a b s t r a c t

Background and purpose: Fibrinogen plays an important role in hemostasis and thrombosis and is proven to have prognostic significance in patients with cardiovascular disease. We examined the utility of fibrinogen as a prog- nostic indicator for patients with type A Acute aortic dissection .

Methods: This study was performed in consecutive patients with type A AAD admitted to our hospital within 24 hours after onset of symptoms. Fibrinogen levels were measured on admission. Baseline clinical characteristics and Laboratory test results were collected. The endpoint was in-hospital mortality.

Results: A total of 143 patients with type A AAD were enrolled. Compared with the survivors, the nonsurvivors had significant lower fibrinogen levels (1.95(1.37, 2.38) vs. 2.37(1.85, 3.15) g/L, p = 0.001). The cutoff level of fibrinogen determined by ROC curve analysis was 2.17 g/L, with a sensitivity, specificity of 71.9%, 60.4% respec- tively, and the area under the ROC curve was 0.686 (95% CI, 0.585-0.768; p = 0.001). After controlling for poten- tially relevant confounding variables, we found an admission fibrinogen level less than 2.17 g/L was associated with an increased risk of in-hospital mortality (odds ratio, 5.527; 95% CI, 1.660-18.401; p = 0.005) compared with those with fibrinogen greater than 2.17 g/L.

Conclusion: Low fibrinogen level on admission is an independent predictor of in-hospital mortality in patients with type A AAD.

(C) 2017

1. Introduction

As a key component of blood clots, fibrinogen plays an important role in hemostasis and thrombosis [1]. Activation of coagulation and the fibrinolytic system are frequently observed in cardiovascular dis- eases. Hence, fibrinogen has been an active focus of investigation in cor- onary artery disease, stroke and peripheral artery disease. Prior studies suggest that fibrinogen mediates leukocyte migration and activates cy- tokine release [2], resulting in the initiation and growth of atheroscle- rotic plaque [3]. Data from clinical studies have shown that fibrinogen level is associated with adverse prognosis and the occurrence of major cardiovascular events in patients with cardiovascular disease [4,5].

Although Coagulation markers such as D-dimer have been shown to predict adverse prognosis among patients with Acute aortic dissection , no study has examined the prognostic significance of fibrinogen in AAD. The primary goal of this study was to investigate whether fibrin- ogen is an independent predictor of survival in type A AAD.

* Corresponding author.

E-mail address: [email protected] (J. Liu).

1. Method
   1. Patients

Patients with type A AAD consecutively admitted to the First Affiliat- ed Hospital of Wenzhou Medical University between January 2006 and January 2017 were enrolled in the present study. The study was reviewed and approved by the ethics committee of the First Affiliated Hospital of Wenzhou Medical University and was waived from in- formed consent due to its retrospective nature.

The diagnosis of type A ADD was confirmed by multidetector com- puted tomography scan. The main inclusion criteria were type A AAD within 24 h after symptom onset and age between 18 and 85 years. Ex- clusion criteria included chronic liver and renal disease, hematologic disorders, malignant tumor, infectious disease, Marfan syndrome and prior history of aortic dissection (AD).

Data collection

Clinical information of enrolled patients was obtained through re- view of medical records. The demographic profile, medical history, vital signs, laboratory results and clinical outcomes were recorded.

https://doi.org/10.1016/j.ajem.2017.10.001 0735-6757/(C) 2017

742 J. Liu et al. / American Journal of Emergency Medicine 36 (2018) 741-744

Endpoint

The study endpoint was defined as all-cause mortality during hospitalization.

Statistical analysis

Distribution of continuous data was assessed using the Shapiro- Wilk test. Continuous variables are presented as the mean +- SD or me- dian and interquartile range according to their distributions. Categorical variables are presented as percentage. Comparisons between groups were performed with unpaired Student’s t-tests for normally distribut- ed continuous variables and Wilcoxon Mann-Whitney tests for no- normally distributed continuous variables. Categorical variables were compared by chi-square tests or Fisher’s exact test when appropriate. Receiver operating characteristic (ROC) analysis was performed to de- termine the cut-off value for fibrinogen in predicting in-hospital mortal- ity with high sensitivity and specificity. Univariate analysis and multiple logistic regression analysis were used to identify the factors related to in-hospital mortality. In order to exclude the impact of age and open surgery on the mortality rates in patients with type A AAD, we stratified the patients by age in subgroup analyses and analyzed the prognostic effects of fibrinogen in patients receiving open surgery. A p value b 0.05 was considered to be statistically significant. All the statistical analyses were performed using SPSS statistical software, version 19.0 (SPSS Inc., Chicago, Illinois).

1. Results
   1. Baseline patients characteristics

In the present study, 125 patients met the enrollment criteria and were included. Clinical features of survivors and non-survivors are sum- marized in Table 1. Compared with the survivors, the non-survivors were older (61(48, 65) vs. 48(41, 58), p = 0.001), had lower fibrinogen levels (1.95(1.37, 2.38) vs. 2.37(1.85, 3.15) g/L, p = 0.001). Besides, the

Table 1

Baseline characteristics between the survivor and non-survivor

Variable	All patients		Survivor		Non-survivor	p
	(n = 143)		(n = 111)		(n = 32)	value
Age (years)	50(43, 62)		48(41, 58)		61(48, 65)	0.001
Male, n(%)	103(72.0)		80(72.1)		23(71.9)	1.000
Hypertension, n(%)	97(67.8)		75(67.6)		22(68.8)	0.900
Coronary artery	2(1.4)		2(1.8)		0(0)	1.000
disease, n(%)
Dyslipidemia, n(%)	2(1.4)		1(0.9)		1(3.1)	0.399
DM, n(%)	10(7.0)		7(6.3)		3(9.4)	0.549
Smoking, n(%)	48(33.6)		36(32.4)		12(37.5)	0.593
Alcohol use, n(%)	27(18.9)		21(18.9)		6(18.8)	1.000
SBP, mm Hg	131 +- 28		133 +- 27		123 +- 30	0.080
DBP, mm Hg	68 +- 18		70 +- 18		69 +- 16	0.708
Heart rate, bpm	80 +- 17		80 +- 16		83 +- 21	0.284
White blood cell,	14.53(11.53,		14.30(11.42,		15.05(12.93,	0.078
x109/L	17.28)		16.88)		19.70)
Neutrophil, x109/L	12.65 +- 3.92		12.35 +- 3.85		13.69 +- 4.06	0.089
Monocyte, x109/L	0.67(0.50,		0.60(0.50,		0.79(0.51,	0.082
	0.90)		0.90)		1.10)
Lymphocyte, x109/L	1.00(0.70,		1.00(0.70,		1.00(0.70,	0.707
	1.35)		1.30)		1.45)
Hemoglobin, g/L	131 +- 16		131 +- 17		129 +- 12	0.552
Platelet, x109/L	162 +- 46		165 +- 45		152 +- 47	0.140
Creatinine, mmol/L	91(68,123)		87(68,119)		111(71,134)	0.105
Fibrinogen, g/L	2.24(1.77,		2.37(1.85,		1.95(1.37,	0.001
	2.94)		3.15)		2.38)
OS	122(85.3)		106(95.5)		16(50.0)	b0.001

Data are median (interquartile range), n(%), or mean +- SD. The bold values indicate statistical significance.

CAD, Coronary artery disease; DM, diabetes mellitus; DBP, diastolic blood pressure; OS, open surgery; SBP, systolic blood pressure; WBC, White blood cell.

non-survivors also had a lower rate of open surgery (50.0% vs. 95.5%, p b 0.001). No significant differences between non-survivors and survivors in gender, hypertension, coronary artery disease, dyslipidemia, diabetes mellitus, smoking, alcohol use, SBP, DBP, heart rate, white blood cell, neutrophil, monocyte and lymphocyte count, hemoglobin, platelet count, as well as creatinine were found.

