Feasibility and utility of arterial spin labelling perfusion in neuro-oncology MRI protocols in the UK

Abstract

Arterial spin labelling, (ASL) is an MRI perfusion technique that uses endogenous contrast and is currently underused in neuro-oncology MRI protocols in the UK. The purpose of this study was to examine the feasibility, ease of implementation and added value of arterial spin labelling (ASL) perfusion imaging in neuro-oncology protocols. Radiographers at a large tertiary MRI centre received in-house training in protocol selection and performing pseudo-continuous ASL (pcASL). During a 10 month period, 104 consecutive patients with gliomas, undergoing diagnostic MRI at 3.0 T, underwent an extended MRI protocol that included pcASL. ASL perfusion weighted images (ASL-PWI) were automatically generated by the scanner and archived together with conventional imaging. A qualitive retrospective survey was performed for feedback on the process. A sub-group of patients with available follow up imaging (n=39) were selected for further analysis (n=39: 20 glioblastoma multiforme, 15 astrocytoma, 4 oligodendroglioma). Two observers examined images and concluded RANO scores in two parts: conventional images alone and conventional imaging with ASL-PWI. Confidence was measured using Likert scales 1–5 (1,very unconfident to 5, very confident) and differences compared. Lastly, all diagnoses were checked against follow up imaging. 90% of radiographers felt either confident or very confident in performing ASL technique. 100% of reviewed PWIs were considered diagnostic quality. A statistically significant increase in average confidence levels was observed for inclusion of ASL (4.44 (SD 0.8)) compared to conventional imaging alone (4.02 (SD 0.65)). ASL is a reliable, non-contrast, low cost and well tolerated technique that can be easily implemented. ASL-PWI can be produced at time of scanning which can be utilised to build confidence in assessing response to treatment in low and high grade gliomas.