Effects of statin therapy on mean platelet volume in patients with risk of cardiovascular diseases: a systematic review and meta-analysis

Many studies have demonstrated the effects of statin therapy on platelet, but it is controversial that whether statin could reduce mean platelet volume (MPV) in patients with the risk of cardiovascular diseases. To further improve the clinical significance of MPV in those patients and explore new function of statin, we conducted this research. Relevant studies were selected by searching electronic databases (PubMed, Embase and Cochrane Library) and reference lists of related articles by hand. Two reviewers independently assessed eligibility and quality of the studies. Eventually, we included ten studies, a total of 1189 patients with the risk of cardiovascular diseases. Consolidating relevant data and comparing the changes of MPV before and after statin treatment, we found that statin could decrease MPV [standard mean difference (SMD) = −0.47 (−0.71–0.23)], which was statistically significant (P=0.0001). Subgroup analysis suggested that when ≥55 years, this decrease did not occur [SMD = −0.06 (−0.18, 0.06)]. Drug type, sample size, ethnicity, mean age and quality of included article were sources of heterogeneity. Therefore, statin therapy could reduce MPV significantly and exhibited antiplatelet activity, which is of great importance in clarifying the clinical significance of MPV in cardiovascular events and the prevention of cardiovascular events.


Introduction
Mean platelet volume (MPV), reflecting the size of the platelets, is a potentially useful marker of platelet activity. Increased MPV level has been identified as an independent risk factor for cardiovascular diseases and vascular risk factors such as coronary heart disease [1], diabetes [2], smoking [3], hypertension [4], dyslipidemia [5] obesity [6] and atrial fibrillation [7].
Statin inhibits 3-hydroxy-3-methylglutaryl co-enzyme A reductase in the mevalonate pathway, simultaneously promotes the low density lipoprotein metabolism and the high density lipoprotein synthesis, which play an essential role in reducing the risk of cardiovascular events. In recent years, many studies have reported possible effects of statin on MPV, but the conclusions have not been uniform and are highly controversial. Some studies have linked statin to lower MPV [8][9][10][11][12][13][14][15], others have found no link [16,17]. So do statin affect MPV? Could MPV be an independent risk factor for patients taking statin with cardiovascular risk factors? Based on this controversial topic and these questions, we conducted a systematic review and meta-analysis to study the relationship between statin therapy and MPV, with a view to providing a reference for clinical practice.

Literature search
This systematic review and meta-analysis are reported in accordance with the Preferred Items for Systematic Reviews and Meta-analysis (PRISMA) Statement. Literature was retrieved by formal search of electronic databases (PubMed, Embase and Cochrane Library) without date limitation. To achieve the maximum sensitivity of the search strategy, we used appropriated free text and thesaurus terms including "Hydroxymethylglutaryl CoA Reductase Inhibitors", "statin", "mean platelet volume". We also searched reference lists of related articles by hand to obtain more studies. All studies were limited to English language and the retrieval strategy of Pubmed as follow: (

Inclusion and exclusion criteria
Inclusion criteria: (1) Patients with the risk of cardiovascular diseases, such as diabetes mellitus, dyslipidemia and hypertension; (2) The data must contain standard mean difference (SMD). If only median and interquartile range (IQR) provided, standard difference (SD) will be calculated according to the Cochrane manual equation: SD = IQR/1.359 [18]; (3) The article must provide baseline data. (4) Abstracts providing necessary data will be included in order to avoid bias.

Data abstraction and quality assessment
Two authors (S.F.J. and B.B.Z.) independently extracted the original data. Disagreement was resolved by discussion. If the two authors could not reach a consensus, the result was reviewed by the third author (X.D.W.). The extracted data were consisted of the follow items: the first author's name, publication year, population (Ethnicity), methods, matching criteria, total number of cases and age (years).
The quality assessment of the included trials was undertaken independently by two review authors (H.S. and L.X.Y.), following Newcastle-Ottawa Scale (NOS), which is composed of three parts: selection, comparability and exposure. It is a semi-quantitative scale, and a score of 0-9 stars was assigned to each study. A total score of <7 was considered poor and 7-9 was deemed high quality.

