Abstract
Background
Chronic kidney disease (CKD) in older adults is associated with high morbidity and mortality, but accurate risk stratification remains challenging. The monocyte-to-high density lipoprotein ratio (MHR), an integrative biomarker of systemic inflammation and lipid metabolism, has been associated with adverse outcomes in younger and mixed-age populations. However, its prognostic value in older adults with CKD, including its association with renal mortality, has not been explored.
Methods
We analyzed 5,115 community-dwelling individuals aged ≥ 70 years with CKD from the National Health and Nutrition Examination Survey (NHANES) 1999–2018 and externally validated findings in 1,684 older individuals with CKD from the 2016 Health and Retirement Study (HRS). In NHANES, CKD was defined by an estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m² and/or albumin-to-creatinine ratio (ACR) ≥ 30 mg/g, whereas in HRS it was defined by eGFR < 60 mL/min/1.73 m². All-cause and renal-specific mortality were ascertained through linkage to the National Death Index and analyzed using survey-weighted Cox proportional hazards models and Fine and Gray competing risk models. To enhance model robustness and reduce overfitting, final multivariable models were informed by covariate selection using least absolute shrinkage and selection operator (LASSO) regression.
Results
During a mean follow-up of 82 months in NHANES, all-cause and renal-specific mortality rates were 72.7 (95% CI: 69.5-75.9) and 1.5 (95% CI: 1.1–2.1) per 1,000 person-years, respectively. Higher MHR was independently associated with an increased risk of all-cause mortality (HR 1.32, 95% CI 1.20–1.45) and renal-specific mortality (subdistribution HR 1.43, 95% CI 1.12–1.82) in fully adjusted models including ACR, eGFR, and relevant clinical and demographic covariates. Feature selection analyses identified MHR, eGFR, and ACR among the strongest predictors of renal-specific mortality. Associations were directionally consistent in the HRS validation cohort, where MHR remained independently associated with both all-cause mortality (HR 1.98, 95%CI 0.76–5.15) and renal-specific mortality (sHR 2.19, 95%CI 1.29–3.74). Stratified analyses suggested that MHR predicted all-cause mortality primarily in individuals with eGFR ≥ 45 mL/min/1.73 m², whereas its association with renal-specific mortality was strongest in those with more advanced CKD.
Conclusions
In older adults with CKD, MHR is an independent predictor of both all-cause and renal-specific mortality and provides prognostic information beyond eGFR and ACR. Given its derivation from routinely available laboratory tests, MHR may help refine risk stratification in older adults with CKD.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12967-026-07745-7.
Keywords: Older, CKD, MHR, Inflammation, HDL, Biomarkers, Machine learning, Prognosis
Background
Chronic kidney disease (CKD) represents a growing global health burden, especially among older adults. Affecting over 850 million people worldwide, CKD contributes significantly to morbidity, reduced quality of life, and premature mortality, with a disproportionate impact on aging populations [1, 2]. In older individuals, CKD frequently coexists with frailty, polypharmacy, and multimorbidity, making early risk stratification critical for improving clinical outcomes [3].
Inflammation is increasingly recognized as a central pathophysiological mechanism in CKD onset and progression, independent of traditional risk factors. Persistent low-grade inflammation may contribute to glomerular and tubulointerstitial injury, accelerated nephron loss and fibrosis through immune cell infiltration, cytokine release, and oxidative stress [4]. These processes are associated with increased serum concentrations of inflammatory mediators such as interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α), transforming growth factor β (TGF-β), and C-reactive protein (CRP), which are elevated from the early stages of the diseases and are linked to CKD progression, end-stage renal disease (ESRD), and death [5].
In parallel, disturbances in lipid metabolism, particularly low levels of high-density lipoprotein cholesterol (HDL-C), are common in CKD [6] and are thought to impair antioxidant and anti-inflammatory defenses within the renal microenvironment [7]; a recent Mendelian randomization study demonstrated that a 17 mg/dl increase in HDL-C concentration was associated with a 15% reduction in CKD risk [8]. Conversely, decreased HDL-C concentrations have been associated with an increased risk of cardiovascular and non-cardiovascular mortality in older patients with non-dialysis dependent CKD [9].
Evidence of the prognostic importance of inflammation and dyslipidaemia has led to the development of composite indices that integrate immune activation and lipid imbalance, providing improved prognostic insight compared to either component alone. Among these, the monocyte-to-HDL cholesterol ratio (MHR) has gained particular interest because it reflects the balance between the pro-inflammatory monocyte activity and the protective and antioxidant effects of HDL-C [10]. Increased MHR values have been associated with CKD presence and progression [11], resistant hypertension among patients with CKD [12], decreased estimated glomerular filtration rate (eGFR) values [13], and a higher prevalence of diabetes, CKD and cardiorenal syndromes in middle-aged adults [14–18]. Recent studies have further shown that MHR may predict prognosis in patients with cardiovascular diseases [19, 20], in CKD patients starting dialysis [11], and in the general adult population [21]; to this regard, a previous analysis conducted using the National Health and Nutrition Examination Survey (NHANES) cohort has shown that increasing MHR values may predict overall and cardiovascular mortality in adults > 20 years [21]. Nevertheless, whether MHR independently predicts adverse outcomes specifically in older adults with CKD, a group in whom immunosenescence, inflammaging, and CKD-related HDL dysfunction may magnify its biological signal, remains unknown. In addition, the association between MHR and renal-specific mortality has not been evaluated. This endpoint is particularly relevant in older patients with CKD, in whom competing non-renal causes of death are frequent and may obscure kidney-related risk. Finally, existing studies have not examined the incremental prognostic value beyond established kidney measures such as eGFR and albumin-to-creatinine ratio (ACR).
To address these gaps, we evaluated the associations between MHR and both all-cause and renal-specific mortality in a cohort of community-dwelling individuals aged ≥ 70 years with CKD from NHANES 1999–2018 and externally validated the findings in the Health and Retirement Study (HRS) population; we further assessed the incremental prognostic value of MHR beyond eGFR and ACR and applied feature-selection approaches to examine the stability of MHR as a predictor.
