Calibrated automated thrombography for monitoring coagulation function in patients with intracerebral haemorrhage

Abstract

Objective

To monitor coagulation function in patients with intracerebral haemorrhage (ICH) using calibrated automated thrombography.

Methods

Patients admitted to hospital with ICH (confirmed within 18 h of symptom onset) were enrolled. Patient history and blood samples were obtained within 6 h of admission; further blood samples were collected on days 4, 8 and 15 (or on discharge between days 9–15: grouped with day 15 data). Blood samples were also collected from age- and sex-matched healthy controls. All samples underwent calibrated automated thrombography.

Results

At admission, thrombin lag time and time to peak was longer, and endogenous thrombin potential and peak height were lower, in patients with ICH (n = 20) than in healthy controls (n = 29). Lag time in patients with ICH gradually decreased, but remained significantly longer than in controls until day 8. Time to peak also gradually decreased, but remained longer in patients than in controls by day 15. Endogenous thrombin potential and peak height gradually increased in patients, but remained lower than in controls on day 15.

Conclusions

Patients with ICH have poorer coagulation function than healthy individuals, but this function gradually recovers during hospitalization.

Keywords

Intracerebral haemorrhage thrombin calibrated automated thrombogram

Introduction

Intracerebral haemorrhage (ICH) is a common type of stroke.^1–3 In patients with ICH, coagulation, anticoagulation and fibrinolysis are disrupted, which triggers a series of reactions to restore balance in the body.⁴ With increase in bleeding and exacerbation of ICH, prothrombin time, activated partial thromboplastin time and thrombin time all increase.⁵ Conventional coagulation tests (e.g. prothrombin time and activated partial thromboplastin time) evaluate the initial stage of clot formation, which only reflects a small part of the coagulation process, as >95% of thrombin is formed after the initial stage.⁶ Calibrated automated thrombography is the only test that can evaluate the entire coagulation process.⁷ In calibrated automated thrombography, the area under the thrombin-generation curve, which represents endogenous thrombin potential,⁸ is calculated, and this test can monitor thrombin generation in real time.⁹

In the present study, the course of coagulation function in patients with ICH was investigated using calibrated automated thrombography.

Patients and methods

Study population

Patients with spontaneous ICH who were treated at the Department of Neurology, Mancheng County Hospital, China between February 2014 and May 2014, were sequentially enrolled into the present study. Inclusion criteria comprised spontaneous ICH confirmed by computed tomography within 18 h of symptom onset, no history of coagulation disorders and no history of taking any drugs affecting coagulation functions within the previous week. One or two age- (±11 years) and sex-matched healthy control subjects recruited from the local population (Beijing or HeBei) were enrolled per patient included in the study.

The study was approved by the ethics committee of Mancheng County Hospital. All patients, or their legal proxies, and all control subjects provided written informed consent before study enrolment.

Data and sample collection

A detailed history was obtained from patients or their legal proxies, and general clinical data were recorded within 6 h of hospital admission. In all patients, blood samples (5 ml) were collected from the median cubital vein into tubes containing 129 mM trisodium citrate (to give a 1:9 [v/v] final ratio), within 6 h of hospital admission and on days 4, 8 and 15 (or on discharge between days 9–15: grouped with day 15 data). Patients’ neurological scores (comprising Modified Rankin Scale; National Institute of Health Stroke Scale; Glasgow Outcome Scale; and Glasgow Coma Scale) were also collected at these timepoints. Blood samples were collected from the control subjects.

The calibrated automated thrombography measurements were performed as described previously.¹⁰ Briefly, all blood samples were centrifuged twice at 2500 g for 15 min at room temperature to obtain platelet-free plasma supernatant, which was analysed within 2 h, or stored at –80℃ if not tested immediately. For every sample, 80 µl of the platelet-free plasma was transferred into each of three wells of a 96-well microplate (Immulon 2HB; Dynex Technologies, Chantilly, VA, USA), then 20 µl of a mixture of tissue factor (5 pmol/l) and phospholipids (4.0 µmol/l PPP reagent; Stago, Asnières sur Seine, France) were added to each well. For every sample, 80 µl of platelet-free plasma was transferred into a further three wells and 20 µl of thrombin calibrator, France) was added instead of tissue factor. Microplates were inserted into a Fluoroskan Ascent™ FL plate reader (Thermolab Systems, Helsinki, Finland), and preheated at 37℃ for 5 min. Experiments were initiated by automatically adding 20 μl of fluorescent substrate and fluorescent buffer (FLUCA KIT; Stago). After plates were shaken for 10 s, the fluorescence generated was measured in real time. Fluorescence intensity readings were used to produce thrombin generation curves with Thrombinoscope™ software, version 3.0.0.29 (Thrombinoscope BV, Maastricht, The Netherlands).

