Sublingual microcirculatory blood flow and vessel density in Sherpas at high altitude

Sherpa highlanders demonstrate extraordinary tolerance to hypoxia at high altitude, yet the physiological mechanisms underlying this tolerance remain unknown. In our prospective study, conducted on healthy volunteers ascending to Everest base camp (5,300 m), we demonstrated that Sherpas have a higher sublingual microcirculatory blood flow and greater capillary density at high altitude than lowlanders. These findings support the notion that the peripheral microcirculation plays a key role in the process of long-term adaptation to hypoxia.

Abstract

Anecdotal reports suggest that Sherpa highlanders demonstrate extraordinary tolerance to hypoxia at high altitude, despite exhibiting lower arterial oxygen content than acclimatized lowlanders. This study tested the hypothesis that Sherpas exposed to hypobaric hypoxia on ascent to 5,300 m develop increased microcirculatory blood flow as a means of maintaining tissue oxygen delivery. Incident dark-field imaging was used to obtain images of the sublingual microcirculation from 64 Sherpas and 69 lowlanders. Serial measurements were obtained from participants undertaking an ascent from baseline testing (35 m or 1,300 m) to Everest base camp (5,300 m) and following subsequent descent in Kathmandu (1,300 m). Microcirculatory flow index and heterogeneity index were used to provide indexes of microcirculatory flow, while capillary density was assessed using small vessel density. Sherpas demonstrated significantly greater microcirculatory blood flow at Everest base camp, but not at baseline testing or on return in Kathmandu, than lowlanders. Additionally, blood flow exhibited greater homogeneity at 5,300 and 1,300 m (descent) in Sherpas than lowlanders. Sublingual small vessel density was not different between the two cohorts at baseline testing or at 1,300 m; however, at 5,300 m, capillary density was up to 30% greater in Sherpas. These data suggest that Sherpas can maintain a significantly greater microcirculatory flow per unit time and flow per unit volume of tissue at high altitude than lowlanders. These findings support the notion that peripheral vascular factors at the microcirculatory level may be important in the process of adaptation to hypoxia.

NEW & NOTEWORTHY Sherpa highlanders demonstrate extraordinary tolerance to hypoxia at high altitude, yet the physiological mechanisms underlying this tolerance remain unknown. In our prospective study, conducted on healthy volunteers ascending to Everest base camp (5,300 m), we demonstrated that Sherpas have a higher sublingual microcirculatory blood flow and greater capillary density at high altitude than lowlanders. These findings support the notion that the peripheral microcirculation plays a key role in the process of long-term adaptation to hypoxia.

anecdotal reports suggest that Sherpa highlanders exhibit extraordinary tolerance to hypoxia at high altitude. Subjective demonstration of their remarkable exercise and endurance abilities may be readily observed by persons trekking and climbing in the Himalayan mountain regions. Because Sherpas have resided at high altitude for 500 generations (2), it is likely that these observations are underpinned by alterations in their genome, adapted through the process of natural selection driven by lifelong environmental exposure to hypobaric hypoxia. While evidence of consequent downstream phenotypic alterations remains limited, intriguingly it has been demonstrated that Sherpas exhibit a lower arterial oxygen content than lowlanders who ascend to comparable altitudes (1, 4, 38, 44). It is thus conceivable that through the comparison of lowlander and highlander genotype-phenotype, the adaptive mechanisms that facilitate their apparent hypoxia tolerance might be uncovered.

The delivery of oxygen to metabolizing tissues is a process of both convective flow within the systemic circulation and diffusion along oxygen partial pressure gradients within the tissues. To date, studies have predominantly focused on the traditionally described aspects of acclimatization involving the restoration of arterial oxygen content and systemic oxygen delivery (4, 6, 15, 38). Although such studies have failed to provide a universally accepted explanation for hypoxia tolerance, little attention has been paid to the tissue components of the oxygen cascade. Within every tissue of the body, the microcirculation [anatomically described as <100-μm blood vessels (26)] regulates localized blood flow to match microregional oxygen demand (9). As the final step in the convective portion of the oxygen cascade, the point at which oxygen diffuses into the surrounding tissues, alterations in the microvasculature may disrupt the balance between oxygen supply and demand at a cellular level, thereby acting as a “bottleneck” in the oxygen cascade. Accordingly, this potential limiting factor may be reduced or obviated by maintaining adequate microcirculatory flow per unit time and/or per unit volume of tissue (functional capillary density); thus the microcirculation should be considered important in the development of hypoxia tolerance (25). This study tested the hypothesis that Sherpas exposed to hypobaric hypoxia on ascent to 5,300 m demonstrate increased microcirculatory blood flow and vessel density as a means to maintain oxygen delivery to the peripheral tissues.

