Elevated hemoglobin is associated with cerebral infarction in Tibetan patients with primary hemorrhagic neurovascular diseases

Elevated hemoglobin is associated with cerebral infarction in Tibetan patients with primary hemorrhagic neurovascular diseases

Clinical Neurology and Neurosurgery 157 (2017) 46–50 Contents lists available at ScienceDirect Clinical Neurology and Neurosurgery journal homepage:...

335KB Sizes 0 Downloads 23 Views

Clinical Neurology and Neurosurgery 157 (2017) 46–50

Contents lists available at ScienceDirect

Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro

Elevated hemoglobin is associated with cerebral infarction in Tibetan patients with primary hemorrhagic neurovascular diseases Ruiqi Chen1, Anqi Xiao1, Lu Ma, Hao Li, Sen Lin, Chao You

MARK



Department of Neurosurgery, West China Hospital, Sichuan University, Chengdu, Sichuan 610041, China

A R T I C L E I N F O

A B S T R A C T

Keywords: Primary hemorrhagic neurovascular diseases Cerebral infarction Tibetan Hemoglobin

Objectives: Although many studies have focused on primary hemorrhagic neurovascular diseases (PHNVDs) in different races, studies of PHNVDs in the plateau area of China are still insufficient. Chinese Tibetan people are the largest population living in the plateau area. Previous studies have shown that Tibetan PHNVD patients have a significantly higher incidence of cerebral infarction, but the mechanism remains uncertain. This study aimed to develop a better understanding on the mechanism of their high risk of cerebral infarction. Patients and methods: In this retrospective case control study, we used a hospital information system to search for consecutive Tibetan patients with PHNVDs from January 2012 to June 2016. Intra-hospital data including baseline information and complications were recorded, and the risk factors for cerebral infarction were analyzed. Results: Univariate analysis and cox proportional hazard multivariate regression analysis revealed that elevated hemoglobin (HGB) concentration was positively associated with an increased incidence of cerebral infarction (P < 0.001). The cutoff value that maximized the ability to predict in-hospital infarction in Tibetans with PHNVDs was 15.2 g/dL. Tibetan PHNVD patients with an increased HGB concentration were more likely to present with cerebral infarction within the first 5 days after onset of PHNVDs, and the probability was highest on the 3rd day. Conclusions: HGB levels could be used to predict in-hospital cerebral infarction in Tibetan patients with PHNVDs. These patients are more likely to develop in-hospital infarction when the HGB concentration is higher than 15.2 g/dL. For Tibetan PHNVD patients with an elevated HGB concentration, most cerebral infarctions occurred within the first five days after onset, with more incidents occurring on the third day.

1. Introduction Primary hemorrhagic neurovascular diseases (PHNVDs) refer to any hemorrhagic events that occur in the central nervous system, mainly including aneurysmal subarachnoid hemorrhage (aSAH), spontaneous intracerebral hemorrhage (sICH) and hemorrhagic arteriovenous malformation (AVM). These events are challenging for neurosurgeons due to their high rates of morbidity and poor prognoses [2,15,19,29,33]. Although many studies have focused on PHNVDs in different races, studies of PHNVDs in the plateau area of China are still insufficient. In China, Tibetan people are the largest population living in the plateau area, estimated at 7.8 million, and the majority inhabits the Southwestern Himalayan Plateau area (mean altitude over 4500 m). Two national surveys have reported that Tibetan people had the highest incidence of stroke among the Chinese population [35,36], and unlike the stroke distribution in other populations in which the ischemic subtype accounts for the majority of cases, hemorrhagic stroke occurs in ⁎

1

the majority (74.1%) of cases in Tibetans [3]. More interestingly, our previous study revealed that Tibetan PHNVD patients had a significantly higher risk of cerebral infarction than contemporary Han Chinese patients (the majority of the Chinese population), which is similar to other reports in the literature [34]. However, to our knowledge, no study has revealed the mechanism for this increased risk of cerebral infarction. Due to the high-altitude, low pressure and an oxygen-thin atmosphere, the serum hemoglobin (HGB) concentration of Tibetan people is increased to compensate for these factors [5,10,20,30]. Several previous studies have provided evidence for increased blood viscosity caused by the increased HGB concentration [7,16,17], and other studies have reported that high blood viscosity is positively associated with the development of cerebral infarction [6,9,14,23]. Here, we propose a hypothesis for the first time that the elevated HBG concentration is correlated with the high incidence of cerebral infarction among Tibetan PHNVD patients. Using the database of the West China Hospital of Sichuan University

