Keywords Acute subdural hematoma - outcome predictive factors - operated acute SDH - Glasgow
outcome score
Introduction
Acute subdural hematoma (aSDH) constitutes 50% to 60% of all subdural hematomas (SDHs).[1 ] They are mostly traumatic[2 ] but may occur spontaneously (or after minor trauma) in patients on chronic anticoagulation
therapy, bleeding diathesis,[3 ] or after rupture of a posterior communicating artery aneurysm.[4 ] Clinically, an acute SDH occurs within 3 days of injury.
Classically, SDHs are caused by rupture of dural bridging veins that lie in subdural
space and drain cortical blood into dural sinuses ([Fig. 1 ]). Acute SDH may also result from bleeding from superficial cortical vessels.[1 ] It may have an underlying burst lobe (complex of SDH and damaged brain cortex).
The associated brain damage is mainly caused by local ischemia due to mass effect,
direct brain injury, or hampered venous outflow.
Fig. 1 Intraoperative image of acute subdural hematoma (SDH) after frontotemporoparietal
craniectomy and incising dura. Note the hematoma of acute SDH overlying brain surface
Classically, acute SDH on noncontrast CT head presents as a crescent-shaped homogeneously
hyperdense extra-axial collection, spreading diffusely over the brain convexity of
affected hemisphere. It is hyperdense relative to the cortex. ([Fig. 2 ]).
Fig. 2 Axial section of noncontrast CT scan brain showing crescent-shaped homogeneously
hyperdense extra-axial collection in left frontotemporoparietal region.
Rapid deterioration may occur, especially if cortical arteries are ruptured. Prognosis
is usually poor in patients of acute SDH, warranting rapid hematoma evacuation and
stopping the cause of bleeding.[5 ]
Ischemic cerebral damage is a very critical prognosticating factor in the pathology
of SDH and may be due to the mass effects of the hematoma and raised intracranial
pressure (ICP), leading to compromised cerebral perfusion pressure (CPP). Timely clot
evacuation (within 4 hours) generally results in significantly improved neurological
outcome.[6 ] The accompanying brain damage besides acute SDH is responsible for poor neurological
function after injury.
Operative outcome of acute SDH depends on multitude of factors, namely, multiple comorbidities,
coexisting brain injury, cerebral ischemia, antithrombotic therapy, geriatric age
group, etc. A correct identification of outcome predictive factors is crucial for
appropriate neurosurgical intervention.
In this article, we analyzed the clinical, surgical, and outcome predictive factors
of operated cases of acute SDH at our tertiary care center. Most of the studies done
earlier are on both conservative and surgically managed patients. In this study, we
studied only the operated cases of acute SDH to get their clinical picture and outcome
predictive factors for better management of neurosurgical patients.
Materials and Methods
This retrospective study includes 114 patients who underwent surgery for acute SDH
in an emergency setup in the Department of Neurosurgery of IMS BHU, Varanasi, India,
a tertiary care center, between 1 August 2018 and 1 November 2019.
This is a retrospective cohort study in two phases: a cross-sectional phase where
the patients included in the sample were evaluated for the following described variables,
and a follow-up phase until hospital discharge or death. Each of the patients were
evaluated in terms of age, sex, mode of injury, localization of hematoma, clinical
presentation, comorbidity, severity of injury, best motor response, CT findings, and
Glasgow outcome scale (GOS) at discharge. The outcome was also evaluated by further
making a dichotomized group using GOS in death/dependent (1–3) versus independent
(4–5).
The inclusion criteria were all patients admitted to the trauma center, IMS BHU, Varanasi,
India and submitted to decompressive craniectomy and surgical evacuation of acute
SDH in the Neurosurgery Department between 1 August 2018 and 1 November 2019. Patients
who underwent surgery in the first 4 hours of reporting to trauma center were included
in the study. The exclusion criterion was unknown data concerning the study variables
and conservatively managed acute SDH cases and those who did not give consent for
surgery. One hundred fourteen patients were included in the study.
Surgical technique used was trauma flap creation, followed by decompressive craniectomy
and acute SDH evacuation and then lax duraplasty using pericranial graft in watertight
fashion, so that a dural expansion is obtained for potential brain parenchyma swelling.
