Practice advisory for intravenous management of headache disorders in hospitalized patients: a review of the evidence and consensus recommendations

Abstract

Introduction

Patients hospitalized for headache treatment pose unique challenges to the healthcare system. Currently, there is a lack of evidence-based guidance on management. This practice advisory aims to fill this critical gap by systematically reviewing the existing literature and providing comprehensive, evidence-based recommendations for managing headache patients during hospitalization.

Methods

In February 2023, the American Society of Regional Anesthesia and Pain Medicine approved this practice advisory proposal. The steering committee selected committee members based on clinical and research expertise in the field of headache medicine. Nine questions were formulated by the committee, and each question was assigned to a group composed of 3–4 members. A systematic literature search for each question was performed in Medline, Embase, Cochrane Database of Systematic Reviews and Web of Science on June 21, 2023. The results from each search were imported into separate Covidence projects for screening, data extraction, and risk of bias assessment. Additionally, relevant systematic reviews (SR) were screened. Each group submitted a structured narrative review along with statements and recommendations based on the US Preventive Services Task Force (USPSTF) format for grading of evidence. While the USPSTF framework was used, including the language in the recommendations, the formal USPSTF methodology, including the SR with meta-analysis and summary tables with forest plots, was not followed because of low overall evidence quality. The interim draft was shared electronically with each collaborator, who was requested to vote anonymously using two rounds of the modified Delphi approach. A consensus recommendation required >75% agreement.

Results

The panel generated 12 statements and 17 recommendations, along with their strength and certainty of evidence. Following two rounds of Delphi voting, a high consensus was achieved for all statements and recommendations. Most statements received a low-to-moderate level of certainty, and all but one recommendation received grade B or C, which was consistent with the lack of randomized controlled trials supporting most of the drugs in this document.

Conclusions

This evidence-based practice advisory provides a foundational step toward standardizing inpatient headache care and highlights existing gaps in the literature that should be addressed through rigorous prospective randomized studies.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Management of hospitalized patients with refractory headache is challenging, and current guidelines are primarily focused on the outpatient population, leaving a knowledge gap when it comes to evidence-based management strategies.

WHAT THIS STUDY ADDS

• This practice advisory provides an up-to-date, evidence-based review of the most commonly used intravenous medications for inpatient refractory headache treatment.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE, OR POLICY

• The assessments and recommendations contained in this document demonstrate that there are multiple potentially effective treatments for refractory headache, and yet there is a great need for additional randomized, double-blind trials that evaluate many of these promising drugs.

Introduction

The prevalence and burden of migraine and other primary headaches in the US adult population are significant. Headache is the fifth most common reason for emergency department (ED) visits, accounting for approximately 3% of all annual ED visits.¹ Approximately 0.2% of patients fail to respond to initial ED treatment and require hospitalization to effectively manage pain.² Patients hospitalized for headache treatment have an average length of hospitalization of 5 days±1.1 SD (range 2–9), resulting in a total annual cost of US$400 million.^{2 3}

Outpatient headache treatment typically includes mostly oral medications, while intravenous infusions are more commonly used in emergency and inpatient settings.⁴ Intravenous administration offers many advantages, including faster onset, improved bioavailability, and potentially greater effectiveness.

Despite extensive literature on pharmacological treatments for primary headache disorders, there is little guidance on the management of the unique challenges of hospitalized patients with headache.⁴,⁶ The refractory nature and complexity of this pathology, coupled with its socioeconomic impact, underscore the urgent need for targeted research and evidence-based guidance. Additionally, many healthcare providers across specialties report limited knowledge regarding effective inpatient headache management.^{7 8} This practice advisory aims to fill this critical gap by providing evidence-based recommendations for managing headache patients during hospitalization.

This document does not establish a standard of care or replace clinical judgment. It does not intend to limit or deny care, affect the rights of patients or providers, or replace clinical judgment. In the imperfect setting of heterogeneous data, limited data, controversial topics, and bias inherent to expert opinion, compliance with the recommendations may not result in improved outcomes compared with personalized medicine.

Methods

The start of this project began with the submission of a proposal to the American Society of Regional Anesthesia and Pain Medicine (ASRA PM) in February 2023. The proposal was reviewed and approved by the Board of Directors. The steering committee (YH, AF, and SN) selected diverse committee members based on clinical and research expertise in the field of headache management (neurologists, pain physicians, anesthesiologists, pain/headache fellowship directors, and a patient representative). Based on the input from all committee members during a virtual meeting on May 30, 2023, the committee selected nine questions that were confirmed. The steering committee assigned 3–4 members to each question based on their expertise, and one person from each group was designated as a group leader. A health sciences librarian conducted a systematic literature search for each question in MEDLINE, Embase, Cochrane Database of Systematic Reviews (SR), and Web of Science on June 21, 2023. The search block for the ‘infusion’ concept and ‘primary headache’, common to all searches, was combined with each specific ‘drug’ (online supplemental appendix A). All types of studies were accepted, except for research in progress, abstracts, and book chapters. A publication date limit of 1999 onwards wasset by the committee, as it was deemed the start of increasing recognition and research of headache. The search was further limited to English language and human subjects of age 18 and above. While studies that included hospitalized patients were prioritized, studies that included patients in the ED were also retrieved and permitted in the consensus discussions if there was a lack of inpatient studies. Questions for which ED studies were permitted are clearly indicated within the responses. Each search result was imported into its own Covidence project for the purposes of deduplication and screening. Each group member independently conducted title, abstract, and full-text screening of the included studies, followed by data extraction in Word tables. Additionally, each group was asked to screen the references of SR relevant to their topic to ensure no critical studies were overlooked. Risk of bias assessment and strength of evidence evaluation across domains were performed (online supplemental appendix B). Each group submitted their findings in the form of a structured narrative review. Recommendations for each topic were formed using the framework of the US Preventive Services Task Force (USPSTF) grading of evidence guidelines.⁹ The grade of evidence was classified as A, B, C, D, or I and was based on group consensus rather than predetermined USPSTF rules because of the flexibility needed in light of the overall low quality of evidence (table 1). Statements were assigned a level of certainty (table 2). Each group’s submission was reviewed and revised by the steering committee (YH, AF, and SN). The interim draft was shared with all committee members electronically, and they were requested to vote anonymously on each recommendation using a modified Delphi approach.¹⁰ We used Microsoft Excel (Microsoft, Redmond, Washington, USA) for two rounds of electronic voting. Each member was requested to ‘agree’, ‘disagree’ or ‘abstain’ from each statement or recommendation.¹¹ Collaborators were also invited to provide narrative comments and reasons for their decision if they disagreed or abstained.

Table 1. US Preventive Services Task Force (USPSTF) grade definitions^*.

Grade definition suggestions for practice
A	The USPSTF recommends the service. There is high certainty that the net benefit is substantial.	Offer or provide this service.
B	The USPSTF recommends the service. There is high certainty that the net benefit is moderate, or there is moderate certainty that the net benefit is moderate to substantial.	Offer or provide this service.
C	The USPSTF recommends selectively offering or providing this service to individual patients based on professional judgment and patient preferences. There is at least moderate certainty that the net benefit is small.	Offer or provide this service for selected patients depending on individual circumstances.
D	The USPSTF recommends against the service. There is moderate or high certainty that the service has no net benefit or that the harms outweigh the benefits.	Discourage the use of this service.
I	The USPSTF concludes that the current evidence is insufficient to assess the balance of benefits and harms of the service. Evidence is lacking, of poor quality, or conflicting, and the balance of benefits and harms cannot be determined.	Read the clinical considerations section of USPSTF Recommendation Statement. If the service is offered, patients should understand the uncertainty about the balance of benefits and harms.

