Influence of sex on the reliability of cerebral blood velocity regulation during lower body negative pressure and supine cycling with considerations of the menstrual cycle

Objective. To evaluate sex differences in the reliability of absolute and relative cerebral blood velocity (CBv) during concurrent supine cycling with lower body negative pressure (LBNP). Approach. A total of 19 participants (11 females; aged 20–33 years) completed five testing sessions, occurring on 7 d intervals. Visit 1 was a maximal-ramp-cycle test to ascertain peak CBv wattage. During visits 2–5, supine cycling protocol occurred at individualized peak CBv wattages with progressive decreases in LBNP from 0 to −20, −40, −60, −70, and −80 Torr. Menstrual cycle day was self-reported via the Rhinessa Women’s Questionnaire. Transcranial Doppler ultrasound insonated bilateral middle cerebral artery velocity (MCAv). Two-way ANOVA assessed potential day- and sex-differences at each LBNP stage. Reliability was determined using intraclass correlation coefficients (ICC) and coefficient of variation (CoV). Main results. For all physiological measures, no main-effects were present for day- or interaction-terms (p > 0.067; negligible-to-small effect sizes), while sex differences were noted for MCAv, blood pressure, and heart rate (p < 0.046). Across visits, males and females displayed excellent and good-to-excellent levels of reliability for MCAv metrics, respectively (ICC range: 0.745–0.999; CoV range: 0.33%–9.90%). Significance. During the current investigation, both relative and absolute CBv demonstrated high reliability in both male and female participants during a supine LBNP cycling protocol. An exploratory analysis revealed increased variance was found in female participants dependent on contraceptive use. Despite this, results indicate future LBNP studies may include females at any menstrual cycle stage.


Background
The ability of the brain to regulate cerebral blood flow (CBF) and perfusion during transient alterations in systemic blood pressure is known as dynamic cerebral autoregulation (dCA) (Willie et al 2014, Brassard et al 2021).When dCA is intact, the cerebrovasculature responds to mitigate both hypo-and hyper-perfusion, in order to attempt to maintain homeostasis (Willie et al 2014).However, an impairment of this regulatory mechanism has been linked to various clinical diseases and/or disorders (e.g.hypotension, exercise intolerance, orthostatic intolerance) (Claassen et al 2021).For example, orthostatic intolerance, a former of transiently impaired dCA, occurs when an individual experiences light-headedness, dizziness, and other symptoms, stemming from a change in posture as a result of the subsequent hypoperfusion (Stewart et al 2018).As the brain has minimal nutrient stores, it heavily relies on consistent CBF to maintain function (Willie et al 2014).If the brain is under-perfused, an individual will often experience syncope and move towards a supine posture.This occurs as the change of body posture increases perfusion of the brain, by placing it at the height of the heart to maintain sufficient CBF/perfusion (Stewart et al 2018).
Orthostatic intolerance has widely been quantified through tilt-table methodology, which results in transient fluid shifting, lower extremity blood pooling, and cerebral hypoperfusion (Teodorovich and Swissa 2016, Jardine et al 2018, Stewart et al 2018).From these results, one can assess the temporary disruption and CBF response to a given degree of tilt change usually within the first 30 s following the postural change (Teodorovich and Swissa 2016).However, with the use of a tilt table, one would be unable to produce a controlled steady-state orthostatic stressor over varying frequencies, as the tilt test relies on rapid and temporary orthostatic challenges (Teodorovich and Swissa 2016).Meanwhile, through the application of lower body negative pressure (LBNP), a steady-state, controlled shift of blood from the upper into the lower extremities can be employed (Goswami et al 2019), allowing for a prolonged orthostatic challenge.
Progressive LBNP has been performed previously to simulate blood loss and has demonstrated the ability to cause symptoms of presyncope (Rickards et al 2015, Rosenberg et al 2021).However, despite the increasing knowledge surrounding the functioning of the cerebrovascular system while undergoing LBNP, a limitation of previous research is the lack of inclusion of female participants (Kennedy et al 2022).The main reason cited to justify the exclusion of females is the unknown impact phase of the menstrual cycle and oral contraceptives have on CBF regulation.The fluctuating ovarian hormones may impact the autoregulatory capabilities as estrogen is thought to decrease cerebral vascular tone while increasing cerebral blood flow (Shamma et al 1992, Krause et al 2006).Therefore, a paucity of sex difference studies exist with respect to the assessment of orthostatic/ autoregulator capabilities during supine cycling.
Several studies to date have combined LBNP with a concurrent cycling exercise (Bonde-Petersen et al 1984, Smirl et al 2016, Linjie et al 2020).Typically, with cycling exercise, CBF follows a parabolic relationship increasing upwards of 20% during mild-to-moderate-intensity exercise before returning to baseline levels with high-intensity/maximal exercise (Ogoh et al 2009, Smith andAinslie 2017).The peak increase in CBF occurs at ∼50%-80% of an individual's maximal oxygen uptake (VO 2max ) due to an elevation in circulating carbon dioxide, which has been termed hyperpnea-induced vasodilation (Ogoh et al 2009, Smith andAinslie 2017).As previously mentioned, LBNP can be used to stimulate blood loss (Rickards et al 2015, Rosenberg et al 2021), therefore acting as a modality to challenge the cerebrovasculature by driving blood from the upper extremities towards the lower extremities.
Therefore, this study sought to delineate the reliability of absolute and relative cerebral blood velocity (CBv) during a progressive steady-state LBNP supine cycling protocol in males and females.Moreover, exploratory analyses were conducted on females not using any contraceptives.It was hypothesized both males and females would display high levels of reliability across the four visits; however, the reliability would be slightly less in females due to the hormonal fluctuations associated with the menstrual cycle.

