Secondary Logo

Journal Logo

Original Article

Postanoxic coma: how (long) should we treat?

Zandbergen, E. G. J.a

Author Information
European Journal of Anaesthesiology: February 2008 - Volume 25 - Issue - p 39-42
doi: 10.1017/S0265021507003468
  • Free



Postanoxic coma is a condition of decreased consciousness due to a global lack of oxygen or blood supply to the brain. Most commonly this concerns patients who have been resuscitated for cardiac arrest and in whom the circulation has been successfully restored, but in whom the circulation has been impaired long enough to damage the brain to such an extent that the patient remains unconscious. Less frequent causes are severe respiratory insufficiency, near-drowning, and severe and long-lasting hypotension or shock.

Outcome after postanoxic coma lasting more than several hours is generally, but not invariably, poor. This means that large numbers of patients with postanoxic coma may be treated despite having suffered severe and often irreversible brain damage. On the other hand, even when only one of four patients would recover, withdrawal of medical support in unselected patients would not be justified without further prognostic information.

In everyday clinical practice, neurologists are often confronted with questions regarding the prognosis of individual patients, and consequently, about the usefulness of continuing treatment in such a patient. Due to limited availability in the ICU, treating physicians would understandably like to have these questions answered as early as possible. Early information about the expected outcome would of course also be desirable for the patient's relatives.

These questions about prognosis in the early stages of the condition are often difficult to answer. In order to be better able to formulate reliable answers to these questions, many studies have been conducted over the years in an effort to reliably predict which patients still have a chance of recovery and which patients do not.

Brief historical perspective

Although resuscitation from respiratory arrest has been reported as early as in biblical times [1], cardiac arrest has been invariably lethal until halfway through the twentieth century. Current techniques of closed chest cardiac massage and defibrillation were introduced in the 1950s and became increasingly prevalent after the landmark publication of Kouwenhoven and colleagues in 1960 [2]. The fact that most cardiac arrests occur outside the hospital subsequently led to the introduction of prehospital emergency medical systems, which were first developed in Belfast. In 1967, Irish investigators reported what was probably the first successful attempt to restore spontaneous circulation by bystander resuscitation [3]. Unfortunately, the patient remained unconscious and died one week later, illustrating the new medical problem of postanoxic brain damage created as a side-effect of this new technique.

Before the introduction of cardiopulmonary resuscitation (CPR), the phenomenon of postanoxic brain damage was not entirely unknown, but it was rare. But from that time on, increasing numbers of patients comatose after ‘successful' resuscitation started to populate the ICUs. Most of these patients never regained consciousness and died within several days or weeks, or became entirely care-dependent in a state of ‘wakefulness without awareness', the so-called vegetative state. However, others did regain consciousness, often even recovering without residual disability [4,5]. This large outcome variation made decisions regarding continuation of supportive treatment in such patients an ever-recurring source of debate.

Studies investigating factors predicting the neurological outcome of postanoxic coma were undertaken as early as in 1965 [6], when Hockaday and colleagues [6] published data on electroencephalographic patterns after CPR and their relation to prognosis. Since then, many studies have followed. One of the most noteworthy was ‘Predicting outcome from hypoxic-ischaemic coma' by Levy and colleagues, published in 1985 [4]. In this study, features of the neurological examination at different time points in 210 patients with postanoxic coma were related to outcome. The results were presented in prognostic algorithms and have since been used as a basis for treatment decisions by neurologists worldwide.

Next to the electroencephalogram (EEG) and features of the clinical neurological examination, many other tests were studied for their prognostic value: evoked potentials (especially somatosensory evoked potentials, SSEP), biochemical variables in serum and cerebrospinal fluid, and several imaging modalities (computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography).

General prognosis

As mentioned before, prognosis in postanoxic coma is generally poor. In this group, 1-yr-survival is around 10-25%, cerebral damage being the principal cause of mortality in 30-40% of those who die [7,8]. Only a small percentage of the survivors (<1%) remains in a prolonged comatose or vegetative state, a more significant number has a more or less severe residual disability. In a large multicentre prospective cohort study, consciousness was recovered in 13% of patients who had been comatose for at least 24 h [9]. After 1 yr, only 10% were still alive (and conscious), about one-third of whom were severely disabled, and another third were moderately disabled. One should realize, however, that severe disability after recovery of consciousness need not be an end stage. Of the people who are severely disabled after 1 month, about 25% continue to recover to a state of independency [9].

