Self Improvement

Relationships between pharmacotherapy-induced metabolic changes and improved psychopathology in schizophrenia: data from a mirtazapine and first-generation antipsychotics combination trial

Relationships between pharmacotherapy-induced metabolic changes and improved psychopathology in schizophrenia: data from a mirtazapine and first-generation antipsychotics combination trial
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Relationships between pharmacotherapy-inducedmetabolic changes and improved psychopathologyin schizophrenia: data from a mirtazapine andfirst-generation antipsychotics combination trial Viacheslav Terevnikov 1  , Jan-Henry Stenberg 2  , Jari Tiihonen 3  , 4  , 5  , Evgeni Chukhin 2  ,Marina Joffe 2  , Mark Burkin 6 and Grigori Joffe 2 1 Kellokoski Hospital, Kellokoski, Finland 2  Helsinki University Central Hospital, Helsinki, Finland 3 Department of Forensic Psychiatry, University of East Finland, Niuvanniemi Hospital, Finland 4 Department of Mental Health and Alcohol Research, National Institute for Health and Welfare, Helsinki, Finland 5 Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden 6 University of Petrozavodsk, Russia Abstract Clinical efficacy and metabolic side-effects of antipsychotics seem to correlate with each other. In this study,interrelationship of similar metabolic effects of mirtazapine and its earlier reported desirable effects on psycho-pathology in first-generation antipsychotics (FGAs)-treated schizophrenia were explored. Symptomatic FGAs-treated patients with schizophrenia received a 6-wk double-blind treatment with add-on mirtazapine ( n = 20) orplacebo( n = 16),followedbya6-wkopen-labelmirtazapinetreatment.Mirtazapine(butnotplacebo)inducedanincrease in body weight and cholesterol levels. The latter was associated with a clinical improvement in all(sub)scales of the Positive and Negative Syndrome Scale [PANSS; an increase of cholesterol by 1 mmol/l pre-dicted 7 points reduction on the PANSS total score ( r = 0.85,  p = 0.001)]. In schizophrenia, mirtazapine-inducedweight gain and increase of total cholesterol are associated with the improved efficacy of mirtazapine-FGAscombination – a novel observation with possible clinical and theoretical implications. Received 18 June 2012; Reviewed 10 August 2012; Revised 16 September 2012; Accepted 11 October 2012;First published online 10 December 2012 Key words : Add-on treatment, antipsychotics, lipids, mirtazapine, schizophrenia. Introduction A large body of evidence suggests that treatment withantipsychotics is associated with an increased risk of metabolic side-effects, such as glucose dysregulation anddiabetes mellitus (Meltzer, 2001), dyslipidaemia (Meyerand Koro, 2004) and weight gain (Ascher-Svanum  et al .2005).Most(thoughnotall)ofthesecond-generationanti-psychotic agents (SGAs) are more likely to produce theseeffects (Liao  et al . 2011) but such risk also exists for thefirst-generation antipsychotics (FGAs; Newcomer, 2007).Clozapine and olanzapine are the most aggressivemetabolic offenders in this manner, yet both appear to bethe most effective antipsychotics (Leucht  et al . 2009). Ithas been repeatedly observed that the more prominentthe antipsychotic-related metabolic side-effects are themore evident is improvement in psychopathology. Thishas been reported earlier for clozapine (Procyshyn  et al .2007), olanzapine and haloperidol (Ascher-Svanum  et al .2005). Recently Hermes  et al . (2011) found similar resultsfor various FGAs and SGAs used in the Clinical Trials of Antipsychotic Treatment Effectiveness. The nature of thisrelationship is obscure (Procyshyn  et al . 2007).Mirtazapine, one of the few existing receptor-blockingantidepressants with a specific psychopharmacologicalprofile,seemstoboostthedesirableeffectsoftheFGAsonthe negative, positive and cognitive symptoms of schizo-phrenia (Joffe  et al . 2009; Stenberg  et al . 2010; Terevnikov et al . 2010; Phan and Kreys, 2011). Mirtazapine also cau-ses weight gain and dyslipidaemia in patients with majordepressive disorder (MDD; Laimer  et al . 2006). In ourprevious study, we reported an increase in total chol-esterol (Joffe  et al . 