Stability of medication in early psychosis: a comparison between second-generation and low-dose first-generation antipsychotics

Stability of medication in early psychosis: a comparison between second-generation and low-dose first-generation antipsychotics
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Original Article Stability of medication in early psychosis: acomparison between second-generation andlow-dose first-generation antipsychotics Stein Opjordsmoen, 1,2 Ingrid Melle, 1,2 Svein Friis, 1,2 Ulrik Haahr, 3 Jan O. Johannessen, 4 Tor K. Larsen, 2,4 Bjørn R. Rund, 5 Erik Simonsen, 3,6 Per Vaglum 7 and Thomas H. McGlashan 8 1 Department of Psychiatry, UllevålUniversity Hospital, Institutes of  2 Psychia-try,  5 Psychology and  7 Behavioural Sciencesin Medicine, University of Oslo, Oslo, 4 Division of Psychiatry, Stavanger Univer-sity Hospital, Stavanger, Norway;  3 Psychi-atric Research Unit, Zealand RegionPsychiatry,  6 Faculty of Health, University ofCopenhagen, Roskilde, Denmark; and 8  Yale Psychiatric Institute, Yale University,New Haven, Connecticut, USA Corresponding author: Dr SteinOpjordsmoen, Department of Researchand Education, Division of Psychiatry,Ullevål University Hospital, 0407 Oslo,Norway. Email: o.s.e.ilner@medisin.uio.noAffiliations exist between organizationsthat have a financial interest in thispaper’s subject matter and the followingauthors: Eli Lilly Pharmaceuticals (DrsMelle, Haahr, Larsen, Simonsen andMcGlashan), Pfizer (Dr Melle), AstraZeneca (Drs Opjordsmoen and Melle),Bristol-Myers Squibb (Dr Opjordsmoen),Lundbeck Pharma (Drs Haahr and Simon-sen) and Janssen-Cilag (Drs Opjordsmoen,Haahr and Simonsen).This paper is part of the TIPS (early Treat-ment and Intervention in PSychosis)project with the following research group:Thomas H. McGlashan, MD, Per Vaglum,MD, Svein Friis, MD, Ulrik Haahr, MD, JanOlav Johannessen, MD, Tor K. Larsen, MD,Ingrid Melle, MD, Stein Opjordsmoen,MD, Bjørn Rishovd Rund, PhD, Erik Simonsen, MD.Received 11 April 2008; accepted 10December 2008 Abstract  Aim:  This naturalistic study aims tocompare discontinuation rates forlow-dose first-generation versussecond-generation antipsychotics infirst-episode psychotic patients. Methods:  The prescription of anti-psychotic medication in 301 con-secutively admitted patients withfirst-episode psychosis from fourcatchment areas is described. Forthe first year of inclusion a first-generation antipsychotic in low dose was recommended as the firstmedication. From the second year asecond-generation antipsychotic wasrecommended as first choice. Switch-ing was allowed when any drug was judged to be ineffective or to haveserious side-effects. Switching during the first 2 years after inclusion isdescribed. Results:  Switching from a low-dosefirst-generation antipsychotic wasmore frequent than from a second-generation antipsychotic (90.7 vs.58.4%). Lack of therapeutic effect andside-effects were the more frequently recorded reasons for changing in thefirst-generation group. Akathisia, par-kinsonism, dyskinesias, dystonia anddysphoria were more often reportedin patients on first-generation drugs. Weight gain and sedation weremore often reported in patients onsecond-generation drugs. Conclusion:  The findings suggest abetter adherence to and tolerability for second-generation antipsychoticsthan for low-dose first-generationantipsychotics in first-episodepsychosis. Key words: first- and second-generation antipsychotics, first-episodepsychosis, schizophrenia, side-effect. INTRODUCTION In patients with schizophrenia, most controlledclinical trials with antipsychotic medications haveshown that antipsychotics are significantly moreeffective than placebo in the acute phase, and they also prevent relapses when used continuously over years. 1 Compliance is important because relapses Early Intervention in Psychiatry   2009;  3:  58–65  doi:10.1111/j.1751-7893.2008.00103.x © 2009 The AuthorsJournal compilation © 2009 Blackwell Publishing Asia Pty Ltd 58  tend to have longer durations and more incompleteremissions than the first episode. 2 Moreover, givenevidence from controlled drug trials in psychoticconditions other than schizophrenia, 3,4 there arereasons to believe that antipsychotics might beuseful for most patients in all categories of psychosis. 