Serum Levels of Vascular Endothelial Growth Factor and Cavity Formation in Active Pulmonary TuberculosisAbe Y.a · Nakamura M.b · Oshika Y.b · Hatanaka H.b · Tokunaga T.b · Ohkubo Y.a · Hashizume T.a · Suzuki K.a · Fujino T.a
aDepartment of Respiratory Disease, National Sanatorium Kanagawa Hospital, Hadano, bDepartment of Pathology, Tokai University School of Medicine, Isehara, Japan
Background: In active pulmonary tuberculosis, certain cytokines have been postulated to be related to cavity formation, although the detailed mechanism of cavity formation is not yet known. Objective: We examined the relationship between cavity formation in pulmonary tuberculosis and vascular endothelial growth factor (VEGF), which functions as an angiogenesis factor. Methods: Forty-eight patients with active pulmonary tuberculosis were divided into two groups according to cavity formation as evaluated by chest high-resolution computed tomography. We evaluated serum VEGF levels by enzyme immunoassay. Results: Group A (with cavities) was comprised of 22 patients and group B (without cavities) was comprised of 26 patients. The serum levels of VEGF were significantly higher in group B (58.733 ± 21.612 pg/ml) than those in normal individuals (8.739 ± 3.656 pg/ml) and in group A (13.053 ± 8.670 pg/ml) (Mann-Whitney U test, p = 0.0149 and p = 0.0481, respectively). Serum levels of interleukin-8 and tumor necrosis factor-α were not significantly different between the two groups. Conclusion: These findings suggested that increased serum VEGF levels subdue cavity formation in active pulmonary tuberculosis.
Copyright © 2001 S. Karger AG, Basel
Tuberculosis is a reemerging disease, affecting patients in both developing and industrialized countries . Cavities in the lungs are most commonly seen in human tuberculosis and still represent a difficult problem in the treatment of tuberculosis. Liquefaction followed by cavity formation perpetuates tuberculosis in humans [2, 3], because coughing aerosolizes tubercle bacilli growing on the cavity wall, thereby enabling them to infect other people. From the cavity, the bacilli enter the bronchial tree and spread to other parts of the lung and also to other people. Levels of certain cytokines in bronchoalveolar lavage are correlated with radiological findings, including cavity formation, in active pulmonary tuberculosis [4, 5, 6], although the detailed mechanisms of cavity formation in pulmonary tuberculosis are not yet known.
Vascular endothelial growth factor (VEGF), known as a vascular permeability factor, is a 34- to 42-kD homodimeric protein. It functions through binding with its fms-like tyrosine kinase receptor (FLT-1), and is believed to function as a tumor angiogenesis factor with mitogenic effects on endothelial cells . There have been a few reports on VEGF expression in infectious diseases [8, 9], especially mycobacterial infection, while many studies have revealed correlations between VEGF and angiogenesis in various neoplasms [10, 11, 12, 13].
In this study, we examined the serum levels of VEGF in patients with pulmonary tuberculosis in order to investigate its relationship with cavity formation in active pulmonary tuberculosis.
patients and methods
To reduce the effects of selection bias, consecutive patients diagnosed as having pulmonary tuberculosis were included in this study. Forty-eight patients were studied (35 males and 13 females), with an age range of 16–89 years (mean 57.7) (table 1). All patients were hospitalized at the National Sanatorium Kanagawa Hospital, and the diagnosis of pulmonary tuberculosis was determined by sputum culture. All serum specimens were obtained from these 48 patients before administration of antitubercular drugs with their informed consent. Forty patients showed only pulmonary lesions, while other tuberculous lesions were noted in 8 patients (4 pleuritis, 2 miliary tuberculosis, 1 pyothorax and 1 colon tuberculosis). Various complications were noted in these patients as follows: diabetes mellitus (7), hypertension (2), ischemic heart disease (2), dementia (2), neoplasm (1), rheumatoid arthritis (1) and miscellaneous (5).
Table 1. Characteristics of the patients with active pulmonary tuberculosis
We examined cavity formation by chest X-ray and high-resolution computed tomography (HRCT) as reported previously . HRCT studies were performed using an Xvision/SR (TSX-002A) (Toshiba, Japan). We divided the 48 patients into two groups according to cavity formation. The parameters used for evaluation were a 1.0-mm section thickness at 10-mm intervals with a 512 × 512 reconstruction matrix, 120 kV and 150 mA, and a high spatial frequency algorithm. All images were obtained at window settings appropriate for lung parenchyma (level –400 to –800 Hounsfield units, width 1,800 to 1,600 Hounsfield units). HRCT scans were evaluated for the presence of cavities. The HRCT scans were analyzed by two independent chest radiologists and final conclusions on the findings were reached by consensus.
Sera were collected after centrifugation and stored at –80°C until analysis. Serum levels of VEGF were measured using a sandwich ELISA kit (IBL, Japan) according to the manufacturer’s instructions. Serum levels of interleukin (IL)-8 and tumor necrosis factor (TNF)-α were also measured using a sandwich ELISA kit (Diaclone, France) according to the manufacturer’s instructions. We also examined serum VEGF levels in 5 healthy adult volunteers.
The Mann-Whitney test was used throughout the study for comparisons between groups, with p values below 0.05 considered significant.
All 48 patients underwent chest X-ray and HRCT analysis in the standing and supine position, respectively. HRCT scans were evaluated for the presence, distribution and extent of the following signs: cavity formation, miliary nodules, nodules, consolidation and ground glass opacity. Twenty-two of the 48 patients showed lesions with cavities (group A) and 26 patients showed no apparent cavity structures in the tuberculoid lesions (group B) (fig. 1).
