- Email: [email protected]
Early CT signs in out-of-hospital cardiac arrest survivors: Temporal profile and prognostic significance
Resuscitation 81 (2010) 534–538
Contents lists available at ScienceDirect
Resuscitation journal homepage: www.elsevier.com/locate/resuscitation
Clinical paper
Early CT signs in out-of-hospital cardiac arrest survivors: Temporal profile and prognostic significance夽 Joji Inamasu a,b,∗ , Satoru Miyatake a , Masaru Suzuki a , Masashi Nakatsukasa b , Hideto Tomioka a , Masanori Honda c , Kenichi Kase a , Kenji Kobayashi a a b c
Department of Emergency Medicine, Saiseikai Utsunomiya Hospital, Utsunomiya, Japan Department of Neurosurgery, Saiseikai Utsunomiya Hospital, Utsunomiya, Japan Department of Diagnostic Radiology, Saiseikai Utsunomiya Hospital, Utsunomiya, Japan
a r t i c l e
i n f o
Article history: Received 26 August 2009 Received in revised form 17 November 2009 Accepted 18 January 2010 Keywords: Early CT sign Loss of boundary Out-of-hospital cardiac arrest Return of spontaneous circulation Sulcal effacement
a b s t r a c t Aim: Although computed tomography (CT) signs of ischaemia, including loss of boundary (LOB) between grey matter and white matter and cortical sulcal effacement, in cardiac arrest (CA) survivors are known, their temporal profile and prognostic significance remains unclear; their clarification is necessary. Methods: Brain CT scans were obtained immediately after resuscitation in 75 non-traumatic CA survivors in a prospective fashion. They were divided into two groups according to the CA-return of spontaneous circulation (ROSC) interval: ≤20 min vs. >20 min. The incidence of the CT signs and predictability of these signs for outcome, assessed 6 months after CA, was evaluated and compared. Results: The incidence of the positive LOB sign was 24% in the ≤20-min group and 83% in the >20-min group, and the difference was statistically significant (p < 0.001). The interval of 20 min seemed to be the time window for the LOB development. The incidence of the positive sulcal effacement sign was 0% in the ≤20 min group and 34% in the >20-min group, and the difference was statistically significant (p = 0.004). A positive LOB sign was predictive of unfavourable outcome with an 81% sensitivity and 92% specificity. A positive sulcal effacement sign was predictive of unfavourable outcome with a 32% sensitivity and 100% specificity. Conclusion: A time window may exist for ischaemic CT signs in CA survivors. The LOB sign may develop when the CA-ROSC interval exceeds 20 min, whereas the sulcal effacement sign may develop later. However, their temporal profile and outcome predictability should be verified by multicentre studies. © 2010 Elsevier Ireland Ltd. All rights reserved.
Treatment of out-of-hospital cardiac arrest (OHCA) survivors has changed dramatically in the past decade. Aggressive therapeutic interventions such as therapeutic hypothermia1,2 and thrombolysis3,4 have been attempted immediately after resuscitation to improve the outcome of OHCA survivors. Although it sounds reasonable that a shorter interval between cardiac arrest (CA) and return of spontaneous circulation (ROSC) results in better prognosis, a therapeutic time window seems to exist: recent studies have shown that the 6-month survival rate of OHCA survivors with the CA-ROSC interval >20 min was significantly lower than those with the interval ≤20 min.5 There are several diagnostic modalities,
夽 A Spanish translated version of the summary of this article appears as Appendix in the final online version at doi:10.1016/j.resuscitation.2010.01.012. ∗ Corresponding author at: Department of Emergency Medicine, Saiseikai Utsunomiya Hospital, 911-1 Takebayashi, Utsunomiya 321-0974, Japan. Tel.: +81 28 626 5500; fax: +81 28 626 5594. E-mail address: [email protected] (J. Inamasu). 0300-9572/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.resuscitation.2010.01.012
which may predict the outcome of OHCA survivors, including blood markers and electrophysiologic studies.6,7 The low sensitivity of the blood markers is an obstacle for their widespread use,6 while the electrophysiologic studies may not be logistically feasible in the acute phase of resuscitation.7 The capability of imaging studies for outcome prediction has also been explored for several years.8 There are two known computed tomography (CT) signs associated with ischaemic brain damage, that is, loss of boundary (LOB) between grey matter and white matter; and cortical sulcal effacement.9–12 Both LOB and cortical sulcal effacement signs may be associated with unfavourable outcome,9,12 and it is expected that these signs be used as early prognosticators in OHCA survivors.9–12 In particular, the density ratio of grey matter to white matter (GM/WM ratio) seems to be an accurate index.10,12 However, these CT signs have been insufficiently investigated compared with those of acute ischaemic stroke patients. Since 2004, we perform a brain CT scan for all non-traumatic OHCA survivors immediately after resuscitation, with the intention of identifying those who sustain irreversible brain damage as early as possible.13 This study was conducted to
J. Inamasu et al. / Resuscitation 81 (2010) 534–538
535
delineate the temporal profile of ischaemic CT signs in OHCA survivors and their potential role in predicting outcome. 1. Methods This study was conducted from July 2006 to June 2008 in a single institution, which is a tertiary referral centre covering a local population of approximately 500,000. The study protocol was approved by our institution’s internal review board. Cardiopulmonary resuscitation (CPR) for OHCA victims was performed following the latest advanced cardiac life support guidelines.14 Pre-hospital data were documented in the Utstein style,15 and a 3-lead electrocardiograph (ECG) was recorded routinely by paramedics. After arrival in the emergency department (ED), the temporal sequence of the resuscitative events was recorded on a minute-by-minute basis by emergency medicine residents. Immediately after successful CPR, OHCA survivors were brought to a CT suite adjacent to the ED. The room was equipped with a General Electric Light-Speed 16detector row CT scanner, and contiguous non-helical 5-mm slices were obtained. Exclusion criteria for the brain CT-scan protocol included the following: (1) a blood pressure not stabilised even with the use of a maximal dose of vasopressors; (2) non-traumatic OHCA due to extrinsic causes, such as asphyxia and near-drowning; (3) when the collapse of survivors was not witnessed; and (4) OHCA survivors with CT evidence of an intracranial haemorrhage. During the 2-year period, 75 witnessed non-traumatic OHCA survivors with a presumed cardiac aetiology underwent a CT scan immediately after resuscitation. Their charts were reviewed thoroughly. To clarify the temporal profile of the CT signs, the 75 OHCA survivors were divided into two groups according to the CA-ROSC interval: ≤20 min vs. >20 min. An inter-group comparison was made for the incidence of the CT signs as well as for various demographic variables, including age, male:female ratio, history of cardiac diseases, percentage of ventricular fibrillation (%VF) as the initial rhythm, frequency of bystander CPR, door-to-CT time, body temperature on arrival and unfavourable outcome rate. The outcomes were evaluated 6 months after CA for long-term survivors with a Pittsburgh cerebral performance category (CPC) score. Those with a CPC1–2 and CPC3–5 were defined as having favourable and unfavourable outcomes, respectively. In addition, demographic variables were compared between those with favourable and unfavourable outcomes. 1.1. Evaluation of CT signs The temporal profile was described separately for each CT sign, that is, LOB between grey matter and white matter and cortical sulcal effacement. The presence of the LOB sign was evaluated at the level of the basal ganglia, and that of the sulcal effacement sign was evaluated at the level of the centrum semiovale.10 The CT findings were available to the clinical team caring for the patients. A boardcertified radiologist and a board-certified neurosurgeon, both of whom are involved in interpreting CT scans of acute stroke patients for whom thrombolysis is considered, independently read all the CT scans without being notified of the patient’s status. They determined the presence of the CT signs in each OHCA survivor. When there was a disagreement, the CT scans were reviewed together by them and a decision was made. The inter-observer concordance was calculated along with the kappa coefficient.