Calibrated automated thrombography

Calibrated automated thrombography included the following parameters for analyses: (1) lag time, time from the beginning of the reaction to the beginning of thrombin generation; (2) peak height, maximum amount of thrombin; (3) time to peak, time from the beginning of the reaction to peak height; (4) endogenous thrombin potential, area under the thrombin-generation curve, reflecting the total amount of thrombin generated during the test.

Statistical analyses

Measurement data are presented as mean ± SD and all data were processed using SPSS®, version 13.0 (SPSS Inc., Chicago, IL, USA). Differences in haematological indices between the patient group (at admission) and the control group were analysed using Student’s t-test. Continuous variables were analysed for normality and homogeneity of variance using Kolmorogov–Smirnov test. Analysis of variance (ANOVA) was used to compare the means of multiple data groups that were normally distributed. Fisher’s least significant difference test was then used for one–one comparison of the data, to analyse which groups were different from which. A value of P < 0.05 was considered statistically significant.

Results

A total of 20 patients with ICH and 29 healthy controls were included in the present study. For all patients included, ICH was confirmed within 5 h of symptom onset. The mean age of the patient group was 59.1 ± 9.4 years (12 male patients, mean age, 60.2 ± 8.8 years and eight female patients (mean age, 57.4 ± 10.6 years). The control group comprised 13 male subjects (mean age, 57.8 ± 14.2 years) and 16 female subjects (mean age, 60.8 ± 10.5 years). There were no statistically significant between-group differences in terms of age or sex. All 20 patients with ICH had a history of hypertension and three of the 20 patients had a history of coronary heart disease. Haematological indices of the ICH and control groups were found to be within normal parameters (Table 1). The neurological scores of patients with ICH, at admission to hospital, are shown in Table 2, and the trends in neurological scores in patients with ICH at days 4, 8 and 15 (or discharge between days 9–15: grouped with day 15 data) are shown in Figure 1.

Figure 1.

Trends in neurological scores in patients with intracerebral haemorrhage (mean age, 59.1 ± 9.4 years) at days 4, 8 and 15 (or on discharge between days 9–15: grouped with day 15 data). mRS, Modified Rankin Scale; NIHSS, National Institute of Health Stroke Scale; GOS, Glasgow Outcome Scale; GCS, Glasgow Coma Scale.

Table 1.

Haematological indices and haematoma volume in patients with intracerebral haemorrhage (mean age, 59.1 ± 9.4 years) at admission to hospital, and haematological indices in age- and sex-matched healthy controls.

	Parameter
Study group	FIB, g/l	INR	APTT, S	WBC, 10⁹/l	PLT, 10⁹/l	GLU, mmol/l	Cr, µmol/l	TG, mmol/l	Haematoma volume, ml
Patients with ICH, n = 20	2.5 ± 0.7	1.6 ± 2.6	25.5 ± 3.8	6.9 ± 2.1	206.5 ± 52.0	7.1 ± 1.5	62.0 ± 18.5	1.7 ± 1.5	12.3 ± 7.8
Control subjects, n = 29	3.1 ± 0.4	1.0 ± 0.1	29.0 ± 1.8	7.0 ± 0.9	243.7 ± 24.2	5.8 ± 0.5	54.6 ± 6.9	1.4 ± 0.5	——
Statistical significance	P = 0.002	P = 0.262	P = 0.001	NS	P = 0.006	P = 0.001	NS	NS	——

Data presented as mean ± SD.

ICH, Intracerebral haemorrhage; FIB, fibrinogen; INR, international normalized ratio; APTT, activated partial thromboplastin time; WBC, white blood cell; PLT, platelet; GLU, glucose; Cr, creatinine; TG, triglyceride.

NS, no statistically significant between-group difference (P > 0.05, Student’s t-test).

Table 2.

Neurological scores of 20 patients (mean age, 59.1 ± 9.4 years) with intracerebral haemorrhage (ICH) at admission to hospital.

