Adjunctive aripiprazole is an effective and well‐tolerated means of reducing prolactin levels in patients with hyperprolactinaemia due to haloperidol, say investigators (Am J Psychiatry 2007;164:1404–10).

While they admit it may be better to stop haloperidol and substitute aripiprazole, they point out that this may not be practicable for some patients whose schizophrenia is well controlled with the older drug.

They randomised 56 patients taking haloperidol (mean dose 2025mg daily) with hyperprolactinaemia (women, 95ng per ml; men, 57ng per ml) to placebo or treatment with aripiprazole 15–30mg daily.

After eight weeks, prolactin levels were normalised in 85 per cent of patients taking aripiprazole and 3.6 per cent (one patient) of those assigned to placebo; there was no difference in response rates between men and women. Menstruation was normalised in 64 per cent of women with menstrual disturbances at baseline, with no change with placebo. Galactorrhoea ceased in one of two women taking aripiprazole but continued in three assigned to placebo. There were no changes in schizophrenia symptoms. Aripiprazole was associated with insomnia, dr y mouth and headache.

Oxford researchers have calculated numbers‐needed‐to‐treat (NNT) or harm (NNH) in a metaanalysis of randomised trials of atypical antipsychotics in the treatment of bipolar disorder (BMC Psychiatry 2007;7:40; doi:10.1186/ 1471–244X–7–40).

In five eight‐week trials involving 2206 patients presenting with depression, quetiapine was superior to olanzapine in achieving response (NNT 5.4 vs 11.6) and remission (NNT 4.3 vs 12.1). Withdrawals due to adverse reactions were more frequent with quetiapine (NNH 9.4 vs 23) but not significantly so. For patients presenting with mania, there were no significant differences in efficacy between olanzapine, quetiapine, aripiprazole and risperidone (NNTs for response and remission approximately 5). Compared with haloperidol or lithium, atypicals were similarly effective but less likely to cause withdrawals due to adverse events (NNT 10).

It is not cost effective to prescribe an atypical antipsychotic to reduce the risk of tardive dyskinesia, an economic analysis suggests (Br J Psychiatry 2007;191:238–45).

The study assumed that the rate of tardive dyskinesia with older antipsychotics is 5.4 per cent compared with 0.8 per cent with atypicals, but noted that is likely to be an over‐optimistic figure for atypicals based on biased trials. Most cases are mild and cause limited impairment or distress; recovery may occur in up to 80 per cent of patients within a year, declining to only 15 per cent in later years. Compared with schizophrenia symptoms, tardive dyskinesia is reported to have a small impact on quality of life.

Using a model that accounted for a switch from older to atypical agents in about half of patients (as per the CATIE trial), the cost per quality‐adjusted life‐year (QALY) of prescribing an atypical to reduce tardive dyskinesia risk was estimated at £75 000 to £193 000. This is unlikely to be considered cost effective at a time when NICE uses a cost effectiveness threshold of £20 000 to £30 000.

Reward‐seeking behaviour may be associated with dysfunction of glutamate in the nucleus accumbens, US researchers believe, prompting a trial of N‐acetyl cysteine (NAC) to treat pathological gambling (Biol Psychiatry 2007;62:652–7).

They randomised 27 compulsive gamblers to placebo or NAC, gradually increase the dose from 600 to 1800mg daily. After eight weeks' non‐blinded treatment, four had withdrawn from the trial; mean gambling scores decreased by 42 per cent, and 59 per cent of patients met predefined response criteria. The mean effective dose of NAC was 1477mg daily.

Thirteen of the responders agreed to a further double‐blind continuation study. After six weeks, 83 per cent of those taking NAC still met response criteria compared with 27 per cent of those assigned to placebo.

Bupropion and the antiepileptic drug zonisamide are both associated with weight loss, say US investigators. Could they offer a new strategy to reduce weight in obese adults (J Clin Psychiatry 2007; 68:1226–9)?

Eighteen obese women (mean BMI 37kg per m2) were randomised to treatment with zonisamide 100–400mg daily, alone or in combination with bupropion 100200mg daily, plus a diet delivering a 500 calorie per day deficit. After 12 weeks, mean weight loss was 2.9kg with zonisamide alone but 7.2kg with both drugs.

