Angiography Suite Cone-Beam Computed Tomography Perfusion Imaging in Large-Vessel Occlusion Patients Using RAPID Software: A Pilot Study

Multidetector computed tomography (CT) perfusion (MDCTP) volume estimation using RAPID software (iSchemaView, Menlo Park, CA) has been validated in clinical trials for endovascular thrombectomy (EVT) selection.^1,2 The availability of cone-beam CT perfusion (CBCTP) in the angiosuite might shorten workflow times in late-window patients transferred from primary centers without CTP capabilities. We evaluate whether quantitative analysis of MDCTP and CBCTP using RAPID software would lead to comparable ischemic core and hypoperfused tissue volumes in patients undergoing EVT.

Methods

Anonymized data are available upon reasonable request. This single-center study was approved by our Institutional Review Board and all participants/proxies signed a written informed consent form. We conducted a prospective, single-arm interventional study including patients with anterior circulation large-vessel occlusion stroke eligible for EVT. MDCTP and CBCTP datasets were processed using RAPID software (Data Supplement).

Ischemic core was defined as relative cerebral blood flow of <30% on MDCTP and <45% on CBCTP of the corresponding contralateral territory. The hypoperfused tissue was evaluated using time-to-maximum >6.0 s and time-to-maximum >10.0 s thresholds. We compared MDCTP and CBCTP volumes using intraclass correlation coefficient, Bland-Altman agreement plots, and Pearson and Spearman correlations. Finally, we compared CBCTP ischemic core and the final infarct volume in reperfused patients.

Results

Thirteen patients were included in the final analysis (Figure I and Table I in the Data Supplement). We found a strong correlation between MDCTP and CBCTP for the ischemic core (Pearson=0.91, Spearman=0.87) and time-to-maximum >6.0 s (Pearson=0.90, Spearman=0.85). Bland-Altman analysis showed 92% agreement for both perfusion parameters (Figure). We observed an intraclass correlation coefficient of 0.89 (95% CI, 0.67–0.96) for the ischemic core and 0.86 (95% CI, 0.55–0.96) for time-to-maximum >6.0 s (Table II in the Data Supplement).

• Download figure
• Download PowerPoint

Comparing CBCTP ischemic core and the final infarct volume, we found a strong correlation (Pearson=0.87, Spearman=0.87) and 90% of agreement. We also observed an intraclass correlation coefficient of 0.81 (95% CI, 0.48–0.94).

Discussion

This study demonstrates that automated analysis of ischemic core and hypoperfused tissue volumes derived from MDCTP and CBCTP in patients with stroke eligible for EVT provides comparable results.

In our study, the contrast dose received in CBCTP after MDCTP was relatively low, and the median time between studies was short, which could reduce perfusion variations between modalities. Potential infarct growth between modalities was considered and 91% (10/11) of cases had concordant results using the DEFUSE 3 trial (Endovascular Therapy Following Imaging Evaluation for Ischemic Stroke) eligibility criteria for EVT.²

An automated real-time perfusion volumetric processing in the angiosuite would allow bypassing the emergency department for transferred large-vessel occlusion late-window patients. Thus, this technology might reduce reperfusion times and improve patients’ functional outcomes. In addition, CBCTP images could provide a more accurate assessment of perfusion parameters during and immediately after EVT. Downsides of transferring patients from primary centers to angiosuite might include EVT ineligibility due to large core infarct, low stroke severity, distal occlusion, among others.³

The study is limited by the small sample size. Additional limitations are included in the Data Supplement.

Supplemental Materials

Expanded Methods, Results, and Discussion

Online Figure I

Online Tables I and II

Footnotes