Examination of Sex-Specific Participant Inclusion in Exercise Physiology Endothelial Function Research: A Systematic Review

Background

To combat historical underrepresentation of female participants in research, guidelines have been established to motivate equal participation by both sexes. However, the pervasiveness of female exclusion has not been examined in vascular exercise physiology research. The purpose of this study was to systematically quantify the sex-specific prevalence of human participants and identify the rationales for sex-specific inclusion/exclusion in research examining the impact of exercise on vascular endothelial function.

Methods

A systematic search was conducted examining exercise/physical activity and vascular endothelial function, assessed via flow mediated dilation. Studies were categorized by sex: male-only, female-only, or mixed sex, including examination of the sample size of males and females. Analysis was performed examining sex-inclusion criteria in study design and reporting and rationale for inclusion/exclusion of participants on the basis of sex. Changes in proportion of female participants included in studies were examined over time in 5 year cohorts.

Results

A total of 514 studies were identified, spanning 26 years (1996–2021). Of the total participants, 64% were male and 36% were female, and a male bias was identified (32% male-only vs. 12% female-only studies). Proportions of female participants in studies remained relatively constant in the last 20 years. Male-only studies were less likely to report sex in the title compared to female-only studies (27 vs. 78%, p < 0.001), report sex in the abstract (72 vs. 98%, p < 0.001) and justify exclusion on the basis of sex (15 vs. 55%, p < 0.001). Further, male-only studies were more likely to be conducted in healthy populations compared to female-only studies (p = 0.002). Qualitative analysis of justifications identified four themes: sex-specific rationale or gap in the literature, exclusion of females based on the hormonal cycle or sex-differences, maintaining congruence with the male norm, and challenges with recruitment, retention and resources.

Conclusions

This systematic review provides the first analysis of sex-based inclusion/exclusion and rationale for sex-based decisions in human vascular exercise physiology research. These findings contribute to identifying the impact of research guidelines regarding inclusion of males and females and the perceived barriers to designing studies with equal sex participation, in an effort to increase female representation in vascular exercise physiology research.

Systematic Review Registration:

CRD42022300388.

Reliability of the hyperaemic response to passive leg movement in young, healthy women

NEW FINDINGS

What is the central question of this study? This is the first study to assess the day-to-day reliability of passive leg movement-induced hyperaemia (PLM-H), an index of lower-limb microvascular function, in young, healthy women. What is the main finding and its importance? Passive leg movement-induced hyperaemia demonstrated good day-to-day reliability, comparable to other common indices of endothelial function, supporting the use of PLM-H to assess lower-limb microvascular function in women.

ABSTRACT

Passive leg movement-elicited hyperaemia (PLM-H) provides an index of lower-limb microvascular function. However, there is currently limited information regarding the reliability of PLM-H and no reliability information specific to women. The purpose of this study was to determine the reliability of PLM-H in women on two separate days. Seventeen young, healthy women [22 ± 3 years old (mean ± SD)] participated in two identical visits including three trials of PLM. Using duplex ultrasound, PLM-H was characterized by six indices: peak leg blood flow (LBF) and vascular conductance (LVC), peak change above baseline (Δpeak) for LBF and LVC, and area under the curve above baseline (AUC) during the first 60 s of PLM for LBF and LVC. The results demonstrated good day-to-day reliability of PLM-H characterized as peak LBF [r = 0.84, P < 0.001; intraclass correlation coefficient (ICC) = 0.84; coefficient of variation (CV) = 13.2%], peak LVC (r = 0.82, P < 0.001; ICC = 0.79; CV = 14.4%), Δpeak LBF (r = 0.83, P < 0.001; ICC = 0.82; CV = 17.8%) and Δpeak LVC (r = 0.83, P < 0.001; ICC = 0.80; CV = 16.5%). Characterization of PLM as AUC demonstrated moderate day-to-day reliability: AUC LBF (r = 0.71, P < 0.05; ICC = 0.70; CV = 31.2%) and AUC LVC (r = 0.78, P < 0.001; ICC = 0.74; CV = 27.1%). In conclusion, this study demonstrates that PLM-H has good reliability as an index of microvascular function; however, characterization of PLM-H as peak, Δpeak LBF and LVC is more reliable than AUC.

The impact of exercise training on endothelial function in postmenopausal women: a systematic review

NEW FINDINGS

What is the topic of this review? The aim of this systematic review was to evaluate and summarize all published literature examining the impact of various exercise training interventions on endothelial function in postmenopausal women. What advances does it highlight? There was a moderate effect of training on macrovascular and microvascular endothelial function and just under two-thirds of studies demonstrated a significant increase in at least one measure of endothelial function in postmenopausal women. Factors including exercise intensity and duration, vessel type, clinical status, hormone therapy, and menopausal status may influence the effects of training on endothelial function in postmenopausal women.

