Do cholinesterase inhibitors enhance cognition in schizophrenia?

Other studies of AChEI augmentation of typical or atypical antipsychotics have been:

• equivocal, reporting benefits in some but not all patients (with no clear statistical or clinical conclusions) or in schizophrenia patients with comorbid dementia^11-14
  
• decisively negative, showing no benefits, particularly in comparatively larger, randomized, placebo-controlled trials (Table 2).^15-19
  

Meta-analysis power. In an attempt to understand these wide-ranging results, Chouinard et al²⁰ performed an elegant meta-analysis of oral AChEI augmentation therapies for cognitive enhancement in schizophrenia. This review emphasized the available studies’ complexity, small number and sample sizes, and small benefit effect sizes.

The authors concluded that—based on preliminary data—adjunctive AChEIs seemed to have “some beneficial effects” on attention and memory for schizophrenia patients.

The last word? Within weeks, however, results of a large multicenter trial by Keefe et al²¹ showed that donepezil augmentation was no more effective than placebo in improving cognition in patients with schizophrenia or schizoaffective disorder. In this 38-center, randomized, double-blind, placebo-controlled, parallel design study, 250 patients with mild to moderate cognitive impairment received adjunctive donepezil—5 mg/d for 6 weeks, then 10 mg/d for 6 weeks—or placebo for 12 weeks.

Both the treatment and placebo groups experienced statistically and clinically significant benefits from baseline in measures of cognition, positive symptoms, and negative symptoms. For all measures, placebo augmentation was equal to or superior to donepezil augmentation.

Table 2

Controlled trials: No benefit from AChEIs in schizophrenia

Study design	Subjects	Drug (dosage)	Results
Friedman et al (2002),15 double-blind, placebo-controlled	36 patients with schizophrenia	Donepezil, 5 or 10 mg/d for 12 weeks	Neither dose produced significant improvement in any cognitive measure
Tugal et al (2004),16 double-blind, placebo- controlled, crossover	12 patients with stable schizophrenia	Donepezil, 5 mg/d for 6 weeks, with crossover to placebo for 6 weeks	Treatment effect was not significant in any cognitive measure
Freudenreich et al (2005),17 double-blind, placebo-controlled	36 stable outpatients with schizophrenia	Donepezil, ≤10 mg/d for 8 weeks	No improvement in cognition or psychopathology measures
Sharma et al (2006),18 randomized, double-blind, placebo-controlled	21 patients with stable schizophrenia	Rivastigmine, 12 mg/d for 24 weeks	No significant improvement in any cognitive measure
Fagerlund et al (2007),19 double-blind, placebo-controlled	21 patients enrolled, 11 completed	Donepezil, 5 or 10 mg/d for 4 months added to ziprasidone	No differences in changes on PANSS scores or a global cognitive score
Keefe et al (2007),21 randomized, double-blind, placebo-controlled	250 stable outpatients with schizophrenia or schizoaffective disorder	Donepezil, 5 mg for 6 weeks then 10 mg for 6 weeks	Donepezil was well-tolerated but did not improve cognition any more than placebo
PANSS: Positive and Negative Syndrome Scale

Analyzing trial results

The large, well-designed clinical trial by Keefe et al²¹ suggests conclusively that donepezil augmentation is not more effective than placebo in most stable schizophrenia or schizoaffective disorder patients with mild to moderate cognitive impairment.

Even so, it is arguably difficult to “prove a negative.” For example:

• Different dosages might have been more effective.
  
• Longer treatment (>3 months) might have been necessary for donepezil to “surpass” the large placebo effect.
  
• Other AChEIs—such as galantamine, which stimulates nicotinic receptors—might be more effective than donepezil, which is predominantly muscarinic.
  

‘Subgroup’ hypothesis. Finally, if schizophrenia’s pathophysiology is extremely heterogeneous, AChEI augmentation might benefit only the small subgroup of patients with decreased cholinergic activity. Most other patients—without decreased cholinergic activity—would not benefit or might even worsen. In support of the “subgroup” hypothesis, Miller²² has reported that many augmentation agents have efficacy in schizophrenia—but only in a minority of patients.

If this hypothesis is true, clinicians would need to differentiate patients before giving them trials of AChEIs or other augmentation therapies. Genetic testing might identify different pathophysiologies among patients, but these technologies are not yet clinically available.

Recommendations

Clinical experience, case reports, and small case series indicate that occasional patients may benefit from AChEI augmentation. On the other hand, the only large, multi-center, placebo-controlled, parallel-design study found no difference between donepezil and placebo augmentation of atypical antipsychotics.²¹

Thus this review of available evidence does not support the routine use of AChEI augmentation of typical or atypical antipsychotics as a viable psychopharmacologic strategy. Until more supportive evidence has been reported, this reviewer cannot recommend AChEIs as a “first line” augmentation strategy. Furthermore, because these medications do not have an FDA-approved indication in schizophrenia and are expensive, a cost-benefit appraisal also would not support their routine use.

Nevertheless, AChEIs are relatively safe and occasionally have been dramatically effective in a small subgroup of schizophrenia patients when used as augmentation. They may represent a reasonable approach: