Decompressive Surgery Cuts Death in Traumatic Brain Injury

September 08, 2016

Patients with traumatic brain injury and raised intracranial pressure who underwent decompressive craniectomy — in which a large section of the skull is removed to allow the brain to expand — had a far lower mortality rate but were more likely to be left with severe disability than those treated medically, according to a new randomized study.

The results, from the Randomised Evaluation of Surgery with Craniectomy for Uncontrollable Elevation of Intracranial Pressure (RESCUEicp) trial, were published online in The New England Journal of Medicine on September 7.

"This is groundbreaking as it is the first intervention that has shown a major difference in outcome in this population — in particular a large and dramatic survival benefit," lead author, neurosurgeon Peter Hutchinson, FRCS, commented to Medscape Medical News.

Mortality was reduced from 48.9% in the control group to 26.9% in the surgery group.

However, the concern is that patients whose lives have been saved by this procedure are generally left with a severe level of disability, with more patients in a vegetative state or with lower severe disability (dependent on others for care) or upper severe disability (able to live independently but requiring support to go out), Professor Hutchinson added.

The rates of moderate disability and good recovery were similar in the two groups, he said, "so the big question is, 'Is it worth it?' That is the fundamental issue."

"There is no doubt this surgery saves lives — but we have to look very carefully at the quality of survival to give information to families on the pros and cons of performing this surgery," he said. "This is not a black and white decision. This surgery is already taking place in practice, but now we have more information to guide our decisions. The data are now out there and can be discussed, and the neurosurgeons need to interpret it for themselves. I believe some will take this study as a reason to do more of these procedures. Others may be more concerned about the increase in patients left in a vegetative state and decide to do less."

He noted that these patients have sustained traumatic brain injury mainly as a result of road traffic accidents, falls, or assaults. They are generally young — the median age was 33 years (range, 10 to 65 years). They had refractory elevated intracranial pressure (>25 mm Hg) for 1 to 12 hours.

So far, no treatment has shown evidence of benefit in terms of outcome for these patients, he said. "We try to bring intracranial pressure down with drugs such as barbiturates, but this has not been tested in an outcomes study. Decompressive surgery is also performed in some cases, but again there has been no evidence of benefit until now in an outcomes study."

Professor Hutchinson explained that that there are two types of decompressive surgery: primary (hematoma is removed on admission and part of the skull is not replaced after the operation) and secondary (conducted later when there is diffuse brain swelling despite medical treatment).

"This trial tested secondary surgery," he said. "All patients received medical treatment. If this did not work and the brain was still swollen, then the surgery was performed."

A previous trial of decompressive surgery in trauma patients — DECRA — did not show a benefit. "In DECRA, the mortality was the same and there were more unfavorable outcomes in the surgery group," Professor Hutchinson commented. "But they had different inclusion criteria and timing of surgery — in our study the threshold for intracranial pressure was higher and surgery was performed later. It is possible that in DECRA the surgery was performed too early and some patients therefore received it who did not need it but still experienced the complications of surgery."

This is a similar procedure that is sometimes performed in patients who have had a very large stroke. Professor Hutchinson noted that stroke patients normally have just one side of the skull removed, but in this trial individual surgeons decided whether to perform unilateral or bifrontal surgery; 63% chose bifrontal.

"The patients were also much younger than stroke patients, and a subgroup analysis showed a tendency to a better outcome with surgery in those aged under 40," he added.

For the study, 408 patients were randomly assigned to undergo decompressive surgery or ongoing medical care.

The primary endpoint was the rating on the Extended Glasgow Outcome Scale (GOS-E) at 6 months. The GOS-E distribution differed between the two groups (P < .001).

Breakdown of the results shows an almost halving of death rate with surgery but more patients being left in a vegetative state or with lower severe disability.

Table 1. GOS-E Distribution at 6 Months

Endpoint Surgery (%) Medical (%)
Death 26.9 48.9
Vegetative state 8.5 2.1
Lower severe disability (dependent on others for care) 21.9 14.4
Upper severe disability (independent at home but requiring support to go out) 15.4 8.0
Moderate disability 23.4 19.7
Good recovery 4.0 6.9


Table 2. GOS-E Distribution at 12 Months

Endpoint Surgery (%) Medical (%)
Death 30.4 52.0
Vegetative state 6.2 1.7
Lower severe disability 18.0 14.0
Upper severe disability 13.4 3.9
Moderate disability 22.2 20.1
Good recovery 9.8 8.4


Surgical patients had fewer hours with intracranial pressure above 25 mm Hg after randomization than did medical patients (median, 5.0 vs 17.0 hours) but had a higher rate of adverse events (16.3% vs 9.2%).

In a prespecified sensitivity analysis, the authors compared the proportion of patients in each group who had outcomes of upper severe disability or better (GOS-E score of 4 to 8), which were defined as "favorable outcomes."

The dichotomized GOS-E results did not show a significantly higher percentage of patients with such a favorable outcome in the surgical group vs the medical group at 6 months (42.8% and 34.6%, respectively; P = .12), but there was a significant difference at 12 months (45.4% vs 32.4%; P = .01).

Professor Hutchinson commented: "There seemed to be some continued improvement in the surgery group after the 6-month time point. At 12 months for every 100 patients receiving the surgery, 22 more survived, 5 more were in a vegetative state, 4 more had lower severe dependency, and 13 more had upper severe dependency or better."

The authors conclude: "Our trial provides quantitative evidence to inform the debate around historical concerns that decompressive craniectomy simply increases the number of patients who survive in a vegetative state. The survival advantage of craniectomy in this trial was translated to both dependent and independent living."

But in an accompanying editorial, Lori A. Shutter, MD, University of Pittsburgh, Pennsylvania, and Shelly D. Timmons, MD, PhD, Penn State University Milton S. Hershey Medical Center, Hershey, Pennsylvania, note that although the study found decompressive surgery to be lifesaving, this does not ensure a return to normal functioning.

"The findings of this trial argue for more investigation into the nuances of selecting patients for decompressive craniectomy after traumatic brain injury and for the development of more refined clinical decision-making tools," they write.

"Quality of life is an individual determination, and it is important to engage patients' surrogates in discussions that focus on the patients' previously stated wishes and personal values," they add. "We must acknowledge the importance of shared decision making in discussions with surrogates about potential outcomes of therapeutic options, prolonged recovery times, and the expected quality of life after neurologic injury."

The trial was supported by the UK Medical Research Council, the University of Cambridge, and Cambridge University Hospitals NHS Foundation Trust. Dr Hutchinson reports serving as director of Medicam and Technicam, a manufacturer of a cranial access device (no financial remuneration), and coauthoring a report on the intracranial-pressure consensus conference in Milan (the meeting was financially supported by Codman with an unconditional grant).

N Engl J Med. Published online September 7, 2016. Abstract, Editorial

For more Medscape Neurology news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.