Rapid Tranquillisation of Violent Or Agitated Patients in a Psychiatric

BRITISH JOURNAL OF PSYCHIATRY (2004), 185, 63^69

Rapid tranquillisation of violent or agitated patients Department of Psychiatry at the Christian Medical College, in Vellore in the southern in a psychiatric emergency setting Indian state of Tamil Nadu. The majority of patients presenting to the psychiatric emergency services of this 1800-bed teach- Pragmatic randomised trial of intramuscular lorazepam v. ing hospital were accompanied by family haloperidol plus promethazine members and were either brought directly, or were referred by general practitioners in the town or adjoining towns and villages JACOB ALEXANDER, PRATHAP THARYAN, CLIVE ADAMS, THOMAS JOHN, and from emergency services of this and CARINA MOL and JONCY PHILIP other hospitals.

Background The pharmacological Violent or aggressive behaviour is a Patient selection management of violence in people with common reason for emergency psychiatric presentations, with assaultive behaviour Consecutive patients were assessed and psychiatric disordersis under-researched. seen in 3–10% of psychiatric patients were eligible for trial entry if the attending physician felt that intramuscular sedation Aims To compare interventions (Tardiff & Sweillam, 1982; Tardiff & Koenigsberg, 1985). A haloperidol– was clearly indicated because of agitation, commonly used for controlling agitation or promethazine mix is commonly used for aggression or violent behaviour, and if the violence in people with serious psychiatric rapid tranquillisation of agitated or violent physician did not feel that either one of disorders.disorders. patients in India and Brazil (Huf et aletal,, the interventions posed an additional risk 20022002aa). Promethazine is an antihistamine for the patient. In keeping with prevailing MethodMethod Werandomised 200 peopleto that adds to the sedative effect of haloperi- clinical practice in this country, consent receive intramuscularlorazepam (4 mg) dol and prevents acute dystonic reactions was obtained from a responsible relative if patients refused, or lacked capacity to con- or intramuscular haloperidol (10 mg) plus otherwise common with the intramuscular use of haloperidol (Van Harten et aletal,, sent to treatment by virtue of severe mental promethazine (25^50 mg mix). 1999). As a haloperidol–promethazine illness. For this trial relatives were fully informed and their written consent ResultsResults At blinded assessments 4 h mix had not been evaluated previously in the context of a randomised trial, two prag- obtained; patients without a responsible later (99.5% follow-up), equalnumbersin matic trials were designed by the Tranquili- relative were excluded. This trial compared both groups (96%) were tranquiltranquilor or zazaccaoˇcçaõRa Rapida-Ensaio´pida-Ensaio ClıClınicńic TREC two low-risk interventions in common use, asleep.However, 76% given the Collaborative Group to assess this combi- the relative benefits of which are unknown. The institutional research and ethics haloperidol^promethazine mix were nation against intramuscular benzodiaze- committee approved the trial design, the asleep comparedwith 45% ofthose pines. One trial was conducted in Rio de Janeiro (TREC–Rio) and the other in India consent procedure and the form used. allocatedlorazepam (RR 2.29,95% CICI2.29,95% ¼ (TREC–India). Here we present results of 1.59^3.39;NNT¼3.2,95% CI 2.3^5.4).The TREC–India and comment briefly on those Sample size haloperidol^promethazine mix produced of TREC–Rio. Intramuscular lorazepam is From the existing literature, with tranquilli- a faster onset of tranquillisation/sedation used increasingly to handle psychiatric sation of 73% of people given benzodiaze- emergencies in India, costs the same as the and more clinical improvement over the pines and 57% given typical antipsychotics haloperidol–promethazine combination first 2 h.Neither intervention differed (Battaglia(Battaglia et aletal, 1997; JoyJoy,1997; et aletal, 2003), with (£0.17) and offers the advantage of produ- a power of 80% at 95% confidence significantly in the need for additional cing no dystonic or extrapyramidal adverse intervals and an expected precision of intervention or physicalrestraints, effects; however, its efficacy in relation to 20%, the minimum sample size required numbers absconding, or adverse effects. the antipsychotic–antihistamine combina- was 90 people per arm. tion is unclear. TREC–India was a pragma- Conclusions Both interventions are tic randomised trial undertaken in real- Randomisation and interventions effective for controlling violent/agitated worldworldconditionsconditions comparing an intramuscular combination of haloperidol plus Eligible patients were randomised to behaviour.If speed of sedationis required, promethazine v.v. intramuscular lorazepam. receive either intramuscular haloperidol the haloperidol^promethazinehaloperidol ^ promethazine (10 mg) and promethazine (25 or 50 mg) combination has advantages over mixed in the same syringe, or intramuscular lorazepam.lorazepam. METHOD lorazepam (4 mg). All doses were at the discretion of the attending doctor, although Declaration of interest None.None. SettingSetting the recommended dose was 10 mg halo- This pragmatic randomised controlled trial, peridol plus 50 mg promethazine, or 4 mg designed to include patients typical of those lorazepam. These doses were arrived at by presenting to emergency services and to prevailing clinical practice and a pilot study interfere little with routine practice, was that showed that at least 4 mg lorazepam conducted in the emergency services of the was required to achieve a similar degree

