Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Tips for migraine sufferers (with visual aura)
- Adrian S. Blaj (20 November 2004)
-
Meta-analysis of role of metoclopramide for aborting migraine attacks: quo vadis?
- Vinod K Gupta (26 November 2004)
-
Number not to respond ?
- Axel Ellrodt (10 December 2004)
-
Beware parenteral metoclopramide
- Barry G Smiler (13 December 2004)
-
Metoclopramide not "the drug of first choice"
- Peter J Effeney (14 December 2004)
-
Is there a possibility of bias?
- Ramesh Donepudi (14 December 2004)
-
Need to carry out detailed work on Metoclopramide
- Dr Buddhi P Paudyal (15 December 2004)
-
First-line use of intravenous metoclopramide for migraine: an unsubstantiated proposition
- Shilpa Patel (10 January 2005)
-
At last!
- Diane-Marie Campbell (12 January 2005)
|
|
|||
|
Adrian S. Blaj, Psychiatrist London, England
Send response to journal:
|
Madam: I have read with the great interest the article 'Parenteral metoclopramide for acute migraine: meta-analysis of randomised controlled trials'. Partly because I am a sufferer of migraine myself since age 9. The authors have calculated that there is an average of 36 episodes per person per year which in itself leads to a staggering loss of productivity and increase in the utilisation of health resources given the fact that migraine is found in 15 - 17 per cent of women and 6 per cent of men --- not counting the impaired quality of life, the pressure put on the personal and professional relationships and living with the constant anxiety of the next attack of migraine. The authors indicate that parenteral metoclopramide is effective for the treatment of acute attacks of migraine given its non-narcotic and antiemetic properties. It is interesting to note that in the late 1970's, anecdotal reports suggested that patients receiving metoclopramide for nausea complained of less pain even before being given pain relief. Metoclopramide appears to relieve gastric stasis potentially improving gastric reabsorption. And here comes my own empirical experience; about ten years ago I read about the theory that once the splitting headache is established, there is a secondary gastric stasis which naturally affects the absorption of painkillers (I always wondered why, no matter how much analgesic tablets I used in the past, it seemed to make no difference). As I am one of those 'fortunate' migraine sufferers who develop a visual aura some 20 - 30 minutes before the splitting headache, I simply take a baby dose of soluble paracetamol and subsequently am able to carry on with my work almost normally, without headache (just a slight feeling of cranial pressure). The only problem which I cannot resolve is the occasional attack of migraine which starts during sleep - fortunately this occurrence happens only a few times a year. I can vouch that my simple tactic of taking analgesia well before the migraine is established has enormously improved my quality of life. Hope that my tip is going to help some of the migraine sufferers out there - 'unfortunately' not everyone has a warning aura... adrian>blaj@beh-mht.nhs.uk Competing interests: None declared |
|||
|
|
|||
|
Vinod K Gupta, Physician Dubai Police Medical Services, P.O. Box 12005, Dubai, United Arab Emirates
Send response to journal:
|
Colman and colleagues present a treatise on the role of parenteral metoclopramide in the management of migraine attacks that is methodologically sound and, at first reading, appears logical and defensible. (1) Meta-analyses, however, are susceptible to the inherent biases of all observations; trials included in meta-analyses frequently do not ask the same questions. (2, 3) Meta-analysis can distance readers from original data while the statistical combination of studies -- that meta- analyses rest upon -- is rarely appropriate in observational research. (4) Such models may be statistically valid, but might be biologically or clinically invalid and some advocates of meta-analysis reject them (5, 6). A danger inherent in the use of meta-analysis to evaluate large effects is that its seeming legitimacy, because it comes up with the "right" answer, can be interpreted as conferring legitimacy to the meta-analysis of small effects. For small effects, meta-analysis may readily lead to spurious conclusions (5) and it is in the evaluation of small effects that meta- analysis is now most commonly being used. (3) When these fundamental limitations of meta-analysis are juxtaposed to the critical limitations of the randomized controlled