5 Nevertheless, one of the most commonproblems in clinical practice is to convince patientsto take medication as prescribed. Non-adherencerates on antipsychotics have been reported to behigh, with an average at about 50%. 6 Dolder  et al.  studied a sample of patients in needof continuous antipsychotic prescription. 7 Thepatients were on average about 50 years old, andtheir adherence was measured by analysing refillrecords for up to 12 months. The authors found amodestly better adherence to second-generation(risperidone, olanzapine, quetiapine) than withfirst-generation drugs (haloperidol, perphenazine).In another study Menzin  et al.  assessed rates of medicationadherenceduringa1-yearperiodamong outpatients who initiated therapy with a first- orsecond-generation antipsychotic (FGA or SGA) inthe last quarter of 1997. 8 Patients had schizophreniaorschizoaffectivedisorder,wereonaverage42 yearsoldandhadnoprescriptionsforthesamedrugintheprecedingyear.Theauthorsfoundthat58%ofthose who started with a first-generation agent discontin-ued the medication versus 33% of the patients whostarted with a SGA. Moreover, switching to anotherantipsychotic was found in 33% in the FGA groupcompared with 18% in the SGA group. The use of novel drugs was also associated with significantly less use of concomitant medications.Second-generation antipsychotics seem now tobe most frequently used in Western societies, butFGAs are still prescribed for a considerable propor-tion of patients. 6,9 Few studies have compared low-dose FGAs with SGAs in the treatment of psychoticpatients. In one study 12, 20 and 24 mg day  - 1 sertin-dole was compared with 4, 8 and 16 mg day  - 1 halo-peridol in chronic schizophrenia patients with ameandurationofillnessof16 years.Eventhelowesthaloperidol dose produced more extrapyramidalsymptoms than sertindole; that turned out to be theoverall better alternative than haloperidol in any dose. 10 In a study of first-episode psychosis, olanza-pine in mean modal dose of 9.1 mg day  - 1  was morefavourable than haloperidol given in mean modaldose of 4.4 mg day  - 1 in the acute phase. 11  Anotherstudy of first-episode psychosis used risperidone inmeanmodaldoseof3.3 mgday  - 1 versushaloperidol2.9 mg day  - 1 , and with better results for the SGA as regards relapse prevention and side-effects. 12  A comparison of a FGA with several SGAs wasaddressed in the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) study  13 in whichalmost 1500 patients with chronic schizophrenia were randomly assigned to receive perphenazine,olanzapine, quetiapine, risperidone or ziprasidone.Overall, 74% of the patients discontinued me-dication before 18 months, with relatively smalldifferences between the above-mentioned drugs.However, it was concluded that olanzapine was themost effective in terms of the rates of discontinua-tion. There were no differences among the othercompounds. However, olanzapine was associated with greater weight gain and increases in measuresof glucose and lipid metabolism.Beasley Jr  et al.  presented discontinuation datafrom double-blind randomized controlled trialscomparing olanzapine with other antipsychotics. 14 They found significantly greater likelihood of dis-continuation among those treated with haloperi-odol, risperidone, ziprasidone and quetiapine thanthose treated with olanzapine. However, no signifi-cant differences were observed between olanza-pine and clozapine, amisulpride, fluphenazine orperphenazine.RecentpublishedworkaboutthebenefitsofSGAsversusFGAsisconflicting.Regardingtreatmentcon-tinuation, for example, the results for perphenazineare remarkably positive. 13,14 In a study wheretreatment-resistant chronic patients were random-ized to either FGAs or SGAs (other than clozapine),FGA patients showed a trend towards greaterimprovementsinqualityoflifeandsymptomscoresafter1 year, 15 althoughmorepatientsrandomizedtoreceiving a SGA remained in their allocated treat-ment group for the whole year (65 vs. 54%). In yetanotherstudySGAswerecomparedwithhaloperidolinnewlyadmittedinpatientswithanexacerbationof psychosis. On the measure of no longer in needof acute inpatient care, haloperidol was as effectiveas olanzapine and risperidone and more effectivethan aripiprazole, quetiapine and ziprasidone. 