Reference: Kahn R, Fleischhacker W, Boter H, et al. Effectiveness of Antipsychotic Drugs in First-Episode Schizophrenia and Schizophreniform Disorder: An Open Randomised Clinical Trial. Lancet 2008 March; 371: 1085-1097.

Clinical Question: Are second generation antipsychotics (SGA) as effective as low dose haloperidol for the treatment of first episode schizophrenia schizophreniform, and schizoaffective disorder when using less restrictive inclusion criteria with long f/u to assess efficacy and tolerability based on continuation of medication?

Background: SGAs intended to be more efficacious with fewer motor side effects. Previous comparison trials have used restrictive inclusion criteria (over-representation of men and under-representation of pts with comorbidities (i.e. SA)), measured treatment in scales rather than effectiveness and tolerability. In addition, previous studies used high doses of haloperidol and had brief study durations of 2 months.

•Randomization blinded to 5 groups: No, pts and treating physicians were not blinded to assigned treatment. Unclear if raters blinded.

•Intention to Treat: Yes, all pt analyzed in groups which randomized and all discontinued pts accounted for. See Figure 1.

•Groups treated equally: Yes, all had f/u at one yr. It did not appear that any groups received co-interventions by study design, however, given non-blinded nature of study it is possible that expectations led to dissimilarities in treatment

•Design: Open randomized trial of haloperidol vs. SGAs from 2002-2006 at 50 sites, including 36 university hospitals in 14 countries

•Inclusion Criteria: 498 pts. 18-40 yrs with MINI Plus confirmed schizophrenia, schizophreniform, or schizoaffective disorder

•Exclusion Criteria: +sxs onset > 2 yrs; any antipsychotic >2wks in prev yr or 6wks any time; intolerance to study drugs; contraindicated.

•Treatment Protocol: 4wks before and 1wk after randomization, baseline data obtained for demographics, dx, tx setting, PANSS, CGI, GAF, CDSS, MANSA, SHRS, selected UKU, and PE with baseline labs. Pt randomized to po daily haloperidol 1-4mg, amisulpride 200-800mg, olanzapine 5-20mg, quetiapine 200-750mg, or ziprasidone 40-160mg. Mood stabilizers, BZD, antidepressants, and anticholinergics allowed. Minimization prevented unequal group sizes. Pt and psychiatrists unblinded. Data collected at 0.5, 1, 1.5, 2, 3, 6, 9, and 12 mos.

•Outcomes: Primary Outcome: all cause treatment d/c. (Dose above/below predefined range, other antipsychotic for 15+ days over 6 mos, parenteral antipsychotic active >14 days.) Secondary Outcome: Reason for d/c (insufficient efficacy, side effects, non-adherence, other); Efficacy outcomes (PANSS, CGI, GAF, CDSS, MANSA, and adherence); Safety and Tolerability outcomes (psych adm, serious AE, selected UKU items, wt, labs, EKG, and use of concomitant drugs)

•Analysis: Kaplan-Meier curves, Cox proportional-hazards regression analysis, Hazard ratios, Longitudinal multilevel linear mixed-effects regression models, Poisson regression analysis, S-Plus, SPSS.

•Tx d/c for any cause (p<0.0001) and for insufficient efficacy (p<0.0001) was substantially lower with SGAs. Olanzapine and quetiapine best tolerated (p=0.023).

•Symptomatic improvement by PANSS and hospital admission rates did not differ, but CGI (p=0.0006) and GAF (p=0.006) differed with most improvement with amisulpride and least with quetiapine and haloperidol.

•Side effects: parkinsonism higher with haldol (p<0.0001); anticholinergic use higher with haldol and amisulpride (p<0.0001); more pt on olanzapine also on AD (p<0.0001); wt gain highest with olanzapine (p<0.0001).

