Global Reach 2018: sympathetic neural and hemodynamic responses to submaximal exercise in Andeans with and without chronic mountain sickness

Abstract

Andeans with chronic mountain sickness (CMS) and polycythemia have similar maximal oxygen uptakes to healthy Andeans. Therefore, this study aimed to explore potential adaptations in convective oxygen transport, with a specific focus on sympathetically mediated vasoconstriction of nonactive skeletal muscle. In Andeans with (CMS⁺, n = 7) and without (CMS⁻, n = 9) CMS, we measured components of convective oxygen delivery, hemodynamic (arterial blood pressure via intra-arterial catheter), and autonomic responses [muscle sympathetic nerve activity (MSNA)] at rest and during steady-state submaximal cycling exercise [30% and 60% peak power output (PPO) for 5 min each]. Cycling caused similar increases in heart rate, cardiac output, and oxygen delivery at both workloads between both Andean groups. However, at 60% PPO, CMS⁺ had a blunted reduction in Δtotal peripheral resistance (CMS⁻, −10.7 ± 3.8 vs. CMS⁺, −4.9 ± 4.1 mmHg·L⁻¹·min⁻¹; P = 0.012; d = 1.5) that coincided with a greater Δforearm vasoconstriction (CMS⁻, −0.2 ± 0.6 vs. CMS⁺, 1.5 ± 1.3 mmHg·mL⁻¹·min⁻¹; P = 0.008; d = 1.7) and a rise in Δdiastolic blood pressure (CMS⁻, 14.2 ± 7.2 vs. CMS⁺, 21.6 ± 4.2 mmHg; P = 0.023; d = 1.2) compared with CMS⁻. Interestingly, although MSNA burst frequency did not change at 30% or 60% of PPO in either group, at 60% Δburst incidence was attenuated in CMS⁺ (P = 0.028; d = 1.4). These findings indicate that in Andeans with polycythemia, light intensity exercise elicited similar cardiovascular and autonomic responses compared with CMS⁻. Furthermore, convective oxygen delivery is maintained during moderate-intensity exercise despite higher peripheral resistance. In addition, the elevated peripheral resistance during exercise was not mediated by greater sympathetic neural outflow, thus other neural and/or nonneural factors are perhaps involved.

NEW & NOTEWORTHY During submaximal exercise, convective oxygen transport is maintained in Andeans suffering from polycythemia. Light intensity exercise elicited similar cardiovascular and autonomic responses compared with healthy Andeans. However, during moderate-intensity exercise, we observed a blunted reduction in total peripheral resistance, which cannot be ascribed to an exaggerated increase in muscle sympathetic nerve activity, indicating possible contributions from other neural and/or nonneural mechanisms.

INTRODUCTION

High-altitude (>2,500 m) residents have undergone various genetic, autonomic, and cardiovascular adaptations that have allowed them to live, work, and exercise at extreme altitudes for generations (1–5). However, some populations, due to their relatively short genetic timeline living at altitude [i.e., Andeans, ∼10,000 year compared with Tibetans, ∼40,000 year (6)], are susceptible to the development of chronic mountain sickness (CMS). This condition is primarily characterized by excessive levels of hemoglobin, hematocrit, and blood viscosity, also known as polycythemia (7–9). Several articles that have defined and evaluated CMS have commented that Andeans with CMS often self-report symptoms of exercise intolerance (10–14). Moreover, because of the excessive erythrocytosis and high blood viscosity in CMS, a decrease in exercise capacity is expected based on the optimal hematocrit hypothesis (15).

However, despite the alleged evidence (10–14) and theoretical predictions (15), maximal oxygen uptake is similar in Andeans suffering from CMS (CMS⁺) compared with otherwise healthy Andeans (CMS⁻) (16). Indeed, data from our cohort confirmed this observation but observed that Andeans with CMS had an attenuated fall in total peripheral resistance and a greater vasoconstrictor response in nonactive skeletal muscle (forearm) during moderate-intensity exercise (17). During submaximal cycling exercise that involves a large muscle mass, appropriate autonomic responses are necessary to redistribute cardiac output (Q̇c) and match oxygen delivery to metabolic demand within the working skeletal muscles (18, 19). Our previous data identified that there were no differences in the sensitivity of vascular adrenergic receptors in CMS. Hence, the greater vasoconstrictor response may be due to an augmented sympathetic outflow during exercise. Indeed, this general pathological sequalae would be reminiscent of cardiovascular diseases at sea level, such as essential hypertension, where higher muscle sympathetic nerve activity (MSNA) (20) and impaired vascular function (21) contribute to an exaggerated blood pressure response to exercise.

When the contribution of MSNA to vasoconstriction during exercise is examined, the intensity domain is an important factor. For example, Saito and colleagues (19) demonstrated that during light intensity cycling exercise, MSNA burst frequency was either reduced or unchanged, despite increases in heart rate (HR) and blood pressure, whereas moderate to high-intensity cycling significantly elevated MSNA burst frequency. Interestingly, acute hypoxia causes an augmented MSNA burst frequency response to even light intensity (40% V̇o_2peak) exercise, although the diastolic blood pressure response is reduced (22). Therefore, with previous studies suggesting an unchanged sensitivity of vascular adrenergic receptors in CMS⁺, it could be hypothesized that even light-intensity exercise could augment sympathetic outflow and result in a greater total peripheral resistance (TPR) and a hypertensive response to exercise. Furthermore, the increase in cardiac afterload during exercise would be expected to impair systolic ejection and stroke volume (SV) even at lighter exercise intensities.

Given the overview above, the aims of this study were to identify whether exercise causes a greater increase in muscle sympathetic nerve activity in CMS⁺ residing in Cerro de Pasco (∼4,300 m) and its relation to the previously observed increase forearm resistance during exercise. Moreover, the impact on components of convective oxygen delivery (i.e., Q̇c, HR, SV, and global oxygen delivery) and hemodynamics (systolic and diastolic arterial blood pressure, and total and local vascular resistances) during multiple submaximal exercise intensities was measured to elucidate a more complete picture of hemodynamic control in this unique population. We hypothesized that CMS⁺ will have an augmented sympathetic outflow during both light and moderate-intensity cycling that causes a greater increase in blood pressure that would impair the expected increase in SV.

METHODS AND MATERIALS

Ethical Approval

The Clinical Research Ethics Board at the University of British Columbia (CREB ID H18-01404) and the Universidad Peruana Cayetano Heredia Comité de Ética (CIEH-UPCH No. 101686) approved all experimental procedures and all protocols were performed in adherence with the principles of the Declaration of Helsinki (except registration in a database). All participants provided verbal and written informed consent after a Spanish translator had thoroughly explained the experimental protocol. This investigation was part of the Global REACH expedition to the Universidad Peruana Cayetano Heredia’s Instituto de Investigacions de Altura (∼4,300 m; Cerro de Pasco, Peru) between June and July 2018. Several participants volunteered for multiple studies conducted throughout the expedition, outlined in a recent overview article (23). However, the a priori research questions in the current investigation were addressed before or at least 24 h after participation in any other study that would alter resting hemodynamics. Some results from this investigation, described below, have been reported elsewhere by our group (17); however, the present investigation addressed its own distinct a priori research question.

