Sex-specific effects of mild kidney dysfunction on vascular function in midlife and older adults

Abstract

Mild kidney dysfunction (MKD) increases cardiovascular disease (CVD) risk. Vascular dysfunction, including vascular endothelial dysfunction and aortic stiffening, is a key antecedent to CVD, but the impact of MKD on vascular function in midlife/older (ML/O) adults is not established. Moreover, sex is a biological variable that influences vascular function, but whether sex modulates the effects of MKD on vascular function is unclear. Vascular endothelial function (brachial artery flow-mediated dilation, FMD_BA) and aortic stiffness (carotid-femoral pulse wave velocity, PWV_CF) were compared in 93 ML/O men and postmenopausal women with MKD (estimated glomerular filtration rate [eGFR]: 60-89mL/min/1.73m²) versus 78 ML/O adults without MKD (healthy controls; eGFR: ≥90mL/min/1.73m²) (age: 50+ years). Circulating markers of inflammation and oxidative stress were also assessed. FMD_BA was lower in men with MKD (4.0±0.3%) versus healthy controls (5.5±0.5%; P=0.0097) and correlated with eGFR (r_s=0.30, P=0.0073). There was no difference in FMD_BA between women with MKD (4.7±0.4%) and healthy controls (4.8±0.5%; P=0.86) and no relation with eGFR. PWV_CF was higher in men with MKD (9.4±0.2m/s) versus controls (8.4±0.3m/s; P=0.030) and correlated with eGFR (r=−0.34, P=0.0013). However, PWV_CF was not different between women with MKD (9.3±0.5m/s) and controls (10.1±0.4m/s; P=0.099) and not related to eGFR. The observed effects of MKD on vascular function were independent of traditional CVD risk factors and medication use. There were no differences in markers of inflammation nor oxidative stress between controls and MKD. Our findings suggest that vascular dysfunction may contribute to increased CVD risk associated with MKD in ML/O men but not postmenopausal women.

Graphical Abstract

NEW AND NOTEWORTHY

Midlife/older (ML/O) adults with mild kidney dysfunction (MKD) are at increased CVD risk compared with ML/O adults with normal kidney function. We assessed whether ML/O men and postmenopausal women with MKD exhibit vascular dysfunction compared with ML/O adults without MKD. We observed MKD-related vascular dysfunction in men but not women. Thus, vascular dysfunction may contribute to MKD-associated increases in CVD risk in men but is unlikely to contribute to the increased risk in postmenopausal women.

INTRODUCTION

Advanced chronic kidney disease (CKD) is a well-established risk factor for cardiovascular diseases (CVD), the leading cause of death in this group¹. Accumulating evidence indicates that individuals with mild, subclinical declines in kidney function, which occurs in ~50% of midlife/older (ML/O) adults², are also at an increased risk for CVD^3-5. For example, in the Framingham Heart Study Offspring cohort, community-dwelling adults with mild kidney dysfunction (MKD) assessed by estimated glomerular filtration rate (eGFR) had a higher incidence of CVD compared with individuals with normal kidney function, even when controlling for markers of kidney damage and traditional risk factors³. However, the physiological mechanisms responsible for increased CVD risk in individuals with MKD are incompletely understood.

Vascular dysfunction, characterized by vascular endothelial dysfunction and large-elastic artery (aortic) stiffening, is a key antecedent to overt CVD^6,7. Patients with advanced CKD demonstrate endothelial dysfunction^8,9 and aortic stiffening^10,11 relative to ML/O adults without CKD. However, whether individuals with MKD exhibit endothelial dysfunction and aortic stiffening in comparison with individuals without MKD is not known. Moreover, sex is a biological variable known to influence vascular function¹²; however, sex-specific effects of vascular function in the context of kidney dysfunction are understudied, due in part to a historical under-representation of women in clinical research¹². As such, it is unknown if biological sex modulates the presence of endothelial dysfunction or arterial stiffness in ML/O adults with MKD. Lastly, systemic inflammation and oxidative stress are key mechanisms of vascular dysfunction with aging and CKD^8,13 but the role of these factors in MKD-related vascular dysfunction is not well understood.