Ethical approval
Ethical approval for this study was provided by the University of Calgary Conjoint Health Research Ethics Board (REB20-1662 andREB 20-2112).During the first visit to the lab, all protocols and equipment were thoroughly explained, and each participant provided written informed consent.All procedures were followed according to institutional guidelines.

Participants and study design
This investigation utilized a repeated measures cohort of 19 healthy adults (11 females and 8 males) between the ages of 20-33.Females were an average age of 23.0 ± 3.5 years and had a body mass index of 24.5 ± 2.5 kg m −2 .Females were eumenorrheic for at least 6 months prior to testing, defined as having a menstrual cycle every 21-35 d (average 27 ± 3 d), with a maximum three day difference between cycle lengths.Of the 11 female participants, two were on birth control, two had intrauterine devices, and six were not using any type of contraceptive.Females were excluded if they were pregnant or attempting to become pregnant.Further, none of the females were post-partum, breastfeeding, or using hormonal replacement therapy.Males were an average age of 24.2 ± 1.6 years and had a body mass index of 25.3 ± 5.3 kg m −2 .Participants were considered for the study if they were free from cardiorespiratory and/or cerebrovascular conditions (and any comorbidities) and were at least 6 months removed from their last concussion (Lapointe et al 2021).Prior to all visits, participants were instructed to abstain from caffeine, alcohol, smoking, vaping, and exercise for a minimum of 8 h prior to the start of data collection (Burma et al 2020).

Experimental protocols
All testing sessions took place in the Cerebrovascular Concussion Laboratory at the University of Calgary, which is located 1,111 metres above sea level.Volunteers were required to visit the laboratory a total of five times.On the first day, participants' sex, age, height, and weight were collected.Participants then completed a maximal stepwise supine-cycling test without the application of LBNP, which determined the individualized wattage associated with hyperpnea-induced vasodilation for each participant (Smith and Ainslie 2017).Each stage lasted one minute in duration, with the wattage being increased every minute based on the participant's body weight and sex, as seen in the following formulas (Miutz et al 2023):

