IV for migraine: what the evidence shows

When a headache needs urgent medical care

The UK National Health Service says an adult or a child with a headache should get emergency help if the headache started suddenly and is extremely painful, or if there was a head injury in the last 3 months, such as from a fall or accident.[8]

The UK National Health Service's headaches page lists these other signs that should prompt emergency help in an adult or a child with a headache: a seizure; numbness or weakness in the body or face; difficulty speaking, balancing, walking or remembering; drowsiness or confusion; loss of vision; a rash that does not fade when a glass is rolled over it; and a very high temperature, a stiff neck or bright lights that bother the person.[8]

The UK National Health Service says to get urgent medical advice for a headache with vision or eye problems, a headache triggered or made worse by coughing, sneezing, bending down or exercising, or a headache with vomiting; for a child's headache that is getting worse or wakes the child at night; for an adult's headache with jaw pain when eating or a sore or tender scalp; and when an adult or a child has any other symptoms that worry them.[8]

The American College of Emergency Physicians' 2019 clinical policy on acute headache is written for physicians working in emergency departments who evaluate adults with a sudden-onset nontraumatic headache and no focal findings on a neurologic exam, and it does not cover people with chronic headaches or pediatric, pregnant or trauma patients.[9] The American College of Emergency Physicians' 2019 clinical policy says that, although most patients with sudden-onset severe headache have benign causes, 'data suggest that between 10% and 15% have serious pathology', most commonly subarachnoid hemorrhage, a bleed in the space around the brain, from an intracranial aneurysm or arteriovenous malformation.[9] The American College of Emergency Physicians' 2019 clinical policy also states that 'additional clinical findings such as fever, severe back pain, or other factors may warrant further additional diagnostic testing'.[9]

The American College of Emergency Physicians' 2019 clinical policy states: 'Regardless of headache etiology, response to treatment should not be solely used to determine whether a cause is benign.'[9] In its discussion of pain medicines, the American College of Emergency Physicians' 2019 clinical policy also states: 'Regardless of the agent used to abort acute headache, pain relief should not be used as a determinant of seriousness of the underlying headache pathology.'[9]

The American College of Emergency Physicians funded its own 2019 clinical policy on acute headache.[9]

If you have severe symptoms, call 911 or seek emergency care. Report side effects to FDA MedWatch.

What studies of IV treatment for migraine found

The American Headache Society's 2025 guideline rates the medicines emergency clinicians give for migraine, which are given by vein, by injection or as nerve blocks; it is based on randomized trials in adults diagnosed with migraine who were treated in emergency departments.[10]

IV fluids

In the records we read from searches of PubMed, Europe PMC and ClinicalTrials.gov on October 1, 2026, we found three published randomized trials that enrolled adults and tested IV saline added to the medicines emergency clinicians give for migraine, in patients with migraine or benign headache in emergency departments: Jones 2019, Zitek 2020 and Turan 2026.

In Jones 2019, a single-center pilot trial in 49 adults with migraine in an emergency department, all of whom also received IV medicines emergency clinicians give for migraine, adding 1 liter of IV saline over 1 hour produced no statistically significant difference in pain improvement from 0 to 60 minutes compared with 10 mL of saline over the same hour.[1] The authors of Jones 2019 wrote that the pilot study 'does not exclude the possibility of a clinically important treatment effect', and the nurses who gave the infusion were not blinded.[1]

In Zitek 2020, a single-blinded randomized trial of 58 patients aged 10 to 65 with benign headache in one emergency department, all of whom also received the same IV medicines, mean pain reduction at 60 minutes differed by 0.4 mm between a group given up to 1 liter of IV saline and a group given 5 mL (95% confidence interval -16.5 to 17.3 mm; p = 0.96).[2] The authors of Zitek 2020 wrote that 'our study lacked statistical power to detect small but clinically significant differences'.[2]

