SAN FRANCISCO – Therapeutic hypothermia is the only therapy proven to decrease mortality and improve neurologic outcomes in comatose patients after cardiac arrest, but an analysis of data on 26,519 U.S. patients suggests that it is used in only 0.35% of cases.
This rate represents a “gross underutilization” of this therapy, said Dr. Pratik Patel, who reported the findings at the annual meeting of the Neurocritical Care Society.
Therapeutic hypothermia has been recommended for comatose patients after cardiac arrest with return of spontaneous circulation since 2003 by the International Liaison Committee on Resuscitation and since 2005 by the American Heart Association.
Other new studies reported at the meeting indicated that despite the benefits of therapeutic hypothermia for comatose patients, about half of patients who undergo it have a poor outcome that may be predicted by several factors, and nearly a quarter of patients experience seizure activity during the postresuscitation period. A separate small study suggested that intracranial multimodal monitoring may be a means to detect secondary injuries, such as seizures, that appear to contribute to poor outcomes.
Dr. Patel and his associates at Rush University, Chicago, analyzed a representative sampling of data from U.S. hospitals (the 2007 Nationwide Inpatient Sample) to identify 26,519 adults who were admitted with a diagnosis of cardiac arrest. Only 92 cases (0.35%) included the ICD-9 code for therapeutic hypothermia.
In general, it’s unclear how many patients with cardiac arrest are comatose immediately following the return of spontaneous circulation. Up to 85% show some signs of cerebral dysfunction. Using the strict criteria of the Hypothermia After Cardiac Arrest study, an estimated 8.6% of patients may be eligible for therapeutic hypothermia, said Dr. Patel, a neurointensivist at Rush.
The use of therapeutic hypothermia was independently associated with younger patient age, admission through an emergency department, centers in western U.S. states, and large hospitals or teaching hospitals.
In contrast, most hospitals in the Netherlands consider therapeutic hypothermia (32?-34? C) to be standard care for comatose patients after cardiac arrest, noted Dr. Janneke Horn of the Academic Medical Center in Amsterdam. Guidelines for determining prognosis in patients with postanoxic coma after cardiac arrest all come from eras before the widespread of use of therapeutic hypothermia, she added.
In a prospective study of 391 patients who were admitted to 10 Dutch centers after CPR, Dr. Horn and her associates found that 51% died within 6 months, most of them within a week. Among the 149 patients (38% of the total) who died in the first week, treatment was withdrawn in 62%.
The study’s definition of poor outcome included severe disability (nine patients, or 2%) or a vegetative state (no patients in this study). Moderate disability remained in 12% of patients, and 32% made a good recovery. Outcomes data were missing in 3% of cases.
Factors that predicted a poor outcome were absent brain stem reflexes, a neuron-specific enolase level higher than 33 mcg/L, and absent somatosensory evoked potentials (SSEP) after restoration of normothermia. These tests can be used shortly after the patient has been rewarmed and the effects of sedative drugs have worn off, she said.
“Absent pupillary reaction 72 hours after CPR and absent cortical responses in SSEP after rewarming reliably predict poor outcome at 6 months in patients who remain comatose after CPR and treatment with hypothermia,” Dr. Horn said.
In a separate study, Dr. Jon Rittenberger of the University of Pittsburgh and his associates added continuous EEG (cEEG) recordings to the postcardiac arrest care of 76 comatose patients who were treated with therapeutic hypothermia, and they found evidence of seizures in 22%. All but one patient died.
Most seizures were evident within 30 minutes of starting cEEG. It took an average of 8 hours after the time of arrest to place the cEEG, suggesting that “many patients may be seizing before cEEG placement,” he noted.
Seizure activity tended to increase with higher categories of cardiac arrest. Seizures were detected in 3 (10%) of 31 patients with category II cardiac arrest, in 4 (40%) of 10 patients in category III, and in 10 (29%) of 35 patients with category IV cardiac arrest.
The study suggests that cEEG monitoring is warranted in these patients, although it’s unclear if early prophylaxis might decrease the incidence of electrographic seizures, or if treating seizures would improve outcomes, Dr. Rittenberger said.
Dr. Stephan A. Mayer, head of the division of critical care neurology at Columbia University, New York, and his associates focused on another unknown: whether the poor outcomes in comatose patients after cardiac arrest who get therapeutic hypothermia result from the original injury, or are affected by secondary injuries in the first few hours after cardiac arrest. Transient global hypoxic-ischemic injury after cardiac arrest may leave the brain vulnerable to delayed, secondary injuries.