J. Biomedical Science and Engineering, 2011, 4, 214-218 JBiSE
doi:10.4236/jbise.2011.43030 Published Online March 2011 (http://www.SciRP.org/journal/jbise/).
Published Online March 2011 in SciRes. http://www.scirp.org/jour nal/JBiSE
The diagnostic significance and the assessment of the value of
vascular endothelial growth factor as a marker for success of
chemical pleurodesis in malignant pleural effusion
Dalokay Kilic1, Alper Findikcioglu1, Goknur Alver2, Tolga Tatar1, Hakan Akbulut3, Ahmet Hatipoglu4
1Department of Thoracic Surgery, Baskent University, School of Medicine, Baskent University Hospital, Ankara, T urkey;
2Department of Thoracic Surgery, Ataturk Hospital, Ankara, Turkey;
3Department of Medical Oncology, Ankara University, School of Medicine, Ankara, Turkey;
4Department of Thoracic Surgery, Ankara University, School of Medicine, Ankara, Turkey.
Received 20 November 2009; revised 12 December 2009; accepted 16 December 2009.
ABSTRACT
Differential diagnosis of pleural effusion is an im-
portant issue, since the treatment modalities and
prognosis strictly depend on early and correct diag-
nosis of the underlying etiology. We assessed the effi-
cacy of vascular endothelial growth factor (VEGF) in
the differential diagnosis of patients with malignant
and non-malignant pleural diseases. And also is as-
sessed of the VEGF as a marker for success of chemi-
cal pleurodesis in malignant pleural effusion. Pleural
effusions of 40 pat ients with a mean age of 55 (range,
26 to 78 years) were examined. A total of 20 patients
had malig nant pleural eff usion; malig nant meso- t he-
lioma (n = 7), lung cancer (n = 5) and metastatic ma-
lignancies (n = 8). Twenty patients had benign pleural
effusion; fibrinous pleuritis (n = 6), tubercu- losis (n
= 3) empyema (n = 5), congestive heart failure (n = 3),
and acute pancreatitis (n = 3). Definitive di- agnosis
was obtained in all cases with blind or open pleural
biopsy, and cytological examination. VEGF levels
were determined by enzyme-linked immu- nosorbent
assay. The VEGF level of pleural effusion was com-
parably higher in the malignant group. The mean
level of VEGF in patients with malignant pleu- ral
effusions (21.7 ± 1.8 ng/ml) was significantly (P <
0.001) higher than that of (13.2 ± 1.5 ng/ml) non-ma-
lignant effusions. No significant difference was found
regarding the VEGF levels and histological types in
malignant pleural effusions. Negative correlation was
observed between success rate of pleurodesis and
VEGF level of pleural effusion (p = 0.015). The
measurement of VEGF levels in pleural effusion may
be useful to differentiate malignant from nonmalig-
nant pleural effusions. VEGF level may also be an
important prognostic marker for ef- fective treatment
of the patients who had malignant pleural effusions
with pleurodesis. It is important issue in here
whether VEGF could be useful in prog- nostication of
outcome of chemical pleurodesis or not.
Keywords: Malig nant Pleural Effusion; Pleural Effusion;
Chemical Pleurodesis; Vascularendothelial Growth Factor
1. INTRODUCTION
Pleural effusion is an important problem in malign ant or
non-malignant pleural disease, causing severe symptoms
such as dyspnea and chest pain. Management of the
pleural effusion (PE) depends on the underlying etiology.
Inflammatory PE can be treated easily with success,
contrary to malignant PE, in which the main goal is to
decrease symptoms and increase the quality of life as
much as possible. This “mandatory” differential diagno-
sis is still difficult, time-consuming and expensive.
Pleural fluid accumulation in malignancy is generally
believed to be secondary to lymphatic obstruction by
malignant cells [1]. However recent studies pointed to
vascular endothelial growth factor (VEGF) as a key
agent in this entity [2,3]. VEGF, produced by malignant
pleural tissue, is thought to both enhance tumor angio-
genesis leading to local growth, and increase vascular
permeability leading to PE [2,3].