Fig. 1. The Hounsfield unit was measured at the grey matter (GM, arrowhead) and white matter (WM, arrow) at the level of the basal ganglia to calculate the GM/WM ratio.
was performed according to the method described by Torbey et al.10 The measuring cursor was configured as a 10-mm2 square surface, and the slice thickness was 5 mm. A representative CT scan illustrating the HU measurement is shown in Fig. 1. The GM/WM ratio was compared between those who were positive and those who were negative for the LOB sign. 1.3. Early CT signs as predictors of unfavourable outcome The rate of unfavourable outcome was compared between those who were positive and negative for the LOB and sulcal effacement signs, respectively. In addition, the sensitivity and specificity of each CT sign to predict unfavourable outcome was calculated. 1.4. Statistical analysis Continuous variables, indicated as mean ± SD, were compared using the Student’s t-test, and categorical variables were compared using Fisher’s exact test. The kappa coefficient was calculated using a 2 × 2 table. Statistical analyses were performed using Statview 5.0, and a p < 0.05 was considered statistically significant. 2. Results
1.2. GM/WM ratio
2.1. Demographics
The Hounsfield unit (HU) of grey matter and white matter was measured at the level of the basal ganglia, and the GM/WM ratio was calculated to ensure that the aforementioned determination of the LOB sign by the two observers was valid. The HU measurement
The number of OHCA survivors with the CA-ROSC interval ≤20 min was 17, while the number of those with the interval >20 min was 58. There was no significant difference in age, male:female ratio, history of cardiac diseases, %VF, frequency of
536
J. Inamasu et al. / Resuscitation 81 (2010) 534–538
Table 1 Demographics of 75 out-of-hospital cardiac arrest survivors classified by cardiac arrest-return of spontaneous circulation interval. CA-ROSC interval
N
Age (y)
M:F
Hx of CVD
VF as initial rhythm
Bystander CPR
Temperature (◦ C)
Door-to-CT time (min)
Favourable outcome at 6 mo
≤20 min >20 min p
17 58
70.2 ± 15.8 67.7 ± 14.6 0.27
8:9 37:21 0.34
7 (41%) 20 (34%) 0.77
8 (47%) 20 (34%) 0.81
8 (47%) 22 (38%) 0.63
35.6 ± 0.9 35.0 ± 1.2 0.06
45.6 ± 25.2 54.6 ± 21.9 0.32
6 (35%) 6 (10%) 0.02
CA, cardiac arrest; CPR, cardiopulmonary resuscitation; CVD, cardiovascular disease; Hx, history; M:F, male: female; NS, not significant; ROSC, return of spontaneous circulation; VF, ventricular fibrillation. Table 2 Demographics of 75 out-of-hospital cardiac arrest survivors classified by 6-month outcome (favourable vs. unfavourable) after cardiac arrest. Outcome at 6 mo
N
Age (y)
M:F
CPC score (N)
Hx of CVD
VF as initial rhythm
Bystander CPR
Temperature Door-to-CT time (min) (C◦ )
CA-ROSC interval (min)
Favourable Unfavourable p
12 63
59.3 ± 13.8 70.0 ± 14.4 0.01
7:5 38:25 0.85
CPC 1: 8, CPC 2: 4 CPC 3: 1, CPC 5: 62 N/A
7 (58%) 20 (32%) 0.10
10 (83%) 18 (29%) 0.006
8 (67%) 22 (35%) 0.06
35.8 ± 1.0 35.1 ± 1.1 0.07
22.7 ± 7.5 35.0 ± 12.2 <0.001
47.3 ± 24.0 54.9 ± 22.2 0.14
CA, cardiac arrest; CPC, cerebral performance category; CPR, cardiopulmonary resuscitation; CVD, cardiovascular disease; Hx, history; M:F, male: female; NS, not significant; ROSC, return of spontaneous circulation; VF, ventricular fibrillation.
bystander CPR, body temperature and door-to-CT time between the two groups (Table 1). The number of OHCA survivors with favourable outcome was six (35%) in those with the interval ≤20 min and six (10%) in those with the interval >20 min. The difference was statistically significant (p = 0.02). In other words, the number of those with favourable outcome was 12, while the number of those with unfavourable outcome was 63. The two groups were also compared, and significant difference was present on age, %VF and the CA-ROSC interval (Table 2). Similarly, the frequency of bystander CPR and body temperature trended to be higher in the group with favourable outcome (Table 2). 2.2. CT signs of ischaemia The number of OHCA survivors who showed a positive LOB sign was four (24%) in those with the interval ≤20 min and 48 (83%) in those with the interval >20 min, and the difference was statistically significant (p < 0.001). The number of survivors who showed a positive sulcal effacement sign was 0 (0%) in those with the interval ≤20 min and 20 (34%) in those with the interval >20 min, and the difference was statistically significant (p = 0.004). The results are summarised in Table 3. Distribution of the 75 OHCA survivors in relation to the CA-ROSC interval was illustrated as a histogram, with the interval on the x-axis and the number of cases on the y-axis (Fig. 2). Those who were positive only for the LOB sign were shown in dark, and those who were positive both for the LOB and sulcal effacement signs were shown in black. There were no survivors who were positive only for the sulcal effacement sign. The number of survivors with positive CT signs seemed to increase substantially when the interval exceeded 20 min (Fig. 2). Inter-observer concordance was 76% (57/75) for the LOB and was 80% (60/75) for the sulcal effacement sign, respectively. The kappa coefficient was 0.51 for the LOB and 0.48 for the sulcal effacement sign, respectively.
Table 3 The incidence of early CT signs in relation to the cardiac arrest-return of spontaneous circulation interval. CA-ROSC interval
N
LOB sign (+)
Sulcal effacement sign (+)
≤20 min >20 min p
17 58
4 (24%) 48 (83%) <0.001
0 (0%) 20 (34%) 0.004
CA, cardiac arrest; LOB, loss of boundary; ROSC, return of spontaneous circulation.
2.3. GM/WM ratio A total of 52 OHCA survivors had a positive LOB sign. Their GM/WM ratio ranged from 0.98 to 1.18, with a mean of 1.10 ± 0.05 (Fig. 3). On the other hand, that of the 23 survivors who had a negative LOB sign ranged from 1.13 to 1.32, with a mean of 1.26 ± 0.06 (Fig. 3). The GM/WM ratio was significantly higher for OHCA survivors with a negative LOB sign (p < 0.01), although there were some overlaps. 2.4. CT signs as predictors of unfavourable outcome Among the 52 OHCA survivors with a positive LOB sign, 50 died after admission, one survived with a CPC 4 and the other survived with a CPC 2; the unfavourable outcome rate was 98%. By contrast, among 23 survivors with a negative LOB, 12 died and the other 11 survivors had either CPC 1 or 2, and the unfavourable outcome rate was 52%. Therefore, a positive LOB sign was predictive of unfavourable outcome with a sensitivity of 81% and specificity of 92% (Table 4). A total of 20 OHCA survivors had a positive sulcal effacement sign, among which none survived till discharge, and the unfavourable outcome rate was 100%. By contrast, 55 survivors had a negative sulcal effacement sign, among which 42 died, one survived with a CPC 4, and the other 12 survived either with CPC 1 or 2; the unfavourable outcome rate was 78%. Thus, a positive sulcal effacement sign was predictive of unfavourable outcome with a sensitivity of 32% and specificity of 100% (Table 4). 3. Discussion Ischaemic CT signs in OHCA survivors have been known for more than two decades,9 and LOB between the grey matter and white matter and cortical sulcal effacement are the signs most frequently documented in the literature.9–12 It is unlikely that such ischaemic CT signs develop if the CA-ROSC interval is short. It is not yet clear, however, how short an interval is safe enough, because few studies investigating the relationship between the interval and incidence of the CT signs have been conducted in OHCA survivors. Moreover, the timing of brain CT scan was not uniform in previous studies, which ranged from several hours to even days after CA.9–12 Furthermore, CT scans for OHCA survivors have not been performed prospectively. Therefore, it remains unclear how early these ischaemic signs become recognisable on CT scans. It also remains unclear if
J. Inamasu et al. / Resuscitation 81 (2010) 534–538
537
Fig. 2. A histogram showing distribution of the 75 OHCA survivors in relation to the CA-ROSC interval. Those who were positive only for a LOB sign are indicated in dark, whereas those who were positive both for LOB and sulcal effacement signs are indicated in black.
a clear-cut time window exists for CT signs in OHCA survivors. Is it really appropriate to call them ‘early’ CT signs, when CT scans might not have been taken early enough? In this study, there was a significant difference in the incidence of CT signs as well as in the unfavourable outcome rate between those with the CA-ROSC interval ≤20 min and those with the interval >20 min (Tables 1 and 3). As illustrated in Fig. 2, an interval of 20 min seemed to be the time window for development of the LOB sign in OHCA survivors, and those with an interval of >30 min developed a positive LOB sign almost invariably. These results are compatible with previous studies, which reported that the CAROSC interval >20–25 min was highly predictive of unfavourable outcome in cardiac OHCA survivors.5,16 It is also likely that development of the LOB sign in OHCA survivors is time-dependent and predictable. As with previous studies, the presence of the LOB sign was evaluated at the level of the basal ganglia, which is one of the brain areas most vulnerable to ischaemic insults.10–12 The sulcal effacement sign may develop in a more delayed fashion and may not be as time dependent and predictable as the LOB sign. Following these results, we believe that it is appropriate to call these
CT signs ‘early’ signs, because they are often recognisable on a CT scan obtained as early as 1 h after CA. Both CT signs may be useful for identifying OHCA survivors who have unfavourable outcome. LOB may be a more sensitive sign of irreversible ischaemia, as it develops earlier in the course and may have a time window. There were two survivors, who showed a positive LOB sign, and yet were discharged alive, and LOB per se may not be a sign of fatal brain injury. By contrast, sulcal effacement seems to be a more specific sign of fatal brain injury, because none of those with positive sulcal effacement sign were discharged alive. The demographic variables, divided by the CA-ROSC interval of 20 min (Table 1), were not significantly different between the two groups, suggesting that the difference in the incidence of the CT signs might not be due to the demographic differences. On the other hand, comparison between those with favourable and unfavourable outcomes (Table 2) revealed that the former group was significantly younger, presented more frequently with VF rhythm, and had a significantly shorter CA-ROSC interval. The latter results are well expected and are compatible with those from literature.6,16 The high inter-observer concordance for both signs indicates good reproducibility of the CT signs for experienced eyes, although the kappa coefficient was not as high as had been expected. The distinct difference in the GM/WM ratio between those with a positive and negative LOB signs (Fig. 3) may also support the accuracy and validity of CT interpretations. The GM/WM ratio may be a reliable prognosticator for OHCA survivors.10,12 Table 4 Sensitivity/specificity of the early CT signs in predicting unfavourable outcome. Early CT signs
Favourable
Unfavourable
LOB sign (+) LOB sign (−)
1 11
51 12
Sulcal effacement sign (+) Sulcal effacement sign (−) Fig. 3. Comparison of the GM/WM ratio between positive (left) and negative (right) LOB, showing that the difference was statistically significant (p < 0.05).
LOB, loss of boundary.