Methods
Study populations
We used data from NHANES, a survey conducted by the National Center for Health Statistics and the Centers for Disease Control and Prevention. Data used in this study were derived from a de-identified and public database (https://www.cdc.gov/nchs/nhanes/index.htm). NHANES protocols were approved by the National Center for Health Statistics Research Ethics Review Board, and all participants provided written informed consent. For the present analysis, we identified 5,946 participants aged ≥ 70 years with a laboratory diagnosis of CKD who were enrolled between the 1999–2000 and 2017–2018 survey cycles. After excluding participants with missing data on MHR (n = 207), eGFR and ACR (n = 274), other covariates (n = 344), and lost to follow-up (n = 6), the final analytical sample comprised 5,115 (NHANES sample). Sampling weights were applied to account for the complex, multistage survey design.
The Health and Retirement Study (HRS) is a national, population-based, multistage area probability household survey of Americans aged ≥ 51 years and their spouses [22]. Our analysis used data from the 2016 wave, which included biomarker assessments from the Venous Blood Study (HRS-VBS). Among 1,806 participants aged ≥ 70 years who underwent venous blood measurements and had CKD defined by eGFR < 60 ml/min/1.73 m2, we excluded those with missing MHR values (n = 56), and missing follow-up data (n = 66), yielding a final analytical sample of 1,684 participants (HRS sample).
Laboratory measurements
NHANES blood samples were collected and analyzed at Mobile Examination Centers between 1999 and 2018, whereas HRS samples were obtained during enhanced face-to-face interviews conducted in participants’ homes. Both studies implemented standardized protocols for morning blood collection, with participants instructed to fast for ≥ 8–9 h, except when clinically contraindicated (e.g., insulin-dependent diabetes). Serum creatinine was measured using isotope-dilution mass spectrometry-traceable (IDMS) methods. In the NHANES cohort, measurements evolved across survey cycles, utilizing the Jaffe rate method on Beckman Synchron LX20 (1999–2006), Beckman Coulter UniCel® DxC800 (2007–2014), and Roche Cobas 6000 (2015–2018) analyzers. HRS-VBS employed similar IDMS-traceable methodology at certified laboratories; eGFR was calculated using the Berlin Initiative Study (BIS) equation [23], which was specifically developed for older individuals. The equation is expressed as: eGFRBIS = 3,736 × SCr–0.87 × age–0.95 [× 0.82 if women], where SCr is serum creatinine. All creatinine measurements were reported in mg/dL, and the resulting eGFR values were expressed in mL/min/1.73 m². Definition of CKD differed between cohorts based on available data. In NHANES, CKD was defined using both eGFR (< 60 mL/min/1.73 m²) and urinary ACR ≥ 30 mg/g, consistent with KDIGO criteria [24], and previous studies in this cohort [25]; in HRS, CKD was defined by eGFR < 60 mL/min/1.73 m² because systematic urine collection was not performed. This eGFR-based definition captures moderate-to-severe CKD (stages 3–5) [24].
Lipid profiles and complete blood counts were assessed using standardized lab procedures in both cohorts. Total cholesterol and triglycerides were measured by enzymatic method. HDL-C was measured directly via homogeneous enzymatic assay. Low-density lipoprotein cholesterol (LDL-C) was derived from laboratory measurements when available. For participants without a directly measured LDL value, LDL-C was estimated using the enhanced Sampson equation. This method was selected because it improves estimation accuracy across a wide range of triglyceride values and outperforms both the Friedewald and Martin–Hopkins formulas, particularly in individuals with low LDL-C or elevated triglycerides [26]. All values were converted to mmol/L.
Complete blood counts were performed on EDTA-anticoagulated whole blood using automated analyzers: Coulter-based methods in NHANES and standardized hematology analyzers in HRS. The neutrophil-to-lymphocyte ratio (NLR) was calculated as the ratio between the absolute neutrophil to lymphocyte counts in NHANES and from differential percentages in HRS. MHR was created by dividing the total monocyte count by HDL-C concentration. To capture glucose metabolism, we also considered glycated hemoglobin (HbA1c) and fasting plasma glucose, when available.
Covariates
Demographic variables included age, sex, and race; lifestyle variables included obesity (defined as a body mass index [BMI] ≥ 30 Kg/m2) and smoking status (former or current); laboratory values included eGFR, ACR, either HbA1c (in NHANES) or fasting glucose (in HRS), total cholesterol, HDL-C, LDL-C, triglycerides, and NLR. Comorbidities were defined using self-reported physician diagnoses and included hypertension, heart disease (congestive heart failure, previous myocardial infarction or coronary heart disease), diabetes, cancer, respiratory disease (chronic bronchitis or emphysema), liver disease, and stroke. All conditions were identified through standardized interviews conducted by trained staff.
Medication use was also assessed. In NHANES we examined the use of lipid-lowering medications (statins, ezetimibe, PCSK9 inhibitors), renin-angiotensin-aldosterone system (RAAS) inhibitors and sodium–glucose cotransporter 2 (SGLT2) inhibitors. In HRS, due to differences in data collection, we evaluated a broader category of heart medications.
Mortality assessments
In NHANES, mortality status was ascertained through linkage to the National Death Index (NDI) using probabilistic matching algorithms based on identifiers such as Social Security Number, name, date of birth, sex, and race/ethnicity). Participants were followed from the baseline examination until death or censoring on December 31, 2019. Cause-specific mortality was classified according to ICD-10 codes, with renal-specific mortality defined by codes N00–N07, N17–N19, and N25–N27.
In HRS, mortality status was ascertained through a combination of biennial participant (or proxy) interviews and NDI linkage. This integrated active and passive follow-up approach has been shown to provide highly accurate and nearly complete ascertainment of mortality. The 2016 survey wave served as the baseline, and participants were followed for mortality outcomes through 2020 (maximum follow-up of approximately 4 years). Causes of death were obtained from the Exit interviews conducted in 2018 and 2020.
Statistical analysis
Because both studies employed a complex, multistage probability sampling design to select nationally representative participants, we incorporated sample weights, clustering, and stratification in all analyses to obtain population-level estimates.
Baseline characteristics of participants were summarized as mean (standard deviation [SD]) or median (interquartile range [IQR]) for continuous variables, and as number (percentage) for categorical variables. The distribution of study variables was assessed using visual inspection and the Shapiro-Wilk test. Skewed covariates (NLR and ACR) were log-transformed before inclusion in regression models. The association between MHR and all-cause mortality was investigated by using Cox proportional hazards models, whereas Fine and Gray competing risk models were used to evaluate the association with renal mortality, treating non-renal deaths as competing events. The exposure variable MHR was analyzed on its original continuous scale to preserve clinical interpretability.