We conducted a study to investigate th e role of VEGF
in differentiating between malignant and non-malignant
pleural effusions in a series of 40 patients.
2. MATERIAL AND METHODS
2.1. Materials
We measured the VEGF levels of pleural effusions in 40
patients consisting of 24 (60%) male and 16 (40%) fe-
male patients with a mean age of 54.5 years (range, 26 to

D. Kilic et al. / J. Biomedical Science and Engineering 4 (2011) 214-218 215
78 years) by enzyme-linked immunosorbent assay
(ELISA). Twenty patients had malignant pleural effusion
including malignant mesothelioma (n = 7), lung cancer
(n = 5), metastasis from genitourinary system malignan-
cies (renal cell Ca, endometrium Ca, ovarian Ca) (n = 3),
metastasis from breast carcinoma (n = 3), adenocarci-
noma metastasis from gastrointestinal system (n = 1),
Non Hodgkin-lymphoma (n = 1). Twenty patients had
benign pleural effusion associated with fibrinous pleu-
ritis (n = 6), tuberculosis (n = 3) empyema (n = 5), con-
gestive heart failure (n = 3), and acute pancreatitis (n =
3). Definitive diagnosis was obtained in all patients with
either open or blind pleural biopsies and cytological
examination (Table 1).
2.2. Analysis of Material Properties and Invitro
Study
VEGF concentrations were measured using an enzyme-
linked immunosorbent assay (ACCUCYTE®, assay sys-
tem, Human VEGF ELISA kit, Cytimmune Sciences, Inc.
Maryland USA). Tecnique of sampling and storage of
PE: Pleural effusion was collected in a sterile centrifuge
tube and centifugated at 3000 rpm for 10 min at 4˚C.
The cell-free supernatant was then separated and stored
immediately at –70˚C until assayed for VEGF. VEGF
level in PE measured in duplicate for each sample with
an ELISA kit that recognizes the soluble isoform VEGF
165. This assay is sensitiv e to 0.195 ng/ml. The kit used
for the detection of total (bound and unbound) VEGF is
designated to capture a specific VEGF complex consist-
ing of VEGF antibody, biotinyl-ated VEGF and sam-
ple/standard. The optical density was measured at 492
Table 1. Definitive diagnosis and VEGF levels in patients with
pleural effusions.
Malignant
effusion n/Mean VEG F level (ng/ml) Nonmalignant effusion n/Mean VEG F level (ng/ml)
Malignant
mesothelioma 7 (20.7 ± 3.8) Fibrinous Pleurit 6 (12.8 ± 2.1)
Lung Cancer
(ADC*) 5 (24.2 ± 2.8) Empyema 5 (20.8 ± 1.8)
Metastatic
Malignancies 8 (20. 9 ± 3.6) Congestive Heart Failure 3 (7.3 ± 1.2)
Breast Cancer 3 Acute
Pancreatitis 3 (12.1 ± 4.4)
Metastasis of
GUS** Cancer 3 Tuberculosis 3 (8.8 ± 2.7)
Metastasis of
GIS*** ADC 1 - -
Non-Hodgkin’s
Lymphoma 1 - -
Total 20 (21.7 ± 1.8) - 20 (13.2 ± 1.5)
ADC*. Adenocarcinoma; GUS**. Genitourinary system; GIS***. Gastro-
intestinal system
nm. The glucose and lactate dehydrogenase (LDH) meas-
urements were obtained from fluid collection in a sterile
tube using an automated analyzer.
2.3. Statistical Analysis
Statistical comparisons of baseline data between groups
were performed by the Mann-Whitney U test as appro-
priate. Data were considered statistically significant if P
values were less than 0.05. Data were expressed as mean
± the standard deviation (SD). Correlations were ana-
lyzed with the Spearman rank order correlation. All sta-
tistical analyses were performed with the Statistical
Package for the Social Sciences (version 11.0; SPSS,
Inc., Chi c ago, Illin ois, USA).
