Is 'Last Seen Normal' a Safe Decision Point?

May 2010

By Mitchel L. Zoler
Elsevier Global Medical News

SAN ANTONIO -- The safety of intravenous thrombolytic therapy in patients with unwitnessed strokes surpassed the safety in patients with witnessed strokes when physicians based treatment decisions on the last time patients were seen in their usual state of health, a review of more than 1,500 stroke patients in Ontario has shown.

To view a video interview with Dr. Sahlas, go to

The finding makes sense, because "last time seen normal" is a conservative estimate of stroke onset, Dr. Demetrios J. Sahlas said at the International Stroke Conference.

"Patients with unwitnessed symptom onset were less likely to undergo hemorrhagic transformation following treatment with intravenous TPA [tissue plasminogen activator]" than patients with witnessed symptom onset, reported Dr. Sahlas, professor of stroke management at McMaster University in Hamilton, Ont.

The report marks the first time researchers have formally examined the safety of intravenous TPA in patients having unwitnessed strokes.

Although the treatment is routinely used on patients with unwitnessed strokes, its safety was never before established, because all of the thrombo-lytic treatment trials exclusively enrolled patients with witnessed strokes so that their time from stroke onset to treatment could be closely controlled.

"No one looked at [patients with unwitnessed strokes] before," Dr. Sahlas said in an interview.

The study results also documented the frequency of unwitnessed stroke among treated patients.

Among 1,562 patients who had an out-of-hospital, first-time ischemic stroke in Ontario during July 2003-March 2008 and received intravenous TPA within 5 hours of either their witnessed symptom onset or their last time seen normal, 480 patients (31%) had unwitnessed strokes.

The unwitnessed subgroup matched the witnessed patients very closely for demographic and clinical para-meters reviewed, including stroke severity, length of stay, mortality, and status at hospital discharge, said Dr. Sahlas at the meeting, which was sponsored by the American Heart Association.

The 1,562 patients underwent treatment at any 1 of the 11 designated stroke centers in Ontario.

Their average age was 73 years, and they split nearly equally between women and men.

During the period reviewed, patients received intravenous TPA if their treatment could be started within 3 hours of either their witnessed symptom onset or their last time seen normal.

In May 2009, an advisory from the American Heart Association and American Stroke Association endorsed administering intravenous TPA to patients within 4.5 hours of their symptom onset or from the last time seen normal (Stroke 2009;40:2945-8) following publication in September 2008 of results from the European Cooperative Acute Stroke Study (ECASS)-3 showing the safety and efficacy of this approach (N. Engl. J. Med. 2008;359:1317-29).

Patients with unwitnessed strokes had an 8% rate of hemorrhagic transformation following TPA treatment, compared with a 12% rate among patents with a witnessed stroke.

In an analysis that adjusted for any baseline differences, the patients with unwitnessed strokes were 38% less likely to have a cerebral hemorrhage than those with a witnessed stroke, a statistically significant difference.

Symptomatic intracerebral hemorrhages occurred in 4% of the unwitnessed patients and 7% of those witnessed, a 39% relative benefit in favor of the unwitnessed patients in the adjusted analysis,and a difference that just missed statistical significance.

Dr. Sahlas said he and his associates had no disclosures.

Click here to
send us feedback