In Turan 2026, a double-blind randomized trial in one emergency department, 125 adults with migraine were analyzed and all received the same injected migraine medicine; the estimated difference in pain reduction at 2 hours between 1,000 mL of IV saline and 10 mL was 10.0 mm on a 100 mm scale (95% confidence interval -2.0 to 20.0 mm).[3] The authors of Turan 2026 wrote that the saline 'did not produce a clear improvement in pain relief at 2 hours', and that lower rescue medicine use and a shorter emergency department stay in the saline group were secondary findings that 'may be influenced by unblinded administering staff and should be interpreted cautiously'.[3]

Each of the three trials found no statistically significant difference in its main pain measure between a group given IV saline in a larger volume and a group given a small control amount, and each was small.[1][2][3]

In Balbin 2016, a post hoc analysis of four emergency department trials in which all 570 adults with migraine were treated with the same IV medicine, the 112 patients who also received IV fluids at their physician's discretion did not improve more than the others; because fluids were not randomized, this is an observational comparison.[11]

In the records we read from searches of PubMed, Europe PMC and ClinicalTrials.gov run on October 1, 2026, we found no randomized trial of IV fluid alone, without the same medicines in both groups, in adults with migraine.

In the records we read from PubMed searches run on October 1, 2026, we found no systematic review or meta-analysis of IV fluids for migraine or headache.

The American Headache Society's 2025 guideline for adults with migraine in emergency departments rates IV normal saline level U, which its abstract gives as 'no recommendation', and says that additional evidence is needed.[10]

The labels for the IV fluid products in the FDA status table do not name headaches or migraine among their approved uses; five describe the product as a source of water and electrolytes (checked October 1, 2026).

IV magnesium

Studies of IV magnesium in adults with acute migraine or headache disagree: a 2014 meta-analysis failed to show a benefit and found more side effects, a 2016 meta-analysis found a benefit, a 2019 systematic review could not draw a firm conclusion, and a 2026 trial found more frequent success with differences its authors judged below the threshold for clinical importance.[4][5][6][7]

The abstracts of the 2014 and 2016 meta-analyses do not state where the people were treated; the 2019 review covered emergency departments and outpatient acute care treatment centers, and the 2021 and 2026 trials took place in emergency departments.[4][5][6][7][12]

A 2014 meta-analysis (Choi 2014) of five randomized, double-blind trials in 295 adults with acute migraine found that the percentage of patients whose headache was relieved 30 minutes after treatment was 7 percentage points lower with IV magnesium than in the control groups (95% confidence interval from 23 points lower to 9 points higher).[4] The authors of Choi 2014 wrote that the analysis 'failed to demonstrate a beneficial effect of intravenous magnesium', and the abstract does not say where the patients were treated or whether the controls received placebo or another medicine.[4]

A 2015 Canadian Headache Society systematic review of treatments for migraine pain in emergency settings weakly recommended against magnesium sulfate, based on moderate-quality evidence.[13]

A 2016 meta-analysis (Chiu 2016) pooled 11 randomized studies of IV magnesium for acute migraine in 948 participants and reported that IV magnesium significantly relieved acute migraine within 15 to 45 minutes, at 120 minutes and at 24 hours; the abstract does not say where the people were treated, and the authors note that some included studies did not use adequate randomization methods.[5]

A 2019 systematic review (Miller 2019) of 7 randomized trials in 545 adults with migraine (6 trials) or benign non-traumatic headache (1 trial), treated in an emergency department or an outpatient acute care treatment center, did not pool the results because of clinical differences between the studies, and its authors wrote that they 'cannot draw a firm conclusion' on the efficacy or benefit of IV magnesium sulfate.[6] The authors of Miller 2019 added that 'the existing evidence indicates potential benefits in pain control beyond 1 hour, aura duration, and need for rescue analgesia'.[6]

In Kandil 2021, a randomized, double-blind trial in one emergency department in which 157 adults with migraine received IV magnesium (61 adults) or one of two other medicines emergency clinicians give for migraine (44 and 52 adults), the authors found no significant difference in the change in pain score at 30 minutes between IV magnesium and the other two medicines.[12] Kandil 2021 had no placebo group, and its authors list early termination because of COVID-19, unequal group sizes and low statistical power among its limitations, so the trial cannot show that the treatments are equivalent.[12] The authors of Kandil 2021 state that the research received no specific grant from funding agencies and that they have no conflicts of interest.[12]

In Messous 2026, a randomized, double-blind, placebo-controlled trial of 1,028 adults with acute nontraumatic headache (not only migraine) in an emergency department, all of whom first received an oral pain medicine, success, defined as a drop of at least 30% in pain score at 30 minutes, was 13.8 percentage points more frequent with IV magnesium sulfate than with placebo (95% confidence interval 8 to 19).[7] The authors of Messous 2026 reported that all timed pain-score differences between the two groups were below the 1.3-point threshold they call accepted for clinical importance, and that rescue pain medicine was needed less often with IV magnesium (difference -8.2 percentage points, 95% confidence interval -12 to -4.3).[7]

Xu 2019 was a retrospective review of 234 migraine patients who received IV magnesium at one university headache clinic, with no comparison group.[14] Xu 2019 reports that many of its patients also received injected medicines as needed, and its authors describe the analysis as exploratory.[14]

The magnesium sulfate injection products whose labels we reviewed are FDA-approved for treating magnesium deficiency or for preventing seizures in preeclampsia and treating them in eclampsia, not for headaches or migraine. Using magnesium sulfate injection for headaches or migraine is off-label. FDA has not reviewed whether it is safe and effective for that purpose.

B vitamins and other reasons people ask

In the 38 records we read from a PubMed search run on October 1, 2026 for B vitamins given by injection, infusion or IV together with migraine or headache terms, we found no randomized trial of them for migraine or headache.

In the first 100 of the 304 Europe PMC records returned by a search run on October 1, 2026, we found no study showing that IV rehydration relieves a headache attributed to dehydration.

Evidence summary by use

Evidence levels in this table follow our grading key. Established means consistent results from randomized trials or a systematic review of them, or that a guideline or expert body says something. Emerging means some controlled human evidence that falls short of that, such as small or short trials or mixed results. Experimental means no controlled human evidence, only case reports, uncontrolled series, laboratory findings or an author's experience. Evidence against means a controlled trial tested the claim as its main result in the people the cell names and did not find the effect, and the cell names no trial that found that effect. 'No support found' means the searches named in that row found no study that supports the claim.

In this table, Not published means we found no published data for that cell.

Emergency department studies and IV lounge or med spa settings

In the records we read from searches of PubMed, Europe PMC and ClinicalTrials.gov run on October 1, 2026, we found no study of IV treatment for headache or migraine in an IV lounge, med spa, mobile IV or wellness-clinic setting, with or without a comparison group; this does not cover hospital-run or specialist headache infusion clinics, which are studied (for example, Xu 2019).

In the records we read from searches of PubMed, Europe PMC and ClinicalTrials.gov run on October 1, 2026, we found no controlled trial of a combination of IV fluids, magnesium and B vitamins in any setting.

In a 2025 JAMA Internal Medicine study, secret shoppers called 87 randomly selected IV hydration spas in the US, and 75 of the 87 (86.2%) recommended specific therapies for the headache and cold symptoms the callers described.[15] In the 2025 JAMA Internal Medicine study of IV hydration spas, 24 of the 87 facilities (27.6%) required consultation with a licensed medical professional before treatment, and 21 of the 87 (24.4%) described potential risks.[15]

The 2025 JAMA Internal Medicine study of IV hydration spas also reviewed 255 spa websites; all included claims for beneficial uses, and only 2 (0.8%) cited sources for those claims.[15]

Limitations

The three IV saline trials described on this page included 49, 58 and 125 patients, and the 2021 trial that compared IV magnesium with two other medicines included 157 adults. Kandil 2021 excluded adults who were pregnant or had kidney impairment, and Turan 2026 excluded adults with dehydration or recent IV fluid use, so neither trial describes those groups.[3][12]

The abstracts of Jones 2019, Turan 2026, Choi 2014, Chiu 2016, Miller 2019, Messous 2026 and Xu 2019 do not report who funded the studies.[1][3][4][5][6][7][14]

Are IV fluids or IV magnesium FDA-approved for migraine?

