Improving TBI outcome

Transcription

1 Improving TBI outcome Dr Peter Smielewski 20/10/2017 Division of Neurosurgery, Department of Clinical Neurosciences

2 Background Stein, S. C., Georgoff, P., et al. (2010). Journal of Neurotrauma Thanks to Joseph Donnelly

3 The impact of specialised neuro-critical care Thanks to David Menon

4 TBI secondary insults Definition of insults ICP > 20 CPP < 60 Potential exacerbating condition PRx > 0.25 Thanks to Joseph Donnelly

5 PRx (a.u.) Associations and consequences of clinical high ICP N= Observat ions on the cerebral effects of refract ory intracranial hypert ension aft er severe t raumat ic brain injury 121 A 0.75 B >90 ICP (mm Hg) >110 CPP (mm Hg) Figure 6.7: PR x response to refract ory int racranial hypert ension expressed relative to changes in ICP (left) and CP P ( right) ( n= 24). Pressure reacitivity increased with increasing ICP Thanks to Joseph Donnelly

6 Associations and consequences of clinical high ICP 124 Chapt er 6. Int racranial sequelae of raised intracranial pressure N=37 A B 30 P BT O 2 (mm Hg) ICP (mm Hg) >110 CPP (mm Hg) Figure 6.9: P BT O 2 response to refract ory intracranial hypertension expressed relative to changes in ICP (left) and CPP (right) (n= 9). When expressed against ICP, P BT O 2 demonstrates a steady decrease. When expressed in relation to changes in CPP, the

7 Associations and consequences of clinical high ICP N=37

8 Impact of ICP insults on outcome N=824 Thanks to Joseph Donnelly

9 Impact of CPP insults on outcome N=824 Thanks to Joseph Donnelly

10 Heatmaps of time profile of ICP and PRx stratified by different levels of functional outcome. N=601 No DC Adams H et al PLoS Med Jul 25;14(7)

11 Time profile of ICP and PRx Variables adjusted for age, sex, best preintubation GCS, primary injury type (diffuse versus mass lesion), surgical interventions (none, craniotomy, primary DC, and secondary DC) N=601 No DC Adjusted for patient, injury, and treatment characteristics. Adams H et al PLoS Med Jul 25;14(7)

12 (ROC) curve analysis for prediction of fatal outcome due to neurological causes. N=601 No DC Adams H et al PLoS Med Jul 25;14(7)

13 Limits of reactivity visualisation Güiza, F. et al., Visualizing the pressure and time burden of intracranial hypertension in adult and paediatric traumatic brain injury. Intensive Care Medicine.

14 Visualising the ICP insults burden N = 261 Güiza, F. et al., Visualizing the pressure and time burden of intracranial hypertension in adult and paediatric traumatic brain injury. Intensive Care Medicine.

15 Duration greater than (mins) Visualising the ICP insults burden (Cambridge cohort) Chapter Novel applications of intracranial monitoring after severe traumatic brain inj ury n= Correlation GOS and insult number Intensity greater than (mm Hg) Figure 7.6: V isualisation of relationship between number of ICP insults (of a particular 24 million duration episodes and intensity) and GOS after severe T BI (748 patients, 24 million insults). Similar to the previous report a transition zone is depicted as a curvilinear function indicating that higher intensity ICP insults can be tolerated for a short period of time without beaing associated with poorer outcome. Above Donnelly J at al, in press

16 Visualising CPP insults Hypo-perfusion Hyper-perfusion Combined (transition curves only) Guiza et al. Cerebral perfusion pressure insults and associations with outcome in adult traumatic brain injury. J Neurotrauma 2017

17 CPP above : 70?...65?... CPP may be low; ICP<15 mmhg Set thresholds: Is it wise? Too low CPP: ischaemia Too high CPP: hyperaemia

18 Individual CPP target based on PRx/CPP relationship CPP ICP PRx CPPopt CPP

19 Risk of managing CPP above or below continuously updated CPP opt N = 100 N = 327 pts

20 Vascular reactivity range Thanks to Joseph Donnelly

21 Most recent data: 2016 J Donnelly et al. Submitted

22 Limits of autoregulation and mortality ΔCPPopt < -10

23 Duration greater than (mins) ICP insults stratified by the limits of reactivity n=748 Chapter A Novel applications of intracranial monitoring after severe traumatic brain inj ury ICP insults with CPP below LLR Correlation GOS and insult number B ICP insults with CPP above LLR Intensity greater than (mm Hg) 3 million episodes Intensity greater than (mm Hg) 21 million episodes Figure7.7: V isualisation of relationship between number of ICP insults (of a particular duration and intensity) and GOS after severe T BI when CPP is below the LLR (A, 2 million insults) or above the LLR (B. 21 million insults). When CPP is below the LLR during an insult, even low intensity ICP insults are associated with worse GOS as denoted by the predominance of red in A. This

