Circulating plasma levels of vascular endothelial
growth factor in patients with sleep disordered
Arschang Valipoura,*, Brigitte Litschauerb, Friedrich Mittermayerb,
Helmuth Rauscherc, Otto Chris Burghubera, Michael Wolztb
aDepartment of Respiratory and Critical Care Medicine and the Ludwig Boltzmann Institute for Chronic
Obstructive Pulmonary Disease, Otto-Wagner-Spital, Vienna, Austria
bDepartment of Clinical Pharmacology, Allgemeines Krankenhaus Wien, Vienna, Austria
cSleep Laboratory, Department of Pulmonary Medicine, Krankenhaus Lainz, Vienna, Austria
Received 16 December 2003; accepted 19 April 2004
Summary Introduction: Cellular vascular endothelial growth factor (VEGF) expres-
sion is increased in response to regional hypoxia, however, contradictory results
were reported on the effects of systemic hypoxemia on circulating VEGF levels. This
study investigated plasma concentrations of VEGF in patients with a variable degree
of overnight hypoxemia due to sleep disordered breathing (SDB).
Methods: VEGF levels were assessed by ELISA in non-activated (VEGFbl) and
thrombin stimulated platelet rich plasma (VEGFprp) of 45 patients with SDB: Group
1patients with obstructive sleep apnea and an apnea–hypopnea index (AHI) 4 15/h;
Group 2 subjects with an AHIo5=h; Group 3patients on CPAP treatment for sleep
Results: 39 patients were included in the final analysis. Patients in Group 1 had a
higher %time of sleep with SaO2o90% and a significantly lower mean and minimum
overnight oxygen saturation than subjects in Group 2 and patients in Group 3
(Po0:05). Despite significant differences in overnight oxygenation, VEGFbl and
VEGFprp concentrations were not significantly different between the three study
groups. However, plasma levels of VEGFbl were significantly higher ðP ¼ 0:02Þ in SDB
patients with arterial hypertension (n ¼ 19; VEGFbl: 14:073:3 pg=ml) than in those
without arterial hypertension (n ¼ 20; VEGFbl: 10:975:2 pg=ml). There were no
relationships between VEGF levels and polysomnographic oxygenation parameters.
In univariate analysis we observed significant relationships for VEGFbl with BMI
(C: 0:393; Po0:05) and serum fibrinogen (C: 0:399; Po0:05).
Conclusions: Circulating plasma VEGF levels in patients with sleep disordered
breathing may be unrelated to night time hypoxemia (257 Words).
& 2004 Elsevier Ltd. All rights reserved.
ARTICLE IN PRESS
Sleep disordered breath-
Obstructive sleep apnea
Abbreviations: VEGF, Vascular endothelial growth factor; SDB, Sleep disordered breathing; OSA, Obstructive sleep apnea; BMI, Body
mass index; CPAP, Continuous positive airway pressure; CVD, Cardiovascular disease; RBC, Red blood cell count; AHI, Apnea–hypopnea-
index; LOQ, Lower detection limit
*Corresponding author. Department of Respiratory and Critical Care Medicine Otto-Wagner-Spital Sanatoriumsstr. 2, 1140 Wien,
Austria. Tel.: þ43-1-91060-41008; fax: þ43-1-91060-49827.
E-mail address: firstname.lastname@example.org (A. Valipour).
0954-6111/$-see front matter & 2004 Elsevier Ltd. All rights reserved.
Respiratory Medicine (2004) 98, 1180–1186
Patients with sleep disordered breathing (SDB) are
at enhanced risk for hypertension and cardiovas-
cular disease (CVD).1,2The risk for CVD, however, is
the same for patients with mild and those with
more severe SDB.2
Thus, it remains unclear
whether repetitive hypoxic episodes during the
night or other mechanisms are responsible for the
link between CVD and SDB.
