Next, we looked over some evidence on the use of magnesium for migraine headaches. The highest-quality randomized, placebo-controlled trial we know about is a study that assessed 2 g magnesium IV as an adjunct to 20 mg of IV metoclopramide.² That study did not bode well for magnesium. Some lower-quality studies conducted since then were a little more encouraging. One study found 1 g of IV magnesium was superior to placebo but only in the subset of patients who had migraines with aura (ie, resolving neurologic symptoms).³ Another study pitted 1 g IV magnesium against 8 mg IV dexamethasone or 10 mg of IV metoclopramide and found that magnesium reduced pain better than the other agents. While this all would seem to add up to modest support for magnesium in some patients, the studies were small and some had issues with methodology. That’s why the American Headache Society makes no recommendation on the use of magnesium for emergency management of headaches, but states that it may be of benefit in patients with migraine with aura.

Lastly, we reviewed evidence on the use of IV magnesium in the management of asthma in the emergency department. The most important insight to remember when evaluating a study of asthma (or any study, in all fairness) is to focus primarily on patient-centered outcomes. Measuring peak flow rates in patients with asthma may give us some sense of the clinical picture, but there are major problems with this tool. First, it is effort related. The outcome and accuracy of the measurement rely on the patient’s willingness and skill to perform the test. Sometimes peak flow measurements are falsely low because patients still feels poorly, and so they simply do not blow with enough force. What we care about are patient-centered outcomes, such as whether a patient required an admission to the hospital. In this regard, the use of IV magnesium early in the emergency department visit appears to have a consistent benefit in decreasing the number of patients who will end up needing admission to the inpatient units. This is true in both adults and children.^4,5 Magnesium seems to benefit those with more severe symptoms. The number of patients needed to treat to decrease an admission seems to range from two to six. Even using the conservative estimate of six is impressive. While the quality of the study designs was variable, the signal seems consistent.