Original Author: Paul Perry, Ph.D, BCPP
Latest Revisers: Paul Perry, Ph.D., BCPP, FACCP and Kathleen
Hradek, Pharm.D.
Creation Date: 1996
Last Revision Date: January 2004
Peer Review Status: Internally Peer Reviewed
The efficacy of maintenance antipsychotic therapy in the prevention of psychotic recurrences and rehospitalization among schizophrenic patients has been well established (Johnson 1985). It is generally agreed that maintenance treatment in chronic schizophrenic patients is justified for a lengthy but still undetermined period since a significant majority of patients do derive therapeutic benefit from antipsychotics. A treatment period in excess of five years has been recommended (Johnson et al 1983). It is obvious that not all patients derive equal benefit from chronic antipsychotic therapy. It is consistently demonstrated in the literature that 30-40% of patients on maintenance therapy will eventually suffer a subsequent acute psychotic episode. However, survival curves illustrate that many of the patients who ultimately relapse have their episode delayed by antipsychotic medication (Hogarty and Ulrich 1977, Johnson 1981, Kane et al 1983). The following discussion raises the questions of who should receive long-term maintenance antipsychotic treatment, what benefits are most likely to be attained, and how the best possible results can be achieved.
APMT FOR MULTI-EPISODE (>2) PATIENTS
Relapse Rates
Viguera et al (1997) described the time course of relapse by combining discontinuation data from multiple clinical studies. A total of 1006 schizophrenic patients were included in this analysis. After abruptly discontinuing conventional antipsychotics the cumulative relapse rates were 12% at 1 month, 22% at 2 months, 34% at 3 months, 46% at 6 months and 54% at 12 months. This data suggests that the majority of schizophrenic patients will not relapse for 2-3 months (or longer) after the discontinuation of conventional antipsychotics agents.
Kissling (1991) surveyed antipsychotic efficacy studies in schizophrenics that had in common a one-year follow-up. Without continuous antipsychotic treatment the patients had a mean relapse rate of 74% within the first year, while on antipsychotic maintenance treatment (APMT) only 16% of the patients relapsed. Table 1 summarizes these findings. Despite these figures from controlled trials, in the naturalistic setting 50% of schizophrenic patients relapse within one year after their last episode. The reason for this exceedingly high relapse rate turns out to be relatively basic. A large number of schizophrenics are either not treated prophylactically with antipsychotics or do so for only a short time. The obvious explanation for this discrepancy is noncompliance. However, it turns outs that noncompliance is a two-edged sword. It includes patients, obviously, but physician noncompliance is just as serious a problem (Kissling et al 1991). A survey of German psychiatrists demonstrated that there was no consensus of agreement as to whether APMT was indicated and if indicated how long it should be continued (see Table 2).
To remedy the lack of consensus, an expert panel in Bruges, Belgium, formulated precise treatment guidelines for antipsychotic maintenance therapy in1989 (Kissling et al 1991). The Bruges panel recommended a minimum antipsychotic treatment duration for relapse prevention of 5 years for multi-episode (> 2) schizophrenic patients. For multi-episode patients, all prospective and controlled discontinuation studies have shown that, even after 5 symptom-free years of APMT, 73% of patients relapse within the first year after discontinuing antipsychotics (Kissling et al 1991). Table 3 demonstrates a mean 70% relapse rate after antipsychotic discontinuation during the next 6 to 24 months for patients receiving APMT for 6 months to 5 years. Additionally, the relapse risk is so high in the absence of APMT that only 10% of patients remain relapse-free at 5-year follow-up (Kissling et al 1991). Unfortunately this group of 10% of patients cannot be prospectively identified (Kissling et al 1991). Patients should be maintained on their effective initial dose that they responded to for a minimum period of 3-6 months. At this point, the consensus conference recommended that the dose be reduced by approximately 20% every 6 months. Table 4 presents minimal effective doses derived from a number of controlled studies involving fluphenazine decanoate (Kane et al 1983, Marder et al 1984, Marder et al 1987, Hogarty et al 1988) and haloperidol decanoate (Eklund and Forsman 1991). These doses should not be decreased before the end of the recommended treatment duration.
Dose versus Relapse Rate
Fluphenazine. Kane et al (1983) in a controlled dose study contrasted the prophylactic effect of a standard dose of fluphenazine decanoate (12.5-50 mg IM q 2 wk) with a low dose (1.25-5 mg IM q 2 wk) among 126 reasonably stable schizophrenic patients. The low-dose patients received approximately 10% of the standard dose. At the end of the one-year study, 56% of the low-dose patients relapsed compared to only 7% of the standard-dose patients. However, Hogarty (1984) reported somewhat different findings regarding low-dose antipsychotic treatment. Stable schizophrenic patients, i.e., 5-6 months post-discharge, were randomly assigned to either a standard-dose group of fluphenazine decanoate (mean = 20 mg IM q 2 wk) or to a low-dose group (mean = 4 mg IM q 2 wk). At the end of one year, 25% (5/20) assigned to the standard-dose group relapsed while a similar number 23% (5/22) assigned to the low-dose relapsed. It may well be that a fluphenazine decanoate dose of 5 mg IM q 2 wk is the minimum effective dose. Marder et al (1984) found similar relapse rates in a fluphenazine decanoate prophylaxis study of 35% and 43% in patients receiving depot doses of 5 mg and 25 mg IM q 2 wk for one year. However, when Marder et al (1987) extended the study to two years in 66 patients, a different conclusion resulted. Evaluation of the relapse rates with each dose revealed no significant difference at one year, but significantly better survival was seen with the higher dose (64%) than the 5 mg dose (31%) at two years. However, there was no significant difference in the survival rates when the clinician was permitted to make a dosage adjustment up to 10 mg in the low-dose group and 50 mg in the high-dose group when the patient demonstrated evidence of a symptomatic exacerbation. Patients assigned to the higher dose were more uncomfortable from the drug during the early months of the study according to adverse effect scores.
Haloperidol. Eklund and Forsman (1991) established the minimal effective dose of haloperidol decanoate APMT dose at 60 mg q 4 wk. Fifty-six chronic schizophrenics, after being stabilized on a haloperidol decanoate dose of 60 mg q 4 weeks for 4 months, were randomized under double-blind condition to either continued haloperidol decanoate (60 mg q 4 weeks) or placebo injections for the subsequent 48 weeks. Eight patients relapsed while another seven refused to participate any further. Of the 41 remaining patients the follow-up relapse rates were 11% (2/18) for the haloperidol group and 70% (16/23) for the placebo group. The mean steady-state haloperidol concentration was 6.7 ng/ml (day 28 trough level). Kane et al (2002) confirmed the prophylactic effectiveness of conservative doses of haloperidol decanoate. In a randomized controlled trial, 105 chronic stable outpatient schizophrenics were randomized to treatment with either haloperidol decanoate 25, 50, 100 or 200 mg/month for one year or until relapse occurred. The relapse rates were similar for the 200 mg/mo (15%), 100 mg/mo (23%), and 50 mg/mo (25%) doses versus a higher rate with the 25 mg/mo (60%) dose. There were no differences in the Simpson-Angus EPS ratings between the three highest doses.