FDA has approved specific IV fluid products, including 0.9% sodium chloride and lactated Ringer's injections, as a source of water and electrolytes (checked October 1, 2026). Specific magnesium sulfate injection products are FDA-approved prescription drugs for treating magnesium deficiency or for preventing seizures in preeclampsia and treating them in eclampsia, according to the labels we reviewed (checked October 14, 2026).

The FDA status table lists IV fluid and magnesium sulfate injection products whose labels we reviewed, with how each is given and the uses in its prescribing information.

Each IV fluid product in the FDA status table is available only with a prescription (checked October 1, 2026).

Each magnesium sulfate injection product in the FDA status table is available only with a prescription (checked October 14, 2026).

Regulatory status checked October 1, 2026. Sources: dailymed.nlm.nih.gov/…/lookup.cfm; dailymed.nlm.nih.gov/…/lookup.cfm; dailymed.nlm.nih.gov/…/lookup.cfm; dailymed.nlm.nih.gov/…/lookup.cfm; dailymed.nlm.nih.gov/…/lookup.cfm; dailymed.nlm.nih.gov/…/lookup.cfm; www.accessdata.fda.gov/…/016677s185,016682s124,018016s071lbl.pdf; www.accessdata.fda.gov/…/016679s122,016682s127,019367s042lbl.pdf; www.accessdata.fda.gov/…/019632s067,019634s058,020002s037lbl.pdf

Regulatory status checked October 14, 2026. Sources: dailymed.nlm.nih.gov/…/lookup.cfm; dailymed.nlm.nih.gov/…/lookup.cfm

Side effects and risks

In the 2014 meta-analysis of five trials (Choi 2014), side effects or adverse events were 37 percentage points more frequent in the groups given IV magnesium than in the control groups (pooled risk difference 0.37, 95% confidence interval 0.06 to 0.68).[4]

In Messous 2026, adverse events were reported more often with IV magnesium than with placebo (difference 4.3 percentage points, 95% confidence interval 0.1 to 8.4), which the authors describe as a modest increase in mild side effects.[7] In Kandil 2021, 5% of the 61 adults given IV magnesium reported adverse events.[12]

In Turan 2026, no serious adverse events occurred among the 125 adults analyzed, who had been screened to exclude dehydration and recent IV fluid use.[3] The abstracts of Jones 2019 and Zitek 2020 do not report adverse events.[1][2]

A 2025 case report (Gorla 2025) describes a woman with one kidney who became short of breath within hours of an elective IV infusion that began with a liter of sterile water, went into cardiac arrest after reaching an emergency department, and then needed dialysis for kidney failure and surgery to remove part of her bowel.[16] The authors of the Gorla 2025 report name suspected destruction of red blood cells, a high dose of one ingredient in the infusion and her single kidney as contributing factors, and state that they received no outside funding.[16] A case report shows that a harm happened after a treatment, not how often it happens or that the treatment caused it. Adverse-event figures from trials describe only the people and settings in those trials.

The American College of Emergency Physicians' 2019 clinical policy states: 'Regardless of headache etiology, response to treatment should not be solely used to determine whether a cause is benign.'[9] In its discussion of pain medicines, the American College of Emergency Physicians' 2019 clinical policy also states: 'Regardless of the agent used to abort acute headache, pain relief should not be used as a determinant of seriousness of the underlying headache pathology.'[9]

Talk to a licensed clinician about whether IV treatment for migraine is appropriate for you.

If you have severe symptoms, call 911 or seek emergency care. Report side effects to FDA MedWatch.

Alternatives

The American Headache Society's 2025 guideline rates the medicines emergency clinicians give for migraine, which are given by vein, by injection or as nerve blocks; it is based on randomized trials in adults diagnosed with migraine who were treated in emergency departments.[10]

Provider directory

Find a Provider

A licensed clinician decides whether treatment is appropriate. A prescription is not guaranteed.