16 Given these recent conflicting results, morestudies are needed, especially naturalistic studies infirst-episode psychosis. We had the opportunity to monitor and comparethe clinical prescription of FGA and SGA in a largestudy of patients with first-episode psychosis (theTIPS: earlyTreatment and Intervention in PSychosisstudy). 17 From January 1997 through December2000 altogether 301 patients with first-episode psy-chosis from four catchment areas (three in Norway and one in Denmark) were recruited consecutively intoastudyofearlydetection, 17,18 andtreatedundernaturalisticconditions.Treatmentwasstandardizedamong sites. In 1997, the first year of inclusion, a  S. Opjordsmoen et al. © 2009 The AuthorsJournal compilation © 2009 Blackwell Publishing Asia Pty Ltd 59  FGA in low dose was the first option in the drug treatment protocol. From 1998 and after the firstoption was a SGA. Change of medication wasallowed if the treating physician judged the effect of the drug to be inadequate or if troublesome side-effects emerged. Every patient was treated andfollowed for a total of at least 2 years.For this study we asked the following questions:(i) Does time to switching drugs differ betweenthose who receive low-dose FGAs and those whoreceive SGAs? (ii) What are the main reasons forswitching in these two groups? METHODSParticipants Detailsofthestudyaredescribedelsewhere, 17 butanoverview is outlined here. Inclusion criteria were:15–65 years of age; meeting Diagnostic and Statisti-cal Manual (of Mental Health Disorders) – FourthEdition (DSM-IV) criteria for schizophrenia, schizo-phreniform disorder, schizoaffective disorder, brief psychotic episode, delusional disorder, affectivepsychosis with mood-incongruent delusions,psychotic disorder not otherwise specified (NOS);actively psychotic; never adequately treated for psy-chosis; no neurological or endocrine disorders withrelationship to the psychosis; no contraindicationstoantipsychoticmedication;abletospeakoneoftheScandinavian languages; IQ over 70. The study wasapproved by the Regional Committee for MedicalResearch Ethics, and all patients gave writteninformed consent. A total of 301 consecutive patients were includedfrom four health-care sectors, three in Norway andone in Denmark. There was a preponderance of male patients (58.5%). In the total sample 93% hada Scandinavian background, and an average of 12 years of education. As to marital status, 72% were single, 10% divorced and only 18% married orcohabitating when included in the study. Thediagnostic distribution was: schizophrenia 28.0%,schizophreniform disorder 22.0%, schizoaffectivedisorder 12.9%, affective disorder with mood-incongruent psychotic features 14.0%, delusionaldisorder 5.6%, brief psychosis 5.6% and psychosisNOS 11.9%.Patients were followed up at 2 years, and thisreport describes the prescription of antipsychoticmedication during that time. Medication Risperidone was introduced into the Norwegianmarket in 1995, and olanzapine in 1997. However,according to the protocol for this study a low-doseFGA was recommended in 1997, with perphenazineas the first priority in Norway and zuclopenthixol inDenmark.In1998andlateraSGAwasrecommendedasfirstchoiceinthestudy.InNorwayolanzapinewasthe first option and risperidone the second. InDenmark sertindole was the first choice until it was withdrawn from the market in 1998. After that theoptions were olanzapine or risperidone. Clozapine was recommended as the third choice at all sites.Switching antipsychotics was proposed for thosepatients where the drug was judged to be ineffectiveor when side-effects were judged to be serious.First drug is shown in Table 1. A FGA was given asthe first drug in 39.2% of patients, a SGA in 57.8%,andin3.0%ofpatientsnoantipsychoticsweregivenduring the 2 years of observation.Before inclusion 84% of patients were drug naïve. Among those exposed to antipsychotics 10% hadreceived clearly inadequate doses (less than 0.33defined daily dosage (DDD); DDD for olanzapine10 mg, risperidone 5 mg, perphenazine 24 mg,zuclopenthixol 24 mg, haloperidol 8 mg). None of those patients had taken drugs for more than12 weeks or until remission of psychotic symptoms,and another 6% had received marginally adequatedoses for much less than 12 weeks (mean 4 weeks).Pharmacotherapy was started during week one in80%ofpatients,andthemeantimefrominclusiontostartofmedicationinthetotalsamplewas2.5 weeks.One case started medication in the second year. Measures The structured clinical interview for DSM-IV wasused for diagnostic purposes. 