Neurological scale
mRS	NIHSS	ICH score	GOS	GCS
4.0 ± 0.6	8.9 ± 4.6	0.4 ± 0.5	3.1 ± 0.2	13.5 ± 1.0

Data presented as mean score ± SD.

mRS, Modified Rankin Scale; NIHSS, National Institute of Health Stroke Scale; GOS, Glasgow Outcome Scale; GCS, Glasgow Coma Scale.

Thrombin lag time

In terms of thrombin lag time, significant differences were found between patients with ICH during hospitalization and healthy controls (P < 0.05; one-way ANOVA). One–one analyses of the different time groups using Fisher’s least significant difference test showed that thrombin lag time in the control group significantly differed from the ICH group on day 1 (admission), and days 4 and 8 (P < 0.001, P = 0.004 and P = 0.032, respectively), however, there was no significant difference between lag time in the control group and in patients with ICH on day 15 (or discharge between days 9–15: grouped with day 15 data). Fisher’s least significant difference test also showed that in the ICH group, thrombin lag time on day 1 (admission) was significantly longer than on day 15 (or discharge between days 9–15: grouped with day 15 data; P = 0.022). Thrombin lag time in patients with ICH gradually decreased during hospitalization, but remained longer than in healthy controls up to day 8 (P < 0.05, Fisher’s least significant difference test; Table 3).

Table 3.

Calibrated automated thrombography results in patients with intracerebral haemorrhage (ICH; mean age, 59.1 ± 9.4 years), and age- and sex-matched healthy controls.

	Study group
	Control subjects, n = 29	Patients with ICH, n = 20
Parameter		Admission Day 1	Day 4	Day 8	Day 15 (or discharge)	F	Statistical significance
Lag time, min	3.26 ± 0.77	9.17 ± 8.90	7.72 ± 5.55	6.56 ± 4.44	5.34 ± 4.20	4.520	P < 0.05^a
ETP, nmol/l × min	2445.60 ± 529.51	209.88 ± 499.20	757.93 ± 609.93	1016.61 ± 536.88	1570.44 ± 400.94	65.287	P < 0.001^b
Thrombin peak, nmol/l	412.12 ± 56.25	59.21 ± 42.58	131.19 ± 91.36	152.13 ± 92.63	210.92 ± 93.47	77.940	P < 0.001^c
ttPeak, min	5.39 ± 1.06	17.50 ± 8.69	12.45 ± 5.54	11.52 ± 4.70	9.62 ± 4.87	16.374	P < 0.01^d

Data presented as mean ± SD.

Overall comparison of control subjects versus patients with ICH during hospitalization (P < 0.05, one-way ANOVA); one–one comparison of control subjects versus ICH patients at days 1, 4 and 8 (P < 0.05, Fisher’s least significant difference test).

Overall comparison of control subjects versus patients with ICH during hospitalization (P < 0.05, one-way ANOVA); one–one comparison of control subjects versus patients at days 1, 4 and 8 (P < 0.001, Fisher’s least significant difference test).

Overall comparison of control subjects versus patients with ICH during hospitalization (P < 0.05, one-way ANOVA); one–one comparison of control subjects versus patients at days 1, 4, and 15 (P < 0.001, Fisher’s least significant difference test).

Overall comparison of control subjects versus patients with ICH during hospitalization (P < 0.05, one-way ANOVA); one–one comparison of control subjects versus patients at days 1, 4, and 15 (P < 0.01, Fisher’s least significant difference test).

ICH, intracerebral haemorrhage; Lag time, time from beginning of reaction to beginning of thrombin generation; ETP, endogenous thrombin potential, area under the thrombin-generation curve (total amount of thrombin generated during the test); Thrombin peak, peak thrombin generation; ttPeak, time to peak thrombin generation, ANOVA, analysis of variance.

Endogenous thrombin potential

In terms of endogenous thrombin potential, significant differences were found between patients with ICH during hospitalization and healthy controls (P < 0.05; one-way ANOVA). Fisher’s least significant difference test showed that endogenous thrombin potential in the control group was significantly different from the ICH group on days 1, 4 and 8 (all P < 0.001). Within the ICH group, endogenous thrombin potential on day 1 was significantly different from endogenous thrombin potential on days 4, 8 and 15 (or discharge between days 9–15: grouped with day 15 data; P = 0.001, P < 0.001 and P < 0.001, respectively). In addition, endogenous thrombin potential on day 15 (or discharge) was significantly different from day 4 (P < 0.001) and day 8 (P = 0.001). In patients with ICH, endogenous thrombin potential gradually increased during hospitalization, but remained lower than in healthy controls before day 15 (or discharge between days 9–15: grouped with day 15 data; Table 3).

