- Email: [email protected]
Relation of Fibrinogen Level With Cardiovascular Events in Patients With Coronary Artery Disease
Relation of Fibrinogen Level With Cardiovascular Events in Patients With Coronary Artery Disease Gjin Ndrepepa, MDa,*, Siegmund Braun, MDa, Lamin King, MDa, Massimiliano Fusaro, MDa, Dritan Keta, MDc, Salvatore Cassese, MDa, Tomohisa Tada, MDa, Albert Schömig, MDa,b, and Adnan Kastrati, MDa Evidence on the usefulness of fibrinogen for the risk stratification of patients with coronary artery disease remains inconclusive. The aims of this study were to investigate the association of fibrinogen with cardiovascular events and to assess whether this biomarker provides additional prognostic information on top of that provided by traditional cardiovascular risk factors. This study included 13,195 patients with angiography-proved coronary artery disease and fibrinogen measurements available. Receiver-operating characteristic curve analysis showed that the best fibrinogen cutoff for mortality prediction was 402.0 mg/dl. On the basis of this cutoff, patients were divided into 2 groups: the group with fibrinogen >402.0 mg/dl (n [ 5,198) and the group with fibrinogen £402.0 mg/dl (n [ 7,997). The primary outcome was 1-year mortality. All-cause deaths occurred in 393 patients with fibrinogen >402.0 mg/dl and in 246 patients with fibrinogen £402.0 mg/dl (Kaplan-Meier estimates of mortality 7.7% and 3.1%, log-rank test p <0.001). The relation between fibrinogen and mortality followed a J-shaped pattern, with lowest mortality in patients with fibrinogen concentrations of 295 to 369 mg/dl. After adjustment for cardiovascular risk factors and relevant clinical variables, fibrinogen remained an independent correlate of all-cause mortality (adjusted hazard ratio 1.07, 95% confidence interval 1.04 to 1.10, p <0.001, for each 50 mg/dl increase in fibrinogen level), but it did not improve the discriminatory power of the model for mortality prediction (integrated discrimination improvement 0.002, p [ 0.32). In conclusion, in patients with coronary artery disease, fibrinogen is an independent correlate of mortality, but it does not provide additional prognostic information on top of that provided by traditional cardiovascular risk factors. Ó 2013 Elsevier Inc. All rights reserved. (Am J Cardiol 2013;111:804e810) Numerous previous studies, recently summarized in several meta-analyses, have demonstrated that plasma fibrinogen shows a moderately strong association with the risk for coronary artery disease (CAD), stroke, and vascular and nonvascular mortality.1,2 Many previous studies, however, have included subjects or patients who are not representative of contemporary Western populations. More recent studies have shown either that the association between fibrinogen and cardiovascular events lost significance when adjusted for cardiovascular risk factors3 or that fibrinogen did not provide additional information to that provided by traditional risk factors.4 Genetic studies and meta-analyses of studies involving b-fibrinogen genotypes suggest that genotypes that produce lifelong differences in fibrinogen concentration are not a major determinant of CAD risk.5 For these reasons, the National Academy of Clinical Biochemistry stated in its recent laboratory a Deutsches Herzzentrum, b1. Medizinische Klinik Rechts der Isar, Technische Universität, Munich, Germany; and cKrankenhaus LandshutAchdorf Medizinische Klinik I, Landshut, Germany. Manuscript received October 19, 2012; revised manuscript received and accepted November 27, 2012. See page 810 for disclosure information. *Corresponding author: Tel: þ49-89-12181535; fax: þ49-8912184053. E-mail address: [email protected] (G. Ndrepepa).
0002-9149/13/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2012.11.060
medicine practice guidelines that evidence for the usefulness of fibrinogen in the risk stratification of patients at risk for cardiovascular events is inconclusive.6 Moreover, other aspects of the association between fibrinogen and cardiovascular events, such as the pattern of fibrinogen-mortality relation or the strength of the association in various subgroups of patients with CAD, remain poorly investigated. We undertook this study with a double objective: (1) to investigate the association between fibrinogen and cardiovascular events in patients with CAD and whether this information is additive to that provided by traditional cardiovascular risk factors and (2) to assess the strength of the association between fibrinogen and mortality in various subgroups of patients. Methods This study included 13,195 patients with angiographyproved CAD who underwent percutaneous coronary intervention (PCI) in the German Heart Center in Munich from March 2000 to December 2009. To be included in the study, patients had to have significant CAD, confirmed by coronary angiography at the time of index hospitalization. Patients who underwent PCI in the setting of failed thrombolysis or routine PCI after thrombolysis, as well as those with no fibrinogen measurements available, acute infections, serum creatinine 2 mg/dl, and known malignant diseases www.ajconline.org
Coronary Artery Disease/Fibrinogen and Cardiovascular Events
805
Table 1 Baseline characteristics Characteristic
Age (yrs) Women Diabetes mellitus Receiving insulin therapy Body mass index (kg/m2) Arterial hypertension Current smoker Hypercholesterolemia (220 mg/dl) Previous myocardial infarction Previous coronary artery bypass surgery Clinical presentation Stable CAD* Acute coronary syndromes Serum creatinine (mg/dl) Glomerular filtration rate (ml/min) Fibrinogen (mg/dl) Number of narrowed coronary arteries 1 2 3 Multivessel coronary disease Left ventricular ejection fraction (%)
Fibrinogen (mg/dl)
p Value
402.0 (n ¼ 7,997)
>402.0 (n ¼ 5,198)
66.2 1,511 1,930 530 26.8 5,490 1,167 5,663 2,580 1,209
(58.4e73.2) (19%) (24%) (7%) (24.6e29.4) (69%) (15%) (71%) (32%) (15%)
69.7 1,574 1,756 650 27.1 3,580 911 3,577 1,601 774
(61.8e76.4) (30%) (34%) (12.5%) (24.7e30.0) (69%) (18%) (69%) (31%) (15%)
5,219 2,778 1.0 83.3 324.0
(65%) (35%) (0.8e1.1) (64.9e104.0) (279.0e358.0)
2,897 2,301 1.0 73.2 483.0
(56%) (44%) (0.8e1.2) (53.5e96.6) (436.0e552.0)
1,519 2,168 4,310 6,478 58.0
(19%) (27%) (54%) (81%) (49.0e63.0)
784 1,331 3,083 4,414 55.0
(15%) (26%) (59%) (85%) (44.0e61.0)
<0.001 <0.001 <0.001 <0.001 <0.001 0.79 <0.001 0.014 0.08 0.72 <0.001 <0.001 <0.001 <0.001 <0.001
<0.001 <0.001
Data are expressed as number (percentage) or as median (interquartile range). Data on body mass index, glomerular filtration rate, and the left ventricular ejection fraction were available for 13,160 patients (99.7%), 13,161 patients (99.7%), and 11,599 patients (87.9%), respectively. * Defined as chest pain that did not change its pattern within the past 2 months with angiographic confirmation of significant CAD.
with life expectancy <1 year, were excluded. All patients gave written informed consent for angiographic examination, performance of PCI, and blood sampling. The study was conducted according to the principles of the Declaration of Helsinki and approved by the institutional ethics committee. Stable CAD was diagnosed if patients presented with chest pain that did not change its pattern within the past 2 months and had angiographic confirmation of CAD. Acute coronary syndromes were diagnosed using pattern-specific criteria.7 In all cases, the diagnosis was confirmed by angiographic criteria, which included the presence of coronary stenoses 50% luminal obstruction in 1 of the 3 major coronary arteries and/or documentation of culprit lesions in case of acute coronary syndromes. The left ventricular ejection fraction was calculated using the area-length method using left ventricular angiography.8 Main cardiovascular risk factors were defined using standard criteria. Hypercholesterolemia was defined as a documented total cholesterol value 220 mg/dl or previous or ongoing treatment with lipid-lowering agents. Arterial hypertension was defined when a patient was under active treatment with antihypertensive drugs or if the systolic blood pressure was 140 mm Hg or the diastolic blood pressure was 90 mm Hg on 2 separate occasions. Current smokers were those with regular smoking in the previous 6 months. The diagnosis of diabetes required active treatment with insulin or an oral hypoglycemic agent, abnormal fasting blood glucose (126 mg/dl), blood glucose >200 mg/dl at any time, or abnormal glucose tolerance according to World Health Organization criteria.
