Skip Ribbon Commands
Skip to main content
SharePoint

PublicComment

 Feedback

If you have any comments or questions on this page, please email us at:
 

Prognostication in Absence of TTM

Question Type:
Prognostic
Full Question:
Among adults who are comatose after cardiac arrest and are not treated with TTM  (P), does does any clinical finding when normal (eg, clinical exam, EEG, SSEPs, imaging, other)  (I), compared with compared with any clinical finding when abnormal  (C), change Survival with Favorable neurological/functional outcome at discharge, 30 days, 60 days, 180 days AND/OR 1 year, Survival only at discharge, 30 days, 60 days, 180 days AND/OR 1 year (O)?
Consensus on Science:
No study on clinical examination reported blinding of the treating team to the results of the index test. Blinding of the treating team is very difficult to achieve for predictors based on clinical examination, which implies a risk of self-fulfilling prophecy. For the critical outcome of survival with unfavorable neurologic status or death at discharge, we identified 2 studies on pupillary reflex and motor response or oculocephalic reflex (151 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision).(Earnest 1979, 56; Bertini 1989, 627) For the critical outcome of survival with unfavorable neurologic status or death at 30 days, we identified 1 study on GCS (97 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision).(Pfeifer 2005, 49) For the critical outcome of survival with unfavorable neurologic status or death at 90 days, we identified 2 studies on corneal reflex, pupillary reflex, motor response, oculovestibular reflex, GCS, or myoclonus (97 patients; very-low-quality evidence downgraded for serious or very serious bias and serious or very serious imprecision).(Young 2005, 159; Topcuoglu 2009, 1635) For the critical outcome of survival with unfavorable neurologic status or death at 180 days, we identified 4 studies on brainstem reflexes, motor response, or myoclonus (650 patients; very-low-quality evidence downgraded for serious or very serious bias and very serious imprecision).(Edgren 1987, 820; Krumholz 1988, 401; Wijdicks 1994, 239; Zandbergen 2006, 62) For the critical outcome of survival with unfavorable neurologic status or death at 1 year, we identified 3 studies on brainstem reflexes, motor response, GCS, or myoclonus (172 patients; very-low-quality evidence downgraded for serious or very serious bias and very serious imprecision).(Bassetti 1996, 610; Thomke 2005, 14; Fischer 2006, 1520) Clinical Examination Pupillary Reflex. In 1 study(Earnest 1979, 56-60) (98 patients; very-low-quality evidence), an absent PLR on hospital admission predicted poor outcome, with 8 (1–25)% FPR and 56 (43–67)% sensitivity. At 24 hours(Bertini 1989, 627; Young 2005, 159; Zandbergen 2006, 62) (3 studies, 496 patients), 48 hours(Edgren 1987, 820; Bertini 1989, 627; Zandbergen 2006, 62) (3 studies, 403 patients), and 72 hours(Bertini 1989, 627; Zandbergen 2006, 62) (2 studies, 382 patients) from ROSC, the FPRs of PLR for prediction of poor outcome were 9 (4–18)%, 4 (0–12)%, and 0 (0–8)%, respectively. Sensitivity ranged from 18 (15–23)% to 21 (17–25)%; (very-low-quality evidence, downgraded for serious or very serious bias and very serious imprecision). Corneal Reflex. In patients who are comatose after resuscitation from cardiac arrest and who are not treated with TTM, an absent corneal reflex at 24 and 48 hours after ROSC predicted poor outcome, with 17 (9–27)% and 7 (2–20)% FPR. Sensitivities were 37 (32–42)% and 30 (25–35)%, respectively(Edgren 1987, 820; Young 2005, 159; Zandbergen 2006, 62) (3 studies, 497 subjects; very-low-quality evidence downgraded for very serious bias, serious inconsistency, and very serious imprecision). Oculovestibular Reflex. In 2 studies(Edgren 1987, 820; Young 2005, 159) (65 patients; very-low-quality evidence downgraded for very serious bias, serious inconsistency, and very serious imprecision), the bilateral absence of oculovestibular reflex at 24 hours from ROSC predicted poor outcome, with 0 (0–18)% FPR and 38 (25–53)% sensitivity. In 1 study(Edgren 1987, 820) (19 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision), the bilateral absence of oculovestibular reflex at 48 hours from ROSC predicted poor outcome, with 0 (0–35)% FPR and 25 (5–57)% sensitivity. Combination of Ocular Reflexes. In 1 study(Zandbergen 2006, 62) (386 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision), the combined absence of both pupillary and corneal reflexes at 24, 48, and 72 hours from ROSC predicted a poor outcome, with 5 (1–17)%, 3 (0–17)%, and 0 (0–15)% FPR, respectively, and 13% to 14% sensitivity. In 1 study(Bassetti 1996, 610-615) (60 patients; very-low-quality evidence downgraded for serious bias and very serious imprecision), the absence