32 (2011) 253–257

Contents lists available at ScienceDirect Peptides

journal homepage: www.elsevier.com/locate/peptides

Change in plasma copeptin level after acute spontaneous basal ganglia hemorrhage

Xiao-Qiao Dong a, Man Huang b, Wen-Hua Yu a,∗, Zu-Yong Zhang a, Qiang Zhu a, Zhi-Hao Che a, Quan Du a, Hao Wang a a Department of Neurosurgery, The First Hangzhou Municipal People’s Hospital, Nanjing Medical University, 261 Huansha Road, Hangzhou 310000, China b Department of Intensive Care Unit, Sir Run Run Shaw Hospital, College of Medicine, Zhejiang University, 3 Qingchun East Road, Hangzhou 310000, China article info abstract

Article history: High plasma copeptin levels are associated with mortality after intracerebral hemorrhage (ICH). However, Received 6 October 2010 there is a paucity of data available on whether copeptin is an independent prognostic marker of mortality. Received in revised form Thus, we sought to furthermore evaluate this relation. Thirty healthy controls and 86 patients with acute 21 November 2010 ICH were included. Plasma samples were obtained on admission and at days 1, 2, 3, 5, and 7 after ICH. Its Accepted 22 November 2010 concentration was measured by enzyme-linked immunosorbent assay. After ICH, plasma copeptin level in Available online 30 November 2010 patients increased during the 6-h period immediately, peaked in 24 h, decreased gradually thereafter, and was substantially higher than that in healthy controls during the 7-day period. A multivariate analysis Keywords: Copeptin showed plasma copeptin level was an independent predictor for 1-week mortality (odds ratio, 1.013; Intracerebral hemorrhage 95% confidence interval (CI), 1.003–1.023; P = 0.009) and positively associated with hematoma volume Prognosis (t = 6.616, P < 0.001). A receiver operating characteristic curve identified that a baseline plasma copeptin level >577.5 pg/mL predicted 1-week mortality with 87.5% sensitivity and 72.2% specificity (area under Biomarkers curve (AUC), 0.873; 95% CI, 0.784–0.935). The AUC of the copeptin concentration was similar to those of Glasgow Coma Scale (GCS) scores and hematoma volumes (P = 0.136 and 0.280). However, copeptin did not statistically significantly improve the AUCs of GCS scores and hematoma volumes (P = 0.206 and 0.333). Hence, increased plasma copeptin level is associated with hematoma volume and an independent prognostic marker of mortality after ICH. © 2010 Elsevier Inc. All rights reserved.

1. Introduction Copeptin levels have also been found to be elevated in ischemic stroke; in these groups of patients, high copeptin levels were Arginine (AVP), produced by hypothalamic neu- highly predictive for poor function outcome and mortality [8,15]. rons, is stored and released from the gland Recently, Kleindienst et al. studied copeptin in brain trauma following different stimuli such as hypotension, hypoxia, hyper- patients. Copeptin levels increased with increasing severity of brain osmolarity, acidosis and infections [6]. AVP has vasoconstrictor injury [10]. Furthermore, it is found that copeptin was significantly and antidiuretic properties and potency to restore vascular tone associated with mortality and with a poor functional outcome in in vasodilatory hypotension [1]. AVP is derived from a larger pre- intracerebral hemorrhage (ICH); in this context, a meaningful mul- cursor (preproAVP) along with two other peptides of unknown tivariate analysis was not allowed because of a small sample size function, neurophysin II and copeptin, the carboxy-terminal part [17]. The present study aimed to investigate changes in plasma of the precursor [4]. Measurement of AVP level has limitations due copeptin levels in patients during the initial 7-day period after ICH to its short half-life and instability. Copeptin is a more stable pep- and also determine whether copeptin is an independent prognostic tide. Copeptin concentrations mirror that of AVP [14]. In critically marker of mortality in a group of ICH patients.. ill patients, copeptin values increased significantly with the sever- ity of the disease [12,13,16]. The role of copeptin is as yet unclear. 2. Materials and methods Copeptin was recently suggested to play an important role in the correct structural formation of the AVP precursor, as a prerequisite 2.1. Study population for its efficient proteolytic maturation [2]. Our target group consisted of consecutive patients with sponta- neous basal ganglia hemorrhage evaluated in the emergency room ∗ Corresponding author. of The First Hangzhou Municipal People’s Hospital within the first E-mail address: [email protected] (W.-H. Yu). 6 h from stroke onset. Between June 2006 and December 2008, a

