Abstract 2813: Acute Tele-Stroke-Service by Stroke Neurologists: Reliability of Clinically Relevant CT-Findings
Background: The Stroke East Saxony Network (SOS-NET) provides tele-consultations for acute stroke patients presenting to local hospitals in eastern Saxony, Germany. Experienced stroke neurologists perform tele-consultations 24/7. They take a clinical history, perform a video examination and a standardized evaluation of acute cerebral CT including Alberta Stroke Program Early CT Score (ASPECTS) before suggesting further management. We assessed the diagnostic accuracy and clinical impact of CT evaluation by stroke neurologists in this acute tele-stroke-service.
Methods: To assess the diagnostic accuracy of stroke neurologists in CT-interpretation, two experienced neuroradiologists re-evaluated all CT scans of tele-consultations in 2009 blinded to clinical findings. We defined discrepant findings as deviations in the ASPECTS >1 or if differential diagnoses like tumor or intracranial hemorrhage (ICH) were not detected by the stroke neurologists. To assess the clinical impact, all discrepant results were subsequently discussed and re-evaluated of all involved neurologists and neuroradiologists regarding its possible influence on therapeutic decisions and on patients’ outcome.
Results: In 2009, we performed 583 tele-consultations (353 ischemic strokes, 119 primary ICH and 111 stroke mimics). In 102 patients with ischemic stroke (29%) thrombolysis was performed with eight bleeding complications (7.8%). The neuroradiologists detected discrepant findings in 44 CT scans (7.5%), however corrected, their image interpretation if unblinded to the clinical information in four patients. We regarded the diagnostic inaccuracy by neurologist as clinically relevant in nine patients (1.5%). In one patient, the stroke neurologist missed a small chronic subdural hematoma. The patient did not receive thrombolysis because of mild clinical findings. Consequently, the false diagnosis did not affect treatment and clinical outcome in this patient. Furthermore, stroke neurologists missed extensive early ischemic changes in the middle cerebral artery territory in eight patients and recommended thrombolytic therapy. We observed symptomatic ICH in five of these patients (62.5%; 0.9% of the total cohort).
Conclusion: Trained stroke neurologists had a high diagnostic accuracy in CT interpretation in acute tele-stroke-service, compared to experienced neuroradiologists. Clinically relevant misinterpretations of the CT scans were rare (1.5%). However, regarding the potential risk of thrombolytics, stroke neurologists should continuously be trained in interpreting CT scans of acute stroke patients.
- © 2012 by American Heart Association, Inc.