Three types of hierarchical models were evaluated: Model A, adjusted for age, sex, and White race; Model B: model A + eGFR and log-transformed ACR; Model C: Model B + obesity, smoking status, diabetes, hypertension, heart disease, stroke, respiratory disease, liver disease, cancer, log-transformed NLR, HDL-C, LDL-C, HbA1c (or fasting glucose in HRS), and use of lipid-lowering or RAAS inhibiting medications (the latter two combined into a single “heart medication” category in the HRS cohort).
Given the limited number of renal mortality events, covariate selection for these models was guided by prior literature to minimize overfitting. Thus, a parsimonious set of clinically established predictors was included in the fully adjusted renal mortality model (model C): age, sex, race, eGFR, ACR, cardiovascular disease, diabetes, and log NLR. As a sensitivity analysis, all models were re-estimated using the log-transformed MHR to assess the robustness of the associations. Furthermore, to explore potential effect modification by CKD severity, stratified analyses by eGFR categories were conducted.
Model development based on feature selection
Analyses were repeated after performing feature selection using the Least Absolute Shrinkage and Selection Operator (LASSO) regression to address multicollinearity and improve model interpretability by shrinking the coefficients of less informative predictors toward zero. LASSO was applied exclusively to the NHANES cohort (derivation sample) to identify the most informative predictors while maintaining model parsimony. This method was chosen because it provides stable coefficient shrinkage, effectively manages multicollinearity, and performs robustly even in low-event settings. Model optimization was guided by evaluating the partial likelihood deviance and tuning the alpha parameter (Figs. 1 and 2, panels A–B). For renal mortality analyses, given the limited number of outcome events, the final regression models were intentionally kept parsimonious to adhere to recommended event-per-variable thresholds.
Fig. 1.
Feature selection for all-cause mortality based on LASSO regression. Panel A: Cross-validated Lasso regression error plot displaying the partial likelihood deviance for different values of lambda. The x-axis indicates the logarithmic transformation of lambda, and the y-axis shows the cross-validated partial likelihood deviance, with error bars representing the standard error of the partial likelihood deviance. The vertical line indicates the lambda value that minimizes the deviance, representing the optimal balance between model complexity and predictive accuracy. Panel B: LASSO regression-selected variables and their coefficients. The horizontal bar plot displays the variables chosen by LASSO regression along with their corresponding coefficients. Positive coefficients (blue bars) indicate a positive association with the outcome, whereas negative coefficients (red bars) indicate a negative association. Variables with larger absolute coefficients have a stronger influence on the model. Panel C: Forest plot showing the results of Cox regression model for all-cause mortality using variables selected through LASSO regression
Fig. 2.
Feature selection for renal mortality based on LASSO regression. Panel A: Cross-validated Lasso regression error plot displaying the partial likelihood deviance for different values of lambda. The x-axis indicates the logarithmic transformation of lambda, and the y-axis shows the cross-validated partial likelihood deviance, with error bars representing the standard error of the partial likelihood deviance. The vertical line indicates the lambda value that minimizes the deviance, representing the optimal balance between model complexity and predictive accuracy. Panel B: LASSO regression-selected variables and their coefficients. The horizontal bar plot displays the variables chosen by LASSO regression along with their corresponding coefficients. Positive coefficients (blue bars) indicate a positive association with the outcome, whereas negative coefficients (red bars) indicate a negative association. Variables with larger absolute coefficients have a stronger influence on the model. Panel C: Forest plot showing the results of Cox regression model for renal mortality based on LASSO regression
Variables selected through LASSO regression when then used to fit Cox proportional hazards models for all-cause mortality and Fine and Gray competing risk models for renal mortality, in the NHANES cohort first (Figs. 1 and 2, panels C), and subsequently validated in the HRS cohort.
All statistical analyses were conducted using R software (version 4.2). All statistical tests were two-sided, and a P value < 0.05 was considered statistically significant.
Results
Descriptive analysis
Demographic, clinical, and laboratory characteristics of the NHANES study population (overall and stratified by survival status), are reported in Table 1. In brief, the study population consisted of 5,115 participants with a mean age of 77 years, of whom 58% were women and 82% were White. Hypertension was the most common disease (67%), followed by obesity (32%), cancer (30%), heart disease (26%), and diabetes (21%). Most participants had an eGFR between 45 and 60 ml/min/1.73 m2 and a urinary ACR < 30 mg/g.
Table 1.
Characteristics of subjects included in the NHANES cohort and stratified according to all-cause mortality
| All (N = 5,115) | Alive (n = 2,566) | Dead (n = 2,549) | |
|---|---|---|---|
| Age, mean (SD) | 77 (4) | 76 (3.7) | 78.4 (4.1) |
| Female sex, n (%) | 2,962 (57.9) | 1,534 (59.7) | 1,428 (56.0) |
| White race, n (%) | 4,217 (82.4) | 2,061 (80.3) | 2,156 (84.6) |
| Smoker, n (%) | 2,465 (48.2) | 1,143 (44.6) | 1,322 (51.8) |
| Obesity, n (%) | 1,662 (32.5) | 922 (35.9) | 740 (29.0) |
| Diabetes, n (%) | 1,102 (21.5) | 528 (20.6) | 574 (22.5) |
| Hypertension, n (%) | 3,417 (66.8) | 1,701 (66.3) | 1,716 (67.3) |
| Heart disease, n (%) | 1,340 (26.2) | 534 (20.8) | 706 (31.6) |
| Stroke, n (%) | 548 (10.7) | 185 (7.2) | 363 (14.2) |
| Respiratory disease, n (%) | 623 (12.2) | 263 (10.2) | 360 (14.1) |
| Liver disease, n (%) | 177 (3.4) | 92 (3.6) | 85 (3.3) |
| Cancer, n (%) | 1,535 (30.0) | 788 (30.7) | 748 (29.3) |
| eGFR, mL/min/1.73 m2, mean (SD) | 51.2 (11.5) | 52.3 (10.3) | 50.0 (12.5) |
| eGFR < 45, n (%) | 1,328 (26.0) | 537 (20.9) | 791 (31.0) |
| ACR, mg/g, median (IQR) | 15.3 (7.2–44.0) | 12.0 (6.4–32.6) | 20.9 (8.6–57.3) |
| ACR ≥ 30 mg/g | 1,764 (34.5) | 697 (27.1) | 1,067 (41.8) |
| Total cholesterol, mmol/L, mean (SD) | 5.0 (1.2) | 4.9 (1.2) | 5.0 (1.1) |
| HDL-C, mmol/L, mean (SD) | 1.4 (0.4) | 1.5 (0.4) | 1.4 (0.4) |
| LDL-C, mmol/L, mean (SD) | 2.8 (1.0) | 2.8 (1.0) | 2.9 (1.0) |
| Triglycerides, mmol/L, mean (SD) | 1.7 (1.0) | 1.7 (1.0) | 1.7 (1.0) |
| HbA1c, mean (SD) | 6 (0.9) | 6 (0.8) | 5.9 (0.9) |
| NLR, median (IQR) | 2.3 (1.7–3.2) | 2.2 (1.6-3.0) | 2.5 (1.8–3.4) |
| MHR, mean (SD) | 0.4 (0.3) | 0.4 (0.2) | 0.5 (0.4) |
| Lipid lowering drugs, n (%) | 2,342 (45.8) | 1,303 (50.8) | 1,039 (40.8) |
| RAAS inhibitors, n (%) | 2,328 (45.5) | 1,216 (47.4) | 1,112 (43.6) |
| SGLT2 inhibitors, n (%) | 3 (0.1) | 2 (0.1) | 1 (0.03) |
Notes: ACR: albumin-to-creatinine ratio; eGFR: estimated glomerular filtration rate; HbA1c: glycated hemoglobin; HDL-C: high-density-lipoprotein cholesterol; IQR: interquartile range; LDL-C: low-density lipoprotein cholesterol; MHR: monocyte-to-HDL ratio; NLR: neutrophil-to-lymphocyte ratio; RAAS: renin angiotensin aldosterone system; SD: standard deviation; SGLT2: Sodium–Glucose Co-Transporter 2
Participants who died during follow-up were older, more commonly former or current smokers, and less commonly obese compared with survivors. They also had a higher prevalence of heart disease, respiratory disease, and stroke, and a more compromised kidney function (lower eGFR and higher ACR), as well as a lower prescription of lipid-lowering and RAAS-inhibiting medications; notably, these participants exhibited higher MHR levels than survivors (Table 1).
