Testing the outcome score of spontaneous intracerebral haemorrhage in haemodialysis patients
Article first published online: 14 OCT 2009
© 2009 The Authors. Journal compilation © 2009 Royal Australasian College of Physicians
Internal Medicine Journal
Volume 39, Issue 10, pages 692–695, October 2009
How to Cite
Huang, B.-R., Lo, Y.-L., Chang, C.-H. and Chen, T.-Y. (2009), Testing the outcome score of spontaneous intracerebral haemorrhage in haemodialysis patients. Internal Medicine Journal, 39: 692–695. doi: 10.1111/j.1445-5994.2009.02015.x
Conflict of interest: None.
- Issue published online: 14 OCT 2009
- Article first published online: 14 OCT 2009
- Received 27 May 2008; accepted 17 March 2009.
- intracranial haemorrhage;
- end-stage renal disease;
The aim of this study is to determine the predictive value of the spontaneous intracerebral haemorrhage (ICH) outcome score (which we described previously) in haemodialysis (HD) patients who develop spontaneous ICH. The validation cohort consisted of all HD patients with spontaneous ICH presenting to Chang Gung Memorial Hospital in Taiwan during 2005–2007. The data were collected from one hospital and prospectively analysed, and the predictive model was tested using an external validation dataset. The prognostic factors were confirmed by chi-squared testing. We calculated the 30-day mortality in different groups of the validation cohort divided according to outcome score and tested the predictive value of the outcome score. The overall mortality rate was 52.6% in 38 HD patients. The originally identified prognostic factors were Glasgow Coma Scale score, age ≥70 years, systolic blood pressure <130 or ≥200 mmHg, ICH volume ≥30 mL, presence of intraventricular haemorrhage and serum glucose ≥8.8 mmol/L. All but one (serum glucose ≥8.8 mmol/L (P= 0.07)) were subsequently found to be associated with 30-day mortality. In patients scoring 6 and 7, mortality was 100%, but in patients scoring 0, it was only 11.1%. The 30-day mortality in the validation cohort increased steadily with score and had good predictive value (correlation coefficient = 0.986, P < 0.001). Similar results in two different cohorts indicate that the ICH outcome score is a reliable outcome measure.