Increased fatigability and impaired skeletal muscle microvascular reactivity in adults with obstructive sleep apnea: a cross-sectional study

Abstract

Background

Sympathetic nervous system hyperactivity and chronic intermittent nocturnal hypoxia in individuals with obstructive sleep apnea (OSA) predispose them to microvascular impairment, which may contribute to increased daytime muscle fatigue. This study aimed to assess microvascular reactivity of the skeletal muscle, examine fatigability, and determine the relationship between fatigability and microvascular reactivity in adults with OSA.

Methods

Twenty-six participants were allocated into two groups—those with OSA and those without i.e. non-OSA. Each group comprised of 13 individuals who underwent an arterial occlusion test on their non-dominant leg. The percentage change of maximal hyperemic response (MHR) and the time to achieve MHR (tM) of both the total myoglobin/hemoglobin (∆[Hbtot]) and the oxygenated myoglobin/hemoglobin (∆[HbO₂]) signals from near-infrared spectroscopy were calculated to examine microvascular reactivity. In addition, a 10-min walk test was performed to assess performance and perceived fatigability.

Results

The OSA group demonstrated a reduced in ∆[Hbtot]MHR (150.9 ± 16.2% vs. 235.8 ± 72.7%, p = 0.006), ∆[HbO₂]MHR (131.4 ± 8% vs. 161.7 ± 10.6%, p = 0.001) and increased ∆[Hbtot]tM (80.5 ± 13.1 s vs. 47.7 ± 9.9 s, p < 0.001), ∆[HbO₂]tM (85.2 ± 22.4 s vs. 52.1 ± 5.9 s, p = 0.001) compared to the non-OSA group. In addition, participants in the OSA group experienced greater perceived (6 ± 1 vs. 2.8 ± 0.1, p = 0.001) and performance fatigability (1.1 ± 0.1 vs. 0.9 ± 0.1, p = 0.001) compared to adults in the non-OSA group. Moreover, both performance and perceived fatigability were significantly associated with microvascular reactivity parameters (all p < 0.05).

Conclusion

Microvascular dysfunction, as determined by an attenuated post-occlusive reactive hyperemia, is observed in individuals with OSA that may contribute to increased fatigability in these individuals.

Background

Obstructive sleep apnea (OSA) is a common sleep breathing disorder characterized by repetitive episodes of partial (hypopnea) or complete (apnea) upper airway closure during sleep that result in intermittent hypoxemia, hypercapnia and poor sleep quality [1]. Epidemiological data indicates that approximately 34% of men and 17% of women in the United States have OSA [2], posing a significant burden on overall health and daily functioning of these individuals. Common symptoms include snoring, frequent arousals during sleep, increased daytime somnolence, and excessive performance fatigue during daily activities. However, the overlapping definitions of daytime sleepiness and fatigue both in research and clinical practice have made it difficult to recognize fatigue in OSA as a major symptom [3].

While excessive daytime sleepiness has historically been emphasized in OSA, recent research indicates that fatigue, characterized by the subjective feelings of tiredness, may be a more prevalent symptom [3]. Indeed, adults with OSA who experience high levels of fatigue often have more severe sleep disturbances and suffer greater overall health consequences compared to those who primarily report increased sleepiness [4]. However, it has also been suggested that the fatigue severity is less likely to be directly related to the OSA severity [5]. This disparity in findings underscores a critical gap in understanding how fatigue manifests differently from sleepiness and its implications for diagnosis and management strategies. One potential approach to address this gap could involve quantifying fatigue rather than relying solely on subjective reporting, thereby improving our ability to manage OSA more effectively. Fatigability is a phenotype of fatigue that measures the change in the feeling of tiredness (perceived fatigability), or the decline in performance (performance fatigability) as a function of the duration, intensity, and/or frequency of an activity [6]. It quantifies how an individual's subjective feeling of tiredness and performance decline over time or with increased demand, providing a normalized assessment of fatigue specific to the task being performed [6]. Offering a description of how physical or mental exertion impacts overall capability to carry-out daily activities, this approach gives a clearer picture of functional implications of tiredness, beyond its subjective experience, in real-world settings. Therefore, fatigability helps us to better understand the interplay between subjective feelings and actual task performance under varying levels of physical and cognitive demand, providing valuable information for optimizing strategies to manage fatigue and improve quality of life (see “Fatigability” section for calculations).

