Risk factors of hospital mortality in chronic subdural hematoma: A retrospective analysis of 1117 patients, a single institute experience

Journal of Clinical Neuroscience 67 (2019) 46–51

Contents lists available at ScienceDirect

Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn

Clinical study

Risk factors of hospital mortality in chronic subdural hematoma: A retrospective analysis of 1117 patients, a single institute experience Shen Wang a,b,1, Yuxiao Ma a,b,1, Xiaochun Zhao c, Chun Yang a,b, Jiacheng Gu a,b, Weiji Weng d, Jiyuan Hui a, Qing Mao a, Guoyi Gao a,b, Junfeng Feng a,b,⇑ a

Department of Neurosurgery, Renji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China Shanghai Institute of Head Trauma, Shanghai, China Department of Neurosurgery, Barrow Neurological Institute, St. Joseph’s Hospital and Medical Center, Phoenix, AZ, USA d Department of Biochemistry and Molecular Cell Biology, Shanghai Key Laboratory for Tumor Microenvironment and Inflammation, Institute of Medical Sciences, School of Medicine, Shanghai Jiao Tong University, Shanghai, China b c

a r t i c l e

i n f o

Article history: Received 11 February 2019 Accepted 9 June 2019

Keywords: Chronic subdural hematoma Hospital mortality Risk factors Prealbumin International normalized rate High sensitivity C-reactive protein

a b s t r a c t Chronic subdural hematoma (CSDH) is not a benign disease in the elderly, and the mortality of CSDH is reported to be up to 32%. The study aimed to analyze hospital mortality and evaluate the risk factors in patients with CSDH. We retrospectively reviewed all patients with CSDH treated in the neurosurgery department of Renji hospital, School of Medicine, Shanghai Jiao Tong University from 10/2003 to 10/2018. Univariate and multivariate logistic regression analyses on the factors, including gender, age, Glasgow Coma Scale (GCS) on admission, main symptoms, history of head trauma, location of hematoma, density of subdural hematoma, laboratory tests on admission, et al, were performed. A total of 1117 patients with CSDH were included in the study, among which 20 patients died (mortality rate: 1.8%). Comparing the survival group and the death group, gender (p < 0.01), GCS on admission (p < 0.01), impaired consciousness as the presenting symptom (p = 0.041), history of head trauma (p = 0.02), location of hematoma (p = 0.03), platelet (p < 0.01), prealbumin (p = 0.046), albumin (p < 0.01), international normalized rate (INR) (p = 0.03), high sensitivity C-reactive protein (hsCRP) (p < 0.01), postoperative hyperthermia (p < 0.01) were shown to have significance. The logistic regression analysis revealed that GCS on admission (odds ratio [OR] 0.077, p < 0.01), bilateral CSDH (OR 0.100, p < 0.01 left CSDH as reference), prealbumin (OR 0.137, p = 0.02), INR (OR 12.959, p < 0.01) and hsCRP (OR 8.397, p < 0.01) were significantly associated with a higher mortality rate in CSDH. GCS on admission, bilateral CSDH, prealbumin, INR and hsCRP might be independent predictors of CSDH mortality. Ó 2019 Elsevier Ltd. All rights reserved.

1. Introduction Chronic subdural hematoma (CSDH) is a frequently encountered neurosurgical disease in elderly patients [1]. The annual incidence of CSDH is estimated to be as high as 14 per 100,000 personyears, and it is higher in the elderly population [1]. At present, Abbreviations: CSDH, chronic subdural hematoma; GCS, Glasgow Coma Scale; INR, international normalized rate; hsCRP, high sensitivity C-reactive protein; OR, odds ratio; CI, confidence interval; CT, computed tomography; MRI, magnetic resonance imaging; PT, prothrombin time; APTT, activated partial thromboplastin time; HU, hounsfield unit; GOS, Glasgow outcome scale; TBI, traumatic brain injury; IL, interleukin. ⇑ Corresponding author at: Department of Neurosurgery, Renji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China. E-mail address: [email protected] (J. Feng). 1 Common first authors. https://doi.org/10.1016/j.jocn.2019.06.026 0967-5868/Ó 2019 Elsevier Ltd. All rights reserved.

multiple surgical techniques can be utilized to treat CSDH including burr hole craniostomy with or without drainage, twist drill craniostomy and craniotomy [1], among which, burr hole with drainage (BHD) is the most widely used technique [2,3]. Previous reports showed that CSDH was not a benign disease in elderly patients [4,5], the mortality rate of CSDH was reported to be up to 32% [1]. The high in-hospital mortality rate makes it a complex clinical scenario in the elderly, an identification of risk factors associated with mortality of CSDH is of paramount importance for the treatment of these patients. Although numerous potential risk factors associated with the outcome of CSDH have been reported, such as GCS, age, frailty, the presence of medical comorbidities and coagulopathy [3,6–14], the results remain elusive. This study is to evaluate the risk factors of hospital mortality by analyzing the clinical data of