Thrombin peak height

Significant differences were found between patients with ICH during hospitalization and healthy controls in terms of thrombin peak height (P < 0.05; one-way ANOVA). One–one analyses of the different time groups using Fisher’s least significant difference test showed that the peak height in the control group was significantly higher than in the ICH group on days 1, 4 and 8 and 15 (or discharge between days 9–15: grouped with day 15 data; all P < 0.001). Within the ICH group, the peak height on day 1 was significantly different from peak heights on days 4, 8 and 15 (P = 0.004, P < 0.001 and P < 0.001, respectively). In addition, the peak height on day 15 differed from that on days 4 (P = 0.001) and 8 (P = 0.017). Peak height was lower in patients with ICH compared with healthy controls. Peak height gradually increased during hospitalization, but on day 15 (or discharge between days 9–15: grouped with day 15 data), remained lower than the control-group value (Table 3).

Thrombin time to peak

One-way ANOVA revealed significant differences in time to peak values between patients with ICH during hospitalization and controls (P < 0.05). Fisher’s least significant difference test showed that time to peak values in the control group were significantly different from patients with ICH on day 1 (admission) and days 4, 8 and 15 (or discharge between days 9–15: grouped with day 15 data; P < 0.001, P < 0.001, P < 0.001, and P = 0.007, respectively). In patients with ICH, time to peak on day 1 (admission) was significantly different from days 4, 8 and 15 (P = 0.003, P = 0.001 and P < 0.001, respectively). Thrombin time to peak was longer in patients with ICH than in healthy controls. The time to peak value gradually decreased during hospitalization, but on day 15 (or discharge between days 9–15: grouped with day 15 data), remained longer than the control-group value (Table 3).

Discussion

Studies have shown that within 24 h following ICH, there is a compensatory increase in secondary fibrinolysis, and the resulting decrease in coagulation function is a risk factor for adverse outcomes in patients with acute ICH.^4,11

Oedema is known to contribute to poor outcomes following spontaneous ICH,¹² and oedema within 24 h following ICH is significantly correlated with the volume of bleeding and platelet count following ICH.^4,11 Factors released from activated platelets at the site of haemorrhage (such as vascular endothelial growth factor), may interact with thrombin to increase vascular permeability and contribute to the development of oedema.⁴ In a meta-analysis of patients with spontaneous ICH, haematoma growth was an independent contributing factor of mortality and poor outcome following ICH.¹³ One study found that fibrinogen levels, antithrombin III and α2-antiplasmin activity, and platelet counts, were significantly lower in patients with haematoma growth than in those with no growth, and found a particularly high likelihood of haematoma enlargement in patients with low fibrinogen levels.¹⁴ Thus, monitoring coagulation function may be an effective measure for predicting outcomes in patients with ICH.^4,11

Coagulation is initiated by the extrinsic pathway interaction of tissue factor (exposed by vascular injury) with plasma factor VIIa, which, in turn, activates factor IX and factor X, resulting in the formation of small amounts of thrombin. Thrombin then primes the intrinsic pathway, resulting in the rapid generation of thrombin, which acts on fibrinogen to form the fibrin clot. These reactions take place on phospholipid surfaces, usually the activated platelet surface.^15–17 Physiological downregulation of the coagulation and fibrinolytic systems is shown in Figure 2; these processes have been described in detail,¹⁷ with CAT being the only test that can evaluate the entire coagulation process.⁷

Figure 2.

The process of coagulation, evaluated using calibrated automated thrombography (CAT). PTT, partial thromboplastin time; PT, prothrombin time; TF, tissue factor; PF3, platelet factor 3; FDPs, fibrin/fibrinogen degradation products; TFPI, tissue factor pathway inhibitor.