Risperidone. Bouchard et al (2000) compared the 12-month effectiveness of risperidone (n=87) to conventional antipsychotics (n=78) in a multi-center open label study. Across the 12 months of the follow-up, the mean risperidone dose was 5.5 mg/d (range=1.2-13.4 mg/d) while the mean conventional antipsychotic dose was 1,006 mg/d (range=10-8,714 mg/d chlorpromazine equivalents). At 12 months, the percentage of responders (>20% PANSS decrease) favored risperidone, 30% versus 15%. Additionally, the total, positive, and negative symptom PANSS change scores favored risperidone.
In a randomized control trial, Csernansky et al (2002) contrasted the prophylactic effectiveness of risperidone (mean, 4.9 mg/d) to haloperidol (mean, 11.7 mg/d) in preventing relapses in 365 stable patients with schizophrenia or schizoaffective (~18%). The Kaplan-Meier estimate of risk of relapse at the end of the study was at 34% for risperidone and 60% for haloperidol. The risk of relapse was similar regardless of diagnosis. Early discontinuation occurred more frequently in the haloperidol group. The mean decrease in the severity of psychotic symptoms and EPS favored risperidone.
Olanzapine. Dellva et al (1997) studied the relapse rates in a 46-week double-blind extension study of 45 olanzapine and 13 placebo patients who met response criteria in the North-American double-blind study. These patients had received olanzapine 2.5 to 17.5 mg/day, or placebo. The mean modal maintenance dose for olanzapine-treated patients was 12.1 ± 4.9 mg/day. There was a significant difference between the survival patterns favoring olanzapine-treated patients (p=0.002). Based on survival curves, the estimated risk of relapse by one year was 29% for olanzapine and 70% for the placebo-treated patients.
Hamilton et al (2000) performed a 46-week follow-up study of schizophrenic patients who completed a 6-week controlled trial of olanzapine (n=319) versus haloperidol (n=104). The doses for both drugs were 5-20 mg/d. The survival curves (time to first hospitalization) favored olanzapine over haloperidol, with absolute re-hospitalization rates in trial completers of about 15% and 40%, respectively, by the end of the trial. The completion rates were quite low, however, at 52.0% for olanzapine and 35.6% for haloperidol. The primary difference was due to more discontinuations due to adverse events in the haloperidol group. The Quality of Life Scale response rate (> 20% improvement) demonstrated an advantage for olanzapine (70%) in contrast to haloperidol (42%). Finally more patients taking olanzapine were working either full or part-time (15% vs 5%) and socializing more than once a month (54% vs 38%).
A summary of the maintenance phase results of the olanzapine versus haloperidol trials was published (Tran et al 1998). In a pooled analysis of patients from three studies, the Kaplan-Meier survival curves for relapse were significantly different, in favor of olanzapine. The estimated one-year risk of relapse for olanzapine was 19.7% and 28% for haloperidol. This difference was found because the Kaplan-Meier curves consider time to relapse in addition to the occurrence of relapse. Relapses with olanzapine tended to occur later in treatment. However, there was no difference in the absolute relapse rates (olanzapine 14%, haloperidol 19%, p=0.097).
Ziprasidone. A 52-week randomized double-blind controlled trial assessed the efficacy of ziprasidone 40mg/d, 80mg/d or 160mg/d and placebo in 293 inpatients with chronic or sub-chronic schizophrenia (Arato et al 2000). Patients had been hospitalized for at least two months and had CGI-S scores of 5 (markedly ill) or less. They entered the study receiving treatment with other antipsychotics and were then switched to ziprasidone or placebo. Relapse was defined as a score of 6 (much worse) or greater on the CGI-I scale or a score of 6 (severe) or greater in either PANSS Item P7 (hostility) or G8 (uncooperativeness) on two successive days. The mean number of psychiatric hospitalizations in the patients prior to the start of the study was 9.9. Patients were either hospitalized or in a sheltered facility with constant medical or nursing supervision. The relapse rates by week 52 were as follows: 41% for the 40mg/d group, 35% for the 80mg/d group, 36% for the 160mg/d group, and 71% for the placebo group. Each ziprasidone group had a significantly lower relapse rate than the placebo group. Each ziprasidone treatment group was also superior to placebo for change scores on the PANSS-Total. PANSS-Negative subscale scores were also reduced in each ziprasidone treatment group versus the placebo group. Kirsch et al (2002) randomized stable chronic schizophrenics to either a 28-week maintenance course of either ziprasidone 80-160 mg/d (mean=110 mg/d) or haloperidol 5-15 mg/d (mean=9 mg/d). At the end of the trial the relapse rates (PANSS-total >60) of 38% for ziprasidone and 45% for haloperidol did not differ. Change scores on the PANSS-total, PANSS-negative, PANSS-positive, BPRS-derived core symptoms, and MADRS also did not differ.
Clozapine. Since the introduction of risperidone there have been reports of switching patients previously controlled with clozapine to a newer atypical agent. With this trend have been several case reports of rapid relapse (i.e. 1-2 weeks) after clozapine discontinuation (Seeman and Tallerico 1999). Relapse rates following discontinuation of clozapine and quetiapine are estimated to be five times greater than expected with other antipsychotics. This clinical observation is supported by receptor binding data (Seeman and Tallerico 1999). In this study, three compounds (raclopride, iodobenzamide, dopamine) were used to displace various radiolabeled antipsychotic agents from cloned D2 binding sites. The investigators found clozapine and quetiapine to be readily displaced, while comparator compounds such as haloperidol, chlorpromazine, and olanzapine were not as readily displaced. For example, dopamine displaced clozapine from its D2 binding sites 100 times faster than it was able to displace conventional agents or olanzapine. It is theorized that when clozapine (and potentially quetiapine) is discontinued, endogenous dopamine quickly displaces clozapine from its binding site leading to rapid deterioration of clinical effect. This finding is clinically significant since it explains why many patients relapsed after being directly switched from clozapine to risperidone. The manufacturer now recommends that the risperidone dose be titrated up the optimum dose of 6 mg/d and then the clozapine be titrated off over approximately a one-month period.
Meta Analysis.
Leucht et al (2003) conducted a systematic review and meta-analysis looking at relapse rates, general treatment failure, and dropout due to adverse events in randomized, controlled trials of the new-generation antipsychotic medications. They found six studies (n=983) which compared atypical antipsychotics to placebo and 11 studies (n=2032) which compared atypical antipsychotics to conventional antipsychotics. Of the studies, three were with clozapine, six with olanzapine, two with risperidone, and one with ziprasidone.
