Overall, 30 studies (n=4,597 participants who received cerebrospinal fluid testing) providing 51 comparisons were included in the review.
Alzheimer’s disease versus healthy or neurological control groups (19 studies, 1,329 patients with probable Alzheimer’s disease, 971 control participants): The pooled estimate of sensitivity was 77.6% (95% CI 70.6 to 83.9) and the pooled estimate of specificity was 87.9% (95% CI 84.3 to 91.1). The I2 test indicated moderate to high heterogeneity. There was no significant difference in sensitivity or specificity between epitopes of cerebrospinal fluid phosphorylated tau. For clinical utility, the positive utility index was 0.70 (good) and negative utility index was 0.64 (good).
Alzheimer’s disease versus other types of dementia (18 studies, 1,304 patients with Alzheimer’s disease, 588 control participants): The pooled estimate of sensitivity was 71.6% (95% CI 63.1 to 79.5) and the pooled estimate of specificity was 77.8% (95% CI 72.3 to 82.7). The I2 test indicated moderate to high heterogeneity. Subgroup analyses of cerebrospinal fluid phosphorylated tau epitopes indicated that p181 may be significantly less sensitive than either p199 or p231, and that p231 may be significantly less specific than either p199 or p181 (both p=0.01), but data were limited. For clinical utility, the positive utility index was 0.62 (satisfactory) and negative utility index 0.45 (poor).
Mild cognitive impairment versus healthy control groups (eight studies, 247 patients with mild cognitive impairment, 200 control participants): The pooled estimate of sensitivity was 79.6% (95% CI 64.2 to 91.5) and the pooled estimate of specificity was 83.9% (95% CI 73.1 to 92.4). The I2 test indicated moderate to high heterogeneity. There were insufficient data for subgroup analyses. For clinical utility, the positive utility index was 0.62 (satisfactory) and negative utility index 0.58 (satisfactory).
Progressive mild cognitive impairment versus stable mild cognitive impairment (six studies, 163 patients with progressive mild cognitive impairment, 225 patients with stable mild cognitive impairment): The pooled estimate of sensitivity was 81.1% (95% CI 69.2 to 90.7) and the pooled estimate of specificity was 65.3% (95% CI 49.6 to 79.5). The I2 test indicated moderate heterogeneity. Data were not significantly different for phosphorylated tau epitope p181 alone. For clinical utility, the positive utility index was 0.51 (satisfactory) and negative utility index 0.54 (satisfactory).
There were no data on cerebrospinal fluid phosphorylated tau for the differential diagnosis of Alzheimer's disease versus mild cognitive impairment.