ROC analysis

The accuracy of fibrinogen to predict in-hospital mortality according to receiver operating characteristics (ROC) is given in Fig. 1. The area under the ROC curve for fibrinogen was 0.686 (optimal cutoff point b 2.17 g/L; sensitivity 71.9%; specificity 60.4%, p = 0.001. Table 2). The patients were classified into two groups according to the best cut-off value of fibrinogen. Among the 67 patients in the low fibrinogen group, 23 (34.3%) died during hospitalization. By contrast, 9 out of 76 patients (11.8%) in the high fibrinogen group died in hospital (Fig. 2). The Chi Square test showed significant difference in mortality between the two groups (p = 0.002).

Fibrinogen for predicting in-hospital mortality

Variables with a P value less than or equal to 0.1 in univariable anal- ysis were selected for logistic regression and the results are shown in Table 3. In univariate regression, low fibrinogen (b 2.17 g/L) was associ- ated with higher in-hospital mortality (crude odds ratio (OR) 3.891; 95%CI, 1.648-9.190; p = 0.002). Besides, age (per one year increase, crude OR 1.064; 95% CI, 1.026-1.104; p = 0.001) were also associated with higher in-hospital mortality. Open surgery (crude OR 0.047; 95% CI, 0.015-0.147; p b 0.001) showed protective effects for patients with type A AAD. In multivariate regression, low fibrinogen was indepen- dently associated with higher in-hospital mortality (adjusted OR 5.527; 95% CI, 1.660-18.401; p = 0.005). As shown in Table 3, age (ad- justed OR 1.067; 95% CI, 1.012-1.125; p = 0.017) and open surgery (ad- justed OR 0.033; 95% CI, 0.007-0.150; p b 0.001) were other independent predictors for outcome in multivariate regression analysis. Given that old age and open surgery have substantial impact on the mortality rates in patients with type A AD, subgroup analyses were per- formed in patients who were >= 65 years old and those receiving open

Fig. 1. Receiver operating characteristic curve of fibrinogen for predicting in-hospital mortality in patients with type A acute aortic dissection.

J. Liu et al. / American Journal of Emergency Medicine 36 (2018) 741-744 743

Table 2

Diagnostic value of fibrinogen for in-hospital mortality.

Variable AUC Cut-off	SE	95% CI	p	Sensitivity	Specificity		Variable	Univariable			Multivariable
value			value					OR 95% CI	p		OR 95% CI	p
Fibrinogen 0.686 2.17	0.052	0.585-0.787	0.001	0.719	0.604				value			value
							Age	1.064 1.026-1.104	0.001		1.067 1.012-1.125	0.017
SBP, mm Hg								0.987	0.972-1.002 0.083	0.979		0.957-1.001 0.065
surgery. In patients aged >=65 years, low fibrinogen was associated with						White blood cell,		1.080	0.984-1.185 0.104	1.011		0.728-1.403 0.948
high risk of in-hospital mortality (OR 6.798, 95% CI 1.308-35.320, p =						x109/L

Table 3

Predictors of in-hospital mortality by logistic regression

0.023). As for the surgery subgroup, low fibrinogen was independently predicted in-hospital mortality (OR6.335, 95% CI 1.426-28.148, p = 0.015) (Fig. 3).

1. Discussion

The current study demonstrated that fibrinogen is a powerful pre- dictor of mortality in patients with type A AAD. Those who had low fi- brinogen level on admission within 24 h after symptom onset had high in-hospital mortality. The predictive value of fibrinogen level was independent of surgery treatment and age. These findings suggest that fibrinogen may serve as a new clinical marker for risk stratification for type A AAD.