Bone was placed in subcutaneous abdominal pouch, which was later utilized for autologous
cranioplasty. In patients on anticoagulation therapy, fresh frozen plasma was used
to correct the international normalized ratio (INR), and in those under antiplatelet
therapy, pools of platelets were used according to surgeon preference. Postoperative
ICU care was provided. Informed consent was taken from all operated cases by patient
or patient attendants (in cases in which patient was not able to consent, namely,
poor Glasgow coma scale [GCS] score)
Data was collected retrospectively by analyzing medical and surgical records of 114
operated patients submitted in the medical record department, IMS BHU, Varanasi, India,
between 1 August 2018 and 1 November 2019. Statistical tests were done using GraphPad
Prism version 8.3.0 software. For dichotomous evaluation, the Pearson and Fisher Chi-square
test and unpaired t -test were used according to data analyzed. For percentage or proportion data, Chi-square
test was used along with unpaired t -test and one-way ANOVA test for testing significance level for difference of means.
p < 0.05 was deemed statistically significant.
Results
Incidence of Age
In the present study, the largest number of patients operated for acute SDH belong
to the age group of 40 to 60 years (n = 45, 39.48%), followed by the 20 to 40 years age group (n = 41, 35.96%), > 60 years (n = 24, 21.05%) and < 20 years (n = 4, 3.51%), respectively ([Table 1 ]). Out of 114 patients, 89 patients (78%) were males and 25 patients (22%) were females
([Table 1 ]). Male to female ratio is 3.56:1.
Table 1
Age, sex distribution, mode of injury, severity of injury, best motor response, clinical
presentation, and comorbidities
Number of patients
%
Abbreviations: GCS, Glasgow coma scale; RTA, road traffic accident.
Age in years
< 20
4
3.51
20–40
41
35.96
40–60
45
39.48
> 60
24
21.05
Total
114
100
Sex
Male
89
78
Female
25
22
Mode of injury
RTA
93
81.58
Assault
4
3.51
Fall from height
15
13.16
Spontaneous/trivial trauma
2
1.75
Severity (GCS)
Mild (14–15)
7
6.14
Moderate (9–13)
21
18.42
Severe (3–8)
86
75.44
Best motor response
M6
2
1.75
M5
28
24.56
M4
21
18.42
M3
28
24.56
M2
30
26.32
M1
5
4.39
Signs/symptoms
Lucid interval
2
1.75
Bradycardia
10
8.77
Headache/vomiting
32
28.07
Altered sensorium
112
98.25
Neurodeficit
30
26.31
Anisocoria
20
17.54
Pupillary changes (B/L)
28
24.56
Black eye
13
11.40
Battle sign
10
8.77
Hemotympanum
4
3.51
Hypoxia/hypotension
5
4.39
Medical history
Hypertension
32
28.07
On anticoagulant therapy
20
17.54
Cardiovascular disease
14
12.28
Diabetes
20
17.54
Psychiatry illness
3
2.63
Respiratory disease
20
17.54
Gastrointestinal disorders
1
0.88
Seizure disorder
2
1.75
Cancer
1
0.88
Stroke (previous history)
5
4.38
Dementia
6
5.26
Previous head injury
2
1.75
Mode of Injury
Mode of injury in majority of patients in this study was road traffic accidents (RTA),
accounting for 93 patients (81.58%); 15 patients (13.16%) suffered a fall from height,
four patients (3.51%) experienced assault, and two patients (1.75%) suffered a spontaneous
fall following trivial trauma ([Table 1 ]).
Severity of Injury
In terms of severity of head injury, which is expressed in terms of GCS scores, majority
of patients (n = 86, 75.44%) were afflicted with a severe head injury and had GCS score of 3 to
8. Twenty-one patients (18.42%) were afflicted a moderate head injury and had GCS
score of 9 to 13, and seven patients were afflicted with a mild head injury (6.14%)
and had GCS score of 14 to 15 ([Table 1 ]).
Best Motor Response
In this study, the most common best motor response was M2 (n = 30, 26.32%) followed by M5 (n = 28, 24.56%), M3 (n = 28, 24.56%), M4 (n = 21, 18.42%), M1 (n = 5, 4.39%) and M6 (n = 2, 1.75%), respectively ([Table 1 ]).
Clinical Presentation
In this study, most of the patients presented with altered sensorium (n = 112, 98.25%). Lucid interval was present in two patients(1.75%), bradycardia in
10 patients(8.77%), headache/vomiting in 32 patients(28.07%), neurodeficit in 30 patients
(26.31%), anisocoria in 20 patients (17.54%), bilateral pupillary changes in 28 patients(24.56%),
Battle’s sign in 10 patients (8.77%), hemotympanum in four patients (3.51%), black
eye in 13 patients (11.40%), and hypoxia or hypotension in five patients (4.39%) ([Table 1 ]).