Participants

Andean participants (n = 16) were recruited for the current study, including nine healthy CMS⁻ (age, 44 ± 15 yr; weight, 62.1 ± 6.9 kg; height, and 1.59 ± 0.04 m) and seven CMS⁺ (age, 43 ± 13 yr; weight, 76.0 ± 10.8 kg; and height, 1.64 ± 0.03 m) participants. Since CMS primarily affects male Andeans and postmenopausal female Andeans, only males were recruited for the current investigation to avoid any confounding effects of age (7, 9, 24). Participants were included if they were between the ages of 18 and 60 yr, without any medical disease or a history of working in the local mines (e.g., lead, cobalt, and sulfur mines) as described in the experimental overview (23). No participants were taking any medication before or during participation. Cerro de Pasco (4,300 m) was chosen because of the high prevalence of CMS where 15% of men aged 30–39 yr and 34% aged 60–69 yr develop CMS (7, 9, 24, 25). All participant characteristic values are presented in Table 1.

Table 1.

Subject demographics of CMS⁻ and CMS⁺

Subjects, n = 16	CMS−, n = 9	CMS+, n = 7	P Value	Cohen’s d
Age, yr	44 ± 15	43 ± 13	0.659	0.7
Weight, kg	62.1 ± 6.9	76.0 ± 10.8	0.023	1.6
Height, m	1.59 ± 0.04	1.64 ± 0.03	0.016	1.2
BMI, m·kg−1	24.7 ± 2.9	28.3 ± 3.8	0.039	1.1
Qinghai CMS score	2 ± 1	7 ± 3	0.0002	2.8
V̇o2peak absolute, mL·min−1	2,082.4 ± 415.1	1,973.2 ± 549.2	0.681	0.2
V̇o2peak relative, mL·kg−1·min−1	33.5 ± 7.5	28.8 ± 10.7	0.351	0.5
Peak workload, W	162 ± 24	149 ± 28	0.148	0.5
30% PPO, W	48.5 ± 7.1	44.6 ± 8.5	0.203	0.5
60% PPO, W	97.0 ± 14.3	89.2 ± 16.9	0.148	0.5
$\text{S}{\text{p}}_{{\text{O}}_2}$, %	87.8 ± 3.7	83.0 ± 2.3	0.014	1.5
$\text{S}{\text{a}}_{{\text{O}}_2}$, %	85.8 ± 3.8	82.3 ± 2.8	0.007	1.6
$\text{P}{\text{a}}_{\text{O}}{}_{{}_2}$, mmHg	50.0 ± 3.9	45.4 ± 3.0	0.029	1.3
Hb, g·dL−1	19.4 ± 1.2	22.7 ± 1.6	<0.0001	3.0
Hct, %	56.0 ± 4.3	67.1 ± 7.1	0.002	2.0
Viscosity, cP	6.1 ± 0.8	8.0 ± 0.6	0.0002	2.6

Values are means ± SD; n = 16 total Andean subjects, n = 9 healthy Andeans (CMS⁻), and n = 7 Andeans with polycythemia (CMS⁺). BMI, body mass index; Hb, hemoglobin; Hct, hematocrit; HR, heart rate; $\text{P}{\text{a}}_{\text{O}}{}_{{}_2}$, resting arterial partial pressure of oxygen; PPO, peak power output; $\text{S}{\text{a}}_{{\text{O}}_2}$, arterial oxygen saturation; $\text{S}{\text{p}}_{{\text{O}}_2}$, peripheral oxygen saturation; V̇o_2peak, peak oxygen uptake. Statistical comparisons performed using two-tailed unpaired t test (CMS⁻ vs. CMS⁺). Effect sizes reported as Cohen’s d.

Qinghai CMS Assessment Questionnaire and Polycythemia Diagnosis

The presence and severity of CMS within the Andeans were assessed using the Qinghai CMS assessment questionnaire (7). The questionnaire is based on the presence and severity of clinical signs and symptoms, with hemoglobin (Hb) concentration and cyanosis as the single objective parameters. Other clinically assessed signs and symptoms include breathlessness, sleep disturbance, dilatation of veins, paresthesia, headache, and tinnitus. Each symptom is rated from 0 (i.e., absent of symptom) to 3 (i.e., severe symptom) and the presence of increased hemoglobin (>21 g·dL⁻) adds 3 to the cumulative score. Diagnosis of CMS based on the total score is as follows: absent (0–5), mild (6–10), moderate (11–14), and severe (>15) (7). In the current cohort, all participants labeled as CMS⁺ had both a hemoglobin ≥21 g·dL⁻¹ and a total clinical score >6. Healthy Andeans were selected based on normal hemoglobin values of <21 g·dL⁻¹ and an absent score on CMS questionnaire.

Experimental Protocol

The general experimental design schematic is presented in Fig. 1. All participants attended the laboratory on two occasions: 1) preliminary visit for maximum cardiopulmonary exercise testing and 2) experimental visit using submaximal cycling exercise intensities. Participants arrived at the laboratory for both testing days having abstained from exercise and alcohol for 24 h, caffeine for 12 h, and having consumed a light meal 4 h before testing. Each testing day was separated by at least 24 h. All experimental testing took place in Cerro de Pasco, Peru (∼4,300 m).

Figure 1.

Schematic outline of the experimental protocol. Instrumentation and catheterization infer the placement of the cardiovascular hemodynamic measurement devices, muscle sympathetic nerve activity (MSNA) electrodes with brachial artery catheter of the participants. Measurements of MSNA are denoted by black arrows and were analyzed during the final minute of baseline and light and moderate exercise intensities, as indicated by the gray line. Arterial blood samples are denoted in red and were taken at baseline and during last 30 s of each exercise stage. Each exercise stage lasted 5 min and occurred individually in succession. Time is displayed in minutes. Subject sample size total, n = 16; CMS⁻, n = 9; CMS⁺, n = 7. MSNA was measured in a sample size of n = 13; CMS⁻, n = 7; CMS⁺, n = 6. CMS, chronic mountain sickness; PPO, peak power output.

Maximal exercise protocol.

During the preliminary visit, all participants performed a stepwise cardiopulmonary exercise test on a custom-built semirecumbent cycle ergometer (Corival Pediatric, Lode; Lode, Groningen, The Netherlands) until volitional exhaustion. After a 2-min warmup (20 W), the test began at 40 W and gradually increased by 20 W every minute. From peak power output (PPO), relative workloads of 30% and 60% TPW were calculated and chosen as the low and moderate workloads for the experimental protocol. At the end of each exercise test, V̇o_2peak was taken as the last 30-s average and presented as both absolute (mL·min⁻¹) and relative (mL·kg⁻¹·min⁻¹) terms.

