Author Year

Country

Research Design

PEDro

Sample Size

Khan et al. (2016)

Pakistan

RCT

PEDro=4

N=59

Population: ABI; Gender: Male=53, Female=6; GCS:3-5=23, 6-8=28, 9-11=8.

Intervention: Expansile Duraplasty and dural-slits were compared for patients who underwent decompressive craniectomy for acute subdural haematoma.

Outcome Measures: Glasgow Outcome Scale (GOS), mortality.

1.        Neither mortality nor postoperative GOS scores differed significantly between groups.

2.        Surgery duration was significantly shorter for the Expansile Duraplasty group (p=0.000).

Moein et al. (2012)

RCT

PEDro=6

N=20

Population: ABI; Treatment Group (TG, n=10): Mean Age=34.6 yr; Mean GCS=5.6. Control Group (CG, n=10): Mean Age=31.8 yr; Mean GCS=5.6.

Intervention: Participants were randomized to receive decompressive craniectomy with standard treatment (TG) or standard treatment alone (CG). Outcomes were assessed at 6 mo.

Outcome Measures: Glasgow Coma Scale (GCS), Glasgow Outcome Scale (GOS).

1.        GCS improved after treatment in the TG when compared to the CG, but the difference was not significant (p=0.087).

2.        Favourable outcome (GOS=4-5) was found in 60% of the TG and 20% of the CG, but the difference was not significant (p=0.85).

3.        Mortality (GOS=1) was found in 10% of the TG and 30% of the CG, but the difference was not significant (p=0.28).

4.        Age group was not significantly correlated with GOS (p=0.57).

Grassner et al. (2018)

Germany & Austria

Case Series

N=9

Population: TBI=6, Intracerebral hemorrhage=2. Extraaxial tumours=2. Mean Age=48yr; Gender: Male=6, Female=3; Mean GCS=7.

Intervention: Patients in need of a DC, where initial skin closure was not possible due to brain swelling or significantly raised ICP were retrospectively analyzed.

Outcome Measures: Synthetic Skin Substitution, Mortality, Complications.

1.        Epiguard (synthetic skin substitute) was used during a primary DC in 5 cases. The remainder were applied in secondary DC, or revision surgeries.

2.        Mortality rates in the TBI population were 50%.

3.        Four patients developed a wound infection.

4.        Average time to skin closure was 21 days.

Kelly et al. (2018)

USA

Case Control

N=1,470

Population: TBI=1,470. Craniotomy Group (n=1,470): Median Age=43yr; Gender: Male=1102, Female: 368; Median GCS=11. Control Group (n=1,470): Mean Age=42yr; Gender: Male= 1,097, Female= 373; Median GCS= 11.

Intervention: Individuals who had received either a Craniotomy (CO) or a Craniectomy (CE) after sustaining a TBI were retrospectively analyzed using the National Institute on Disability, Independent living, and Rehabilitation Research TBI model systems (TBIMS). Outcomes were measured during treatment, rehabilitation, and up to 2yr post intervention.

Outcome Measures: Hospital LOS, Rehospitalization, Mortality, Functional Independence Measure (FIM), Glasgow Outcome Scale-Extended (GOSE) score, Satisfaction with Life Scale.

1.        On average, CE patients had significantly longer hospital LOS compared to CO patients (22 versus 18d, p=0.01).

2.        Rehospitalization rates were significantly higher in the CE group at 1 and 2yr post intervention compared to the CO group (P<0.01).

3.        Mortality was significantly higher in the CE group at 6mo (p=0.09), but not 1 or 2yr compared to the CO group.

4.        The FIM total, motor, and cognitive score was significantly higher in the CO group compaed to the CE group at rehab discharge, 1 and 2 years post treatment (p=0.02).

5.        Significantly more patients in the CE group had GOSE scores ≤4 at 1 and 2 years post treatment compared to the CO group (p=0.02).

