Table 1 Data Extraction of All Studies Included in the Narrative Synthesis, with Data Quality Scores

  Blair and Curran (1999) [38]

Double-blind, independent group design

32 healthy volunteers

Diazepam

15 mg

N

Y

Emotion Processing

FERT

N/A

• Diazepam has a selective effect on the recognition of angry expressions. However, it did not affect the recognition of any of the other five expressions investigated.

• Limited sample size

• Absence of a control group of psychiatric patients

• No follow-up

12

  Coupland et al. (2003) [39]

Randomised, counterbalanced, double-blind, placebo-controlled, within-subjects comparison

28 healthy volunteers

Diazepam

15 mg

N

Y

Emotion Processing

FERT

N/A

• Diazepam produced impairments in emotional recognition accuracy. The processing of surprise and disgust were most affected.

• No follow-up

• Limited sample size

18

  Murphy et al. (2008) [40]

Randomised, between-group, double-blind, placebo-controlled design

24 healthy volunteers

Diazepam

5 mg

N

Y

Emotion Processing

FERT

N/A

• No significant effect of Diazepam on accuracy or reaction times.

• Limited sample size

• Low dosage of Diazepam

19

  Pringle et al. (2016) [41]

Double-blind intervention

36 healthy volunteers

Diazepam

15 mg

N

Y

Emotion Processing

FERT

6, 7 or 8 days

• Diazepam makes participants significantly slower on emotional face recognition than controls.

• Limited sample size

19

  Zangara et al. (2002) [42]

Double-blind independent group design

45 healthy volunteers

Diazepam

Metropolol (selective antagonist of B1 adrenoceptors)

15 mg

50 mg

N

Y

Emotion Processing

FERT

N/A

• Diazepam impairs the ability to recognise angry and fearful expressions.

• No follow-up

• Limited sample size

21

  Nilsonne et al. (2018) [43]

Double-blind randomised controlled experiment.

Wave 1 = 37 healthy volunteers

Wave 2 = 39 healthy volunteers

Oxazepam

25 mg

N

Y (Vitamin D3)

Empathy

Empathy for Pain Questionnaire

N/A

• No significant effect of Oxazepam on empathy

• Demographics of patient sample limits generalisability (all-male, largely university educated)

23

  Zurowska et al. (2018) [44]

Intergroup Difference Study

The sample comprised 43 patients with schizophrenia in three groups: (1) during detoxification from benzodiazepines (N = 13), (2) after detoxification (N = 15), (3) a matched control group (N = 15).

Diazepam

concentrations of BZD differed significantly between patients

N

N

Emotion Processing/Empathy

Computerised emotion recognition task/Empathy Quotient

N/A

• Schizophrenia patients (during detox) addicted to benzodiazepines decreased ability to recognise emotions. Specifically, negative emotions (fear, sadness, and anger) compared to healthy volunteers

• Patients going through detoxification of bzds could be experiencing more severe symptoms than those addicted – may impact general emotional outcomes – no assessment of withdrawal symptoms

• Small sample size

• Did not control for anxiety and depression

12

  Paulus et al. (2005) [45]

Double-blind, placebo-controlled, randomised dose-response study.

15 healthy volunteers

Lorazepam

0.25 or 1 mg

N

Y

Emotion Processing

Emotional Face Assessment Task - fMRI

N/A

• Lorazepam decreased activation in Amygdala and Insula when viewing emotional faces.

• No follow-up

• Limited sample size

20

  Olofsson et al. (2011) [39]

Double-blind experimental task.

45 healthy volunteers

Oxazepam

20 mg

N

Y

Emotion Processing

Affective Processing Task - EEG

1 week

• Oxazepam does not influence electrocortical indexes of emotional perception

• No patient sample

• Only one medication type

14

  Del-Ben et al. (2012) [46]

Randomised, balanced-order, double-blind, placebo-controlled crossover design

12 healthy volunteers

Diazepam

10 mg

N

Y

Emotion Processing

FERT

N/A

• Diazepam impaired the recognition of fear in female faces

• Reduced activation in right Amygdala and right OFC

• Reduced activation of bilateral ACC to angry faces

• Enhanced activation of posterior left Insula

• Limited sample size

• Patients may be aware of treatment arm

17

  Richter et al. (2010) [47]