6363

Downloaded from https://www.cambridge.org/core. 28 Sep 2021 at 05:26:29, subject to the Cambridge Core terms of use. ALEXANDER ET AL

of tranquillisation as the haloperidol– asleep; in addition, the time of onset of follow-up points entered as a covariate. We promethazine mix. tranquillisation and sleep were noted. Parti- used the Mann–Whitney UU-test to compare Randomisation was according to a cipants were considered to be tranquil mean times to tranquillisation and sedation computer-generated random numbers list when they were calm and not exhibiting in the two groups, as the data did not have in varying sized blocks of less than 10 pre- agitated, aggressive or dangerous behav- a normal distribution. The kappa statistic pared by the UK collaborator. This iour. They were considered to be asleep if, was used to evaluate agreement between collaborator worked with a member of on inspection, they appeared to be sound the blinded guesses of the coordinators the TREC–India team who had no clinical asleep and were not aroused by ambient regarding treatment allocation. responsibilities in conducting the trial. The disturbances; the depth of this apparent team member and a pharmacist prepared slumber was not assessed further. They consecutively numbered opaque cardboard were also rated on the Clinical Global RESULTSRESULTS boxes, identical in appearance and weight, Impression – Severity (CGI–S) scale at on the outside of which was a form with entry, and the CGI–Improvement (CGI–I) A total of 221 patients presented to the questions to be completed by the attending scale (Guy, 1976) with respect to aggres- emergency services with disruptive behav- doctor while ‘blind’ to the contents of the sion and violence, the Simpson–Angus iour over a 5-month period in 2002. box. The boxes contained haloperidol extrapyramidal side-effects rating scale Twenty-one people could not be included (5 mg(5mg662 ampoules) plus promethazine (Simpson & Angus, 1970) and the Barnes in the trial for reasons outlined in the (50 mg(50mg661 ampoule) or lorazepam Akathisia Scale (Barnes, 1989) at each CONSORT diagram (Fig. 1). The remain- (4 mg(4mg661 ampoule), as determined by the assessment point; any other clinically im- ing 200 patients were randomised. The randomisation list, one disposable syringe portant adverse effect, especially dystonia, follow-up rate for the primary outcome at and needle and study follow-up forms. All was also noted. These assessments were 4 h was 100% for those given lorazepam those involved clinically in the study had conducted only on participants who were and 99% for the antipsychotic–antihistamine no indication of what medicines were in awake, as extrapyramidal symptoms are combination. Follow-up rates for second- the boxes until they were opened. usually not apparent during sleep or, in ary outcomes and at 2 weeks were 92% the case of dystonia or akathisia, are likely for those allocated to lorazepam and Procedure to prevent sleep. Other outcomes within the 90% for people given the antipsychotic– Once eligibility of a patient was ensured, first 4 h were the use of additional medi- antihistamine combination. the next consecutive box was taken from cation for control of agitated or aggressive All patients allocated to the lorazepam the emergency cupboard and this consti- behaviour, the use of physical restraints, arm received 4 mg of the drug and everyone tuted randomisation. The duty doctor the need for further medical attention and randomised to haloperidol–promethazine recorded the severity of the episode and numbers absconding. Participants were also received 10 mg of haloperidol combined the initial diagnosis on the form stuck to followed up 2 weeks later to check for with 50 mg (96/100) or 25 mg (4/100) the outside of the sealed intervention pack. adverse effects or adverse outcomes and promethazine. The box was then opened and the interven- compliance with oral medication. The Most patients were male, were diag- tion administered. The patient was then primary outcome was ‘tranquil or asleep nosed to have mania (ICD–10; World followed up at 15, 30, 60 and 120 min by by 4 h’.h’.by4 Health Organization, 1992) and were rated as markedly or severely ill (Table 1). Nine- the treatment team and at 240 min and at Data analyses 2 weeks by the study coordinators. Data teen patients with severe depression with were also obtained from the case notes as We used double data entry and analysed agitation, psychotic or suicidal behaviour well as from interviews with relatives and data using the Statistical Package for Social were judged to require parenteral medi- the treatment team. Sciences (SPSS) version 9.0 for Windows. cation to prevent harm to themselves or We assessed the adequacy of randomisation others. Groups were evenly balanced on Blinding by comparing participants’ baseline socio- the numbers on psychotropic medication, demographic and clinical characteristics. mean age and CGI mean scores. The study was blind until the point of treat- We compared proportions tranquillised, The study coordinators accurately ment assignment, which minimised selec- asleep, improved (CGI much and very guessed allocation for 58% of those given tion bias. After assignment, rating for the much improved, stipulated in the trial pro- lorazepam and 33% of those given the first 2 h was not blind as the management tocol), requiring restraints, requiring the antipsychotic–antihistamine combination team had to know the prescribed medica- doctor to be recalled and requiring addi- ((kk770.68).0.68). tions. In any event, TREC–India evaluated tional sedation, all using the chi-squared Equal numbers of people (96%) were real-world interventions that are not given test, with a continuity correction, or Fish- ‘tranquil or asleep’ by 4 h. The combination blind. The study coordinators, however, er’s exact test, as appropriate. We cal- treatment, however, resulted in more who were blind to interventions given, culated relative risks and an absolute people being tranquil/asleep by 15 min, undertook ratings at 240 min. At this time, measure, the number-needed-to-treat 30 min, 1 h and 2 h (Table 2). Haloperidol they also guessed the allocated intervention, (NNT), and their 95% confidence intervals plus promethazine was also superior to to assess their blinding. (Altman, 1998) using intention-to-treat lorazepam in inducing sleep. The 40% analysis. We also used repeated measures difference in favour of the antipsychotic– Outcomes analysis of variance to compare mean antihistamine mix at 15 min increased to Patients were rated at each assessment CGI–I scores between groups across var- 47% by 30 min but receded to 31% by point on whether they were tranquil or ious time points, with being asleep at the 4 h. That the combination treatment

