010302 a comparison of risperidone and haloperidol

The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne A COMPARISON OF RISPERIDONE AND HALOPERIDOL FOR THE PREVENTION
OF RELAPSE IN PATIENTS WITH SCHIZOPHRENIA
JOHN G. CSERNANSKY, M.D., RAMY MAHMOUD, M.D., M.P.H., AND RONALD BRENNER, M.D., ABSTRACT
Background
lifetime prevalence of 0.7 percent in the Unit- of maintenance treatment in patients with psychotic ed States1 and with serious physical, social, and disorders. We performed a long-term comparison of economic consequences.2 The economic bur- a newer, atypical antipsychotic drug, risperidone, and den of schizophrenia on society was estimated as an older, conventional neuroleptic drug, haloperidol, $33 billion in the United States in 1990.3 Much of in terms of the rate of relapse in patients with schizo- this cost was attributed to the consequences of psy- phrenia and schizoaffective disorder.
chotic relapse.4 The course of schizophrenia varies,5 Methods
but most patients have a chronic course with frequent 40 sites, we randomly assigned adult outpatients in relapses, typically characterized by exacerbation of stable condition with chronic schizophrenia or schizo- psychosis and rehospitalization. Successive relapses can affective disorder to receive treatment with flexibledoses of either risperidone or haloperidol for a mini- reduce the degree and duration of the next remis- sion, worsen disability, and increase refractoriness to Results
future treatment.6 To prevent relapse, maintenance domization, data from 2 were excluded because they treatment with antipsychotic drugs is obligatory for did not receive study medication; data from all 30 most patients who have schizophrenia or schizoaffec- patients from one site were excluded by the sponsor, tive disorder.7 Nevertheless, long-term outcomes have the Janssen Research Foundation, because of con- cern about the integrity of the data. The median du- Drugs referred to as atypical antipsychotic drugs ration of treatment was 364 days in the risperidone are antagonists at both dopamine and serotonin re- group and 238 days in the haloperidol group (P= ceptors in the central nervous system; in contrast, 0.02). Of the 177 patients assigned to risperidone andthe 188 assigned to haloperidol who remained in the conventional agents act predominantly at the dopa- analysis, 44.1 percent and 52.7 percent, respectively, mine receptor. Although considered a class of drugs, discontinued treatment for reasons other than relapse.
these atypical agents (clozapine, risperidone, olanza- The Kaplan–Meier estimate of the risk of relapse at pine, and quetiapine) differ in their receptor-binding the end of the study was 34 percent for the risperi- profiles10 and clinical effects. Some authors have ex- done group and 60 percent for the haloperidol group pressed doubt that atypical antipsychotic drugs offer (P<0.001); the risk ratio for relapse with haloperidol, any advantages beyond improved tolerability11,12 and from the Cox model, was 1.93 (95 percent confidence therefore argue for the continued use of convention- interval, 1.33 to 2.80; P<0.001). Early discontinuation of treatment for any reason was more frequent among The results of short-term studies indicate that all haloperidol-treated patients (risk ratio, 1.52; 95 per-cent confidence interval, 1.18 to 1.96). Patients in the available atypical antipsychotic drugs (clozapine, ris- risperidone group had greater reductions in the mean peridone, olanzapine, and quetiapine) are more effec- severity of both psychotic symptoms and extrapy- tive than the conventional antipsychotic drug halo- ramidal side effects than those in the haloperidol peridol for the treatment of “negative” schizophrenic symptoms, such as withdrawal from social interactions Conclusions
and blunted emotional expression.13-16 For “positive” schizophrenia or schizoaffective disorder have a low- symptoms, such as hallucinations and delusions, ris- er risk of relapse if they are treated with risperidone peridone was found to be superior to haloperidol in than if they are treated with haloperidol. (N Engl J Med a combined analysis of large controlled trials.17 We therefore hypothesized that treatment with risperi- Copyright 2002 Massachusetts Medical Society.