^*https://www.uspreventiveservicestaskforce.org/uspstf/about-uspstf/methods-and-processes/grade-definitions.

Table 2. Levels of certainty regarding net benefit^*.

Level of description certainty
High	The available evidence usually includes consistent results from well-designed, well-conducted studies in representative primary care populations. These studies assess the effects of the preventive service on health outcomes. This conclusion is therefore unlikely to be strongly affected by the results of future studies.
Moderate	The available evidence is sufficient to determine the effects of the preventive service on health outcomes, but confidence in the estimate is constrained by such factors as: The number, size, or quality of individual studies. Inconsistency of findings across individual studies. Limited generalizability of findings to routine primary care practice. Lack of coherence in the chain of evidence. As more information becomes available, the magnitude or direction of the observed effect could change, and this change may be large enough to alter the conclusion.
Low	The available evidence is insufficient to assess effects on health outcomes. Evidence is insufficient because of: The limited number or size of studies. Important flaws in study design or methods. Inconsistency of findings across individual studies. Gaps in the chain of evidence. Findings not generalizable to routine primary care practice. Lack of information on important health outcomes. More information may allow estimation of effects on health outcomes.

^*The USPSTF defines certainty as ‘likelihood that the USPSTF assessment of the net benefit of a preventive service is correct’. The net benefit is defined as benefit minus harm of the preventive service as implemented in a general, primary care population. The USPSTF assigns a certainty level based on the nature of the overall evidence available to assess the net benefit of a preventive service.

USPSTF, US Preventive Services Task Force.

After the first round of voting, all statements and recommendations obtained >80% consensus. Recommendations with 100% agreement were accepted, as well as those recommendations with >85% agreement that only needed grammatical improvement. All other statements and recommendations were revised by the steering committee considering contributors’ comments and sent back for a second voting round with the aim to increase consensus in the second round. In the second round, statements and recommendations with a predefined >75% agreement were accepted. Both USPSTF grading and >75% agreement threshold have been previously used in ASRA PM consensus guidelines.^{12 13} Language was further refined considering contributors’ comments. The subsequent draft was shared with all the collaborators for final approval.

Results

The results retrieved from our systematic literature search for each drug are shown in Online supplemental appendix.

After two rounds of voting, all statements and recommendations were approved by >85% consensus. The final manuscript was approved by all committee members and ASRA PM Board of Directors. The statements and recommendations are summarized in table 3.

Table 3. Summary of statements and recommendations.

Drug	Statement	Recommendation
DHE	IV DHE decreases headache frequency and severity in inpatients with CM, status migrainosus cluster headache and/or MOH. Level of certainty: moderate	Preinfusion ECG and hemodynamic monitoring before, during and after the infusion. Grade B Consider IV DHE in select inpatients without significant vascular disease. Grade C
Magnesium	IV Magnesium sulfate may improve migraine severity and associated symptoms with better tolerability than sumatriptan and prochlorperazine. Level of certainty: low	Administer 1–2 g of IV magnesium sulfate infused over 15 min in inpatients with migraine. Grade: I Check pre-infusion magnesium levels and blood pressure, to rule out hypotension. Grade C
Valproic Acid	IV VPA may be beneficial for inpatients with intractable migraine. Level of certainty: low	Consider IV VPA for intractable migraine in select inpatients. Grade C
Ketorolac	IV ketorolac may be an effective and safe treatment option for managing acute migraine or non-NSAID MOH in the inpatient setting. Level of certainty: Low	Consider IV ketorolac for select inpatients with acute migraine or non-NSAID MOH. Grade C.
Steroids	IV steroids may be useful in the inpatient treatment of status migrainosus, MOH, and trigeminal autonomic cephalalgias (TAC). Level of certainty: Moderate	Consider IV dexamethasone or methylprednisolone for 1–3 days in selected inpatients with refractory CM with MOH, status migrainosus, and TACs including CH and HC. Grade C Use GI prophylaxis. Steroids should be used with caution in patients with gastritis. Patients with diabetes mellitus should have blood glucose levels monitored and appropriately treated. Concurrent use of NSAIDs with steroids should be avoided. Grade B
Levetiracetam	IV levetiracetam may be effective and safe to use in the acute management of intractable migraine or status migrainosus under specific circumstances. Level of certainty: Low	Consider IV levetiracetam for inpatient management of acute migraine, when other therapies have been exhausted. Grade C
Dopamine antagonists	IV dopamine antagonists may be effective in the acute treatment of intractable headaches in the inpatient setting. Level of certainty: moderate Side-effects of IV dopamine antagonists can be minimized with slow infusion. Level of certainty: moderate	IV dopamine antagonists should be offered in the acute treatment of inpatients with intractable headaches. Grade B. Home medications should be reviewed for drug interactions resulting in QTc prolongation. Obtain an ECG before using dopamine antagonists. Grade C Slow rate of infusion over the course of 15 min to minimize the extrapyramidal symptoms. Grade B
Lidocaine	IV lidocaine infusion may offer benefit for inpatients with TACs and intractable migraine and may be associated with mild AEs. Level of certainty: low	Duration of IV lidocaine should not routinely exceed 5 days to reduce risks associated with accumulation of a toxic metabolite. If prolonged infusion is warranted, then daily measurement of serum lidocaine levels may be considered. Grade C
Ketamine	Ketamine infusions may provide benefit in the acute treatment of intractable headache. Level of certainty: low Adverse effects can occur even at subanesthetic doses. Level of certainty: high The incidence and severity of psychotomimetic and sympathomimetic effects from ketamine may be reduced by administration of clonidine or benzodiazepines. Level of certainty: moderate	Consider Ketamine infusions in the acute inpatient treatment of refractory headache. Grade C. Ketamine infusion rates should not exceed 1 mg/kg/hour outside of intensive care unit settings. Grade B. Continuous respiratory and hemodynamic monitoring while receiving ketamine infusions. Grade B. Consider administering clonidine and/or a low-dose benzodiazepine to mitigate psychotomimetic and sympathomimetic adverse effects. Grade C

AEs, adverse events; CM, chronic migraine; DHE, dihydroergotamine; HC, hemicrania continua; IV, intravenous; MOH, medication overuse headache; NSAID, non-steroidal anti-inflammatory drug; VPA, valproate.

What is the evidence for the use of intravenous dihydroergotamine for inpatient headache management?

