effect head injury

7/28/2019 effect head injury

1/7

Articles

1538 www.thelancet.com Vol 366 October 29, 2005

IntroductionThe recorded 40% reduction in odds of dying (adjustedfor case mix) after major trauma between 1989 and 1994

has been attributed to improvements in hospital care;specically, implementation of recommendations fromthe Royal Colleges, increased involvement of seniormedical staff, and improved integration of traumaservices.1 Unfortunately, a continuing reduction inmortality, or a change in the process of care has been notbeen apparent since 1994. 2 Head injury remains animportant factor in the cause of death and disability aftertrauma. 2,3

Most head injury deaths occur in those presenting incoma (Glasgow coma score less than 9). In the UK,neurosurgeons have the main responsibility for treatingsuch patients. Since 1948, neurosurgical care in the UKhas been conned to regional neurosurgical centres,necessitating hospitals without neurosurgical serviceson site to seek advice for management (including theneed for transfer) of severely head injured patients. 4,5Within US and Canadian trauma systems, most patientswith severe head injuries are taken directly to anappropriate trauma centre with neurosurgical facilitieson site. However, trauma system coverage is notubiquitous in Europe or North America, and not alltrauma centres offer 24-hour neurosurgical cover. 69 Thissituation suggests that throughout the developed worldmany patients with severe head injuries are initiallymanaged in facilities without continuous access toneurosurgical care.

Treatment of patients with severe head injury inneurosurgical centres is driven by guidelines. The initial

1984 guidelines recommended treatment based on earlyidentication and intervention for neurosurgical masslesions (eg, extradural haematoma, subdural haema-

toma).10

This approach was justied by a review of a1980s head injury cohort that suggested a low rate of apparently preventable deaths from severe head injuryin patients not transferred to neurosurgical centres. 11However, contemporary guidelines have recommendedthat all patients with a severe head injury should betreated in a neurosurgical centre. 3 Despite these updatedrecommendations, many patients with a severe headinjury, particularly those without surgical lesions, arecurrently not treated in or transferred to a neurosurgicalcentre. 12,13

We wished to establish whether case fatality trendswere the same in major trauma patients with andwithout head injury. Our collaboration, whichrepresents the largest trauma registry in Europe,compared the odds of death (adjusted for case mix) inpatients with severe head injury managed inneurosurgical centres as opposed to hospitals withoutneurosurgical services on site, within the context of reporting temporal odds of death trends.

MethodPatientsWe studied patients injured by blunt trauma between1989 and 2003 who were treated by participatinghospitals in the Trauma Audit and Research Network(TARN=60% of trauma receiving hospitals within

England and Wales). TARN includes patients of any agewho sustain injury resulting in immediate admission to

Lancet2005; 366: 153844

SeeComment page 1509

Department of Neurosurgery,Hope Hospital, Salford, UK

(HC Patel PhD, A T King FRCS);and Trauma Audit and Research

Network, University of Manchester, Hope Hospital,Salford, UK (OBouamra PhD,

M Woodford BSc,ProfD W Yates MD,

F E Lecky PhD)

Correspondence to:Dr F E Lecky, Trauma Audit and

Research Network, University of Manchester, Hope Hospital,

Salford M6 8HD, [email protected]

Trends in head injury outcome from 1989 to 2003 and theeffect of neurosurgical care: an observational studyH C Patel, O Bouamra, M Woodford, A T King, D W Yates, F E Lecky, on behalf of the Trauma Audit and Research Network

Background Case fatality rates after all types of blunt injury have not improved since 1994 in England and Wales,possibly because not all patients with severe head injury are treated in a neurosurgical centre. Our aims were toinvestigate the case fatality trends in major trauma patients with and without head injury, and to establish the effectof neurosurgical care on mortality after severe head injury.

Methods We analysed prospectively collected data from the Trauma Audit and Research Network database for patients presenting between 1989 and 2003. Mortality and odds of death adjusted for case mix were compared for patients with and without head injury, and for those treated in a neurosurgical versus a non-neurosurgical centre.

Findings Patients with head injury (n=22 216) had a ten-fold higher mortality and showed less improvement in theadjusted odds of death since 1989 than did patients without head injury (n=154 231). 2305 (33%) of patients withsevere head injury (presenting between 1996 and 2003) were treated only in non-neurosurgical centres; suchtreatment was associated with a 26% increase in mortality and a 215-fold increase (95% CI 177260) in the oddsof death adjusted for case mix compared with patients treated at a neurosurgical centre.