of more than 1 among pupillary, corneal, and oculocephalic reflex at 6 to 12, 24, and 48 hours from ROSC predicted poor outcome, with 0 (0–22)% FPR. Motor Response to Pain. At 24 hours from ROSC, an absent or extensor motor response, corresponding to a motor score 1 or 2 (M1–2) of the GCS, predicted a poor outcome, with 27 (12–48)% FPR and 76 (71–80)% sensitivity(Young 2005, 159; Zandbergen 2006, 62) (2 studies, 462 patients; very-low-quality evidence downgraded for serious bias, serious inconsistency, and serious imprecision). At 72 hours from ROSC, an M1–2 predicted a poor outcome, with 15 (5–31)% FPR and 39 (33–44)% sensitivity(Zandbergen 2006, 62; Topcuoglu 2009, 1635) (2 studies, 322 patients; very-low-quality evidence downgraded for serious bias, serious inconsistency, and very serious imprecision). An absent extensor of abnormal flexion to pain (M1–3) predicted a poor outcome at 12, 24, and 48 hours from ROSC with 57 (37–76)%, 35 (21–52)%, and 10 (3–24)% FPR, respectively(Edgren 1987, 820; Bertini 1989, 627; Bassetti 1996, 610) (3 studies, 120 patients; very-low-quality evidence downgraded for very serious bias, serious inconsistency, and very serious imprecision). At 72 hours, the FPR of this sign was 6 (0–29)%(Bertini 1989, 627) (1 study, 27 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision). GCS. A GCS of 4 or less on admission, at 24 hours, and at 48 hours from ROSC predicted poor outcome, with 40 (19–64)%, 25 (5–57)%, and 0 (0–22)% FPR, respectively(Bassetti 1996, 610; Fischer 2006, 1520) (2 studies, 119 patients; very-low-quality evidence downgraded for serious bias and very serious imprecision). Sensitivity ranged from 54 (37–71)% to 74 (58–86)%. A GCS of 5 or less at 72 hours from ROSC predicted poor outcome, with 75 (63–86)% sensitivity and 7 (1–24)% FPR. Myoclonus and Status Myoclonus. Presence of myoclonus on admission(Wijdicks 1994, 239) (1 study, 107 patients; very-low-quality evidence) or at 24 hours from ROSC(Young 2005, 159) (1 study, 75 patients; very-low-quality evidence) predicts a poor outcome, with 0 (0–14)% and 0 (0–5)% FPR, respectively. A status myoclonus within 24 hours, at 36 to 48 hours, and 72 hours from ROSC predicted a poor outcome, with 0 (0–7)%, 0 (0–5)%, and 0 (0–14)% FPR, respectively(Krumholz 1988, 401; Zandbergen 2006, 62) (2 studies, 464 patients; very-low-quality evidence downgraded for very serious bias and serious imprecision). Sensitivity ranged from 2% to 29%. Electrophysiology Short-Latency SSEPs. For the critical outcome of survival with unfavorable neurologic status or death at discharge, we identified 2 studies on short-latency SSEPs (63 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision)(Brunko 1987, 15; Stelzl 1995, 24) and 3 studies on EEG (46 patients; very-low-quality evidence, downgraded for very serious bias and very serious imprecision).(Vignaendra 1974, 582; Chokroverty 1975, 655; Scollo-Lavizzari 1987, 161) For the critical outcome of survival with unfavorable neurologic status or death at 30 days, we identified 2 studies on SSEPs (80 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision).(Berek 1995, 543; Rothstein 2000, 486) For the critical outcome of survival with unfavorable neurologic status or death at 60 days, we identified 2 studies on EEG (54 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision).(Grindal 1977, 371; Rothstein 1991, 101) For the critical outcome of survival with unfavorable neurologic status or death at 90 days, we identified 2 studies on SSEPs or EEG (102 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision).(Zingler 2003, 79; Young 2005, 159) For the critical outcome of survival with unfavorable neurologic status or death at 180 days, we identified 6 studies on SSEPs or EEG (733 patients; very-low-quality evidence downgraded for serious or very serious bias and serious or very serious imprecision).(Edgren 1987, 820; Madl 2000, 721; Gendo 2001, 1305; Tiainen 2005, 1736; Zandbergen 2006, 583; Zhang 2011, 497) For the critical outcome of survival with unfavorable neurologic status or death at 1 year, we identified 6 studies on SSEPs or EEG (829 patients; low- or very-low-quality evidence downgraded for serious or very serious bias and very serious imprecision).(Bassetti 1996, 610; Kaplan 1999, 205; Berkhoff 2000, 297; Nakabayashi 2001, 1210; Bauer 2003, 283; Zandbergen 2006, 62) Bilateral absence of the N20 wave of short-latency SSEPs predicted death or vegetative state, with 0 (0–12)% FPR as early as 8 hours from cardiac arrest. An FPR of 0% was also confirmed at 24, 48, and 72 hours after ROSC (95% CIs from 0–3 to 0–9) with consistent sensitivity (43%–46%). Among all patients in whom N20 SSEP wave was absent in the first 7 days from cardiac arrest, there was only 1 case of false-positive result.(Young 2005, 159) Quality of evidence was very low in all but 1 study, downgraded for serious or very serious bias and serious or very serious imprecision. Studies assessing the predictive value of a delayed or absent N70 SSEP from 24 hours to 72 hours after ROSC reported a false-positive prediction from 1 (0–7)% to 58 (28–85)%(Madl 2000, 721; Gendo 2001, 1305; Bauer 2003, 283; Young 2005, 159; Zandbergen 2006, 62) (5 studies, 657 subjects; very-low-quality evidence downgraded for serious or very serious bias and serious or very serious imprecision). Blinding of SSEP results, along with criteria for withdrawal of life-sustaining treatment, was not reported in most prognostication studies in resuscitated patients who were not treated with TTM. Electroencephalography. In 1 study(Edgren 1987, 820) (26 patients; very-low-quality evidence downgraded for serious bias and very serious imprecision), an EEG grade 3 to 5 at 24 and 48 hours predicted poor outcome (CPC 3–5), with 0% FPR (95% CIs, 0–22 and 0–24, respectively). An EEG grade 4 to 5 at 72 hours or less from ROSC predicted poor outcome, with 0 (0–11)% FPR and 44 (34–54)% sensitivity(Scollo-Lavizzari 1987, 161; Bassetti 1996, 610; Rothstein 2000, 486) (3 studies, 125 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision). EEG grading systems were not consistent among studies. Presence of burst suppression within 48 hours from ROSC was compatible with recovery of consciousness (FPR 5 [0–26]%(Young 2005, 159); 1 study, 72 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision), while a burst suppression at 72 hours from ROSC predicted poor outcome, with 0 (0–11)% FPR(Zandbergen 2006, 62) (1 study, 277 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision). A low-voltage EEG (20–21 mcV or less) predicted a poor outcome, with 0 (0–15)% FPR within 48 hours from ROSC(Young 2005, 159) (1 study, 72 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision) and with 0 (0–11)% FPR at 72 hours from ROSC(Zandbergen 2006, 62) (1 study, 283 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision). Sensitivity was 15 (7–28)% and 31 (25–37)%, respectively. Presence of alpha coma within 72 hours or from 1 to 7 days after ROSC was not consistently associated with poor outcome (positive predictive value, 96 [80–100]% and 88 [74–96]%)(Vignaendra 1974, 582; Chokroverty 1975, 655; Grindal 1977, 371; Edgren 1987, 820; Kaplan 1999, 205; Berkhoff 2000, 297) (6 studies, 68 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision). Blood Markers In patients who are comatose after resuscitation from cardiac arrest and who are not treated with TTM, high concentrations of biomarkers predict a poor outcome. Advantages of biomarkers over other predictors such as EEG and clinical examination include quantitative results and likely independence from the effects of sedatives. However, the thresholds associated with 0% FPR vary between studies, and S100B thresholds are less well documented than NSE thresholds. The main reasons for the observed variability in biomarkers’ thresholds include the use of heterogeneous measurement techniques,(Bloomfield 2007, 121; Stern 2007, 84; Rundgren 2014, 726) the presence of extraneuronal sources of biomarkers (hemolysis and neuroendocrine tumors for NSE,(Johnsson 2000, 750) muscle and adipose tissue breakdown for S100B,(Anderson 2001, 1255) and the incomplete knowledge of the kinetics of their blood concentrations in the first few days after ROSC. For the critical outcome of survival with unfavorable neurologic status or death at discharge, we identified 2 studies on S100B (99 patients; low- or very-low-quality evidence downgraded for very serious bias and/or very serious imprecision)(Rosen 1998, 473; Hachimi-Idrissi 2002, 251) and 1 study on NSE (73 patients; very-low-quality evidence downgraded for serious bias and very serious imprecision).(Steffen 2010, R69) For the critical outcome of survival with unfavorable neurologic status or death at 90 days, we identified 1 study on NSE (32 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision)(Samaniego 2011, 113) and 1 study on S100B (27 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision).(Zingler 2003, 79) For the critical outcome of survival with unfavorable neurologic status or death at 180 days, we identified 3 studies on NSE or S100B (618 patients; moderate-, low-, or very-low-quality evidence downgraded for serious bias and/or serious or very serious imprecision).(Tiainen 2003, 2881; Zandbergen 2006, 583; Reisinger 2007, 52) For the critical outcome of survival with unfavorable neurologic status or death at 1 year, we identified 2 studies on NSE or S100B (86 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision).(Rosen 2001, 183; Mussack 2002, 2669) Neuron-Specific Enolase. In resuscitated patients with poor neurologic outcome, the blood levels of NSE are higher than those in patients with good neurologic outcome. However, the threshold for prediction of poor outcome with 0% FPR varied between 13.3 and 47.6 mcg/L at 24 hours from ROSC(Tiainen 2003, 2881; Zingler 2003, 79; Zandbergen 2006, 62) (3 studies, 332 patients; very-low-quality evidence), between 8.8 and 65 mcg/L at 48 hours(Rosen 2001, 183; Tiainen 2003, 2881; Zingler 2003, 79; Reisinger 2007, 52) (4 studies, 277 patients; moderate- to very-low-quality evidence), and between 15 and 90.9 mcg/L at 72 hours(Rosen 2001, 183; Zingler 2003, 79; Steffen 2010, R69) (3 studies, 301 patients; low- or very-low-quality evidence). S100B. For S100B, the documented thresholds for 0% FPR ranged between 0.19 and 5.2 mcg/L at 24 hours after ROSC(Tiainen 2003, 2881; Zingler 2003, 79) (2 studies, total 60 patients; very-low-quality evidence) and between 0.12 and 0.8 mcg/L at 48 hours(Rosen 1998, 473; Rosen 2001, 183; Tiainen 2003, 2881; Zingler 2003, 79) (4 studies, 158 patients; very-low-quality evidence). In 1 study (27 patients; very-low-quality evidence), the threshold for prediction of poor outcome with 0% FPR at 72 hours was 0.5 mcg/L. Imaging All prognostication studies on imaging have a small sample size, and in all of them, imaging was performed at the discretion of the treating physician, which may have caused a selection bias and overestimated the performance of index tests. Another limitation is that these methods depend partly on subjective human decision in identifying the region of interest to be studied and in the interpretation of results. For the critical outcome of survival with unfavorable neurologic status or death at discharge, we identified 3 studies on CT (113 patients; very-low-quality evidence)(Morimoto 1993, 104; Torbey 2004, 55; Choi 2008, 666) and 2 studies on MRI (40 patients; very-low-quality evidence).(Berek 1995, 543; Wijdicks 2001, 1561) For the critical outcome of survival with unfavorable neurologic status or death at 90 days, we identified 2 studies on MRI (61 patients; low- or very-low-quality evidence).(Topcuoglu 2009, 1635; Choi 2010, R17) For the critical outcome of survival with unfavorable neurologic status or death at 180 days, we identified 3 studies on MRI (34 patients; very-low-quality evidence).(Els 2004, 361; Wijman 2009, 394; Mlynash 2010, 1665) CT Scan. The main CT finding of global anoxic-ischemic cerebral insult after cardiac arrest is cerebral edema,(Morimoto 1993, 104) which appears as a reduction in the depth of cerebral sulci (sulcal effacement) and an attenuation of the GM/WM interface, due to a decreased density of the GM. This attenuation has been quantitatively measured as the GWR between the GM and the WM densities. In 2 studies(Torbey 2004, 55; Choi 2008, 666) (total 60 patients; very-low-quality evidence), a GWR between the caudate nucleus and the posterior limb of internal capsule (CN/PIC) below 1.22 within 24 hours or below 1.18 within 48 hours from ROSC predicted poor outcome, with 0 (0–28)% and 17 (0–64)% FPR, respectively. At 72 hours from ROSC, the presence of diffuse brain swelling on CT predicts a poor outcome, with 0 (0–45)% FPR and 52 (37–67)% sensitivity(Morimoto 1993, 104) (1 study, 53 patients; very-low-quality evidence). MRI. The main MRI finding of anoxic-ischemic cerebral injury is a hyperintensity in DWI sequences due to cytotoxic edema. In a small study subpopulation(Mlynash 2010, 1665) (12 patients; very-low-quality evidence), presence of diffuse DWI abnormalities in cortex or brainstem at a median of 80 hours from ROSC predicted poor outcome, with 0 (0–35)% FPR. In another small study(Els 2004, 361) (12 patients; very-low-quality evidence), presence of extensive (cortex, basal ganglia, and cerebellum) DWI changes predicted poor outcome, with 0 (0–45)% FPR. Postischemic DWI abnormalities can be quantified by using ADC. ADC values between 700 and 800×10−6 mm2/s are considered normal.(Wijdicks 2001, 1561) In 1 study(Wu 2009, 173) (80 patients; very-low-quality evidence), a whole-brain ADC less than 665×10−6 mm2/s predicted poor outcome, with 0 (0–21)% FPR and 40 (28–53)% sensitivity. In a small subset of another study(Wijman 2009, 394-402) (10 patients; very-low-quality evidence), presence of more than 10% of brain volume with ADC less than 650×10−6 mm2/s predicted poor outcome, with 88 (47–100)% sensitivity and 0 (0–78)% FPR. In another study, an ADC below various thresholds at the level of putamen, thalamus, or occipital cortex at less than 120 hours from ROSC also predicted poor outcome, with 0 (0–31)% FPR. Finally, in 2 studies(Wijdicks 2001, 1561; Topcuoglu 2009, 1635) (total 24 patients; very-low-quality evidence), the presence of extensive cortical global DWI or fluid-attenuated inversion recovery changes within 7 days from arrest predicted poor outcome, with 0 (0–78)% FPR.
Treatment Recommendation:
Clinical Examination We recommend using the absence of PLR (or the combined absence of both pupillary and corneal reflexes) at 72 hours or greater from ROSC to predict poor outcome in patients who are comatose after resuscitation from cardiac arrest and who are not treated with TTM (strong recommendation, very-low quality evidence). We suggest against using an absent or extensor motor response to pain (M≤2) alone to predict poor outcome, given its high FPR (weak recommendation, very-low-quality evidence). However, due to its high sensitivity, this sign may be used to identify the population with poor neurologic status needing prognostication or to predict poor outcome in combination with other more-robust predictors. We suggest using the presence of myoclonus or status myoclonus within 72 hours from ROSC in combination with other predictors to predict poor outcome in comatose survivors of cardiac arrest (weak recommendation, very-low-quality evidence). We suggest prolonging the observation of clinical signs when interference from residual sedation or paralysis is suspected, so that the possibility of obtaining false-positive results is minimized (weak recommendation, very-low-quality evidence). Electrophysiology We recommend using bilateral absence of the N20 SSEP wave within 72 hours from ROSC to predict poor outcome in patients who are comatose after cardiac arrest and who are not treated with TTM (strong recommendation, very-low quality evidence). SSEP recording requires appropriate skills and experience, and utmost care should be taken to avoid electrical interference from muscle artifacts or from the ICU environment. We suggest using the presence of burst suppression on EEG at 72 hours from ROSC in combination with other predictors for prognosticating a poor neurologic outcome in patients who are comatose after cardiac arrest and who are not treated with TTM (strong recommendation, very-low-quality evidence). We suggest against using EEG grades for prognostication due to the inconsistencies in their definitions (weak recommendation, very-low-quality evidence). We suggest against using low-voltage EEG for prognostication, given the potential interferences of technical factors on EEG amplitude (weak recommendation, very-low-quality evidence). Blood Markers We suggest using high serum values of NSE at 24 to 72 hours from ROSC in combination with other predictors for prognosticating a poor neurologic outcome in patients who are comatose after cardiac arrest and who are treated with therapeutic hypothermia (weak recommendation, very-lowquality evidence). However, no threshold-enabling prediction with 0 FPR can be recommended. We suggest using utmost care and preferably sampling at multiple time points when assessing NSE, to avoid false-positive results due to hemolysis. Imaging We suggest using the presence of a marked reduction of the GM/WM ratio on brain CT within 48 hours after ROSC or the presence of extensive reduction in diffusion on brain MRI at 2 to 6 days after ROSC only in combination with other more established predictors for prognosticating a poor neurologic outcome in patients who are comatose after resuscitation from cardiac arrest and who are not treated with TTM (weak recommendation, very-low-quality evidence). We suggest using brain-imaging studies for prognostication only in centers where specific experience is available (weak recommendation, very-low-quality evidence).
CoSTR Attachments:
ALS 713 Prognostication in non-TTM-treated Plenary and TF.pdf    
COSTR Draft Prognostication in Normothermia 2015 - 2.pdf    
Evidence Profile Table 3a NT.pdf    
Evidence Profile Table 3b NT.pdf    
Evidence Profile Table 3c NT.pdf    
Evidence Profile Table 3d NT.pdf    

 Contact Us

 
If you have any comments or questions on this page, please email us at: