Abstract
Background
Reduced kidney function is a major contributor to cardiovascular morbidity and mortality worldwide, particularly among individuals with hypertension. The triglyceride–glucose (TyG) index, a surrogate marker of insulin resistance, has been associated with renal dysfunction in multiple populations; however, data from Sub-Saharan Africa remain limited. We examined the association between the TyG index and reduced kidney function among hypertensive Nigerian adults.
Methods
In this cross-sectional study, 1,470 consecutively presenting hypertensive adults aged ≥ 18 years at a cardiac center in Lagos, Nigeria (2009–2024) were analyzed. Individuals with known diabetes, fasting glucose ≥ 126 mg/dL, pregnancy, acute kidney injury, or incomplete data were excluded. Reduced kidney function was defined as estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m² using the CKD-EPI 2021 race-free equation. The TyG index was analyzed as quartiles and as a standardized continuous variable. Multivariable logistic regression models were constructed. Sensitivity analyses included adjustment for HbA1c and exclusion of participants with HbA1c ≥ 6.5%. Restricted cubic spline and receiver operating characteristic analyses were performed. Estimated 10-year cardiovascular risk was calculated using the Framingham Risk Score.
Results
Reduced kidney function was present in 36.7% of participants. Compared with the lowest TyG index quartile, individuals in the highest quartile had higher odds of reduced kidney function (fully adjusted OR 3.14; 95% CI 2.25–4.39). Each 1-standard deviation increase in TyG index was associated with higher odds (adjusted OR 1.49; 95% CI 1.31–1.70). The association remained robust after adjustment for HbA1c and exclusion of participants with elevated HbA1c. Restricted cubic spline analysis demonstrated a log-linear relationship (p for non-linearity = 0.43). The TyG index showed modest discrimination (AUC 0.664).
Conclusions
Higher TyG index values are independently associated with reduced kidney function in hypertensive Nigerian adults, supporting a graded relationship between metabolic dysfunction and renal impairment. The TyG index may serve as a simple adjunct tool for cardio–renal risk stratification in resource-limited settings.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12872-026-05985-5.
Keywords: Reduced kidney function, Triglyceride–glucose index, Hypertension, Insulin resistance, Cardiometabolic risk, Nigeria
Introduction
Cardiovascular disease (CVD), chronic kidney disease (CKD), and type 2 diabetes are interrelated conditions that share common pathophysiological pathways rooted in insulin resistance and metabolic dysfunction. CKD alone affects approximately 8–13% of the global population and is independently associated with increased cardiovascular risk, healthcare utilization, and premature mortality [1, 2]. In Nigeria, community- and hospital-based studies estimate CKD prevalence between 11% and 15% in the general population, with substantially higher rates among individuals with hypertension and diabetes, of hypertension is a major risk factors [3–6]. Assessment of estimated glomerular filtration rate (eGFR) and urine albumin-to-creatinine ratio (UACR) is recommended for kidney disease evaluation and risk stratification [7, 8]. However, these measures typically identify established structural or functional renal injury rather than early upstream markers of metabolic risk.
The triglyceride–glucose (TyG) index, calculated from fasting triglyceride and glucose concentrations, has emerged as a practical surrogate marker of insulin resistance [9]. Compared with traditional measures such as HOMA-IR, the TyG index demonstrates strong correlations with metabolic syndrome and type 2 diabetes across diverse populations [10]. Insulin resistance promotes renal injury through diverse mechanisms including glomerular hyperfiltration, endothelial dysfunction, oxidative stress, inflammation, and lipid-mediated nephrotoxicity [11–13], providing biological plausibility for a link between elevated TyG values and impaired kidney function.
Observational studies and meta-analyses from Asia, Europe, and North America have reported associations between higher TyG index values and reduced kidney function, incident CKD, accelerated renal function decline, and adverse cardiovascular outcomes [14–23]. However, evidence from African populations remains limited. Given the high prevalence of hypertension and metabolic risk factors in Sub-Saharan Africa, evaluation of the TyG index within this context is particularly relevant.
In Nigeria, fasting triglyceride and glucose measurements are routinely obtained during hypertension evaluation, creating an opportunity to assess the TyG index without additional laboratory cost. Prior work among hypertensive Nigerian adults demonstrated an association between TyG index and estimated 10-year cardiovascular risk [24]. Nevertheless, its relationship with reduced kidney function in African hypertensive cohorts has not been well characterized.
Accordingly, this study aimed to examine the association between the triglyceride–glucose index and reduced kidney function (eGFR < 60 mL/min/1.73 m²) among hypertensive Nigerian adults and to evaluate its discriminatory performance for identifying reduced kidney function. By providing region-specific evidence, this study seeks to inform cardio-renal risk stratification strategies in resource-limited settings.
Methods
Study design and reporting standards
This cross-sectional study examined the association between the triglyceride–glucose (TyG) index and reduced kidney function among Nigerian adults with hypertension. The study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.
Study setting and population
This study involved a retrospective review of clinical and laboratory records of hypertensive adults presenting to a cardiac center in Lagos, Nigeria, between January 2009 and December 2024. Data extraction and verification were conducted between January 2023 and June 2024.