Interpretation Since 1989 trauma system changes in England and Wales have delivered greater benet to patientswithout head injury. Our data lend support to current guidelines, suggesting that treatment in a neurosurgical centrerepresents an important strategy in the management of severe head injury.


2/7


3/7

Articles


(Ws scores) of survivors with severe head injury in bothgroups. 17 Logistic regression model performance wasdescribed in terms of area under the receiver operatorcurve (AROC).

TARN publishes quarterly Ws scores for all traumacases to participating hospitals aiding clinical gover-nance. 2 This benchmarking activity is supported by theHealthcare Commission without specic informedpatient consent or ethical approval because no patientidentiers are retained by TARN electronically or onpaper. Data for this study came exclusively from theTARN database without author access to patient recordsand we used the same principles toward patient consentand ethical approval as described for the TARNbenchmarking role.

Role of the funding sourceThe sponsor had no role in study design, data collection,data analysis, data interpretation, or the writing of thereport. The corresponding author had full access to allthe data in the study and had nal responsibility for thedecision to submit for publication

Results22 216 patients (13% of 176 447 eligible patients) had ahead injury AIS greater than 2. Patients with head injurywere on average 914 years younger than those withoutsuch injury, although age ranges overlappedconsiderably. Head injured patients were much morelikely to be male, have more severe injury, and havehigher mortality (table 1) than those without head injury.

01

02

04

06

08

10

20

252 374 637 718 683 1083 1319 1304 1137 1035 1033 1095 943 963 912

1551 2207 3918 5002 5012 8758 9227 9590 9039 8463 9068 9384 9030 8937 8096

198901

O d d s o

f d e a

t h ( l o g s c a

l e )

O d d s o

f d e a

t h ( l o g s c a

l e )

02

04

06

08

10

20

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

Year of admission

Figure:Yearly odds of death compared with 1989 baseline adjusted for age, ISS, and RTSUpper=patients with head injury (n=13 490). Lower=patients without head injury (n=107 282). Number of patients shown. Vertical bars=95% CI.


4/7

Articles

www.thelancet.com Vol 366 October 29, 2005 1541

Patients with head injury constituted less than an eighth

of the patients on the database. However, such injurieswere present in 6169 (61%) of all deaths, and wereassociated with a ten-fold increase in mortality comparedwith patients without head injury.

There was a signicant trend in odds of death adjustedfor case mix for patients with head injury between 1989and 2003 (15% reduction per year, p=0008,AROC=091) (gure). However, all the actual reductionoccurred before 1994, so the trend after 1994 was notsignicant (p=0845). A similar but lesser trend wasrecorded when 8726 additional patients with incompleteRTS were included (01% reduction per year between1989 and 2003, p=0032, AROC=087).

In 2003 the odds of death (case mix adjusted) forpatients without head injury was almost half that in 1989(odds ratio 055; 95% CI 04078, crude mortality27%) (gure). The trend analysis showed a signicantreduction of 2% per year in the odds of death adjustedfor case mix (p=0003, AROC 092). Although a 10%reduction by year occurred between 1989 and 1994(p=005), the 19942003 trend was not signicant(p=04). The trend remained signicant after 46 949patients with missing RTS were included (04% per yearbetween 1989 and 2003; p=001, AROC=091).

6921 patients were identied as having had severehead injury between 1996 and 2003. Age, medianGlasgow coma score, and ISS were much the same

between the two groups (table 2). 53% (2677/4982) of patients presenting to a non-neurosurgical centre weretransferred to a neurosurgical centre (table 2). Patientstreated in a non-neurosurgical centre were less likely tohave isolated head injury and to have normal bloodpressure at rst hospital presentation than those in aneurosurgical centre. Mortality was 26% higher than forthose treated in neurosurgical centres (p=0000). Anadditional 660 severe head injury patients were excludedbecause of a secondary transfer to non-TARN hospitalsand consequent lack of outcome data. Their injurycharacteristics (median age 33, median ISS 16, medianGlasgow coma score 5) were similar to the study group,making this exclusion unlikely to cause selection bias(data not shown).

The case mix adjusted odds of death after injury forpatients with severe head injury with completephysiological data who were treated in a non-neurosurgical centre was 215 (95% CI 177260,AROC=087) times that of patients who were treated in aneurosurgical centre. However, slightly fewer than 50%of these patients had a component of the RTS missing(not Glasgow coma score and usually respiratory rate).Imputing a median value for the RTS in such patientsdid not substantially change the result (239 [211269,AROC 082]). Ws scores for severe head injury(standardised observedexpected survival rates) were 6% ( 5% to 8%) for neurosurgical centres and 10% ( 9 to 12%) for non-neurosurgical centres.

Patients (aged 1665 years) with an isolated non-surgical head injury treated in a non-neurosurgicalcentre had a similar median age, Glasgow coma score,and ISS, to those treated in a neurosurgical centre. An8% (p=0005) mean increase in mortality was seen inpatients treated in a non-neurosurgical centre versusthose treated at a neurosurgical centre (table 2), and the

adjusted odds of death in patients with complete datatreated in a non-neurosurgical centre was 192(111330, AROC 087) times that of patients in aneurosurgical centre. These ndings were againunchanged (208 [142-302]) when a median value forthe RTS was inserted for patients with incomplete datato enable their inclusion in the model.

DiscussionIn patients presenting with blunt trauma, we observed aten-fold increased mortality in those who had sustaineda head injury (of any severity visible on CT) comparedwith those who had not. Further, although substantialimprovement in the odds of death adjusted for case mixwas recorded from 1989 to 2003 in patients without headinjury, a less pronounced improvement was seen inthose with head injury. For patients with severe headinjury, the overall mortality was 44%. This crudemortality was signicantly higher (26% p=001) in the33% of patients who were treated in a non-neurosurgicalcentre than in those who were treated in a neurosurgicalcentre.

The observational approach used to assess the effect of neurosurgical care could be biased by case selection. Indeciding whether to proceed with the analysis, weassessed whether 6000 patients with severe head injurycould be randomised to different locations of care. Wethought this randomisation process unlikely, since theCRASH trial 19 recruited only 2000 patients with severe

All SHI patients n=6921 (age 1665)

Neurosurgical centres Nonneurosurgical centres

Number of patients 4616 2305Age (years, median, IQR) 28 (1648) 34 (2058)Male (%, 95% CI) 3448 (75%, 7376) 1642 (71%, 6973)ISS (median, IQR) 25 (1833) 26 (1835)GCS (median, IQR) 3 (36) 4 (36)Isolated head injury (95% CI) 2054 (44%, 4346) 899 (39%, 3741)SBP 90 mm Hg (95% CI) 383 (8%, 89) 434 (19%, 1720)Transferred (95% CI) 2665 (58%, 5659) 302 (13%, 1214)Deaths (95% CI) 1624 (35%, 3437) 1406 (61%, 5963)Isolated, non-surgical SHI n=894 (age 1665)

Number of patients 552 342Age (years, median, IQR) 33 (2347) 31 (2246)ISS (median, IQR) 16 (1025) 16 (1025)GCS (median, IQR) 4 (37) 5 (37)SBP 90 mm Hg (%, 95% CI) 21 (4%, 25) 29 (9%, 612)Patients transferred (%, 95% CI) 311 (56%, 5260) 23 (7%, 49)Deaths (%, 95% CI) 142 (26%, 2229) 118 (34%, 2940)

SHI=severe head injury. GCS=Glasgow coma score. SBP=systolic blood pressure.

Table 2:Patient characteristics after severe head injury according to treatment centre


5/7

Articles


head injury (Glasgow coma score 9) over 4 years,

despite a worldwide recruitment drive. Further, weacknowledge that routinely collected data might beuseful for assessment of interventions, provided thatthey are of sufcient scope and quality, since they arelikely to cover the range of conditions that interventionsapply to, which is not always the case in randomisedclinical trials. TARN has regularly used case-mixanalysis to compare the effectiveness of trauma care insingle or groups of hospitals across Europe. Thesedifferences between neurosurgical and non-neuro-surgical outcomes are substantially larger than anypreviously recorded within TARN. The inclusion of propensity score in our analysis of the benet of neurosurgical care is a valid method to keep bias to aminimum in this type of study. 18

Current practice is such that patients who are nottransferred to a neurosurgical centre have injuriesdeemed incompatible with life, or do not requireoperative neurosurgical treatment. Therefore, thedifferences in mortality might indicate good neuro-surgical triage, which further justies current practice,especially since pupillary responses (which help identifypatients with non-survivable head injury) were notavailable. However, in this large study adjustment of case mix with other important prognostic factors such asage, RTS, and ISS 20 supplemented by the use of propensity scores, we have shown that care in a non-

neurosurgical centre was associated with a two-foldincrease in odds of death, suggesting that the mortalitydata do indicate less than optimum outcomes. Thatpatients with an isolated non-surgical severe head injurytreated in a non-neurosurgical centre also had an almosttwo-fold increase in odds of death adjusted for case mixfurther suggests that this current neurosurgical triagepractice is unsatisfactory.

Our study has some weaknesses, since the physio-logical data were incomplete in 40% of head injury casesand 30% of other cases. This scarcity of data was mostcommonly attributable to the absence of presentingrespiratory rate recordings. However, the respiratoryrate has the lowest weighting within the RTS, and ourdatabase modelling has shown it can be ignored withinTRISS UK21 without loss of model integrity. Ouradjustments for incomplete physiological data have beenestablished in previous publications and have nomaterial effect on the results when they allow all-caseinclusion. 1 Notably, the number of neurosurgical centreswithin TARN (and England and Wales) remained stableover the study period and severe head injury casestransferred out from TARN were similar in injury anddemographic characteristics to those studied.

We have shown that only 53% of patients with severehead injury in England and Wales are transferred toreceive neurosurgical or neurointensive care. 60% of hospitals receiving trauma patients submit data toTARN. Therefore, the national proportion of patients

with severe head injury not treated in a neurosurgical

centre might be different from this gure. However,national surveys lend support to the observation that asubstantial proportion of patients with severe headinjury are treated in units without neurosurgicalfacilities.12,13 Our odds of death estimate for centreswithout neurosurgical facilities is probably typical, sinceour membership comes from all regions of England andWales, representing district general teaching hospitalsand 50% of neurosurgical centres.

International reports of temporal trends have shownan overall reduction in age-adjusted mortality associatedwith traumatic brain injury, which does not accord withthe trends we describe. 22,23 However, these previousstudies have shown a reduction in mortality secondary topenetrating brain injury, a small increase in mortalityfrom fall-related head injuries, and no change inmortality related to road trafc accidents. This varianceprobably relates to the fact that our analysis was based onall patients with injury secondary to non-penetratinginjury.

The lack of improvement in head injured patients istypied by the apparent overall lack of progress in headinjury care, which is suggested by the failure to identify asingle therapy to improve outcome despite over250 randomised controlled trials. 24 However, severalstudies have shown that the institution of packages of specialist neurosurgical or neurocritical care is

associated with improved outcomes.25,26

The reason for the reluctance to accept and treat allhead injuries in neurosurgical centres is unknown, andmight indicate the lack of good quality data for the effectof neurosurgical care. Within Europe, decisions aboutthe transfer of patients are often based on early initial CTscan ndings and whether patients have a surgicallesion, and therefore are most likely to benet fromneurosurgical intervention. 5,11 Admissions are alsodriven by the availability of the necessary facilities forneurosurgical intensive care. 5 Therefore, patients withsurgical lesions have justiably been given preferencebecause of the shortage of neurosurgical expertise innon-neurosurgical centres (most of which have intensivecare facilities). However, non-surgical head injuriesconsist of up to 55% of patients with severe headinjuries, and patients with non-surgical head injuriessimilarly have a high mortality to those with surgicallesions. 27 Furthermore, contemporary studies haveshown that protocol driven therapy based on intracranialpressure and cerebral perfusion targets in the specialistsetting can improve outcome after severe head injury,even in patients with non-surgical injuries. 25,26

Non-surgical lesions are generally accepted to evolveinto those requiring neurosurgical intervention, andincreasingly, interventions such as a decompressivecraniectomy in patients with non-surgical head injuryhave been benecial. 28,29 Although monitoring of intra-cranial pressure is possible in non-neurosurgical


6/7

Articles

www.thelancet.com Vol 366 October 29, 2005 1543

centres, such facilities are not widely available in the

UK.13

There is also evidence that outcome after traumain comatose patients is better in high volume centresthan in those seeing fewer head injury patients, whichargues against setting up protocols for monitoring of intracranial pressure in centres that on average treat lessthan 16 patients per year. 30 These data are in accordancewith our nding that care in a neurosurgical centre isimportant for patients with a non-surgical head injury.

This analysis strongly suggests that improvement of care for patients with severe head injury represents thebest strategy for reduction of case fatality in thosehospitalised after blunt trauma (which has plateauedsince 1994 in England and Wales) and thatneurosurgical or neurointensive care intervention ispivotal to such a strategy. The lack of evidencesurrounding single neuroprotective interventions makesit unlikely that changes in care for patients with severehead injury in hospitals without neurosurgical serviceswould deliver outcome improvements of the magnitudethat could follow signicant expansion in neurointensivecare facilities. Therefore, in accordance with currentguidelines all patients with severe head injury should betransferred to and treated in a setting with 24-hourneurosurgical facilities.Conict of interest statementWe declare that we have no conict of interest

AcknowledgmentsOB, MW, and DWY are employed by the University of Manchester as aresult of TARN funding. FEL has received travel support from TARN topresent these ndings.Participating Hospitals since 1989 Addenbrookes Hospital Cambridge, Heatherwood & Wexham ParkHospital Slough, Rotherham District General Hospital, The PrincessRoyal Hospital Shropshire, Airedale General Hospital Yorkshire,Hillingdon Hospital Middlesex, Royal Albert Edward Inrmary Wigan,Torbay Hospital Devon, Arrowe Park Hospital Merseyside,Hinchingbrooke Hospital Cambridgeshire, Royal Berkshire HospitalReading, Trafford General Hospital Manchester, Ashford GeneralHospital London, Homerton Hospital London, Royal Bolton HospitalFarnworth, University Hospital Lewisham London, Atkinson MorleysHospital London, Hope Hospital Salford, Royal Cornwall HospitalTruro, University Hospital of Hartlepool, Barnsley District GeneralHospital Yorkshire, Hudderseld Royal Inrmary, Royal Devon &Exeter Hospital University Hospital, North Staffordshire, BasildonHospital Essex, Hull Royal Inrmary North Humberside, Royal GwentHospital Newport, University Hospital of North Tees Cleveland,Bassetlaw Hospital Nottinghamshire, Ipswich Hospital Suffolk, RoyalHallamshire Hospital Shefeld, University Hospital of Wales Cardiff,Bedford Hospital, James Cook University Hospital Cleveland, RoyalHampshire County Hospital Winchester, University Hospital, AintreeLiverpool, Birmingham Heartlands Hospital, James Paget HospitalNorfolk, Royal Lancaster Inrmary, Walton Centre for NeurologyLiverpool, Blackburn Royal Inrmary Lancashire, Jersey GeneralHospital, Royal Liverpool Childrens Hospital (Alder Hey), WansbeckGeneral Hospital Northumberland, Blackpool Victoria Hospital, JohnCoupland Hospital Royal Liverpool University Hospital, WarringtonHospital Cheshire, Booth Hall Childrens Hospital Manchester, JohnRadcliffe Hospital Oxon, Royal London Hospital, Warwick HospitalWarwick, Bradford Royal Inrmary Yorkshire, Kent & CanterburyHospital, Royal Manchester Childrens Hospital Pendlebury, WaterfordRegional Hospital Ireland, Bristol Royal Inrmary, Kent & SussexHospital, Royal Oldham Hospital, Watford General Hospital Herts,Bromley Hospital Kent, Kettering General Hospital Northamptonshire,