Vascular endothelial growth factor (VEGF) is a
cytokine with potent angiogenic properties, indu-
cing proliferation of endothelial cells and modulat-
ing thrombogenicity.3Increased VEGF expression
occurs following hypoxia4by enhanced transcrip-
tion of the VEGF gene mediated by the hypoxia
inducible factor-1 (HIF-1). Hypoxia may be involved
in the development of CVD via increased VEGF
expression inhuman atherosclerotic
Although cellular VEGF expression is stimulated by
regional hypoxia, contradictory results were re-
ported on the effects of systemic hypoxemia on
circulating VEGF. Maloney et al.6did not observe
increased VEGF levels in subjects exposed to
hypobaric hypoxia. Similarly, several other investi-
gators reported unchanged VEGF levels despite
exposure to a variable degree of either short or
long-term hypoxia.7–9In contrast, recent studies
have reported elevated blood VEGF levels in
patients with episodic hypoxia due to sleep
In most of these reports VEGF was measured in
serum rather than plasma, which may not reflect
VEGF synthesis by peripheral tissue.14Plasma VEGF,
however, may reflect more closely free, circulating
VEGF.15,16Therefore, the aim of our study was to
investigate plasma concentrations of VEGF in
patients with a variable degree of nocturnal
hypoxia due to sleep disordered breathing.
The study protocol was approved by the Ethics
Committee of the Vienna City Council. Informed
consent was obtained from each subject prior to
Forty-five participants including patients with
sleep apnea (Group 1), subjects without sleep
apnea (Group 2) and patients on continuous
positive airway pressure (CPAP) treatment for sleep
apnea (Group 3) were studied. Standard polysom-
nography (Jaeger SleepLab Pro, Germany) was
performed including EEG, EOG, submental EMG,
ECG, oro-nasal airflow using thermistors, induc-
tance plethysmography to assess thoraco-abdom-
oxyheamoglobin saturation (SaO2). Apneas and
hypopneas were defined according to standard
criteria.17The total number of apneas and hypop-
neas were recorded (AHtot) and divided by the
total sleep time (TST) to give an apnea–hypopnea
index (AHI) per hour of sleep (AHI/h). Measure-
ments such as %time of sleep o 90% SaO2, sleep
efficiency, arousal index,18mean low and minimum
oxygen saturation were obtained after careful
staging. Standard pulmonary function testing and
arterial blood gas analysis were performed. Exclu-
sion criteria were: a history of cancer, myocardial
infarction within the past 12 months, stroke or
severe heart failure, preceding thoracic surgery,
anemia (Hbo12g/dl), presence of autoimmune,
renal or liver disease and concomitant oral antic-
Patients with a diagnosis of OSA, based on
symptoms, clinical findings and an AHI of greater
than 15/h were included in Group 1 (n ¼ 13).
Subjects with a history of loud snoring, an AHI of
less than 5/hour and evidence of snoring, served as
controls and were recruited in Group 2 (n ¼ 13).
Patients with an AHI between 5 and 15/h were
excluded from the analysis in order to avoid an
overlap between the groups (n ¼ 4).
In addition we studied 15 patients with pre-
viously diagnosed OSA, who were on nasal CPAP
therapy for at least 12 months. These patients were
age and body weight matched to patients in Group
1. Their sleep study was performed with the CPAP
pressure set to the level used for home therapy.
Compliance data was available from the patient’s
device. Subjects with an average compliance of less
than 4.5h/night(n ¼ 1)
AHI45=h (n ¼ 1) despite CPAP therapy were ex-
cluded. Data is reported on the remaining 13
Measurements of VEGF
Venous blood samples were collected between 6.00
and 6.30a.m. at the end of the polysomnography.