Studies involving the monitoring of coagulation function have traditionally focused on prothrombin time, activated partial thromboplastin time, international normalized ratio, whole blood clotting time and other conventional coagulation indices that evaluate coagulation function on the basis of the time required for fibrin clot formation.^18,19 When <5% of the total thrombin generated during coagulation has been produced, the conversion of fibrinogen to fibrin begins, initiating blood clot formation. This is the endpoint of tests for prothrombin time and activated partial thromboplastin time, and consequently, such tests can only evaluate the beginning of the coagulation process. Over 95% of thrombin is generated after this initial coagulation phase.^6,20 Much information about coagulation function is lost if coagulation time is used as the test index, because most thrombin generation occurs after blood clot formation. Thrombin is central to the mechanisms of haemostasis and thrombosis, in that scores of factors influence thrombin formation, thrombin itself has numerous actions on blood and vessel components, and no pathways bypass thrombin. Thrombin generation, therefore, reflects much – if not all – of the thrombotic–haemostatic functions of blood.⁹ The catalytic effects of thrombin, and thrombin formation ability, determine the capacity of coagulation. Measurement of thrombin generation, in particular, the enzymatic catalysis of thrombin, provides a quantitative method for evaluating coagulation function comprehensively. Many methods for evaluating coagulation function through thrombin detection have been developed.^21–26 Thrombin generated during coagulation is neutralized by antithrombin as soon as thrombin is activated, making direct detection of thrombin very difficult. Specific coagulation markers are generated to activate the process of thrombin generation, and the majority of thrombin-detection methods indirectly measure thrombin levels by detecting intermediate-stage factors in plasma, such as prothrombin fragment 1 + 2, fibrinopeptide A, soluble fibrin monomer complex and thrombin–antithrombin complex.^21–26 The method used in the present study detects thrombin directly through the examination of a fluorescent substrate.⁹

The present study showed that patients with ICH had poor coagulation function at admission to hospital, but that this recovered during hospitalization. By day 15 (or discharge between days 9–15), coagulation function remained inferior in patients with ICH compared with healthy controls, but had almost returned to normal. To the authors’ knowledge, the present study is the first to demonstrate hypocoagulability using calibrated automated thrombography in patients with ICH.

Treatment with recombinant factor VIIa within 4 h following onset of ICH may limit the growth of haematoma and reduce mortality, despite a small increase in the frequency of thromboembolic adverse events.^27–29 The computed tomography angiography spot sign is a predictor of haematoma expansion, and the spot sign is recommended as an entry criterion for future trials of haemostatic therapy in patients with acute ICH.³⁰ Use of calibrated automated thrombography to monitor coagulation function during treatment in patients with ICH may increase in future, due to its advantages over traditional tests such as prothrombin time and activated partial thromboplastin time. To the authors’ knowledge, the present study provides the first published evidence of the efficacy of calibrated automated thrombography in monitoring coagulation function in patients with ICH. Future studies should investigate the parameters of calibrated automated thrombography in patients with ICH and haematoma enlargement.

The present study is limited by the fact that none of the patients developed haematoma enlargement, so the conclusions are only applicable to patients with ICH without haematoma enlargement. In addition, the sample size was relatively small, and future studies with larger sample sizes are required to validate the present results.

Footnotes

Declaration of the conflicting interest

The authors declare that there are no conflicts of interest.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Feigin

Lawes

Bennett

. Worldwide stroke incidence and early case fatality reported in 56 population-based studies: a systematic review. Lancet Neurol 2009; 8: 355–369.

Feigin

Lawes

Bennett

. Stroke epidemiology: a review of population-based studies of incidence, prevalence, and case-fatality in the late 20th century. Lancet Neurol 2003; 2: 43–53.

Qureshi

Mendelow

Hanley

. Intracerebral haemorrhage. Lancet 2009; 373: 1632–1644.

Sansing

Kaznatcheeva

Perkins

. Edema after intracerebral hemorrhage: correlations with coagulation parameters and treatment. J Neurosurg 2003; 98: 985–992.

Cho

Lee

Park

. Relationship between systemic thrombogenic or thrombolytic indices and acute increase of spontaneous intracerebral hemorrhage. J Cerebrovasc Endovasc Neurosurg 2014; 16: 159–165.

Mann

Brummel

Butenas

. What is all that thrombin for? J Thromb Haemost 2003; 1: 1504–1514.

Regnault

Hemker

Wahl

. Phenotyping the haemostatic system by thrombography—potential for the estimation of thrombotic risk. Thromb Res 2004; 114: 539–545.

Mannucci

. The measurement of multifactorial thrombophilia. Thromb Haemost 2002; 88: 1–2.

