Prognostic Value of Fibrinogen in Patients With Coronary Artery Disease and Prediabetes or Diabetes Following Percutaneous Coronary Intervention: Five-year Findings From a Large Cohort Study


 Background: Fibrinogen (FIB) is an independent risk factor for mortality and cardiovascular events in the general population. However, the relationship between FIB and long-term mortality among CAD patients undergoing PCI remains unclear, especially in individuals complicated with diabetes mellitus (DM) or prediabetes (Pre-DM). Methods: 6,140 patients with CAD undergoing PCI were included in the study and subsequently divided into three groups according to FIB levels (FIB-L, FIB-M, FIB-H). These patients were further grouped by glycemic status [normoglycemia (NG), Pre-DM, DM]. The primary endpoint was all-cause mortality. The secondary endpoint was cardiac mortality. Results: FIB was positively associated with hemoglobin A1c (HbA1c) and fasting blood glucose (FBG) in CAD patients with and without DM (P <0.001). During a median follow-up of 5.1 years, elevated FIB was significantly associated with long-term all-cause mortality (adjusted HR: 1.86; 95% CI: 1.28-2.69; P=0.001) and cardiac mortality (adjusted HR: 1.82; 95% CI: 1.15-2.89; P=0.011). Similarly, patients with DM, but not Pre-DM, had increased risk of all-cause and cardiac mortality compared with NG group (all P<0.05). When grouped by both FIB levels and glycemic status, diabetic patients with medium and high FIB levels had higher risk of mortality [(adjusted HR: 2.57; 95% CI: 1.12-5.89), (adjusted HR: 3.04; 95% CI: 1.35-6.82), all P<0.05]. Notably, prediabetic patients with high FIB also had higher mortality risk (adjusted HR: 2.27; 95% CI: 1.01-5.12). Conclusions: FIB was strongly associated with long-term all-cause and cardiac mortality among CAD patients undergoing PCI, especially in those with DM and Pre-DM. FIB test may help to identify high-risk individuals in this specific population.


Background
Despite advances in revascularization strategies over recent decades, the clinical outcomes remain unfavorable in patients with coronary artery disease (CAD), especially when complicated with diabetes mellitus (DM) [1]. Regarded as a pivotal component of coagulation as well as a biomarker of in ammation, brinogen (FIB) plays a crucial role in the pathophysiological process of thrombosis and atherosclerosis [2][3][4][5]. Previous evidences suggested that FIB was an independent risk factor of CAD development and cardiovascular events in the general population [6,7]. Similar results on the prognostic value of FIB were also observed in patients with CAD[8-10].
Glycemic metabolism abnormality, including DM and prediabetes (pre-DM), is increasingly prevalent on a global scale. By 2045, over 600 million individuals are projected to develop DM, with about the same number developing pre-DM [11]. The cardiovascular disease (CVD) risk, disability and mortality brought by glycemic metabolism abnormality is undisputedly a serious public health concern. Interestingly, FIB level was found to be higher in diabetic and prediabetic patients, and was involved in glycemic metabolism abnormality and insulin resistance [12,13]. Moreover, recent studies reported that FIB was positively related with the glycemic metabolism (hemoglobin A1c [HbA1c] and fasting blood glucose [FBG]) in patients with acute coronary syndrome (ACS) or stable CAD [9,10]. However, few data are available examining the correlation between FIB and glycemic metabolism in CAD patients undergoing percutaneous coronary intervention (PCI). Furthermore, the association of FIB with long-term outcomes in this population was far less investigated, particularly in those with impaired glycemic metabolism.
In light of the above, we aimed to evaluate the relationship between FIB and glycemic metabolism, and further determine the combined effect of FIB and impaired glycemic metabolism on long-term all-cause and cardiac mortality in CAD patients undergoing PCI.