Double-blind independent group design

6 catatonic schizophrenia patients (recovered) 16 healthy controls (8 placebo/8 Lorazepam)

Lorazepam

A dose of lorazepam 1–2.5 mg was administrated intravenously 2–4 times (mean: 5.2 mg)

N

Y (saline)

Emotion Processing

IAPS - fMRI analysis

N/A

• High signal decreases in OFC and MPFC in catatonic patients during negative emotional stimulation after Lorazepam administration

• Limited sample size

• Absence of a control group of psychiatric patients

• fMRI measurements covered only the frontal lobe – so relationship between amygdala and MPFC regarding emotional processes remains unclear

18

  Lawrence et al. (2002) [48]

2 experimental test conditions (drug vs. placebo) - crossover study design - participants who took Sulpiride in week 1 testing took placebo in week 2 testing, and vice versa

14 healthy volunteers

Sulpiride

400 mg

Testing commenced 100 min following tablet (drug or placebo) ingestion in order to maximise drug levels during test administration. In order to provide an adequate washout period, two test sessions were separated by a median interval of 3 weeks. In each of the two testing sessions, participants completed a test of emotion recognition from the face and a control task of unfamiliar face matching (the Benton task).

Y (lactose)

Emotion Processing

FERT

baseline, ~ 3 weeks

• Following Sulpride use, recognition of anger facial expression at follow-up was impaired compared to baseline, other emotions intact

• Limited sample size

• Short follow-up time

13

  Rock et al. (2016) [49]

Between-subject, randomised, double-blind, placebo-controlled design

40 healthy volunteers

Quetiapine

150 mg

27 received Quetiapine for 7 days - dropout to n = 20 for Emotion Processing task

Y

Emotion Processing

FERT

baseline, one week

• No effect of Quetiapine on emotion processing ability in healthy participants at one week, compared to baseline

• No compliance measure

• Healthy volunteers only

• One-week duration only

• Modest sample size

• Dropout in Quetiapine arm (reduction of power)

• Authors consultants for pharmaceutical company

22

  Behere et al. (2009) [50]

Short-term treatment follow-up

55 antipsychotic-naïve schizophrenia patients30 healthy controls

Risperidone

4 mg/daily

25 drug-naïve schizophrenia (DSM-IV) patients

N

Emotion Processing

TRENDS

Not specified (short-term)

• Schizophrenia patients showed impairments in emotion processing at baseline compared to healthy controls

• Risperidone use in schizophrenia patients resulted in improvements in patient scores on the emotion processing task, when comparing their scores at baseline and follow-up

• Non-specified follow-up duration - may be practice effects

• Only one antipsychotic type

• Non-randomised design

16

  Gaebel et al. (1992) [51]

Experimental task

23 schizophrenia patients 21 MDD 15 healthy volunteers

13 Perazine 10 Haloperidol (schizophrenia patients only)

The mean daily/cumulative dosages were 376/10160 mg CPZE and 445/16400 mg CPZE respectively.

11/23 schizophrenia patients’ drug-naïve, remaining 12 drug-free

N

Emotion Processing

FERT

baseline and 4 weeks

• Schizophrenia patients showed impairments in emotion processing at baseline compared to healthy controls

• Both schizophrenia patient and healthy control groups improved at follow-up, larger improvements in schizophrenia patient group

• Pratice effects due to short follow-up time

• Mixture of drug-naïve and drug-free patients

18

  Olivier et al., (2015) [52]

Case-control design over 12 months.

92 FEP patients 100 healthy volunteers

Flupenthixol Decanoate (LAI)

10 mg

<  4 weeks of treatment (not a statistically significant difference at baseline, but difference is present)

N

Emotional Intelligence

MCCB

6-month, 12 month

• FEP performed significantly worse at baseline in all cognitive domains bar social cognition compared to healthy controls

• FEP significantly improved in all MCCB domains (including social cognition) between baseline and 6 months.

• No further improvements were seen in social cognition at 12 months in the FEP group, suggesting stability of emotional intelligence over time.