Weight and height were measured during the index hospitalization, and body mass index was calculated. The glomerular filtration rate was estimated using the CockcroftGault equation.9 PCI and periprocedural care were performed according to standard criteria. Antithrombotic therapy included clopidogrel (300 or 600 mg as a loading dose followed by 75 mg/day for 1 month) and aspirin (200 mg/day administered orally and continued indefinitely). Blood samples were obtained before angiography. Venous blood was collected using the S-Monovettes blood collection system (Sarstedt, Sarstedt, Germany) containing 0.106 mol/L sodium citrate. Nine volumes of blood were mixed with 1 volume of sodium citrate solution. Plasma was separated after immediate centrifugation at 1,500g for 10 minutes. Fibrinogen was determined in citrated plasma by a modification of the Clauss method using the BCS analyzer (Multifibren U; Siemens Healthcare, Erlangen, Germany). The measurement range lies between 80 and 1,200 mg/dl. Expected values range from 180 to 350 mg/dl. Heparin concentration in plasma <2 U/ml does not affect the test. Creatinine was measured using a kinetic colorimetric assay on the basis of the compensated Jaffe method. Laboratory personnel were unaware of clinical, angiographic, or followup information. The primary outcome was all-cause mortality at 1 year after PCI. Secondary outcomes included 1-year occurrences of cardiac mortality, nonfatal myocardial infarction, and stroke. Information on mortality was obtained from hospital records, death certificates, or phone contact with relatives of the patients or referring physicians. Cardiac death was
806
The American Journal of Cardiology (www.ajconline.org) Table 2 One-year mortality according to fibrinogen deciles Decile of Fibrinogen
Fibrinogen Level (mg/dl)
Number of Patients
1st 2nd 3rd 4th 5th 6th 7th 8th 9th 10th
<262 262 to <295 295 to <323 323 to <346 346 to <369 369 to <400 400 to <435 435 to <482 482 to <551 551 to 1,190
1,278 1,341 1,324 1,330 1,288 1,351 1,307 1,324 1,332 1,320
Mortality All-Cause 49 40 31 34 35 53 54 66 96 181
(3.83) (2.98) (2.34) (2.56) (2.72) (2.92) (4.13) (4.98) (7.21) (13.71)
Cardiac 35 28 25 27 27 34 37 50 64 124
(2.74) (2.09) (1.89) (2.03) (2.10) (2.52) (2.83) (3.78) (4.80) (9.39)
Figure 1. Fibrinogen levels in nonsurvivors and survivors. Data are medians with 25th to 75th and 5th to 95th percentiles.
Figure 3. One-year mortality in various fibrinogen deciles.
Figure 2. Kaplan-Meier curves of all-cause (A) and cardiac (B) mortality. Fib ¼ fibrinogen.
defined according to Academic Research Consortium criteria.10 The diagnosis of myocardial infarction required the development of new abnormal Q waves in 2 contiguous precordial or 2 adjacent limb leads or an elevation of creatine kinase-MB >2 times (>3 times for the 48 hours after a PCI procedure) the upper limit of normal. Definite stent thrombosis was defined according to the Academic Research
Consortium criteria. Stroke required the occurrence of acute neurologic deficits that were confirmed by computed tomography or magnetic resonance imaging of the head. The follow-up protocol after discharge included a phone interview at 1 month, a visit at 6 months, and a phone interview at 12 months. Follow-up information and adjudication of adverse events was performed by medical staff members unaware of clinical diagnosis or fibrinogen level. Data are presented as median (interquartile range [IQR]), number of patients or events, or proportions. The Kolmogorov-Smirnov test was used to assess the normality of data distribution. Continuous data were compared using KruskalWallis rank-sum test. Categorical data were compared using chi-square test. Receiver-operating characteristic curve analysis was performed to determine the best fibrinogen cutoff value for the prediction of all-cause mortality while maximizing sensitivity and specificity. Survival analysis was performed using the Kaplan-Meier method, and differences in survival were compared using the log-rank test. The association between fibrinogen and mortality (allcause or cardiac) was tested using the multivariate Cox proportional-hazards model. All variables listed in Table 1 were entered into the model. The discriminatory power of the multivariate model regarding prediction of mortality was assessed by calculating the integrated discrimination improvement (IDI) according to Pencina et al.11 Differences in the association between fibrinogen and survival in various
Coronary Artery Disease/Fibrinogen and Cardiovascular Events
807
Figure 4. All-cause deaths in subgroups of patients. Deaths are reported as numbers with Kaplan-Meier estimates obtained by survival analysis. ACS ¼ acute coronary syndromes; BMI ¼ body mass index; GFR ¼ glomerular filtration rate; LVEF ¼ left ventricular ejection fraction; Pint ¼ p value for interaction.
subgroups of patients (obtained by dichotomization of patients according to age [cutoff 65 years], gender, diabetes, arterial hypertension, smoking status, hypercholesterolemia, clinical presentation [stable CAD vs acute coronary syndromes], glomerular filtration rate [cutoff 60 ml/min], body mass index [cutoff 30 kg/m2], and the left ventricular ejection fraction [cutoff 50%]) were investigated by performing interaction testing. Analyses were performed using S-Plus (Insightful Corporation, Seattle, Washington). Twotailed p values <0.05 were considered to indicate statistical significance. Results The receiver-operating characteristic curve analysis showed that fibrinogen predicted 1-year all-cause mortality with an area under the curve of 0.652 (95% confidence interval [CI] 0.628 to 0.677). The best cut-off value of fibrinogen for the prediction of 1-year all-cause mortality, while maximizing sensitivity and specificity, was 402.0 mg/dl. Using this value, patients were divided into 2 groups: those with fibrinogen >402.0 mg/dl (n ¼ 5,198) and those with fibrinogen 402.0 mg/dl (n ¼ 7,997). Baseline characteristics of patients are listed in Table 1. With the exception of proportions of patients with arterial hypertension, previous myocardial infarctions, and histories of coronary artery bypass surgery, all other variables appeared to differ significantly between patients in the 2 groups. Of note, patients with fibrinogen >402.0 mg/dl showed a markedly more adverse cardiovascular risk profile compared to those with fibrinogen 402.0 mg/dl. In patients with 1-, 2-, and 3-vessel disease, median fibrinogen levels were 355.0 mg/dl (IQR 293.0 to 437.0), 364.0 mg/dl (IQR
303.0 to 447.0), and 376.0 mg/dl (IQR 318.0 to 466.0), respectively (p <0.001). Patients with acute coronary syndromes had higher fibrinogen levels than those with stable CAD (median 383.0 mg/dl [IQR 316.0 to 488.0] vs 362.0 mg/dl [IQR 307.0 to 440.0], p <0.001). Coronary stents were implanted in 90.5% of patients (n ¼ 11,938). Drug-eluting stents were used in 9,023 patients (76%). Statins at discharge were prescribed in 12,041 patients (91.2%), b blockers in 12,401 patients (94%), and angiotensin-converting enzyme inhibitors in 12,225 patients (93%). Within the first year after PCI, there were 639 deaths. The median fibrinogen level was 366.0 mg/dl (IQR 308.0 to 450.0) among survivors and 449.0 mg/dl (IQR 347.0 to 576.7) among nonsurvivors (p <0.001; Figure 1). All-cause deaths occurred in 393 patients with fibrinogen >402.0 mg/dl and 246 patients with fibrinogen 402.0 mg/dl (Kaplan-Meier estimates of mortality 7.7% and 3.1%, respectively, unadjusted hazard ratio [HR] 2.52, 95% CI 2.16 to 2.94, p <0.001; Figure 2). Deaths of cardiac origin occurred in 451 patients: 273 among patients with fibrinogen >402.0 mg/dl and 178 among those with fibrinogen 402.0 mg/dl (Kaplan-Meier estimates of mortality 5.4% and 2.3%, respectively, unadjusted HR 2.41, 95% CI 2.01 to 2.90, p <0.001; Figure 2). Nonfatal myocardial infarctions occurred in 213 patients with fibrinogen >402.0 mg/dl and 204 patients with fibrinogen 402.0 mg/dl (Kaplan-Meier estimates of nonfatal myocardial infarction 4.2% and 2.6%, respectively, unadjusted HR 1.63, 95% CI 1.35 to 1.97, p <0.001). Definite stent thrombosis occurred in 50 patients with fibrinogen >402.0 mg/dl and 35 patients with fibrinogen 402.0 mg/dl (1.0% vs 0.4%, respectively, unadjusted HR 2.24, 95% CI 1.47 to 3.41, p <0.001). Stroke occurred in 42 patients with fibrinogen >402.0 mg/dl and 51 patients with fibrinogen level 402.0 mg/dl (Kaplan-Meier