Statistical analysis
We measured the statin effect on continuous outcomes (e.g. change of mean platelet volume) by SMD with 95% confidence interval (CI). We used Review manager 5.3 and Stata14.0 software to perform the meta-analysis in the present study. Sensitivity analysis was performed by changing effect model and remerging data after excluding abstracts. Begg's test and egger's test were used to detect the asymmetry of the funnel plot, which (P<0.05) were considered to be representative of statistically significant publication bias. Heterogeneity among studies was assessed by I 2 statistic. I 2 > 50% indicated evidence of heterogeneity. If heterogeneity existed among the studies, the random effects model was used to estimate the pooled effect size. Otherwise, the fixed effects model was adopted. Subgroup analyses regarding drug type, follow-up, sample size, ethnicity, mean age and NOS score, were also performed to explore source of heterogeneity.

Literature search
After initial retrieval, 167 articles were obtained. 15 articles were removed after repeated examination. 152 articles were excluded after reading the title and abstract, which included four case reports, six animal researches, ten reviews and letters, and 122 unrelated studies such as relationship between statin and cancers, researches about platelet function. The remaining ten articles included eight case control studies [8,[10][11][12][14][15][16][17] and two abstracts [9,13]. Two of these studies [10,11] reported the effects of two statins on MPV, so we considered a total of 12 studies to be included in the data analysis. The flowchart of literature inclusion was shown in Figure 1.

Meta-analysis of statin therapy and MPV
The pooled analysis is shown in Figure 2. The results showed that the mean platelet volume significantly down-regulated after statin treatment [SMD = −0.47 (−0.71-0.23)], which was statistically significant (P=0.0001). There was significant heterogeneity among studies (P<0.00001, I 2 = 87%), so the random-effect model was used.

Subgroup analysis
Subgroup analysis based on drug type, follow-up, sample size, ethnicity, mean age and NOS score, was performed and the results are shown in Table 2. In terms of specific drug administration, the mean platelet volume could be

Assessment of publication bias and linear correlation of MPV changes by intensity statin
A funnel plot (Figure 3), Begg's test and Egger's test were performed to investigate the potential publication bias. The results showed no significant publication bias existed (Begg's P=0.451>0.05, Egger's P=0.271>0.05). Linear correlation of MPV changes by intensity statin is shown in Figure 4, and the result suggested that there was a positive correlation.