Nice effectiveness trial with long treatment duration and broad inclusion criteria, including women and substance abusers, however it has important limitations. The biggest is its non-blinded design. Though study coordinator expectations are assessed, they may have had a larger impact as indicated by differences in the subjective GAF & CGI results vs. no diff in the more objective PANSS. And, what about pt expectations? Use of haldol, a high potency antipsychotics as the representative FGA may have skewed results towards SGAs. Why did they not use a mid-potency antipsychotic as in the CATIE trial’s perphenazine, which may have lead to better outcomes for the FGA, or used an add’l FGA agent? Also, did the dose limitation of haldol impact the results? The limits of med doses may have impacted generalizability. Given that dose increase above the level prescribed by the trial was counted as discontinuation, this could have important impact on its findings. Also, generalizability limited to pts with first episode schizophrenia and notably 2/5 of pts met criteria for schizophreniform only, thus may be more likely to respond to tx. Does not account for tx differences between Europe/Israel and the U.S.

Clinical Bottom Line: In this effectiveness trial, although high continuation rates for most SGAs suggest clinically meaningful long-term antipsychotic tx is achievable in first episode schizophrenia, it cannot be concluded that SGAs are more efficacious than FGA, low dose haloperidol.

AUTHOR
Deepmala Singh

Reduced Sleep Spindle Activity in Schizophrenia Patients

Chairman’s Rounds, November 5, 2007

Citation: Ferrarelli, F et al. “Reduced Sleep Spindle Activity in Schizophrenia Patients.” Am J Psychiatry 2007; 164:483-492.

Clinical Question: Whether sleep rhythms differ between schizophrenia subjects, healthy individuals and a psychiatric control group with a history of depression?

Background information: The study of spontaneous neural activity during sleep offers some important advantages for investigating brain function in subjects with schizophrenia. Past EEG studies focused on the overall architecture of sleep in schizophrenia population. Several studies have reported a reduction of slow wave sleep (Sleep stage 3 & 4). But because of the heterogeneity in patient selection the results were not observed in all studies. Nor were they specific of Schizophrenia. Few studies went beyond sleep architecture and only three studies specifically examined spindle activity in schizophrenic subjects.

Why sleep spindles? The generation of sleep spindles requires inhibitory cells in the reticular thalamic nucleus. The role of reticulo-thalamic circuits in attentional gating of sensory information is particularly relevant in view of reports that schizophrenia involves deficits in these functions. A well-known example of such deficits is the evoked response potential to auditory stimuli. The best characterized of these deficits are impairments in the P50 and P300 evoked response potentials. These deficits have been seen in acutely psychotic and remitted patients with schizophrenia as well as in unaffected relatives, suggesting that these defects might be heritable markers of schizophrenia predisposition, rather than measures of active illness.

Patient/population: Healthy comparison subjects (N=17), medicated schizophrenia patients (N=18), and subjects with a history of depression (N=15) were recruited from local mental health providers or by advertisements in a local newspaper.

Inclusion criteria---All subjects were between 18 and 55 years of age. All outpatients with stable chronic illness with a mean duration of 16.7 and mean total PANSS score of 84.8.

Exclusion criteria---They were excluded if they had substance abuse or dependence within the last 6 months, an identifiable neurological disorder, insulin-dependent diabetes, a recent heart attack or cancer, or a diagnosed sleep disorder; had worked night shifts; or had traveled across time zones in the last month. The healthy subjects were excluded if they were taking psychotropic medications or had first-degree relatives with psychiatric diagnoses.

Intervention: Subjects were recorded during the first sleep episode of the night with a 256-electrode high-density EEG.

Outcomes: Recordings were analyzed for changes in EEG power spectra, power topography, and sleep-spindles.

Validity Criteria

Main Results sleep architecture, sleep EEG power spectra, and spindle parameters between groups compared with one- and two-way analyses of variance (ANOVAs) with Bonferroni correction for multiple comparisons, followed by post hoc two-tailed unpaired t tests. For some parameters Cohen’s d was calculated to determine effect sizes.

1. Sleep Architecture--- total sleep time, time spent in each sleep stage, and sleep maintenance were not different among the three groups. Increased sleep onset latency in both schizophrenia (p<0.01) and depression (p<0.05) subjects in relation to comparison subjects.

2. EEG Power Analysis-- an exploratory analysis of the EEG power for non-REM stages 2–4 in the 0.75–40.00 Hz frequency range ---a significant difference between the comparison and schizophrenia subjects (p<0.001) and the depressed and schizophrenia subjects (p<0.01), but not between the comparison and depressed subjects.