808
The American Journal of Cardiology (www.ajconline.org)
Figure 5. Cardiac deaths in subgroups of patients. Deaths are reported as numbers with Kaplan-Meier estimates obtained by survival analysis. ACS ¼ acute coronary syndromes; BMI ¼ body mass index; GFR ¼ glomerular filtration rate; LVEF ¼ left ventricular ejection fraction; Pint ¼ p value for interaction. Table 3 Results of multivariate Cox proportional-hazards model regarding all-cause and cardiac mortality Characteristic
All-Cause Mortality Adjusted HR (95% CI)
Fibrinogen (for each 50 mg/dl increase) Age (for 10-yr increase) Female gender Body mass index (for each 5 kg/m2 increase) Diabetes mellitus Arterial hypertension Hypercholesterolemia Current smoking Previous myocardial infarction Previous coronary artery bypass surgery Clinical presentation (acute coronary syndromes vs stable CAD) Glomerular filtration rate (for each 30 ml/min decrease) Left ventricular ejection fraction (for each 10% decrease) Multivessel disease (vs single-vessel disease)
estimates of stroke 0.8% and 0.6%, respectively, unadjusted HR 1.29, 95% CI 0.86 to 1.94, p ¼ 0.22). The association pattern of fibrinogen with mortality was assessed by analyzing the mortality rates in each decile of fibrinogen concentration (Table 2). The lowest rates of allcause (2.34% to 2.72%) and cardiac (1.89% to 2.10%) mortality were observed in the third to fifth deciles of fibrinogen (concentration 295 to 369 mg/dl). Mortality increased on both sides of this concentration range, showing a J-shaped fibrinogen-mortality relation (Figure 3). The relation between fibrinogen and mortality was analyzed in various subgroups of patients (see “Methods” for criteria of cutoff definition). Results of subgroup analyses
1.07 1.15 1.08 1.11 1.53 0.61 0.74 1.05 0.88 0.93 2.03 2.27 1.54 1.70
(1.04e1.10) (1.01e1.30) (0.87e1.33) (0.98e1.26) (1.26e1.86) (0.51e0.75) (0.61e0.90) (0.79e1.40) (0.72e1.08) (0.71e1.22) (1.65e2.49) (1.51e2.70) (1.44e1.65) (1.23e2.36)
Cardiac Mortality p Value <0.001 0.03 0.49 0.107 <0.001 <0.001 0.002 0.72 0.23 0.59 <0.001 <0.001 <0.001 0.001
Adjusted HR (95% CI) 1.05 1.07 1.09 1.18 1.53 0.55 0.77 0.94 0.82 0.79 2.27 2.32 1.64 1.89
(1.01e1.09) (0.93e1.24) (0.85e1.39) (1.02e1.36) (1.22e1.93) (0.43e0.69) (0.61e0.97) (0.67e1.32) (0.64e1.05) (0.56e1.11) (1.76e2.91) (1.92e2.86) (1.51e1.78) (1.27e2.81)
p Value 0.008 0.36 0.51 0.03 <0.001 <0.001 0.024 0.72 0.12 0.17 <0.001 <0.001 <0.001 0.002
are shown in Figures 4 and 5. The analysis showed that fibrinogen predicted an increased risk for mortality across all subgroups. An interaction was found between fibrinogen and arterial hypertension, revealing a stronger association between fibrinogen and all-cause mortality in patients with arterial hypertension than in patients without arterial hypertension (p for interaction ¼ 0.02; Figure 4). There was another fibrinogen-by-gender interaction showing a stronger association between fibrinogen and cardiac mortality in men than in women (p for interaction ¼ 0.01; Figure 5). The Cox proportional-hazards model was used to assess the association between fibrinogen level and mortality while adjusting for potential confounders (see “Methods” for
Coronary Artery Disease/Fibrinogen and Cardiovascular Events
variables entered into the model). After adjustment in the Cox model, the association between fibrinogen and mortality remained significant, showing increases in the adjusted risk for death of 7% and 5% for all-cause and cardiac mortality, respectively, for each 50 mg/dl increase in fibrinogen level. Results of the Cox model are listed in Table 3. The inclusion of fibrinogen in the multivariate model did not improve the discriminatory power of the model regarding prediction of all-cause (absolute IDI 0.002, relative IDI 1.4%, p ¼ 0.32) or cardiac (absolute IDI 0.0003, relative IDI 0.25%, p ¼ 0.84) mortality. Discussion The main findings of this study can be summarized as follows: (1) In patients with confirmed CAD, elevated plasma fibrinogen levels predicted an increased risk for allcause and cardiac mortality independent of traditional cardiovascular risk factors, extent of angiographic CAD, the left ventricular ejection fraction, and renal function, (2) The association between plasma fibrinogen and all-cause or cardiac mortality followed a J-shaped pattern, with the lowest mortality at fibrinogen concentrations of 295 to 369 mg/dl. For fibrinogen values <295 and >369 mg/dl, allcause and cardiac mortality increased, (3) Elevated fibrinogen levels predicted an increased risk for death across all subsets of patients with CAD. A fibrinogen-by-gender and a fibrinogenebyearterial hypertension interaction demonstrating a stronger association of fibrinogen with cardiac mortality in men than in women and a stronger association between fibrinogen and all-cause mortality in patients with hypertension than in patients without arterial hypertension were observed, and (4) Fibrinogen did not improve the discriminatory power of the multivariate model(s) for the prediction of mortality. The association of elevated fibrinogen levels with myocardial infarction was described in 1954 by Losner et al.12 Numerous subsequent studies have shown that fibrinogen levels are related to cardiovascular risk factors, incident cardiovascular disease, or cardiovascular adverse events.1,2 An array of mechanisms by which fibrinogen may promote atherosclerosis and/or atherothrombosis and consequently cardiovascular events have been described and reviewed.13,14 Concerns have also been raised that many previous studies have included subjects or patients who are not representative of contemporary Western populations.4 Moreover, the results of studies investigating the association between fibrinogen and cardiovascular events are affected by the degree of adjustment in multivariate models and the use (or not) of the appropriate tests to assess the impact of fibrinogen on the discriminatory power of the multivariate models for risk prediction. Several recent studies including more contemporary series of patients or studies that have used advanced research methods came to opposite conclusions. In the Prospective Epidemiological Study of Myocardial Infarction (PRIME) study, which included middle-aged men free of CAD at entry, adjustment for cardiovascular risk factors and interleukin-6 abolished the independent association of fibrinogen or C-reactive protein with the risk for myocardial infarction and coronary
809
death.3,15 The AtheroGene, study, which included patients with confirmed stable CAD, showed that fibrinogen and C-reactive protein predicted future cardiovascular risk independent of cardiovascular risk factors, but they did not provide further information on top of that provided by traditional cardiovascular risk factors.4 The International Studies of Infarct Survival (ISIS) and a meta-analysis of 19 other studies of b-fibrinogen genotypes suggested that genotypes that produce lifelong differences in fibrinogen concentration did not influence the incidence of CAD.5 Our data seem to support the findings of the AtheroGene study in showing that plasma fibrinogen is independently associated with the increased risk for mortality and that fibrinogen did not improve the discriminatory power of multivariate models for the prediction of mortality. Although the exact reasons for the failure of fibrinogen to provide prognostic information on top of that provided by traditional cardiovascular risk factors remain unknown, 2 putative explanations may be offered. First, close associations of fibrinogen with an array of cardiovascular risk factors have been described.16 Consequently, elevated fibrinogen level may be considered an associate of increased cardiovascular risk. Second, in contrast to many previous studies, in the present study, statins (known for their strong anti-inflammatory actions) were used in >90% of the patients. These agents through their anti-inflammatory actions may attenuate the