Published online before print November 10, 2005, doi: 10.1161/01.STR.0000190091.36915.84
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2005;36:2530-a.)
© 2005 American Heart Association, Inc.
Letters to the Editor |
Trial Design in "Magnesium Sulphate in Aneurysmal Subarachnoid Hemorrhage: A Randomized Controlled Trial"
Division of Neurosurgery, Prince of Wales Hospital, Chinese University of Hong Kong, Hong Kong, China
Department of Anaesthesia and Intensive Care, Prince of Wales Hospital, Chinese University of Hong Kong, Hong Kong, China
To the Editor:
We read with great interest the study by W.M. van den Bergh regarding the randomized controlled trial on Magnesium Sulfate in Aneurysmal Subarachnoid Hemorrhage1. In essence, it was a negative study using delayed cerebral ischemia (DCI) and poor outcome at three months from the data provided.
The methodology raised concerns us. The author, for some unexplained reason, truncated out the aspirin group’s case selection as well as data to carry out the analysis. As noted in the MASH group’s trial protocol published in the Internet Stroke Center,2 the way would be to add aspirin to the group that had aneurysm intervention within 4 days. If the aspirin group were actually excluded, the study population would be undoubtedly focused on patients with delayed aneurysm treatment. The study used DCI as the primary outcome measure and used CT evidence of infarct.
Magnesium is thought to be neuroprotective and reverse cerebral vasospasm in animal models. Thus, it would be of interest to look at their clinical vasospasm data and blood flow study data as Transcranial Doppler USG and SPECT. It would certainly enlighten about the possible actions of magnesium. The other reasons for the above data would be that some patterns of CT infarct may not be related to cerebral vasospasm, and some infarct may actually be masked by edema or metallic artifact as quoted by Rabinstein et al.3
The outcome was made in 3 months. From our experience, some of the poor-grade subarachnoid hemorrhage patients would show progressive improvements beyond 3 months after the initial insult. We proposed to also assess the 1-year outcome for more definitive conclusions. Our group is currently in the midst of conducting a multicenter trial directly looking at the administration of Magnesium Sulfate in patients with aneurysmal subarachnoid hemorrhage, aiming for a total patient recruitment of 340. The interim analysis was presented in the ICP Meeting held in Hong Kong last August.4
References
1. van den Bergh WM; on behalf of the MASH Study Group. Magnesium Sulfate in Aneurysmal Subarachnoid Hemorrhage. Stroke. 2005; 36: 1011–1015.
2. van den Bergh WM (contact). Magensium and Acetylsalicylic acid in Subarachnoid Hemorrhage (MASH). Stroke. 2002; 33: 646–655.
3. Rabinstein AA, Weigand S, Atkinson JLD, Wijdicks EFM. Patterns of Cerebral Infarction in Aneurysmal Subarachnoid Hemorrhage. Stroke. 2005; 36: 992–997.
4. Boet R, Chan MTV, Poon WS, Wong GKC, Wong HT, Gin T. Intravenous Magnesium Sulphate to improve outcome after aneurysmal subarachnoid hemorrhage. Acta Neurochir. 2005 (In press).
Response:
We thank the responders for their interest in our study. This trial was designed to study whether magnesium reduces the occurrence of delayed cerebral ischemia (DCI). It was not the aim of our study to clarify the proposed mechanism of action of magnesium in subarachnoid hemorrhage (SAH). After finding that magnesium reduces DCI, the next step is to study whether magnesium improves eventual outcome. For such a study, much larger numbers of patients are needed: based on the current data, at least 1200. We are currently embarking on such a trial.
Dr Gupta notes that the clinical outcome of reduction of DCI remains, however, uncertain. He provides extensive pharmacological reasoning why magnesium might be detrimental in many conditions including eclampsia. Reasoning will not always give a proper or right solution in medicine; the proof is in the eating of the pudding. Nimodipine was developed to reduce vasospasm. It works, but not through reducing vasospasm.1 With respect to eclampsia, there is ample clinical evidence regarding the beneficial effect of magnesium.2–4
Dr Gupta raises the question whether the proposed mechanism of action of magnesium in SAH is correct given the limited brain bioavailibilty of peripherally administered magnesium sulfate in patients with acute brain injury. However, a study in neurosurgical patients showed that peripherally administered magnesium sulfate does lead to a significant increase in the CSF magnesium concentration.5
The suggestion that hypomagnesemia might be beneficial is in contradiction with our observations in 107 consecutive patients admitted within 48 hours after SAH, the adjusted hazard ratio of hypomagnesemia for DCI being 1.9; 95% CI 0.7 to 4.7, which is hardly compatible with a benefit of low magnesium levels.6 Rather, the results of that study suggest a causal relation between hypomagnesemia and the occurrence of DCI, although we also feel that it is more likely that hypomagnesemia is just part of a more complex pathophysiological mechanism.
