Social workers merit much more respect than that, Phil
Re: 'Does aspartame trigger headaches?' (Clinical, May 28). It is unfortunate that the authors, Moore and McQuay, leading exponents in the application of evidence-based medicine, use anecdotes (evidence-base level <4), case="" reports="" (level="" 3)="" and="" two="" poorly="" conducted="" trials="" (level="">4),>
1-) to conclude that there is a link between aspartame and headache. They even coin a new, and unsubstantiated, condition of 'aspartame-triggered migraine'.
High-grade evidence from a well-conducted study and systematic reviews (level 1+) do not support this conclusion.
The Centers for Disease Control reported in 1984 that their 'investigation of consumer complaints of symptoms experienced after consumption of aspartame-containing products identified no specific constellation of symptoms clearly related to aspartame consumption1'.
They went on to say: 'Reports are undoubtedly due to the coincidence of symptoms and aspartame consumption, and others may be due to the suggestibility of some persons. Still others may be attributable to some as yet undefined sensitivity of some individuals to aspartame in commonly consumed amounts. The only way these possibilities can be thoroughly evaluated would be through focused clinical studies.'
In response, Schiffman et al2 carried out a study in 40 (not 'a few tens') of subjects who specifically had linked their headaches to aspartame (and had reported their complaints to a manufacturer of aspartame). As figure 1 (unreferenced and uncited in the article) shows, exposure to placebo caused more headache (and all symptoms) than aspartame given as three successive 10mg doses.
The author calculated a 95 per cent confidence interval for the difference in incidence between aspartame and placebo as -35 per cent to +12 per cent.
Moore and McQuay state that a completers analysis of studies with a near 50 per cent drop-out rate is problematic3,4. It is worse, making these studies of little value.
One must assume an intention to treat analysis would fail to show a significant effect, further compounded by the fact that in both trials the blinding was broken by some subjects. Furthermore, in the van den Eeden study, the four subjects with migraine had no excess of headaches with aspartame4.
While we clearly do not have data to exclude rare cases of headache being precipitated by aspartame, the balance of evidence does not suggest this as a cause.
If dietary factors are to be considered as environmental triggers of headache it would be better to consider missing meals, alcohol, cheese, citrus fruits and chocolate5, caffeine and foods containing high levels of amines, nitrites and nitrates before implicating aspartame.
There is a real danger in misdiagnosing headaches to a food substance whose safety has recently been reviewed and reaffirmed6.
Dr N Finer
Senior Clinical Research Associate
University of Cambridge Clinical School of Medicine
1. Centers for Disease Control. Morbidity and Mortality Weekly Report (1984). 33(43);605-7
2. Schiffman SS et al(1987). Aspartame and susceptibility to headache. N Engl J Med 317:1181-1185
3. Koehler SM and Glaros A (1988). The effect of aspartame on migraine headache. Headache 28: 10-14
4. Van den Eeden SK et al(1994). Aspartame ingestion and headaches: a randomised crossover trail. Neurology 44: 1787-1793
5. British Association for the Study of Headache. Guidelines for All Doctors in the Diagnosis and Management of Migraine and Tension-Type Headache (2004). Writing Committee: TJ Steiner,
EA MacGregor, PTG Davies http://220.127.116.11/NS_BASH/BASH%20guidelines%2025Feb05.pdf
6. European Commission. Health and Consumer Protection Directorate-General, Scientific Committee on Food. Opinion of the scientific committee on food: update on the safety of aspartame. SCF, December 10, 2002 http://europa.eu.int/comm/food/fs/sc/scf/index_en.html
Andrew Moore and Henry McQuay reply: Aspartame is mentioned as a potential trigger for migraine in a number of studies. In a beautiful example of a case report1, migraine experts clearly demonstrated sensitivity to tiny amounts of aspartame from a wafer formulation of rizatriptan in two patients.
Oral rizatriptan aborted migraine attacks, while the wafer formulation made migraine worse, despite the presence of rizatriptan. The wafers contained 3.8mg of aspartame, a fraction of that in a diet drink, and a tiny fraction of that in some people's diet.
Case reports, when done well, can tell us things that randomised trials and even systematic reviews of randomised trials cannot, especially about rare events.
The old rule of three states that if no event is found in 1,000 people then it is unlikely to occur more often than one in 333. To be sure that an event occurs less frequently than one in 10,000 people, we would need information on 30,000 people.
For aspartame, we have evidence that it can occur in some. We also know from some randomised challenge tests that it probably does not occur in more than one in 30 people (the 100 in the three trials, divided by three).
This is what we know, and the EC report was based on no more information. We said: 'The lesson was probably to consider aspartame in some people with migraine who consume lots of diet drinks or other diet products.'
In the current state of knowledge, and without a remarkably larger body of evidence than currently available, that still seems sensible.
1 Newman LC, Lipton RB. Migraine MLT-down: an unusual presentation of migraine in patients with aspartame-triggered headaches. Headache. 2001;41:899-901