Early Treatment of TBI A Prospective Study from Austria

Transcription

1 Early Treatment of TBI A Prospective Study from Austria Walter Mauritz MD, PhD Dept. of Anaesthesiology & Critical Care Trauma Hospital XX, 1200 Vienna, Austria International Neurotrauma Research Organisation, 1080 Vienna, Austria

2 Background In a study done , we found that the only guideline that had significant effects upon outcomes after severe TBI was the guideline that recommends to restore oxygenation and perfusion as rapidly as possible Rusnak M, Janciak I, Majdan M, Wilbacher I, MAURITZ W: Severe Traumatic Brain Injury in Austria VI: Effects of guideline-based management. Wien Klin Wochenschr. 2007; 119:64-71

3 New Study Thus, between 4/2009 und 4/2010, 16 Austria centers enrolled 446 TBI patients into an observational study of prehospital as well as early hospital care (first phase) The study was funded by the Ministry of Health and the AUVA (Austrian Trauma Insurance)

4 METHODS OBSERVATIONAL STUDY, NO CONTROL GROUP

5 Data Collected For each patient Prehospital status & treatment Status & treatment in the Trauma Room Times ICU treatment summary Hospital & 6-months outcome

6 For each patient Data Calculated Intervals (EMS-Hosp; Arr-CT, CT-OR, etc) Probability of mortality = P M Probability of poor outcome = P P Hukkelhoven CW, Steyerberg EW, Habbema JD, et al. Predicting outcome after traumatic brain injury: development and validation of a prognostic score based on admission characteristics. J Neurotrauma 2005;22(10):

7 IMPACT-Score Prognosis of death or poor outcome for patients with traumatic brain injury (%) validated in >5000 pts! points P mort (%) 4,3 5,7 7,1 10,0 12,8 18,6 24,3 32,0 41,5 P poor (%) 7,9 12,9 21,0 32,0 47,4 60,0 72,9 82,6 90,0 points P mort (%) 51,9 61,4 69,0 77,0 82,7 87,4 91,1 95,9 P poor (%) 93,1 95,9 97,6 99,0

8 For treatment options Data Calculated P M and P P were calculated (mean P of all patients who DID or DID NOT have the treatment) The ratio between observed mortality or rate of poor outcome and predicted probability was calculated for all treatment options (O/E ratio)

9 O/E Ratio If the O/E ratio is >1, a treatment option is associated with a higher mortality or a higher rate of poor outcome than expected If the O/E ratio is <1, a treatment option is associated with a lower mortality or a lower rate of poor outcome than expected

10 RESULTS PREHOSPITAL TREATMENT OPTIONS

11 Intervall EMS Arr. Hospital Arr. % mort O/E R mort % unfav O/E R unfav 0-45 min 32,0 1, ,4 1, min 33,8 1, ,1 0, min 25,0 1, ,0 0,8581 total 32,3 1, ,7 0,9440 Short interval was associated with lower mortality

12 Definitions of Treatment Efforts Maximal: Venous access, intubation, ventilation, monitoring incl. capnography and blood pressure, infusion 500+ ml Standard: Venous access, intubation, ventilation, basic monitoring (ECG, pulse oximetry), Infusion Minimal: Venous access y/n, monitoring y/n, infusion y/n

13 Prehospital Treatment Effort % mort O/E R mort % unfav O/E R unfav maximal 31,9 1, ,8 0,9210 standard 31,3 1, ,5 1,1108 minimal 31,8 0, ,2 0,7616 Maximal effort was associated with better outcomes

14 Transport % mort O/E R mort % unfav O/E R unfav Patients (all) 31,8 1, ,7 0,9440 Helicopter 31,0 1, ,7 0,9918 Ambulance 34,9 1, ,0 0,9658 Treatment and transport by aeromedical teams were associated with lower mortality

15 Treatment Heli vs. Ambulance Heli Amb total maximal standard minimal total The better outcomes of patients treated by aeromedical teams was due to differences in treatment; the intervals EMS arrival hospital arrival were similar (49 vs. 50 min)

16 Intubation & Ventilation % mort O/E R mort % unfav O/E R unfav Patients (all) 31,8 1, ,7 0,9440 Intubation no 25,6 1, ,6 0,8858 yes 36,1 1, ,5 0,9842 pco2 <34 mmhg 33,8 1, ,8 0, mm Hg 27,6 0, ,2 0, mmhg 18,2 0, ,0 1,3996