3. RESULTS
3.1. Definitive Diagnosis and VEGF Levels in
Patients with Pleural Effusions
Twenty patients had malignant pleural effusion associ-
ated with malignant mesothelioma (20.7 ± 3.8 ng/ml),
lung cancer (24.2 ± 2.8 ng/ml), metastasis from genitou-
rinary system carcinomas (renal cell Ca, endometrium
Ca, ovarian Ca) (21.3 ± 6.9 ng/ml) metastasis from breast
carcinoma (19.7 ± 4.3 ng/ml), adenocarcinoma metasta-
sis from gastrointestinal system (23.8 ng/ml), and Non-
Hodgkin’s Lymphoma (21.1 ng/ml). Twenty patients had
benign pleural effusion associated with fibrinous pleu-
ritis (12.8 ± 2.1 ng/ml), tuberculosis (8.8 ± 2.9 ng/ml)
empyema (20.8 ± 1.8 ng/ml), congestive heart failure
(7.3 ± 1.2 ng/ml), and acute pancreatitis (12.1 ± 4.4
ng/ml) (Table 1).
3.2. Laboratory Results
The mean level of VEGF in patients with malignant
pleural effusions (21.7 ± 1.8 ng/ml) was significantly
higher than that of non-malignant (13.2 ± 1.5 ng/ml)
pleural effusions (p < 0.001) (Figure 1). No significant
differences were observed in concentration of pleural
VEGF in different histological types of malignant pleu-
ral effusion (malignant mesothelioma, lung cancer (p =
0.482) and malignant mesothelioma, other metastatic car-
cinomas (excluding lung cancer) (p = 0.354). Mean LDH
level of PE was 614.32  288.62 IU/L and mean glucose
level of PE was 77.0  30.6 mg/dL. The closest correla-
tion was between the pleural effusion VEGF level and
the LDH level (r = 0.894, p < 0.001) ( Figure 2). No sig-
nificant correlation existed between the pleural effusion
VEGF level and the glucose level (r = 0.079, p = 0.628).
3.3 Comparison of Success of Chemical
Pleurodesis and VEGF Level
Seventeen patients who had malignant pleural effusion
underwent chemical pleurodesis. Five of the patients

D. Kilic et al. / J. Biomedical Science and Engineering 4 (2011) 214-218 216
Figure 1. Comparison of VEGF levels of pleural effusion be-
tween the patients with malignant and nonmalignant disease.
Figure 2. Relationship between pleural fluids LDH levels and
VEGF levels are shown.
(29%) were readmitted with reaccumulation in 60 days.
The mean VEGF level of PE (28.3  3.7 ng/ml) of these
re- admitted patients were higher than th at of (17.4  6.7
ng/ml) other patients (p = 0.015). Negative correlation
was observed between success rate of pleurodesis and
VEGF level of PE (Figure 3).
4. DISCUSSION
Pleural fluid accumulation is a common clinical problem
in pleural diseases. Treatment of PE in inflammatory
diseases is much easier than in malignant PE. The pri-
mary target in malignant PE, except for early stages of
malignant mesothelioma is to prevent further accumula-
tion of the effusion and to increase the quality of life,
which means that “cure” is no more expected. Surgical
treatment including decortication or pleuropneumonec-
tomy has been proven to increase the survival rates in
early stages of malign mesothelioma [4]. This difference
among treatment approaches in benign and malignant
Figure 3. Negative correlation between VEGF levels of PE
and success rate of pleurodesis in patients with “group 1” or
without “group 2” recurrent pleural effusion.
diseases makes the accurate and early diagnosis manda-
tory, in order not to let medicine harm the patient.
