ORIGINAL ARTICLE

rrmn:

controlled trial

Charles Lung-Cheng Huang , Tzung-Jeng Hwang ,

National Taiwan University, Taipei, Taiwand Department of Psychiatry, Chang Bing Show Chwan Memorial Hospital, Changhua County, Taiwan

n intramuscular (IM)atients with moder-

ly agitated patientsd to receive 10 mg(n Z 30). Agitationponent (PANSS-EC)nd at 24 hours afternd Barnes Akathisia

Otsuka Taiwan, and served on an advisory board for dementia research from Eli Lilly and Company. The other authors declare no financialrelationships relevant to the subject of this article.* Corresponding author. Department of Psychiatry, National Taiwan University Hospital, Number 7 Chung-Shan South Road, Taipei 10002,

Taiwan.E-mail address: tjhwang@ntu.edu.tw (T.-J. Hwang).

g These authors contributed equally as first authors to this work.

http://dx.doi.org/10.1016/j.jfma.2015.01.0180929-6646/Copyright 2015, Elsevier Taiwan LLC & Formosan Medical Association. All rights reserved.

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.j fma-onl ine.com

Journal of the Formosan Medical Association (2015) 114, 438e445Conflicts of interest: Dr. Tzung-Jeng Hwang has received honoraria for continuous medical education from Pfizer, AstraZeneca, ande Institute of Statistics, National Chiao Tung University, Hsinchu, Taiwanf Division of Biostatistics, Graduate Institute of Epidemiology and Preventive Medicine,College of Public Health, National Taiwan University, Taipei, Taiwan

Received 7 October 2014; received in revised form 24 January 2015; accepted 29 January 2015

KEYWORDSagitation;haloperidol;injection;lorazepam;olanzapine;schizophrenia

Background/Purpose: To compare the efficacy and safety profile betweeolanzapine and IM haloperidol plus IM lorazepam in acute schizophrenic pate to severe agitation.Methods: This was a prospective, randomized, open-label study. Acutewith schizophrenia or schizoaffective disorder (n Z 67) were randomizeIM olanzapine (n Z 37) or 5 mg IM haloperidol plus 2 mg IM lorazepamwas measured with Positive and Negative Syndrome Scale Excited Comand AgitationeCalmness Evaluation Scale (ACES) during the first 2 hours athe first injection. Safety was assessed using the SimpsoneAngus Scale aYi-Hsing Chen d, Guan-Hua Huang e, Ming H. Hsieh c,Hsiu-Hsi Chen f, Hai-Gwo Hwu c

a Department of Psychiatry, Chi-Mei Medical Center, Tainan, Taiwanb Department of Social Worker, Chia Nan University of Pharmacy and Science, Tainan, Taiwanc Department of Psychiatry, National Taiwan University Hospital and Collage of Medicine,a,b,g c,*,gIntramusculaintramusculafor the treatwith agitatioolanzapine versushaloperidol plus lorazepament of acute schizophreniaAn open-label, randomized

g aritys dp 80% of agreement amongraters was achieved. In terms of efficacy and safetyassessment, ratings were completed by one psychiatrist intwo centers, and by two attending psychiatrists in the thirdcenter. All raters were not blinded to the medications. Tofurther minimize bias in rating, each unblinded psychiatristcompleted the ratings together with the respectiveattending nurse who observed the patients condition mostclosely.

Efficacy assessments

Patients were assessed by the study investigators at thescreening visit and at 15 minutes, 30 minutes, 60 minutes,and 120 minutes after the first injection. The primary ef-ficacy measure was PANSS-EC, which was derived from thePANSS by its originators using a principal-components factoranalysis, and includes the items of tension, uncooper-ativeness, hostility, poor impulse control, and excite-ment.25,26 The score of each item ranges from 1 (normal) to7 (most severe), with a total sum score ranging from 5 to35. Agitation was further assessed by the AgitationeCalm-ness Evaluation Scale (ACES) (Copyright 1998), a single-itemscale developed by Eli Lilly and Company, where1 Z marked agitation; 2 Z moderate agitation; 3 Z mildagitation; 4 Z normal; 5 Z mild calmness; 6 Z moderatecalmness; 7 Z marked calmness; 8 Z deep sleep; and9 Z unable to be aroused. The Clinical Global Impression-Severity27 (CGI-S) scale was used to assess the generalpsychiatric condition. For each patient, the same raterconducted the assessment throughout the study.