Site of aSDH
In the present study, the most common site of acute SDH was frontotemporoparietal
(n = 92, 80.70%), followed by frontotemporal (n = 9, 7.89%), temporoparietal (n = 3, 2.63%), frontoparietal (n = 3, 2.63%), temporal (n = 2, 1.75%) and parietal (n = 1, 0.88%), respectively ([Table 2 ]).
Table 2
Site, side and CT findings of aSDH
Site
Number of patients
%
Abbreviations: aSDH, acute subdural hematoma; MLS, midline shift.
Fronto-temporo-parietal
92
80.70
Posterior fossa
4
3.52
Temporoparietal
3
2.63
Frontotemporal
9
7.89
Temporal
2
1.75
Parietal
1
0.88
Frontoparietal
3
2.63
Side of aSDH
Left
65
57
Right
49
43
CT findings
Clot thickness (mm)
< 10
2
1.75
10–15
28
24.56
15–20
49
42.98
≥ 20
35
30.70
MLS > 5 mm
105
92.10
Herniation
66
57.89
Side of aSDH
In our study, out of 114 patients, 65 patients (57%) had acute SDH on left side, 49
patients (43%) on right side ([Table 2 ]).
CT Findings
CT brain findings of acute SDH and its effects are as follows: Clot thickness less
than 10 mm was found in two patients (1.75%), 10 to 15 mm was found in 28 patients
(24.56%), 15 to 20 mm was found in 49 patients (42.98%), and more than or equal to
20 mm was found in 35 patients (30.70%).
Midline shift was noted in 105 patients (92.10%). Brain herniation was present in
66 patients (57.89%) ([Table 2 ]).
Comorbidity
In the present study, comorbidities associated with acute SDH patients were hypertension
(n = 32, 28.07%), diabetes (n = 20, 17.54%), on anticoagulant therapy (n = 20, 17.54%), respiratory disease (n = 20, 17.54%), cardiovascular disease (n = 14, 12.28%), dementia (n = 6, 5.26%), stroke (n = 5, 4.38%), psychiatry illness (n = 3, 2.63%), seizure disorder (n = 2, 1.75%), previous head injury (n = 2, 1.75%), gastrointestinal disorders (n = 1, 0.88%), and cancer (n = 1, 0.88%) ([Table 1 ]).
Glasgow Outcome Score (GOS)
In the present study, 30 patients (26.32%) recovered well with GOS score of 5 (good
recovery), 24 patients (21.05%) had moderate disability with GOS = 4, 23 patients
(20.17%) had severe disability with GOS = 3, 25 patients (21.93%) were in persistent
vegetative state GOS = 2, and 12 patients (10.53%) was dead with GOS = 1 ([Table 3 ]).
Table 3
GOS
GOS
Number of patients
%
Abbreviation: GOS, Glasgow outcome scale.
5 (good recovery)
30
26.32
4 (moderate disability)
24
21.05
3 (severe disability)
23
20.17
2 (persistent vegetative state)
25
21.93
1 (Dead)
12
10.53
Variables of Operated Acute SDH cases–Glasgow Outcome Scale (GOS)
In this study, we found GCS at admission (p < 0.0001), severity of injury(p < 0.0001), pupillary changes (p < 0.0001), and best motor response (p < 0.0001) were significantly (p < 0.05) associated with outcome following surgery for acute SDH. In contrast, age
and gender of patients were not found to be significantly associated with outcome
following surgery. ([Table 4 ])
Table 4
Variables of operated aSDH cases–GOS
Characteristics
1 (n = 12)
2 (n = 25)
3 (n = 23)
4 (n = 24)
5 (n = 30)
p -Value
Abbreviations: aSDH, acute subdural hematoma; GCS, Glasgow outcome scale; FFH, fall
from height; RTA, road traffic accident.
a Statistically significant (<0.05).