• Additional oral flupenthixol was prescribed at the discretion of the investigator

• Not all patients were tested in their first language

• Patients were not necessarily antipsychotic naïve

• One antipsychotic type

• FEP only

16

  Zhou et al. (2017) [53]

12-week treatment study

56 schizophrenia inpatients 28 healthy volunteers

haloperidol (n = 12), fluphenazine (n = 8), chlorpromazine (n = 6), or trifluoperazine (n = 2).

Risperidone (n = 28)

The mean chlorpromazine-equivalent dose was 502.0 ± 198.3 mg/d.

The mean (±standard deviation) dose of risperidone was 4 ± 1.5 mg/d.

In the risperidone treatment group, 19 patients were drug-naive and 9 were drug-free (5 for at least 6 months and 4 for at least 1 month). In the typical antipsychotic treatment group, 17 patients were drug-naive and 11 were drug-free (8 for at least 6 months and 5 for at least 1 month).

N

Emotion Processing

FEDT

baseline, 4 weeks, 12 weeks

• Schizophrenia patients showed impairments in emotion processing at baseline compared to healthy controls

• Risperidone improved social cognition in schizophrenia patients after 12 weeks compared to baseline, but not at 4 weeks.

• Mixture of drug-naïve and drug-free patients

18

  Lewis et al. (1995) [54]

Experimental task

18 psychosis patients 10 healthy volunteers

Haloperidol

5-20 mg

Drug-free at baseline (for an unspecified time period)

N

Emotion Processing

FERT

baseline and 2 weeks

• Schizophrenia patients showed impairments at emotion processing at baseline compared to healthy controls

• Haloperidol had no effect on patient performance at follow-up compared to baseline scores

• Small sample size

• Short follow-up time period - practice effects

• Did not subtype psychotic patients

• Patients were not antipsychotic naïve

16

  Wölwer et al. (1996) [55]

Experimental task

32 acute schizophrenia inpatients (S/a) 36 remitted schizophrenic patients (S/r) 21 healthy volunteers

Perazine

Haloperidol

Chlorpromazine

Clozapine

S/r and S/a only

The S/a were orally treated with either perazine (n = 20) or haloperidol (n = 12). The average daily dosage in chlorpromazine equivalents (CPZE) in the T0-T1 interval did not differ significantly (perazine: 436 + 217 mg CPZE; haloperidol: 531 + 313 mg CPZE). Among S/r 10 patients were treated with clozapine (mean daily dosage = 426 − +  144 mg CPZE), 21 received typical neuroleptic drugs either orally or as depot (mean daily dosage = 477 + 430 mg CPZE)

Among S/r 10 patients were treated with clozapine (mean daily dosage = 426 − +  144 mg CPZE), 21 received typical neuroleptic drugs either orally or as depot (mean daily dosage = 477 + 430 mg CPZE) and 5 patients were drug-free in the T0”-T1” interval. Five S/a, but none of the S/r, received anticholinergic medication.

N

Emotion Processing

FERT

baseline and 4 weeks

• Acute and remitted schizophrenic patients demonstrated a stable deficit in emotion recognition compared to healthy controls.

• Antipsychotic medication had no effect on patient performance at follow-up compared to baseline scores.

• Non-randomised design

• Short follow-up - practice effects

• Patients were not antipsychotic naïve

15

  Herbener et al. (2005) [56]

Short-term follow-up study

13 schizophrenia patients 13 healthy volunteers

Risperidone

Ziprasidone

Aripiprazole

Haloperidol

mean dose

R = 3.38 mg

Z = 140 mg

A = 30 mg

H = 4.5 mg

<  4 weeks prior antipsychotic treatment in lifetime

N

Emotion Processing

CNB

baseline, average 31.3 days later (where clinically stable)

• Schizophrenia patients showed impairments at emotion processing at baseline compared to healthy controls• Antipsychotic medication had no effect on patient performance at follow-up compared to baseline scores

• Limited sample size

• Non-randomised design

• Short follow-up time - practice effects

11

  Daros et al. (2014) [57]

Blocked experimental task

54 Healthy volunteers

29 Schizophrenia patients

28 Bipolar Disorder patients

Schizophrenia

Risperidone (79.2%)

Aripiprazole (12.5%)

Haloperidol (8.3%)

Ziprasidone (4.2%).

