A Phase 3 Trial of Hypothermia Following TBI Is Stopped Early, Finding It Worsens Outcomes
ARTICLE IN BRIEF
Therapeutic hypothermia to control intracranial hypertension following traumatic brain injury significantly worsened, rather than improved at six months outcomes compared with standard therapy, according to findings from a phase 3 study. But independent neurocritical care experts said the study findings are not reason enough to abandon the therapy entirely.
Therapeutic hypothermia to control intracranial hypertension following traumatic brain injury significantly worsened, rather than improved, six-month outcomes compared with standard therapy, a phase 3 trial has found.
The study of 387 patients at 47 centers in 18 European countries was suspended early after researchers found that a favorable outcome based on the Extended Glasgow Outcome Scale (GOS-E) occurred in 26 percent of patients in the hypothermia group and in 37 percent of the control group (p=0.03).
An accompanying editorial in the December 17 issue of the New England Journal of Medicine said the findings “may lead to the demise of hypothermia as yet another pillar of therapy for intracranial hypertension.”
In the editorial, Claudia S. Robertson, MD, a professor of neurosurgery at Baylor College of Medicine, and Allan H. Ropper, MD, FAAN, executive vice chair and professor of neurology at Harvard Medical School, wrote: “The effects on mortality and functional outcome of all the treatments we currently use — hypothermia, craniectomy, drainage of cerebrospinal fluid, hyperventilation, and even hyperosmolar therapy — are probably smaller than we think, because severe head injury is an overwhelming and complex process involving neurons, glia, and vasculature. It may be that raised intracranial pressure compresses cerebral veins and creates a self-generating cycle of even higher pressure so that medical management becomes futile. This is not an excuse to allow elevated intracranial pressure to go unchecked, but it is an admonition to be discerning in the use of medical and surgical treatment, especially hypothermia.”
But three prominent neurocritical care specialists in the United States emphasized in interviews with Neurology Today that, for now, hypothermia should not be thrown out with the frigid bathwater. The editorial similarly noted that the findings, although disappointing, need to be interpreted with caution.
The editorialists wrote: “In the hypothermia group of the study, hyperosmolar infusions were implemented only if cooling failed to control intracranial pressure, but we do not have details pertaining to the manner and quantity of mannitol and hypertonic saline that were used. Furthermore, hypothermia resulted in a greatly decreased need for pentobarbital-induced coma. This may indicate that barbiturates provided the same or better metabolic suppression and neuroprotection as compared with hypothermia. For these reasons, it would be difficult to appreciate an effect of hypothermia alone on outcome.”
The Eurotherm3235 Trial Collaborators, led by Peter J.D. Andrews, MD, of the Centre for Clinical Brain Sciences at the University of Edinburgh in Scotland, tested the effects of lowering body temperature to 32° to 35°C, plus standard care, when intracranial pressure (ICP) reached 20 mm Hg.
In the control group, stage 2 treatments such as osmotherapy were added as needed to control ICP, whereas in the hypothermia group stage 2 treatments were added only if ICP remained elevated. In both groups, stage 3 treatments involving barbiturates and decompressive craniectomy were used if all stage 2 treatments failed to control ICP.
The first patient in the study was enrolled in November of 2009, and the trial was stopped early in October of 2014, when it became clear that the hypothermia group had significantly worse outcomes than the controls. Stage 3 treatments to control ICP, for instance, were necessary in 54 percent of the controls, compared with 44 percent of the hypothermia group.
Likewise, six months after injury, the distribution of GOS-E scores was shifted in an unfavorable direction in the hypothermia group (adjusted common odds ratio, 1.53; 95% CI, 1.02 to 2.30; p=0.04). Favorable outcomes, representing a GOS-E score of 5 to 8, occurred in 49 of 191 patients (25.7 percent) in the hypothermia group and in 69 of 189 patients (36.5 percent) in the control group (p=0.03). The risk of death was also increased in the hypothermia group (hazard ratio, 1.45; 95% CI, 1.01 to 2.10; p= 0.047).
Commenting on the study, Romergryko G. Geocadin, MD, FAAN, a professor of neurology, anesthesiology, critical care medicine and neurosurgery at the Johns Hopkins University School of Medicine, emphasized that the study reversed the standard treatment sequence in the United States, where hyperosmolar therapy ( i.e. hypertonic saline or mannitol) is used as a tier 2 treatment, followed by hypothermia as a tier 3 treatment only if ICP remains elevated. Not only did the Eurotherm3235 trial use hypothermia prior to hyperosmolar therapy, it also withheld barbituates unless the cooling treatment failed to control ICP.
“This study is not just about the effect of hypothermia on TBI,” said Dr. Geocadin, who is the immediate past president of the Neurocritical Care Society. “The study affected several aspects in the prevailing management of TBI, especially the earlier use of hyperosmolar therapy and even the use of pentobarbital. So we cannot confidently conclude whether it was the hypothermia that resulted in worse outcomes, or the deviation from common practice. We just don't know.”
Dr. Geocadin said he was concerned that clinicians might stop using hypothermia as a result of the trial, contending that it has value in highly selected patients.
“We ought to be very careful before we bury the idea of hypothermia,” he said. “If you're fighting for your patients and have exhausted other treatments, don't be afraid to try therapeutic hypothermia. It's an option after you've exhausted tier 1 and tier 2 treatments. That has not changed. If nothing else has worked, I think hypothermia can still improve outcomes.”
Another leader in neurocritical care suggested that rather than reserving hypothermia only as a late option, it should perhaps be considered earlier, before ICP rises to a dangerously high level.
“In the setting of the Eurotherm3235 trial, hypothermia initiated to control ICP didn't improve outcomes,” said William M. Coplin, MD, an associate professor of neurology and neurological surgery at Wayne State University, and director of neurotrauma and the neurological intensive care unit and chief of neurology at Detroit Receiving Hospital. “Is that because once ICT develops, the horse is out of the barn and there's nothing we can do to fix the rest of the problem? I am a proponent of not waiting for people to get into trouble.”
Edward M. Manno, MD, FAAN, head of neurocritical care at the Cleveland Clinic and the current president of the Neurocritical Care Society, agreed that waiting for intracranial hypertension to develop before initiating hypothermia may be a mistake. “I think the key is that we need to move beyond just thinking about intracranial pressure as the modality we need to treat,” he said.
A more promising factor to focus on, he said, may be maintenance of adequate oxygenation. He pointed to the BOOST 2 trial results, reported at the annual meeting of the Neurocritical Care Society in October of 2014, which found that monitoring brain tissue oxygenation reduced the duration of measured brain tissue hypoxia.
“The BOOST 2 results were positive, and BOOST 3 is underway,” Dr. Manno said. “That's a potential bright spot.”
EXPERTS: ON HYPOTHERMIA FOLLOWING TRAUMATIC BRAIN INJURY