Furthermore, the findings of epidemiological studies demonstrating a link between fatigue and ischemia [7, 8] suggest the plausibility of microvascular dysfunction as one of the potential contributing mechanisms of fatigue. In fact, endothelial cells have been proposed to be an important sensor of fatigue in healthy adults [9]. In OSA, the apnea–hypopnea induced hypoxia and hypercapnia, acting via the chemoreflex, elicit a concomitant surge of sympathetic nervous system (SNS) activity that modulates a rapid cardiorespiratory response to normalize blood gasses [10]. This sustained excitation of the SNS persists during the day and is believed to promote vascular alterations that are associated with cardiovascular comorbidities in adults with OSA [11]. However, there is inconclusive evidence regarding the presence of microvascular dysfunction in OSA. Despite the results of skeletal muscle biopsies showing increased microvascularization in adults with OSA as compared to healthy controls [12], the assessment of endothelial function of resistance vessels by strain gauge venous occlusion plethysmography showed reduced blood forearm flow following acetylcholine infusion, indicative of endothelial microvascular dysfunction in this population [13].

Additionally, while reduced exercise capacity attributed to factors such as fatigue and dyspnea is well documented in OSA [14], the underlying mechanisms linking these exercise-limiting symptoms to the microvascular alterations remain incompletely understood. It is plausible that the deterioration in muscle oxygenation secondary to microvascular dysfunction may predispose individuals with OSA to a greater overall fatigability while carrying out various physical activities. However, as discussed above, the conflicting results in the literature showing both increased microvascularization and signs of endothelial dysfunction in adults with OSA warrant the need of comprehensive investigations focused on exploring the microvascular function in skeletal muscle and its implications for fatigability. Non-invasive assessment of microvascular function such as the evaluation of microvascular reactivity, could provide a promising approach. Microvascular reactivity, as measured by post-occlusive reactive hyperemia (PORH), is a vital mechanism of microvascular function that enables a vascular response to stimuli requiring adjustments of blood flow and alterations of vessel tone [15]. PORH represents a global assessment of microvascular function as it involves metabolic vasodilators, endothelial vasodilators, myogenic response to shear stress, and release of vasoactive factors from perivascular nerves [16]. Thus, impaired microvascular reactivity or a diminished PORH response in skeletal muscle leading to reduced tissue oxygenation or oxygen availability may contribute significantly to increased fatigability.

To further our knowledge about the plausible mechanisms associated with increased fatigability in OSA, assessment of the skeletal muscle microvasculature is critical for understanding the relationship between sleep disorders and vascular control in OSA. Therefore, the present study aimed to: (i) determine perceived and performance fatigability; (ii) assess skeletal muscle microvascular reactivity during resting condition following a brief vascular occlusion; and (iii) explore the relationship between fatigability and microvascular reactivity in individuals with OSA. By advancing our understanding of these interactions, we aim to pave the way for targeted therapeutic interventions aimed at improving both symptoms and quality of life in individuals with OSA.

Methods

Ethical approval

The study protocol and procedures were approved by the Institutional Review Board of George Mason University, Fairfax, Virginia (IRB # 535275-1). Before beginning any data collection, details of the protocol including the risks and benefits involved with testing procedures were discussed and a written informed consent was obtained from all the volunteers participating in the study.

Study design

This prospective observational study used a cross-sectional design and a convenience sampling method that compared two groups: (1) an experimental group (i.e. OSA group) consisting of individuals diagnosed with OSA having an apnea hypopnea index (AHI) score ≥ 5 events per hour of sleep, and (2) a control group (i.e. non-OSA group) comprising of healthy adults without signs or symptoms of OSA or any other known sleeping disorder.

Study participants

All the participants were recruited from the surrounding areas of Northern Virginia and Metropolitan Washington, D.C. Inclusion criteria for the OSA group included: adults aged between 18 and 60 years, and body mass index (BMI) < 40 kg/m² with OSA diagnosed by a third party polysomnography study conducted within the past five years of enrollment according to the clinical guidelines of the American Academy of Sleep Medicine [17]. Participants in the non-OSA group were required to have no night-time signs or symptoms of OSA and/or an AHI between 0 and 4 on a polysomnography study in addition to the criteria mentioned above. In addition, all the participants had to be free of any known cardiovascular, metabolic, or lung diseases, engaging in less than 150 min of physical activity per week, and be non-smokers. Participants were excluded from the study if they were determined to be at high risk for a cardiac event as assessed by the American College of Sports Medicine (ACSM) risk stratification algorithm [18], or if they reported that they had been or are presently being treated by a physician for any form of cardiovascular disorder, respiratory ailments, metabolic disease including diabetes, any bleeding disorder or vascular disease, or if they were pregnant. Since hypertension is a common condition associated with OSA, individuals with controlled hypertension were not excluded from the study.

Study procedure

After being provided details about the study objectives and procedures, all participants completed a general health history questionnaire and the ACSM risk stratification algorithm [18]. Following these assessments, the participants signed a written informed consent form and were formally enrolled in the study. To determine their physical activity status and number of sleep hours, the seven-day physical activity recall interview was conducted. The duration of the physical activity was directly calculated as minutes per week, while the physical activity dose in metabolic equivalent hours per week (MET hours/week) was estimated by multiplying the hours spent on moderate, vigorous, and very vigorous-intensity activity by the corresponding MET: 4, 6, and 10, respectively [19]. Thereafter, the participants’ height, weight, and neck circumference were measured using standard criteria [20]. In addition, skinfold thickness was measured three times according to the international standards [21] at the level of the largest circumference of the left lateral gastrocnemius using a Champion’s baseline standard skinfold caliper (#1249835). The adipose tissue thickness covering the measuring site of the muscle, that has been shown to influence the accuracy of the near-infrared spectroscopy signal [22], was then determined by dividing the mean skinfold thickness measurement by two. Also, resting heart rate and blood pressure were measured in the seated position prior to beginning the testing protocol. Moreover, participants in the OSA group were asked about their continuous positive airway pressure (CPAP) usage, determined by the total number of days per week and total hours for each use through a self-report interview.

Fatigability

Fatigability is a quantifiable phenotype of fatigue that captures the change in perceived tiredness (perceived fatigability) or the change in performance (performance fatigability) over time, and is measured as a function of the duration, intensity or frequency of exertion, representing an important aspect of functionality [23]. To measure perceived and performance fatigability, a standardized 10 min walk test (10MWT) was conducted in a 25 m enclosed corridor on a flat course. All participants were instructed to walk at their self-selected pace to cover as much distance as possible in 10 min. They were allowed to stop before 10 min if they had symtoms like chest pain, nausea or dizziness or if they were unwilling to continue. To avoid interference with their performance, participants walked unaccompanied and no encouragement was given. Before beginning the walking assessment, each participant was asked to report their current level of tiredness on a seven-point scale that ranged from 1 (indicating extremely energetic) to 7 (indicating extremely tired) (Table 1) [24]. Each participant was again asked to rate their change in tiredness following 10 min of walking on the same seven-point scale. This subjective rating was considered a measure of perceived fatigue. Both perceived and performance fatigability scores were multiplied by 1000 for reporting purposes.

Table 1 Fatigue and fatigability scale

Perceived fatigability assessment

Perceived fatigability was determined by dividing each participant’s perceived rating of change in tiredness following the 10MWT by the total distance covered in 10 min. For example, a participant who reported a scale of 5 after walking 300 m in 10 min would have a perceived fatigability score of 5 divided by 300, i.e., 0.016. A participant who reported a similar level of change in tiredness but walked 250 m would have a fatigability score of 0.02. Thus, the higher the score the greater the perceived fatigability.

Performance fatigability measurement

Performance fatigability was measured by dividing the walking speed (m/s) over the entire 10 min by the walking speed in the first 2.5 min. This number was then divided by the total distance walked in 10 min to describe the change in performance [24]. The higher score reflects greater performance fatigability.

Near-Infrared spectroscopy (NIRS)

NIRS is a non-invasive technique that has been used to measure tissue oxygenation (saturation/perfusion) and hemodynamics in various human tissues including the brain and skeletal muscle [25]. Its utility lies in the assessment of the contribution of oxygen delivery and utilization in skeletal muscle at rest and during exercise [25]. In the present study, a continuous wave NIRS system (Oxymon MK-III, Artinis Medical Systems, The Netherlands) was used tot measure the concentration changes (Δ) of the signals. Please note that the delta symbol (Δ) preceding [HbO₂] and [Hbtot] in this manuscript represents the relative change in the concentration of these signals from baseline. It does not indicate changes during occlusion or hyperemia, as the continuous wave NIRS system is designed to measure changes relative to baseline values rather than absolute concentrations. Prior to each data collection, the NIRS optode was calibrated per manufacturer’s recommendations. Following the calibration procedures, the skin-fold measurement was obtained, and the NIRS optode was placed and secured on the belly of the left lateral gastrocnemius muscle using tape. As a default set-up to maximize light penetration, all data collections were conducted on a 40-mm transmitter–receiver channel using a sampling rate of 10 Hz per second. Data were collected using the software Oxysoft 4.7 (Artinis Medical System) and were exported in Microsoft Excel format for off-line analysis.

Microvascular reactivity by post-occlusive reactive hyperemia (PORH)

A rapid cuff inflation system (Hokanson) was used during the occlusion test. The cuff was placed around the left quadriceps, inflated to 220 mmHg, and was sustained for maximum of 8 min. To quantify the PORH response and thus the microvascular function and reactivity, the following NIRS parameters for both the oxygenated myoglobin/hemoglobin (∆[HbO₂]) and the total myoglobin/hemoglobin (∆[Hbtot]) signals were measured and are presented in Fig. 1:

• ΔC_OCC, maximal change during occlusion, presents the maximal change in the signal from the beginning of the occlusion to the end of occlusion (Fig. 1E).

• ΔC_RH, maximal change during reactive hyperemia, indicates the maximal change in the signal during the phase of reactive hyperemia was calculated as the (maximal response value – end occlusion value) of the signal after cuff deflation (Fig. 1F).

• MHR, maximal hyperemic response, depicts the maximal change in the signal after the release of the cuff, expressed as the percentage of the signal change after cuff deflation, and was calculated as [ΔC_RH / (resting value – end occlusion value)] *100 (Fig. 1G).

• tM, time to hyperemic response (peak value): the time interval in seconds after release of the cuff until 100% of the signal’s peak value is reached (Fig. 1H).

Fig. 1

Measures obtained from NIRS during arterial occlusion test. Panel-I represents measures obtained from oxygenated myoglobin/hemoglobin (Δ[HbO₂]) tracing. Panel-II represents measures obtained from total myoglobin /hemoglobin (Δ[Hbtot]) tracing. The graphs are depicting the comparison of NIRS parameters between a representative participant with OSA (OSA: red tracing) and a healthy adult without OSA (Non-OSA: blue tracing) measured during an arterial occlusion assessment of the left lateral gastrocnemius. A Baseline measured as the average of 30 s before the start of occlusion; B start of occlusion (time at 0 s); C end of occlusion; D peak hyperemic response; E ΔCOCC—maximal change in signal during occlusion; F ΔCRH—maximal change in signal during reactive hyperemia; G MHR%—maximal hyperemic response in percentage; H tM—time to hyperemic response (in seconds)

Data analysis

For all the analyses, signals were averaged into 1 s time bins and time aligned to the start of occlusion. The baseline response was an average of the 30 s before the start of occlusion, while 5 s averages immediately following the start of occlusion, end of occlusion, and post-occlusion maximal response was used for the remaining events. Hypertension was created as a binary variable with values ‘0’ and ‘1’ representing its absence and presence, respectively. Individuals were designated as hypertensive according to the American Heart Association’s guidelines of categorizing hypertension i.e., if their resting blood pressure was ≥ 130/80 mmHg.

Differences in baseline characteristics between the two groups were calculated using a two-sample Student’s t-test. Analysis of Covariance (ANCOVA) was used to compare microvascular reactivity and fatigability measures between the two groups after controlling for age, sex, BMI, race/ethnicity, sleep duration and total physical activity (MET hours/week). Multivariate linear regression models were used to assess the relationship between fatigability and microvascular reactivity after controlling for the effects of the above-mentioned covariates. Previous research indicates that these factors i.e., age, sex, BMI, physical activity and sleep duration can influence fatigability and microvascular function by impacting metabolic rate and vascular recovery [26,27,28,29,30]. We used a stepwise regression approach to include the variables based on their significance and given theoretical relevance. Initially, all variables were entered into the model to assess their overall impact. We then performed stepwise selection (both forward and backward) to determine which variables had the most significant impact on fatigability. Statistical significance was set at p < 0.05. All the data analyses were performed using STATA (StataCorp LP, Stata Statistical Software: Release 14. College Station, TX).