Royal Preston Hospital, West Cumberland Hospital Cumbria,Broomeld Hospital Essex, Kings College Hospital London, RoyalShrewsbury Hospital Shropshire, West Middlesex University Hospital,Burnley General Hospital, Kings Mill Hospital Nottinghamshire, RoyalSurrey County Hospital, West Wales General Hospital Dyfed,Calderdale Royal Hospital Halifax, Leeds General Inrmary RoyalSussex County Hospital Brighton, Weston General Hospital Avon,Cheltenham General Hospital, Leicester Royal Inrmary, Royal UnitedHospital Bath, Weymouth & District Hospital Dorset, Chestereld &Nth Derbyshire Royal Hospital, Leigh Inrmary, Royal VictoriaHospital, Belfast N Ireland, Whipps Cross Hospital London, ChorleyDistrict General Hospital, Leighton Hospital Cheshire, Royal VictoriaInrmary, Newcastle Upon Tyne, Whiston Hospital Liverpool, CityHospital Birmingham, Lincoln County Hospital, Sandwell DistrictGeneral Hospital West Midlands, William Harvey Hospital Kent,Colchester General Hospital Essex, Maidstone General Hospital Kent,Scarborough Hospital North Yorkshire, Withington HospitalManchester, Conquest Hospital East Sussex, Manchester RoyalInrmary, Scunthorpe General Hospital South Humberside, WithybushGeneral Hospital Dyfed, Countess of Chester Hospital, MedwayHospital Kent, Selly Oak Hospital Birmingham, Worcester RoyalInrmary, County Hospital Hereford, Milton Keynes Hospital, ShefeldChildrens Hospital, Worthing Hospital West Sussex, Coventry &Warwickshire Hospital, Morriston Hospital Swansea, Skegness &District Hospital Lincolnshire, Wrexham Maelor Hospital Clwyd,Craigavon Area Hospital Co Armagh, Nevill Hall Hospital Wales, SouthTyneside District Hospital Tyne & Wear, Wycombe Hospital HighBucks, Crawley Hospital West Sussex, Newcastle General Hospital,Southampton General Hospital, Wythenshawe Hospital Manchester,Cumberland Inrmary Cumbria, Norfolk & Norwich General Hospital,Southend Hospital Essex, York District Hospital, Daisy Hill HospitalCounty Down Northern Ireland, North Manchester General Hospital,Southmead Hospital Bristol, Ysbyty Gwynedd District General DarrentValley Hospital Kent, North Tyneside General Hospital Tyne & Wear,Southport & Formby District General Hospital, Derbyshire RoyalInrmary, Northampton General Hospital, St Bartholomews HospitalLondon, Derriford Hospital Plymouth Northern General HospitalShefeld, St Georges Hospital London, Dewsbury District HospitalYorkshire, Northwick Park Hospital Middlesex, St Helier HospitalSurrey, Diana Princess of Wales Childrens Hospital Birmingham,Nottingham University Hospital, St James University Hospital Leeds,Diana, Princess of Wales Hospital South Humberside, Ormskirk &District Hospital, St Marys Hospital London, Doncaster RoyalInrmary, Peterborough District Hospital, St Peters Hospital Surrey,Ealing Hospital Middlesex, Pilgrim Hospital Lincs, St Thomas HospitalLondon, East Surrey Hospital Redhill Surrey, Pinderelds GeneralHospital Wakeeld, Stepping Hill Hospital Stockport, EastbourneDistrict General Hospital East Sussex, Pontefract General Inrmary,Stoke Mandeville Hospital Buckinghamshire, Epsom Hospital Surrey,Queen Elizabeth Hospital Kings Lynn, Sunderland Royal Hospital,Faireld General Hospital Bury, Queen Elizabeth, Queen MotherHospital Kent, Tameside General Hospital Ashton Under Lyne,Hammersmith Hospital London, Regional Spinal Injuries UnitSouthport Merseyside, Taunton & Somerset Hospital, Harrogate DistrictHospital Yorkshire, Rochdale Inrmary Lancashire, The HortonHospital Oxfordshire.

References1 Lecky F, Woodford M, Yates DW. Trends in trauma care in England

and Wales 198997. UK Trauma Audit and Research Network.Lancet 2000; 355: 177175

2 Lecky FE, Woodford M, Bouamra O, Yates DW. Lack of change intrauma care in England and Wales since 1994. Emerg Med J 2002;19: 52023

3 Royal College of Surgeons. Report of the working party on themanagement of patients with head injury. London, UK: TheCollege, 1999.

4 Cairns HWB, Jefferson G, Dott N. Notes on the neurosurgicalneeds of the population and the training of the neurosurgeon.Report from the planning committee of the Society of Britishneurological surgeons 19451947. Manchester UK: Hotspur Press,1947.


7/7

Articles


5 Crimmins DW, Palmer JD. Snapshot view of emergencyneurosurgical head injury care in Great Britain and Ireland. J Neurol Neurosurg Psychiatry 2000; 68: 813.