From Washington University School of Medicine and Metropolitan St. Louis Psychiatric Center, St. Louis (J.G.C.); Janssen Research Founda-tion, Titusville, N.J. (R.M.); and St. John’s Episcopal Hospital, Far Rockaway,N.Y. (R.B.). Address reprint requests to Dr. Csernansky at the Department ofPsychiatry, Washington University School of Medicine, Box 8134, 660 S. Eu-clid Ave., St. Louis, MO 63110-1081, or at csernanj@medicine.wustl.edu.
*Investigators in the study are listed in the Appendix.
16 · N Engl J Med, Vol. 346, No. 1 · January 3, 2002 · www.nejm.org
Downloaded from www.nejm.org by MR KEVIN PATRICK on October 07, 2003.
Copyright 2002 Massachusetts Medical Society. All rights reserved.
R I S P E R I D O N E O R H A L O P E R I D O L F O R T H E P R EV E N T I O N O F R E L A P S E I N PAT I E N T S W I T H S C H I Z O P H R E N I A
done would be superior to haloperidol in reducing Assessment of Outcomes
the risk of relapse among outpatients with schizo- We calculated relapse rates and time to first relapse. Relapse was defined by any one of the following: psychiatric hospitalization;an increase in the level of psychiatric care (e.g., from clinic visits to day treatment) and an increase of 25 percent from base line inthe total score on the Positive and Negative Syndrome Scale,20 or Patients
an increase of 10 points if the base-line score was 40 or less (total Eligibility criteria included an age of 18 to 65 years, a diagnosis possible scores range from 30 to 210, with higher scores indicat- of schizophrenia or schizoaffective disorder according to the cri- ing greater severity of symptoms); deliberate self-injury; suicidal teria of the Diagnostic and Statistical Manual of Mental Disorders, or homicidal ideation that was clinically significant in the investi- fourth edition (DSM-IV),19 and inpatient psychiatric hospitaliza- gator’s judgment; violent behavior resulting in clinically signifi- tion, daytime psychiatric hospitalization, outpatient crisis man- cant injury to another person or property damage; or substantial agement, or short-term treatment in a psychiatric hospital emer- clinical deterioration, defined as a change score of 6 (“much worse”) gency room within the 24 months before study entry. All patients or 7 (“very much worse”) on the Clinical Global Impressions had received a stable dose of antipsychotic medication for at least Scale (possible scores range from 1 to 7, with a score of 4 indi- 30 days before entry, had resided at the same address for at least cating no change, 1 to 3 improvement, and 5 to 7 worsening).
30 days before entry, and were judged clinically stable by the prin- Secondary outcome measures included the total score and five cipal investigator at each site. Exclusion criteria included another factor scores from the Positive and Negative Syndrome Scale, in- current DSM-IV Axis I diagnosis, an Axis II diagnosis of border- cluding positive symptoms, negative symptoms, anxiety–depression, line personality disorder or antisocial personality disorder, current hostility–excitement, and disorganized thoughts.
substance dependence or abuse, clinically significant or unstable Safety assessments included monitoring for adverse events, a bat- medical illness, current treatment with clozapine, a history of refrac- tery of standard laboratory tests, electrocardiography, and physical toriness to antipsychotic drugs, and treatment with depot neuro- examination. Extrapyramidal symptoms were assessed with the Ex- leptic injections within one treatment cycle before screening. Pa- trapyramidal Symptom Rating Scale (total possible scores range tients who were allergic to either risperidone or haloperidol and from 0 to 162, with higher scores indicating greater severity of women who were pregnant or nursing were also excluded.
adverse effects). A patient questionnaire is included within this The study protocol was approved by an institutional review board scale; the scores range from 0 to 36, with higher scores indicating at each site, and each patient or his or her legal guardian provided greater perception of adverse effects.
written informed consent for participation.