Mechanism of action

Dihydroergotamine (DHE) is a semisynthetic derivative of ergotamine, synthesized by Stoll and Hofmann in 1943.^{1 14} It has an affinity for serotonin, dopamine, and adrenergic receptors, with its antimigraine effect thought to be exerted through 5HT1B/5HT1D receptor activity.^{15 16} In addition, DHE seems to inhibit trigeminal pain processing in the brainstem and neurogenic inflammation in the trigeminovascular system.^{17 18} DHE is frequently used by headache practitioners due to its higher bioavailability and more favorable side effect profile in comparison to its predecessor (ergotamine) with less arterial vasoconstriction, lower likelihood of medication overuse headache (MOH), and lack of retroperitoneal fibrosis.¹⁹

Study findings

Our systematic literature search yielded 15 retrospective observational studies (n=1213) on the use of intravenous DHE infusions in the setting of inpatient headache management. All studies were open-label studies of variable quality. No ED studies were included. The target population was generally primary headache syndromes with possible MOH. There was significant heterogeneity across studies in terms of headache type. Chronic daily headache (CDH) comprised the largest group with over 500 patients studied.²⁰,²² CDH, including chronic migraine (CM), hemicrania continua (HC), and new daily persistent headache (NDPH), is currently no longer used terminology for these headache disorders.

The overwhelming headache subtype was CM, comprised 269 patients.²²³,²⁵ Older studies used evolving terminology including transformed migraine (n=37), drug rebound headache (n=71), and chronic intractable headache.^{22 26 27} Other diagnoses included in studies were NDPH and cluster headache (CH), both episodic and chronic,^{2 20 25 26 28} and one study investigating short-lasting unilateral neuralgiform headache with conjunctival injection and tearing (SUNCT) and short-lasting unilateral neuralgiform headache with autonomic symptoms (SUNA) in patients with coexisting chronic cluster or CM and tension-type headache.²⁹ Across studies, heterogeneity in terms of duration of pain was noted, ranging from months to years.

Repetitive intravenous DHE was first evaluated in a study by Raskin.²⁷ 55 patients with chronic intractable headache were treated with intravenous DHE and metoclopramide every 8 hours, whereas an age-matched and sex-matched control group received intravenous diazepam every 8 hours. 49 of the 55 DHE patients became headache-free within 48 hours, and 39 of them had sustained benefit at mean follow-up time of 16 months. In contrast, in the intravenous diazepam-treated group, only seven patients became headache-free within 3–6 days. The main critique of this study is that the outcomes are not representative of intravenous DHE use alone. All DHE patients were given metoclopramide, which has also been shown to be effective for migraine. Entry into the study required cessation of overused medications, and patients were placed on a regimen of propranolol and ergotamine in addition to DHE suppositories at posthospital discharge.³⁰,³²

Subsequent case series have concluded that intravenous DHE is effective for CM, CDH, and CH. Most studies looked at pain relief during the allotted time of hospitalization, typically at 2–3 days.^{2224 27},³⁰ Other studies looked at headache improvement during the hospitalization without a fixed endpoint.^{2 20 25} Five studies²¹²⁵,²⁸ offered long-term follow-up, and two studies^{21 25} examined functional metrics suggesting improvement in work absences, activity levels, and quality of life. It is important to note, however, that these studies were not a clear comparison of DHE to placebo. Rather, these studies investigated a regimen that involved cessation of previous medications that could trigger a rebound headache, commonly referred to as MOH today.^{21 24 26 27}

Concurrent medications during the intravenous treatment included metoclopramide and often ended with the transition onto a new or different preventive medication.^{2 20 21 25 27} Results from NDPH suggested a lack of benefit²⁵ or less robust results as compared with migraine,² although the sample size was much smaller. DHE was ineffective for SUNCT and SUNA and made the condition worse in some patients.²⁹

Both continuous and repetitive intravenous DHE dosing were used in the included studies, and both were shown to be effective.²² The continuous regimen is 3 mg of DHE in 1 L of normal saline administered at 42 mL/hour in an intravenous infusion pump over 24 hours, with 10 mg of intravenous metoclopramide given every 8 hours as needed. The dose can be lowered for problematic side effects and is decreased as the headache subsides. The dosing of repetitive DHE did differ in the literature from the standard ‘Raskin Protocol’ (0.5 mg every 8 hours with 10 mg of intravenous metoclopramide given pro re nata prior to each DHE infusion). This protocol was initially intended to run for seven doses (including an initial dose that is broken up into two parts for assessment of tolerability).²⁷ However, this has gradually changed over the years and evolved into 9 doses,²³ or 10 doses,²⁹ or until headache improvement with taper when headache is abated.^{2 20 25 28} Interestingly, a study by Pringsheim and Howse dosed DHE 4 times in 24 hours for an unclear duration.²¹

Adverse events

The most common side effects in the included studies were nausea and vomiting. Other reported side effects included leg cramps and diarrhea.²²⁰,^{23 27 28} Transient headache exacerbation was also reported with DHE infusions.³³ However, this is not a reason to stop infusions, as a 2016 study showed that headache exacerbation was not an independent predictor of treatment outcome via a multivariate logistic regression model.³⁴ Vasospastic adverse events (AEs) are rare but possible; hence, DHE is contraindicated in patients with cardiac, cerebral or peripheral vascular disease or uncontrolled hypertension. A low threshold should be used when considering screening for catheter-associated venous thrombosis as DHE may increase its risk.³⁵ DHE is also contraindicated in pregnancy, hemiplegic migraine, and migraine with brain stem aura.

In summary, based on expert opinion, it appears that intravenous DHE given in hospital, either via repetitive or continuous dosing, is relatively safe and tolerated. DHE frequently improves headache frequency and severity, in conjunction with a protocol to minimize MOH and implement migraine or CH prevention. Overall, the committee would recommend offering it to a select patient population depending on the individual circumstances.

Statement

• Intravenous DHE decreases headache frequency and severity in inpatients with CM, status migrainosus CH and/or MOH. Level of certainty: moderate

Recommendations

• Based on vasospastic AEs, the committee recommends preinfusion ECG and hemodynamic monitoring (blood pressure and heart rate) before, during, and after the infusion. Grade B.

• The committee recommends offering intravenous DHE in select inpatients without significant vascular disease being treated for status migrainosus. Grade C.

What is the evidence for the use of intravenous magnesium sulfate for inpatient headache management?

Mechanism of action

Magnesium has multiple roles in the central and peripheral nervous systems, including reducing cortical hyperexcitability, partly due to modulating N-methyl-D-aspartate (NMDA) sensitivity, modulating neurotransmission, and regulating vascular tone, all involved in migraine pathophysiology.³⁶ Patients with migraine demonstrate lower plasma and brain intracellular magnesium levels compared with controls.^{37 38}

Study findings

At the time of writing, no randomized controlled trials (RCTs) or observational studies assessed intravenous magnesium for the treatment of patients admitted with headache. Studies that included ED patients were included because of a lack of inpatient studies.