Eligibility criteria
Eligible participants were adults aged ≥ 18 years with a documented diagnosis of hypertension and complete clinical, anthropometric, and biochemical data required for TyG index calculation and kidney function assessment. Participants were excluded if they had a known diagnosis of diabetes (whether historically documented or based on current use of glucose-lowering agents), fasting blood glucose ≥ 126 mg/dL, or pregnancy. Additional exclusions included acute kidney injury, defined according to KDIGO guidelines [25], and laboratory values considered physiologically implausible. Implausible values were defined a priori as fasting glucose < 40 mg/dL or > 600 mg/dL, triglycerides < 20 mg/dL or > 1,000 mg/dL, or serum creatinine values inconsistent with life, as these were considered likely to represent measurement or transcription errors. A total of 1,612 records were screened, and 1,470 participants met inclusion criteria and were included in the final analysis.
Data collection and measurements
Clinical and anthropometric data
Sociodemographic variables (age, sex), anthropometric measures (body mass index [BMI], waist circumference), and lifestyle characteristics (smoking status, alcohol use) were extracted from medical records. Blood pressure was measured by trained personnel using standardized protocols.
Laboratory measurements
Fasting blood samples were obtained after an overnight fast of 10–12 h. Fasting glucose and lipid parameters were measured using enzymatic colorimetric methods on an automated chemistry analyzer (Mindray BA-88 A), with adherence to internal and external quality control procedures. Serum creatinine was measured using the Jaffe method. Estimated glomerular filtration rate (eGFR) was calculated using the CKD-EPI 2021 race-free equation [26].
Definition of variables
Exposure variable: TyG index
The TyG index was calculated using the validated formula:
 |
TyG index values were categorized into quartile and also analyzed as a standard continuous variable (per 1-standard deviation increase).
Outcome variable: reduced kidney function
Reduced kidney function was defined as eGFR < 60 mL/min/1.73 m², consistent with KDIGO criteria for decreased kidney function [27]. Given the cross-sectional design and reliance on a single creatinine measurement, chronicity could not be confirmed.
Cardiovascular risk assessment
Estimated 10-year cardiovascular risk was calculated using the online version Framingham Risk Score (FRS), which estimates the probability of a first major cardiovascular event over 10 years based on age, sex, systolic blood pressure, treatment for hypertension, total cholesterol, HDL cholesterol, smoking status, and diabetes status. The Framingham model is widely used for global cardiovascular risk stratification in clinical and epidemiologic studies [28]. Although originally derived in a predominantly Western population, the Framingham Risk Score remains widely applied in Sub-Saharan African clinical practice for cardiovascular risk estimation. For this study, the estimated 10-year risk was expressed as a percentage and analyzed as a continuous variable when comparing participants with and without reduced kidney function. The Framingham risk score was not included in multivariable regression models evaluating the association between the TyG index and reduced kidney function, as it represents a composite downstream cardiovascular risk estimate rather than a biological confounder of the TyG index-eGFR relationship.
Causal framework and covariates selection
A directed acyclic graph (DAG) was constructed to conceptualize the relationships between the triglyceride–glucose (TyG) index and reduced kidney function and to guide covariate selection (Supplementary Figure S1). Variables considered potential confounders defined as factors associated with both the exposure and the outcome were selected a priori based on biological plausibility and existing literature. These included age, sex, body mass index, blood pressure, lipid parameters, smoking status, and alcohol consumption. Variables considered mediators were excluded from the adjustment models to avoid over-adjustment.
Bias minimization
Measurement bias was minimized through standardized blood pressure measurement procedures and routine laboratory quality control. This was achieved by training observers on the universal standard for blood pressure measurement which includes ensuring participants are seated 5 min before measurement is taken with arm positioned at the heart level using the appropriately sized cuff on the non-dominant arm and a calibrated sphygmomanometer was used. To ensure reliability, 2–3 readings were taken and an average was used. All laboratory personnel were adequately trained, and standard reagents and kits from validated sources were used throughout the study to ensure consistency across measurements. Fasting samples were collected after 10–12 h of fasting and processed within a standardized timeframe. Consecutive sampling was used to reduce selection bias, whereby all eligible patients presenting during the study period were included without predefined sampling intervals. Data entry was double-checked to minimize transcription errors.
Statistical analysis
Continuous variables were summarized as mean ± standard deviation (SD) or median (interquartile range [IQR]) as appropriate. Categorical variables were presented as frequencies and percentages. Between-group comparisons were performed using Student’s t-test or Mann–Whitney U test for continuous variables and chi-square test for categorical variables. Multivariable logistic regression models were constructed to evaluate the association between TyG index (as quartiles and as standardized continuous variable) and reduced kidney function. Models were specified as follows: Model 1: Unadjusted. Model 2: Adjusted for age and sex. Model 3 (Fully Adjusted): Adjusted for age, sex, BMI, SBP, DBP, HDL-c, LDL-c, smoking status, and alcohol use. Model 4 (Sensitivity Model): Fully adjusted model plus HbA1c. Odds ratios (ORs) with 95% confidence intervals (CI) were reported. To address potential residual confounding from undiagnosed dysglycemia, an additional sensitivity analysis was performed excluding participants with HbA1c ≥ 6.5%, consistent with diagnostic threshold for diabetes. Multivariable logistic regression models were re-estimated in this restricted cohort to evaluate the robustness of the association between TyG index quartiles and reduced kidney function. Restricted cubic spline regression with knots at the 5th, 35th, 65th, and 95th percentiles was used to explore potential non-linear associations. Multicollinearity was assessed using variance inflation factors (VIF), with values < 2 indicating no significant collinearity.