Results

A total of sixteen individuals in the OSA group and fourteen in the non-OSA group were screened for participation in the present study. Three individuals in the OSA group were excluded based on the preliminary screening through a phone interview as they did not meet inclusion criteria. During the in-person evaluation, one individual in the non-OSA group was determined to be at high risk for a cardiac event as assessed by the ACSM risk assessment algorithm [18] and was excluded from the study. A total of 26 individuals (13 in each group) completed the in-person screening, signed the informed consent, and were enrolled in the study.

Baseline characteristics

The baseline characteristics of all the study participants are presented in Table 2. Participants were middle-aged overweight (BMI: 26.5 ± 4.9 kg/m²) adults consisting of 73% males. The average apnea/hypopnea index for participants in the OSA group was 46.8 ± 26.2 events/hour. Four participants in the OSA group were clinically diagnosed with hypertension, with a mean resting systolic blood pressure of 134 ± 4.3 mmHg, mean resting diastolic blood pressure of 83.5 ± 8.4 mmHg, but were not on any antihypertensives medication whereas all the participants in the non-OSA group were non-hypertensive. Also, six individuals in the OSA group used CPAP with an average usage of 46.5 ± 8.8 h/week. There were no significant differences at baseline between the OSA and the non-OSA group in age, sex, adipose tissue thickness at the gastrocnemius, sleep duration and physical activity status level.

Table 2 Baseline characteristics of the participants in OSA and Non-OSA groups

Microvascular function

The values of PORH variables representing microvascular reactivity for both the OSA and the non-OSA group are presented in Table 3. The absence of a significant change in the ∆[Hbtot] signal from the beginning to the end of occlusion in both the groups (OSA group = −7.05 ± 3.8 and −7.23 ± 3.9, respectively, p = 0.91; non-OSA group = −7.06 ± 4.2 and −7.35 ± 4.3, respectively, p = 0.87) was an indicator that a complete arterial occlusion has been achieved. Conversely, the change in Δ[Hbtot] from the end of occlusion to the period of hyperemia reflects microvascular capacity. It is important to note that these two are separate measurements, each providing a distinct insight into microvascular function, and are not comparable. Further, after controlling for age, sex, race/ethnicity, BMI, sleep duration and level of physical activity, the amplitude changes for ∆[HbO₂] and ∆[Hbtot] during reactive hyperemia (i.e., MHR%) were significantly lower whereas the time taken to achieve the maximal response (tM) was significantly greater in individuals with OSA as compared to their healthy counterparts (Fig. 2; Table 3).

Table 3 Differences in post-occlusive reactive hyperemia between OSA and non-OSA groups

Fig. 2

Graphs showing differences in the NIRS parameters between OSA and non-OSA groups. A Represents the difference in the maximal hyperemic response of both the oxygenated (Δ[HbO₂]) and the total myoglobin/hemoglobin measurements (Δ[Hbtot]) between the OSA and the non-OSA participants. * indicates p < 0.05. B Represents the difference in the time to reach the maximal hyperemic response of both the oxygenated (Δ[HbO₂]) and the total myoglobin/hemoglobin measurements (Δ[Hbtot]) between the OSA and the non-OSA participants. * indicates p < 0.05

Fatigability

Eight individuals (61.3%) in the OSA group reported feeling tired after completion of the 10MWT. Nine participants (69.3%) in the non-OSA group reported being more energetic, compared to only one (8%) in the OSA group. Four individuals (30.7%) in each group reported feeling “neither more tired nor energetic” at the end of 10MWT. After controlling for age, sex, race/ethnicity, BMI, sleep duration and level of physical activity, individuals with OSA were found to have greater perceived, as well as performance fatigability as compared to non-OSA group (Table 4).