Statistical Analysis
Procedure
The sample sizes were selected to make possible the detection of a 15 percent difference in relapse rates after one year with 80 During the period from May 1996 to September 1998, pa- percent power and a two-tailed level of significance of 0.05 (165 tients were randomly assigned to double-blind treatment with ris- patients per treatment group). The sample sizes were then increased peridone (Risperdal, Janssen, Titusville, N.J.) or haloperidol (Hal- to 207 per group to account for a 20 percent rate of dropout for dol, Ortho-McNeil, Raritan, N.J.). To minimize possible adverse reasons other than relapse. Primary analyses were performed on all events associated with the discontinuation of medication that the subjects who underwent randomization and were assessed at least patients were taking before the trial began, the gradual reduction once during drug treatment. Base-line characteristics and duration of such antipsychotic drugs was permitted during days 1 to 7. Dur- of treatment were compared between the two groups by analysis ing days 1 to 3, doses were increased from 1 mg to 4 mg of ris- of variance or the Cochran–Mantel–Haenszel test.23 Duration of peridone per day and from 2 mg to 10 mg of haloperidol per day.
treatment and follow-up were summarized by descriptive statistics.
Both drugs were given once daily. From day 8 onward, investiga- Differences in time to relapse between treatment groups were tors at each site adjusted the doses of identical-appearing tablets analyzed with use of a Cox proportional-hazards model and a log- (range of permissible doses, 2 to 8 mg of risperidone per day and rank test23 with control for site and sex. Follow-up for end points 5 to 20 mg of haloperidol per day) to maximize clinical benefits ceased after the discontinuation of treatment, and data for the and minimize adverse events. Lower doses (e.g., 1 mg of risperi- analysis of time to relapse were therefore censored at the time of done per day or 2.5 mg of haloperidol per day) were permitted discontinuation of treatment. Risk ratios and reductions in risk (with in rare cases. Concomitant medications were not allowed, except corresponding confidence intervals) were derived with use of the for antacids; acetaminophen; propranolol, benztropine mesylate, Cox model. Differences between the groups in the degree of change biperiden, or procyclidine for extrapyramidal symptoms caused by in scores on the Positive and Negative Syndrome Scale and Ex- treatment; chloral hydrate, zolpidem, or flurazepam to improve trapyramidal Symptom Rating Scale from base line to the last as- sleep; and lorazepam for agitation, in doses not exceeding 4 mg sessment were studied by analysis of covariance. All statistical tests per day for no more than four days in any seven-day period.
Patients were assessed weekly during the first four weeks of the trial and then every four weeks until the last patient enrolled had completed one year of treatment. Because of concern that the pros-pect of discontinuation of treatment and dismissal from the study A total of 397 patients were recruited at 40 sites could influence reporting of relapses, a patient who had a relapse and randomly assigned to double-blind treatment with could attend the remaining scheduled study visits if both the pa- either risperidone or haloperidol. The data on two tient and the investigator wished. However, all patients who hada second relapse were removed from the trial and no longer fol- patients assigned to risperidone who did not receive study medication were excluded from analysis. The The investigators at participating academic institutions contrib- data from all 30 patients at another site were also ex- uted to the design of the study, had full access to the study data cluded, because the principal investigator at that site base, and were involved in data analysis and interpretation. Inves- did not conduct the trial in a manner that was con- tigators from the Janssen Research Foundation, the developer ofboth risperidone and haloperidol, also participated in the design, sistent with restrictions previously placed on him by analysis, and reporting of the trial.
the Food and Drug Administration (FDA) after an N Engl J Med, Vol. 346, No. 1 · January 3, 2002 · www.nejm.org · 17
Downloaded from www.nejm.org by MR KEVIN PATRICK on October 07, 2003.
Copyright 2002 Massachusetts Medical Society. All rights reserved.
The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne inspection by the FDA. The restrictions were in-tended to ensure the integrity of the data in the trial.
TABLE 1. BASE-LINE CHARACTERISTICS OF THE 365 PATIENTS.*
The decision to exclude the data from this site wasmade by the quality-assurance department of Jans- RISPERIDONE
HALOPERIDOL
sen Research Foundation and not by a committee of CHARACTERISTIC
the authors. The quality-assurance department did not know to which treatment patients had been as- signed. Exclusion of the data from this site did not change the findings regarding the primary end point (data not shown). Data from the remaining 365 pa- tients were included in the final analysis.
Characteristics of Patients and Withdrawals
The characteristics of the 365 patients in the two treatment groups were similar (Table 1). Most patients had had a diagnosis of schizophrenia and had been ill for more than a decade. The median duration of treat- ment was 364 days (range, 3 to 799) in the risperi- done group and 238 days (range, 4 to 794) in the hal- operidol group (P=0.02 by analysis of variance).
The rate of premature discontinuation of study *Plus–minus values are means ±SD. Differences were not significant be- treatment for any reason was higher in the haloper- tween the groups. Possible scores on the Positive and Negative Syndrome idol group than in the risperidone group (risk ratio, Scale for Schizophrenia (PANSS) range from 30 (no symptoms) to 210(extremely severe symptoms). Possible scores for positive symptoms range 1.52; 95 percent confidence interval, 1.18 to 1.96).
from 8 to 56, scores for negative symptoms from 7 to 49, scores for disor- The reasons for discontinuation, other than relapse, ganized thoughts from 7 to 49, scores for hostility–excitement from 4 to28, and scores for anxiety–depression from 4 to 28. In each case, higher were similar in the two treatment groups: the patient’s scores indicate greater severity of symptoms. DSM-IV denotes Diagnostic choice in 18.1 percent in the risperidone group and and Statistical Manual of Mental Disorders, fourth edition.
17.6 percent in the haloperidol group; adverse eventsin 12.4 percent and 15.4 percent, respectively; lossto follow-up in 5.1 percent and 4.8 percent; poorcompliance in 2.8 percent and 5.3 percent; inade-quate response in 1.1 percent and 3.7 percent; admin- cent of patients in the haloperidol group (75 of 188) istrative reasons in 2.8 percent and 1.1 percent; and had relapsed (Fig. 1). The Kaplan–Meier estimate of other reasons in 1.7 percent and 4.8 percent.
the risk of relapse was 34 percent (95 percent con-fidence interval, 28 to 43 percent) in the risperidone Dosage and Compliance
group and 60 percent (95 percent confidence interval, The means (±SD) of the modal daily doses were 50 to 70 percent) in the haloperidol group (P<0.001).
4.9±1.9 mg of risperidone and 11.7±5.0 mg of hal- The risk of relapse was substantially higher among operidol. The modal daily doses of risperidone were patients assigned to haloperidol (risk ratio, 1.93; 95 less than 4 mg in 12.4 percent of patients, 4 to 6 mg percent confidence interval, 1.33 to 2.80; P<0.001).
in 68.4 percent, and 8 mg in 19.2 percent. The For patients assigned to risperidone, the relapse rate modal daily doses of haloperidol were 2.5 to 5 mg represented a 48 percent reduction in risk (95 per- in 19.1 percent of patients, 7.5 to 10 mg in 44.1 per- cent confidence interval, 25 to 64 percent) as com- cent, and 15 to 20 mg in 36.7 percent.
Almost all patients were found to be compliant The Kaplan–Meier estimates of the relapse rates with the study medication regimen, as assessed by pill were similar in patients with schizophrenia (34 per- count (97.0 percent of patients taking risperidone and cent for the risperidone group and 59 percent for the 96.0 percent of patients taking haloperidol). Com- haloperidol group) and in those with schizoaffective pliance was also high according to pill count in the disorder (34 percent and 62 percent, respectively).