47

S. Wang et al. / Journal of Clinical Neuroscience 67 (2019) 46–51

CSDH patients in our department and may guide future treatment strategy in terms of reducing mortality. 2. Methods We retrospectively reviewed all patients with the diagnosis of CSDH admitted to the Department of Neurosurgery, Renji Hospital, School of Medicine, Shanghai Jiao Tong University between October 2003 and October 2018. Patients with CSDH were diagnosed by head computed tomography (CT) and magnetic resonance imaging (MRI). Patients older than 18 years old were included. All patients were treated with BHD under anesthesia. Local or general anesthesia was selected depending on their general status and ability of cooperation. We perforated the burr hole at the thickest part of the CSDH; after incising the dura, a silicon drainage tube was placed into the hematoma cavity and irrigation was then carried out with warm sterile saline. Air was flushed out with copious irrigation. 200 ml of the solution was used in one-side irrigation. The draining catheters were always connected to a closed draining system for several days. Most catheters were removed after 72 h after the operation. CT scans were performed routinely at the time of catheter removal and prior to the discharge, which was usually postoperative day 7. Patients were followed up at 1, 3 and 6 months at the outpatient clinic. The following demographic characteristics were collected: the age, gender, Glasgow Coma Scale (GCS) on admission, major presenting symptoms (headache, hemiparesis, gait disturbance, consciousness impairment, memory impairment, speech disturbance, seizure and visual disturbance), history of head trauma, usage of anticoagulant or antiplatelet medication and comorbidities such as hypertension, diabetes mellitus and cardiovascular disease. The results of the laboratory tests on admission including hemoglobin, platelet, blood glucose, prealbumin, albumin, prothrombin time (PT), activated partial thromboplastin time (APTT), international normalized rate (INR), and high sensitive C-reactive protein (hsCRP) were taken in to analysis. Surgical factors including the type of operation, anesthesia methods and duration of subdural drainage were investigated. The location of CSDH, the distance of midline shift and the density of subdural hematoma were extracted by reviewing the initial CT scan. The hematoma was classified as hypo-dense (<30 hounsfield unit [HU]), iso-dense (30–60HU), hyper-dense (>60HU) or mixed based on the density difference of the hematoma with the brain parenchyma. Patients with bilateral CSDH which had different types were arbitrated according to which reflected the greater bleeding tendency [15]. The postoperative hyperthermia and corticosteroids administration were also analyzed as the potential risk factors. The outcome was evaluated using the Glasgow outcome scale (GOS) [16]. GOS was assessed by analyzing the patient’s functional status at discharge. And in this study, the recurrence rate of CSDH was also investigated. The recurrence of CSDH was defined as the re-accumulation of hematoma in the ipsilateral subdural space seen on the CT scan which required operation within 90 days after the index surgery [17]. 3. Statistical analysis Variables considered in the statistical analysis included gender, age, GCS on admission, the main symptoms (headache, hemiparesis, gait disturbance, consciousness impairment, memory impairment, speech disturbance, seizure and visual disturbance), history of head trauma, antiplatelet or anticoagulant therapy, comorbidities (hypertension, diabetes mellitus and cardiovascular disease), location of CSDH, distance of midline shift, density of sub-

dural hematoma, laboratory data on admission (hemoglobin, platelet, blood glucose, prealbumin, albumin, PT, APTT, INR, hsCRP), type of operation, anesthesia methods, duration of subdural drainage, postoperative hyperthermia and administration of corticosteroids. For the statistical analysis, SPSS 22.0 (SPSS, INC; Chicago, USA) was used. We conducted a univariate analysis to evaluate the relaTable 1 Clinical characteristics of patients. Characteristics

Value, n (%)

Number of patients

1117

Gender, n (%) Male Female Age, mean ± SD (years)

885 (79.2) 232 (20.8) 68.25 ± 13.21

Initial GCS, n (%) 15 13–14 8–12 3–7

918 (82.2) 142 (12.7) 50 (4.5) 7 (0.6)

The main symptoms Headache Hemiparesis Gait disturbance Consciousness impairment Memory impairment Speech disturbance Seizure Visual disturbance History of head trauma Antiplatelet or anticoagulant drugs

620 (55.5) 317 (28.4) 76 (6.8) 30 (2.7) 53 (4.7) 16 (1.4) 3 (0.3) 2 (0.2) 722 (64.6) 130 (11.6)

