Do antipsychotics differ in efficacy? Eur Neuropsychopharmacol. 2014 Jul.

The idea that all antipsychotics are of same efficacy ( or comparable efficacy )  is widely accepted.  Chlorpromazine (CPZ)  has been the gold standard to compare antipsychotic effect and It would be interesting to see how all other antipsychotics fare against CPZ.

 Myrto T. Samara, Haoyin Caob, Bartosz Helferb, John M. Davisd & Stefan Leucht report the results of a meta analysis of all  randomised controlled trials that compared oral formulations of chlorpromazine with any other oral antipsychotic for the treatment of schizophrenia or related disorders.

The primary outcome was response to treatment defined  as at least 50% reduction of rating scales.


128 randomized trials published over a period of 55 years from 1956 to 2011 were included.Chlorpromazine was compared with 43 other antipsychotics.  ( risperiodne only 1 study, quetiapine 4 studies , olanzapine 4 studies, Zotepine 2, clozapine 7). The mean/median doses in the chlorpromazine group ranged between 50 and 2000 mg, with a median of 525 mg.

Chlorpromazine was significantly more efficacious than four antipsychotic drugs : butaperazine, mepazine, oxypertine, and reserpine. After Bonferroni correction ( correction for multiple comparisons) only reserpine remained statistically less efficacious than chlorpromazine.

CPZ was significantly less efficacious than four antipsychotic drugs: clomacran, clozapine, olanzapine, zotepine.

The sensitivity analysis  (exclude single blind studies, open label studies,   studies with unclear level of blindness):  confirmed the differences of chlorpromazine compared to butaperazine, clomacran, clozapine and zotepine.

There were no statistically significant differences between chlorpromazine and the remaining 28 antipsychotics.


Small sample sizes in most of the trials ( median :50)  A clinically meaningful responder difference of 10% (which would lead to 100 additional responders among 1000 treated), assuming the average response rate in this meta-analysis (47%), would require a total sample size of 390 which is not met by most comparisons.

The studies varied substantially in design, patient populations ,dosing and other factors. Most of the included studies were carried out more than 30 years ago.

Authors argue that the “equal efficacy statement” regarding antipsychotics is not evidence based.  The observation of some antipsychotics being different support their argument.  A recent Cochrane review that compared haloperidol with all other first-generation antipsychotics yielded similar results (Dold and Leucht, 2012).

One could argue that the present analysis show that CPZ do not differ in efficacy with majority of antipsychotics. Among antipsychotics in use there is difference in efficacy with clozpaine and zotepine only.

‘Equal efficacy argument ‘  has always made exemptions. (mainly for Clozapine).  This analysis also supports that clozapine is superior .

Other recent Meta analysis have shown that some SGAs have small superiority over other FGA s and SGAs.

Comments: This analysis in general support the idea that antipsychotics in general are of comparable efficacy. Clozapine’s  superiority  is supported by other evidence.

Summary of the article:

Chlorpromazine versus every other antipsychotic for schizophrenia: A systematic review and meta-analysischallenging the dogma of equal efficacy of antipsychotic drugs.

Samara MT, Cao H, Helfer B, Davis JM, Leucht S. Eur Neuropsychopharmacol. 2014 Jul;24(7):1046-55.

Genetic breakthrough in Schizophrenia? Nature. July 2014


Schizophrenia is a highly heritable disorder. Around 30 schizophrenia-associated loci have been identified through GWAS (genome-wide association study). Schizophrenia Working Group of the Psychiatric Genomics Consortium (PGC) was set up to study genetic factors  using larger samples. The group  combined  all  available schizophrenia samples with published or unpublished GWAS genotypes into a single, systematic analysis and the results are now published.

The analysis included 49 ancestry matched, non-overlapping case-control samples ( 34,241 cases and 45,604 controls) and 3 family-based samples.


108 loci were identified , 83 of this were not previously identified. 75% of this include protein-coding genes . This include DRD2 (the target of all effective antipsychotic drugs) and many genes  involved in glutamatergic neurotransmission and synaptic plasticity. Genes that encode voltage-gated calcium channel subunits are also involved.

Other interesting findings include:

1.Copy Number Variants associated with schizophrenia overlap with those associated with autism spectrum disorder  and intellectual disability.

2.Schizophrenia associations were more relevantly shown in tissues with important immune functions.

This is the largest molecular genetic study of any neuropsychiatric disorder, ever conducted and it  has produced exciting converging evidence that will soon illuminate the biology of schizophrenia and may be more effective treatments in due course.

Summary of the article:

Biological insights from 108 schizophrenia-associated genetic loci. Schizophrenia Working Group of the Psychiatric Genomics Consortium.  NATURE | VOL 511 | JULY 2014.

What matters for population well being? GDP or Gini? Jl of happiness studies. July.2014


What factors decide/ influence/ define that a society a ‘better’ one ? Economists argue that GDP is a good measure of well-being of a population.over the many years, social scientists have expanded the idea of wellbeing to include satisfaction with career, social interactions, financial status, physical health, life evaluation, emotional health, work environment, access to basic needs.. to mention a few. Rawl’s theory ( justice) focus on the redistribution of wealth to equalize access of opportunities and provision of public/social goods. The fact that countries with progressive income taxes measure better on wellbeing is very much supportive of this theory.

US, richest country in the world, is well behind in terms of wellbeing. Many studies using US data, have confirmed that it is relative wealth/ income that is more important ( than absolute ) and income disparity with in a community is excessively corrosive for wellbeing of the whole community.

Vivian Grace Valdmanis of the Health Policy and Public Health at Philadelphia analysed well-being in each of the fifty states in the US using a mixed panel model approach (2009–2011) including typical economic and political measures.


1. Well-being is positively and statistically significantly correlated with the Gini coefficient i.e. the more the income inequality in a state, less well-being it has.

2. The more politically conservative a state is, less is the well-being.
3. Per capita income itself did not explain the well-being differences. it is likely that post 2008 economic state might have unduly influenced this. However Gini index was a sensitive measure explaining wellbeing.

This is not a new finding. Inequality with in a society has harmful effects on population well-being. Political conservatism also limits wellbeing of the population.

Comment: In countries where all focus is on economic growth ( i.e. GDP), neglecting “Gini” could have serious and long-lasting negative consequences.

Summary of the article

Vivian Grace Valdmanis.Factors Affecting Well-Being at the State Level in the United States.J Happiness Stud. DOI 10.1007/s10902-014-9545-0

Do psychotherapy reduce relapse rates in depression? Jl Aff disorder. July.2014


Depression is a highly recurrent disorder. Rates  of recurrence range between 35% and 85% over a time span of 15 years  depending on nature of the sample. Psychotherapy is effective in depression in short or medium term. However some studies suggest that half of patients who responded to such therapies relapse in two years though CBT is found in some studies as providing more enduring effect.

Christiane Steinert  and colleagues asked the question: In studies of sufficient follow-up ( i.e. more than two years), what is the overall likelihood for a depressive relapse post therapy? 

The search was comprehensive and the PRISMA guidance were adhered to. All RCTs on  Adult unipolar non psychotic depression  where psychotherapy was compared against a non-psychotherapeutic comparison condition  were included. From 2985 abstracts 60 were subjected to full review and 11 were included in the meta analysis.


Long-term follow-up data were available for 966 patients. Follow-up times varied between 34 and 72 months post-therapy, with an average of about 4.4 years. All patients in the comparison conditions received some form of active alternative treatment (pharmacotherapy or some form of psychotherapy ) that was designed not to contain ingredients that are specific for psychotherapy.

Pooled overall event rate  ( observer rated relapse ) was 0.39 (95% CI 0.29, 0.50 ). Self-reported non-remission of the sample was  0.43 (95% CI 0.33, 0.53)  i.e. about 40% of patients previously treated with psychotherapy had at least one relapse (or were still judged as depressed) by the end of the follow-up. Heterogeneity of results was large ( i.e. much of the observed difference between study results were due to non random differences between studies. i.e. quality, methodology, sampling etc ).

 0.53 (95% CI 0.37, 0.68) of patients in the psychotherapy conditions and 0.71 (95% CI 0.58, 0.82) of patients in the non- psychotherapeutic comparison conditions had a relapse.  Resulting NNT was 6 i.e.  6 patients had to be treated with psychotherapy to achieve one less relapse as compared to other treatments.

Limitations: Though most studies were graded as good quality,  studies scored poorly on treatment integrity and therapist competence.  All studies were originally designed as RCTs, but  the follow-up of most of these trials has to be regarded naturalistic. Five studies only followed-up patients who were remitted after receiving different kinds of treatment in the first non-remitted patients were excluded from the follow-ups undermining the RCT’s core assumptions. Psychotherapy was delivered in different stages of a depressive disorder (remission, episode) and this limits comparability.

Conclusions: On average of 4.5 year follow-up, 40% of patients experienced a relapse post therapy. Psychotherapeutic interventions yielded a more favorable long-term outcome. Results has to be interpreted with caution given the limitations.

Summary of the article: 

Relapse rates after psychotherapy for depression – stable long-term effects? A meta-analysis. Christiane Steinert , Mareike Hofmann , Johannes Kruse , Falk Leichsenring. 2014. Journal of Affective Disorders 168 (2014) 107–118. ( Online: July 2014)


Is serum BDNF concentration a peripheral marker of depression? Molecular Psych.2014.July.


Neurotrophin hypothesis of depression was originally formulated in 1997 by Duman, Heninger and Nestler. Stress induced aberrations in neural plasticity is thought to result in depression. This is reflected in lower BDNF . Antidepressants increase the BDNF levels and correct this aberration. In 2002 , low BDNF was shown in depressed patients (Karege et al. 2002). In 2003, an increase in BDNF with antidepressant treatment was demonstrated (Shimizu et al 2003). Meta analysis have  confirmed these observations.

Molendijk  et al report current state of knowledge by a new meta analysis . From 55  studies, they  extracted 124 between-group effect- size estimates and 55 correlation coefficients.


Antidepressant-free depressed patients had lower BDNF concentrations as compared with healthy controls (d=  0.71, 95% CI = -0.89 to  -0.53).  Heathy controls and antidepressant treated groups did not differ in BDNF levels. Higher symptom severity was associated with a lower BDNF levels in untreated patients. 

Differences were still significant after correcting for publication bias.

The effects sizes are smaller than previous analysis , mainly due to  adjusting for between-study heterogeneity and publication bias.

This analysis confirms that  altered serum BDNF concentrations are peripheral manifestations of depression.  Authors raise the possibility that BDNF  cab  used  as a treatment biomarker helping us to predict the outcomes.  Another interesting question: What is the difference between pro and mature BDNF as it is thought that they have opposing actions. Would the ratio tell us more about depression or  treatment response ?

 Summary of the article: 

Serum BDNF concentrations as peripheral manifestations of depression: evidence from a systematic review and meta-analyses on 179 associations (N=9484).

Molendijk ML, Spinhoven P, Polak M, Bus BA, Penninx BW, Elzinga BM. Mol Psychiatry. 2014 Jul;19(7):791-800.

Can antibiotics cause Psychosis? J Clin Psychopharmacol. 2014 Aug.


Psychotic symptoms emerging after use of antibiotics  have been reported occasionally.  This include psychosis as well as mania with psychotic symptoms ( = antibiomania). Penicillin associated psychosis  is also known as Hoigne syndrome.  How antibiotics cause psychosis is not well understood. It is possible that GABA antagonism ( mainly by fluroquinolone ) , inhibition of prostaglandin synthesis ( = increase in dopamine) , and N-methyl-D-aspartate (NMDA) receptor hypo functioning ( due to depletion of D alanine producing intestinal flora) may be involved.

Infection itself can be associated with psychotic symptoms. UTI in the context of delirium and dementia is a well-known/ common example. Studies suggest that UTI can be the commonest cause for unrecognised medical condition on admission in geriatric patients with psychotic symptoms.

Safinaz Mostafa, and Brian J. Miller conducted a systematic review of cases of antibiotic-associated acute psychosis during treatment of a UTI and evaluated the strength of the association for each case using the guidelines for evaluation of drug-associated events.

They identified 15 cases of antibiotic- associated psychosis during treatment of UTI. These were associated with: fluoroquinolones (n = 8), trimethoprim-sulfamethoxazole  (n = 5), and penicillins (n = 2). Patients were on antibiotics for a mean 4.8 days before psychosis emerged. Two had a previous history of antibiotic related psychosis. All patients had some form of psychiatric history.

Psychotic symptoms included hallucinations (n = 12),  delusions (n = 9), disorganization (n = 5), and catatonia (n = 3) .

80% of cases with trimethoprim- sulphamethoxazole was considered as  highly suggestive of causation. 60% of all cases were considered as  strongly suggesting antibiotic to cause psychosis. 33% were moderately suggestive of this.

In most cases onset and resolution of psychosis occurred within 1 week of initiation or discontinuation of the antibiotics.  Half did not require antipsychotic treatment. In 3 cases, a recurrence of psychosis after re challenge with the same antibiotic provides strong support to the notion that antibiotics could cause psychotic states.

Limitations: Small case series. Cases associated with UTI treatment alone included.

Summary of the article:

Antibiotic-associated psychosis during treatment of urinary tract infections: a systematic review.

Mostafa S, Miller BJ. J Clin Psychopharmacol. 2014 Aug;34(4):483-90

Do statin reduce depression risk ? J Affective Disorders. 2014.May.


Statins are widely prescribed. Side effects as well as benefits of statins have recently been a focus of hot debate. Low serum cholesterol has been associated with antisocial personality, violent behaviour, suicide, and aggressive conduct. Depression is also considered as a possible consequence of statins. Although  many studies have reported benefits of statin in depression , negative associations  have also been reported.

Do statins increase the risk of depression?

Ajay K. Parsaik et al report the results of a comprehensive meta analysis  answering this question in this article. They included all comparative studies of any design . Newcastle–Ottawa Scale was used to assess quality of non randomised studies .From initial search yield of 979 articles only 7 were included in the final analysis. Analysis thus had  a total of 9187 participants.  Mean age of participants were 64+ -5 years.


Overall pooled OR for depression among statins users was 0.68 (95% confidence interval 0.52–0.89) indicating  protective effect of statins against depression. Modest heterogeneity was observed between the studies ( 55% of difference between study results were due to true differences rather than chance/random).Exclusion of one particular study contributing to this  did not change the pooled effect much.

Limitations: Adjustment for confounders varied across the studies. All studies used different scales for reporting depression.

How statin might work? Ability to reduce oxidative stress and inflammatory cytokines may be the basis of this association.

Conclusions: Statin users are less likely to be depressed.statins have a protective effect on depression.

Summary of the article:

Statins use and risk of depression: a systematic review and meta-analysis.

Parsaik AK, Singh B, Hassan Murad M, Singh K, Mascarenhas SS, Williams MD, Lapid MI, Richardson JW, West CP, Rummans TA.

J Affect Disord. 2014 May;160:62-7