To address these research gaps, the primary goals of this study were to determine whether otherwise healthy ML/O men and postmenopausal women with MKD (eGFR: 60-89 mL/min/1.73m²) exhibit endothelial dysfunction and aortic stiffness compared with ML/O adults with normal kidney function (eGFR: ≥90 mL/min/1.73m²) independent of traditional risk factors. Because of the biomedical imperative to assess the impact of sex as a biological variable¹², we assessed if the presence of vascular dysfunction in MKD is sex-dependent. To provide initial insight into potential mechanisms, we assessed circulating markers of inflammation and oxidative stress.

MATERIALS AND METHODS

All procedures were approved by the University of Colorado Boulder Institutional Review Board. Measurements were taken at the University of Colorado Boulder at the Clinical Translational Research Center (CTRC) after written informed consent was obtained. The intent, purpose of the measurements and the risks and benefits were explained to all participants.

Participants

Participants enrolled in previously completed or ongoing studies (2015-2025) in the Integrative Physiology of Aging Laboratory were identified through the laboratory database and assigned to each group based on pre-specified criteria: controls had an eGFR of ≥90 mL/min/1.73m² and individuals with MKD had an eGFR of 60-89 mL/min/1.73m². Fasting serum creatinine was measured to calculate eGFR using the Chronic Kidney Disease Epidemiology Collaboration equation (2021 CKD-EPI, race free)¹⁴.

Participants were men and women age ≥50 years, free from overt CVD as determined by medical history, physical examination, and resting 12-lead electrocardiogram. All women were postmenopausal based on self-report of amenorrhea for ≥1 year. Potential participants were excluded if they had an eGFR <60 mL/min/1.73m², systolic (SBP) >160 mmHg or diastolic BP (DBP) >100 mmHg, cholesterol >240 mg/dL, were currently or in the past 6 months on hormone therapy (HT), had alcohol dependence or uncontrolled thyroid disease, severe obesity (body mass index [BMI] >40 kg/m²), were engaging in high levels of endurance exercise, and/or were not weight stable for at least 3 months (2.5 kg change in body mass). Participants on antihypertensive medications or other prescription medications were enrolled if their treatment regimen had been stable for >3 months.

Measurements

Measurements were performed following a ≥5 hour fast from food and caffeine, ≥24 hours abstention from alcohol, physical activity, tobacco/marijuana, and ≥48-hour abstention from over-the-counter medications and supplements. Prescribed medications were taken as normal.

Resting, seated BP was assessed in triplicate over the brachial artery using a semi-automated device (Dinamap XL, Johnson & Johnson). BMI was determined by anthropometry¹⁵. Physical activity levels (MET hours/week) were measured using the Modifiable Activity Questionnaire (MAQ)¹⁶. Serum creatinine and lipids (total cholesterol, high- and low-density lipoprotein [HDL, LDL] and triglycerides), and plasma glucose were assessed in venous blood at the Boulder Community Hospital Clinical Laboratory. The University of Colorado Anschutz Colorado CTRC assessed circulating markers of inflammation and oxidative stress, including c-reactive protein (immunoturbidimetric; Beckman Coulter), interleukins-6 and -10 (Luminex FlexMap; R&D Systems) and oxidized LDL (ELISA; Mercodia). Gynecological history was assessed in women using a questionnaire based on the Stages of Reproductive Aging Workshop (STRAW) +10 Guidelines^17,18.

Vascular measurements

Endothelial function was assessed as brachial artery flow-mediated dilation (FMD_BA)¹⁹ using high-resolution ultrasonography and simultaneous acquisition of pulsed-wave Doppler velocity as described previously¹⁹. Brachial artery diameters and velocity were captured and analyzed by Vascular Research Tools 5.10.9 (Medical Imaging Applications). Shear rate area under the curve to peak dilation (SR_AUC), SR_AUC-adjusted FMD and endothelium-independent dilation were analyzed as previously described¹⁹ in a subset with available data.

Aortic stiffness was assessed via carotid-femoral pulse wave velocity (PWV_CF) by applanation tonometry with simultaneous electrocardiogram gating of the R-wave to measure the time delay between the base of the carotid and femoral arterial pressure waves⁷. PWV_CF was calculated as the distance between arterial sites (m) divided by the pressure wave transit time (s) (Non-Invasive Hemodynamics Workstation, Cardiovascular Engineering Inc).

Statistical analyses

A 2-way ANOVA with Fisher’s LSD post-hoc testing when a sex x condition interaction was observed was used to compare continuous variables across men and women. An independent t-test (2-tailed) was used to compare female-specific factors (continuous) between women with and without MKD. A Chi-Squared Test was used to compare categorical variables. A univariate ANCOVA assessed the influence of possible confounding variables (SBP, DBP, physical activity, antihypertensive medication use, LDL cholesterol, time since menopause) on endothelial function and aortic stiffness (β-coefficients, 95% confidence intervals [CI] and P values of the regression model and adjusted mean±SEM are reported). Associations between vascular and kidney function were assessed using simple linear regression (Pearson’s correlation coefficient (r) for parametric distributions and Spearman’s correlation coefficient (r_s) for non-parametric distributions based on the Shapiro-Wilk normality test). Outliers identified with the ROUT test (Q=1%) (2 men for FMD and PWV_CF, 2 postmenopausal women for PWV_CF) were not included in final analyses. Analyses were conducted using GraphPad Prism (10.0.0; Boston, MA) and Statistical Package for Social Sciences (IBM SPSS; 29.0.0.0; Armonk, NY). Data are presented as mean±SEM unless otherwise noted. Statistical significance was set a priori at α≤0.05.

RESULTS

Participant characteristics

One hundred and seventy-one ML/O men and postmenopausal women (age: 53-86 years; eGFR: ≥60 mL/min/1.73m²) were divided into two groups based on kidney function. Seventy-eight ML/O adults (men: age: 66±1 years, eGFR: 98±1 mL/min/1.73m², n=39; postmenopausal women: age: 67±1 years, eGFR: 97±1 mL/min/1.73m², n=39) were identified as healthy controls. Ninety-three ML/O adults (men: age: 67±1 years, eGFR: 77±1 mL/min/1.73m², n=51; women: age: 69±1 years, eGFR: 75±1 mL/min/1.73m², n=42) were identified with MKD (Table 1).

Table 1: Participant characteristics.

Data are reported as mean±SD or number (n) and percentage (%) of group. ****P<0.0001, **P<0.01, *P<0.05 vs. healthy controls. ^P<0.01 vs. MKD, ^&P<0.01 vs. controls via Fisher’s LSD post hoc testing; ^#P values represent the sex * condition interaction derived from a 2-way ANOVA, independent t-test (2-tailed) or Chi-squared test. MKD, mild kidney dysfunction; eGFR, estimated glomerular filtration rate; MET, metabolic equivalent of task; LDL and HDL, low-and high-density lipoprotein; BMI, body mass index.

	Men		Postmenopausal Women
	Healthy controls	MKD	Healthy controls	MKD	P value#
Cardiometabolic factors
Number of participants	39	51	39	42	-
Age, yrs	66±7	67±7	67±7	69±6	0.91
eGFR, mL/min/1.73m2	98±5	77±8****	97±4	75±7****	0.76
Systolic blood pressure, mmHg	124±11	133±16**	130±15	124±15	0.0024
Diastolic blood pressure, mmHg	77±9	81±9	76±8	74±10^	0.037
BMI, kg/m2	26.9±3.4	26.8±3.9	25.7±4.8	25.7±4.6	0.87
Physical activity, MET hrs/wk	50±39	47±29	42±25	41±29	0.82
Total cholesterol, mg/dL	181±32	178±44	217±37	201±36	0.31
Triglycerides, mg/dL	110±61	119±57	113±42	121±52	0.86
HDL, mg/dL	54±14	51±12	67±18&	66±16^	0.66
LDL, mg/dL	106±30	104±38	127±37	113±30^	0.28
Glucose, mg/dL	97±12	97±12	95±9	94±12	0.74
Race and ethnicity, n (%)
Hispanic	0 (0)	1 (2)	0 (0)	2 (5)	0.44
Non-Hispanic White	37 (95)	46 (90)	38 (97)	39 (93)	0.60
Non-Hispanic Black	1 (3)	1 (2)	0 (0)	0 (0)	0.85
Non-Hispanic Asian	0 (0)	1 (2)	1 (3)	0 (0)	0.85
More than 1 race	1 (3)	2 (4)	0 (0)	1 (2)	0.91
Not reported	0 (0)	0 (0)	0 (0)	0 (0)	>0.99
Current medication use, n (%)
Lipid lowering	9 (23)	18 (35)	10 (26)	12 (29)	0.19
Anti-hypertensive	12 (31)	23 (45)	16 (41)	17 (40)	0.58
Thyroid regulating	3 (8)	6 (12)	11 (28)	9 (21)	0.061
Gynecological history
Number of participants	-	-	39	37	-
Time since menopause, yrs	-	-	15±8	18±10	0.14
Age at menopause, yrs	-	-	52±3	51±5	0.33
Natural menopause, n (%)	-	-	36 (92)	32 (86)	0.88
Prior therapy use, n (%)	-	-	13 (33)	20 (54)	0.80
Type, n (%)	-	-	-	-
Estradiol	-	-	4 (31)	9 (45)	0.13
Testosterone	-	-	0 (0)	0 (0)	>0.99
Progesterone	-	-	0 (0)	0 (0)	>0.99
Estradiol + progesterone	-	-	2 (15)	0 (0)	0.26
Unknown	-	-	7 (54)	11 (55)	0.93
Number of participants	-	-	17	7	-
Age at menarche, yrs	-	-	13±1	14±3	0.42
Number of births	-	-	2±2	1±1	0.28
Mother’s age at delivery, yrs	-	-	31±5	31±5	0.68
Gestation at delivery, wks	-	-	39±4	39±4	0.55

Participant characteristics were similar between ML/O adults with and without MKD within sex as age, BMI, physical activity, total cholesterol, HDL, LDL, triglycerides, glucose and medication use were not different within sex (Table 1). However, SBP and DBP were higher in men with MKD compared with controls. Gynecological history including time since menopause, age at menopause and menarche, prior HT use, and pregnancy history was also not different within women (Table 1).

Vascular endothelial function

There was a significant sex x condition interaction effect for FMD_BA (% change: P=0.045; absolute change: P=0.014). FMD_BA was lower in men with MKD (4.0±0.3%) compared with control men (5.5±0.5%, P=0.0097) when expressed as a % change (Figure 1A) and absolute change (Supplemental Table 1). There was no difference in FMD_BA between women with (4.8±0.4%) versus without MKD (4.7±0.5%; P=0.86) (Figure 1B, Supplemental Table 1). There were no differences in resting brachial artery parameters between groups (Supplemental Table 1).

Figure 1. Endothelial function in healthy controls compared with midlife/older (ML/O) men and postmenopausal women with mild kidney dysfunction (MKD).

Endothelial function assessed by brachial artery flow-mediated dilation (FMD_BA) in ML/O men and women with (n=43, 41) and without (n=37, 36) MKD (A and B). Association of kidney function, assessed as estimated glomerular filtration rate (eGFR; CKD-EPI), with FMD_BA in men (n=80) (C) and women (n=77) (D). Data reported as mean±SEM. P values derived from Fisher’s LSD post hoc (2-way ANOVA) (A and B). Correlations reported as r (Pearson’s; parametric) or r_s (Spearman’s; non-parametric) based on normality (Shapiro-Wilk) (C and D). ns, not significant.