association between fibrinogen and cardiovascular events. To our knowledge, the finding of a J-shaped relation between fibrinogen level and mortality is novel. The finding may be important because it suggests the most optimal fibrinogen levels associated with lowest mortality. Although multiple mechanisms may explain increased coronary risk in patients with elevated fibrinogen levels, reasons for increased risk for death in patients with low fibrinogen levels remain unknown. Factors or measures that reduce fibrinogen level are not considered to increase cardiovascular risk.13,14 Alternatively, low fibrinogen levels may reflect hidden co-morbidities (such as hepatic diseases) that might have remained unaccounted for. Subgroup analyses showed that fibrinogen predicted an increased risk for cardiac and all-cause mortality across all subsets of patients with CAD. Of note, a fibrinogen-by-gender interaction showing a stronger association between fibrinogen and cardiac mortality in men than in women and a fibrinogenebyearterial hypertension interaction revealing a stronger association between fibrinogen and all-cause mortality in patients with hypertension than in patients without arterial hypertension were observed. A recent study has reported an interaction between gender and fibrinogen (or C-reactive protein), suggesting that these inflammatory markers improve cardiovascular risk discrimination in men but not in women.2 In the Cardiovascular Health Study (CHS), which included subjects aged 65 years, fibrinogen predicted mortality in men but not in women.17 Although we found a significantly stronger association between fibrinogen and cardiac mortality in men than in women and a stronger association between fibrinogen and mortality in patients with hypertension than in patients without arterial hypertension, no good explanation for these findings can currently be offered. Our study had limitations. First, the follow-up period of 1 year might be short. However, at 1 year, time-to-event
810
The American Journal of Cardiology (www.ajconline.org)
curves were still expanding, which may suggest a further accentuation of the differences in survival over a longer follow-up period. Second, when analyzing the results of subgroup analyses, caution is warranted because of possible errors introduced by multiple testing. Third, these findings belong to patients confirmed to have CAD and may not be extrapolated to subjects in the general population. Disclosures The authors have no conflicts of interest to disclose. 1. Danesh J, Lewington S, Thompson SG, Lowe GD, Collins R, Kostis JB, Wilson AC, Folsom AR, Wu K, Benderly M, Goldbourt U, Willeit J, Kiechl S, Yarnell JW, Sweetnam PM, Elwood PC, Cushman M, Psaty BM, Tracy RP, Tybjaerg-Hansen A, Haverkate F, de Maat MP, Fowkes FG, Lee AJ, Smith FB, Salomaa V, Harald K, Rasi R, Vahtera E, Jousilahti P, Pekkanen J, D’Agostino R, Kannel WB, Wilson PW, Tofler G, Arocha-Pinango CL, Rodriguez-Larralde A, Nagy E, Mijares M, Espinosa R, Rodriquez-Roa E, Ryder E, Diez-Ewald MP, Campos G, Fernandez V, Torres E, Marchioli R, Valagussa F, Rosengren A, Wilhelmsen L, Lappas G, Eriksson H, Cremer P, Nagel D, Curb JD, Rodriguez B, Yano K, Salonen JT, Nyyssonen K, Tuomainen TP, Hedblad B, Lind P, Loewel H, Koenig W, Meade TW, Cooper JA, De Stavola B, Knottenbelt C, Miller GJ, Bauer KA, Rosenberg RD, Sato S, Kitamura A, Naito Y, Palosuo T, Ducimetiere P, Amouyel P, Arveiler D, Evans AE, Ferrieres J, Juhan-Vague I, Bingham A, Schulte H, Assmann G, Cantin B, Lamarche B, Despres JP, Dagenais GR, Tunstall-Pedoe H, Woodward M, Ben-Shlomo Y, Davey Smith G, Palmieri V, Yeh JL, Rudnicka A, Ridker P, Rodeghiero F, Tosetto A, Shepherd J, Ford I, Robertson M, Brunner E, Shipley M, Feskens EJ, Kromhout D, Dickinson A, Ireland B, Juzwishin K, Kaptoge S, Memon A, Sarwar N, Walker M, Wheeler J, White I, Wood A. Plasma fibrinogen level and the risk of major cardiovascular diseases and nonvascular mortality: an individual participant meta-analysis. JAMA 2005;294:1799e1809. 2. Kaptoge S, Di Angelantonio E, Pennells L, Wood AM, White IR, Gao P, Walker M, Thompson A, Sarwar N, Caslake M, Butterworth AS, Amouyel P, Assmann G, Bakker SJ, Barr EL, Barrett-Connor E, Benjamin EJ, Bjorkelund C, Brenner H, Brunner E, Clarke R, Cooper JA, Cremer P, Cushman M, Dagenais GR, D’Agostino RB Sr, Dankner R, Davey-Smith G, Deeg D, Dekker JM, Engstrom G, Folsom AR, Fowkes FG, Gallacher J, Gaziano JM, Giampaoli S, Gillum RF, Hofman A, Howard BV, Ingelsson E, Iso H, Jorgensen T, Kiechl S, Kitamura A, Kiyohara Y, Koenig W, Kromhout D, Kuller LH, Lawlor DA, Meade TW, Nissinen A, Nordestgaard BG, Onat A, Panagiotakos DB, Psaty BM, Rodriguez B, Rosengren A, Salomaa V, Kauhanen J, Salonen JT, Shaffer JA, Shea S, Ford I, Stehouwer CD, Strandberg TE, Tipping RW, Tosetto A, Wassertheil-Smoller S, Wennberg P, Westendorp RG, Whincup PH, Wilhelmsen L, Woodward M, Lowe GD, Wareham NJ, Khaw KT, Sattar N, Packard CJ, Gudnason V, Ridker PM, Pepys MB, Thompson SG, Danesh J. C-reactive protein, fibrinogen, and cardiovascular disease prediction. N Engl J Med 2012;367:1310e1320.
3. Luc G, Bard JM, Juhan-Vague I, Ferrieres J, Evans A, Amouyel P, Arveiler D, Fruchart JC, Ducimetiere P. C-reactive protein, interleukin6, and fibrinogen as predictors of coronary heart disease: the PRIME study. Arterioscler Thromb Vasc Biol 2003;23:1255e1261. 4. Sinning JM, Bickel C, Messow CM, Schnabel R, Lubos E, Rupprecht HJ, Espinola-Klein C, Lackner KJ, Tiret L, Munzel T, Blankenberg S. Impact of C-reactive protein and fibrinogen on cardiovascular prognosis in patients with stable angina pectoris: the AtheroGene study. Eur Heart J 2006;27:2962e2968. 5. Keavney B, Danesh J, Parish S, Palmer A, Clark S, Youngman L, Delepine M, Lathrop M, Peto R, Collins R. Fibrinogen and coronary heart disease: test of causality by “Mendelian randomization.” Int J Epidemiol 2006;35:935e943. 6. Myers GL, Christenson RH, Cushman M, Ballantyne CM, Cooper GR, Pfeiffer CM, Grundy SM, Labarthe DR, Levy D, Rifai N, Wilson PW. National Academy of Clinical Biochemistry laboratory medicine practice guidelines: emerging biomarkers for primary prevention of cardiovascular disease. Clin Chem 2009;55:378e384. 7. Ndrepepa G, Braun S, Haase HU, Schulz S, Ranftl S, Hadamitzky M, Mehilli J, Schomig A, Kastrati A. Prognostic value of uric acid in patients with acute coronary syndromes. Am J Cardiol 2012;109: 1260e1265. 8. Sandler H, Dodge HT. The use of single plane angiocardiograms for the calculation of left ventricular volume in man. Am Heart J 1968;75: 325e334. 9. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976;16:31e41. 10. Cutlip DE, Windecker S, Mehran R, Boam A, Cohen DJ, van Es GA, Steg PG, Morel MA, Mauri L, Vranckx P, McFadden E, Lansky A, Hamon M, Krucoff MW, Serruys PW. Clinical end points in coronary stent trials: a case for standardized definitions. Circulation 2007;115: 2344e2351. 11. Pencina MJ, D’Agostino RB Sr, D’Agostino RB Jr, Vasan RS. Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyond. Stat Med 2008;27:157e172. 12. Losner S, Volk BW, Wilensky ND. Fibrinogen concentration in acute myocardial infarction; comparison of the clot density determination of fibrinogen with the erythrocyte sedimentation rate. AMA Arch Intern Med 1954;93:231e245. 13. Kamath S, Lip GY. Fibrinogen: biochemistry, epidemiology and determinants. QJM 2003;96:711e729. 14. Koenig W. Fibrin(ogen) in cardiovascular disease: an update. Thromb Haemost 2003;89:601e609. 15. Empana JP, Jouven X, Canoui-Poitrine F, Luc G, Tafflet M, Haas B, Arveiler D, Ferrieres J, Ruidavets JB, Montaye M, Yarnell J, Morange P, Kee F, Evans A, Amouyel P, Ducimetiere P. C-reactive protein, interleukin 6, fibrinogen and risk of sudden death in European middleaged men: the PRIME study. Arterioscler Thromb Vasc Biol 2010;30: 2047e2052. 16. Folsom AR. Fibrinogen and cardiovascular risk markers. Blood Coagul Fibrinolysis 1999;10(suppl):S13eS16. 17. Tracy RP, Arnold AM, Ettinger W, Fried L, Meilahn E, Savage P. The relationship of fibrinogen and factors VII and VIII to incident cardiovascular disease and death in the elderly: results from the cardiovascular health study. Arterioscler Thromb Vasc Biol 1999;19: 1776e1783.