6.        Satisfaction with Life Scale score was not different for either group at 1 or 2 years post-treatment.

Nasi et al. (2018)

Italy

Pre-Post test

N= 190

Population: TBI=190; Mean Age=50yr; Gender: Male=149, Female=41; Mean Time Post Injury to Intervention= 24.5hr; GCS: 3-5= 129, 6-8= 59, 8+=2.

Intervention: Patients receiving a decompressive craniectomy (primary or secondary) following a severe TBI were retrospectively analyzed. Outcomes were assessed before and after the intervention.

Outcome Measures: Glasgow Outcome Scale (GOS), Mortality, Complications.

1.        Primary DC was performed on 109 patients, whereas secondary DC was performed on 81.

2.        Overall mortality was 46.8%, with 31.6% of all patients dying within 30d.

3.        Age over 65, lower GCS at admission, absence of bilateral pupil reactivity, and SDH and brain contusion were associated with higher risk of 30d mortality.

4.        At 6mo after DC, 51.5% of patients who were alive had a poor outcome (GOS severe disability, persistent vegetative state)

5.        The overall rate of complication was 46.8%. The most common complication was the development of Hydrocephalus (19.5% of all patients).

Jehan et al. (2017)

USA

Case Control

N=99

Population: TBI; DC (n=33): Mean Age=47.9 yr; Gender: Male=23, Female=10; Mean Time Post Injury=3.8 hr; Median GCS=9. CO (n=66): Mean Age=49.2 yr; Gender: Male=46, Female=20; Mean Time Post Injury=3.5 hr; Median GCS=11.

Intervention: Patients with TBI underwent decompressive craniectomy (DC) in conjunction with evacuation of intracranial hemorrhage (ICH) to control intracranial pressure (ICP). Controls were treated by craniotomy (CO) in conjunction with ICH evacuation.

Outcome Measures: Mortality, Glasgow Coma Scale (GCS), Glasgow Outcome Scale (GOS), Complications, Length of Stay (LOS).

1.        There were no significant differences between groups for mortality, LOS, GOS or GCS.

2.        The DC group had significantly more returns to the OR (p=0.04), events of hydrocephalus requiring shunt (p=0.03) and median days on ventilator (p=0.02) than the CO group.

Population: TBI; with Emergency Trepanation (n=305): Median Age=62 yr; Gender: Male=207, Female=98; Median GCS=4. without Emergency Trepanation (n=1086): Median Age=65 yr; Gender: Male=717, Female=369; Median GCS=6.

Intervention: Patients with thick subdural hemorrhages who had undergone craniotomy were retrospectively identified. Those who underwent trepanation in an emergency room were compared with those who did not.

Outcome Measure: Survival at Discharge.

Tapper et al. (2017)

Australia

Case Series

N=822

Population: TBI; No operation (n=363): Median Age=57 yr; GCS:3-8=133, 9-12=83, 13-15=147. Craniotomy (n=401): Median Age=61 yr; GCS:3-8=183, 9-12=100, 13-15=118. Craniectomy (n=58): Median Age=45 yr; GCS:3-8=42, 9-12=13, 13-15=3.

Intervention: Patients admitted to a neurosurgical ICU during 2009-2012 were divided into 3 groups (decompressive craniectomy, craniotomy, and conservative).

Outcome Measures: Glasgow Outcome Scale (GOS), Mortality.

1.        Patients in the decompressive craniectomy group (60%) were more likely to have an unfavourable outcome when compared to the craniotomy (53%) and conservative group (41%) (p<0.001).

2.        Decompressive craniectomy was not associated with increased mortality at 6mo compared to conservatively treated patients. However, it was associated with increased 6mo mortality compared to patients in the craniotomy group (p=0.029).

Rush et al. (2016)

USA & Canada

Case Series

N=1940

Population: TBI; Craniotomy (n=1763): Mean Age=68.9 yr; Gender: Male=1150, Female=613. Craniectomy (n=177): Mean Age=49.5 yr; Gender: Male=127, Female=50.