30 days before relapse among patients who had relaps- The subtypes of relapse were similar in the two groups es (95.3 percent of the risperidone group and 95.2 of patients: psychiatric hospitalization in 44 percent percent of the haloperidol group were compliant).
and 48 percent, respectively; substantial clinical de-terioration in 36 percent and 29 percent; an increase in level of care in 18 percent and 19 percent; and sui- At the end of the study, 25.4 percent of patients cidal or homicidal ideation in 2 percent and 4 percent.
in the risperidone group (45 of 177) and 39.9 per- Before the study, 25 percent of patients were re- 18 · N Engl J Med, Vol. 346, No. 1 · January 3, 2002 · www.nejm.org
Downloaded from www.nejm.org by MR KEVIN PATRICK on October 07, 2003.
Copyright 2002 Massachusetts Medical Society. All rights reserved.
R I S P E R I D O N E O R H A L O P E R I D O L F O R T H E P R EV E N T I O N O F R E L A P S E I N PAT I E N T S W I T H S C H I Z O P H R E N I A
Figure 1. Kaplan–Meier Analysis of Time to Relapse in Patients Assigned to Risperidone or Haloperidol.
ceiving risperidone, 27 percent haloperidol, and 48 toms were not improved over base line. In post hoc percent other conventional antipsychotic medications.
analyses, patients in the risperidone group had a sig- To assess the effect of the change in treatment caused nificant improvement from base line in symptoms af- by random assignment to another drug in these sta- ter one week of double-blind treatment: the changes ble outpatients, patients assigned to a new treatment in total scores on the Positive and Negative Syndrome were compared with patients whose treatment was Scale were a decrease of 3.7 in the risperidone group unchanged. Switching therapy had no effect on the (P<0.001) and a decrease of 1.1 in the haloperidol estimated relapse rates at the end of the study; 29 percent of patients who were switched from haloper-idol to risperidone had relapsed, as compared with Extrapyramidal Symptoms
60 percent of patients switched from risperidone to The severity of extrapyramidal symptoms was re- haloperidol; 28 percent of patients who continued duced from base line to the last recorded value in to take risperidone had relapsed, as compared with the risperidone group and increased in the haloperi- 60 percent who continued to take haloperidol.
dol group (Table 2). Differences between the groupswere significant on each measure on the Extrapyram- Symptom Scales
idal Symptom Rating Scale. Antiparkinsonian drugs The overall reduction in symptoms was smaller were prescribed for 30 consecutive days for twice as than that reported in previous studies of patients many patients assigned to haloperidol (33 of 188 with acute exacerbations of schizophrenia, as would [17.6 percent]) as patients assigned to risperidone be expected in patients with stable disease. Significant (16 of 177 [9.0 percent], P=0.02 by the Cochran– differences between subjects assigned to risperidone Mantel–Haenszel test). The new onset of tardive dys- and those assigned to haloperidol were seen in total kinesia was reported in one patient assigned to risperi- scores on the Positive and Negative Syndrome Scale done (0.6 percent) and five assigned to haloperidol and in four of the five factor scores at the last study rating (Fig. 2). In the risperidone group, improve-ments from base line to one year or to the last study Adverse Events and Body-Weight Changes
rating were seen in total scores and in positive symp- Adverse events were identified in 89.8 percent of toms, negative symptoms, disorganized thoughts, and patients assigned to risperidone and 91.0 percent of anxiety–depression. In the haloperidol group, symp- those assigned to haloperidol. The events reported N Engl J Med, Vol. 346, No. 1 · January 3, 2002 · www.nejm.org · 19
Downloaded from www.nejm.org by MR KEVIN PATRICK on October 07, 2003.
Copyright 2002 Massachusetts Medical Society. All rights reserved.
The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne Figure 2. Mean (±SE) Changes from Base Line to the End of the Study in Total and Factor Scores on the Positive and
Negative Syndrome Scale for Schizophrenia in Patients Assigned to Risperidone or Haloperidol.
