This study includes results from 21 participants, comprising 9 males (age: 21.6 years ±1.7, height: 177.4 cm ± 2.8, and weight: 82.9 kg ± 4.8) and 12 females (age: 21.3 years ±1.1, height: 167.9 cm ± 1.7, and weight: 71.0 kg ± 4.6). The male participants were treated as the control group in this experiment due to the absence of fluctuations in the levels of progesterone and estradiol (Barbieri, 2014). An a priori sample size calculation indicated a minimum of 20 individuals for a two-group (males vs. females), three-timepoint (menstrual cycle phases) study to indicate differences using a within–between repeated measures ANOVA with an α value of 0.05, effect size of 0.30, and power of 0.80 (Lebrun et al., 1995; Gordon et al., 2018). Additionally, in this study, male and female participants were required to be free from external hormonal influences (hormonal contraceptives, antidepressants, or other medications) and were required to maintain the current physical activity levels for the duration of the study. The female participants were required to have a regular menstrual cycle while not using oral hormonal contraceptives, intrauterine devices, implants, injections, or any other hormonal birth control methods for at least 1 year. All female participants reported a regular cycle lasting ≥21 days and ≤35 days (cycle length: 29.33 days ±3.30) and a luteal phase of ≥10 days based on ovulation detected using basal body temperature (BBT) measurements. Participants were free from metabolic, respiratory, neurological, and cardiovascular diseases, and free from any debilitating musculoskeletal injury (within 1 year). In addition, none of the participants were engaged in a regular training regimen, which was defined as not meeting ACSM recommendations for physical activity. All participants provided written informed consent before participating in the study. This study was approved by the local Institutional Review Board and complied with the Declaration of Helsinki.

This study utilized a repeated measures design, across three timepoints—during menses (day 0–3), which was considered the early follicular phase (EF), the ovulation phase (O), and the mid-luteal phase (ML; between 7 and 9 days following the spike in BBT)—with similar time intervals between visits for the male participants. The participants completed a Profile of Mood States (POMS) at all three visits. In addition, the female participants received a menstrual cycle history questionnaire. The first visit included a familiarization with the GXT protocol and equipment. Following the familiarization trial, visits were scheduled to control for the time of day (within a 2-h start range). The participants were instructed to refrain from vigorous exercise for 48 h and fast 3 h prior to data collection and were also advised not to consume any caffeine within 6 h prior to testing.

The female participants were scheduled for visits during menses as an indication of an early follicular phase (EF; within the first 3 days of menstruation, day 0–3), ovulatory (O; within 24 h of a spike in BBT ≥ 0.3°C), and mid-luteal (ML; 7–9 days following the spike in BBT) phases. BBT was tracked in all participants using a digital thermometer (iProvèn Model BBT-113Ai, Beaverton, OR, USA). BBT was measured immediately after waking to ensure the most accurate reading. Participants were required to track and report BBT each morning for 4 weeks before and throughout the study. This established a baseline of BBT values for each individual and aided in the confirmation of ovulation (McClure Brown, 1973). Further visits were scheduled according to which phase occurred first following the familiarization visit. This ensured that the visit order was randomized across female participants, which served to minimize the influence of the learning effect on the test results. For the female participants, four participants had their first visit during EF, two participants had their first visit during O, and six participants had their first visit during L. The time between visits varied between subjects due to differences in female participants’ cycle and availability (Barbieri, 2014). Visit dates for male subjects were randomly assigned to correspond to the time between visits for the average female 28-day cycle. For example, if a male participant was assigned a visit equivalent to EF first, his visit equivalent to O occurred approximately 14 days after, and his visit equivalent to ML was approximately 7–9 days later. The male participants were randomly assigned to visit on analogous days according to a predicted 28-day cycle. For the male participants, three had their first visit scheduled as “EF,” four had their first visit scheduled as “O,” and two participants started testing in “ML.”

The GXT protocol was performed on a cycle ergometer (Lode Excalibur Sport, The Netherlands). Participants were fitted with a Polar heart rate sensor (Polar Inc. Model H1, Bethpage, NY, USA) to track the heart rate. Before GXT, participants performed a 5-min warm-up at a self-selected intensity. GXT consisted of a ramp protocol starting at 0.5 W/kg of body weight and increasing by 0.5 W/kg every 60 s (Larson et al., 2015). The participants were required to maintain a self-selected cycling cadence; if the cadence declined more than 10 rpm, then the test was terminated. Following the cessation of the test, the participants performed a cooldown at 25 W until their heart rate dropped below 130 bpm.

Expired gases were collected and analyzed using a metabolic cart (ParvoMedics TrueOne 2400, Sandy, UT, USA). VO_2max was recorded breath-by-breath continuously during the exercise test using a 30-s average. During GXT, RPE and lactate were measured at the end of each minute of exercise. At the end of the test, the last stage HR (HR_max) respiratory exchange ratio (RER), and peak power (power achieved during the last stage) were recorded.

To ensure a true maximal value for oxygen consumption, the participants rested for 20 min before performing a verification test on the cycle ergometer. The load of the verification test was set at the peak power of the previously administered GXT. The verification test began with a 10-s warm-up at 0.5 W per kilogram of the participant’s body weight and increased every 5 s, ensuring that the participants reached the max load within 30 s of cycling. If the participant maintained a cadence of at least 40 rpm for the majority of the 5 min, or if the participant’s VO₂ value exceeded the previous measure (±3%) (Nolan et al., 2014), then the participant’s previous maximal graded exercise test was deemed invalid. If these criteria were not met and the participant stopped cycling or cycled at a cadence <40 rpm within 5 min, then the original maximal test was considered valid, and all measured parameters were recorded. All of the participants met the criteria for a valid test. Cooldown was set at 25 W, and subjects were free to dismount the bike when their HR measured ≤130 bpm and they felt comfortable walking and exiting the laboratory safely.

The blood lactate level was assessed by performing a finger stick drawing of ∼1.0 μL blood; samples were analyzed using a portable lactate analyzer (Lactate Plus, Nova Biomedical, Waltham, MA, USA). Blood lactate was collected before and immediately after the GXT.

The participants rated their exertion based on a modified Borg scale ranging from 0 (no exertion at all) to 10 (maximal exertion) (Borg, 1982). They were instructed to point to the number corresponding to their overall RPE.

Twenty-four hours after each visit, the participants were asked to complete a perceived recovery status (PRS) scale, which ranged from 0 (very poorly recovered and extremely tired) to 10 (very well recovered and highly energetic) (Laurent et al., 2011).

A shortened version of the POMS questionnaire was used for the study (Shacham, 1983). The questionnaire consists of a series of descriptions that describe the feelings that the participants may have. The participants self-reported on each of these areas for how they were feeling on the day of each test. The scoring ranged from 0 to 4, with 0 being “not at all” and 4 being “extremely.” The categories of interest were dealing with feelings of fatigue and vigor on the day of each test. POMS was used as a way to examine if there were changes in moods over the menstrual cycle for the female participants and over time in the male participants.

Statistical analysis for this study was performed using IBM SPSS Statistics version 28 (IBM Corp., Armonk, NY, USA). A two-group (female and male) by two-condition (GXT and verification) by three-timepoint (MC phase/time point) repeated measures ANOVA was used to examine differences in VO_2max. A two-group (female and male) by three-timepoint (MC phase/time point) repeated measures ANOVA was performed for the objective (HR_max, RER, and blood lactate) and subjective measures (overall RPE, local RPE, and PRS). Pairwise comparisons were performed using a Bonferroni correction. The effect size for each variable was calculated using partial eta-squared (η_p ²), in which small effect = 0.01–0.05, medium effect = 0.06–0.13, and a large effect size ≥ 0.14 (Cohen, 1988).

Intraclass correlation coefficients (ICCs) and their 95% confidence intervals for the objective parameters (VO_2max, RER, HR_max, and lactate) were calculated based on a mean-rating (k = 3), absolute agreement, two-way mixed effects model (Koo and Li, 2016). ICCs were interpreted as follows: values less than 0.50 indicated poor reliability, values between 0.50 and 0.75 indicated moderate reliability, values between 0.75 and 0.90 indicated good reliability, and values greater than 0.90 indicated excellent reliability (Koo and Li, 2016).

