We read with interest the paper by Simler et al investigating angiogenic cytokines in patients with idiopathic interstitial pneumonia.¹ We were surprised by their reported high levels of vascular endothelial growth factor (VEGF) in the plasma in the normal control group. Several other groups—including the manufacturers of the ELISA (R&D Systems)—have previously quoted normal plasma VEGF levels in the range 36–76 pg/ml.^2,3 Indeed, one of the authors of the paper previously quoted normal VEGF levels as 76 pg/ml using a matched pair ELISA.⁴ It is clear therefore that the levels of 648 pg/ml quoted for normal controls are nearly 10 times higher than those reported previously.

One possible explanation is the low centrifugal force used for preparation of the plasma (300g for 12 minutes). The manufacturer of the ELISA recommends 1000g for 15 minutes to reduce platelet contamination of the plasma. Platelet secretion of VEGF is the reason for increased serum levels of VEGF compared with plasma and might explain the extraordinarily high levels of VEGF found in these normal subjects.⁴ Interestingly, 14 of the 49 patients (28.5%) were on immunosuppressant drugs which potentially reduce the platelet count. This may be an alternative explanation as to why there was no difference between normal patients and those with pulmonary fibrosis, in contrast to earlier reports in patients with connective tissue disease related pulmonary fibrosis.⁵

Although the plasma levels of VEGF correlated with fibrosis based on the CT score, it is difficult fully to appreciate the relevance of this finding without knowing the concentration of VEGF within the lung compartment because, in normal individuals, epithelial lining fluid levels of VEGF at 9–11 ng/ml are several orders of magnitude greater than that found in the circulation.^6,7 Furthermore, previous investigators have reported reduced levels of alveolar VEGF in patients with idiopathic pulmonary fibrosis.⁸ Low levels of VEGF are also seen in the bronchoalveolar lavage (BAL) fluid of patients with acute lung injury, sarcoidosis, emphysema, and lung transplants. It would therefore appear that a reduced alveolar level of VEGF is a common feature of diseases associated with alveolar epithelial damage. Indeed, in ARDS, alveolar levels of VEGF are lowest in those with the worst lung injury.⁹ This is probably a result of reduced epithelial cell secretion of VEGF and increased expression of its soluble receptor, sVEGFR-1, which acts as a natural inhibitor to the bioactivity of VEGF.

The trophic role of VEGF within the lung is supported by the fact that VEGF acts as a proliferative factor for fetal pulmonary epithelial cells¹⁰ and lung targeted VEGF inactivation leads to an emphysema phenotype in mice.¹¹ These studies suggest that reduced alveolar levels of VEGF may inhibit epithelial repair in a wide variety of lung diseases.

In summary, we have some concerns about the validity/reproducibility of the VEGF levels reported in the study by Simler et al. Furthermore, based on the available evidence, we believe it is inappropriate to suggest that antagonising VEGF would be a successful potential treatment for patients with pulmonary fibrosis. On the contrary, we believe this would hasten epithelial cell apoptosis and promote alveolar septal cell loss with resultant honeycombing and functional deterioration.