ABSTRACT

Women experience a rapid decline in endothelial function during menopause. Therefore, it is important to explore interventions, such as exercise training, that may prevent endothelial dysfunction in postmenopausal women. The aim of this systematic review was to evaluate and summarize all published literature examining the impact of various exercise training interventions on endothelial function in postmenopausal women. Three electronic databases (MEDLINE, EMBASE and Web of Science) were used to systematically select studies related to exercise training, endothelial function and postmenopausal women. The major initial and secondary update systematic searches yielded 502 unique articles that were screened for eligibility. Thirty-five studies were included in the systematic review. Two-thirds of all studies demonstrated a group-level increase in at least one measure of endothelial function with training. Most studies investigating biomarkers of endothelial function showed improvement in at least one measured biomarker post-training. There was a moderate effect of training on both macrovascular and microvascular endothelial function in observational and randomized intervention studies. Variability in study designs, training protocols and participant characteristics make it difficult to directly compare studies. Factors including exercise intensity and duration, vessel type, clinical status, hormone therapy, and menopausal status may contribute to the inconsistent effects of training on endothelial function in postmenopausal women. Future research is needed in this population to understand the mechanisms driving inter-study and inter-individual differences in training-induced changes in endothelial function.

The acute effect of a laboratory shame induction protocol on endothelial function in young, healthy adults

NEW FINDINGS

What is the central question of this study? Shame is a form of social stress which involves internalizing social devaluations imposed by others. The aim of this study was to determine, for the first time, how acutely experiencing shame impacts endothelial function. What is the main finding and its importance? Brachial artery flow-mediated dilation, an index of endothelial function, was impaired following an intervention that acutely increased self reported shame. This occurred without increases in cortisol or tumor necrosis factor alpha receptor binding. Frequent or prolonged shame induced endothelial dysfunction could have important cardiovascular consequences.

ABSTRACT

Objective The objective of this study was to examine the impact of a shame induction protocol on endothelial function. Methods Fifteen participants (n = 7 men, n = 8 women) completed both a written shame induction and control protocol on two different experimental days. Pre- and post-protocol we assessed: 1) Endothelial function and arterial shear rate via a standard brachial artery reactive hyperemia flow-mediated dilation (FMD) test across two post-intervention time points (15 and 35-min post); 2) Perceived shame via the Experiential Shame Scale (ESS), and; 3) Cortisol and sTNFαRII (soluble tumor necrosis factor alpha receptor) through oral fluid analysis. Results Shame increased after the shame induction protocol (pre: 2.9±.6 vs. post: 3.7±.5, p<.001) but not the control protocol (pre: 3.0±.5 vs. post: 2.8±.5, p = .15) (protocol by time interaction: p<.001). When all three time points were included in the analysis, %FMD did not change over time. Considering only the lowest point, %FMD significantly decreased in response to the shame protocol (pre: 4.8±1.9 vs. post: 3.2±1.6, p<.001) but not the control protocol (4.2±1.8 vs. post: 3.8±1.5, p = .45) (protocol by time interaction: p = .035). Covariation of the shear rate stimulus for FMD did not alter the FMD results. When including both the control and shame protocol, but not the shame protocol alone, increased shame was significantly associated with decreased FMD (r = -.37, p<.046). There were no significant time by protocol interaction effects for cortisol or sTNFαRII. Conclusions Temporary increases in shame may cause transient endothelial dysfunction which, if chronically repeated, could manifest as reduced vasoprotection against atherosclerosis. This article is protected by copyright. All rights reserved.

Individual variation of follicular phase changes in endothelial function across two menstrual cycles

NEW FINDINGS

What is the central question of this study? The purpose of this study was to determine intra-individual reproducibility of follicular phase changes in endothelial function (flow-mediated dilatation) over two menstrual cycles in healthy, premenopausal women. What is the main finding and its importance? Phase changes in endothelial function were not consistent at the individual level across two menstrual cycles, which challenges the utility of interpreting individual responses over one cycle.