MATERIALS AND METHODS

Participant Selection

Approval for this study was obtained from both the University College London Research Ethics Committee and the Nepal Health Research Council as part of the Xtreme Everest 2 research expedition (18). Healthy Sherpa and lowlander volunteers were recruited, and written consent was obtained from all participants. Sherpas were defined as direct descendants (for ≥2 generations) of Nepali Sherpas, drawn from communities in the Solukhumbu and Rolwaling valleys. Lowlanders were recruited in the United Kingdom; they were not descendants of a native high-altitude population (e.g., Tibetan, Andean, Ethiopian), and all were born and lived below 1,000 m.

Study Setting

The Xtreme Everest 2 research expedition (29) was conducted from December 2012 to May 2013 and was among the individual studies conducted on the research expedition. Sublingual microcirculatory data were collected at three locations: baseline (BL), Everest base camp (EBC, 5,300 m), and on descent in Kathmandu (KTM, 1,300 m). BL testing was conducted in London (LON, 35 m) for lowlanders and in KTM (1,300 m) for Sherpas. Having departed from KTM, all participants followed an identical ascent and descent profile: a flight from KTM to an altitude of 2,800 m, followed by an 11-day trek to EBC, 3 nights at 5,300 m, and, finally, descent to KTM in 5 days.

Observation of the Sublingual Microcirculation

The sublingual microcirculation was visualized using an incident dark-field (IDF) imaging video microscope (Cytocam, Braedius Medical, Huizen, The Netherlands) (3). Prior to its use in the study, thorough assessment of this new video microscope and its automated analysis software was undertaken. Results demonstrated that the IDF camera provided improved image acquisition of human sublingual microcirculation compared with the sidestream dark-field (SDF) video microscope (17). The camera uses polarized green light (548-nm wavelength) to illuminate the observed tissue. This light corresponds to one of the isobestic points of oxy- and deoxyhemoglobin and, thus, ensures optimal absorption by red blood cells within the microvasculature, regardless of oxygenation status (20). Absorption of light by hemoglobin (Hb), but not by surrounding tissue, creates a distinct contrast of dark and light color, respectively; thus, red blood cells moving through the mucosal microcirculation appear as dark globules moving along the axis of flow.

At each measurement point, participants were required to rest for 10 min in the supine position before images were obtained following the standard operating guidelines of Trzeciak et al. (41), whereby the investigator positioned and focused the IDF camera under the participant’s tongue. Ten seconds of video footage were digitally recorded onto the computer, where images were stored for later analysis. This process was repeated on each participant until five good-quality recordings had been acquired from separate areas of the sublingual region. Studies were conducted during the day, and subjects were sheltered from extremes of temperature. All images were obtained by one of three researchers, all of whom were experienced in using the IDF video microscope.

Analysis and Scoring of Microcirculatory Video Images

IDF data analysis was conducted by two researchers using the AVA 3.0 microcirculatory analysis software (MicroVision Medical, Amsterdam, The Netherlands) (10). To avoid observer bias during analysis of microcirculatory films, investigators were blinded to both the study location and cohort identity by assignment of random codes to identify films. To assess for interobserver variability, the two observers evaluated a selection of IDF videos (30 films). Each video was deemed appropriate for analysis only if it adhered to the “microcirculation image quality scoring system” (31), whereby stability, illumination, duration, focus, content, and pressure artifact are assessed. Videos were subsequently corrected for background variation; image contrast was optimized, and, to compensate for movement artifact, all video images underwent image stabilization by the analysis software. After initial automated vessel detection, every film was checked visually, whereupon incorrectly identified blood vessels were deleted and undetected vessels were drawn manually. Additionally, incorrectly disconnected segments of vessels were “chained,” and erroneously connected segments were “unchained.”

In keeping with the consensus statement published in 2007 (8), the mean score from each of the five measurements recorded at each altitude was used. IDF variables included the microvascular flow index (MFI), heterogeneity index (HI), and vessel density.

Microvascular flow index.

The magnitude of microvascular perfusion is commonly evaluated by a semiquantitative scoring system referred to as the MFI (5, 12). The MFI is based on determination of the average or predominant flow type in the field of view at a given time point. It is quantified using an ordinal scale as follows: 0 = no flow, 1 = intermittent flow, 2 = sluggish flow, 3 = continuous flow. The “vessel-by-vessel” approach to MFI calculation (11, 12) was utilized in this study, in which the mean value of the MFIs in each individual vessel is calculated. This approach has been shown to provide the best correlation with both the erythrocyte velocity and the proportion of perfused small vessels (36) and, furthermore, demonstrates the closest intraobserver reliability for vessel detection and flow classification (35).

Heterogeneity index.

The flow HI provides information relating to the presence of microcirculatory distributive alterations and shunting (13). It is calculated as the highest site flow velocity (i.e., the MFI) minus the lowest site flow velocity divided by the mean flow velocity of all sublingual sites at that time point (42).

Vessel density.

Microcirculatory density is assessed as the vessel density. The total length of the vessel is divided by the total surface of the analyzed vessel (mm/mm²).

In each instance, both small (<25-μm-diameter) and large (>25-μm-diameter) vessel density values are reported. Although small vessels relate to capillaries (and, thus, contribute principally to organ perfusion and are, arguably, the vessels of greatest significance), large vessel density values are reported, as they are used as a quality-control measure to ensure that excessive pressure was not used in obtaining the videos.

Physiological Measurements

Hb concentration (HemoCue, Ängelholm, Sweden) and hematocrit (Hct) values (Sigma 1-14 microcentrifuge, Sigma, Osterode, Germany) were obtained from whole blood samples. Peripheral oxygen saturation (Onyx 9500, Nonin Medical, Minneapolis, MN), heart rate (HR), and blood pressure (model M3H, Omron Healthcare, Kyoto, Japan) were recorded after 10 min in the seated position at rest. Mean arterial pressure (MAP) was calculated from systolic and diastolic blood pressure (SBP and DBP) values. Tympanic temperature was measured from the ear canal (model 4020, Braun, Kronberg, Germany).

Statistical Analysis

All data were assessed for normality. The Shapiro-Wilk test (P > 0.05) and visual inspection of their histograms, normal Q-Q plots, and box plots showed that the data were not normally distributed. Nonparametric tests were therefore used for statistical analysis, with values summarized as median and interquartile ranges. Related-samples Friedman’s two-way analysis of variance by ranks tests (>2 sites) and related-samples Wilcoxon’s signed rank test (between 2 sites) with Bonferroni’s correction were used to assess the effect of hypoxia on the peripheral microcirculation. Sherpa and lowlander cohorts were compared using the unpaired Mann-Whitney U-test. Data are presented as box-whisker plots. The relationship between microcirculatory flow and other physiological variables was assessed individually using Spearman’s rank correlation coefficient (r). Interobserver variability for analysis of the IDF images was assessed by calculation of the intraclass correlation coefficient. All statistical calculations were performed on SPSS version 21 (IBM, New York, NY), and P < 0.05 was taken to indicate statistical significance.

RESULTS

Of the 133 participants (64 Sherpas and 69 lowlanders) who underwent BL testing, 131 (63 Sherpas and 68 lowlanders) completed testing at EBC and 83 (17 Sherpas and 66 lowlanders) in KTM. The demographics of the participants are shown in Table 1, and information relating to the laboratory environments is shown in Tables 2 and 3. At each altitude, HR, SBP, DBP, MAP, Hb concentration, Hct, peripheral oxygen saturation, and core temperature were similar between the two cohorts (Table 4).

Table 1.

Demographic summary of participants

	Sherpas (n = 64)	Lowlanders (n = 69)
Sex, %male	47	39
Age, yr	27.9 ± 6.9	41.3 ± 13.9
Height, cm	160 ± 6	171 ± 10
Weight, kg	71.1 ± 13.5	61.3 ± 8.9
Smokers, %	14 ± 21	6 ± 8.6

Values are means ± SD.

Table 2.

Laboratory environmental conditions

Laboratory	Altitude, m	Barometric Pressure, kPa	Temperature, °C	Humidity, %	Po2, kPa
LON	35	100.6 ± 0.2	16.9 ± 1.8	35.4 ± 6.5	21.0
EBC	5,300	53.0 ± 0.2	12.9 ± 8.2	37.8 ± 17.5	11.0
KTM	1,300	86.8 ± 0.4	23.8 ± 3.4	47.4 ± 15.7	18.1

Values are means ± SD. Data were recorded during laboratory testing in the field. Po₂ was calculated from barometric pressure, with the assumption that inspired oxygen fraction = 0.209. LON, London; EBC, Everest base camp; KTM, Kathmandu.

Table 3.

Laboratory temperature and Po₂ according to study cohort

	BL		EBC		KTM (descent)
	Sh	LL	Sh	LL	Sh	LL
Laboratory temperature, °C	16.9 ± 1.8	22.6 ± 3.2	12.6 ± 8.4	12.9 ± 7.9	23.8 ± 3.3	24.1 ± 3.1
$\mathrm{P}{\mathrm{I}}_{{\mathrm{O}}_{\mathrm{2}}}$, kPa	16.8	19.8	9.8	9.8	16.8	16.8

Values are means ± SD. Baseline (BL) testing was carried out in LON for lowlanders (LL) and in KTM for Sherpas (Sh). Both cohorts were tested in KTM on descent. $\mathrm{P}{\mathrm{I}}_{{\mathrm{O}}_{\mathrm{2}}}$, inspired Po₂.

Table 4.

Physiological variables for participants during the study

	HR, Beats/min		SBP, mmHg		DBP, mmHg		MAP, mmHg		SpO2, %		Hb, g/l		Hct, %		Core Temperature, °C
Laboratory	Sh	LL	Sh	LL	Sh		Sh	LL	Sh	LL	LL	LL	Sh	LL	Sh	LL
BL	69 ± 10	64 ± 9	121 ± 10	127 ± 19	81 ± 10	79 ± 10	94 ± 9	95 ± 13	97 ± 1	98 ± 1.2	137 ± 16	141 ± 14	43 ± 4.5	43 ± 3.9	36.3 ± 0.5	36.3 ± 0.4
EBC	87 ± 10	77 ± 14	125 ± 13	132 ± 16	89 ± 11	86 ± 8	101 ± 11	101 ± 10	78 ± 5	79 ± 5.3	151 ± 17	153 ± 20	49 ± 4.5	49 ± 5.7	36.2 ± 0.6	36.1 ± 0.7
KTM	75 ± 13	68 ± 12	112 ± 8	122 ± 15	75 ± 9	80 ± 8	87 ± 8	95 ± 10	95 ± 7	97 ± 1.4	140 ± 13	145 ± 20	43 ± 3.4	46 ± 5.8	36.1 ± 0.4	36.2 ± 0.5

Values are means ± SD. HR, heart rate; SBP, systolic blood pressure, DBP, diastolic blood pressure; MAP, mean arterial pressure; SpO₂, peripheral oxygen saturation; Hb, hemoglobin concentration; Hct, hematocrit.

MFI and HI were used to provide indexes of microcirculatory flow. The MFI for small (<25-μm-diameter) vessels did not differ between Sherpas and lowlanders at BL [2.81 (2.60–2.98) and LL 2.96 (2.62–3.00), respectively] or in KTM [2.97 (2.75–3.00) and 2.84 (2.52–3.00), respectively]; however, at EBC, Sherpas had a significantly higher MFI [3.00 (2.88−3.00) vs. 2.66 (2.45–2.97), P < 0.001; Fig. 1]. The MFI for large (>25-μm-diameter) vessels did not differ between Sherpas and lowlanders at any of the three measurement points.

Fig. 1.

Ascent to high altitude caused microvascular flow index (MFI) to increase from baseline (BL) in Sherpas, while MFI decreased in lowlanders. *Significant difference between cohorts. ^Significantly different from BL.

There was no difference in the small vessel HI between Sherpas and lowlanders at BL [0.386 (0.336–0.402) and 0.359 (0.336–0.667), respectively]; however, lowlander values were significantly greater than Sherpa values at EBC [0.408 (0.374–0.724) vs. 0.341 (0.333–0.390), P < 0.001] and on descent to KTM [0.392 (0.352–0.667) vs. 0.333 (0.333–0.470), P = 0.010; Fig. 2].

Fig. 2.