0196-9781/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.peptides.2010.11.021 254 X.-Q. Dong et al. / Peptides 32 (2011) 253–257 total of 100 patients with spontaneous basal ganglia hemorrhage 2.6. Statistical analysis were initially evaluated. Exclusion criteria were existing previous neurological disease, head trauma, use of antiplatelet or anticoagu- Statistical analysis was performed with SPSS 10.0 (SPSS, Inc., lant medication, presence of other prior systemic diseases including Chicago, IL, USA) and MedCalc 9.6.4.0 (MedCalc Software, Mari- uremia, liver cirrhosis, malignancy, and chronic heart or lung dis- akerke, Belgium). The normality of data distribution was assessed ease, with the exceptions of diabetes mellitus and hypertension. by the Kolmogorovor–Smirnov test or Shapiro–Wilk test. All val- The patients who did not have serial copeptin measurements avail- ues are expressed as median (lower quartile and upper quartile), able before death were also excluded. Finally, 86 patients were mean ± standard deviation or counts (percentage) unless oth- included. erwise specified. Comparisons were made using (1) chi-square A control group consisted of 30 healthy age- and sex-matched test or Fisher exact test for categorical data, (2) unpaired Stu- subjects with normal results on brain magnetic resonance imaging dent t test for continuous normally distributed variables, and and without vascular risk factors. (3) the Mann–Whitney U-test for continuous non-normally dis- Informed consent to participate in the study was obtained from tributed variables. Correlations of copeptin with other variables them or their relatives. This protocol was approved by the Ethics were assessed by Spearman’s correlation coefficient and multivari- Committee before implementation. ate linear regression. The relation of copeptin to 1-week mortality was assessed in a logistic-regression model. For multivariate anal- 2.2. Clinical and radiological assessment ysis, we included the significantly different outcome predictors as assessed in univariate analysis. A receiver operating characteris- On arrival to the emergency department, a detailed history of tic curve was configured to establish the cutoff point of plasma vascular risk factors, concomitant medication, Glasgow Coma Scale copeptin with the optimal sensitivity and specificity for predict- (GCS) score, body temperature, heart rate, respiratory rate, and ing 1-week mortality. A P value of less than 0.05 was considered blood pressure was taken. statistically significant. All computerized tomography scans were performed accord- ing to the neuroradiology department protocol. Investigators who read them were blinded to clinical information. Hematoma vol- 3. Results ume was measured according to the previously reported formula A × B × C × 0.5 [11]. Hydrocephalus and the presence of intraven- 3.1. Patient characteristics tricular hemorrhage (IVH) and hematoma enlargement were also recorded. The hydrocephalus was assessed by two doctors. Eighty-six patients were enrolled, including 66 men and 20 women. The median age was 65 years (range, 42–80 years). 2.3. Intracerebral hemorrhage management Eighty patients (93.0%) presented with hypertension. Twenty- four patients (27.9%) suffered from diabetes mellitus. The median The treatments included surgical therapy, mechanical venti- admission time was 2.7 h (range, 0.3–6 h). On admission, the lation, blood pressure control, intravenous fluids, hyperosmolar median arterial pressure was 129 mmHg (range, 91–160 mmHg), agents, H2 blockers, early nutritional support, and physical ther- the median GCS score was 8 (range, 5–13), the median hematoma apy. The decision to intubate and use mechanical ventilation was volume was 48 mL (range, 20–80 mL), 60 patients (69.8%) com- based on the individuals’ level of consciousness, ability to protect plicated with IVH and 38 patients (44.2%) complicated with their airway and arterial blood gas levels [5]. When clinical and hydrocephalus. All patients underwent surgical treatment. The radiological examinations provide some estimation of elevation of surgical techniques included evacuation of hematoma (73 cases, intracranial pressure, the osmotherapy in the form of intravenous 84.9%), external decompression (46 cases, 53.3%), and external ven- mannitol were administered. The mean arterial pressure below tricular drainage (52 cases, 60.5%). The median time to surgery 130 mmHg (systolic arterial pressure below 170 mmHg) was main- was 3.9 h (range, 1.5–8.8 h). The other demographic, clinical, lab- tained [3]. The patients underwent evacuation of hematoma and/or oratory and radiological data from baseline CT scans are provided external ventricular drainage as appropriate. The choices of exter- in Tables 1 and 2. nal decompression depended on the neurosurgeons’ preference and intracranial pressure.