17 Monitoring % mort O/E R mort % unfav O/E R unfav Patients (all) 31,8 1, ,7 0,9440 Pulse oximetry no 32,1 1, ,7 1,4347 yes 31,8 1, ,1 0,9793 Capnography no 32,4 1, ,2 1,0561 yes 32,9 1, ,1 0,9669

18 Volume Replacement % mort O/E R mort % unfav O/E R unfav Patients (all) 31,8 1, ,7 0,9440 Lact. Ringer s no 31,4 1, ,8 0,9718 yes 33,3 1, ,5 1,0168 Ringer s no 32,9 1, ,8 1,0073 yes 30,0 1, ,5 0,9409 HES no 30,6 1, ,5 1,0125 yes 34,6 1, ,0 0,9175 NaCl hyperton no 31,2 1, ,9 0,9952 yes 43,5 1, ,1 0,8843

19 Errors 29 patients with a GCS <9 were not intubated on the scene O/E-Ratio for mortality was 1,1874 (slightly higher than average) O/E-Ratio for poor outcome was 0,8162 (15% better than averge!)

20 RESULTS TRAUMA ROOM TREATMENT OPTIONS

21 Interval Hospital Arr. CT scan % mort O/E R mort % unfav O/E R unfav CT first 22,6 0, ,5 0, min 23,2 0, ,5 0, min 38,9 1, ,6 1, min 29,4 1, ,2 0, min 37,0 1, ,6 1, min 29,5 1, ,9 0, min 34,5 1, ,0 1,1092

22 Interval CT OP % mort O/E R mort % unfav O/E R unfav 0-30 min 31,6 0, ,0 0, min 38,2 1, ,4 1, min 30,8 0, ,0 0, min 33,3 1, ,1 1, min 19,0 0, ,0 1, h 30,6 1, ,0 1,1946

23 Definitions of Treatment Efforts Maximal: Venous access, intubation, ventilation, monitoring incl. capnography, blood pressure, TEG, infusion 500+ ml Standard: Venous access, intubation, ventilation, basic monitoring (ECG, pulse oximetry), Infusion Minimal: Venous access y/n, monitoring y/n, infusion y/n

24 Treatment Effort % mort O/E R mort % unfav O/E R unfav maximal 20,0 0, ,0 0,7705 standard 36,7 1, ,8 1,0007 minimal 44,1 1, ,6 1,1661 The Trauma Room is the place where treatment efforts should be maximized

25 ASS, Clopidogrel & Co. alive dead total % mort % unfav Patients (all) ,8 47,7 No medication (%) 82,9 76,8 80,9 30,2 45,6 1 medication (%) 16,8 18,3 17,3 33,8 53,5 2 medications (%) 0,3 4,9 1,8 87,5 87,5 total 100,0 100,0 100,0 31,8 47,7 Incidence: 0-40 years: 1% years: 9% years: 38% years: 62%

26 TEG % mort O/E R mort % unfav O/E R unfav Patients (all) 31,8 1, ,7 0,9440 no 35,0 1, ,4 0,9879 yes 24,3 0, ,9 0,8719 Use of thrombelastography in the trauma room was associated with a significant reduction in mortality

27 Corticosteroids % mort O/E R mort % unfav O/E R unfav Patients (n) 31,8 1, ,7 0,9440 no 29,1 0, ,5 0,8464 yes 52,9 2, ,8 1, patients received corticosteroids: their mortality was significantly higher (p < 0,05)! Corticosteroids more than doubled the odds for mortality!

28 Cooling % mort O/E R mort % unfav O/E R unfav Patients (all) 31,8 1, ,7 0,9440 no 29,9 1, ,4 0,8769 yes 32,4 1, ,5 1,1520 Contrary to patients after CPR cooling was associated with higher mortality

29 Outcomes alive dead n % alive ,2 Death before ICU ,2 Death at ICU ,6 Death after ICU 9 9 2,0 total ,0

30 Mortality vs. Prognosis % Prognose % Mortalität

31 Poor Outcome vs. Prognosis % Prognose % unfavorable

32 Conclusions In addition to the results shown here, this study confirmed that age severity of TBI (= GCS score, pupils), and severity of trauma are the most important factors after TBI; together these factors explain >80% of the outcomes!