ABSTRACT

Evidence regarding the impact of menstrual phase on endothelial function is conflicting, and studies to date have examined responses only over a single cycle. It is unknown whether the observed inter-individual variability of phase changes in endothelial function reflects stable, inter-individual differences in responses to oestrogen (E₂ ; a primary female sex hormone). The purpose of this study was to examine changes in endothelial function from the early follicular (EF; low-E₂ ) phase to the late follicular (LF; high-E₂ ) phase over two consecutive cycles. Fourteen healthy, regularly menstruating women [22 ± 3 years of age (mean ± SD)] participated in four visits (EF_Visit 1 , LF_Visit 2 , EF_Visit 3 and LF_Visit 4 ) over two cycles. Ovulation testing was used to determine the time between the LF visit and ovulation. During each visit, endothelial function [brachial artery flow-mediated dilatation (FMD)], E₂ and progesterone were assessed. At the group level, there was no impact of phase or cycle on FMD (P = 0.48 and P = 0.65, respectively). The phase change in FMD in cycle 1 did not predict the phase change in cycle 2 (r = 0.03, P = 0.92). Using threshold-based classification (2 × typical error threshold), four of 14 participants (29%) exhibited directionally consistent phase changes in FMD across cycles. Oestrogen was not correlated between cycles, and this might have contributed to variability in the FMD response. The intra-individual variability in follicular fluctuation in FMD between menstrual cycles challenges the utility of interpreting individual responses to phase over a single menstrual cycle.

Consistency of endothelial function across two consecutive oral contraceptive pill cycles

Oral contraceptive pills (OCPs), composed of an active pill (AP; synthetic hormone) and a placebo pill (PP; synthetic hormone-free) phase, might impact endothelial function across the OCP cycle depending on the synthetic hormone composition (type and dose). Only one study has investigated very low-dose second-generation OCP users, finding impaired endothelial function in the AP versus PP phase. No studies have reported individual changes in endothelial function across OCP phases, and no studies have examined repeatability of endothelial function across multiple OCP cycles. Owing to the consistency of synthetic hormone exposure in OCP users, we hypothesized that group and individual flow-mediated dilatation (FMD) responses to the OCP phase would be consistent across two OCP cycles. Endothelial function was assessed by FMD via Duplex ultrasound in 17 very low-dose second-generation OCP users (19 ± 2 years of age) during the AP phase and PP phase for two consecutive OCP cycles. Individual responses were classified using a threshold of ±2 × typical error. There was a main effect of phase such that FMD was lower in the AP versus PP phase (P = 0.022; AP = 4.3% ± 1.3%, PP = 5.4% ± 1.4%). Threshold analysis revealed no consistent responders, and there was no relationship between Δ%FMD in cycle 1 and cycle 2 (P = 0.220; r = -0.314). Overall, these results suggest that exposure to the synthetic hormones in second-generation OCPs might be detrimental to vascular function, although this was not demonstrated to be a consistent trait-like response at the individual level over two cycles.

The 'normal' adjustment of oxygen delivery to small muscle mass exercise is not optimized for muscle contractile function

Oxygen delivery is viewed as tightly coupled to demand in exercise below critical power because increasing oxygen delivery does not increase V O 2 ${V_{{O_2}}}$ . However, whether the 'normal' adjustment of oxygen delivery to small muscle mass exercise in the heavy intensity domain is optimal for excitation-contraction coupling is currently unknown. In 20 participants (10 female), a remote skeletal muscle (i.e. tibialis anterior) metaboreflex was (Hyperperfusion condition) or was not (Control condition) activated for 4 min during both force of contraction (experimental model 1) and muscle activation-targeted (experimental model 2) rhythmic forearm handgrip exercise. Analysis was completed on the combined data from both experimental models. After 30 s of remote skeletal muscle metaboreflex activation, mean arterial blood pressure, forearm blood flow and muscle oxygenation were increased and remained increased until metaboreflex discontinuation. While oxygen delivery was elevated, the muscle activation to force of contraction ratio was improved. Upon metaboreflex discontinuation, forearm oxygen delivery and the muscle activation and force of contraction ratio rapidly (within 30 s) returned to control levels. These findings demonstrate that (a) the metaboreflex was effective at increasing forearm muscle oxygen delivery and oxygenation, (b) the muscle activation to force of contraction ratio was improved with increased oxygen delivery, and (c) in the heavy exercise intensity domain, the normal matching of oxygen delivery to metabolic demand is not optimal for muscle excitation-contraction coupling. These results suggest that the nature of vasoregulation in exercising muscle is such that it does not support optimal perfusion for excitation-contraction coupling. KEY POINTS: Oxygen delivery is viewed as tightly coupled to demand in exercise below critical power because increasing oxygen delivery does not increase the rate of oxygen uptake. Whether the 'normal' adjustment of oxygen delivery in small muscle mass exercise below critical power is optimal for excitation-contraction coupling is not known. Here we show in humans that increasing oxygen delivery above 'normal' improves excitation-contraction coupling. These results suggest that, in the heavy exercise intensity domain, the 'normal' matching of oxygen delivery to metabolic demand is not optimal for muscle excitation-contraction coupling. Therefore, the nature of vasoregulation in exercising muscle is such that it does not support optimal perfusion for excitation-contraction coupling.

