Evidence of reduced peripheral microvascular function in young Black women across the menstrual cycle

Abstract

Black women (BLW) have a higher prevalence of cardiovascular disease (CVD) morbidity and mortality compared with White women (WHW). A racial disparity in CVD risk has been identified early in life as young adult BLW demonstrate attenuated vascular function compared with WHW. Previous studies comparing vascular function between premenopausal WHW and BLW have been limited to the early follicular (EF) phase of the menstrual cycle, which may not reflect their vascular function during other menstrual phases. Therefore, we evaluated peripheral microvascular function in premenopausal WHW and BLW using passive leg movement (PLM) during three menstrual phases: EF, ovulation (OV), and mid-luteal (ML). We hypothesized that microvascular function would be augmented during the OV and ML phases compared with the EF phase in both groups, but would be attenuated in BLW compared with WHW at all three phases. PLM was performed on 26 apparently healthy premenopausal women not using hormonal contraceptives: 15 WHW (23 ± 3 yr), 11 BLW (24 ± 5 yr). There was a main effect of race on the overall change in leg blood flow (ΔLBF) (P = 0.01) and leg blood flow area under the curve (LBF AUC) (P = 0.02), such that LBF was lower in BLW. However, there was no effect of phase on ΔLBF (P = 0.69) or LBF AUC (P = 0.65), nor an interaction between race and phase on ΔLBF (P = 0.37) or LBF AUC (P = 0.75). Despite peripheral microvascular function being unchanged across the menstrual cycle, a racial disparity was apparent as microvascular function was attenuated in BLW compared with WHW across the menstrual cycle.

NEW & NOTEWORTHY This is the first study to compare peripheral microvascular function between young, otherwise healthy Black women and White women at multiple phases of the menstrual cycle. Our novel findings demonstrate a significant effect of race on peripheral microvascular function such that Black women exhibit significant attenuations in microvascular function across the menstrual cycle compared with White women.

INTRODUCTION

Black women (BLW) have a higher prevalence of cardiovascular disease (CVD) morbidity and mortality compared with White women (WHW) (1, 2). Lifestyle and sociocultural factors play a large role in CVD risk and incidence, yet the racial disparity in CVD remains even after controlling for differences in modifiable traditional CVD risk factors (3–5). This has led to various studies investigating this disparity through nontraditional risk factors that may be contributing to the discrepancy in CVD risk, incidence, and outcomes between races.

Vascular dysfunction is a nontraditional risk factor for CVD (6) and previous research demonstrates that attenuations in vascular function are observable in Black Americans compared with White Americans as early as adolescence (7). Furthermore, macrovascular function assessed by brachial artery flow-mediated dilation (FMD) has been shown to be lower in otherwise healthy premenopausal BLW compared with age-matched White men and WHW (8, 9). More recently, microvascular function has also been shown to be reduced in young adult BLW compared with WHW when assessed by skin blood flow (10) and reactive hyperemia during FMD (11). Although the exact mechanism(s) driving these observable differences in vascular function remain speculative (10), impairments in the nitric oxide (NO)-mediated pathway resulting in reduced NO bioavailability have been emphasized as a major contributing factor (9, 12), as evidenced by reduced NO substrate bioavailability (13, 14) and increased reactive oxygen species (15) in Black individuals as compared with White individuals.

The leg blood flow (LBF) response to passive leg movement (PLM) is a noninvasive assessment of peripheral microvascular function that is primarily NO-mediated (16) and the hyperemic response to PLM is largely reflective of downstream microvessel dilation (17). Reductions in microvascular function are known to precede reductions in macrovascular function, as well as can be predictive of future CVD development (18, 19). Accordingly, microvascular disturbances are more likely to be detected in younger populations consisting of otherwise healthy individuals. Taken together, PLM presents as a useful method for evaluating microvascular function in a young adult population.

