDL’s role in T2D risk, its relationship has yet to be conclusively established through meta-analyses. Therefore, our present systematic review and meta-analysis aim to synthesize and pool available data to investigate the link between non-HDL and T2D, potentially furnishing a more scientifically grounded basis for enhanced T2D prevention and management strategies.

Methods

Protocol and registration

The current study adhered to the guidelines outlined in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) (Supplementary Table 1). Furthermore, this study was registered with the International Prospective Register of Systematic Reviews (PROSPERO) under the registration number CRD42024499526.

Search strategy

We systematically searched multiple electronic databases, including PubMed, EMBASE, Medline, Web of Science, and the Cochrane Library, from inception until December 2023 to identify relevant studies (Supplementary Table 2). The search terms, connected by Boolean logical operators “OR” or “AND,” encompassed: (1) non-HDL-C, non-HDL cholesterol, non-HDL-cholesterol, non-HDL-cholesterol, non-HDL-cholesterol, non-HDL-cholesterol, non-HDL-cholesterol, and non-HDL-cholesterol, cholesterol, non-high-density lipoprotein cholesterol, non-HDL; (2) type 2 Diabetes Mellitus, type 2 Diabetes, T2D, Diabetes Mellitus Type II, type II Diabetes Mellitus, Diabetes Type II. Two investigators independently conducted a systematic search of all relevant references. There were no additional restrictions regarding time, country, or language. Furthermore, we analyzed the reference lists of relevant original articles and review articles.

Study selection

Two evaluators will independently screen titles and abstracts to identify potentially eligible studies based on the inclusion criteria. Inclusion criteria included the following: (1) cohort studies published in full text; (2) examination of the association between non-HDL levels and the risk of T2D; (3) reporting of the association between non-HDL and the risk of T2D in terms of odds ratios (ORs) or hazard ratios (HRs) with their 95% confidence intervals (CIs). Exclusion criteria consist of: (1) conference proceedings; (2) experimental and interventional studies, reviews, and case reports; (3) studies with incomplete information; (4) studies not published in English.

Data extraction and quality assessment

Two investigators independently extracted information from the literature based on predefined inclusion criteria. The extracted data included: (1) the name of the first author, year of publication, and country where the study was conducted; (2) characteristics of the study design; (3) participant characteristics, such as study population, sample size, age, and gender; (4) duration of study follow-up; (5) definition and diagnosis of outcome; (6) confounders adjusted for in multivariate analyses; (7) adjusted risk estimates (OR or HR values) and their 95% confidence intervals. Two researchers critically reviewed all relevant studies, extracted potentially eligible data, and thoroughly resolved any uncertainties or discrepancies. The quality of cohort studies was assessed using the Newcastle-Ottawa Scale (NOS), which ranges from 0 to 9. Studies with a NOS score of ≥6 were deemed to have a low risk of bias [22].

Statistical analysis

Systematic evaluation and meta-analysis were performed using Stata 14.0 software. The multivariate-adjusted risk estimates (OR or HR values) and their 95% confidence intervals (CIs) were collected from all individual studies. Statistical heterogeneity was analyzed using Cochran’s Q statistic and I² statistic [23]. An indicator of significant heterogeneity was considered if I² was greater than 75%. Then the analysis was conducted using the random effects model. Subgroup analysis was used to explore the heterogeneity of included studies and meta-regression analysis was conducted to test for the source of heterogeneity among subgroups [24]. Sensitivity analysis was used to further test the impact of individual studies on the overall results and to analyze whether the results were robust and reliable [25]. The presence of potential publication bias was assessed by funnel plot, Begg’s test, and Egger’s test [26]. The trim and fill method was utilized to correct for biased results and to assess the impact of bias on the combined risk estimate [27].

Results

Study selection

A total of 5,218 articles were identified through searches conducted in PubMed, EMBASE, Medline, Web of Science, and the Cochrane Library databases. Among these articles, 358 studies were excluded due to duplication or irrelevant topics. Subsequently, after reviewing titles and abstracts, 4,795 articles were excluded. Upon full-text examination, 57 records were further excluded. Ultimately, only 8 studies met our selection criteria and were included in the quantitative assessment of the meta-analysis (Fig. 1).