Muscle contraction force conforms to muscle oxygenation during constant activation voluntary forearm exercise

NEW FINDINGS

What is the central question of this study? In electrically stimulated skeletal muscle, force production is downregulated when oxygen delivery is compromised and rapidly restored upon oxygen delivery restoration. Whether "oxygen conforming" of force production occurs during voluntary muscle activation in humans and whether it is exercise intensity dependent remains unknown. What is the main finding and its importance? Here we show in humans that force at a given voluntary muscle activation does conform to a decrease in oxygen delivery and rapidly and completely recovers with restoration of oxygen delivery. This oxygen conforming response of contraction force appears to happen only at higher intensities.

ABSTRACT

In electrically stimulated skeletal muscle, force production is downregulated when oxygen delivery is compromised and rapidly restored upon oxygen delivery restoration in the absence of cellular disturbance. Whether this "oxygen conforming" response of force occurs and is exercise intensity dependent during stable voluntary muscle activation in humans is unknown. In 12-participants (6-female), handgrip force, forearm muscle activation (electromyography; EMG), muscle oxygenation, and forearm blood flow (FBF) were measured during rhythmic handgrip exercise at forearm EMG achieving 50, 75 or 90% critical impulse (CI). 4-min of brachial artery compression to reduce FBF by ∼60% (Hypoperfusion) or sham compression (adjacent to artery; Control) was performed during exercise. Sham compression had no effect. Hypoperfusion rapidly reduced muscle oxygenation at all exercise intensities, resulting in contraction force per muscle activation (force/EMG) progressively declining over 4 min by ∼16% in 75 and 90% CI. No force/EMG decline occurred in 50% CI. Rapid restoration of muscle oxygenation post-compression was closely followed by force/EMG such that it was not different from Control within 30-sec for 90% CI and after 90-sec for 75% CI. Our findings reveal an oxygen conforming response does occur in voluntary exercising muscle in humans. Within the exercise modality and magnitude of fluctuation of oxygenation in this study, the oxygen conforming response appears to be exercise intensity dependent. Mechanisms responsible for this oxygen conforming response have implications for exercise tolerance and warrant investigation. This article is protected by copyright. All rights reserved.

Does perceived caloric and nutrient intake influence the acute effect of beverage consumption on cardiovascular function?

Our objective was to explore whether consuming the same high-fat/sugar beverage affects endothelial function differently depending on whether it is presented as "unhealthy" [accurate high calorie (kcal), fat, and sugar information displayed] versus "healthy" (inaccurate low kcal, fat, and sugar information displayed). Twenty-five, young (21 ± 2 yr), healthy, food-stress/shame-prone women completed three conditions: milkshake consumption (540 kcal, 80 g sugar, and 14 g fat) where correct, "unhealthy" nutritional information was shown to participants (milkshake condition), consumption of the same milkshake but with incorrect, "healthy" information shown to participants (100 kcal, 3 g sugar, and 4 g fat; sham-nutrishake condition), and water consumption (control condition). Pre- and postbeverage we assessed 1) endothelial function via standard brachial artery flow-mediated dilation (FMD); 2) perceived shame, stress, beverage healthiness, and harm; and 3) blood (plasma) glucose, insulin, triglycerides and oral fluid cortisol, and tumor necrosis factor-alpha (TNFα) receptor binding. Glucose, triglycerides, and insulin increased in the milkshake and sham-nutrishake conditions (P < 0.05). The milkshake was perceived as less healthy (P < 0.001) and more harmful (P < 0.001) than the sham-nutrishake. Shame, stress, oral fluid cortisol and TNFα receptor binding did not increase postconsumption. FMD decreased after the milkshake condition (pre: 7.4 ± 3.3%; post-60 min: 4.9 ± 2.9%; post-90 min: 4.5 ± 3.1%, P < 0.001) but not the sham-nutrishake (pre: 5.7 ± 2.2%; post-60 min: 5.5 ± 2.6%; post-90 min: 5.0 ± 2.4%, P = 0.43) or control conditions (pre: 7.0 ± 2.6%; post-60 min: 6.6 ± 4.1%; post-90 min: 6.0 ± 3.2%, P = 0.29). Shear rate stimulus covariation did not alter FMD results. Lower perceived beverage healthiness was significantly associated with a greater reduction in FMD (ρ = 0.36, P = 0.002). In conclusion, a high-fat/sugar milkshake reduced FMD only when presented as high in fat, sugar, and calories. This suggests that perceptions about nutritional information contribute to the impact of food intake on endothelial function and that nocebo effects could be involved in cardiovascular disease etiology.NEW & NOTEWORTHY This was the first study to investigate how perceived nutritional content influences the impact of a high-sugar/fat beverage on endothelial function. We found that a high-sugar/fat beverage only reduced endothelial function when it was presented to participants as high in calories, fat, and sugar. This suggests that perceived nutritional information contributes to the impact of high sugar and fat intake on endothelial function.