3.2. Mortality prediction 2.4. Determination of copeptin in plasma

Higher baseline plasma copeptin level was associated with 1- The informed consents were obtained from study population week mortality, as well as other variables shown in Tables 1 and 2. or family members in all cases before the blood were collected. In A multivariate analyses selected GCS (odds ratio, 0.151; 95% the control group, venous blood was drawn at study entry. In the confidence interval, 0.045–0.510; P = 0.002) and blood glucose ICH patients, venous blood was drawn on admission (defined as level (odds ratio, 1.425; 95% confidence interval, 1.055–1.925; day 0) and at 8:00 am at days 1, 2, 3, 5, and 7 after ICH. The blood P = 0.021) and baseline plasma copeptin level (odds ratio, 1.013; samples were immediately placed into sterile EDTA test tubes and ◦ 95% confidence interval, 1.003–1.023; P = 0.009) as the independent centrifuged at 1500 × g for 20 min at 4 C to collect plasma. Plasma ◦ predictors for 1-week mortality. was stored at −70 C until assayed. The concentration of copeptin in plasma was analyzed by enzyme-linked immunosorbent assay using commercial kits (Cusabio biotech Co. Ltd., Wuhan, Hubei Province, China) in accordance with the manufactures’ instructions. 3.3. The serial change in the plasma copeptin level in ICH patients

2.5. End point After ICH, plasma copeptin level in patients increased during the 6-h period immediately, peaked in 24 h, decreased gradually there- Outcome was assessed as mortality in 1 week. Cause of death after, and was substantially higher than that in healthy controls during the study for all patients was ICH. during the 7-day period (Fig. 1). X.-Q. Dong et al. / Peptides 32 (2011) 253–257 255

Table 1 Baseline clinical and radiological characteristics and factors associated with 1-week mortality.

All patients Survival group Non-survival group Univariate analysis (n = 86) (n = 54) (n = 32) P value

Gender 66/20 41/13 25/7 0.816 (male/female) Age (years) 65 (57, 77) 64 (57, 76) 66 (58, 78) 0.273 Hypertension 80 (93.0%) 50 (92.6%) 30 (93.8%) 0.839 Diabetes mellitus 24 (27.9%) 9 (16.7%) 15 (46.9%) 0.003 GCS score 8 (6, 10) 10 (8, 11) 6 (5, 6) 0.000 Presence of IVH 60 (69.8%) 31 (57.4%) 29 (90.6%) 0.001 Hematoma volume 48.0 (33.8, 65.0) 40.0 (25.0,48.0) 70.0 (60.0, 73.8) 0.000 (mL) Hydrocephalus 38 (44.2%) 16 (29.6%) 22 (68.8%) 0.000 Hemorrhage 13 (15.1%) 9 (16.7%) 4 (12.5%) 0.602 growth Rebleeding 6 (7.0%) 4 (7.4%) 2 (6.3%) 0.837 Seizure 6 (7.0%) 5 (9.3%) 1 (3.1%) 0.280 Pneumonia 10 (11.6%) 4(7.4%) 6 (18.7%) 0.113 Deep vein 7 (8.1%) 2 (3.7%) 5 (15.6%) 0.051 thrombosis Mechanical 47 (54.7%) 17 (31.5%) 30 (93.8%) 0.000 ventilation Admission time (h) 2.7 (1.4, 3.8) 2.7 (1.4, 3.7) 2.7 (1.2, 4.2) 0.858 Time to surgery (h) 3.9 (2.4, 5.5) 4.2 (2.3, 5.6) 3.7 (2.4, 5.5) 0.582 Plasma-sampling 3.1 (1.6, 4.5) 3.1 (1.7, 4.4) 3.1 (1.6, 4.7) 0.711 time (h) Evacuation of 73 (84.9%) 41 (75.9%) 32 (100.0%) 0.003 hematoma External 46 (53.5%) 17 (31.5%) 29 (90.6%) 0.000 decompression External 52 (60.5%) 24 (44.4%) 28 (87.5%) 0.000 ventricular drainage

Numerical variables were presented as median (lower quartile and upper quartile). Categorical variables were expressed as counts (%). Numerical variables were analyzed by Mann–Whitney U-test. Categorical variables were analyzed by chi-square test or Fisher exact test. n indicates number of patients; GCS, Glasgow Coma Scale; IVH, intraventricular hemorrhage.