Downloaded from https://www.cambridge.org/core. 28 Sep 2021 at 05:26:29, subject to the Cambridge Core terms of use. RRAPID APID TRTRANQUILLISATION ANQUILLIS ATION

produced faster onset of tranquillisation/ sedation was additionally evident when mean time to onset of tranquillisation/ sleep and mean time to sleep were compared (Table 3). Four people given lorazepam were never tranquil, compared with one allocated to the haloperidol– promethazine mix. Twenty-three people given lorazepam failed to sleep at all during the 4-h follow-up compared with only eight in the combination group. The haloperidol–promethazine combination also resulted in greater numbers of people being rated as clinically improved (Table 2). Compared with lorazepam, the 31% difference at 15 min in favour of the antipsychotic–antihistamine combination receded to 14% at 2 h. By 4 h there was no difference in CGI scores between the two interventions. The mean scores on the CGI–I scale over the 4 h of follow-up were entered into repeat measures analysis of variance (Table 4). The CGI scores showed significant differences over time as well as between groups. When being asleep was entered as a covariate to control for differential seda- Fig. 11Fig. CONSORT diagram forTREC^India. tive effects between drugs on clinical improvement ratings, CGI scores continued to show differences over time, but the Ta b l e 11Tab Baseline demographic and clinical characteristics differences in CGI scores between drugs administered was not significant. LorazepamLorazepam Haloperidol+promethazine Despite the superiority of the anti- ((nn¼100) ((nn¼100) psychotic–antihistamine combination in producing sedation, the two interventions Male 64 55 did not differ in proportions of people Clinical diagnosis (ICD^10) requiring restraint, exhibiting further epi- Schizophrenia 17 20 Acute psychosis 717 155sodes of agitation or violence and needing Mania 53 44 additional medication, or in requiring the Depression 11 8 duty doctor to be recalled (Table 2), nor Substance misuse 828 2 were there differences for the outcomes of OtherOther 414 111admitted or discharged after 4 h, and lost to follow-up over 4 h and 2 weeks. No On medication differences were evident between inter- Anticonvulsants 787 8 ventions in those with different clinical Anticholinergics 777 7 Antidepressants 10 7 diagnoses, or with respect to age or gender Antipsychotics 27 26 (data available on request). None of those Benzodiazepines 14 9 given the combination reported any adverse Beta-blockers 101 0 effects, whereas one person given loraze- Lithium 777 7 pam, who had a history of bronchial asthma, complained of moderate worsening Clinical Global Impression ^ Severity of respiratory difficulty and another re- Moderately ill 10 19 ported nausea and dizziness following the Markedly illillMarkedly 55 52 Severely ill 35 27 administration of the benzodiazepine. Extremely ill 020 2 Two people given lorazepam scored 10 and 18 on the Simpson–Angus scale for extrapyramidal side-effects before the intervention, with no change in scores post- Age (mean (s.d.)) 32.2 (10.6)(10.6)32.2 30.9 (8.7)(8.7)30.9 intervention. No other patient scored above CGI^Severity score (mean (s.d.)) 5.25 (0.63)(0.63)5.25 5.125.12 (0.80)(0.80) zero on the extrapyramidal or akathisia