0002-9149/13/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2012.11.060
medicine practice guidelines that evidence for the usefulness of fibrinogen in the risk stratification of patients at risk for cardiovascular events is inconclusive.6 Moreover, other aspects of the association between fibrinogen and cardiovascular events, such as the pattern of fibrinogen-mortality relation or the strength of the association in various subgroups of patients with CAD, remain poorly investigated. We undertook this study with a double objective: (1) to investigate the association between fibrinogen and cardiovascular events in patients with CAD and whether this information is additive to that provided by traditional cardiovascular risk factors and (2) to assess the strength of the association between fibrinogen and mortality in various subgroups of patients. Methods This study included 13,195 patients with angiographyproved CAD who underwent percutaneous coronary intervention (PCI) in the German Heart Center in Munich from March 2000 to December 2009. To be included in the study, patients had to have significant CAD, confirmed by coronary angiography at the time of index hospitalization. Patients who underwent PCI in the setting of failed thrombolysis or routine PCI after thrombolysis, as well as those with no fibrinogen measurements available, acute infections, serum creatinine 2 mg/dl, and known malignant diseases www.ajconline.org
Coronary Artery Disease/Fibrinogen and Cardiovascular Events
805
Table 1 Baseline characteristics Characteristic
Age (yrs) Women Diabetes mellitus Receiving insulin therapy Body mass index (kg/m2) Arterial hypertension Current smoker Hypercholesterolemia (220 mg/dl) Previous myocardial infarction Previous coronary artery bypass surgery Clinical presentation Stable CAD* Acute coronary syndromes Serum creatinine (mg/dl) Glomerular filtration rate (ml/min) Fibrinogen (mg/dl) Number of narrowed coronary arteries 1 2 3 Multivessel coronary disease Left ventricular ejection fraction (%)
Fibrinogen (mg/dl)
p Value
402.0 (n ¼ 7,997)
>402.0 (n ¼ 5,198)
66.2 1,511 1,930 530 26.8 5,490 1,167 5,663 2,580 1,209
(58.4e73.2) (19%) (24%) (7%) (24.6e29.4) (69%) (15%) (71%) (32%) (15%)
69.7 1,574 1,756 650 27.1 3,580 911 3,577 1,601 774
(61.8e76.4) (30%) (34%) (12.5%) (24.7e30.0) (69%) (18%) (69%) (31%) (15%)
5,219 2,778 1.0 83.3 324.0
(65%) (35%) (0.8e1.1) (64.9e104.0) (279.0e358.0)
2,897 2,301 1.0 73.2 483.0
(56%) (44%) (0.8e1.2) (53.5e96.6) (436.0e552.0)
1,519 2,168 4,310 6,478 58.0
(19%) (27%) (54%) (81%) (49.0e63.0)
784 1,331 3,083 4,414 55.0
(15%) (26%) (59%) (85%) (44.0e61.0)
<0.001 <0.001 <0.001 <0.001 <0.001 0.79 <0.001 0.014 0.08 0.72 <0.001 <0.001 <0.001 <0.001 <0.001
<0.001 <0.001
Data are expressed as number (percentage) or as median (interquartile range). Data on body mass index, glomerular filtration rate, and the left ventricular ejection fraction were available for 13,160 patients (99.7%), 13,161 patients (99.7%), and 11,599 patients (87.9%), respectively. * Defined as chest pain that did not change its pattern within the past 2 months with angiographic confirmation of significant CAD.
with life expectancy <1 year, were excluded. All patients gave written informed consent for angiographic examination, performance of PCI, and blood sampling. The study was conducted according to the principles of the Declaration of Helsinki and approved by the institutional ethics committee. Stable CAD was diagnosed if patients presented with chest pain that did not change its pattern within the past 2 months and had angiographic confirmation of CAD. Acute coronary syndromes were diagnosed using pattern-specific criteria.7 In all cases, the diagnosis was confirmed by angiographic criteria, which included the presence of coronary stenoses 50% luminal obstruction in 1 of the 3 major coronary arteries and/or documentation of culprit lesions in case of acute coronary syndromes. The left ventricular ejection fraction was calculated using the area-length method using left ventricular angiography.8 Main cardiovascular risk factors were defined using standard criteria. Hypercholesterolemia was defined as a documented total cholesterol value 220 mg/dl or previous or ongoing treatment with lipid-lowering agents. Arterial hypertension was defined when a patient was under active treatment with antihypertensive drugs or if the systolic blood pressure was 140 mm Hg or the diastolic blood pressure was 90 mm Hg on 2 separate occasions. Current smokers were those with regular smoking in the previous 6 months. The diagnosis of diabetes required active treatment with insulin or an oral hypoglycemic agent, abnormal fasting blood glucose (126 mg/dl), blood glucose >200 mg/dl at any time, or abnormal glucose tolerance according to World Health Organization criteria.
Weight and height were measured during the index hospitalization, and body mass index was calculated. The glomerular filtration rate was estimated using the CockcroftGault equation.9 PCI and periprocedural care were performed according to standard criteria. Antithrombotic therapy included clopidogrel (300 or 600 mg as a loading dose followed by 75 mg/day for 1 month) and aspirin (200 mg/day administered orally and continued indefinitely). Blood samples were obtained before angiography. Venous blood was collected using the S-Monovettes blood collection system (Sarstedt, Sarstedt, Germany) containing 0.106 mol/L sodium citrate. Nine volumes of blood were mixed with 1 volume of sodium citrate solution. Plasma was separated after immediate centrifugation at 1,500g for 10 minutes. Fibrinogen was determined in citrated plasma by a modification of the Clauss method using the BCS analyzer (Multifibren U; Siemens Healthcare, Erlangen, Germany). The measurement range lies between 80 and 1,200 mg/dl. Expected values range from 180 to 350 mg/dl. Heparin concentration in plasma <2 U/ml does not affect the test. Creatinine was measured using a kinetic colorimetric assay on the basis of the compensated Jaffe method. Laboratory personnel were unaware of clinical, angiographic, or followup information. The primary outcome was all-cause mortality at 1 year after PCI. Secondary outcomes included 1-year occurrences of cardiac mortality, nonfatal myocardial infarction, and stroke. Information on mortality was obtained from hospital records, death certificates, or phone contact with relatives of the patients or referring physicians. Cardiac death was
806
The American Journal of Cardiology (www.ajconline.org) Table 2 One-year mortality according to fibrinogen deciles Decile of Fibrinogen
Fibrinogen Level (mg/dl)
Number of Patients
1st 2nd 3rd 4th 5th 6th 7th 8th 9th 10th
<262 262 to <295 295 to <323 323 to <346 346 to <369 369 to <400 400 to <435 435 to <482 482 to <551 551 to 1,190
1,278 1,341 1,324 1,330 1,288 1,351 1,307 1,324 1,332 1,320
Mortality All-Cause 49 40 31 34 35 53 54 66 96 181
(3.83) (2.98) (2.34) (2.56) (2.72) (2.92) (4.13) (4.98) (7.21) (13.71)
Cardiac 35 28 25 27 27 34 37 50 64 124
(2.74) (2.09) (1.89) (2.03) (2.10) (2.52) (2.83) (3.78) (4.80) (9.39)
Figure 1. Fibrinogen levels in nonsurvivors and survivors. Data are medians with 25th to 75th and 5th to 95th percentiles.
