Page 598

Karger_ESC London_2013

22. European Stroke Conference Cerebrovascular Diseases
Jour-nal 598 © 2013 S. Karger AG, Basel Scientific Programme 30 Intracerebral/subarachnoid haemorrhage and venous diseases Validation of ICH and ICH-GS scores in Indian patients with Intracerebral Haemorrhage R. Bhatia1, V. Sreenivas2, S. Singh3, V. Rai4, M.V. Padma5, K. Prasad6 All India Institute of Medical Sciences, New Delhi, INDIA1, All India Institute of Medical Scienc-es, New Delhi, INDIA2, All India Institute of Medical Sciences, New Delhi, INDIA3, All India Insti-tute of Medical Sciences, New Delhi, INDIA4, All India Institute of Medical Sciences, New Delhi, INDIA5, All India Institute of Medical Sciences, New Delhi, INDIA6 Background Prognostic scores have been devised to predict mortality in Intracerebral hemorrhage (ICH). We aimed at validating the ICH score and ICH grading scale (ICH–GS) for in-hospital mortality in our ICH population without any DNR orders. Methods ICH and ICHGS scores were calculated for each patient as defined. The discriminatory ability of scores was determined using ROC curve analysis. Best threshold values were determined, and sen-sitivity, specificity, PPV and NPV along with 95% CI were calculated. Mortality with individual scores observed was compared with the expected using Chi-Square test. Analysis was performed for the entire group, patients treated medically and medical+surgical intervention. Results Among the 214 patients with ICH, 70(32.7%) died during the hospital stay. 95 (44.4%) patients underwent neurosurgical intervention. The ICH scores showed increasing mortality with score increase: 0 (6.25%),1(0%), 2(25%),3(36.2%), 4(80.2%).The AUC (area under curve) was 76.2% (95%CI: 69.8 - 82.5%). ICH score 3 had a sensitivity of 71.4% (59.4% - 81.6%), specificity 65.3% (56.9% - 73%), PPV 50% (39.8% - 60.1%) and NPV of 82.5% (74.2% - 88.9).The ICH-GS scores also showed increasing mortality with score increase: 5 (0%), 6 (11.1%),7(10.34%),8(23.7%),9(2 5.9%),10 (40.5%),11(82.4%),12 (90.0%) and13(100%).The AUC was 75.5% (68.5 - 82.9%) and was significantly lower than previously reported (p < 0.05). ICH-GS score 10 had a sensitivity of 58.6% (46.2% - 70.2%), specificity 79.9% (72.4% - 86.1%), PPV 58.6% (46.2% - 70.2%) and NPV of 79.9% (72.4% - 86%). For both scores, there was a significant difference in the mortality propor-tions with each score value in our data compared to the expected (p < 0.001). Similar results were observed for the medically and surgically treated groups. Conclusions The discriminatory ability of the scores in our cohort though good, is lower than reported previously and the mortality observed with each individual score is significantly different. No difference was observed among patients treated medically or surgically in performance of the scores. 29 Intracerebral/subarachnoid haemorrhage and venous diseases Association of intracranial pressure and cerebral perfusion pressure with outcome in comatose patients with intracerebral hemorrhage M. Sykora1, S. Steinmacher2, T. Steiner3, S. Poli4, J. Diedler5 Dept. of Neurology, University Heidelberg, Heidelberg, GERMANY1, Dept. of Neurology, Uni-versity Heidelberg, Heidelberg, GERMANY2, Dept. of Neurology, University Heidelberg, Heidel-berg, GERMANY3, Dept. of Neurology, University Heidelberg, Heidelberg, GERMANY4, Dept. of Neurology, University Heidelberg, Heidelberg, GERMANY5 Background – Therapeutic targets for intracranial pressure (ICP) and cerebral perfusion pressure (CPP) in patients with severe intracerebral hemorrhage (ICH) are approximated from data of trau-matic brain injury. However, specific data for ICH are lacking. Here, we aimed to identify or con-firm critical thresholds of ICP and CPP with respect to functional outcome following severe ICH. Methods – We analyzed 143 consecutive ICH patients in whom ICP monitoring was applied. Out-come at 3 months was assessed using the modified Rankin Score and dichotomized into acceptable (mRS 0-4) versus poor outcome (mRS 5-6). Frequencies of ICP and CPP values relative to the total monitoring time were calculated and categorized into ranges. Multivariate logistic regression was used in order to identify the best prediction model fit by testing different ICP and CPP thresholds. Results: Mean monitoring duration was 177 (SD 67.5) hours. In the multivariate model with the best predictive power, age (OR 0.96, CI 0.93-0.99, p=0.009), Graeb score (OR 0.83, CI 0.73-0.95, p=0.005), hemorrhage volume (OR 0.99, CI 0.97-0.99, p=0.03), relative frequency of ICP values < 15 mmHg (OR 1.05, CI 1.02-1.09, p=0.04) and relative frequency of CPP values >60 mmHg (OR 1.1, CI 1.03-1.14, p=0.003) were related to acceptable outcome at 3 months. Conclusions: Our data confirm CPP and ICP thresholds in ICH patients previously found in other pathologies. In the context of other predictors as age, hemorrhage volume and intraventricular hem-orrhage, thresholds ICP<15 mmHg and CPP > 60 mmHg seemed to best discriminate between ac-ceptable and poor outcome after ICH. Abbreviation: Cerebro-vasc 1015-9770 1421-9786 karger.com/


Karger_ESC London_2013
To see the actual publication please follow the link above