Intervention: Patients data was gathered retrospectively to compare craniectomy and craniotomy for the management of acute subdural hemorrhage (ASDH).

Outcome Measures: Mortality, Length of Hospital Stay (LOS).

1.        Compared to craniotomy (22%), craniectomy (32%) was associated with increased hospital mortality (p=0.044).

2.        LOS is significantly higher in craniectomy patients (14.3 days) compared to craniotomy (10.9 days) (p<0.0001).

Quintard et al. (2015)

France

Case Control

N=25

Population: TBI; Treatment Group (TG, n=13): Mean Age=32 yr; Gender: Male=11, Female=2; Mean GCS=8. Control Group (CG, n=12): Mean Age=30 yr; Gender: Male=10, Female=2; Mean GCS=8.

Intervention: Participants who received decompressive craniectomy (TG) were compared to those who received standard care (CG).

Outcome Measure: Glasgow Outcome Scale (GOS).

1.        At baseline, the CG had significantly lower ICP than the TG.

2.        No significant differences were found between groups on GOS (p>0.05).

Gong et al. (2014)

China

Case Series

N=72

Population: TBI; Mean Age=46 yr; Gender: Male=47, Female=25; Mean GCS=5.

Intervention: Participants who received decompressive craniectomy were retrospectively analyzed.

Outcome Measure: Glasgow Outcome Scale (GOS), Mortality.

1.        Mortality at was 23% at 1 mo and 29% at 6mo.

2.        Favourable outcome (GOS=4-5) at 6 mo was seen in 39%.

3.        GOS was significantly associated with GCS on admission.

Nirula et al. (2014)

USA

Case Control

N=420

Population: TBI; Mean Age=40yr; Gender: Male=330, Female=90; Mean GCS=7.

Intervention: Participants received standard care (SC) and early decompressive craniectomy (DC, n=210) or SC alone (n=210).

Outcome Measure: Mortality.

Galal et al. (2013)

Egypt

Case Series

N=37

Population: ABI; Age Range=14-65 yr; Gender: Male=30, Female=7; GCS<8.

Intervention: Participants who received decompressive craniectomy were retrospectively analyzed.

Outcome Measure: Glasgow Outcome Scale (GOS).

1.        At discharge, 70% of participants had favourable outcomes (GOS=4-5).

2.        At discharge, 14% of participants had died (GOS=1).

Huang et al. (2013)

China

Case Series

N=201

Population: TBI; Mean Age=45.78 yr; Gender: Male=144, Female=57; Mean GCS=6.95.

Intervention: Participants who received decompressive craniectomy were retrospectively analyzed.

Outcome Measure: Mortality.

1.        Overall, mortality was 26.4% at 30 days.

2.        In univariate analysis, significant predictors of 30 days mortality included age (OR=1.020, p=0.014) and GCS (OR=0.5444, p<0.001).

3.        In multivariate analysis, significant predictors of 30 days mortality included age (OR=1.035, p=0.018) and GCS (OR=0.0769, p=0.041).

Limpastan et al. (2013)

Thailand

Case Series

N=159

Population: TBI; Mean Age=37 yr; Gender: Male=130, Female=29; Mean GCS=6.

Intervention: Participants who received decompressive craniectomy were retrospectively analyzed.

Outcome Measure: Glasgow Outcome Scale (GOS), Mortality.

1.        Mortality at discharge was 45%.

2.        Favourable outcome (GOS=4-5) was seen in 13% at discharge and 24% at 6 mo.

3.        GOS was significantly associated with age and GCS.

Yuan et al. (2013)

China

Case Series

N=164

Population: TBI; Mean Age=47.60 yr; Gender: Male=123, Female=41; GCS: 3-5=51, 6-8=52, 9-12=61.

Intervention: Participants were included in retrospective analysis following decompressive craniectomy with (n=93) or without (n=71) mass evacuation.

