Skip navigation

Paracetamol for acute migraine

Clinical bottom line:

Paracetamol 1000 mg was more effective than placebo in one trial with an NNT for at least 50% pain relief by 2 hours of 7.8 (4.8 to 21). The lower dose, 650 mg, was not effective. Compared with NSAIDs, three out of five trials showed that the NSAID was significantly better at reducing migraine pain than paracetamol 500 to 1000 mg. Overall, the trials were of poor methodological quality.

In the absence of a systematic review of paracetamol for the acute treatment of migraine, Bandolier pulled together the evidence. Paracetamol is often recommended as the first line of treatment for acute migraine attacks, and patients frequently self-medicate with over the counter drugs. It would be useful to have some idea of the relative efficacy of paracetamol.

Inclusion criteria were treatment of acute migraine with paracetamol by any route; randomised allocation to treatment groups; double-blind design; adult population and headache outcomes.


Comprehensive searches of the following databases were made: MEDLINE (1966 - July 2000), EMBASE (1980 - June 2000), Cochrane Library (Issue 3, 2000) and the Oxford Pain Relief Database (1950 - 1994). A series of free text searches were undertaken, using generic and trade names for paracetamol. There was no restriction to language. Retrieved reports were searched for additional trials. Neither individual authors nor pharmaceutical companies were contacted for unpublished data.


Eight randomised, double blind trials were found with paracetamol given at doses ranging from 500 to 1000 mg ( Table ) [1-8]. Seven were of oral paracetamol, one of intramuscular (IM) paracetamol. Two were placebo controlled and six used active control groups. One trial scored two, six scored three and one scored four out of a maximum of five using a scale that rates trial quality based on randomisation, blinding and withdrawals. However, overall the trials were of low methodological quality based on other aspects of trial design. The trials were all of a crossover design, and many of the authors failed to either report on the absence or presence of carryover effects, or describe a minimum period between attacks to avoid these effects. It was not clear if all patients had sufficient pain at baseline, which is of particular importance in the absence of a placebo group to demonstrate trial assay sensitivity. All of the studies treated multiple attacks, and there is a high proportion of withdrawals and dropouts in many of the trials, data for which were not included in the final analysis. This may bias the results in favour of the responders to treatment. Most of the studies pre-date International Headache Society (IHS) migraine diagnostic criteria and all use slightly different criteria for migraine diagnosis.

The trials were disparate with respect dosing regimes, patient diagnosis and outcome measures precluding pooling of data for quantitative analysis. One trial was excluded from the review due to invalid data (Million et al 1984). It clearly stated in the report that the two treatment groups were not similar at baseline with respect to pain intensity, therefore, the results 24 hours after treatment are not valid. A descriptive summary of the seven included trials is provided in the Table , and a summary of the overall results below. Trials for which dichotomous data could be extracted are shown in the Figure.



Placebo controlled trials

One trial found no difference between paracetamol 650 mg and placebo for reduction in migraine severity, and the other reported a significant difference between paracetamol 1000 mg and placebo. The number-needed-to-treat (NNT) for one patient to achieve at least 50% pain relief by 2 hours with paracetamol 1000 mg over placebo was 7.8 (4.8 to 21).

Active controlled trials

Based on a vote counting exercise, three of the active controlled trials showed significantly better pain relief with ibuprofen, tolfenamic acid and ketoprofen than paracetamol. Two trials showed no significant difference between paracetamol and tofenamic acid and paracetamol combined with domperidone for the main outcome of headache severity.

Adverse effects

No serious adverse effects were reported.


1 Diamond S. Treatment of migraine with isometheptene, acetaminophen, and dichloralphenazone combination: a double blind, crossover trial. Headache 1976; 15:282-7

2 Hoernecke R, Doenicke A. [Treatment of migraine attacks: combination of dihydroergotamine tartrate and paracetamol in comparison with individual drugs and placebo] Behandlung des Migraneanfalls: die Kombination Dihydroergotamintartrat und Paracetamol im Vergleich zu den Einzelsubstanzen und Placebo. Med Klin 1993; 88:642-8

3 Karabetsos A, Karachalios G, Bourlinou P, Reppa A, Koutri R, Fotiadou A. Ketoprofen versus paracetamol in the treatment of acute migraine. Headache 1997; 37:12-4

4 Larsen BH, Christiansen LV, Andersen B et al . Randomized double-blind comparison of tolfenamic acid and paracetamol in migraine. Acta Neurol Scand 1990; 81:464-7

5 MacGregor EA, Wilkinson M, Bancroft K. Domperidone plus paracetamol in the treatment of migraine. Cephalalgia 1993; 13:124-7

6 Norrelund N, Christiansen LV, Plantener S. [Tolfenamic acid versus paracetamol in migraine attacks. A double blind study in general practice] Tolfenamsyre versus paracetamol ved migraeneanfald. En dobbeltblind undersogelse i almen praksis. Ugeskr Laeger 1989; 151:2436-8

7 Pearce I, Frank GJ, Pearce JM. Ibuprofen compared with paracetamol in migraine. Practitioner 1983; 227:465-7 Peatfield RC, Petty RG, Rose FC. Double blind comparison of mefenamic acid and acetaminophen (paracetamol) in migraine. Cephalalgia 1983; 3:129-34

8 Million R, Finlay BR, Whittington JR. Clinical trial of flupirtine maleate in patients with migraine. Curr Med Res Op 1984 9: 204-212.