Overall mortality
Over a mean (SD) follow-up time of 82 (53) months, there were 2,549 deaths (49.8%), corresponding to an all-cause mortality rate of 72.7 (95% CI: 69.5–75.9) per 1,000 person-years. In survey-weighted Cox proportional hazards models, MHR was consistently associated with higher all-cause mortality in both bivariate and multivariate analyses (Table 2). Other significant positive predictors included age, ACR, smoking, hypertension, stroke, respiratory disease, and NLR; conversely, female sex, higher eGFR, and use of lipid-lowering medications were protective. These associations remained robust in sensitivity analyses using log-transformed MHR (Supplementary Table 1).
Table 2.
Survey-weighted Cox proportional hazards models showing the associations between MHR and all-cause mortality
| Model A, HR (95%CI) | Model B, HR (95%CI) | Model C, HR (95%CI) | |
|---|---|---|---|
| MHR | 1.19 (1.07–1.33) | 1.18 (1.06–1.31) | 1.32 (1.20–1.45) |
| Age | 1.12 (1.11–1.13) | 1.11 (1.10–1.12) | 1.11 (1.10–1.12) |
| Female sex | 0.75 (0.68–0.82) | 0.76 (0.70–0.83) | 0.89 (0.80–0.98) |
| White race | 1.00 (0.90–1.12) | 1.11 (1.00–1.23) | 1.03 (0.92–1.16) |
| Log ACR | - | 1.29 (1.25–1.33) | 1.24 (1.20–1.29) |
| eGFR | - | 0.99 (0.98–0.99) | 0.99 (0.99–1.00) |
| Obesity | - | - | 0.95 (0.85–1.07) |
| Smoking | - | - | 1.24 (1.13–1.35) |
| Hypertension | - | - | 1.03 (0.93–1.15) |
| Heart disease | - | - | 1.38 (1.26–1.53) |
| Diabetes | - | - | 1.09 (0.96–1.24) |
| Stroke | - | - | 1.47 (1.31–1.66) |
| Respiratory disease | - | - | 1.22 (1.05–1.41) |
| Liver disease | - | - | 1.00 (0.77–1.29) |
| Cancer | - | - | 1.07 (0.98–1.16) |
| Log NLR | - | - | 1.36 (1.23–1.49) |
| HDL-C | - | - | 1.03 (0.91–1.17) |
| LDL-C | - | - | 0.94 (0.89–0.99) |
| HbA1c | - | - | 1.05 (0.99–1.11) |
| Lipid lowering medications | - | - | 0.79 (0.71–0.89) |
| RAAS inhibitors | - | - | 1.07 (0.98–1.18) |
Notes: ACR: urinary albumin-to-creatinine ratio; eGFR: estimated glomerular filtration rate; HbA1c: glycated hemoglobin; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; MHR: monocyte-to-HDL ratio; NLR: neutrophil-to-lymphocyte ratio; RAAS: renin angiotensin aldosterone system
LASSO regression identified 17 predictors for all-cause mortality (Table 3; Fig. 1, Panels A and B): 13 with a positive association, and 4 with a negative association. When they were entered into a final multivariable Cox model (Fig. 1C), stroke, heart disease, NLR, MHR, smoking, age, and ACR emerged as significant positive predictors; significantly and positively associated with all-cause mortality; conversely, female sex, eGFR, and lipid-lowering medications were protective factors. Importantly, MHR was selected over HDL-C in both the feature selection and final models, confirming its independent and clinically relevant association with all-cause mortality.
Table 3.