0
20
12
43
Sensitivity
Specificity
81%
92%
32%
100%
538
J. Inamasu et al. / Resuscitation 81 (2010) 534–538
Although the ratio was significantly higher for those who survived than those who died,10,12 this study is the first to show that the GM/WM ratio was significantly higher for those with a negative LOB sign. We presumed that most survivors for whom CT scan was obtained had sustained a CA with cardiac aetiology. We made every effort to exclude those with asphyxia, in whom the degree of hypoxia is more severe and CT signs may develop earlier than those with cardiac aetiology. The temporal profile of the CT signs in cardiac OHCA survivors is evidently different from that of acute stroke patients in whom the ischaemic interval and development of CT signs is not necessarily correlated.17,18 The lack of a temporal correlation in acute ischaemic stroke patients may be explained by significant individual variability in the collateral brain circulation.17,18 Compared with acute stroke patients, OHCA survivors are more uniform in the degree of ischaemia they sustain because there is a total caessation of blood flow to the brain invariably after CA. The CA-ROSC interval is the most important factor in determining the severity of ischaemic brain damage in OHCA survivors. This uniformity in the degree of ischaemia after CA may explain why the relationship between the ischaemic interval and incidence of early CT signs in OHCA survivors is more time dependent and predictable than that in acute stroke patients. It remains unclear what is the best imaging modality to detect and evaluate CA-induced ischaemic brain damage in humans. By analogy from acute stroke patients, magnetic resonance imaging (MRI), particularly a diffusion-weighted imaging sequence, may detect areas of brain damage earlier and more clearly than a CT scan in OHCA survivors.19,20 However, because of the feasibility and safety concerns, the role of MRI in the acute phase of resuscitation may be limited in many OHCA survivors, who require a ventilator. Although selected survivors may benefit from undergoing MRI in the sub-acute to chronic phases for precise assessment of brain damage, its routine use immediately after resuscitation may neither be practical nor recommended, and CT scan may be the imaging modality of choice. There are several limitations in this study. First, accuracy of the CA-ROSC interval may be hard to verify because the information on the time when the patient collapsed relied solely on the memory of witnesses. Information on the time when ROSC was achieved was given either by paramedics or by emergency medicine residents depending on where ROSC was achieved. Both of these timings may be subject to reporting or measurement errors, and therefore, the interval might not have been as accurate as in an experimental setting. Second, we assumed that most OHCA survivors had a cardiac aetiology. Although those with asphyxia were excluded from the study, it is probable that those with non-cardiac causes of CA, such as a ruptured aortic aneurysm or a pulmonary embolism, might have been included. Brain ischaemia in those with a ruptured aortic aneurysm or a pulmonary embolism develops differently compared with those with a cardiac aetiology.21,22 It remains unknown how many of the 75 survivors actually sustained a non-cardiac CA, because routine autopsies were not performed. In addition, the neurological status of each OHCA survivor, believed to reliably predict outcome,23 was not taken into account in this study. When neurological status and imaging studies are combined, better prediction of outcome is possible and their combined efficacy should be addressed in future studies. Finally, the number of 75 survivors evaluated in this study may not be large enough for statistical power, despite the fact that previous studies comprised less than 30 cases.10–12 CT scans for OHCA survivors immediately after resuscitation may be obtained safely without much ethical and economical concern,13 and we expect that the reproducibility of the temporal profile and the prognostic significance of early CT signs reported here will be verified by multicentre studies.
4. Conclusions This retrospective study indicates that a time window may exist for early CT signs in CA survivors. The LOB sign may develop when the CA-ROSC interval exceeds 20 min, whereas the sulcal effacement sign may develop later. The former seems to be predictive of unfavourable outcome with high sensitivity, while the latter seems to be a prognosticator with high specificity. Conflict of interest statement None. There are no financial and personal relationships with other people or organisations that could inappropriately influence our work. References 1. Hypothermia after Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med 2002;346:549–56. 2. Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med 2002;346:557–63. 3. Böttiger BW, Arntz HR, Chamberlain DA, et al. Thrombolysis during resuscitation for out-of-hospital cardiac arrest. N Engl J Med 2008;359:2651–62. 4. Mysiak A, Nowicki P, Kobusiak-Prokopowicz M. Thrombolysis during cardiopulmonary resuscitation. Cardiol J 2007;14:24–8. 5. Belliard G, Catez E, Charron C, et al. Efficacy of therapeutic hypothermia after out-of-hospital cardiac arrest due to ventricular fibrillation. Resuscitation 2007;75:252–9. 6. Oksanen T, Tiainen M, Skrifvars MB, et al. Predictive power of serum NSE and OHCA score regarding 6-month neurologic outcome after out-ofhospital ventricular fibrillation and therapeutic hypothermia. Resuscitation 2009;80:165–70. 7. Koenig MA, Kaplan PW, Thakor NV. Clinical neurophysiologic monitoring and brain injury from cardiac arrest. Neurol Clin 2006;24:89–106. 8. Kandiah P, Ortega S, Torbey MT. Biomarkers and neuroimaging of brain injury after cardiac arrest. Semin Neurol 2006;26:413–21. 9. Kjos BO, Brant-Zawadzki M, Young RG. Early CT findings of global central nervous system hypoperfusion. Am J Roentgenol 1983;141:1227–32. 10. Torbey MT, Selim M, Knorr J, Bigelow C, Recht L. Quantitative analysis of the loss of distinction between gray and white matter in comatose patients after cardiac arrest. Stroke 2000;31:2163–7. 11. Yanagawa Y, Un-no Y, Sakamoto T, Okada Y. Cerebral density on CT immediately after a successful resuscitation of cardiopulmonary arrest correlates with outcome. Resuscitation 2005;64:97–101. 12. Choi SP, Park HK, Park KN, et al. The density ratio of grey to white matter on computed tomography as an early predictor of vegetative state or death after cardiac arrest. Emerg Med J 2008;25:666–9. 13. Inamasu J, Miyatake S, Tomioka H, et al. Subarachnoid haemorrhage as a cause of out-of-hospital cardiac arrest: a prospective computed tomography study. Resuscitation 2009;80:977–80. 14. International Liaison Committee on Resuscitation. 2005 International consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations. Part 4: Advanced life support. Resuscitation 2005;67:213–47. 15. Langhelle A, Nolan J, Herlitz J, et al. Recommended guidelines for reviewing, reporting, and conducting research on post-resuscitation care: the Utstein style. Resuscitation 2005;66:271–83. 16. Oddo M, Ribordy V, Feihl F, et al. Early predictors of outcome in comatose survivors of ventricular fibrillation and non-ventricular fibrillation cardiac arrest treated with hypothermia: a prospective study. Crit Care Med 2008;36:2296–301. 17. Moulin T, Cattin F, Crépin-Leblond T, et al. Early CT signs in acute middle cerebral artery infarction: predictive value for subsequent infarct locations and outcome. Neurology 1996;47:366–75. 18. Kucinski T, Koch C, Grzyska U, Freitag HJ, Krömer H, Zeumer H. The predictive value of early CT and angiography for fatal hemispheric swelling in acute stroke. Am J Neuroradiol 1998;19:839–46. 19. Arbelaez A, Castillo M, Mukherji SK. Diffusion-weighted MR imaging of global cerebral anoxia. Am J Neuroradiol 1999;20:999–1007. 20. Barrett KM, Freeman WD, Weindling SM, et al. Brain injury after cardiopulmonary arrest and its assessment with diffusion-weighted magnetic resonance imaging. Mayo Clin Proc 2007;82:828–35. 21. Virkkunen I, Paasio L, Ryynänen S, et al. Pulseless electrical activity and unsuccessful out-of-hospital resuscitation: what is the cause of death? Resuscitation 2008;77:207–10. 22. Pierce LC, Courtney DM. Clinical characteristics of aortic aneurysm and dissection as a cause of sudden death in outpatients. Am J Emerg Med 2008;26:1042–6. 23. Booth CM, Boone RH, Tomlinson G, Detsky AS. Is this patient dead, vegetative, or severely neurologically impaired? Assessing outcome for comatose survivors of cardiac arrest. JAMA 2004;291:870–9.