One gram of intravenous magnesium sulfate administered over 15 min was superior to placebo in migraine pain relief, pain freedom, and freedom from other migraine-associated symptoms, including photophobia, phonophobia, and nausea.³⁹ Magnesium sulfate infusion was superior to dexamethasone/metoclopramide in pain severity reduction at 20 min, 1 hour, and 2 hours postinfusion.⁴⁰ Two grams of intravenous magnesium sulfate demonstrated similar effectiveness to sumatriptan, metoclopramide, and prochlorperazine.^{41 42}

Patients with low ionized magnesium serum levels (<0.54 mmol/L) demonstrated longer-lasting migraine pain relief than patients with normal levels.⁴³ In a few studies, only patients with migraine with aura showed a significantly greater reduction in pain and other migraine-associated symptoms with magnesium sulfate compared with controls.³⁶⁴⁴,⁴⁶ Another double-blind RCT of intravenous magnesium sulfate versus placebo (normal saline) revealed significant improvement in migraine pain and nausea in patients with migraine with aura. Pain improvement (although not statistically significant) was found among patients with migraine without aura, with a significant improvement in photo and phonophobia.⁴⁷ Two case reports demonstrated clinical improvement within 24 hours in patients with prolonged status migrainous aura/hemiplegic migraine with prochlorperazine and 1 g of intravenous magnesium sulfate given every 12 hours.⁴⁸

The most common dose was 1–2 g of intravenous magnesium sulfate infused over 15 min. There are conflicting data regarding adjuvant therapy, with a study demonstrating the added benefit of combining magnesium sulfate with metoclopramide.⁴⁹ In contrast, another study suggested that adding magnesium sulfate attenuated the effectiveness of metoclopramide.⁴⁵ Other studies have combined magnesium sulfate with 10 mg of prochlorperazine and meperidine as a rescue medication.^{36 48}

Adverse events

Intravenous magnesium sulfate was better tolerated than sumatriptan and prochlorperazine, with the most common side effects including sensation of warmth at the intravenous infusion site, flushing, mild burning sensation in the face and neck, and asymptomatic drop in systolic blood pressure.^{41 42} Repeated doses can be administered daily. Hemodynamic monitoring is recommended due to the potential for hypotension with rapid infusion due to smooth muscle relaxation. Magnesium should be avoided in patients with neuromuscular disease, heart block, or previous hypersensitivity. We recommend serum levels be tracked in these patients to avoid hypermagnesemia.

In summary, we extrapolated our recommendations primarily from studies that included ED patients. Based on evidence from ED literature, the committee recommends 1–2 g of intravenous magnesium sulfate infused over 15 min in inpatients with migraine (with or without aura), without hypermagnesemia and without a history of hypersensitivity, neuromuscular disease, or heart block. It is recommended to perform hemodynamic monitoring and to check magnesium levels prior to daily infusions.

Statement

• Intravenous magnesium sulfate may improve migraine severity and associated symptoms in inpatients, with better tolerability than sumatriptan and prochlorperazine. Level of certainty: low

Recommendations

• The committee recommends using 1–2 g of intravenous magnesium sulfate infused over 15 min in inpatients with migraine (with or without aura). Grade: I.

• The committee recommends checking preinfusion magnesium levels and blood pressure to rule out hypotension. Grade C.

What is the evidence for the use of intravenous valproate sodium for inpatient headache management?

Mechanism of action

Valproate sodium (VPA) is an intravenous injection approved by the US Food and Drug Administration for seizure management. In contrast, divalproex sodium is an oral formula indicated for managing not only seizures, but also bipolar disorder and migraine. VPA has multiple mechanisms of action, including potentiation of γ-aminobutyric acid and blockade of sodium channels, NMDA-mediated excitation, and T-type calcium channels.

Importantly, VPA use is contraindicated in several conditions, including migraine prevention during pregnancy due to the risk of neural tube defects.⁵⁰

Study findings

The American Headache Society (AHS) endorsed oral divalproex sodium use with established efficacy for migraine prevention,⁵¹ whereas intravenous VPA may be offered to adults with acute migraine in the ED (Level C).⁵² A recent meta-analysis on the single use of intravenous VPA in the ED, including 7 double-blinded RCTs of 682 patients, noted inferiority in effectiveness relative to other treatments (eg, metoclopramide, ketorolac, prochlorperazine) for acute migraine, but similarity to dexamethasone for status migrainosus.⁵³ A more recent RCT reported the superiority of intravenous VPA over intravenous ibuprofen in the ED.⁵⁴ In the clinic setting, 400 mg of intravenous VPA was as effective as 6 mg of subcutaneous sumatriptan in a double-blinded RCT.⁵⁵ In an open-label randomized study, 500 mg of intravenous VPA was similarly effective to 10 mg of intramuscular metoclopramide with 1 mg of DHE.⁵⁶ In another open-label observational study, 75% of migraine patients reported pain relief in 60 min on 900–1200 mg of intravenous VPA.⁵⁷ While single-dose intravenous VPA appeared somewhat effective for acute migraine use, repetitive intravenous VPA use in the inpatient setting has not been properly studied.

At the time of writing, no RCTs or prospective observational studies have examined the role of repetitive intravenous VPA for inpatient headache management. We found two small low-quality case series (n=24) which reported on the use of intravenous VPA in patients admitted briefly to hospitals.^{58 59} No ED studies were included. Schwartz et al studied 10 patients with CDH or transformed migraine and/or MOH, who received a 15 mg/kg loading dose of intravenous VPA infused over 30 min, followed by 5 mg/kg infused every 8 hours for 4.3 days on average.⁵⁸ Eight patients had headache improvement, with four having total or near-total headache resolution. There were minimal AEs, with one patient having dysarthria, dysmetria, and nystagmus resulting in a 25% dose reduction. Notably, patients also had their preventive medications adjusted and were taught how to avoid MOH with the correct use of acute medications.

Riggins et al retrospectively reported on 14 patients with CM.⁵⁹ Over 4-day of treatment, patients received 250 mg of intravenous VPA infusions over 60 min, repeated every 8 hours, with a total dose ranging from 1125 to 2250 mg. They noted that 69% had improvement in their headache frequency, intensity, and/or use of acute medications 4–6 weeks following admission. One patient had to discontinue the protocol due to nausea/vomiting and vertigo, and another patient had restlessness/anxiety that improved by prolonging the infusion to 120 min.

We found a retrospective study (n=160) on the pharmacokinetics of intravenous VPA in pediatric and adult patients admitted to hospital (most commonly for seizures and migraine).⁶⁰ This study identified 160 adult migraine patients who received intravenous VPA (loading dose of 20–40 mg/kg over 30–60 min, followed by a continuous infusion of 1 mg/kg/hour with a target concentration of 50–100 µg/mL). They reported that 63.3% of patients had a >50% reduction in pain score, with the remaining 36.7% having <30% reduction in pain score (no patients were categorized as having a 30%–50% reduction in pain score).

Adverse events

Common AEs included gastrointestinal (GI) complaints (7.3%), hyperammonemia (3.6%), elevated liver enzymes >3 times the upper limit of normal (1.2%), and platelets <1x10ˆ5/cc (1.8%). It is worth noting that while the maximal tolerated dose of intravenous VPA is 140 mg/kg in healthy volunteers, intravenous VPA-related drug interactions (eg, metabolic inhibitions, displacement of protein binding sites) may lead to serious drug toxicity, especially under higher intravenous VPA dosage.⁶¹

In summary, the available evidence for repetitive intravenous VPA in managing inpatient headache is limited. Although it has been used for managing CDH or CM, its optimal dosing regimen has not been determined. Nevertheless, intravenous VPA may be a consideration for the management of inpatient headache depending on individual patient and contextual factors.

Statement

• Intravenous VPA may be beneficial for inpatients with intractable migraine. Level of certainty: low Recommendation:

• The committee recommends offering intravenous VPA for intractable migraine in select inpatients. Grade C.