Figure 3. One-year mortality in various fibrinogen deciles.
Figure 2. Kaplan-Meier curves of all-cause (A) and cardiac (B) mortality. Fib ¼ fibrinogen.
defined according to Academic Research Consortium criteria.10 The diagnosis of myocardial infarction required the development of new abnormal Q waves in 2 contiguous precordial or 2 adjacent limb leads or an elevation of creatine kinase-MB >2 times (>3 times for the 48 hours after a PCI procedure) the upper limit of normal. Definite stent thrombosis was defined according to the Academic Research
Consortium criteria. Stroke required the occurrence of acute neurologic deficits that were confirmed by computed tomography or magnetic resonance imaging of the head. The follow-up protocol after discharge included a phone interview at 1 month, a visit at 6 months, and a phone interview at 12 months. Follow-up information and adjudication of adverse events was performed by medical staff members unaware of clinical diagnosis or fibrinogen level. Data are presented as median (interquartile range [IQR]), number of patients or events, or proportions. The Kolmogorov-Smirnov test was used to assess the normality of data distribution. Continuous data were compared using KruskalWallis rank-sum test. Categorical data were compared using chi-square test. Receiver-operating characteristic curve analysis was performed to determine the best fibrinogen cutoff value for the prediction of all-cause mortality while maximizing sensitivity and specificity. Survival analysis was performed using the Kaplan-Meier method, and differences in survival were compared using the log-rank test. The association between fibrinogen and mortality (allcause or cardiac) was tested using the multivariate Cox proportional-hazards model. All variables listed in Table 1 were entered into the model. The discriminatory power of the multivariate model regarding prediction of mortality was assessed by calculating the integrated discrimination improvement (IDI) according to Pencina et al.11 Differences in the association between fibrinogen and survival in various
Coronary Artery Disease/Fibrinogen and Cardiovascular Events
807
Figure 4. All-cause deaths in subgroups of patients. Deaths are reported as numbers with Kaplan-Meier estimates obtained by survival analysis. ACS ¼ acute coronary syndromes; BMI ¼ body mass index; GFR ¼ glomerular filtration rate; LVEF ¼ left ventricular ejection fraction; Pint ¼ p value for interaction.
subgroups of patients (obtained by dichotomization of patients according to age [cutoff 65 years], gender, diabetes, arterial hypertension, smoking status, hypercholesterolemia, clinical presentation [stable CAD vs acute coronary syndromes], glomerular filtration rate [cutoff 60 ml/min], body mass index [cutoff 30 kg/m2], and the left ventricular ejection fraction [cutoff 50%]) were investigated by performing interaction testing. Analyses were performed using S-Plus (Insightful Corporation, Seattle, Washington). Twotailed p values <0.05 were considered to indicate statistical significance. Results The receiver-operating characteristic curve analysis showed that fibrinogen predicted 1-year all-cause mortality with an area under the curve of 0.652 (95% confidence interval [CI] 0.628 to 0.677). The best cut-off value of fibrinogen for the prediction of 1-year all-cause mortality, while maximizing sensitivity and specificity, was 402.0 mg/dl. Using this value, patients were divided into 2 groups: those with fibrinogen >402.0 mg/dl (n ¼ 5,198) and those with fibrinogen 402.0 mg/dl (n ¼ 7,997). Baseline characteristics of patients are listed in Table 1. With the exception of proportions of patients with arterial hypertension, previous myocardial infarctions, and histories of coronary artery bypass surgery, all other variables appeared to differ significantly between patients in the 2 groups. Of note, patients with fibrinogen >402.0 mg/dl showed a markedly more adverse cardiovascular risk profile compared to those with fibrinogen 402.0 mg/dl. In patients with 1-, 2-, and 3-vessel disease, median fibrinogen levels were 355.0 mg/dl (IQR 293.0 to 437.0), 364.0 mg/dl (IQR
303.0 to 447.0), and 376.0 mg/dl (IQR 318.0 to 466.0), respectively (p <0.001). Patients with acute coronary syndromes had higher fibrinogen levels than those with stable CAD (median 383.0 mg/dl [IQR 316.0 to 488.0] vs 362.0 mg/dl [IQR 307.0 to 440.0], p <0.001). Coronary stents were implanted in 90.5% of patients (n ¼ 11,938). Drug-eluting stents were used in 9,023 patients (76%). Statins at discharge were prescribed in 12,041 patients (91.2%), b blockers in 12,401 patients (94%), and angiotensin-converting enzyme inhibitors in 12,225 patients (93%). Within the first year after PCI, there were 639 deaths. The median fibrinogen level was 366.0 mg/dl (IQR 308.0 to 450.0) among survivors and 449.0 mg/dl (IQR 347.0 to 576.7) among nonsurvivors (p <0.001; Figure 1). All-cause deaths occurred in 393 patients with fibrinogen >402.0 mg/dl and 246 patients with fibrinogen 402.0 mg/dl (Kaplan-Meier estimates of mortality 7.7% and 3.1%, respectively, unadjusted hazard ratio [HR] 2.52, 95% CI 2.16 to 2.94, p <0.001; Figure 2). Deaths of cardiac origin occurred in 451 patients: 273 among patients with fibrinogen >402.0 mg/dl and 178 among those with fibrinogen 402.0 mg/dl (Kaplan-Meier estimates of mortality 5.4% and 2.3%, respectively, unadjusted HR 2.41, 95% CI 2.01 to 2.90, p <0.001; Figure 2). Nonfatal myocardial infarctions occurred in 213 patients with fibrinogen >402.0 mg/dl and 204 patients with fibrinogen 402.0 mg/dl (Kaplan-Meier estimates of nonfatal myocardial infarction 4.2% and 2.6%, respectively, unadjusted HR 1.63, 95% CI 1.35 to 1.97, p <0.001). Definite stent thrombosis occurred in 50 patients with fibrinogen >402.0 mg/dl and 35 patients with fibrinogen 402.0 mg/dl (1.0% vs 0.4%, respectively, unadjusted HR 2.24, 95% CI 1.47 to 3.41, p <0.001). Stroke occurred in 42 patients with fibrinogen >402.0 mg/dl and 51 patients with fibrinogen level 402.0 mg/dl (Kaplan-Meier
808
The American Journal of Cardiology (www.ajconline.org)
Figure 5. Cardiac deaths in subgroups of patients. Deaths are reported as numbers with Kaplan-Meier estimates obtained by survival analysis. ACS ¼ acute coronary syndromes; BMI ¼ body mass index; GFR ¼ glomerular filtration rate; LVEF ¼ left ventricular ejection fraction; Pint ¼ p value for interaction. Table 3 Results of multivariate Cox proportional-hazards model regarding all-cause and cardiac mortality Characteristic
All-Cause Mortality Adjusted HR (95% CI)
Fibrinogen (for each 50 mg/dl increase) Age (for 10-yr increase) Female gender Body mass index (for each 5 kg/m2 increase) Diabetes mellitus Arterial hypertension Hypercholesterolemia Current smoking Previous myocardial infarction Previous coronary artery bypass surgery Clinical presentation (acute coronary syndromes vs stable CAD) Glomerular filtration rate (for each 30 ml/min decrease) Left ventricular ejection fraction (for each 10% decrease) Multivessel disease (vs single-vessel disease)