Estradiol is a sex hormone that has been shown to have beneficial effects on the vasculature, mostly due to its ability to upregulate endothelial NO synthesis and release in vascular endothelial cells (20–23). Similarly, progesterone is a sex hormone that has also been shown to play a role in regulating vascular endothelial cell function via enhancing both transcriptional and nontranscriptional NO synthesis (24). In premenopausal women, estradiol, progesterone, and other sex hormones fluctuate cyclically throughout the natural menstrual cycle (25). However, the effect of fluctuations in the concentrations of these hormones on measures of NO-mediated vascular function in premenopausal women remain inconclusive with some studies demonstrating elevations in vascular function during the high hormone phases as compared with the low hormone phase (26, 27), whereas other studies did not detect any differences in vascular function across the menstrual phases (28, 29). Therefore, vascular function studies conducted in this population are traditionally restricted to the low hormone early follicular (EF) phase of the menstrual cycle to control for potential hormonal influences (10, 11). Consequently, this has led to a lack of understanding of vascular function in premenopausal women beyond the EF phase, which only encompasses the first ∼5–7 days of each menstrual cycle and may not reflect the vascular function of women for the remaining ∼20–30 days of each cycle. To our knowledge, no studies have explored the racial disparity in vascular function between premenopausal BLW and WHW beyond the EF phase of the menstrual cycle.

Therefore, the aim of this study was to evaluate the potential differences in peripheral microvascular function between WHW and BLW across three distinct phases of the menstrual cycle. We hypothesized that peripheral microvascular function as assessed by LBF responses to PLM would be elevated during the ovulation (OV) and mid-luteal (ML) phases as compared with the EF phase for both groups, and that microvascular function would be attenuated in BLW compared with WHW during each menstrual phase.

METHODS

Study Participants and Protocol

This study was approved by the Institutional Review Board at the University of Delaware (IRB Study No. 941369) and was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants. Apparently, healthy women between the ages of 18–35 that self-identified their race as either non-Hispanic White or non-Hispanic Black were recruited from the University of Delaware and the surrounding Newark, DE region. All participants were naturally cycling (i.e., not taking any form of hormonal contraceptives for at least six months before their participation) and had regularly occurring menstrual cycles. Participants were nonobese defined as a body mass index <30 kg/m², normotensive defined as a blood pressure <130/80 mmHg, nontobacco users defined as <1 cigarette in the past month, and free of any chronic diseases.

Following consenting procedures, participants completed a review of medical history to ensure they were free of overt CVD and not taking any medications or supplements that may interfere with vascular function or their natural menstrual cycle. A menstrual cycle questionnaire was used to self-report cycle length, regularity, and number of days per cycle spent menstruating. Cycle length was recorded as the number of days between the two most recent consecutive menstrual cycles for each participant. The Global Physical Activity Questionnaire (GPAQ) was used to assess self-report physical activity reported in hours per week of moderate-vigorous physical activity (MVPA). Body mass index and body fat percentage were assessed using bioelectrical impedance (Tanita TBF-300A, Arlington Heights, IL).

Participants had three peripheral vascular function testing visits that occurred during three distinct menstrual phases across one complete menstrual cycle. The menstrual phases were identified as EF, OV, and ML. The timing of the three visits was determined for each individual participant based on their cycle length, days per cycle spent menstruating, and the day of ovulation determined by an ovulation prediction test. The EF phase was defined as the first 5 days following the onset of menstruation, which was self-reported by the participants, and the EF visit was scheduled accordingly. Starting on day 7 of their cycle, participants were instructed to use a daily at-home ovulation prediction test (Clearblue Advanced Digital Ovulation Test, Geneva, Switzerland) and report their results (low fertility, high fertility, or peak fertility) to the research staff; the test was taken until the participant reported a peak fertility reading. When participants reported their first high fertility reading, their OV visit was scheduled within the next 1 to 5 days, depending on their cycle length and the day of their cycle they were currently on. The OV visit occurred on or immediately before the luteinizing hormone surge (peak fertility) detected by the ovulation kit. The ML visit was scheduled between 7- and 10-days postluteinizing hormone surge, depending on cycle length. If peak fertility was not detected by the ovulation kit, it was assumed that the participant had an anovulatory cycle and they were not included in analyses. At each visit, PLM was performed in a temperature-controlled laboratory (∼23°C). Participants were instructed to report to the laboratory fasted for ≥6 h, without caffeine, alcohol, exercise, or over-the-counter medication for ≥24 h before the visit, and to avoid taking any vitamins or supplements on the morning of the visit. Upon arrival to the laboratory, resting blood pressure was assessed (Omron 5 Series, BP7200) and intravenous blood sampling was performed for the clinical assessment of fasting blood glucose and a lipid panel (Quest Diagnostics, Inc., Philadelphia, PA). Additional serum samples were stored at −80°C for the later assessment of estradiol and progesterone concentrations. Participants then rested in the seated position for a minimum of 15 min before the start of PLM.