Submaximal exercise protocol.

Upon arrival for the experimental visit, participants were asked to lie on the semirecumbent cycle ergometer previously used for cardiopulmonary exercise testing. After catheterization and instrumentation, a minimum of 45 min of rest occurred to allow for hemodynamic stabilization and adjustment to the semirecumbent position, Thereafter, a blood sample was drawn and baseline hemodynamic measurements were recorded, which included electrocardiogram, arterial blood pressure, muscle sympathetic nerve activity (MSNA), and forearm blood flow.

Participants performed cycling at a light exercise intensity of 30% PPO (48.5 ± 7.1 W) and at a moderate intensity of 60% PPO (97.0 ± 14.3 W). Participants performed each cycling exercise intensity for 5 min beginning with the lowest relative workload for each participant to establish a hemodynamic steady state (18, 19, 26). We chose both the light relative cycling exercise workload of 30% PPO to match hemodynamic responses to activities of daily living (i.e., walking) (27) and the moderate-intensity workload of 60% PPO to elicit a robust sympathetic and cycling exercise pressor response (18, 19, 26). MSNA was recorded during all relative intensities. Beat-by-beat HR and arterial blood pressure were recorded continuously throughout each exercise intensity and arterial blood samples were drawn during the final 30 s of each exercise intensity. Forearm blood flow was only obtained at 60% PPO because of the expected and previously shown vasoconstrictor responses at this intensity (17, 28). Participants’ arms were abducted at 90° laterally and supported at the level of the heart.

Arterial Catheterization, Arterial Blood Pressure, Blood Samples, and Heart Rate

Under local anesthesia (2% lidocaine), 20-gauge, 7.6-cm catheter (Arrow, Markham, ON, Canada) was inserted under aseptic conditions into the brachial artery of the nondominant arm using ultrasound guidance. The catheter was connected to an arterial blood sample kit (VAMP adult; Edwards Lifesciences, Irvine, CA) for repeated arterial blood sampling and flushing with 0.9% heparinized saline. The VAMP blood sampling kit was connected to a blood pressure amplifier (ADInstruments, FE117, Sydney, Australia) and was zeroed at the level of the right atrium as estimated from the fourth intercostal space for continuous measurements of arterial blood pressure. Mean (MAP), systolic (SAP), and diastolic (DAP) arterial blood pressure were derived from the arterial pressure waveform (29) and calibrated using a water manometer. During cycling exercise, the brachial arterial waveform was used to calculate SV via the Model-flow technique and LabChart Noninvasive Cardiac Output software analysis (LabChart NICO extension v. 8.1, ADInstruments, Sydney, Australia). From the product of HR and SV, Q̇c (L·min⁻¹) was calculated. TPR was derived from the quotient of MAP and Q̇ and expressed as mmHg·L⁻¹·min⁻¹ (30). Total peripheral conductance (TPC) was derived as a quotient of Q̇c and MAP and expressed L⁻¹·min⁻¹·100 mmHg (30). Forearm vascular resistance (FVR), measured at the brachial artery, was calculated as a quotient of MAP and forearm blood flow (FBF), measured using duplex Doppler ultrasound (Vivid 7, General Electric, Milwaukee, WI) (FVR = MAP/FBF), and expressed as mmHg·mL⁻¹·min⁻¹. Forearm vascular conductance (FVC) was calculated as a quotient of FBF and MAP (FVC = FBF/MAP) and expressed as mL⁻¹·min⁻¹·100 mmHg. Both FVR and FVC were published elsewhere by our group (17). Cardiac work was calculated using Q̇c and SAP (cardiac work = Q̇c × SAP) and expressed as L·mmHg⁻¹·min⁻¹. Arterial blood samples were drawn using a 1.0 mL safePICO syringe and the partial pressure of oxygen ($\text{P}{\text{a}}_{\text{O}}{}_{{}_2}$), oxygen saturation ($\text{S}{\text{a}}_{{\text{O}}_2}$), and total hemoglobin (Hb) along with various skeletal muscle metabolic by-products (i.e., Lactate and pH) were determined via an ABL90 FLEX blood gas analyzer (Radiometer Medical ApS, Brønshøi, Denmark) configured to body temperature. Arterial oxygen content ($\text{C}{\text{a}}_{\text{O}}{}_{{}_2}$) was calculated as $\text{C}{\text{a}}_{\text{O}}{}_{{}_2}$ = (Hb × 1.31) × ($\text{S}{\text{a}}_{{\text{O}}_2}$ × 0.01) + ($\text{P}{\text{a}}_{\text{O}}{}_{{}_2}$ × 0.0225) with 1.31 as Hüfner’s constant and 0.0225 as the solubility coefficient of oxygen at body temperature (31) and expressed as mL·dL⁻¹. Whole body global Do₂ was calculated as Do₂ = (Q̇c × $\text{C}{\text{a}}_{\text{O}}{}_{{}_2}$) × 10 and expressed as mL·min⁻¹. Heart rate was measured via three-lead electrocardiogram (ADInstruments, Sydney, Australia). Venous whole blood viscosity was measured in duplicate at a shear rate of 225 s⁻¹ at 37°C using a cone and plate viscometer (DV2T Viscometer, Brookfield Amtek) and circulating heating water bath (TC-150, Brookfield Amtek) (32, 33).

Muscle Sympathetic Nerve Activity

Multiunit MSNA were measured in both Andean participant groups during rest, light, and moderate relative intensity cycling exercise (i.e., 30 and 60% PPO). Acceptable MSNA recordings were established within 13 Andean subjects (CMS⁻, n = 7; CMS⁺, n = 6), with three participants excluded because of background noise and movement artifacts. Neural activity was recorded from the radial nerve by an experienced microneurographer (GM) using standardized ultrasound-guided microneurographic techniques (18, 34). Nerve signals were amplified (1,000× preamplifier and 100× variable gain isolated amplifier, Neuroamp Ex, ADInstruments, Sydney, Australia) bandpass filter (300–5,000 Hz), rectified, and integrated (time decay constant = 0.1 s) (LabChart Pro V8.3.1, ADInstruments, Sydney, Australia). Multiunit bursts of MSNA were identified using standardized guidelines (i.e., 3:1 signal-to-noise ratio) by a single trained observer (GM) (35, 36). MSNA was averaged over the final 1 min when subjects had reached steady-state exercise and quantified as burst frequency (bursts·min⁻¹) and burst incidence (bursts·100 heartbeats⁻¹). In addition, as an index of the transduction of MSNA into vascular tone, we calculated the change in the nonactive forearm skeletal muscle vascular resistance over the change in MSNA (ΔFVR/ΔMSNA) and expressed as mmHg·mL⁻¹·min⁻¹/bursts⁻¹·min⁻¹.

Data Acquisition and Analyses

All cardiovascular variables were sampled at 1 KHz, with MSNA sampled at 10 KHz, via an analog-to-digital converter (PowerLab, 16/30, ADInstruments, Sydney, Australia), displayed on LabChart (v. 8.1; ADInstruments, Sydney, Australia) and analyzed offline.

Absolute changes in the hemodynamic responses to each exercise intensity were calculated as:

ΔTPR = TPR_30%/60% – TPR_baseline

The same calculation was made for TPC, MAP, SAP, DAP, HR, SV, FVR, and global Do₂.

Absolute and relative change in autonomic response to both low and moderate intensity exercise was calculated as:

ΔMSNA BF = MSNA BF_30%/60% – MSNA BF_baseline

%ΔMSNA BF = (MSNA BF_30%/60% – MSNA BF_baseline)/MSNA BF_baseline·100

All absolute hemodynamics and MSNA data were compared between CMS⁻ and CMS⁺ during exercise using a 2 × 3 repeated-measured analysis of variance (ANOVA) (group, CMS⁻ vs. CMS⁺ × time, baseline, 30%, 60%), and when significant interactions were seen (P < 0.05), post hoc planned contrasts between CMS⁻ and CMS⁺ were performed using the Fisher’s least significant difference (LSD) test. Our data support the previous observation that V̇o_2peak is similar between CMS⁻ and CMS⁺ (16). Thus, all workloads were statistically treated independently, and only planned follow-up contrasts between CMS⁻ and CMS⁺ at each workload were compared. Data normality was confirmed using a Shapiro–Wilks test on all absolute data, when normality was not reached, a nonparametric Mann–Whitney test was used to detect changes. All change scores for exercise hemodynamic data were analyzed using two-tailed unpaired t tests with a Welch correction to accommodate unequal variances (CMS⁻ vs. CMS⁺). All statistical analyses were completed using Prism GraphPad (v. 8, GraphPad Software, San Diego, CA) and are reported as means ± SD. Statistical significance was accepted at P < 0.05.

RESULTS

Subject Characteristics and Maximum Cardiopulmonary Exercise Testing

Table 1 provides subject characteristics and maximum cardiopulmonary exercise test values for both CMS⁻ and CMS⁺ Andean groups. On average, CMS⁺ subjects were heavier (P = 0.023, d = 1.6) and taller (P = 0.016, d = 1.2) resulting in a slightly larger BMI (P = 0.039, d = 1.1) when compared with CMS⁻ (Table 1). By design, the Qinghai CMS questionnaire score was significantly higher in CMS⁺ when compared with CMS⁻ (P = 0.0002, d = 2.8). Furthermore, as expected, at rest, CMS⁺ had elevated hemoglobin (P < 0.0001, d = 3.0), hematocrit (P < 0.0001, d = 2.0) with reduced resting $\text{S}{\text{a}}_{{\text{O}}_2}$ (P = 0.007, d = 1.6) and $\text{P}{\text{a}}_{\text{O}}{}_{{}_2}$ (P = 0.029, d = 1.3) when compared with CMS⁻ (Table 1). In correspondence to both elevated hemoglobin and hematocrit, we found that blood viscosity was elevated in CMS⁺ (P = 0.0002, d = 0.9; Table 1). Maximal oxygen uptake, in both absolute (P = 0.681, d = 0.2) and relative (P = 0.351, d = 0.5) terms, and PPO (P = 0.148, d = 0.5) did not differ between CMS⁻ and CMS⁺.

Convective Oxygen Delivery during Cycling Exercise

Light intensity exercise.

All CMS⁻ and CMS⁺ variables related to convective oxygen delivery during cycling exercise at 30% PPO are presented in Table 2. Cycling at 30% PPO caused similar increases in ΔHR (P = 0.150, d = 0.8), ΔSV (P = 0.287, d = 0.5), ΔQ̇c (P = 0.361, d = 1.0), and ΔDo₂ (P = 0.853, d = 0.1) between CMS⁻ and CMS⁺ (Table 2; Fig. 2, A–D).

Table 2.

Convective oxygen delivery between CMS⁻ and CMS⁺ during light and moderate intensity submaximal cycling exercise

Subjects, n =16	Baseline	30%	60%	2 × 3 ANOVAP Value
HR, beats/min
CMS−	70.6 ± 11.2	107.0 ± 9.0	131.0 ± 13.4	0.384
CMS+	69.9 ± 15.3	98.7 ± 14.7	121.7 ± 18.5
Cohen’s d	0.1	0.7	0.6
SV, mL
CMS−	52.7 ± 13.4	56.2 ± 16.7	59.0 ± 17.4	0.088
CMS+	58.0 ± 10.9	59.1 ± 11.9	58.0 ± 14.5
Cohen’s d	0.4	0.2	0.1
Q̇C, L·min−1
CMS−	3.7 ± 1.0	6.0 ± 1.8	7.7 ± 2.4	0.442
CMS+	4.1 ± 1.2	5.9 ± 1.9	7.2 ± 2.9
Cohen’s d	1.4	1.3	1.2
Cardiac work, L·mmHg−1·min−1
CMS−	443.0 ± 152.0	913.2 ± 297.7	1,412.0 ± 506.2	0.936
CMS+	473.6 ± 141.7	895.1 ± 370.6	1,438.7 ± 656.4
Cohen’s d	0.2	0.1	0.1
$\text{C}{\text{a}}_{\text{O}}{}_{{}_2}$, mL·dL−1
CMS−	21.5 ± 1.1	20.9 ± 1.3	20.7 ± 1.5	0.315
CMS+	24.8 ± 0.8	24.9 ± 1.0	24.9 ± 1.6
Cohen’s d	2.3	3.4	2.8
Do2, mL·min−1
CMS−	812.5 ± 234.1	1,304.9 ± 399.2	1,659.3 ± 546.0	0.115
CMS+	986.8 ± 294.1	1,439.1 ± 440.7	1,741.7 ± 631.8
Cohen’s d	0.6	0.3	0.1
$\text{S}{\text{a}}_{{\text{O}}_2}$, %
CMS−	85.8 ± 3.8	81.4 ± 5.0	80.2 ± 5.3	0.189
CMS+	82.3 ± 2.8	80.7 ± 2.4	79.6 ± 2.8
Cohen’s d	1.6	0.2	0.2
$\text{P}{\text{a}}_{\text{O}}{}_{{}_2}$, mmHg
CMS−	50.0 ± 3.9	46.6 ± 5.0	44.7 ± 5.8	0.745
CMS+	45.4 ± 3.0	44.8 ± 3.2	43.3 ± 3.5
Cohen’s d	1.0	0.4	0.3
$\text{P}{\text{a}}_{\text{C}\text{O}}{}_{{}_2}$, mmHg
CMS−	30.8 ± 5.3	30.5 ± 3.4	28.3 ± 4.0	0.946
CMS+	32.6 ± 2.3	32.1 ± 3.9	29.6 ± 2.9
Cohen’s d	0.4	0.4	0.4
pH
CMS−	7.5 ± 0.1	7.3 ± 0.1	7.4 ± 0.1	0.078
CMS+	7.4 ± 0.1	7.4 ± 0.0	7.4 ± 0.1
Cohen’s d	1.1	0.7	0.2
Lactate, mmol·L−1
CMS−	1.0 ± 0.5	2.7 ± 0.9	4.2 ± 1.5	0.338
CMS+	1.1 ± 0.5	3.4 ± 1.4	5.2 ± 1.7
Cohen’s d	0.1	0.6	0.7