Corresponding author at: No. 37 Guo Xue Xiang, Chengdu, Sichuan 610041, China. E-mail address: [email protected] (C. You). These authors contributed equally to the manuscript.

http://dx.doi.org/10.1016/j.clineuro.2017.03.025 Received 20 February 2017; Received in revised form 25 March 2017; Accepted 28 March 2017 Available online 30 March 2017 0303-8467/ © 2017 Elsevier B.V. All rights reserved.

Clinical Neurology and Neurosurgery 157 (2017) 46–50

R. Chen et al.

divided by the cutoff value of HGB. Significance was defined as P < 0.05, and 95% confidence intervals (CI) were calculated for each variable.

(WCH), the largest medical center in the West China area, we performed a retrospective analysis of 284 Tibetan PHNVD patients to determine whether the HGB concentration could be considered a predictive indicator of in-hospital cerebral infarction. The results of our study will facilitate the understanding of PHNVDs in the plateau area of China.

3. Results 3.1. Patients characteristics

2. Patient and methods A total of 284 Tibetan PHNVD patients were enrolled with a mean age of 48.3 ± 14.7 years, and the participants included 110 aSAH, 122 sICH and 52 AVM patients. Males accounted for 66.9% (n = 190) of all patients, and the mean HGB concentration was 162.4 ± 14.6 (g/L). Among all patients, 53.5% (n = 152), 12% (n = 34), 15.1% (n = 43), 6.7% (n = 19) and 17.6% (n = 50) had a history of hypertension, diabetes mellitus, coronary artery disease, hemorrhagic stroke and hypercholesterolemia, respectively. At the time of onset, 34.5% (n = 98) of patients were smokers and 30.6% (n = 87) used alcohol. The mean time of admission delay was 16.9 ± 14.6 h. Regarding treatment, 62.7% (n = 178) of patients underwent a surgical approach, and the remaining patients chose a conservative method. Infarction occurred in 131 (46.1%) patients during hospitalization. Additionally, 19.4% (n = 55), 5.7% (n = 17), 17.6% (n = 50), and 3.9% (n = 11) of Tibetan PHNVD patients presented with rebleeding, hydrocephalus, pulmonary infection and seizures during hospitalization, respectively.

2.1. Study design and participants This study was a retrospective case control study that was approved by the institutional review board (IRB) of WCH. We used the hospital information system (HIS) to obtain data on Tibetan PHNVD patients who were admitted in the Department of Neurosurgery of WCH from January 2014 to June 2016. The inclusion criteria were patients identified by the discharge diagnoses in the HIS database using the key words “Tibetan”; “aneurysmal subarachnoid hemorrhage”; “spontaneous intracerebral hemorrhage” and “arteriovenous malformation”. One neurosurgeon and one neuroradiologist confirmed the diagnoses of the selected patients. Those with severe illnesses in other systems or incomplete medical profiles were excluded. 2.2. Data collection and grouping We used the HIS to collect all data. The medical data collected included baseline information, such as age, gender, HGB concentration on admission, history of hypertension, diabetes mellitus, hypercholesterolemia, prior hemorrhagic stroke, current smoking, current drinking and admission delay in hours; treatment information, such as surgical treatment or conservative treatment; and complications during hospitalization, such as rebleeding, cerebral infarction, hydrocephalus, seizures, pulmonary infection, gastrointestinal bleeding and length of stay (LOS). Comparisons were made between patients with cerebral infarction (case group) and those without cerebral infarction (control group).

