Bilateral Hemicraniectomy in Non-Penetrating Traumatic Brain Injury

Abstract

Traumatic brain injury is a heterogeneous entity that encompasses both surgical and non-surgical conditions. Surgery may be indicated with traumatic lesions such as hemorrhage, fractures, or malignant cerebral edema. However, the neurological exam may be clouded by the effects of medications administered in the field, systemic injuries, and inaccuracies in hyperacute prognostication. Typically, neurological injury is considered irreversible if diffuse loss of grey/white matter differentiation or if brainstem hemorrhage (Duret hemorrhage) exists. We aim to characterize a cohort of patients undergoing bilateral hemicraniectomy for severe traumatic brain injury. A retrospective consecutive cohort of adult patients undergoing craniectomy for trauma was established between the dates of January 2008 and November 2011. The primary outcome of the study was in-hospital mortality. Secondary outcomes were ICU length of stay, surgical complications, and Glasgow Outcome Score at most recent follow-up. During the study period, 210 patients undergoing craniectomy for traumatic mass-occupying lesion (epidural hematoma, subdural hematoma, or parenchymal contusion) were analyzed. Of those, 9 met study criteria. In-hospital mortality was 67% (6 of 9 patients). The average ICU length of stay was 12 days. The GOS score was 3 in surviving patients. Bilateral hemicraniectomy is a heroic intervention for patients with severe TBI, but can be a life-saving procedure.

Introduction

S evere traumatic brain injury (TBI) in the United States results in 56,000 deaths and 80,000 impaired individuals annually, at an estimated cost of $56 billion USD. (Waxweiler et al., 1995). Severe TBI can be dichotomized into the categories of surgical and non-surgical, according to pathology that reflects lesions that are amenable to surgical intervention as opposed to those that are not. A variety of factors contribute to practice patterns, and vary from surgeon to surgeon. While recent randomized trials have attempted to establish the efficacy of decompressive craniectomy in diffusely-injured states (Cooper et al., 2011), randomized trials for acute mass-occupying lesions such as acute subdural hematomas do not exist. Prognostic indicators, particularly in the preoperative evaluation, may be useful in determining which patients will benefit from surgical intervention. (Karasu et al., 2010) For instance, it is generally accepted that advanced age, bilateral fixed and dilated pupils, and extended time to evaluation (>4 h) are associated with poorer outcomes (Bullock et al., 2006) However, other factors that influence outcome are less clear. While an accurate neurological exam can be adversely affected by pharmacology, a non-contrast head CT is one of the few studies that can be obtained that is unaffected by pharmacology. In addition to characterizing lesion size and location for diagnostic purposes, efforts have been made at correlating radiographic findings with outcomes. (Domenicucciet al., 1998; Lobato et al., 1983; Zumkeller et al., 1996). Other characteristics, such as the presence of diffuse loss of grey/white matter differentiation or brainstem hemorrhage (Duret hemorrhage) are poor prognostic signs. Ultimately, the decision whether to operate on a patient is made on a case-by-case basis, taking into account clinical data with an understanding of the practice patterns and the literature.

In general, acute subdural hematoma (SDH) represents the TBI subtype with the highest mortality, which is estimated at 40–60%. (Bullock, 2006; Fell et al., 1975; Gabl et al., 1989; Haselsberger et al., 1988; Hatashita et al., 1993; Koc et al., 1997; Servadei et al., 1998; Zumkeller et al., 1996), Among those patients presenting to the hospital in a coma (i.e. Glasgow Coma Scale [GCS] score <8), and then undergoing surgical intervention, mortality rates are as high as 68%. (Domenicucci et al., 1998; Gennarelli et al., 1982; Kotwica and Brzezinski, 1993; Lobato et al., 1983; Seelig et al., 1981; Wilberger et al., 1990,1991). However, little is known about the most efficacious treatment for patients with bilateral SDH, or a combination of other types of acute intra-axial or extra-axial mass-occupying lesions. A case series of bilateral craniectomies for penetrating head injury from combatants in the conflicts in Iraq and Afghanistan has been reported; however, pathophysiology from penetrating injury differs from that of closed-head injury (Ecker et al., 2011). Another study suggested that upfront bilateral craniectomy was preferable to delaying the procedure until intracranial pressure became refractory (Akyuz et al., 2010). The optimal management strategy and related outcomes for patients who present with acute, bilateral intra-axial or extra-axial mass-occupying lesions (such as large SDH, epidural hematoma [EDH], or parenchymal contusion) after closed-head injury is largely unknown. We review data from our high-volume neurotrauma center to isolate a case series of this infrequent entity.