Comorbidities, n (%) Hypertension DM Cardiovascular disease

361 (32.3) 240 (21.5) 111 (9.9)

Location of hematoma, n (%) Left Right Bilateral

489 (43.8) 348 (31.2) 280 (25.0)

Midline shift, n (%) 1 cm <1 cm

642 (57.5) 475 (42.5)

Hematoma density on CT Hypo-dense Iso-dense Hyper-dense Mixed dense

227 (20.3) 317 (28.4) 72 (6.4) 501 (44.9)

Laboratory investigation (mean ± SD) Hemoglobin Platelet Blood glucose Prealbumin Albumin PT APTT INR High Sensitive C-reactive protein

148.09 ± 28.57 153.83 ± 52.64 6.34 ± 2.17 270.22 ± 64.63 39.48 ± 4.26 11.00 ± 1.14 29.20 ± 2.79 1.01 ± 0.10 2.61 ± 4.66

Type of operation Urgent operation Elective operation Anesthesia methods General anesthesia Local anesthesia

270 (24.2) 847 (75.8)

Duration of subdural drainage 3 day >3 day 1–2 day Postoperative hyperthermia Administration of corticosteroids

338 356 423 421 314

588 (52.6) 529 (47.4)

(30.2) (31.9) (37.9) (37.7) (28.1)

Abbreviations: SD, standard deviation; GCS, Glasgow Coma Scale; DM, Diabetes mellitus; CT, computed tomography; PT, prothrombin time; APTT, activated partial thromboplastin time; INR, international normalized rate.

48

S. Wang et al. / Journal of Clinical Neuroscience 67 (2019) 46–51

tionship between each variable and the mortality using the chisquare test. Then, all significant factors were analyzed in a multivariate logistic regression. A p value <0.05 was considered to be statistically significant. The statistical power of the risk factors of mortality was calculated by R 3.5.2, The power computation employs sample size, p0 [Prob (Y = 1|X = 0): the probability of observing 1 for the outcome variable Y when the predictor X equals 0], p1 [Prob (Y = 1|X = 1): the probability of observing 1 for the outcome variable Y when the predictor X equals 1), and sig. level = 0.05, the procedure introduced by Demidenko is adopted here for computing the statistical power [18]. 4. Results 4.1. Demographic data A total of 1117 patients (885 male and 232 female) were included in our study (Table 1). The average age was [mean ± SD] 68.25 ± 13.21 years old, ranging from 19 to 94 years. An initial GCS score of 15 was noted in 918 patients (82.2%), 199 patients (17.8%) had initial GCS scores less than 15. The most common symptom was headache, followed by hemiparesis. Several symptoms were less common such as gait disturbance, consciousness impairment, memory impairment, speech disturbance, episodes of seizure and visual disturbance. 722 patients (64.6%) reported the history of head trauma and 130 patients (11.6%) had a history of anticoagulants usage. Hypertension, diabetes mellitus, cardiovascular disease was present as comorbidities in 361 (32.3%), 240 (21.5%), 111 (9.9%) patients, respectively. The hematoma was seen on the left side in 489 patients (43.8%), on the right side in 348 patients (31.2%), and bilateral in 280 patients (25%). The most common hematoma density on CT was the mixed density (501 patients, 44.9%). The hypo-dense type was observed in 227 patients (20.3%), iso-dense type in 317 patients (28.4%), and hyper-dense type in 72 patients (6.4%). All patients underwent BHD, of these, 588 patients received urgent operations after admis-