Pleural fluid cytology and blind pleural biopsy are the
methods most commonly used but are inadequate pro-
cedures for the diagnosis. In some studies, blind pleural
biopsy has been reported to be inadequate in up to 40%
of the patien ts [5,6]. Th is situatio n puts forward the n eed
for a different method directed to pleural fluid. Certain
molecular markers, if proven to be sensitiv e and specific
enough, can help the physician decide whether the pa-
tient should have further investigation or not to d iagnose
a suspected malignancy, i.e. open pleural biopsy (VATS,
mini- thoracotomy) or not. Among these biomarkers, in-
sulin growth fa ct or, hepatocyte gro wt h facto r and Simian
Virus-40 have been proven to play an important role in
the development and progression of malignant meso-
thelioma [7-9]. Likewise, a recent study has shown that
plasma N/CD-13 activity had a strong correlation with
tumor load in malignant pleural diseases [10].
VEGF is another molecule that has been expressed,
and has been taking an important role in the develop-
ment of malignant pleural effusion. Although some au-
thors have claimed that lymphatic blockage resulting
from tumor cells is the main contributing factor for the
development of malignant pleural effusion, others have
accused of VEGF at the first place [11-13]. VEGF is a
disulfide-bonded dimeric glycoprotein. Its molecular
weight is 34-45 kD, and the most common types are
VEGF 165 and VEGF 121 [14]. The suspected molecu-
lar mechanisms are increase in local vascular permeabil-
ity and stimulation of tumor cell growth by angiogenesis,
both stimulated by VEGF produced by tumor cells
[1-3,11,12]. Increase in capillary permeability functions
through binding with fms-like tyrosine kinase receptor
(FLT-1). The FLT-1 VEGF receptors have been identi-

D. Kilic et al. / J. Biomedical Science and Engineering 4 (2011) 214-218 217
fied in pleural mesothelial cells and vascular endothelial
cells also high densities in infiltrating malignant tissue
[13]. It has been shown to be important regulatory sys-
tems for angiogenesis and vasculogenesis [12].
Many studies have shown that several types of tumor
cells express VEGF. VEGF is an important cytokine in
lung cancer [15,16 ] Matsuyama et al. reported a positive
correlation between serum VEGF and stage progression
of the disease [15]. Measurement of serum VEGF levels
was suggested to be useful to evaluate lung cancer pro-
gression. Similar data are obtained from studies directed
to gastric, colorectal and renal cell carcinomas [17-19].
It also seems to be an important determinant in malign-
nant pleural disease. Tickett et al. have reported that me-
dian VEGF levels of 2500 pg/ml in malignant PE were
significantly higher than of 305pg/ml in the non-malig-
nant group [13]. Our study is in accordance with these
previous data, showing an increased level of VEGF in
malignant effusion. Lim et al. compared VEGF levels of
tuberculous PE (median, 994 pg/ dl) and malignant PE
(2418 pg/ dl), reaching to similar results obtained in our
study [20]. In our se ries, VEGF of PE levels in empyem a
patients were higher than other nonmalignant groups (p
= 0.02). There were no significant differences between
VEGF levels in empyema and in lung cancer (p = 0.349).
Similar results have been reported by Thickett [13]. The
known biological functions of VEGF may promote the
accumulation of pleural fluid and increase the loculation
and organization of empyema [21]. In our study, when
VEGF levels found in empyema cases were excluded,
the levels were more significantly higher in malignant
than in non-malignan t PE (P < 0.0001) similar to Thick-
ett’s study [13].
If patients having high levels of VEGF in suspicious
PE that cannot be diagnosed with blind biopsy of the
pleura or cytological techniques, patients should undergo
an early open biopsy. Thickett et al. reported false nega-
tive rate of 50 % in the malignant group after blind bi-
opsy with a median level of VEGF higher than the level
in the benign group. However, in the abovementioned
study, additional invasive procedures were needed for
de finitive diagnosis [13]. In our study, the results obtained
via pleural fluid cytology in addition to blind biopsy
were false negative in 25% (n = 5) of the malignant
group mean with a mean VEGF level of PE = 21.2 ± 1.4
ng/ml and definitive diagnosis had to be made by VATS
and mini-thoracotomy.
VEGF and LDH in pleural effusion correlated because
they are both crude markers of the inflammatory re-
sponse [22].