Methods

Following approval by the Massachusetts General Hospital (MGH) Institutional Review Board (protocol #2010P002080), consecutive adult patients (age >17 years) undergoing craniectomy for mass-occupying lesions by all providers were retrospectively collected and observed from January 2008 to November 2011. Inclusion criteria included a bilateral cranial operation characterized by the removal of a large fronto-temporal-parietal bone window, with the bones being stored either in a freezer or subcutaneously in the patient's abdomen. Exclusion criteria were defined as any bifrontal operation or any operation in which the bone flap was replaced on the cranium in situ (craniotomy). Additionally, patients were excluded if their contralateral craniectomy was separated in time by more than 8 h.

Details of the patient's presentation were recorded, and included demographics, admission GCS score, presence of a third-nerve palsy, and admission serum alcohol level. Radiographic findings from their initial non-contrast head CT scan were recorded. Details of the operative procedure were previously recorded electronically in the anesthesia and surgical records, including the type of procedure performed, estimated blood loss, and length of surgery. The intensive care unit (ICU) courses of patients meeting entry criteria were retrospectively studied; each patient's electronic inpatient medical record (including daily progress notes, event notes, medication records, radiology records, and physician orders) was extracted for data.

Surgical complications were defined as the development of a clinically significant post-operative hematoma, ischemic stroke, systemic infection, surgical site infection, myocardial infarction, and death, when patient comfort was not the main treatment objective. The Glasgow Outcome Scale (GOS) score was used as a secondary outcome.

Results

During the study period, 210 sequential patients undergoing craniectomy for traumatic mass-occupying lesion (EDH, SDH, or cerebral contusion) were analyzed. Of those, 9 met study criteria (7 male, 2 female). The average age was 40 years (range 19–67 years). Indications for surgery included severe traumatic brain injury with a combination of bilateral SDH, EDH, or parenchymal contusion (Fig. 1). All patients underwent bilateral hemicraniectomy, consisting of a large fronto-temporal-parietal bone flap removal (Fig. 2). This was accomplished either through a single bicoronal incision (Fig. 3), or via separate, standard trauma “reverse question mark” incisions. In-hospital mortality was 67% (6 of 9 patients). ICU length of stay averaged 12 days (range 1–32 days). Complications most commonly included respiratory failure and pneumonia; in addition there were other medical complications, as well as a case of post-operative hemorrhage re-accumulation requiring emergent surgery (Table 1). GOS score for survivors was 3 in all cases.

FIG. 1.

Preoperative head computed tomography (CT) scan. This axial non-contrast head CT obtained upon arrival demonstrates a bilateral severe traumatic brain injury. Surgical lesions are identified as a right-sided temporal contusion (star), right-sided subdural hematoma (arrow), and left-sided frontal contusion (lightning bolt), resulting in brainstem compression.

FIG. 2.

Post-operative head computed tomography (CT) scan reconstruction. Three-dimensional reconstruction of axial-acquired non-contrast head CT scan demonstrates large fronto-temporal-parietal craniectomies.

FIG. 3.

Intra-operative photograph. In this photograph, the patient is positioned supine with the anterior superior aspect of their cranium at the top of the image. Large bilateral craniectomies were performed with the peri-sagittal dural leaflets tacked to each other over a strip of bone that covers the superior sagittal sinus.

Table 1.

Bilateral Craniectomy Patient Characteristics and Outcomes

Patient	Age and gender	Injury; mechanism	Time from injury to operation (hours)	Preoperative third-nerve palsy	ICP pre-op/post-op (cm H₂O)	ED GCS	Ethanol level (mg/L)	Length of operation (hours:minutes)	Estimated blood loss (mL)	ICU length of stay (days)	Hospital length of stay (days)	GCS score on discharge	Hospital disposition	Complications	Glasgow Outcome Score
1	50 M	Bilateral subdural hematoma; blunt trauma (assault)	2	Yes, bilateral	^* / 20	5T	0	2:29	1200	3	3	Expired	Expired	Respiratory failure
2	67 F	Bilateral subdural hematoma; blunt trauma (fall down stairs)	0.5	Yes, bilateral	^* /22	5T	0	3:04	1500	1	1	Expired	Expired	Care withdrawn immediately post-operatively
3	36 M	Bilateral subdural hematoma; blunt trauma (bicycle versus car)	0.75	Yes, bilateral	^* / ^*	3T	0	0:30	1400	1	1	Expired	Expired	Care withdrawn immediately post-operatively
4	29 F	Bilateral subdural hematoma; blunt trauma (bicycle versus car)	0.5	Yes, bilateral	^* / 55	4T	0	3:07	900	3	3	Expired	Expired	Immediate return to operating room for post-operative hematoma re-accumulation
5	50 M	Right epidural/subdural hematoma, left subdural hematoma; blunt trauma (fall down stairs)	3.7 (patient transferred from referring hospital)	No	^* / 3	6T	2332	2:00	600	9	19	11T	Rehabilitation	Alcohol withdrawal, respiratory failure requiring tracheostomy, pneumonia	Expired (goals of care changed to comfort measures only on subsequent admission)
6	34 M	Bilateral subdural hematoma, bilateral contusions >30 mL volume; blunt trauma (assault)	8 days post-injury	No	27 / 2	3T (pharmacologic paralysis)	0	3:30	1000	12	12	Expired	Expired	Anuric renal failure, hyperkalemia, fatal cardiac arrhythmia
7	19 M	Left epidural/subdural hematoma and depressed skull fracture, bilateral hemorrhagic contusions >30 mL volume; blunt trauma (motor vehicle collision)	Unknown; found down	No	^* / 20	6T	0	2:40, 1:28	3300	32	55	9T	Rehabilitation	Urinary tract infection, wound infection requiring operation, pneumonia, diabetes insipidus, respiratory failure requiring tracheostomy	3 (alert, conversant, able to walk with assistance at 1 year follow-up)
8	19 M	Bilateral subdural hematoma; blunt trauma (motor vehicle collision)	4.75 (patient transferred from referring hospital)	No	60 / 5	7T	0	2:59	600	18	27	10T	Rehabilitation	Pneumonia, respiratory failure requiring tracheostomy	3 (alert, follows commands at 3 months post-injury)
9	54 M	Bilateral subdural hematoma; blunt trauma (assault)	0.75	Yes, unilateral	^* / 26	3T	3327	3:00, 2:30	10,500	30	30	Expired	Expired	Respiratory failure requiring tracheostomy, pneumonia, transfusion related acute lung injury, coagulopathy, liver failure

GCS, Glasgow Coma Scale (“T” indicates intubated); ICU, intensive care unit; ED, emergency department; ICP, intracranial pressure.

No ICP monitor in place.

Discussion

The decision to operate emergently on traumatic brain injury patients is determined by limited objective data points such as radiographic evidence of mass-occupying lesions and neurological exam results. It is difficult to determine those patients who will ultimately benefit from surgical intervention. Therefore our protocol is to err on the side of surgical intervention when any opportunity for survival and recovery is present. It should be noted that all survivors in our series had an emergency department GCS score of at least 6T and no evidence of a midbrain compression (third-nerve palsy). Alternatively, bilateral closed intracranial injuries causing mass effect, combined with evidence of midbrain compression and poor neurological exam results (GCS score 3T) were associated with universal mortality in our series. In general, prognostication is aided by some factors such as advanced age (Hukkelhoven et al., 2003; Mosenthal et al., 2002) and Duret hemorrhages (Parizel et al., 2002), tend to predict poor outcomes from severe traumatic brain injury. However, survival has been reported after Duret hemorrhage (Stiver et al., 2009), and even patients older than 80 years may be able to return to pre-injury functional baselines (Lau et al., 2012). Other prognostic indicators we use include the early evidence of large hypodensities on the initial non-contrast head CT scan. These patients are expected to have poor outcomes secondary to neuronal death and subsequent secondary brain injury from cerebral edema, as a result of either hemispheric (Duhaime and Durham, 2007; Lobato et al., 1988) or global ischemia (Yoshino et al., 1985). While not all patients with severe traumatic brain injury undergo CT angiography as the initial evaluation, cerebrovascular injury is a common entity and has been shown to affect outcomes (Miller et al., 2002). Occasionally, we have also used CT angiography to demonstrate impaired cerebral blood flow in cases of presumed extremely elevated intracranial pressure or occlusive causes of stroke.

In patients with bilateral operative severe traumatic brain injuries, diagnosis, prognostication, and treatment must be performed almost simultaneously. There is no role for placement of intracranial pressure monitors when the patient needs to be transported immediately to the operating room for a decompressive procedure. En route, several things can be done that may help the patient. For one, optimization of Pco ₂ to acutely reduce intracranial pressure can be employed (hyperventilation). While longer-term hyperventilation is generally not recommended, as it leads to impaired cerebral blood flow, it can be used en route to the operating room to transiently and acutely reduce intracranial pressure (Bratton et al., 2007b). Additionally, hyperosmolar therapy should be administered in all patients in the form of mannitol and/or 23.4% hypertonic saline (Bratton et al., 2007a; Koenig et al., 2008). Also, vasopressin-receptor antagonists such as conivaptan have been used off-label and have been effective in reducing intracranial pressure (Galton et al., 2011). It is not uncommon for the patient experiencing a brain herniation syndrome to exhibit a Cushing reflex (profound hypertension and bradycardia) (Wan et al., 2008). One patient in this series developed a Cushing reflex response and subsequently experienced several episodes of asystole refractory to resuscitative medications while in transport to the operating room. Since the presumed cardiac failure was related to a brain herniation syndrome, and a craniotomy could not be performed for several more minutes, a transorbital ventricular decompression was successfully performed with return of cardiac function (Madrazo Navarro et al., 1981; Tubbs et al., 2009). While ventricular size is typically very small and distorted secondary to the traumatic lesions, this technique can prove useful in dire circumstances.

This cohort is limited to surgical patients, not taking into consideration patients who were deemed nonsurgical candidates upon initial assessment. Selection bias is also present by excluding patients whose bone flap was replaced, such as in several patients with contralateral epidural hematomas. In these cases, it is likely that less severe injuries existed for which the surgeon did not feel that cerebral edema would be clinically significant. For these reasons, it is likely that the mortality associated with significant bilateral acute lesions is unknown given the size and selection of our cohort.

Bilateral hemicraniectomy is a heroic but potentially life-saving procedure for patients with severe traumatic brain injury following closed-head injury. It is difficult to prognosticate outcome in the hyperacute preoperative evaluation. Therefore, we favor aggressive surgical intervention in the form of bilateral hemicraniectomy when bilateral lesions warrant treatment, particularly in younger patients. Awareness of the associated morbidity and mortality must be conveyed to the patient's representatives when they are available in the preoperative period.

Footnotes

Author Disclosure Statement

No competing financial interests exist

References

Akyuz

, Ucar

, Acikbas

, Kazan

, Yilmaz

, Tuncer

2010. Effect of early bilateral decompressive craniectomy on outcome for severe traumatic brain injury. Turkish Neurosurg., 20:382–389.

Bratton

S.L.

, Chestnut

R.M.

, Ghajar

, McConnell Hammond

F.F.

, Harris

O.A.

, Hartl

, Manley

G.T.

, Nemecek

, Newell

D.W.

, Rosenthal

, Schouten

, Shutter

, Timmons

S.D.

, Ullman

J.S.

, Videtta

, Wilberger

J.E.

, Wright

D.W.

2007a. Guidelines for the management of severe traumatic brain injury. II. Hyperosmolar therapy. J. Neurotrauma, 24,Suppl. 1:S14–S20.

Bratton

S.L.

, Chestnut

R.M.

, Ghajar

, McConnell Hammond

F.F.

, Harris

O.A.

, Hartl

, Manley

G.T.

, Nemecek

, Newell

D.W.

, Rosenthal

, Schouten

, Shutter

, Timmons

S.D.

, Ullman

J.S.

, Videtta

, Wilberger

J.E.

, Wright

D.W.

2007b. Guidelines for the management of severe traumatic brain injury. XIV. Hyperventilation. J. Neurotrauma, 24,Suppl. 1:S87–S90.

Bullock

M.R.

, Chesnut

, Ghajar

, Gordon

, Hartl

, Newell

D.W.

, Servadei

, Walters

B.C.

, Wilberger

J.E.

2006. Surgical management of acute subdural hematomas. Neurosurgery, 58:S16–S24discussion Si–Sv.

Cooper

D.J.