Atypical antipsychotics vs. placebo. The pooled results for olanzapine from studies by Beasley et al (2000) and Dellva et al (1997) found the risk difference for raw relapse rates to be -0.24 (CI -0.33 to -0.16, p<0.00001). The risk difference from survival curve estimates was found to be -0.45 (CI -0.55 to -0.36, p<0.00001). Olanzapine was significantly better than placebo in terms of relapse. Arato et al (2002) found the risk difference for raw relapse rates with ziprasidone to be -0.26 (CI -0.39 to -0.13, p=0.00009). The risk difference for survival curve estimates was -0.39 (CI -0.51 to -0.27, p<0.00001). Relapse rates for ziprasidone were significantly less than for placebo. In terms of overall treatment failure, 43 % of patients on atypical antipsychotics left the study early for any reason, compared to 72% of placebo treated patients. No significant differences were found in adverse effects between olanzapine and placebo or ziprasidone and placebo.
Atypical antipsychotics vs conventional antipsychotics. The pooled results for clozapine studies by Essock et al (1996), Rosenheck et al (1999 and 2000) and Tamminga et al (1994) found a risk difference for relapse of -0.08 (CI -0.19 to 0.04, p=0.18). Clozapine was not found to be significantly better than conventional antipsychotics in terms of relapse. The results of three olanzapine studies by Tran et al (1998) were pooled and for raw relapse rates the risk difference was -0.05 (CI -0.11 to 0.01, p=0.11). For survival curve estimates the risk difference was -0.09 (CI -0.16 to -0.02, p=0.01). Relapse rate for Olanzapine was only found to be significantly better than conventional antipsychotics when survival curve estimates were used. The pooled results for risperidone studies by Csernansky et al (2002) and Marder et al (2003) found the risk difference for raw relapse rates to be -0.10 (CI -0.18 to -0.02, p=0.01). The risk difference for survival curve estimates was -0.15 (CI -0.24 to -0.06, p=0.0006). Relapse rates with risperidone were significantly better than conventional antipsychotics regardless of the method used to determine relapse rate. When looking at dropouts for any reason, 49% of subjects receiving atypical antipsychotics left the study early compared to 66% of subjects taking conventional antipsychotics. There was no significant superiority found for any atypical antipsychotic compared to conventional antipsychotics for adverse effects.
Pharmacoeconomics of APMT. Essock et al (2000) conducted a 2-year open-label randomized controlled trial that compared clozapine (n=138) with physicians'-choice medications (n=89) among long-term state hospital inpatients. Both groups incurred similar costs during the study with a trend for clozapine to be less costly in the second year One important finding was that clozapine reduced the risk of rehospitalization for patients who were discharged during the observation period. It is important to remember, however, that clozapine is the most costly of the atypical antipsychotics. Thus, if other atypicals were to perform similarly to clozapine from the standpoint of effectiveness, they might show a statistically significant reduction in cost vs. typical antipsychotics.
APMT FOR FIRST-EPISODE PATIENTS
Relapse Rates
The APMT data for first-episode schizophrenics suggest that it is effective and indicated. One to two years of APMT are recommended after the remission of the first episode. Using a single blind design the Scottish Schizophrenia Research Group reported a 65% relapse rate in 23 first episode patients over the first year after discontinuing antipsychotic treatment (1988). However, of 19 other first episode patients who entered a 1-year antipsychotic maintenance trial comparing pimozide to flupenthixol, only one patient (5%) relapsed and was rehospitalized. In a double-blind, placebo-controlled trial of maintenance drug therapy in patients with a final clinical diagnosis of schizophrenia, the drug effect over 1 year was significant (p<0.01), with a relapse rate of 41% in the placebo group versus no relapses in the drug-treated group (Kane et al 1982). A subgroup of patients meeting the research diagnostic criteria of Spitzer et al had a similar result, with no relapses in the fluphenazine treated group compared with a 46% relapse rate in the placebo group. An antipsychotic discontinuation study by Johnson (1979) found that patients with first-episode schizophrenia diagnosed by Schneiderian criteria had a better prognosis for the ensuing two years if they received at least 12 months of antipsychotic medication. The relapse rate for the first-illness schizophrenics was 32% over two years after discontinuation, whereas it was 71% for the untreated chronic schizophrenics. These data are summarized in Table 5. These data however may be overly optimistic. Gitlin et al withdrew 50 stable outpatients diagnosed with first-episode schizophrenia. The patients had remained stable for approximately two years while being treated with fluphenazine decanoate. Using sensitive measures for symptom reemergence such as a 6 or 7 score on the BPRS items of hallucinations, unusual thought content, or disorganized thinking, 78% of the patients relapsed within one year while 96% relapsed within two years. The mean time to relapse was 235 days. Because of the sensitive measure of relapse and rapid reinstitution of medication only 13% (6/45) of the regularly followed patients required hospitalization.
Lieberman (1993) suggests that there are three major biological predictors of schizophrenic relapse. These include 1) a psychotic behavioral response to dopamine agonist (IV methylphenidate); 2) the presence of tardive dyskinesia; and 3) increases in anterior pituitary hormones, as shown by growth hormone increases in response to IV methylphenidate and low levels of prolactin after withdrawal from a conventional antipsychotic. All of these predictors reflect enhanced dopamine activity in the striatal, mesocortical and mesolimbic systems.
Summary
The Bruges APMT consensus conference recommended at least 1-2 years of APMT for first-episode schizophrenics. Among placebo patients, the reported average relapse rate was 57% during the first year and 60 to 90% in the second year for first-episode schizophrenics. The conference limited their duration of treatment recommendation for "at least 1-2 years" only because prospective, controlled studies over longer periods are not presently available. Thus patients must be informed that even after 1-2 years of successful APMT they still have a high relapse risk if the drug is discontinued. The Conference's recommended duration of antipsychotic treatment for relapse prevention are presented in Table 6.
IMPACT OF MAINTENANCE THERAPY
To judge the value of maintenance therapy in schizophrenic patients, one must recognize the serious consequences of a relapse in these patients. The duration of a relapse is not confined to a discrete period of rehospitalization following the reemergence of positive symptoms or behavioral changes. Some patients never regain their premorbid levels of functioning or suffer irreversible social consequences of additional episodes of illness. Johnson (1979) found that less than half of patients were able to regain their pre-relapse level of functioning within a 12-month period. Antipsychotic discontinued patients not only suffered a greater number of relapses, but the form of the relapse was more severe. These patients experienced greater risk of self-harm, antisocial behavior, rehospitalization, and decreases in social and work functioning during the subsequent 18-month period (Johnson 1983). In an acute psychotic episode it is the positive symptoms, i.e., hallucinations, delusions, thought disorder, not the negative symptoms, e.g., social withdrawal, that the antipsychotics control. Survival curve studies suggest that, even when relapse occurs, the appearance of new symptoms is probably delayed by maintenance therapy.