What is the evidence for the use of intravenous ketorolac for inpatient headache management?

Mechanism of action

Ketorolac is a potent analgesic non-steroidal anti-inflammatory drug (NSAID) used for treating migraine. Its mechanism of action involves the inhibition of spinal cyclooxygenase activity, which is responsible for prostaglandin biosynthesis. This inhibitory effect terminates migraines associated with central sensitization and allodynia.⁶² Studies have also shown that ketorolac can reduce mast cell degranulation and block dural macrophage activation.^{63 64} Richer et al found that ketorolac effectively reduced central sensitization in animal models of migraine and in adult subjects, highlighting its potential as a treatment option.⁶⁵ Intramuscular or intravenous ketorolac exhibits a rapid onset of action, with effects achieved within 10 min and peak effect reached after 75–150 min.⁶⁶

Study findings

Our literature search did not find any studies on the use of ketorolac for the inpatient setting. Studies that included ED patients were included because of lack of inpatient studies. Intravenous ketorolac has been studied for the treatment of migraine in the ED. When compared with other parenteral drugs, such as VPA and prochlorperazine, intravenous ketorolac was found to be effective in this setting.⁶⁷,⁶⁹

Adverse events

The administration of ketorolac may be associated with GI bleeding, nephropathies, hematological dysfunction, and increased risk of cardiovascular events, including thrombotic events, myocardial infarction, stroke, and heart failure, and should be avoided in patients with or at high risk for the aforementioned conditions. Dosage adjustment and short-term uses can help minimize these severe events.⁷⁰,⁷² Other possible side effects may include dizziness, nausea, or vomiting, which are typically of mild intensity and tolerable in patients.⁷⁰

The utility of ketorolac is its ease of use, as it does not require hemodynamic monitoring or other specialized monitoring. For patients less than 65 years old, and provided no GI, renal, or cardiovascular contraindications, intravenous ketorolac is dosed at 10–30 mg once every 6–8 hours. The total time for therapy with ketorolac is not to exceed 5 days for risk mitigation.⁷² While 30 mg is an often-used dose, the analgesic effect of intravenous ketorolac exhibits a ceiling effect at 10 mg. Using lower doses is thus effective and may further reduce adverse effects.⁷³

In summary, there are no published studies on intravenous ketorolac use for migraine and MOH in non-ED inpatient settings. Notably, the AHS has rated ketorolac as probably effective for acute migraine management in the ED.^{52 74} More studies are needed to validate the effectiveness of ketorolac for the management of migraine in the inpatient setting. From extrapolation of ED studies, there is a consensus among medical experts that intravenous ketorolac could be an effective treatment option for managing acute migraine and NSAID MOH in inpatient settings.

Statement

• Intravenous ketorolac may be an effective and safe treatment option for managing acute migraine and non-NSAID MOH in the inpatient setting. Level of certainty: Low

Recommendation

• The committee recommends offering intravenous ketorolac for select inpatients with acute migraine or non-NSAID MOH. Grade C.

What is the evidence for the use of intravenous steroids for inpatient headache management?

Mechanism of action

Systemic corticosteroids have a long history in headache management as monotherapy or adjunctive. The main analgesic mechanisms of corticosteroids include decreased inflammation due to inhibition of prostaglandin synthesis, reduced vascular permeability and tissue edema, suppression of stress responses in the hypothalamic–pituitary axis, and modulation of neuroplasticity.⁷⁵ Administration of steroids for headache management is typically limited to short duration due to the well-known systemic side effects of long-term corticosteroid use, including Cushing syndrome, immunosuppression, neuropsychiatric dysfunction, osteoporosis, and GI bleeding.⁷⁶ Simultaneous use of corticosteroids with NSAIDs should be avoided due to increased risk of gastritis and GI bleed.

Study findings

The European Federation of Neurological Societies (EFNS) migraine treatment guideline recommends use of corticosteroids for status migrainosus by expert consensus, acknowledging that the data from RCTs are mixed regarding effectiveness in acute migraine treatment and recurrence prevention.⁷⁷ Similarly, the AHS assessed 4–16 mg of intravenous dexamethasone for acute migraine treatment as level C (possibly effective or ineffective).⁷⁴ Both the European Academy of Neurology and EFNS guidelines for CH concluded that corticosteroids are effective for CH prevention, based primarily on data from open-label studies and clinical experience, although this also includes oral and locally injected steroid.^{78 79} The overwhelming majority of evidence from RCTs supporting the above guidelines was performed in the outpatient or ED setting.

Focusing specifically on inpatient headache management, nine papers on intravenous steroids met criteria for inclusion; three case reports, three case series, two observational studies, and a preliminary report from an RCT (n=141). No ED studies were included. The preliminary report from the RCT⁸⁰ and both observational studies^{81 82} assessed MOH. We found three case reports⁸³,⁸⁵ investigating the use of intravenous steroids for patients with SUNCT, two case series^{86 87} for CH management, and one case series⁸⁸ for HC.

For MOH, preliminary results of the multicenter RCT reported on the first 23 patients enrolled at 2 of 10 planned Italian centers.⁸⁰ The patients were randomized to receive either 125 mg of methylprednisolone plus 10 mg of diazepam (14 patients) or 10 mg of diazepam alone (9 patients). A significant decrease was observed in the whole sample (p=0.005), but no significant difference in effectiveness was observed between the treatment groups, likely due to the small sample size (p=0.76). Both cohort studies were also performed in Italy, using multimodal treatment regimens lasting 7–10 days including intravenous anxiolysis (eg, daily diazepam), prophylactic medications (eg, amitriptyline, topiramate), and 4 mg⁸³ or 8 mg⁸¹ of intravenous dexamethasone daily.

At 1 month, 26 out of 32 patients (81%) in one study⁸¹ and 46 out of 52 patients (89%) in the other study⁸² no longer had medication overuse, although these rates decreased substantially in long-term follow-up (45% at 6 months).⁸¹

In all three cases of intravenous steroids for SUNCT, patients were administered 1 g of intravenous methylprednisolone for 3 days, in conjunction with other medications for headache prophylaxis, most notably membrane stabilizers such as carbamazepine^{83 85} or topiramate.⁸⁴ Similarly, the two case series for refractory CH used 1 g intravenous methylprednisolone for 3 days⁸⁷ or 125–500 mg for 2–3 days.⁸⁶ The two patients in the HC case series received repeated infusions of 500 mg of intravenous methylprednisolone for 1–3 days, with complete headache response.⁸⁸ Long-term remission was reported in all cases.

Adverse events

Three reports specifically stated no apparent adverse effects;⁸⁶,⁸⁸ the remaining cases did not comment on AEs.

In summary, there is limited evidence on the effectiveness of intravenous steroids in the inpatient treatment of acute headache and some evidence for their effectiveness in the prevention of headache recurrence.

Statement

• Intravenous steroids may be useful in the inpatient treatment of status migrainosus, MOH, and trigeminal autonomic cephalalgias (TACs). Level of certainty: Moderate

Recommendations

• The committee recommends using intravenous dexamethasone or intravenous methylprednisolone for 1–3 days in selected inpatients with refractory CM with MOH, status migrainosus, and TACs including CH and HC. Grade C

In patients who are administered intravenous steroids for inpatient headache, the committee recommends use of GI prophylaxis. Steroids should be used with caution in patients with gastritis, and patients with diabetes mellitus should have blood glucose levels monitored and appropriately treated until glucose levels return to baseline. Concurrent use of NSAIDs with steroids should be avoided. Grade B

What is the evidence for the use of intravenous levetiracetam for inpatient headache management?

Mechanism of action

Levetiracetam is a broad-spectrum antiepileptic medication against various seizure types and used as a second-line agent in the treatment of status epilepticus.^{89 90} Known for its effectiveness and good tolerability, levetiracetam operates through a unique mechanism by interacting with the synaptic vesicle protein 2A. Its actions extend to calcium homeostasis, GABA-ergic system modulation, and targeting α-Amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors, contributing to its antiepileptogenic, anti-inflammatory, neuroprotective, and antioxidant properties.⁹¹

Levetiracetam has a favorable pharmacokinetic profile with rapid and complete absorption following oral administration, excellent bioavailability, quick achievement of steady state concentrations, linear kinetics, and minimal plasma protein binding.⁹² It has no hepatic metabolism and is renally excreted in the form of inactive metabolites. This makes it a preferred option for patients with hepatic disease, but caution is advised in those with renal impairment.⁹³

Study findings

As of now, there are no publications specifically addressing the inpatient management of migraine with intravenous levetiracetam. The existing evidence primarily relies on a case report and a small open-label study that investigated intravenous levetiracetam use in ED and outpatient settings.

A case report documented a positive outcome in treating status migrainosus in the ED with a single dose of 1000 mg levetiracetam infusion in a patient with a history of migraine with aura and complex partial epilepsy.⁹⁴ In an open-label study focusing on outpatient intravenous levetiracetam treatment for acute intractable migraine, half of the patients achieved complete migraine resolution, with an average decrease in severity of 77.2%. The administered dose was 400 mg of intravenous levetiracetam every 5 min, up to a maximum effect, resulting in an average total dose of 4167 mg (range 400–11200 mg) over 56 min.⁹⁵

A proposed approach to ED and inpatient headache management in adults suggested the use of intravenous levetiracetam at a dose of 500–1500 mg as a second-line option, particularly for patients experiencing migraine aura status or prolonged aura.⁴ This recommendation is based on results from a small open-label study of 1000 mg oral levetiracetam treatment over 6 months in high-frequency migraine with aura, resulting in significant reductions in headache severity, duration, and aura symptoms.⁹⁶

Indirect evidence supporting the effectiveness of levetiracetam in migraine comes from its impact as migraine preventive medication. SR and meta-analysis on the effectiveness of levetiracetam for migraine prevention demonstrated the effectiveness of oral levetiracetam in reducing headache frequency and severity for adults with episodic migraine. The achievement of a 50% reduction in frequency ranged from 57.9% to 100%, with levetiracetam dosages varying from 500 to 3000 mg per day. Notably, inconsistent results were observed in the preventive treatment of CM.^{97 98}

The recent Canadian Headache Society guidelines give levetiracetam a weak recommendation with low quality of evidence for episodic migraine prevention and recommend against use for CM.⁹⁹

Adverse events

Importantly, levetiracetam exhibits no clinically significant pharmacokinetic interactions and presents with mild adverse effects. Intravenous administration aligns with the safety and pharmacokinetic properties established for the oral formulation.¹⁰⁰ Common adverse effects of levetiracetam include somnolence, dizziness, and asthenia. While depression and anxiety are possible, psychosis is rare.^{89 92 93}

In summary, while oral levetiracetam has demonstrated some evidence of effectiveness and safety as a preventive treatment for episodic migraine, the use of intravenous levetiracetam for migraine management in an inpatient setting is characterized by uncertainty and insufficient evidence. The current knowledge gap underscores the need for further research to provide a more comprehensive understanding of the potential benefits and risks of intravenous levetiracetam in this specific scenario. Based on extrapolation from ED studies, the committee recommends:

Statement

• Intravenous levetiracetam may be effective and safe to use in the acute management of intractable migraine or status migrainosus under specific circumstances. Level of certainty: Low

Recommendation

• The committee recommends that intravenous levetiracetam may be considered for inpatient management of acute migraine, when other therapies have been exhausted. Grade C

What is the evidence for the use of intravenous dopamine antagonists for inpatient headache management?

Mechanism of action

Dopamine antagonists are often used for acute treatment of headache given their antiemetic, sedating, and headache-relieving effects. Patients with migraine have been shown to have higher peripheral dopamine receptor density, dopamine agonist hypersensitivity, and ictal dopamine surges that are correlated with pain severity, which suggests that those with classic dopamine-mediated symptoms (eg, premonitory yawning, nausea, and vomiting) may derive greater benefit. The central and analgesic effects of dopamine antagonists are less clear due to complex pathophysiology but are potentially related to actions on dopaminergic inhibition of the trigeminocervical complex via hypothalamic projections and effects on ictal dopamine release.¹⁰¹,¹⁰⁴

Examples of common dopamine antagonists used in headache management include metoclopramide, prochlorperazine, chlorpromazine, promethazine, haloperidol, and droperidol. SR by the AHS and American Academy of Neurology have provided guidance regarding infusion dosage on the most studied dopamine antagonists giving level B evidence (probably effective) for 12.5 mg of chlorpromazine, 10 mg of metoclopramide, and 10 mg of prochlorperazine.^{74 105}

Study findings

Limited evidence is present regarding the effectiveness of intravenous dopamine antagonists for inpatient headache management specifically: we found one RCT, one prospective and two retrospective observational studies (n=275). No ED studies were included. Most studied medications in this setting are metoclopramide, prochlorperazine, and chlorpromazine, which have been shown to reduce headache intensity prior to discharge. Dosing frequency included once daily, every 8–12 hours, and a single repeat dose after 1 hour.

In addition to metoclopramide,¹⁰⁶ prochlorperazine,¹⁰⁷ and chlorpromazine,¹⁰⁸,¹¹¹ RCTs in the ED have assessed the effectiveness of intravenous droperidol, promethazine, and haloperidol for the acute treatment of headache. Similar effectiveness was observed with droperidol and promethazine when compared with prochlorperazine.¹¹²,¹¹⁴ Haloperidol was shown to have improved outcomes when compared with placebo¹¹⁵ and similar outcomes to metoclopramide.¹¹⁶

When given as part of a multidrug regimen over the course of an admission, along with implementation of preventive treatment and counseling, the use of intravenous dopamine antagonists has shown reductions in headache frequency, analgesic use, and acute care utilization following hospital discharge.

Adverse events

AEs with intravenous dopamine antagonists include sedation, hypotension, QTc prolongation,³⁷ and extrapyramidal symptoms (akathisia or dystonic reactions), with some studies showing a reduction in extrapyramidal symptoms when infused at a slower rate over 15 min.¹¹⁷,¹²²

In summary, depending on patient and clinicians’ preferences, intravenous dopamine antagonists (metoclopramide, prochlorperazine, chlorpromazine, droperidol, or haloperidol) should be offered to patients admitted to the hospital for acute treatment of headache, with the best evidence supporting the use of prochlorperazine and metoclopramide. Extrapyramidal symptoms can be minimized with a slower rate of infusion over the course of 15 min, which may prevent the need for the use of further sedating antihistamine medications. Home medications should be reviewed for drug interactions resulting in QTc prolongation with concomitant use. An ECG should be obtained prior to the use of any dopamine antagonists. For droperidol, ECG monitoring should also be obtained until 2–3 hours after completing treatment

Statements

• Intravenous dopamine antagonists may be effective in the acute treatment of intractable headaches in the inpatient setting. Level of certainty: moderate

• Side effects of intravenous dopamine antagonists can be minimized with slow infusion. Level of certainty: moderate

Recommendations

• The committee recommends that intravenous dopamine antagonists should be offered in the acute treatment of intractable headaches in the inpatient setting. Grade B.

• The committee recommends that home medications should be reviewed for drug interactions resulting in QTc prolongation with concomitant use. An ECG should be obtained prior to the use of any dopamine antagonists. Grade C.

• To minimize the extrapyramidal symptoms, the committee recommends a slower rate of infusion over the course of 15 min. Grade B.

What is the evidence for the use of intravenous lidocaine for inpatient headache management?

Mechanism of action

Intravenous lidocaine has been demonstrated to be both safe and efficacious in various chronic pain disorders,¹²³ but its study in CDH is more limited. Lidocaine is a sodium channel blocker, and its mechanism of action in the treatment of CDH has been suggested to be multifaceted, with mention of antinociceptive, anti-hyperalgesic, as well as anti-inflammatory effects playing a role in relieving symptoms.¹²⁴

Study findings

A systematic literature search yielded 13 papers (n=953). One non-RCT, seven observational studies, and five case reports/series, all addressing the effectiveness and safety of lidocaine for the treatment of CDH in the inpatient setting. No ED studies were included. The non-randomized open-label study examined the utility of intravenous lidocaine in patients with SUNCT or SUNA. The observational studies all examined migraine as the primary headache disorder,^{2123 125},¹²⁸ two of which also described patients with a subset of TAC.^{123 129} The case reports described intravenous lidocaine in SUNCT (6/8 cases),¹³⁰,¹³² HC¹³³ and SUNA¹³⁴ (one each) respectively. Most captured studies described a patient population that was disproportionately female¹²³¹²⁷,^{129 135} with reported mean ages from 39 to 45. Of all included studies, only one observational study used a comparator arm in their design.

All studies investigating migraine were observational and used varying doses, duration of treatment, and follow-up intervals. While the majority of studies did not use a weight-based dosing protocol, doses ranged from 0.5 to 1.5 mg/kg/hour among studies that did incorporate weight into dosing.^{130 131 134} Non-weight-based regimens ranged from approximately 0.5–4.0 mg/min with most studies using between 1 and 2 mg/min.¹²⁶,¹²⁸¹³⁰

Based on the available evidence, intravenous lidocaine may be indicated for inpatient management of TACs and migraine. The single non-randomized open-label prospective study by Lambru et al found that 91% of patients (52/58) were responders (ie, >50% benefit), with the vast majority of responders experiencing maximal benefit at infusion rates of 120 mg/hour.¹³⁵ The mean duration of benefit was classified as 22.1 days, but half of the responders suffered from a recurrence of attacks immediately after termination of the infusion. Three patients discontinued the infusion due to side effects (cognitive slowing and paranoid thoughts).

Of the observational studies, only Schwenk et al examined outcomes beyond discharge,^{126 127} with one study demonstrating a sustained response rate of 43%¹²⁶ at 1-month postdischarge, while the other demonstrated no sustained benefit.¹²⁷ Many studies reported AEs, with the most reported ones being vomiting, hallucinations, agitation, and cardiovascular changes.

Several studies examined outcomes at hospital discharge. Kiarashi noted that 7 out of 11 patients reported improvement in headache symptoms, 1 reported worsening of symptoms, and 3 reported no change at discharge. Patients also received other intravenous medications during the infusion, thus potentially confounding the results.²

Ray et al reported 51% of patients achieving a 50% or greater reduction in pain after a mean lidocaine infusion of 4.5 days and 31% of patients achieving pain freedom with a mean lidocaine infusion duration of 6.2 days at discharge.¹²⁸

Hand and Stark demonstrated lidocaine infusions were successful in alleviating headache symptoms with four infusions in three patients, whereas one infusion was deemed partially successful. Median response to achieve analgesia was 24 hours. While outcome measures were reported beyond the discharge period, the reported data did not clearly differentiate between varying headache diagnoses.¹²⁵

Adverse events

Most studies reported minor AEs including nausea, vomiting, cardiovascular issues, and blurry vision. Some AEs required temporary cessation of the lidocaine infusion until symptoms resolved. A few studies used continuous telemetry^{129 131 134}; however, some studies reported baseline ECG and intermittent monitoring.^{2 125}

In summary, the literature suggests a dosing of 1–2 mg/min of intravenous lidocaine infusion in the inpatient setting, with an ideal dose range of 0.5 mg/min to 4 mg/min. There is some evidence that after 5 days of infusion, there is an increased risk of accumulation of the toxic metabolite monoethylglycinexylidide.¹³⁶ Intravenous lidocaine infusion may offer short-term benefit for patients with TACs and migraine but may be associated with mild AEs.

Prior to infusion, patients should undergo 12-lead ECG, liver function tests, and liver enzyme tests. Intravenous lidocaine is contraindicated for patients with liver or renal failure. On-site anesthesiologist, local anesthetic systemic toxicity protocol, intralipid, and cardiovascular monitoring should be exclusively and immediately available for patients undergoing continuous lidocaine infusion; providers may also consider monitoring plasma concentrations if available and if results would return in time to affect management.

Statement

• Intravenous lidocaine infusion may offer benefit for inpatients with TACs and intractable migraine and may be associated with mild AEs. Level of certainty: low

Recommendation

• Duration of intravenous lidocaine should not routinely exceed 5 days to reduce risks associated with accumulation of a toxic metabolite. If prolonged infusion is warranted, then daily measurement of serum lidocaine levels may be considered. Grade C.

What is the evidence for the use of intravenous ketamine for inpatient headache management?

Mechanism of action

Ketamine is a dissociative anesthetic agent that possesses potent analgesic effects at subanesthetic doses. It is a non-competitive NMDA antagonist but interacts with multiple other receptors, including kappa-opioid and mu-opioid receptors, nicotinic and muscarinic cholinergic receptors, D2-dopamine receptors, and voltage-gated calcium channels, among others.¹³⁷ It has also been shown to have anti-inflammatory properties.¹³⁸

Study findings

A systematic literature search found five retrospective observational studies (n=183) investigating the effect of ketamine for inpatients specifically. No ED studies were included. It has primarily been studied for headache in patients refractory to standard treatments.^{127128 139},¹⁴¹ While only a few studies have been published, there is some evidence for a dose–response relationship in that longer infusions at higher infusion rates have been suggested to produce greater relief than shorter infusions at lower doses.^{128 140}

In a study by Ray et al, 34.4% of patients reported at least 50% relief after a multiday infusion, while in a study by Schwenk et al the mean pain of 7.5/10 decreased to 3.4/10 at the end of infusion.^{128 140}

Outcome measures across studies have not been consistent, with one study measuring the proportion of patients with a 50% reduction in pain¹²⁸ and the others reporting changes in the Numerical Rating Scale.¹²⁷,¹²⁹¹⁴² This may be related to variation in study design, including dosing and patient selection. It is not clear which patients would benefit most from ketamine infusion in terms of overall severity and status of disease (ie, stable vs acute exacerbation). Because primarily refractory patients have been studied, baseline pain levels of those treated with multi-day infusions have generally been in the severe range; in a study of six patients, they all had Migraine Disability Assessment scores of grade 4 at the time of treatment, indicating the greatest degree of disability.¹⁴⁰ Similar to the issue of timing, the frequency of repeated treatments has not been directly addressed.