Discriminatory performance
Receiver operating characteristic (ROC) curve analysis was performed to evaluate the discriminatory ability of the TyG index for identifying reduced kidney function. The area under the curve (AUC) with 95% confidence intervals was calculated. The optimal cut-off value was determined using the Youden index. Sensitivity and specificity were reported. All analyses were conducted using SPSS version 28.0 (IBM Corp., Armonk, NY, USA). A two-tailed p-value < 0.05 was considered statistically significant.
Effect modification
Sex-stratified and age-stratified analyses were conducted to assess potential effect modification based on prior evidence of sex- and age-related differences in TyG-index associated metabolic risk.
Ethical considerations
All patient data were anonymized prior to analysis. The Health Research Ethics Committee of Lagos University Teaching Hospital waived the requirement for individual informed consent due to the retrospective nature of the study and the use of de-identified records.
Ethics approval and consent to participate
This study was conducted in accordance with the ethical principles of the Declaration of Helsinki (as revised). Ethical approval was obtained from the Health Research Ethics Committee of Lagos University Teaching Hospital (Approval No. ADM/DSCT/HERC/APP/6494). The requirement for individual informed consent was waived due to the retrospective design of the study and the use of fully anonymized, de-identified patient records prior to analysis.
Results
Participants characteristics
A total of 1,612 records were screened, of which 142 were excluded based on predefined criteria. The final analytic sample comprised 1,470 hypertensive adults. A flow diagram illustrating participant selection is presented in Fig. 1. There was a slight male preponderance of 52.7% and a study population mean age of 55.8 ± 13.9 years. Reduced kidney function (eGFR < 60 mL/min/1.73 m²) was identified in 36.7% participants. Compared with participants with preserved kidney function, those with reduced kidney function were significantly older and more frequently female (both p < 0.001). They demonstrated a less favorable cardiometabolic profile, characterized by higher body mass index, greater waist circumference, and elevated systolic blood pressure. Markers of metabolic dysregulation including fasting glucose, HbA1c, triglycerides, uric acid, and TyG index were significantly higher among individuals with reduced kidney function (all p < 0.01), whereas mean HDL cholesterol level was modestly lower (p = 0.029). In contrast, alcohol consumption and smoking were more commonly reported among participants with preserved kidney function. Age-stratified distributions demonstrated a progressive increase in reduced kidney function across advancing age categories (p < 0.001). Median estimated 10-year cardiovascular risk scores were significantly higher among participants with reduced kidney function compared with those with preserved eGFR (p < 0.001). The overall staging distribution of kidney function within the cohort is illustrated in Fig. 2.
Fig. 1.
Flow diagram of participants recruitment, eligibility assessment and inclusion in the analysis. Percentages are shown relative to the previous step. eGFR, estimated glomerular filtration rate; TyG, triglyceride-glucose index
Fig. 2.
Distribution of KDIGO eGFR categories in the participants
Prevalence of reduced kidney function across TyG index quartiles
Participants with reduced kidney function had significantly higher TyG index values than those without (Table 1). When categorized into quartiles, the prevalence of reduced kidney function increased progressively across TyG index categories, with the highest burden observed in Quartile 4 (Fig. 3; p-trend < 0.001).
Table 1.
Baseline demographic, clinical, and biochemical characteristics stratified by estimated glomerular filtration rate
| Variable | Overall (n = 1470) |
eGFR < 60 mL/min/1.73 m² (n = 540) |
eGFR ≥ 60 mL/min/1.73 m² (n = 930) |
p-value |
|---|---|---|---|---|
| Anthropometric and haemodynamic variables | ||||
| Body mass index, kg/m² (mean ± SD) | 31.73 ± 6.6 | 32.28 ± 6.9 | 31.41 ± 6.3 | 0.014 |
| Waist circumference, cm (mean ± SD) | 105.48 ± 15.9 | 106.86 ± 15.8 | 104.67 ± 15.9 | 0.011 |
| Systolic BP, mmHg (mean ± SD) | 154.81 ± 24.3 | 157.96 ± 26.1 | 152.98 ± 23.0 | < 0.001 |
| Diastolic BP, mmHg (mean ± SD) | 89.31 ± 14.5 | 88.82 ± 15.9 | 89.60 ± 13.5 | 0.318 |
| Biochemical variables | ||||
| Uric acid, mg/dL (mean ± SD) | 7.25 ± 2.3 | 7.80 ± 2.8 | 6.93 ± 1.8 | < 0.001 |
| LDL-C, mg/dL (median [IQR]) | 122.3 [97.0–146.6] | 118.7 [92.3–147.5] | 124.3 [100.0–146.2] | 0.033 |
| HDL-C, mg/dL (median [IQR]) | 50.2 [42.3–59.7] | 49.3 [40.8–58.5] | 50.9 [43.1–60.2] | 0.029 |
| Total cholesterol, mg/dL (median [IQR]) | 196.6 [171.5–227.6] | 193.6 [167.4–221.5] | 198.1 [172.9–230.0] | 0.077 |
| VLDL-C, mg/dL (median [IQR]) | 22.3 [16.8–30.1] | 22.7 [17.9–30.6] | 22.1 [16.6–29.5] | 0.005 |
| Triglycerides, mg/dL (median [IQR]) | 119.1 [88.9–155.6] | 123.5 [94.4–159.4] | 116.0 [87.2–151.7] | 0.003 |
| Fasting blood glucose, mg/dL (mean ± SD) | 108.9 ± 28.3 | 117.4 ± 24.5 | 105.0 ± 23.2 | < 0.001 |
| HbA1c, % (mean ± SD) | 6.3 ± 1.4 | 6.59 ± 1.6 | 6.11 ± 1.4 | < 0.001 |
| 10-year CVD risk, % (median [IQR]) | 6.0 [2.0–14.0] | 10.0 [4.0–17.0] | 5.0 [2.0–11.0] | < 0.001 |
| TyG index (mean ± SD) | 8.72 ± 0.6 | 8.84 ± 0.6 | 8.66 ± 0.5 | < 0.001 |
Fig. 3.
