TABLE 5

¹⁸F-FDG PET and AD Diagnostic Studies Reporting Likelihood Ratios

Reference and study type	Diagnostic standard	Subjects	Likelihood ratios
Panegyres et al. 2009 (35), prospective cohort study of diagnostic utility of ¹⁸F-FDG PET	Clinical diagnosis based on longitudinal long-term assessment	Community-dwelling subjects presented to primary care center for cognitive complaints. Final clinical diagnosis was early-stage AD (n = 49), non-AD dementia (n = 29), depression (n = 11), or miscellaneous (n = 13).	Positive likelihood ratio for ¹⁸F-FDG PET scan considered consistent with AD was 4.11 (95% CI, 2.29–7.32), suggesting increase in likelihood of final diagnosis of AD when diagnosed on ¹⁸F-FDG PET with AD. Negative likelihood ratio for AD was 0.27 (95% CI, 0.16–0.46), suggesting more significant decrease in likelihood of final diagnosis of AD when ¹⁸F-FDG PET findings are negative for AD. Probability before ¹⁸F-FDG PET that patient had AD was 48%. After ¹⁸F-FDG PET, probability increased to 79%, indicating that ¹⁸F-FDG PET increases diagnosis probability of early-onset AD from 48% to 79%.
Jagust et al., 2007 (39), historical cohort study	Postmortem diagnosis	Forty-four individuals with dementia, cognitive impairment, or normal cognitive function underwent clinical initial evaluation and PET and were followed up for approximately 4 y until final evaluation and 5 y until death and autopsy. Clinical, pathologic, and imaging diagnoses were categorized as AD or not AD.	Positive likelihood ratio of ¹⁸F-FDG PET for AD diagnosis was 3.2, and negative likelihood ratio was 0.21.
Silverman et al., 2001 (42), multicenter retrospective analysis	Postmortem diagnosis	Multicenter retrospective analysis was performed on heterogeneous patient population.	Positive likelihood ratio of ¹⁸F-FDG PET for AD diagnosis was 3.5, and negative likelihood ratio was 0.08. Positive likelihood ratio of ¹⁸F-FDG PET for presence of neurodegenerative disease of any kind was 4.2, and negative likelihood ratio was 0.075.