kilogram kg of body weight
Females: 0.15 kg of body weight.
During the exercise tests, participants maintained a cadence of 60-80 revolutions per minute (rpm).Researchers terminated the test when participants reached volitional exhaustion or the inability to maintain at least 60 rpm.From this test, CBv was extracted at each stage to determine the wattage with the greatest increase in CBv.The following formula was then used to calculate the individual wattage for the second to fifth visits: -LBNP wattage peak CBV wattage 2.5x wattage increase .
The wattage was reduced from the peak stage for several reasons.First, as the duration of the LBNP protocol could last a total of 42 min in length, the wattage was reduced slightly to ensure each volunteer would have minimal fatigue and be able to sustain the wattage throughout the duration of the full LBNP protocol.Second, the peak associated with hyperpnea-induced vasodilation peak typically occurs at a respiratory exchange ratio of ∼1.0 (Beaver et al 1986, Ogoh et al 2009).To safeguard the likelihood of participants crossing their anaerobic threshold, the wattage was reduced by a few stages.Finally, as CBv follows a parabolic curve during cycling exercise, the calculated stage would fall within the plateau (Smith and Ainslie 2017), thus producing a nearmaximal dilatory response.
The second to fifth visits each occurred a week apart, resulting in four testing sessions being completed over the span of a month.At the start of each visit, females completed questions 1-11 on the Rhinessa Women's Questionnaire (Rezaie-Chamani et al 2019) to obtain information about each participant's menstrual cycle.Following experimental setup (figure 1) participants completed the same LBNP protocol over these visits.Baseline data were collected from participants in a supine position, as all exercise was completed in this position.Following baseline, participants began cycling at zero negative pressure for a duration of six minutes, eliciting an expected elevation in CBv (Smith and Ainslie 2017).Subsequently, decreasing intensities of LBNP were applied for six minutes at stages of −20, −40, −60, −70, and −80 Torr.If participants successfully completed stage −80 Torr, the test was terminated, and the negative pressure was released.If an individual did not reach −80 Torr, testing was terminated if participants voluntarily ceased the test and/or the participants began to display pre-syncope signs (e.g.cardiac collapse) (Kay and Rickards 2015).All symptoms from the negative pressure dissipated within 5 min of LBNP termination.

Instrumentation
Participants performed all supine cycling exercises inside a custom-built LBNP chamber, containing a supine cycling ergometer (figure 1).To ensure an airtight seal with the chamber was created, a fitted vertical wooden slide and placed in the opening of the chamber, around the participant's iliac crest.Prior to beginning the cycling protocol, a plastic sleeve with a neoprene band was secured around the waist.Transcranial Doppler ultrasound (TCD) (DWL USA, Inc., San Juan Capistrano, CA, USA), was used to measure CBv responses during both the maximal exercise test without LBNP changes, as well as during the four LBNP supine cycling protocol visits.The two 2 MHz ultrasound probes were placed over the transtemporal acoustic windows to continuously quantify CBv in the left and right middle cerebral arteries (MCA).During the initial visit, researchers located the vessels and confirmed the correct vessels based on the expected signal depth, velocity, and signal responses to carotid compressions (Willie et al 2011).Individual depths and velocities were recorded for each participant to be used for the subsequent visits to ensure maximal reliability (Willie et al 2011).Moreover, all testing sessions occurred in the presence of a highly trained sonographer to further augment the between-day reliability, ensure the same vessel was insonated, and ensure similar velocities were found.Once the correct vessels were found, the probes were locked in the correct position with a TCD headframe (DWL USA, Inc., San Juan Capistrano, CA, USA).A finger photoplethysmography device with a height correction unit (Finometer NOVA; Finapres Medical Systems, Amsterdam, The Netherlands) was used to measure beat-to-beat arterial BP.A 3-lead electrocardiogram was used to capture individual PQRST waveforms and heart rate was calculated by the intervals between the R-waves (FE 231 BioAmp; AD Instruments, Colorado Springs, CO, USA).Participants also wore a heart rate monitor, placed below the sternum as another way of capturing heart rate (Polar H10, Kempele, Finland).Breath-to-breath end-tidal pressure of carbon dioxide (P ET CO 2 ) values were measured with a mouthpiece and inline gas analyzer (ML206; AD Instruments).All data were simultaneously sampled at 1000 Hz via an analog-to-digital converter (PowerLab 16/35 PL3516; AD Instruments, Colorado Springs, CO, USA).Data were stored for analysis with commercially available software (LabChart Pro Version 8, AD Instruments).