Outcome Measure: Glasgow Outcome Scale (GOS), Mortality.

1.        At 60 days, good outcome (GOS>4) was found in 42%, poor outcome in 36% (GOS=2-3), and death (GOS=1) in 22% of patients.

2.        Predictors of death at 60 days included age>50 yr (OR=2.36, p=0.047) and mass evacuation (OR=0.31, p=0.014).

3.        Predictor of good outcome at 60 days was GCS=9-12 (OR=2.43, p=0.002).

Agrawal et al. (2012)

India

Case Series

N=273

Population: TBI; Mean Age=34 yr; Gender: Male=229, Female=44; Mean GCS=6.

Intervention: Participants who received decompressive craniectomy were retrospectively analyzed.

Outcome Measure: Glasgow Outcome Scale (GOS), Mortality.

1.        In-hospital mortality rate (GOS=1) was 54%.

2.        At discharge, 22% of participants showed favourable outcome (GOS=4-5).

Li et al. (2012)

UK

Case-Control

N=91

Population: ABI;Treatment Group 1 (TG1, n=36): Median Age=59 yr; Gender: Male=18, Female=18; Median GCS=9.5. Treatment Group 2 (TG2, n=49): Median Age=49 yr; Gender: Male=33, Female=16; Median GCS=5.

Intervention: Participants were compared on treatment received for elevated intracranial pressure (ICP): craniotomy (TG1) versus decompressive craniectomy (TG2). Outcomes were assessed at 6 mo.

Outcome Measures: Glasgow Outcome Scale (GOS), Mortality.

1.        Initially, TG2 were significantly older (p=0.015), more severely injured (p=0.001), and had a higher rate of extracranial injury (33% versus 3%, p=0.001) than TG1.

2.        At 6mo, there was no significant difference between TG1 and TG2 for mortality (32% versus 38%, p=0.65).

3.        At 6mo, there was no significant difference between TG1 and TG2 for poor outcome (GOS<4; 55% versus 58%, p=0.83).

Chen et al. (2011)

Taiwan

Case-Control

N=102

Population: ABI; Treatment Group 1 (TG1, n=42): Mean Age=47.4 yr; Gender: Male=21, Female=21; Mean Time Post Injury=6.7 hr; Mean GCS=6.3. Treatment Group 2 (TG2, n=60): Mean Age=41.2 yr; Gender: Male=41, Female=19; Mean Time Post Injury=5.8 hr; Mean GCS=5.9.

Intervention: Participants were compared based on treatment for ICP: craniotomy (TG1) or decompressive craniectomy (TG2). Outcomes were assessed at 1 yr.

Outcome Measures: Glasgow Outcome Scale (GOS), Complications.

1.        Mortality (GOS=1) was significantly higher in TG2 than TG1 (23.3% versus 7.1%, p=0.04).

2.        There was no significant difference between groups in GOS scores (p=0.21), functional survival (GOS>4; p=0.78), non-functional survival (GOS=2-3; p=0.57), or poor outcome (GOS<4; p=0.78).

3.        There were no significant differences between groups in clinical features of injury (e.g. severity, timing, CT scan).

4.        There were no significant differences between groups in postoperative complications.

Chibbaro et al. (2011)

France

Post-Test

N=147

Population: TBI; Median Age=39 yr; Gender: Male=111, Female=36; Median GCS=6.

Intervention: Participants received decompressive craniectomy (DC) within 28 hr and cranioplasty within 12 wk. Outcomes were assessed at a mean follow-up of 26 mo.

Outcome Measure: Glasgow Outcome Scale (GOS).

1.        Outcome was good (GOS>4) in 67% of participants and poor (GOS=2-3) in 19%; 14% died (GOS=1).

2.        Significant predictors of positive outcome were age <50 yr (p<0.0001) and DC <9 hr post injury (p<0.03).

Ho et al. (2011)

Australia

Post-Test

N=104

Population: TBI; Mean Age=33yr; Gender: Male=86, Female=18; Mean GCS=7.

Intervention: Participants who received decompressive craniectomy and had moderate to severe neurological disability at 6mo were recruited and analyzed. Outcomes were assessed at 6, 12, and 18 mo.

Outcome Measure: Glasgow Outcome Scale (GOS).

1.        At 6mo, 43% had good outcomes (GOS>4) and 57% had poor outcomes (GOS<4).

2.        At 12mo, 54% had good outcomes, 43% had poor outcomes, and 1% died; 2% were lost to follow-up.

3.        At 18mo, 55% had good outcomes, 37% had poor outcomes, and 2% died; 8% were lost to follow-up.

4.        GCS score was a predictor of improvement from poor to good outcome over 6-18 mo (OR=1.44, p=0.18) and of good outcome at 18 mo (OR=1.47, p=0.001).

Wen et al. (2011)

China

Cohort

N=44

Population: TBI; Treatment Group (TG1, n=25):  Mean Age=46.7 yr; Gender: Male=17, Female=8; Mean GCS=6.0. Treatment Group 2 (TG2, n=19): Mean Age=50.2 yr; Gender: Male=15, Female=4; Mean GCS=6.6.

Intervention: Participants who received decompressive craniectomy (DC) were recruited and compared based on timing of DC: early (<24 hr, TG1) or late (>24 hr, TG2). Outcomes were assessed at 1 mo and 6 mo.

Outcome Measures: Glasgow Outcome Scale (GOS), Mortality.

1.        There was no significant difference between TG1 and TG2 in good outcome (GOS>4) at 1 mo (28% versus 37%, p=0.533) or 6 mo (52% versus 63%, p=0.459).

2.        There was no significant difference between TG1 and TG2 in mortality at 1 mo (16% versus 15.8%, p=0.985) or 6 mo (20% versus 21%, p=0.932).

Otani et al. (2010)

Japan

Case Control

N=80

Population: TBI; Mean Age=37.3 yr; Gender: Male=60, Female=20; Severity: Severe=40, Moderate=16, Mild=24.

Intervention: Participants underwent either Hematoma Evacuation (HE) or HE with Decompressive Craniectomy (HE+DC).

Outcome Measure: Glasgow Outcome Scale (GOS).

1.        Significant differences in patient characteristics were noted between groups.

2.        Favorable outcome on GOS was noted in 78.2% of patients in the HE group, while only in 55.8% of the HE+DC group.

Rubiano et al. (2009)

Colombia

Case Control

N=36

Population: TBI; Mean Age=20yr; Gender: Male=21, Female=15; Mean GCS=4.5.

Intervention: Participants received early frontotempoparietal Decompressive Craniectomy (DC, n=16) or Standard Care (SC, n=10).

Outcome Measures: Glasgow Outcome Scale (GOS), Mortality.

1.        Mortality was significantly higher in the DC group than in the SC group (25% versus 65%, p<0.05).

2.        Mean GOS was significantly higher in the DC group than in the SC group (3.1 versus 1.6, p=0.0002).

Williams et al. (2009)

USA

Case Series

N=171

Population: TBI; Median Age=35yr; Gender: Male=137, Female=34; Median GCS=8.

Intervention: Participants who received decompressive craniectomy were retrospectively analyzed.

Outcome Measures: Glasgow Outcome Score Extended (GOSE), Mortality.

1.        Thirty-two percent of patients died in hospital.

2.        Of the survivors, 82% achieved good outcomes (GOSE=5-8).

3.        Patients who experienced good outcomes were younger (26yr versus 43yr, p=0.0028), had higher pre-surgery GCS (7 versus 5, p<0.001), and experienced a greater reduction in ICP post surgery (23mmHg versus 10 mmHg, p<0.0001).

Flint et al. (2008)

USA

Pre-Post

N=40

Population: TBI; Mean Age=43 yr; Gender: Male=30, Female=10; Median GCS=7.5.