For basal plasma VEGF concentrations (VEGFbl),
blood was drawn into ice-cooled tubes containing
an anticoagulant solution. Platelet-free plasma
samples were prepared by centrifugation, the
supernatant was collected, aliquoted and frozen
at ?801C until assayed. For stimulated VEGF in
platelet rich plasma (VEGFprp), blood was drawn
into sodium citrate containing tubes, centrifuged
and activated with 2–4U/ml of thrombin. After
clotting, sampleswere centrifugedandthe
ARTICLE IN PRESS
Circulating plasma levels of VGEF in patients with SDB 1181
supernatant was removed, aliquoted and stored at
–801C. VEGF was measured in duplicate determina-
tions by a quantitative ELISA (R&D Systems,
Germany). The assay is calibrated with a mono-
VEGF165, the major isoform of VEGF. The lower
detection limit (LOQ) of the system is 9.0pg/ml,
the intra- and inter-assay variation is 6.7% and
8.8%, respectively. Plasma concentrations below
LOQ were set to LOQ/2. Platelet VEGF content
(VEGFct) was defined as the amount of VEGFprp per
platelets ?106, obtained from the platelet count
of the respective blood sample. Full blood count
and serum fibrinogen were also determined. La-
boratory staff was blinded to clinical data.
All data are reported as means (SD). Due to a non-
homogenous distribution of the primary outcome
parameters a Van-der-Waerden test was used to
assess differences in VEGFbl, VEGFprp and VEGFct
between the study groups. Between group analysis
for other physiological and biological parameters
were performed using a Student’s paired t-test in
normally distributed data and a Mann–Whitney test
if data suggested a non-homogenous distribution.
Univariate linear regression analysis was performed
to determine relationships between plasma VEGF
concentrations and parameters of polysomnogra-
phy, blood gas analysis, chemistry, body mass
index (BMI), age and lung function parameters.
A stepwise multiple regression analysis was then
used to record independent associations of the
above mentioned variables with the outcome
parameters. A P-value less than 0.05 was consid-
Baseline characteristics of the study population are
given in Table 1. The three groups were similar with
regard to age, BMI and smoking history. However,
patients in Group 1 had a significantly higher red
blood cell count (RBC) and haemoglobin concentra-
There was a large variability between patients in
VEGF measurements. Plasma levels of VEGFbl
ranged from 4.5 to 25.2pg/ml, VEGFprp from 11
to 303pg/ml and calculated VEGFct from 0.376 to
sults, daytime arterial paO2 values and plasma
concentrations of VEGFbl, VEGFprp and VEGFct for
the three study groups are summarized in Table 2.
Patients in Group 1 had significantly lower over-
night oxygen saturation parameters and a higher
AHI, AHtot and %time of sleep with SaO2o90%:
Despite significant differences in overnight oxyge-
nation between the groups, there were no differ-
ences in plasma concentrations of VEGFbl, VEGFprp
Plasma levels of VEGFbl, however, were higher
(P ¼ 0:02) in SDBpatients
14:073:3 pg=ml) than in those without arterial
hypertension (VEGFbl : 10:975:2 pg=ml), without
significant differences in SDB severity.
Using linear regression analysis we found sig-
nificant relationships for plasma VEGFbl with BMI
(Fig. 1, Correlation coefficient C: 0:393; Po0:05)
and serum fibrinogen (Fig. 2, C: 0:399; Po0:05).
Subgroup analysis (Group 1 plus Group 2) also
revealed a significant correlation between VEGFbl
and the RBC (C ¼ 0:407; Po0:05). Stepwise multi-
ple regression analysis, however, did not identify
independent predictors of circulating plasma VEGF
ARTICLE IN PRESS
and laboratory parameters of the study population.
Clinical characteristics, FEV1/FVC ratio
(n ¼ 13)
(n ¼ 13)
(n ¼ 13)
(% of total)
(% of total)
BMI, Body mass index; RBC, Red blood count; WBC, White
blood cell count; Hb, haemoglobin. Values are given as
nPo0:05 versus patients in Group 1.
1182 A. Valipour et al.
The present study investigated plasma concentra-
tions of the circulating, unbound form of soluble
VEGF (VEGFbl), VEGF in platelet rich plasma
(VEGFprp) and calculated VEGF platelet content
(VEGFct) in patients with a variable degree of
episodic hypoxemia due to sleep disordered breath-
ing. The major findings are: (1) Lack of a difference
in VEGF measurements between the study groups,
despite significant differences in overnight oxyge-
nation;(2) Higher VEGFbl levels in SDB patients with
hypertension than in those without;(3) Evidence of
a modest relationship of plasma VEGFbl with the
BMI, serum fibrinogen and RBC.