6565

Downloaded from https://www.cambridge.org/core. 28 Sep 2021 at 05:26:29, subject to the Cambridge Core terms of use. ALEXANDER ET AL

Ta b l e 2 Main outcomes

OutcomesHaloperidol+promethazine LorazepamLorazepam PP Relative risk NNTNNT ((nn¼100) ((nn¼100) (95% CI)CI)(95% (95% CI)CI)(95%

Tranquil/asleep 15 min 89 78 0.041.1 (1.01^1.29)(1.0 1^1.29) 9.1 (4.7^??)) 30 min 9595 81 550.011.2 (1.06^1.30)7.1 (4.4^19.2) 60 min 98 9090 0.041.1 (1.01^1.17)12.5 (6.4^77.7) 120 min 97 88 0.031.1 (1.02^1.19)11.1 (5.9(5.9^62.4) ^ 62.4) 240 min 96 9696 1.001.0 (0.94^1.06) Asleep 15 min 45 55 550.019.0 (3.73^21.73)2.5 (2.0^3.5) 30 min 69 22 550.013.1 (2.12^4.64)2.1 (1.7^2.9) 60 min 67 32 550.012.1 (1.52^2.88)2.9 (2.1^4.6) 120 min 69 3939 550.00.01 11.8 (1.34^2.34)3.3 (2.3^5.9) 240 min 76 45 550.011.7 (1.32^2.15)3.2 (2.3^5.7) Clinically improved11 15 min61 30 550.012.0 (1.45^2.85)3.2 (2.3^5.8) 30 min 83 58 550.00.01 11.4 (1.18^1.73)4.0 (2.7^8.0) 60 min 80 6060 550.00.01 11.3 (1.11^1.61)5.0 (3.1^13.7) 120 min 88 74 0.011.2 (1.04^1.37)7.1 (4.1^32.2) 240 min 87 86 0.841.0 (0.91^1.13)100 (9.4^??)) In physical restraints 15 min 11 19 0.110.6 (0.29^1.15)49.6 (5.6^??)) 30 min 11 20 0.120.120.6 (0.28^1.09) 88.4 (5.2^(5.2^88.4 ??)) 60 min 10 1818 0.100.6 (0.27^1.14)57.3 (5.6^??)) 120 min 10 13 0.510.8 (0.35^1.67) 16.4 (8.2^(8.2^16.4 ??)) 240 min 919 1110.810.82 (0.35^1.89) 15.0 (9.3^(9.3^15.0 ??)) Additional medicines 15 min 000 0 30 min 010 1 60 min 131 300.03.030.3 (0.04^3.15) 50.0 (13.3^(13.3^50.0 ??)) 120 min 343 411.00.000.8 (0.17^3.27)100 (14.0^??)) 240 min 898 911.00.000.9 (0.36^2.21)100 (10.9^??)) Other outcomes within 4 h Doctor recalled 13 18 0.300.300.7 (0.37^1.39) 20 (6.7^20(6.7^??)) Any adverse effect 202 0 Missing data 15 min 010 1 30 min 010 1 60 min 101 0 120 min 353 500.72.720.6 (0.15^2.44) 240 min 101 0 After 4 h Admitted 43 51 0.260.260.8 (0.63^1.13)17.4 (4.7^??)) Discharged 52 46 0.321.1 (0.85^1.50) 16.7 (5.2^(5.2^16.7 ??)) Further observation 434 311.00.001.3 (0.31^5.81)100 (14.0^??)) Lost to follow-up 101 0 At 2 weeks Lost to follow-up 10 808 0.62.621.3 (0.51^3.04)50 (13.7^??)) No serious adverse outcome 90 92 0.811.0 (0.90^1.07)50 (13.7^??)) Taking oral medication 87 92 0.360.95 (0.86^1.04)27 (7.3^??))