trial (RCT) itself, (7, 8) the clinical conviction with which biologically small but statistically significant changes can be viewed is necessarily limited. Meta-analysis seeks or forces consensus through mathematical logic in the state of biological confusion. (9) With a topic like migraine, the pathophysiology of which is in its infancy, use of meta-analysis should probably be avoided because: (i) all current therapy is empirical; (ii) end-points in clinical trials are subjective and soft; (iii) the migraine attacks is essentially self- limited; and (iv) key pharmacological tenets, including blood-brain barrier permeability of anti-migraine agents, have been largely if not completely neglected. (10, 11) Both the RCT as well as meta-analysis underscore that procedure has become primary in medical research; it isn’t what you, it is how you do it. Means not ends are what matters. Since there is no such thing as a “purely objective observation”, (12) the RCT and the meta-analysis both spring from a preconceived notion – an intrinsic investigator-related “need-to-believe” bias. Colman et al conclude that metoclopramide is an effective treatment for pain of migraine headache that should be considered as a primary line of therapy in emergency departments, alone or in combination with other therapies. (1) Since prediction is impossible without generalization, (13) it cannot be over-emphasized that outside of primary headache – particularly migraine – research, there is no reference to a clinically useful analgesic property of metoclopramide. Even in migraine, the antinociceptive “effect” of metoclopramide is far from having been seen uniformly across different clinical studies. If only one out of four patients treated with parenteral metoclopramide can achieve significant reduction in pain, (1) the putative antinociceptive action of metoclopramide is highly unpredictable. Even more disconcerting is the absence of any clear influence of metoclopramide, alone or in combination with other therapies, upon the nausea of migraine. Further trials with multiple arms, increasing focus on adequate randomization, and increasingly rigorous concealment of allocation, matching of controls and blinding of researchers, as suggested, (1) are simplistic methodological refinements that do not address the questions that lie at the heart of the matter. Any inquiry must begin with the right question that remains valid after the data are received. (2) For migraine pathophysiology – and eventually therapy with drugs like metoclopramide – the crucial question is the biological nature of nausea and / or vomiting. All documented phenomenological – clinical or laboratory – perturbations of migraine, including nausea and vomiting, have been illogically consigned to presumed pathogenetic algorithms. Unless patients are viewed as being balanced between opposing physiological forces pushing them towards health or disease -- headache or headache-free state -- the biology of migraine will remain mysterious. (14) The key question related to use of parenteral metoclopramide for managing attacks of migraine is the biological role of nausea / vomiting. Given that vomiting generally diverts the pain or renders it more moderate, nausea and / or vomiting probably represent an important anticipatory or attack-related neuroendocrinological mechanism to limit the impact of headache in migraine patients. (14, 15) The benefit of metoclopramide to nausea in migraine is, at best, equivocal. Nausea, even without vomiting, is a potent, dominant and specific stimulant for vasopressin (AVP) release. (15) Vasopressin likely constitutes an important neuroendocrine adaptive mechanism that can significantly delay the onset of headache or abort the headache attack by improving vasomotor control, antinociception and behaviour control. (16) Therapeutic attenuation of nausea during migraine attacks should, therefore, not be viewed as a primary or highly desirable management strategy. Given this background, it should not be surprising that metoclopramide does not appear to have a definitive role in management of nausea of migraine. Finally, metoclopramide itself stimulates vasopressin secretion in man, particularly in the elderly, possibly through cholinergic mechanisms. (17,18) Nausea and metoclopramide, therefore, both activate the same neuroendocrinal system. Metoclopramide probably cannot stimulate AVP secretion beyond that already achieved by migraine attack-associated nausea. Use of metoclopramide in migraine patients with nausea – a significantly large group -- does not appear to be scientifically logical. For migraine patients without nausea / vomiting, metoclopramide has yet to be shown to be clearly superior to conventional agents such as parenteral non-steroidal anti-inflammatory agents or ergot preparations or triptans. The case for considering metoclopramide as a primary therapeutic agent during migraine attacks is weak. Much more conceptual groundwork and pathophysiological consensus need to be evolved before further trials of metoclopramide in acute attacks of migraine might be regarded as being scientifically and ethically appropriate. References 1. Colman I, Brown MD, Innes GD, Grafstein E, Roberts TE, Rowe BH. Parenteral metoclopramide for acute migraine: meta-analysis of randomized controlled trials. BMJ, doi:10.1136/bmj.38281.595718.7C (published 18 November 2004). 2. Horton RC, Kendall MJ. Clinical pharmacology and therapeutics. Postgrad Med J 1991;67:1042-54. 3. Kaufman DW, Shapiro S. Narrative review should have been used. BMJ 1997;314:445--6. 4. Smith GD, Egger M. Meta-analyses of observational data should be done with due care. BMJ 1999;318:56. 5. Shapiro S. Meta-analysis/shmeta-analysis. Am J Epidemiol 1994;140:771--7. 6. Greenland S. Can meta-analysis be salvaged? Am J Epidemiol 1994;140:783--7. 7. Gupta VK. Randomized controlled trials: the hijacking of basic sciences by mathematical logic. BMJ Online. Available at: http://bmj.bmjjournals.com/cgi/eletters/329/7456/2#65969. 8. Gupta VK. Randomized controlled trials versus clinical realities: prostate cancer screening. BMJ Published online 11 February 2004. Available at: http://bmj.bmjjournals.com/cgi/eletters/328/7435/301#50002. 9. Gupta VK. Does magnesium supplementation have any role in acute myocardial infarction? No. Cardiovasc Drugs Therapy 1996;10:303-305 (responses 297-301;307). 10. Gupta VK. Classification of primary headaches: pathophysiology versus nosology? BMJ 2004; published online Jan 22. Available at: http://bmj.bmjjournals.com/cgi/eletters/328/7432/119. 11. Gupta VK. Bureaucratization of migraine. Lancet Neurology. 2004;3:396. 12. Medawar PB. The Art of the Soluble. London: Metheun, 1967. 13. Poincare H. Science and hypothesis. New York: Dover, 1952: 142. 14. Gupta VK. Conceptual divide between adaptive and pathogenetic phenomena in migraine: nausea and vomiting. Brain 2004; 127: E18. 15. Gupta VK. Cyclic vomiting syndrome: anticipatory stress response in migraine? Headache 2004;44:106-107. 16. Gupta VK. A clinical review of the adaptive role of vasopressin in migraine. Cephalalgia 1997;17:561-9. 17. Norbiato G, Bevilacqua M, Chebat E, et al. Metoclopramide increases vasopressin secretion. J Clin Endocrinol Metab 1986;63:747-50. 18. Bevilacqua M, Norbiato G, Chebat E et al. Osmotic and nonosmotic control of vasopressin release in the elderly: effect of metoclopramide. J Clin Endocrinol Metab 1987;65:1243-7. Competing interests: None declared |
|||
|
|
|||
|
Axel Ellrodt, MD American Hospital of Paris 92202 Neuilly France
Send response to journal:
|
This review leaves me a bitter taste. That metoclopramide can be statistically shown more effective than placebo is indeed good news. But the number needed to treat is 4. I understand 3 patients will not be relieved.I cannot be convinced that I should use this treatment alone in my emergency department. We need randomised controlled trials of triptans vs cheap, widely used emergency department treatments: IV NSAIDs, neuroleptics, IV paracetamol, dihydroergotamin and their combinations. Our ED migraineurs probably are not similar to most patients enrolled in RCTs led by neurologists. Neurologists treat migraine headaches that have not yet occurred and that they do not see, whereas emergency physicians actually see and treat ongoing migraines that probably are different since the patient goes to the ED. Competing interests: None declared |
|||
|
|
|||
|
Barry G Smiler, Anesthesiologist Sarasota Fla 34236,USA
Send response to journal:
|
The use of parenteral metoclopramide may indeed alleviate migraine pain. However, it is well known that intravenous metoclopramide is associated with akathisia, ventricular dysrrthymias and mood alterations. Patients have been known to bolt from the operating suite and disavow planned surgery after intravenous metoclopramide. Low dose droperidol is also well known to relieve the migraineur at a lower overal side effect risk. Barry Smiler MD
Competing interests: None declared |
|||
|
|
|||
|