24 How can we use the information at the bedside? Time spent below estimated lower limit of reactivity associates with patient outcome Having a CPP above the estimated lower limit of reactivity is protective during episodes of raised ICP

25 16th International Symposium on Intracranial Pressure and Neuromonitoring CPP optimal in real time on NICU How I do it with "ICM+" in Moscow Neurosurgery Institute named after Burdenko, Moscow Oshorov AV Boston

26 Methods of ICP corrections relatively short-term HOB EVD HV Sedation Hyperosmolar therapy relatively long-term removal of mass-effect optimization of CPP Hypothermia DC Barbiturates

27 Optimization of CPP Primary: through manipulation with ABP Secondary: through decreasing ICP

28 Data of rcbf, ICP and Prx in severe TBI pts (GCS < 9) ICP =48 Prx 0,17 ICP =26 Prx 0,45 ICP =16 Prx 0,11 ICP =24 Prx 0,0 ICP =30 Prx 0,3 18,8 26,7 45,8 44, ,8 24, ,2 55,1 18,3 10,1 34,9 30,7 47,1 47,5 42,3 40,9 30,6 35,1 51,3 45,3 32,8 29,5 19,5 36,1 52,1 71,4 36,9 34,4 ICP =23 Prx -0,21 21,4 28,2 49, ,9 20,8 ICP =23 Prx -0,07 20, ,2 19, ICP =28 Prx 0,02 15,9 12,1 41,6 38,2 25,3 27,9 ICP =18 Prx -0,19 25, ,6 53,2 29,8 26,5 ICP =26 Prx 0,1 24 5,8 52,3 49,8 49,5 34,8 ICP =28 Prx 0, ,2 59,2 35,4 29,4 ICPm28 Prx 0,07 17,8 23, ICP =27 Prx-0,15 24,3 24,6 51,6 47,3 32,1 29,9 ICP=49 Prx 0, ,6 53,3 38,4 29,4 ICP =56 Prx 0,27 1,8 11,4 55,9 49,2 ICP =26 Prx-0,02 28,9 26,6 54,3 51,8 35,6 43,5 ICP =25 Prx 0, ,3 33,9 69,5 38,4 36,4 23,1 29, ,6 ICP =25 Prx 0,1 ICP=28 Prx 0,0 32,5 20,7 59,2 62,3 36,2 29,2 ICP =28 Prx 0, ,9 62,1 68,9 29,2 32,7 ICP =32 Prx 0,14 41,2 43,5 70, ,6 ICP =27 Prx -0,05 42,1 18,5 88,3 75,6 69,5 74,3 41,6 39,6 47,6 41,

29 Data of rcbf in patient with focal injury in left hemispheric GCS 7, GOS 3 (severe disability) Mean CBF in intact region: 46,0±10,0 ml100g/min Mean CBF in region of contusions: 16,3±6,0 ml/100g/min (p <0,01) Potapov AA, Zaharova NE, Pronin IN et al. (2011) Zh Vopr Neirohir Im N N Burdenko

30 Monitoring Autoregulation ~ 8 h Admission to NICU pts with Severe TBI ABP ICP CPP + ICM Plus (Prx) Neurosurgical operation : remove hematoma, craniotomy, decompression, EVD and other Autoregulation was preserved Prx (-1; 0) Autoregulation was partially failured Prx (0; 0,2) Autoregulation was completely failured Prx (0,2; 1)

31 1) Autoregulation was preserved Prx (-1; 0)

32 2) Autoregulation was partially failure Prx (0; 0,2)

33 3) Autoregulation was completely failure Prx (0,2; 1)

34 Monitoring Autoregulation 8 h Admission to NICU pts with Severe TBI ABP ICP CPP + ICM Plus (Prx) Neurosurgical operation : remove hematoma, craniotomy, decompression, EVD and other Autoregulation was preserved Prx (-1; 0) Autoregulation was partially failured Prx (0; 0,2) Autoregulation was completely failured Prx (0,2; 1) ICP < 20 ICP > 20 ICP < 20 ICP > 20 ICP < 20 ICP > 20 CPP CPP CPP CPP CPP CPP 50-70

35 Optimization CPP in patient with preserved autoregulation from 65 to mmhg

36 Optimization CPP in patient with partially failured autoregulation from 70 to 90 mmhg

37 Patients Patients Historical group* (N=60) ICP/CPP 15% 3% 28% 28% 25% Tendency (but p>0,05) Decreasing death from 15% to 9% GOS 1-death; 2-vegetative; 3-severe disability; 4-moderate disability; 5-good recovery Increasing Favorable outcome from 53% to 61% Prospective group* (N=96) ICP/CPP + Prx 9% 1% 29% 38% 23% GOS In conclusion: we need more data to define a best strategy * - pts with * decompression were excluded from analysis

38 CPPopt guided-therapy in TBI: one center data Celeste Dias C E L E S T E. D I A M E D. U P. P T

39 CPPopt - decision steps 30º head up elevation and ABP transducer at heart level. CPPopt value and curve, updated every minute, in a 4 hr calculation window. at least 75% of time good recordings of CPP and ICP values available in the 4hr calculation window average PRx values had to be < 0.25 the past 4hrs select the CPP value with most negative PRx value covered by the curve. U-shaped, ascending and descending curves were accepted in case the overall PRx<0.25. CPPopt guided-therapy in TBI: one center data

40 CPPopt demographic data From Jul 2011 to Jan 2016 n; % median (min-max) n 53 Age (years) 44 (20-88) Gender (n; %male) 47; 89% M in local GCS 7 (3-13) SAPSII 43 (22-66) SAPSII mortality (%) 31 (5-78) NCCU mortality rate 7; 13% Hospital mortality rate 8; 15% LOS at NCCU (days) 20 (5-65) LOS at Hospital (days) 37 (8-138) GOS at 3M 3 (1-5) Decompressive craniectomy (n; %) 15; 28% CPPopt guided-therapy in TBI: one center data

41 CPPopt vs real CPP 30º head up elevation and ABP transducer at heart level.

42 PRx, delta-cpp and mortality at NCCU CPPopt guided-therapy in TBI: one center data

43 Conclusions Online monitoring data (CPP, ICP) is suitable for online assessment of autoregulation (PRx) CPPopt guided-therapy needs to applied with a strict protocol CPPopt algorithm may be applied by trained nurses Results show that patients with impaired autoregulation have worse outcomes CPP-CPPopt around zero is related to best outcomes Thanks to Dr Celeste Dias, Porto

44 Pressure reactivity index and CPPopt work so far 174 articles with PRx 70 articles with CPPopt Pubmed Oct 2017

45 Unanswered fundamental questions How does CPPopt behave prospectively? Is CPP management according to CPPopt safe? How do clinicians interpret and act on CPPopt recommendations? What is the best end point for the subsequent RCT?

46 Further questions to address Should CPPopt be targeted outright or be a guide given other parameters (and if so, how)? Is CPPopt the most appropriate target or some other associated parameter (such as the lower limit of autoregulation)? Should we target CPPopt even if autoregulation at CPPopt is still absent? Is CPPopt guided therapy beneficial in all TBI patients? Is CPPopt guided therapy equally beneficial in contusional, pericontusional and normal TBI brain

47 Upcoming CPP feasibility and safety study CPPOpt Guided Therapy: Assessment of Target Effectiveness A randomized trial assessing the safety and effect of optimal cerebral perfusion pressure directed treatment.

48 The study protocol N = 30 N = 30 Covering 5 days after admission

49 COGiTATE tool - Randomisation

50 COGiTATE tool CPP treatment arm

51 COGiTATE tool 4 hourly review

52 COGiTATE tool CPP Opt treatment arm

53 COGiTATE tool 4 hourly review

54 COGiTATE tool review with no CPPopt target

55 Treatment (local) protocol CPP target ICP Action Interventions > 20 mmhg Decrease ICP ICH treatment < 20 mmhg Increase ABP Fluids Vasopressor (as per clinician) > 20 mmhg Decrease ABP Vasopressor < 20 mmhg Decrease ABP Vasopressor

56 Study main endpoints Feasibility Differences in CPP and CPPopt in both groups. Mean PRx Mean difference in CPP-CPPopt Safety A change in treatment intensity level score (TIL) score of 3 is representative of an escalation of treatment effectiveness from basic ICP management to second tier therapies which are known to be more harmful and therefore this is a significant effect

57 Secondary aims/endpoints Between group differences in ICP variability Frequency and average duration of spikes > 20mmHg Mean daily RAP (cerebral compliance index). Mean daily vasopressor dose. Incidence of troponin rise stratified by day. Mean daily fluid balance. Mean daily P/F ratio (pulmonary complications). Survival and GOSE at 3 months.

58 Physiological effectiveness / effect of targeting CPPop Differences in Mean daily MAP. Mean daily PRx at CPPopt. Mean daily L/P ratio, PbO2 (and PbO2/PaO2 ratio), brain tissue glucose

59 COGiTATE website