Safety assessments

During the 24-hour treatment period, safety was assessedby clinical examination, recording spontaneously reportedadverse events, and completing the SimpsoneAngus Scale28

(SAS) and Barnes Akathisia Rating Scale.29

Statistical analysis

The efficacy analyses were based on the intent to treatpopulation defined as all of the randomized patients. Thelast observation carried forward dataset was used to esti-mate the missing data. Demographic characteristics andclinical parameters at baseline were compared by thetreatment group using the t test for continuous variablesand Chi-square test for categorical variables. The primaryendpoint was the change in PANSS-EC score 2 hours afterthe first injection, and the secondary endpoints included

change in ACES score 2 hours after the first injection, andchange in PANSS-EC, ACES, and CGI-S scores 24 hours afterthe first injection. For the primary endpoint, the lower limitof noninferiority was defined a priori as 40% of the observedmean change from baseline to 2 hours following the firstolanzapine injection. A lower boundary of the one-sided97.5% confidence interval of zero or less but greater thanthe lower limit indicated no between-treatment differenceand noninferiority, i.e., haloperidol plus lorazepam was notinferior to olanzapine. Within-group comparison was per-formed using paired t test. Responder was defined as thosewith at least 40% reduction from baseline on the PANSS-ECat 2 hours. To compare the number of differences inadverse events and response rate between the two treat-ment groups, Fishers exact test was used. Data wereanalyzed using the statistical program R Language version2.8.0 (http://www.r-project.org/).

Results

Demographic characteristics and clinicalparameters at baseline

A total of 294 patients received an initial assessment afteracute admission. After exclusion, 67 patients (NTUH, 6;

NTUH Yun-Lin branch, 21; Yu-Li Psychiatric Hospital, 40)were entered into the randomization phase. The patientflow is shown in Fig. 1. The main reasons for exclusion weredid not sign informed consent and having receivedrecent depot injections or newly added benzodiazepines orantipsychotics. Another 180 patients did not enter therandomization phase because of inadequate severity ofagitation as judged by the raters. The 67 patients, including58 individuals with schizophrenia and nine with schizo-affective disorder, were randomized to the two treatmentgroups (37 in the olanzapine group, 30 in the haloperidolplus lorazepam group), and they all completed the trial.Overall, the patients were moderately to severely agitated(mean PANSS-EC score: 21.4 4.2). There were no signifi-cant between-group differences in sex, age, age of onset,length of current psychotic episode, or baseline PANSS-ECscore (Table 1).

Efficacy

In terms of primary endpoint, the PANSS-EC scores decreasedsignificantly at 2 hours following the first injection in bothgroups (olanzapine: e10.2 6.5, t Z 9.750, p < 0.001;haloperidol lorazepam:e9.9 5.6, tZ 9.900, p< 0.001).The difference between haloperidol plus lorazepam and

ry o

Intramuscular olanzapine for acute schizophrenia 441Figure 1 Summa f participant flow.

olanzapinewas 0.3 units favoring olanzapine (with one-sidedlower 97.5% confidence limit Z 3); therefore non-inferiority (3 vs.10.2 0.4Z4.1) could be concluded.Regarding sedative effect, the ACES scores increasedsignificantly at 2 hours in both groups (olanzapine: 2.1 1.7,t Z 7.225, p < 0.001; haloperidol lorazepam: 2.2 1.7,p < 0.001). There were no significant differences in PANSS-EC or ACES scores between the two groups at 15 minutes,30 minutes, 60 minutes, and 120 minutes following the firstinjection (Figs. 2 and 3), although the olanzapine grouptended to have faster action in the 1st hour after the firstinjection. The percentage of responder (defined as at least

haloperidol lorazepam group; Fishers exact test,pZ 0.323]. The percentage of participants with ACES of 8 or9 was also not significantly different between the two groups[3 (8%) in the olanzapine group vs. 4 (13%) in thehaloperidol lorazepam group; Fishers exact test,p Z 0.692]. The changes in CGI-S and PANSS-EC from base-line to 24 hours after the first injection showed no significantdifference between the two groups (Table 2). Five patients(13.5%) received a second injection, and two patients (5.4%)received a third injection in the olanzapine group, whereasin the haloperidol plus lorazepam group, five (16.7%) pa-tients received a second injection, and no participant

Table 1 Demographic characteristics and clinical parameters at baseline.

