No Correlation between Plasma Levels of Vascular Endothelial Growth Factor or its Soluble Receptor and Acute Mountain Sickness

Abstract

Schommer, Kai, Neele Wiesegart, Christoph Dehnert, Heimo Mairbäurl, and Peter Bärtsch. No correlation between plasma levels of vascular endothelial growth factor or its soluble receptor and acute mountain sickness. High Alt. Med. Biol. 12:323–327.—Increased plasma levels of vascular endothelial growth factor (VEGF) due to lower levels of its soluble receptor (sFlt-1) had been suggested to cause vasogenic brain edema and thereby to cause the symptoms of acute mountain sickness (AMS). We tested this hypothesis after active ascent to high altitude. Plasma was collected from 31 subjects at low altitude (100 m) before (LA1) and after (LA2) 4 weeks of aerobic exercise training in normobaric hypoxia or normoxia, and one night after ascent to high altitude (4559 m). Training modalities (hypoxia or normoxia) did not influence VEGF- and sFlt-1-levels. Therefore, data of both training groups were analyzed together. After one night at 4559 m, 18 subjects had AMS (AMS+), 13 had no AMS (AMS−). In AMS+ and AMS−, VEGF was 110±75 (SD) pg/ml vs. 104±82 (p=0.74) at LA1, 63±40 vs. 73±50 (p=0.54) at LA2, and 88±62 vs. 104±81 (p=0.54) at 4559 m, respectively. Corresponding values for sFlt-1 in AMS+ and AMS− were 81pg/ml±13.1 vs. 82±17 (p=0.97), 79±11 vs. 80±16 (p=0.92) and 139±28 vs. 135±31 (p=0.70), respectively. Absolute values or changes of VEGF were not correlated and those of sFlt-1 slightly correlated with AMS scores. These data provide no evidence for a role of plasma VEGF and sFlt-1 in the pathophysiology of AMS. They do, however, not exclude paracrine effects of VEGF in the brain.

Introduction

Acute mountain sickness (AMS) is characterized by symptoms of the central nervous system such as headache, nausea, and dizziness, and occurs at altitudes above 2500 m (Hackett and Roach, 2001). Absolute altitude, the rate of ascent, individual susceptibility, and degree of preacclimatization are its major determinants (Schneider et al., 2002). There is also evidence that exercise increases the risk for development of AMS (Roach et al., 2000). Despite extensive research in this field, the pathophysiology of AMS is still not completely understood (Bartsch et al., 2004). Hypoxia stabilizes hypoxia-inducible factors that stimulate the transcription of various genes involved in vascular reactivity, angiogenesis, and arteriogenesis (Rey and Semenza, 2010). One of the upregulated target hormones of this activation is the vascular endothelial growth factor (VEGF), which is one of the most potent hormones in increasing vascular permeability and in neoangiogenesis (Carmeliet and Collen, 2000). Its soluble receptor (sFlt-1) binds circulating VEGF and thereby reduces available VEGF. It is conceivable that the VEGF-mediated increase in vascular permeability contributes to the pathophysiology of AMS either because of greater upregulation of VEGF due to more severe hypoxemia in AMS (Bartsch et al., 2002) or because of lower levels of sFlt-1 resulting in an increased bioavailability of VEGF (Tissot van Patot, et al., 2005).

Several investigations reported an increase in plasma levels of VEGF at high altitude but found no correlation between plasma levels and AMS (Dorward et al., 2007; Maloney et al., 2000; Nilles et al., 2009; Palma et al., 2006; Walter et al., 2001). Tissot van Patot et al. (2005) reported that low plasma levels of sFlt-1 were associated with higher AMS scores in 19 subjects after passive ascent to 4300 m, suggesting that the ratio of VEGF vs. sFlt-1 was essential since the ratio of VEGF and sFlt-1 determines the bioavailabilty of VEGF. They concluded that this supports the hypothesis of a VEGF-mediated vasogenic brain edema that causes the symptoms of AMS.

We therefore examined the association between plasma VEGF- and sFlt-1- levels and AMS in a larger group in a setting which involved an active ascent. A study investigating the effects of training in normobaric hypoxia for the prevention of AMS after rapid ascent to 4559 m (Schommer et al., 2010) provided the opportunity to test this hypothesis.

Material and Methods

Plasma VEGF- and sFlt-1- levels were measured in a double-blind placebo-controlled study on 42 subjects which primarily aimed at investigating the effects of training in normobaric hypoxia for prevention of AMS. For details of the study design we refer to the original publication (Schommer et al., 2010). Written informed consent was obtained from the subjects and the study was approved by the Ethics Committee of the Medical Faculty of the University of Heidelberg.

Blood samples were withdrawn from resting subjects at low altitude (100 m) before (LA1) and after (LA2) a 4-week period of aerobic bicycle training (three times per week over 60 min in hypoxia or normoxia, intensity corresponded approximately to 60% of VO₂max) and in the morning after the night spent at 4559 m (HA). Subjects ascended by cable car from low altitude to 3000 m and then further by foot to the overnight stay at 3611 m. On the next morning, the final ascent to 4559 m was performed within 4–5 hours. The entire ascent from low altitude to 4559 m was performed within 20 hours. The primary study endpoint was the occurrence of AMS defined as a Lake Louise (LL) score (Roach 1993) ≥5 and an AMS−C score of the Environmental Symptom Questionnaire (Sampson et al., 1983) ≥0.70 after one night spent at 4559 m. Eleven subjects were excluded from analysis due to violation of study protocol in two cases, high altitude pulmonary edema in one case, and early termination because of severe AMS in eight cases. All of the subjects were low-altitude residents and had no altitude exposure higher than 2000 m for at least 2 months prior to the study. Anthropometric data and AMS history are given in Table 1. Subjects were considered susceptible or not susceptible to AMS according to previously described criteria (Schneider et al., 2002). Oxygen saturation was measured by pulse oximetry (MasimoSet Radical, Masimo Corporation, Irvine, CA) and blood sampling from a radial artery for blood gas analysis on a Rapidlab 840 (Bayer Diagnostics, Sudbury, UK) was performed in resting subjects at 4559 m.

Table 1.

Anthropometric Data and AMS History

N	31
Male (n)	20
Age (years)	30.8 (20–55)
BMI (kg/m²)	22.7 (18.8–27.7)
No previous exposure >3000 m (n)	27
Not AMS susceptible (n)	3
AMS susceptible (n)	1

Values are given as means (range). AMS susceptibility was defined following the criteria of Schneider et al. (2002).

Whole blood was collected from a cubital vein into EDTA-containing tubes (Monovette, Sarstedt, Nümbrecht, Germany) and centrifuged at 4°C for 20 min at 400 g. Plasma was collected, flash frozen in liquid nitrogen, and stored at −80°C until analysis. Free plasma VEGF was measured using the Human VEGF Quantikine ELISA Kit (catalog no DVE00, R&D Systems, Minneapolis, MN) following manufacturers instructions. The Human sVEGF R1/Flt-1 Quantikine ELISA Kit (catalog No. DVR100B, R&D Systems) was used for measurement of sFlt-1. All measurements were performed in duplicate.

A two-way ANOVA for repeated measurements was performed using the SAS/STAT 8.0 (SAS Institute Inc., Cary, NC) software to detect interactions of VEGF- and sFlt-1-levels with the training modality (normoxia or hypoxia) or with the presence or absence of AMS and the time of sampling. Correlations between plasma levels of VEGF or sFlt-1 with AMS scores, Sao₂ and arterial blood gases were analyzed by Pearson correlation (SigmaStat 3.0, SPSS Inc., Chicago, Il). Oxygen saturation between subjects with and without AMS were compared by Student`s t test. Significance was set at p≤0.05 (two-sided) for all statistical analyses. Values are presented as mean±standard deviation.