Heinrichs et al (1985) proposed that early recognition of decompensation of schizophrenic patients is possible and that this clinical strategy can reduce rehospitalization rates. Schizophrenics, prior to decompensation, experience a prodromal symptom phase characterized by symptoms of dysphoria, anorexia, insomnia, decreased concentration, and social withdrawal. During the prodromal period most of the patients still have enough insight into their illness to recognize that they are beginning to relapse. In a population of 32 schizophrenic and six schizoaffective patients, 24 (63%) demonstrated early insight, and of these only two (8%) were rehospitalized. Conversely, seven of 14 (50%) of the uninsightful patients required hospitalization (p<0.006). The two groups were similar with respect to age, sex distribution, social economic status, drug status, hospitalizations, and relapse severity. At the time of decompensation, 23 patients were taking antipsychotics while 16 were drug free. In either case, intervention treatment consisted of initiating or increasing antipsychotic drug therapy, psychosocial support and environmental manipulation where appropriate. Thus, clinicians ought to describe the relapse prodrome to patients so that they can contact their physician when the prodrome occurs.
The Heinrichs (1985) study posed the question as to whether patients could discontinue their medication and would only be required to restart at the first sign of the relapse prodrome. Multiple studies have concluded that the answer to this question is no. Jolley et al (1989) followed two groups of 27 schizophrenics. One group continued to receive fluphenazine decanoate at their pretrial dose while the intermittent treatment group received placebo injections. If the latter group experienced two or more days of neurotic or dysphoric symptoms they were immediately started on oral haloperidol at usually a dose of 5-10 mg/d. Unfortunately, intermittent therapy failed. These patients experienced a 30% (8/27) relapse rate in contrast to a 7% (2/27) rate in the control group. Carpenter et al (1990) also confirmed the inadequacy of intermittent therapy with in schizophrenic patients. In a 2-year single-blind study of 116 chronic schizophrenics, his group found that continuous antipsychotic medication was superior to targeted medication in preventing decompensations, hospitalizations and in extent of employment at 2 years. The targeted treatment group received medication only when they experienced prodromal symptoms of a psychotic episode. Of the targeted-medication patients 53% (30/57) were hospitalized while only 36% (21/59) of the continuous-medication patients required hospitalization. Additionally, for the 19th to the 24th month of follow-up, continuous treatment patients were employed approximately 50% of the time, which was significantly greater than the other group. Herz et al (1991), using a double-blind design, followed 101 chronic schizophrenics for two years. They were treated with either continuous antipsychotic maintenance therapy or intermittent therapy using either haloperidol, chlorpromazine, trifluoperazine, or fluphenazine. The intermittent group was treated with antipsychotics only if the patients showed any prodromal signs of relapse (nonpsychotic dysphoric symptoms and/or mild exacerbation of schizophrenic symptoms). Patients were dropped out of the study if they experienced 3 prodromal episodes in 1 year or if one episode lasted more than 9 weeks. The dropout rate for the maintenance treatment group was 14% in contrast to 46% for the intermittent treatment group. The relapse rate (GAS score < 30) was favorable in the chronic treatment group at 16% (8/51) in contrast to 30% (15/50) for the intermittent treatment group. Thus intermittent therapy has clearly been shown to be of no value in the treatment of schizophrenic patients. Schooler et al (1997) randomized 313 schizophrenic or schizoaffective patients into three double-blind fluphenazine treatment groups, continuous moderate dose decanoate (12.5-50mg q 2 weeks), continuous low dose decanoate (2.5-10mg q 2 weeks), and targeted fluphenazine therapy (only received when symptomatic). The principle finding was an increased rate of hospitalization at 2 years in patients randomized to targeted therapy; (46%) compared to 25% in both the continuous groups (p < 0.001).
When patients present themselves with the relapse prodrome, it is questionable whether the depressive symptoms should be treated with antidepressants drugs. Studies consistently demonstrate that the addition of a tricyclic antidepressant to the medication regimen of a chronic schizophrenic has either no advantage over antipsychotic treatment alone or actually can result in a worsening of the patient's thought disorder. Prusoff et al (1979), in a controlled blinded study, treated 35 chronic schizophrenics with significant depressive symptoms, as measured by the Raskin depression scale (³7, range 3-15), with either perphenazine 16-48 mg/d or perphenazine plus amitriptyline 100-200 mg/d for 5 months. The depression rating scale scores at the end of the treatment period were not different. However, the thought disorder factor (conceptual disorganization, grandiosity, hallucinations, and unusual thought content) measured on the BPRS was significantly worse in the group exposed to amitriptyline. Likewise, Kramer et al (1989) studied 84 DSM-III defined actively psychotic chronic schizophrenics with baseline BPRS scores of ³ 30 and a HAM-D scores of ³ 17. Initially, the patients were treated with 7 weeks of haloperidol 0.4 mg/kg/d. Patients with BPRS and HAM-D scores that remained elevated were treated with either haloperidol and placebo or haloperidol and a tricyclic (amitriptyline or desipramine 3.5 mg/kg/d initially for 4 weeks). There were no differences in the HAM-D and BPRS scores between the treatment groups. However, a subgroup of patients whose scores were ³ 30% worse than baseline was entirely made up of patients in the haloperidol/tricylic group. Thus in schizophrenic patients who are either in a relapsing or residual state the addition of a TCA has either no beneficial effect or a deleterious effect on the patients primary illness. Unlike the TCAs, the SSRIs, without any catecholamine agonist activity, could possibly be of benefit to schizophrenic patients on APMT. In a double-blind placebo controlled study, Spina et al (1994) randomly added either fluoxetine 20 mg/d or placebo for 12 weeks to the therapy of 34 chronic schizophrenic patients receiving APMT with conventional antipsychotics. The addition of the SSRI improved the negative symptoms (SANS and HAM-D) and had no effect on the positive symptoms (PANS). However, there were three dropouts in the fluoxetine treatment group. Thus the data suggest that SSRI antidepressant may be of benefit to some schizophrenic patients. In a controlled trial, Goff et al (1995) randomized 41 schizophrenic or schizoaffective patients to fluoxetine 20 mg/d or placebo for 6 weeks. The patients had been maintained on stable doses of either depot haloperidol or fluphenazine for the 6 months prior to randomization. Negative symptom ratings on the BPRS decreased (p<0.05) while psychosis, depression, global functioning, and EPS did not change although the overall trend was toward improvement. Interestingly, fluoxetine's metabolite was responsible for a 65% increase in the fluphenazine concentrations and a 20% increase in the haloperidol concentrations by noncompetitive inhibition. Therefore, any benefit demonstrated by the addition of fluoxetine may be a result of increased antipsychotic levels and not by direct action of the drug. Buchanan et al (1996) conducted a treatment trial in 33 schizophrenic or schizoaffective patients inadequately responding to clozapine. The patients were augmented with fluoxetine 20-80 mg/d or placebo for 8 weeks. No significant changes in the positive symptoms, negative symptoms, depressive symptoms or OCD symptoms were observed. The negative finding resulting from fluoxetine augmentation was not a function of too small a population, pharmacokinetic drug interactions or pharmacodynamic drug interactions.
