Background: Intracerebral hemorrhage (ICH) is a major public health problem. This subset of stroke often coexists with other serious medical problems such as hypertension, diabetes, and obesity. Management of hemorrhagic stroke is controversial and there is no standardized system for assessing presentation and predicting outcome of this disease. We propose a new grading system based on clinical and radiologic factors important in influencing outcome in ICH that can be used by the entire health care team.
Methods: We conducted a retrospective study of the last 50 patients who presented with hypertensive ICHs to Louisiana State University Health Sciences Center in Shreveport during 2001 to 2003. Significant predictors of outcome at 6 months as measured by the Glasgow outcome score (GOS) were determined and a grading system based on clot volume, hydrocephalus on initial computed tomographic scan, and focal neurologic deficit was formulated.
Results: Three factors observed to have significant association with GOS were presence of a focal neurologic deficit on initial presentation (P = .003), presence of hydrocephalus on initial computed tomographic scan (P < .0001), and clot volume (P = .003). Patients were scored on these variables as follows: neurologically intact (0 point), any focal neurological deficit (1 point); absence of hydrocephalus (0 point), presence of hydrocephalus (1 point); and clot volume less than 20 mL (1 point), 20 to 50 mL (2 points), and greater than 50 mL (3 points) [corrected] The scores were summed to assign an ICH grade to each patient for predicting his GOS at 6 months. Given the nonsignificant difference between a patient's grade and his actual observed GOS (mean difference 0.04, P = .79), as well as their significant correlation (correlation coefficient = 0.76, P < .0001), we believe our grading system is useful for predicting a patient's GOS.
Conclusion: An accurate and reliable grading scale for ICH is helpful in standardizing the management of ICH, improving communication of patient presentation among health care workers, and predicting outcomes.