Adverse events

Ketamine has been associated with transient worsening of liver function in some patients, which may limit the frequency of repeated treatments.^{143 144} We recommend performing a baseline hepatic function panel prior to treatment and repeating this at the end of treatment at a minimum. More frequent monitoring may be indicated in patients with a history of elevated liver enzymes or in those who are also prescribed intranasal ketamine on an outpatient basis. While the clinical significance of liver enzyme elevation from ketamine is not entirely known, we suggest decreasing the dose in patients with mild to moderate increases (≤10 times the upper limit of normal) and stopping the infusion for those with marked elevation (>10 times the upper limit of normal). There have been documented cases of symptomatic liver injury from prolonged ketamine infusions, all of which improved on ketamine cessation.¹⁴⁴ The side effect profile of ketamine includes sympathomimetic effects (hypertension and tachycardia), psychomimetic effects (anxiety, hallucinations, nightmares, vivid dreams, and out-of-body experiences), hepatic toxicity, nausea and vomiting, blurry vision, and urinary retention, among others.¹⁴⁰¹⁴⁵,¹⁴⁷ Several authors and reviews have reported that ketamine’s psychomimetic adverse effects can be mitigated by benzodiazepines and clonidine.^{148 149} Some experts recommend the use of a clonidine patch for all normotensive patients and PRN benzodiazepines for more severe adverse effects.¹⁴⁰

In one study by Schwenk et al, about half of the patients treated with a 5-day ketamine infusion experienced sedation, 37.7% had nausea or vomiting, 37.7% had blurry vision, and 27.9% reported hallucinations.¹⁴⁰ Patients in that study received a mean ketamine infusion rate of 0.76 mg/kg/hour. In healthy volunteers, ketamine increases heart rate and blood pressure but usually does not lead to hypertensive events.¹⁵⁰ Nevertheless, hypertensive events with ketamine infusions may occur, and this can occasionally limit the tolerated maximal dose.

In terms of suggested personnel and monitoring, the committee recommends following the recommendations proposed in the 2018 ASRA PM guidelines for ketamine infusions for chronic pain, which stated that the ordering physician should be an anesthesiologist, critical care physician, or pain physician experienced with ketamine and trained in both Advanced Cardiac Life Support and administration of moderate sedation.¹³⁷

While the guidelines did not specify exact monitors to be used for doses under 1 mg/kg/hour, the committee recommends telemetry be used for all patients unless close observation can be assured by an acute pain service or other personnel experienced with ketamine’s adverse effects. In addition to telemetry, blood pressure and pulse oximetry should be checked at a minimum of every 4 hours.

In summary, based on available evidence, the committee recommends ketamine infusions for patients with refractory headache be offered selectively based on physician and treatment center experience.

Statements

• Ketamine infusions may provide benefit in acute treatment of intractable headache. Level of certainty: low

• Adverse effects can occur even at subanesthetic doses. Level of certainty: high

• The incidence and severity of psychotomimetic and sympathomimetic effects from ketamine may be reduced by administration of clonidine or benzodiazepines. Level of certainty: moderate

Recommendations

• Ketamine infusions may be considered in the acute inpatient treatment of refractory headache. Grade C.

• Ketamine infusion rates should not exceed 1 mg/kg/hour outside of intensive care unit settings. Grade B.

• Patients should be under continuous respiratory and hemodynamic monitoring while receiving ketamine infusions. Grade B.

• Consider administering clonidine and/or a low-dose benzodiazepine to partially mitigate psychotomimetic and sympathomimetic adverse effects of ketamine. Grade C.

Discussion

Principal findings

The goal of this document was to provide structured and evidence-based statements and recommendations on intravenous drug administration specifically for the hospitalized headache patient. Although numerous headache management articles are available, they are primarily focused on the outpatient population. The evidence level for most of the medications included in this consensus document is low, and as a result, the strength for most recommendations is Grade C, indicating that the drug could be considered for selected patients in specific circumstances.

When there were limited data available from the inpatient literature, we extrapolated data from relevant ED literature and noted these instances.

Interpretation

Despite the overall weak evidence base, this review has provided valuable insights into the current evidence for each drug treatment in the inpatient setting and highlighted areas where more research is needed. Ultimately, this practice advisory aims to provide the first evidence-based recommendations for intravenous headache management specifically tailored to the inpatient population.

Clinical implications

Parenteral pharmacological treatment typically includes a combination of DHE, NSAIDs, antiemetics, antiepileptics, magnesium, corticosteroids, lidocaine, and ketamine, all tailored to the specific headache patient.¹⁵¹ These recommendations aim to provide additional options for providers to consider. While opioids are commonly prescribed for acute headache treatment, they should be avoided due to their limited effectiveness in pain relief, association with increased ED return rates, and long-term risk of tolerance, dependence, and addiction.^{152 153}

Our literature review suggests that intravenous inpatient pharmacological management is effective, often producing rapid results. However, the goals of admission should include validation of correct diagnosis and restoration of function and quality of life, in addition to the reduction of pain and other disabling headache-related symptoms. Key objectives include aborting the current headache attack, preventing headache recurrence, addressing associated symptoms, optimizing preventive and acute home treatment, and incorporating biobehavioral and nonpharmacological treatments.¹⁵⁴ Postdischarge follow-up is essential to ensure the effectiveness of new or adjusted medications and to provide patient support and education.

Limitations

While many of the medications have an established track record in treating headache in the outpatient setting, this review specifically examined their use in the inpatient population.

Unfortunately, the quality of the included studies was often low to moderate. Most studies were observational and limited by sample sizes. Further, heterogeneity in study population, intervention, and duration of follow-up, as well as lack of standardized outcome measures and publication bias, made it challenging to provide recommendations with a high level of certainty. For some widely recognized and well-established drugs, the evidence grading was lower than anticipated as compared with their effectiveness witnessed in the outpatient headache setting.

Additionally, ED data were included in several questions to allow for conclusions to be drawn, and this indirectness could have reduced the overall quality of the recommendation.

Future directions

These findings, therefore, underscore the need for targeted high-quality research in this specific patient population. Prospective trials are needed to assess and compare these intravenous infusions, alone or in combination, their dose–response relationship, and effect on pain-associated domains, specifically in the inpatient setting. More studies are also needed to evaluate the effect of single and repeated hospitalizations for infusion therapy with assessment of subacute to long-term outcomes.

Conclusions

Hospitalized patients with refractory headaches present unique challenges to the healthcare system. This evidence-based practice advisory provides a foundational step toward standardizing inpatient headache care and highlights existing gaps in our knowledge and the literature that should be addressed through rigorous prospective randomized studies.