Abstract Background and aims Observational data suggest that mannitol could paradoxically increase haematoma expansion after acute intracerebral haemorrhage (ICH). We assessed these effects using data from the Mannitol for cerebral oedema after intracerebral haemorrhage trial (MACE-ICH). Methods 46 ICH participants with cerebral oedema or at risk presenting within 72 hours of ictus were randomised to 10% mannitol repeated at 24 hours (n=15), 10% single dose mannitol (n=15) or standard care (n=16). CT scans were assessed by adjudicators blinded to clinical and treatment allocation. ICH volume was segmented using a machine-based deep learning method and expansion was defined as increase between baseline scan and day 5 follow-up CT of 6ml or 33%. We analysed whether mannitol was associated with ICH expansion. Results The mean age of participants was 74 (s.d.12) years, NIHSS was 12.5 (10.6) and the median time to randomisation was 22.9 hours. Most haematomas were lobar in location (58.7%), and the mean ICH volume was: 27.1 (22.3) ml in the group randomised to two doses of mannitol; 32.4 (23.3) ml in the single dose mannitol group and 27.8 (25.6) ml in the standard care group. At day 5, ICH expansion occurred in 7 participants and there was no significant difference between the three groups Odds Ratio OR 0.74; 95% CI (0.22, 2.56); p=0.64. In adjusted analysis, there was no difference in ICH volume at day 5 between the groups mean difference MD -2.01 (-4.96, 0.94); p=0.18. Conclusions In this trial of moderate size ICH, Mannitol was not associated with ICH expansion. Conflict of interest
Krishnan et al. (Fri,) studied this question.