sion based on the CT scan and surgeons’ decision. In 421 patients, the temperature exceeded 38.5 degrees Celsius postoperatively and 314 patients were treated with corticosteroids. 4.2. Factors related to CSDH mortality Twenty of 1117 patients died (mortality rate: 1.8%, Table 2). Among these, 68 patients (6%) experienced a recurrence of CSDH. Univariate analysis (Table 3) indicated the gender was significantly related to the mortality of CSDH (p < 0.01). Preoperative factors including the symptom of consciousness impairment (p = 0.04), GCS on admission (p < 0.01), history of head trauma (p = 0.02), and location of the hematoma (p = 0.03) were significantly related to the mortality of CSDH. Laboratory tests such as blood platelet (p < 0.01), prealbumin (p = 0.046), albumin (p < 0.01), INR (p = 0.03), hsCRP (p < 0.01) were significantly related to mortality of CSDH. Postoperative hyperthermia (p < 0.01) was also considered to be associated with the mortality of CSDH. In addition, we found that age (p = 0.88), antiplatelet or anticoagulant therapy (p = 0.64), distance of midline shift (p = 0.82), density of subdural hematoma (p = 0.92), type of operation (p = 0.111), anesthesia methods (p = 0.93), duration of subdural drainage (p = 0.59) and administration of corticosteroids (p = 0.23) were not related to the mortality of CSDH. A multivariate logistic regression was also performed. The independent factors of hospital mortality in 1117 patients with CSDH were summarized in Table 4. We reported GCS on admission (OR, 0.08; 95% CI, 0.016–0.371; p < 0.01), bilateral CSDH (OR, 0.100; 95% CI, 0.018– 0.553; p < 0.01, left CSDH as reference), prealbumin (OR, 0.137; 95% CI, 0.028–0.676; p = 0.02), INR (OR, 12.959; 95% CI, 2.031–82.673; p < 0.01), and hsCRP (OR, 8.397; 95% CI, 2.014–35.005; p < 0.01) were significantly associated with mortality of CSDH. 4.3. Statistical power The statistical power of risk factors was calculated. The statistic power of GCS was 0.97, of INR was 0.84, and of

Table 2 Mortality description. Code

Gender

Age

GCS

The main symptom

History of head trauma

Location of hematoma

Midline shift

Hematoma density

Type of operation

Anesthesia methods

Causes of death

001 002

Male Female

83 94

12 7

No No

Left Bilateral

1 cm <1 cm

Iso Mixed

Elective Urgent

Local General

003 004 005

Male Male Male

74 59 76

15 15 14

Hemiparesis Consciousness impairment Headache Hemiparesis Headache

Yes Yes No

Bilateral Right Left

<1 cm 1 cm 1 cm

Mixed Hypo Mixed

Urgent Elective Elective

Local General Local

006

Female

47

14

Yes

Bilateral

1 cm

Hypo

Elective

Local

007 008

Female Female

27 82

15 13

Memory impairment Headache Headache

Intracerebral hemorrhage Pulmonary infection, heart failure Rebleeding, pulmonary infection Cardiac arrest Coronary heart disease, heart failure Acute cerebral infarction

No No

Bilateral Bilateral

<1 cm <1 cm

Mixed Iso

Urgent Elective

General Local

009 010

Female Male

60 64

14 15

Yes No

Right Bilateral

1 cm <1 cm

Hyper Iso

Urgent Elective

Local Local

011

Female

65

15

Headache Memory impairment Hemiparesis

No

Left

1 cm

Mixed

Elective

Local

012 013 014

Male Female Female

50 81 79

15 15 15

Headache Headache Headache

Yes No Yes

Left Right Bilateral

1 cm 1 cm <1 cm

Hyper Hypo Mixed

Urgent Elective Elective

Local Local Local

015 016 017 018 019

Male Female Male Male Male

81 81 73 80 69

13 11 15 12 7

Yes No No Yes No

Right Left Bilateral Bilateral Bilateral

1 cm 1 cm 1 cm <1 cm <1 cm

Iso Mixed Iso Mixed Mixed

Elective Urgent Elective Elective Urgent

Local Local General Local General

020

Male

72

13

Hemiparesis Headache Headache Headache Consciousness impairment Gait disturbance

Rebleeding, coagulation disorders Acute subdural hemorrhage Heart failure Acute respiratory distress syndrome Acute pulmonary embolism Chronic renal failure Tension pneumocephalus Acute subdural hemorrhage Cardiac arrest

No

left

1 cm

Hypo

Elective

Local

Intracerebral hemorrhage

Chronic renal failure Acute respiratory distress syndrome Rebleeding, heart failure Acute myocardial infarction

49

S. Wang et al. / Journal of Clinical Neuroscience 67 (2019) 46–51 Table 3 Univariate analysis of factors for mortality of CSDH.

Table 4 Logistic regression analysis of factors for mortality of CSDH.

Characteristics

Deceased

Survive

Number of patients

20

1097

Gender, n (%) Male Female

11 (55.0) 9 (45.0)

874 (79.7) 223 (20.3)

Age(years) 68 <68

13 (65.0) 7 (35.0)

731 (66.6) 366 (33.4)

P value

0.007*

0.878

0.000*

Initial GCS 15 3-14

9 (45.0) 11 (55.0)

909 (82.9) 188 (17.1)

The main symptoms Headache Hemiparesis Gait disturbance Consciousness impairment Memory impairment Speech disturbance Seizure Visual disturbance

11 (55.0) 4 (20.0) 1 (5.0) 2 (10.0) 2 (10.0) 0 (0.0) 0 (0.0) 0 (0.0)