3.2. Univariate analysis When divided patients into two groups based on cerebral infarction (infarction group and non-infarction group), univariate analysis showed several significant differences between two groups. Patients in cerebral infarction group got significantly higher age (55.6 ± 15.8 vs 42.0 ± 13.8, P < 0.000), more with hypertension (61.1% vs 47.1%, P = 0.025), more presented with seizures (7.6% vs 0.7%, P = 0.006) and higher serum hemoglobin levels (181.3 ± 15.9 vs 146.2 ± 13.6, P < 0.000) compared with those in non-cerebral infarction group. Two groups did not differ significantly in other baseline information (p > 0.05). (Table 1)

2.3. Outcomes 3.3. Multivariate analysis The primary outcome was the incidence of cerebral infarction defined as radiographic evidence of stroke due to any cause during hospitalization. Cerebral infarction was confirmed by head Computed Tomography (CT) or diffusion-weighted magnetic resonance imaging (MRI) during hospitalization. The secondary outcomes were LOS and other complications during hospitalization, including rebleeding, hydrocephalus, seizures, pulmonary infection and gastrointestinal bleeding.

Then we did Cox proportional hazard regression analysis (multiTable 1 Comparisons of variables between cerebral infarction and non-cerebral infarction groups in Tibetan PHNVDs: univariate analysis.

2.4. Statistical analysis Mean age years (mean ± SD) Gender, male, n (%) Hypertension, n (%) Diabetes mellitus, n (%) Hypercholesterolemia, n (%) Prior hemorrhagic stroke, n (%) Prior coronary artery disease, n (%) Platelet, ×109 (mean ± SD) Current smoking, n (%) Current drinking, n (%) Admission delay, h (mean ± SD) Treatment method, surgery, n (%) Rebleeding, n (%) Hydrocephalus, n (%) Pulmonary infection, n (%) Seizures, n (%) Hemoglobin, g/L (mean ± SD)

SPSS statistical software (version 22.0; SPSS Inc., Chicago, Illinois, USA) and MedCalc statistical software (version 15.2; MedCalc Software, Mariakerke, Ostend, Belgium) were used for all statistical analyses. The mean ± standard deviation (SD) was reported for quantitative data. Categorical data were expressed as frequencies and percentages. The Cox proportional hazard regression model was performed for multivariate analysis. If the OR of a test is larger than 1, it is considered as risk factor, and protective factor if the OR is less than 1. The receiver operating characteristic (ROC) curve and area under the ROC curve (AUROC) were used to evaluate the ability of HGB to predict the risk of cerebral infarction. The z statistic was used to calculate the difference in the AUROC derived from the same cases, and the cutoff value that maximized the ability of the HGB concentration to predict in-hospital infarction in Tibetan PHNVD patients was calculated. Moreover, Kaplan-Meier curves were used to show and compare the in-hospital cerebral infarction curves of elevated HGB and low-normal HGB groups

Cerebral infarction group (n = 131)

Non-cerebral infarction group (n = 153)

P value

55.6 ± 15.8 91 (69.5) 80 (61.1) 14 (10.7) 26 (19.8) 11 (8.4) 21 (16.0) 207.3 ± 76.3 52 (39.7) 39 (29.8) 15.8 ± 12.2 84 (64.1) 27 (20.6) 10 (7.6%) 27 (20.6) 10 (7.6) 181.3 ± 15.9

42.0 ± 13.8 99 (64.7) 72 (47.1) 20 (13.1) 24 (15.7) 8 (5.2) 22 (14.4) 216.4 ± 64.7 46 (30.1) 48 (31.4) 17.8 ± 15.7 94 (61.4) 28 (18.3%) 7 (4.6%) 23 (15.0) 1 (0.7) 146.2 ± 13.6

0.000 0.470 0.025 0.664 0.446 0.408 0.825 0.278 0.115 0.871 0.238 0.731 0.734 0.405 0.283 0.006 0.000

PHNVDs indicate primary hemorrhagic neurovascular diseases; SD indicates standard deviation.

47

Clinical Neurology and Neurosurgery 157 (2017) 46–50

R. Chen et al.

concentrations (HGB > 15.2 g/dL). The Kaplan-Meier curves for inhospital cerebral infarction (Fig. 2) showed statistically significant differences between the groups (P < 0.01).