Intervention: Participants received decompressive craniectomy.

Outcome Measures: Glasgow Outcome Score (GOS), Mortality, Contusions.

1.        New or expanded hemorrhagic contusions of ≥5cc were observed in 58% of participants.

2.        Mean volume of increase hemorrhage in these participants was 37.1cc.

3.        Contusions expanded >20 cc post treatment were strongly associated with mortality and poor outcome on the GOS at 6 mo, even after controlling for age and GCS.

Huang et al. (2008)

Taiwan

Case Control

N=54

Population: TBI; Mean Age=43.37 yr; Median Time Post Injury=48.65 min; Mean GCS=7.98.

Intervention: Participants who received standard craniotomy (n=16) or Decompressive Craniectomy (DC, n=38) for hemorrhagic contusions were retrospectively analyzed.

Outcome Measures: Glasgow Outcome Score-Extended (GOSE), Length of Stay (LOS), Reoperation Rate, Mortality.

1.        Reoperation rates (7.9% versus 37.5%, p<0.05) were significantly lower after DC than craniotomy.

2.        GOSE scores at 6 mo were significantly higher in DC than craniotomy (5.55 versus. 3.56, p<0.01)

3.        LOS and mortality were not significantly different between groups.

Li et al. (2008)

China

Case Control

N=135

Population: TBI; Mean Age=46.3 yr; Gender: Male=91, Female=44; GCS≤8

Intervention: Participants received routine (10x15cm, n=128) or large (n=135) Decompressive Craniectomy (DC).

Outcome Measures: Glasgow Outcome Score (GOS), Complications, Recurrent Surgery.

1.        Satisfactory outcomes (GOS=3-5) were obtained from 71.1% of the large DC group, compared to 58.6% of routine DC group (p<0.05).

2.        Large DC was more effective in treating very severe TBI than routine DC (p<0.01).

3.        Large DC was associated with a lesser need for recurrent surgery and fewer complications than routine DC.

Meier et al. (2008)

Germany

Case Series

N=131

Population: TBI; Mean Age=36 yr; Gender: Male=99, Female=32; Mean Time Post Injury=49 mo; GCS≤8.

Intervention: Participants who received decompressive craniectomy were retrospectively analyzed.

Outcome Measure: Glasgow Outcome Scale (GOS).

Morgalla et al. (2008)

Germany

Case Series

N=33

Population: TBI; Mean Age=36.3 yr; Gender: Male=20, Female=13; Mean Time Post Injury=3 yr.

Intervention: Participants who received decompressive craniectomy were retrospectively analyzed.

Outcome Measure: Barthel Index (BI), Mortality.

1.        Twenty percent died and 20% remained in a vegetative state.

2.        Thirteen of the surviving patients made a full recovery (BI=90-100).

Chibbaro et al. (2008)

USA

Case Series

N=80

Population: TBI; Mean Age=35 yr; Gender: Male=58, Female=22; Mean GCS=5.

Intervention: Participants who received decompressive craniectomy were retrospectively analyzed.

Outcome Measure: Glasgow Outcome Score (GOS).

1.        Seventy-five percent of patients had a favourable outcome.

2.        Younger age and earlier operations were associated with better outcomes.

3.        Preoperative GCS had no effect on outcome.

Yang et al. (2008)

China

Case Series

N=108

Population: TBI; Mean Age=44.3 yr; Gender: Male=74, Female=34; Severity: Severe=94, Moderate/Mild=14.

Intervention: Participants who received decompressive craniectomy were retrospectively analyzed.

Outcome Measures: Glasgow Outcome Score (GOS), Mortality, Complications.

1.        Twenty-five patients died within 1 mo.

2.        Lower GCS was associated with poorer outcomes on GOS.

3.        Complications secondary to surgery occurred in 50%, 28% of whom developed >1 complication.

4.        Older patients and more severe injuries were associated with more complications.