Age
49.83
(12.22)
50.52
(13.68)
48.48
(11.88)
44.5
(17.40)
40.42
(16.55)
0.0926
Gender (%)
0.1966
Male
9 (75%)
18 (72%)
16 (69.56%)
23 (95.83%)
23 (76.67%)
Female
3 (25%)
7 (28%)
7 (30.44%)
1 (4.17%)
7 (23.33%)
Mode of injury
RTA
FFH
Assault
Spontaneous
10 (83.34%) 0
1 (8.33%)
1 (8.33%)
20 (80%)
4 (16%)
1 (4%)
0
18(78.26%)
5 (21.74%)
19 (79.16%)
3 (12.5%)
1 (4.17%)
1 (4.17%)
26 (86.67%)
3 (10%)
1 (3.33%)
0.5995
Best motor response
1.75
(0.62)
2.24
(0.60)
3.13
(0.62)
3.67
(0.82)
4.93
(0.52)
< 0.0001a
Pupillary changes
12 (100%)
24 (96%)
10 (43.47)
2 (8.33%)
0
< 0.0001a
Severity of injury
< 0.0001a
Mild
0
0
0
0
7 (23.33%)
Moderate
0
0
0
2 (8.33%)
19 (63.34%)
Severe
12 (100%)
25 (100%)
23 (100%)
22 (91.67%)
4 (13.33%)
GCS at admission
3.75
(0.62)
4.52
(0.96)
5.69
(1.06)
6.71
(1.90)
11
(2.49)
< 0.0001a
Variables of Operated Acute SDH cases - Glasgow Outcome Scale Categories (1–3 versus
4–5)
On evaluating dichotomized outcome classified in death/dependent (1–3) versus independent
(4–5), we found significant association of age (p = 0.0089), best motor response (p < 0.0001), pupillary changes (p < 0.0001), severity of injury (p < 0.0001), and GCS at admission with outcome following surgery. Gender and mode of
injury were not significantly associated ([Table 5 ])
Table 5
Variables of operated aSDH cases–GOS categories (1–3 versus 4–5)
Characteristics
1–3 (n = 60)
4–5 (n = 54)
p -Value
Abbreviations: aSDH, acute subdural hematoma; GCS, Glasgow outcome scale; FFH, fall
from height; RTA, road traffic accident.
a Statistically significant (<0.05).
Age
49.6
(12.55)
42.23
(16.89)
0.0089a
Gender (%)
0.0815
Male
Female
43 (71.67%)
17 (28.33%)
46 (85.18%)
8 (14.82%)
Mode of injury
0.9439
RTA
FFH
Assault
Spontaneous
48 (80%)
9 (15%)
2 (3.33%)
1 (1.67%)
45 (83.33%)
6 (11.12%)
2 (3.70%)
1 (1.85%)
Best motor response
2.48
(0.81)
4.37
(0.92)
<0.0001a
Pupillary changes
46 (76.67%)
2 (3.70%)
< 0.0001a
Severity of injury
< 0.0001a
Mild
Moderate
Severe
0
0
60 (100%)
7 (12.96%)
21 (38.89%)
26 (48.15%)
GCS at admission
4.82
(1.20)
9.09
(3.10)
<0.0001a
Discussion
This study included only operated cases of acute SDH. The most common age group operated
upon in this study was 40 to 60 years (n = 45, 39.48%), closely followed by 20 to 40 years (n = 41, 35.96%), thus affecting the most productive working population. Acute SDH is
more common in geriatric patients due to age-related cerebral atrophy, increased subdural
space, stretching of bridging veins, and higher risk for trauma due to gait or orthopedic-related
problems.[7 ]
[8 ] As the age of population increases, acute SDH also increases in its frequency.[9 ]
In this study, males (n = 89, 78%) were affected more than females (n = 25, 22%) with male is to female ratio of 3.56:1. Studies in operated acute SDH
have shown male predominance.[10 ] It may be attributable to our social culture in which most of the females are nonworking
and so are not susceptible to external works.
In this study, RTA (n = 93, 81.58%) was the most common mode of injury. Studies with comatose patients
have demonstrated RTA as the mechanism of injury in most acute SDH cases.[11 ] Most acute SDHs are caused by RTA and falls.[12 ] Frequency of other mechanisms is very less. RTA is most frequent in the 15 to 30
years age group. Injury due to falls is most frequent in the 45 to 80 years age group.[13 ]
Most of the operated cases of acute SDH were severe (GCS 3–8) in traumatic brain injury
(TBI) severity scale (n = 86, 75.44%). M2 (n = 30, 26.32%) was the most common best motor response in this study. It is similar
to earlier studies having 35% to 80% of patients with acute SDH, presenting with GCS
score of 8 or less.[14 ]
[15 ]
The most common clinical presentation was altered sensorium (n = 112, 98.25%), followed by headache/vomiting (n = 32, 28.07%) which may be attributed to poor GCS score and raised ICP due to acute
SDH mass effect and secondary brain injury.
