Increased Platelet Reactivity Due to Platelet Receptor Polymporphisms? Not in the Real World
To the Editor:
We read with interest the article by Pontiggia et al,1 discussing the possible functional implications of double homozygosity for two “prothrombothic” platelet glycoprotein receptor polymorphisms (PlA1/A2 of GpIIIa and C807T of GpIa) in a family with a strong history of premature cardiovascular events. We would like to report our experience in 90 stable coronary artery disease (CAD) patients, studied in the “real world” while taking their usual medications, in whom we tried to recognize environmental and genetic determinants of high-shear platelet aggregation, measured by collagen-ADP PFA-100 closure time (mimicking the high-shear conditions of diseased arteries).
Using a multivariate analysis approach, we recorded clinical variables such as age, sex, smoking, diabetes, previous myocardial infarction, and drug therapy, including use of thyenopiridines (aspirin use was not considered as it is known not to influence collagenADP PFA-100 values). We also measured laboratory variables (platelet and white blood cell count, mean platelet volume, cholesterol, triglycerides, von Willebrand factor activity as Ristocetin Cofactor Activity [vWf RCA]) and determined GpIIIa PlA1/A2 and GpIa C807T genotypes.
On univariate analysis, increased vWf RCA (r=−0.372, P=0.0001) and platelet count (r=−0.207, P=0.04), reduced triglyceride levels (r=0.318, P=0.002), male sex (89±18 vs 105±23 s, P=0.003) and nonusage of thyenopiridines (84±16 vs 96±20 s, P=0.02) were associated with increased platelet reactivity (ie, reduced PFA-100 closure times). In contrast, the genetic polymorphisms were not. No double homozygotes for the variant genotypes were observed in our population, and double heterozygotes did not differ from the general population (Table).
On stepwise multivariate linear regression, all the nongenetic predictors maintained their significant relation with closure time (with vWf RCA remaining the strongest predictor), with an R2 (total explained variance) of 0.52, indicating a good predictive power. When the genetic predictors were forced into the model, no significant increase in the explained variance occurred.
Thus, in a general population of patients with stable CAD taking their usual medications, the presence of putative “prothrombotic” GpIa and GpIIIa polymorphisms is unlikely to exert important effects on platelet reactivity. We believe the results by Pontiggia et al1 are very convincing. However, the authors do not claim universal applicability of their findings. Moreover, some old-fashioned predictors of platelet reactivity, such as platelet count, were not considered in their analysis, because these authors correctly used platelet-rich plasma (at a fixed concentration) for their functional tests and for PFA-100 closure times.
- ↵Pontiggia L, Lassila R, Pederiva S, Schmid HR, Burger M, Beer JH. increased platelet-collagen interaction associated with double homozygosity for receptor polymorphisms of platelet GPIa and GPIIIa. Arterioscler Thromb Vasc Biol. 2002; 22: 2093–2098.
We appreciate the contribution by Dr Porto et al, and we agree that “basic“parameters, such as the platelet count and the level of von Willebrand factor RCA, are of importance and need to be considered, particularly in the global tests, such as the PFA-100, as has been shown before.1,2⇓ The influence of the triglyceride levels is an interesting observation. The fact that thienopyridines (but not aspirin alone) prolong the closure times in the collagen-ADP-assays confirms previous studies. In combination with clopidogrel, however, aspirin seems to prolong the PFA-100-closure times significantly3 and it may not really be justified not to consider it. The finding by Porto et al that single homozygous or double heterozygous patients for GPIIIa PlA2 and GPIa 807T patients did not have statistically shorter PFAs than the GPIIIa PlA1 or the GPIa 807Cs appears not surprising given the relatively small numbers and the huge variability (SDs of up to 28.5 s even in samples sizes of n=10), which reflects the multitude of genetic and acquired (and “procedural”) risk factors. Unfortunately, a double homozygous case was not found.
In some contrast to Porto et al, di Paola et al4 found that the collagen-receptor density of GPIa (which is regulated by the GPIa 807 C/T polymorphism) correlated well with closure times. Thus, other groups that used the same method did find indeed an measurable influence. In our study,5 we did attempt to find a model case of double homozygosity with a clinical endpoint (family history), but without recognizable acquired risk factors or medications affecting the results. This case was then analyzed by a series of independent assays. As stated in our report, we cannot exclude other until present unknown prothrombotic factors affecting this family.
Overall, we believe that the findings of the two groups are not mutually exclusive. Whether the “real world“is better reflected by negative global test results affected by a multitude of acquired and genetic factors or by a selected model case without recognizable risk factors analyzed by several independent methods seems a rather philosophical question.
- ↵Grau AJ, Reiners S, Lichy C, Buggle F, Ruf A. Platelet function under aspirin, clopidogrel, and both after ischemic stroke: a case-crossover study. Stroke. 2003; 34: 849–854.
- ↵Di Paola J, Federici AB, Mannucci PM, Canciani MT, Kritzik M, Kunicki TJ, Nugent D. Low platelet alpha2beta1 levels in type I von Willebrand disease correlate with impaired platelet function in a high shear stress system. Blood. 1999; 93: 3578–3582.
- ↵Pontiggia L, Lassila R, Pederiva S, Schmid HR, Burger M, Beer JH. Increased platelet-collagen interaction associated with double homozygosity for receptor polymorphisms of platelet GPIa and GPIIIa. Arterioscler Thromb Vasc Biol. 2002; 22: 2093–2098.