19 Symptoms were TABLE 1. First medication during 2 years n  (%)SGA Olanzapine 140 (46.5)Risperidone 28 (9.3)Sertindole 6 (2.0)FGA Perphenazine 68 (22.6)Zuclopenthixol 23 (7.6)Haloperidol 16 (5.3)Chlorpromazine 5 (1.7)Chlorprothixene 4 (1.3)Levomepromazin 1 (0.3)Pimozide 1 (0.3)No medication 9 (3.0)301 (99.9) FGA,first-generationantipsychotic;SGA,second-generationantipsychotic. Antipsychotics in first psychosis 60  © 2009 The AuthorsJournal compilation © 2009 Blackwell Publishing Asia Pty Ltd  measured with Positive and Negative SyndromeScale (PANSS). 20 Global functioning was measuredby the Global Assessment of Functioning Scale(GAF), and the scores were split into symptomscores (GAFs) and function scores (GAFf) toimprove psychometric properties. 21  All raters weretrained to reliability in the use of study instrumentsby rating previously prepared case notes andaudiotapes/videotapes before entering the study assessment teams. Reliability of measurementsrangedfromfairtoverygood. 22 Side-effectsofmedi-cationwererecordedbythetreatingphysicianusing a standardized form. Extrapyramidal symptoms were measured using the St. Hans Rating Scale forExtrapyramidal Syndromes. 23 Statistics  Analyses were made with the statistical package spss  for Windows (version 11) (SPSS, Chicago, IL,USA). Yates-corrected  c 2 -tests were used forcategorical variables, Mann–Whitney   U  -test wasused for duration of untreated psychosis (DUP) and t  -tests were used for continuous variables. A Kaplan–Meier survival analysis was carried out forthe assessment of time to switching class of drug.Two-sided tests were used, and significance wasdefined at the 5% level. RESULTS Patients in the FGA group were older and hadshorterDUPthanSGApatientsatintake(mean(SD)age 29.6 (9.5) vs. 26.5 (9.6) years,  P   <  0.01; mean(median) 40 (5) vs. 54 (12) weeks,  P   <  0.05, respec-tively). Moreover, FGA patients were more symp-tomatic according to GAF scores (GAFs mean (SD)27.6 (7.4) vs. 30.4 (6.5),  P   <  0.01; GAFf 30.0 (11.1) vs.32.6 (9.7),  P   <  0.05). No significant differences onthe PANSS positive, negative or general scores werefound. There were some more male patients in theSGA group (64 vs. 52%,  P   <  0.06). No significantdifferences were found when the two largest treat-mentgroups,thosewhoreceivedperphenazineandolanzapine as the first drug, were compared on thesame variables.Dosesinthemostuseddrugcategoriesduringthe2-year period of observation are shown in Table 2.The first-generation compounds were given in low doses, the only exception being haloperidol (notshown in the table). This drug was given to 16excitedand/oraggressivepatientsforashortperiodof time in doses up to 30 mg. By 3-month follow uponly one patient was still using haloperidol, at thattime 2 mg a day.The result of the survival analysis is shown inFigure 1, showing at which point in time treatment was switched with first- and second-generationdrugs. A second-generation drug was more likely tobe used throughout the 2-year period (log rank sta-tistic,  P   <  0.001). Figure 2 compares the findings forthe most used medications in each group, namely  TABLE 2. Doses by most used drugsAt 3 months During 2 years At 2 yearsMax dose Most used doseMean Median Median MeanOlanzapine 11.5 15.0 10.0 10.0Risperidone 3.3 4.3 4.0 3.4Perphenazine 11.8 16.0 10.0 7.0Zuclopenthixol 12.9 20.0 16.0 17.1FIGURE 1. Time in weeks to switch in first-generation antipsy-chotic (FGA) ( n  =  118) and second-generation antipsychotic(SGA) ( n  =  168) groups. 