Our methodology involved the assessment of approximate previous duration of pain (T-0) and of pain and associated symptom intensity at T0, 30 and 60 minutes and within 24 hours after drug administration. The fact that we limited our first evaluation of efficacy to only one hour may be criticized. However, this option was dictated by the needs of the health service where the study was conducted. Since this was a public health unit in a developing country, with overcrowding, excess demand and all other problems presented in these places, we though it would be inappropriate to maintain a patient under observation for more than one hour unless his clinical needs required it. We tried to minimize this limitation by re-contacting the patients 24 hours after drug administration. This procedure, in spite of minimizing the above drawbacks, has its own limitations. After 24 hours there is an increase in subjectivity with a proportional decrease in the precision of information, a fact that might generate possible deviations and distortions. Although the guidelines of the International Headache Society recommend an evaluation of two hours20 , we consider our procedure to be justified. We also guarantee that, although the methodology was not ideal, the study was conducted under the closest possible conditions of real utilization of these medications in the most part of the emergency rooms. We also tried to minimize another bias, excluding patients with history of migraine. Treating the patients 4.1 to 4.2 hours, in mean, after the beginning of the pain, almost exclude the possibility of having treated the beginning.