Statistical contribution to prognostic factors in spontaneous intracerebral hematomas. Of v-cramer coefficient and discriminative analysis

Spontaneous intracerebral hematomas (ICH) present a major problem with a high mortality rate and significant functional impairment in survivors. Surgical treatment of spontaneous ICH is still a matter of controversy. The indications for surgical intervention and evacuation of ICH are still undetermined. More data is therefore required concerning the natural history of ICH and factors predicting its outcome. We had tried to identify those factors predicting outcome and further define surgical indications. In the present study, we have identified factors available on the day of admission that are associated with outcome. A total of 182 patients with spontaneous ICH had been analysed in relation to the acute findings. Out of this group 142 patients were operated upon and 40 underwent conservative treatment. The results one month after treatment were analysed by the use of Glasgow Outcome Scale (GOS). In the surgically treated group good recovery (GOS I and II) was noted in 53.6% of the cases and mortality was 39.4%. In the conservatively treated group good outcome was obtained in 81.5% and mortality was 15%. In the whole group good outcome after one month was obtained in 57.9% and mortality was 34.1%. A statistical analysis of the relationship between the effect of the treatment and the variables considered in the present research demonstrates that the factors of the greatest importance in forseeing the results of the treatment applied are: GCS, diameter and volume of haematoma, age, blood pressure and clinical condition (H-H), especially with regard to patients treated surgically. A comparative analysis of the results of surgical and conservative treatment in different subgroups (localization, etiology etc.) proves that better results were obtained through the conservative treatment in the whole group. That conclusion ought to be approached with caution in the light of the fact that the groups were not entirely homogeneous. The methods in question facilitate taking decision regarding the management of ICH and seem noteworthy also in appraising great numbers of patients with different clinical entities and lesions.