Significance of Soluble P-Selectin, Von Willebrand Factor, and Other Adhesion Molecules in Hypercholesterolemia and Peripheral Artery Disease
To The Editor:
We recently read with interest the paper by Davı̀ and colleagues,1 who reported raised levels of both soluble P-selectin and von Willebrand factor in the plasma of 20 patients with hypercholesterolemia compared with 20 normocholesterolemic controls, with a strong (P=0.0001) correlation between the molecules. We are delighted that they have confirmed our previous and recent observations of raised levels of von Willebrand factor among patients with this risk factor.2 However, we are unable to confirm their finding of raised soluble P-selectin in hypercholesterolemia in a cross-sectional study of similar population and design.2 The reasons for this difference are unclear but are unlikely to be simply related to the use of different reagents than those of Davı̀ and colleagues.1
Davı̀ and colleagues1 hypothesize that circulating levels of P-selectin may represent, in this clinical setting, an in vitro marker of endothelial cell and/or platelet activation. We beg to differ. It has been hypothesized that soluble P-selectin is a marker of platelet activation and does not appear to represent endothelial cell activation.3 4 In support of this concept, it can be noted that the administration of DDAVP (desmopressin, a substance that stimulates the endothelium) to healthy subjects results in increased levels of von Willebrand factor but no change in levels of soluble P-selectin in the plasma.5 Thrombocytosis is often present in cardiovascular disease, so that a platelet count should be included because it correlates with levels of soluble P-selectin.4 Furthermore, several groups have been unable to correlate levels of von Willebrand factor (probably the “gold standard” in vivo marker of endothelial cell activation and damage) with those of soluble P-selectin in various studies that have included patients with diabetes, thrombocytopenia purpura, hypertension, and atherosclerosis.
Therefore, in order for the study of these molecules to provide something of substance, we feel that a consensus needs to be achieved. Until several large, hopefully prospective, comparative studies are published, it seems likely that the routine measurement of soluble adhesion molecules has little to offer practicing physicians toward attaining their goal of improving patient care. Additional studies may tell us if the measurement of soluble adhesion molecules will have more concrete answers to offer rather than simply asking more questions and offering inconsistencies.
- Copyright © 1999 by American Heart Association
Davı̀ G, Romano M, Mezzetti A, Procopio A, Iacobelli S, Antidormi T, Bucciarelli T, Alessandrini P, Cuccurullo F, Bon GB. Increased levels of soluble P-selectin in hypercholesterolemic patients. Circulation. 1998:97;953–957.
Blann AD, Goode GK, Miller JP, McCollum CN. Soluble P-selectin in hypercholesterolemia with and without symptomatic vascular disease: relationship with von Willebrand factor. Blood Coagul Fibrinolysis. 1997:8;200–204.
Blann AD, Lip GYH. Is soluble P-selectin a new marker of platelet activation? Atherosclerosis. 1997:128;135–138.
Jilma B, Eichler HG, Vondrovec B, Breiteneder H, Kyrle PA, Kitzweger E, Kapiotis S, Speiser W. Effects of desmopressin on circulating P-selectin. Br J Haematol. 1996:93;432–436.
We thank Dr Blann and colleagues for their interesting comments on our article.
Unfortunately, we don’t have a clear explanation for the apparent discrepancy between some of our results and data from his group, but we agree that it should not be related to the reagent source.
Dr Blann and colleagues propose that soluble P-selectin should rather be considered a marker of platelet activation. It is also our opinion that plasma P-selectin may reflect platelet activation, and recently we have described a significant correlation between levels of soluble P-selectin and urinary 11-dehydro-TXB2, an established marker of in vivo platelet activation, in hypercholesterolemia (M. Romano, MD, et al, unpublished data). However, we believe that more convincing evidence should be obtained to completely rule out the contribution of a dysfunctional endothelium to circulating P-selectin levels.
We think that definitive conclusions cannot be drawn from studies with DDAVP administration to healthy subjects because they may not be directly applicable to patients with vascular damage. It is known that when endothelial cells are exposed to various stimuli, von Willebrand factor (vWF) and P-selectin are mobilized from Weibel-Palade bodies to different compartments: vWF is mainly released, whereas P-selectin remains bound to the cell surface.R1 DDAVP also induces P-selectin translocation to the endothelial cell plasma membrane.R2 Thus, it can be predicted that DDAVP administration may determine an increase in plasma vWF without modifying soluble P-selectin.
However, an altered endothelium may behave differently. In particular, the endothelium overlying atherosclerotic plaques overexpresses P-selectinR3 and is likely to be subjected to interactions with inflammatory cells and exposure to inflammatory mediators, eg, cytokines and proteolytic enzymes, that might induce P-selectin release. In pathological conditions, then, P-selectin and vWF may undergo different mechanisms of release, and this might explain why it is not always possible to observe a clear correlation between vWF and soluble P-selectin levels. This would imply that vWF and P-selectin may monitor different aspects of the functional status of the endothelium.
Therefore, although our current knowledge may suggest that platelets represent the principal source of soluble P-selectin under physiological conditions, it cannot be excluded that an altered endothelium might also contribute to increases in P-selectin levels in selected pathological states. The extent and relevance of this contribution and therefore the specificity of P-selectin as an endothelial index remain to be established, and we agree with Dr Blann and colleagues that additional studies are needed to provide consistent criteria for a correct evaluation of circulating adhesion molecule measurements.