Passive Leg Movement Protocol

The PLM technique initiates the mechanoreflex via controlled passive movement of the lower leg at the knee joint through a 90° range of motion (90°–180°–90°) at a frequency of 1 Hz. The LBF response induced by PLM is ∼80% NO-mediated as evidenced by a N^G-monomethyl-l-arginine (l-NMMA) blockade (16), however, other mechanism(s) responsible for the remaining 20% of the LBF response have yet to be elucidated (30). PLM was performed as previously described (16) and in accordance with current guidelines (31). Briefly, in the upright seated position, the PLM protocol consisted of 60 s of baseline measurements immediately followed by a 60 s bout of passive leg flexion and extension at the knee joint. Participants were instructed to remain passive and relaxed throughout the duration of the protocol. Leg blood velocity and femoral artery diameter were recorded in the right common femoral artery, distal to the inguinal crease but proximal to the femoral artery bifurcation, via duplex ultrasound imaging (Logic e, General Electric Medical Systems, Milwaukee, WI) using a linear array ultrasound probe (12 Hz). Passive movement was achieved by a member of the research team moving the participant’s lower leg at the knee joint through a 90° to 180° range of motion at a rate of 1 Hz, while movement cadence was maintained by a metronome. Femoral artery diameter was measured during baseline, while leg blood velocity was measured throughout baseline as well as the passive movement of the leg. Throughout the duration of the protocol, the unaffected leg remained extended and fully supported.

Leg Blood Flow Calculations and Analyses

Measurements of leg blood velocity and femoral artery diameter were performed on the passively moved leg using a Logic e ultrasound system (General Electric Medical Systems, Milwaukee, WI) equipped with a linear array transducer operating at an imaging frequency of 12 MHz. Femoral artery diameter was determined at a perpendicular angle along the central axis of the scanned area and leg blood velocity was obtained using the same transducer with a Doppler frequency of 5 MHz with the probe appropriately positioned to maintain an insonation angle of 60° or less. The sample volume was maximized to vessel size and centered within the vessel based on real-time ultrasound visualization. Femoral artery diameter was measured by the investigators and mean velocity (V_mean) values [angle-corrected and intensity-weighted area under the curve (AUC)] were automatically calculated by the Doppler ultrasound system. Using arterial diameter and V_mean, LBF was mathematically calculated as follows: blood flow = V_mean π (vessel diameter/2)² × 60, where blood flow is in milliliters per minute in a custom Excel spreadsheet designed for PLM analysis.

LBF was calculated from anterograde and retrograde blood velocities achieved during PLM using continuous ultrasound Doppler imaging. Baseline LBF was calculated as a 60-s average of anterograde and retrograde blood velocities, whereas second-by-second analysis of anterograde and retrograde blood velocities was used to determine LBF during the movement phase of PLM, using the blood flow equation previously described. Peak LBF was calculated as the maximal value achieved during the first 30 s of PLM. The absolute change in LBF from baseline flow to peak flow (ΔLBF) was calculated as peak LBF − baseline LBF. Cumulative area under the curve (AUC) for values of blood flow were determined and interpreted to indicate the overall increase in blood volume achieved during movement. AUC was calculated as the sum of LBF above baseline for each second during the 60 s movement phase of PLM, according to the trapezoidal rule and using the equation as follows: Σ(y_i(x_{(i + 1)} − x_i) + (1/2)(y_{(i + 1)} − y_i)(x_{(i + 1)} − x_i)).

Measurement of Serum Estradiol and Progesterone

Estradiol and progesterone concentrations were measured in a subset of participants’ blood serum by enzyme-linked immunosorbent assays (ELISAs) (ALPCO, Salem, NH) (Estradiol ELISA Cat. No.: 11-ESTHU-E01, Progesterone ELISA Cat. No.: 11-PROHU-E01). Serum samples were run in duplicate. The sensitivity of the estradiol assay was 10 pg/mL and the sensitivity of the progesterone assay was 0.1 ng/mL. For estradiol, the intra-assay coefficient of variation (CV) was <5% and the interassay CV was <6%. For progesterone, the intra-assay CV was <12% and the interassay CV was <4%.

Statistical Analyses

Group differences in participant characteristics were analyzed using independent samples t tests. Mixed-model two-way ANOVAs were used to detect any differences in LBF responses to PLM and estradiol and progesterone concentrations between races and across the menstrual phases. Pearson correlation coefficients were used to assess associations between hormone concentrations and LBF responses to PLM across the menstrual phases, and to assess the association between the change in hormone concentrations and the change in LBF responses to PLM from EF to OV (ΔEF to OV) calculated as OV values − EF values. Normality was confirmed for PLM variables and estradiol and progesterone concentrations using Shapiro–Wilk tests. In a secondary analysis, the effect size of race on LBF responses to PLM was calculated within each phase using Hedge’s g for small sample sizes. Based on the effect size of race and clear visual differences in LBF responses between groups, independent samples t tests were performed between groups within each phase as exploratory analyses. The effect size of race was calculated as Hedge’s g using a supplementary spreadsheet (32) and all other analyses were performed using the Statistical Package for the Social Sciences (SPSS version 26.0, IBM, NY). Statistical significance was set a priori at P ≤ 0.05. Data are presented as means ± standard deviation (SD).

RESULTS

Participant Characteristics

A total of 36 participants enrolled in the study; two WHW were excluded from the analyses due to anovulatory cycles detected by the at-home ovulation prediction kit, two BLW were withdrawn due to not completing all three visits, one BLW was excluded from the final analyses due to having a blood pressure out of normal range, and five (4 WHW, 1 BLW) were unable to complete the study protocol due to a COVID-19-related research shutdown during their enrollment. A total of 26 women (15 WHW, 11 BLW) completed the study protocol and were included in the final analyses. Participant characteristics for each group are displayed in Table 1. The groups were of similar ages, body compositions, participated in comparable amounts of MVPA, and had similar cycle lengths. Vascular testing visits took place on similar days within each phase for each group. Resting diastolic blood pressure (mmHg) was higher in BLW compared with WHW (P = 0.04), but was within normal range for both groups. Fasting blood lipid and glucose profiles were obtained from 21 participants (13 WHW, 8 BLW) and were within normal ranges for healthy young adults; however, fasting blood triglycerides (mg/dL) were lower in BLW compared with WHW (P < 0.01).

Table 1.

Participant and menstrual cycle characteristics

	WHW	BLW
Participant characteristics
n	15	11
Age, yr	23 ± 3	24 ± 5
BMI, kg/m²	23.8 ± 3.2	23.7 ± 5.3
Body fat, %	29.4 ± 5.7	29.6 ± 9.1
Resting SBP, mmHg	112 ± 6	112 ± 8
Resting DBP, mmHg	69 ± 7	74 ± 6*
MVPA, h/wk	9.0 ± 6.2	8.2 ± 5.4
Total-C, mg/dL†	140 ± 20	162 ± 32
LDL-C, mg/dL†	69 ± 15	86 ± 30
HDL-C, mg/dL†	57 ± 14	66 ± 13
Triglycerides, mg/dL†	69 ± 30	41 ± 18*
Glucose, mg/dL†	86 ± 6	86 ± 5
Menstrual cycle characteristics
Cycle length, days	30 ± 2	30 ± 3
EF day of testing	3 ± 1	3 ± 1
OV day of testing	14 ± 4	15 ± 3
ML day of testing	23 ± 2	23 ± 4

Values presented as means ± SD. Independent samples t tests were used to assess group differences (n = 26, 15 WHW and 11 BLW). BLW, Black women; BMI, body mass index; DBP, diastolic blood pressure; EF, early follicular phase; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; ML, mid-luteal phase; MVPA, moderate to vigorous physical activity; OV, ovulation phase; SBP, systolic blood pressure; WHW, White women.

^*P < 0.05, significantly different from WHW; †blood chemistry values were obtained in a subset of participants (n = 21, 13 WHW and 8 BLW).