A new experimental study for the treatment of PE in-
cludes VEGF receptor (receptor tyrosine kinase) block-
age model for human tumors [23,24]. VEGF receptor
blockage model may also provide a new therapeutic ap-
proach for pleural malignancies. Further studies are nec-
essary to outline the feasibility of VEGF receptor block-
age model as a therapeutic modality.
As a conclusion, malignant pleural effusions show
significantly higher VEGF levels compared with non-
malignant pleural effusions. Thus, assessment of VEGF
levels may be used to differentiate malignant from non-
malignant pleural effusions as an adjunct to conventional
differential diagnostic techn iques. Low lev el of VEGF in
malignant PE patients may be a good prognostic marker
for effective treatment of malignant PE with pleurodesis.
REFERENCES
[1] Sahn, S.A. (1997) Pleural diseases related to metastatic
malignancies. European Respiratory, 10, 1907-1913.
doi:10.1183/09031936.97.10081907
[2] Fontanini, G., Vignati, S., Boldrini L., Chinè, S., Silvestri,
V., Lucchi, M., Mussi, A., Angeletti, C.A. and Bevilac-
qua, G. (1997) Vascular endothelial growth factor is as-
sociated with neovascularization and influences progres-
sion of non-small cell lung carcinoma. Cliical Cancer
Research, 3, 861-865.
[3] Yanagawa, H., Takeuchi, E., Suzuki, Y., Ohmoto, Y.,
Bando, H. and Sone, S. (1999) Vascular endothelial
growth factor in with malignant pleural effusion associ-
ated with lung cancer. Cancer Immunology Immuno-
therapy, 48, 396-400. doi:10.1007/s002620050592
[4] Rusch, V.W. and Venkatraman, E.S. (1999) Important
prognostic factors in patients with malignant pleur al meso-
thelioma, managed surgically. Annalls of Thoraic Surgery,
68, 1799-1804. doi:10.1016/S0003-4975(99)01038-3
[5] Poe, R.H., Isreal, R.H. and Utell, M.J. (1984) Sensitivity,
spesificity, and predictive values of closed pleural biopsy.
Arcives of Internal Medicine, 144, 325-328.
doi:10.1001/archinte.144.2.325
[6] Sahn, S.A. (1988) The pleura. The American Review of
Respiratory Disease, 138, 184-234.
[7] Lee, T.C.Y., Zhang, C., Aston R., Hintz, R., Jagirdar, J.,
Perle, M.A., Burt, M. and Rom, W.N. (1993) Normal hu-
man mesothelial cells and mesothelioma cell lines ex-
press insulin like growth factor I and associated mole-
cules. Cancer Research, 53, 2858-2864.
[8] Harvey, P., Warn, S.A., Dobbin, S., Arakaki, N., Daiku-
hara, Y., Jaurand, M.C. and Warn, R.M. (1998) Expres-
sion of HGF/SF in mesothelioma cell lines and its effects
on cell motility, proliferation and morphology. British
Journal of Can cer, 77, 1052-1059.
doi:10.1038/bjc.1998.176
[9] Cacciotti, P., Strizzi, L., Vianale, G., Iaccheri, L., Libener,
R., Porta, C., Tognon, M., Gaudino, G. and Mutti, L.
(2002) The presence of simian-virus 40 sequences in
mesothelioma and mesothelial cells is associated with
high levels of vascular endothelial growth factor. Ameri-
can Journal of Respirartory Cell and Molecular Biology,
26, 189-193.
[10] Hensbergen, Y.V., Broxterman, H.J., Hanamaaijer, R.,
Jorna, A.S., van Lent, N.A., Verheul, H. M., Pi nedo, H.M.
and Hoekman, K. (2002) Soluble aminopeptida se N/CD13
in malignant and nonmalignant effusion and intratumoral

D. Kilic et al. / J. Biomedical Science and Engineering 4 (2011) 214-218 218
fluid. Clinical Cancer Research, 8, 3747-3754.