NNT, number needed to treat. 1. Clinical Global Impression ^ Improvement scale dichotomised; much and very much improved.

Downloaded from https://www.cambridge.org/core. 28 Sep 2021 at 05:26:29, subject to the Cambridge Core terms of use. RRAPID APID TRTRANQUILLISATION ANQUILLIS ATION

Ta b l e 33Tab Time until onset of tranquillisation and sleep agitated patients likely to be seen in every- day clinical practice. It compared two in- Time, min (mean (s.d.))Mann^Whitney U PP expensive, commonly used interventions for clinically relevant outcomes and lost LorazepamHaloperidol+promethazine data on only one person (0.5%) for the primary outcome and on 18 people for Time to tranquillisation47.8 (46.7) 29.7 (35.6)(35.6)29.7 327.00.0001 the 2-week follow-up (9%). Although both Time to sleep 80.6 (64.3) 37.4 (42.9)(42.9)37.4 1893.5 550.0001 interventions are effective for controlling agitated or violent behaviour, with over 75% in each group tranquil/asleep within Ta b l e 4 Clinical Global Improvement (CGI) scale and antipsychotic and benzodiazepine com-com- 15 min of administration and 96% in each scores over 4 h binations (Allen, 2002; McAllister- group tranquil/asleep by 4 h, 10 mg of Williams & Ferrier, 2002). More recent intramuscular haloperidol combined with strategies include longer-acting drugs CGI scores (mean (s.d.))11 25–50 mg promethazine is superior to such as zuclopenthixol acetate (Coutinho 4 mg intramuscular lorazepam in the speed LorazepamLorazepam Haloperidol+ et aletal, 2000) and rapidly acting intra- of onset of sleep and thereby clinical nn¼100100 promethazine muscular formulations of the atypical improvement. If lorazepam is used alone a nn¼100 antipsychotics olanzapine (Jones et aletal,, significant proportion of people remain 2001) and ziprasidone (Brook et aletal, 2000).,2000). awake for longer, potentially exposing 15 min2.97 (1.01)(1.0 1) 2.48 (0.85)(0.85)2.48 The recommendations of guidelines for everyone to danger. TREC–India did not 30 min2.49 (1.10) 1.89 (0.70)(0.70)1.89 the management of psychiatric emergencies find, however, that the reduced ability of 6060min min 2.42 (0.88)(0.88)2.42 2.09 (0.59)(0.59)2.09 (Royal College of Psychiatrists, 1998; lorazepam to sedate compared with the Expert Consensus Guideline Group, 1999) 120 min2.24 (1.07) 2.01 (0.95)(0.95)2.01 combination was accompanied by a greater are not evidence-based (Allen, 2002), or 240 minmin240 1.91 (0.67) 1.82 (0.99)(0.99)1.82 need for subsequent intervention or harm are they followed uniformly (Pilowsky etet to the patient or others. 1. Repeated measures analysis of variance; alal, 1992; Cunnane, 1994; Binder & TREC–Rio (Huf et aletal, 2002,2002aa,,bb; TREC;TREC difference in CGI scores over time: FF¼35.57, PP550.001; McNeil, 1999; Huf et aletal, 2002,2002aa). Evidence).Evidence difference in CGI scores between groups: FF¼3.74, Collaborative Group, 2003) randomised PP¼0.005; withwith‘being ‘being asleep’asleep’entered entered as a covariate from randomised trials and systematic 301 people over 6 months to receive either difference in CGI scores over time: FF¼32.46,32.46, PP550.001; reviews is limited and does not indicate difference in CGI scores between groups: FF¼1.8 6,6,1.8 a haloperidol–promethazine mix or intra- PP¼0.115. the superiority of zuclopenthixol acetate muscular midazolam. Though midazolam over conventional antipsychotics (Fenton is available in India, it is five-times as scales before or after the intervention. No et aletal, 2003), or the commonly used combi- expensive as the haloperidol and pro- patient developed dystonia. nation of haloperidol and benzodiazepines methazine mix and is not in common use over haloperidol alone (Battaglia et aletal,, in psychiatry. 1997). There is a suggestion that benzo- DISCUSSION Midazolam consistently induced more diazepines are superior to typical anti- rapid tranquillisation and sedation than the psychotics (Battaglia et aletal, 1997; Allen, It is estimated that approximately 15 haloperidol–promethazine mix. However, 2002) and that haloperidol is superior to million people in India suffer from serious the combination treatment in TREC–Rio placebo (Joy et aletal, 2003).,2003). mental disorders (schizophrenia 2.7/1000, was less tranquillising/sedating than in affective disorders 12.3/1000 and organic TREC–India (Table 5). Similar numbers of psychoses 0.4/1000; Reddy & Chandrase- Management of violence in middle- people in both trials were male, markedly khar, 1998). Although systematically ascer- and low-income countries ill and psychotic. In TREC–India, however, tained prevalence data for violence among The management of aggressive or violent everyone allocated combination treatment people with psychiatric disorders in low- psychiatric patients in India includes ‘talking receivedreceived 10 mg haloperidol. In the TREC– and middle-income countries are scant, down’ techniques, physical restraint and Riohaloperidol–promethazine arm, 77/148 there is no evidence to suggest that the seclusion, as well as the use of medication. (52%) were given 5 mg haloperidol and prevalence of violent or agitated behaviour The initial minutes and hours are crucial 71/148 (48%) were given 10 mg. Most is likely to be any less in low-income and drugs that rapidly render people tranquil people allocated to the combination countries such as India than elsewhere. and/or sedated without producing distressing treatment in both studies were given The magnitude of the problem faced by or dangerous adverse effects are desirable. In 50 mg promethazine. Subgroup analysis of emergency services in India is therefore low-income countries such as India, the high the two different doses of haloperidol in readily apparent. cost of zuclopenthixol acetate precludes its TREC–Rio, however, did not suggest widespread use; intramuscular atypical anti- differences in numbers tranquil/asleep psychotics are not available and are likely (Evandro Coutinho, personal communica- Management of agitation to be prohibitively expensive. tion, 2003). However, evaluation of and violence in emergency whether the dose of haloperidol matters settings: lack of consensus will require a direct comparison with an Drugs commonly used to manage agitation TREC ^India^ India and TREC ^Rio^ Rio adequately large sample. and violence in emergency situations world- TREC–India, the largest and only study for TREC–India did not compare halo- wide include antipsychotics, benzodiazepinespinesbenzodiaze this comparison, randomised violent or peridol alone with a haloperidol–lorazepam