Heterogeneity index (HI) decreased in Sherpas at Everest base camp (EBC) and in Kathmandu (KTM) compared with BL, while HI increased in lowlanders at EBC. *Significant difference between cohorts. ^Significantly different from BL.

Density of small (<25-μm-diameter) vessels was not different between the two cohorts at BL or in KTM, but Sherpas had a significantly greater small vessel density at EBC [13.83 (11.41–14.52) vs. 10.52 (8.90–11.34) mm/m², P = 0.047; Fig. 3]. There was no difference in large (>25-μm-diameter) vessel density between Sherpas and lowlanders at any site.

Fig. 3.

Small vessel density increased in Sherpas at EBC and remained higher than BL values on return to KTM. Vessel density in lowlanders increased at EBC and returned to BL values on return to KTM. *Significant difference between cohorts. ^Significantly different from BL.

There was no correlation between small vessel MFI, HI, or vessel density and any of the measured physiological variables (Hb, Hct, HR, SBP, DBP, MAP, and peripheral oxygen saturation).

The intraclass correlation coefficient was used to assess interobserver variability in IDF image analysis between two investigators. A strong correlation [0.89 (95% confidence interval 0.83–0.96)] was demonstrated.

DISCUSSION

This study demonstrated differences between Sherpa and lowlander microcirculatory responses to sustained hypobaric hypoxia at high altitude. Although microcirculatory blood flow and capillary density did not differ between cohorts in normoxia (BL), upon exposure to hypoxia, Sherpas demonstrated significantly greater values for both indexes. Hypoxia caused Sherpas to increase both microcirculatory blood flow and capillary density, whereas lowlanders decreased flow, but increased density, however, to a lesser extent than the Sherpas (Fig. 4). On descent to KTM, the relative increase in vessel density for both cohorts persisted; however, blood flow returned to previous BL values.

Fig. 4.

On ascent to high altitude, small vessel density and microvascular flow in Sherpas increased dramatically in a uniform and homogenous manner. On reexposure to normoxia, flow returned to previous BL values; however, while vessel density decreased, it remained greater than initial BL values. On ascent to high altitude, small vessel density in lowlanders increased, but to a far lesser extent than in Sherpas. In lowlanders, microvascular flow decreased, but not in a uniform manner, such that it became heterogeneous in nature. On reexposure to normoxia, vessel density and flow returned to BL values.

Numerous studies have attempted to determine the genetic and physiological differences between the indigenous high-altitude Sherpa (and Tibetan) people and those who live at low altitude (19). Few of these studies, however, have revealed marked differences that might explain how this high-altitude population not only lives, but seemingly thrives, so effectively under conditions of chronic environmental hypoxia. In 2007, Erzurum et al. (14) explored the possibility that peripheral blood flow was an important determinant in long-term adaptation to hypoxia. Venous occlusion plethysmography was utilized to measure blood flow in the forearm of 88 Tibetans at 4,200 m and 50 sea-level residents at 206 m. Their results demonstrated that Tibetans have more than double the forearm blood flow of American controls. Although these results support earlier studies relating to blood flow (39) and skeletal muscle capillary density (22), the venous occlusion plethysmography data in the study of Erzurum et al. relate to total blood flow in the forearm, rather than in the microcirculation. The first description of in vivo microcirculatory changes on ascent to high altitude coincided with the introduction of SDF imaging (21). On ascent to 4,900 m, blood flow in the sublingual vessel decreased significantly in 12 lowland subjects (30), and similar data were recorded in another group of 24 lowland subjects on ascent to 5,300 m (28).

The lowlander MFI data support the findings of Martin et al. (28, 30), who reported that flow decreased on ascent to high altitude. They theorized that the slowing microcirculatory blood flow could, in fact, be an adaptive response, applied to increase the erythrocyte tissue transit time and improve oxygen diffusion. This is conceivable, since a prolonged course through the capillary network may enhance offloading of oxygen in the presence of a reduced partial pressure gradient between the capillary and mitochondria, particularly when cardiac output is high, as is the case during exercise. Sherpas, by contrast, seem to utilize brisk flow to maintain localized oxygen delivery. This, in turn, may explain the lower Hb concentration in this population after prolonged exposure to hypobaric hypoxia (4). Undoubtedly, increased Hb concentration augments arterial oxygen content; however, elevated Hct increases blood viscosity, alters its rheology, and, at levels >50%, may decrease cardiac output and oxygen delivery (43). Furthermore, elevated Hct levels in South American resident populations are associated with an increased prevalence of chronic mountain sickness and related embolic or thrombotic events (33). In contrast, it seems that Sherpas favor a blunted erythropoietic response, thereby allowing for brisk microvascular blood flow.

The speed of microcirculatory blood flow per se may also be less important than its nature. Maintaining a homogenous microcirculatory blood flow, irrespective of the speed at which the contained blood may flow, could be crucial to tissue perfusion. In this study, ascent to EBC was associated with a fall in the HI in Sherpas and an increase in lowlanders, such that a significant difference is evident between cohorts at high altitude. A lower HI equates to more homogenous flow, and the importance of this may be highlighted in the clinical setting, where dysregulated, heterogeneous microvascular flow is a fundamental mechanism through which tissue hypoxia occurs in sepsis (7). In either case, whether the important determinant of tissue oxygenation relates to the speed and/or the homogenous nature of blood flow, Sherpas demonstrate superiority in both.

The descent data observed in this study are also novel. The similarity in MFI values in the two cohorts at BL and in KTM suggests that the physiological basis underpinning the ability of the Sherpas to maximize microcirculatory blood flow at altitude is transient and hypoxia-dependent. We do, however, appreciate that only a small number of Sherpas were studied on their return to KTM; thus we are cautious in our interpretation of these data.

The data illustrating the effects of hypoxic exposure on capillary density in Sherpas are the first of their kind. Although no difference in small vessel density was evident between cohorts at BL, these data demonstrate that Sherpas have a substantial capacity to increase their capillary numbers. An increase in sublingual vessel density on ascent to altitude has been reported by Martin et al. (28) in their study of 21 lowlanders ascending to 5,300 m: it was not the density of small (<25-μm-diameter) but, rather, the density of larger (>25-μm-diameter) vessels that changed at high altitude. Although the actual values reported by Martin et al. for small vessel density were similar to those reported above, our data contrast with our previous finding of an increase in small vessel density in lowlanders on ascent to 5,300 m (Fig. 3; P = 0.020) but no change in their large vessel density. Although both studies used a very similar ascent profile, the discrepancy between our findings may be due to the increased statistical power of this study and/or our use of the IDF, rather than the SDF, video microscope (17).

While both cohorts demonstrated increased capillary density on ascent to high altitude, Sherpas did so to a much greater degree. At EBC, the capillary network was ~30% denser in Sherpas than lowlanders. Vessel recruitment due to elevated Hb and Hct might have accounted for the rise in capillary density in both groups (16, 34, 37, 45). These values, however, were similar between the two cohorts upon arrival at EBC, so it seems unlikely that this explains the difference between them, unless Sherpas have a much larger unrecruited (and, thus, unseen) reservoir in normoxia. This is certainly plausible, and, as with flow, it is likely that the difference is ultimately underpinned by genetic differences.

This is the first study of Sherpa microcirculation on ascent to, and descent from, high altitude. A large number of participants were studied, and >98% successfully ascended to EBC following an identical ascent profile. This matched ascent profile, along with serial measurements, controls for variability of exposure to hypoxia and, thereby, enables valid interindividual comparison of hypoxia responses while amplifying the signal-to-noise ratio (24). The newly released Cytocam video microscope was used to obtain images of the sublingual microcirculation, and our assessment of this IDF video microscope before the expedition demonstrated its superior capabilities regarding image acquisition compared with its predecessor, the SDF video microscope (17). Unfortunately, no validation of the camera in a hypobaric hypoxic environment was performed before the expedition, and this is a limitation of the study. Further limitations include potential recruitment bias, confounding factors within laboratories, the different altitude for BL testing in Sherpas (1,300 m) and lowlanders (35 m), and the small number of Sherpas tested on descent. Although they were recruited through open advertisement and word of mouth, the participants were self-selecting, in that this research expedition involved opportunistic observation for individuals with a desire to visit the study environment and, thus, may not be truly representative of a “normal” Sherpa or lowlander population. The demographic data (Table 1) demonstrate that approximately equal numbers of participants were compared, with a similar male-to-female ratio in each group; however, the age of participants was markedly younger in the Sherpa cohort, while the percentage of smokers was higher. Smoking is known to affect the vasculature and, thus, could be a confounding factor in the results (32). Despite our best efforts to minimize temperature differences between laboratories, disparities were seen. This could affect microvascular flow due to cold-induced vasoconstriction (23, 40). There were, however, no significant differences between the environmental temperatures to which both cohorts were exposed within each individual laboratory (Table 3). Additionally, as the sublingual circulation is within the oral cavity and, as such, is regarded as being at a temperature similar to core temperature, the data demonstrated no differences in tympanic temperatures between the two cohorts (Table 4). Other potential confounding factors specific to the high-altitude environment include hydration status, which in turn may affect Hct values and, thus, alter blood rheology. We minimized the effect of this potential confounding factor by conducting studies at all altitudes after a period of overnight rest and by ensuring that the subjects had free access to oral fluids and were actively encouraged to drink enough fluid to produce normal volumes of clear urine. Finally, ascent to altitude may cause tissue edema (27), which, in the sublingual mucosa, could theoretically affect image quality and lead to false measurements of flow and density. BL testing was conducted in LON for lowlanders (35 m) and in KTM (1,300 m) for Sherpas due to logistical reasons. First, it would not have been pragmatic or financially viable to fly all Sherpas to London for their BL testing. Second, data from Caudwell Xtreme Everest 2007 (24) had failed to identify significant differences in participants’ physiology between the laboratory at sea level and the laboratory in KTM (1,300 m); thus we believed it to be scientifically appropriate to use these two distinct locations for BL testing. Lastly, the notable deficit in Sherpas tested on descent in KTM should be highlighted: 46 Sherpas, having previously been tested at EBC, were not tested in KTM due to logistical constraints.

In conclusion, this study suggests that adaptation to hypoxia in the Sherpa sublingual microcirculation involves increasing both microcirculatory blood flow and capillary density. In turn, teleological reasoning would suggest that these alterations result in a greater oxygen delivery both per unit time and per unit volume of tissue. It remains unclear whether these microvascular alterations are restricted to the sublingual microcirculation or what underlying biochemical and physiological factors facilitate the changes in blood flow and vessel density. Further work is required to explore these questions.

GRANTS

Xtreme Everest 2 is a research project coordinated by the Xtreme Everest Hypoxia Research Consortium, a collaboration between the University College London Centre for Altitude, Space, and Extreme Environment Medicine, the Centre for Human Integrative Physiology at the University of Southampton and Duke University Medical Center. Xtreme Everest 2 was supported by the Royal Free Hospital NHS Trust Charity, the Special Trustees of University College London Hospital NHS Foundation Trust, the Southampton University Hospital Charity, the University College London Institute of Sports Exercise and Health, The London Clinic, University College London, University of Southampton, Duke University Medical School, the United Kingdom Intensive Care Society, the National Institute of Academic Anaesthesia, the Rhinology and Laryngology Research Fund, The Physiological Society, Smiths Medical, Deltex Medical, Atlantic Customer Solutions, and the Xtreme Everest 2 volunteer participants who trekked to Everest base camp. Some of this work was undertaken at University College London Hospital-University College London National Institute for Health Research (NIHR) Biomedical Research Centre, which received a portion of funding from the United Kingdom Department of Health’s NIHR Biomedical Research Centres funding scheme. Some of this work was undertaken at University Hospital Southampton-University of Southampton NIHR Respiratory Biomedical Research Unit, which received a portion of funding from the United Kingdom Department of Health’s NIHR Biomedical Research Units funding scheme.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

E.T.G.-K., M.F., D.L., M.M., M.P.G., and D.S.M. conceived and designed the research; E.T.G.-K., J. Coppel, J. Court, J.V.D.K., and A.V. performed the experiments; E.T.G.-K. and J. Coppel analyzed the data; E.T.G.-K. and D.S.M. interpreted the results of the experiments; E.T.G.-K. and D.S.M. prepared the figures; E.T.G.-K., M.P.G., and D.S.M. drafted the manuscript; E.T.G.-K., J. Coppel, J. Court, J.V.D.K., A.V., M.F., D.L., M.M., M.P.G., and D.S.M. edited and revised the manuscript; E.T.G.-K., J. Coppel, J. Court, J.V.D.K., A.V., M.F., D.L., M.M., M.P.G., and D.S.M. approved the final version of the manuscript.