609 (55.5) 313 (28.5) 75 (6.8) 28 (2.6) 51 (4.6) 16 (1.5) 3 (0.3) 2 (0.2)

0.963 0.402 0.746 0.041* 0.265 0.586 0.815 0.848

8 (40.0) 3 (15.0)

714 (65.1) 127 (11.6)

0.020* 0.636

Comorbidities Hypertension DM Cardiovascular disease

10 (50.0) 6 (30.0) 2 (10.0)

351 (32.0) 234 (21.3) 109 (9.9)

0.088 0.350 0.992

Location of hematoma, n (%) Left Right Bilateral

6 (30.0) 4 (20.0) 10 (50.0)

483 (44.0) 344 (31.4) 270 (24.6)

Midline shift, n (%) 1 cm <1 cm

12 (60.0) 8 (40.0)

630 (57.4) 467 (42.6)

Hematoma density on CT Hypo-dense Iso-dense Hyper-dense Mixed dense

4 5 2 9

223 (20.3) 312 (28.4) 70 (6.4) 492 (44.8)

Laboratory investigation Hemoglobin Platelet Blood glucose Prealbumin Albumin PT APTT INR High Sensitive C-reactive protein

142.10 ± 16.68 105.20 ± 25.96 6.39 ± 2.25 198.90 ± 95.42 33.28 ± 4.34 11.04 ± 1.30 30.26 ± 3.17 1.13 ± 0.18 10.95 ± 6.91

148.20 ± 28.73 154.72 ± 52.59 6.34 ± 2.17 271.52 ± 63.26 39.59 ± 4.18 11.00 ± 1.14 29.18 ± 2.78 1.00 ± 0.10 2.46 ± 4.47

Type of operation Urgent operation Elective operation

7(35.0) 13(65.0)

581 (53.0) 516 (47.0)

Anesthesia methods General anesthesia Local anesthesia

5(25.0) 15(75.0)

265 (24.2) 832 (75.8)

Duration of subdural drainage 3 day >3 day 1-2 day

4(20.0) 7(35.0) 9(45.0)

334 (30.4) 349 (31.8) 414 (37.7)

Postoperative hyperthermia Administration of corticosteroids

17(85.0) 8(40.0)

404 (36.8) 306 (27.9)

History of head trauma Antiplatelet or anticoagulant drugs

0.034*

0.818

0.923 (20.0) (25.0) (10.0) (4.0)

0.425 0.000* 0.973 0.046* 0.004* 0.169 0.062 0.031* 0.000* 0.111

0.930

0.593

0.000* 0.233

Abbreviations: GCS, Glasgow Coma Scale; DM, Diabetes mellitus; CT, computed tomography; PT, prothrombin time; APTT, activated partial thromboplastin time; INR, international normalized rate. * p < 0.05 was considered significant.

hsCRP was 0.97, which were all higher than 0.8. However, the statistic power of prealbumin and bilateral hematoma

Risk factors

B value

Initial GCS 2.562 Location of hematoma Left Ref Right 1.261 Bilateral 2.307 Prealbumin 1.987 INR 2.562 hsCRP 2.128

OR

95% CI

P value

0.077

0.016–0.371

0.001*

Ref 3.527 0.100 0.137 12.959 8.397

Ref 0.527–23.588 0.018-0.553 0.028–0.676 2.031–82.673 2.014–35.005

Ref 0.194 0.008* 0.015* 0.007* 0.003*

Abbreviations: OR, odds ratio; CI, confidence interval; GCS, Glasgow Coma Scale; INR, international normalized rate. * p < 0.05 was considered significant.

were 0.48 than 0.8.