3.4. Correlation of baseline plasma copeptin level with hematoma mortality of patients with 87.5% sensitivity and 72.2% specificity volume (area under curve, 0.873; 95% confidence interval, 0.784–0.935) (Fig. 3). The predictive value of the copeptin concentration was A significant correlation emerged between baseline plasma thus similar to that of GCS scores (area under curve, 0.941; copeptin level and hematoma volume, as well as other variables 95% confidence interval, 0.868–0.980) (P = 0.136) and that of shown in Table 3. When the above variables were introduced hematoma volumes (area under curve, 0.925; 95% confidence inter- into the linear regression model, plasma copeptin level remained val, 0.848–0.971) (P = 0.280). Copeptin improved the area under positively associated with hematoma volume (Fig. 2)(t = 6.616, curve of GCS scores to 0.964 (95% confidence interval, 0.899–0.992) P < 0.001). and that of hematoma volumes to 0.946 (95% confidence interval, 0.874–0.983), but the differences were not significant (P = 0.206 and 0.333). 3.5. The predictive significance of plasma copeptin level for 1-week mortality of patients 4. Discussion A receiver operating characteristic curve identified that a baseline plasma copeptin level >577.5 pg/mL predicted 1-week The current study demonstrated that plasma copeptin levels in the patients were significantly higher than those in healthy controls

1500 1500

1000 1000 (pg/mL)

500 (pg/mL) 500 Plasma copeptin level

0 Control Day 0 Day 1 Day 2 Day 3 Day 5 day 7 0 group Time after intracerebral hemorrhage Plasma copeptin level on admission 0 30 60 90 hematoma volume on admission (mL) Fig. 1. Graph showing serial changes of plasma copeptin concentration in patients with intracerebral hemorrhage. Data are expressed as median (lower quartile and Fig. 2. Graph showing correlation between hematoma volume and plasma copeptin upper quartile). level in patients with intracerebral hemorrhage. 256 X.-Q. Dong et al. / Peptides 32 (2011) 253–257

Table 2 Baseline laboratory characteristics and factors associated with 1-week mortality.

All patients Survival group Non-survival group Univariate analysis (n = 86) (n = 54) (n = 32) P value

Systolic arterial 184 (160, 207) 183 (159, 207) 186 (168, 208) 0.549 pressure (mmHg) Diastolic arterial 100.7 ± 13.0 100.1 ± 15.0 101.7 ± 8.8 0.538 pressure (mmHg) Mean arterial 128.1 ± 16.8 127.2 ± 18.4 129.6 ± 13.8 0.492 pressure (mmHg) Heart rate 81.1 ± 16.7 80.5 ± 15.7 82.2 ± 18.6 0.669 (beats/min) Body temperature 36.9 (36.6, 37.1) 36.9 (36.6, 37.1) 36.8 (36.6, 37.2) 0.978 (◦C) Respiratory rate 18 (16, 20) 18 (16, 20) 19 (16, 21) 0.806 (respirations/min) Blood white blood 9.8 (6.7, 14.2) 9.6 (7.2, 14.1) 10.6 (6.5, 14.4) 0.754 cell count (×109/L) Blood hemoglobin 130.0 (119.8, 127.5 (121.8, 133.0 (106.3, 0.652 level (g/L) 145.0) 145.3) 143.0) Blood platelet 169.5 (140.8, 164.5 (139.8, 176.5 (141.5, 0.971 count (×109/L) 204.5) 205.5) 204.0) Blood glucose level 9.7 (8.3, 13.3) 9.0 (7.2, 10.5) 10.9 (9.9, 14.8) 0.000 (mmol/L) Blood sodium level 141 (137, 142) 140 (137, 142) 141 (137, 144) 0.904 (mmol/L) Blood potassium 3.8 (3.5, 4.2) 3.8 (3.5, 4.2) 3.8 (3.6, 4.5) 0.511 level (mmol/L) Prothrombin time 13.0 (11.8, 14.6) 12.5 (11.3, 14.5) 13.7 (12.4, 14.8) 0.090 (s) Thrombin time (s) 17.7 (15.6, 19.5) 17.3 (15.6, 19.0) 18.3 (15.6, 19.6) 0.209 Partial 36.6 ± 5.6 35.9 ± 6.1 37.8 ± 4.4 0.104 thromboplastin time (s) Plasma copeptin 587.5 (336.5, 382.9 (323.1, 741.6 (622.1, 0.000 level (pg/mL) 687.5) 613.5) 899.2) Plasma C-reactive 7.2 (5.6, 12.2) 6.4 (5.3, 9.2) 10.4 (6.6, 13.8) 0.004 protein level (mg/L) Plasma fibrinogen 4..0 (2.9, 5.6) 3.2 (2.8, 4.4) 5.0 (3.6, 7.0) 0.002 level (g/L) Plasma D-dimer 2.0 (1.4, 2.5) 1.7 (1.3, 2.1) 2.3 (1.5, 3.4) 0.001 level (mg/L)