Results

The training modality in normoxia or hypoxia had no influence on VEGF- and sFlt-1 levels. After the training period, plasma VEGF was decreased and sFlt-1 levels measured at low altitude were unchanged in both the normoxic and the hypoxic training groups (Table 2). Therefore, data from both groups were pooled and analyzed together. At high altitude, 18 subjects had AMS (AMS+), 13 did not (AMS−). AMS-C scores were 1.72±0.87 in AMS+ and 0.28±0.21 in AMS−. The corresponding values for the Lake Louise scores were and 8.1±2.9 vs. 3.4±2.9, respectively. AMS+ had significantly lower oxygen saturation compared to AMS−. Values obtained by pulse oximetry were 71.8±7.1% vs. 77.7±7.5% (p=0.035), and those obtained by arterial blood gas analysis were 71.3±6.9% vs. 77.5±7.7% (p=0.026), respectively.

Table 2.

VEGF and sFlt-1 Before (LA1) and After (LA2) the Training Period

	Normoxic training group	Hypoxic training group
	n=14	n=17
VEGF (pg/ml)
LA1	112.9±82.0	102.8±69.6
LA2	67.6±36.9	69.8±52.4
group p=0.51, time p=0.002^*, group×time p=0.998
sFlt-1 (pg/ml)
LA1	80.5±11.3	82.6±17.1
LA2	75.1±11.1	83.8±11.9
group, time, group x time p>0.90

Italic fonts show the p values of an analysis of variance for repeated measurements. LA1, low altitude 1 (before training); LA2, low altitude 2 (after training); VEGF, vascular endothelial growth factor; sFlt-1, soluble VEGF receptor.

Compared to the values obtained at low altitude prior to ascent (LA2), VEGF and sFlt-1 increased at high altitude, but were not different between AMS±and AMS−(Table 3). Neither absolute values for VEGF at high altitude nor their changes from baseline correlated significantly with the AMS scores (r values between 0.06 and 0.10). Plasma levels of sFlt-1 did not correlate with AMS-C scores (r values 0.20 for absolute values and for changes from base line). However, there was a significant correlation between the Lake Louise Score and the absolute values of sFlt-1 at high altitude (r=0.39, p=0.02), and with changes of sFlt-1 from baseline (r=0.45, p<0.05). Furthermore, there was an inverse correlation between sFlt-1 values and oxygen saturation at high altitude (for pulse oximetry r=−0.50, p<0.001, for arterial blood gas analysis r=−0.45, p=0.01). VEGF values did not correlate with oxygen saturation (p>0.10). Results of the statistical analysis did not change when the values were corrected for plasma volume (data not shown).

Table 3.

Changes of VEGF and sFlt-1 from Lowland to High Altitude in Subjects with (AMS+) and without AMS (AMS−)

	AMS+	AMS−
	n=18	n=13
VEGF (pg/ml)
LA2	63.4±40.4	73.3±49.7
HA	87.9±61.6	103.9±80.7
group p=0.67, time p=0.035^*, group x time p=0.435
sFlt-1 (pg/ml)
LA2	79.1±10.5	79.6±15.9
HA	138.9±28.4	134.7±30.7
group p=0.918, time p<0.001^*, group x time p=0.743

Italic fonts show the p values of an analysis of variance for repeated measurements. AMS, acute mountain sickness; HA, high altitude; LA2, lowland; sFlt-1, soluble VEGF receptor; VEGF; vascular endothelial growth factor.

Discussion

Our data show that plasma levels of VEGF and sFlt-1 increase after active ascent to 4559 m. Plasma concentrations of VEGF and sFlt-1 were not different between the AMS and non-AMS group. In contrast to what has been suggested (Tissot van Patot et al., 2005), we found that a more severe hypoxemia and higher Lake Louise scores at high altitude were associated with higher plasma levels of sFlt-1. Taken together, this study demonstrates that AMS was not associated with altered VEGF or sFlt-1 plasma concentrations. Therefore, it does not support the concept of increased plasma levels of VEGF causing vasogenic brain edema in AMS.

The lack of correlation between plasma levels of VEGF at high altitude with AMS scores is in line with most of the available studies on this subject (Dorward et al., 2007; Maloney et al., 2000; Nilles et al., 2009; Palma et al., 2006; Walter et al., 2001). One study on 19 subjects (Tissot van Patot et al., 2005) showed higher VEGF in subjects with AMS compared to those without AMS and showed lower sFlt-1 at low and high altitude in individuals who developed AMS. We did not find significant differences in plasma levels of sFlt-1 at baseline or at high altitude between the AMS+ and AMS− groups. On the contrary, those individuals with higher LL scores also had higher sFlt-1 in plasma as shown by a significant correlation with an r value of 0.5. The major differences between these two studies were the mode of ascent (active in our study vs. passive), the altitude of residency of the subjects (100 m vs. 1370 - 1645m, respectively), and the sample size (n=31 vs. n=19, respectively), while assessment of the LL score was comparable and the assay kits used for the measurement of VEGF and sFlt-1 were identical.

The lack of correlation between AMS scores and sFlt-1 despite both parameters being associated with more severe hypoxemia can be explained by the considerable overlap of individual values of Sao₂ in the groups with and without AMS as demonstrated by the large standard deviations. Furthermore, the correlation coefficient for the correlation between sFlt-1 and Sao₂ was 0.45 which gives an r² value of 0.20, indicating that merely 20% of the variation of sFlt-1 is explained by the changes in Sao₂. Both the large overlap of individual values of Sao₂ in AMS+ and AMS−, as well as the rather weak correlation between sFlt-1 and Sao₂, can explain the lack of a correlation between AMS scores and plasma levels of sFlt-1.

The mechanisms that account for changes of VEGF and sFlt-1 at high altitude are not clear. It is also not known whether changes depend on the duration of high altitude stay, on the exercise during ascent, or exercise in combination with hypoxia. Hypoxia is certainly a major stimulus for VEGF production (Rey and Semenza, 2010). There are no studies that adequately investigated the effect of altitude of residency on plasma levels of VEGF and sFlt-1 and their changes upon ascent to high altitude.

Little is known about the effects of exercise on plasma levels of these two parameters in otherwise healthy human. Plasma VEGF has been shown to decrease after a marathon race at high altitude, but it remained unchanged after survival training at low altitude (Gunga et al., 1999). Similarly, acute short-time hypoxia for 30 min decreased plasma VEGF but did not influence sFlt-1 (Oltmanns et al., 2006). Sixty minutes of aerobic exercise corresponding to 50% Vo₂max in normoxia increased plasma VEGF immediately and 2 h after exercise termination only in trained individuals, but not in sedentary people (Kraus et al., 2004). In our study, we demonstrate a reduction of free plasma VEGF and no changes in sFlt-1 at rest after a 4-week period of aerobic exercise, which was independent of the environmental conditions where the training had been performed (hypoxia or normoxia).

The difference in results between studies on VEGF and sFlt-1 might also be due to detection of the state of VEGF and its receptor in the assays (i.e., total vs. free VEGF or sFlt-1). The assay used for this study, which was also the one used by Tissot et al. (2005), detects the free VEGF, which is the form not bound to sFlt-1. The sFlt-1 assay kit measures the soluble receptor bound and unbound to VEGF. In this setting, the plasma concentration of VEGF reflects the biologically available VEGF in plasma, which is independent of the soluble receptor and its changes at high altitude. Taken together, it remains unclear which particular circumstances between the studies account for the discrepant results. The lack of difference in VEGF between AMS+ and AMS – seems to exclude a role of plasma VEGF in AMS in our study. Our study does, however, not exclude the possibility that VEGF produced in the brain and acting locally may play a role in the pathophysiology of AMS.

It was postulated that symptoms of AMS are caused by brain swelling due to vasogenic VEGF-induced brain edema, since an increased VEGF expression was shown to parallel brain edema in hypoxic mice (Xu and Severinghaus, 1998). In addition, pretreatment of mice with VEGF neutralizing antibodies completely blocked hypoxia-induced increase of vascular permeability (Schoch et al., 2002). Although the lack of a significant difference of the increase in VEGF at altitude between subjects with and without AMS does not support a role of VEGF-induced brain edema in the pathophysiology of AMS, measurements in plasma do, however, not necessarily exclude paracrine effects of VEGF in AMS, which we cannot detect in our setting. Indirect arguments against a role of VEGF in AMS can be taken also from recent MRI studies, which show that brain swelling is minimal and not different between subjects with and without AMS (Fischer et al., 2004; Kallenberg et al., 2007; Schoonman et al., 2008). These studies also show that AMS scores were correlated with indicators of cytotoxic but not vasogenic edema (Kallenberg et al., 2007; Schoonman et al., 2008), suggesting that vasogenic edema is not involved in the pathophysiology of AMS. In addition, it was shown that the concentration of VEGF in the cerebrospinal fluid is not different between subjects with and without AMS (Bailey et al., 2006). While all these studies do not support the concept of VEGF-induced brain edema causing AMS, they do not rule out paracrine intracerebral effects of VEGF in AMS, and an animal model (Schoch et al. 2002) strongly suggests that VEGF may play an important pathophysiological role in clinically manifest high altitude cerebral edema (HACE).

Conclusion

In summary, our study does not provide evidence for a role of plasma VEGF and its soluble receptor sFlt-1 in the pathophysiology of AMS, because their absolute values at high altitude and their changes from low to high altitude in plasma were not different between subjects with and without AMS. We can, however, not rule out paracrine effects of VEGF at the tissue level in the brain in AMS. The use of neutralizing monoclonal anti-VEGF antibodies could help answering this question. Such drugs are on the market, but considerable side effects prevent at present their use in studies on the pathophysiology of AMS.

Footnotes

Author Disclosure Statement

All authors declared no financial disclosure.

Acknowledgments

The authors thank the study participiants and mountain guides; the hut keepers and the Sezione Varallo of the Club Alpino Italiano for providing an excellent research facility at the Capanna Regina Margherita and Capanna Gnifetti; Hermann Buhl and Ute Haas for their assistance during training and providing blood samples before and after training at low altitude.

References

Bailey

, Roukens

, Knauth

, Kallenberg

, Christ

, Mohr

, Genius

, Storch-Hagenlocher

, Meisel

, McEneny

et al. 2006. Free radical-mediated damage to barrier function is not associated with altered brain morphology in high-altitude headache. J Cereb Blood Flow Metab, 26:99–111.

Bartsch

, Bailey

, Berger

, Knauth

, Baumgartner

. 2004. Acute mountain sickness: Controversies and advances. High Alt Med Biol, 5:110–124.

Bartsch

, Swenson

, Paul

, Julg

, Hohenhaus

. 2002. Hypoxic ventilatory response, ventilation, gas exchange, and fluid balance in acute mountain sickness. High Alt Med Biol, 3:361–376.

Carmeliet

, Collen

. 2000. Molecular basis of angiogenesis. Role of VEGF and VE-cadherin. Ann NY Acad Sci, 902:249–262discussion 262–264.

Dorward

, Thompson

, Baillie

, MacDougall

, Hirani

. 2007. Change in plasma vascular endothelial growth factor during onset and recovery from acute mountain sickness. Respir Med, 101:587–594.

Fischer

, Vollmar

, Thiere

, Born

, Leitl

, Pfluger

, Huber

. 2004. No evidence of cerebral oedema in severe acute mountain sickness. Cephalalgia, 24:66–71.

Gunga

, Kirsch

, Rocker

, Behn

, Koralewski

, Davila

, Estrada

, Johannes

, Wittels

, Jelkmann

. 1999. Vascular endothelial growth factor in exercising humans under different environmental conditions. Eur J Appl Physiol Occup Physiol, 79:484–490.

Hackett

, Roach

. 2001. High-altitude illness. N Engl J Med, 345:107–114.