and

0.53,

respectively,

which

were

less

5. Discussion Through the retrospective analysis of 1117 patients with CSDH, we reported that the hospital mortality of CSDH was related to the level of consciousness on admission, the location of hematoma, prealbumin, INR and the level of hsCRP. CSDH is a common form of intracranial hemorrhage [19]. BHD is the most common used technique [2,3]. A few studies have also found that middle meningeal artery embolization also has a positive therapeutic effect on CSDH and can be more effective than the conventional treatments [20,21]. The mortality of CSDH was reported to be up to 32%, while in our study, the mortality rate was 1.8% in the 1117 CSDH patients, which was lower than that of some previous studies [22–24]. A few risk factors for mortality of CSDH had been described in the literature, such as GCS, age, frailty, and coagulopathy. Table 5 summarizes studies that have investigated the risk factors of poor outcome in patients with CSDH. But there was a lack of predictive factors on CSDH mortality in terms of the laboratory tests on admission. In the presenting study, the preoperative laboratory results including hemoglobin, platelet, blood glucose, prealbumin, albumin, PT, APTT, INR and hsCRP were analyzed. The hemoglobin, blood glucose, PT and APTT were not associated with the mortality of CSDH. The multivariate analysis revealed that prealbumin, INR and hsCRP were independent predictors on the mortality of CSDH. To the best of our knowledge, this was the first study to investigate the association between early markers of malnutrition (i.e. prealbumin) and the outcome in CSDH patients. Prealbumin, a plasma protein synthesized by the liver, is an important marker of the nutritional status. A decline of prealbumin reflects the possible inadequate intake of nutrients. Baltazar [25] found the malnutrition on admission was associated with poor outcomes after severe traumatic brain injury (TBI), in which study, both initial albumin and prealbumin in the survival group were significantly higher than those in the death group. The outcome of patients with severe TBI may improve if the intrinsic correction of the albumin level was achieved. Attenuated immune function can be a possible reason of the poor prognosis seen in patients with malnutrition. This can be also be the reason why the complications such as pneumonia are common in the elderly. The extravasation of albumin is associated with the progression of CSDH [26]. Fujisawa et al. [26] reported that the hematoma density shown on CT scans correlates with the level of the protein in the hematoma. Therefore, the exudation of the serum protein (i.e. albumin) is associated with the change of the hematoma density. The level of protein in patients with malnutrition is relatively low, which may affect the absorption and the pro-

50

S. Wang et al. / Journal of Clinical Neuroscience 67 (2019) 46–51

Table 5 Published studies with factors of poor outcome in CSDH. Authors & Year

No. of Pts Poor outcome

Kitya et al., 2018 Shimizu et al., 2018 Hussain et al., 2017

205 211 267

Katsigiannis et al., 2017

197

Kwon et al., 2017

154

Agawa et al., 2016 Choi et al., 2016

368 502

Leroy et al., 2015

140

Hyeong et al., 2015

130

Rovlias et al., 2015

986

Ishibashi et al., 2011 Miranda et al., 2011

92 209

Ramachandran et al., 2006 647

Older age and low admission GCS score frailty Age at admission, the median length of hospital stay, number of comorbidities, GCS at discharge and discharge destination age, pre-operative neurological status, surgical technique and recurrence age, midline shift, haematoma thickness, admission GCS and the presence of a motor deficit Bilateral hematoma Preoperative high-sensitivity C-reactive protein age > 75 years, residual hematoma thickness > 14 mm, GCS < 15 Age > 75 years, severe preoperative neurologic deficits, the density of hematoma neurological status on admission, age, brain atrophy, thickness and density of hematoma, subdural accumulation of air, and antiplatelet and anticoagulant therapy Burr hole drainage alone The inhospital mortality rate was solely related to presenting neurological condition whereas the postdischarge mortality rate was related to age and discharge to another nursing facility (as opposed to home) Age, GCS at presentation, and associated illnesses such as cardiac and renal problems

Consistent with the previous reports, we found that the bilateral CSDH and GCS on admission were associated with the poor clinical outcome [6,34,35]. Agawa [34] reported that bilateral CSDH had a higher mortality rate compared with unilateral cases. Under such a scenario, the brain herniation can be significantly more possible than in the unilateral hematoma. Huang et al. [35] found that focal neurological deficits were less common in patients with bilateral CSDH, and the bilateral CSDH was found to be thicker than unilateral hematoma, which can reduce the autonomic capacity to cushion the increased intracranial pressure. These factors can confound the diagnosis and delay the treatment which may result in a poor outcome in CSDH. The relationship between INR and poor prognosis of CSDH has also been reported [36]. As is well-known, INR presents the coagulation function in patients. Oden et al. [37] reported that the mortality rate of patients treated with oral anticoagulants was the lowest at an INR value of 2.2–2.3 and the mortality rate increased substantially at higher INR levels. Yasuda et al. [36] also suggested that when INR values were greater than 1.25, the CSDH mortality would increase. In our study, multivariate analysis showed that a high INR was associated with the in-hospital mortality of CSDH. Thus, surgery should be postponed in patients with high INR. In addition, we found that gender, age, comorbidities, duration of subdural drainage were not related to the mortality of CSDH. This study has several limitations. This is a retrospective study, in which the selection bias is inevitable. This study is a singlecenter study, which may affect the generalization of the results. In addition, there was a lack of sufficient statistical power of a significant difference in some of the factors. Therefore, these particular results should be regarded somewhat conservatively, and a further prospective study with a large number of cases is merited to confirm these findings.

Abbreviations: GCS, Glasgow Coma Scale.

6. Conclusion gression of the subdural hematoma. This may subsequently affect the poor prognosis. CSDH is a common neurosurgical disease in the elderly. Elderly CSDH patients tend to have poor prognosis. It has been a routine to perform the preoperative examinations, however, the indicators such as the prealbumin and albumin, which are early markers of nutritional status, were easily omitted. The prealbumin is more sensitive in assessing the recent nutritional status. In our study, patients with low prealbumin have higher mortality in CSDH. Both Prealbumin and albumin tests are also clinically available, and through dynamic monitoring, we can provide patients with appropriate nutrition to improve their status. Therefore, we suggest that low prealbumin levels should be corrected in CSDH to reduce the risk of mortality in these patients. The hsCRP is a stable biomarker of inflammatory processes [27]. A few studies have found that in patients with ischemic stroke and head trauma, the mortality rate of patients with high hsCRP is higher than that of patients with low hsCRP [28–30]. Choi et al. [31] have investigated the relationship between the preoperative hsCRP and the mortality of CSDH, he found that the serum hsCRP was the only laboratory factor that predicted in-hospital mortality and morbidity after BHD in CSDH. In our study, both univariate and multivariate logistic analyses revealed that hsCRP was positively correlated with the mortality rate. A high level of inflammatory factors such as interleukin (IL)-6 and IL-8 are associated with the recurrence of the CSDH [32]. The prevalence of the CSDH is high in the elderly, chronic and complex comorbidities among whom are somewhat common. hsCRP is an important inflammatory indicator, which can be used to evaluate the severity of the comorbidities [33]. An elevation of the hsCRP is associated with the progression and recurrence of the CSDH.

The presenting study drew the conclusion that the GCS on admission, bilateral CSDH, prealbumin, INR, and hsCRP were related to CSDH mortality. Malnutrition status, which is presented by the level of prealbumin, can be an important factor which may affect the treatment timing and strategy in CSDH. Declaration of Competing Interest None. Acknowledgment None. Funding This work was supported by SMC-Star Award for Young Scholars (B), and Shanghai Talent Development Fund (2018077). References [1] Kolias AG, Chari A, Santarius T, et al. Chronic subdural haematoma: modern management and emerging therapies. Nat Rev Neurol 2014;10:570–8. [2] Lind CR, Lind CJ, Mee EW. Reduction in the number of repeated operations for the treatment of subacute and chronic subdural hematomas by placement of subdural drains. J Neurosurg 2003;99:44–6. [3] Ramachandran R, Hegde T. Chronic subdural hematomas – causes of morbidity and mortality. Surg Neurol 2007;67:367–72. discussion 372-363. [4] Miranda LB, Braxton E, Hobbs J, et al. Chronic subdural hematoma in the elderly: not a benign disease. J Neurosurg 2011;114:72–6. [5] Uno M, Toi H, Hirai S. Chronic subdural hematoma in elderly patients: is this disease benign? Neurol Med Chir (Tokyo) 2017;57:402–9.

S. Wang et al. / Journal of Clinical Neuroscience 67 (2019) 46–51 [6] Kitya D, Punchak M, Abdelgadir J, et al. Causes, clinical presentation, management, and outcomes of chronic subdural hematoma at Mbarara Regional Referral Hospital. Neurosurg Focus 2018;45. E7. [7] Rovlias A, Theodoropoulos S, Papoutsakis D. Chronic subdural hematoma: surgical management and outcome in 986 cases: a classification and regression tree approach. Surg Neurol Int 2015;6:127. [8] Leroy HA, Aboukais R, Reyns N, et al. Predictors of functional outcomes and recurrence of chronic subdural hematomas. J Clin Neurosci 2015;22:1895–900. [9] Kwon CS, Al-Awar O, Richards O, et al. Predicting prognosis of patients with chronic subdural hematoma: a new scoring system. World Neurosurg 2018;109:e707–14. [10] Shimizu K, Sadatomo T, Hara T, et al. Importance of frailty evaluation in the prediction of the prognosis of patients with chronic subdural hematoma. Geriatr Gerontol Int 2018;18:1173–6. [11] Gelabert-Gonzalez M, Iglesias-Pais M, Garcia-Allut A, et al. Chronic subdural haematoma: surgical treatment and outcome in 1000 cases. Clin Neurol Neurosurg 2005;107:223–9. [12] Adhiyaman V, Asghar M, Ganeshram KN, et al. Chronic subdural haematoma in the elderly. Postgrad Med J 2002;78:71–5. [13] van Havenbergh T, van Calenbergh F, Goffin J, et al. Outcome of chronic subdural haematoma: analysis of prognostic factors. Br J Neurosurg 1996;10:35–9. [14] Delgado PD, Cogolludo FJ, Mateo O, et al. Early prognosis in chronic subdural hematomas. Multivariate analysis of 137 cases. Rev Neurol 2000;30:811–7. [15] Stanisic M, Pripp AH. A reliable grading system for prediction of chronic subdural hematoma recurrence requiring reoperation after initial burr-hole surgery. Neurosurgery 2017;81:752–60. [16] Teasdale G, Jennett B. Assessment of coma and impaired consciousness. A practical scale. Lancet 1974;2:81–4. [17] Hori YS, Ebisudani Y, Aoi M, et al. Elevated serum fibrinogen degradation products on admission is a novel predictive factor for recurrence of chronic subdural hematoma. World Neurosurg 2018;118:e753–7. [18] Demidenko E. Sample size determination for logistic regression revisited. Stat Med 2007;26:3385–97. [19] Motiei-Langroudi R, Stippler M, Shi S, et al. Factors predicting reoperation of chronic subdural hematoma following primary surgical evacuation. J Neurosurg 2017:1–8. [20] Ban SP, Hwang G, Byoun HS, et al. Middle meningeal artery embolization for chronic subdural hematoma. Radiology 2018;286:992–9. [21] Link TW, Boddu S, Paine SM, et al. Middle meningeal artery embolization for chronic subdural hematoma: a series of 60 cases. Neurosurgery 2018. [22] Ernestus RI, Beldzinski P, Lanfermann H, et al. Chronic subdural hematoma: surgical treatment and outcome in 104 patients. Surg Neurol 1997;48: 220–5.