Our findings are in contrast with previous
reports10,11,13in which VEGF levels were higher in
patients with OSA than in controls. These studies
ARTICLE IN PRESS
in patients with sleep disordered breathing.
Results of polysomnography, daytime arterial paO2and plasma VEGFbl, VEGFprp and calculated VEGFct
Group 1 Group 2 Group 3P
Total sleep time
%time of sleep with
Mean low SaO2
paO2, partial arterial oxygen pressure; AHtot (n/TST), total number of apneas and hypopneas during total sleep time; AHI,
Apnea–hypopnea index; SaO2, arterial oxygen saturation; n.s., non-significant.
Data presented as mean (SD). P-value for between group analysis using a Van der Waerden test for VEGF data and a Student’s t-
test or Mann–Whitney test for other parameters depending on data distribution.
Circulating plasma levels of VGEF in patients with SDB1183
may suffer from two main limitations: First, control
groups were not body-weight matched, which
might have resulted in higher VEGF levels in the
more obese OSA patients.19Second, these reports
have measured serum VEGF, which is an inaccurate
indicator of circulating VEGF because of its variable
release from platelets
tion.14,16Serum VEGF reflects platelet count rather
than VEGF synthesis by peripheral tissues with large
inter-individual variations in VEGF generation in
clotted samples.15,20,21Plasma VEGF, on the other
hand, reflects free, circulating and thus newly
produced VEGF after equilibration with platelet
The only other report that has investigated into
the relationship between plasma VEGF and sleep
apnea was performed by Lavie et al.22They found
higher VEGF levels in patients with OSA than
controls and a reduction of VEGF concentrations
with CPAP therapy in a subgroup of patients.
ARTICLE IN PRESS
Correlation coefficient; CI ¼ 95% confidence interval; PI ¼ 95% prediction interval.
Univariate correlation between individual VEGFbl values and BMI for the whole study population. C ¼
Figure 2 Univariate correlation between individual VEGFbl values and serum fibrinogen concentrations for the whole
study population. C ¼ Correlation coefficient; CI ¼ 95% confidence interval; PI ¼ 95% prediction interval.
1184A. Valipour et al.
Noteworthy, there was no reported relationship
between measurements of overnight hypoxemia
and plasma VEGF and, consistent with our findings,
a strong link between the BMI and VEGF levels.
The latter observation supports recent findings of
enhanced VEGF expression in obesity19and may
support the hypothesis of an AHI-independent risk
of CVD in the mostly overweight patients with SDB.1
Several reasons might account for the lack of a
difference in VEGF concentrations between the
study groups. First, it has to be acknowledged that
the majority of patients were suffering from
moderate OSA and only few were exposed to severe
episodic hypoxemia. These patients might have
failed to induce the O2-sensing mechanism for the
induction of VEGF, particularly in those with low
constitutive levels of VEGF.4,23However, if VEGF
had been the link between OSA and CVD, the
patients in this study would have been sufficiently
‘‘severe’’ to show increased VEGF levels.2Second,
there is a pre-defined, inter-individual variation in
the VEGF response to hypoxia.24Given this varia-
tion in VEGF expression, the number of patients in
the respective subgroups might have been too low
to detect inter-group differences. Third, we can
not rule out that the systemic hypoxic response
occurred on the erythropoietin rather than on the
VEGF level. The inductions of VEGF and erythro-
poietin expression are both mediated by HIF-1
regulated genes.25This is supported by the associa-
tion between VEGFbl and the RBC. As patients with
OSA had a higher RBC and Hb, it is conceivable that
the systemic adjustment to hypoxemia was an
increased erythropoietin rather than VEGF expres-
Other than hypoxia associated stimuli for the
VEGF production may be also relevant. Several pro-
inflammatory cytokines such as Interleukin 6 (IL-6)
and TNF-Alpha can potentate VEGF production.26
We observed a link between VEGFbl and serum
fibrinogen,a marker of inflammation
correlated with IL-6 levels.27The role of VEGF in
this context is not entirely clear, however, it may
contribute to the relationship between CVD and
systemic inflammatory processes.