Fig. 1  Flowchart of database search and study identification

Study characteristics and quality assessment

We included eight cohort studies conducted in China, Korea, Canada, and Iran, published between 2011 and 2022. The total participant count across these studies was 251,672, comprising 34,382 patients with T2D and 217,290 patients without T2D. Four studies were conducted in Korea, two in Iran, one in China, and one in Canada. The study with the largest sample size, conducted by You-Cheol Hwang et al. in Korea, included 118,429 participants, while the study with the smallest sample size, by Ley et al., included 492 participants. Primary data sources included the Korean Genome and Epidemiology Study (KoGES), the Tehran Lipid and Glucose Study (TLGS), the Isfahan Diabetes Prevention Study, and hospital or community-based population studies. Baseline characteristics, such as study source, sample size, and other relevant details, were summarized in Table 1. Newcastle-Ottawa Scale (NOS) assessment revealed that two studies achieved a NOS score of 9, three studies scored 8, and three studies scored 7 (Supplementary Table 3).

Table 1 Characteristics of the included observational studies

Study	Country	Design	Study population	Male%	Mean age	Study time	Non-HDL analysis	Definition and diagnosis of outcome	Variables adjusted
S. H. Ley [14]	Canada	cohort	492 participants without diabetes	42.1%	NA	10-year	continuous	Incidence of T2D was defined as: fasting plasma glucose ≥7.0 mmol/L or 2h postload glucose ≥11.1 mmol/L on a 2h-OGTT; current use of insulin or oral hypoglycaemic agents; a positive response to the question: Have you ever been diagnosed with diabetes by a nurse or doctor?	age, sex, hypertension, ever smoking and log C-reactive protein, waist circumference
Mi Hae Seo [15]	Korea	retrospective, longitudinal	5,577 participants without diabetes	70.3%	44.5	4-year	continuous	Incidence of T2D was defined as: fasting plasma glucose 126 mg/dL or greater; or a HbA1c 6.5% or greater; subjects who had a history of diabetes or currently used insulin or oral antidiabetic drugs based on a self-report questionnaire at each visit	age and gender, BMI, smoking status, hypertension, fasting glucose, fasting insulin
You-Cheol Hwang [16]	Korea	retrospective longitudinal	84,394 participants without diabetes	57.9%	38.4	3.3-year	continuous	Incidence of T2D was defined as: fasting glucose levels 126 mg/dL (7.0 mmol/L); taking oral hypoglycemic agents or insulin therapy; a self-reported history of diabetes; HbA1c levels ≥6.5%	Age, sex, waist circumference, fasting serum insulin, TG, HDL-C, family history of diabetes, and systolic blood pressure
Erfan Sadeghi [17]	Iran	follow-up	1,222 participants without diabetes	26.8%	42.9	14-year	continuous	Incidence of T2D was defined as: FPG ≥126 mg/dL; 2h plasma glucose ≥200 mg/dL; HbA1c ≥6.5%	NA
Pegah Khaloo [18]	Iran	prospective	5,474 participants without diabetes	42.4%	41.3	8.9-year	continuous	Incidence of T2D was considered to be present: using antidiabetic drugs; FPG was ≥7 mmol/L or if the 2h-PLG was ≥11.1 mmol/L	age sex, education, lipid drug use, family history of T2DM, history of cardiovascular disease, hypertension and baseline measurements of FPG, BMI, waist circumferences and lipid profile, BMI change, FPG change
In-Ho Seo [19]	Korea	prospective cohort	7,608 participants without diabetes	46.9%	51.7	14-year	Q4:Q1	Incidence of T2D was defined as: FPG ≥126 mg/dL; a plasma glucose level ≥200 mg/dL at two hours after a 75-g OGTT; an HbA1c ≥6.5%; or current treatment with oral anti-diabetic medications or insulin therapy	age, sex, waist circumference, smoking status, alcohol intake, physical activity, mean arterial blood pressure, family history of diabetes and HOMA-IR
Tong Yang [20]	China	retrospective cohort	28,476 CHD participants without diabetes	44.07%	64	6.75-year	continuous	Incidence of T2D was defined as: FBG ≥7.0 mmol/L; HbA1c ≥6.5%	age, sex, SBP, smoking, hypertension, family history of T2DM, current antilipidemic medication, current antihypertensive medication
Y.-C. Hwang [21]	Korea	longitudinal	118,429 participants without diabetes	59.9%	39.6	3.1-year	continuous	Incidence of T2D was defined as: fasting glucose levels ≥7.0 mmol/L; taking oral hypoglycaemic agents or insulin therapy; a self-reported history of diabetes; HbA1c levels ≥48 mmol/mol (6.5%)	age and sex waist circumference, FPG, fasting serum insulin, HbA1c, family history of diabetes and SBP, TG and HDL-C

BMI, body mass index; OGTT, oral glucose tolerance test; HbA1c, Hemoglobin A1C; FPG, fasting plasma glucose; FBG, fasting blood glucose; Q4:Q1, the 4th quintile vs. The 1st quintile; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; SBP, systolic blood pressure; PLG, postload glucose; HOMA-IR, homeostatic model assessment for insulin resistance; NA, not applicable; CHD, coronary heart disease.

The association between non-HDL and T2D in the general population

This meta-analysis, comprising eight studies [14-21], indicates that elevated non-HDL levels are associated with an increased risk of developing T2D compared to the general population with a lower non-HDL levels (pooled effect size 1.16, 95% CI 1.079–1.251) (Fig. 2). However, significant heterogeneity was observed in the analysis (I² = 97.3%, p < 0.001).

Fig. 2  Forest plots for the meta-analysis of the association between non-HDL and T2D

The sensitivity analysis revealed no extreme results affecting the pooled estimates (Supplementary Fig. 1 and Supplementary Table 4). Furthermore, subgroup analyses were conducted by region, sample size, and proportion of males (Table 2). Within the subgroup analyses by region, both Asia (effect size: 1.15, 95% CI: 1.06–1.24) and North America (effect size: 1.43, 95% CI: 1.078–1.897) yielded consistent results. Moreover, subgroup analyses based on sample sizes indicated that non-HDL were associated with T2D (sample size >8,000 [effect size: 1.146, 95% CI: 1.128–1.165], sample size ≤8,000 [effect size: 1.203, 95% CI: 1.036–1.397]). Subgroup analyses concerning the proportion of male participants exhibited consistent associations for male proportions >50% [effect size: 1.142, 95% CI: 1.113–1.172] and male proportions ≤50% [effect size: 1.166, 95% CI: 1.047–1.299]. Additionally, given that BMI, waist circumference, hypertension, and family history of diabetes are also recognized risk factors for T2D, stratified subgroup analyses were conducted based on whether these variables were adjusted in the included studies. The results remained consistent with the pooled effect size.

Table 2 Results of subgroup analyses

Analysis	Number of studies	Effect Size (95%CI)	p (Z-test)	I2(%)	p (Q-test)
Region
Asia	7	1.150 (1.066–1.240)	<0.001	97.6%	<0.001
North America	1	1.430 (1.078–1.897)	0.013	—	—
Sample size
>8,000	3	1.146 (1.128–1.165)	<0.001	91%	0.375
≤8,000	5	1.203 (1.036–1.397)	0.015	0%	<0.001
Male%
>50%	3	1.142 (1.113–1.172)	0.005	21.2%	0.281
≤50%	5	1.166 (1.047–1.299)	<0.001	97.2%	<0.001
Adjustment for hypertension
Yes	4	1.160 (1.067–1.261)	<0.001	63.8%	0.040
No	4	1.155 (1.048–1.273)	0.004	97.9%	<0.001
Adjustment for waist circumference
Yes	5	1.171 (1.096–1.251)	0.065	80.3%	<0.001
No	3	1.121 (0.993–1.267)	<0.001	98.3%	<0.001
Adjustment for family history of diabetes
Yes	5	1.153 (1.105–1.203)	<0.001	79%	<0.001
No	3	1.177 (0.961–1.442)	0.115	88%	0.001
Adjustment for BMI
Yes	2	1.133 (0.949–1.352)	0.168	79.8%	0.026
No	6	1.172 (1.077–1.276)	<0.001	98%	<0.001

The results of univariate meta-regression analyses revealed that the incorporation of studies based on region, study design, sample size, the proportion of males, and various adjustment factors such as BMI, hypertension, family history of diabetes, and waist circumference did not significantly influence the association between non-HDL cholesterol and T2D (all p-values >0.10) (Table 3). These findings imply that disparities in these characteristics may not serve as major contributors to heterogeneity. The Begg’s test and Egger’s test indicated potential publication bias (p = 0.536 and p = 0.020, respectively). The funnel plot illustrated that some studies tended to skew towards the right side of the funnel (Supplementary Fig. 2). Subsequently, the trim and fill method was conducted, revealing no significant change in the effect size (adjusted pooled estimate: 1.130, 95% CI: [1.052, 1.213], p = 0.001), suggesting that publication bias had little effect on the results.

Table 3 Impact of study characteristics on the association between non-HDL and T2D: univariate meta-regression analysis

Variables	Coefficient (95%CI)	SE	p
Region	0.208 (–0.278,0.707)	0.203	0.328
Male%	0.005 (–0.252,0.263)	0.105	0.958
Sample size	0.041 (–0.211,0.294)	0.103	0.703
Adjustment for BMI	0.045 (–0.245,0.336)	0.118	0.714
Adjustment for waist circumference	–0.070 (–0.316,0.176)	0.101	0.511
Adjustment for hypertension	–0.011 (–0.265,0.243)	0.103	0.919
Adjustment for family history of diabetes	–0.012 (–0.281,0.256)	0.109	0.914

Discussion

In this meta-analysis of cohort studies, we found that compared to the general population with a lower group of non-HDL, higher non-HDL was more likely to increase the risk of T2D. Furthermore, the association between non-HDL and T2D remained consistent in subgroup analyses according to the study design, region, proportion of males, sample size, and adjustment for several confounding factors such as BMI, hypertension, waist circumference, and family history of diabetes.

Multiple studies suggest that non-HDL cholesterol serves as a superior indicator of cardiovascular disease progression in comparison to conventional lipid parameters. While conventional lipid parameters, including LDL cholesterol, are well-established markers associated with incident T2D, limited evidence exists regarding the association between non-HDL cholesterol and incident T2D. Higher non-HDL cholesterol signifies the presence of remnants of atherogenic lipoproteins, which promote the development of small, dense low-density lipoprotein cholesterol particles in individuals with insulin resistance, metabolic syndrome, obesity, and diabetes [28]. Recent guidelines propose that non-HDL cholesterol stands as a more robust independent risk factor for cardiovascular disease and should be regarded as a risk marker and secondary therapeutic target for cardiovascular disease management [29]. While the precise mechanism by which non-HDL impacts T2D remains unclear, several potential mechanisms have been proposed. Aberrations in lipid levels may precipitate insulin resistance and atherosclerosis, which are pivotal contributors to diabetes [30-32]. Within this cascade, the vasoactive hormonal pathway, particularly the renin-angiotensin-aldosterone system (RAAS), assumes significance as a crucial system for regulating plasma sodium concentration, arterial blood pressure, and extracellular volume [33]. An imbalance between renin and angiotensin II may instigate various chronic and acute diseases [34], with early plaque formation induced by angiotensin II standing out as a key function of RAAS in atherosclerosis [35]. In pathological contexts, the RAAS also contributes, either directly or indirectly, to the progression of atherosclerosis and its diverse complications through its interactions with other systems [36]. Moreover, an imbalance between renin and angiotensin II within the detrimental axis of RAAS amplifies the release of inflammatory cytokines and fosters oxidative stress [37]. These pathological alterations further stimulate atherosclerosis formation, exacerbate insulin resistance, and diminish insulin secretion, thereby fostering the onset and progression of diabetes mellitus.

Given the significant heterogeneity of this study, we used subgroup analysis and meta-regression to probe into sources of heterogeneity. Due to the restricted number of included studies, subgroup analyses were performed based on a few methodological and clinical characteristics. The analysis of eight cohort studies indicated that elevated non-HDL levels were linked to an increased risk of T2D. However, caution is warranted in interpreting this finding due to the limited number of studies encompassing different study types. Subgroup analyses based on various geographical regions revealed a notable association between non-HDL and T2D in Asia and North America. Notably, our study comprised only one investigation conducted in North America, while the remaining seven studies were conducted in Asia, thereby constraining the regional diversity of our findings. While most articles have adjusted for certain risk factors for T2D, including gender, age, hypertension, smoking, alcohol consumption, body mass index, and family history of cerebrovascular disease, other potential confounding variables such as sedentary lifestyle, dietary habits, history of gestational diabetes, presence of diabetes-related end-organ damage, presence of cardiovascular risk factors, and additional T2D risk factors have not been consistently accounted for. Consequently, prospective studies with appropriate controls that comprehensively adjust for potential confounders are urgently required to further explore the association between non-HDL and T2D.

Our study highlights the significance of utilizing non-HDL as a potential indicator of T2D risk in the general population. Given its ready availability, cost-effectiveness, and routine incorporation into clinical practice, non-HDL emerges as a convenient tool for assessing T2D risk. Nonetheless, our study is not devoid of limitations. The predominantly observational nature of the included studies entails a lower level of evidence compared to randomized controlled trials. Furthermore, aside from non-HDL, residual confounders such as diet and physical activity may also exert an influence on T2D risk. The exact nature of the relationship between non-HDL and T2D risk remains uncertain due to limitations in the number of studies. Moreover, the wide variation in sample sizes across the included studies may introduce unforeseen effects on the findings. Therefore, further large-scale cohort studies and foundational investigations are imperative to procure more definitive evidence.

In summary, the findings of our meta-analysis suggest a plausible association between elevated non-HDL levels and heightened risk of incidence of T2D. More prospective cohort studies and fundamental investigations are indispensable. Also, further studies are needed to elucidate the underlying mechanisms between non-HDL and T2D.

Author contributions

MQH performed the literature search and drafted the manuscript. YS and CH screened the literature. XG and XCJ extracted data and assessed the quality. HYG analyzed and interpreted the data. YLJ and HY reviewed the manuscript. All authors listed have made a substantial, direct, and intellectual contribution to the work, and approved it for publication.

Funding

This study was supported by grants from Horizontal Project of Wannan Medical College (2309-510-JXKY).

Ethics approval and consent to participate

The study does not involve ethical approval and consent.

Conflict of interest

The authors declare that they have no conflict of interest.

Data availability

Not applicable.

Acknowledgments

We’d like to thank all the authors who participated in the article.

References

• 1   Pearson  ER (2019) Type 2 diabetes: a multifaceted disease. Diabetologia 62: 1107–1112.
• 2   Wu  Y,  Ding  Y,  Tanaka  Y,  Zhang  W (2014) Risk factors contributing to type 2 diabetes and recent advances in the treatment and prevention. Int J Med Sci 11: 1185–1200.
• 3   Ogurtsova  K,  Guariguata  L,  Barengo  NC,  Ruiz  PL,  Sacre  JW, et al. (2022) IDF diabetes atlas: global estimates of undiagnosed diabetes in adults for 2021. Diabetes Res Clin Pract 183: 109118.
• 4   Ahmad  E,  Lim  S,  Lamptey  R,  Webb  DR,  Davies  MJ (2022) Type 2 diabetes. Lancet 400: 1803–1820.
• 5   Ji  L (2023) Current challenges of diabetes and metabolic disorders in China. Diabetes Obes Metab 25: 3–4.
• 6   Nelson  AJ,  Rochelau  SK,  Nicholls  SJ (2018) Managing dyslipidemia in type 2 diabetes. Endocrinol Metab Clin North Am 47: 153–173.
• 7   Al Amri  T,  Bahijri  S,  Al-Raddadi  R,  Ajabnoor  G,  Al Ahmadi  J, et al. (2019) The association between prediabetes and dyslipidemia among attendants of primary care health centers in Jeddah, Saudi Arabia. Diabetes Metab Syndr Obes 12: 2735–2743.
• 8   Unger  RH,  Zhou  YT (2001) Lipotoxicity of beta-cells in obesity and in other causes of fatty acid spillover. Diabetes 50: S118–S121.
• 9   Roden  M,  Price  TB,  Perseghin  G,  Petersen  KF,  Rothman  DL, et al. (1996) Mechanism of free fatty acid-induced insulin resistance in humans. J Clin Invest 97: 2859–2865.
• 10   Lokpo  SY,  Laryea  R,  Osei-Yeboah  J,  Owiredu  WKBA,  Ephraim  RKD, et al. (2022) The pattern of dyslipidaemia and factors associated with elevated levels of non-HDL-cholesterol among patients with type 2 diabetes mellitus in the Ho municipality: a cross sectional study. Heliyon 8: e10279.
• 11   Aggarwal  DJ,  Kathariya  MG,  Verma  DPK (2021) LDL-C, NON-HDL-C and APO-B for cardiovascular risk assessment: looking for the ideal marker. Indian Heart J 73: 544–548.
• 12   Vega  GL,  Wang  J,  Grundy  SM (2020) Temporal decline in non-high-density lipoprotein cholesterol in subjects with diabetes mellitus without atherosclerotic cardiovascular disease. J Clin Lipidol 14: 425–430.
• 13   Banu  S,  Jabir  NR,  Manjunath  NC,  Firoz  CK,  Kamal  MA, et al. (2014) Comparative study of non-high density lipoproteins cholesterol level and lipid profile in pre-diabetic and diabetic patients. CNS Neurol Disord Drug Targets 13: 402–407.
• 14   Ley  SH,  Harris  SB,  Connelly  PW,  Mamakeesick  M,  Gittelsohn  J, et al. (2012) Utility of non-high-density lipoprotein cholesterol in assessing incident type 2 diabetes risk. Diabetes Obes Metab 14: 821–825.
• 15   Seo  MH,  Bae  JC,  Park  SE,  Rhee  EJ,  Park  CY, et al. (2011) Association of lipid and lipoprotein profiles with future development of type 2 diabetes in nondiabetic Korean subjects: a 4-year retrospective, longitudinal study. J Clin Endocrinol Metab 96: E2050–E2054.
• 16   Hwang  YC,  Ahn  HY,  Park  SW,  Park  CY (2014) Apolipoprotein B and non-HDL cholesterol are more powerful predictors for incident type 2 diabetes than fasting glucose or glycated hemoglobin in subjects with normal glucose tolerance: a 3.3-year retrospective longitudinal study. Acta Diabetol 51: 941–946.
• 17   Sadeghi  E,  Hosseini  SM,  Vossoughi  M,  Aminorroaya  A,  Amini  M (2020) Association of lipid profile with type 2 diabetes in first-degree relatives: a 14-year follow-up study in Iran. Diabetes Metab Syndr Obes 13: 2743–2750.
• 18   Khaloo  P,  Hasheminia  M,  Tohidi  M,  Abdi  H,  Mansournia  MA, et al. (2018) Impact of 3-year changes in lipid parameters and their ratios on incident type 2 diabetes: Tehran lipid and glucose study. Nutr Metab (Lond) 15: 50.
• 19   Seo  IH,  Son  DH,  Lee  HS,  Lee  YJ (2022) Non-HDL cholesterol as a predictor for incident type 2 diabetes in community-dwelling adults: longitudinal findings over 12 years. Transl Res 243: 52–59.