Values are means ± SD; n = 16 total Andean subjects, n = 9 healthy Andeans (CMS⁻), and n = 7 Andeans with polycythemia (CMS⁺). $\text{C}{\text{a}}_{\text{O}}{}_{{}_2}$, arterial oxygen content; Do₂, oxygen delivery; HR, heart rate; $\text{P}{\text{a}}_{\text{O}}{}_{{}_2}$, arterial blood partial pressure of oxygen; $\text{P}{\text{a}}_{\text{C}\text{O}}{}_{{}_2}$, arterial blood partial pressure of carbon dioxide; Q̇_C, cardiac output; $\text{S}{\text{a}}_{{\text{O}}_2}$, arterial blood oxygen saturation; SV, stroke volume.. Statistical comparisons performed using 2 × 3 repeated-measures analyses of variance (ANOVA) with the interaction reported. Effect sizes reported as Cohen’s d.

Figure 2.

Changes in convective oxygen delivery during submaximal cycling exercise in Andeans with and without chronic mountain sickness. A–D: absolute changes in convective oxygen delivery during light and moderate intensity [30% and 60% peak power output (PPO)] cycling comparing healthy Andeans (CMS⁻) denoted in yellow to Andeans with CMS and polycythemia (CMS⁺) denoted in blue. Individual and mean absolute change in cardiac output (Q̇_C) (A), heart rate (HR) (B), stroke volume (SV) (C), and global oxygen delivery (Do₂) (D). Statistical comparisons performed using two-tailed unpaired t tests with a Welch correction to accommodate unequal variances (CMS⁻ vs. CMS⁺). Effect sizes are reported as Cohen’s d. Subject sample size, total, n = 16; CMS⁻, n = 9; CMS⁺, n = 7. *Statistical difference P < 0.05. CMS, chronic mountain sickness.

Moderate intensity exercise.

All CMS⁻ and CMS⁺ variables related to convective oxygen delivery during cycling at 60% PPO are presented in Table 2. Moderate intensity cycling caused a slightly attenuated ΔSV in CMS⁺ (CMS⁻, 6.3 ± 7.0 vs. CMS⁺, −0.1 ± 6.6 mL; P = 0.085, d = 0.9), yet similar increases in ΔHR (P = 0.358, d = 0.5), ΔQ̇c (P = 0.388, d = 1.0), and Δglobal Do₂ (P = 0.718, d = 0.9) in both Andean groups (Table 2; Fig. 2, A–D).

Hemodynamic Responses during Cycling Exercise

Light intensity exercise.

All CMS⁻ and CMS⁺ hemodynamic variables during 30% cycling are presented in Table 3, respectively. We found that, when cycling at 30% PPO, ΔMAP (P = 0.952, d = 0.4), ΔSAP (P = 0.999, d = 0.6), and ΔDAP (P = 0.300, d = 0.6) did not differ between CMS⁻ and CMS⁺ (Table 3; Fig. 3, A–C). In addition, we observed no differences in ΔTPR (P = 0.131, d = 1.1) and ΔTPC (P = 0.156, d = 0.9) between both Andean groups at 30% cycling intensity (Table 3; Fig. 3, D and E).

Table 3.

Hemodynamics between CMS⁻ and CMS⁺ during light and moderate-intensity submaximal cycling exercise

Subjects, n = 16	Baseline	30%	60%	2 × 3 ANOVA P Value
Mean BP, mmHg
CMS−	96.9 ± 12.1	115.3 ± 16.0	122.8 ± 16.5	0.076
CMS+	90.7 ± 8.0	109.3 ± 12.2	124.8 ± 8.3
Cohen’s d	0.6	0.4	0.2
Systolic BP, mmHg
CMS−	132.8 ± 16.1	168.4 ± 24.7	182.7 ± 29.0	0.049
CMS+	129.7 ± 9.5	165.3 ± 19.4	195.8 ± 17.5
Cohen’s d	0.2	0.1	0.5
Diastolic BP, mmHg
CMS−	75.6 ± 9.3	83.9 ± 12.9	89.8 ± 10.5	0.086
CMS+	70.1 ± 6.7	82.2 ± 9.9	91.8 ± 7.0
Cohen’s d	0.7	0.2	0.2
TPR, mmHg·L−1·min−1
CMS−	28.1 ± 8.7	21.0 ± 7.5	17.5 ± 6.0	0.005
CMS+	24.1 ± 7.7	19.9 ± 5.6	19.3 ± 5.6
Cohen’s d	0.5	0.2	0.3
TPC, L−1·min−1·100 mmHg
CMS−	3.9 ± 1.2	5.4 ± 1.9	6.4 ± 2.1	0.049
CMS+	4.5 ± 1.3	5.4 ± 1.5	5.8 ± 2.1
Cohen’s d	0.5	0.02	0.3
FVR, mmHg·ml−1·min−1
CMS−	3.5 ± 3.1		4.5 ± 6.4	0.727
CMS+	2.4 ± 1.0		3.9 ± 2.1
Cohen’s d	1.1		0.6
FVC, mL·min−1·100 mmHg
CMS−	42.0 ± 21.0		45.5 ± 26.3	0.002
CMS+	48.8 ± 22.8		33.4 ± 17.7
Cohen’s d	1.4		0.5
Skin conductance, AU
CMS−	22.3 ± 34.4	50.3 ± 28.1	85.2 ± 66.8	0.739
CMS+	30.0 ± 18.4	71.2 ± 44.1	78.9 ± 47.7
Cohen’s d	0.3	0.6	0.2
Skin temperature, °C
CMS−	26.0 ± 3.5	26.1 ± 3.2	26.1 ± 3.1	0.028
CMS+	26.8 ± 2.2	25.8 ± 2.7	23.7 ± 2.4
Cohen’s d	0.3	0.1	0.8

Values are means ± SD; n = 16 total Andean subjects, n = 9 healthy Andeans (CMS⁻), and n = 7 Andeans with polycythemia (CMS⁺). AU, arbitrary units; diastolic BP, diastolic arterial blood pressure; FVC, forearm vascular conductance; FVR, forearm vascular resistance; Mean BP, mean arterial blood pressure; systolic BP, systolic arterial blood pressure; TPC, total peripheral conductance; TPR, total peripheral resistance; . Statistical comparisons performed using 2 × 3 repeated-measures analyses of variance (ANOVA) with the interaction reported. For comparisons for forearm vascular resistance and conductance, we used a 2 × 3 repeated-measures ANOVA. Effect sizes reported as Cohen’s d.

Figure 3.

Changes in arterial blood pressure responses to submaximal cycling exercise in Andeans with and without chronic mountain sickness. A–E: absolute change in arterial blood pressure responses to light and moderate intensity [30% and 60% peak power output (PPO)] cycling comparing healthy Andeans (CMS⁻) denoted in yellow to Andeans with CMS and polycythemia (CMS⁺) denoted in blue. Individual and mean absolute change in mean arterial pressure (MAP) (A), systolic arterial pressure (SAP) (B), diastolic arterial pressure (DAP) (C), total peripheral resistance (TPR) (D), and total peripheral conductance (TPC) (E). Individual absolute changes in local forearm vascular resistance (F) and conductance (G) during 60% PPO, previously published data from our group (17). Statistical comparisons performed using two-tailed unpaired t tests with a Welch correction to accommodate unequal variances (CMS⁻ vs. CMS⁺). Effect sizes are reported as Cohen’s d. Subject sample size, total, n = 16; CMS⁻, n = 9; CMS⁺, n = 7. CMS, chronic mountain sickness.

Moderate intensity exercise.

All CMS⁻ and CMS⁺ hemodynamic variables during 60% cycling exercise intensity are presented in Table 3. Moderate intensity cycling exercise resulted in a trend for a higher ΔMAP (P = 0.068, d = 0.9) and ΔSAP (P = 0.069, d = 1.0) and a significantly augmented ΔDAP (P = 0.023, d = 1.2) response in CMS⁺ compared with CMS⁻ (Table 3, Fig. 3, A–C). Furthermore, the fall in ΔTPR was significantly lower at 60% PPO in CMS⁺ (CMS⁻, −10.7 ± 3.8 vs. CMS⁺, −4.9 ± 4.1 mmHg·L⁻¹·min⁻¹; P = 0.012, d = 1.5; Table 3, Fig. 3D), which corresponded with a trending attenuation in the increase in ΔTPC (P = 0.064, d = 1.0; Table 3, Fig. 3E). This attenuation of the fall in ΔTPR and lower ΔTPC is supported by a greater increase in local forearm vascular resistance (FVR) and a reduced forearm vascular conductance (FVC) during 60% TPW (ΔFVR, P = 0.008, d = 1.7; ΔFVC, P = 0.006, d = 2.0; Fig. 3, F and G).

Autonomic Responses during Cycling Exercise

Light intensity exercise.

All CMS⁻ and CMS⁺ autonomic variables during 30% cycling are presented in Table 4. Light intensity cycling exercise elicited no change in MSNA burst frequency similarly between both Andean groups (CMS⁻, −0.7 ± 9.6 vs. CMS⁺, −1.3 ± 11.9 bursts·min⁻¹; P = 0.921, d = 0.1; Fig. 4, A and C); and because of similar elevations in HR (P = 0.150, d =.8), Δburst incidence decreased similarly between both Andean groups (CMS⁻, −11.1 ± 10.7 vs. CMS⁺, −13.2 ± 9.1 bursts·100 heartbeats⁻¹; P = 0.719, d = 0.2; Table 4, Fig. 4B).

Table 4.

Muscle sympathetic nerve activity during light and moderate submaximal cycling between CMS⁻ and CMS⁺

Subjects, n = 13	Baseline	30%	60%	Two-Way ANOVA P Value
MSNA burst frequency, burst·min−1
CMS−	23 ± 7	22 ± 9	36 ± 7	0.241
CMS+	36 ± 15	35 ± 15	41 ± 14
Cohen’s d	1.2	1.1	0.5
MSNA burst incidence, burst·100 heartbeats−1
CMS−	32 ± 12	20 ± 10	28 ± 6	0.052
CMS+	49 ± 12	36 ± 14	34 ± 8
Cohen’s d	1.5	1.4	0.8

Values are means ± SD; n = 13 total Andean subjects, n = 7 healthy Andeans (CMS⁻), and n = 6 Andeans with polycythemia (CMS⁺). MSNA, muscle sympathetic nerve activity. Statistical comparisons performed using 2 × 3 repeated-measures analyses of variance (ANOVA) with the interaction reported. Effect sizes reported as Cohen’s d.

Figure 4.

Muscle sympathetic nerve activity during submaximal cycling exercise in Andeans with and without chronic mountain sickness. Individual and mean absolute change in muscle sympathetic nerve activity (MSNA) burst frequency (A) and burst incidence (B) during low and moderate intensity [30% and 60% peak power output (PPO)] submaximal cycling comparing healthy Andeans (CMS⁻) denoted in yellow to Andeans with CMS and polycythemia (CMS⁺) denoted in blue. C: mean absolute values for MSNA burst frequency at baseline, 30%, and 60% PPO. D: individual and mean absolute change in forearm vascular resistance (ΔFVR)/ΔMSNA comparing CMS⁻ to CMS⁺ at 60% PPO. E: raw tracings of mean arterial blood pressure (MAP) and MSNA at baseline, 30% and 60% PPO in both CMS⁻ and CMS⁺. To reduce oversized analysis files, data were extracted at 0.2 s or 5 Hz from LabChart when creating the raw tracings, which is indicated by the “smoothing effect.” Statistical comparisons performed using two-tailed unpaired t tests with a Welch correction to accommodate unequal variances (CMS⁻ vs. CMS⁺) and 2 × 3 repeated-measures analysis of variance (ANOVA). Effects sizes are reported as Cohen’s d. Subject sample size, total, n = 13; CMS⁻, n = 7; CMS⁺, n = 6. *Statistical difference P < 0.05. CMS, chronic mountain sickness.

Moderate intensity exercise.

All CMS⁻ and CMS⁺ autonomic variables during 60% cycling exercise intensity are presented in Table 4. During moderate-intensity cycling exercise, Δburst frequency increased by ∼66% in CMS⁻ but only by ∼18% in CMS⁺ (CMS⁻, 13.1 ± 8.9 vs. CMS⁺, 5.3 ± 11.1 bursts·min⁻¹; P = 0.199, d = 0.8; Fig. 4, A and C). Furthermore, despite similar increases in HR with moderate intensity cycling exercise (P = 0.358, d = 0.5), Δburst incidence was reduced by ∼7% within CMS⁻ compared with the ∼31% reduction within CMS⁺ (CMS⁻, −3.7 ± 9.7 vs. CMS⁺, −15.5 ± 7.1 bursts·100 heartbeats⁻¹; P = 0.028, d = 1.4; Table 4, Fig. 4B). Raw tracings of blood pressure and MSNA neurogram within both CMS⁻ and CMS⁺ at rest. 30% and 60% cycling exercise (Fig. 4E).

