Pina Christine Sanelli MD, Presenter: Nothing to Disclose

Carl E. Johnson MD, Abstract Co-Author: Nothing to Disclose

Linda Ann Heier MD, Abstract Co-Author: Nothing to Disclose

Joseph Peter Comunale MD, Abstract Co-Author: Nothing to Disclose

Apostolos J. Tsiouris MD, Abstract Co-Author: Research Consultant, BioClinica, Inc

Robert D. Zimmerman MD, Abstract Co-Author: Nothing to Disclose

Philip Stieg, Abstract Co-Author: Nothing to Disclose

Alvin Mushlin MD, Abstract Co-Author: Nothing to Disclose

To prospectively evaluate the diagnostic accuracy of CT perfusion (CTP) for vasospasm and determine a quantitative threshold in aneurysmal subarachnoid hemorrhage (A-SAH) patients.

A-SAH patients were prospectively enrolled and monitored according to clinical and imaging exams. IRB approval, HIPAA compliance and informed consent were obtained. CTP was typically performed between days 6-8 following aneurysm rupture. Quantitative cerebral blood flow(CBF), cerebral blood volume(CBV), and mean transit time(MTT) values were obtained using standard region-of-interest placement sampling the gray matter. Reference standard consisted of both imaging and clinical criteria for determination of vasospasm. Additional analysis was performed using digital subtraction angiography as a single reference standard. Ninety-seven patients were included in the data analysis for the study population and 57 patients in the subgroup analysis. Receiver operating characteristic(ROC) curves were analyzed to determine diagnostic accuracy using the area under the curve. Optimal threshold values were determined by several methods, including the northwest corner, predetermined 95% sensitivity or specificity, and maximizing patient population utility.

There were statistically significant differences (p<0.05) in the CBF, CBV and MTT values for the vasospasm and no vasospasm patients. ROC curves reveal CBF with the highest overall diagnostic accuracy (93%), followed by MTT (88%) and CBV (72%). Using different methods, optimal CBF threshold values were calculated between 27-37mL/100gm/min with varying sensitivity and specificity values. Overall, analysis of the subgroup using DSA as the reference standard demonstrates almost identical results as the study population.

CBF emerged as the CTP parameter with the greatest discrimination ability for vasospasm diagnosis. Choosing an optimal threshold for CBF depends on each physician’s values in managing A-SAH patients and minimizing errors. To improve the average state of health for this patient population, a CBF threshold of 35mL/100gm/min (90% sensitivity,68% specificity) is recommended based on the population utility method.

This study provides a guide for physicians to optimally utilize CTP data in clinical practice by selecting important threshold values that will direct the management and treatment of A-SAH patients.

Sanelli, P, Johnson, C, Heier, L, Comunale, J, Tsiouris, A, Zimmerman, R, Stieg, P, Mushlin, A, Diagnostic Accuracy of CT Perfusion for Vasospasm in Aneurysmal Subarachnoid Hemorrhage: A Prospective Study.  Radiological Society of North America 2010 Scientific Assembly and Annual Meeting, November 28 - December 3, 2010 ,Chicago IL. http://archive.rsna.org/2010/9005280.html