, Rosenfeld

J.V.

, Murray

, Arabi

Y.M.

, Davies

A.R.

, D'Urso

, Kossmann

, Ponsford

, Seppelt

, Reilly

, Wolfe

2011. Decompressive craniectomy in diffuse traumatic brain injury. N. Engl. J. Med., 364:1493–1502.

Domenicucci

, Strzelecki

J.W.

, Delfini

1998. Acute posttraumatic subdural hematomas: “intradural” computed tomographic appearance as a favorable prognostic factor. Neurosurgery, 42:51–55.

Duhaime

A.C.

, Durham

2007. Traumatic brain injury in infants: the phenomenon of subdural hemorrhage with hemispheric hypodensity (“Big Black Brain”) Prog. Brain Res., 161:293–302.

Ecker

R.D.

, Mulligan

L.P.

, Dirks

, Bell

R.S.

, Severson

M.A.

, Howard

R.S.

, Armonda

R.A.

2011. Outcomes of 33 patients from the wars in Iraq and Afghanistan undergoing bilateral or bicompartmental craniectomy. J. Neurosurg., 115:124–129.

Fell

D.A.

, Fitzgerald

, Moiel

R.H.

, Caram

1975. Acute subdural hematomasReview of 144 cases.J. Neurosurg., 42:37–42.

10.

Gabl

, Mohsenipour

, Benedetto

1989. [Acute post-traumatic subdural hematoma in advanced age] Unfallchirurgie, 15:273–278.

11.

Galton

, Deem

, Yanez

N.D.

, Souter

, Chesnut

, Dagal

, Treggiari

2011. Open-label randomized trial of the safety and efficacy of a single dose conivaptan to raise serum sodium in patients with traumatic brain injury. Neurocrit. Care, 14:354–360.

12.

Gennarelli

T.A.

, Spielman

G.M.

, Langfitt

T.W.

, Gildenberg

P.L.

, Harrington

, Jane

J.A.

, Marshall

L.F.

, Miller

J.D.

, Pitts

L.H.

1982. Influence of the type of intracranial lesion on outcome from severe head injury. J. Neurosurg., 56:26–32.

13.

Haselsberger

, Pucher

, Auer

L.M.

1988. Prognosis after acute subdural or epidural haemorrhage. Acta Neurochirurgica, 90:111–116.

14.

Hatashita

, Koga

, Hosaka

, Takagi

1993. Acute subdural hematoma: severity of injury, surgical intervention, and mortality. Neurologia Medico-chirurgica, 33:13–18.

15.

Hukkelhoven

C.W.

, Steyerberg

E.W.

, Rampen

A.J.

, Farace

, Habbema

J.D.

, Marshall

L.F.

, Murray

G.D.

, Maas

A.I.

2003. Patient age and outcome following severe traumatic brain injury: an analysis of 5600 patients. J. Neurosurg., 99:666–673.

16.

Karasu

, Civelek

, Aras

, Sabanci

P.A.

, Cansever

, Yanar

, Saglam

, Imer

, Hepgul

K.T.

, Taviloglu

, Canbolat

2010. Analyses of clinical prognostic factors in operated traumatic acute subdural hematomas. Ulusal travma ve acil cerrahi dergisi=Turkish J. Trauma Emerg. Surg., 16:233–236.

17.

Koc

R.K.

, Akdemir

, Oktem

I.S.

, Meral

, Menku

1997. Acute subdural hematoma: outcome and outcome prediction. Neurosurgical Rev., 20:239–244.

18.

Koenig

M.A.

, Bryan

, Lewin

J.L.

3rd , Mirski

M.A.

, Geocadin

R.G.

, Stevens

R.D.

2008. Reversal of transtentorial herniation with hypertonic saline. Neurology, 70:1023–1029.

19.

Kotwica

, Brzezinski

1993. Acute subdural haematoma in adults: an analysis of outcome in comatose patients. Acta Neurochirurgica, 121:95–99.

20.

Lau

, El-Sayed

A.M.

, Ziewacz

J.E.

, Jayachandran

, Huq

F.S.

, Zamora-Berridi

G.J.

, Davis

M.C.

, Sullivan

S.E.

2012. Postoperative outcomes following closed head injury and craniotomy for evacuation of hematoma in patients older than 80 years. J. Neurosurg., 116:234–245.

21.

Lobato

R.D.

, Cordobes

, Rivas

J.J.

, de la Fuente

, Montero

, Barcena

, Perez

, Cabrera

, Lamas

1983. Outcome from severe head injury related to the type of intracranial lesion. A computerized tomography study. J. Neurosurg., 59:762–774.

22.

Lobato

R.D.

, Sarabia

, Cordobes

, Rivas

J.J.

, Adrados

, Cabrera

, Gomez

, Madera

, Lamas

1988. Posttraumatic cerebral hemispheric swelling. Analysis of 55 cases studied with computerized tomography. J. Neurosurg., 68:417–423.

23.

Madrazo Navarro

, Garcia Renteria

J.A.

, Rosas Peralta

V.H.

, Dei Castilli

M.A.

1981. Transorbital ventricular puncture for emergency ventricular decompression. Technical note. J. Neurosurg., 54:273–274.

24.

Miller

P.R.

, Fabian

T.C.

, Croce

M.A.

, Cagiannos

, Williams

J.S.

, Vang

, Qaisi

W.G.

, Felker

R.E.

, Timmons

S.D.

2002. Prospective screening for blunt cerebrovascular injuries: analysis of diagnostic modalities and outcomes. Ann. Surg., 236:386–393discussion 393–385.

25.

Mosenthal

A.C.

, Lavery

R.F.

, Addis

, Kaul

, Ross

, Marburger

, Deitch

E.A.

, Livingston

D.H.

2002. Isolated traumatic brain injury: age is an independent predictor of mortality and early outcome. J. Trauma, 52:907–911.

26.

Parizel

P.M.

, Makkat

, Jorens

P.G.

, Ozsarlak

, Cras

, Van Goethem

J.W.

, van den Hauwe

, Verlooy

, De Schepper

A.M.

2002. Brainstem hemorrhage in descending transtentorial herniation (Duret hemorrhage) Intensive Care Med., 28:85–88.

27.

Seelig

J.M.

, Becker

D.P.

, Miller

J.D.

, Greenberg

R.P.

, Ward

J.D.

, Choi

S.C.

1981. Traumatic acute subdural hematoma: major mortality reduction in comatose patients treated within four hours. N. Engl. J. Med., 304:1511–1518.

28.

Servadei

, Nasi

M.T.

, Cremonini

A.M.

, Giuliani

, Cenni

, Nanni

1998. Importance of a reliable admission Glasgow Coma Scale score for determining the need for evacuation of posttraumatic subdural hematomas: a prospective study of 65 patients. J. Trauma, 44:868–873.

29.

Stiver

S.I.

, Gean

A.D.

, Manley

G.T.

2009. Survival with good outcome after cerebral herniation and Duret hemorrhage caused by traumatic brain injury. J. Neurosurg., 110:1242–1246.

30.

Tubbs

R.S.

, Loukas

, Shoja

M.M.

, Cohen-Gadol

A.A.

2009. Emergency transorbital ventricular puncture: refinement of external landmarks. J. Neurosurg., 111:1191–1192.

31.

Wan

W.H.

, Ang

B.T.

, Wang

2008. The Cushing Response: a case for a review of its role as a physiological reflex. J. Clin. Neurosci., 15:223–228.

32.

Waxweiler

R.J.

, Thurman

, Sniezek

, Sosin

, O'Neil

1995. Monitoring the impact of traumatic brain injury: a review and update. J. Neurotrauma, 12:509–516.

33.

Wilberger

J.E.

Jr. , Harris

, Diamond

D.L.

1990. Acute subdural hematoma: morbidity and mortality related to timing of operative intervention. J. Trauma, 30:733–736.

34.

Wilberger

J.E.

Jr. , Harris

, Diamond

D.L.

1991. Acute subdural hematoma: morbidity, mortality, and operative timing. J. Neurosurg., 74:212–218.

35.

Yoshino

, Yamaki

, Higuchi

, Horikawa

, Hirakawa

1985. Acute brain edema in fatal head injury: analysis by dynamic CT scanning. J. Neurosurg., 63:830–839.

36.

Zumkeller

, Behrmann

, Heissler

H.E.

, Dietz

1996. Computed tomographic criteria and survival rate for patients with acute subdural hematoma. Neurosurgery, 39:708–712discussion 712–703.