Characteristic Olanzapine 10 mg(n Z 37)

Haloperidol 5 mg lorazepam2 mg (n Z 30)

p

Male/female 22/15 19/11 0.750Mean age (y) 37.1 10.8 41.3 11.3 0.127Age of onset (y) 22.3 9.1 19.3 3.8 0.075Length of current psychotic episode (d) 20.4 28.1 49.0 13.4 0.274PANSS-EC scorea 21.1 4.4

(range 14e32)20.9 3.6(range 15e29)

0.783

ACES scoreb 2.0 0.6(range 1e3)

2.1 0.7(range 1e4)

0.327

a Positive and Negative Syndrome Scale-Excited Component; five-term scale, with a maximum total score of 35.b AgitationeCalmness Evaluation Scale; one-item scale, with a range of 1e9.

442 C.L.-C. Huang et al.40% reduction from baseline on the PANSS-EC at 2 hours)was not significantly different between the two groups[19 (51%) in the olanzapine group vs. 11 (37%) in theFigure 2 Changes in Positive and Negative Syndrome ScaleExcited Component (PANSS-EC) score within 2 hours after thefirst injection. There were no significant differences in PANSS-EC scores between the two groups at 15 minutes, 30 minutes,60 minutes, and 120 minutes following the first injection.received a third injection. However, the injection fre-quencies between the two groups were not significantlydifferent.Figure 3 Change in AgitationeCalmness Evaluation Scale(ACES) score within 2 hours after the first injection. There wereno significant differences in ACES scores between the twogroups at 15 minutes, 30 minutes, 60 minutes, and 120 minutesfollowing the first injection.

Adverse effects

treatment of acute agitation, as revealed by the changes in

the PANSS-EC score within the 1st hour after the initial in-

Table 2 Comparisons of mean changes in PANSS-EC, ACES, CGI, SAS, and BARS from baseline to 24 hours after the firstinjection.

Olanzapine Haloperidol 5 mg lorazepam 2 mg pa

Score at baseline Score at 24 h Paired difference Score at baseline Score at 24 h Paired difference

PANSS-EC 21.1 4.4 12.0 4.9 9.2 5.4 20.9 3.6 12.7 4.0 8.1 3.1 0.450ACES 2.0 0.6 3.2 0.9 1.2 0.9 2.1 0.7 3.5 0.9 1.3 1.0 0.393CGI-S 5.0 0.8 4.2 0.8 0.8 0.6 5.2 0.7 4.4 0.9 0.8 0.7 0.896SAS 13.4 4.3 13.0 3.8 0.4 1.3 13.9 5.5 13.2 3.8 0.6 2.8 0.765BARS 2.4 2.7 1.6 2.3 0.8 1.7 2.1 2.5 1.6 2.3 0.6 2.1 0.551ACES Z Agitation Calmness Evaluation Scale; BARSZ Barnes Akathisia Rating Scale; CGI-SZ Clinical Global ImpressionsdSeverity ofIllness; PANSS-EC Z Positive and Negative Syndrome Scale-Excited Component; SASZ SimpsoneAngus Scale.a p value for comparing paired differences between two treatment groups, adjusting for centers.

Intramuscular olanzapine for acute schizophrenia 443the PANSS-EC scores within and at 2 hours after the firstinjection. Furthermore, the treatment regimens demon-strated similar efficacy as measured by injection frequencyand the changes in CGI-S score from baseline to 24 hoursafter the first injection.

In previous studies comparing with 7.5 mg IM haloperidolin the treatment of acute agitation of schizophrenia, 10 mgIM olanzapine repeatedly showed a more rapid reduction inThe changes in SAS and Barnes Akathisia Rating Scale scoresfrom baseline to 24 hours after the first injection showed nosignificant differences between the two groups (Table 2).The incidences of adverse reactions were also not signifi-cantly different between the two groups (Table 3). How-ever, acute dystonia only occurred in the haloperidol pluslorazepam group.