Table 2 Multivariate analysis on risk factors of cerebral infarction. Variables

Gender Age Hypertension Diabetes mellitus Hypercholesterolemia Prior hemorrhagic stroke Prior cardiovascular diseases Platelet Current smoking Current drinking Admission delay Treatment method Rebleeding Hydrocephalus Pulmonary infection Seizures Hemoglobin

OR

0.577 1.025 0.588 1.273 0.911 0.760 3.407 1.152 0.698 1.672 1.014 1.268 0.704 0.775 0.613 1.125 1.237

95% CI

P-value

Lower

Upper

0.327 1.004 0.313 0.628 0.342 0.148 0.659 0.837 0.374 0.838 0.979 0.747 0.339 0.240 0.152 1.020 1.224

1.017 1.047 1.105 2.415 1.479 2.707 31.164 1.404 1.403 3.210 1.055 1763 1.386 2.706 2.609 1.574 1.251

4. Discussion 0.087 0.018 0.099 0.336 0.602 0.923 0.531 0.094 0.479 0.079 0.752 0.234 0.418 0.844 0.415 0.009 0.000

This study is the first retrospective observational study to examine the mechanism responsible for the high risk of cerebral infarction in Tibetan PHNVD patients. Based on our results, elevated HGB concentration is positively associated with the increased incidence of cerebral infarction. Therefore, the HGB concentration could be used to predict in-hospital cerebral infarction in Tibetan PHNVD patients. The cutoff value of HGB was 15.2 g/dL, indicating that the risk of cerebral infarction during a hospital stay could be significantly higher if patients presented with a HGB concentration higher than 15.2 g/dL. Tibetan PHNVD patients with an elevated HGB concentration were more likely to present with cerebral infarction within the first 5 days after onset of PHNVDs, especially on the third day. The results of our study may provide a better understanding of the relationship between HGB and cerebral infarction in Tibetan PHNVD patients. According to our study, the average HGB value in Tibetan PHNVD patients was higher than that reported in the literature [1,13]. The inhospital cerebral infarction rate was also significantly higher in Tibetan PHNVD patients than in other races reported in the literature [4,11,18,25,26,28,37]. Based on our multivariate analysis, an elevated HGB concentration could be a risk factor for in-hospital cerebral infarction in Tibetan PHNVD patients. Therefore, hemodilution therapy aiming to decrease the HGB concentration is important to reduce the risk of cerebral infarction and improve the prognosis. However, according to some recent studies, lower mean HGB concentrations could also be related to brain infarction, unfavorable outcomes and a higher rate of mortality in PHNVD patients via anemia and oxygen deficiency [12,21,22,31]. Thus, a “seesaw” relationship might exist between HGB and cerebral infarction in Tibetan PHNVD patients; either excessively high or low HGB concentrations may lead to cerebral infarction. According to the cutoff value of the ROC curve, Tibetan PHNVD patients with HGB levels higher than 15.2 g/dL are more likely to develop in-hospital infarction. Therefore, an optimal treatment goal to minimize the incidence of cerebral infarction in Tibetan PHNVD patients might be a HGB concentration lower than 15.2 g/dL. However, further prospective, multicenter studies with larger sample sizes are warranted to further verify the accuracy of this cutoff value. According to the Kaplan-Meier curves for in-hospital infarction, most infarctions occurred within the first 7 days after onset, particularly during the first 5 days, which was similar to the results of several previous studies [8,24,27,32]. Notably, our study first reported the time-cerebral infarction relationship between high and low HGB concentration groups. For patients with an elevated HGB concentration, the zenith of the growth rate of cerebral infarction was day 3, which was followed by a gradual decline from day 4 to day 7. However, for the low HGB group, the number of patients who had a cerebral infarction gradually increased from day 1 to day 7. Thus, we found that for Tibetan PHNVD patients with a high HGB concentration (higher than 15.2 g/dL), cerebral infarction was more likely to occur on the third day after onset of PHNVD. The results of our study could help provide a better understanding of how infarction rates change over time among patients with different HGB concentrations, which could be useful for clinical workers to predict and prevent cerebral infarction. Based on the positive relationship between cerebral infarction and the HGB concentration in Tibetan PHNVD patients, we propose the “hemoglobin, blood viscosity, tiny thrombosis-based cerebral infarction” hypothesis, which might be a possible mechanism to explain the high rate of cerebral infarction in Tibetan PHNVD cases. To be more accurate, the compensatory high HGB concentration caused by the oxygen-thin atmosphere of the plateau area functions as the beginning of the transduction axis and then contributes to an increase in blood

CI indicates confidence intervals.