Table 4 10MWT variables and fatigability between OSA and non-OSA groups

In addition, on multivariate regression, after controlling for the impacts of age, sex, race/ethnicity, BMI, sleep duration and physical activity level, the performance fatigability was found to be significantly associated with ∆[Hbtot] MHR (β = −0.002, R² = 0.703, p < 0.01), ∆[Hbtot] tM (β = 0.005, R² = 0.685, p < 0.01), ∆[HbO₂] MHR (β = −0.006, R² = 0.636, p < 0.01) and ∆[HbO₂] tM (β = 0.004, R² = 0.629, p < 0.01). Likewise, the perceived fatigability was significantly associated with ∆[Hbtot] MHR (β = −0.033, R² = 0.662, p < 0.01), ∆[Hbtot] tM (β = 0.059, R² = 0.491, p < 0.01), ∆[HbO₂] MHR (β = −0.092, R² = 0.618, p < 0.01) and ∆[HbO₂] tM (β = 0.039, R² = 0.368, p = 0.046). Please note that the R² values provided represent the overall model fit and not the partial correlations. Figures 3 and 4 further present the relationship between the measurements of fatigability and NIRS parameters.

Fig. 3

Relationship between performance fatigability and NIRS parameters. The scatter plots (A, B) present the relationship between the performance fatigability and oxygenated myoglobin/hemoglobin (Δ[HbO₂]). The scatter plots (C, D) present the relationship between the performance fatigability and total myoglobin/hemoglobin (Δ[Hbtot]). Please note that R represents the Pearson’s coefficient of correlation and p value depicts the level of significance

Fig. 4

Relationship between perceived fatigability and NIRS parameters. The scatter plots (A, B) present the relationship between the perceived fatigability and oxygenated myoglobin/hemoglobin (Δ[HbO₂]). The scatter plots (C, D) present the relationship between the perceived fatigability and total myoglobin/hemoglobin (Δ[Hbtot]). Please note that R represents the Pearson’s coefficient of correlation and p value depicts the level of significance

Moreover, we did not observe any significant impact of hypertension or hours of CPAP use on microvascular function or fatigability measures in participants in the OSA group. Specifically, there were no significant associations between CPAP use and microvascular function indicators: ∆[HbO₂] MHR vs. CPAP use (β = 0.044, R² = 0.368, p = 0.697), ∆[HbO₂] tM vs. CPAP use (β = −0.246, R² = 0.321, p = 0.463), ∆[Hbtot] MHR vs. CPAP use (β = 0.119, R² = 0.49, p = 0.566), or ∆[Hbtot] tM vs. CPAP use (β = −0.005, R² = 0.321, p = 0.979). Similarly, there were no significant correlations between CPAP use and measures of performance fatigability (performance fatigability index vs. CPAP use: β = −0.001, R² = 0.776, p = 0.11) or perceived fatigability (perceived fatigability index vs. CPAP use: β = −0.001, R² = 0.492, p = 0.887).

Discussion

The present study assessed fatigability in adults with OSA and examined muscle microvascular reactivity using NIRS which, to the best of our knowledge, have not been examined previously. The primary findings of this study are that individuals with OSA: (i) reported higher levels of both perceived and performance fatigability compared to adults without OSA, and (ii) demonstrated reduced magnitude as well as slowed maximal hyperemic response after arterial occlusion, suggesting lower maximal vasodilator capacity and impaired microvascular reactivity suggesting that these individuals with OSA have microvascular dysfunction at rest compared to their healthy counterparts. In addition, perceived and performance fatigability were associated with microcirculatory alterations as measured by the hyperemic response. Specifically, higher (greater MHR%) and faster (lower tM) hyperemic responses were significantly associated with both perceived and performance fatigability.

Fatigability

This study is the first to demonstrate higher levels of both perceived and performance fatigability in individuals with OSA as compared to non-OSA adults. Fatigue is often cited as common sequelae of OSA in sleep literature [3]. However, the traditional method of measuring this important clinical symptom through subjective report of perceived feeling of tiredness introduces response bias and does not account for the physical activity level of the individual making it difficult to draw comparison across populations. The concept of fatigability overcomes this limitation by normalizing the change in perceived feeling of tiredness or change in performance of a task to the metabolic demand of the activity [24]. Thus, the quantitative measurement of fatigability in OSA provides a better understanding of the phenomenon of fatigue and could be used as a more meaningful tool for assessing the effectiveness of the treatment in clinical settings.