Data processing
As stated, MCA velocities were averaged across each 60 s stage during the stepwise exercise test, to determine the wattage associated with peak CBv.The sampling frequency for all physiological variables (i.e.MCA velocities, heart rate, blood pressure, P ET CO 2 , and respiration rate) were collected at 1000 Hz, thereafter, the mean for all traces were taken as the sum of all data points collected from each trace.Despite each stage being six minutes in duration, the first few minutes of each transition involve a physiological adjustment period to enable the body time to adapt and stabilize after each increase in negative pressure (Smith et al 1994).Therefore, only data collected in the last two minutes of each stage were extracted and used for analysis.All physiological variables were extracted as absolute values; however, MCA velocity was also calculated as a relative percent change from baseline.Bilateral MCAs were insonated; however, three participants had one vessel drop out due to technological limitations of the TCD associated with exercise.Therefore, bilateral recordings were collected for 16 participants, while unilateral were collected for the final three (two lefts and one right).Data were filtered and Figure 1.This figure provides information regarding the protocol used for the maximal wattage test, the lower body negative pressure (LBNP) exercise protocol as well as the different modalities used to capture physiological variables important to the research study.(A) provides a photo of a participant during the LBNP cycling protocol and demonstrates the instrumentation used to measure physiological changes throughout the exercise test.Of importance, a 3-lead electrocardiogram was used to capture heart rate, a finger photoplethysmography device provided information regarding blood pressure and heart rate, transcranial doppler ultrasound was used to insonate the left and right middle cerebral artery velocities (MCAv), and a capnography device provided end-tidal carbon dioxide (CO2) levels.(B) demonstrates the supine maximal wattage cycling test, which was used to find the individualized wattage associated peak cerebral blood velocity.The relationship between end-tidal CO2 and MCAv throughout the increasing stages.Stages increased in wattage every minute based on the participant's sex and weight.(C) demonstrates the protocol used for the LBNP supine cycling exercise test.Stages lasted six minutes each, whereby after the pressure would progressively decrease in the chamber with every stage, up to a maximum of −80 millimeters of mercury (Torr).The protocol also demonstrates the relationship between stages, MCAv, and mean arterial pressure (MAP).Middle cerebral artery (MCA), centimeter (cm), second (s), watt (W), cerebral blood velocity (CBv).
cleaned with interpolated R-R data points to correct for ectopic beats and/or misaligned r-spikes, which occurred in less than 0.1% of the data.

Independent and dependent variables
The main independent variable throughout this protocol was the day of testing, while the main dependent variables were the absolute and relative change in MCAv to the subsequent decreases in LBNP.Relative MCA velocities taken throughout the different pressure stages were calculated in comparison to each volunteer's daily baseline CBv values.Female participants did not begin the protocol on a certain day or stage of their cycle; however, they recorded their associated cycle day during each LBNP protocol testing day (visits 2-5) using the Rhinessa Woman's Questionnaire.Each visit was spaced out a week apart and exercise tests were scheduled at the same time of day for each visit.
Statistical analyses R-Studio (Version 2022.7.1.554)(R Core Team.R: A, 2020) was used to perform all statistical analyses.To assess the reliability between the four visits, intraclass correlation coefficients (ICC) and coefficient of variation (CoV) metrics were used.Inferences for both metrics were computed based on their estimates and the associated 95% confidence intervals based on standards set forth within the literature (Koo and Li 2016).More specifically, the ICC were computed using a 2-way mixed-effect model, absolute agreement, and a mean-rating (k = 3) (Koo and Li 2016).Thresholds for the ICC were set a priori at <0.50 (poor), 0.50-0.75(moderate), 0.75-0.90(good), and >0.90 (excellent) (Koo and Li 2016).The ICC and CoV values were calculated for the right and left MCAv vessels independently, as well as the combination of the two vessels for the 16 participants with bilateral recordings.These were calculated for both males and females and were additionally stratified for females using no contraceptives.This was an exploratory analysis to better understand the influence of the menstrual cycle on the outcomes of interest.Within-subject, between-day CoV values were produced as the quotient of the standard deviation and the mean CBv values between the four testing sessions.The 95% confidence intervals for these calculations were computed using bootstrap analysis with 10 000 resamples.Thresholds for the CoV were set a priori at <5% (excellent), 5%-10% (good), 10%-20% (acceptable), and >20% (unacceptable) (Hopkins 2000).Likewise, the CoV calculations were produced for both females and males but additionally stratified based on contraceptive usage in females.
Based on the reliability analyses, the two vessels were averaged across each day, creating one MCAv value (n = 16), while the unilateral value was used for the remaining 3 participants.To determine if differences existed between days at each stage and between sexes, a 2 × 4 analysis of variance (ANOVA) was utilized for all absolute and relative CBv values and absolute physiological parameters.To understand the level of variance homogeneity between groups for each variable, Levene's tests were run, of which no comparison violated this assumption.Previous modelling has demonstrated the use of ANOVAs to be highly robust in the face of violated assumptions (e.g.unequal variances, differences in sample sizes between groups) (Blanca et al 2017).If a significant main effect was present, Tukey's post-hoc pairwise comparisons were conducted to assess which groups were different.Effect sizes were calculated for the ANOVAs and Tukey comparisons using generalized eta squared (η 2 G ) and Cohen's d effect sizes, respectively.For the former, thresholds of <0.02 (negligible), 0.02-0.14(small), 0.14-0.26(moderate), and >0.26 (large) were used.Cohen's d thresholds of <0.20 (negligible), small (0.20-0.50), moderate (0.50-0.80), and large (>0.80) were used.Alpha was set a priori at 0.05, with inferences being made from both p-value and effect sizes.