2009) in FGAs-treated patients withschizophrenia whose positive, negative and cognitivesymptoms improved after treatment with add-on mirta-zapine.The aim of the present study was to more thoroughlyanalyse the metabolic effects of add-on mirtazapine inFGAs-medicatedschizophrenia.Inparticular,wehypoth-esized that the presumable metabolic side-effects of  Address for correspondence: V. Terevnikov, MD, PhD, KellokoskiHospital, 04500 Kellokoski, Finland. Tel. : + 358503250557  Fax : + 358927163461 Email : International Journal of Neuropsychopharmacology  (2013),  16 , 1661–1666. f CINP 2012doi:10.1017/S146114571200137X BRIEF REPORT  add-on mirtazapine correlate with mirtazapine-inducedimprovement in psychopathology as appears to occurwith antipsychotics. Method and materials Subjects The eligibility criteria for this study have been describedelsewhere (Terevnikov  et al . 2010). The FGAs-treated,stabilized but highly symptomatic patients with schizo-phrenia were recruited from the psychiatric hospitalof the Karelian Republic (in-patients) and from the day-treatment unit (out-patients) in Petrozavodsk, RussianFederation. All patients provided written informed con-sent, and the study was carried out in accordance withthe Declaration of Helsinki and good clinical practice. Study design The study was a double-blind, add-on, randomizedplacebo-controlled trial with a placebo run-in period andan open-label extension phase. Patients were randomizedto receive double-blind treatment with either add-onmirtazapine (30 mg/d; mirtazapine group) or add-onplacebo (placebo group) for the first 6 wk of the study.Completers of the double-blind phase who volunteeredto continue in this study received add-on mirtazapine(30 mg/d) in an open-label fashion for an additional6 wk (open-label phase). Only data for completers of bothphases were analysed. Doses of current FGAs remainedunchanged. Study assessments Clinical efficacy was assessed with the Positive andNegative Syndrome Scale (PANSS; Kay  et al . 1987)atweeks0,1,2,4and6(double-blindphase)andweeks7,8, 10 and 12 (open-label phase). Body weight, bloodpressure, fasting glucose and total cholesterol (metabolicparameters) were measured at weeks 0, 6 and 12. Dueto technical limitations of the local laboratory, a moredetailed lipid profile panel was unavailable. Statistical analysis Cross-sectional statistical differences and a comparisonof changes in the variables between the study groupswere calculated using the independent samples  t  test.Within-group changes in the variables over time weretested with the paired sample  t  test. Pearson’s correlationanalysis was performed for calculations of correlations between changes in metabolic parameters and changeson the PANSS (sub)scales. Two-tailed  p  values  < 0.05were considered statistically significant. Relationships between the changes in the metabolic parameters andPANSS-measured changes of psychopathology wereinvestigated with analysis of variance (ANOVA). TheANOVA model included changes in each of the PANSS(sub)scales as dependent variables and duration of disease, current dose of antipsychotic, change in bodyweight, fasting glucose and total cholesterol levels asindependent variables.Statistical analyses were performed using SPSS forWindows version 14.0. Results Subject disposition, baseline demographic and clinicalcharacteristics The study population of the double-blind and open-labelphases has been described in detail (Joffe  et al . 2009;Terevnikov  et al . 2010). Of the initial 41 patients enrolled,40 patients completed the double-blind phase and 39patients were enrolled in the subsequent open-labelphase. Of these, 20 patients formed the mirtazapinegroup and 19 patients formed the placebo group. Afterpremature discontinuations and data cleaning, 20patients in the mirtazapine group and 16 patients in theplacebo group were eligible for, and subjected to, thecompleter analysis. Patients demonstrated a high degreeof chronicity and treatment resistance. Mean durationof illness was19.9 yr in the mirtazapine group and 23.8 yrin the placebo group (  p = 0.79). Patients received therapywith chlorprotixene ( n = 4), flupentixol ( n = 2), fluphena-zine decanoate ( n = 7), haloperidol ( n = 14), haloperidoldecanoate ( n = 8), levomepromazine ( n = 5), periciazine( n = 2), sulpiride ( n = 1), trifluoperazine ( n = 12),zuclopentixol ( n = 4) or zuclopentixol decanoate ( n = 5).The patients had received numerous antipsychotic trialsduring previous years of their treatment (mean  n = 8.50in mirtazapine group and  n = 8.17 in placebo group,  p = 0.63). The majority were receiving a combination of aforementioned FGAs. Mean daily dose of antipsychoticswas 325.8 and 313.8 mg chlorpromazine equivalents forthe mirtazapine group and the placebo group, respect-ively (  p = 0.21).At baseline (week 0), the mirtazapine group showedhigher PANSS positive scores (22.1  vs.  18.3,  p = 0.02). Interms of other PANSS (sub)scales, age and gender, thegroups did not differ (data not shown).  Metabolic parameters At baseline, there were no statistically significant differ-ences between the groups for any metabolic parameters(see Table 1).Duringthedouble-blindphase,therewasastatisticallysignificant increase in body weight and total cholesterolin the mirtazapine group. In the placebo group, no meta- bolic parameters changed. The between-group com-parison of longitudinal changes revealed a statisticallysignificant difference in the same two parameters. Fastingglucose level did not change in either group. A growthtrend in both systolic and diastolic blood pressure wasobserved in the mirtazapine group, but did not reachstatistical significance.1662  V. Terevnikov et al.  Table 1.  Metabolic changes in first-generation antipsychotic-treated patients with schizophrenia during double-blind treatment with add-on mirtazapine or placebo (weeks 0–6) and open-labelextension add-on mirtazapine treatment phase (weeks 6–12)ParameterBaseline Change, week 6 minus week 0 Change, week 12 minus week 6Mirtazapinegroup a Placebogroup  b Mirtazapine  vs. placebo groupWithin mirtazapinegroupWithin placebogroupMirtazapine  vs. placebo groupMirtazapinegroupPlacebogroupMirtazapine  vs. placebo groupWeight mean ( S . D .) 67.13 (5.97) 73.06 (12.52)  t =x 1.534 3.72 c 0.13  t =x 3.927 0.93 2.72  t = 1.623  p = 0.135  t = 4.729  t = 0.314  p < 0.005  t = 1.374  t = 3.038  p = 0.114  p < 0.005  p = 0.758  p = 0.186  p = 0.01Systolic BP mean ( S . D .) 119.21 (10.17) 122.19 (11.25)  t =x 0.822 3.68  x 1.25  t =x 1.717  x 3.16 4.06  t = 2.565  p = 0.417  t = 1.895  t =x 0.591  p = 0.095  t =x 1.679  t = 1.932  p = 0.015  p = 0.074  p = 0.564  p = 0.111  p = 0.072Diastolic BP mean ( S . D .) 76.32 (5.97) 77.19 (4.46)  t =x 0.481 3.42 1.25  t =x 1.009  x 2.9 0.94  t = 1.609  p = 0.634  t = 1.998  t = 1.074  p = 0.32  t =x 1.874  t = 0.510  p = 0.117  p = 0.061  p = 0.300  p = 0.077  p = 0.617Glucose mean ( S . D .) 4.96 (0.69) 4.93 (0.73)  t = 0.125  x 0.15  x 0.07  t = 0.547 0.16 0.07  t =x 0.262  p = 0.901  t =x 0.885  t = 0.034  p = 0.588  t = 0.655  t = 0.253  p = 0.795  p =x 0.387  p = 0.973  p = 0.521  p = 0.804Total cholesterol mean ( S . D .) 4.68 (0.99) 5.28 (0.87)  t =x 1.887 0.95  x 0.21  t =x 3.525 0.08 0.51  t = 1.271  p = 0.068  t = 3.515  t =x 1.462  p < 0.005  t = 0.368  t = 2.042  p = 0.213  p = 0.002  p = 0.164  p = 0.717  p = 0.059 a Mirtazapine group ( n = 20): patients treated with add-on mirtazapine during weeks 0–12.  b Placebo group ( n = 16): patients treated with placebo during weeks 0–6 and with add-on mirtazapine during weeks 6–12. c Weight gain in mirtazapine group and placebo group was 5.5 and 0.12%, respectively.  M i    r  t   a  z  a   p i    n e  a  n d   m e  t   a  b   o  l    i    c  c  h   a  n  g e  s  i    n s  c  h   i    z  o   p h   r  e  n i    a   1    6    6    3     During the open-label phase, body weight and dia-stolic blood pressure increased with statistical signifi-cance in the former placebo group, while no furtherchanges occurred in the former mirtazapine group. Relationships between metabolic and psychometric parameters Correlation analysis During the double-blind phase, a change in total choles-terol negatively correlated with changes in all (sub)scalesof PANSS for the mirtazapine group; i.e. total cholesterolincreased with decreases in the PANSS scores (Table 2).In the placebo group, a change in total cholesterol corre-lated similarly with changes on the PANSS positive sub-scale. In the mirtazapine group, the change in bodyweight correlated negatively with changes on the PANSSgeneral scores and PANSS total scores. In the placebogroup, glucose level change correlated negatively withchanges in the negative, general and total PANSS (sub)-scale scores.During the open-label phase, no statistically significantcorrelations between the measured metabolic variablesand PANSS (sub)scale emerged in either group (data notshown). Linear regression analysis During the double-blind phase, changes in total choles-terol were associated with changes in all PANSS (sub)-scales scores in the mirtazapine group. An increase of cholesterol by 1 mmol/l predicted a reduction of 7 points( r = 0.85,  p = 0.001), 1.7 points ( r = 0.72,  p = 0.03), 1.8 points( r = 0.73,  p = 0.004) and 3.5 points ( r = 0.83,  p = 0.005) onthe PANSS total, positive, negative and general scores,respectively. No other statistically significant relation-ships between the measured metabolic variables andPANSS (sub)scale were found. In the placebo group, novariables showed statistically significant associations.In the open-label phase, no significant relationships ineither group were observed. Discussion In our previous publications a clear-cut effect of add-onmirtazapine on schizophrenia symptoms was reported(Joffe  et al . 2009; Stenberg  et al . 2010; Terevnikov  et al .2010), where significant improvement was observed inthe main clinical domains of schizophrenia, as measured by PANSS total, positive, negative and general (sub)-scales and abattery of neurocognitive tests. In the presentstudy, we explored the relationships between the meta- bolic and psychotropic effects of mirtazapine.As expected, mirtazapine induced weight gain (5.5%in the mirtazapine group  vs.  1.2% in the placebo group)with an increase in total cholesterol level (0.95 mmol/l  vs. 0.2 mmo/l, respectively) during the 6-wk double-blindphase. This is in line with previous studies on major de-pression,inwhichmirtazapinenoticeably increasedbodyweight and cholesterol – mostly during the first 4 wk of treatment (Thase  et al . 2001). Similar to some prior reports(Montgomery, 1995), mirtazapine did not cause anyclinically or statistically significant changes in bloodpressure or fasting glucose levels during our study.Furthermore, our data show that an increase in totalcholesterol is directly associated with an improvement ina broad range of schizophrenia symptoms. Regressionanalysis revealed that an increase of total cholesterol by1 mmol/lpredicted7pointsreduction inthe PANSStotalscore. In a study by Procyshyn  et al . (2007), the clozapine-induced increase inserumlipids wasalsoassociated withan improvement in psychopathology of schizophrenia.However, in that study an increase in total cholesterolwas only associated with an improvement in PANSS Table 2.  Correlations between changes in metabolic parameters and changes in the Positive and Negative Syndrome Scale (PANSS)scores in first-generation antipsychotics-treated patients with schizophrenia during the double-blind phase of an add-on mirtazapinetrialMetabolicparameterschangePANSS positive change PANSS negative change PANSS general change PANSS total changeMirtazapinegroup a Placebogroup  b MirtazapinegroupPlacebogroupMirtazapinegroupPlacebogroupMirtazapinegroupPlacebogroupBody weight  x 0.39 c  p = 0.112 x 0.14  p = 0.601 x 0.19  p = 0.441 x 0.22  p = 0.419 x 0.50  p = 0.033 x 0.46  p = 0.072 x 0.48  p = 0.045 x 0.45  p = 0.08Glucose  x 0.08  p = 0.7360.30  p = 0.2660.04  p = 0.083 x 0.62  p = 0.0110.40  p = 0.080 x 0.58  p = 0.0180.23  p = 0.327 x 0.51  p = 0.042Total cholesterol  x 0.54  p = 0.015 x 0.52  p = 0.037 x 0.57  p = 0.0090.23  p = 0.397 x 0.53  p = 0.017 x 0.04  p = 0.896 x 0.63  p = 0.003 x 0.13  p = 0.628 a Mirtazapine group ( n = 20): patients treated with add-on mirtazapine during weeks 0–12.  b Placebo group ( n = 16): patients treated with add-on placebo during the double-blind phase and with add-on mirtazapine during theopen-label phase. c Negative correlations mean improvement on the PANSS scores in increase of body weight, blood glucose or total cholesterol. 1664  V. Terevnikov et al.  negative scores and not PANSS total scores. At the sametime, the drop in PANSS total scores corresponded withan increase in serum triglyceride level while the observedincrease in serum lipids did not predict improvement oneither PANSS positive or PANSS general subscales.Recently, Hermes  et al . (2011) did not find any significantassociation between lipid (i.e. cholesterol and triglycer-ide) levels and change in PANSS during treatment withdifferent FGAs and SGAs. However, the antipsychoticsused in their study did not include clozapine – the drugwith probably the most noticeable relationship betweenits clinical and metabolic effects (Procyshyn  et al . 2007).Our study failed to show statistically significant associ-ations between body weight gain and improvement inpsychopathology, which was reported earlier with anti-psychotic compounds (Meltzer, 2001; Ascher-Svanum et al . 2005; Hermes  et al . 2011). However, there was asimilar trend for this finding in our data. It is plausiblethat statistical significance was not reached in the presentstudy due to the small sample size and hence insufficientstatistical power.Clozapine (Lieberman  et al . 2003) and olanzapine(Leucht  et al . 