Peter J Effeney, Anaesthetic SHO Anaesthetic Department, Dorset County Hospital, Dorchester, Dorset, DT1 2JY
Send response to journal:
|
This paper seems to confirm, despite its limitations, that parenteral metoclopramide is a reasonably effective treatment for acute migraine. (1) In fact, the authors are quite cautious in their estimation of it's treatment effect and even note that other agents, particularly chlorpromazine, seem to offer greater efficacy. The editorial team at the BMJ is somewhat less reserved as reflected in the title of my response. In my own experience parenteral chlorpromazine (0.1-0.2 mg/kg) diluted in crystalloid (10 ml/kg) in conjunction with a quiet, darkened cubicle seems to offer the greatest treatment benefit in patients presenting to emergency departments with severe migraine. A quick review of recent literature seems to support the case for using chlorpromazine. A recent randomised control trial (RCT) found significant improvements in pain intensity, nausea, photophobia and phonophobia at 30 and 60 minutes, as well as decreased recurrence at 24 hours, following chlorpromazine administration (0.1 mg/kg). (2) The calculated number needed to treat (NNT) was two. (2) A recent literature review of emergency department treatment of migraine suggested chlorpromazine, among others, as the most useful agent with greater than 70% efficacy in some studies reviewed. (3) Parenteral chlorpromazine is at very least, as efficacious as metoclopramide. (4) The disadvantages of chlorpromazine are its anticholinergic and antidopaminergic side effects. These include sedation, dry mouth, postural hypotension, and extra-pyramidal reactions. Metoclopramide shares the propensity to produce dystonic reactions but is generally less sedating. These side effects can be managed with relative ease in emergency departments. A final caution regarding chlorpromazine is its ability to provide symptom relief in patients with more serious causes of headache including subarachnoid and subdural haemorrhage. (5) Metoclopramide is one treatment available to emergency physicians for use in acute severe migraine. Performing a meta-analysis of available data does not make it "the" treatment. References 1. Colman I, Brown MD, Innes GD, Grafstein E, Roberts TE, Rowe BH. Parenteral metoclopramide for acute migraine: meta-analysis of randomized controlled trials. BMJ, doi:10.1136/bmj.38281.595718.7C (published 18 November 2004). 2. Bigal ME, Bordini CA, Speciali JG. Intravenous chlorpromazine in the emergency department treatment of migraines: a randomised control trial. J Emerg Med 2003 Oct;25(3):322-3 3. Kelly AM. Migraine: pharmacotherapy in the emergency department. West J Med 2000 Sep;173(3):189-93 4. Cameron JD, Lane PL, Speechley M. Intravenous chlorpromazine vs intravenous metoclopramide in acute migraine headache. Acad Emerg Med. 1995 Jul;2(7):597-602. 5. Barclay CL, Shuaib A, Montoya D, Seland TP, Thomas HG. Response of non- migrainous headaches to chlorpromazine. Headache. 1990 Jan;30(2):85-7. Competing interests: None declared |
|||
|
|
|||
|
Ramesh Donepudi, SpR Stoke on Trent
Send response to journal:
|
As the authors rightly mentioned, Metoclopramide reduces gastric stasis and improves the transit time of gastric contents.Conventional wisdom suggests using a prokinetic agent along with simple analgesics when treating acute migraine. But most patients would have used simple analgesics often in good measure before attending Emergency departments. Is there a possibility, that the resolution of symptoms after using Metoclopramide as a single agent to treat an acute attack the result of better absorption of the analgesics? Competing interests: I myself am a fan of giving Metoclopramide along with simple analgesics in Acute Migraine. |
|||
|
|
|||
|
Dr Buddhi P Paudyal, Physician Patan Hospital, Lalitpur, Nepal
Send response to journal:
|
The meta-analysis by Colman and others on efficacy of parenteral metoclopramide in migraine headache is really encouraging. As in other parts of world, migraine is a frequent cause of emergency and outpatient department visits in our setting. Due to lack of standard treatment protocol, there seems to be heterogeneity in treatment patterns- different patients receiving different medicines for the same problem. We frequently prescribe metoclopramide in migraine headache but mainly as an antiemetic agent. This meta-analysis has uncovered the potential use of this cheap and less toxic but highly efficacious drug in aborting acute attacks of migraine. I think further large scale randomized controlled trials are necessary to compare this agent with standard treatment options that are existing at the moment. Similarly the use of oral preparation should also be explored as its use at home/ workplace may avoid many unnecessary emergency department visits. Competing interests: None declared |
|||
|
|
|||
|
Shilpa Patel, PRHO, Respiratory/General Medicine Whittington Hospital, Highgate Hill, London N19 5NF
Send response to journal:
|
Sir I cannot see how metoclopramide could even be considered as a first- line treatment for migraine on the data you have presented. The only two sections in your article that could back this are the comparisons with non -antiemetics and other agents. In clinical practice the vast majority of patients who receive antiemetics are given them as an adjunct to treat the nausea associated with migraine rather than the migraine itself and rarely do they have them without simple analgesia. This renders the comaparison with antiemetics irrelevant to your main proposition, leaving you with 9 studies comprising only 271 patients. And as you point out most are relatively poor studies with little scope for statistical pooling. This is a weak basis for advocating a new first-line treatment to such a major problem. Furthermore side-effects are poorly reported in many of the studies. The most worrying ones are neurological. Although the actual incidence is rare, the relative risk could be as high as 4.0 for drug-induced acute extrapyramidal movements and 1.67 for tardive dyskinesia (1). Given the large numbers of patients involved (7% in men and 18% of women in the United States (2)) that could be a major problem. Larger studies are needed before metoclopramide could be even considered as a first line treatment for migraine. 1 Ganzini L, Casey DE, Hoffman WF, McCall AL. The prevalence of metoclopramide-induced tardive dyskinesia and acute extrapyramidal movement disorders. Arch Intern Med. 1993 Jun 28;153(12):1469-75. 2 Hu XH, Markson LE, Lipton RB, Stewart WF, Berger ML. Burden of migraine in the United States: disability and economic costs. Arch Intern Med 1999;159: 813-8. Competing interests: None declared |
|||
|
|
|||
|
Diane-Marie Campbell, locum emergency physician Bunbury Regional Hospital, West Australia, 6230
Send response to journal:
|
I was pleased to see an article in the BMJ finally acknowledging that metoclopramide works for migraine, particularly as the last BMJ review of migraine (1) had perpetuated the myth that metoclopramide was a mere adjunct for treatment of associated nausea. Emergency physicians have been using this modality, as Colman et al acknowledge, for some 30 years, generating RCT's which have been in the literature nearly as long. The intravenous route may be more effective for migraine (2) although intramuscular, rectal and oral administration relieve nausea. Perhaps the NNT would decrease if the intravenous route were used. Chlorpromazine is indeed effective, but somewhat problematical in departments still relying on junior staff. In spite of well-written protocols and verbal instruction in words of one syllable, chlorpromazine is occasionally given as an intravenous bolus to a patient who has not had a fluid load. Admittedly the headache has often been relieved by the time the patient is resuscitated - provided he can be, and I am aware of one instance at least, in a hospital where I was working, which resulted in death. A standard of care which demands a NNT of less than 4 for interventions would be difficult to meet in most emergency departments, and cause widespread therapeutic nihilism. No thrombolysis. No PPI's. (3) No codeine- paracetamol (4) and almost no antimicrobials, according to the BMJ POEMs(5,6) and Clinical Evidence Reviews (7) because a lot of antivirals and antibiotics, either don't work (6) or have NNT of more than 4. (5,7) Critics should audit their own departments to find how many patients who present with pain have significant relief when they go to the ward or are discharged. (8,9) One in four with total resolution is better than average. Of course we should do better - but we don't. And there is nothing in Colman et al's review to prevent one adding simple analgesics to the metoclopramide. Simple, safe, inexpensive, and effective. Side-effects are readily treatable, likely to manifest while the patient is still being observed (if given I/v) and metoclopramide won't interfere with any additional drugs should they be needed. References
Competing interests: None declared |
|||