Physiological data
Table 1 displayed the raw data for females, males, and the combination of both sexes for MCAv, P ET CO 2 , respiration rate, MAP, SBP, DBP, and heart rate metrics across each of the four visits.Additionally, table 1 also highlights the CoV for all physiological parameters stratified by biological sex.Individual absolute and relative MCAv data for females are displayed in Supplemental figures 1 and 2, respectively across the self-reported day of their menstrual cycle.

Between-day reliability
The absolute MCAv ICC and associated 95% confidence intervals displayed moderate-to-excellent levels of reliability for males, females, and females using no contraceptives for the right and left MCAv; however, when combined, all groups displayed excellent reliability (figure 2).Conversely, the relative MCAv ICC and 95% confidence intervals largely displayed good-to-excellent levels of reliability for each vessel independently, as well as the combination of both for all groupings (figure 2).When controlling for each individual's absolute value, the Table 1.Cardiorespiratory variables collected during a progressive lower body negative pressure (LBNP) cycling protocol.Each cardiovascular variable displays raw data for all participants (n = 19), and females (n = 11) and males (n = 8) independently.Data are displayed for each stage of the LBNP protocol and test statistics combined to show a between day, between sexes, and a day-by-sex interaction.These were assessed using a two-factorial repeated measures analysis of variance with generalized eta-squared effect sizes (n 2 G ). Coefficient of variation values were also computed to show to levels of within-subject between-day variability.CoV for all participants was good-to-excellent for both absolute (figure 3) and relative (figure 4) MCAv parameters across all stages.Moreover, all physiological metrics displayed excellent-to-acceptable levels of CoV within both absolute (table 1) and relative domains.Finally, figures S1 and S2 display each participant's raw absolute MCAv and relative MCAv data, respectively across their four visits and the associated day of their menstrual cycle.

Absolute physiological and physiological parameters main effects
No main effects of day were present for all absolute physiological variables: MCAv (F  The data displays differences in both males (n = 8), all females (n = 11), and females not on contraceptives were stratified as an exploratory analysis (n = 6) across all stages of the lower body negative pressure (LBNP) protocol.ICC values were attained through a 2-way mixed-effect model, absolute agreement, and a mean-rating (k = 3).Thresholds were set at <0.50 (poor; red), 0.50-0.75(moderate; orange), 0.75-0.90(good; yellow), and >0.90 (excellent; green).Males exhibited excellent levels of reliability across all stages for both relative and absolute MCAv, while females displayed excellent levels of reliability for absolute MCAv and good-toexcellent for relative MCAv.ICC values were attained through a 2-way mixed-effect model, absolute agreement, and a mean-rating (k = 3).Middle cerebral artery (MCA), centimeters (cm), second (s), percent (%), stages 0 to −60 represent pressure changes in millimeters of mercury (Torr).

Discussion
This study sought to understand sex differences in the reliability of an individualized supine exercise test during progressive decreases in stepwise LBNP.The exploratory analysis stratified female participants into contraceptive types to determine how hormonal fluctuation modifies and/or confounds the response of the MCAv to the above test.The main findings were: (1) the between-day reliability for physiological measures was excellent for males and good-to-excellent for females; and (2) females not using contraceptives displayed goodto-excellent reliability within this small sample, highlighting hormonal fluctuation likely has some but minimal impact on the protocol utilized.The collective findings of this study in conjunction with previous reports (Favre andSerrador 2019, Korad et al 2022), highlights the exclusion of females from physiological studies due to the menstrual cycle appears to be inappropriate, as despite testing occur over a month, females displayed good-toexcellent levels of reliability.

Comparisons with previous literature
There is a lack of studies investigating reliability measures of MCAv in response to exercise.However, a prior investigation by Billinger and colleagues (Billinger et al 2017) investigated the MCAv responses to moderateintensity exercise in 10 healthy subjects.Of importance, using CoV measures, the authors observed the MCAv values for baseline, response parameters, and exercising steady-state and determined values were not systematically different between the right and left MCAv values as CoV was found to be between 8% and 23% (Billinger et al 2017).However, the aforementioned study utilized a ramp-incremental protocol with wattage increases every 10 s until the target wattage was reached, from there, subjects cycled for six minutes at steadystate.In comparison, the current investigation utilized a stepwise protocol and observed good-to-excellent Figure 5. Absolute middle cerebral artery velocities (MCAv) seen between days and across stages during a lower body negative pressure (LBNP) cycling protocol.Data are displayed as boxplots with interquartile ranges, from 19 participants (11 females, 8 males).Four boxplots per stage are used to visualize between day differences in both males and females.A 2 × 4 analysis of variance (ANOVA) and Tukey's post-hoc comparisons were used to assess differences in MCAv between days and sexes.Statistically significant differences were observed when comparing males and females absolute MCAv values in every stage except −70, with an asterix ( * ) denoting the presence of sex differences.Middle cerebral artery (MCA), centimetres (cm), second (s), day one of the LBNP protocol (One), day two of the LBNP protocol (Two), day three of the LBNP protocol (Three), day four of the LBNP protocol (Four), 0 to −70 represent pressure changes measured in millimetres of mercury (Torr).
absolute and relative MCAv values throughout all stages and across all days.Furthermore, a previous study investigating the influence of aging on MCAv observed higher MCAv values in females compared to males (Alwatban et al 2021).These results agree with the findings from the current investigation which noted sex differences between baseline MCAv parameters, as female participants displayed higher absolute values compared to males.However, there is a paucity of exercise-based MCAv studies that compared the reliability between sexes, denoting the importance of the current sex-based findings.
A previous study by Draper and colleagues (Draper et al 2018) observed female participants' hormones across the menstrual cycle through blood and urine samples, finding wide variation in numerous hormones: progesterone (Cohen's d range: −1.51-2.88ng l −1 ), estradiol (Cohen's d range: −1.19-1.89pmol l −1 ), follicular stimulating hormone (Cohen's d range: −1.16-0.42IU l −1 ), luteinizing hormone (Cohen's d range: −1.40-1.39IU l −1 ).Additionally, a couple of studies, namely Favre and Serrador (2019) and Korad et al (2022) observed the impact of the menstrual cycle on mechanisms associated with dCA using repeated squat-to-stand maneuvers.These authors reported similar fluctuations in hormone levels throughout the study, with estradiol concentrations increasing throughout the cycle; however, it was noted by the authors this did not impact CBF regulation (Favre andSerrador 2019, Korad et al 2022).Moreover, Favre and Serrador (2019) observed differences in hormonal concentrations with the lowest for estrogen occurring in the early follicular phase, (1.83 +/− 0.58 pg ml −1 ), the highest for estradiol in the mid-luteal phase (2.46 +/− 0.60 pg ml −1 ) and progesterone peaking within the mid-luteal phase compared to both the early and late follicular phases (Favre and Serrador 2019).Therefore, despite our study not directly measuring hormone levels throughout the visits, having a similar population to these prior studies (Draper et al 2018, Favre and Serrador 2019, Korad et al 2022) and estimating cycle day through the Rhinessa Women's Questionnaire, enables us to assume hormonal fluctuations occurred across testing sessions (figures 5-6).For example, the ICC and associated 95% CI ranged Figure Relative middle cerebral artery velocities (MCAv) observed between days, stages and between sexes.Data are displayed as boxplots with interquartile ranges from 19 participants (11 females, 8 males).The four boxplots per stage are used to represent the between day differences in relative MCAv values.Comparisons for between days and sexes relative MCAv were performed through a 2 × 4 analysis of variance (ANOVA) and Tukey's post-hoc comparisons.Outliers were represented by the * seen outside of the boxplot ranges.No significant differences were found regarding relative MCAv between males and females across any days and stages (p > 0.05).Middle cerebral artery (MCA), centimetres (cm), second (s), day one of the LBNP protocol (One), day two of the LBNP protocol (Two), day three of the LBNP protocol (Three), day four of the LBNP protocol (Four), 0 to −70 represent pressure changes measured in millimetres of mercury (Torr).
from good-to-excellent, highlighting the very strong reliability of all participants within the current study (figure 4).Furthermore, as demonstrated by the CoV metrics, all participants displayed excellent levels of reliability and minimal variation (figures 2-3).While there was a very small sample size involved in this exploratory analysis, it appeared those not using contraceptives had slightly increased variability, albeit this was minimal compared to other physiological confounding influences.Future research is warranted to examine this potential relationship.

Future implications and directions
Overall, both males and females displayed good-to-excellent levels of reliability between testing sessions.These results highlight future studies employing similar methods can perform longitudinal testing, where differences across time points and/or groups would likely be due to experimental interventions, rather than variability across time.Considering the lack of female participants in previous cerebrovascular research (Kennedy et al 2022), these findings will help ensure females are represented in subsequent investigations.However, the current investigation only quantified the response of the MCAv (i.e.anterior circulation).Future investigations should further these findings within the posterior circulation (i.e.posterior cerebral artery) as some studies have noted regional differences in cerebrovascular regulation (Willie et al 2012, Smith et al 2016, Labrecque et al 2020).
The majority of female participants in this study were Caucasian, between the ages of (18-33), had a healthy BMI of (20.1-28.6 kg m −2 ), and were regularly menstruating.Future investigations should consider exploring more diverse female populations, including females who do not have a regular menstrual cycle and postmenopausal women.Finally, as this was a pilot study, the current investigation did not exclude female participants on different types of contraceptives.Despite results indicating that all female participants, regardless of contraceptive use, displayed good-to-excellent levels of reliability, females not on any form of contraceptives did display greater variability than the female group (n = 11).While the current investigation did not have the power to detect differences between contraceptive usage, subsequent investigations could benefit from larger sample sizes with the ability to stratify based on contraceptives.

Limitations
One of the major limitations of this study surrounds the usage of TCD to insonate the left and right MCA during exercise.TCD has shown an inability to quantify diameter, therefore relying upon the supposition diameter remains constant to provide a surrogate measure of CBF (Ainslie and Hoiland 2014).Moreover, the results in the current investigation highlight the excellent levels of reliability this measure showed over four testing sessions, demonstrating the utility of this measure.As previously mentioned, other limitations of this study include a small sample size of young, healthy females experiencing a regular menstrual cycle.The current investigation did not directly quantify hormone levels and we cannot be certain that female participants experienced a menstrual cycle resulting in ovulation.Thus, as only indirect self-reported measures were utilized, there may be a larger margin of error concerning the phase of the menstrual cycle collected for each participant.Hormone profiling while using a similar design would result in a more robust protocol and increased confidence regarding the influence of the menstrual cycle on orthostatic intolerance.Nevertheless, these results demonstrate a promising foundation to build upon, as LBNP cycling appears to have high reproducibility as an exercise modality with both females and males displaying good-to-excellent reliability.

Conclusions
The current investigation aimed to understand the reliability of relative and absolute CBv throughout a progressive steady-stage LBNP supine cycling exercise test.The findings of the preliminary study demonstrated high reliability in relative and absolute MCAv in both male and female participants.Throughout the LBNP cycling protocol, male volunteers displayed excellent levels of reliability, while female participants, despite known fluctuating ovarian hormone levels, also observed good-to-excellent levels of reliability.An exploratory analysis was performed on female participants not using any contraceptives, denoting slightly increased variability, albeit good-to-excellent values were still observed.This pilot study indicates that future LBNP cycling protocols may be performed across multiple days with deviations attributable to an experimental intervention or clinical prognosis, rather than between-day reliability.

Figure 4 .
Figure 4. Coefficient of variation (CoV) metrics demonstrating the validity of relative middle cerebral artery velocity (MCAv) values during a progressive lower body negative pressure (LBNP) cycling protocol.CoV data are presented for each participant involved in the study, females have further been stratified based on contraceptive use.A p-value of 0.05 as well as effect sizes were used for interpretations.The CoV thresholds were set at: <5% (excellent), 5%-10% (good), 10%-20% (acceptable), and >20% (unacceptable).Regardless of biological sex or use of contraceptives, all participants displayed excellent CoV levels across all stages.Centimeters (cm), second (s), s01 to s19 represent participants for the study, millimeters of mercury (Torr).