51

[23] Mellergard P, Wisten O. Operations and re-operations for chronic subdural haematomas during a 25-year period in a well defined population. Acta Neurochir (Wien) 1996;138:708–13. [24] Merlicco G, Pierangeli E, di Padova PL. Chronic subdural hematomas in adults: prognostic factors. Analysis of 70 cases. Neurosurg Rev 1995;18:247–51. [25] Baltazar GA, Pate AJ, Panigrahi B, et al. Malnutrition as measured by albumin and prealbumin on admission is associated with poor outcomes after severe traumatic brain injury. Am Surg 2015;81:E61–63. [26] Fujisawa H, Nomura S, Tsuchida E, et al. Serum protein exudation in chronic subdural haematomas: a mechanism for haematoma enlargement? Acta Neurochir (Wien) 1998;140:161–5. discussion 165–166. [27] Pober JS, Cotran RS. Cytokines and endothelial cell biology. Physiol Rev 1990;70:427–51. [28] Huang Y, Jing J, Zhao XQ, et al. High-sensitivity C-reactive protein is a strong risk factor for death after acute ischemic stroke among Chinese. CNS Neurosci Ther 2012;18:261–6. [29] Sogut O, Guloglu C, Orak M, et al. Trauma scores and neuron-specific enolase, cytokine and C-reactive protein levels as predictors of mortality in patients with blunt head trauma. J Int Med Res 2010;38:1708–20. [30] Hergenroeder G, Redell JB, Moore AN, et al. Identification of serum biomarkers in brain-injured adults: potential for predicting elevated intracranial pressure. J Neurotrauma 2008;25:79–93. [31] Choi JJ, Kim HS, Lee KC, et al. Prediction of in-hospital mortality and morbidity using high-sensitivity C-reactive protein after burr hole craniostomy. J Anesth 2016;30:956–60. [32] Frati A, Salvati M, Mainiero F, et al. Inflammation markers and risk factors for recurrence in 35 patients with a posttraumatic chronic subdural hematoma: a prospective study. J Neurosurg 2004;100:24–32. [33] Nouvenne A, Ticinesi A, Lauretani F, et al. The prognostic value of highsensitivity C-reactive protein and prealbumin for short-term mortality in acutely hospitalized multimorbid elderly patients: a prospective cohort study. J Nutr Health Aging 2016;20:462–8. [34] Agawa Y, Mineharu Y, Tani S, et al. Bilateral chronic subdural hematoma is associated with rapid progression and poor clinical outcome. Neurol Med Chir (Tokyo) 2016;56:198–203. [35] Huang YH, Yang KY, Lee TC, et al. Bilateral chronic subdural hematoma: what is the clinical significance? Int J Surg 2013;11:544–8. [36] Yasuda CL, Morita ME, Nishimori FY, et al. Chronic subdural hematoma: study of 161 patients and the relationship with coagulation abnormalities. Arq Neuropsiquiatr 2003;61:1011–4. [37] Oden A, Fahlen M. Oral anticoagulation and risk of death: a medical record linkage study. BMJ 2002;325:1073–5.