[11] Yano, S., Shinohara, H., Herbst, R.S., Kuniyasu, H., Bu-
cana, C.D., Ellis, L.M. and Fidler, I.J (2000) Production
of experimental malignant pleural effusions is dependent
on invasion of the pleura and expression of vascular en-
dothelial growth factor/vascular permeability factor by
human lung cancer cells. American Journal of Pathol-
ogy, 157, 1893-1903.
doi:10.1016/S0002-9440(10)64828-6
[12] Strizzi, L., Catalano, A., Vianale, G., Orecchia, S., Casal-
ini, A., Tassi, G., Puntoni, R., Mutti, L. and Procopio, A.
(2001) Vascular endothelial growth factor is an autocrine
growth factor in human malignant mesothelioma. Jour-
nal of Pathology, 193, 468-475. doi:10.1002/path.824
[13] Thickett, D.R., Armstrong, L. and Millar, A.B. (1999)
Vascular endothelial growth factor (VEGF) in inflamma-
tory and malignant pleural effusions. Thorax, 54,
707-710. doi:10.1136/thx.54.8.707
[14] Ferrara, N. (1999) Molecular and biological properties of
vascular endothelial growth factor. J. Mol. Med, 77,
527-543. doi:10.1007/s001099900019
[15] Matsuyama, W., Hashiguchi, T., Mizoguchi, A., Iwami,
F., Kawabata, M., Arimura, K. and Osame, M. (2000)
Serum levels of vascular endothelial growth factor de-
pendent on the stage progression of lung cancer. Chest,
118, 948-951. doi:10.1378/chest.118.4.948
[16] Mattern, J., Koomagi, R. and Volm, M. (1997) Coexpres-
sion of VEGF and bFGF in human epidermoid lung car-
cinoma is associated with increased vessel density.
Anti-cancer Research, 17, 2249-2252.
[17] Eroglu, A., Demirci, S., Ayyildiz, A., Kocaoglu, H.,
Akbulut, H., Akgul, H. and Elhan, H.A. (1999) Serum
concentration of vascular endothelial growth factor and
nitrite as an estimate of in vivo nitric oxide in patients
with gastric cancer. British ournal of. Cancer, 80,
1630-1634. doi:10.1038/sj/bjc/6690573
[18] Akbulut, H., Altuntas, F., Akbulut, K.C., Ozturk, G.,
Cindoruk, M., Unal, E. and Icli, F. (2002) Prognostic role
of serum vascular endothelial growth factor, basic fibro-
blast growth and nitrit oxide in patients with colorectel
carcinoma. Cytok ine, 20, 184-190.
doi:10.1006/cyto.2002.1993
[19] Paradise, V., Lagha, N.B. and Zeimoura, L. (2000) Ex-
pression of vascular endothelial growth factor in renal
cell carcinomas. Vir ch ows A rchiv, 436, 351-356.
doi:10.1007/s004280050458
[20] Lim, S.C., Jung, S.I., Kim, Y.C. and Park, K.O. (2000)
Vascular endothelial growth factor in malignant and tu-
berculous pleural effusions. Journal of Korean Medicine
Science, 15, 279-283.
[21] Nehls, V. and Herrmann, R. (1998) The configuration of
fibrin clots determines capillary morphogenesis and endo-
thelial cell migration. Microvascular Research, 2, 9-20.
[22] Taichman, N.S., Young, S., Cruchley, A.T., Taylor, P. and
Paleolog, E. (1997) Human neutrophils secrete vascular
endothelial growth factor. Journal of Leukocyte Biology,
62, 397-400.
[23] Masood, R., Tong, Zheng, J.C., Smith, D.L., Hinton, D.R.
and Gill, P.S. (2001) Vascular endothelial growth factor
(VEGF) is an autocrine growth factor for VEGF recep-
tor-positive human tumors. Blood, 98, 1904-1913.
doi:10.1182/blood.V98.6.1904
[24] Verheul, H.M.W., Hoekman, K., Jorna, A.S., Smit, E.F.
and Pinedo, H.M. (2000) Targeting vascular endothelial
growth factor blockade: Ascites and pleural effusion
formation. The Oncologist, 5, 45-50.
doi:10.1634/theoncologist.5-suppl_1-45