estimates of stroke 0.8% and 0.6%, respectively, unadjusted HR 1.29, 95% CI 0.86 to 1.94, p ¼ 0.22). The association pattern of fibrinogen with mortality was assessed by analyzing the mortality rates in each decile of fibrinogen concentration (Table 2). The lowest rates of allcause (2.34% to 2.72%) and cardiac (1.89% to 2.10%) mortality were observed in the third to fifth deciles of fibrinogen (concentration 295 to 369 mg/dl). Mortality increased on both sides of this concentration range, showing a J-shaped fibrinogen-mortality relation (Figure 3). The relation between fibrinogen and mortality was analyzed in various subgroups of patients (see “Methods” for criteria of cutoff definition). Results of subgroup analyses
1.07 1.15 1.08 1.11 1.53 0.61 0.74 1.05 0.88 0.93 2.03 2.27 1.54 1.70
(1.04e1.10) (1.01e1.30) (0.87e1.33) (0.98e1.26) (1.26e1.86) (0.51e0.75) (0.61e0.90) (0.79e1.40) (0.72e1.08) (0.71e1.22) (1.65e2.49) (1.51e2.70) (1.44e1.65) (1.23e2.36)
Cardiac Mortality p Value <0.001 0.03 0.49 0.107 <0.001 <0.001 0.002 0.72 0.23 0.59 <0.001 <0.001 <0.001 0.001
Adjusted HR (95% CI) 1.05 1.07 1.09 1.18 1.53 0.55 0.77 0.94 0.82 0.79 2.27 2.32 1.64 1.89
(1.01e1.09) (0.93e1.24) (0.85e1.39) (1.02e1.36) (1.22e1.93) (0.43e0.69) (0.61e0.97) (0.67e1.32) (0.64e1.05) (0.56e1.11) (1.76e2.91) (1.92e2.86) (1.51e1.78) (1.27e2.81)
p Value 0.008 0.36 0.51 0.03 <0.001 <0.001 0.024 0.72 0.12 0.17 <0.001 <0.001 <0.001 0.002
are shown in Figures 4 and 5. The analysis showed that fibrinogen predicted an increased risk for mortality across all subgroups. An interaction was found between fibrinogen and arterial hypertension, revealing a stronger association between fibrinogen and all-cause mortality in patients with arterial hypertension than in patients without arterial hypertension (p for interaction ¼ 0.02; Figure 4). There was another fibrinogen-by-gender interaction showing a stronger association between fibrinogen and cardiac mortality in men than in women (p for interaction ¼ 0.01; Figure 5). The Cox proportional-hazards model was used to assess the association between fibrinogen level and mortality while adjusting for potential confounders (see “Methods” for
Coronary Artery Disease/Fibrinogen and Cardiovascular Events
variables entered into the model). After adjustment in the Cox model, the association between fibrinogen and mortality remained significant, showing increases in the adjusted risk for death of 7% and 5% for all-cause and cardiac mortality, respectively, for each 50 mg/dl increase in fibrinogen level. Results of the Cox model are listed in Table 3. The inclusion of fibrinogen in the multivariate model did not improve the discriminatory power of the model regarding prediction of all-cause (absolute IDI 0.002, relative IDI 1.4%, p ¼ 0.32) or cardiac (absolute IDI 0.0003, relative IDI 0.25%, p ¼ 0.84) mortality. Discussion The main findings of this study can be summarized as follows: (1) In patients with confirmed CAD, elevated plasma fibrinogen levels predicted an increased risk for allcause and cardiac mortality independent of traditional cardiovascular risk factors, extent of angiographic CAD, the left ventricular ejection fraction, and renal function, (2) The association between plasma fibrinogen and all-cause or cardiac mortality followed a J-shaped pattern, with the lowest mortality at fibrinogen concentrations of 295 to 369 mg/dl. For fibrinogen values <295 and >369 mg/dl, allcause and cardiac mortality increased, (3) Elevated fibrinogen levels predicted an increased risk for death across all subsets of patients with CAD. A fibrinogen-by-gender and a fibrinogenebyearterial hypertension interaction demonstrating a stronger association of fibrinogen with cardiac mortality in men than in women and a stronger association between fibrinogen and all-cause mortality in patients with hypertension than in patients without arterial hypertension were observed, and (4) Fibrinogen did not improve the discriminatory power of the multivariate model(s) for the prediction of mortality. The association of elevated fibrinogen levels with myocardial infarction was described in 1954 by Losner et al.12 Numerous subsequent studies have shown that fibrinogen levels are related to cardiovascular risk factors, incident cardiovascular disease, or cardiovascular adverse events.1,2 An array of mechanisms by which fibrinogen may promote atherosclerosis and/or atherothrombosis and consequently cardiovascular events have been described and reviewed.13,14 Concerns have also been raised that many previous studies have included subjects or patients who are not representative of contemporary Western populations.4 Moreover, the results of studies investigating the association between fibrinogen and cardiovascular events are affected by the degree of adjustment in multivariate models and the use (or not) of the appropriate tests to assess the impact of fibrinogen on the discriminatory power of the multivariate models for risk prediction. Several recent studies including more contemporary series of patients or studies that have used advanced research methods came to opposite conclusions. In the Prospective Epidemiological Study of Myocardial Infarction (PRIME) study, which included middle-aged men free of CAD at entry, adjustment for cardiovascular risk factors and interleukin-6 abolished the independent association of fibrinogen or C-reactive protein with the risk for myocardial infarction and coronary
809
death.3,15 The AtheroGene, study, which included patients with confirmed stable CAD, showed that fibrinogen and C-reactive protein predicted future cardiovascular risk independent of cardiovascular risk factors, but they did not provide further information on top of that provided by traditional cardiovascular risk factors.4 The International Studies of Infarct Survival (ISIS) and a meta-analysis of 19 other studies of b-fibrinogen genotypes suggested that genotypes that produce lifelong differences in fibrinogen concentration did not influence the incidence of CAD.5 Our data seem to support the findings of the AtheroGene study in showing that plasma fibrinogen is independently associated with the increased risk for mortality and that fibrinogen did not improve the discriminatory power of multivariate models for the prediction of mortality. Although the exact reasons for the failure of fibrinogen to provide prognostic information on top of that provided by traditional cardiovascular risk factors remain unknown, 2 putative explanations may be offered. First, close associations of fibrinogen with an array of cardiovascular risk factors have been described.16 Consequently, elevated fibrinogen level may be considered an associate of increased cardiovascular risk. Second, in contrast to many previous studies, in the present study, statins (known for their strong anti-inflammatory actions) were used in >90% of the patients. These agents through their anti-inflammatory actions may attenuate the association between fibrinogen and cardiovascular events. To our knowledge, the finding of a J-shaped relation between fibrinogen level and mortality is novel. The finding may be important because it suggests the most optimal fibrinogen levels associated with lowest mortality. Although multiple mechanisms may explain increased coronary risk in patients with elevated fibrinogen levels, reasons for increased risk for death in patients with low fibrinogen levels remain unknown. Factors or measures that reduce fibrinogen level are not considered to increase cardiovascular risk.13,14 Alternatively, low fibrinogen levels may reflect hidden co-morbidities (such as hepatic diseases) that might have remained unaccounted for. Subgroup analyses showed that fibrinogen predicted an increased risk for cardiac