To determine the clinical efficacy of piracetam for the features of dementia or cognitive impairment, classified according to the major subtypes of dementia: vascular, Alzheimer's disease or mixed vascular and Alzheimer's disease, or unclassified dementia, or cognitive impairment not fulfilling the criteria for dementia.
The trials were identified from a search of the Specialized Register of the Cochrane Dementia and Cognitive Improvement Group on 10 November 2000 using the term spiracetam, nootropic and 2-Oxo-1-pyrrolidine. In addition the pharmaceutical company responsible for marketing most of the piracetam worldwide, UCB Pharma, provided a comprehensive list of abstracts, which included many unpublished studies. As many of these unpublished, placebo-controlled studies will be reviewed as possible.
All unconfounded trials specified as randomized in which treatment with piracetam was administered for more than a day and compared with placebo in patients with dementia of Alzheimer type, vascular dementia,or mixed vascular and Alzheimer's disease, or unclassified dementia, or cognitive impairment not fulfilling the criteria for dementia.
Data were extracted independently by two reviewers. Each study was independently verified as fulfilling the inclusion criteria. Studies were rated for methodological quality by assessment of blinding and loss before analysis as described by Jadad et al. (1996). Studies were pooled if appropriate and possible, and the pooled odds ratios (95%CI) or the average differences (95%CI) were estimated. Where possible, intention-to-treat analyses were undertaken. Sensitivity analyses were performed to determine if successive elimination of those studies performing most poorly on these quality criteria changed the effect estimate.
Unfortunately, many of the studies were of cross-over design and first-phase data were unavailable, or could not be extracted. Global Impression of Change was the only outcoeme for which there was a significant volume of evidence from the pooled data. There was evidence of heterogeneity in the results from the individual studies, chi-square test = 20.8 (df=5). Using a fixed effects model the odds ratio for improvement in the piracetam group compared with the placebo group was 3.55, [95% CI][2.45, 5.16]. If a random effects model was used the odds ratio was 3.47 [1.29, 9.30]. If one single-blind study was excluded, the fixed effects model yielded an odds ratio of 3.36 [2.29, 4.99] and if a random effects model was applied then the odds ratio was 2.89 [1.01, 8.24]. The evidence of effects on cognition and other measures, was inconclusive.
At this stage the evidence available from the published literature does not support the use of piracetam in the treatment of people with dementia or cognitive impairment. Although effects were found on global impression of change, no benefit was shown by any of the more specific measures. There is a need for further evaluation of piracetam by : 1) Obtaining the data from the identified studies for an individual patient database review, 2) Performing a randomized trial of piracetam in patients with diagnoses made by currently accepted diagnostic criteria. The trial should extend over for a period of at least 6 months and preferably longer. Specific cognitive instruments which are sensitive to change, Clinician Global Impression of Change, levels of dependency and caregiver quality of life scales should also be incorporated in such a study.