There may be several possible interpretations as to why low fibrino- gen was associated in-hospital mortality. After Vascular injury in AAD, plasma factor (f) VIIa gains access to tissue factor, leading to activation of the coagulation cascade [6]. During blood clotting formation process, fibrinogen is cleaved by thrombin to form fibrin, resulting in consump- tion of fibrinogen. Hence, AAD is accompanied by thrombus formation and excessive fibrinolysis. Paparella et al. evaluated the biomarkers that were specific for thrombin generation (prothrombin fragment 1.2, F1.2), and fibrinolysis activation (plasmin-antiplasmin complex (PAP)) in patients with AAD within 32 h from the onset of symptoms. They found that both F1.2 and PAP were significantly increased among these patients, suggesting intense activation of the coagulation and fi- brinolytic system [7]. Besides, numerous studies have demonstrated an increased level of D-dimer and Fibrin degradation products (FDP) in AAD, providing clear evidence for hemostasis system activation [8-10]. Moreover, previous studies indicated that the amount of coagu- lation and fibrinolytic activation was proportionally related to the ana- tomic extent of the dissection [8,11,12]. Ohlmann et al. observed that D-dimer levels were correlated with the number of segment of dissect- ed aorta and tended to be higher in DeBakey I than in DeBakey II and DeBakey III [8]. Taking together, intimal tearing in aortic dissection causes subendothelial elements exposure, which leading to tissue factor release and initiates coagulation cascade. This process consumes large amounts of fibrinogen and results in decreased level of the later in bloodstream. So lower fibrinogen level would indicate greater extent in- jury of aorta and suggests poorer prognosis.

Neutrophil, x109/L	1.091	0.986-1.207	0.092	1.020	0.731-1.422	0.909
Monocyte, x109/L	2.619	0.869-7.894	0.087	2.457	0.423-14.266	0.316
Lower fibrinogen	3.891	1.648-9.190	0.002	5.527	1.660-18.401	0.005
OS	0.047	0.015-0.147	b 0.001	0.033	0.007-0.150	b0.001

Abbreviations as in Table 1. The bold values indicate statistical significance.

Besides, low fibrinogen may be a surrogate marker for consumption coagulopathy in patients with dissecting aorta. In 1967, Fine et al. first reported a patient with a Dissecting aneurysm of the aorta accompanied by disseminated intravascular coagulation [13]. In this case, de- creased fibrinogen, platelets, factor II, factor V, and Factor VIII were ob- served. Since then, a number of clinical case studies have provided evidence to substantiate the association between low fibrinogen and co- agulopathy in aortic dissection [14-17]. This aortic dissection-induced coagulopathy is associated with increased Transfusion requirements, risk of complications, and mortality [7].

Considering the prognostic value of fibrinogen level in patients with type A AAD, we suggested that fibrinogen should be evaluated immedi- ately after admission and used as a biomarker in triage of those patients. Large prospective cohort studies should be performed to explore the role of fibrinogen in guiding type A AAD therapy in emergency setting.

Limitations

The present study had several limitations. First, D-dimer is a proven prognostic factor for AAD and not included in our data analysis. In our hospital, the upper limit of determination for D-dimer was 20 g/L and more than half of the cases had a value exceeding the upper determina- tion limit. So we decided to exclude D-dimer analysis in this study. Sec- ond, we only evaluated the relationship between fibrinogen and in- hospital mortality and long term impact of fibrinogen in patients with AAD is unknown. Third, the rates of CAD and dyslipidemia were very low in our study, so this limits extrapolation to patients with high rates of CAD and dyslipidemia. Fourth, since patients included were of Chinese Han descent, results from this study should be validated in other ethnic background. Finally, the sample size in our study was rela- tively small, larger sample size studies are needed to confirm our findings.

Fig. 2. Distribution of the mortality rate according to fibrinogen categories.

Declaration of interest“>744 J. Liu et al. / American Journal of Emergency Medicine 36 (2018) 741-744

Fig. 3. Odds ratios of fibrinogen for predicting in-hospital mortality in subgroup analysis.

