Hypothermia for Prevention of Cerebral Injury After
Cardiac Arrest
Cardiac arrest is estimated to occur in 375,000 patients annually,
[159]
with survival rates ranging from 5% to 35%.
[160]
The wide variation in survival depends on
the underlying rhythm that the patient presents with; survival is improved if the
initial rhythm is ventricular fibrillation or ventricular tachycardia rather than
asystole.[161]
[162]
During cardiac arrest, lack of cerebral blood flow leads to generation of free radicals
and other mediators that result in cerebral injury during reperfusion.[163]
In patients who are resuscitated and admitted to the ICU, neurologic injury from
anoxia is a major cause of morbidity and mortality that occurs despite the level
of care provided.[164]
Because cerebral ischemia
may exist for hours after resuscitation,[165]
the
use of hypothermia to decrease cerebral oxygen demand has been proposed as a treatment
option.[166]
[167]
Other possible beneficial effects of hypothermia include retardation of enzymatic
reactions, suppression of free-radical formation, reduction of intracellular acidosis,
and inhibition of excitatory neurotransmitters.[168]
[169]
[170]
Two studies were undertaken to examine the efficacy of hypothermia
after witnessed cardiac arrest. The Hypothermia after Cardiac Arrest Study Group
studied 275 patients resuscitated after a witnessed ventricular fibrillation arrest.
Patients were randomized to usual care or hypothermia (32°C to 34°C) achieved
with external cooling. The goals were to achieve the target temperature within 8
hours of randomization and maintain hypothermia for 24 hours, followed by passive
rewarming. There was a favorable neurologic outcome with good functional status
in 55% of the hypothermia group compared with 39% in the control group (P
= .009). The mortality rate was reduced from 55% to 41% (P
= .02) for the hypothermia group. Complications were largely insignificant except
for a trend toward greater occurrence of sepsis in the hypothermia group, but the
difference was not statistically significant.[171]
In the second study, Bernard and coworkers[172]
studied 77 patients resuscitated after ventricular fibrillation who
Figure 74-4a
Protocol for "tight" glycemic control in critically ill
patients in use at the University of California, San Francisco. This protocol requires
significant nursing intervention, because changes in the infusion rate require rapid
measurement of fingerstick glucose concentrations. Changes in the insulin infusion
rate are based on the rate of change of glucose concentration over the preceding
interval.
Figure 74-4b
were randomized to hypothermia or normothermia. Hypothermia was induced with exposure
in the field and supplemented with ice packs in the emergency department. The target
temperature was achieved within 2 hours and then continued for 12 hours. After 18
hours of hypothermia, patients were actively rewarmed over 6 hours using heated-air
blankets. The difference in mortality (51% for hypothermia versus 68% for normothermia)
did not reach statistical significance, but 49% of the hypothermia group versus 26%
of the normothermia group was considered to have a good neurologic outcome (P
= .046). There was no increase in the frequency of adverse events with hypothermia.
[172]
These studies are intriguing in that hypothermia appears beneficial
with very few adverse effects and does not increase the number of patients resuscitated
with poor functional status. Additional studies will be needed to determine the
benefit of this treatment protocol for patients presenting with cardiac arrest due
to causes other than ventricular fibrillation. In contrast to these positive findings,
a study of therapeutic hypothermia in patients after traumatic brain injury showed
no improvement in neurologic outcome.[173]
These
differences may be explained by different pathogenesis of the central nervous system
injury (i.e., global versus focal ischemia) or possibly by the duration of hypothermia
(i.e., 48 hours for the traumatic brain injury group versus less than 24 hours for
the cardiac arrest group). Remaining challenges for future randomized, controlled
trials will be to determine the appropriate level and duration of hypothermia, but
current data are suggestive enough to recommend attempting to achieve hypothermia
in patients presenting with spontaneous circulation after ventricular fibrillation
or ventricular tachycardia from cardiac ischemia. Development of new hypothermia
catheter technology will allow more rapid and precise control of the hypothermic
period.
 |