In the present study, both perceived and performance fatigability were significantly associated with microvascular function as measured during PORH by NIRS. Our results suggest that impaired microvascular function in OSA attenuates the perfusion and oxygenation of skeletal muscle, which may predispose these individuals to fatigue and lower physical activity participation. The reduced maximal vasodilator capacity and the prolonged time to achieve maximal vasodilation in skeletal muscle microcirculation may result in an imbalance between the demand and supply of oxygen to the exercising muscle, thus limiting performance and precipitating greater fatigability.

Post-Occlusive reactive hyperemia and microvascular function

The PORH describes a transient increase in blood flow following a brief arterial occlusion and serves as an index of reactivity and maximal vasodilator capacity of the microcirculatory system [31]. The attenuated PORH response in skeletal muscle found in individuals with OSA may be explained by various pathophysiological mechanisms associated with the disorder. First, as assessed in rat models, intermittent hypoxia (IH), a hallmark of sleep apnea, may result in an increase in myogenic tone secondary to enhancement of the vasoconstrictor activity of endothelin [32]. Additionally, endothelial dysfunction potentially caused by oxidative stress, inflammation, and chronic IH may amplify the myogenic arteriolar constriction of skeletal muscles in OSA [33]. Moreover, a reduction in the calcium sparks released by vascular smooth muscles found in IH-exposed rats has been demonstrated to inhibit depolarization, thereby increasing vascular myogenic tone [34]. Therefore, alterations in myogenic tone could substantially change the architectural and tensile properties of the microvasculature after IH exposure and explain the possibility of an impaired PORH response in our group of adults with OSA who experienced chronic IH. Second, it is well documented that individuals with OSA have impaired metabolic vasodilator pathways [35]. Lower nitric oxide bioavailability, elevated C-reactive protein, decreased expression of endothelial nitric oxide synthase [36] and abnormally increased plasma endothelin [37] in OSA may cause an imbalance in the microvascular milieu. The diminished PORH response may thus, be partially attributed to a disproportion in the vasoactive mediators, impeding the endothelial-dependent metabolic vasodilation pathway. Third, because of prolonged apnea- and hypopnea-induced arousal and chronic nighttime activation of the SNS, several studies have demonstrated daytime SNS hyperactivity, possibly causing endothelial dysfunction and increasing the risk of cardiovascular disease in OSA [36, 38, 39]. The SNS stimulation causes vasoconstriction in skeletal muscle. Therefore, the elevated SNS activity could result in heightened vasoconstriction in the gastrocnemius muscle and disrupt the blood flow response to arterial occlusion in our group of participants with OSA.

Our findings of attenuated PORH response in OSA are in accordance with the work of Imadojemu and colleagues who demonstrated a reduced peak reactive hyperemic forearm blood flow measured by venous-occlusion plethysmography in adults with OSA [40]. However, the application of occlusion at supra-systolic pressure in their experiment violated one of the four basic assumptions of venous-occlusion plethysmography technique of not affecting the arterial pressure or inflow by the pneumatic cuff [41]. In contrast, studies investigating the effects of systemic hypoxia on blood flow report a preserved vasodilatory response in individuals with OSA when compared to the healthy controls [36, 42]. The discrepancy in these findings could be attributed to the methodological differences and the type of vasodilator stimulus used (limb ischemia vs. systemic hypoxia).

Furthermore, a few studies assessing cutaneous microvasculature employing acetylcholine iontophoresis have reported impaired microvascular endothelial function, as demonstrated by reduced blood flow in OSA [13, 43, 44]. However, it is important to consider some key points when interpreting these findings. First, the mechanisms regulating the microvascular function vary substantially across vascular beds, and studying human cutaneous circulation as a representative model of generalized microvascular function is debatable [45]. The response mechanism of vasodilation or vasoconstriction in cutaneous circulation could not be isolated to vascular-signaling pathways. Given a high vasodilatory reserve capacity, influence of thermal feedback, and non-specificity of vascular response, skin microcirculation measurements may not be an appropriate surrogate measure of microvascular function in skeletal muscle [46]. Second, the assumption that acetylcholine iontophoresis is a specific test for skin endothelial function is questionable. It has been demonstrated that acetylcholine mediated cutaneous vasodilation remains unchanged after nitric oxide synthase inhibition and induces axon reflex that contributes to the skin blood flow [47]. Third, the regulation of blood flow in the microcirculation is reflected by the functional interactions between skeletal muscle fibers, and the respective smooth muscle cells, endothelial cells and neural projections which comprise and regulate the vascular supply [48]. However, previous studies have examined only the endothelial-dependent function of the microvasculature in adults with OSA, which limits our understanding of the overall microvascular function. Therefore, our findings add to the available knowledge regarding the responses of skeletal muscle microvasculature to vasodilatory stimulus in this subset of population.