Contents lists available at ScienceDirect
Resuscitation journal homepage: www.elsevier.com/locate/resuscitation
Clinical paper
Early CT signs in out-of-hospital cardiac arrest survivors: Temporal profile and prognostic significance夽 Joji Inamasu a,b,∗ , Satoru Miyatake a , Masaru Suzuki a , Masashi Nakatsukasa b , Hideto Tomioka a , Masanori Honda c , Kenichi Kase a , Kenji Kobayashi a a b c
Department of Emergency Medicine, Saiseikai Utsunomiya Hospital, Utsunomiya, Japan Department of Neurosurgery, Saiseikai Utsunomiya Hospital, Utsunomiya, Japan Department of Diagnostic Radiology, Saiseikai Utsunomiya Hospital, Utsunomiya, Japan
a r t i c l e
i n f o
Article history: Received 26 August 2009 Received in revised form 17 November 2009 Accepted 18 January 2010 Keywords: Early CT sign Loss of boundary Out-of-hospital cardiac arrest Return of spontaneous circulation Sulcal effacement
a b s t r a c t Aim: Although computed tomography (CT) signs of ischaemia, including loss of boundary (LOB) between grey matter and white matter and cortical sulcal effacement, in cardiac arrest (CA) survivors are known, their temporal profile and prognostic significance remains unclear; their clarification is necessary. Methods: Brain CT scans were obtained immediately after resuscitation in 75 non-traumatic CA survivors in a prospective fashion. They were divided into two groups according to the CA-return of spontaneous circulation (ROSC) interval: ≤20 min vs. >20 min. The incidence of the CT signs and predictability of these signs for outcome, assessed 6 months after CA, was evaluated and compared. Results: The incidence of the positive LOB sign was 24% in the ≤20-min group and 83% in the >20-min group, and the difference was statistically significant (p < 0.001). The interval of 20 min seemed to be the time window for the LOB development. The incidence of the positive sulcal effacement sign was 0% in the ≤20 min group and 34% in the >20-min group, and the difference was statistically significant (p = 0.004). A positive LOB sign was predictive of unfavourable outcome with an 81% sensitivity and 92% specificity. A positive sulcal effacement sign was predictive of unfavourable outcome with a 32% sensitivity and 100% specificity. Conclusion: A time window may exist for ischaemic CT signs in CA survivors. The LOB sign may develop when the CA-ROSC interval exceeds 20 min, whereas the sulcal effacement sign may develop later. However, their temporal profile and outcome predictability should be verified by multicentre studies. © 2010 Elsevier Ireland Ltd. All rights reserved.
Treatment of out-of-hospital cardiac arrest (OHCA) survivors has changed dramatically in the past decade. Aggressive therapeutic interventions such as therapeutic hypothermia1,2 and thrombolysis3,4 have been attempted immediately after resuscitation to improve the outcome of OHCA survivors. Although it sounds reasonable that a shorter interval between cardiac arrest (CA) and return of spontaneous circulation (ROSC) results in better prognosis, a therapeutic time window seems to exist: recent studies have shown that the 6-month survival rate of OHCA survivors with the CA-ROSC interval >20 min was significantly lower than those with the interval ≤20 min.5 There are several diagnostic modalities,
夽 A Spanish translated version of the summary of this article appears as Appendix in the final online version at doi:10.1016/j.resuscitation.2010.01.012. ∗ Corresponding author at: Department of Emergency Medicine, Saiseikai Utsunomiya Hospital, 911-1 Takebayashi, Utsunomiya 321-0974, Japan. Tel.: +81 28 626 5500; fax: +81 28 626 5594. E-mail address: [email protected] (J. Inamasu). 0300-9572/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.resuscitation.2010.01.012
which may predict the outcome of OHCA survivors, including blood markers and electrophysiologic studies.6,7 The low sensitivity of the blood markers is an obstacle for their widespread use,6 while the electrophysiologic studies may not be logistically feasible in the acute phase of resuscitation.7 The capability of imaging studies for outcome prediction has also been explored for several years.8 There are two known computed tomography (CT) signs associated with ischaemic brain damage, that is, loss of boundary (LOB) between grey matter and white matter; and cortical sulcal effacement.9–12 Both LOB and cortical sulcal effacement signs may be associated with unfavourable outcome,9,12 and it is expected that these signs be used as early prognosticators in OHCA survivors.9–12 In particular, the density ratio of grey matter to white matter (GM/WM ratio) seems to be an accurate index.10,12 However, these CT signs have been insufficiently investigated compared with those of acute ischaemic stroke patients. Since 2004, we perform a brain CT scan for all non-traumatic OHCA survivors immediately after resuscitation, with the intention of identifying those who sustain irreversible brain damage as early as possible.13 This study was conducted to
J. Inamasu et al. / Resuscitation 81 (2010) 534–538
535
delineate the temporal profile of ischaemic CT signs in OHCA survivors and their potential role in predicting outcome. 1. Methods This study was conducted from July 2006 to June 2008 in a single institution, which is a tertiary referral centre covering a local population of approximately 500,000. The study protocol was approved by our institution’s internal review board. Cardiopulmonary resuscitation (CPR) for OHCA victims was performed following the latest advanced cardiac life support guidelines.14 Pre-hospital data were documented in the Utstein style,15 and a 3-lead electrocardiograph (ECG) was recorded routinely by paramedics. After arrival in the emergency department (ED), the temporal sequence of the resuscitative events was recorded on a minute-by-minute basis by emergency medicine residents. Immediately after successful CPR, OHCA survivors were brought to a CT suite adjacent to the ED. The room was equipped with a General Electric Light-Speed 16detector row CT scanner, and contiguous non-helical 5-mm slices were obtained. Exclusion criteria for the brain CT-scan protocol included the following: (1) a blood pressure not stabilised even with the use of a maximal dose of vasopressors; (2) non-traumatic OHCA due to extrinsic causes, such as asphyxia and near-drowning; (3) when the collapse of survivors was not witnessed; and (4) OHCA survivors with CT evidence of an intracranial haemorrhage. During the 2-year period, 75 witnessed non-traumatic OHCA survivors with a presumed cardiac aetiology underwent a CT scan immediately after resuscitation. Their charts were reviewed thoroughly. To clarify the temporal profile of the CT signs, the 75 OHCA survivors were divided into two groups according to the CA-ROSC interval: ≤20 min vs. >20 min. An inter-group comparison was made for the incidence of the CT signs as well as for various demographic variables, including age, male:female ratio, history of cardiac diseases, percentage of ventricular fibrillation (%VF) as the initial rhythm, frequency of bystander CPR, door-to-CT time, body temperature on arrival and unfavourable outcome rate. The outcomes were evaluated 6 months after CA for long-term survivors with a Pittsburgh cerebral performance category (CPC) score. Those with a CPC1–2 and CPC3–5 were defined as having favourable and unfavourable outcomes, respectively. In addition, demographic variables were compared between those with favourable and unfavourable outcomes. 1.1. Evaluation of CT signs The temporal profile was described separately for each CT sign, that is, LOB between grey matter and white matter and cortical sulcal effacement. The presence of the LOB sign was evaluated at the level of the basal ganglia, and that of the sulcal effacement sign was evaluated at the level of the centrum semiovale.10 The CT findings were available to the clinical team caring for the patients. A boardcertified radiologist and a board-certified neurosurgeon, both of whom are involved in interpreting CT scans of acute stroke patients for whom thrombolysis is considered, independently read all the CT scans without being notified of the patient’s status. They determined the presence of the CT signs in each OHCA survivor. When there was a disagreement, the CT scans were reviewed together by them and a decision was made. The inter-observer concordance was calculated along with the kappa coefficient.