Hemker

Giesen

AIDieri

. The calibrated automated thrombogram (CAT): a universal routine test for hyper- and hypocoagulability. Pathophysiol Haemost Thromb 2002; 32: 249–253.

10.

Zhao

Zhang

. Thrombin Generation Increasing with Age and Decreasing with Use of Heparin Indicated by Calibrated Automated Thrombogram Conducted in Chinese. Biomed Environ Sci 2014; 27: 378–384.

11.

Wang

Kong

. Retrospective analysis of the predictive effect of coagulogram on the prognosis of intracerebral hemorrhage. Acta Neurochir Suppl 2011; 111: 383–385.

12.

Qureshi

Mendelow

Hanley

. Intracerebral haemorrhage. Lancet 2009; 373: 1632–1644.

13.

Davis

Broderick

Hennerici

. Hematoma growth is a determinant of mortality and poor outcome after intracerebral hemorrhage. Neurology 2006; 66: 1175–1181.

14.

Fujii

Takeuchi

Sasaki

. Multivariate analysis of predictors of hematoma enlargement in spontaneous intracerebral hemorrhage. Stroke 1998; 29: 1160–1166.

15.

Hoffbrand

Pettit

. Essential hematology, 3rd ed. Oxford, London: Blackwell Scientific Publications, 1993.

16.

Goodnight

Jr Hathaway

. Disorders of hemostasis and thrombosis, a. clinical guide, 2nd ed. New York, NY: McGraw-Hill, 2001.

17.

Johari

Loke

. Brief overview of the coagulation cascade. Dis Mon 2012; 58: 421–423.

18.

Krajewski

Krauss

Kurz

. Real-time measurement of free thrombin: evaluation of the usability of a new thrombin assay for coagulation monitoring during extracorporeal circulation. Thromb Res 2014; 133: 455–463.

19.

Theusinger

Baulig

Seifert

. Changes in coagulation in standard laboratory tests and ROTEM in trauma patients between on-scene and arrival in the emergency department. Anesth Analg 2015; 120: 627–635.

20.

Brummel

Paradis

Butenas

. Thrombin functions during tissue factor-induced blood coagulation. Blood 2002; 100: 148–152.

21.

Borgen

Dahl

Reikeras

. Biomarkers of Coagulation and Fibrinolysis during Cemented Total Hip Arthroplasty with Pre- versus Postoperative Start of Thromboprophylaxis. Thrombosis 2013; 2013: 563217–563217.

22.

Soya

Terasawa

Okumura

. Fibrinopeptide A release is necessary for effective B:b interactions in polymerisation of variant fibrinogens with impaired A:a interactions. Thromb Haemost 2013; 109: 221–228.

23.

Sadanaga

Mitamura

. Soluble fibrin monomer complex levels during oral anticoagulant therapy do not predict subsequent thromboembolic events in patients with permanent atrial fibrillation. Int J Cardiol 2013; 168: 578–580.

24.

Sakata

. Thrombin, antithrombin, thrombin-antithrombin complex (TAT). Nihon Rinsho 2004; 62(Suppl 12): 626–628. in Japanese, English asbstract).

25.

Kawai

. Soluble fibrin monomer complex (SFMC), soluble fibrin monomer (SFM). Nihon Rinsho 2004; 62(Suppl 12): 600–603. in Japanese, English abstract.

26.

Brummel-Ziedins

. Models for thrombin generation and risk of disease. J Thromb Haemost 2013; 11(Suppl 1): 212–223.

27.

Mayer

Brun

Begtrup

. Recombinant activated factor VII for acute intracerebral hemorrhage. N Engl J Med 2005; 352: 777–785.

28.

Mayer

Brun

Begtrup

. Efficacy and safety of recombinant activated factor VII for acute intracerebral hemorrhage. N Engl J Med 2008; 358: 2127–2137.

29.

Mayer

Davis

Skolnick

. Can a subset of intracerebral hemorrhage patients benefit from hemostatic therapy with recombinant activated factor VII? Stroke 2009; 40: 833–840.

30.

Demchuk

Dowlatshahi

Rodriguez-Luna

. Prediction of haematoma growth and outcome in patients with intracerebral haemorrhage using the CT-angiography spot sign (PREDICT): a prospective observational study. Lancet Neurol 2012; 11: 307–314.