Cranial dural arteriovenous fistula: diagnosis and classification with time-resolved MR angiography at 3T.

BACKGROUND AND PURPOSE The diagnosis of dural arteriovenous fistula (DAVF) remains one of the few uncontested indications for catheter based cerebral angiography. We report our experience of using a commercially available form of time-resolved MR angiography (trMRA) at 3T for the diagnosis and classification of a cranial DAVF compared with the reference standard of digital subtraction angiography (DSA). MATERIALS AND METHODS A retrospective review of our patient records identified patients who had undergone trMRA at 3T and DSA for the evaluation of DAVF. The trMRA consisted of whole-head, contrast-enhanced "time-resolved imaging of contrast kinetics" (TRICKS) MRA. Image sets were independently reviewed by 3 readers for the presence, location, and classification of a DAVF. The reported result of the DSA was used as the gold standard against which the performance of the trMRA was measured. RESULTS Forty patients were identified who had undergone DSA and trMRA for evaluation of DAVF, yielding a total of 42 cases. On DSA, the results of 7 cases were normal, 15 cases were performed for surveillance of a previously cured fistula, and a new fistula (14) or persistent (6) fistula was found in 20 cases. Of these 20 fistulas, on DSA, 13 were Borden I, 2 were Borden II, and 5 were Borden III. In 93% (39/42) of DAVF cases, the 3 readers were unanimous and correct in their independent interpretation of the trMRA, correctly identifying (or excluding) all fistulas and accurately classifying them when encountered. CONCLUSIONS In this small series, trMRA at 3T seems be a reliable technique in the screening and surveillance of DAVF in specific clinical situations.