Page 801

Karger_ESC London_2013

London, United Kingdom 2013 Poster Session Blue Cerebrovasc Dis 2013; 35 (suppl 3)1-854 801 859 Intracerebral/subarachnoid haemorrhage and venous diseases Assessing and improving the Intracerebral Haemorrhage (ICH) score’s predictive value for 30 day mortality C.J. Sinclair-Wilson1 St. George’s University of London, London, UNITED KINGDOM1 Background The ICH score is a 6 point score made up of 5 variables and is designed to predict mor-tality at 30 days. The aim of this study was to assess the validity of the ICH scoring system at pre-dicting 30-day mortality in a recent sample of patients with ICH, as well as to identify whether the inclusion of high blood pressure at presentation improved the score’s accuracy. Methods Data was gathered from the St. George’s Hospital stroke registry which contains details of 3,692 strokes from October 2003 to July 2010 of which 314 patients had ICH. Of these, 217 patients were repatriated to local hospitals shortly after admission, leaving 97 patients with complete follow up data for inclu-sion in the study. The data collected included: patient demographics, risk factors, size and location of haemorrhage on initial brain imaging and mortality rates. The predictive value of the ICH score and the modified score were analysed using ROC curves in SPSS version 16.0, as was statistical analysis of patient characteristics. Meanwhile tabulation was done using Microsoft Excel 1997. Re-sults From the sample, 28% died within 30 days. Of their presenting features systolic blood pressure (SBP) ≥200mmHg showed the most promise as an additional variable to improve the current ICH score. The risk of mortality within 30 days did not increase in an entirely linear fashion and had a reduction at ICH score 4. With SBP ≥200mmHg included as a variable in the ICH score the linearity improved. The area under the ROC curve for the original ICH score was 0.81 while the addition of SBP ≥200mmHg in the modified score increased the area to 0.83. Conclusion - The risk of mortality within 30 days increases with ICH score. - The current ICH score is highly predictive of mortality within 30 days. - The incorporation of a systolic blood pressure ≥200mmHg at presentation into the score increases its accuracy but only marginally (+2%). 860 Intracerebral/subarachnoid haemorrhage and venous diseases Intensive BP monitoring in acute intracranial hemorrhage improves high BP detection and HIC prognosis prediction M. Rubiera1, D. Rodriguez-Luna2, M. Ribo3, J. Pagola4, A. Flores5, M.A. Muchada6, P. Meler7, E. San Juan8, C.A. Molina9 Hospital Vall d´Hebron, Barcelona, SPAIN1, Hospital Vall d’Hebron, Barcelona, SPAIN2, Hos-pital Vall d´Hebron, Barcelona, SPAIN3, Hospital Vall d’Hebron, Barcelona, SPAIN4, Hospital Vall d’Hebron, Barcelona, SPAIN5, Hospital Vall d’Hebron, Barcelona, SPAIN6, Hospital Vall d´Hebron, Barcelona, SPAIN7, Hospital Vall d’Hebron, Barcelona, SPAIN8, Hospital Vall d’Hebron, Barcelona, SPAIN9 High blood pressure(BP) is related to hematoma growth(HG), early clinical deterioration(ECD) and poor outcome in acute intracranial hemorrhage(ICH). High BP detection is crucial for adjusting an-tihypertensive treatment and may influence outcomes. We aim to evaluate the efficacy of high BP detection at different time-points as recommended by guidelines(BPg) compared with continuous BP monitoring(BP-M). METHODS: ICH patients <6h from symptoms onset were studied. BP evaluation and treatment ac-cording to BPg were as follows: every 15min <1sth, every 1h < 6h and every 6h < 24h. Targets were Systolic(SBP)<180 and Mean(MBP)<130. Simultaneously, automatic non-invasive BP-M every 15 min during 24h was performed. BP-M recordings were blindly analyzed later and did not influence management. Patients received a 24h control CT. Serial NIHSS <24h and mRS at 3mo were per-formed to evaluate clinical outcome. RESULTS: 89 patients were included, 58.8% males, mean age 71.5. Mean time from symptoms: 118 min. Median NIHSS: 17(IQR 8). Mean baseline ICH vol: 28.7. BP-M identified a significantly(p<0.05) higher number of patients with high SBP and MBP readings than BPg, especially between 6-24 hours from symptoms onset(Figure). According to BPg, 32(40.5%) patients received antihypertensive drugs. Among them, good SBP control(defined by less than 3 readings>180) was achieved in 37.1% according to BPg, but only in 14.3% on BP-M. Good MBP control(<3 readings >130) was observed in 45.7% patients according to BPg, and only in 17.1% on BP-M. BP-M was more accurate than BPg on predicting ICH outcomes. Patients with high SBP readings on the first 6h on BP-M had higher rate of ECD(p=0.016). High SBP in 24h on BP-M was associated with HG(p=0.024) and poor outcome(p=0.032). BPg showed similar trends without statistical signif-icance. CONCLUSIONS: Intensive BP monitoring improves detection of ICH patients with poor BP con-trol, especially beyond 6h, and increases the accuracy of HIC prognosis prediction.


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