In addition, evaluation of the slope of the decline of Δ[HbO₂] trace during arterial occlusion demonstrated that the participants in the OSA group have a blunted slope as compared to those in the non-OSA group (0.031 ± 0.01 vs 0.044 ± 0.02, p = 0.022). A blunted slope indicates a slower rate of decline in oxyhemoglobin concentration, suggesting a reduced rate of oxygen utilization during periods of restricted blood flow. This may reflect reductions in muscle oxidative capacity and/or to impaired oxygen delivery and utilization, which could occur due to alterations in the skeletal muscle and its microvasculature in OSA secondary to chronic intermittent hypoxia and sleep fragmentation associated with the condition [5, 12]. This finding is further supported by our previously published results of a reduced peak oxygen consumption (VO₂ peak) in the OSA group compared to the non-OSA group, which reflects a diminished overall aerobic capacity [49] and is indicative of impaired aerobic endurance and reduced muscle oxidative capacity. This decreased aerobic capacity observed in the OSA group can lead to a compromised ability to sustain high-intensity exercise and appropriate energy production. This, in turn, is likely to affect muscle oxygenation, as indicated by the blunted Δ[HbO₂] slope during arterial occlusion. Specifically, impaired aerobic capacity can result in decreased oxidative enzyme activity and lower mitochondrial density, which collectively contribute to reduced oxygen extraction and utilization by the muscles [50]. Consequently, individuals with OSA may experience a more rapid onset of fatigability and reduced endurance during performance of physical activities. [27].

In summary, despite an adaptive angiogenic response to chronic exposure to intermittent hypoxia a lower peak hyperemic response suggests a functional impairment and incompetence of microvasculature that might indicate a perfusion deficit or an impaired blood flow response to skeletal muscle contractions in adults with OSA, thus with a likely abnormal response to metabolic demand and impaired oxidative capacity of skeletal muscle resulting in greater fatigability.

Limitations

There are a few limitations to the present study. First, although the contribution of skin overlying muscle has been shown to be less than 5% of the total NIRS signal [51], we could not estimate the influence of cutaneous circulation on skeletal muscle microvascular assessment. Second, we had a heterogeneous sample with an unequal distribution for the presence of hypertension and CPAP use in the OSA group. Although we did not find any significant effect of hypertension and hours of CPAP use on microvascular function in our study sample of participants with OSA, caution should be used when generalizing the findings of this study given the small sample of 6 out 13 individuals used CPAP. Third, because polysomnograms in clinical settings are typically re-evaluated after five years, we included OSA participants who had an OSA diagnosis made within five years of their enrollment in the study. Thus, we could not comment on acute and chronic effects of OSA on microvascular function. However, it is seen that this disorder remains undiagnosed for many years, thus this is a practical limitation to the study. Lastly, despite efforts to match our groups on weight distribution, participants in the OSA group had higher BMI than participants in the non-OSA group. Nonetheless, the inclusion of BMI as a covariate in all the analyses would possibly have mitigated the influence of BMI on the outcomes of our study.

Conclusion

In the present study, we report a greater fatigability in individuals with OSA compared to healthy adults. In addition, we observed a diminished degree of maximal vasodilatory capacity after arterial occlusion in these participants, indicating impaired microvascular reactivity and microcirculatory autoregulation. Likewise, the presence of lower oxidative metabolism of the skeletal muscle in OSA predisposes these individuals to a reduced oxygen extraction capacity. Collectively, these results suggest that negative modifications in the microvascular milieu may significantly contribute to greater fatigability in adults with OSA indicating that microvascular function might be a potential modulator of fatigability in this clinical population.