There were no observed main effects of the MC phase for the female participants or main effects of time for the male participants on the objective measures (VO_2max, RER, HR_max, or peak blood lactate) (Table 1). In addition, there were no interactions between MC phase/time and sex for any of the objective measures (Table 1). There were no observed differences within sex for VO_2max recorded from GXT and verification protocols (Table 1). There was an effect of sex on VO_2max (F_1,19 = 318.896, p < 0.001, ηp² = 0.944) and peak power (F_1,19 = 548.377, p < 0.001, ηp² = 0.967), where the male participants had a higher VO_2max (mean difference: 9.44 mL kg^-1 ∙ min^-1; p = 0.026, ηp² = 0.236) and higher peak power (mean difference: 86.96 W; p < 0.001, ηp² = 0.472) than the female participants (Table 1). The objective measures were observed to have moderate to excellent test–retest reliability (Table 2).

There were no main effects or interactions for any of the POMS categories for either sex. There was a phase-by-sex interaction for overall RPE (F_2,38 = 3.543, p = 0.039, ηp² = 0.157). Overall RPE was higher at EF (8.92 ± 0.79; ηp² = 0.416) than O (7.67 ± 1.23; p = 0.008) and ML (7.75 ± 1.06; p = 0.013) for the female participants (Figure 1). A medium effect of sex at each timepoint (EF: 0.09, O: 0.09, and L: 0.08) was observed, with the male participants reporting lower RPE at EF and higher RPE at O and L compared to the female participants (Figure 1). There was a phase-by-sex interaction for perceived recovery (F_2,38 = 21.972, p < 0.001, ηp² = 0.536). Perceived recovery was lower (i.e., they felt less recovered) following EF (6.83 ± 0.94; ηp² = 0.825) compared to O (8.83 ± 1.12; p < 0.001) and ML (8.67 ± 0.65; p < 0.001) for the female participants (Figure 2). The female participants (6.83 ± 0.94) had lower perceived recovery after GXT during EF than the male participants (9.00 ± 1.00; p < 0.001, ηp² = 0.577) (Figure 2). There were no observed differences between visits for the subjective parameters for the male participants (Figure 1; Figure 2). Individual data for RPE are shown in Figure 3, and individual data for PRS are shown in Figure 4.

The objective measures of aerobic performance did not significantly differ across the menstrual phases that were examined, which is in agreement with the results of previous studies of untrained, active, and trained female participants (Dombovy et al., 1987; Bemben et al., 1995; Smekal et al., 2007). The high test–retest reliability observed in the present study show that the objective aerobic parameters (VO_2max, RER, peak blood lactate, and peak power) are not affected by the menstrual cycle in the female participants or tests performed at similar time intervals in the male participants.

The major findings from this study would suggest that during menses, the female participants felt as if they were exerting more effort during the maximal graded exercise test and felt less recovered 24 h following the maximal aerobic exercise bout. These findings are in contrast with those of some previous research studies, which found no changes in RPE across the menstrual cycle phases (Bailey et al., 2000; Lara et al., 2020). This may be due to differences in the training status of the individuals in the present study compared to the previous studies. VO_2peak for the participants in the present study was ∼30 mL/kg/min compared to ∼50 mL/kg/min (Bailey et al., 2000) and ∼48 mL/kg/min (Lara et al., 2020). Individuals with low fitness or sedentary individuals have been shown to perceive greater exertion during exercise compared to trained individuals (Travlos and Marisi, 1996; Chen et al., 2002). Endurance training enhances oxidative enzyme activity and mitochondrial content, which results in a different metabolic profile compared with other types of training (Egan and Zierath, 2013; Al-Khelaifi et al., 2018). This could impact how female athletes perceive effort compared to untrained female participants. In addition, elite and high-level athletes have been shown to have increased pain tolerance, higher pain thresholds, and lower pain intensity (Pettersen et al., 2020). In addition, endurance athletes have been shown to have better pain tolerance and lower pain intensity than non-athletes (Pettersen et al., 2020).

