Misdiagnosis of stroke in tissue plasminogen activator-treated patients: characteristics and outcomes.

STUDY OBJECTIVE: Misdiagnosis of acute ischemic stroke is a risk inherent in treating patients with acute deficits, yet few data exist on the problem. We report an evaluation of emergency department misdiagnoses in patients treated with tissue plasminogen activator for acute ischemic stroke.

METHODS: We conducted an observational study of 151 consecutive patients treated without an acute stroke team. Initial diagnosis was compared with interim and hospital discharge diagnoses. Separate analyses were performed for patients without a final diagnosis of acute ischemic stroke and for those without a final diagnosis of either acute ischemic stroke or transient ischemic attack, combined.

RESULTS: Ten of 151 patients (7%; 95% confidence interval [CI] 3% to 12%) had final diagnoses that did not include acute ischemic stroke. Six of 151 (4%; 95% CI 1% to 8%) had a final diagnosis other than acute ischemic stroke or transient ischemic attack (conversion disorder [4], complex migraine [1], and Todd's paralysis [1]). These "stroke mimics" had no intracranial hemorrhage (0%; 95% CI 0% to 31%), had less disability at discharge (modified Rankin Scale score mean+/-SD, 2.2+/-1.3 versus 3.2+/-1.8), and were younger (mean age+/-SD, 47+/-21 years versus 68+/-15 years) than patients with acute ischemic stroke or transient ischemic attack. An additional 4 (3%) patients had interim diagnoses other than acute ischemic stroke, all subsequently changed to acute ischemic stroke after magnetic resonance imaging or computed tomography.

CONCLUSION: These data show that in a 4-hospital system without an acute stroke team, thrombolytic treatment of patients with diagnoses mimicking stroke was infrequent, and hemorrhagic complications did not occur in any patients without an acute ischemic stroke. However, because the number of mimics was small, safety cannot be ensured with statistical confidence.