Discussion

To the best of our knowledge, this study is the first ran-domized controlled trial in a Chinese population comparingIM olanzapine with a frequently used combination of IMhaloperidol plus IM lorazepam in the treatment of acuteschizophrenia with moderate to severe agitation. The re-sults showed that 5 mg IM haloperidol plus 2 mg IM loraz-epam were not inferior to 10 mg IM olanzapine for theTable 3 Incidence of treatment emergent adverse events.

Primary term Olanzapine 10 mg(n Z 37)

All adverse events 9Dizziness 1Drowsiness 3Malaise 1Nausea 1Palpitation 1Anxiety 1Headache 0Acute dystonia 0Restlessness 1jection, but the difference became smaller at 2hours.15,17,22,30 Similar phenomena were observed in thetreatment of agitated patients with dementia or maniawhen IM olanzapine was compared with IM lorazepam.19,20

Thus, one interesting clinical question would be how effi-cacious and fast-acting it would be when 10 mg IM olan-zapine is compared with 5 mg IM haloperidol plus 2 mglorazepam. Our data showed that there was still a trend ofa slightly faster action of IM olanzapine treatment asobserved by the changes in the PANSS-EC scores within 1hour after the first injection. However, the differences didnot reach statistical significance and became nearly zero at2 hours after the first injection, suggesting that IM olanza-pine may have faster action, but IM haloperidol plus lor-azepam are not inferior at 2 hours following the firstinjection (Fig. 2). A study comparing 10 mg IM olanzapinewith 5 mg IM haloperidol plus 15 mg midazolam in thetreatment of agitated psychiatric patients also showed thatolanzapine had a more rapid action in the first 2 hours afterthe initial intervention.31 However, the effect may berelated to the dosage chosen. If a lower dose of olanzapineis chosen for comparison, the difference between the twogroups may be less.17

In terms of calming effect (as measured by the ACES),although our study did not show significant between-groupdifferences during the first 2 hours or at the end of 24hours, IM olanzapine seemed to have slightly faster actionin the 1st hour; however, IM haloperidol plus lorazepamHaloperidol 5 mg lorazepam2 mg (n Z 30)

p

10 0.4302 0.5835 0.4510 > 0.990 > 0.990 > 0.990 > 0.991 0.4481 0.4481 > 0.99

figure was lower than 4.0e7.1% reported in association with

treatment of acute agitation in schizophrenia. Arch Gen Psy-chiatry 2002;59:441e8.

444 C.L.-C. Huang et al.IM 7.5 mg haloperidol in previous studies.15,17,22 Moreover,there seemed to be a higher incidence of dizziness/drowsiness in the haloperidol plus lorazepam group. Bycontrast, only one patient in the olanzapine group reportedpalpitations. Previous studies have demonstrated that,compared with IM haloperidol, olanzapine causes a lowerdegree of dystonia and parkinsonism, but a greater degreeof hypotension.15,17,18 Compared with IM lorazepam, olan-zapine has been reported to cause a similar incidence ofsomnolence and dizziness.19,20 When haloperidol is com-bined with lorazepam, the additional effects of benzodi-azepines may allow for a lower incidence of acute dystonia;however, the potential risk of synergistic sedative effectsmay be raised.12,13

There are several limitations to this study. First, this isan open-label trial, so some bias may have been introducedinto the evaluation process. We tried to avoid this by ratertraining and joint evaluation by the attending physician andnursing specialist. Furthermore, the rating of the primaryoutcome was behavior-based rather than interview-based,thus minimizing subjective bias. Second, there was noplacebo group, although the efficacy of these drugs hasalready been proven, which cannot be explained by a pla-cebo effect. Third, our sample size may not be largeenough to have adequate power to detect differences be-tween the two treatment regimens. Fourth, the severity ofagitation of the 180 excluded patients was based on theraters clinical judgment rather than formal rating using thePANSS-EC scale; thus, some of them might have beenqualified for enrollment if they had received formal ratings.However, this would not significantly bias the resultsbecause we adopted a randomization procedure in thisstudy. Further research needs to be undertaken with well-designed, double-blind, placebo-controlled studies with alarger sample size to provide more evidence for the efficacyand safety of these treatment regimens.

