We used IVs of the four platelet indices and BP in the current study to explore their association. In the univariable MR, we found that the level of PLT and PCT predicted by host genetics was causally associated with an increase in BP (SBP, DBP). PDW was associated with DBP, and the magnitude of its association with SBP was similar but nonsignificant. After the collinearity correction in MVMR, the effect of PLT and PDW remained robust, while the associations of PLT and PCT with BP got nonsignificant. These findings suggest that PLT and PDW are the critical traits underlying the positive associations of platelet activation with elevated BP. In the reverse MR, we observed the positive effect of SBP on PLT and PCT. It provides evidence of a causal association between the genetically determined BP level and platelet indices, although the magnitudes of the associations were small enough that they could be neglected.
Several studies have reported that platelet activation is related to cardiovascular morbidity and mortality [23, 32]. Platelet indices are blood-based parameters related to platelet morphology and proliferation kinetics, and they can be used to reflect the activation of platelets. Increased platelet indices were observed in multiple diseases, such as ischemic heart disease and ischemic strokes [19, 24] [33, 34]. Our findings of the univariable MR study align with earlier studies on PLT concerning the elevation of BP [22, 14]. To the best of our knowledge, no MR study evaluated the effect of the other platelet indices on BP, whereas the association of hypertension with them has been reported in observational studies [14–16]. Some studies also showed that MPV is associated with the incidence of hypertension. Still, it is worth noting that the conclusions from observational studies are prone to be biased by confounding effects and reverse causation. Our study found no association between MPV and BP, except for PCT.
Studies on the comparative effects of platelet indices on BP are limited. Given the co-association among platelet indices, it is necessary to explore which trait or traits predominated the influence of platelet on BP. We then performed the MVMR to disentangle the comparative effects of platelet indices. The present MR study extended the evidence to show that PLT alone with PDW exhibited independent effects. PLT, as we all know, measures platelet counts per unit volume of blood, PCT is the volume occupied by platelets in blood expressed as a percentage, and MPV is a marker of the average size of platelets. So PCT is analogous to the total platelet volume in some way, and is correlated to the product of the MPV and PLT. The results of MVMR excluding PCT after the co-linear test provide evidence that heterogeneity of platelet size, i.e. PDW, is a marker of more significant importance among PCT and MPV. So high PLT and PDW might be an indication to recognize the part of platelet activation in interpreting the process of hypertension.
In vitro studies, the increased shear force that platelets are exposed to due to the high BP could lead to platelet activation, and the more giant platelets are more aggregate and reactive compared to the smaller size[35, 36]. In the reverse direction, the present study found that SBP influences PLT and PCT. However, a previous MR study based on a smaller sample size revealed that hypertension has no effect on PLT [22]. The discrepancy might be caused by inadequate power due to limited phenotypic variance explained by used genetic variants and/or small sample size for outcomes. Nevertheless, whether BP had effects on platelet activation needs more study.
The benefit obtained from a reduction in blood pressure is undisputed[37, 38]. So the findings of our study are relevant in both clinical and public health terms. We revealed that platelet activation is one of the causes of elevated BP, mirrored by platelet indices. They are potentially valuable markers for evaluating hypertension progression and the early diagnosis of thromboembolic disease. In four platelet indices, PLT and PDW plays an independent and dominant role which should be emphasized in clinical practice. Reversely, BP also affected platelet, as previous studies have shown that platelet activation plays a significant role in embolic disease; the use of antiplatelet management therapies in the high BP group should be considered [12, 39]. The latest review showed no evidence that antiplatelet therapy modifies mortality in patients with elevated BP for primary prevention [13]. The reason for this result is that bleeding events outweighed the potential benefits. However, it does not mean that antiplatelet therapy is meaningless in people with hypertension. On the contrary, small doses of antiplatelet therapy may benefit a specific population if BP is controlled smoothly and the risk of bleeding and embolism is accurately assessed. All in all, further randomized controlled trials (RCTs) of antithrombotic therapy with complete documentation of all benefits and harms are required in patients with elevated BP.
A consistent limitation of all the observational studies discussed is that platelet indices are influenced by various conditions, which can confound analysis. The strengths of our study are, therefore, that it overcomes this potential source of bias through the use of genetic variants as instruments for platelet indices and explores the possibility of reverse causation [28, 29, 40]. Furthermore, the multivariable MR method was the major strength, which compared the roles of different correlated platelet traits in BP elevation. We also took a rigorous methodological approach, incorporating a range of sensitivity analyses to explore potential bias due to genetic pleiotropy.
However, it also has limitations. First, the GWAS of traits associated with platelets was performed in the UK Biobank population, and data on blood pressure data were available from meta-analyses published by the International Consortium of Blood Pressure (ICBP) and UK Biobank studies. Therefore, there is the potential for overlap in samples and bias related to this[31]. Second, we added a more significant number of SNPs as IVs with no horizontal pleiotropy to increase the R2. As a result, the improvement of SNPs in the instrument came increasingly due to heterogeneity. Third, We only included four platelet indices in the study; there may be other hematological markers related to platelet; this means that the dominant role of the PLT and PDW could be adjusted when including the other traits, such as platelet larger cell ratio (P-LCR).
MR assumes a linear relationship between the exposure and outcome [41], in our case, genetically determined platelet indices and blood pressure level. For this reason, the results of our MR analysis should not be extrapolated to extremes of platelet indices and blood pressure. Of particular note is the possibility that platelet may be a causal risk factor for the progression rather than the onset of hypertension; there could be distinct causes for the initiation and progression of a disease, which means that the causal exposures for disease onset may not necessarily be causal for disease progression (and vice versa) [42].