Fourteen studies (1,134 participants, range 5 to 343) were included in the review. Withdrawals were not explained in 30% of included test accuracy studies. Approximately 73% of studies did not report whether the reference standard was interpreted blind to index test results. The description of the index test was unclear in approximately 15% of studies. Other QUADAS criteria were met by all studies.
Seven studies (926 participants) evaluated D-dimer accuracy in patients with suspected cerebral vein thrombosis. The pooled estimate of sensitivity from these studies was 93.9% (95% CI 87.5 to 97.1) and the pooled estimate of specificity was 89.7% (95% CI 86.5 to 92.2).
Potential risk factors for false negative D-dimer results included isolated headache (two out of three studies), longer duration of symptoms (two out of four studies), and limited sinus involvement (combined results of three studies).
A sensitivity estimate derived from all study participants with objectively confirmed cerebral vein thrombosis was also reported.