6767

Downloaded from https://www.cambridge.org/core. 28 Sep 2021 at 05:26:29, subject to the Cambridge Core terms of use. ALEXANDER ET AL

Ta b l e 5 Proportion asleep if given haloperidol plus promethazine CLINICAL IMPLICATIONS

TREC^IndiaTREC^Rio && Intramuscular lorazepam (4 mg) is as effective as haloperidol (10 mg) plus promethazine (25/50 mg) in controlling violence or agitation in people with mental nn (%) nn (%)(%) disorders.

15 min 45 (45)45(45) 20 min20min 43 (29)(29)43 && If rapid sedation is required, the haloperidol^promethazine combination is 30 min69 (69)40min 40 min 69 (46) superior to lorazepam. 6060min min 67 (67)(67)67 60 min83 (55) && Pragmatic randomised trials of interventions relevant to low-income countries, 120 min 88 (88)88(88) 120 minmin120 95 (63)95(63) with limited funding, clinically meaningful outcomes and low attrition rates, are possible within the field of mental health.

LIMITATIONS

combination, but adding promethazine to && Assessments over the first 2 h were not blind and were carried out by multiple haloperidol could be superior to adding raters.raters. lorazepam for promoting tranquillisation and sedation and superior to haloperidol && The effects of both interventions could be dose-related. alone for preventing extrapyramidal && Haloperidol alone or in combination with a benzodiazepine was not evaluated. adverse effects such as acute dystonia or akathisia (Salzman et aletal, 1991; Battaglia et aletal, 1997; Brook et aletal, 2000). No serious adverse effects, particularly those related to the extrapyramidal system, were reported for either treatment, although the JACOB ALEXANDER, MD,PRATHAP THARYAN, MRCPsych, Department of Psychiatry,Christian Medical moderate worsening of respiratory diffi- College,Vellore, India; CLIVE ADAMS, MRCPsych,Cochrane Schizophrenia Group and Academic Unit of Psychiatry and Behavioural Sciences,University of Leeds,UK;THOMAS JOHN, MD,Department of Psychiatry, culty reported with lorazepam is in keeping Christian Medical College,Vellore, India; CARINA MOL, Medical Documentation Programme,University of with the known association of benzodiaze- Ulm,Germany; JONCY PHILIP,MA, Department of Psychiatry,Christian Medical College,Vellore, India pines with respiratory depression. About 15% of people in this trial were CorrespCorrespondence:ondence: Dr Jacob Alexander,Department of PsychPsychiatry,Christianiatry,Christian Medical College,Vellore physically restrained and less than 10% 632002,Tamil Nadu,Nadu,India.Tel: India.Tel: 91 416 2262603, ext. 4259; fax: 91 416 2261632; e-mail: were given additional medication over the dralexander______in@@yahoo.comyahoo.com 4 h. This common practice in India and (First received 30 June 2003, final revision 26 November 2003, accepted 15 December 2003) Brazil (Huf et aletal, 2002,2002aa) of physically restraining disruptive patients after administration of a parenteral drug as opposed to administering additional medication REFERENCES Cunnane, J. G.(19 (1994) 94) Drug management of disturbed requires evaluation. behaviour by psychiatrists. Psychiatric Bulletin,, 1818,, 138 ^ 139. The two treatment regimens evaluated Allen, M. H.(2002) Managing the agitated psychotic patient: a reappraisal of the evidence. Journal of Clinical in this study are inexpensive, effective and PsychiatryPsychiatry,, 6161 (suppl.14),11^20. Expert Consensus Guideline Group(1999) available worldwide. Where rapid sedation Treatment of schizophrenia 1999.The expert consensus Altman, D. G.(19 (1998) 9 8) Confidence intervals for the guideline series. Journal of Clinical Psychiatry,, 60 (suppl. is needed a combination of intramuscular number needed to treat. BMJ,, 317317,1309^1312.,1309^1312. 11), 33^80. ^ 8 0. haloperidol and promethazine is superior Barnes, T. R. E.(19 (1989) 8 9) A rating scale for drug-induced Fenton, M., Coutinho, E. S. F. & Campbell, C. (2003)(2003) to intramuscular lorazepam. akathisia. British Journal of Psychiatry,, 154,672^676., 672^676. Zuclopenthixol acetate in the treatment of acute Battaglia, J., Moss, S., Rush, J., et aletal (19 9 7) schizophrenia and similar serious mental illnesses. Haloperidol, lorazepam, or both for psychotic agitation? Cochrane Library,issue1.Oxford:UpdateSoftware. A multicenter, prospective, double-blind, emergency Guy,W.Guy, W. (19 (1976) 76) ECDEU Assessment Manual for ACKNOWLEDGEMENTS department study. American Journal of Emergency Psychopharmacology revised edn. Rockville, MD: National MedicineMedicine,, 1515, 335^340.,335^340. Institute of Mental Health. The TREC^India team acknowledges the contribu- Binder,R.L.&McNeil,D.E.(1999)Binder,R.L.&McNeil,D.E.(1999) EmergencyEmergency tion of Giselle Huf in the design of TREC^ Rio, on psychiatry: contemporary practices in managing acutely Huf,G.,Coutinho,E.S.F.,Fagundes,H.M.Jr,Huf, G., Coutinho, E. S. F., Fagundes, H. M. Jr, et aletal which this trial is modelled.We thank Evandro Cou- violent patients in 20 psychiatric emergency rooms. (2002aa)) Current practices in managing acutely tinho for sharing unpublished data from TREC ^ Rio. Psychiatric Services,, 50,1553^1554., 1553^1554. disturbed patients at three hospitals in Rio de Janeiro ^ Brazil: a prevalence study. BMC Psychiatry,, 22,4. This trial would not have been possible but for the Brook, S., Lucey, J.V. & Gunn, K. P., et aletal (2000) considerable support and involvement of many Intramuscular ziprasidone compared with intramuscular Huf, G., Coutinho, E. S. F. & Adams, C. E. (2002bb)) doctors, nurses and support staff at the department haloperidol in the treatment of acute psychosis. Journal ofofJournal TREC-Rio trial: a randomised controlled trial for rapid of psychiatry.We thank Mr N. Ravi for help with Clinical Psychiatry,, 6161,,933^941. 933^941. tranquillisation for agitated patients in emergency preparing the TREC boxes. This trial was funded by psychiatric rooms [ISRCTN44153243]. BMC Psychiatry,, Coutinho, E., Fenton, M., Adams, C., et aletal (2000) 22,,11. 11. intramural research grants from the Fluid Research Zuclopenthixol acetate in psychiatric emergencies: Fund of the Christian Medical College, Vellore and looking for evidence from clinical trials. Schizophrenia Jones, B.,B.,Taylor,C. Taylor,C. C. & Meehan, K. (2001)(2001) TheThe the Cochrane Schizophrenia Group general fund. Research,, 46,111^118. efficacy of a rapid-acting intramuscular formulation of