Passive Leg Movement

Second-by-second LBF responses to PLM are displayed in Fig. 1 for each race at each menstrual phase. LBF responses to PLM for each race across three phases of the menstrual cycle are displayed in Fig. 2. There was a main effect of race on ΔLBF (P = 0.01) (Fig. 2A) and LBF AUC (P = 0.02) (Fig. 2B) such that LBF responses were lower in BLW as compared with WHW. However, there was no effect of phase on ΔLBF (P = 0.69) or LBF AUC (P = 0.65) nor an interaction effect between race and phase on ΔLBF (P = 0.37) or LBF AUC (P = 0.75). There was an effect of race on baseline femoral artery diameter such that arterial diameters were smaller in the BLW (P = 0.03), however there was no effect of phase (P = 0.40) nor an interaction effect (P = 0.53) on resting femoral artery diameter; EF (WHW: 0.81 ± 0.06 cm, BLW: 0.73 ± 0.10 cm), OV (WHW: 0.81 ± 0.08 cm, BLW: 0.76 ± 0.10 cm), ML (WHW: 0.80 ± 0.08 cm, BLW: 0.75 ± 0.07 cm). There was no effect of race on baseline LBF (P = 0.14) nor an interaction effect (P = 0.72), however there was a main effect of phase on baseline LBF such that baseline LBF was higher during the OV phase as compared with the EF phase (P = 0.01); EF (WHW: 198 ± 61 mL/min, BLW: 228 ± 106 mL/min), OV (WHW: 259 ± 66 mL/min, BLW: 293 ± 113 mL/min), ML (WHW: 209 ± 101 mL/min, BLW: 268 ± 78 mL/min).

Figure 1.

Second-by-second leg blood flow responses to passive leg movement (PLM). Tracings of leg blood flow from baseline through 60-s of PLM are shown for White women (WHW, n = 15) and Black women (BLW, n = 11) during the early follicular (A), ovulation (B), and mid-luteal (C) phases of the menstrual cycle for illustrative purposes. Data displayed as means ± SD.

Figure 2.

Leg blood flow responses to passive leg movement (PLM). The overall change in leg blood flow (mL/min) from baseline to peak leg blood flow achieved during the first 30 s of movement (A) and leg blood flow area under the curve (mL) (B) are shown for White women (WHW, n = 15) and Black women (BLW, n = 11) at each menstrual phase. A mixed-model two-way ANOVA was used to assess group differences across three time points. AUC, area under the curve; EF, early follicular; ML, mid-luteal; OV, ovulation. Data displayed as means ± SD. *P ≤ 0.05.

Secondary analyses of the effect size of race (i.e., group mean differences) within each phase indicated a large effect of race on ΔLBF during the EF (g = 0.92) and OV (g = 0.89) phases, a medium effect of race on LBF AUC during the EF (g = 0.69), OV (g = 0.67), and ML (g = 0.53) phases, and a small effect of race on ΔLBF (g = 0.49) during the ML phase. Exploratory analyses using independent samples t tests to evaluate potential group differences within each menstrual phase indicated that ΔLBF was attenuated in BLW compared with WHW during the EF and OV phases (P = 0.01 and P = 0.02, respectively), but not the ML phase (P = 0.26), and LBF AUC was attenuated in BLW compared with WHW at the EF phase (P = 0.05), but not the OV or ML phases (P = 0.07 and P = 0.17, respectively).

Estradiol and Progesterone Concentrations

Measurement of serum estradiol and progesterone concentrations at each of the three visits occurred in a subset of participants (n = 15; 10 WHW, 5 BLW) largely due to participant unwillingness to consent to multiple intravenous blood samples. There was a main effect of race (P = 0.02) and a main effect of phase (P < 0.01) on serum estradiol such that estradiol was lower in BLW and estradiol was lower during the EF phase as compared with the OV and ML phases, but there was no interaction effect (P = 0.37); EF (WHW: 54.1 ± 21.8 pg/mL, BLW: 28.4 ± 12.6 pg/mL), OV (WHW: 93.8 ± 33.2 pg/mL, BLW: 49.0 ± 23.6 pg/mL), ML (WHW: 98.9 ± 37.1 pg/mL, BLW: 64.5 ± 36.2 pg/mL). There was no effect of race (P = 0.56) nor an interaction effect (P = 0.78) on serum progesterone, but there was a main effect of phase on serum progesterone such that progesterone was higher during the ML phase as compared with the EF and OV phases (P < 0.01); EF (WHW: 0.47 ± 0.38 ng/mL, BLW: 0.15 ± 0.0 ng/mL), OV (WHW: 0.82 ± 1.08 ng/mL, BLW: 0.41 ± 0.63 ng/mL), and ML (WHW: 4.13 ± 1.80 ng/mL, BLW: 3.96 ± 2.33 ng/mL).

