Is measurement of serum vascular endothelial growth factor reliable in patients with active pulmonary tuberculosis?

In a recent issue of the CHEST (June 2004), Alatas et al1 measured serum vascular endothelial growth factor (VEGF) concentrations in subjects with active pulmonary tuberculosis, and compared the changes in concentration before and after therapy. We would like to bring to your attention some methodological aspects of this study that might have caused relevant bias in the conclusion. It is well-known that VEGF resides in the -granules of platelets and is released during blood clotting.2 Therefore, serum VEGF level is not a reliable indicator of the circulating extracellular VEGF levels in vivo, and it principally reflects platelet count rather than VEGF synthesis by peripheral tissues. The authors reported that serum VEGF levels are increased in patients with active pulmonary tuberculosis when compared with those of subjects with inactive tuberculosis and those of healthy subjects. Thrombocytosis is common in subjects with active pulmonary tuberculosis, and it has been shown that, in these patients, platelets aggregate excessively in vitro.3–5 As a consequence, we think that the higher serum VEGF levels observed by the authors in the patients with active pulmonary tuberculosis were caused by increased platelet counts and increased platelet degranulation in vitro rather than by the fact that the “inflammatory mass stimulates the production of VEGF in order to supplant the needed blood supply by the formation of new blood vessels,” as suggested by the authors. Our hypothesis is supported by the authors’ observation that VEGF levels decrease after treatment when platelet count typically declines.5 Considering that a direct correlation between platelet count and serum VEGF level has been described,6 it is advisable to correct VEGF serum levels for variations in platelet count.7 Furthermore, it must be pointed out that the authors did not report the length of time between venipuncture and the separation of serum from blood cells. This interval should always be standardized and declared because serum VEGF concentrations increase in a time-dependent manner.8 In particular, allowing the whole-blood sample to clot for between 2 and 6 h before serum is collected may reduce the time-dependent, nonuniform release of VEGF.6 In plasma, platelet degranulation is minimized by adding anticoagulants to blood samples, and, as a consequence, plasma VEGF concentrations are up to 20 times lower than the matched serum VEGF concentrations.8 Citrate/theophylline/adenosine/ dipyridamole plasma should be used when circulating extracellular VEGF levels are measured.9 In conclusion, we think that the increase in serum VEGF levels observed by the authors in subjects with active pulmonary tuberculosis was caused by higher platelet counts and enhanced platelet degranulation in vitro, and does not reflect increased circulating extracellular VEGF levels in vivo. A meticulous standardization of blood sampling is mandatory when serum VEGF levels are measured. Serum VEGF concentrations should always be corrected for the variations in platelet count. Finally, it would be interesting to confirm the results reported by the authors with citrate/theophylline/adenosine/dipyridamole plasma samples.