The present study investigated whether there was a causal association between circulating GDF-15 levels and nine CVDs including AIS, CES, LAS, SVS, AF, HF, NICM, CAD and MI by using two-sample MR based on 5 GWASs and 2.6 million cases. The result suggested that circulating GDF-15 level had a significant causal association with the incidence of CES, AF, CAD and MI, whereas no significant relationship could be concluded when coming to AIS, LAS, SVS, HF and NICM. In addition, a higher circulating GDF-15 was associated with the higher risk of CES and AF and lower risk of CAD and MI. Therefore, our study indicated a crucial role of circulating GDF-15 levels for the prevention and intervention of CVDs.
GDF-15 and CES
CES has tripled in the past few decades and would persistently triple by 2050 worldwide. It’s the predominantly increasing proportion in ischemic strokes and eventually causes more severe strokes than other ischemic stroke subtypes [26]. The present result showed that patients with higher circulating levels of GDF-15 had more risk of CES. In ENGAGE AF-TIMI 48 trial, GDF-15 levels elevating from the baseline of 1661 pg/mL to 12 months of 1711 pg/mL were independently associated with a 2-fold higher rate of stroke or systemic embolic events in AF patients [27]. Likely, after 1.9 years of follow-up in ARISTOTLE trial, an annual rate of stroke or systemic embolic events was 2.03% in AF patients with the highest quartile of GDF-15 (>2052 ng/L) compared to 0.90% in those with the lowest quartile of GDF-15 (≤977 ng/L) [28]. As AF is one of the most common risk factors for CES, some studies showed that GDF-15 could promote thrombosis in patient with AF by inhibiting anticoagulation. AF patients with high serum GDF-15 levels (range from 1661.0 to 5163.0 pg/mL) had 19.2% longer clot lysis time (CLT) (range from 94.3 to 117.5 min), indicating that elevation of serum GDF-15 levels drives anti-fibrinolytic reactions unrelated to increased plasminogen activator inhibitor-1 (PAI-1) or antiplasmin [29]. Taken together, circulating GDF-15 levels might exert a predictive value of CES in AF patients attributed to its prothrombotic effects.
GDF-15 and AF
AF is the most common arrhythmia, currently affecting over 33 million individuals worldwide. The prevalence of AF is expected to more than double in adults over the next 40 years. AF is associated with a 2-fold increase in premature mortality and major adverse cardiovascular events such as HF, stroke and MI. Our result suggested that the circulating GDF-15 level was positive associated with the risk of AF. In community-based Individuals[30], postoperative patients[31] and hypertrophic cardiomyopathy (HCM) patients[32], those who had the higher GDF-15 levels were more vulnerable to AF than the lower. In addition, circulating GDF-15 level was reckoned independently associated with paroxysmal AF after multivariable analyses. In the previous study, the significantly higher serum levels of GDF-15 were found in patients with paroxysmal AF than those of controls (1473.14 ± 628.52 vs. 1233.592 ± 262.76 pg/ml, p<0.05) [33]. Furthermore, serum GDF-15 level was associated with incident AF with the hazards ratios of 1.31 after adjusted for age and gender [30]. However, some inconsistent studies showed that serum GDF-15 level was not influenced by the presence of AF in patient with HF from the BIOSTAT-CHF trial [9] and patients with chronic kidney disease (CKD) from the CRIC study [34]. Potential pathologic mechanisms of AF include atrial remodeling, reentry, pulmonary vein trigger, abnormal autonomic nerve modulation and inflammation. Otherwise, atrial structural remodeling is closely related to atrial fibrosis and extracellular matrix. Some evidences showed that GDF-15 could enhance activation of M2 macrophages and induce fibroblasts to participate in the progression of cardiac fibrosis [35]. In addition, GDF-15 might be involved in the process of inflammatory and oxidative stress, which also could promote the occurrence of AF. Taken together, circulating GDF-15 level might be a potentially biomarker and therapeutic target for AF.
GDF-15 and CAD
CAD is the most prevalent and important cause of ischemic heart disease, with major implications on global morbidity and mortality. The present result suggested that higher circulating level of GDF-15 was related to the lower risks of CAD and MI. GDF-15 may have a protective effect on atherosclerosis process. Consistently, overexpression of GDF-15 in macrophages significantly attenuated atherosclerotic lesions in the apoE knockout mouse model of atherosclerosis [36]. In addition, GDF-15 showed anti-apoptotic effect against ischemia reperfusion (I/R) and reduced the MI area mediated by activating intracellular phosphoinositide 3-kinase (PI3K) - protein kinase B (Akt)-dependent signaling [37]. Furthermore, GDF-15 could reduce platelet aggregation by inhibiting platelet integrin activation. GDF-15 knockout mice displayed an accelerated systemic thrombus formation and a reduced survival rate, while exogenous recombinant GDF-15 administration alleviated thrombus formation [38]. And the increased GDF-15 level may exert beneficial roles during process of CVD and MI as the FRISC-II trial suggested [39]. Taken together, GDF-15 might play multiple protective roles in the development and progression of CVD and MI.
GDF-15 and HF
Though the present MR study did not support a causality of the circulating GDF-15 with the risk of HF, there are some reports that GDF-15 may be related to disease severity and prognosis HF patients [40]. In Val-HeFT trial, serum GDF-15 levels at baseline were abnormally high (>1200 ng/L) in 85% of HF patients. Baseline serum GDF-15 levels (per 100 ng/L) were associated with the risks of mortality (hazard ratio 1.017; p<0.001) and first morbid event (hazard ratio 1.020; p<0.001). Increases in GDF-15 over 12 months were independently associated with the risks of future mortality and first morbid event. Taken together, serum GDF-15 concentration had not only a promising value of diagnosis but also a superior prognostic biomarker. However, patient with HF usually have some comorbidities such as hypertension, AF, stroke, diabetes, and chronic renal dysfunction, which may influence the circulating GDF-15 level and need to be eliminated.
Strengths and limitations
This MR study firstly described the causal association between GDF-15 and 9 CVDs. Our work provided with a new perspective to clarify the role of GDF-15 in the development of CVDs. However, limitations should be considered while interpreting the results. The MR method have some common defects [41]. Firstly, the SNPs we selected could not satisfy the demand of independence principle. In addition, the MR is not sensitive to confounders from environmental exposures and might violate exclusion restriction unless we took into consideration all influence factors of GDF-15. Since our summary statistic was based on European populations, limiting the generalizability of our work, other high qualified databases should be enrolled to further strengthen the results.