When looking at adding an antidepressant to antipsychotic maintenance therapy in a patient with schizophrenia it is first important to determine exactly what is being treated. It must be determined if it is negative symptoms that are being treated or if it is a major depressive episode. This can be difficult since the two often present in a very similar way.
Several double-blind studies have looked at treating negative symptoms with antidepressants and for these studies subjects without depression were selected. Spina et al (1994) randomly added either fluoxetine 20 mg/d or placebo for 12 weeks to the therapy of 34 chronic schizophrenia patients treated with conventional antipsychotics. The addition of fluoxetine improved the negative symptoms but had no effect on positive symptoms. The risk of a drug-drug interaction in this study was high due to the fact that fluoxetine can increase antipsychotic blood levels. Therefore, it is difficult to determine if the positive effect is due to the antidepressant or to increased levels of antipsychotic. A study by Salokangas et al (1996) looked at 90 stable patients with schizophrenia on typical antipsychotics and added up to 40mg/d of citalopram or placebo for 12 weeks. There was an improvement in positive and negative symptoms in both the citalopram and placebo group, but no difference between groups. Silver and Shmugliakov (1998) studied 38 subjects with schizophrenia on a typical antipsychotic and an anticholinergic. Either fluvoxamine or maprotiline was added at a dose of up to 100mg/d for 6 weeks. They found a decrease in negative symptoms in the fluvoxamine group only. The possibility of a drug-drug interaction is also high in this study, along with fluoxetine; fluvoxamine can increase antipsychotic blood levels. Lee et al (2000) added sertraline 50mg/d or placebo to haloperidol treatment in 36 subjects with chronic schizophrenia for 8 weeks. There was no change in positive or negative symptoms in this study. A final study by Buchanan et al (1996) added fluoxetine up to 80mg/d or placebo to the treatment of 34 patients with chronic schizophrenia on clozapine. It was an 8 week study. There was no difference in positive or negative symptoms. The information from these 5 studies indicates that antidepressants, specifically SSRIs, are unlikely to improve negative symptoms in patients with schizophrenia on antipsychotic maintenance therapy. If a benefit is seen it could be due to a drug-drug interaction rather than to the antidepressant itself.
A recent systematic review and meta-analysis conducted by Whitehead et al (2003) looked at eleven small studies which investigated the use of antidepressants to treat depression in patients with schizophrenia. A meta-analysis of six studies was conducted which examined standardized mean difference in HAM-D scores. The antidepressants used in the studies were sertraline, trazodone, bupropion, or tricyclics. Using a random effects model, the standardized mean difference in HAM-D score was -0.27 (95% CI -0.69 to 0.15, p=0.2). This was not a significant difference in HAM-D scores between the placebo and antidepressant groups. There was no evidence of worsening of psychosis when the antidepressant was used during the stable phase of schizophrenia. The authors concluded that the literature available for meta-analysis was generally of poor quality. From the results of this meta-analysis it seems that the efficacy of antidepressants in treating depression co-occurring with schizophrenia remains to be proven.
OUTCOME PREDICTORS OF ANTIPSYCHOTIC MAINTENANCE THERAPY
Approximately one-third of antipsychotic-treated patients will relapse within two years of recovery from an acute episode. Also, 10-30% of chronic schizophrenic patients can survive without the benefit of antipsychotics, free of symptoms and at an equivalent level of social and work function over a similar two-year period (Rushkin and Nyman 1991). With these two radically different outcomes possible it is important to be able to prospectively identify patients who will eventually require maintenance therapy in the future. Leff and Wing (1991) reported that schizophrenic patients with a good prognosis experience more auditory hallucinations, more features of endogenous depression, shorter periods of illness and had good premorbid personalities. Thus it would seem that patients who do not fit this mold ought to be placed on antipsychotic maintenance therapy. These findings were fortified by Fenton and McGlashan (1987) who were able to demonstrate that these classic predictors of good prognosis schizophrenia were also useful in predicting sustained remission without medication. Twenty-three drug-free chronic schizophrenics who had a sustained remission following hospitalization were contrasted to 140 other outpatient schizophrenics who required medication. Significant differences were noted between the two groups in their levels of premorbid functioning. Prior to the onset of their illness, the drug-free patients functioned significantly better than the control group with respect to the social functions of heterosexual relations, social relations in latency, quality and stability of premorbid work functioning, and accrued psychosocial competence as reflected by acquisition of skills and interests. The last of these skills was the best predictor of a good prognosis. Additionally, the drug-free group demonstrated fewer hebephrenic traits and more affect as manifested by dysphoria and derealization on readmission to the hospital. A caveat to the observation of dysphoric affect on readmission is Johnson's (1988) finding that depression developing after a stable interval of greater than one year following recovery from an acute relapse predicted a significant increase in the risk of a further relapse within 2 years as compared with post-psychotic depressions occurring within the first year or no depression at all. Thus one can conclude that good prognosis schizophrenics may be able to survive without medication but this is not an absolute.
Heresco-Levy et al (1995), using discriminant function analyses, were able to characterize schizophrenic patients who were least likely to tolerate dose reduction of their APMT. Forty-one stable schizophrenics (DSM-III-R) who were maintained on APMT for at least 5 months were studied. The patients were classified as either remitted or chronically psychotic, i.e., ³ 1 psychotic symptom present such as suspiciousness, conceptual disorganization, hallucinations or unusual thoughts. In phase one of the study fluphenazine decanoate (FD) dose was decreased. If the dose was greater than FD 35 mg/4 wk it was reduced to 35 mg/4 wk and if it was < 35 mg/4 wk it was reduced to 10 mg/4 wk. The relapsers had been ill on average for 16 years versus 22 for the non-relapsers. Three relapse predictors were identified. In order of their importance they included (1) being chronically psychotic, (2) male, and (3) having an illness of a shorter duration.
RISK EVALUATION IN ANTIPSYCHOTIC MAINTENANCE THERAPY
The primary objection to placing all schizophrenic patients on APMT is the concern for increasing the risk of tardive dyskinesia (TD) as well as the other extrapyramidal adverse effects (EPS). Johnson (1973) demonstrated that with the depot antipsychotics EPS was dose dependent. Two-thirds of the EPS were abolished by simple dose adjustment without loss of therapeutic control during a 15-month follow-up study. After six months in the community, nearly half of these patients had their maintenance dose reduced significantly (Johnson 1975). After 12 months from hospital discharge they estimated that two-thirds of the patients who had survived free of relapse required no more than half their dose at the time of discharge. Garvey et al (1992) confirmed Johnson's dosage observations. An examination of haloperidol plasma levels in 51 stable schizophrenic outpatients found that 53% had levels at or below 5 ng/ml whereas only 16% had levels greater than 17 ng/ml, the putative upper limit for acute treatment.
Johnson et al (1983) discontinued maintenance therapy in a group of patients who had been stable for one to four years on antipsychotic medication. At the end of 12 months, 65% of the drug-discontinued sample had resumed medication. The average total dose prescribed for patients who resumed medication was 86% of the mean total dose prescribed for the control group who remained on drug. At the end of 18 months, 80% of the discontinuation group had resumed maintenance with depot antipsychotics. The mean total dose being prescribed for these patients was 20% greater than that for the control group. Additionally, the excessive doses prescribed for the discontinuation group were likely to be continued for many months. An analysis of extrapyramidal adverse effects (Table 7) and their treatment showed no difference between groups in prevalence of symptoms or the number of patients requiring anticholinergic medication. Both the prevalence of tardive dyskinesia and its rate of development (see Table 7) were also similar in both groups. Thus, in reality the prescribing qualms of the practicing psychiatrists were not realized by using maintenance antipsychotics in all schizophrenic patients. Thus drug holidays are of no value in reducing the risk of TD and EPS in this patient population.
SUMMARY
The recommended duration of antipsychotic treatment for relapse prevention is summarized in Table 7. These recommendations are based on the following considerations:
1. Because of an approximately 65% annual risk of relapse, untreated chronic schizophrenics are considered to be at a high risk of relapse following remission. Antipsychotic maintenance therapy for at least the first 5 years of illness is now the standard of care.
2. The relapse rates of chronic schizophrenic patients have consistently been shown to increase with treatment measures such as drug holidays, intermittent therapy, or targeted medication. Recognition of the prodromal symptoms of schizophrenic relapse can prevent relapses if social intervention and increasing of the antipsychotic dose are quickly instituted.
3. Patients receiving APMT with oral antipsychotics consistently demonstrate higher rates of relapse than patients receiving long-acting depot antipsychotics.
4. Depressive symptoms of acutely ill schizophrenic patients should not be treated with tricyclic antidepressants since they worsen or have no effect on patients' symptoms of schizophrenia. Often depressive symptoms are the first symptoms of impending relapse. If not the result of a drug interaction increasing antipsychotic blood levels, SSRI antidepressants combined with conventional antipsychotics e.g., haloperidol and fluphenazine may possibly be of value in improving negative symptoms. However, this is not the case with the atypical antipsychotics. Despite increasing the blood levels due to drug interactions, the SSRIs have no effect on symptoms of schizophrenic patients.
5. There is a subgroup of chronic schizophrenics who can survive without maintenance medication. These patients are characterized by having good premorbid histories from the standpoint of social and work functioning and heterosexual relationships.
6. In many patients, it is possible to reduce the hospital discharge dose by 20% every six months in the community without increasing the risk of relapse. Reducing the dose can improve compliance and reduce the risk of EPS and TD. Compliance may also be improved through the use of depot antipsychotics, particularly for chronic schizophrenics who lack insight into their illness and who are not motivated to take oral medications.
COST OF IM APMT
Table 8 below contrasts the cost of 4 weeks of IM APMT for haloperidol decanoate versus fluphenazine decanoate. The mg/mg ratio of 2.5 (2.5 mg: 1 mg) of haloperidol to fluphenazine was based on the results of an 8-month double blind clinical trial of APMT in 72 schizophrenic outpatients (Chouinard et al 1989). In these patients, the mean 4-week dose of haloperidol decanoate was 385 mg versus 154 mg for fluphenazine decanoate.
|
Table 1. Efficacy of antipsychotic relapse prevention in chronic schizophrenics: one year follow-up (Kissling 1991) |
|||
|
Study |
n |
placebo (%) |
antipsychotic (%) |
|
Trochinsky et al 1962 |
43 |
63 |
4 |
|
Leff and Wing 1971 |
35 |
80 |
35 |
|
Hogarty et al 1974 |
374 |
68 |
31 |
|
Chien 1975 |
47 |
86 |
12 |
|
Rifkin et al 1977 |
73 |
75 |
5 |
|
Mueller 1982 |
50 |
72 |
8 |
|
Total/Mean |
622 |
74% |
16% |
|
Table 2. Survey of German psychiatrists (n=117) concerning indication and duration of NMT (Kissling 1991) |
||
|
Case History |
|
|
|
Acute First Episode |
|
|
|
|
|
|
|
Two episodes in 1 year |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Table 3. Relapse rate following drug discontinuation among patients in long-term remission. |
||||
|
|
|
(years) |
(months) |
(%) |
|
Hogarty et al 1976 |
|
|
|
|
|
Johnson 1976 |
|
|
|
|
|
Dencker, Lepp, and Malm 1980 |
|
|
|
|
|
Cheung 1981 |
|
|
|
|
|
Johnson 1979 |
|
|
|
|
|
Wistedt 1981 |
|
|
|
|
|
Odejide and Aderounmu 1982 |
|
|
|
|
|
Summary |
|
|
|
|
|
Table 4. Minimum effective dose for NMT |
||
|
Drug |
|
|
|
fluphenazine decanoate |
|
|
|
fluphenazine |
|
|
|
haloperidol decanoate |
|
|
|
oral haloperidol |
|
|
|
Table 5. Relapse rates (%) following discontinuance of depot injections and comparison with control group (on drug) for chronic patients. |
|||
|
RELAPSE RATE (%) |
|
|
|
|
one year |
|
|
|
|
two years |
|
|
|
|
total (maximum = 4 yr) |
|
|
|
|
Table 6. Significance of Drug Discontinuation: ADRs Over 18 Months |
||
|
Item |
discontinued (N=60) |
(N=56) |
|
EPS |
|
|
|
Anticholinergic drugs |
|
|
|
TD at onset |
|
|
|
TD at 18 months |
|
|
|
All comparisons nonsignificant. |
||
|
Table 7. Recommended duration of neuroleptic treatment for relapse prevention. From Kissling (1991) |
|
|
Category |
Recommendation |
|
First-episode schizophrenia |
at least 1-2 years |
|
Multi-episode schizophrenia |
at least 5 years |
|
Patients who pose a danger to themselves or others during any one episode |
Indefinitely |
|
Table 8. Monthly cost comparison of IM depot antipsychotics using a fluphenazine decanoate to haloperidol decanoate dose ratio of 2.5 (Chouinard et al, 1989) based on 2003 drugstore.com retail prices. |
|||
|
Drug |
Cost (drugstore.com) |
Monthly Dose |
Monthly Cost |
|
Fluphenazine decanoate 25 mg/ml |
$21/5 ml vial |
25 mg q 2 wks |
$8.4/4 weeks |
|
Haloperidol Decanoate 100 mg/ml |
$184/5 ml vial |
60mg q 4 wks |
$22/4 weeks |
REFERENCES
Arato M, O'Connor R, Bradbury JE, et al. Ziprasidone in the long-term treatment of negative symptoms and prevention of exacerbation of schizophrenia. American Psychiatric Association 153rd Annual Meeting, Chicago, 2000.