P values were derived by analysis of covariance. Improvements in symptoms are shown as negative change scores andworsening of symptoms as positive change scores.
in more than 10 percent of subjects in at least one both early and late improvements in symptoms over- group were somnolence (14 percent with risperidone all, as well as an amelioration of extrapyramidal symp- and 25 percent with haloperidol), agitation (10 per- toms. In contrast, patients receiving haloperidol had cent and 18 percent, respectively), and hyperkinesia a slight worsening of both psychotic and extrapy- (5 percent and 20 percent). There was a mean in- ramidal symptoms. These findings are similar to those crease in body weight of 2.3 kg (5.0 lb) in patients previously reported for an eight-week comparative assigned to risperidone — similar in magnitude to trial of risperidone and haloperidol.14,17 In our trial, the weight gain seen in short-term studies — and a compliance with the study medication, which can also mean decrease of 0.73 kg (1.6 lb) in patients as- influence the rate of relapse,25 was similar in the two treatment groups. Improvements in cognition or oth-er symptoms of schizophrenia produced by risperi- DISCUSSION
done but not well assessed by the Positive and Neg- Among patients with clinically stable chronic schizo- ative Syndrome Scale may also have contributed to phrenia or schizoaffective disorder, the risk of relapse the reductions in the rate of relapse.
was significantly lower during treatment with ris- Receptor profiles and mechanisms of action vary peridone than during treatment with haloperidol. The among atypical antipsychotic agents.10 Therefore, oth- benefit with risperidone was substantial. The means er agents should be assessed individually with regard of the modal daily doses of risperidone (4.9 mg) and to their ability to prevent relapses.
haloperidol (11.7 mg) were similar to those used in The longer duration of drug treatment with ris- clinical practice. The relapse rate among subjects re- peridone than with haloperidol appeared to be the ceiving haloperidol (39.9 percent) was similar in mag- result of lower rates of relapse in patients receiving nitude to that found previously among patients re- risperidone, since the rates of other reasons for dis- ceiving conventional antipsychotic agents.4,24 continuation were similar in the two treatment groups.
The reduced risk of relapse with risperidone could Analysis of the time to relapse (Fig. 1) suggested that be due to that drug’s superior efficacy, better tolera- the benefits of risperidone over haloperidol appeared bility, or both. Patients who received risperidone had early and grew progressively larger throughout the 20 · N Engl J Med, Vol. 346, No. 1 · January 3, 2002 · www.nejm.org
Downloaded from www.nejm.org by MR KEVIN PATRICK on October 07, 2003.
Copyright 2002 Massachusetts Medical Society. All rights reserved.
R I S P E R I D O N E O R H A L O P E R I D O L F O R T H E P R EV E N T I O N O F R E L A P S E I N PAT I E N T S W I T H S C H I Z O P H R E N I A
Supported by Janssen Research Foundation. Dr. Csernansky has received research grant support and honorariums for delivering lectures from Jans- TABLE 2. SCORES ON THE EXTRAPYRAMIDAL SYMPTOM RATING
sen Research Foundation, Eli Lilly, AstraZeneca Pharmaceuticals, and Pfiz- SCALE AT BASE LINE AND LEAST-SQUARES MEAN CHANGES er Pharmaceuticals. Dr. Brenner has received research support from Janssen FROM BASE LINE TO THE LAST RECORDED VALUE.* Research Foundation, Eli Lilly, AstraZeneca Pharmaceuticals, and PfizerPharmaceuticals.
RISPERIDONE
HALOPERIDOL
APPENDIX
P VALUE†
The following investigators participated in the Risperidone-USA-79 Study: F. Adan, Miami; D. Brown, Austin, Tex.; J. Chou, New York; C.
Cohn, Houston; M. DePriest and B. Cole, Las Vegas; L. Dunn, Durham, N.C.; W. Goodman and M. Byerly, Gainesville, Fla.; A.I. Green and D.A.
