| Summary: | Spontaneous intracerebral haemorrhage (ICH) is a form of stroke which often leads to severe disability and death. ICH-related poor outcomes are significantly due to haematoma expansion and perihaematomal oedema (PHO). The ability to identify patientswho are most likely to experience haematoma expansion and PHO and benefit from treatments targeting expansion or oedema remains an unmet need. In this thesis data from the 'Tranexamic Acid in Intracerebral Haemorrhage -2' (TICH-2) was used to· analyse the relationship of the baseline imaging characteristics to haematoma expansion and PHO expansion in patients with spontaneous ICH. Methods: Intra-and inter-observer reliability of haematoma and PHO volume measurement methods were first established using 20 samples randomly selected from 'Efficacy of Nitric Oxide in Stroke Trial' (ENOS). This confirmed that both techniques of haematoma volume segmentation, ABCj2 formula and semi-automated segmentation (SAS), and manual delineation of PHO volume were reliable methods for measuring both haematoma volumes and PHO volumes with high intra- and inter observer reliability. In the TICH-2 trial data, haematoma expansion was defined as an absolute increase in haematoma volume of more than 6 mL or relative increases of more than 33% in follow-up scan compared to the baseline haematoma volume. PHO expansion was defined as relative increase of more than 20% in PHO volume from baseline to follow-up. Follow-up scan was done at 24 hours after baseline scan. Participants were classified as favourable clinical outcome for mRs <3 and poor clinical outcome for mRs ~3.
|
|---|