Strokes Too Mild for IV tPA May Still Have Severe Outcomes

Abstract & Commentary

By Alan Z. Segal, MD Assistant Professor, Department of Neurology, Weill-Cornell Medical College, Attending Neurologist, NewYork-Presbyterian Hospital. Dr. Segal is on the speaker's bureau of Boehringer Ingelheim.

Synopsis: A substantial minority of patients deemed too good for intravenous tPA were unable to be discharged home. A re-evaluation of the stroke severity criteria for tPA eligibility may be indicated.

Source: Smith EE et al. Poor Outcomes in Patients Who Do Not Receive Intravenous Tissue Plasminogen Activator Because of Mild or Improving Ischemic Stroke. Stroke. 2005;36:2497-2499.

Intravenous tPA treatment for acute ischemic stroke carries a significant risk of intracerebral hemorrhage. Patients who have a relatively small stroke with a benign natural history, may not justify this therapy. In the NINDS tPA study, patients were excluded if they had "rapidly improving or minor symptoms." This lead to a subcategory of acute stroke patients referred to as "too good to treat" or TGT. Many of these patients also showed "rapid improvement" or RI, hopefully indicative of a continued upward trajectory. Smith and colleagues, however, report that a substantial subset of TGT patients, especially those with RI, may still go on to poor outcomes after being excluded from tPA treatment.

Among 128 patients presented to the Massachusetts General Hospital within the 3-hour time window for acute stroke, 71 were deemed ineligible for treatment. Among these 71 patients, the majority (41) were excluded because they were TGT. Half of these 41 showed RI in the emergency room, with improvements on their National Institute of Health Stroke Scale (NIH-SS) from approximately 5 to 3. Patients who showed the greatest acute improvement in their NIH-SS (< 4 points) prior to the tPA decision were more likely to have a subsequent neurological deterioration (with a relative risk of 4.1, P = 0.05). These patients were more likely to show large vessel occlusions on vascular imaging, compared with those who had a more modest or absent RI. Overall, 27% of TGT patients (11/41) were determined to have an unfavorable outcome, either because they showed neurological worsening or had a persistent neurological deficit preventing them from being discharged directly to home.


One of my trainees once referred to the decision to give tPA for acute stroke as "the most risky thing we do in medicine, based on the least information." These data suggest, however, that a decision not to give tPA based on a perception that the patient will do well without it, is perhaps equally risky.

The issue of rapid improvement is a thorny one. Stroke patients are moving targets no matter what their underlying vascular pathology may be. Lacunar stroke patients may stutter, with apparent stability or even transient resolution proceeding towards much more dense defects hours later. Patients with large vessel strokes (as shown in Smith et al data) may fluctuate even more widely. Acute stroke therapy, unfortunately, does not give us time to make these observations. We must base our decision to treat or not to treat on a single assessment, move forward from there, and not function with the "retrospect-oscope."

There is also ambiguity in our terminology. A mild deficit is generally thought to refer to an NIH-SS < 4. A patient with an expressive aphasia and a facial droop, however, may only have an NIH-SS of 3, yet might have a potentially disabling infarct.

Given these uncertainties, whether we treat or not, it is crucial that we communicate effectively with patients (and even more so their families). Whenever possible, they should be included in the decisions that we make, and together, we will live with the consequences.