33 Conclusions Significant effect of time: Rapid transport to the best facility available CT scan prior to or immediately after trauma room admission Short interval between CT scan and start of neurosurgery

34 Adequate monitoring: Conclusions Use of capnography in all patients who are ventilated Early use of TEG to optimize coagulation (especially important in patients aged > 60 years!)

35 Volume therapy: Avoid lactated Ringer s Conclusions Use of Ringer s or HES is recommended Hypertonic NaCl should be considered in all patients with shock

36 Adequate ventilation: Conclusions The goal is normoventilation Hyperventilation must be avoided Ventilation must be monitored be capnography Corticosteroids must no be used

37 Second Phase Guidelines were written, distributed to the centers, and implemented by the local EMS and centers In April 2011, a second round of data collection was done, to investigate Whether guidelines were actually used Whether use of guidelines was associated with better outcomes

38 Second Phase Data was collected between 4/2011 and 3/2012; 329 patients were enrolled by 15 centers Long-term follow-up interviews were done between 8/2012 and 10/2012 Data collection and analysis were similar to phase 1 Data from 318 patients were analysed

39 RESULTS PREHOSPITAL TREATMENT OPTIONS

40 Interval EMS Arr. Hospital Arr. % Mort O/E Mort % unfav O/E unfav 0 to 30 min 32,1 1, ,5 0, to 60 min 23,7 0, ,7 0, to 90 min 25,6 0, ,2 0, min 21,7 0, ,4 0,6518 total 26,0 0, ,0 0,7004 The shortest interval was associated with higher mortality, probably due to greater trauma severity; outcomes were better than expected

41 Prehospital Treatment Effort % mort O/E R mort % unfav O/E R unfav maximal 29,8 0, ,2 0,6999 standard 18,9 0, ,5 0,6567 minimal 30,3 0, ,3 0,7362 Maximal & standard effort were associated with better outcomes; standard effort was associated with better outcomes than during phase 1

42 Transport % mort O/E R mort % unfav O/E R unfav Patients (all) 26,0 0, ,0 0,7004 Helicopter 23,8 0, ,9 0,6193 Ambulance 27,0 0, ,1 0,7689 Treatment and transport by aeromedical teams were associated with lower mortality but similar O/E ratios (phase 1: 1,11 vs 1,22)

43 Treatment Heli vs. Ambulance Heli Amb total maximal standard minimal total The intervals EMS arrival hospital arrival were similar (49 vs. 50 min); compared to phase 1, more patients had maximal effort in ambulance cars

44 Intubation & Ventilation % mort O/E R mort % unfav O/E R unfav Patients (all) 26,0 0, ,0 0,7004 Intubation no 23,3 0, ,8 0,7618 yes 27,0 0, ,8 0,6672 pco2 <34 mmhg 26,1 0, ,8 0, mm Hg 25,0 0, ,5 0, mmhg 22,0 0, ,3 0,5193

45 Intubation and Ventilation Compared to phase 1 Prehospital intubation was associated with better outcomes (mortality 27 vs. 36%) Lower rate of hyperventilation (17 vs. 46%) Higher rate of hypoventilation (36 vs. 10%)

46 Monitoring % mort O/E R mort % unfav O/E R unfav Patients (all) 26,0 0, ,0 0,7004 Pulse oximetry no 28,6 0, ,48 0,71112 yes 25,6 0, ,63 0,69874 Capnography no 36,4 0, ,36 0,65083 yes 29,6 0, ,18 0,68572

47 Compared to phase 1 Monitoring Pulse oximetry was used more frequently (87 vs 82% of all patients) Capnography was used more frequently (91 vs 60% of all ventilated patients) Again, patients who were not monitored had worse outcomes

48 Errors 16/118 patients (13,6%) with GCS scores 3-5 were not intubated; mortality was 50% (O/E-Ratio 0,9377). In intubated patients with GCS scores 3-5 mortality was 39,1% (O/E-Ratio 0,7834). 23/73 patients (31,4%) with GCS scores 6-8 were not intubated; mortality was 26,1% (O/E-Ratio 0,8418). In intubated patients with GCS scores 6-8 mortality was 14,1% (O/E-Ratio 0,5352).