Feature selection through LASSO regression to investigate the prediction of all-cause mortality
| Lasso selection | Lasso coefficient | Fully-adj HR (95%CI) | |
|---|---|---|---|
| Stroke | Yes | 0.294 | 1.49 (1.32–1.68) |
| Heart disease | Yes | 0.286 | 1.40 (1.27–1.54) |
| Log NLR | Yes | 0.283 | 1.36 (1.23–1.49) |
| MHR | Yes | 0.200 | 1.30 (1.19–1.43) |
| Smoker | Yes | 0.197 | 1.23 (1.13–1.35) |
| Log ACR | Yes | 0.188 | 1.25 (1.21–1.29) |
| Respiratory disease | Yes | 0.154 | 1.22 (1.05–1.41) |
| Lipid lowering | Yes | -0.151 | 0.80 (0.72–0.89) |
| Female sex | Yes | -0.137 | 0.89 (0.81–0.98) |
| Age | Yes | 0.093 | 1.11 (1.10–1.12) |
| White race | Yes | 0.060 | 1.04 (0.93–1.16) |
| HbA1c | Yes | 0.013 | 1.07 (1.02–1.12) |
| LDL-C | Yes | -0.032 | 0.93 (0.89–0.98) |
| eGFR | Yes | -0.007 | 0.99 (0.98-1.00) |
| Diabetes | Yes | 0.082 | - |
| Cancer | Yes | 0.064 | - |
| Obesity | No | - | - |
| RAAS inhibitors | No | - | - |
| Liver disease | No | - | - |
| Hypertension | No | - | - |
| HDL-C | No | - | - |
Notes: ACR: urinary albumin-to-creatinine ratio; eGFR: estimated glomerular filtration rate; HbA1c: glycated hemoglobin; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; MHR: monocyte-to-HDL ratio; NLR: neutrophil-to-lymphocyte ratio; RAAS: renin angiotensin aldosterone system
Renal mortality
During follow-up, 56 participants (1.1%) of the NHANES cohort died from renal causes, corresponding to a renal mortality rate of 1.5 (95% CI, 1.1–2.1) per 1,000 person-years. In Fine and Gray competing risk models, higher MHR was significantly associated with increased renal mortality (Table 4). Additional predictors included older age and higher ACR (positive associations), and higher eGFR (negative association).
Table 4.
Fine and Gray competing risk models investigating the association between MHR and renal-specific mortality in the NHANES cohort
| Model A, sHR (95%CI) | Model B, sHR (95%CI) | Model C, sHR (95%CI) | |
|---|---|---|---|
| MHR | 1.31 (1.13–1.52) | 1.32 (1.13–1.55) | 1.43 (1.12–1.82) |
| Age | 1.09 (1.02–1.16) | 1.05 (0.98–1.12) | 1.06 (0.99–1.14) |
| Female sex | 0.76 (0.46–1.26) | 0.78 (0.45–1.31) | 0.78 (0.46–1.31) |
| White race | 0.94 (0.56–1.59) | 1.47 (0.83–2.58) | 1.39 (0.77–2.53) |
| Log ACR | - | 1.35 (1.15–1.57) | 1.30 (1.10–1.53) |
| eGFR | - | 0.94 (0.92–0.96) | 0.94 (0.92–0.96) |
| Heart disease | - | - | 1.09 (0.64–1.85) |
| Diabetes | - | - | 1.69 (0.94–3.05) |
| Log NLR | - | - | 1.22 (0.77–1.95) |
Notes: ACR: urinary albumin-to-creatinine ratio; eGFR: estimated glomerular filtration rate; MHR: monocyte-to-HDL ratio; NLR: neutrophil-to-lymphocyte ratio; sHR: subdistribution hazard ratio
Feature selection using LASSO regression produced similar findings (Fig. 2; Table 5): MHR, ACR, and eGFR were identified as the most informative predictors of renal mortality.
Table 5.
Feature selection through LASSO regression to investigate the prediction of renal-specific mortality
| Lasso selection | Lasso coefficient | sHR (95%CI) | |
|---|---|---|---|
| Log ACR | Yes | 0.32 | 1.29 (1.10–1.51) |
| MHR | Yes | 0.25 | 1.37 (1.17–1.59) |
| Diabetes | Yes | 0.063 | 1.74 (1.01–2.98) |
| Age | Yes | 0.049 | 1.06 (0.99–1.14) |
| eGFR | Yes | 0.006 | 0.94 (0.92–0.96) |
| Log NLR | Yes | 0.075 | - |
| Heart disease | Yes | 0.006 | - |
| Female Sex | No | - | - |
| White race | No | - | - |
Notes: ACR: urinary albumin-to-creatinine ratio; eGFR: estimated glomerular filtration rate; MHR: monocyte-to-HDL ratio; NLR: neutrophil-to-lymphocyte ratio; sHR: subdistribution hazard ratio
Sensitivity analysis by eGFR classes
Stratified analyses revealed differential associations between MHR and outcomes according to CKD severity (Supplementary Tables 2–3). For all-cause mortality, MHR was significantly associated with death among participants with eGFR ≥ 45 mL/min/1.73 m² (HR, 1.38 [95% CI, 1.24–1.54]) but not among those with eGFR < 45 mL/min/1.73 m² (HR, 1.12 [95% CI, 0.95–1.32]). For renal mortality, the pattern was reversed. Among participants with eGFR < 45 mL/min/1.73 m², MHR showed a strong and significant association (sHR, 2.34 [95% CI, 1.49–3.68]). In contrast, among those with eGFR ≥ 45 mL/min/1.73 m², the association was attenuated and non-significant (sHR, 1.26 [95% CI, 0.85–1.87]).
Validation in the HRS cohort
Models derived from the NHANES cohort were finally validated in the HRS cohort. Baseline characteristics of the validation cohort (overall and by mortality status at 4 years) are reported in Supplementary Table 4. Demographic, clinical, and laboratory profiles were generally similar to those of NHANES participants. Over a mean (SD) follow-up of 40.8 (13.3) months, 435 participants (25.8%) died. Survey-weighted Cox regression models confirmed the independent association between MHR and all-cause mortality (Supplementary Table 5) in models A and B; although the association in the fully adjusted Model C did not reach statistical significance, the direction and magnitude of the estimates (HR: 1.98 [95% CI, 0.76–5.15]) remained consistent with those observed in less adjusted models.
Mortality due to renal causes occurred in 30 participants (1.8%). In parsimonious Fine and Gray models adjusted for age, eGFR, diabetes, heart disease, and NLR, MHR remained independently associated with renal mortality (Supplementary Table 6). The consistency of results across two independent cohorts, despite low absolute event numbers in each, strengthens the robustness of the observed MHR-renal mortality association.
Discussion
In this large and nationally representative cohort of community-dwelling older adults with CKD, we demonstrated that elevated MHR, a composite marker of systemic inflammation and lipid metabolism, was independently associated with both all-cause mortality and renal-specific mortality.
In the NHANES derivation cohort, higher MHR was associated with a 32% increased risk of all-cause mortality (HR 1.32, 95% CI 1.20–1.45) and a 43% increased risk of renal-specific mortality (sHR 1.43, 95% CI 1.12–1.82). These associations were replicated in the HRS validation cohort, where MHR showed consistent directional effects for both all-cause mortality (HR 1.98, 95% CI 0.76–5.15) and renal mortality (sHR 2.19, 95% CI 1.29–3.74), with the latter reaching statistical significance. Although wider confidence intervals in HRS reflect its smaller sample size and shorter follow-up, the similar direction and magnitude of effect estimates reinforce the robustness of MHR as a mortality-associated marker in older adults with CKD. Our findings extend prior studies linking MHR to cardiovascular events and mortality in broader populations [19, 21, 27], representing the first evaluation of MHR as a mortality predictor specifically among older individuals with CKD, a population particularly prone to multimorbidity, frailty, and accelerated cardiovascular and renal decline.
Physiologically, MHR integrates two biologically opposing elements: monocytes contribute to vascular inflammation and renal injury through cytokine release, oxidative stress, and macrophage activation [28, 29]; HDL-C exerts anti-inflammatory and vasoprotective effects through cholesterol efflux, endothelial stabilization, and inhibition of oxidative processes [30, 31]. Beyond lipid transport, HDL-C modulates circulating monocyte levels by suppressing activation via down-regulation of CD11b and limiting progenitor cell proliferation [10], thereby attenuating endothelial oxidative injury. Consequently, a higher MHR may capture both the detrimental effects of increased monocyte activity and the loss of HDL-C’s protective functions, reflecting an immune-metabolic shift toward a pro-inflammatory, atherogenic state that predisposes to adverse outcomes. The mechanistic basis for this association may involve several interlinked pathways: endothelial dysfunction and vascular calcification, promoted by monocyte adhesion, transmigration, and reactive oxygen species generation, leading to plaque instability and microvascular rarefaction [32]; impaired reverse cholesterol transport and HDL dysfunction, limiting antioxidant capacity and amplifying systemic oxidative stress [30, 33]; finally, the feedback between low HDL-C and monocyte proliferation, perpetuating a vicious cycle of decreased vascular protection and exacerbated inflammation. Importantly, the association between MHR and mortality persisted even after adjustment for traditional risk factors, renal function markers (eGFR and ACR), and other inflammatory indices such as NLR. In addition, application of regularized feature selection methods (LASSO regression) consistently identified MHR as a stable and informative predictor.
Together, these results suggest that MHR may serve as a clinically meaningful, integrative biomarker that may enhance mortality risk stratification in older adults with CKD. Its incorporation into clinical practice could improve prognostic accuracy and potentially support personalised management strategies targeting inflammatory and metabolic pathways in older patients with CKD.
Another important finding of this study is that MHR was identified as one of the key predictors of renal mortality, together with ACR and eGFR, two well-established markers of kidney damage and function. This represents a notable novelty, as previous studies have not specifically examined MHR in relation to renal-specific mortality, an outcome of particular relevance in older individuals with CKD, where competing non-renal causes of death often obscure kidney-related risks [34]. Previous studies have reported associations between higher MHR values, lower eGFR, and greater albuminuria [13, 14, 16]. Both eGFR and ACR are consistently recognized as key predictors of renal mortality. A decrease in eGFR, especially below 60 mL/min/1.73 m² and more markedly below 45 mL/min/1.73 m², correlated with an increasingly higher risk of kidney-related death [35]. Similarly, even modest increases in albuminuria were independently associated with increased mortality in patients with CKD, with risk rising continuously as ACR increases [36–38]. The coexistence of decreased eGFR and elevated albuminuria multiplies the overall risk, underscoring the synergistic impact of these abnormalities [38].
Within this framework, the inclusion of additional biomarkers such as MHR may enhance risk stratification by capturing pathophysiological processes, notably inflammation and lipid dysregulation, that are not fully reflected by traditional renal markers. The persistence of MHR as a predictor even after regularization via LASSO regression reinforces its independent and complementary value relative to ACR and eGFR. Because MHR is easily derived from standard laboratory tests, it could serve as a cost-effective adjunct to established measures such as eGFR and albuminuria, refining risk prediction rather than replacing existing diagnostic biomarkers.
Finally, stratified analyses showed that the prognostic significance of MHR may vary according to CKD severity. For all-cause mortality, MHR was significantly associated with death among participants with eGFR ≥ 45 mL/min/1.73 m², likely reflecting its role in capturing systemic cardiovascular and inflammatory risk. In contrast, for renal-specific mortality, MHR demonstrated a strong association specifically among individuals with eGFR < 45 mL/min/1.73 m², suggesting it may reflect kidney-specific pathophysiologic mechanisms that accelerate progression to end-stage kidney disease [39]. This differential pattern carries important biological and clinical implications. In moderate CKD, where cardiovascular mortality predominates and patients are more likely to “die with” rather than “die from” kidney disease, MHR appears to represent systemic immune-metabolic dysregulation underlying competing mortality risks. Conversely, in advanced CKD, MHR may more directly reflect renal inflammatory injury, driven by monocyte/macrophage infiltration promoting glomerulosclerosis and tubulointerstitial fibrosis [40, 41], combined with HDL dysfunction that compromises renal endothelial integrity and podocyte stability. Consistent with this interpretation, traditional kidney disease markers (ACR and eGFR) emerged as significant predictors of renal mortality only in participants with more compromised eGFR, supporting the concept that kidney-specific pathophysiologic pathways dominate in this population.
This study has several notable strengths. It leveraged large, nationally representative cohorts of older adults with CKD, included a long follow-up, and applied a complementary analytical framework combining traditional regression models with advanced feature selection algorithms to enhance model stability and interpretability. Findings were further confirmed in sensitivity analyses using logarithmically transformed MHR and validated in an independent cohort, strengthening the robustness of the results.
Nevertheless, some limitations should be acknowledged. First, the observational design precludes causal inference, and residual confounding from unmeasured inflammatory mediators (e.g., IL-6, oxidized LDL, HDL subfractions) cannot be excluded. Second, MHR was calculated from single baseline measurements of monocytes and HDL-C, introducing potential random misclassification and regression dilution bias; however, such bias generally attenuates associations, suggesting that the true relationship between MHR and mortality may be even stronger. Longitudinal studies with repeated measurements are warranted to determine whether dynamic changes in MHR may improve prognostic precision. Third, CKD classification was based on a single laboratory assessment without confirmation of persistence over ≥ 3 months. While this approach is consistent with major epidemiological studies [25, 42, 43], it may have diluted associations by including individuals with transient abnormalities. Fourth, the history of chronic diseases other than CKD was based on self-report, which may have introduced recall bias and affected prevalence estimates. Finally, for renal mortality, the absolute number of events was limited in both cohorts; this is consistent with other community-based CKD studies, where death from competing causes substantially exceeds progression to end-stage renal disease, particularly in older adults [34].
Conclusions
This study demonstrates that MHR is a significant predictor of both all-cause and renal-specific mortality in older adults with CKD. MHR provides prognostic information that may complement traditional renal measures such as eGFR and ACR. Our findings, strengthened by advanced feature selection methods, highlight the potential clinical utility of MHR as an accessible and integrative biomarker for risk stratification in the aging CKD population. Incorporating MHR into routine assessment could facilitate earlier identification of high-risk individuals and inform personalized management strategies aimed at improving outcomes in this vulnerable group.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
This research study uses data from the NHANES, which has been conducted by the National Center for Health Statistics and Centers for Disease Control and Prevention, and the HRS, sponsored by the National Institute on Aging and conducted by the University of Michigan.
Abbreviations
- ACR
Albumin-to-creatinine ratio
- CKD
Chronic kidney disease
- eGFR
Estimated glomerular filtration rate
- HDL
High density lipoprotein
- IQR
Interquartile range
- MHR
Monocyte-to-HDL ratio
- NHANES
National Health and Nutrition Examination Survey
- NLR
Neutrophil-to-lymphocyte ratio
- SD
Standard deviation
Authors’ contributions
Conceptualization: G. G. and L. S.; data curation: L. S.; Formal Analysis: G. G and L. S.; Investigation: G. G., L. S., A. Co., and D. S.; writing original draft: G. G., L. S., A. Co., and D. S.; writing-review & editing: G. G., L. S., F. Lu., A. Ca., L. F., L. B., C. C., L. M., M. P., F. La., E. V. S., M. G., A. Co., and D. S.; supervision or mentorship: F. La., A. Co., and D. S.
Funding
None.
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request and available publicly at https://www.cdc.gov/nchs/nhanes/index.htm and at https://hrs.isr.umich.edu/data-products.
Declarations
Ethics approval and consent to participate
NHANES is approved by the National Center for Health Statistics Research Ethics Review Board, and all participants provided written consent. The HRS and HRS-VBS were approved by the Health Sciences and Behavioral Sciences institutional review board at the University of Michigan.
Competing interests
The authors declare no conflict of interest to disclose.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Guido Gembillo and Luca Soraci contributed equally to first authorship.
Andrea Corsonello and Domenico Santoro contributed equally to last authorship.
Contributor Information
Guido Gembillo, Email: guidogembillo@live.it.
Luca Soraci, Email: l.soraci@inrca.it.
References
- 1.Wong G, Bernier-Jean A, Rovin B, Ronco P. Time for action: recognizing chronic kidney disease as a major noncommunicable disease driver of premature mortality. Kidney Int. 2024;105:1144–6. [DOI] [PubMed] [Google Scholar]
- 2.Walther CP, Gregg LP, Navaneethan SD. Cardiovascular disease risk estimates in the US CKD population using the PREVENT equation. Am J Kidney Dis. 2025;86:10–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Chesnaye NC, Ortiz A, Zoccali C, Stel VS, Jager KJ. The impact of population ageing on the burden of chronic kidney disease. Nat Rev Nephrol. 2024;20:569–85. [DOI] [PubMed] [Google Scholar]
- 4.Kadatane SP, Satariano M, Massey M, Mongan K, Raina R. The role of inflammation in CKD. Cells. 2023;12. [DOI] [PMC free article] [PubMed]
- 5.Petreski T, Piko N, Ekart R, Hojs R, Bevc S. (2021). Review on inflammation markers in chronic kidney disease. 9:182. [DOI] [PMC free article] [PubMed]
- 6.Wang Y, Zhang L, Zhang W, Tang M, Cui H, Wu X, et al. Understanding the relationship between Circulating lipids and risk of chronic kidney disease: a prospective cohort study and large-scale genetic analyses. J Translational Med. 2023;21:671. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Pavanello C, Ossoli A. HDL and chronic kidney disease. Atheroscler Plus. 2023;52:9–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lanktree MB, Thériault S, Walsh M, Paré G. HDL Cholesterol, LDL Cholesterol, and triglycerides as risk factors for CKD: a Mendelian randomization study. Am J Kidney Dis. 2018;71:166–72. [DOI] [PubMed] [Google Scholar]
- 9.Navaneethan SD, Schold JD, Walther CP, Arrigain S, Jolly SE, Virani SS, et al. High-density lipoprotein cholesterol and causes of death in chronic kidney disease. J Clin Lipidol. 2018;12:1061–e10711067. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ganjali S, Gotto AM Jr., Ruscica M, Atkin SL, Butler AE, Banach M, et al. Monocyte-to-HDL-cholesterol ratio as a prognostic marker in cardiovascular diseases. J Cell Physiol. 2018;233:9237–46. [DOI] [PubMed] [Google Scholar]
- 11.Kim D, Kim DW, Lee YH, Park SY, Song YW, Shin H, et al. Relationships between monocyte count to high-density lipoprotein cholesterol ratio and cardiovascular outcomes in patients commencing Dialysis. J Int Med Res. 2021;49:3000605211058861. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Gembillo G, Siligato R, Cernaro V, Satta E, Conti G, Salvo A, et al. Monocyte to HDL ratio: a novel marker of resistant hypertension in CKD patients. Int Urol Nephrol. 2022;54:395–403. [DOI] [PubMed] [Google Scholar]
- 13.Kanbay M, Solak Y, Unal HU, Kurt YG, Gok M, Cetinkaya H, et al. Monocyte count/HDL cholesterol ratio and cardiovascular events in patients with chronic kidney disease. Int Urol Nephrol. 2014;46:1619–25. [DOI] [PubMed] [Google Scholar]
- 14.Xu L, Li D, Song Z, Liu J, Zhou Y, Yang J, et al. The association between monocyte to high-density lipoprotein cholesterol ratio and chronic kidney disease in a Chinese adult population: a cross-sectional study. Ren Fail. 2024;46:2331614. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Lin J, Li Z, Xu J, Pan M, Yin T, Wang J, et al. Independent and joint associations of monocyte to high-density lipoprotein-cholesterol ratio and body mass index with cardiorenal syndrome: insights from NHANES 2003–2020. Lipids Health Dis. 2024;23:153. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Song J, Xu Z, Yu H, Li A, Liu Y, Jin M. (2025). Association of three composite inflammatory and lipid metabolism indicators with cardiovascular-kidney-metabolic syndrome: a cross-sectional study based on NHANES 1999–2020. Mediat Inflamm. 2025:6691516. [DOI] [PMC free article] [PubMed]
- 17.Wu D, Chen G, Lan Y, Chen S, Ding X, Wei C, et al. Measurement of cumulative high-sensitivity C-reactive protein and monocyte to high-density lipoprotein ratio in the risk prediction of type 2 diabetes: a prospective cohort study. J Translational Med. 2024;22:110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Wu D, Lan Y, Ding X, Balmer L, Li X, Wang W, et al. Imbalances in Circulating monocyte and high-density lipoprotein cholesterol exacerbates the residual risk of incident myocardial infarction beyond LDL-C: a real-life, prospective cohort study. J Translational Med. 2025;23:1433. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Pei G, Liu R, Wang L, He C, Fu C, Wei Q. Monocyte to high-density lipoprotein ratio is associated with mortality in patients with coronary artery diseases. BMC Cardiovasc Disord. 2023;23:451. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Tian S, Li Z, Bi Y, Che X, Feng A, Wang Y. Identifying pathways to cardiovascular mortality by causal graphical models and mediation analysis among hypertensive patients: insights from a prospective study. J Translational Med. 2025;23:690. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Jiang M, Yang J, Zou H, Li M, Sun W, Kong X. Monocyte-to-high-density lipoprotein-cholesterol ratio (MHR) and the risk of all-cause and cardiovascular mortality: a nationwide cohort study in the united States. Lipids Health Dis. 2022;21:30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Sonnega A, Faul JD, Ofstedal MB, Langa KM, Phillips JW, Weir DR. Cohort profile: the health and retirement study (HRS). Int J Epidemiol. 2014;43:576–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Schaeffner ES, Ebert N, Delanaye P, Frei U, Gaedeke J, Jakob O, et al. Two novel equations to estimate kidney function in persons aged 70 years or older. Ann Intern Med. 2012;157:471–81. [DOI] [PubMed] [Google Scholar]
- 24. KDIGO 2024 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int. 2024;105:S117–314. [DOI] [PubMed]
- 25.Coresh J, Selvin E, Stevens LA, Manzi J, Kusek JW, Eggers P, et al. Prevalence of chronic kidney disease in the united States. JAMA. 2007;298:2038–47. [DOI] [PubMed] [Google Scholar]
- 26.Sampson M, Ling C, Sun Q, Harb R, Ashmaig M, Warnick R, et al. A new equation for calculation of Low-Density lipoprotein cholesterol in patients with normolipidemia and/or hypertriglyceridemia. JAMA Cardiol. 2020;5:540–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Cetin MS, Ozcan Cetin EH, Kalender E, Aydin S, Topaloglu S, Kisacik HL, et al. Monocyte to HDL cholesterol ratio predicts coronary artery disease severity and future major cardiovascular adverse events in acute coronary syndrome. Heart Lung Circ. 2016;25:1077–86. [DOI] [PubMed] [Google Scholar]
- 28.Swirski FK, Nahrendorf M. Leukocyte behavior in atherosclerosis, myocardial infarction, and heart failure. Science. 2013;339:161–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Nikolic-Paterson DJ, Wang S, Lan HY. (2014). Macrophages promote renal fibrosis through direct and indirect mechanisms. Kidney Int Suppl. 2011;4:34–38. [DOI] [PMC free article] [PubMed]
- 30.Rohatgi A, Khera A, Berry JD, Givens EG, Ayers CR, Wedin KE, et al. HDL cholesterol efflux capacity and incident cardiovascular events. N Engl J Med. 2014;371:2383–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Besler C, Heinrich K, Rohrer L, Doerries C, Riwanto M, Shih DM, et al. Mechanisms underlying adverse effects of HDL on eNOS-activating pathways in patients with coronary artery disease. J Clin Invest. 2011;121:2693–708. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Libby P. Inflammation in atherosclerosis. Nature. 2002;420:868–74. [DOI] [PubMed] [Google Scholar]
- 33.Marsche G, Heine GH, Stadler JT, Holzer M. Current Understanding of the relationship of HDL Composition, structure and function to their cardioprotective properties in chronic kidney disease. Biomolecules. 2020. 10. [DOI] [PMC free article] [PubMed]
- 34.Dalrymple LS, Katz R, Kestenbaum B, Shlipak MG, Sarnak MJ, Stehman-Breen C, et al. Chronic kidney disease and the risk of end-stage renal disease versus death. J Gen Intern Med. 2011;26:379–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Wang J, Wang F, Saran R, He Z, Zhao MH, Li Y, et al. Mortality risk of chronic kidney disease: A comparison between the adult populations in urban China and the united States. PLoS ONE. 2018;13:e0193734. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Consortium CKDP. Estimated glomerular filtration Rate, Albuminuria, and adverse outcomes: an Individual-Participant data Meta-Analysis. JAMA. 2023;330:1266–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Astor BC, Matsushita K, Gansevoort RT, van der Velde M, Woodward M, Levey AS, et al. Lower estimated glomerular filtration rate and higher albuminuria are associated with mortality and end-stage renal disease. A collaborative meta-analysis of kidney disease population cohorts. Kidney Int. 2011;79:1331–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.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;375:2073–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Huimin C, Yuxin Z, Peng W, Wei G, Hong L, Na L, et al. Bioinformatics analysis and experimental validation of potential targets and pathways in chronic kidney disease associated with renal fibrosis. J Transl Med. 2025;23:387. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Zhao L, Zhang Y, Wang R, Yang J, Wu R, Wu Z, et al. Engineered macrophage membrane-coated Dihydroartemisinin nanoparticles with enhanced CCR2 expression improved symptoms in MRL/LPR mice by metabolic reprogramming of Proinflammatory macrophages. J Transl Med. 2025;23:608. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Chai S, Zeng Z, Shang J, Xu Y, Han X, Ma B, et al. Histone modification-mediated epigenetic regulation in AKI-to-CKD transition: a cell-type-specific perspective. J Transl Med. 2025;23:1346. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Go Alan S, Chertow Glenn M, Fan D, McCulloch Charles E, Hsu C. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med. 351:1296–305. [DOI] [PubMed]
- 43.Matsushita K, Coresh J, Sang Y, Chalmers J, Fox C, Guallar E, et al. Estimated glomerular filtration rate and albuminuria for prediction of cardiovascular outcomes: a collaborative meta-analysis of individual participant data. Lancet Diabetes Endocrinol. 2015;3:514–25. [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 data that support the findings of this study are available from the corresponding author upon reasonable request and available publicly at https://www.cdc.gov/nchs/nhanes/index.htm and at https://hrs.isr.umich.edu/data-products.