ISC: Maintaining life support in ICH patients may lead to more recoveries

The prognosis for intracerebral hemorrhage (ICH) patients may not be as dire as many treating physicians assume, and a significant percentage of such patients may recover an acceptable level of function over time, according to a study comparing ICH patients whose life support was withdrawn to similar patients who remained on life support. The findings were presented Feb. 7 at the American Stroke Association's International Stroke Conference 2013 in Honolulu.

David Tirschwell, MD, of the University of Washington Stroke Center in Seattle, and colleagues began with a cohort of 590 ICH patients at a single center in Seattle. The researchers created a propensity score to determine the likelihood that life support would be removed for a given patient, based on age, functional status before the ICH, level of consciousness, amount of bleeding, pre-existing hypertension, diabetes, atrial fibrillation, initial temperature and whether the patient was intubated.

The researchers found 78 pairs of propensity score-matched patients in which life support was withdrawn for one patient and the other remained on life support. The primary endpoint of the study was modified Rankin Scale (mRS) at hospital discharge. For the purposes of the study, the researchers assumed an “acceptable” outcome was an mRS score of 4 or less. They acknowledged that this definition included disability so severe as to render the patients in need of assistance for activities of daily living, and that many people may not consider such an outcome “acceptable.”   

The mRS scores in both groups ranged from 3 to 6. In the group from whom life support was withdrawn, 1.3 percent scored 3, 2.6 percent scored 4, 6.4 percent scored 5 and 90 percent scored 6 (death). In the group who were maintained on life support, 6.4 percent scored 3, 32 percent scored 4, 24 percent scored 5 and 37 percent scored 6. Thus, among patients from whom life support was withdrawn, 3.9 percent survived with an acceptable outcome, compared with 38.4 percent of the patients who were maintained on life support.

The researchers speculate that treating physicians may present a negative prognosis to families prematurely, with the result that up to a third of patients from whom life support is withdrawn may have been able to recover to an mRS of at least 4 had support been maintained.

Noting that recovery in these extremely ill patients progresses slowly, and that end-of-life decisions are often made in the first days of hospitalization, Tirschwell urged treating physicians and loved ones to allow some time to pass before considering withdrawal of life support. “Greater patience and less pessimism may be called for in making these life and death decisions,” he said in a release.