and all-cause mortality across all subsets of patients with CAD. Of note, a fibrinogen-by-gender interaction showing a stronger association between fibrinogen and cardiac mortality in men than in women and a fibrinogenebyearterial hypertension interaction revealing a stronger association between fibrinogen and all-cause mortality in patients with hypertension than in patients without arterial hypertension were observed. A recent study has reported an interaction between gender and fibrinogen (or C-reactive protein), suggesting that these inflammatory markers improve cardiovascular risk discrimination in men but not in women.2 In the Cardiovascular Health Study (CHS), which included subjects aged 65 years, fibrinogen predicted mortality in men but not in women.17 Although we found a significantly stronger association between fibrinogen and cardiac mortality in men than in women and a stronger association between fibrinogen and mortality in patients with hypertension than in patients without arterial hypertension, no good explanation for these findings can currently be offered. Our study had limitations. First, the follow-up period of 1 year might be short. However, at 1 year, time-to-event
810
The American Journal of Cardiology (www.ajconline.org)
curves were still expanding, which may suggest a further accentuation of the differences in survival over a longer follow-up period. Second, when analyzing the results of subgroup analyses, caution is warranted because of possible errors introduced by multiple testing. Third, these findings belong to patients confirmed to have CAD and may not be extrapolated to subjects in the general population. Disclosures The authors have no conflicts of interest to disclose. 1. Danesh J, Lewington S, Thompson SG, Lowe GD, Collins R, Kostis JB, Wilson AC, Folsom AR, Wu K, Benderly M, Goldbourt U, Willeit J, Kiechl S, Yarnell JW, Sweetnam PM, Elwood PC, Cushman M, Psaty BM, Tracy RP, Tybjaerg-Hansen A, Haverkate F, de Maat MP, Fowkes FG, Lee AJ, Smith FB, Salomaa V, Harald K, Rasi R, Vahtera E, Jousilahti P, Pekkanen J, D’Agostino R, Kannel WB, Wilson PW, Tofler G, Arocha-Pinango CL, Rodriguez-Larralde A, Nagy E, Mijares M, Espinosa R, Rodriquez-Roa E, Ryder E, Diez-Ewald MP, Campos G, Fernandez V, Torres E, Marchioli R, Valagussa F, Rosengren A, Wilhelmsen L, Lappas G, Eriksson H, Cremer P, Nagel D, Curb JD, Rodriguez B, Yano K, Salonen JT, Nyyssonen K, Tuomainen TP, Hedblad B, Lind P, Loewel H, Koenig W, Meade TW, Cooper JA, De Stavola B, Knottenbelt C, Miller GJ, Bauer KA, Rosenberg RD, Sato S, Kitamura A, Naito Y, Palosuo T, Ducimetiere P, Amouyel P, Arveiler D, Evans AE, Ferrieres J, Juhan-Vague I, Bingham A, Schulte H, Assmann G, Cantin B, Lamarche B, Despres JP, Dagenais GR, Tunstall-Pedoe H, Woodward M, Ben-Shlomo Y, Davey Smith G, Palmieri V, Yeh JL, Rudnicka A, Ridker P, Rodeghiero F, Tosetto A, Shepherd J, Ford I, Robertson M, Brunner E, Shipley M, Feskens EJ, Kromhout D, Dickinson A, Ireland B, Juzwishin K, Kaptoge S, Memon A, Sarwar N, Walker M, Wheeler J, White I, Wood A. Plasma fibrinogen level and the risk of major cardiovascular diseases and nonvascular mortality: an individual participant meta-analysis. JAMA 2005;294:1799e1809. 2. Kaptoge S, Di Angelantonio E, Pennells L, Wood AM, White IR, Gao P, Walker M, Thompson A, Sarwar N, Caslake M, Butterworth AS, Amouyel P, Assmann G, Bakker SJ, Barr EL, Barrett-Connor E, Benjamin EJ, Bjorkelund C, Brenner H, Brunner E, Clarke R, Cooper JA, Cremer P, Cushman M, Dagenais GR, D’Agostino RB Sr, Dankner R, Davey-Smith G, Deeg D, Dekker JM, Engstrom G, Folsom AR, Fowkes FG, Gallacher J, Gaziano JM, Giampaoli S, Gillum RF, Hofman A, Howard BV, Ingelsson E, Iso H, Jorgensen T, Kiechl S, Kitamura A, Kiyohara Y, Koenig W, Kromhout D, Kuller LH, Lawlor DA, Meade TW, Nissinen A, Nordestgaard BG, Onat A, Panagiotakos DB, Psaty BM, Rodriguez B, Rosengren A, Salomaa V, Kauhanen J, Salonen JT, Shaffer JA, Shea S, Ford I, Stehouwer CD, Strandberg TE, Tipping RW, Tosetto A, Wassertheil-Smoller S, Wennberg P, Westendorp RG, Whincup PH, Wilhelmsen L, Woodward M, Lowe GD, Wareham NJ, Khaw KT, Sattar N, Packard CJ, Gudnason V, Ridker PM, Pepys MB, Thompson SG, Danesh J. C-reactive protein, fibrinogen, and cardiovascular disease prediction. N Engl J Med 2012;367:1310e1320.
3. Luc G, Bard JM, Juhan-Vague I, Ferrieres J, Evans A, Amouyel P, Arveiler D, Fruchart JC, Ducimetiere P. C-reactive protein, interleukin6, and fibrinogen as predictors of coronary heart disease: the PRIME study. Arterioscler Thromb Vasc Biol 2003;23:1255e1261. 4. Sinning JM, Bickel C, Messow CM, Schnabel R, Lubos E, Rupprecht HJ, Espinola-Klein C, Lackner KJ, Tiret L, Munzel T, Blankenberg S. Impact of C-reactive protein and fibrinogen on cardiovascular prognosis in patients with stable angina pectoris: the AtheroGene study. Eur Heart J 2006;27:2962e2968. 5. Keavney B, Danesh J, Parish S, Palmer A, Clark S, Youngman L, Delepine M, Lathrop M, Peto R, Collins R. Fibrinogen and coronary heart disease: test of causality by “Mendelian randomization.” Int J Epidemiol 2006;35:935e943. 6. Myers GL, Christenson RH, Cushman M, Ballantyne CM, Cooper GR, Pfeiffer CM, Grundy SM, Labarthe DR, Levy D, Rifai N, Wilson PW. National Academy of Clinical Biochemistry laboratory medicine practice guidelines: emerging biomarkers for primary prevention of cardiovascular disease. Clin Chem 2009;55:378e384. 7. Ndrepepa G, Braun S, Haase HU, Schulz S, Ranftl S, Hadamitzky M, Mehilli J, Schomig A, Kastrati A. Prognostic value of uric acid in patients with acute coronary syndromes. Am J Cardiol 2012;109: 1260e1265. 8. Sandler H, Dodge HT. The use of single plane angiocardiograms for the calculation of left ventricular volume in man. Am Heart J 1968;75: 325e334. 9. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976;16:31e41. 10. Cutlip DE, Windecker S, Mehran R, Boam A, Cohen DJ, van Es GA, Steg PG, Morel MA, Mauri L, Vranckx P, McFadden E, Lansky A, Hamon M, Krucoff MW, Serruys PW. Clinical end points in coronary stent trials: a case for standardized definitions. Circulation 2007;115: 2344e2351. 11. Pencina MJ, D’Agostino RB Sr, D’Agostino RB Jr, Vasan RS. Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyond. Stat Med 2008;27:157e172. 12. Losner S, Volk BW, Wilensky ND. Fibrinogen concentration in acute myocardial infarction; comparison of the clot density determination of fibrinogen with the erythrocyte sedimentation rate. AMA Arch Intern Med 1954;93:231e245. 13. Kamath S, Lip GY. Fibrinogen: biochemistry, epidemiology and determinants. QJM 2003;96:711e729. 14. Koenig W. Fibrin(ogen) in cardiovascular disease: an update. Thromb Haemost 2003;89:601e609. 15. Empana JP, Jouven X, Canoui-Poitrine F, Luc G, Tafflet M, Haas B, Arveiler D, Ferrieres J, Ruidavets JB, Montaye M, Yarnell J, Morange P, Kee F, Evans A, Amouyel P, Ducimetiere P. C-reactive protein, interleukin 6, fibrinogen and risk of sudden death in European middleaged men: the PRIME study. Arterioscler Thromb Vasc Biol 2010;30: 2047e2052. 16. Folsom AR. Fibrinogen and cardiovascular risk markers. Blood Coagul Fibrinolysis 1999;10(suppl):S13eS16. 17. Tracy RP, Arnold AM, Ettinger W, Fried L, Meilahn E, Savage P. The relationship of fibrinogen and factors VII and VIII to incident cardiovascular disease and death in the elderly: results from the cardiovascular health study. Arterioscler Thromb Vasc Biol 1999;19: 1776e1783.