120.00100.0080.0060.0040.0020.000.00 Time in weeks    C  u  m   s  u  r  v   i  v  a   l Survival functions SGAFGA  S. Opjordsmoen et al. © 2009 The AuthorsJournal compilation © 2009 Blackwell Publishing Asia Pty Ltd 61  perphenazine and olanzapine. Perphenazine wasmore likely to be discontinued ( P   <  0.001).Table 3 shows stability of the used medicationover time. At 1-year follow up more than half thepatients who started on a SGA still used the samedrug, whereas switching to another SGA had takenplace in nearly every fifth patient. Only 14.4% of thepatients who started treatment with a FGA usedthe same drug at 1-year follow up, whereas half of the patients then used a SGA. At 2-year follow up41.6% of patients who started with a SGA still usedthe same drug whereas 26% had switched toanother SGA, and 21.4% had stopped taking drugs.Out of those who started taking a FGA, 9.3% stillused the same drug at 2-year follow up, whereas43.2% used a SGA, and 34.7% had stopped taking medication.Reasons for switching were recorded, and theresults are given in Table 4. Lack of effect and side-effects were more likely registered as reasons forchange in the FGA group. We cannot rule out thatsome of the switches from the FGA group for ‘otherreasons’ in 1997 could be associated with the intro-duction of olanzapine, and that the therapist,patient or his or her family wanted to give it a try.Table 5 shows side-effects recorded in the twogroups. Side-effects were reported as present or not.In many patients they were judged to be mild, anddid not lead to discontinuation of the antipsychoticdrug. Akathisia, parkinsonism, dyskinesias, dysto-nia and dysphoria were significantly more oftennoticed in the FGA group whereas weight gain andsedation were significantly more often recorded inthe SGA group. Anticholinergic drugs were more likely given topatients in the FGA group, otherwise there was nodifference between the groups as regards additionalmedication (Table 6). DISCUSSION The main finding in this largely representativesample of first-episode psychosis is more frequentswitching from a low-dose FGA than from a SGA.Lack of effect and side-effects were more likely to bereported as reasons for change in patients treated with FGAs than in those who received a SGA as firstmedication. Akathisia, parkinsonism, dyskinesias,dystonia and dysphoria were more likely to bereported in patients on FGAs.Weight gain and seda-tion were more likely to be reported in those whoused SGAs. Reported side-effects could not beexplained by any drug combination.McCombs  et al.  studied the use patterns for FGAsspecifically. 24 Out of about 2000 patients on average42 years old they found that only 12% recorded1 yearofuninterruptedmedication.Thisisinaccor-dance with our finding in those patients whereinitiation of therapy was with a FGA.In a series of 400 patients with early psychosisrandomized to either olanzapine, quetiapine orrisperidone, no significant differences as to discon-tinuation rates between groups were found. Only about 30% continued treatment for 1 year, 25  whichis lower than what we found for our patients treated with SGAs. Kinon  et al.  analysed treatment comple-tion in four randomized, controlled studies witholanzapine versus risperidone, quetiapine andziprasidone (flexible and fixed dose), respectively. 26 During 24–28 weeks 54% of all olanzapine-treatedpatients were completers, versus 47% in the risperi-done, 38% in the quetiapine and 37% in the ziprasi-done groups. Taken together, 48% of patients stillused the same drug after 24–28 weeks. This is alsosomewhat lower than what we found for our SGA patients at the same point in time.Our investigation is different from the CATIEstudy  13 inthatwedealtexclusivelywithfirst-episodepsychotic patients, whereas CATIE involved only patients with established schizophrenia. Our study is naturalistic and does not use a randomizedcomparison of first- and second-generation drugs.Nevertheless, like CATIE, we found greater ad-herence to a second-generation medication overtime and similar differences in side-effect profiles FIGURE 2. Time in weeks to switch in perphenazine ( n  =  66) andolanzapine ( n  =  135) groups. 120.00100.0080.0060.0040.0020.000.00 Time in weeks    C  u  m   s  u  r  v   i  v  a   l Survival functions OLAPER  Antipsychotics in first psychosis 62  © 2009 The AuthorsJournal compilation © 2009 Blackwell Publishing Asia Pty Ltd
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks