• 
• 

Three Part Question

In [adult patients agitated secondary to acute psychosis] is [IM olanzapine or IM haloperidol] better at [providing safe and effective sedation]?

Clinical Scenario

An agitated 24-year-old patient with a history of schizophrenia enters the emergency department, and becomes aggressive and hostile towards staff and patients. De-escalation techniques fail, and the patient will not take oral medication or allow intravenous cannulation. You wonder whether intramuscular (IM) haloperidol or olanzapine is the best drug to rapidly, safely and effectively calm the patient.

Search Strategy

Medline 1950 to October week 2 2010 using the OVID interface, EMBASE 1980 to week 24 2010 using the OVID interface, PsycINFO 1806 to June week 3 2010 using the OVID interface:
(‘olanzapine’.mp/OR ‘atypical antipsychotic’.mp/OR ‘atypical antispychotic$’) AND (exp antipsychotic/OR exp haloperidol/OR ‘haloperidol’.mp) AND (exp psychomotor agitation/OR ‘agitation’.mp/OR ‘aggress$’/OR ‘agitat$’) LIMIT to human and adult and English language and therapy (sensitivity).

References of identified papers were also searched for relevant articles.

Search Outcome

Three hundred and forty seven papers were found, of which three were relevant high-quality studies

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Wright et al, 2001, Canada	Participants: 311 adults aged over 18 years and with schizophrenia schizophreniform disorder, or schizoaffective disorder (DSM-IV criteria). Setting: psychiatric inpatients in South Africa, and 12 European and North American countries. Inclusion criteria: score of >14 on the Positive and Negative Syndrome Scale–Excited Component (PANSE). Exclusion criteria: lactation, pregnancy, medical comorbidity. Study groups: Group 1: 10.0 mg IM olanzapine (n=131) Group 2: 7.5 mg IM haloperidol (n=126) Group 3: placebo IM (n=54)	Double blind randomised placebo controlled clinical trial.Level 1	Primary outcome: pair-wise comparisons of change in PANSE over 24 h	Olanzapine vs haloperidol: p=0.76. Olanzapine vs placebo: p<0.001. Haloperidol vs placebo: p<0.001	Randomisation method not stated. All participants signed consent forms, so degree of agitation uncertain. Limited data reported comparing baseline characteristics of different study groups. Protocol violations and additional medications prescribed not stated. Adverse effects of placebo-treated group not fully reported
			Secondary outcomes: mean change in PANSE at 15, 30, 45 min, 2 h and 24 h	Significantly greater mean change in olanzapine-treated patients than in haloperidol-treated patients seen at 15, 30 and 45 min, p<0.01. Otherwise no significant difference between haloperidol and olanzapine at other time points. Both olanzapine and haloperidol significantly better than placebo in reducing PANSE (eg, −7.7 vs −7.6 vs −3.6, at 2 h)
			Incidence of adverse effects in first 24 h of treatment for olanzapine vs haloperidol	No significant QTc changes in either group
Breier et al, 2003, USA	Participants: 270 adults aged over 18 years and with schizophrenia schizophreniform disorder, or schizoaffective disorder (DSM-IV criteria). Setting: psychiatric inpatients in 4 sites in Croatia, 1 site in Italy, 3 sites in Romania and 6 sites in South Africa. Inclusion criteria: score of >14 on PANSE. Exclusion criteria: drug dependency, violence, need for physical restraint, significant comorbidity. Study groups: Group 1: 2.5 mg, 5 mg, 7.5 mg, or 10.0 mg IM olanzapine (n=185) Group 2: 7.5 mg IM haloperidol (n=40) Group 3: IM placebo (n=45)	Double blind randomised placebo controlled clinical trial. Level 1	Primary Outcome: Mean change in PANSE at 2 h post treatment	Olanzapine (2.5 mg = -5.5; 5.0 mg= -8.1; 7.5 mg = -8.7; 10.0 mg = -9.4) were superior to placebo (-2.9; p= 0.01 vs olanzapine at 2.5 mg; p<0.001 for each other olanzapine dose) but not with haloperidol (-7.5; p>0.05).	Randomisation and allocation concealment methods not fully described. All participants signed consent forms, so degree of agitation uncertain. Power calculation not reported. Study only powered to investigate difference between drugs and placebo, not different drugs. Therefore possibility of a type II error exists. Limited data reported comparing baseline characteristics of different study groups. Wide exclusion criteria limits external validity
			Secondary outcome: incidence of adverse effects in first 24 h of treatment for olanzapine vs haloperidol	Acute dystonia: 0% vs 5.0%, p<0.01. Extrapyramidal symptoms: 2.9% vs 16.7%, p=0.01. No significant QTc changes in either group
Raveendran et al, 2009, India	Participants: 300 adults with psychiatric illness. Setting: Indian teaching hospital Inclusion criteria: patient judged to require IM sedation by managing doctors. Exclusion criteria: unable to gain consent from responsible adult. Patient groups: Group 1: 10.0 mg IM olanzapine (n=150). Group 2: 10.0 mg IM haloperidol+50.0 mg promethazine (n=150)	Pragmatic randomised controlled trial. Level 1	Primary outcome: Proportion ‘tranquil’ or asleep at 15 min and 4 h	15 min: olanzapine 87% vs haloperidol 91% (RR 0.96, 95% CI 0.34 to 1.47) 4 h: olanzapine 96% vs haloperidol 97% (RR 0.99, 95% CI 0.98 to 1.03)	Power calculation reported to assess treatment difference between drugs. Study not powered to assess safety of either intervention. No information on QTc interval reported. Primary outcomes assessed blinded to treatment allocation. Secondary outcome assessors unblinded. External validity to UK setting uncertain, although pragmatic design more consistent with everyday practice.
			Secondary outcomes: requirement for restraint within 4 hr. Requirement for additional drugs within 4 h. Mean time to produce tranquilisation Side effects within 4 h	Olanzapine 16% vs haloperidol 10% (RR 1.6, 95% CI 0.88 to 2.93). Olanzapine 43% vs haloperidol 31% (RR 2.07, 95% CI 1.43 to 2.97). Olanzapine 20.5 min vs haloperidol 12.8 min (p=0.4). Adverse effects uncommon in both groups. Olanzapine 2 patients with akathesia. Haloperidol: 1 patient with hypotension.