Numerical variables were presented as median (lower quartile and upper quartile) or mean ± standard deviation. Numerical variables were analyzed by Mann–Whitney U-test or unpaired Student t test. n indicates number of patients.

on admission until a week; and in patients who died in a week, the [7,9]. Copeptin is known to have prognostic value in a variety of dis- copeptin levels on admission were significantly higher compared eases, as it reflects disease severity and thus the chance of recovery. with levels in survivors. In multivariate logistic regression models For example, copeptin levels are independent predictors of survival of predictors of death, the copeptin levels on admission were an in critically ill patients suffering from hemorrhagic and septic shock independent predictor. [12]. Furthermore, copeptin levels have prognostic implications in Copeptin is co-synthesized with AVP in the patients with acute [13,16]. Therefore, it has been and is released into the portal circulation of the neurohy- hypothesized that the close and reproducible relation of copeptin pophysis. AVP contributes to the regulation of osmotic and cardiovascular homeostasis [6,1]. In addition, AVP activates the hypothalamo–pituitary–adrenal axis through potentiation of corticotrophin-releasing-hormone-induced ACTH secretion and 1.0 thus reflects the individual response at a hypothalamic level

Table 3 Baseline clinical, radiological and laboratory factors correlated with plasma copeptin 0.5 level.

r value P value

Glasgow Coma Scale score −0.557 0.000 Sensitivity of baseline plasma copeptin Hematoma volume (mL) 0.552 0.000 Blood glucose level (mmol/L) 0.257 0.017 0.0 Plasma C-reactive protein level (mg/L) 0.266 0.013 0.0 0.5 1.0 Plasma fibrinogen level (g/L) 0.403 0.000 1 - Specificity of baseline plasma copeptin Plasma D-dimer level (mg/L) 0.275 0.010

Correlations of plasma copeptin level with other variables were analyzed by Spear- Fig. 3. Graph showing the predictive significance of plasma copeptin level for 1- man test. week mortality of patients with intracerebral hemorrhage. X.-Q. Dong et al. / Peptides 32 (2011) 253–257 257 levels to the degree of activation of the stress axis is the basis of References its unique usefulness as a prognostic biomarker [8]. In our study, copeptin was closely related to initial hematoma volume in a mul- [1] Asfar P, Hauser B, Radermacher P, Matejovic M. Catecholamines and vaso- pressin during critical illness. Crit Care Clin 2006;22:131–49. tivariate regression model, which in turn is associated with clinical [2] Barat C, Simpson L, Breslow E. Properties of human vasopressin precursor con- severity and outcome. In accordance with this hypothesis, an earlier structs: inefficient monomer folding in the absence of copeptin as a potential study showed a correlation between the severity of hemorrhagic contributor to . Biochemistry 2004;43:8191–203. [3] Butcher K, Laidlaw J. Current intracerebral haemorrhage management. J Clin brain injury and copeptin level on admission [17]. Overall, our data Neurosci 2003;10:158–67. suggest that copeptin level in this early period might reflect the [4] De Bree FM, Burbach JP. Structure–function relationships of the vasopressin initial hemorrhagic insult (as demonstrated by the close relation prohormone domains. Cell Mol Neurobiol 1998;18:173–91. between copeptin and hematoma volume). [5] Gujjar AR, Deibert E, Manno EM, Duff S, Diringer MN. Mechanical ventilation for ischemic stroke and intracerebral hemorrhage: indications, timing, and Copeptin comes from the same precursor as mature AVP, outcome. Neurology 1998;51:447–51. which is already well associated with hemodynamic changes [6] Itoi K, Jiang YQ, Iwasaki Y, Watson SJ. Regulatory mechanisms of corticotrophin- and patient outcome. The measurement of mature AVP, how- releasing hormone and vasopressin gene expression in the hypothalamus. J Neuroendocrinol 2004;16:348–55. ever, is subject to considerable challenges, and has therefore not [7] Itoi K, Seasholtz AF, Watson SJ. Cellular and extracellular regulatory mech- reached clinical routine in the context of rapid measurements in anisms of hypothalamic corticotropin-releasing hormone neurons. Endocr J the critically ill patients. Here, the stability and longer ex vivo 1998;45:13–33. [8] Katan M, Fluri F, Morgenthaler NG, Schuetz P, Zweifel C, Bingisser R, et al. half-life of copeptin is a practical advantage, which makes it eas- Copeptin: a novel, independent prognostic marker in patients with ischemic ier to determine in the clinical laboratory. The current study stroke. Ann Neurol 2009;66:799–808. demonstrated for the first time that copeptin is an independent [9] Katan M, Morgenthaler N, Widmer I, Puder JJ, König C, Müller B, et al. Copeptin, a stable derived from the vasopressin precursor, correlates with the mortality marker in patients with ICH and its discriminative power individual stress level. Neuro Endocrinol Lett 2008;29:341–6. (reflected by area under curve) was in the range of GCS score [10] Kleindienst A, Brabant G, Morgenthaler NG, Dixit KC, Parsch H, Buchfelder M. and hematoma volume which are known to be a strong indi- Following brain trauma, copeptin, a stable peptide derived from the AVP pre- cursor, does not reflect osmoregulation but correlates with injury severity. Acta vidual outcome predictor. However, copeptin did not statistically Neurochir Suppl 2010;106:221–4. significantly improve the AUCs of GCS scores and hematoma vol- [11] Kothari RU, Brott T, Broderick JP, Barsan WG, Sauerbeck LR, Zuccarello M, umes. Therefore, our findings might have some potential clinical et al. The ABCs of measuring intracerebral hemorrhage volumes. Stroke applications such as the use of copeptin level as a surrogate 1996;27:1304–5. [12] Morgenthaler NG, Müller B, Struck J, Bergmann A, Redl H, Christ-Crain M. marker in future trials testing hemostatic agents or neuroprotective Copeptin, a stable peptide of the arginine vasopressin precursor, is elevated drugs. in hemorrhagic and septic shock. Shock 2007;28:219–26. [13] Stoiser B, Mörtl D, Hülsmann M, Berger R, Struck J, Morgenthaler NG, et al. Copeptin, a fragment of the vasopressin precursor, as a novel predictor of 5. Conclusion outcome in heart failure. Eur J Clin Invest 2006;36:771–8. [14] Struck J, Morgenthaler NG, Bergmann A. Copeptin, a stable peptide derived In this study, increased plasma copeptin level is associated with from the vasopressin precursor, is elevated in serum of sepsis patients. Peptides 2005;26:2500–4. hematoma volume and an independent prognostic marker of mor- [15] Urwyler SA, Schuetz P, Fluri F, Morgenthaler NG, Zweifel C, Bergmann A, et al. tality after ICH. Prognostic value of copeptin: one-year outcome in patients with acute stroke. Stroke 2010;41:1564–7. [16] Voors AA, von Haehling S, Anker SD, Hillege HL, Struck J, Hartmann O, et al. Acknowledgement C-terminal provasopressin (copeptin) is a strong prognostic marker in patients with heart failure after an acute : results from the OPTI- The authors thank all staffs in Central Laboratory and Depart- MAAL study. Eur Heart J 2009;30:1187–94. [17] Zweifel C, Katan M, Schuetz P, Siegemund M, Morgenthaler NG, Merlo A, et al. ment of Neurosurgery of The First Hangzhou Municipal People’s Copeptin is associated with mortality and outcome in patients with acute Hospital for their assistance. intracerebral hemorrhage. BMC Neurol 2010;10:34.