variate analysis) adjusted for confounding factors which showed that the HGB concentration was still positively associated with in-hospital infarction in Tibetan PHNVD patients (OR [95% CI], 1.237 [1.224–1.251], P = 0.000). (Table 2) 3.4. AUROC curve for HGB cutoff value The cutoff value that maximized the ability to predict in-hospital infarction in Tibetan PHNVD patients was 15.2 g/dL. As shown in Fig. 1, the AUROC curve for the ability of HGB to predict in-hospital infarction was 0.775 (95% CI, 0.696–0.841, P < 0.0001). 3.5. Kaplan-Meier curve for in-hospital cerebral infarction We divided all patients into two groups according to the cutoff value. Finally, 148 (52.1%) patients were identified as having high HGB

Fig. 1. Receiver operating characteristic curve.Receiver operating curves of the ability of hemoglobin to predict in-hospital infarction in Tibetan PHNVD patients (n = 284). AUROC = Area under the Receiver Operating Characteristic Curve, HGB = hemoglobin.

48

Clinical Neurology and Neurosurgery 157 (2017) 46–50

R. Chen et al.

Fig. 2. Kaplan-Meier curves for in-hospital infarction.Kaplan-Meier curves for in-hospital infarction in Tibetan PHNVD patients with low or high hemoglobin concentrations (P < 005). Hb = hemoglobin.

Funding

viscosity. High blood viscosity subsequently leads to the development of tiny thromboses inside the cerebral circulation system, which ultimately results in cerebral infarction. Several previous studies have provided evidence for an increase in blood viscosity due to high HGB levels [7,16,17]. Moreover, other studies have reported a relationship between high blood viscosity and infarction in which tiny thromboses form in the cerebral circulation system and may play an important role in infarction [6,9,14,23]. Although supported by the literature mentioned above, additional research is still needed to confirm this hypothesis. Our conclusions are limited by some weaknesses. First, the study design was a nonrandomized, retrospective, observational study in a single institution. Second, all data collected were based on inpatient records. Thus, the mechanism by which HGB concentrations predicted the long-term prognosis remains uncertain. Third, PHNVD is a combination of several different hemorrhagic diseases; research aiming to analyze each subtype of PHNVD in Tibetan patients will constitute our future research plans.

This work was supported by the National Key Technology R & D Program for the 12th Five-year Plan of P.R China (grant number: 2011BAI08B05). Disclosure The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. Acknowledgment Authors would like to express our sincere appreciation to Dr. Xie Hong for her valuable comments on our paper. References [1] H. Chaplin Jr., M. Cassell, G.E. Hanks, The stability of the plasma hemoglobin level in the normal human subject, J. Lab. Clin. Med. 57 (1961) 612–619. [2] S.A. Dijkland, B. Roozenbeek, P.A. Brouwer, H.F. Lingsma, D.W. Dippel, L.J. Vergouw, et al, Prediction of 60-day case fatality after aneurysmal subarachnoid hemorrhage: external validation of a prediction model, Crit. Care Med. 44 (8) (2016) 1523–1529. [3] J. Fang, C. Zhuo-Ga, Y. Zhao, F. Kong, Y. Si, M. Liu, et al., Characteristics of stroke in tibet autonomous region in china: a hospital-based study of acute stroke, Eur. Neurol. 66 (3) (2011) 151–158. [4] S. Ferguson, R.L. Macdonald, Predictors of cerebral infarction in patients with aneurysmal subarachnoid hemorrhage, Neurosurgery 60 (4) (2007) 658–667. discussion 667. [5] A.R. Frisancho, Frisancho Developmental functional adaptation to high altitude: review, Am. J. Hum. Biol. 25 (2) (2013) 151–168. [6] J. Grotta, R. Ackerman, J. Correia, G. Fallick, J. Chang, Whole blood viscosity parameters and cerebral blood flow, Stroke 13 (3) (1982) 296–301. [7] A. Herrmann, P. Muller, Muller correlation of the internal microviscosity of human erythrocytes to the cell volume and the viscosity of hemoglobin solutions, Biochim. Biophys. Acta 885 (1) (1986) 80–87. [8] R. Jabbarli, M. Reinhard, W.D. Niesen, R. Roelz, M. Shah, K. Kaier, et al., Predictors and impact of early cerebral infarction after aneurysmal subarachnoid hemorrhage, Eur. J. Neurol. 22 (6) (2015) 941–947. [9] K.M. Jan, S. Chien, J.T. Bigger Jr., Observations on blood viscosity changes after acute myocardial infarction, Circulation 51 (6) (1975) 1079–1084.