Fig. 1. The Hounsfield unit was measured at the grey matter (GM, arrowhead) and white matter (WM, arrow) at the level of the basal ganglia to calculate the GM/WM ratio.
was performed according to the method described by Torbey et al.10 The measuring cursor was configured as a 10-mm2 square surface, and the slice thickness was 5 mm. A representative CT scan illustrating the HU measurement is shown in Fig. 1. The GM/WM ratio was compared between those who were positive and those who were negative for the LOB sign. 1.3. Early CT signs as predictors of unfavourable outcome The rate of unfavourable outcome was compared between those who were positive and negative for the LOB and sulcal effacement signs, respectively. In addition, the sensitivity and specificity of each CT sign to predict unfavourable outcome was calculated. 1.4. Statistical analysis Continuous variables, indicated as mean ± SD, were compared using the Student’s t-test, and categorical variables were compared using Fisher’s exact test. The kappa coefficient was calculated using a 2 × 2 table. Statistical analyses were performed using Statview 5.0, and a p < 0.05 was considered statistically significant. 2. Results
1.2. GM/WM ratio
2.1. Demographics
The Hounsfield unit (HU) of grey matter and white matter was measured at the level of the basal ganglia, and the GM/WM ratio was calculated to ensure that the aforementioned determination of the LOB sign by the two observers was valid. The HU measurement
The number of OHCA survivors with the CA-ROSC interval ≤20 min was 17, while the number of those with the interval >20 min was 58. There was no significant difference in age, male:female ratio, history of cardiac diseases, %VF, frequency of
536
J. Inamasu et al. / Resuscitation 81 (2010) 534–538
Table 1 Demographics of 75 out-of-hospital cardiac arrest survivors classified by cardiac arrest-return of spontaneous circulation interval. CA-ROSC interval
N
Age (y)
M:F
Hx of CVD
VF as initial rhythm
Bystander CPR
Temperature (◦ C)
Door-to-CT time (min)
Favourable outcome at 6 mo
≤20 min >20 min p
17 58
70.2 ± 15.8 67.7 ± 14.6 0.27
8:9 37:21 0.34
7 (41%) 20 (34%) 0.77
8 (47%) 20 (34%) 0.81
8 (47%) 22 (38%) 0.63
35.6 ± 0.9 35.0 ± 1.2 0.06
45.6 ± 25.2 54.6 ± 21.9 0.32
6 (35%) 6 (10%) 0.02
CA, cardiac arrest; CPR, cardiopulmonary resuscitation; CVD, cardiovascular disease; Hx, history; M:F, male: female; NS, not significant; ROSC, return of spontaneous circulation; VF, ventricular fibrillation. Table 2 Demographics of 75 out-of-hospital cardiac arrest survivors classified by 6-month outcome (favourable vs. unfavourable) after cardiac arrest. Outcome at 6 mo
N
Age (y)
M:F
CPC score (N)
Hx of CVD
VF as initial rhythm
Bystander CPR
Temperature Door-to-CT time (min) (C◦ )
CA-ROSC interval (min)
Favourable Unfavourable p
12 63
59.3 ± 13.8 70.0 ± 14.4 0.01
7:5 38:25 0.85
CPC 1: 8, CPC 2: 4 CPC 3: 1, CPC 5: 62 N/A
7 (58%) 20 (32%) 0.10
10 (83%) 18 (29%) 0.006
8 (67%) 22 (35%) 0.06
35.8 ± 1.0 35.1 ± 1.1 0.07
22.7 ± 7.5 35.0 ± 12.2 <0.001
47.3 ± 24.0 54.9 ± 22.2 0.14
CA, cardiac arrest; CPC, cerebral performance category; CPR, cardiopulmonary resuscitation; CVD, cardiovascular disease; Hx, history; M:F, male: female; NS, not significant; ROSC, return of spontaneous circulation; VF, ventricular fibrillation.
bystander CPR, body temperature and door-to-CT time between the two groups (Table 1). The number of OHCA survivors with favourable outcome was six (35%) in those with the interval ≤20 min and six (10%) in those with the interval >20 min. The difference was statistically significant (p = 0.02). In other words, the number of those with favourable outcome was 12, while the number of those with unfavourable outcome was 63. The two groups were also compared, and significant difference was present on age, %VF and the CA-ROSC interval (Table 2). Similarly, the frequency of bystander CPR and body temperature trended to be higher in the group with favourable outcome (Table 2). 2.2. CT signs of ischaemia The number of OHCA survivors who showed a positive LOB sign was four (24%) in those with the interval ≤20 min and 48 (83%) in those with the interval >20 min, and the difference was statistically significant (p < 0.001). The number of survivors who showed a positive sulcal effacement sign was 0 (0%) in those with the interval ≤20 min and 20 (34%) in those with the interval >20 min, and the difference was statistically significant (p = 0.004). The results are summarised in Table 3. Distribution of the 75 OHCA survivors in relation to the CA-ROSC interval was illustrated as a histogram, with the interval on the x-axis and the number of cases on the y-axis (Fig. 2). Those who were positive only for the LOB sign were shown in dark, and those who were positive both for the LOB and sulcal effacement signs were shown in black. There were no survivors who were positive only for the sulcal effacement sign. The number of survivors with positive CT signs seemed to increase substantially when the interval exceeded 20 min (Fig. 2). Inter-observer concordance was 76% (57/75) for the LOB and was 80% (60/75) for the sulcal effacement sign, respectively. The kappa coefficient was 0.51 for the LOB and 0.48 for the sulcal effacement sign, respectively.
Table 3 The incidence of early CT signs in relation to the cardiac arrest-return of spontaneous circulation interval. CA-ROSC interval
N
LOB sign (+)
Sulcal effacement sign (+)
≤20 min >20 min p
17 58
4 (24%) 48 (83%) <0.001
0 (0%) 20 (34%) 0.004
CA, cardiac arrest; LOB, loss of boundary; ROSC, return of spontaneous circulation.
