Migraine league table:
acute treatments and two hour pain free
There are a number of treatments available for acute migraine. This page summarises their efficacy in several ways. The information is drawn predominantly from systematic reviews of comparable trials. The important points that make the treatments comparable are as follows:
- Trials enrolled similar patients, with migraine defined according to IHS criteria .
- Trials enrolled adults.
- Pain had to be moderate to severe intensity before treatments were used.
- The outcome used was pain free at two hours, so that pain had to be completely gone at two hours.
- Trials were randomised, and double blind.
- An intention-to-treat approach was taken that included all patients randomised and who took the treatment, but not those randomised who never had a migraine.
- Only the first attack information was included.
- Treatments have all been compared with placebo.
So what we have here is a consistent data set. We can tell it is consistent by examining the placebo responses for the treatments - they are all about 8%.
Dangers of over-interpretation
Information from league tables is sometimes over-interpreted. It should not be used to exclude treatments from formularies, for instance. rather we should celebrate the fact that we have so many effective remedies. Some arguments against over-interpretation follow:
- This league table looks only at one outcome, that of pain free at two hours.
- The outcome used is refers only to pain. What about nausea, vomiting, photophobia, phonophobia, or other outcomes that are important in migraine?
- We cannot give information in a league table about adverse effects. Some patients will "do well" on one treatment but not another.
- We cannot here give a league table for cost. That will differ in different places around the world.
Patient choice, and professional choice, will be influenced by many factors. Analgesic efficacy is only one. Moreover, relative efficacy should be used as a tool to guide personal and professional choice, and not be used as a rule to exclude certain types of treatment because of cost or convenience.
So please use the league table as you would a walking stick. Use it to help you and not to beat others.
Information provided
The information here has been collected from reviews. The table gives all the information we thought relevant, including the number of trials and patients from which the table is drawn. The two figures show the results graphically for NNTs, and for the proportion of patients with pain free at two hours. In the figures subcutaneous treatments are in red , and intranasal treatments in green ; all other treatments are oral, with oral treatments that are not triptans in yellow.
Table of relative efficacy of acute migraine treatments - pain free at two hours
|
Treatment |
Placebo |
||||||
|
Treatment |
Route |
Number of trials |
Number/Total |
Percent |
Number/Total |
Percent |
NNT (95% CI) |
| Sumatriptan 6 mg | Subcutaneous |
6 |
206/345 |
60 |
41/338 |
12 |
2.1 (1.9 to 2.4) |
| Rizatriptan 10 mg | Oral |
7 |
720/1783 |
40 |
83/987 |
8 |
3.1 (2.9 to 3.5) |
| Eletriptan 80 mg | Oral |
6 |
386/1221 |
32 |
38/779 |
5 |
3.7 (3.2 to 4.2) |
| Zolmitriptan 5 mg | Oral |
4 |
298/943 |
32 |
17/285 |
6 |
3.9 (3.4 to 4.6) |
| Eletriptan 40 mg | Oral |
6 |
335/1224 |
27 |
38/779 |
5 |
4.5 (3.9 to 5.1) |
| Sumatriptan 20 mg | Intranasal |
3 |
182/536 |
34 |
33/276 |
12 |
4.6 (3.6 to 6.1) |
| Sumatriptan 100 mg | Oral |
8 |
415/1461 |
28 |
53/744 |
7 |
4.7 (4.1 to 5.5) |
| Rizatriptan 5 mg | Oral |
4 |
284/933 |
30 |
65/713 |
9 |
4.7 (4.0 to 5.7) |
| Zolmitriptan 2.5 mg | Oral |
2 |
109/438 |
25 |
17/213 |
8 |
5.9 (4.5 to 8.7) |
| Sumatriptan 50 mg | Oral |
4 |
131/711 |
18 |
21/372 |
6 |
7.8 (6.1 to 11) |
| Aspirin 900 mg + metoclopramide 10 mg | Oral |
3 |
69/378 |
18 |
25/375 |
7 |
8.6 (6.2 to 14) |
| Naratriptan 2.5 mg | Oral |
1 |
44/213 |
21 |
9/107 |
8 |
8.2 (5.1 to 21) |
| Eletriptan 20 mg | Oral |
2 |
53/349 |
15 |
17/353 |
5 |
10 (7 to 17) |
Figure 1: Numbers needed to treat for two hour pain free for each treatment compared with placebo (bars are 95% confidence interval of the NNT)
|
Figure 2: Percentage of patients with two hour pain free for each treatment (bars are 95% confidence interval of the percentage) |
|