Study population
This study was based on a prospective, observational, single-center cohort. From January 2013 to and Low-density lipoprotein cholesterol (n=153) values were excluded. A total of 6,140 patients were ultimately included in the analysis. The study protocol was approved by the Institutional Review Board of Fuwai Hospital and complied with the Declaration of Helsinki. All patients provided written informed consent before the intervention. Regular follow-up assessment of patients was performed at ve time points (1-month, 6-month, 12-month, 2-year, and 5-year after the discharge). Follow-up data were collected through medical records and telephone interview. The primary endpoint was all-cause mortality. The secondary outcome was cardiac mortality. Mortality that could not be attributed to a noncardiac etiology was considered cardiac mortality. All endpoints were adjudicated centrally by 2 independent cardiologists, and disagreement was resolved by consensus.

Procedure and medications
Before the procedure, patients receiving selective PCI were treated with aspirin (300mg) and ticagrelor (loading dose, 180mg) or clopidogrel (loading dose, 300mg), except for patients already on dual antiplatelet therapy; for patients with ACS receiving emergency PCI, the same dose of aspirin and ticagrelor or clopidogrel (loading dose, 300-600 mg) were administered as soon as possible. All patients were administered with unfractionated heparin (100 U/kg), and interventional cardiologist decided whether to use glycoprotein IIb/IIIa antagonist according to the clinical conditions and coronary lesions during the procedure. After the procedure, the dual antiplatelet therapy including aspirin (100mg daily), ticagrelor (90mg, twice daily) or clopidogrel (75mg, daily) were recommended for at least 1 year. The choice of equipment and techniques during PCI was at the discretion of the physicians.

De nition of clinical status
Glycemic categories were based on the guideline recommendations of American Diabetes Association [14]. Diabetes mellitus (DM) was de ned by HbA1c levels ≥6.5%, or fasting blood glucose (FBG) ≥7.0mmol/L, or 2-hour blood glucose levels of oral glucose tolerance test ≥11.1 mmol/L), or current use of hypoglycemic medications. Prediabetes (Pre-DM) was de ned as nondiabetic patients with FBG ranging from 5.6 to 6.9 mmol/L, HbA1c levels ranging from 5.7% to 6.4%. Patients without Pre-DM or DM were de ned as normoglycemia (NG). Hypertension was de ned as systolic blood pressure≥140 mmHg and/or diastolic blood pressure≥90 mmHg and/or current use of antihypertensive drugs. Lowdensity lipoprotein cholesterol ≥3.4mmol/L, fasting total cholesterol ≥5.2 mmol/L, triglyceride≥1.7 mmol/L, high-density lipoprotein cholesterol <1.0 mmol/L and/or chronic use of lipid-lowering drugs were considered criteria for dyslipidemia. Left main disease was de ned as stenosis of ≥50% in left main coronary artery, and three-vessel disease was de ned as stenosis of ≥50% in all three main coronary arteries (right coronary artery, left circum ex artery and left anterior descending artery) con rmed by coronary angiography.

Laboratory analysis
Fasting blood samples were drawn from all patients within 24 hours after admission. Enzymatic hexokinase method was used to measure the concentrations of blood glucose. Tosoh Automated Glycohemoglobin Analyzer (HLC-723G8) was used to measure the HbA1c levels. Stago autoanalyzer with the STA brinogen kit (Diagnostica Stago, Taverny, France) was used to measure the concentrations of FIB. All other laboratory measurements were conducted at the biochemistry center of Fu Wai Hospital by standard biochemical techniques.