Prevalence of reduced kidney function across the TyG index quartiles
Sex-stratified prevalence patterns demonstrated heterogeneity (Table 2). Among men, reduced kidney function was disproportionately concentrated in the highest TyG quartile. Among women, a graded increase in prevalence was observed across quartiles, (p-trend < 0.001).
Table 2.
Sex-stratified distribution of reduced kidney function across TyG index quartiles
| TyG Quartile | eGFR<60 ml/min n (%) |
eGFR≥60 ml/min n (%) |
p-value |
|---|---|---|---|
| Male | < 0.001 | ||
| Q1 (< 8.38) | 50 (22.3) | 17 (21.3) | |
| Q2 (8.38–8.70) | 52 (23.2) | 133 (24.2) | |
| Q3 (8.71–8.94) | 32 (14.3) | 160 (29.1) | |
| Q4 (> 8.94) | 90 (40.2) | 140 (25.5) | |
| Female | < 0.001 | ||
| Q1 (< 8.38) | 60 (19.0) | 140 (36.8) | |
| Q2 (8.38–8.70) | 70 (22.2) | 113 (29.7) | |
| Q3 (8.71–8.94) | 88 (27.8) | 87 (22.9) | |
| Q4 (> 8.94) | 98 (31.0) | 40 (10.5) |
Sex and age stratified analyses
Sex-stratified regression analyses demonstrated heterogeneity in the association between TyG index and reduced kidney function (Table 3). Among men, only the highest TyG quartile was significantly associated with reduced kidney function (OR 1.85; 95% CI 1.16–2.91). In contrast, among women, a graded association was observed across increasing TyG quartile indices. Compared with Quartile 1, women in Quartile 2 had 1.58-fold higher odds (95% CI 1.00–2.47), those in Quartile 3 had 2.67-fold higher odds (95% CI 1.71–4.19), and those in Quartile 4 had 5.63-fold higher odds of reduced kidney function (95% CI 3.41–9.28). Age-stratified analyses also revealed variation in effect size. Among participants aged < 40 years, those in the highest TyG quartile had markedly increased odds of reduced kidney function (OR 34.22; 95% CI 5.49–201.03), although confidence intervals were wide due to small group size. In the 40–59-year age group, Quartile 3 (OR 1.82; 95% CI 1.09–3.05) and Quartile 4 (OR 4.80; 95% CI 2.92–7.88) were significantly associated with reduced kidney function. Among participants aged ≥ 60 years, the association was attenuated and did not reach statistical significance in the highest quartile (OR 1.60; 95% CI 0.97–2.63).
Table 3.
Sex- and age-stratified multivariable logistic regression analyses of the association between triglyceride–glucose index quartiles and reduced kidney function
| Overall | |||
|---|---|---|---|
| TyG Quartile | Adjusted OR | 95% CI | p-value |
| Q1 (< 8.38) | 1.00 (Reference) | — | — |
| Q2 (8.38–8.70) | 1.28 | 0.92–1.78 | 0.149 |
| Q3 (8.71–8.94) | 1.37 | 0.98–1.91 | 0.091 |
| Q4 (> 8.94) | 3.20 | 2.29–4.47 | < 0.001 |
| Sex-Stratified Analyses* | |||
| Male | |||
| Q1 | 1.00 | — | — |
| Q2 | 0.96 | 0.59–1.57 | 0.868 |
| Q3 | 0.56 | 0.33–1.04 | 0.131 |
| Q4 | 1.85 | 1.16–2.91 | 0.010 |
| Female | |||
| Q1 | 1.00 | — | — |
| Q2 | 1.58 | 1.00–2.47 | 0.048 |
| Q3 | 2.67 | 1.71–4.19 | < 0.001 |
| Q4 | 5.63 | 3.41–9.28 | < 0.001 |
| Age-Stratified Analyses* | |||
| < 40 Years | |||
| Q1 | 1.00 | — | — |
| Q2 | 3.78 | 0.74–19.22 | 0.109 |
| Q3 | 10.05 | 2.02–49.49 | 0.005 |
| Q4 | 34.22 | 5.49–201.03 | < 0.001 |
| 40–59 Years | |||
| Q1 | 1.00 | — | — |
| Q2 | 1.08 | 0.63–1.86 | 0.774 |
| Q3 | 1.82 | 1.09–3.05 | 0.022 |
| Q4 | 4.80 | 2.92–7.88 | < 0.001 |
| ≥ 60 Years | |||
| Q1 | 1.00 | — | — |
| Q2 | 0.82 | 0.50–1.35 | 0.443 |
| Q3 | 1.05 | 0.64–1.72 | 0.843 |
| Q4 | 1.60 | 0.97–2.63 | 0.064 |
Association between TyG index and reduced kidney function
Table 4 presents multivariable logistic regression models examining the association between TyG index and reduced kidney function. In unadjusted analysis (Model 1), participants in the highest TyG quartile had significantly higher odds of reduced kidney function compared with the lowest quartile (OR 2.44; 95% CI 1.80–3.30). After adjustment for age and sex (Model 2), the association strengthened (OR 3.31; 95% CI 2.38–4.62). In the fully adjusted model (Model 3), controlling for age, sex, BMI, systolic and diastolic blood pressure, HDL-c, LDL-c, smoking status, and alcohol use, participants in Quartile 4 retained over threefold higher odds of reduced kidney function (adjusted OR 3.14; 95% CI 2.25–4.39). When analyzed as a continuous variable, each 1-standard deviation increase in the TyG index was associated with higher odds of reduced kidney function (adjusted OR 1.49; 95% CI 1.31–1.70), supporting a dose-response relationship (Table 5).
