Authors (including presenting author) :
So MH (1), Ng J (1)(2), Tang CW (1), Au-Yeung WS (1), Cheng KK (1)
Affiliation :
(1) Department of Diagnostic and Interventional Radiology , Kwong Wah Hospital, (2) Department of Radiology, Queen Elizabeth Hospital
Introduction :
Mechanical thrombectomy for large vessel occlusion ischaemic stroke has become the mainstream treatment. Post-intervention single energy CT (SECT) brain is performed to exclude haemorrhagic transformation and guide treatment with antiplatelet or anticoagulants, particularly in patients with tandem stenosis and carotid stenting. A common phenomenon with reported incidence of 30-60% is post-interventional parenchymal contrast staining which appears hyperdense like acute blood on CT, posing a clinical dilemma when hyperdense areas in the territory of arterial occlusion are seen. Contrast staining can take up to 12-48 hours to wash out.
When there is suspicion of haemorrhage in the post-intervention CT brain, an early follow-up CT is arranged in 6 hours at our centre to exclude an enlarging haematoma. Another follow up CT the next day is also performed to monitor the size of the hyperdensity and progression of infarct. With the advent of dual energy CT (DECT) and post-processing algorithms, contrast staining and haemorrhage can be reliably differentiated.
Objectives :
1. Exclude haemorrhage in post-intervention CT brain to guide treatment.
2. Reduce follow-up CT to minimize radiation exposure to the patient and the discomfort of transfer.
Methodology :
All post-intervention CT brain studies were performed on single source DECT (Definition AS+, Siemens Healthcare) at 140kV and 80kV. Images were post-processed by using dedicated software (Syngo Dual Energy Brain Hemorrhage, Siemens Healthcare). Virtual unenhanced images and iodine overlay images were created. All studies were interpreted by neuroradiologists. CTDIvol of the DECT scans were also recorded to compare with SECT scans performed on the same CT machine.
Result & Outcome :
A total of 9 patients underwent mechanical thrombectomy with DECT. 6 of them showed hyperdense areas corresponding to sites of vascular occlusion in the post-intervention CT and all of them were determined to be parenchymal staining after post-processing. Out of these 6, only 2 had follow-up CT in 6 hours, thus 4 CT scans were avoided. The CTDIvol of DECT was also significantly lower compared to the SECT (38.5Gy vs 63.9Gy).