Arato M, O'Connor R, Meltzer H (2002). A 1-year, double-blind, placebo-controlled trial of ziprasidone 40, 80 and 160 mg/day in chronic schizophrenia: the ziprasidone extended use in schizophrenia (ZEUS) study. Int Clin Psychopharmacol 17: 207-215.
Beasley C, Hamilton S, Dossenbach M (2000). Relapse prevention with olanzapine. (abstract) Eur Neuropsychopharmacol 10 (suppl 3): s304.
Bouchard RH, Merette C, Pourcher E, et al (2000). Longitudinal comparative study of risperidone and conventional antipsychotics for treating patients with schizophrenia. J Clin Psychopharmacol 20:295-304.
Buchanan RW, Kirkpatrick B, Bryant N, et al (1996). Fluoxetine augmentation of clozapine treatment in patients with schizophrenia. Am J Psychiatry 153:1625-7.
Carpenter WT, Hanlon TE, Heinrichs DW, et al (1990). Continuous versus targeted medication in schizophrenic outpatients: outcome results. Am J Psychiatry 147:1138-48.
Cheung HK (1981). Schizophrenics fully remitted on antipsychotics for 3-5 years--to stop or continue drugs? Br J Psychiatry 138:490-4.
Chouinard Y, Annable L, Campbell D (1989). Randomized clinical trial of haloperidol decanoate and fluphenazine decanoate in the outpatient treatment of schizophrenia. J Clin Psychopharmacol 9:247-53.
Csernansky JG, Mamoud R, Brenner R, et al (2002). A comparison of risperidone and haloperidol for the prevention of relapse in patients with schizophrenia. N Eng J Med 346:16-22.
Dellva MA, Tran P, Tollefson GD, et al (1997). Standard olanzapine versus placebo and ineffective-dose olanzapine in the maintenance treatment of schizophrenia. Psychiatr Serv 48:1571-7.
Dencker SJ, Lepp M, Malm U (1980). Clopenthixol and flupenthixol depot preparations in outpatient schizophrenics. I. A one year double-blind study of clopenthixol decanoate and flupenthixol palmitate. Acta Psychiatr Scand Suppl 279:10-28.
Eklund K, Forsman A (1991). Minimal effective dose and relapse - Double-blind trial: haloperidol decanoate versus placebo. Clin Neuropharmacol 14(suppl 2):S7-S15.
Essock SM, Frisman LK, Covell NH, et al (2000). Cost-effectiveness of clozapine compared with conventional antipsychotic medication for patients in state hospitals. Arch Gen Psychiatry 57:987-94.
Essock SM, Hargreaves WA, Covell NH, et al (1996). Antipsychotics in research and clinical settings- Clozapine's effectiveness for patients in state hospitals: results from a randomized trial. Psychopharmacology Bull 32: 683-697.
Fenton WS, McGlashan TH (1987). Sustained remission in drug-free schizophrenic patients. Am J Psychiatry 144:1306-9.
Garvey M, Wakefield DS, Perry P, et al (1992). Assessment of prophylactic haloperidol plasma levels in stable outpatient schizophrenics. Ann Clin Psychiatry 4:121-6.
Gitlin M, Nuechterlein K, Subotgnik KL (2001). Clinical outcome following neuroleptic discontinuation in patients with remitted recent-onset schizophrenia. Am J Psychiatry 158:1835.
Goff DC, Midha KK, Sarid-Segal O, et al (1995). A placebo-controlled trial of fluoxetine added to antipsychotic in patients with schizophrenia. Psychopharmacology 117:417-23.
Hamilton SH, Edgell ET, Revicki, et al (2000). Functional outcomes in schizophrenia: a comparison of olanzapine and haloperidol in a European sample. Int Clin Psychopharmacol 15:245-55.
Heinrichs DW, Cohen BP, Carpenter WT (1985). Early insight and the management of schizophrenic decompensation. J Nerv Ment Dis 173:133-8.
Heresco-Levy U, Greenberg D, Lerer B, et al (1995). Two-year trial of maintenance antipsychotic dose reduction in schizophrenic out-patients: predictors of relapse. Is J Psychiatry Relat Sci 32:268-75.
Herz MI, Glazer WM, Mostert MA et al (1991). Intermittent vs maintenance medication in schizophrenia: two-year results. Arch Gen Psychiatry 48:333-9.
Hirsch SR, Kissling W, Baumi J, et al (2002). A 28-week comparison of ziprasidone and haloperidol in outpatients with stable schizophrenia. J Clin Psychiatry 63:516-23.
Hogarty GE, Schooler NR, Uhlrich R, et al (1974). Fluphenazine and social therapy in the aftercare of schizophrenic patients. III. Adjustment of nonrelapsed patients. Arch Gen Psychiatry 31:609-18.
Hogarty GE, Ulrich RF, Mussare F (1976). Drug discontinuation among long term, successfully maintained schizophrenic outpatients. Dis Nerv Syst 37:494-500.
Hogarty GE, Ulrich RF (1977). Temporal effects of drug and placebo in delaying relapse in schizophrenic outpatients. Arch Gen Psychiatry 34:297-301.
Hogarty GE (1984). Depot antipsychotics: the relevance of psychosocial factors - a United States perspective. J Clin Psychiatry 45 (sec 2):36-42.
Johnson DAW (1973). The side effects of fluphenazine decanoate in maintenance therapy of schizophrenia.. Br J Psychiatry 123;519-22.
Johnson DAW (1975). Observations on the dose regimen of fluphenazine decanoate in maintenance therapy of schizophrenia. Br J Psychiatry 126;457-61.
Johnson DAW (1979). Further observations on the duration of depot antipsychotic maintenance therapy in schizophrenia. Br J Psychiatry 135:524-30.
Johnson DAW (1981). Studies of depressive symptoms in schizophrenia. Br J Psychiatry 139:89-101.
Johnson DAW, Pasterski G, Ludlow JM, et al (1983). The discontinuation of maintenance antipsychotic therapy in chronic schizophrenic patients: drug and social consequences. Acta Psychiatr Scand 65:339-52.
Johnson DAW (1985). Antipsychotic medication: clinical guidelines for maintenance therapy. J Clin Psychiatry 46(suppl):6-15.
