A systematic review including 48 studies (41 diagnostic cohorts and 7 diagnostic case-controls) with a total of 2 274 subjects was abstracted in DARE. Patients with Alzheimer disease (AD) were also included. The diagnostic performance of 99mtechnetium-hexamethyl-propylenamine oxime (Tc-HMPAO)-SPECT imaging in distinguishing Alzheimer disease (AD) from other dementias was assessed. The reference standard used to determine the clinical diagnosis was not specified. In 13 studies (n=768) the accuracy of SPECT in distinguishing AD from vascular dementia was evaluated. The pooled estimates of sensitivity and specificity were 71.3% (95% CI 67.5 to 75.2) and 75.9% (95% CI 70.8 to 81.1), respectively. In 7 studies (n=424) the accuracy of SPECT in distinguishing AD from fronto-temporal dementia was evaluated. The pooled estimates of sensitivity and specificity were 71.5% (95% CI 66.3 to 76.7) and 78.2% (95% CI 71.2 to 85.2), respectively. In 13 studies (n=1082) the accuracy of SPECT in distinguishing AD from non-dementia patients was evaluated. The pooled estimates of sensitivity and specificity were 65.7% (95% CI 62.2 to 69.3) and 79.1% (95% CI 75.1 to 83.1), respectively.
Comment: The quality of evidence is downgraded by limitations in review quality (inappropriate pooling of heterogenous data, poorly defined reference standard, conclusions rely upon data published elsewhere and not presented in the review). The benefit for patients of slightly increasing or decreasing the probability of Alzheimer’s disease is questionable. The sensitivity of SPECT is so poor that Alzheimer’s disease cannot be excluded, and the effect of cholinergic drugs often has to be determined in a therapeutic trial.