Dr Gupta ends with remarking that we may have to reexamine our belief that cerebral vasospasm in SAH can be usefully modified by systemic vasodilators and that regional or local measures at the level of the CSF might yield better results to manage the blood-related vasospasm. We do not aim to treat vasospasm but to prevent and treat DCI. Because only 70% of patients with vasospasm develop DCI and 30% of the patients with DCI have no vasospasm,7 we feel that focusing on vasospasm may yield results that are clinically less relevant. This is the very reason why we do not have the Transcranial Doppler USG and SPECT data Dr Wong et al are asking for. Wong and others conclude that this was a negative study. This comment has to do with the general way of presenting the results of a clinical trial. We strongly believe that a clinical trial first and for all should be seen as a tool to quantify a treatment effect and to accompany this treatment effect with a measure of its precision. We consider emphasis on hypothesis testing less appropriate because this leads to an unnecessary dichotomy of trial results into "positive" and "negative," whereas there is much more than that in a trial. Had the trial been twice as large, we would have reached statistical significance for all outcome measures even if the effect size would be the same. In that situation the responders undoubtedly would have classified the trial to be "positive", with the same effect size. Hence, the problem is the precision of effect estimate. We fully acknowledge that the final word has not yet been said; for that reason we plan a second, larger trial. But to label our current trial as "negative" is too simplistic an approach. If we consider this phase II trial as "negative", no further phase III trials would be performed. This implies that a safe and inexpensive treatment that very well may prove to be effective would not be further studied.
Further, Wong and others are concerned that the aspirin group has been excluded from our analyses on the effect of magnesium. Moreover, they worry that we focused on patients with delayed aneurysm treatment. Our trial assessed the effects of 2 different agents, magnesium and aspirin, by means of a factorial design, implying that patients were randomized twice (if eligible for each of the agents), once for magnesium versus placebo, and once for aspirin versus placebo. In our recent article, only the results of the magnesium part of the study were presented. We did not exclude the aspirin group, but at the time of publication the results for that part of the trial were still unknown. Patients allocated to aspirin, however, were equally distributed among the patients on magnesium or placebo-magnesium.
We agree that it can be difficult to distinguish ischemic cerebral lesions on brain CT from other causes of hypodense lesions. For that reason we also imbedded clinical information to be as accurate as possible about the origin of new hypodense lesions on brain CT, but we may have misjudged some of the lesions.
The experience from the authors that the outcome after SAH is still improving after 3 months is in line with our own finding.8 Further improvement after 3 months would probably dilute the difference between both groups; however, immediately returning home after discharge instead of first spending a couple of months in a rehabilitation facility is in our opinion a clinically relevant difference, even if the eventual results after 1 year or more are similar. Moreover, many SAH trials use a primary outcome at 3 months.
We are looking forward to the results of the clinical trial (iMASH) Dr. Wong and others are performing.
Received June 8, 2005; accepted September 28, 2005.
References
1. Rinkel GJ, Feigin VL, Algra A, van den Bergh WM, Vermeulen M, van Gijn J. Calcium antagonists for aneurysmal subarachnoid haemorrhage. Cochrane Database Syst Rev. 2005; CD000277.
2. Altman D, Carroli G, Duley L, Farrell B, Moodley J, Neilson J, Smith D; Magpie Trial Collaboration Group. Do women with pre-eclampsia, and their babies, benefit from magnesium sulphate? The Magpie Trial: a randomised placebo-controlled trial. Lancet. 2002; 359: 1877–1890.[CrossRef][Medline] [Order article via Infotrieve]
3. Which anticonvulsant for women with eclampsia? Evidence from the Collaborative Eclampsia Trial. Lancet. 1995; 345: 1455–1463.[Medline] [Order article via Infotrieve]
4. Lucas MJ, Leveno KJ, Cunningham FG. A comparison of magnesium sulfate with phenytoin for the prevention of eclampsia. N Engl J Med. 1995; 333: 201–205.
5. Fuchs-Buder T, Tramer MR, Tassonyi E. Cerebrospinal fluid passage of intravenous magnesium sulfate in neurosurgical patients. J Neurosurg Anesthesiol. 1997; 9: 324–328.[Medline] [Order article via Infotrieve]
6. van den Bergh WM, Algra A, van der Sprenkel JW, Tulleken CA, Rinkel GJ. Hypomagnesemia after aneurysmal subarachnoid hemorrhage. Neurosurgery. 2003; 52: 276–282.[CrossRef][Medline] [Order article via Infotrieve]
7. Rabinstein AA, Friedman JA, Weigand SD, McClelland RL, Fulgham JR, Manno EM, Atkinson JL, Wijdicks EF. Predictors of cerebral infarction in aneurysmal subarachnoid hemorrhage. Stroke. 2004; 35: 1862–1866.
8. Hop JW, Rinkel GJ, Algra A, van Gijn J. Quality of life in patients and partners after aneurysmal subarachnoid hemorrhage. Stroke. 1998; 29: 798–804.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||