1. Conclusions

Low fibrinogen on admission is an independent predictor of in- hospital mortality in patients with type A AAD. Our results seem to sug- gest that the preservation of the Hemostatic system should be one of the objectives in the treatment of AAD.

Sources of grants

The Natural Science Foundation of Zhejiang Province (LQ17H02005).

Declaration of interest

The authors declare that they have no conflict of interest.

Acknowledgment

This study was supported by grant from the Natural Science Founda- tion of Zhejiang Province (LQ17H020005).

Reference

1. Kattula S, Byrnes JR, Wolberg AS. Fibrinogen and fibrin in hemostasis and thrombo- sis. Arterioscler Thromb Vasc Biol 2017;37:e13-21.
2. Xiao Q, Danton MJ, Witte DP, et al. Fibrinogen deficiency is compatible with the de- velopment of atherosclerosis in mice. J Clin Invest 1998;101:1184-94.
3. Lou XJ, Boonmark NW, Horrigan FT, et al. Fibrinogen deficiency reduces vascular ac- cumulation of apolipoprotein(a) and development of atherosclerosis in apolipoprotein(a) transgenic mice. Proc Natl Acad Sci U S A 1998;95:12591-5.
4. Peng Y, Wang H, Li YM, et al. Relation between admission plasma fibrinogen levels and mortality in Chinese patients with coronary artery disease. Sci Rep 2016; 6:30506.
5. Wannamethee SG, Whincup PH, Papacosta O, et al. Associations between blood co- agulation markers, NT-proBNP and risk of incident heart failure in older men: the British Regional Heart Study. Int J Cardiol 2017;230:567-71.
6. Sorensen B, Larsen OH, Rea CJ, et al. Fibrinogen as a hemostatic agent. Semin Thromb

Hemost 2012;38:268-73.

1. Paparella D, Rotunno C, Guida P, et al. Hemostasis alterations in patients with acute aortic dissection. Ann Thorac Surg 2011;91:1364-9.
2. Ohlmann P, Faure A, Morel O, et al. Diagnostic and prognostic value of circulating D- dimers in patients with acute aortic dissection. Crit Care Med 2006;34:1358-64.
3. Tian L, Fan X, Zhu J, et al. Plasma D-dimer and in-hospital mortality in patients with Stanford type A acute aortic dissection. Blood Coagul Fibrinolysis 2014;25:161-6.
4. Wen D, Du X, Dong JZ, et al. Value of D-dimer and C reactive protein in predicting inhospital death in acute aortic dissection. Heart (British Cardiac Society) 2013;99: 1192-7.
5. Weber T, Hogler S, Auer J, et al. D-dimer in acute aortic dissection. Chest 2003;123: 1375-8.
6. Jo Y, Anzai T, Ueno K, et al. Re-elevation of D-dimer as a predictor of re-dissection and venous thromboembolism after Stanford type B acute aortic dissection. Heart Vessels 2010;25:509-14.
7. Fine NL, Applebaum J, Elguezabal A, et al. Multiple coagulation defects in association with dissecting aneurysm. Arch Intern Med 1967;119:522-6.
8. Bieger R, Vreeken J, Stibbe J, et al. Arterial aneurysm as a cause of consumption co- agulopathy. N Engl J Med 1971;285:152-4.
9. ten Cate JW, Timmers H, Becker AE. Coagulopathy in ruptured or dissecting aortic aneurysms. Am J Med 1975;59:171-6.
10. Nakajima T, Kin H, Minagawa Y, et al. Coagulopathy associated with residual dissec- tion after surgical treatment of type A aortic dissection. J Vasc Surg 1997;26:609-15.
11. Naser MNM, Chong BP, Karim NKA. thoracic aortic dissection mimicking primary hematological disorder: a case report. Case reports in internal medicine, 3. ; 2016.

p. 10.