1999) may have advantages in terms of effi-cacy over antipsychotics with less protuberant metabolicside-effects. A mechanism for the putative relationships between blood lipid levels, symptoms of schizophreniaand mechanism of action of antipsychotics remains ob-scure. It has been reported that low cholesterol levels can be related to maladaptive behaviours, such as aggression,violence (Golomb, 1998), suicide and self-aggression(Hawthon  et al . 1993) in some groups of psychiatricpatients. It has been hypothesized that low serum chol-esterol may be accompanied with parallel changes in thecomposition of neuronal membranes and a consequentattrition of serotonin (5-HT) receptors (Hawton  et al .1993). This, in turn, may contribute to a poor suppressionof aggressive behaviour and, in the light of recent datareportedbyStone et al .(2010),alsointreatment-resistancein schizophrenia. Another mechanism possibly explain-ing the correlation between metabolic and psychotropiceffects of mirtazapine is the relationship between 5-HT 2 -antagonism and insulin sensitivity. In a study by Gilles et al . (2005) a 5-HT 2 -antagonist ketanserin impairedinsulin sensitivity which, in turn, seems to be strongly as-sociated with hypercholesterolaemia and other forms of dyslipidaemia (Bonora  et al . 1998). From this point of view, the clinical efficacy of mirtazapine can be explained by its influence on each of the earlier-mentioned mechan-isms;i.e.changesinlipidmetabolismandadirecteffecton5-HT receptors. Similarly to clozapine, mirtazapine in-hibits also the histamine H 1  receptors. This effect may beresponsible not only for the observed metabolic changes but might also be involved in the modulation of under-lying pathophysiological processes (Mancama  et al . 2002).Even though mirtazapine in itself probably lacks anti-psychotic properties, it seems that the metabolic side-effects that mirtazapine shares with the most potentantipsychotics, whether receptor-mediated or not, may be a cause or at least a marker of additive antipsychoticefficacy.Interestingly, in a MDD study by Hummel  et al .(2011), the relationship was quite the opposite i.e. inpatients who responded to mirtazapine, both LDL chol-esterol and LDL:HDL ratio decreased. However, sincesimilar changes also occurred in responders to venlafax-ine, the question was rather about a non-drug specificmechanism of recovery from depression (e.g. HPA-mediated) than about a direct influence of mirtazapineupon the lipid levels.Although probably effective as an add-on medicationfor schizophrenia (Phan and Kreys, 2011), mirtazapineapparently increases body weight and level of chol-esterol – both proven to be cardiovascular disease riskfactors. Thus, a clinician should carefully weight thesepros and cons when considering mirtazapine add-ontreatment in each particular case.The association between metabolic side-effects andparallel improvement in the symptoms of schizophrenia,resembling those observed with clozapine were in thisstudy caused by a receptor-blocking antidepressant, notan antipsychotic – to the knowledge of the authors, anobservation never previously reported. This finding mayhave both practical (e.g. acceptance of the metabolicchanges as a tribute to an effective treatment of difficult-to-treat psychoses) and theoretical (e.g. understanding of the nature of psychosis and mechanism of action of anti-psychotics) implications.The main limitation of the present study is the smallsample size, as well as a lack of head-to-head comparisongroups of patients treated with weight gain-inducingantipsychotics e.g. clozapine or olanzapine, but withoutadd-on mirtazapine. Also, the analysis of lipids waslimited to total cholesterol, while evaluations of bodycomposition, leptin, insulin resistance index and otherparameters of the metabolic syndrome could not be con-ducted. Furthermore, the double-blind treatment phasewas relatively short. Conclusions This study demonstrates that in FGAs-treated patientswith chronic schizophrenia, mirtazapine can induceweight gain and increased total cholesterol levels, anal-ogous to those observed in earlier studies in major de-pression. To our knowledge, this is the first study todemonstrate that an increase in cholesterol level, induced bymirtazapineaddedtoFGAs,isdirectlyassociatedwithanimprovementinthepsychopathologyofschizophrenia,whichalsoseemstooccurduringmonotherapywithsomeantipsychotics. This finding may have both theoreticalandpracticalimplications.Futureresearchshouldincludea broader panel of lipid tests, a control group of patientstreated with weight gaining antipsychotics, a larger sam-ple size and a prolonged double-blind treatment.  Mirtazapine and metabolic changes in schizophrenia  1665
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks