Page 35 - DECISION-MAKING IN SEVERE TRAUMATIC BRAIN INJURY PATIENT OUTCOME, HOSPITAL COSTS, AND RESEARCH PRACTICE

Type of GCS score.
Initial day 1 post- resuscitation
Admission
ED
Initial
Presentation
Best in first 12 hours.
Admission Admission
Outcome measure
14-day mortality
ICU/Hospital mortality
Inpatient mortality 2-week mortality
Mortality (in hospital)/ FIM (good)
Hospital and 28- day mortality
Hospital mortality GOS (>6M)
Results
Mortality (OR; 95% CI; P) 2 GCS6-8 vs. GCS3-5 = (0.44; 0.36 - 0.53; <0.0001)
ICP monitoring is a statistically significant predictor of 2-week
mortality: (0.63; 0.41-0.94; 0.02)*
Mortality: GCS3 (N=796): ICU: 48.5%, Hospital: 51.1% GCS>3: ICU: 24.8%, Hospital: 29.3%
Age 65 and GCS=3: ICU 67%; Hospital 71.1% Numbers irrespective the presence of ICP monitoring
Mortality:
GCS(3): -13.3% (95% CI: -6.0 to -20.5). P:<0.001 GCS(3)+ High ISS (>25): -32.9% (95% CI: -20.3 to -45.4) P:<0.001**
Overall mortality: ICP: 24%, no-ICP: 50.9% (p=0.025) Mortality: GCS3-5: 57.8%
GCS3-5: Crude OR 3.625 (1.406-9.343)***
Adjusted OR: 2.506 (0.712-8.822)***
Mortality Overall ICP/no-ICP: 27.2% / 22.4%
FIM (good) Overall ICP/no-ICP: 17.8% / 28.7% Mortality: GCS3-5: 26.3%. Overall: 22.9%
GCS3-5: Independent predictor of mortality: OR1.84
Hospital mortality (OR; 95% CI; P): GCS<6: 0.76; 0.18–3.2; 0.71, GCS≥6: 5.6; 1.7-18.4; <0.01
28-day mortality (OR; 95% CI; P): GCS<6: 0.47; 0.11–2.1; 0.32, GCS≥6: 5.0; 1.5-16.7; <0.01
Mortality ICP/No-ICP, (OR; 95%CI; P)
GCS3-4: 12.9%/ 24.5%, (0.51; 0.17-1.59; 0.25) GCS5-6: 18.2%/ 7%, (3.74; 0.61-22.82; 0.15) GCS7-8: 50%/ 7.2%, (12.89; 3.14-52.95; 0.0004)
GCS3-5: GOS1= 16%, GOS2=12%, GOS3= 24%, GOS4= 10%, GOS5=38%
GCS6-8: GOS1= 4%, GOS2=4%, GOS3= 13%, GOS4= 10%, GOS5=69%
Decision-making in very severe traumatic brain injury
Multiple factors seemed likely to be associated with more frequent placement of an ICP monitoring device, including age (<65 years), female gender, the presence of at least one reactive pupil and more isolated TBI with a higher Abbreviated Injury Scale (AIS) head score and higher Injury Severity Score (ISS).22, 23 Increased likelihood ratios for ICP monitoring were also found when the CT-scan showed subdural hematoma, cerebral contusion or diffuse mass effect.23 Reasons for not providing ICP monitoring included higher age,21-23 pupillary abnormalities,21 history of cancer,22 cardiac insufficiencies,22 alcoholism, coagulopathy or injury from a fall 23 and a higher estimated mortality as assessed by the treating physician.22 A cohort of 1856 patients, showed ICP rates rise with TBI severity, but interestingly again decreased for vs-TBI.22
33