EFFECT OF INTRAVENOUS CORTICOSTEROIDS ON DEATH WITHIN 14 DAYS IN 10,008 ADULTS WITH CLINICALLY SIGNIFICANT HEAD INJURY (MRC CRASH TRIAL):RANDOMISED PLACEBO- CONTROLLED TRIAL Dr. Hussein Ssenyonjo Resident, Neurosurgery Mulago Hospital

Summary Cortical steroids have been used to treat head injuries for more than 30 years.In 1997,findings after systematic review suggested that these drugs reduce the risk of death by 1-2%.The CRASH trial- a muliticentre international collaboration aimed to confirm or refute such an effect by recruiting 20,000 patients.In May,2004, the data monitoring committee disclosed the unmasked results to the steering committee,which stopped recruitment.

Patients and methods Adults 16 yrs and above who presented within 8 hrs of injury and were noted in hospital to be having GCS of 14 and less.These were recruited if the treating doctor was uncertain whether or not to use steroids-(UNCERTANITY PRINCIPLE)

Procedures Patients randomised in one of two ways. 1.reliable telephone access 2.Local pack system We randomly allocated patients to 48 h infusion of either methylprednisolone or placebo.Loading dose 2g over 1 h in 100mls of infusion.Maintenance dose 0.4g/h for 48h at rate of 20ml/h. Primary outcome measures were death from any cause within 2 weeks of injury and death or disability at 6 months.

Statistical analysis Estimated risk of death allocated to placebo –15%. 2% survival difference would be clinically important so the trial size had to be large enough to detect a difference of this size.A trial of 20,000 patients would have a good chance of showing a 2% survival difference at convincing levels of significance ie >90% power to achieve p<0.01 Precision-95%CI for overall risk and 99%CI for subgroup results Homogenicity in treatment effects assessed within subgroups by X2 test at 5% significance level.

Results 239 hospitals from 49 countries. 10,008 patients randomised to steroid or placebo infusion 62 8h after injury, 3 stopped at request of a relative.

Baseline characteristics Age <25 STEROID n= (30%) PLACEBO n= (29%) (23%)1217(24%) (31%)1485(30%) >55821(16%)849(17%)

GCSCP Severe(3-8)1985(40%)1981(39%) Moderate(9-12) Mild(13-14) 1557(31%) 1465(29%) 1483(30%) 1537(31%) Baseline characteristics-cont

Time of injury <11350(27%)1347(27%) >1<31532(31%)1483(30%) >3<82125(42%)2087(42%) Baseline characteristics-cont

Cause of Injury R-T Crash3249(65%)3169(63%) Fall>2m608(12%)699(14%) other1085(22%)1053(21%) Not Known65(1%)80(2%) Baseline characteristics-cont

Brain CT Scan Done Yes 3,916(78%)3,896(78%) No/Not Known1,091(22%) 1,105(22%) Baseline characteristics-cont

CT Scan ResultsCP Normal23%22% 1 petechial Hrrge29%28% 3 rd V and basal cistern obliteration23%24% SAH31%32% Shift>514%15% Non evacuated haematoma27% Evacuated H12%13% Cortical contusion > 1cm22%23% Baseline characteristics-cont

Adherence to treatment-9,848(98%) 9,748(99%)received full loading dose. Mortality data obtained for 9,964 patients,4,985- CS M-1.052(21%) in 2wks, 4,979-P M- 893(18%) RR of death from all causes in 2wks in CS group with placebo 1.18 RR of death at 2wks didn’t differ by injury severity (p=0.22) or time since injury(p=0.05) Baseline characteristics-cont

Effects of CS allocation on death in 2wks,by injury severity Injury severityC n=(5007)P=(5001) GCS(3-8)785/1972(39.8%)687/1972(34.8%) Moderate (9-12)205/1553(13.2%0143/1476(9.7%) Mild(13-14)62/1460(4.2%)63/1531(4.1%) Time since Injury <1h227/1341(16.9%)209/1342(15.6%) >1<3h308/15309(20.14)296/1560(19%) >3<8h X2 6.03,p= /2114(24.5%)388/2077(18.7%) All patients 95%CI 1052/4985(21.1%)893/4979(17.9%)

Effects of CS allocation on death from all causes in 2 weeks by CT scan results. Normal Ct CS n= /897(4.7%) P n= /878(3.6%) One or more petechial 313/1139(27.5%)270/1098(24.6%) Obliteration 3 rd V or basal cistern 396/906(43.7%)381/920(41.4%) SAH406/1226(33.1%)378/1231(30.7%) Shift >5mm279/556(50.2%)269/579(46.5%) Non evacuated H366/1061(34.5%)322/1050(30.7%) Evacuated H137/486(28.2%)144/500(28.8%) Cortical contusion> 1289/869(33.3%)252/886(28.4%)

Comment on results Relative Risk of death at 2wks wasn’t different in any of the 8 Ct diagnosis subgroups examined

Conclusion Corticosteroids should not be used routinely to use had injury whatever the severity.