Responses between Hormone Concentrations and LBF Response to PLM

Pearson correlation coefficient results between a subset of estradiol and progesterone concentrations and LBF responses to PLM are displayed in Table 2. Estradiol concentration was significantly positively associated with ΔLBF and LBF AUC at the ovulation phase of the menstrual cycle. In addition, the change in estradiol concentration from EF to OV was significantly positively associated with the change in ΔLBF from EF to OV. Furthermore, when all three phases were combined, estradiol concentration was positively associated with ΔLBF and LBF AUC for all participants (Fig. 3). Progesterone concentration was significantly positively associated with ΔLBF at the EF phase of the menstrual cycle.

Table 2.

Associations between hormone concentrations and PLM variables

	Estradiol, pg/mL		Progesterone, ng/mL
	r	P	r	P
EF
ΔLBF, mL/min	0.17	0.55	0.65	0.01*
LBF AUC, mL	0.04	0.89	0.34	0.21
OV
ΔLBF, mL/min	0.77	<0.01*	0.24	0.39
LBF AUC, mL	0.57	0.03*	0.02	0.94
ML
ΔLBF, mL/min	0.22	0.42	−0.35	0.20
LBF AUC, mL	0.21	0.45	0.31	0.26
ΔEF to OV
ΔLBF, mL/min	0.51	0.05*	0.01	0.98
LBF AUC, mL	0.37	0.18	0.13	0.65

Pearson correlation coefficients were used to evaluate relations between hormone concentrations and LBF responses to PLM in a subset of participants (n = 15, 10 WHW and 5 BLW). AUC, area under the curve; BLW, Black women; EF, early follicular phase; LBF, leg blood flow; ML, mid-luteal phase; OV, ovulation phase; PLM, passive leg movement; r, Pearson correlation coefficient; WHW, white women.

^*Significance (P < 0.05).

Figure 3.

Correlation analyses between estradiol concentration and leg blood flow (LBF) responses to passive leg movement (PLM). Pearson correlation coefficients were used to evaluate the relation between estradiol concentration and ΔLBF (A) and estradiol concentration and LBF area under the curve (AUC) (B) in a subset of participants [n = 15; 10 White women (WHW), 5 Black women (BLW)]. Each figure includes values from all three menstrual cycle phases, with each participant appearing three times. *Significance, P < 0.05. r, Pearson correlation coefficient.

DISCUSSION

The main finding of the present study is that peripheral microvascular function was attenuated in BLW compared with WHW across the menstrual cycle as assessed by LBF responses to PLM. Contrary to our hypothesis, microvascular function was similar across the three phases of the menstrual cycle for each race. Collectively, the novel findings from this study suggest that racial disparities in microvascular function between WHW and BLW are present across the menstrual cycle, which may contribute to racial differences in long-term CVD risk and outcomes.

Peripheral Microvascular Function Is Attenuated in BLW Compared with WHW across the Menstrual Cycle

Microvascular dysfunction precedes the development of macrovascular dysfunction and overt CVD (18, 19, 33). Detection of early disturbances in microvascular function may provide insight into the racial disparity in CVD risk, incidence, and outcomes observed between WHW and BLW. In the present study, BLW demonstrated attenuated peripheral microvascular function compared with WHW across the menstrual phases, with ΔLBF being ∼48% lower at the EF phase, ∼53% lower at the OV phase, and ∼26% lower at the ML phase. In addition, LBF AUC was ∼59% lower at the EF phase, ∼55% lower at the OV phase, and ∼33% lower at the ML phase in BLW compared with WHW. Further calculation of the effect size of race within each phase demonstrated a large effect of race on ΔLBF at the EF and OV phases and a medium effect of race on LBF AUC at all phases, with only a small effect of race on ΔLBF during the ML phase. Based on these effect sizes, in combination with visual inspection of our data, we broke from traditional statistical post hoc standards, and conducted exploratory pairwise comparisons and found that ΔLBF was significantly attenuated in BLW compared with WHW during the EF and OV phases, and LBF AUC was significantly attenuated in BLW compared with WHW at the EF phase. Conversely, statistically similar LBF responses to PLM were observed between BLW and WHW at the ML phase. To our knowledge, this is the first study to demonstrate that differences in peripheral microvascular function between young premenopausal WHW and BLW are present across the menstrual cycle, but may be largely driven by group differences at the EF and OV phases. These results extend our laboratory’s previous findings which demonstrated attenuated microvascular function in young adult BLW compared with WHW at the EF phase of the menstrual cycle as assessed by reactive hyperemia area under the curve (RH AUC) following occlusion cuff release during the brachial artery FMD assessment (11). In addition, these findings are in agreement with and extend those by Patik et al. (10) that found cutaneous microvascular function assessed by local skin heating to 39°C was attenuated in young adult BLW compared with WHW at the EF phase of the menstrual cycle (10). RH AUC following occlusion cuff release during the brachial artery FMD assessment (34), skin blood flow responses to local skin heating to 39°C (35), and LBF responses to PLM (16) have all been found to be, at least in part, mediated by the potent vasodilator NO. Taken together, these current findings further suggest that an impairment in the NO-pathway may be responsible for the attenuated microvascular function observed in BLW compared with WHW.