2.3. GM/WM ratio A total of 52 OHCA survivors had a positive LOB sign. Their GM/WM ratio ranged from 0.98 to 1.18, with a mean of 1.10 ± 0.05 (Fig. 3). On the other hand, that of the 23 survivors who had a negative LOB sign ranged from 1.13 to 1.32, with a mean of 1.26 ± 0.06 (Fig. 3). The GM/WM ratio was significantly higher for OHCA survivors with a negative LOB sign (p < 0.01), although there were some overlaps. 2.4. CT signs as predictors of unfavourable outcome Among the 52 OHCA survivors with a positive LOB sign, 50 died after admission, one survived with a CPC 4 and the other survived with a CPC 2; the unfavourable outcome rate was 98%. By contrast, among 23 survivors with a negative LOB, 12 died and the other 11 survivors had either CPC 1 or 2, and the unfavourable outcome rate was 52%. Therefore, a positive LOB sign was predictive of unfavourable outcome with a sensitivity of 81% and specificity of 92% (Table 4). A total of 20 OHCA survivors had a positive sulcal effacement sign, among which none survived till discharge, and the unfavourable outcome rate was 100%. By contrast, 55 survivors had a negative sulcal effacement sign, among which 42 died, one survived with a CPC 4, and the other 12 survived either with CPC 1 or 2; the unfavourable outcome rate was 78%. Thus, a positive sulcal effacement sign was predictive of unfavourable outcome with a sensitivity of 32% and specificity of 100% (Table 4). 3. Discussion Ischaemic CT signs in OHCA survivors have been known for more than two decades,9 and LOB between the grey matter and white matter and cortical sulcal effacement are the signs most frequently documented in the literature.9–12 It is unlikely that such ischaemic CT signs develop if the CA-ROSC interval is short. It is not yet clear, however, how short an interval is safe enough, because few studies investigating the relationship between the interval and incidence of the CT signs have been conducted in OHCA survivors. Moreover, the timing of brain CT scan was not uniform in previous studies, which ranged from several hours to even days after CA.9–12 Furthermore, CT scans for OHCA survivors have not been performed prospectively. Therefore, it remains unclear how early these ischaemic signs become recognisable on CT scans. It also remains unclear if
J. Inamasu et al. / Resuscitation 81 (2010) 534–538
537
Fig. 2. A histogram showing distribution of the 75 OHCA survivors in relation to the CA-ROSC interval. Those who were positive only for a LOB sign are indicated in dark, whereas those who were positive both for LOB and sulcal effacement signs are indicated in black.
a clear-cut time window exists for CT signs in OHCA survivors. Is it really appropriate to call them ‘early’ CT signs, when CT scans might not have been taken early enough? In this study, there was a significant difference in the incidence of CT signs as well as in the unfavourable outcome rate between those with the CA-ROSC interval ≤20 min and those with the interval >20 min (Tables 1 and 3). As illustrated in Fig. 2, an interval of 20 min seemed to be the time window for development of the LOB sign in OHCA survivors, and those with an interval of >30 min developed a positive LOB sign almost invariably. These results are compatible with previous studies, which reported that the CAROSC interval >20–25 min was highly predictive of unfavourable outcome in cardiac OHCA survivors.5,16 It is also likely that development of the LOB sign in OHCA survivors is time-dependent and predictable. As with previous studies, the presence of the LOB sign was evaluated at the level of the basal ganglia, which is one of the brain areas most vulnerable to ischaemic insults.10–12 The sulcal effacement sign may develop in a more delayed fashion and may not be as time dependent and predictable as the LOB sign. Following these results, we believe that it is appropriate to call these
CT signs ‘early’ signs, because they are often recognisable on a CT scan obtained as early as 1 h after CA. Both CT signs may be useful for identifying OHCA survivors who have unfavourable outcome. LOB may be a more sensitive sign of irreversible ischaemia, as it develops earlier in the course and may have a time window. There were two survivors, who showed a positive LOB sign, and yet were discharged alive, and LOB per se may not be a sign of fatal brain injury. By contrast, sulcal effacement seems to be a more specific sign of fatal brain injury, because none of those with positive sulcal effacement sign were discharged alive. The demographic variables, divided by the CA-ROSC interval of 20 min (Table 1), were not significantly different between the two groups, suggesting that the difference in the incidence of the CT signs might not be due to the demographic differences. On the other hand, comparison between those with favourable and unfavourable outcomes (Table 2) revealed that the former group was significantly younger, presented more frequently with VF rhythm, and had a significantly shorter CA-ROSC interval. The latter results are well expected and are compatible with those from literature.6,16 The high inter-observer concordance for both signs indicates good reproducibility of the CT signs for experienced eyes, although the kappa coefficient was not as high as had been expected. The distinct difference in the GM/WM ratio between those with a positive and negative LOB signs (Fig. 3) may also support the accuracy and validity of CT interpretations. The GM/WM ratio may be a reliable prognosticator for OHCA survivors.10,12 Table 4 Sensitivity/specificity of the early CT signs in predicting unfavourable outcome. Early CT signs
Favourable
Unfavourable
LOB sign (+) LOB sign (−)
1 11
51 12
Sulcal effacement sign (+) Sulcal effacement sign (−) Fig. 3. Comparison of the GM/WM ratio between positive (left) and negative (right) LOB, showing that the difference was statistically significant (p < 0.05).
LOB, loss of boundary.