The residual disability in this group of patients is usually a combined motor and cognitive disability, with both memory and executive impairments [10].

Prognostic variables

Over the years, many studies have aimed to identify prognostic variables in postanoxic coma. Many variables have been studied but most studies focussed on only one or two groups of variables, e.g. only neurological examination, or only clinical neurophysiological variables. More recently, several systematic reviews have been carried out, combining the results from different studies for one or more variables [11-15]. In the first of these reviews [11] both clinical neurological and neurophysiological variables were studied. Bilateral absence of the early cortical response (N20) of the median nerve SSEP in the first week proved to be the most reliable predictor of poor outcome, defined as death or prolonged coma/vegetative state (0% false-positive rate (FPR), CI 0-2%). Absent pupillary reactions to light on day 3 and absent motor response to pain on day 3 were also invariably associated with poor outcome, but due to small numbers of patients, CIs of the FPR were much wider for these variables.

The findings for the SSEP were confirmed by the later reviews of Carter and colleagues [12] and Robinson and colleagues [13]. Booth and colleagues [14] focussed on the predictive value of the neurological examination with poor outcome defined as death, coma or severe disability. They found absent pupillary reactions, absent motor response to pain and absent corneal reflexes to be the most reliable prognostic variables, but do not mention whether false prediction of poor outcome can occur when these variables are used.

A separate systematic review of potential biochemical predictors of poor outcome showed that the value of these variables was still unclear [15].

In 2006, a multicentre prospective cohort study on prognosis in postanoxic coma (PROPAC) was published. In this study, components of the neurological examination, neurophysiological tests and biochemical variables were registered prospectively in 407 patients with postanoxic coma of at least 24 h duration. Outcome was defined as death or persistent coma after 1 month and follow-up was continued until 1 yr after resuscitation. Based on the previous literature, the predictive value of the SSEP for poor outcome was considered reliable enough to justify treatment limitations based on its absence. In the study, absent SSEP after 24 h and after 48 h proved to be as reliable as the 72-h SSEP. In addition, other potential prognostic variables were studied. A serum neuron-specific enolase (NSE) level >33 μg L−1 emerged as the most reliable predictor of poor outcome in this study apart from the SSEP. However, other studies have shown that its predictive value is not quite as good as it seemed to be, as recovery of consciousness has been described in spite of higher serum NSE levels. Other factors that predicted poor outcome with a 0% FPR in the PROPAC study were absence of both pupillary reactions and corneal reflexes after 72 h and a myoclonic status, especially within the first 24 h (at longer time intervals the numbers of patients are so low that CIs become unacceptable). Low-voltage (<20 μV) EEG or burst-suppression pattern on EEG and a serum S100 >0.7 μg L−1 after 72 h are other good predictors of poor outcome (Table 1).

Table 1
Table 1:
. Prognostic variables PROPAC study.

In July 2006, the American Practice Parameter on Postanoxic coma was published [16]. In this work, all potential prognostic variables that have been studied are described with a summary of their value and recommendations for use. For the evaluation, poor outcome was defined as death, coma or severe disability after 6 months. In summary, the conclusions are as follows:

  • Variables that predict poor outcome reliably:
    • ○ myoclonic status epilepticus within the first 24 h;
    • ○ absent pupillary reactions to light days 1-3;
    • ○ absent corneal reflexes days 1-3;
    • ○ absent or extensor motor response to pain day 3;
    • ○ bilateral absence of N20 on the median nerve SSEP days 1-3.
  • Variables that are related to poor outcome, but do not have sufficient predictive value:
    • ○ characteristics of resuscitation, such as initial heart rhythm, cause and duration of anoxia and duration of resuscitation;
    • ○ hyperthermia;
    • ○ isolated myoclonus or seizures;
    • ○ EEG characteristics (generalized suppression <20 μV; burst-suppression pattern with generalized epileptiform activity, or generalized periodic complexes on a flat background are strongly, but not invariably, associated with poor outcome);
    • ○ neurological examination earlier then the time interval mentioned above.
  • Variables for which the prognostic value is still unclear:
    • ○ biochemical variables such as serum NSE;
    • ○ neuro-imaging (CT-brain or MRI-brain).

Effect of treatment?

After restoration of circulation, prognosis depends mostly on the patient's neurological condition. In an effort to improve the neurological prognosis, several studies have been done in search of treatment options. Examples are barbiturates, calcium-antagonists, N-methyl-D-aspartate (NMDA) blockers and corticosteroids, none of which had any influence on the prognosis. This changed in February 2002, when two studies on hypothermia in postanoxic coma were published in the New England Journal of Medicine. Both studies showed that mild hypothermia can positively influence outcome in postanoxic coma, at least in those patients with ventricular fibrillation as the initial rhythm [17,18].

During hypothermia, patients are treated with sedatives and muscle relaxants, and therefore cannot be examined neurologically. The SSEP may be an exception to this rule, because the presence or absence of N20 is not influenced by this grade of hypothermia. Moreover, the results from a study by Tiainen and colleagues [19] indicate that the predictive value of the SSEP for poor outcome is likewise not influenced by the hypothermia.

In spite of these considerations, evaluation of patients with postanoxic coma requires at least the performance of a reliable neurological examination and it seems unwise to take irreversible steps like the decision to limit or stop treatment solely on the basis of an ancillary examination such as the SSEP when a patient cannot be examined clinically. As a result, evaluation of a patient's prognosis will usually be postponed to about 36 h after resuscitation.

Funding for research by the author presented within this manuscript: Dutch Heart Foundation and Brain Foundation The Netherlands.


1. The Bible. 1 Kings 17: 17-22.
2. Kouwenhoven WB, Jude JR, Knickerbocker GG. Closed chest cardiac massage. JAMA 1960; 173: 94-97.
3. Pantridge JF, Geddes JS. A mobile intensive-care unit in the management of myocardial infarction. Lancet 1967; 2: 271-273.
4. Levy DE, Caronna JJ, Singer BH, Lapinski RH, Frydman H, Plum F. Predicting outcome from hypoxic-ischemic coma. JAMA 1985; 253: 1420-1426.
5. Edgren E, Hedstrand U, Kelsey S, Sutton-Tyrrell K, Safar P. Assessment of neurological prognosis in comatose survivors of cardiac arrest. Lancet 1994; 343: 1055-1059.
6. Hockaday JM, Potts F, Epstein E, Bonazzi A, Schwab RS. Electroencephalographic changes in acute cerebral anoxia from cardiac or respiratory arrest. Electroenceph Clin Neurophysiol 1965; 18: 575-586.
7. Thomassen A, Wernberg M. Prevalence and prognostic significance of coma after cardiac arrest outside intensive care and coronary care units. Acta Anaesth Scand 1979; 23: 143-148.
8. Longstreth WT Jr, Diehr P, Inui TS. Prediction of awakening after out-of-hospital cardiac arrest. N Eng J Med 1983; 308: 1378-1382.
9. Zandbergen EGJ, Hijdra A, Koelman JHTM et al.. Prediction of poor outcome within the first three days of postanoxic coma. Neurology 2006; 66: 62-68.
10. Lim C, Alexander MP, LaFleche G, Schnyer DM, Verfaellie M. The neurological and cognitive sequelae of cardiac arrest. Neurology 2004; 63: 1774-1778.
11. Zandbergen EGJ, De Haan RJ, Stoutenbeek CP, Koelman JHTM, Hijdra A. Systematic review of early prediction of poor outcome in anoxic-ischaemic coma. Lancet 1998; 352: 1808-1812.
12. Carter BG, Butt W. Review of the use of somatosensory evoked potentials in the prediction of outcome after severe brain injury. Crit Care Med 2001; 29: 178-186.
13. Robinson LR, Micklesen PJ, Tirschwell DL, Lew HL. Predictive value of somatosensory evoked potentials for awakening from coma. Crit Care Med 2003; 31: 960-967.
14. Booth CM, Boone RH, Tomlinson G, Detsky AS. Is this patient dead, vegetative, or severely neurologically impaired? JAMA 2004; 291: 870-879.
15. Zandbergen EGJ, De Haan RJ, Hijdra A. Systematic review of prediction of poor outcome in anoxic-ischaemic coma with biochemical markers of brain damage. Intensive Care Med 2001; 27: 1661-1667.
16. Wijdicks EFM, Hijdra A, Young GB, Bassetti CL, Wiebe S. Practice parameter: Prediction of outcome in comatose survivors after cardiopulmonary resuscitation (an evidence-based review). Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 2006; 67: 203-210.
17. 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: 557-563.
18. The Hypothermia after Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med 2002; 346: 549-556.
19. Tiainen M, Kovala TT, Takkunen OS, Roine RO. Somatosensory and brainstem evoked potentials in cardiac arrest patients treated with hypothermia. Crit Care Med 2005; 33: 1736-1740.


© 2008 European Society of Anaesthesiology