Systematic review of clinical prediction rules for neuroimaging in the evaluation of dementia

Background: Clinical practice guidelines for dementia do not recommend rout ine neuroimaging but vary in their recommended clinical prediction rules to identify patients who should undergo neuroimaging for potentially reversib le causes of dementia. Methods: Using a MEDLINE search supplemented by other strategies, we identi fied studies from January 1, 1983, through December 31, 1998, that evaluate d the diagnostic performance of a clinical prediction rule. We calculated t he sensitivity and specificity of each rule, then evaluated their diagnosti c performance in a hypothetical cohort of 1000 patients with dementia, vary ing the prevalence of potentially reversible dementia from 1% to 15%. Results: We identified 7 studies that evaluated at least 1 of 6 different c linical prediction rules. Only one rule consistently had high sensitivity ( >85%) across all stud-ies; none consistently had high specificity (>85%). S ix of the 7 studies included less than 15 cases of potentially reversible d ementia; thus the sensitivity and specificity for each rule bad relatively wide confidence intervals. At a 5% prevalence of potentially reversible dem entia, all rules had low positive predictive value ((15%) in our hypothetic al cohort. Depending on the rule, our analysis predicts 6 to 44 of the 50 p atients with potentially reversible dementia (5% prevalence in cohort of 10 00 patients) would not undergo imaging. Conclusions: There is considerable uncertainty in the evidence underlying c linical prediction rules to identify which patients with dementia should un dergo neuroimaging. Application of these rules may miss patients with poten tially reversible causes of dementia.