Discussion
Our study shows that statin do reduce mean platelet volume in patients with cardiovascular disease risk factors, which indirectly suggests that statin could inhibit platelet function. To our knowledge, this is the first study to demonstrate the effect of statin on platelets through systematic review and meta-analysis, providing a powerful supplement for the diversity of clinical effects of statin.
In clinical practice, cardiovascular disease is a kind of common and multiple diseases, with a long course and rapid progress. Platelets play an important role in the development of cardiovascular diseases, especially atherosclerosis. Cardiovascular disease is mainly caused by thrombosis. In the vascular lumen, platelet aggregation and adhesion lead to thrombosis, which causes narrowing and obstruction of the lumen, myocardial ischemia, hypoxia and eventually necrosis. Obesity, diabeteshypertension and dyslipidemia are risk factors for cardiovascular diseases. Many studies have shown that platelets are highly reactive in these risk factors, which in turn can accelerate the development of cardiovascular diseases [19][20][21][22]. Platelet activation is often accompanied by an increase in volume, i.e., increased MPV. Coban et al. [23] observed a mean MPV significantly higher in the group of obese women, in comparison with the non-obese group (8.18 + − 1.09 vs 8.01 + − 0.95 fL, P=0.004). In addition, Coban et al. [24] in another case-control study with 200 subjects and Ozkan et al. [25] in a case-control study with 108 children aged 6-16 years reached similar conclusions.
Patients with Type 2 DM have a higher risk of coagulation abnormalities and thromboembolic events [26,27], and elevated MPV and its significance as marker in Type 2 DM have been elaborated in many studies [28][29][30][31]. The same is true for those with Arterial hypertension or dyslipidemia [32,33]. Therefore, MPV is an important marker for patients with cardiovascular disease risk factors and represents abnormal platelet function. Improving platelet function and reducing MPV are important in preventing cardiovascular events Statin can significantly reduce blood cholesterol levels and the risk of cardiovascular disease and ischemic stroke [34,35]. In addition to inhibiting lipid synthesis, statin also has antioxidant [36], anti-inflammatory [37] and antithrombotic effects [38]. Our study showed that statin can reduce MPV and regulate platelet function, which may be mainly through multiple mechanisms to combat platelet activation. CD36 (glycoprotein IV) is expressed in a variety of cells on the surface of the single glycoprotein, and ox-LDL binds to platelet membrane CD36 to generate endogenous ROS, which can promote platelet activation and thrombosis [39]. Simvastatin and pravastatin can inhibit platelet activation by directly interacting with CD36 or by affecting CD36 intracellular signal transduction pathways [38]. Peroxisome proliferator-activated receptor (PPAR) is a member of the nuclear steroid hormone receptor superfamily, and its biological functions of transcriptional regulation are mainly performed by binding specific DNA reaction elements and retinol X heterodimer receptors [40]. Simvastatin is dose-dependent on the induction of PPARα and PPARγ activation, and the up-regulation of PPARγ expression also inhibited collagen-induced plaque aggregation. Meanwhile, elevated expression of CD62, bispecific phosphatase 2 and Ca2 + mobilization induced by simvastatin could further increase antiplatelet activity [41]. Lee et al. [42] reported effect of cyclicAMP-eNOS/NO-cyclicAMP pathway in antiplatelet activity induced by statin. Activation of cyclicAMP-eNOS/NO-cyclicAMP pathway could cause phospholipase Cγ2-Protein kinase C-mitogen activated protein kinase-TXA2 cascade reactions inhibition, which suggests antiplatelet activity. Ni et al. [43] reported atorvastatin can increase the level of guanosine cyclophosphate in platelets and delay the maximum activation rate of p-selectin and CD41, which inhibited activation of platelet induced by H-Gly-Tyr-Pro-GlyLys-Phe-OH (GYP) and thrombin. Serebruany et al. [44] demonstrated that multiple statin can reduce the expression of thrombin receptor protein kinase-activated receptor-1, and reduce platelet activation and thrombosis in the evaluation of the primary prevention effect of statin therapy in patients with metabolic syndrome. Therefore, it is not hard to understand why statin could reduce MPV.
In the ten studies, we included the results of Gungoren et al. [17] and Kucera et al. [16] indicated that statin therapy may not affect MPV. After comprehensive analysis, we found that the biggest difference between these two articles and other articles is the original data type of MPV, both of two using median and IQR. For this reason, we removed these two articles and recombined the data, and the result [SMD = −0.58 (−0.82, −0.34), P=0.000] verified our idea to a certain degree.
Our study indicates that statin therapy could reduce MPV regardless of drug type, follow-up time, sample size, ethnicity, and <55 years subgroup, and we also find that drug type, ethnicity, age, and literature quality are sources of heterogeneity. Sensitivity analysis and publication bias test show that our results are stable and reliable. The age group ≥55 years shows no effect [SMD = −0.06 (−0.18, 0.06)], and we fail to explain it.
Of course, we need to point out the limitations of our research. First of all, the time interval of blood sample collection and analysis, and the use of different anticoagulants (such as citrate and EDTA) may be the influencing factors, but due to insufficient data, it cannot be verified by subgroup analysis; Second, the follow-up time of the included studies was relatively long, which could not guarantee that the patients did not have behaviors that interfered with the study results; Then, we could not perform subgroup analysis about pravastatin and simvastatin; Finally, there is no way to know if there is a drug dose effect because the doses overlap and cannot be analyzed further.
Fortunately, our study demonstrated that statin therapy may reduce MPV, indirectly demonstrating platelet inhibition. Given the importance of MPV as a marker of activated platelets and as a predictor of vascular events, this effect of statin therapy is encouraging. What is more, our study suggests when we use MPV to diagnose and monitor patients with cardiovascular disease, we need to pay attention to patients' statin use.