49 RESULTS TRAUMA ROOM TREATMENT OPTIONS

50 Interval Hospital Arr. CT scan % mort O/E R mort % unfav O/E R unfav CT first 29,4 0, ,9 0, min 25,8 0, ,9 0, min 18,8 0, ,3 0, min 27,3 0, ,9 0, min 23,5 0, ,3 0, min 35,2 0, ,2 0, min 24,1 0, ,0 0,5478

51 Interval CT OP % mort O/E R mort % unfav O/E R unfav 0-60 min 28,2 0, ,7 0, to 120 min 19,0 0, ,2 0, to 240 min 17,6 0, ,2 0,6598 > 240 min 13,2 0, ,0 0,7536 total 26,0 0, ,0 0,7004

52 Compared to phase 1 Intervals CT scans were done faster; 63 vs 60% of patients had their first CT scan within 30 min No differences in O/E ratios

53 Treatment Effort % mort O/E R mort % unfav O/E R unfav maximal 26,0 0, ,5 0,6781 standard 25,3 0, ,8 0,7198 minimal 29,2 1, ,0 0,7560 Results comparable to phase 1: maximal treatment is the best option

54 TEG % mort O/E R mort % unfav O/E R unfav Patients (all) 26,0 0, ,0 0,7004 no 26,7 0, ,9 0,7053 yes 25,2 0, ,5 0,6952 Compared to phase 1, TEG was used more frequently (45 vs 23%). Results are comparable: patients had better outcomes with TEG monitoring. The nearly identical mortality rates suggest that almost all patients who needed TEG analysis got one

55 Intubation % Mort O/E Mort % unfav O/E unfav prehosp intubation 25,8 0, ,2 0,6710 ER intubation 31,7 0, ,9 0,8353 No intubation 23,1 1, ,2 0,6739 Compared to phase 1, prehospital intubation was now associated with better outcomes, due to better prehospital ventilation

56 ICP Monitoring % mort O/E R mort % unfav O/E R unfav Patients (n) 26,0 0, ,0 0,7004 no 31,2 1, ,9 0,7662 yes 22,1 0, ,6 0,6534 Compared to phase 1, ICP monitoring was now associated with better outcomes (p = )

57 Cooling % mort O/E R mort % unfav O/E R unfav Patients (all) 26,0 0, ,0 0,7004 no 26,9 0, ,2 0,6858 yes 20,0 0, ,9 0,8020 Compared to phase 1, cooling was now associated with better outcomes

58 Compared to phase 1 Hospital Treatment Steroids were used in only 4 patients (vs 17) ICP monitoring and cooling were beneficial, because the (more seriously injured) patients who needed these treatment had had better prehospital and ER treatment, and were able to reap a benefit from these options

59 Outcomes alive dead n % alive ,7 Death before ICU 5 5 1,6 Death at ICU ,7 Death after ICU (hosp.) ,4 Death after discharge 1 1 0,3 total ,0

60 Outcomes Phase 1: mortality 31,8%, 11% higher than expected; 79% favorable outcome in survivors; 11 patients with expected unfavorable outcomes had good outcomes Phase 2: mortality 26,5% (reduction by 16,7%), 20% lower than expected; 77% favorable outcome in survivors; 50 patients with expected unfavorable outcomes had good outcomes

61 O/E Ratio for mortality Phase 1 Phase % Prognose % Mortalität %Mortalität %Prognose

62 O/E Ratio unfavorable outcome Phase 1 Phase % Prognose % unfavorable %unfavorable %Prognose

63 Conclusions (1) The new guidelines were used; main changes were Better prehospital management, especially by ambulance teams More patients with maximal treatment Increased use of capnography Better prehospital ventilation Increased use of TEG

64 Conclusions (2) Compliance with the new guidelines was associated with Lower mortality Comparable rate of good outcome in survivors The IMAPCT score corrects for differences in age and TBI severity The improved outcomes are, therefore, due to the implementation of the new guidelines

65 Conclusions (3) The new guidelines can be implemented on a broader scale they are safe & effective A number of questions need to be addressed by detailed analyses: Prehospital intubation and ventilation Volume replacement (which fluid, how much, to whom?)