3. Topography of 13.75–15.00 Hz EEG power-- The topography of non-REM sleep EEG power in the high sigma range showed a typical frontal-parietal peak in the comparison and depression groups, which was greatly reduced in the schizophrenia group. No electrodes had statistically different EEG power values in the comparison between the depression and the comparison groups.

4. Sleep spindle analysis-- Spindle number, amplitude, and duration were calculated for each group. Each of these parameters was significantly reduced at the electrodes, with a significant power reduction in schizophrenia versus comparison and schizophrenia versus depression groups, but not the comparison versus the depression group. The integrated spindle activity was markedly reduced in amplitude and number of detections during the first sleep episode in the schizophrenia group in relation to the comparison and depression groups . Furthermore, integrated spindle activity had the largest effect size of all spindle parameters investigated, which corresponded to 93.0% and 90.2% separation of the schizophrenia from the comparison and depression groups, respectively.

There were no significant correlations between integrated spindle activity values and PANSS subscores, medication dose (chlorpromazine equivalents), age, education, body mass index, or duration of illness. We did not have enough statistical power to stratify the results with respect to gender.

Conclusion: Strengths of study---Hypothesis generating research, Use of high density EEG, Explanation of various questions, large effect size.

Weaknesses--- limited group size, Results by chance, Questionable reproducibility and stability over time, and baseline difference.

Synopsis: No single genetic, electrophysiological or cognitive measure has been identified in all subjects with schizophrenia. The heterogeneity of the schizophrenia syndrome has made the search for schizophrenia trait markers challenging. Similarly, results indicate that impairments resulting in decreased spindle activity during the first sleep episode are present in most—but not all—of the schizophrenia subjects reported here. Whole-night sleep recordings on multiple nights will be necessary to determine the stability of spindle activity patterns over time. Further studies with larger groups, especially with unmedicated schizophrenia patients or with nonschizophrenia patients treated with antipsychotics, will be needed to confirm the specificity of the present findings. Studies with larger populations and with first-degree relatives will also be needed to establish whether reduced sleep spindle activity in schizophrenia varies with diagnostic subcategory, symptom severity, or gender and whether it may represent a trait or a state marker of disease.

Real- World Trial of Family Psychoeducation for Schizophrenia

Chairman’s Rounds, October 1, 2007

Clinical Question: Is psychoeduacational family intervention effective for schizophrenia on patient’s personal and social functioning as well as on the relatives’ burden and perceived support?

Background Info: Several studies conducted since 1980s have demonstrated the efficacy of psycho educational family intervention for the treatment of schizophrenia. This study addresses a frequently emphasized weakness of clinical trial involving patients with schizophrenia which often use relapse as the outcome variable, whereas consumers and their families are more concerned about issues as housing, employment and social relationships.

Patients/Population: 71 families of consumers with DSM IV diagnosis of Schizophrenia selected by clinicians at each site and randomized; 62 analyzed (N=36 in intervention group, N=26 in control group).

Inclusion criteria---clinically stable; in treatment with local center for 6 months; living with at least one adult relative; patient with alcohol or drug abuse or physical illness were not excluded.

Intervention/exposure: Family Psychoeducation (three one hr sessions a month for each family = 18 sessions over 6 months)

1. Patient’s clinical status and personal and social functioning---- assessed by Brief Psychiatric rating scale ( BPRS) and Assessment of disability (AD) ;

2. Patient’s and relatives’ social resources ---- evaluated by social network questionnaire(SNQ);

3. Burden of care of each family member---- by Family Problem Questionnaire (FPQ);

Follow-up: 6 patients in the case group and 3 in the control group were lost to follow up and not included in the analysis. In the intervention group The families who dropped out differed from the other families only in regards to patients’ mean±SDscore on BPRS mania-hostility subscale

Randomization: pts were randomized through computerized system in blocks of 5 with a 3:2 treatment to control ratio.

Intention to treat: pts were randomized in the groups to which they were randomized; however only completers were included in analysis

Similar groups: As regards the sociodemographic characteristic there was significant between-group difference in context of employement (chi square=4.5, df=1, p<0.05). Clinical characteristic was the same in both the group. Significant difference in the treatment of family/patient in prior 6 months is summarized in the table backside.

Equal treatment: all the patients were on antipsychotics, and efforts were made to provide a uniform intervention. Some important differences in non-study interventions are summarized in the table backside

---Baseline to endpoint changes compared with patient in the control group, those in the intervention group demonstrated lower levels of disability in self care ( F=5.4, df=1 and 60, p<.05) and in behaviour in emergencies(F=7.1, df=1and 57, p<.01)

----Baseline to endpoint changes in the intervention group was significant as regards poor global functioning (47% Vs 25%, z= -3.3, p<.001) ; moderate to severe social withdrawal (36% Vs 19%, z = -2.1, p< .03) ; Difficulties in managing friction in social & interpersonal relationships( 19% Vs 5 %, z=-2.4, p<.01); Difficulties maintaining interests ( 64% Vs 39%, z=2.0, p<.05); Improvement in their social relationship (75% Vs 58 %, z=-2.1, p<.03)

--- Baseline to endpoint changes in the control group---The only significant improvement was in participation in household activities which was poor in 23% at baseline and 15% at follow up(z=-2.0, p<.05)

2. Family Burden, social network, and professional support----The average level of family burden significantly improved in both the control and the intervention group.

----In the intervention group---baseline to end point changes as regards neglecting their hobbies decreased (51% Vs 21% z=-2.6, p<.01), crying and feeling depressed decreased( 46% Vs 28% z=-2.5, p<.02), feeling embarrassed in public decreased(21% Vs 9%, z=-3.4; p<.001; relative’s social contacts & perceived help in emergencies increased(15% Vs 44% z=-2.9, p<.003); relatives perception of the level of professional support increased(65% Vs 93 %, z=-4.4, p<.001)

---- In the control group baseline to end point changes as regards neglecting their hobbies decreased (32% Vs 21% z=-2.0, p <.05); relatives’ perceived social support increased (37% Vs 53% z=-2.8,p<.005)

Conclusions: Strengths of the study---Important topic, Routine clinical setting, focused on the weakness of other studies by addressing to more practical problems, sample was representative of the heterogeneity of patients, good attempt to compare the baseline.

Weaknesses---small sample size, possible selection bias, relatively brief period of follow up, assessment tools were administered by the same professionals who provided the intervention, concerns of co-intervention, inter-rater variation, questions regarding generalizability.

Synopsis: In this 6 months RCT of 62 patients with schizophrenia the results of family psychoeducation on patient’s functioning and family burden seem to be ambiguous. Considering the various weakness of the study more confirmatory studies are needed especially in the clinical setting of US.

Significant Baseline differences between the control & the intervention group

Characteristic Intervention (N=36) control (N=26)

---Patient’s information sessions on 11 26 9 31
Schizophrenia and drugs in prior
6 months
__________________________________________________________________________________________________

Significant difference in the treatment of patient during the period of 6 months

Characteristic Intervention (N=36) control (N=26)

______________________________________________________________________________________________

Citation: Patil S, et al. Activation of mGlu2/3 receptors as a new approach to treat schizophrenia; a randomized phase 2 clinical trial. Nature Medicine. September 2007;13: 1102-1107.

Clinical Question: Is LY2140023 for 28 days superior to placebo or olanzapine in the treatment of patients with schizophrenia as measured by the PANSS score?

Background information: While altered glutamate neurotransmission has long been linked to schizophrenia, all commonly prescribed antipsychotics act on the dopamine receptor. LY404039 is a selective agonist for metabotropic glutamate 2/3 (mGlu2/3) receptors and has shown antipsychotic potential in prior animal studies. Animal knockout studies using the mouse PCP model have shown that this drug works in a mechanism different than that of olanzapine and other atypical antipsychotics. However, oral absorption of LY404039 in humans is low. A methionine amide of that drug, LY2140023 was found to be an effective oral prodrug in humans, which after absorbed is hydrolyzed to produce the active mGlu2/3 receptor agonist LY404039 with bioavailability of 49%.

Study Design: RCT (Phase 2 double blind, parallel, placebo controlled fixed dose study- proof of concept study to test the efficacy of LY2140023)

Patients/Population: Supplementary Table 2/Figure 2. 234 entered Period I, 38 failed screening Total N= 196 (N=98 new drug, N=63 placebo, N=34 olanzapine). One passed screening but had adverse event before study drug started. Demographics, baseline severity of illness, baseline EPS scores are similar among all three groups. Notably, >75% of subjects are male.

Inclusion criteria: men and non-child bearing women 18-65 with DSM-IV dx of schizophrenia confirmed by SCID, willing to be inpatient for 8-12 weeks, PANSS score >/= 4 on at least two positive subscale items extracted from the PANSS at screening (conceptual disorganization, hallucinatory behavior, suspiciousness, unusual thought content), CGI-S (Clinical Global Impression-Severity) score >/= 4 (moderately ill), BMI </= 32, clinical lab test result wnl.

Exclusion criteria: comorbid Axis I d/o, prolactin >200 in absence of risperdone, diagnosis of Parkinsons disease, dementia related psychosis, any depot injections w/in 4 weeks, current mood stabilizer treatment, SI/HI, olanzapine use w/in 6 months or had h/o non-response to adequate trial of olanzapine, substance abuse diagnosis, positive UDS, concomitant centrally acting meds, women of child bearing potential

Intervention: Inpatient treatment with either LY2140023 40 mg BID, olanzapine 15 mg daily, or placebo for one month.

Statistics: Primary measures included MMRM to evaluate safety measures, ANOVA for baseline patient characteristics, Fischer’s exact test for patient characteristics, patient disposition, adverse events. Used one-sided alpha of 0.05.

Follow-up: Supplementary Table/Figure 2. 196 enrolled, 1 lost before treatment, 118 completed. It looks like all patients were accounted for, except that Supp Fig 2 reports one more assigned to treatment group than Table 1 in paper- typo?

Intention to treat: all patients were analyzed in the group they were randomized to and all patients’ data was analyzed. Those who discontinued treatment carried forward their last data point.

Similar groups: all groups similar in prognostic factors, demographic factors

Blinding: states double blinded, but does not comment on assessors. Does not comment on the preservation of blinding or provide information regarding the rating.

Main Results:

-Response rates (25% decrease in PANSS score): 32% (new drug) v. 3.2% (placebo) v. 41.2% (olanzapine)

-Table 1/Figure 2 PANSS total score -13.1 (p<0.001) in new drug v. +7.63 (p=0.034) in placebo v. -19.12 (p<0.001) in olanzapine. No statistically significant difference b/w new drug and olanzapine group, but obviously both do much better than placebo (-20.8 and -26.7 on PANSS, respectively) PANSS positive scores -4.62(p<0.001) v. 1.73(p=0.16) v. -6.91(p<0.001). PANSS negative scores -3.33(p<0.001) v. 1.36(p=0.15) v. -3.87(<0.001). CGI-S scores -0.62 (p<0.001) v. 0.35(p=0.11) v. -0.97(p<0.001)

- Supp Table 2. Completion rates 66.3% new drug v. 41.3% placebo v. 79.4% olanzapine. That is more than 10% difference between new drug and olanzapine. 16.3% of pts who d/c’d new drug did so 2/2 lack of efficacy v. 5.9% of olanzapine. Overall p-value describes comparison among all three groups but does not tell about new drug v. olanzapine.

- Table 2. Adverse event- noted mood lability, increases in CPK, myocardial ischemia in new drug group. Needs further evaluation. Noted decrease in wt (-0.51 kg) and decrease in prolactin in new drug group. Weight gain (+0.74 kg) in olanzapine group as we would expect.

Conclusions: This is an important study because it suggests that a new class of drug with a mechanism and side effect profile different than typical and atypical antipsychotics currently used to treat schizophrenia may be nearly as efficacious. Strengths of the study include double blinded RCT, adequate treatment with olanzapine, similar baseline demographics/severity of illness. Weaknesses include small sample size, male predominance, short duration, no information on blinding, drug company sponsored. Exclusion of non-responders to olanzapine and women of child bearing potential may reduce the generalizability of the study. Additionally, the inclusion of a placebo arm in this study is ethically questionable and one must wonder how much less impressive the result would appear had LY2140023 been compared to olanzapine and a typical antipsychotic rather than placebo. Little information obtained from placebo arm; would have been interesting to see LY2140023 compared to olanzapine and a typical antipsychotic.

Synopsis: In this 4 week double blinded RCT of 196 patients testing the efficacy of a new drug targeting the mGlu2/3 receptor as compared to placebo and olanzapine in the treatment of schizophrenia, the new drug seems to be nearly as efficacious as olazapine in terms of reducing total, positive, and negative PANSS scores as well as CGI-S scores. Due to the male predominance among all groups it may be difficult to generalize these results to the general population. Given that this study is an efficacy study with numerous inclusion and exclusion criteria, generalizability is limited. Nonetheless, a different target for treating schizophrenia is promising.

Citation: Rosenheck RA, Leslie DL, Sindelar J, et al. Cost-Effectiveness of Second-Generation Antipsychotics and Perphenazine in a Randomized Trial of Treatment for Chronic Schizophrenia. Am J Psychiatry 2006. 163: 2080-2089

Clinical question: How cost-effective are the atypical antipsychotics compared to first-generation drugs?

Background: Less than 1/3 of patients with schizophrenia are prescribed first-generation antipsychotics. The US spends over ten billion dollars on atypical antipsychotics as drugs of choice for schizophrenia, but recent studies have failed to find strong advantages for the newer drugs in effectiveness, Parkinsonism, or cost.

Methods: This is an analysis of data from the CATIE trial, a 57-center study using an established algorithm to compare treatment with olanzapine, perphenazine, quetiapine, risperidone, and, later, ziprasidone. Study patients were men and women aged 18 to 65 with diagnosis of schizophrenia. Exclusion criteria were: diagnosis of schizoaffective disorder or mental retardation, unstable medical condition, previous adverse reaction to one of the study meds, treatment-resistant schizophrenia, first episode psychosis, and pregnancy/breastfeeding. They were treated with 1-4 capsules daily of an antipsychotic.

This was double-blind, but patients with established tardive dyskinesia could not be assigned to perphenazine (15% of the sample). Patients who discontinued their first treatment could be switched to any other atypical if they wanted.

This article used total health costs and quality-adjusted life year (QALY) ratings as primary outcomes. To assess costs, the study used self-report of any hospital days, nursing homes, group homes, use of outpatient care, and ER service use, multiplied by estimated local unit cost. Study medicine costs were determined by published wholesale costs and discounting appropriately for patients who would have used Medicaid or the VA. QALY was determined by using specific factors on the Positive and Negative Syndrome Scale (PANSS) to develop script and video materials about major health impairments, schizophrenia symptoms, and side effects. 620 lay people saw these videos and rated quality of life from 0 (worst imaginable) to 1 (perfect health). Final QALY is the product of the rating for schizophrenia state and for each side effect.

Statistical analysis was done on four datasets. Data reported in the paper come from a set excluding patients with TD or assigned to ziprasidone. Data reported about ziprasidone was taken from a set excluding those patients with TD.

Validity criteria: Blinding – double-blind, doctors could give 1 to 4 capsules daily

Intent to Treat – patients were followed for up to 18 months. 45.7% were still participating at 18 months and analyzed in their original assigned groups

Randomization/Baseline – when TD is excluded, the groups were equal at baseline and randomized to any of the five meds

Main results: 1,424 of the original 1,493 randomized were analyzed, and the cost for drugs in the patients assigned to perphenazine were 40-50% lower than the other groups with no significant difference in the proportion of patients receiving inpatient care each month. Group-by-time interactions were likewise similar. Summed, average total monthly health costs were 20-30% lower in the perphenazine group even when accounting for VA or Medicaid discounts. Olanzapine carried higher drug costs but lower inpatient and outpatient costs, so it was not significantly different.

All five drugs were associated with significant improvement in QALY (0.683 to 0.747), but the only statistically significant difference was that perphenazine had better QALY than risperidone. This was consistent before and after the first switch of medications.

Conclusions: This study shows that for the population in the CATIE trial, initial assignment to perphenazine was associated with reduction in cost but not reduction in effectiveness compared to four atypical antipsychotics at 18 months. 99% of the patients who switched from perphenazine went to an atypical, but the time spent taking a less-costly drug resulted in long-term health care savings. Of note, this study did not include patients with first-break psychosis and cannot be applied to patients with pre-existing tardive dyskinesia, and an 18-month follow-up is not likely to pick up on tardive dyskinesia. This study would also need to be recalculated in a few years as atypicals go off-patent, and it only addresses one first-generation antipsychotic. The main practical application of this study is to seriously consider a trial of a typical antipsychotic early in the treatment of a patient with schizophrenia.

Citation: Harris M G, et al. The relationship between duration of untreated psychosis (DUP) and outcome: An eight-year prospective study. Schizophrenia Research. 79 (2005) 85-93

Clinical Question: Does DUP predict clinical and functional outcomes, independent of potential confounding variables, in a large first episode patient cohort followed up 8 years after first psychiatric treatment?

Background Information: Preventing an illness from occurring is inherently better than having to treat the illness after its onset. This approach is unfortunately not currently an option in most of the serious mental illnesses. For this reason there has been an emergence of secondary prevention oriented mental health research in the past decade. Secondary prevention can be seen as an approach aimed at early detection and prompt intervention to control disease and minimize disability. In terms of first episode psychosis, there is evidence to suggest that the DUP correlates moderately with short term symptomatic and functional outcomes, and that the association is independent of potential confounding factors. Few prospective studies have examined this association over the longer term. Based on current evidence, early intervention in psychosis programs have been established worldwide.

Setting: Patients treated at a specialist early psychosis service in Melbourne, Australia: “The pre-EPIC cohort” 200 patients from consecutive admissions to the Aubrey Lewis Unit for First Episode Psychosis, Royal Park Hospital, between March 1989 and April 1992. “The EPIC cohort” 359 patients who commenced treatment at the Early Psychosis Prevention and Intervention Centre between January 1993 and July 1997.

Patients: Four groups: Schizophrenia spectrum (schizophrenia and schizophreniform disorder); Affective disorders (Bipolar disorder and Depressive psychosis); Schizoaffective disorder; and Mixed psychotic disorders (Delusional disorder; Psychotic disorder NOS, Brief psychotic disorder).

T4 assessments were conducted between January 1998 and April 2005. At T4 65.7% (367) of the 559 participants were assessed via interview. 16.5% (92) refused invitation to be interviewed, 5.7 % (32) were known to be deceased, 11.6 % (65) could not be located, and 0.5 % (3) were not approached as they had previously refused all further follow-up. Of those interviewed at T4, 13.4 % (49) were excluded from analysis due to missing data on any one of the predictor variables – leaving a final group of 318 (56.9%). Table 1 comparison of completers and non-completers, no statistically significant differences were present between the groups.

Results:
R-square value is an indicator of how well the model fits the data (e.g., an R-square close to 1.0 indicates that we have accounted for almost all of the variability with the variables specified in the model).

The regression model coefficients for the QLS - Table 3. The strongest predictor was premorbid adjustment, followed by DUP.

DUP was the strongest predictor of positive symptoms in both the total and the schizophrenia-spectrum subsample. A dose response effect may be present, as a longer DUP resulted in significantly worse symptoms.

SANS – no statistically significant differences between short and long DUP was present.

The model showed reasonably similar explanatory power in the total sample at 8 year follow up; and slightly reduced power in the schizophrenia-spectrum subsample for the QLS and SANS. The independent contribution of DUP diminished at 8 year follow up. Greater explanatory power was seen in the BPRS-PS at 8 year follow up.

In summary - The results suggest that a shorter DUP is associated with decreased severity of positive symptoms and enhanced social and occupational functioning and quality of life, although it is suggested that it may have a reduced effect on negative symptoms.

Based on these results is there a need for assertive early detection strategies to facilitate treatment to those with emerging psychotic illness in order to reduce the risk of deleterious outcome?

1 - Australia - McGorry et al - the PACE and EPPIC programs

2 - Canada - numerous programs (Addington et al; Norman & Malla; etc)

3 - UK - several programs, especially in London (AESOP, LEO)

4 - Western Europe - Norway (TIPS), Denmark (OPUS), France, Germany

5 - US - several programs (Hillside, Pittsburgh, California)

Median DUP in this study was 40.5 days – utility of early intervention program

AUTHOR
Doug Burgess

Reference: Honer WG, et al. Clozapine alone versus clozapine and risperidone with refractory Schizophrenia. New England Journal of Medicine 2006; 354(5): 472-82.

Clinical Question: In patients with chronic schizophrenia refractory to a therapeutic trial of clozapine, does the addition of risperidone to clozapine improve symptoms?

While on Williams Ward, a patient treated concomitantly with risperidone and clozapine was admitted secondary to recent decompensation following non compliance to medication regimen. This raised the question as to whether the patient was likely to benefit from treatment with both medications.

Therapy Question Validity: Follow up was complete and all patients were accounted for at the end of the trial. The patients were randomized by computer and treated within the groups they were originally assigned. The only significant difference between treatment groups was the duration of prior clozapine treatment. The mean number of weeks of prior clozapine treatment was greater in the treatment group. All patients, clinicians and assessors were blinded throughout the trial. With the exception of the intervention, there was no evidence that the treatment and control groups were treated differently. The study did acknowledge that Janssen-Ortho provided the risperidone and matching placebo. The company was also allowed to review the study protocol but there were no requests for amendment. The only study data provided to the company consisted of serious adverse events.

Randomized Controlled Trial criteria for appropriate design: The trial was supported by a grant from the Stanley Medical Research institute and the risperidone and placebo used in the study were provided by Janssen-Ortho. Following the exclusion period, all participants remained blinded throughout the study. The patients were followed over the course of eight weeks at which time they were offered an optional 18-week period of non-blinded augmentation with risperidone. The participants were enrolled from seven institutions in Canada, Germany, China and United Kingdom similar to academic medical centers in the United States. 595 candidates were evaluated for enrollment into the study. 458 did not meet eligibility requirements and 69 declined to participate leaving a total of 71 participants. The participants were mostly male (74%) and had been hospitalized an average of 5.4 times. The number of outpatient to inpatient participants was 42 to 26 respectively. The intervention studied in this trial was augmentation of clozapine therapy with risperidone. The researchers’ primary outcome measured severity of symptoms using the total PANSS score. They also evaluated frontal lobe cognitive function using the Brown-Peterson procedure and Letter-Number sequencing task. Finally, side effects and adverse events were measured using the ESRS, BAS and a standardized 42 item general side effect scale. 100% of patients were accounted for at the end of the study.

Results: At eight weeks and 26 weeks, the total PANSS score did not differ significantly between the risperidone and placebo groups. Both groups improved significantly from baseline, but the mean difference in the change of PANSS scores between the groups was 0.1 (95% C.I., -7.3 to 7.0). When analyzed separately, there was also no significant difference in improvement observed between inpatients and outpatients. Secondary outcome measures revealed no significant differences in CGI severity or improvement scores. There was also no significant change in the composite score of the verbal working memory; however, the amount of change between baseline and eight weeks differed significantly with performance slightly increased in the placebo group and slightly decreased in the risperidone group. There were no significant differences between the two groups in the incidence of side effects.

Conclusions: The study demonstrated that there is no added benefit in augmenting clozapine treatment of chronic schizophrenia with risperidone. In addition, there is evidence that frontal lobe function was actually impaired in patients receiving risperidone in addition to clozapine. Given the apparent frontal lobe impairment in conjunction with the lack of evidence for any improvement in symptoms, there does not appear to be an indication for risperidone augmentation in the treatment of clozapine resistant schizophrenia. Interestingly, both groups demonstrated an improvement in PANSS score from baseline suggesting a nonspecific effect of simply participating in a treatment trial.

What medications are useful in augmenting treatment of patients with clozapine refractory schizophrenia?
Is it surprising that the trial did not reveal a significant increase of side effects within the treatment group?
Is the PANSS scale an appropriate measure of clinically relevant improvement?

AUTHOR
Steven Vas

Citation: “Mifepristone vs Placebo in the Treatment of Psychosis in Patients with Psychotic Major Depression.” DeBattista et al. Biological Psychiatry 2006; 60:1343-1349;

Clinical Question Does mifepristone cause rapid and sustained reduction of psychotic ssx in patients with psychotic major depression (PMD)?

AUTHOR Deepmala Singh