Early decompressive surgery in malignant infarction of the middle cerebral artery: a pooled analysis of three randomised controlled trials.
This review may be edited
Posted By: Gordon Doig
Posted Date: 11 July 2007
Title: Early decompressive surgery in malignant infarction of the middle cerebral artery: a pooled analysis of three randomised controlled trials.
Authors: Vahedi K, Hofmeijer J, Juettler E, Vicaut E, George B, Algra A, Amelink GJ, Schmiedeck P, Schwab S, Rothwell PM, Bousser MG, van der Worp HB, Hacke W; DECIMAL, DESTINY, and HAMLET investigators.
Reference: Lancet Neurol. 2007 Mar;6(3):215-22.
Link: Click here for a direct link to the paper. A password may be required for access to fulltext.
Abstract: BACKGROUND: Malignant infarction of the middle cerebral artery (MCA) is associated with an 80% mortality rate. Non-randomised studies have suggested that decompressive surgery reduces this mortality without increasing the number of severely disabled survivors. To obtain sufficient data as soon as possible to reliably estimate the effects of decompressive surgery, results from three European randomised controlled trials (DECIMAL, DESTINY, HAMLET) were pooled. The trials were ongoing when the pooled analysis was planned.
METHODS: Individual data for patients aged between 18 years and 60 years, with space-occupying MCA infarction, included in one of the three trials, and treated within 48 h after stroke onset were pooled for analysis. The protocol was designed prospectively when the trials were still recruiting patients and outcomes were defined without knowledge of the results of the individual trials. The primary outcome measure was the score on the modified Rankin scale (mRS) at 1 year dichotomised between favourable (0-4) and unfavourable (5 and death) outcome. Secondary outcome measures included case fatality rate at 1 year and a dichotomisation of the mRS between 0-3 and 4 to death. Data analysis was done by an independent data monitoring committee.
FINDINGS: 93 patients were included in the pooled analysis. More patients in the decompressive-surgery group than in the control group had an mRS less than or equal to 4 (75%vs 24%; pooled absolute risk reduction 51% [95% CI 34-69]), an mRS less than or equal to 3 (43%vs 21%; 23% [5-41]), and survived (78%vs 29%; 50% [33-67]), indicating numbers needed to treat of two for survival with mRS less than or equal to 4, four for survival with mRS less than or equal to 3, and two for survival irrespective of functional outcome. The effect of surgery was highly consistent across the three trials.
INTERPRETATION: In patients with malignant MCA infarction, decompressive surgery undertaken within 48 h of stroke onset reduces mortality and increases the number of patients with a favourable functional outcome. The decision to perform decompressive surgery should, however, be made on an individual basis in every patient.
Are the Results Valid?
1. Was the assignment of patients to treatments randomized? ( Was allocation concealment maintained?)
Two of the three pooled trials used central randomisation however the third remains uncertain.
The authors describe that DECIMAL was 'centrally randomised' in blocks of four. Using consistent block sizes in an unblinded trial can lead to guessing or anticipating upcoming group assignments every fourth patient. For example, if two of the first three patients were control and one was decompression, all investigators would know the fourth patient MUST be decompression so that the 'block of four' balances.
In DESTINY, although the authors report that the randomisation list was generated centrally, they do not explicitly report that patients were randomised centrally.
In HAMLET, the authors report the use of central computer radomisation, using a minimisation approach.
2. Were all patients who entered the trial properly accounted for and attributed at its conclusion?
2a. Was followup complete?
No. The investigators only included patients randomised to all three trials before 1 Nov 2005. This decision may have resulted in the exclusion of available patient outcomes from HAMLET.
Because HAMLET was still recruiting patients at the time of this analysis, and because both other trials were stopped early, we should treat the results of this pooled analysis as we would any other interim analysis. This extends to the application of a very conservative threshold to the p-value of this analysis.
2b. Were patients analyzed in the groups to which they were randomized?
One crossover (from standard care to decompression) was reported in DESTINY. It appears this patient was analysed as standard care (the group to which they were first randomised).
3. Were patients, health workers, and study personnel blind to treatment?
4. Were the groups similar at the start of the trial?
Baseline balance is reported as being 'broadly' similar (Table).
5. Aside from the experimental intervention, were the groups treated equally?
Not explicitly reported for each trial.
What are the Results?
1. How large was the treatment effect?
Primary outcome
Significantly more patients in the decompressive-surgery group than in the control group had an mRS less than or equal to 4 (favourable outcome):
(75% vs 24%; pooled absolute risk reduction 51% [95% CI 34-69], p less than 0.00001)
Secondary outcome
Significantly more decompression patients survived:
(78% vs 29%; 50% [33-67], p less than 0.00001).
2. How precise was the estimate of the treatment effect?
See 95% confidence limits, above.
Will the Results Help Me In Caring For My Patients?
1. Can the results be applied to my patient care?
Yes, as long as these patients can be identified early enough and as long as the remaining HAMLET patients are consistent with these findings.
Before we apply these results however, we should review the a priori stopping rules for DECIMAL and DESTINY. Since both were stopped early, it is important to know whether unbiased, objective and a priori rules were used to stop these two trials.
2. Were all clinically important outcomes considered?
Survival and quality of survival at one year are likely the two most important outcomes in these patients. There is an increase in the number of survivors and in favourable outcome (fewer disabilities).
3. Are the likely treatment benefits worth the potential harms and costs?
Although costs were not specifically calculated, the increase in favourable outcomes are likely worth any additional costs.

What other people had to say about:
Early decompressive surgery in malignant infarction of the middle cerebral artery: a pooled analysis of three randomised controlled trials.

Add Your Comment
[ Back ]

Any questions or comments please contact Gordon S. Doig
Implemented and designed by John Soer and Gordon Doig
Page last modified on Friday August 10, 2001