Analysis of prescribing statistics and suicide rates in the USA and the Netherlands has shown that warnings about prescribing antidepressants for young people has not saved lives (Am J Psychiatry 2007;164:1356–63).

In 2003/04, regulatory authorities in the US and Europe, concerned about a possible increased risk of suicidal ideation, advised against prescribing antidepressants for the under–18s. In the USA, SSRI prescribing in the under–14s subsequently decreased from 2003 to 2005 by 20 per cent overall and 30 per cent for new prescriptions. Suicide rates in 5 to 19–year‐olds increased by 14 per cent between 2003 and 2004 after years of decline, the largest increase on record.

In the Netherlands, SSRI prescribing for the under–20s decreased by 22 per cent between 2003 and 2005. By contrast, the suicide rate increased by 49 per cent and by more than a factor of four in boys aged under 15 years (though numbers were small ‐ from two to 11).

In theory, selegiline's possible neuroprotective activity could reduce neuronal damage due to oxidative stress; in practice, it may not work.

Comparing a new transdermal formulation of selegiline delivering 3 or 6mg daily with placebo in 128 patients with HIV‐associated cognitive impairment, US investigators found similar improvements in a composite score of neurocognitive function in all groups after six months. Selegiline was, however, well tolerated (Neurology 2007;69: published online ahead of print).

Relapse rates are quite high in patients taking antidepressants as continuation therapy. But, US psychiatrists argue, any group of people with depression who appear to have responded to treatment includes some who actually have a placebo response, some with a ‘true’ response to the drug, and some with both types of response. If these groups could be differentiated, what would be the ‘true’ relapse rate from drug therapy? (J Clin Psychiatry 2007;68:1271–6).

The analysis included four six to eight‐week trials in which patients were randomised to placebo or treatment with a newer antidepressant, then continued on the treatment to which they had responded. Overall relapse rates were 7.4 per cent with antidepressants and 24 per cent with placebo.

The relapse rate during drug treatment that was attributed to the loss of the placebo effect (inferred from the relapse rate among placebo responders) was 8.2 per cent. This was higher than the actual rate observed in the trials, leading the authors to suggest that much of the observed relapse rate in continuation trials may be due to a loss of placebo effect rather than a true loss of treatment effect.

Patients who develop depression after suffering a myocardial infarction are more likely to experience new cardiac events if they do not respond to antidepressant therapy, a new study has confirmed (Am J Psychiatry 2007;164:1371–8).

In MIND‐IT (Myocardial Infarction and Depression ‐ Intervention Trial), 94 patients with post‐MI depression were randomised to placebo or treatment with mirtazepine 30mg daily; non‐responders in both groups received open‐label citalopram after eight weeks. After 24 weeks, 39 per cent of patients treated with an antidepressant were classed as responders. The rate of cardiac events was 26 per cent among non‐responders, 7.4 per cent among responders and 11 per cent among untreated controls.

Swiss specialists have proposed a new way of looking at the response to antidepressants (J Clin Psychiatry 2007;68:1195–1205). They point out that analyses in clinical trials obscure wide differences in individual times to response. Further, the fixed time points for assessing patients in trials may wrongly categorise the actual time of the response. These features of trial design have led to the belief that the onset of effect of antidepressants is delayed by at least two weeks. In fact, they argue, time to response is continuously distributed over time.

They suggest that antidepressants may act by triggering a ‘resilience’‐like response that controls recovery from depression; subsequently, the pattern of recovery is independent of the antidepressant and similar to that observed with placebo. Separating the trigger and response phases in this way may facilitate new approaches to treatment, they say.

Given the demands of conducting scientifically rigorous clinical trials, it is not surprising that their findings can be difficult to extrapolate to everyday clinical practice. To find out how closely trials of disease‐modifying drugs (DMDs) in multiple sclerosis (MS) correspond to the real world, Canadian epidemiologists assessed changes in the Expanded Disability Status Scale (EDSS) for all Nova Scotia patients with MS who received a DMD between 1988 and 2004 (Neurology 2007;69:published online ahead of print).

They found that the delay in disease progression was similar to that reported in clinical trials. Compared with an expected increase in EDSS score in untreated patients, initial treatment for patients with relapsing‐remitting MS and mild to moderate disability arrests progression and may slightly reduce disability (effect size, ES, 100–112 per cent). Effectiveness against relapsing onset MS was slightly lower (ES 90 per cent), and was much less against secondary progressive MS (ES 8–21 per cent).

Change in EDSS was greater after switching drug treatment and termination of treatment, suggesting that factors such as adverse reactions, antibody development or comorbidities may correspond to increased progression.

All the available cholinesterase inhibitors improve cognitive and neuropsychiatric measures in patients with dementia with Lewy bodies (DLB), a new review suggests, and there is little difference between them (Int J Geriatr Psychiatry 2007;22:890–5).

The analysis of two non‐blinded trials of donepezil and galantamine, and a placebo‐controlled trial of rivastigmine involved a total of 106 patients. MMSE scores improved by 0.5–3.9 over 20 weeks, with wide variation about the means. Galantamine was associated with less severe Parkinsonian symptoms and donepezil reduced neuropsychiatric symptoms more. These findings justify a formal trial, the authors say.

Pregabalin may improve the cerebellar signs of cortical cerebellar atrophy (CCA), according to a Spanish report (Acta Neurol Scand 2007;116:235–8).

In this single‐blind pilot study, two patients received pregabalin 225mg daily or placebo for 15 days, separated by a one‐week washout period. Scores on the Scale for the Assessment and Rating of Ataxia (SARA) improved significantly after pregabalin but not after placebo. This was reflected in improved gait and ability to stand, clearer speech and reduced tremor. Both patients elected to continue treatment, the only recorded adverse effect being slight weight gain in one.

Risperidone has replaced thioridazine as the antipsychotic most frequently prescribed for older people living in care homes (Psychiatric Bull 2007;31:329–32).

This UK survey of 65 care homes in Leeds found that 44 per cent of residents had been diagnosed with dementia, 7 per cent with a psychotic disorder and 75 per cent with cognitive impairment of some degree. A quarter of residents had a documented medication review by the GP within the previous year.

Twenty per cent of the 331 residents had been prescribed an antipsychotic; of these, 15 per cent had a diagnosis of psychotic disorder and 69 per cent had a diagnosis of dementia. Of all patients with dementia, about one‐third had been prescribed an antipsychotic. Risperidone was prescribed for 57 per cent of patients receiving an antipsychotic; the next most popular were haloperidol (13 per cent) and olanzapine (12 per cent).

The authors say there is little evidence that antipsychotics are effective in patients with dementia but they may be prescribed in desperation in the face of challenging behaviour.

A small US study suggests there may, however, be some purpose in prescribing antipsychotics for older people with dementia (J Geriatr Psychiatry Neurol 2007;20:178;doi:10.1177/ 0891988707303335). This retrospective study reviewed case notes for 16 patients with dementia who had been prescribed clozapine for treatment‐resistant agitation. On average, they had undergone six trials of psychotropic medication.

After treatment with clozapine at a mean dosage of 41–48mg daily, 10 patients (63 per cent) were rated ‘much improved’ and two were rated ‘much or minimally worse’ for global improvement. Specific measures of agitation were also improved but there was no change in Mini Mental State Examination (MMSE). Thirteen patients (81 per cent) were discharged on clozapine. Two patients discontinued clozapine after developing delirium; one developed leucopenia and stopped treatment.

Eisai has launched rufinamide (brand named Inovelon) for adjunctive therapy in the treatment of seizures associated with Lennox‐Gastaut Syndrome (LGS) in patients 4 years and older. The company says it is the first medicine licensed by the European Agency for the Evaluation of Medicinal Products (EMEA) specifically for the treatment of LGS.

Eisai says that rufinamide produced a median reduction in tonic‐atonic (drop attack) seizure frequency relative to baseline as compared with placebo (42.5 per cent versus ‐1.4 per cent; p<0.0001). And rufinamide reduced the occurrence of total seizures compared with placebo (32.7 per cent versus 11 per cent; p=0.0015). Results from a follow‐up extension study suggested that seizure reduction was maintained for up to two years. Copyright © 2007 Wiley Interface Ltd