Kallenberg

, Bailey

, Christ

, Mohr

, Roukens

, Menold

, Steiner

, Bartsch

, Knauth

. 2007. Magnetic resonance imaging evidence of cytotoxic cerebral edema in acute mountain sickness. J Cereb Blood Flow Metab, 27:1064–1071.

10.

Kraus

, Stallings

3rd , Yeager

, Gavin

. 2004. Circulating plasma VEGF response to exercise in sedentary and endurance-trained men. J Appl Physiol, 96:1445–1450.

11.

Maloney

, Wang

, Duncan

, Voelkel

, Ruoss

. 2000. Plasma vascular endothelial growth factor in acute mountain sickness. Chest, 118:47–52.

12.

Nilles

, Sayward

, D'Onofrio

. 2009. Vascular endothelial growth factor and acute mountain sickness. J Emerg Trauma Shock, 2:6–9.

13.

Oltmanns

, Gehring

, Rudolf

, Schultes

, Hackenberg

, Schweiger

, Born

, Fehm

, Peters

. 2006. Acute hypoxia decreases plasma VEGF concentration in healthy humans. Am J Physiol Endocrinol Metab, 290:e434–439.

14.

Palma

, Macedonia

, Deuster

, Olsen

, Mozayeni

, Crutchfield

. 2006. Cerebrovascular dynamics and vascular endothelial growth factor in acute mountain sickness. Wilderness Environ Med, 17:1–7.

15.

Rey

, Semenza

. 2010. Hypoxia-inducible factor-1-dependent mechanisms of vascularization and vascular remodelling. Cardiovasc Res, 86:236–242.

16.

Roach

, Maes

, Sandoval

, Robergs

, Icenogle

, Hinghofer-Szalkay

, Lium

, Loeppky

. 2000. Exercise exacerbates acute mountain sickness at simulated high altitude. J Appl Physiol, 88:581–585.

17.

Roach

RCB

, Hackett

, Oelz

. 1993. The Lake Louise acute mountain sickness scoring system. Hypoxia and molecular medicine. Sutton

JRH

CS , Coates

. 272–274.

18.

Sampson

, Cymerman

, Burse

, Maher

, Rock

. 1983. Procedures for the measurement of acute mountain sickness. Aviat Space Environ Med, 54:1063–1073.

19.

Schneider

, Bernasch

, Weymann

, Holle

, Bartsch

. 2002. Acute mountain sickness: Influence of susceptibility, preexposure, and ascent rate. Med Sci Sports Exerc, 34:1886–1891.

20.

Schoch

, Fischer

, Marti

. 2002. Hypoxia-induced vascular endothelial growth factor expression causes vascular leakage in the brain. Brain, 125:2549–2557.

21.

Schommer

, Wiesegart

, Menold

, Haas

, Lahr

, Buhl

, Bartsch

, Dehnert

. 2010. Training in normobaric hypoxia and its effects on acute mountain sickness after rapid ascent to 4559 m. High Alt Med Biol, 11:19–25.

22.

Schoonman

, Sandor

, Nirkko

, Lange

, Jaermann

, Dydak

, Kremer

, Ferrari

, Boesiger

, Baumgartner

. 2008. Hypoxia-induced acute mountain sickness is associated with intracellular cerebral edema: A 3 T magnetic resonance imaging study. J Cereb Blood Flow Metab, 28:198–206.

23.

Tissot van Patot

, Leadbetter

, Keyes

, Bendrick-Peart

, Beckey

, Christians

, Hackett

. 2005. Greater free plasma VEGF and lower soluble VEGF receptor-1 in acute mountain sickness. J Appl Physiol, 98:1626–1629.

24.

Walter

, Maggiorini

, Scherrer

, Contesse

, Reinhart

. 2001. Effects of high-altitude exposure on vascular endothelial growth factor levels in man. Eur J Appl Physiol, 85:113–117.

25.

, Severinghaus

. 1998. Rat brain VEGF expression in alveolar hypoxia: Possible role in high-altitude cerebral edema. J Appl Physiol, 85:53–57.