Downloaded from https://www.cambridge.org/core. 28 Sep 2021 at 05:26:29, subject to the Cambridge Core terms of use. RRAPID APID TRTRANQUILLISATION ANQUILLIS ATION

olanzepine for positive symptoms. Journal of Clinical Royal College of Psychiatrists(19 (1998) 9 8) Management of Tardiff, K. & Koenigsberg, H.W.(19 (1985) 8 5) AssaultiveAssaultive Psychiatry,, 62 (suppl. 2), 22^24. Imminent Violence: Clinical Practice Guidelines to Support behavior among psychiatric outpatients. American Mental Health Services (Occasional Paper OP41). Journal of Psychiatry,, 142142, 960^963.,960^963. Joy, C. B., Adams, C. E. & Lawrie, S. M. (2003)(2003) London: Royal College of Psychiatrists. Haloperidol versus placebo for schizophrenia. CochraneCochrane Library,issue1.Oxford:UpdateSoftware. TREC Collaborative Group (2003)(2003) RapidRapid Salzman, C., Solomon, D., Miyawaki, E., et aletal (19 91) tranquillisation for agitated patients in emergency McAllister-Williams, R. H. & Ferrier, I. N.(2002) Parenteral lorazepam versus parenteral haloperidol for psychiatric rooms: a randomised trial of midazolam Rapid tranquillisation: time for a reappraisal of options the control of psychotic disruptive behavior. Journal ofofJournal versus haloperidol plus promethazine. BMJBMJ,, 327327,, for parenteral therapy. British Journal of Psychiatry,, 180180,, Clinical Psychiatry,, 5252,177^180.,177^180. 708^713.708^713. 485^489.485^489.

Pilowsky,L.Pilowsky, L. S., Ring, H., Shine, P.J., et aletal (19 92)92)(19 RapidRapid Simpson, E. N. & Angus, J.W. F. (1970)(1970) Aratingscale Van Harten, P.N., Hoel, H.W. & Kahn, R. S.(1999) tranquillisation. A survey of emergency prescribing in a for extrapyramidal side-effects. Acta Psychiatrica Acute dystonia induced by drug treatment. BMJBMJ,, 319319,, general psychiatric hospital. British Journal of Psychiatry,, Scandinavica Supplementum,, 212212,11^19. 623^626. 160,,831^835. 831^835. Reddy, M.V. &Chandrasekhar,C. R.(19 (1998) 9 8) Prevalence Tardiff, K. & Sweillam, A.(19 (1982) 82) Assaultive behavior WorldHealthOrganization(1992)WorldHealthOrganization(19 92) The ICD ^10^10The of mental and behavioural disorders in India: a meta- among chronic inpatients. American Journal of Psychiatry,, Classification of Mental and Behavioural Disorders: Clinical analysis. Indian Journal of Psychiatry,, 4040,149^157.,149^157. 139,212^215.,212^215. Descriptions and Diagnostic Guidelines..Geneva:WHO. Geneva: WHO.

Downloaded from https://www.cambridge.org/core. 28 Sep 2021 at 05:26:29, subject to the Cambridge Core terms of use.