5. Conclusions To date, studies of PHNVDs in the Tibetan population are still lacking. As the first retrospective, observational study focusing on the mechanism responsible for the high risk of cerebral infarction in this population, our research revealed that the HGB concentration has a relatively good predictive ability for in-hospital infarction in Tibetan PHNVD patients. These patients are more likely to develop in-hospital infarction when the HGB concentration is higher than 15.2 g/dL. Tibetan PHNVD patients with elevated HGB concentrations are more likely to develop cerebral infarction within the first 5 days onset, especially on the third day. However, more randomized, prospective studies with larger sample sizes and a multicenter design are needed to provide more robust evidence.

49

Clinical Neurology and Neurosurgery 157 (2017) 46–50

R. Chen et al.

hemorrhage: a critical review, Stroke 43 (8) (2012) 2258–2263. [25] S. Prabhakaran, R. Gupta, B. Ouyang, S. John, R.E. Temes, Y. Mohammad, et al., Acute brain infarcts after spontaneous intracerebral hemorrhage: a diffusionweighted imaging study, Stroke 41 (1) (2010) 89–94. [26] A.A. Rabinstein, J.A. Friedman, S.D. Weigand, R.L. McClelland, J.R. Fulgham, E.M. Manno, et al., Predictors of cerebral infarction in aneurysmal subarachnoid hemorrhage, Stroke 35 (8) (2004) 1862–1866. [27] Y.B. Roos, R.J. de Haan, L.F. Beenen, R.J. Groen, K.W. Albrecht, M. Vermeulen, Complications and outcome in patients with aneurysmal subarachnoid haemorrhage: a prospective hospital based cohort study in the Netherlands, J. Neurol. Neurosurg. Psychiatry 68 (3) (2000) 337–341. [28] J.M. Schmidt, K.E. Wartenberg, A. Fernandez, J. Claassen, F. Rincon, N.D. Ostapkovich, et al., Frequency and clinical impact of asymptomatic cerebral infarction due to vasospasm after subarachnoid hemorrhage, J. Neurosurg. 109 (6) (2008) 1052–1059. [29] S.F. Shakur, K. Liesse, S. Amin-Hanjani, A.P. Carlson, V.A. Aletich, F.T. Charbel, et al., Relationship of cerebral arteriovenous malformation hemodynamics to clinical presentation, angioarchitectural features, and hemorrhage, Neurosurgery 63 (Suppl. 1) (2016) 136–140. [30] T.S. Simonson, Y. Yang, C.D. Huff, H. Yun, G. Qin, D.J. Witherspoon, et al., Genetic evidence for high-altitude adaptation in Tibet, Science 329 (5987) (2010) 72–75. [31] M. Stein, L. Brokmeier, J. Herrmann, W. Scharbrodt, V. Schreiber, M. Bender, et al., Mean hemoglobin concentration after acute subarachnoid hemorrhage and the relation to outcome mortality, vasospasm, and brain infarction, J. Clin. Neurosci. 22 (3) (2015) 530–534. [32] M.D. Vergouwen, D. Ilodigwe, R.L. Macdonald, Cerebral infarction after subarachnoid hemorrhage contributes to poor outcome by vasospasm-dependent and -independent effects, Stroke 42 (4) (2011) 924–929. [33] C. Weimar, J. Kleine-Borgmann, Epidemiology, prognosis and prevention of nontraumatic intracerebral hemorrhage, Curr. Pharm. Des. (2016). [34] A. Xiao, R. Chen, H. Li, L. Ma, S. Lin, X. Hu, et al., Primary hemorrhagic neurovascular diseases in Tibetans: a retrospective observational study, World Neurosurg. 96 (2016) 423–428. [35] G.B. Xue, The incidence rates in 1986 of stroke in urban and rural areas of the People's Republic of China, Zhonghua Yu Fang Yi Xue Za Zhi 25 (4) (1991) 196–200. [36] G.B. Xue, B.X. Yu, X.Z. Wang, G.Q. Wang, Z.Y. Wang, Stroke in urban and rural areas of China, Chin. Med. J. (Engl.) 104 (8) (1991) 697–704. [37] E. Zia, B. Hedblad, H. Pessah-Rasmussen, G. Berglund, L. Janzon, G. Engstrom, Blood pressure in relation to the incidence of cerebral infarction and intracerebral hemorrhage. Hypertensive hemorrhage: debated nomenclature is still relevant, Stroke 38 (10) (2007) 2681–2685.

[10] A.R. Jha, D. Zhou, C.D. Brown, M. Kreitman, G.G. Haddad, K.P. White, White shared genetic signals of hypoxia adaptation in drosophila and in high-altitude human populations, Mol. Biol. Evol. 33 (2) (2016) 501–517. [11] C.H. Kim, J.S. Kim, Development of cerebral infarction shortly after intracerebral hemorrhage, Eur. Neurol. 57 (3) (2007) 145–149. [12] A.H. Kramer, M.J. Gurka, B. Nathan, A.S. Dumont, N.F. Kassell, T.P. Bleck, Bleck complications associated with anemia and blood transfusion in patients with aneurysmal subarachnoid hemorrhage, Crit. Care Med. 36 (7) (2008) 2070–2075. [13] J. Leonova, E.G. Kazanetz, N.S. Smetanina, A.D. Adekile, G.D. Efremov, T.H. Huisman, Variability in the fetal hemoglobin level of the normal adult, Am. J. Hematol. 53 (2) (1996) 59–65. [14] R.Y. Li, Z.G. Cao, Y. Li, R.T. Wang, Increased whole blood viscosity is associated with silent cerebral infarction, Clin. Hemorheol. Microcirc. 59 (4) (2015) 301–307. [15] J. Liu, D. Wang, R. Yuan, Y. Xiong, M. Liu, Prognosis of 908 patients with intracerebral hemorrhage in Chengdu, Southwest of China, Int. J. Neurosci. (2016) 1–6. [16] O.G. Luneva, N.A. Brazhe, O.E. Fadyukova, M.Y. Akhalaya, A.A. Baizhumanov, E.Y. Parshina, et al., Changes in plasma membrane viscosity and hemoporphyrin conformation in erythrocyte hemoglobin under the conditions of ischemia and reperfusion of rat brain, Dokl. Biochem. Biophys. 405 (2005) 465–467. [17] B.L. McClain, I.J. Finkelstein, M.D. Fayer, Dynamics of hemoglobin in human erythrocytes and in solution: influence of viscosity studied by ultrafast vibrational echo experiments, J. Am. Chem. Soc. 126 (48) (2004) 15702–15710. [18] M. Mijiti, P. Mijiti, A. Axier, M. Amuti, Z. Guohua, C. Xiaojiang, et al., Incidence and predictors of angiographic vasospasm, symptomatic vasospasm and cerebral infarction in chinese patients with aneurysmal subarachnoid hemorrhage, PLoS One 11 (12) (2016) e0168657. [19] T. Miyazaki, F. Ohta, K. Moritake, A. Nagase, T. Kagawa, The key to improving prognosis for aneurysmal subarachnoid hemorrhage remains in the pre-hospitalization period, Surg. Neurol. 65 (4) (2006) 360–365. discussion 365-366. [20] L.G. Moore, Human genetic adaptation to high altitude, High Alt. Med. Biol. 2 (2) (2001) 257–279. [21] A.M. Naidech, J. Drescher, M.L. Ault, A. Shaibani, H.H. Batjer, M.J. Alberts, Higher hemoglobin is associated with less cerebral infarction, poor outcome, and death after subarachnoid hemorrhage, Neurosurgery 59 (4) (2006) 775–779. discussion 779-780. [22] A.M. Naidech, B. Jovanovic, K.E. Wartenberg, A. Parra, N. Ostapkovich, E.S. Connolly, et al., Higher hemoglobin is associated with improved outcome after subarachnoid hemorrhage, Crit. Care Med. 35 (10) (2007) 2383–2389. [23] E.O. Ott, H. Lechner, A. Aranibar, High blood viscosity syndrome in cerebral infarction, Stroke 5 (3) (1974) 330–333. [24] S. Prabhakaran, A.M. Naidech, Naidech Ischemic brain injury after intracerebral

50