0
20
12
43
Sensitivity
Specificity
81%
92%
32%
100%
538
J. Inamasu et al. / Resuscitation 81 (2010) 534–538
Although the ratio was significantly higher for those who survived than those who died,10,12 this study is the first to show that the GM/WM ratio was significantly higher for those with a negative LOB sign. We presumed that most survivors for whom CT scan was obtained had sustained a CA with cardiac aetiology. We made every effort to exclude those with asphyxia, in whom the degree of hypoxia is more severe and CT signs may develop earlier than those with cardiac aetiology. The temporal profile of the CT signs in cardiac OHCA survivors is evidently different from that of acute stroke patients in whom the ischaemic interval and development of CT signs is not necessarily correlated.17,18 The lack of a temporal correlation in acute ischaemic stroke patients may be explained by significant individual variability in the collateral brain circulation.17,18 Compared with acute stroke patients, OHCA survivors are more uniform in the degree of ischaemia they sustain because there is a total caessation of blood flow to the brain invariably after CA. The CA-ROSC interval is the most important factor in determining the severity of ischaemic brain damage in OHCA survivors. This uniformity in the degree of ischaemia after CA may explain why the relationship between the ischaemic interval and incidence of early CT signs in OHCA survivors is more time dependent and predictable than that in acute stroke patients. It remains unclear what is the best imaging modality to detect and evaluate CA-induced ischaemic brain damage in humans. By analogy from acute stroke patients, magnetic resonance imaging (MRI), particularly a diffusion-weighted imaging sequence, may detect areas of brain damage earlier and more clearly than a CT scan in OHCA survivors.19,20 However, because of the feasibility and safety concerns, the role of MRI in the acute phase of resuscitation may be limited in many OHCA survivors, who require a ventilator. Although selected survivors may benefit from undergoing MRI in the sub-acute to chronic phases for precise assessment of brain damage, its routine use immediately after resuscitation may neither be practical nor recommended, and CT scan may be the imaging modality of choice. There are several limitations in this study. First, accuracy of the CA-ROSC interval may be hard to verify because the information on the time when the patient collapsed relied solely on the memory of witnesses. Information on the time when ROSC was achieved was given either by paramedics or by emergency medicine residents depending on where ROSC was achieved. Both of these timings may be subject to reporting or measurement errors, and therefore, the interval might not have been as accurate as in an experimental setting. Second, we assumed that most OHCA survivors had a cardiac aetiology. Although those with asphyxia were excluded from the study, it is probable that those with non-cardiac causes of CA, such as a ruptured aortic aneurysm or a pulmonary embolism, might have been included. Brain ischaemia in those with a ruptured aortic aneurysm or a pulmonary embolism develops differently compared with those with a cardiac aetiology.21,22 It remains unknown how many of the 75 survivors actually sustained a non-cardiac CA, because routine autopsies were not performed. In addition, the neurological status of each OHCA survivor, believed to reliably predict outcome,23 was not taken into account in this study. When neurological status and imaging studies are combined, better prediction of outcome is possible and their combined efficacy should be addressed in future studies. Finally, the number of 75 survivors evaluated in this study may not be large enough for statistical power, despite the fact that previous studies comprised less than 30 cases.10–12 CT scans for OHCA survivors immediately after resuscitation may be obtained safely without much ethical and economical concern,13 and we expect that the reproducibility of the temporal profile and the prognostic significance of early CT signs reported here will be verified by multicentre studies.
4. Conclusions This retrospective study indicates that a time window may exist for early CT signs in CA survivors. The LOB sign may develop when the CA-ROSC interval exceeds 20 min, whereas the sulcal effacement sign may develop later. The former seems to be predictive of unfavourable outcome with high sensitivity, while the latter seems to be a prognosticator with high specificity. Conflict of interest statement None. There are no financial and personal relationships with other people or organisations that could inappropriately influence our work. References 1. Hypothermia after Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med 2002;346:549–56. 2. Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med 2002;346:557–63. 3. Böttiger BW, Arntz HR, Chamberlain DA, et al. Thrombolysis during resuscitation for out-of-hospital cardiac arrest. N Engl J Med 2008;359:2651–62. 4. Mysiak A, Nowicki P, Kobusiak-Prokopowicz M. Thrombolysis during cardiopulmonary resuscitation. Cardiol J 2007;14:24–8. 5. Belliard G, Catez E, Charron C, et al. Efficacy of therapeutic hypothermia after out-of-hospital cardiac arrest due to ventricular fibrillation. Resuscitation 2007;75:252–9. 6. Oksanen T, Tiainen M, Skrifvars MB, et al. Predictive power of serum NSE and OHCA score regarding 6-month neurologic outcome after out-ofhospital ventricular fibrillation and therapeutic hypothermia. Resuscitation 2009;80:165–70. 7. Koenig MA, Kaplan PW, Thakor NV. Clinical neurophysiologic monitoring and brain injury from cardiac arrest. Neurol Clin 2006;24:89–106. 8. Kandiah P, Ortega S, Torbey MT. Biomarkers and neuroimaging of brain injury after cardiac arrest. Semin Neurol 2006;26:413–21. 9. Kjos BO, Brant-Zawadzki M, Young RG. Early CT findings of global central nervous system hypoperfusion. Am J Roentgenol 1983;141:1227–32. 10. Torbey MT, Selim M, Knorr J, Bigelow C, Recht L. Quantitative analysis of the loss of distinction between gray and white matter in comatose patients after cardiac arrest. Stroke 2000;31:2163–7. 11. Yanagawa Y, Un-no Y, Sakamoto T, Okada Y. Cerebral density on CT immediately after a successful resuscitation of cardiopulmonary arrest correlates with outcome. Resuscitation 2005;64:97–101. 12. Choi SP, Park HK, Park KN, et al. The density ratio of grey to white matter on computed tomography as an early predictor of vegetative state or death after cardiac arrest. Emerg Med J 2008;25:666–9. 13. Inamasu J, Miyatake S, Tomioka H, et al. Subarachnoid haemorrhage as a cause of out-of-hospital cardiac arrest: a prospective computed tomography study. Resuscitation 2009;80:977–80. 14. International Liaison Committee on Resuscitation. 2005 International consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations. Part 4: Advanced life support. Resuscitation 2005;67:213–47. 15. Langhelle A, Nolan J, Herlitz J, et al. Recommended guidelines for reviewing, reporting, and conducting research on post-resuscitation care: the Utstein style. Resuscitation 2005;66:271–83. 16. Oddo M, Ribordy V, Feihl F, et al. Early predictors of outcome in comatose survivors of ventricular fibrillation and non-ventricular fibrillation cardiac arrest treated with hypothermia: a prospective study. Crit Care Med 2008;36:2296–301. 17. Moulin T, Cattin F, Crépin-Leblond T, et al. Early CT signs in acute middle cerebral artery infarction: predictive value for subsequent infarct locations and outcome. Neurology 1996;47:366–75. 18. Kucinski T, Koch C, Grzyska U, Freitag HJ, Krömer H, Zeumer H. The predictive value of early CT and angiography for fatal hemispheric swelling in acute stroke. Am J Neuroradiol 1998;19:839–46. 19. Arbelaez A, Castillo M, Mukherji SK. Diffusion-weighted MR imaging of global cerebral anoxia. Am J Neuroradiol 1999;20:999–1007. 20. Barrett KM, Freeman WD, Weindling SM, et al. Brain injury after cardiopulmonary arrest and its assessment with diffusion-weighted magnetic resonance imaging. Mayo Clin Proc 2007;82:828–35. 21. Virkkunen I, Paasio L, Ryynänen S, et al. Pulseless electrical activity and unsuccessful out-of-hospital resuscitation: what is the cause of death? Resuscitation 2008;77:207–10. 22. Pierce LC, Courtney DM. Clinical characteristics of aortic aneurysm and dissection as a cause of sudden death in outpatients. Am J Emerg Med 2008;26:1042–6. 23. Booth CM, Boone RH, Tomlinson G, Detsky AS. Is this patient dead, vegetative, or severely neurologically impaired? Assessing outcome for comatose survivors of cardiac arrest. JAMA 2004;291:870–9.