Statistical analysis
Continuous variables were presented as mean±standard deviation, while categorical variables were presented as frequency and percentage. Differences of continuous and categorical variables were analyzed by analysis of variance or Kruskal-Wallis test and c2 test or Fisher's exact test, as appropriate.
Linear regression analysis was performed to evaluate the correlation between FIB and glycemic metabolism (HbA1c and FBG). In survival analysis, the association between FIB and clinical endpoint was initially examined using restricted cubic splines. The FIB was subsequently analyzed as both a continuous and a categorical variable. For categorical analysis, patients were grouped according to tertiles of the distribution [FIB-L(<2.98mg/dL), FIB-M(2.98-3.59mg/dL), FIB-H(<3.59mg/dL)]. Survival distributions were presented by Kaplan-Meier curves and compared by log-rank test. Cox regression analyses were performed to calculate the hazard ratios (HRs) and 95% con dence interval (CI).
Proportional hazards assumption was veri ed by Schoenfeld residuals. In multivariate Cox analyses, covariates including age, sex, body mass index (BMI), hypertension, family history of CAD, prior PCI/CABG, LVEF, LDL-C, creatine, DES implantation, clopidogrel, ACEI/ARB were adjusted because of their statistical signi cance in univariate analysis or clinical importance. The prognostic impact of glycemic metabolism status (NG, Pre-DM and DM) for all-cause mortality was also assessed by using the model mentioned above. Patients were further divided into 9 groups by both FIB levels and glycemic metabolism status to calculate HRs for all-cause mortality using FIB-L plus NG as reference. Statistical analyses were conducted with SPSS version 25.0 (IBM Corp., Armonk, N.Y., USA), R Programming Language version 4.0.0 (R Core Team, 2014), and GraphPad Prism version 7.0.0 for windows (GraphPad Software, San Diego, California USA). A two-tailed P value < 0.05 was considered statistically signi cant.

Baseline characteristics of patients with different FIB levels
Among the 6140 patients, the mean age was 58.4±10.4 years, and 4771(77.7%) were male. The baseline characteristics of patients according to the tertiles of FIB are summarized in Table 1. Patients with higher FIB levels were older and less likely to be male (all P< 0.05). In addition, they had higher proportion of diabetes, hypertension, prior stroke, ACS, and left main or three-vessel disease (all P< 0.05). Moreover, patients with elevated FIB levels had higher FBG, HbA1c, TC, LDL cholesterol, creatinine, lesion vessels, SYNTAX score, but lower LVEF and lower rate of complete revascularization (all P< 0.05). No signi cant differences were noted regarding dyslipidemia, family history of CAD, smoking, HDL cholesterol, number of stents and DES implantation among these groups (all P> 0.05).

Comparison of clinical data among groups with different glycemic metabolism status
In Table 2, Patients were divided into three subgroups based on different glycemic metabolism status. In general, the DM and pre-DM group had a less favorable cardiovascular risk pro le. Patients with DM or pre-DM tended to be older and female, with a larger burden of concomitant diseases, such as hypertension, dyslipidemia and prior stroke compared with those in NG group (all P< 0.05). Additionally, the prevalence of prior PCI/CABG and left main or three-vessel disease was higher in the DM and pre-DM group (all P< 0.05). Meanwhile, there were also higher BMI, FBG, HbA1c, TG, number of diseased vessels, SYNTAX score, number of stents, and lower LVEF, HDL cholesterol, complete revascularization, DES implantation in the DM group (all P< 0.05). Notably, FIB levels were signi cantly elevated from NG to DM group (P< 0.001).

Relationship between HbA1c/FBG and FIB
Linear regression analysis was used to assess the correlation between glycemic metabolism and FIB (Table 3). The results showed that both admission HbA1c (R 2 =0.018, Standard β=0.133, P<0.001) and FBG (R 2 =0.012, Standard β=0.111,P<0.001) were positively associated with FIB in the whole cohort. In The median follow-up time was 5.1 years (interquartile range 5.0-5.2 years), and the response rate was 91.2% ( Figure 1). During this period, 214 (3.5%) patients experienced all-cause mortality, of which 127 was cardiac mortality (accounted for 58.5%). The incidence of all-cause mortality in FIB-L, FIB-M and FIB-H group was 2.1%, 3.7% and 4.7%, respectively. Restricted cubic splines visualized a positive relation between FIB on a continuous scale with long-term risk of all-cause mortality and cardiac mortality (all P for non-linearity>0.05) (Additional le 1: Figure S1). The Kaplan-Meier survival curve revealed that patients with higher FIB levels had signi cantly increased risk of all-cause mortality and cardiac mortality (all logrank P<0.001) (Figure 3a, Additional le 1: Figure S2a) (Table 4).