6 Hesdorffer DC, Ghajar J, Iacono L. Predictors of compliance withthe evidence-based guidelines for traumatic brain injury care: asurvey of United States trauma centers. J Trauma 2002; 52:120209.

7 MacKenzie EJ. Review of evidence regarding trauma systemeffectiveness resulting from panel studies. J Trauma 1999; 47:S34S41.

8 Sampalis JS, Lavoie A, Boukas S, et al. Trauma center designation :initial impact on trauma-related mortality. J Trauma 1995; 39:23237.

9 Sherman HF, Landry VL, Jones LM. Should Level I trauma centersbe rated NC-17? J Trauma 2001; 50: 78491.

10 Guidelines for initial management after head injury in adults.Suggestions from a group of neurosurgeons.BrMedJ (Clin Res Ed)1984; 288: 98385.

11 Gentleman D, Jennett B, MacMillan R. Death in hospital after head

injury without transfer to a neurosurgical unit: who, when, andwhy? Injury 1992; 23: 47174.12 Matta B, Menon D. Severe head injury in the United Kingdom and

Ireland: a survey of practice and implications for management.Crit Care Med 1996; 24: 174348.

13 McKeating EG, Andrews PJ, Tocher JI, Menon DK. The intensivecare of severe head injury: a survey of non-neurosurgical centres inthe United Kingdom. Br J Neurosurg 1998; 12: 714.

14 Champion HR, Sacco WJ, Copes WS, Gann DS, Gennarelli TA,Flanagan ME. A revision of the Trauma Score. J Trauma 1989; 29:62329

15 Committee on Injury Scaling, Association for the advancement of automotive medicine. The Abbreviated injury scale 1990 revision.Des Plaines, Illinois: 1990.

16 Baker SP, ONeill B. The injury severity score : an update. J Trauma 1976; 16: 88285.

17 Hollis S, Yates DW, Woodford M, Foster P. Standardizedcomparison of performance indicators in trauma: a new approachto case-mix variation. J Trauma 1995; 38: 7636

18 Agostino Jr RB. Propensity score methods for bias reduction in thecomparison of a treatment to a non-randomized control group.Stat Med 1998; 17: 226581.

19 CRASH Trial Collaborators. Effect of intravenous corticosteroidson death within 14 days in 10 008 adults with clinically signicanthead injury (MRC CRASH trial): randomised placebo-controlledtrial. Lancet 2004; 364: 132128.

20 The Brain Trauma Foundation. The American Associationof Neurological Surgeons. The Joint Section on Neurotrauma andCritical Care. Trauma systems. J Neurotrauma 2000; 17: 45762.

21 The Trauma Audit and Research Network. Ps/04. http://www.tarn.ac.uk/resources/modelsummary.htm

22 Adekoya N, Thurman DJ, White DD, Webb KW. Surveillance fortraumatic brain injury deathsUnited States, 19891998.MMWR Surveill Summ 2002; 51: 114.

23 Gennarelli TA, Champion HR, Sacco WJ, Copes WS, Alves WM.Mortality of patients with head injury and extracranial injurytreated in trauma centers. J Trauma 1989; 29: 1193201.

24 Narayan RK, Michel ME, Ansell B, et al. Clinical trials in head

injury. J Neurotrauma 2002; 19: 50357.25 Patel HC, Menon DK, Tebbs S, Hawker R, Hutchinson PJ,Kirkpatrick PJ. Specialist neurocritical care and outcome from headinjury. Intensive Care Med 2002; 28: 54753.

26 Stocchetti N, Rossi S, Buzzi F, Mattioli C, Paparella A, Colombo A.Intracranial hypertension in head injury: management and results.Intensive Care Med 1999; 25: 37176.

27 Marshall LF, Gautille T, Klauber MR, et al. The outcome of severeclosed head injury. J Neurosurg 1991; 75: S28S36.

28 Piek J. Decompressive surgery in the treatment of traumatic braininjury. Curr Opin Crit Care 2002; 8: 13438.

29 Whiteld PC, Patel H, Hutchinson PJ, et al. Bifrontaldecompressive craniectomy in the management of posttraumaticintracranial hypertension. Br J Neurosurg 2001; 15: 50007.

30 Nathens AB, Jurkovich GJ, Maier RV, et al. Relationshipbetween trauma center volume and outcomes. JAMA2001; 285:116471

effect head injury

Documents