Bipolar Disorder

Risperidone (86.7%)

Olanzapine (6.7%)

Drugs in chlorpromazine equivalents was 326.9 mg (SD = 218.9; range: 34.4–907.8 mg) for SCZ patients and 154.4 mg (SD = 125.7; range: 34.4–524.6 mg) for BP patients.

FEP patients.

At study entry, some patients with SCZ and BP had previously been exposed to atypical antipsychotics (45.0%), antidepressants (30.0%), typical antipsychotics (15.0%), mood stabilizers/anticonvulsants (12.5%), and stimulants (12.5%), typically for brief periods of time in the months preceding their participation. No patient had taken a dose of any of these medications within three days of assessments, with the exception of BP (6.3%) and SCZ (12.5%) patients who were on maintenance antidepressant treatment started prior to study entry. Up to four weeks of prior cumulative lifetime antipsychotic treatment was allowed.

Y

Emotion Processing

CNB

baseline and an average of 6.8 weeks

• Schizophrenia patients showed impairments on emotion processing at baseline compared to healthy controls

• Compared with controls, schizophrenia patients were worse at recognising mildly and moderately sad expressions at follow-up.

• At follow-up, schizophrenia and bipolar disorder patients did not significantly differ from each other on any emotion category.

• Non-randomised design

• Authors consult for pharmaceutical company

12

  Kucharska-Pietura et al. (2012) [58]

Naturalistic treatment conditions

100 schizophrenia patients 50 healthy volunteers

Typical

Atypical

Not stated

Twenty-eight (13 males) were treated with FGAs (perphenazine, n = 14; haloperidol, n = 14) and 56 (31 males) were treated with SGAs (olanzapine, n = 28; clozapine, n = 28). All patients were clinically stable after 3–4 weeks of antipsychotic treatment.

N

Emotion Processing

Theory of Mind

Empathy

FERT

‘Reading in the Minds Eye’ test

Balanced Emotional Empathy Scale

N/A - cross-sectional

• Schizophrenia patients showed impairments at emotion processing at baseline compared to healthy controls

• Antipsychotic medication had no effect on patient performance compared to healthy controls

• Non-randomised design

• No follow-up evaluation

• Antipsychotic medication not specified

15

  Kee et al. (1998) [59]

Baseline phase, brief placebo washout, and two double-blind phases 8 weeks double blind

18 schizophrenia patients

Haloperidol

Risperidone

15 mg

6 mg

During baseline, patients received 15–30 mg/day of haloperidol for 3 weeks. This phase was followed by a period of 3–7 days of placebo wash-out. Upon entering the subsequent double-blind phases, patients first were randomly assigned to receive either 6 mg/day of risperidone or 15 mg/day of haloperidol for 4 weeks (fixed- dose phase). In the second double-blind phase, which also lasted for 4 weeks, medication doses from the previous phase could be changed according to symptom and side-effect considerations (flexible-dose phase).

N

Emotion Processing

FEIT

baseline and 8 weeks

• Risperidone improved the ability to perceive emotions compared to Haloperidol at follow-up compared to baseline

• Small sample size

• Short follow-up time period - practice effects

• Patients were not antipsychotic naïve

19

  Harvey et al. (2006) [60]

8 week, multicentre, double-blind, parallel-designed, randomised, flexible-dose study

166 Schizophrenia patients

Risperidone

Quetiapine

2-8 mg/daily

200-800 mg/daily

Sleep medication and benzodiazepines were allowed as needed but were not allowed within 24 h of clinical or neuropsychological assessments. Participants were taking antipsychotic medication at the start of the study and there was no titration period.

N

Emotion Processing

CNB

baseline, 8 weeks

• No significant differences associated with antipsychotic treatment at follow-up compared to baseline

• Supported by pharma company

• Patients were not antipsychotic naïve

• Short follow-up time period - practice effects

• High drop-out at follow-up (%) – low generalisability

18

  Mizrahi et al. (2007) [61]

Cross sectional study and a longitudinal study

17 FEP patients

Clozapine

Risperidone

Olanzapine

Loxapine

Clozapine = 300 (n = 1) and 225 mg (n = 1)

Risperidone = 4 mg (n = 4), 3 mg (n = 1), 3.5 mg (n = 1), or 1 mg (n = 1).

Olanzapine = 10 mg (n = 4), 20 mg (n = 1), 15 mg (n = 1), or 2.5 mg (n = 1).

Loxapine = 35 mg (n = 1)

Most subjects were started on atypical antipsychotic medications, except for two patients who were restarted on their previous clozapine dose (300 and 225 mg). The rest were started on risperidone 4 mg (n = 4), 3 mg (n = 1), 3.5 mg (n = 1), 1 mg (n = 1) or olanzapine 10 mg (n = 4), 20 mg (n = 1), 15 mg (n = 1), 2.5 mg (n = 1), and one patient was restarted on her previous 35 mg of loxapine.

N

ToM

Hinting Task

baseline - 6 weeks

(measured every 2 weeks)

• Greatest improvement in ToM occurred during first 2 weeks of antipsychotic treatment, compared to baseline

• FEP patients only

• Mixture of antipsychotic-naïve and drug-free patients

• Non-randomised design

• Short follow-up time period - practice effects

18

  Sergi et al. (2007) [62]

8, week double blind, randomised study

73 outpatients with schizophrenia-spectrum disorder

Risperidone

Olanzapine

Haloperidol

4 mg

15 mg

8 mg

Patients were initially enrolled and tested at baseline on their pre-study medication; there was no medication washout period.

N

Emotion Processing/Social Perception

Half-profile of non-verbal sensitivity/IPT-15

baseline, 8 weeks

• No significant changes in social cognition associated with treatment over an 8-week study period.

•Pharmaceutical funding- medications for the study were provided by pharmaceutical companies

• Modest group size and two random assignment paths - limited statistical power

• Short follow-up time period - practice effects

• Patients were not antipsychotic-naïve (no washout period)

21

Mizrahi et al. (2008) [63]

Cross sectional study and a longitudinal study

17 FEP patients

Typical Atypical

Not stated

The study was a cohort of consecutively admitted antipsychotic-free patients to the inpatient and outpatient Schizophrenia program who were willing to start antipsychotic medication. Patients had previously untreated psychosis and were antipsychotic-naïve at the beginning of the study, or had started or changed medication to improve symptoms in the previous 48 h.

N

Attribution Style

IPSAQ

baseline, 6 weeks

• Attributional style scores did not change during 6 weeks of antipsychotic treatment

• Small longitudinal cohort – may not have sufficient power

• Short follow-up time period - practice effects

• FEP patients only

• Antipsychotic medication not specified

11

  Fakra et al. (2009) [64]

Controlled, open, randomised and prospective design.

25 schizophrenia patients

Haloperidol

Risperidone

Not stated

Followed a wash-out period of at least 1 week for prior antipsychotic treatment. Random assignment to Haloperidol or Risperidone treatment groups.Use of other antipsychotics or long-life benzodiazepines was prohibited. Benzodiazepines were not administered for a minimum of eight hours before emotional testing.

N

Emotion Processing

FEDT

baseline, 2 weeks, 4 weeks

• Greater beneficial effect of Risperidone than Haloperidol in schizophrenic patients’ ability to discriminate facial emotions at follow-up compared to baseline

• Small sample size

• Patients were not antipsychotic-naïve

• Short follow-up time period - practice effects

17

Penn et al. (2009) [65]

Random assignment to double-blind intervention

873 schizophrenia patients

Olanzapine

Quetiapine

Fumarate

Risperidone

Ziprasidone

Perphenazine

(Zyprexa, Eli Lilly) (7.5 mg), (Seroquel, AstraZeneca) (200 mg)

(Risperdal, Janssen Pharmaceuticals) (1.5 mg)

(Trilafon, Schering-Plough) (8 mg)

(Geodon, Pfizer) (40 mg)

Overlap in the administration of the antipsychotic agents that patients received before study entry was permitted for the first four weeks after randomization to allow a gradual transition to study medication. Concomitant medications were permitted throughout the trial, except for additional antipsychotic agents.

N

Emotion Processing

FEDT

baseline and 2 months

• Patients in all treatment groups (with the exception of Ziprasidone) showed small, non-significant improvements in emotion perception from baseline to two months

• Authors consult for pharma companies

• Medications provided by pharma companies

• Patients were not antipsychotic-naïve (medication was gradually titrated over 4 weeks following randomisation)

22

  Roberts et al. (2010) [66]

Randomised, double-blind clinical trial.

