Improvement of the Diagnostic Accuracy of MRA with Subtraction Technique in Cerebral Vasospasm

ABSTRACT PURPOSE Vasospasm has been considered the most severe acute complication after subarachnoid hemorrhage (SAH). MRA is not considered ideal for detecting cerebral vasospasm because of background including the hemorrhage. The aim of this study is to evaluate the efficacy of Subtraction MRA (SMRA) by comparing it to that of conventional MRA (CMRA) for diagnosis of cerebral vasospasm. METHODS Arteries were assigned to one of three categories based on the degree of MRA diagnostic quality of vasospasm (quality score): 0, bad … 2, good. Furthermore each artery was assigned to one of four categories based on the degree of vasospasm severity (SV score): 0, no vasospasm … 3, severe. The value of the difference between DSA‐SV score and MRA‐SV score was defined as the DIF score. CMRA and SMRA were compared for each arterial region with regard to quality score and DIF score. RESULTS The average CMRA and SMRA quality score were 1.46 and 1.79; the difference was statistically significant. The average CMRA and SMRA DIF score were 1.08 and .60; the difference was statistically significant. CONCLUSIONS Diagnosis of cerebral vasospasm is more accurate by SMRA than by CMRA. The advantages are its noninvasive nature and its ability to detect cerebral vasospasm.