In a subset of participants with available data, SR_AUC (n=11-20/group) did not differ based on sex nor MKD status (Supplemental Table 1). There was not a significant sex x condition interaction for SR_AUC-adjusted FMD_BA (P=0.18) (Supplemental Table 1). However, despite the lack of a significant interaction, some evidence of a similar pattern in SR_AUC-adjusted FMD_BA as non-shear rate adjusted FMD was observed (Supplemental Table 1). In men, FMD_BA was correlated with eGFR (r_s=0.30; P=0.0073) (Figure 1C). FMD_BA was not correlated with eGFR in women (r=−0.0078; P=0.95) (Figure 1D). When controlling for possible confounding effects of age, SBP, DBP, LDL cholesterol, physical activity and antihypertensive medication use, FMD_BA remained significantly different between men with versus without MKD (controls: 5.5±0.4%, MKD: 4.0±0.4%; β=−1.5; 95% CI: −2.7 - −0.3; P=0.019). There was no effect of controlling for these factors and time since menopause on FMD_BA in women (controls: 4.6±0.5%, MKD: 4.5±0.5%; β=0.14; 95% CI: −1.4 – 1.6; P=0.85).

There were no differences in endothelium-independent dilation (n=8-9/group) between men with MKD (26.7±1.3%) and controls (22.8±2.6%; P=0.45) nor women with (23.8±5.4%) and without MKD (26.6±3.4%; P=0.59). FMD_BA in these subsets was representative of the larger groups for men (controls: 6.4±1.5%, MKD: 3.8±0.6%) and women (controls: 5.1±1.1%, MKD: 4.6±1.1%).

Large-elastic artery stiffness

There was a significant sex x condition interaction effect for PWV_CF (P=0.0074). PWV_CF was higher in men with MKD (9.4±0.2 m/s) compared with controls (8.4±0.3 m/s, P=0.030) (Figure 2A). PWV_CF was not different between postmenopausal women with (9.3±0.5 m/s) and without MKD (10.1±0.4 m/s; P=0.099) (Figure 2B). In men, PWV_CF was moderately correlated with eGFR (r=−0.34; P=0.0013) (Figure 2D). PWV_CF remained higher in men with MKD versus men without MKD when controlling for age, SBP, DBP, LDL cholesterol, physical activity and antihypertensive medication use (controls: 8.5±0.3 m/s, MKD: 9.2±0.2 m/s; β=0.75; 95% CI: 0.8-14.1; P=0.029). The lack of a difference in PWV_CF between postmenopausal women with and without MKD persisted (controls: 10.6±0.6 m/s, MKD: 9.6±0.6 m/s; β=−1.0; 95% CI: −2.9 – 0.86; P=0.29) when controlling for these factors and time since menopause.

Figure 2. Aortic stiffness in healthy controls compared with midlife/older (ML/O) men and postmenopausal women with mild kidney dysfunction (MKD).

Aortic stiffness assessed by carotid-femoral pulse wave velocity (PWV_CF) in men and women with (n=49,39) and without MKD (n=38,38) (A and B). Association of kidney function, assessed as estimated glomerular filtration rate (eGFR; CKD-EPI), with PWV_CF in men (n=87) (C) and women (n=77) (D). Data reported as mean±SEM. P values derived from Fisher’s LSD post hoc (2-way ANOVA) (A and B). Correlations reported as r (Pearson’s; parametric) or r_s (Spearman’s; non-parametric) based on normality (Shapiro-Wilk) (C and D). ns, not significant.

Circulating markers of inflammation and oxidative stress

There were no differences in serum c-reactive protein, interleukin-6, nor interleukin-10 (Table 2) between men and women with and without MKD. Oxidized LDL also did not differ between groups (Table 2).

Table 2: Markers of inflammation and oxidative stress in men and postmenopausal women with and without mild kidney dysfunction (MKD).

Data are reported as mean±SEM. ^#P values represent the sex * condition interaction derived from a 2-way ANOVA. n=14-35/group.

	Men		Postmenopausal Women
	Healthy controls	MKD	Healthy controls	MKD	P value#
C-reactive protein, mg/L	1.13±0.26	1.19±0.17	1.51±0.66	1.53±0.25	0.95
Interleukin-6, pg/mL	1.21±0.15	1.35±0.20	1.02±0.21	1.30±0.4	0.72
Interleukin-10, pg/mL	0.78±0.14	0.59±0.08	0.54±0.12	0.58±0.13	0.32
Oxidized low-density lipoprotein, U/L	66.1±4.5	62.5±5.7	73.3±4.2	69.6±5.1	0.99

DISCUSSION

Here, we demonstrate sex-specific effects of MKD on vascular function in ML/O adults. We found ML/O men with MKD had lower endothelial function and greater aortic stiffness compared with control men largely independent of CVD risk factors, and that the degree of vascular dysfunction was associated with kidney function. In contrast, our findings suggest postmenopausal women do not exhibit additional MKD-related impairments in vascular function, as endothelial function and aortic stiffness were similar between age-matched women with and without MKD. These effects on vascular function are not accompanied by differences in circulating markers of inflammation and oxidative stress. Taken together, our observations suggest MKD-associated impairments in vascular function may contribute to elevated risk of CVD in men, and we provide new insight into how sex as a biological variable may modulate the effects of MKD on vascular function.

Accumulating evidence supports that MKD, prior to overt CKD⁴, increases the risk of CVD^3-5. Endothelial function is a predictor of future CV events in individuals with CKD²⁰, but independent effects of MKD on endothelial function have not been clearly demonstrated. In the current study, we identified a cohort of healthy ML/O adults without kidney dysfunction and ML/O adults with MKD, but otherwise relatively similar in terms of age, traditional CVD risk factor profile and medication use. This approach allowed us to largely isolate the effects of declines in kidney function on vascular function independent of co-morbid conditions that often accompany reduced kidney function²¹. We found that ML/O men with mild, subclinical kidney dysfunction exhibited a 1.5 unit-percentage-unit lower FMD_BA values compared with control men and that this difference persisted even when controlling for traditional CVD risk factors and antihypertensive medication use. The observed difference in endothelial function is clinically meaningful as meta-analyses show a 1% lower FMD_BA is associated with a ~15% higher CVD risk²². Endothelial function was positively related to eGFR in men, suggesting that the degree of kidney dysfunction may influence the extent of endothelial dysfunction in men. The effects of MKD on endothelial function were likely not a result of differences in smooth muscle sensitivity to NO or shear stress, as endothelium-independent dilation and SR_AUC were not different between groups. These results extend observations of impaired endothelial function in moderate to advanced CKD^8,9 and support endothelial dysfunction as a potential contributing factor to the elevated risk of CVD observed in ML/O men with MKD.

Aortic stiffness is a major risk factor for the development of clinical CVD, kidney function decline and CKD progression⁷. We observed that in comparison with control men, ML/O men with MKD exhibited a 1 m/s higher PWV_CF, which is associated with 15% higher risk of CV mortality²³. Moreover, PWV_CF was positively related to kidney function in ML/O men. Men with MKD also exhibited higher BP compared with men without MKD, which is in line with previous findings demonstrating BP is a stronger correlate of kidney dysfunction in men with CKD compared with women²⁴. Although BP influences aortic stiffness⁷, the higher PWV_CF in men with versus without MKD was independent of higher SBP and DBP in MKD and anti-hypertensive medication use. As such, our findings suggest that kidney dysfunction per se may induce aortic stiffening. Greater arterial stiffening in men with MKD may be related to potential differences in structural proteins in the arterial wall (e.g., increased collagen, decreased elastin), which has been observed in advanced CKD²⁵. Regardless, our findings identify increased aortic stiffness as an additional manifestation of vascular dysfunction that may contribute to higher CVD risk in ML/O men with MKD.

