Agitation is a common presenting or comorbid condition among patients in psychiatric emergency settings (1) and exists on a continuum of severity, ranging from irritability to violence (2–4). Concurrent with management of symptoms, attempting to identify the etiology of acute agitation is crucial for guiding ongoing treatment (5, 6). Recognizing and managing acute agitation effectively is essential to maintaining both patient and staff safety (7), and first-line management starts with verbal de-escalation (8).

When behavioral approaches alone are insufficient, pharmacologic interventions may be required. A significant portion of acutely agitated patients, though, do not respond to initial pharmacologic interventions (7, 9–12). An understanding of what medications may be safely administered in a sequential fashion is thus a fundamental component of acute agitation management protocols. While multiple guidelines are available that discuss first-line medications for acute agitation (7, 9, 13, 14), there is limited guidance regarding steps one should take pharmacologically when an initial medication has failed (15).

This manuscript aims to assist clinicians with managing patients with persistent acute agitation in psychiatric emergency settings by bridging the gap in the literature on this topic. Our focus is on pharmacological management of persistent acute agitation when sequential medications are needed. We discuss oral (PO), inhalational and intramuscular (IM) routes of administration, as these are most commonly available in psychiatric emergency settings. Using available evidence in the literature, we provide our own potential clinical algorithms for managing patients with persistent acute agitation and conclude with future research recommendations.

There is a substantial body of literature reviewing safety and efficacy of medications for acute agitation from which expert consensus guidelines have been developed. A representative sample of recent guidelines are provided in Table 1, which demonstrate that even among expert bodies, there is no consensus regarding any single medication that should be used first-line to manage acute adult agitation. This variability across expert recommendations is likely driven by studies to date demonstrating that multiple approaches appear to be efficacious for managing agitation. Additionally, many current guidelines have not been revised for years and do not incorporate newer medication options, more comprehensive data regarding side effects, or the importance of certain prophylactic agents that are now considered standard of care (16).

Table 2 highlights key factors to consider when deciding to administer sequential medication to treat persistent acute agitation. Staff members should have adequate training in identification and non-pharmacologic management of acute agitation (8, 17). Whenever a patient is being evaluated for necessity of medication, standardized measures to assess acute agitation severity should be utilized (18). Reassessment time points should be clearly defined (e.g., every 15 min), as should the goal time period to achieve calm (e.g., 45–60 min). The goal of treating acute agitation with medication should be calmness, not oversedation or obtundation from medication combinations (19, 20).

When a patient is unable to consent to medication, a restraint checklist should be used to reduce the risk that bias or provider limitations are impacting the decision to administer medications in an emergency setting (21). Oral pharmacological formulations are effective (22, 23) and should be utilized whenever possible prior to consideration of use of IM medications. If the IM route is indicated, staff must be properly trained in injection technique to avoid unpredictable absorption patterns and inadequate response (24).

Patient characteristics such as age and sex are important for informing treatment decisions. Certain medications are relatively contraindicated in elderly individuals due to an increased risk of adverse events in this age group (25, 26). Pediatric populations necessitate a multimodal approach with modified dosing (27), and only a subset of discussed medications are considered appropriate for use in pregnant patients (28).

A thorough history, review of medical conditions and current medications, and physical examination is important, although often not possible at the initial assessment time point due to the severity of agitation and should be reattempted when appropriate to do so. Assessing recent substance use is critical to ensure accurate diagnosis and mitigate interactions between substances and medications used to treat acute agitation.

Combining medications with similar side effect profiles may increase risk for adverse effects, and during sequential dosing, close attention should be paid to what a patient has already received to avoid or mitigate these cumulative risks. Documentation of the expected time to effect of medications already administered (29) and likelihood of requiring subsequent dosing (30, 31) can help avoid premature dosing and polypharmacy.

Current evidence from research examining first-line interventions should guide sequential medication selection (7, 9, 14, 29, 32–34). Evidence-based antipsychotics for acute agitation include loxapine, haloperidol, droperidol, olanzapine, risperidone, ziprasidone, aripiprazole, and asenapine (13, 35–53). Evidence-based benzodiazepines for acute agitation include midazolam and lorazepam (29, 32, 33, 54, 55). In the United States (U.S.), IM droperidol and midazolam are not typically used in psychiatric emergency settings unless embedded in medical emergency departments due to monitoring requirements.

Current guidelines counsel against co-administration of IM olanzapine and IM lorazepam, although some have argued the risk associated with such is overstated (56) or most relevant for patients with alcohol intoxication given elevated risk of respiratory depression in this population (57, 58). Promethazine and diphenhydramine have also been used in combination with antipsychotics to treat acute agitation (31, 42, 59). Evidence is lacking to support the combination use of a first-generation antipsychotic (FGA), anticholinergic and benzodiazepine concurrently (60). Compared to antipsychotics and benzodiazepines, ketamine has a greater incidence of adverse events when used for acute agitation (61, 62) and should be avoided in psychiatric emergency settings.

There is limited evidence examining specific sequential medication options for managing persistent acute agitation, although many studies examining initial medications allowed for rescue sedation if indicated. One contributor to this limited evidence may be the United States Food and Drug Administration's restrictions on conducting clinical trials of agitation (63, 64). The majority of studies examining initial interventions for acute agitation did not evaluate associations between sequential dosing or polypharmacy and efficacy or adverse events. We summarize notable findings, while emphasizing limitations on interpretation.

In a prospective observational study of 1,403 participants receiving IM droperidol (65), 31% of participants received additional sedation including droperidol, midazolam, ketamine, diazepam and dexmedetomidine. Oversedation was significantly associated with three or more attempts at sedation (but not if droperidol was the only medication used), and with the use of benzodiazepines. Five percentage of participants experienced other adverse events (hypotension and respiratory depression were the most common).

A randomized controlled trial (RCT) of 167 participants receiving IM midazolam, olanzapine or haloperidol allowed for repeated dosing and alternative medications (66). Of participants receiving midazolam, 23% received a second dose, and 16% received alternative medication; of those receiving olanzapine, 28% received a second dose, and 11% received alternative medication; of those receiving haloperidol, 32% received a second dose, and 12% received alternative medication. There was one episode of dystonia and one fatal cardiac arrest in the haloperidol arm, the latter of which was thought partially due to stimulant intoxication.

In a prospective observational study of 2011 participants examining IM olanzapine, haloperidol, and zuclopenthixol (12, 67), ≤ 67% of participants required at least one additional dose of medication, and 62.7% received ≥2 antipsychotics during the acute phase. Participants initially receiving IM olanzapine had fewer adverse events and required fewer anticholinergics/anxiolytics. Sedation was common, and haloperidol was associated with extrapyramidal symptoms (EPS). Another observational study examining olanzapine (68) allowed for sequential dosing with additional parenteral medications. 10/489 participants receiving the IM formulation experienced respiratory depression, including five requiring intubation.

One RCT of 270 participants examining repeated IM olanzapine dosing compared to repeated dosing of IM haloperidol or placebo found that haloperidol was associated with greater incidence of EPS (69); those randomized to olanzapine 10 mg or haloperidol required fewer doses. Another RCT of 300 participants receiving IM olanzapine or IM haloperidol and promethazine found that the olanzapine group required more additional dosing (43 vs. 21%), and there were no significant between-group differences in EPS (59). An RCT of 488 participants receiving IM haloperidol, aripiprazole or placebo, allowed up to 3 injections/24 h (70); the haloperidol group had greater incidence of EPS. An RCT of 376 participants comparing sequential dosing of IM ziprasidone and haloperidol found ziprasidone was associated with fewer adverse events, and haloperidol associated with greater incidence of EPS (71).

A retrospective study of 15,918 participants received IM droperidol, olanzapine, and haloperidol examined rescue sedation requirements (31). Haloperidol was associated with greater use of rescue sedation medications. Forty nine intubations were documented; one fatality occurred in a participant who received olanzapine, haloperidol, and ketamine and was later found to have a subarachnoid hemorrhage. A prospective study examining IM ziprasidone, midazolam, haloperidol and olanzapine found 20–40% of 737 participants required sequential dosing (30). There were four intubation events, but no association between such and specific medication combinations received was reported.

An RCT of 115 participants comparing IM droperidol, lorazepam, and ziprasidone (72) and allowing for rescue sedation noted lower respiratory depression with droperidol than with ziprasidone or lorazepam. An RCT of 359 participants comparing inhaled loxapine and IM aripiprazole found that repeat administrations of both were well-tolerated, loxapine had a faster onset, and participant satisfaction was higher in the loxapine-treated group (73). An RCT of 344 participants examining repeated dosing of inhaled loxapine found that administration of two to three doses of 5–10 mg each was well-tolerated and efficacious (74). An RCT of 124 participants found sequential dosing of risperidone (total 24-h dose >6 mg) was associated with EPS (43). A retrospective analysis of 388 participants found that repeated dosing of IM antipsychotics, but not PO antipsychotics, was associated with a longer length of stay in the psychiatric emergency department (11).

Overall, while numerous studies of acute agitation management have allowed sequential medication administration, few have assessed what risks are associated with specific sequential options. Risks that have been reported include increased cardiac risk from combining QTc prolonging agents and more common known side effects with higher doses of specific agents (e.g., EPS when haloperidol is administered without an agent for EPS prophylaxis). The incidence of serious events (intubation, serious arrhythmia, or death) are quite rare even when multiple medications are used concurrently. A study of 904 participants to comprehensively assess risk factors for adverse events found that administering multiple medications within a 60-min period, alcohol intoxication, and age >65 were associated with adverse events (25), suggesting sequential medication use does increase risk of adverse events and should be used cautiously.

Since the need for guidance remains surrounding how to approach sequential medications in practice, we offer our own suggested sequential treatment algorithms (Figure 1) for persistent acute agitation that are consistent with data from available studies, using agents that are available in the United States.

Since polypharmacy is one of the main predictors of adverse events (25), an approach that focuses on limiting polypharmacy should be prioritized. Assessment of acute agitation severity using a standardized measure (18) should be completed at each time point. Behavioral interventions should be re-attempted prior to consideration of subsequent medication administration. Vital sign and side effect monitoring should be done to reduce risk of adverse events. Oral routes should be utilized over parenteral routes when possible. The goal of intervention should be to assist the patient in achieving a state of calmness, not oversedation or obtundation from medication combinations. At each step, consider expected time to onset of the medication being used, and take this into account when determining how soon another dose may be needed (32). At each step, continue to assess whether a non-psychiatric medical condition is causing or contributing to continued agitation, and whether transfer to a medical emergency department might be indicated.

This paper aims to fill a gap in the acute agitation literature regarding how to treat persistent acute agitation in psychiatric emergency settings when an initial medication has failed. Given the paucity of studies in this area, it is not possible to provide a single best-practices algorithm for management. The suggestions we put forth are our own clinical suggestions and are created from available data gleaned from studies of initial medication interventions for acute agitation. The general principle of using the lowest effective dose (starting at low doses and providing adequate time to effect between doses) applies, as well as incorporating new information as it becomes available to guide treatment. Aiming to achieve a state of calmness, but not oversedation or obtundation from medication combinations, should be prioritized as additional medications are added over a multi-hour timeframe. Similarly, assessment of vital signs, side effects, treatment response, and utility of behavioral interventions should be undertaken at each time point that a sequential medication is being considered. Future research in this area is clearly needed, including formal testing of repeated medications using explicit algorithms (e.g., re-dosing with the same medication vs. switching medications), with subsequent comprehensive analyses of associations between specific sequential options and both treatment response and adverse events.

VS completed the background literature search and wrote the first draft of the article. All authors have approved the final article.

VS was supported by the National Institute of Mental Health grant R25MH06048.

The authors alone are responsible for the content and writing of this article. The views expressed in this article do not represent the views of the Department of Veterans Affairs or the U.S. government.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The reviewer SZ declared a shared affiliation, with no collaboration, with the authors VS, EH, and RK at the time of the review.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.