Early detection of cerebral arterial occlusion on magnetic resonance angiography: predictive value of the baseline NIHSS score and impact on neurological outcome.

The presence of a thrombus on initial arteriography is directly related to the baseline NIHSS score. Magnetic resonance angiography (MRA) offers a noninvasive and rapid assessment of large cerebral vessel patency. We aimed at evaluating (1) the baseline NIHSS score as a tool for predicting the likelihood of an occluded artery on MRA and (2) the course of stroke within the first week according to the presence of a cerebral arterial occlusion. Patients were enrolled in this prospective study according to the following criteria: (1) acute cerebral ischemia with a neurological deficit lasting >1 h, and (2) brain MRI performed within 24 h of stroke onset. The NIHSS score assessment was performed on admission and at day 1 and day 7. The MRI protocol included: (1) T2-weighted Turbo spin echo, (2) echo-planar imaging isotropic diffusion, (3) T2*-gradient echo sequence, and (4) time of flight MRA (3D TOF Turbo MRA). The presence of a symptomatic cerebral arterial occlusion on MRA was systematically screened. Fifty-four patients were studied. Median age was 60 years. Mean time from stroke onset to NIHSS assessment was 170 +/- 95 min. The mean baseline NIHSS score was 13.5 +/- 7.3. The mean time from stroke onset to MRI was 384 +/- 171 min. MRA was readable in 50 cases. An arterial occlusion was detected in 23 patients (46%). The median baseline NIHSS score was significantly higher in the group of patients with occlusion than in the group of patients without occlusion (18 vs. 7, p = 0.01). The predictive probability to demonstrate an arterial occlusion was related to the baseline NIHSS score. None of the patients with an NIHSS score of 1-6 (11 patients) had visible occlusion, whereas 9 (43%) out of 21 patients with an NIHSS score of 7-15 and 14 (78% ) out of 18 patients with an NIHSS score above 16 had an arterial occlusion. For an increase by one point in the NIHSS score, the odds ratio for the presence of occlusion was 1.28 (95% CI: 1.11-1.46). The course of the stroke as assessed by follow-up NIHSS score was significantly more severe if an occlusion was detected. Median day 0, day 1 and day 7 NIHSS score were, respectively, 18, 16 and 13 in patients who had an occlusion versus 7, 4 and 0 in patients who had no visible occlusion (p < 0.01). A direct relation between the baseline NIHSS score and the likelihood of the presence of an occlusion on initial MRA is demonstrated. The presence of a cerebral arterial occlusion on MRA is significantly linked to a poor neurological outcome. Copyright 2002 S. Karger AG, Basel