Is there evidence for late cognitive decline in chronic schizophrenia?

Schizophrenia (SZP) has been historically referred to as "dementia praecox" because of the recognition that its onset is associated with deficits in memory, attention and visuospatial orientation. We wondered whether there is evidence for additional cognitive decline late in the course of chronic SZP. This review examined the evidence (1) for cognitive decline late in the course of chronic SZP, (2) for how often the late cognitive decline occurs, and (3) whether the cognitive decline in late-life SZP is related to pathophysiology of SZP versus the superimposition of another type of dementia. A PUBMED search was performed combining the MESH terms schizophrenia and dementia, cognitive decline, cognitive impairment and cognitive deficits. A manual search of article bibliographies was also performed. We included longitudinal clinical studies employing standard tests of cognition. Cross-sectional studies and those that did not test cognition through standard cognitive tests were excluded. The initial search produced 3898 studies. Employing selection criteria yielded twenty-three studies. Our data extraction tool included the number of patients in the study, whether a control group was present, the age of patients at baseline and follow-up, the study setting (inpatients versus outpatients), the cognitive tests employed, study duration, and results. Only three longitudinal studies tested for dementia using Diagnostic and statistical manual of mental disorder (DSM) or International classification of disease (ICD) criteria and compared them to controls: two studies demonstrated an increase in the prevalence of dementia and one did not. Twenty longitudinal studies tested for one or more cognitive domains without employing standard criteria for dementia: twelve studies demonstrated a heterogeneous pattern of cognitive decline and eight did not. Studies generally did not control for known risk factors for cognitive impairment such as education, vascular risk factors, apolipoprotein (ApoE) genotype and family history. The evidence for late cognitive decline in SZP is mixed, but, slightly more studies suggest that it occurs. If it occurs, it is unclear whether it is related to SZP or other risks for cognitive impairment. Hence, prospective, longitudinal, controlled studies are needed to confirm that there is progressive cognitive decline in chronic SZP which occurs independent of other risk factors for cognitive impairment.