Comment(s)

Rapid IM tranquilisation may be necessary to control agitated psychotic patients, where de-escalation techniques have been unsuccessful and administration of oral medication or cannulation is not possible. However, the most appropriate drug is uncertain, with the National Institute for Health and Clinical Excellence guidelines on the short-term management of disturbed behaviour advising consideration of haloperidol, olanzapine or lorazepam.

All three randomised controlled trials comparing the efficacy of haloperidol and olanzapine have methodological flaws; however, the study by Raveedran et al appears valid and is likely to have the greatest generalisability given its pragmatic design. Addition of promethazine appears to reduce the side effect profile of haloperidol alone, while producing effective sedation with reduced requirements for additional drugs compared with olanzapine.

The final choice of drug should be tailored to individual patients, depending on contraindications, with intensive clinical and physiological monitoring necessary after administration. Note should also be taken of the manufacturer's Summary of Product Characteristics, which recommend a baseline ECG before treatment with haloperidol and warn of increased risk of death in using olanzapine in increased doses and for off-licence indications.

Unfortunately there have been no studies directly comparing olanzapine, haloperidol and lorazepam in agitation secondary to psychosis. Furthermore, the setting for published trials is restricted to psychiatric inpatients, predominantly outside of the UK. Further research is therefore necessary to determine a ‘gold standard’ regimen for rapid tranquilisation in UK emergency departments.

Clinical Bottom Line

Haloperidol and olanzapine both effectively reduce agitation secondary to psychosis. Combination of haloperidol and promethazine may produce a longer lasting effect with comparative risk of adverse effects to olanzapine.

References

1. Wright P, Birkett M, David SR, et al. Double-blind, placebo-controlled comparison of intramuscular olanzapine and intramuscular haloperidol in the treatment of acute agitation in schizophrenia. Am J Psychiatry 2001;158:1149–51.
2. Breier A, Meehan K, Birkett M, et al. A double-blind, placebo-controlled dose-response comparison of intramuscular olanzapine and haloperidol in the treatment of acute agitation in schizophrenia. Arch Gen Psychiatry 2002;59:441–8.
3. Raveendran NS, Tharyan P, Alexander J, et al. TREC-India II Collaborative Group. Rapid tranquillisation in psychiatric emergency settings in India: pragmatic randomised controlled trial of intramuscular olanzapine versus intramuscular haloperidol BMJ 2007;335:865–73.
4. National Institute for Health and Clinical Excellence. Violence: The Short-Term Management of Disturbed/Violent Behaviour in Psychiatric In-Patient Settings and Emergency Departments. Clinical Guideline 25 2005. London: National Institute for Health and Clinical Excellence.