Intracerebral Hemorrhage (ICH) Score

Why did you develop the ICH Score? Was there a clinical experience that inspired you to create this tool for clinicians?

The ICH Score was developed as a clinical grading scale and communication tool. As a young neurointensivist, I noticed that we used clinical grading scales for communication across providers regarding traumatic brain injury (GCS), ischemic stroke (NIHSS), subarachnoid hemorrhage (Hunt-Hess and WFNS), and AVM (Spetzler-Martin), but we did not have a standard way to communicate about ICH. Thus, I hoped for something informative yet simple and easy to determine. It was never intended to be used as a formal prognostic outcome prediction model.

What pearls, pitfalls and/or tips do you have for users of the ICH Score? Are there cases in which it has been applied, interpreted, or used inappropriately?

I have been overall surprised and disappointed at how many people have used the point estimates from the original ICH Score publication to presumably predict outcome and to communicate that to patients and their families. Ironically, in the first draft of the manuscript, I did not even include these numbers, just an overall graph. But one of the reviewers demanded they be put in and, as a young investigator wanting to get published, I complied. It has been extremely disappointing when I hear that physicians have chosen to not treat a patient aggressively or transfer to a higher level of care hospital because of a high ICH Score. I actually recall a conversation at the International Stroke Conference around 2003, when an ED physician in a community hospital thanked me for developing the ICH Score, because now he had a reason to avoid accepting transfers from smaller community hospitals for patients with ICH Scores of 4 or higher because they would always do poorly. This saddened me. Ironically this concern led me and others to the study of the self-fulfilling prophecy of poor outcome in ICH and other neurocritical care conditions if early DNR or withdrawal of support is undertaken. And this has hopefully overall increased the emphasis on aggressive care for ICH and other conditions such as TBI, SAH, and cardiac arrest.

What recommendations do you have for health care providers once they have applied the ICH Score? Are there any adjustments or updates you would make to the score given recent changes in medicine?

Utilize the ICH Score and other clinical grading scales for their intended use. Patient stratification and communication among providers and with patients and surrogates. But do not hang your hat on the point estimates. I certainly hope that the editors and users of MDCalc are not doing this. We do not assign a specific mortality risk to a GCS of 6 in TBI or a Hunt-Hess score of 3. Don't do it for the ICH Score either. Numerous studies have shown two things regarding the ICH Score: it is valid as an overall scale (meaning a 4 is worse than a 3 is worse than a 2, etc.) and the point estimates are confounded by early care limitations. And if you just can't help yourself, then use confidence intervals. But overall avoid the incorrect temptation to use any mathematical developed on a population of patients as a way to precisely predict outcome for the purposes of clinical decision-making in an individual patient I remain surprised at how many investigators continue on this false quest in ICH, SAH, and other conditions.

As the ICH Score has become more widely adopted, do you anticipate further uses of the score?

The American Heart Association is emphasizing the use of a baseline severity score as part of the initial evaluation of ICH patients. And the Joint Commission requires this for comprehensive stroke centers. This is overall a good thing for standardization. The ICH Score is one of these scales and is probably the most widely used and validated. So use it, but as intended.

What are the main differences you see between the ICH score and FUNC score? In your mind, how can the two be used together to help clinicians?

The FUNC score is focused on functional independence and, I think, is overall intended as more of a prediction tool. I would leave the issue of point estimate validity to the FUNC score authors. But I think they can be used in conjunction to provide a communication tool for providers and give an overall sense of ICH severity.

Other comments? Any new research or papers on this topic in the pipeline?

Take a look at the study we published last year that addresses some of the concerns above. Morgenstern, L. B., et al. (2015). Full medical support for intracerebral hemorrhage. Neurology 84(17): 1739-1744. And as always, use the ICH Guidelines as a framework for aggressive care of the ICH patient. Hemphill, J. C., 3rd, et al. (2015). Guidelines for the Management of Spontaneous Intracerebral Hemorrhage: A Guideline for Healthcare Professionals From the American Heart Association/American Stroke Association. Stroke 46(7): 2032-2060.

It continues to humble me to see how, even without a “magic bullet,” there remains an art to being a good physician and aggressive care does matter. This is a very empowering message that should encourage us to identify the aspects of guideline-concordant care that improve patient outcomes.

J. Claude Hemphill, III, MD, MAS, is a professor of medicine and clinician in the Department of Neurology at the University of California, San Francisco. He is also the director of the Neurocritical Care Program at San Francisco General Hospital Medical Center. Dr. Hemphill cares for patients with stroke and traumatic brain injury and has a special research interest in intracerebral hemorrhage.