Finally, increased plasma levels of VEGF in
hypertensive patients confirm recent findings28
and may have been an important bias in earlier
studies.10–13Enhanced VEGF production in these
subjects may be involved in the pathogenesis of
complications related to hypertension and merits
exploration in future studies.
In conclusion, monitoring of circulating VEGF
plasma concentrations seems not to reflect sys-
temic tissue hypoxia in patients with SDB. Our
results suggest that a number of alternative
mechanisms of VEGF gene expression need to be
considered. In this context the relationship be-
tween hypoxia, obesity and inflammation might
have a pivotal role on VEGF induction and should
trigger new studies which focus not only on the
stimulating but also on the inhibiting factors of
We appreciate the help of the sleep laboratory staff
of Lainz Hospital. We would like to thank Prof. J.
Strametz-Juranek and Prof. S.G. Spiro for helpful
1. Peppard PE, Young T, Palta M, Skatrud J. Prospective study of
the association between sleep-disordered breathing and
hypertension. N Engl J Med 2000;342:1378–84.
2. Shahar E, Whitney CW, Redline S, Lee ET, Newman AB, Nieto
FJ, O‘Connor GT, Boland LL, Schwartz JE, Samet JM. Sleep
disordered breathing and cardiovascular disease: cross-
sectional results from the sleep heart health study. Am
J Respir Crit Care Med 2001;163:19–25.
3. Ferrara N. Role of vascular endothelial growth factor in
regulation of physiological angiogenesis. Am J Physiol Cell
4. Minchenko A, Bauer T, Salceda S, Caro J. Hypoxic stimulation
of vascular endothelial growth factor expression in vitro and
in vivo. Lab Invest 1994;71(3):374–9.
5. Inoue M, Itoh H, Ueda M, et al. Vascular endothelial growth
factor (VEGF) expression in human coronary atherosclerotic
lesions: possible pathophysiological significance of VEGF in
progression of atherosclerosis. Circulation 1998;98(20):
6. Maloney J, Wang D, Duncan T, Voelkel N, Rouss S. Plasma
vascular endothelial growth factor in acute mountain
sickness. Chest 2000;118:47–52.
7. Gunga HC, Kirsch K, R. ocker L, Behn C, Koralewski E, Davila
EH, Estrada MI, Johannes B, Wittels P, Jelkmann W. Vascular
endothelial growth factor in exercising humans under
different environmental conditions. Eur J Appl Physiol
Occup Physiol 1999;79(6):484–90.
8. Pavlicek V, Marti HH, Grad S, Gibbs JSR, Kol C, Wenger RH,
Gassmann M, Kohl J, Maly FE, Oelz O, Koller EA, Schirlo C.
Effects of hypobaric hypoxia on vascular endothelial growth
factor and the acute phase response in subjects who are
susceptible to high-altitude pulmonary oedema. Eur J Appl
9. Walter R, Maggiorini M, Scherrer U, Contesse J, Reinhart
WH. Effects of high-altitude exposure on vascular endothe-
lial growth factor levels in man. Eur J Appl Physiol
10. Imagawa S, Yamaguchi Y, Higuchi M, Neichi T, Hasegawa Y,
Mukai HY, Suzuki N, Yamamoto M, Nagasawa T. Levels of
vascular endothelial growth factor are elevated in patients
with obstructive sleep apnea–hypopnea syndrome. Blood
11. Schulz R, Hummel C, Heinemann S, Seeger W, Grimminger F.
Serum levels of vascular endothelial growth factor are
ARTICLE IN PRESS
Circulating plasma levels of VGEF in patients with SDB1185
elevated in patients with obstructive sleep apnea and severe Download full-text
12. Gozal D, Lipton AJ, Jones KL. Circulating vascular endothe-
lial growth factor levels in patients with obstructive sleep
apnea. Sleep 2002;25(1):59–65.
13. Teramoto S, Kume H, Yamamoto H, Ishii T, Miyashita A,
Matsuse T, Akishita M, Toba K, Ouchi Y. Effects of oxygen
administration on the circulating vascular endothelial
growth factor (VEGF) levels in patients with obstructive
sleep apnea syndrome. Intern Med 2003;42:681–5.
14. Webb NJA, Bottomley, Watson CJ, Brenchley PEC. Vascular
endothelial growth factor (VEGF) is released from platelets
during blood clotting: implications for measurement of
circulating VEGF levels in clinical disease. Clin Sci 1998; 94:
15. George ML, Eccles SA, Tutton MG, Abulafi AM, Swift RI.
Correlation of plasma and serum vascular endothelial
growth factor levels with platelet count in colorectal
cancer: clinical evidence of platelet scavenging? Clin Cancer
16. Jelkmann W. Pitfalls in the measurement of circulating
17. Phillipson EA, Bowes G. Sleep disorders. In: Update:
pulmonary diseases and disorders. New York: McGraw-Hill;
1982. p. 256–73.
18. American Sleep Disorders Association (ASDA) Report. EEG-
arousals: scoring rules and examples. A preliminary report
from the Sleep Disorders Atlas Task Force of the ASDA. Sleep
19. Miyazawa-Hoshimoto S, Takahashi K, Bujo H, Hashimoto N,
Saito Y. Elevated serum vascular endothelial growth factor is
associated with visceral fat accumulation in human obese
subjects. Diabetologia 2003;46:1483–8.
20. Verheul HM, Hoekman K, Luykx-de Bakker S, Eekman CA,
Folman CC, Broxterman HJ, Pinedo HM. Platelet: transpor-
Respir CritCare Med
ter of vascular endothelial growth factor. Clin Cancer Res
21. Gunsilius E, Petzer A, Stockhammer G, Nussbaumer W,
Schumacher P, Clausen J, Gastl G. Thrombocytes are the
major source for soluble vascular endothelial growth factor
in peripheral blood. Oncology 2000;58:169–74.
22. Lavie L, Kraiczi H, Hefetz A, Ghandour H, Perelman A,
Hedner J, Lavie P. Plasma vascular endothelial growth factor
in sleep apnea syndrome: effects of nasal continuous
positive air pressure treatment. Am J Respir Crit Care Med
23. Koehne P, Willam C, Strauss E, Schindler R, Eckardt KU,
B. uhrer C. Lack of hypoxic stimulation of VEGF secretion
from neutrophils and platelets. Am J Physiol Heart Circ
24. Schultz A, Lavie L, Hochberg I, Beyar R, Stone T, Skorecki K,
Lavie P, Roguin A, Levy AP. Interindividual heterogeneity in
the hypoxic regulation of VEGF: significance for the
development of the coronary artery collateral circulation.
25. Goldberg MA, Schneider TJ. Similarities between the
oxygen-sensing mechanisms regulating the expression of
vascular endothelial growth factor and erythropoietin. J Biol
26. Cohen T, Nahari D, Cerem LW, Neufeld G, Levi BZ.
Interleukin 6 induces the expression of vascular endothelial
growth factor. J Biol Chem 1996;271(2):736–41.
27. Kanabrocki EL, Sothern RB, Messmore HL, Roitman-Johnson
B, McCormick JB, Dawson S, Bremner FW, Third JL,
Nemchausky BM, Shirazi P, Scheving LE. Circadian inter-
relationships among levels of plasma fibrinogen, blood
platelets,and serum interleukin-6. Clin Appl Thromb Hemost
28. Belgore FM, Blann AD, Li-Saw-Hee Fl, Beevers DG, Lip GY.
Plasma levels of vascular endothelial growth factor and its
soluble receptor (SFlt-1) in essential hypertension. Am J
ARTICLE IN PRESS
1186 A. Valipour et al.