• 20   Yang  T,  Liu  Y,  Li  L,  Zheng  Y,  Wang  Y, et al. (2022) Correlation between the triglyceride-to-high-density lipoprotein cholesterol ratio and other unconventional lipid parameters with the risk of prediabetes and type 2 diabetes in patients with coronary heart disease: a RCSCD-TCM study in China. Cardiovasc Diabetol 21: 93.
• 21   Hwang  YC,  Ahn  HY,  Yu  SH,  Park  SW,  Park  CY (2014) Atherogenic dyslipidaemic profiles associated with the development of type 2 diabetes: a 3.1-year longitudinal study. Diabet Med 31: 24–30.
• 22   Wells  GA,  Shea  B,  O’Connell  D,  Peterson  J,  Welch  V, et al. The Newcastle-Ottawa scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. https://www.ohri.ca/programs/clinical_epidemiology/oxford.asp accessed on December, 15, 2023.
• 23   Higgins  JP,  Thompson  SG (2002) Quantifying heterogeneity in a meta-analysis. Stat Med 21: 1539–1558.
• 24   Michaelis  R,  Tang  V,  Wagner  JL,  Modi  AC,  LaFrance WC  Jr, et al. (2018) Cochrane systematic review and meta-analysis of the impact of psychological treatments for people with epilepsy on health-related quality of life. Epilepsia 59: 315–332.
• 25   Patsopoulos  NA,  Evangelou  E,  Ioannidis  JP (2008) Sensitivity of between-study heterogeneity in meta-analysis: proposed metrics and empirical evaluation. Int J Epidemiol 37: 1148–1157.
• 26   Egger  M,  Davey Smith  G,  Schneider  M,  Minder  C (1997) Bias in meta-analysis detected by a simple, graphical test. BMJ 315: 629–634.
• 27   Duval  S,  Tweedie  R (2000) Trim and fill: a simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics 56: 455–463.
• 28   Alé  MC,  Echeverría  G,  Jugo  A,  Villarroel  L,  Maiz  A, et al. (2019) Non-HDL cholesterol levels in Chilean population and their association with diabetes mellitus and cardiovascular disease. Rev Med Chil 147: 1365–1373 (In Spanish).
• 29   Yang  S,  Jiao  X,  Huo  X,  Zhu  M,  Wang  Y, et al. (2018) Association between circulating full-length angiopoietin-like protein 8 and non-high-density lipoprotein cholesterol levels in Chinese non-diabetic individuals: a cross-sectional study. Lipids Health Dis 17: 161.
• 30   Kane  JP,  Pullinger  CR,  Goldfine  ID,  Malloy  MJ (2021) Dyslipidemia and diabetes mellitus: role of lipoprotein species and interrelated pathways of lipid metabolism in diabetes mellitus. Curr Opin Pharmacol 61: 21–27.
• 31   Zhang  L,  Chen  S,  Deng  A,  Liu  X,  Liang  Y, et al. (2015) Association between lipid ratios and insulin resistance in a Chinese population. PLoS One 10: e0116110.
• 32   Poznyak  A,  Grechko  AV,  Poggio  P,  Myasoedova  VA,  Alfieri  V, et al. (2020) The diabetes mellitus-atherosclerosis connection: the role of lipid and glucose metabolism and chronic inflammation. Int J Mol Sci 21: 1835.
• 33   Patel  S,  Rauf  A,  Khan  H,  Abu-Izneid  T (2017) Renin-angiotensin-aldosterone (RAAS): The ubiquitous system for homeostasis and pathologies. Biomed Pharmacother 94: 317–325.
• 34   Ahmadian  E,  Pennefather  PS,  Eftekhari  A,  Heidari  R,  Eghbal  MA (2016) Role of renin-angiotensin system in liver diseases: an outline on the potential therapeutic points of intervention. Expert Rev Gastroenterol Hepatol 10: 1279–1288.
• 35   Poznyak  AV,  Bharadwaj  D,  Prasad  G,  Grechko  AV,  Sazonova  MA, et al. (2021) Renin-angiotensin system in pathogenesis of atherosclerosis and treatment of CVD. Int J Mol Sci 22: 6702.
• 36   Durante  A,  Peretto  G,  Laricchia  A,  Ancona  F,  Spartera  M, et al. (2012) Role of the renin-angiotensin-aldosterone system in the pathogenesis of atherosclerosis. Curr Pharm Des 18: 981–1004.
• 37   Favre  GA,  Esnault  VL,  Van Obberghen  E (2015) Modulation of glucose metabolism by the renin-angiotensin-aldosterone system. Am J Physiol Endocrinol Metab 308: E435–E449.