Klegon, Boston; R.M. Hamer and M. Menza, Piscataway, N.J.; M. Ham- ner, Charleston, S.C.; H. Harsch, Milwaukee; G.G. Jaskiw, Brecksville, Ohio; B. Johnson, Houston; A. Kiev, Englewood, N.J.; I.S. Kolin, Winter Park, Fla.; M.A. Knesevich, Dallas; A. Kopelowicz, Mission Hills, Calif.; D. Levinson, Philadelphia; H.E. Logue, Birmingham, Ala.; M.J. Miller, In- dianapolis; R. Nakra, Chesterfield, Mo.; G. Pahl, Oklahoma City; M. Plop- per, San Diego, Calif.; S. Preskorn, Wichita, Kans.; G. Simpson, Los An- geles; S. Strakowski, Cincinnati; M. Thomas, Denver; S.A. West, Orlando, Fla.; J. Yaryura-Tobias, Great Neck, N.Y.; and T.P. Yoo, Detroit.
REFERENCES
1. Kendler KS, Gallagher TJ, Abelson JM, Kessler RC. Lifetime preva-
lence, demographic risk factors, and diagnostic validity of nonaffective psy-chosis as assessed in a US community sample: the National Comorbidity *Possible scores on the total Extrapyramidal Symptom Rating Scale Survey. Arch Gen Psychiatry 1996;53:1022-31.
range from 0 to 162, scores on the parkinsonism subscale from 0 to 48, 2. Buckley PF. Treatment of schizophrenia: let’s talk dollars and sense. Am
Global Impression scores for severity of parkinsonism from 0 to 8, Global J Manag Care 1998;4:369-83. [Erratum, Am J Manag Care 1998;4:611.] Impression scores for severity of dyskinesia from 0 to 8, and scores on the 3. Rice DP. The economic impact of schizophrenia. J Clin Psychiatry
patient questionnaire from 0 to 36. In all cases, higher scores indicate more severe symptoms. Values for scores are means ±SD. Values for changes in 4. Weiden PJ, Olfson M. Cost of relapse in schizophrenia. Schizophr Bull
scores are means ±SE. Data were available for 173 patients in the risperi- done group and 187 patients in the haloperidol group for all scores except 5. Andreasen NC. Symptoms, signs, and diagnosis of schizophrenia. Lan-
the patient-questionnaire scores, for which data were available for 173 pa- tients in the risperidone group and 183 patients in the haloperidol group.
6. Wyatt RJ. Neuroleptics and the natural course of schizophrenia.
Schizophr Bull 1991;17:325-51.
†P values are for differences between the groups and were derived by 7. Kane JM. Pharmacologic treatment of schizophrenia. Biol Psychiatry
‡The total score is for parkinsonism plus dystonia plus dyskinesia.
8. Herz MI, Lamberti JS, Mintz J, et al. A program for relapse prevention
§P=0.02 for the comparison with the base-line value.
in schizophrenia: a controlled study. Arch Gen Psychiatry 2000;57:277-83.
9. Ayuso-Gutiérrez JL, del Rio Vega J. Factors influencing relapse in the
¶P=0.002 for the comparison with the base-line value.
long-term course of schizophrenia. Schizophr Res 1997;28:199-206.
10. Richelson E. Preclinical pharmacology of neuroleptics: focus on new
generation compounds. J Clin Psychiatry 1996;57:Suppl 11:4-11.
11. Mattes JA. Risperidone: how good is the evidence for efficacy?
Schizophr Bull 1997;23:155-61.
12. Idem. Olanzapine on trial. Am J Psychiatry 1998;155:153.
trial. Thus, the full clinical benefits of treatment with 13. Kane J, Honigfeld G, Singer J, Meltzer H. Clozapine for the treat-
an antipsychotic drug should be judged over an ex- ment-resistant schizophrenic: a double-blind comparison with chlorpro- mazine. Arch Gen Psychiatry 1988;45:789-96.
