Patients with a variety of intracranial disorders — including traumatic brain injury, stroke, subarachnoid hemorrhage, intracerebral hemorrhage, and brain tumors — often present with a progressive increase in intracranial pressure, leading to clinical deterioration and ultimately to death. Medical therapy1 can help to mitigate such increases in pressure, but despite the use of the best available measures, intracranial hypertension becomes life-threatening in some patients. More than a century ago, it was suggested that it might be beneficial to “decompress the brain by widely opening the skull to decrease the pressure”2 in patients with severe traumatic brain injury. This procedure, called decompressive craniectomy, has been shown to decrease intracranial pressure, but since there is no evidence of an association with a better clinical outcome, the procedure is considered optional.3

In the past 15 years, the use of decompressive craniectomy has increased substantially. There has also been a tremendous increase in the number of articles that have been published on the subject, mainly retrospective reviews of a limited number of cases. A PubMed search identified 143 articles that were published on this topic for a variety of intracranial disorders in 2009 and 2010.

The list of publications is much shorter, however, when only randomized, controlled trials are considered. In the field of traumatic brain injury, the results of only one small, prospective, randomized trial have been published.4 This trial involving 27 children showed promising results in favor of decompressive craniectomy. However, the surgical procedure that was used (bitemporal decompression without opening of the dura) is not the standard approach.

In this issue of the Journal, Cooper et al. report the results of the Decompressive Craniectomy (DECRA) trial,5 which investigated the role of early decompressive craniectomy in adults with severe head injury (Glasgow Coma Scale score of 3 to 8). Patients with refractory intracranial hypertension (defined as an intracranial pressure higher than 20 mm Hg for more than 15 minutes despite medical therapy) were randomly assigned either to receive standard care or to undergo standard care plus bifrontotemporoparietal decompression craniectomy. The study showed a significant decrease in intracranial pressure in patients in the surgical group, as was expected. However, clinical outcomes, as assessed by scores on the Extended Glasgow Outcome Scale, were worse in the surgical group than in the standard-care group, a finding that went against expectations.

There are a couple of important concerns regarding this otherwise valuable study. First, most neurosurgeons and intensivists dealing with traumatic brain injury will not consider decompressive craniectomy in patients who have an intracranial pressure of around 20 mm Hg for such a short time. This aggressive approach may be justified in order to decompress the brain as soon as possible, but in patients with diffuse injury without mass lesions, physicians in many centers would use medical therapy for a longer period, leaving decompressive craniectomy as a last resort.

Second, the screening of 3478 patients over a 7-year period to enroll only 155 study patients indicates that the results of this study are limited to a restricted subpopulation of patients with traumatic brain injury. Most of the patients who were excluded from this trial either had mass lesions (for which a different surgical approach might be appropriate) or had successful control of intracranial hypertension with medical management (thus not requiring surgical intervention at all).

An important question arising from the DECRA study is whether it is now appropriate to continue an ongoing trial of craniectomy, called the Randomized Evaluation of Surgery with Craniectomy for Uncontrollable Elevation of Intracranial Pressure (RESCUEicp; Current Controlled Trials number, ISRCTN66202560).6 As of March 16, 2011, a total of 294 patients had been enrolled in this trial, with an enrollment goal of 400 patients. However, the design of RESCUEicp differs from that of the DECRA study in some important ways. In RESCUEicp, patients are randomly assigned either to undergo craniectomy or to receive standard care (including the use of barbiturates) when maximal medical therapy cannot control intracranial pressure, with a threshold of 25 mm Hg (rather than 20 mm Hg) for more than 1 to 12 hours (rather than 15 minutes) at any time after injury. In addition, in RESCUEicp, as compared with the DECRA study, previous evacuation of a hematoma is allowed before randomization, and the permitted surgical techniques include both bifrontal decompression and unilateral wide decompression. The primary end point is the assessment of outcome at discharge and at 6 months. Thus, because of such differences in trial design, it seems that RESCUEicp should continue.

Another important question is, How do the DECRA and RESCUEicp studies relate to the real practice of decompression in patients with traumatic brain injury? As noted above, the exclusion criteria for the DECRA study were such that the data do not apply to the majority of patients. In a multicenter study involving 729 patients with intradural mass lesions after traumatic injury,7 we found that about one third of patients undergoing surgery for an intracranial hematoma also required a decompressive procedure. In most of these cases, the decompression was unilateral and associated with hematoma evacuation. Such patients are not considered at all in the DECRA study and are only partially included in the RESCUEicp study. We therefore probably need another study of early decompression associated with hematoma evacuation, as has been suggested previously,8 since this procedure is a common one.

The main lesson from the DECRA study is that surgical reduction of intracranial pressure by the technique that was used by the investigators does not necessarily result in better outcomes for patients and indeed appears to worsen them in at least some circumstances. However, it is important that the procedure not be simply abandoned on the basis of these data. Rather, we must think more carefully about the risks and benefits of the decompressive craniectomy before performing the procedure and must work to define appropriate clinical settings for this procedure.

source: NEJM