When calculating an index of sympathetic vascular transduction (ΔFVR/ΔMSNA), during moderate-intensity cycling exercise, we observed that sympathetic transduction was significantly greater in CMS⁺ compared with CMS⁻ (P = 0.044, d = 1.4; Fig. 4D), indicating that a smaller sympathetic neural response is required to elicit a given change in FVR in CMS⁺.

DISCUSSION

To our knowledge, this is the first investigation to report the interaction of convective oxygen delivery, hemodynamic, and autonomic responses during submaximal cycling exercise in Andeans with and without CMS and polycythemia. We identified that during light-intensity exercise (30% PPO), there were no major differences between CMS⁻ and CMS⁺. However, moderate-intensity exercise (60% PPO) unmasked early signs of impairment in cardiac function (i.e., SV), and an exaggerated vasoconstrictor response in nonactive tissue, which caused an attenuated fall in TPR and an augmented arterial blood pressure response. Interestingly, the greater vasoconstrictor effect cannot be explained by exaggerated sympathetic neural activity directed toward the nonactive skeletal muscle as MSNA burst frequency only increased by ∼18% in CMS⁺, whereas in healthy Andeans MSNA burst frequency increased by ∼66%. Collectively, global oxygen delivery is maintained in persons with CMS, despite increased vascular resistance and reduced SV, during moderate-intensity submaximal cycling. Thus, it appears that the ability to perform submaximal exercise is maintained within Andeans who suffer from CMS and polycythemia.

Convective Oxygen Delivery during Cycling Exercise

Convective oxygen delivery is composed of Q̇c and $\text{C}{\text{a}}_{\text{O}}{}_{{}_2}$, which determines global Do₂ to the periphery during exercise (37). In the current study, we found that Q̇c and its components HR and SV were similar between CMS⁻ and CMS⁺ during light cycling exercise. At 60% PPO, Q̇c was modestly lower in CMS⁺ compared with CMS⁻. The trend toward a lower Q̇c in CMS⁺ was due to an unaltered SV (∼≤1%), whereas at the same workload, SV increased by ∼11% in CMS⁻ (Table 2; Fig. 3C). The mechanism(s) contributing to the reduction in SV during exercise is unknown but may be multifactorial. For example, individuals with CMS generally have a steeper increase in pulmonary pressure during exercise (16), which would increase right ventricular afterload and potentially impair contractility. Although recent stress-echocardiographic data do not support this hypothesis (38), exercise intensities were only prescribed at 25 W and 50 W, and our data highlight these workloads are insufficient to unmask potential impairments in cardiac function and thus SV. Both blood viscosity and $\text{C}{\text{a}}_{\text{O}}{}_{{}_2}$ increase with polycythemia whereby several studies have linked the increase in blood viscosity with a decreased Q̇c because of a rise in TPR and increased cardiac work and/or a fall in venous return (39). However, an elegant study by Lindenfeld and colleagues (40) separated the effect of viscosity and $\text{C}{\text{a}}_{\text{O}}{}_{{}_2}$ via the transfusion of packed red blood cells containing methemoglobin and the conversion of methemoglobin to oxyhemoglobin via Methylene blue. Albeit during acute polycythemia, these data imply that systemic vascular resistance is increased and Q̇c reduced during exercise primarily because of an increase in $\text{C}{\text{a}}_{\text{O}}{}_{{}_2}$ as opposed to viscosity. Alternatively, the increase in $\text{C}{\text{a}}_{\text{O}}{}_{{}_2}$ may reduce the vasodilatory stimulus within active skeletal muscle also resulting in a reduced fall in resistance and rise in Q̇c. Future research is needed to separate these mechanisms in humans.

Hemodynamic and Autonomic Responses during Cycling Exercise

At sea level, light intensity cycling (i.e., 20%–40% V̇o_2peak) decreases MSNA burst frequency due in part to cardiopulmonary loading via an increase in venous return (19, 41). In contrast, MSNA burst frequency is augmented during light intensity cycling (i.e., 40% V̇o_2peak) when performed in acute hypoxia (22) and unchanged in Andeans who are chronically adapted to hypoxia irrespective of CMS. Mechanistically it is unclear why the MSNA responses to light intensity cycling differ between normoxia, acute, and chronic hypoxemia, but a potential contributing factor is differences in oxygen carrying capacity. Acute exposure to hypoxia causes a reduction in $\text{C}{\text{a}}_{\text{O}}{}_{{}_2}$, systemic vasodilation, and exaggerated exercise hyperemia, which must be actively restrained by sympathetic activation (22, 42, 43). In contrast, oxygen-carrying capacity and Do₂ are restored in Andeans, which limits the need for compensatory vasodilation and sympathetic restraint during exercise. Although an attractive hypothesis, other factors such as baroreflex control of sympathetic activity should not be ruled out as diastolic blood pressure falls during light exercise in acute hypoxia (22), whereas it is increased in Andeans with CMS.

During higher intensity exercise, metabolic vasodilation is profound with a fall in TPR that must be restrained by sympathetic vasoconstriction in active and nonactive tissues (28, 44, 45). The increase in resistance maintains a normal blood pressure and aids the redistribution of blood volume toward active skeletal muscle (46). Moreover, it is hypothesized that when exercising after acclimatization to high altitude, due to the elevations in $\text{C}{\text{a}}_{\text{O}}{}_{{}_2}$, an increased sympathetic restraint remains necessary to match oxygen delivery with demand within active tissues (47). As reported previously in this cohort, moderate-intensity cycling caused a greater increase in FVR (an index of nonactive tissue vasoconstriction) (17), an attenuated fall in TPR, and an augmented arterial blood pressure response, specifically the diastolic blood pressure, compared with CMS⁺. In this study, our novel findings are that 1) light intensity exercise elicited no differences in the peripheral resistance and MSNA and 2) the greater increase in TPR and the hypertensive response was not due to an exaggerated sympathetic neural drive as the rise in MSNA burst frequency was lower in CMS⁺ compared with CMS⁻. Since there is no indication of exaggerated α₁-adrenergic responsiveness within CMS⁺ (17), an explanation may point toward alternative vasoconstrictor mechanisms such as circulating angiotensin II (18) or endothelin-1 during exercise. Alternatively, blood viscosity can increase vascular resistance and consequently arterial blood pressure (48), independent of vasoconstrictor mechanism(s). Interestingly, experimental polycythemia causes a disproportionate reduction in blood flow to nonactive muscle compared with exercising muscle and may provide a simple explanation (49). It must be noted that our measurement of FVR is only an index of nonactive tissue vasoconstriction, which cannot exclusively explain the increase in blood pressure during exercise. Therefore, it is likely that a greater vasoconstrictor response also occurs in other vascular beds/organs that were not measured (45).

Finally, CMS⁺ had a quantitively lower MSNA burst frequency, which did not reach the classic statistical threshold, and a statistically lower burst incidence response to moderate intensity cycling which could be due to 1) an upward resetting of the arterial baroreflex control over sympathetic neural activity (50–52) perhaps resulting in the lower MSNA burst incidence response during exercise; 2) a blunted peripheral chemoreceptor drive to hypoxemia in CMS⁺ (53–55); and/or 3) nonneural mechanisms that increase peripheral resistance (e.g., local vasoconstriction, blood viscosity) and reduce the need for sympathetic neural discharge to maintain blood pressure. Interestingly in this investigation, peripheral sympathetic transduction, indexed as the change in forearm vascular resistance for a given change in MSNA burst frequency during steady-state exercise, was greater in CMS⁺ (Fig. 4D). These data indicate that CMS⁺ had a greater vasoconstrictor response to a given change in MSNA burst frequency. However, our previous data in the same population demonstrate that α₁-adrenergic receptor responsiveness is not exaggerated in CMS⁺, and α-adrenergic restraint of nonactive tissue is not different from CMS⁻ (17). Thus, collectively these data highlight that in CMS⁺, other sympathetic vasoconstrictor mechanisms [e.g., NPY (56)] and/or local nonneural mechanisms (e.g., angiotensin II, ET-1, blood viscosity) likely contribute to increased vascular resistance during exercise.

Limitations

First, we acknowledge that TPR was indirectly assessed from the calculation of SV and Q̇c via the model flow algorithm. Yet the separate assessment of forearm vascular resistance lends support to this observation. Albeit the model flow algorithm is not the gold-standard technique to assess SV, Siebenmann et al. (57) demonstrated a strong comparison between the use of the Innocor device and the model flow algorithm when determining Q̇c during exercise. Indeed, it is highly unlikely that changes in vascular resistance localized to smaller muscle mass, such as the forearm, would cause systemic changes in arterial blood pressure alone (58), thereby indirectly indicating heightened resistance within other nonactive vascular beds, such as organs and viscera (45). Second, the differences in height, weight, and ultimately BMI between subject groups perhaps influenced the hemodynamic and autonomic responses to dynamic leg cycling exercise (59). It has previously been demonstrated that body weight and BMI can influence resting sympathetic activity by ∼10% independent of blood pressure (60) with reduced sympathetic responsiveness during sympathoexcitation via a blunted metaboreceptor responsiveness (61). Sympathetic hyperactivity within overweight individuals has been linked to alterations in arterial baroreflex response and arterial distensibility (62). Third, the current study did not assess burst amplitude or area and encompasses a relatively small sample (n = 16; CMS⁻, n = 9; CMS⁺, n = 7). Quantification of burst amplitude and total activity is extremely difficult during exercise because electromyographic efferent and afferent activity influence the baseline neurogram (63, 64). Nonetheless, our interpretation that the sympathetic nervous system is not responsible for the greater vasoconstrictor response to exercise should be treated with caution as axonal recruitment patterns (65) and neurotransmitter release is not assessed via the simplified calculation of burst frequency. Regarding sample size, the primary outcomes included gold-standard measurements of MSNA, via microelectrodes, and arterial blood pressure, via intra-arterial blood pressure recordings, yet examination of effect sizes suggests that we may be slightly underpowered to observe differences at lighter-intensity workloads (i.e., effect size 1.1 for total peripheral resistance). Fourth, the current study only assessed males within the Andean population, thus these findings cannot be generalized to females. Although female representation is substantially lacking within high-altitude research, to our knowledge, males and postmenopausal females have a higher risk of developing polycythemia and CMS compared with premenopausal females who seem to be hormonally protected from the condition (24, 66). Fifth, the CMS⁺ subjects within the current study only had mild-to-moderate subjective symptoms reported via the CMS questionnaire. However, objective measurements such as an elevated hemoglobin concentration (>21 g·dL⁻¹) and high blood viscosity were established in all CMS Andean subjects. The progression of disease severity may induce further physiological alterations in the hemodynamic and autonomic responses to submaximal cycling exercise; which theoretically may result in the aforementioned exercise intolerance, and therefore further investigations are warranted. Finally, caution is warranted when extrapolating these findings to exercise intolerance during longer duration exercise since the current study only utilized 5-min bouts of cycling. Longer durations of exercise would elicit further alterations in thermoregulatory, hemodynamic, and autonomic responses (41), which may encroach on an already over taxed hemodynamic reserve and explain the previously subjective reports of exercise intolerance.

Conclusion

Our data demonstrated that, despite a lower SV response, Andeans with mild to moderate CMS have a comparable convective oxygen delivery during light and moderate-intensity submaximal exercise to those without CMS. Furthermore, we identified that the exaggerated TPR and blood pressure response to moderate-intensity exercise are not attributed to greater sympathetic outflow and therefore suggesting that other neural and/or nonneural factors are involved.

GRANTS

The work contained in this study was supported by the Department of Sport Science, University of Innsbruck (to A. B. Hansen, S. B. Amin, F. Hofstätter, L. L. Simpson, H. Mugele, and J. S. Lawley), the Wilderness Medical Society research intraining grant (to G. Moralez), and National Heart, Lung, and Blood Institute Grant 1F32HL137285-0 (to C. M. Hearon, Jr.). Some of the work contained in this study was supported by a Canada Research Chair (to P. N. Ainslie) and the Natural Sciences and Engineering Research Council of Canada (to M. M. Tymko and P. N. Ainslie).

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

C.M.H.Jr., J.S.L., and G.M. conceived and designed research; A.B.H., S.B.A., F.H., L.L.S., C.G., T.G.D., M.M.T., C.M.H.Jr., J.S.L., and G.M. performed experiments; A.B.H., H.M., L.L.S., J.S.L., and G.M. analyzed data; A.B.H., C.M.H.Jr., J.S.L., and G.M. interpreted results of experiments; A.B.H. prepared figures; A.B.H. drafted manuscript; A.B.H., S.B.A., F.H., H.M., L.L.S., C.G., T.G.D., M.M.T., P.N.A., F.C.V., C.M.H.Jr., J.S.L., and G.M. edited and revised manuscript; A.B.H., S.B.A., F.H., H.M., L.L.S., C.G., T.G.D., M.M.T., P.N.A., F.C.V., C.M.H.Jr., J.S.L., and G.M. approved final version of manuscript.