Expression and predictive value of leucine rich α-2 glycoprotein in ischemic heart disease


 Background : The expression and predictive value of leucine rich α-2 glycoprotein (LRG) in ischemic heart disease, including coronary artery disease (CAD), myocardial infarction (MI) and chronic total occlusion (CTO), remain unclear.Methods : A total of 231 patients were included and divided into 4 groups: control group (N=39), CAD group (N=73), MI group (N=67) and CTO group (N=52). LRG was determined and compared within groups. MI group was then divided into 2 subgroups according to LRG expression. The association between LRG expression and clinical characteristics, peak Troponin T (TNT), left ventricular ejection fraction (LVEF) and in-hospital stay was investigated.Results : LRG expression in MI group was significantly higher than that in control group (37.9±12.8 vs 25.3±8.9, P<0.001). LRG expression in CAD and CTO group were similar with that in control group (CAD vs Con, 23.1±6.8 vs 25.3±8.9, P=0.233; CTO vs Con, 22.0±7.8 vs 25.3±8.9, P=0.082). In MI group, patients with increased LRG had higher peak TNT (4.80 (2.34-10.0) vs 2.46 (0.89-4.98), P=0.002), lower LVEF (48.7±10.2 vs 54.8±8.4, P=0.009) and longer in-hospital stay (10.4±5.4 vs 8.0±2.9, P=0.027). Linear regression showed that LRG was independently associated with in-hospital stay. ROC curve showed the area of LRG under the curve (AUC) was 0.859.Conclusions : LRG expression was elevated in MI patients compared with controls, angiography confirmed CAD and CTO patients. In the MI subgroup, patients with elevated LRG had higher peak TNT, lower LVEF and longer in-hospital stay. Whether LRG is associated with major adverse cardiovascular events and long term prognosis needs further investigation.


Background
Cardiovascular disease is the leading cause of death and disability worldwide. Coronary artery disease (CAD) with severe stenosis, myocardial infarction (MI) and chronic total occlusion (CTO) are different stages and represent different pathophysiological processes. MI leads to intense inflammation, which plays a role in recovery and remodeling [1]. Although advances in early reperfusion therapies reduced mortality, reperfusion of the myocardial tissue might be hindered by dysfunction of the microcirculation [2]. Chronic total coronary occlusion generally results to good collateral circulation, which improve cellular function and global myocardial performance [3]. leucine rich α-2 glycoprotein (LRG) was initially purified from healthy individuals and subsequent studies have shown that LRG expression increases in response to mediators of acute phase response [4]. It was found to be elevated in patients with autoimmune diseases such as RA, Crohn's disease and ulcerative colitis [5,6]. Then LRG was demonstrated to promote angiogenesis by modulating endothelial TGF-b signaling [7]. LRG levels are also increased in patients with heart failure [8]. In our previous study, we found LRG was elevated in MI patients compared with healthy controls [9]. However, the prognostic value of LRG in MI patients and its expression in CAD and CTO patients has not been investigated. In this study, we detected serum LRG expression in controls, CAD patients, MI patients and CTO patients. We found that LRG was elevated in MI patients compared with controls, CAD and CTO patients. The differences of expression of LRG in controls, CAD and CTO patients were not significant. In MI subgroup, patients with elevated LRG had higher peak TNT, lower EF and longer in-hospital stay.

Study design and patient selection
A total of 231 patients were included in this prospective study. Of these patients, 112 received angiography due to suspected coronary artery disease. After angiography, they were divided into 2 groups: control group (39patients, 3 main vessels stenosis <50%) and CAD group (73 patients, at least 1 main vessel stenosis ≥50%). The MI group included 67 patients (48 STEMI and 19 NSTEMI) who were diagnosed with myocardial infarction according to the fourth universal definition of myocardial infarction [10]. The CTO group included 52 patients who had been confirmed as CTO by previous angiography and were hospitalized for interventional therapy.

Laboratory assays and Clinical end points
A venous blood sample was collected for all patients upon admission (within 24 hours) for routine laboratory assays performed by central laboratory at our hospital. Troponin T (TNT) of MI patients was determined by Roche assays upon admission and then every morning in CCU until they began to reduce from peak value. Another 2 ml venous blood sample was collected and centrifuged at 1,500 g for 10 min to precipitate blood cells. The serum was then frozen at -80℃ until required for LRG analysis. LRG was determined by Elisa kit purchased from Raybiotech (GA, USA). Echocardiography for MI patients was performed within 24 hours of admission and left ventricular ejection fraction (LVEF) was determined.

Statistical analysis
Normally distributed data are expressed as mean ± SD, and were compared using the Student's t-test (2 groups) or 1-way ANOVA (n groups). LSD method was used for comparison among specific 2 groups. Skewed variables are expressed as median and inter quartile range and Mann-Whitney U test was used. Categorical data are expressed as number (percentage) and were compared using the chi-square test. Receiver operating characteristic (ROC) curve and a under the curve (AUC) were used to evaluate predictive value for MI occurrence. Linear regression was used to evaluate association between LRG and other factors. All statistical analyses were performed using SPSS 17.0 (SSPS Inc., Chicago, IL, USA). A value of P<0.05 was considered as statistically significant.

Results
The baseline characteristics of the 4 groups are shown in Table 1. Patients in CAD and MI group were older than control and CTO group. The percentages of male, diabetes mellitus (DM), hypertension and percutaneous coronary intervention (PCI) history were significantly lower in control group. LRG expression in MI group was significantly higher than that in control group (37.9±12.8 vs 25.3±8.9, P<0.001). LRG expression in CAD and CTO group were similar with that in control group (CAD vs Con, 23.1±6.8 vs 25.3±8.9, P=0.233; CTO vs Con, 22.0±7.8 vs 25.3±8.9, P=0.082) (shown in Figure 1). We divided the MI patients into 2 groups according to LRG expression and their characteristics are shown in Table 2. There were no differences of age, male, DM, hypertension and PCI history between the 2 groups. Patients with increased LRG had higher peak TNT Linear regression showed that LRG was independently associated with in-hospital stay (Table 3). LRG was not independently associated with other factors, including gender, age, DM, Hypertension, PCI history LVEF and peak TNT. We plotted ROC curve to evaluate the ability of LRG to predict MI. As shown in Figure 2, the area of LRG under the curve (AUC) was 0.859. The best cut off value was 28.3ug/mL with the biggest sum of sensitivity and specificity (0.826 and 0.795, respectively).

Discussion
In this study, LRG expression was demonstrated to be elevated in MI patients compared with controls, angiography confirmed CAD and CTO patients. There was no significant difference of LRG expression in controls, angiography confirmed CAD and CTO patients. In the MI subgroup, patients with elevated LRG had higher peak TNT, lower LVEF and longer in-hospital stay. Linear regression analysis revealed an independent association between LRG and in-hospital stay. LRG belongs to the leucine-rich repeat (LRR) family and is expressed during granulocyte differentiation [10]. LRG plays important roles in protein-protein interaction, signaling, and cell adhesion [11]. Then it was found to increase in response to mediators of acute phase response [4] and promote angiogenesis [7]. Thus LRG level was elevated in inflammatory diseases [12] and some cancers [13][14][15]. Recent study also found elevation of LRG is associated with arterial stiffness, endothelial dysfunction, peripheral vascular disease and CAD [16][17][18]. Myocardial infarction leads to comprehensive pathophysiological process, including intense inflammation and angiogenesis [19]. In our previous study [9], we performed a proteomics research in early-onset myocardial infarction patients using iTRAQ coupled with LC-MS/MS. Among the 538 proteins identified and quantified, LRG was up-regulated in the early-onset MI group compared with healthy controls. During the validation stage, LRG was confirmed to be elevated in MI group. We also found that LRG was positively correlated with C-reactive protein. In this current study, we further compared LRG expression in angiography confirmed CAD and MI patients with non-CAD controls. The result of LRG expression in MI patients compared with control and CAD patients was similar with the previous study. The elevated LRG expression might promote angiogenesis at border zone. A previous study showed that angiogenesis and cardiac function were impaired in LRG-deficient mice. Prognosis was improved by transplantation of bone marrow cells, which can produce LRG protein [20]. However, the predictive value of LRG in MI patients has not been investigated. We divided the MI patients into 2 groups according to LRG expression and found higher LRG group had increased peak TNT, decreased EF and longer in-hospital stay. Further analysis revealed an independent association between LRG level and in-hospital stay. Future studies investigating predictive value for clinical events are warranted. It sounds inconsistent that LRG promote angiogenesis while, at the same time, correlates with worse cardiac function and longer in-hospital stay. We think it might be just like BNP, which is associated with heart failure and able to act as a therapeutic agent. Whether administration of LRG can improve prognosis needs to be studied. CTO is defined as an atherosclerotic complete vessel occlusion with Thrombolysis in Myocardial Infarction (TIMI) grade 0 flow within the occluded segment for more than 3 months. Its prevalence ranges from 30-50% [21] and around 50% CTO patients had MI history [22]. There is no evidence of association between CTO and inflammation. The most distinct character of CTO is coronary collateralization (including arteriogenesis and angiogenesis), which can improve prognosis [23]. The coronary collateralization of CTO patients are influenced by some factors, such as cytokines, miRNAs and diabetes [24]. In this study, we for the first time investigated LRG expression in CTO patients and found no difference between CTO patients and control/CAD patients.

Study limitations
The present study has several limitations. The sample size was relatively small and we did not investigate the major adverse cardiovascular events (MACE), such as cardiac death, non-fatal acute myocardial infarction, revascularization and stroke. The MACE rate was quite low during in-hospital follow up. The study observation time was limited to in-hospital stay; therefore, analysis of long-term prognosis is warranted.

Conclusions
In this study, LRG expression was demonstrated to be elevated in MI patients compared with controls, angiography confirmed CAD and CTO patients. There was no significant difference of LRG expression in controls, angiography confirmed CAD and CTO patients. In the MI subgroup, patients with elevated LRG had higher peak TNT, lower LVEF and longer in-hospital stay. Linear regression analysis revealed an independent association between LRG and in-hospital stay. Whether LRG is associated with major adverse cardiovascular events and long term prognosis needs further investigation.

Fig. 2
ROC curve of LRG to predict MI. The area of LRG under the curve (AUC) was 0.859. The best cut off value was 28.3ug/mL with the biggest sum of sensitivity and specificity (0.826 and 0.795, respectively).