Title: Which factors are associated with the natural rate of haematoma resolution after clinical ICH? A pooled analysis of medical patients from the MISTIE and CLEAR trials 

Authors: Lilli Nelson1,2, Calvin Heal 3, W Andrew Mould 4, Wendy Ziai 4, Daniel F Hanley 4, Adrian R Parry-Jones1,2, MISTIE & CLEAR Trial Investigators
1 Salford Royal NHS Foundation Trust, Manchester Academic Health Science Centre, Stott Lane, Salford, M6 8HD, UK 
2 Division of Cardiovascular Sciences, School of Medical Sciences, Faculty of Biology, Medicine and Health, University of Manchester, Manchester Academic Health Science Centre, Manchester M13 9PL, UK
3 Centre for Biostatistics, School of Health Sciences, Faculty of Biology, Medicine and Health, University of Manchester, Manchester Academic Health Science Centre, Manchester M13 9PL, UK
4 Division of Brain Injury Outcomes, Johns Hopkins Medical Institutions, Baltimore, MD, USA

Introduction:
Experimental evidence suggests that the immune system facilitates haematoma resolution after intracerebral haemorrhage (ICH) and prompt clearance may improve outcomes by limiting injury to the perihaematomal brain. We sought to determine factors associated with the rate of haematoma resolution in clinical ICH, including markers of the peripheral and central inflammatory response.

Methods:
Patients randomised to standard care in MISTIE II, MISTIE III and CLEAR III were included in this retrospective analysis. Perihaematomal oedema (PHO) at day 3 and haematoma volume on all CT scans from the diagnosis to day 30 were measured by threshold-based, semi-automated segmentation. Haematoma surface area at baseline was calculated by the sum of [region-of-interest perimeter x slice thickness] for each haematoma. To remove early haematoma expansion, volumes prior to day 3 were excluded. A linear regression line was then fitted to volume cube roots to determine each patient’s resolution rate.  A multifactorial linear regression model was then used to determine factors associated with resolution.

Results:
Of 416 patients, we excluded 61 due to missing data and 89 without haematoma volumes after day 20 post-onset, leaving 266 in the final analysis. The mean rate of clearance was 0.89 cm3/day (SD: 0.58 cm3/day). Surface area:volume ratio was associated with haematoma resolution rate (p=0.014) but no other factors were significantly associated, including age, baseline white blood cell count, platelet count, baseline haematoma volume and PHO volume.

Conclusion: 
Our study demonstrates that only the shape of the haematoma (as described by surface area:volume ratio) significantly influences resolution rate.

WHICH FACTORS ARE ASSOCIATED WITH THE NATURAL RATE OF HAEMATOMA RESOLUTION AFTER ICH? A POOLED ANALYSIS OF MEDICAL PATIENTS FROM THE MISTIE AND CLEAR TRIALS