Glycemic metabolism, FIB levels and occurrence of all-cause mortality
The prevalence of all-cause mortality in NG, Pre-DM and DM group was 2.4%, 3.3% and 4.6%, respectively.
The Kaplan-Meier curve demonstrated that patients with DM had signi cantly increased risk of all-cause mortality and cardiac mortality among the three groups (all log-rank P<0.05) (Figure 3b, Additional le 1: Figure S2b). Univariate Cox analysis revealed that DM group had 1.91-fold higher risk of all-cause mortality (HR: 1.91; 95% CI: 1.28-2.84; P=0.001) and 2.18-fold higher risk of cardiac mortality (HR: 2.18; 95% CI: 1.28-3.37; P=0.004) when compared with NG group. And this signi cant association remained unchanged after adjustment for other covariates. However, Pre-DM group did not increase the risk of allcause mortality and cardiac mortality compared with NG group (Figure 4, Additional le 1: Figure S3).
When patients were evaluated according to both glycemic metabolism and FIB levels, the Kaplan-Meier curve showed that those with DM and FIB-H levels had signi cantly highest risk of all-cause mortality compared with the reference group (NG plus FIB-L group, log rank P<0.001). Furthermore, NG plus FIB-H, Pre-DM plus FIB-M, Pre-DM plus FIB-H and DM plus FIB-M groups also had signi cantly increased risk of all-cause mortality than the reference group (NG plus FIB-L group, all log rank P<0.05) (Figure 3c)