The most common site of operated acute SDH cases was frontotemporoparietal region
(n = 92, 80.70%) with left side predilection (n = 65, 57%). It is usually hemispheric over convexity of brain. In this study, clot
thickness of 15 to 20 mm (n = 49, 42.98%) was found most commonly. Most of the operated cases had a midline shift
(n = 105, 92.10%) and herniation (n = 66, 57.89%) before surgery.
The most common comorbidity associated with operated acute SDH was hypertension (n = 32, 28.07%). Hypertension, as the most common comorbidity in acute SDH, is also
seen in other study.[16 ]
In this study, a GOS score of 5 was most common (n = 30, 26.32%), followed by a GOS score of 2 (n = 25, 21.93%). There were 12 (10.53%) mortalities in this study. Many studies had
shown mortality between 30 to 60% in acute SDH patients with a GCS score of 3–15 [17 ]
[18 ] or mortality between 55 to 70%with a GCS score of 8 or less.[11 ]
[19 ] However, a GOS score of 2 (persistent vegetative state) was present in 21.93% (n = 25), showing high-morbidity rate in this study.
On analyzing variables on GOS, we found that GCS at admission, severity of injury,
pupillary changes, and best motor response were significantly (p < 0.05) associated with outcome following surgery for acute SDH. In contrast, age
and gender of patients were not found to be significantly associated with outcome
following surgery. Studies had shown that patient with age more than 65 years are
associated with poorer outcomes.[19 ]
Studies have shown gender influence on prognosis in TBI.[20 ] Poorer outcome had been reported in females surviving severe TBI when compared with
males.[21 ] In this study, gender is not associated with poor outcome in the GOS or dichotomized
analysis.
Acute SDH due to spontaneous mechanism of injury was a poor prognostic factor in various
studies. Oral anticoagulation therapy and disorders of bleeding diathesis are the
most common causes related to spontaneous aSDH.[22 ]
[23 ]
[24 ]
Nevertheless, a reliable GCS score may be hard to get in acute situations, according
to medical sedation, paralysis and/or intoxication.[25 ] Motor response is the most reproductive element in trauma patients, as the other
responses may be blunt.[26 ] Keeping these limitations in mind, the GCS in this study was evaluated at the time
of hospital admission. GCS at admission was significantly associated with outcome
in both GOS and dichotomized analysis.
Pupillary reactivity changes or abnormalities reflect brainstem compression or lesion
and are indicative of poorer outcome.[27 ] Many studies have shown that pupillary reactivity when compared with GCS is more
reliable after TBI, as it is less effected by sedation and paralysis[28 ]
However, on analyzing dichotomized outcome, namely, death/dependent (1–3) versus independent
(4–5), we found that, in addition to above mentioned variables, age was also significantly
(p < 0.05) associated with outcome following surgery.
Some studies recommend prompt evacuation of acute SDH with indications for surgery,
as this reduces brain damage resulting from secondary ischemic injury, due to mass
effect and/or raised ICP.[5 ]
[29 ]
[30 ] However, most studies done to evaluate correlation between early surgery and outcome
have not shown a correlation with outcome.[11 ]
[14 ]
[18 ]
[31 ]
[32 ]
Cognitive and neurological changes following large craniectomy have been attributed
to many factors including atmospheric pressure, venous return obstruction, cerebral
blood flow, metabolic changes and cerebrospinal fluid (CSF) changes. There is reduction
in cerebral blood flow due to atmospheric pressure and reduced cerebral metabolic
rate.[33 ] This may certainly affect the outcome, producing a bias in the presented results.[33 ]
[34 ]
Conclusion
Acute SDH is a frequent neurosurgical condition responsible for significant morbidity
and mortality. It affects the working productive age group (20–60 years), with males
more affected than females. The most common associated comorbidity was hypertension.
RTA was the most common mode of injury. Altered sensorium is the most common clinical
presentation. It most commonly affects the frontotemporoparietal region, with greater
predilection to the left side. GCS at admission, severity of injury, pupillary changes,
and best motor response were significantly (p < 0.05) associated with outcome following surgery for acute SDH. In contrast, age
and gender of patients were not found to be significantly associated with outcome
following surgery. However, when comparing dichotomized outcome, death/dependent (1–3)
versus independent (4–5), age was also significantly (p < 0.05) associated with outcome following surgery.