223 Schizophrenia-spectrum patients

Olanzapine

Quetiapine

olanzapine mean dose = 15.6 mg quetiapine mean dose = 455.8 mg

Chlorpromazine equivalents of these doses are 312 mg/day and 607.7 mg/day, respectively.

Participants entered a 2-week titration period during which they were switched from their current medication to Olanzapine or Quetiapine.

N

Social Perception

SCRT

baseline and 6 months

• Olanzapine and Quetiapine significantly improve performance on 3/4 social cue recognition tasks at follow-up compared to baseline

• Patients were not antipsychotic-naïve (medication was titrated over 2 weeks following randomisation)

• Pharmaceutical funding

20

  Maat et al. (2013) [67]

8 week, randomised, multicentre, open-label study

48 schizophrenia patients

Aripiprazole

Risperidone

maximum 30 mg

maximum 6 mg

Overlap in the administration of the antipsychotic agents that patients received before study entry was permitted for the first 2 weeks after randomisation to allow for gradual transition. Concomitant medication other than antipsychotics was permitted throughout the trial; the dosage was restricted to a maximum of 30 mg diazepam or equivalent, 120 mg propranolol, and 12 mg biperiden or equivalent.

N

Emotion Processing

FERT

baseline, 8 weeks

• No significant effect of medication-group on endpoint performance on social cognition at follow-up compared to baseline

• High drop-out rate (few follow-ups)

• Short follow-up time period - practice effects

• Funded by pharma company

• Patients were not antipsychotic-naïve

17

  Shi et al. (2016) [68]

Single-arm, open-label study

95 Schizophrenia patients

Paliperidone

3-12 mg/daily

Single antipsychotic usage for at least 4 weeks before study.

N

Emotional Intelligence

MCCB

baseline, 6 months

• Treatment associated with improvements in 5/6 cognitive domains, but not social cognition

• Funding from pharma company

• Open-label, single-arm design (efficacy bias)

19

  Koshikawa et al. (2016) [69]

6 month pilot, open-label, randomised controlled study

21 Schizophrenia-spectrum patients

Paliperidone Palmitate Risperidone (LAI)

PP- doses of the drug were adjusted according to clinical status, upper limit of 50 mg /2 weekly. R (LAI)-The dose was determined depending on patient’s clinical status, with an upper limit of 150 mg/monthly

Inclusion:

Having received risperidone long-acting injection for 2 months or longer.

Exclusion: Current treatment with oral risperidone or oral palmitate risperidone.

Current treatment with multiple oral antipsychotics.

N

Emotion Processing

SECT

baseline, 6 months

• No significant differences between the two groups in terms of the SECT accuracy at follow-up

• Small sample size

• Patients were not antipsychotic-naïve (excluded if they were not currently being treated with antipsychotic medication)

21

  Gultekin et al. (2017) [70]

Longitudinal naturalistic study

19 Schizophrenia-spectrum patients

Clozapine Risperidone

CPZE equivalent = 600 mg/day CPZE equivalent = 800 mg/day

being under current antipsychotic treatment included in inclusion criteria

N

Emotion Processing

FERT

baseline, 16–20 weeks

• Ability to recognise disgust faces poorer by a significant amount in the Risperidone group compared to the Clozapine group at baseline and significantly poorer after treatment with Risperidone then with Clozapine at follow-up.

• Mean responses to facial emotions significantly shorter after Clozapine and Risperidone than at baseline

• Small sample size

• Patients were not antipsychotic-naïve

16

  Savina et al. (2007) [71]

Experimental task

Naturalistic design

84 schizophrenia-spectrum patients 24 healthy volunteers

clozapine (n = 18)

olanzapine (n = 20)

risperidone (n = 23)

perphenazine (n = 2)

fluphenazine (n = 8)

flupentixol (n=6)

zuclopenthixol (n = 4)

stelazine (n = 1)

haloperidol (n = 2)

Not stated

received clozapine (n = 18), olanzapine (n = 20), risperidone (n = 23) or typicals (n = 23), including perphenazine (n = 2), fluphenazine (n = 8), flupentixol (n = 6), zuclopenthixol (n = 4), stelazine (n = 1) and haloperidol (n = 2), for at least 4 months. Most were also receiving mood stabilizers or other medications, but these were not systematically recorded. However, treating physicians were asked not to refer patients who received anticholinergic medication.

N

ToM

First-order Belief Task

N/A - cross-sectional

• Olanzapine and Clozapine groups performed similar to healthy controls on ToM task.

• Risperidone and typical antipsychotic groups performed worse on ToM task (compared to healthy controls)

• Non-randomised design

• No follow-up evaluation

• Patients were not antipsychotic-naïve

13

  Kucharska-Pietura et al. (2012) [72]

Naturalistic, pragmatic sample

84 Schizophrenia-spectrum patients

FGAs and SGAs

Not stated

39 patients were treated using conventional antipsychotic drugs (perphenazine, perazine, fluphenazine, haloperidol) and 61 were treated with atypical antipsychotic drugs (olanzapine, risperidone, amisulpride, clozapine and quetiapine). All patients were clinically stable after 4 weeks of antipsychotic use

N

Emotion Processing

FERT

N/A - cross-sectional

• No significant differences in performance between typical and atypical treatment groups.

• Non-randomised

• No follow-up evaluation

15

 Labuschagne et al. (2013) [73]

Experimental task

113 Early HD patients

Neuroleptics

Not specified

Not stated

Of those taking neuroleptics (n1/429) almost all of the patients were on atypical neuroleptics except for one patient; the most common neuroleptic taken was olanzapine (14 patients). The neuroleptic daily dose range (expressed as the equivalent dose of chlorpromazine) was 50–800 mg. These patients may have been taking additional medications such as SSRI’s that were not fully listed. Adjusted for stage of disease

N

Emotion Processing

FERT

N/A

• In early HD neuroleptic use was associated with worse facial emotion recognition compared to those not using neuroleptics

• Emotion recognition deficits in HD may be due to facial perception impairments

• Time constraints in testing – presenting only 10 stimuli per emotion

• Single channel of emotion processing – faces only

13

  Sumiyoshi et al. (2009) [74]

Longitudinal treatment design

20 outpatients with schizophrenia

Perospirone

Dose adjusted to optimise improvement in symptoms. Subjects who had already been treated with antipsychotic drugs, had medication switched stepwise to Perospirone monotherapy during the initial 6 weeks.

7/20 drug-free, 13/20 on antipsychotic medication

N

Social Perception

Script Tasks

baseline, 6 months

• Perospirone was associated with an increase in P300 ERP in the left PFC.

• Performance on script tasks (social cognitive task) was improved during treatment, positively correlated with P300 changes.

• Subjects heterogeneous in terms of premedication

• Small sample size due to large drop-out rate

• Funding from pharmaceutical company

15

  Takahashi et al. (2005) [75]

Single-blind, randomised, placebo-controlled design study.

13 healthy volunteers

Sultopride Fluoxetine (antidepressant)

25 mg

50 mg

N

Y (lactose)

Emotion Processing

Affective Processing Task - fMRI

Not specified 3 sessions

• After antipsychotic administration healthy volunteers showed decreased BOLD responses in limbic areas when viewing emotional stimuli

•Pharmacological actions may be on vascular and respiratory systems which in turn effect BOLD

• Only healthy volunteers used

• Pharmacological changes did not represent the minimal behavioural changes

14

  Franken et al. (2008) [76]

Randomised, double-blind, placebo-controlled crossover design.

32 healthy volunteers

Bromocriptine (Beta-Blocker)

Haloperidol

2.5 mg

2 mg

All subjects received a single oral dose of placebo (lactose), bromocriptine (2.5 mg), and haloperidol (2 mg) in a counterbalanced order. The medication was provided by the pharmacy of the Erasmus Medical Centre in indistinguishable capsules.

Y (lactose)

Emotion Processing

Affective Processing Task - EEG

weekly (for each condition −3 weeks total)

• Low dose haloperidol and bromocriptine did not change ERPs towards affective stimuli.

• Substantial dropout in Bromocriptine group – lower generalisability

• Low doses of medication - due to unwanted side effects

• Some participants received Domperidone to treat nausea

21