Acknowledgments

The study was supported by a grant from Mr Ya-Jen ChengsFoundation, Taipei, Taiwan. The preparation of thismanuscript was supported by a grant from National ScienceCouncil (97-2314-B-002-129-MY3). We are grateful to Dr Yi-Fang Chuang and Ms Shuang-Sui Chen for their help in thedata collection of this study, Eli Lilly and Company forsupplying short-acting injectable olanzapine, and Dr SusanShur-Fen Gau for her comments on this paper. The authorsalso acknowledge statistical assistance provided by thecaught up 2 hours after the initial injection (Fig. 3). Thefindings were compatible with those of recent studiescomparing 10 mg IM olanzapine with 2.5e5 mg IM halo-peridol plus 7.5e15 mg midazolam.31,32 Furthermore, ourstudy showed these two treatments both demonstrated anacceptable calming effect without too much sedation in thefirst 2 hours after injection.

Although our study found no significant differences inadverse events between the two groups at the end of 24hours (Table 3), their profiles were slightly different. Forexample, acute dystonia only occurred in the haloperidolplus lorazepam group (1 of 30 patients, 3.3%), and the18. Wright P, Lindborg SR, Birkett M, Meehan K, Jones B, Alaka K,et al. Intramuscular olanzapine and intramuscular haloperidolin acute schizophrenia: antipsychotic efficacy and extrapyra-midal safety during the first 24 hours of treatment. Can JPsychiatry 2003;48:716e21.Taiwan Clinical Trial Bioinformatics and Statistical Center,Training Center, and Pharmacogenomics Laboratory (whichwas founded by the National Research Program for Bio-pharmaceuticals at the Ministry of Science and Technologyof Taiwan; MOST 103-2325-B-002-033).

References

1. Liu CC, Chen CH, Hwu HG, Shiu SY, Hua MS, Chen CH, et al.Medium-term course and outcome of schizophrenia depictedby the sixth-month subtype after an acute episode. J FormosMed Assoc 2012;111:265e74.

2. Noble P, Rodger S. Violence by psychiatric in-patients. Br JPsychiatry 1989;155:384e90.

3. Krakowski M, Czobor P. Violence in psychiatric patients: therole of psychosis, frontal lobe impairment, and ward turmoil.Compr Psychiatry 1997;38:230e6.

4. Allen MH. Managing the agitated psychotic patient: a reap-praisal of the evidence. J Clin Psychiatry 2000;61:11e20.

5. Battaglia J. Pharmacological management of acute agitation.Drugs 2005;65:1207e22.

6. Thomas P, Alptekin K, Gheorghe M, Mauri M, Olivares JM,Riedel M. Management of patients presenting with acute psy-chotic episodes of schizophrenia. CNS Drugs 2009;23:193e212.

7. Wilson MP, Pepper D, Currier GW, Holloman Jr GH, Feifel D.The psychopharmacology of agitation: consensus statement ofthe American Association for Emergency Psychiatry projectBeta psychopharmacology workgroup. West J Emerg Med 2012;13:26e34.

8. Battaglia J, Moss S, Rush J, Kang J, Mendoza R, Leedom L, et al.Haloperidol, lorazepam, or both for psychotic agitation? Amulticenter, prospective, double-blind, emergency depart-ment study. Am J Emerg Med 1997;15:335e40.

9. Binder RL, McNiel DE. Emergency psychiatry: contemporarypractices in managing acutely violent patients in 20 psychiatricemergency rooms. Psychiatr Serv 1999;50:1553e4.

10. Allen MH, Currier GW. Use of restraints and pharmacotherapyin academic psychiatric emergency services. Gen Hosp Psy-chiatry 2004;26:42e9.

11. Currier GW, Trenton A. Pharmacological treatment of psy-chotic agitation. CNS Drugs 2002;16:219e28.

12. Salzman C, Green AI, Rodriguez-Villa F, Jaskiw GI. Benzodiaz-epines combined with neuroleptics for management of severedisruptive behavior. Psychosomatics 1986;27:17e22.

13. Yildiz A, Sachs GS, Turgay A. Pharmacological management ofagitation in emergency settings. Emerg Med J 2003;20:339e46.

14. Hughes DH, Kleespies PM. Treating aggression in the psychiatricemergency service. J Clin Psychiatry 2003;64:10e5.

15. Wright P, Birkett M, David SR, Meehan K, Ferchland I, Alaka KJ,et al. Double-blind, placebo-controlled comparison of intra-muscular olanzapine and intramuscular haloperidol in thetreatment of acute agitation in schizophrenia. Am J Psychiatry2001;158:1149e51.