Table 4.
Multivariable association between TyG index quartiles and reduced kidney function
| TyG Quartile | Model 1 Unadjusted OR (95% CI) | Model 2 (Age & Sex Adjusted) | Model 3 Fully Adjusted† | Model 4 (+ HbA1c) | Model 5 (Excluding HbA1c ≥ 6.5%) |
|---|---|---|---|---|---|
| Q1 | Reference | Reference | Reference | Reference | Reference |
| Q2 | 1.16 (0.85–1.58) | 1.32 (0.95–1.84) | 1.26 (0.90–1.76) | 1.25 (0.89–1.75) | 1.14 (0.77–1.69) |
| Q3 | 1.14 (0.83–1.55) | 1.40 (1.00–1.96) | 1.38 (0.99–1.93) | 1.29 (0.92–1.81) | 1.76 (1.18–2.64) |
| Q4 | 2.44 (1.80–3.30) | 3.31 (2.38–4.62) | 3.14 (2.25–4.39) | 2.70 (1.92–3.81) | 2.46 (1.59–3.80) |
†Fully adjusted model includes Age, Sex, BMI, SBP, DBP, HDL-c, LDL-c, smoking status, and alcohol use
Table 5.
Association between TyG index (continuous) and reduced kidney function
| Variable | Adjusted OR (95% CI) |
|---|---|
| TyG (per 1 SD increase) | 1.49 (1.31–1.70) |
HbA1c sensitivity and glycemic exclusion analyses
To evaluate potential confounding from chronic hyperglycemia, HbA1c was incorporated into the fully adjusted model (Model 4). The association was modestly attenuated but remained robust (adjusted OR 2.70; 95% CI 1.92–3.81) (Table 4). Further restriction of the cohort to participants with HbA1c < 6.5% (Model 5) yielded consistent findings. In this glycemically restricted population, Quartile 3 was associated with 1.76-fold higher odds (95% CI 1.18–2.64;), and Quartile 4 with 2.46-fold higher odds (95% CI 1.59–3.80) of reduced kidney function compared with Quartile 1 (Table 4).
Spline and discriminatory analyses
Restricted cubic spline analysis demonstrated a log-linear association between the TyG index and reduced kidney function, with no evidence of non-linearity (p for non-linearity = 0.43) (Fig. 4).
Fig. 4.
Restricted cubic spline analysis - TyG index and reduced kidney function. Adjusted for age, sex, BMI, SBP, DBP, HDL-c, LDL-c, smoking, alcohol | 4 knots | p for non-linearity = 0.43 (linear relationship)
Discriminatory performance of TyG index
Receiver Operating Characteristics (ROC) analysis showed modest but statistically significant discrimination of the TyG index for reduced kidney function (overall AUC 0.664; p < 0.001), with slightly higher performance in women (AUC 0.748) than men (AUC 0.709). Sensitivity and specificity at the optimal cut-off were 67.3% and 65.1%, respectively (Table 6 and Fig. 5).
Table 6.
Sex-specific and overall Receiver Operating Characteristic (ROC) analysis of the triglyceride–glucose index for identifying reduced kidney function
| Group | AUC (95% CI) | p-value | Sensitivity | Specificity | Youden Index |
|---|---|---|---|---|---|
| Male | 0.709 | < 0.001 | 0.731 | 0.684 | 0.415 |
| Female | 0.748 | < 0.001 | 0.764 | 0.622 | 0.386 |
| Overall | 0.664 | < 0.001 | 0.673 | 0.651 | 0.324 |
Fig. 5.
Discriminatory performance of the triglyceride–glucose index for reduced kidney function: sex-stratified ROC analysis
Discussion
In this cross-sectional study of hypertensive Nigerian adults, we found that higher triglyceride–glucose (TyG) index values were independently associated with reduced kidney function. The association was consistent across categorical, continuous, and spline-based analyses, demonstrating a graded dose-response relationship. Notably, individuals in the highest TyG index quartile had more than threefold higher odds of reduced kidney function(adjusted OR 3.14), and each 1-standard deviation increase in TyG was associated with a 49% increase in odds. These findings highlight the relevance of metabolic dysfunction in renal impairment among hypertensive populations. Importantly, further adjustment for HbA1c resulted in only modest attenuation (adjusted OR 2.70), indicating that the observed association extends beyond glycemic exposure alone and likely reflects broader metabolic dysregulation (Fig. 4).
These findings are consistent with a growing body of international evidence linking the TyG index to renal dysfunction [12–14]. However, the magnitude of association observed in this study appears greater than that reported in several meta-analyses, which typically demonstrate 1.5-2.5-fold increases in risk [15–23]. This difference may reflect a higher burden of untreated metabolic abnormalities, delayed diagnosis, and clustering of cardiometabolic risk factors in Sub-Saharan African populations (Fig. 5).
Reduced kidney function is not merely a renal endpoint but a recognized amplifier of cardiovascular risk. Large epidemiologic studies have demonstrated that even moderate reductions in eGFR are independently associated with increased cardiovascular mortality, major adverse cardiovascular events, heart failure, and vascular dysfunction [29–34]. In our cohort, participants with reduced kidney function also exhibited significantly higher estimated 10-year cardiovascular risk scores, reinforcing the established relationship between impaired renal function and heightened cardiovascular vulnerability. The clustering of metabolic dysregulation, reduced kidney function, and elevated global cardiovascular risk further supports the positioning of the TyG index within the broader cardio-renal risk spectrum [32–34]. Together, these findings suggest that metabolic markers associated with reduced kidney function may have important implications for cardiovascular risk assessment among hypertensive populations.
The biological plausibility of the observed association is well established. Insulin resistance, as reflected by the TyG index, contributes to renal injury through multiple interrelated mechanisms, including endothelial dysfunction, oxidative stress, inflammation, and lipid-induced toxicity [12–14]. These processes promote glomerular hyperfiltration, mesangial expansion, and progressive nephron loss. The absence of non-linearity in spline analysis further supports a log-linear relationship, suggesting that incremental increases in metabolic dysfunction are associated with progressively higher odds of renal impairment.
Sex-stratified analyses revealed notable heterogeneity. Among women, a graded relationship was observed across increasing TyG index quartiles, with substantially elevated odds in the highest category. In contrast, among men, significant association was confined to the highest quartile. These differences may reflect sex-related variations in adiposity distribution, hormonal regulation, and metabolic vulnerability, and are consistent with prior reports describing sex-specific cardiometabolic and renal risk patterns [21, 22]. The stronger gradient observed among women warrants further exploration in longitudinal cohorts.
Age also modified the association. Although markedly elevated odds were observed among participants younger than 40 years, wide confidence intervals indicate limited precision in this subgroup. Significant associations were also present among individuals aged 40–59 years, whereas the relationship was attenuated among those aged ≥ 60 years. In older adults, age-related nephron loss and the cumulative burden of comorbidities may diminish the relative contribution of metabolic markers to renal impairment [8]. These findings suggest that elevated TyG values in younger hypertensive individuals may identify a particularly adverse metabolic phenotype with potential long-term cardio-renal implications [21].
Receiver operating characteristic analysis demonstrated modest but statistically significant discriminatory ability of the TyG index for identifying reduced kidney function (overall AUC 0.664; p < 0.001), with slightly higher discriminatory performance in women than in men. While TyG is not intended to replace direct assessment of kidney function, its balanced sensitivity and specificity suggest potential utility as a simple adjunct marker in cardiovascular risk evaluation, particularly in resource-limited settings where early identification of high-risk individuals remains challenging.
The high prevalence of reduced kidney function observed in this cohort is consistent with previous reports from Nigeria and other Sub-Saharan African settings documenting substantial renal disease burden among hypertensive individuals [3–6, 31]. Limited access to early screening and delayed presentation contribute to adverse outcomes. In this context, leveraging routinely available metabolic parameters to enhance cardio-renal risk stratification may have practical clinical relevance.
Several limitations merit consideration. The cross-sectional design precludes causal inference and does not establish temporality. Reduced kidney function was defined using a single eGFR measurement; therefore, chronicity could not be confirmed. Albuminuria data were unavailable, limiting comprehensive kidney disease classification. Although extensive covariate adjustment was performed including sensitivity analyses excluding individuals with HbA1c ≥ 6.5%, residual confounding cannot be excluded. Additionally, the hospital-based nature of the cohort may limit generalizability to community populations. Residual confounding from unmeasured variables, including medication use and socioeconomic status, is acknowledged.
Despite these limitations, this study provides region-specific evidence linking metabolic dysregulation, as reflected by the TyG index, with reduced kidney function in a hypertensive African population. The consistency of associations across adjusted, stratified, and restricted analyses reinforces the relevance of metabolic risk clustering within the cardio-renal axis in this setting.
Prospective longitudinal studies are warranted to determine whether elevated TyG index values predict incident decline in kidney function and cardiovascular events, and to clarify their potential role within integrated cardiovascular and renal risk assessment frameworks.
Conclusion
In hypertensive Nigerian adults, higher triglyceride–glucose (TyG) index values were independently associated with reduced kidney function, even after extensive adjustment for traditional cardiovascular and metabolic risk factors. The persistent association after accounting for glycemic exposure support the role of broader metabolic dysregulation beyond overt hyperglycemia.
Given the established relationship between reduced kidney function and adverse cardiovascular outcomes, these findings reinforce the relevance of metabolic risk clustering within the cardio-renal continuum. The TyG index, derived from routinely measured fasting triglyceride and glucose concentrations, represents a simple and accessible metabolic marker that may enhance early cardio-renal risk stratification in hypertensive populations, particularly in resource-constrained settings.
Prospective longitudinal studies are needed to determine whether elevated TyG index values predict incident renal decline and cardiovascular outcomes in African populations.
Supplementary Information
Acknowledgements
The authors wish to thant the research assistants and the staff of the Meridian Cardiac Center, Lagos Nigeria for their support.
Abbreviations
- Abbreviation
Full Meaning
- AUC
Area Under Curve
- BMI
Body mass index
- BP
Blood pressure
- CI
Confidence interval
- CKD
Chronic kidney disease
- CKD-EPI
Chronic Kidney Disease Epidemiology Collaboration
- CVD
Cardiovascular disease
- DBP
Diastolic blood pressure
- eGFR
Estimated glomerular filtration rate
- FBG
Fasting blood glucose
- HDL-c
High-density lipoprotein cholesterol
- IQR
Interquartile range
- IR
Insulin resistance
- KDIGO
Kidney Disease: Improving Global Outcomes
- LDL-c
Low-density lipoprotein cholesterol
- OR
Odds ratio
- ROC
Receiver operating characteristic
- SBP
Systolic blood pressure
- SD
Standard deviation
- TC
Total cholesterol
- TG
Triglyceride
- TyG
Triglyceride–glucose index
- UACR
Urine albumin-to-creatinine ratio
- VLDL
Very-low-density lipoprotein
Authors’ contributions
CEA: Conceptualization, study design, data analysis, manuscript drafting, overall supervision; POA: Data collection, data curation, manuscript review; MO: Data collection, analysis support, manuscript review; JNA and ACM: Manuscript review.
Funding
This study received no external funding from public, commercial, or not-for-profit agencies.
Data availability
The datasets generated and/or analysed during the current study are available from the corresponding author on reasonable request. Due to institutional and ethical restrictions involving patient-level clinical data, the data are not publicly available but may be shared in anonymized form upon reasonable request and with appropriate ethical approval.
Declarations
Ethics approval and consent to participate
This study was conducted in accordance with the ethical principles of the Declaration of Helsinki (as revised). Ethical approval was obtained from the Health Research Ethics Committee of Lagos University Teaching Hospital (Approval No. ADM/DSCT/HERC/APP/6494). The requirement for individual informed consent was waived due to the retrospective design of the study and the use of fully anonymized, de-identified patient records prior to analysis.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Hill NR, Fatoba ST, Oke JL, Hirst JA, O’Callaghan CA, Lasserson DS, et al. Global prevalence of chronic kidney disease: a systematic review and meta-analysis. PLoS ONE. 2016;11(7):e0158765. 10.1371/journal.pone.0158765. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.GBD Chronic Kidney Disease Collaboration. Global, regional, and national burden of chronic kidney disease, 1990–2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet. 2020;395(10225):709–33. 10.1016/S0140-6736(20)30045-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Chukwuonye II, Ogah OS, Anyabolu EN, Ohagwu KA, Nwabuko OC, Onwuchekwa U, et al. Prevalence of chronic kidney disease in Nigeria: systematic review of population-based studies. Int J Nephrol Renovasc Dis. 2018;11:165–72. 10.2147/IJNRD.S162230. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Alebiosu CO, Ayodele OE, Abbas A, Olutoyin AI. Chronic renal failure at the Olabisi Onabanjo University Teaching Hospital, Sagamu. West Afr J Med. 2006;25(3):190–3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Ulasi II, Ijoma CK, Onodugo OD, Arodiwe EB, Ifebunandu NA, Okoye JU. Towards prevention of chronic kidney disease in Nigeria: a community-based study in Southeast Nigeria. Kidney Int Suppl (2011). 2013;3(2):195–201. 10.1038/kisup.2013.4.
- 6.Olanrewaju TO, Aderibigbe A, Popoola AA, Braimoh KT, Buhari MO, Adedoyin OT, et al. Prevalence of chronic kidney disease and risk factors in North-Central Nigeria: a population-based survey. BMC Nephrol. 2020;21(1):467. 10.1186/s12882-020-02126-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Okpechi IG, Chukwuonye II, Tandi T, et al. Early detection of chronic kidney disease in Africa: a systematic review. BMC Nephrol. 2020;21:460. 10.1186/s12882-020-02061-7.33148221 [Google Scholar]
- 8.Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group. KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int Suppl (2011). 2013;3(1):1–150. 10.1038/kisup.2012.73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Khandpur S, Choi AI. Challenges in early detection and risk prediction of chronic kidney disease. Curr Opin Nephrol Hypertens. 2021;30(3):289–96. 10.1097/MNH.0000000000000697. [Google Scholar]
- 10.Simental-Mendía LE, Rodríguez-Morán M, Guerrero-Romero F. The triglyceride–glucose index as a surrogate marker for insulin resistance. Metab Syndr Relat Disord. 2008;6(4):299–304. 10.1089/met.2008.0034. [DOI] [PubMed] [Google Scholar]
- 11.Park HM, Kim HC, Kim DJ. Triglyceride–glucose index as a reliable marker for predicting metabolic syndrome and insulin resistance. Endocrinol Metab (Seoul). 2019;34(2):180–7. 10.3803/EnM.2019.34.2.180. [Google Scholar]
- 12.Wu T, Gao J, Chen G, et al. Hyperglycemia-induced kidney injury: mechanisms and therapeutic strategies. Front Endocrinol (Lausanne). 2022;13:834323. 10.3389/fendo.2022.834323. [Google Scholar]
- 13.Ren L, Jin S, Tong Y. Lipotoxicity and kidney disease: mechanistic insights and therapeutic opportunities. Nephrol Dial Transpl. 2021;36(12):2150–8. 10.1093/ndt/gfaa307. [Google Scholar]
- 14.Jin Q, Wang Q, Zhang H, et al. Oxidative stress, inflammation, and diabetic kidney disease: molecular mechanisms and potential interventions. J Diabetes Res. 2020;2020:2954879. 10.1155/2020/2954879. [Google Scholar]
- 15.Li H, Zuo Y, Wang L, et al. Association between the triglyceride–glucose index and chronic kidney disease: a systematic review and meta-analysis. J Clin Endocrinol Metab. 2023;108(4):e1231–42. 10.1210/clinem/dgac706. [Google Scholar]
- 16.Ali M, Khan M, Riaz M, et al. Triglyceride–glucose index and risk of chronic kidney disease: a meta-analysis. Clin Chim Acta. 2022;530:25–32. 10.1016/j.cca.2022.03.015. [Google Scholar]
- 17.Li X, Wang L, Zhou H, Xu H. Association between triglyceride–glucose index and chronic kidney disease: results from NHANES 1999–2020. Int Urol Nephrol. 2024;56(12):3605–16. 10.1007/s11255-024-04103-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Pasdar Y, Mohammadi S, Hamzeh B, et al. Triglyceride–glucose index and kidney dysfunction: findings from a large cohort study. Nutr Metab Cardiovasc Dis. 2021;31(10):2907–15. 10.1016/j.numecd.2021.07.013. [Google Scholar]
- 19.Ren X, Chen Z, Liu Y, et al. Triglyceride–glucose index and chronic kidney disease risk: a cohort study and updated meta-analysis. Front Endocrinol (Lausanne). 2022;13:919812. 10.3389/fendo.2022.919812. [Google Scholar]
- 20.Ali M, Meng X, Wang X. Triglyceride–glucose index as a predictor of chronic kidney disease: a systematic review. Front Endocrinol (Lausanne). 2022;13:883005. 10.3389/fendo.2022.883005. [Google Scholar]
- 21.Wang J, Li M, Song Y, et al. Prognostic significance of the triglyceride–glucose index in patients with chronic kidney disease. Clin Chim Acta. 2021;518:1–7. 10.1016/j.cca.2021.03.021.33741356 [Google Scholar]
- 22.Hong S, Han K, Park CY, et al. Relationship between the triglyceride–glucose index and kidney function in Korean adults. Endocrinol Metab (Seoul). 2020;35(1):137–47. 10.3803/EnM.2020.35.1.137. [Google Scholar]
- 23.Zhao S, Yu S, Chi C, et al. Association between the triglyceride–glucose index and renal function in Chinese adults: a population-based study. J Diabetes Investig. 2019;10(3):686–93. 10.1111/jdi.12948. [Google Scholar]
- 24.Amadi CE, Okorafor UC, Okorafor CI, Okwah MN, Akanbi PO, Okam OV, et al. Association between triglyceride–glucose index and estimated 10-year risk of cardiovascular disease in Nigerian non-diabetic hypertensives: a cross-sectional study. PLoS Glob Public Health. 2025;5(7):e0004760. 10.1371/journal.pgph.0004760. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Kidney Disease. Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group. KDIGO clinical practice guideline for acute kidney injury. Kidney Int Supplements. 2012;2(1):1–138. [Google Scholar]
- 26.Inker LA, Eneanya ND, Coresh J, Tighiouart H, Wang D, Sang Y, et al. New Creatinine- and Cystatin C–Based Equations to Estimate GFR without Race. N Engl J Med. 2021;385:1737–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kidney Disease. Improving Global Outcomes (KDIGO) CKD Work Group.KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int Supplements. 2013;3(1):1–150. [Google Scholar]
- 28.D’Agostino RB, Sr, Vasan RS, Pencina MJ, Wolf PA, Cobain M, Massaro JM, et al. General cardiovascular risk profile for use in primary care: the Framingham Heart Study. Circulation. 2008;117(6):743–53. 10.1161/CIRCULATIONAHA.107.699579. [DOI] [PubMed] [Google Scholar]
- 29.Matsushita K, van der Velde M, Astor BC, Woodward M, Levey AS, de Jong PE, et al. Association of estimated glomerular filtration rate and albuminuria with all-cause and cardiovascular mortality in general population cohorts: a collaborative meta-analysis. Lancet. 2010;375(9731):2073–81. 10.1016/S0140-6736(10)60674-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Fox CS, Matsushita K, Woodward M, Bilo HJ, Chalmers J, Heerspink HJ, et al. Associations of kidney disease measures with mortality and end-stage renal disease in individuals with and without hypertension: a meta-analysis. Lancet. 2012;380(9854):1662–73. 10.1016/S0140-6736(12)61272-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Okoye O, Oviasu E, Ijoma C, et al. Prevalence of chronic kidney disease and associated risk factors among adults in a rural Nigerian community. Int J Med Health Dev. 2020;25(2):105–15. 10.4103/ijmh.IJMH_47_19. [Google Scholar]
- 32.London GM. Arterial stiffness in chronic kidney disease and end-stage renal disease. Blood Purif. 2018;45(1–3):154–8. 10.1159/000485732. [DOI] [PubMed] [Google Scholar]
- 33.Damman K, Valente MA, Voors AA, O’Connor CM, van Veldhuisen DJ, Hillege HL. Renal impairment, worsening renal function, and outcome in patients with heart failure: an updated meta-analysis. Eur Heart J. 2014;35(7):455–69. 10.1093/eurheartj/eht386. [DOI] [PubMed] [Google Scholar]
- 34.Townsend RR, Anderson AH, Chirinos JA, Feldman HI, Grunwald JE, Nessel L, et al. Association of pulse wave velocity with chronic kidney disease progression and cardiovascular outcomes: the CRIC Study. Hypertension. 2018;71(6):1101–7. 10.1161/HYPERTENSIONAHA.117.10648. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated and/or analysed during the current study are available from the corresponding author on reasonable request. Due to institutional and ethical restrictions involving patient-level clinical data, the data are not publicly available but may be shared in anonymized form upon reasonable request and with appropriate ethical approval.