Johnson DA (1988). The significance of depression in the prediction of relapse in chronic schizophrenia. Br J Psychiatry 152:320-3.
Jolley AG, Hirsch SR, McRink A, et al (1989). Trial of brief intermittent antipsychotic prophylaxis for selected schizophrenic outpatients: clinical outcome at one year. Br Med J 298:986-90.
Kane JM, Rifkin A, Quitkin F, et al (1982). Fluphenazine vs. placebo in patients with remitted, acute first-episode schizophrenia. Arch Gen Psychiatry 39:70-3.
Kane JM, Rifkin A, Woerner M, et al (1983). Low-dose antipsychotic treatment of outpatient schizophrenics. Arch Gen Psychiatry 40:893-6.
Kane JM, Davis JM, Schooler N, et al (2002). A multidose study of haloperidol decanoate in the maintenance treatment of schizophrenia. Am J Psychiatry 159:554-60.
Kissling W (1991). The current unsatisfactory state of relapse prevention in schizophrenic psychoses - suggestion for improvement. Clin Neuropharmacol 14(suppl 2):S33-S44.
Kissling W, Kane JM, Barnes TRE, et al (1991). Guidelines for antipsychotic relapse prevention in schizophrenia toward a consensus view. In: Kisslling W, ed.. Guidelines for Antipsychotic Relapse Prevention in Schizophrenia. Proc Consensus Conf, April 19-20, 1989, Bruges, Belgium. Berlin: Springer-Verlag.
Kramer MS, Vogel WH, DiJohnson C, et al (1989). Antidepressants in depressed schizophrenic inpatients: a controlled trial. Arch Gen Psychiatry 46:922-8.
Lee MS, Kim YK, Lee SK et al (1998). A double-blind study of adjunctive sertraline in haloperidol-stabilized patients with chronic schizophrenia. J Clin Psychopharmacol 18: 399-403.
Leff JP, Wing JK (1971). Trial of maintenance therapy in schizophrenia. Br Med J 3:599-604.
Leucht S, Barnes TRE, Kissling W (2003). Relapse prevention in schizophrenia with new-generation antypsychotics: a systematic review and exploratory meta-analysis of randomized, controlled trials. Am J Psychiatry 160: 1209-1222.
Lieberman JA (1993). Prediction of outcome in first-episode schizophrenia. J Clin Psychiatry 54(suppl 3):13-7.
Marder SR, Glynn SM, Wirshing DA et al (2003). Maintenance treatment of schizophrenia with risperidone or haloperidol: 2-year outcomes. Am J Psychiatry 160: 1405-1412.
Marder SR, Van Putten T, Mintz J, et al (1984). Cost and benefits of two doses of fluphenazine. Arch Gen Psychiatry 41:1025-29.
Marder SR, Van Putten T, Mintz J, et al (1987). Low and conventional dose maintenance therapy with fluphenazine decanoate: Two year outcome. Arch Gen Psychiatry 44:518-21.
Odejide OA, Aderounmu AF (1979). Double-blind placebo substitution: withdrawal of fluphenazine decanoate in schizophrenic patients. J Clin Psychiatry 43:195-6.
Prusoff BA, Williams DH, Weissman MM, et al (1979). Treatment of secondary depression in schizophrenia: a double-blind, placebo-controlled trial of amitriptyline added to perphenazine. Arch Gen Psychiatry 36:569-75.
Rifkin A, Quitkin F, Rabiner CJ, Klein DF (1977). Fluphenazine decanoate, fluphenazine hydrochloride given orally, and placebo in remitted schizophrenics. I. Relapse rates after one year. Arch Gen Psychiatry 34:43-7.
Rosenheck R, Chang S, Choe Y, et al (2000). Medication continuation and compliance: a comparison of patients treated with clozapine and haloperidol. J Clin Psychiatry 61: 382-386.
Rosenheck R, Evans D, Herz L, et al (1999). How long to wait for a response to clozapine: a comparison of time course of response to clozapine and conventional antipsychotic medication in refractory schizophrenia. Schizophrenia Bull 25: 709-719.
Rushkin PE, Nyman G (1991). Discontinuation of antipsychotic medication in older, outpatient schizophrenics. J Nerv Ment Dis 179:212-4.
Salokangas RKR, Saarijarvi S, Taiminen T et al (1996). Citalopram as an adjuvant in chronic schizophrenia: a double-blind placebo-controlled study. Acta Psychiatr Scand 94: 175-180.
Schooler NR, Keith SJ, Severe JB, et al (1997). Relapse and rehospitalization during maintenance treatment of schizophrenia: the effects of dose reduction and family treatment. Arch Gen Psychiatry 54:453-63.
Scottish Schizophrenia Research Group (1988). The Scottish first episode schizophrenia study. V. One-year follow-up. Br J Psychiatry 152:470-6.
Seeman P, Tallerico T (1999). Rapid release of antipsychotic drugs from dopamine D2 receptors: an explanation for low receptor occupancy and early clinical relapse upon withdrawal of clozapine or quetiapine. Am J Psychiatry 156:876-84.
Silver H, Shmugliakov N (1998). Augmentation with fluvoxamine but not maprotiline improves negative symptoms in treated schizophrenia: evidence for a specific serotonergic effect from a double-blind study. J Clin Psychopharmacol 18: 208-211.
Spina E, DeDomenioc P, Ruello C, et al (1994). Adjunctive fluoxetine in the treatment of negative symptoms in chronic schizophrenic patients. Int Clin Psychopharmacol 9:281-5.
Sptizer RL, Endicott J, Robins E (1977). Research Diagnostic Driteria for a Selected Group of Functional Disorders, ed 3. New York, Biometrics Research Division, New York State Psychiatric Institute.
Tamminga CA, Thaker GK, Moran M, et al (1994). Clozapine in tardive dyskinesia: observations from human and animal model studies. J Clin Psychiatry 55 (suppl B): 102-106.
Tran PV, Dellva MA, Tollefson GD, et al (1998). Oral olanzapine versus oral haloperidol in the maintenance treatment of schizophrenia and related psychoses. Br J Psychiatry 172:499-505.
Viguera AC, Baldessarini RJ, Hegarty JD, et al (1997). Clinical risk following abrupt and gradual withdrawal of maintenance antipsychotic treatment. Arch Gen Psychiatry 54:49-55.
Whitehead C, Moss S, Cardno A et al (2003). Antidepressants for the treatment of depression in people with schizophrenia: a systematic review. Psychological Medicine 33: 589-599.
Wistedt B (1981). A depot antipsychotic withdrawal study. A controlled study of the clinical effects of the withdrawal of depot fluphenazine decanoate and depot flupenthixol decanoate in chronic schizophrenic patients. Acta Psychiatr Scand 64:65-84.