16. Jones B, Taylor CC, Meehan K. The efficacy of a rapid-actingintramuscular formulation of olanzapine for positive symp-toms. J Clin Psychiatry 2001;62:22e4.

17. Breier A, Meehan K, Birkett M, David S, Ferchland I, Sutton V,et al. A double-blind, placebo-controlled doseeresponsecomparison of intramuscular olanzapine and haloperidol in the

19. Meehan K, Zhang F, David S, Tohen M, Janicak P, Small J, et al.A double-blind, randomized comparison of the efficacy andsafety of intramuscular injections of olanzapine, lorazepam, orplacebo in treating acutely agitated patients diagnosed withbipolar mania. J Clin Psychopharmacol 2001;21:389e97.

20. Meehan KM, Wang H, David SR, Nisivoccia JR, Jones B,Beasley Jr CM, et al. Comparison of rapidly acting intramus-cular olanzapine, lorazepam, and placebo: a double-blind,randomized study in acutely agitated patients with demen-tia. Neuropsychopharmacology 2002;26:494e504.

21. Allen MH, Currier GW, Carpenter D, Ross RW, Docherty JP. Theexpert consensus guideline series. Treatment of behavioralemergencies 2005. J Psychiatr Pract 2005;11. 5e108; quiz110e2.

22. Chan HY, Ree SC, Su LW, Chen JJ, Chou SY, Chen CK, et al. Adouble-blind, randomized comparison study of efficacy andsafety of intramuscular olanzapine and intramuscular halo-peridol in patients with schizophrenia and acute agitatedbehavior. J Clin Psychopharmacol 2014;34:355e8.

23. Tulloch KJ, Zed PJ. Intramuscular olanzapine in the manage-ment of acute agitation. Ann Pharmacother 2004;38:2128e35.

24. American Psychiatric Association. Diagnostic and statisticalmanual of mental disorders. 4th ed (DSM-IV). Washington, DC:APA; 1994.

25. Kay SR, Sevy S. Pyramidical model of schizophrenia. SchizophrBull 1990;16:537e45.

26. Lindenmayer JP, Brown E, Baker RW, Schuh LM, Shao L,Tohen M, et al. An excitement subscale of the Positive andNegative Syndrome Scale. Schizophr Res 2004;68:331e7.

27. Guy W. ECDEU Assessment manual for psychopharmacology.Bethesda, MD: US Dept of Health, Education, and Welfare;1976.

28. Simpson GM, Angus JW. A rating scale for extrapyramidal sideeffects. Acta Psychiatr Scand Suppl 1970;212:11e9.

29. Barnes TR. A rating scale for drug-induced akathisia. Br JPsychiatry 1989;154:672e6.

30. Hsu WY, Huang SS, Lee BS, Chiu NY. Comparison of intramus-cular olanzapine, orally disintegrating olanzapine tablets, oralrisperidone solution, and intramuscular haloperidol in themanagement of acute agitation in an acute care psychiatricward in Taiwan. J Clin Psychopharmacol 2010;30:230e4.

31. Baldacara L, Sanches M, Cordeiro DC, Jackoswski AP. Rapidtranquilization for agitated patients in emergency psychiatricrooms: a randomized trial of olanzapine, ziprasidone, halo-peridol plus promethazine, haloperidol plus midazolam andhaloperidol alone. Rev Bras Psiquiatr 2011;33:30e9.

32. Mantovani C, Labate CM, Sponholz Jr A, de AzevedoMarques JM, Guapo VG, de Simone Brito dos Santos ME, et al.Are low doses of antipsychotics effective in the management ofpsychomotor agitation? A randomized, rated-blind trial of 4intramuscular interventions. J Clin Psychopharmacol 2013;33:306e12.

Intramuscular olanzapine for acute schizophrenia 445

Intramuscular olanzapine versus intramuscular haloperidol plus lorazepam for the treatment of acute schizophrenia with agit ...IntroductionMethodsPatientsRandomization and allocation concealmentProceduresEfficacy assessmentsSafety assessmentsStatistical analysis

ResultsDemographic characteristics and clinical parameters at baselineEfficacyAdverse effects

DiscussionAcknowledgmentsReferences