Discussion
Using a large, real-world, prospective cohort sample, we found that FIB positively correlated with glycemic metabolism in CAD patients undergoing PCI. Moreover, higher FIB levels, analyzed as continuous or categorical variables, were strongly associated with increased risk of long-term all-cause and cardiac mortality. Furthermore, poorer long-term outcomes were also found in diabetic patients, but not in prediabetic patients. Interestingly, when patients were categorized into 9 groups according to both FIB levels and glycemic metabolism status, patients with pre-DM plus high FIB levels, DM plus medium FIB levels and DM plus high FIB levels had increased risk of all-cause mortality than those with NG and low FIB levels. For the rst time, our study demonstrated that FIB might affect the long-term prognosis in CAD patients with pre-DM undergoing PCI, and indicated a joint prognostic value of FIB levels and impaired glycemic metabolism on mortality in CAD patients undergoing PCI.
FIB is a crucial glycoprotein consisting of three different polypeptides, which is mainly synthesized in the liver [15]. Upon action of thrombin, FIB is transformed into brin monomer which then crosslinks platelets, increases blood viscosity and ultimately leads to clot formation [3]. Besides, FIB levels are elevated in response to various chronic in ammatory conditions, including DM, obesity and atherosclerosis [7,12,16]. It is also directly involved in the pathogenesis of atherosclerosis through multiple mechanisms, such as inducing endothelial dysfunction, stimulating smooth muscle cell proliferation and migration, facilitating monocyte or macrophage adhesion and in ltration of atherosclerotic lesions, which will jointly potentiate plaque evolution [17].
To date, studies have been conducted on the prognostic value of FIB in different clinical settings. Aside from the positive association with all-cause and CVD mortality in general individuals [6,18,19], FIB was reported to be an independent risk factor of the occurrence and severity of CAD [20]. Further, both small sample and large epidemiological studies showed that FIB was associated with worse clinical outcomes in patients with stable CAD [10,21,22]. A recent prospective study from China indicated elevated FIB was also strongly associated with MACE risk in ACS patients, especially when complicated with DM [9]. Similarly, the present study found FIB had an independent association with long-term all-cause and cardiac mortality in CAD patients undergoing PCI. Instead, the ADVANCE study showed FIB was not an independent predictor of macrovascular events and mortality in diabetic patients with documented CVD or risk factors [23]. The PRIME study and the EPIC-Norfolk study found FIB was not a long-term predictor of all-cause or cardiovascular mortality in subjects free of CAD [24,25]. The controversial results may result from the heterogeneity of the study population. In spite of the con icting ndings mentioned above, data from the latest clinical trials con rmed the bene t of anti-in ammatory effect both in patients with chronic coronary disease and acute MI [26,27]. Considering the role of FIB as an in ammatory biomarker, additional studies are warranted to further evaluate whether FIB could be helpful to identify high-risk individuals in CAD patients.
Currently, glycemic metabolism abnormality including DM and pre-DM is prevalent in clinical practice, especially in patients with established CAD [11]. It has been previously demonstrated that DM independently increased the risk of adverse CVD events in CAD patients [28]. Notably, CAD patients with pre-DM seemed to share similar clinical outcomes with those with normoglycemia [10,29]. However, when combined with other disorders, such as dyslipidemia or hypertension, prediabetic patients with CAD were demonstrated to have signi cantly less favorable prognosis[30-32]. Interestingly, a large-sample observational study recently reported that elevated FIB increased the MACE risk in patients with stable CAD only in the presence of DM and pre-DM, indicating FIB to be valuable for prognostic assessment in prediabetic patients with stable CAD [10]. However, the combined value of FIB and impaired glycemic metabolism on prediction of mortality in CAD patients undergoing PCI is still unclear. In this study, we observed that among CAD patients undergoing PCI, diabetic individuals with high or medium FIB levels had 2.57-fold and 3.04-fold higher risk of mortality respectively during a median follow-up of 5.1 years. Furthermore, prediabetic patients also had higher mortality risk in the subgroup of high FIB levels, indicating that FIB may be useful for further risk strati cation in CAD patients with mild impaired glycemic metabolism after PCI.
Patients with DM were con rmed to have higher levels of plasma FIB [12]. Chronic mild in ammation is a recognized pathological mechanism of DM[33]. Elevated FIB existing in diabetic patients aggravates the in ammatory process and the burden of atherosclerosis [4,34]. FIB also involves in insulin resistance and impair the normal glycemic regulation [13]. Moreover, elevated FIB could weaken platelet inhibition with clopidogrel in the presence of DM [35]. And this effect is mediated through its direct interaction with the GP IIb/IIIa receptor, which is independent from in ammation. Indeed, our study found that the average levels of FIB elevated from NG, pre-DM to DM. Moreover, FIB was also positively associated with glycemic metabolism (HbA1c and FBG) both in CAD patients with or without DM, which was basically consistent with the prior studies [9,10]. Collectively, although without established causality, the present study revealed a robust association between FIB and glycemic metabolism, as well as the long-term mortality in CAD patients undergoing PCI. Given the relatively simple and cost-effective test of FIB, these ndings encourage its potential value as a biomarker in this speci c population to identify high-risk patients, especially in those with DM and pre-DM. Meanwhile, the importance of routine screening for impaired glycemic metabolism also cannot be neglected. In addition, whether this speci c population could bene t from lowering FIB levels through lifestyle modi cation or pharmacologic intervention warrants future studies. This study has some limitations. First, FIB test was only performed at baseline. Data on FIB level uctuation during follow-up was not available. Second, due to the observational design, potential confounders cannot be fully controlled. Third, a direct causal link between FIB and mortality risk was not established by this study.

Conclusions
FIB was strongly associated with long-term all-cause and cardiac mortality among CAD patients undergoing PCI, especially in those with DM and Pre-DM. FIB test may help to identify high-risk individuals in this speci c population. The Ethical Review Board of Fuwai Hospital approved the study protocol in accordance with the Declaration of Helsinki, and written informed consent was obtained from all participants.

Consent for publication
The manuscript was approved by all authors for publication.

Availability of data and materials
Due to ethical restrictions related to the consent given by subjects at the time of study commencement, our datasets are available from the corresponding author upon reasonable request after permission of the Institutional Review Board of Fuwai Hospital.

Figure 2
Linear regression analysis of the relationship between glycemic metabolism and FIB. a Linear regression analysis of the relationship between glycemic metabolism (HbA1c and FBG) and FIB in whole patients. b Linear regression analysis of the relationship between glycemic metabolism (HbA1c and FBG) and FIB in patients with DM. c Linear regression analysis of the relationship between glycemic metabolism (HbA1c