SAN FRANCISCO—Decompressive surgery reduces the risk of death in patients who have a malignant infarction of the middle cerebral artery (MCA) without increasing the number of severely disabled survivors, a pooled analysis of three trials of 93 randomized patients suggests.
While physicians have accepted the view that surgery to relieve intracranial pressure decreases mortality in these patients, the procedure has not been widely adopted by the medical community, said Werner Hacke, MD, PhD, a neurologist at the University of Heidelberg in Germany.
The reason, he said, is concern that patients who survive the procedure would be left in a chronic vegetative state.
The new analysis suggests that the “surgery results in a 25 percent absolute risk reduction in severe disability…and is likely to improve functional outcome among survivors,” Dr. Hacke said here at a late-breaking session at the International Stroke Conference in February.
However, the results published in Lancet Neurology (2007;6:215–222) also showed there were some survivors with moderate disability who needed help with activities of daily living.
Malignant MCA infarction is defined by stroke with increased intracranial pressure; it occurs in about 5 percent of ischemic stroke patients and, because it is often uncontrollable, the mortality rate is 80 percent, Dr. Hacke said.
The reason patients die is not stroke per se, but the edema that goes with it, he explained. With a small stroke, it doesn't matter, but if it's a big stroke, edema is life threatening “due to increased pressure in the skull,” he said.
“You need to relieve the pressure, but medical therapy has consistently failed,” he continued. Decompressive surgery removes a piece of the skull and dura, allowing the brain to expand, Dr. Hacke said.
Early uncontrolled studies suggested that mortality could be drastically reduced with decompressive surgery, from 80- to 40-percent or less, Dr. Hacke said. Performing a randomized controlled trial has been difficult, however, because of the fear that patients would be in a vegetative state, he said.
Nevertheless, three European study groups undertook trials: the Hemicraniectomy After Middle Cerebral Artery Infarction with Life-Threatening Edema trial, which is being conducted in the Netherlands; the German Decompressive Surgery for the Treatment of Malignant Infarction of the Middle Cerebral Artery trial; and the French Decompressive Craniectomy in Malignant Middle Cerebral Artery Infarcts trial.
Plagued by slow accrual, however, the investigators decided to take a novel approach and pool data — even before publication. The protocol was designed prospectively when the trials were still recruiting, so the investigators could define outcomes while blinded to the results of the individual trials, Dr. Hacke said.
He noted that the study, the first in the field of stroke in which a pooled analysis of individual patient data from three independent randomized trials was planned while the trials were still in progress, has the obvious advantage of keeping the number of patients to a minimum. It also allowed for publication of the results several years earlier than would have been possible based on the individual trials alone.
The final analysis included 93 patients, about one-third from each trial, randomized to surgery or conservative therapy within 30, 36, or 99 hours of stroke.
To be included, patients needed to be 18 to 60 years and have signs indicative of MCA infarction with a score on the National Institutes of Health Stroke Scale (NIHSS) of more than 15 points. Additionally, patients had to have signs on CT of an infarct of at least 50 percent, with or without additional infarction in the territory of the anterior or posterior cerebral artery on the same side, or infarct volume of greater than 145 cm3 on diffusion MRI. All patients were randomized within 45 hours of stroke onset.
The primary endpoint was the score on the modified Rankin Scale (mRS) at one year, dichotomized between favorable (0 to 4 points) — indicating no disability to moderately severe disability, meaning the patient was unable to walk without assistance and unable to attend to own bodily needs without assistance — and unfavorable (5 points) indicating severe disability, in which the patient was bedridden, incontinent and required constant nursing care and attention, and death (6 points). Secondary outcome measures included dichotomization of the mRS between 0 and 3, and 4 to death, and case fatality at one year.
Dr. Hacke said that although a score on the mRS of 3 points or less is generally accepted as a favorable outcome in stroke research, an mRS of 4 points or more was chosen as the primary outcome because survival with no or only slight disability after large MCA infarction is rare. As a result, the primary aim of the study was to assess whether decompressive surgery reduced mortality without an increase in the number of severely disabled survivors — that is, patients with an mRS score of 5 points, he said.
Results showed that decompression surgery reduced the risk of complete dependency or death, Dr. Hacke said. For the primary outcome measure, 75 percent of patients survived and had good outcomes at one year, as reflected by a modified Rankin Scale Score of 0 to 4 points, compared with 24 percent of control patients.
The number needed to treat to save one life with a modified Rankin score of 4 points or less was two, he said.
When the researchers looked at how many survived with less severe disability, the difference between surgically treated and unoperated groups wasn't as large, although it was still statistically significant. Specifically, 43 percent of patients who had surgery were alive with a modified Rankin score of 3 points or less at one year versus 21 percent of controls.
Four patients must have surgery for one patient to survive with a modified Rankin score of 3 points or less, Dr. Hacke said.
As for survival, 78 percent of patients randomized to surgery were alive at one year, compared with 29 percent of control patients, again a significant difference. The number needed to treat to save one patient was two.
Dr. Hacke noted that the effect of surgery was highly consistent in all three trials. Surgery was also significantly beneficial in all predefined subgroups: patients younger and older than 50 years, those with and without aphasia, and those randomized within or after 24 hours of stroke onset. While the results are encouraging, the decision to perform decompressive surgery should be made on an individual basis depending on willingness to accept survival with moderate disability, Dr. Hacke said.
Robert J. Adams, MD, Presidential Distinguished Chair, Regents and professor of neurology at the Medical College of Georgia in Augusta, said he is “very excited” about the results.
The question we have had for some time is how aggressively we should be treating this group of patients. This research not only answers that question but tells us exactly who should get it and when.
“As defined by the inclusion criteria, surgery should be done early, probably within the first 24 hours and at least by 45 hours. Patients should be selected based on a combination of clinical status – they should be scored with more than 15 points on the NIH Stroke Scale, be less than 60 years, and should have neuroimaging showing a large MCA stroke,” Dr. Adams said.
ARTICLE IN BRIEF
A pooled analysis of three large European trials showed that decompressive surgery reduced the risk of complete dependency or death in patients who have a malignant infarction of the middle cerebral artery without increasing the number of very severely disabled survivors.
DECOMPRESSIVE SURGERY: THE PROCEDURE
Decompressive surgery consisted of a duraplasty and the creation of a large bone flap. A large skin incision in the shape of a question mark based at the ear was made. A bone flap with a diameter of at least 12 cm (always including the frontal, temporal, and parietal bones) was removed. Additional temporal bone was removed so that the floor of the middle cerebral fossa could be reached. The dura was opened and a dural patch, consisting of pericranium or a commercially available dura substitute, was inserted and secured to enlarge the intradural space.
To prevent epidural bleeding, dural tacking sutures were used as necessary. The temporal muscle and the skin flap were then reapproximated and sutured. Infarcted brain tissue was not resected. In surviving patients, cranioplasty was undertaken after at least six weeks with the stored bone flap or acrylate. After surgery, patients were transferred to an intensive-care unit.
Source: Lancet Neurol 2007;6:215–222.
TO REACH NEUROLOGY TODAY
For Editorial, Permissions, or Publishing Matters:
Lippincott Williams & Wilkins
333 Seventh Ave., 19th Floor
New York, NY 10001
Phone: (646) 674–6545
Fax: (646) 674–6500
For Subscription Information and Address Changes:
American Academy of Neurology
1080 Montreal Ave.
St. Paul, MN 55116
Phone: (800) 879–1960
For Display Advertising:
Kathleen A. Harrison
Pharmaceutical Media, Inc.
30 East 33rd St., 4th Floor
New York, NY 10016
Phone: (212) 685–5010
Fax: (212) 685–6126