On the other hand, the present findings are analogous to results found in other studies, which had individuals with lower reported physical activity levels (Caldwell Hooper et al., 2011). One study measured RPE and perceived pain during a treadmill exercise protocol at 65% of measured VO_2max in regularly menstruating female participants who participated in low levels of physical activity (Caldwell Hooper et al., 2011). The authors found that RPE and perceived pain ratings were higher in the female participants during the EF phase when compared to the late follicular and luteal phases. In addition, this study found a moderately strong positive correlation (r = 0.548) between perceived pain and RPE (Caldwell Hooper et al., 2011). This could be due to the fact that the female participants were shown to have increased clinical and experimental pain during the EF phase, which is associated with the declining levels of estrogen and progesterone (Martin, 2009). The authors concluded that RPE and perceived pain ratings were higher in the female participants during the EF phase, and this change is attributed to a drop in concentrations of both sex hormones, estradiol and progesterone, during this phase (Caldwell Hooper et al., 2011). However, this study did not measure hormone levels and used a calendar-based method for estimating the MC phase (Caldwell Hooper et al., 2011). Given the prevalence of menstrual cycle disturbances (Janse De Jonge et al., 2019), a calendar-based method would allow for the possibility that interindividual variations in hormone concentrations and hormone fluctuations across the MC could affect RPE.

There are a few possible explanations for the differences in subjective parameters across the menstrual cycle during exercise including pain perception, which in turn can influence RPE (Borg, 1982). Changes in mechanical and metabolic activities stimulate group III/II afferent neurons, which are involved in the perception of fatigue and pain (Jankowski et al., 2013; Pollak et al., 2014; Amann and Light, 2015). Increased concentrations of metabolites (ATP, lactate, and inorganic phosphate) during exercise have been shown to increase the perception of efforts during exercise (Silva et al., 2021; da Silva et al., 2019), which could be perceived as fatigue and/or pain. Given that the female participants achieved similar peak power, VO_2max, RER, and blood lactate levels, it would be safe to assume that there was a similar level of mechanical and metabolic stimulation across the MC phases in the female participants. A possible explanation for the difference in RPE during exercise across the MC could be due to female participants having a higher sensitivity to a painful stimulus (lower pain threshold) during the follicular phase compared to other phases due to greater activation of brain areas related to general body awareness during the follicular phase (Veldhuijzen et al., 2013). This study found that the brain activity during a painful stimulus was found in the brain regions associated with general body awareness and motor skills and that areas associated with pain modulation were deactivated during the follicular phase (Veldhuijzen et al., 2013). A second possible explanation is the hypoactivity of the brain circuitry systems related to the inhibition of painful stimuli in the female participants. These systems have different activity levels throughout the menstrual cycle and operate more effectively around ovulation when estrogen is high and progesterone is low (Tousignant-Laflamme and Marchand, 2009). Previous research has shown that there is a correlation between pain and RPE during exercise (Rejeski, 1981). Our findings support the complicated interaction of physiological elements to perceive noxious stimuli and perception of efforts across the menstrual cycle.

Lastly, the female participants reported significantly lower PRS (recovery) values during EF compared to both O and ML. Previous research has shown that 24 h post-endurance exercise biomarkers (creatine kinase and IL-6) were higher during the mid-follicular phase than during the luteal phase (Hackney et al., 2019). Considering the antioxidant aspects and reduced levels of estrogen during the follicular phase (Tiidus, 2000), the biomarker results (Hackney et al., 2019), in addition to the findings from the present study, could indicate that there may be a need for greater recovery time in female participants following intense exercise in the early and mid-follicular phases. In trained individuals, greater sleep disturbance in the mid-luteal phase compared to the late-follicular phase was observed (De Martin Topranin et al., 2023). However, in another study, the female participants reported a lower quality of sleep during the 4 days of menses compared to mid-follicular and luteal phases (Baker and Driver, 2004). Decreased sleep quality may have impacted the recovery of female participants, following exercise in the EF phase.