Estradiol binding to estrogen receptor (ER)-α on vascular endothelial cells has the ability to directly influence vascular function through the NO pathway by upregulating endothelial nitric oxide synthase (eNOS) activity (36, 37), with estradiol bioavailability being directly linked to NO production in endothelial cells (38). Furthermore, there is evidence that estradiol concentration is positively associated with vascular endothelial cell ER-α expression (39, 40). In a subset of participants in the present study, both estradiol concentration and microvascular function were attenuated in BLW compared with WHW across the menstrual cycle, suggesting that reduced estradiol bioavailability may be contributing to the blunted vascular response in these young BLW and introduces the possibility that lower estradiol concentration in BLW may be downregulating endothelial cell ER-α expression, reducing estrogen-mediated NO synthesis via this pathway. To the best of our knowledge, no study that has previously demonstrated reductions in vascular function in BLW as compared with WHW has considered estradiol concentration as potentially having influence on their findings. Correlation analyses demonstrated that both ΔLBF and LBF AUC were significantly positively related to estradiol concentration at the OV phase and when all phases were combined. In addition, the change in estradiol concentration from the EF to the OV phase was positively related to the change in ΔLBF from the EF to the OV phase, further indicating a link between estradiol concentration and vascular function.

Progesterone has also been shown to enhance both transcriptional and nontranscriptional NO synthesis at the cellular level (24). In the present study, progesterone concentration was assessed in a subset of our participants and there was no effect of race but there was an effect of phase. Visually, a large rise in progesterone was observed at the ML phase as compared with the EF and OV phases for both groups and, as described earlier, similar LBF responses to PLM were observed between groups at the ML phase. These data suggest that progesterone-mediated NO synthesis may be driving the similar LBF responses to PLM observed in the BLW and WHW at the ML phase. However, recognizing that our hormone data only consisted of a subset of participants, these data should be interpreted with that in mind and be used to generate hypotheses for future research.

Other potential mechanisms may be contributing to the attenuated LBF responses observed in BLW as compared with WHW in the present study. Limited previous research suggests that the attenuated vasodilation observed in young adult BLW compared with WHW is likely a result of endothelial dysfunction rather than smooth muscle dysfunction, as endothelium-independent vascular function seems to be preserved in this WHW population (9, 41). Specifically, the contribution of NO to vasodilator tone has been found to be reduced in Black Americans as compared to White Americans (12). Elevated reactive oxygen species (ROS) have been a proposed mechanism driving this disparity (15), however, recent evidence suggests that ROS may be less of a contributing factor to the lower endothelial function observed in young BLW than previously thought (10). In addition, although outside the scope of the current study, another potential mechanism that could explain our observed racial disparity in microvascular function is bioavailability of l-arginine, an essential substrate for NO synthesis. l-Arginine is converted into l-citrulline in the vasculature by eNOS and NO is produced as a byproduct of the conversion (42). l-Arginine supplementation improved coronary vascular endothelial function to a significantly greater extent in a small cohort of middle-age and older adult BLW as compared with WHW (13), and improved cutaneous microvascular function in a combined cohort of young adult Black men and women, but not White men and women (14) suggesting that the bioavailability of l-arginine may be a limiting factor for NO production, leading to declines in endothelial function in BLW. Future research should investigate whether differences in l-arginine bioavailability contribute to the racial disparity in NO-mediated vascular function specifically in young BLW.

Peripheral Microvascular Function Is Unchanged across the Menstrual Cycle

Contrary to our hypothesis, peripheral microvascular function was unchanged across the menstrual phases for both groups. Previous studies evaluating the influence of fluctuations in sex hormones across the menstrual cycle on overall vascular function remain inconclusive (26, 29, 43, 44). However, a recent meta-analysis suggests that peripheral microvascular function remains stable across the menstrual cycle potentially due to the microvasculature not being as susceptible to hormonal influence as the macrovasculature might be (45). Data from the present study align with this finding such that no menstrual cycle-dependent variation in microvascular function was observed in young, apparently healthy BLW and WHW.

Experimental Considerations

Predictable fluctuations in estradiol concentration occur across a regular, natural menstrual cycle. Although this allows researchers to estimate phase timing based on cycle length, the most precise method to determine menstrual phase is to directly measure daily blood hormone concentrations. When daily blood samples are not feasible, at-home digital ovulation tests, such as those used in the present study, provide highly accurate, easily interpretable estimations of ovulation (46). At-home ovulation kits detect the rises in luteinizing hormone and estradiol that occur during the transition from the EF to the OV phase to best predict the OV phase. Furthermore, a positive ovulation test is indicative of a rise in both estradiol and progesterone during the ML phase which, in combination with knowing the participant’s cycle length, allows for accurate prediction of the timing of each menstrual phase. To accurately estimate the timing of each visit and limit participant burden, all visits were conducted in the same order (EF, OV, ML) across a single menstrual cycle. Participants were not familiarized to PLM before their first visit but they were given a clear explanation of how to perform the assessment and if there was any indication of muscle activation or resistance to the movement, the PLM was repeated with additional coaching. Also considering that the study protocol did not include continuous blood pressure measurement throughout PLM, leg vascular conductance could not be calculated. Therefore, the impact of mean arterial pressure (MAP) on the LBF responses in this study and between races remains unknown; however, we do not have any evidence to suggest that the MAP response to a brief bout of PLM would differ between these young, healthy women. Moreover, there is evidence that LBF responses to PLM in women may be locally driven as evidenced by an attenuated mechanoreflex with similar LBF responses in women as compared with men (47). In addition, we did not quantify endothelium-independent vasodilation. However, previous studies have utilized sodium nitroprusside (SNP) as an exogenous NO donor and have found that endothelium-independent vascular function seems to be preserved in young premenopausal BLW despite reductions in endothelium-dependent vasodilation compared with WHW (9, 41). In addition, our sample size was greatly limited by only including women who were naturally cycling; therefore, future research should consider including women taking hormonal contraceptives.

Conclusions

These data suggest that a racial disparity in peripheral microvascular function is present across the phases of the menstrual cycle between premenopausal WHW and BLW. Findings from the present study align with previous work indicative of a racial disparity in vascular function between premenopausal WHW and BLW during the EF phase of the menstrual cycle, but newly demonstrate that this racial disparity extends beyond this low hormone phase. These findings have important clinical implications as CVDs are the leading cause of death in women, with BLW having the greatest prevalence of CVD morbidity and mortality (1). Ultimately, this study has further elucidated the vascular function profile of young, premenopausal BLW and emphasizes the need for research evaluating potential mechanism(s) contributing to the observed impairment in NO-mediated vascular function across the menstrual cycle.

GRANTS

This work was supported in part by National Institute of General Medical Sciences Grant P20GM113125.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

M.M.W. and M.A.W. conceived and designed research; M.N.D., E.K.H., A.E.H., and A.V.K. performed experiments; M.N.D., E.K.H., and A.E.H. analyzed data; M.N.D., A.V.K., S.M.R., M.M.W., and M.A.W. interpreted results of experiments; M.N.D. prepared figures; M.N.D. drafted manuscript; E.K.H., F.R.B., A.V.K., S.M.R., M.M.W., and M.A.W. edited and revised manuscript; M.N.D., E.K.H., F.R.B., A.E.H., A.V.K., S.M.R., M.M.W., and M.A.W. approved final version of manuscript.