Sex is a biological variable that influences CV function¹² and risk of CVD²⁶. However, the influence of sex as a biological factor on vascular function in the context of MKD is not well established due, in part, to a historical underrepresentation of women in research¹². In contrast with our observed effects of MKD on vascular function in ML/O men, the otherwise healthy women with MKD in our study exhibited similar endothelial function and aortic stiffness compared with the postmenopausal women without MKD even after controlling for CVD risk and female-specific factors. At first glance these findings appear to be difficult to reconcile with epidemiological observations, including findings from the Framingham Heart Study Offspring cohort, which did not suggest any sex differences in the impact of MKD on CVD risk³. However, the Framingham analysis was not restricted to postmenopausal women and the interaction between MKD, menopause and/or other female-specific factors on CVD risk was not assessed. It is possible that any additional vascular dysfunction and/or increase in CVD risk associated with MKD is not detectable because of the significant adverse impact of menopause on vascular aging and CVD risk in women²⁷. This idea is consistent with the observation that the women in our study exhibited vascular dysfunction based on published normative values (i.e., FMD_BA <6%, PWV_CF >9 m/s²⁷), but more studies on the CV effects of MKD in postmenopausal women are needed. Taken together, our observations support the idea of sex-specific modulation of the effects of MKD on vascular function in ML/O adults and underscore the importance of assessing sex as a biological variable in clinical research.

Although determining the complete mechanisms responsible was beyond the scope of the present investigation, we sought to gain initial insight into whether inflammation and oxidative stress may be contributing mechanisms of the observed sex-specific effects of MKD on vascular function. Although C-reactive protein, interleukin-6 and −10, and oxidized LDL did not differ between ML/O men and women with and without MKD, these circulating markers may lack the sensitivity to reliably reflect the mechanisms at play within the vasculature as previously observed^28-30. Thus, further investigation of the cellular mechanisms involved in the observed sex-specific effects of MKD on vascular dysfunction is needed.

We would like to acknowledge that our findings may not be applicable to all racial groups, as most participants in our analysis self-reported as non-Hispanic White. Whether there are sex-specific effects of MKD on vascular dysfunction in racial groups that are at a disproportionately higher risk of CVD should be the focus of future studies. Additionally, we focused on FMD_BA (expressed as % and absolute change) and PWV_CF as the primary clinically-relevant measures in our analysis given the links between these outcomes and risk of CVD^6,7. A few supporting measures such as endothelium-independent dilation, SR_AUC and SR_AUC-adjusted FMD_BA were only available in a subset of participants. Thus, our statistical power was reduced for these analyses, and therefore, firm conclusions on the potential effects of SR_AUC and smooth muscle sensitivity to NO cannot be made. Lastly, as this was a cross-sectional analysis, the directionality of the observed relation between vascular dysfunction and MKD cannot be determined. It is possible that MKD precedes vascular dysfunction, endothelial dysfunction and/or arterial stiffness may lead to declines in kidney function or a combination of both.

Conclusions

Here, we demonstrate for the for the first time a sex-specific effect of MKD on vascular dysfunction. Specifically, ML/O men with mild declines in kidney function exhibit endothelial dysfunction and aortic stiffening relative to ML/O men without MKD, whereas postmenopausal women display no additional vascular dysfunction with MKD. Collectively, our findings suggest vascular dysfunction may be an important mechanism by which MKD increases the risk of CVD in ML/O men and provides novel insight into how biological sex might modify the relation between kidney and vascular dysfunction with advancing age prior to established CKD.

Supplementary Material

Supplemental Table 1: https://doi.org/10.6084/m9.figshare.29162030

DATA AVAILABILITY

Raw data for this study are available upon request from the corresponding author.