Fig. 1. Typical chest HRCT findings in the two groups of patients. a A case in group A with cavity formation. b A case in group B showing consolidation without cavity formation.
Serum levels of VEGF ranged from 0.02 to 511.2 pg/ml (mean 37.796 ± 12.095 pg/ml) as determined by enzyme immunoassay in the patients with active tuberculosis, while those of normal individuals were 0.005–24.81 pg/ml (mean 8.739 ± 3.656 pg/ml). The serum levels of VEGF showed marked variations in both group A (0.02–31.97 pg/ml, mean 13.053 ± 8.670 pg/ml) and group B (0.002–288.5 pg/ml, mean 58.733 ± 21.612 pg/ml). Group B patients showed significantly higher serum levels of VEGF than normal healthy individuals and group A patients (Mann-Whitney U test, p = 0.0149 and p = 0.0481, respectively) (fig. 2a).
Fig. 2. Serum levels of each factor were examined by enzyme immunoassay. a Serum levels of VEGF. Various serum levels of VEGF were observed in both group A (with cavities) (0.02–31.97 pg/ml, mean 13.053 ± 8.670 pg/ml) and group B (without cavities) (0.002–288.5 pg/ml, mean 58.733 ± 21.612 pg/ml). Serum levels of VEGF were significantly increased in group B compared to those in normal individuals and in group A (Mann-Whitney U test, p = 0.0149 and p = 0.0481, respectively). b Serum levels of TNF-α were 0–748.5 pg/ml (mean 70.985 ± 45.059 pg/ml) in group A and 0–96.28 pg/ml (mean 11.568 ± 4.739 pg/ml) in group B. The serum levels of TNF-α were not significantly different between the two groups.
IL-8 was detected in the serum in only 3 of the 48 patients (1 in group A and 2 in group B) (32–73 pg/ml). Serum levels of TNF-α were 0–748.5 pg/ml (mean 70.985 ± 45.059 pg/ml) in group A and 0–96.28 pg/ml (mean 11.568 ± 4.739) in group B (fig. 2b). The serum levels of IL-8 and TNF-α were not significantly different between the two groups.
In this study, we examined the serum levels of VEGF in 48 patients divided into two groups, i.e. those with and those without cavity formation as evaluated by HRCT. Group A (with cavities) was comprised of 22 patients and group B (without cavities) was comprised of 26 patients. Enzyme immunoassay showed significantly higher serum levels of VEGF in group B than in normal individuals (p = 0.0149) and group A (p = 0.0481). These findings suggested that serum VEGF levels are correlated with cavity formation in active pulmonary tuberculosis.
There have been a few previous reports concerning VEGF in infectious diseases [8, 9, 14]. Kraft et al.  reported that serum VEGF levels in patients with acute infections were elevated compared with those in healthy individuals. Our findings suggested that the increased blood and oxygen supply caused by the expression of VEGF inhibits cavity formation in active pulmonary tuberculosis, while another study indicated that Mycobacterium tuberculosis and its components might contribute to cavity formation by stimulating macrophages to release matrix metalloproteinases that digest collagen type I–IV .
In this study, serum levels of IL-8 and TNF-α were not significantly different between the two groups of patients, while some cytokines (IL-8 and TNF-α) have been reported to be related to immunity to active tuberculosis [16, 17, 18, 19]. Casarini et al.  reported that the levels of IL-6, IL-8, IL-12, TNF-α and interferon-=γ, not in serum but in bronchoalveolar lavage fluid, were correlated with radiological findings in active pulmonary tuberculosis, while other cytokines have also been reported to be correlated with activity in pulmonary tuberculosis [20, 21, 22].
Cavity formation has been shown to be rare in active pulmonary tuberculosis in patients with acquired immune deficiency syndrome, which is an immune disease of T lymphocytes [23, 24]. Ischemia is known to induce VEGF to mediate active neovascularization in patients with diabetic retinopathy , and TNF-α also affects VEGF expression through T lymphocytes in human tumors . In active pulmonary tuberculosis, VEGF expression may affect cavity formation through local immunity. In this study, 6 of the 26 patients in group B (without cavities) showed significantly increased levels of VEGF (p = 0.0481). We may need to evaluate a larger number of pulmonary tuberculosis patients to clarify this point in detail.
This work was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Education, Science and Culture: 09680835 (Y.O.), 09670239 (H.H.) and 09670205 (M.N.). We are grateful to two radiologists, Dr. Yuya Yamazaki and Dr. Kohzo Satoh, for HRCT analysis. We thank Mr. Daisuke Yamada, Mr. Kazuaki Nakano and Mr. Keisuke Takahashi for technical assistance.
Yoshiyuki Abe, MD
Department of Respiratory Disease
National Sanatorium Kanagawa Hospital
666-1, Ochiai, Hadano, Kanagawa 257-8585 (Japan)
Tel. +81 463 81 1771, Fax +81 463 82 7533, E-Mail firstname.lastname@example.org
Received: Received: July 3, 2000
Accepted after revision: March 7, 2001
Number of Print Pages : 5
Number of Figures : 2, Number of Tables : 1, Number of References : 26
Respiration (International Review of Thoracic Diseases)
Founded 1944 as ‘Schweizerische Zeitschrift für Tuberkulose und Pneumonologie’ by E. Bachmann, M. Gilbert, F. Häberlin, W. Löffler, P. Steiner and E. Uehlinger, continued 1962–1967 as ‘Medicina Thoracalis’
Vol. 68, No. 5, Year 2001 (Cover Date: September-October 2001)
Journal Editor: C.T. Bolliger, Cape Town
ISSN: 0025–7931 (print), 1423–0356 (Online)
For additional information: http://www.karger.com/journals/res