14. Marder SR , Meibach RC. Risperidone in the treatment of schizophre-
A limitation of our study is the number of patients nia. Am J Psychiatry 1994;151:825-35.
who left the trial. To the extent that such discontin- 15. Beasley CM Jr, Tollefson G, Tran P, Satterlee W, Sanger T, Hamilton
S. Olanzapine versus placebo and haloperidol: acute phase results of the
uation is dependent on the treatment assignment, bias North American double-blind olanzapine trial. Neuropsychopharmacology may be introduced. The reasons for discontinuation did not differ between treatment groups, and the dif- 16. Arvanitis LAS, Miller BG. Multiple fixed doses of “Seroquel” (que-
tiapine) in patients with acute exacerbation of schizophrenia: a comparison
ferences between the groups in the effects of treat- with haloperidol and placebo. Biol Psychiatry 1997;42:233-46.
ment appeared early, grew consistently with time, and 17. Marder SR , Davis JM, Chouinard G. The effects of risperidone on the
five dimensions of schizophrenia derived by factor analysis: combined re-sults of the North American trials. J Clin Psychiatry 1997;58:538-46.
In our study, patients with chronic but stable [Erratum, J Clin Psychiatry 1998;59:200.] schizophrenia were randomly assigned to treatment, 18. Marder SR. Antipsychotic drugs and relapse prevention. Schizophr
Res 1999;35:Suppl:S87-S92.
in contrast to studies in which patients successfully 19. Diagnostic and statistical manual of mental disorders, 4th ed.:
treated in a short-term trial were continued on the DSM-IV. Washington, D.C.: American Psychiatric Association, 1994.
same antipsychotic agent.26,27 Our results demonstrate 20. Kay SR , Fiszbein A, Opler LA. The Positive and Negative Syndrome
Scale (PANSS) for schizophrenia. Schizophr Bull 1987;13:261-76.
that substantial reductions in the risk of relapse can 21. Guy W, ed. ECDEU assessment manual for psychopharmacology. Rev.
be achieved in such patients with the use of risperi- ed. Rockville, Md.: National Institute of Mental Health, Psychopharmacol- done, even in comparison with the use of an effec- ogy Research Branch, 1976. (DHEW publication no. (ADM) 76-338.)
22. Chouinard G, Ross-Chouinard A, Annable L, Jones BD. Extrapyram-
idal Symptom Rating Scale. Can J Neurol Sci 1980;7:233. abstract.
N Engl J Med, Vol. 346, No. 1 · January 3, 2002 · www.nejm.org · 21
Downloaded from www.nejm.org by MR KEVIN PATRICK on October 07, 2003.
Copyright 2002 Massachusetts Medical Society. All rights reserved.
The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne 23. Lang TA, Secic M. How to report statistics in medicine: annotated
26. Moller H-J, Gagiano CA, Addington DE, von Knorring L, Torres-
guidelines for authors, editors, and reviewers. Philadelphia: American Col- Plank JF, Gaussares C. Long-term treatment of chronic schizophrenia with risperidone: an open-label, multicenter study of 386 patients. Int Clin Psy- 24. Hogarty GE, Ulrich RF. The limitations of antipsychotic medication
on schizophrenia relapse and adjustment and the contributions of psycho- 27. Dellva MA, Tran P, Tollefsen GD, Wentley AL, Beasley CM Jr. Stand-
social treatment. J Psychiatr Res 1998;32:243-50.
ard olanzapine versus placebo and ineffective-dose olanzapine in the main- 25. Fenton WS, Blyler CR , Heinssen RK. Determinants of medication
tenance treatment of schizophrenia. Psychiatr Serv 1997;48:1571-7.
compliance in schizophrenia: empirical and clinical findings. Schizophr Bull 1997;23:637-51.
Copyright 2002 Massachusetts Medical Society.
22 · N Engl J Med, Vol. 346, No. 1 · January 3, 2002 · www.nejm.org
Downloaded from www.nejm.org by MR KEVIN PATRICK on October 07, 2003.
Copyright 2002 Massachusetts Medical Society. All rights reserved.

Source: http://www.hawaii.edu/hivandaids/Comparison%20of%20the%20Prevention%20of%20Relapse%20in%20Patients%20with%20Schizophrenia.pdf