Neurologist Experience Affects Stroke Thrombolysis Outcome

Susan Jeffrey

June 23, 2015

BERLIN, Germany — A new study suggests thrombolysis is most beneficial in the hands of the most experienced neurologists, but outcomes are within the range seen in clinical trials even for those who are less experienced.

Researchers report that patients treated by less experienced neurologists had similar odds of having a modified Rankin Scale (mRS) score of 0 to 1 or 0 to 2 (indicating no or mild disability at 3 months) as patients enrolled in the major thrombolytic randomized trials. This finding "suggest[s] a benefit for these patients," said lead author Amélie Tuffal, MD, from the Department of Neurology, University of Lille, France.

They also had similar rates of symptomatic intracranial hemorrhage (sICH) and death, as seen in these trials of recombinant tissue plasminogen activator (rt-PA), she noted, suggesting a good safety profile.

However, the outcome of patients treated by more experienced neurologists was slightly better, with a lower risk for dependency at 3 months, she noted.

Dr Amélie Tuffal

These findings nevertheless argue against limiting the use of thrombolytics to the most experienced neurologists, Dr Tuffal said, "in order to increase the number of physicians who can give rt-PA."

Dr Tuffal presented the results here at the first Congress of the European Academy of Neurology (EAN). The paper was published recently in the Journal of Neurology.

24/7 Basis

Thrombolysis with rt-PA should be available on a 24/7 basis in hospitals treating patients with stroke, the authors note. This retrospective, single-center study assessed the effect of neurologist experience on patient outcome after thrombolysis.

The researchers analyzed outcomes for patients consecutively treated with intravenous rt-PA for cerebral ischemia at the Lille University Hospital over a 10-year period between September 2003 and March 2014. Neurologist experience was defined as the number of previous rt-PA treatments they had administered as the neurologist on duty at the hospital.

The researchers then performed logistic regression analyses to see what influence, if any, that level of experience had on the patient's outcome, assessed by using the mRS after 3 months; scores on the mRS of 0 to 1 indicate independence and scores of 0 to 2 indicate the absence of disability. They also recorded incidence of sICH and death at 3 months.

Forty-five neurologists treated 800 patients during the 10 years of the study. Of these, three were vascular neurologists who had experience with more than 50 cases. Less experienced vascular neurologists treated fewer cases, as did neurologists from other subspecialties who were treating stroke patients only on off-hours.

"Experienced" neurologists were defined as having administered at least 35 previous treatments, and "inexperienced" neurologists as having treated no more than four cases. The 800 patients were divided into quartiles based on neurologist experience.

Results showed "high" experience neurologists had treated 63 cases (range, 47 to 87), "median high" neurologists had treated 21 cases (range, 17 to 27), "median low" neurologists had treated 8 cases (range, 6 to 11), and those with "low" experience had treated only 2 cases (range, 1 to 3).

Baseline characteristics of the patients stratified by neurologist experience showed experienced neurologists were more likely to treat patients who were older, had an unknown time of stroke onset (probably due to more patients with "wake up" strokes arriving at hospital during working hours), and those who already had some stroke comorbidity before the current stroke, Dr Tuffal noted. They were more likely to use bridging therapy after the tPA treatment.

In unadjusted analyses, the number of patients who fell into mRS categories of 0 to 1 and 0 to 2 did not significantly differ between those in the quartiles of neurologists with the most and least experience (P = .123 for mRS score of 0 to 1 and P = .143 for mRS score of 0 to 2), However, numerically more patients reached these outcomes in the most experienced vs the least experienced neurologist group (approximately 50% vs 40% for mRS score of 0 to 1 and 60% vs 50% for mRS score of 0 to 2).

The results achieved by the least experienced neurologists were nevertheless better than those among patients receiving placebo in the randomized trials of rt-PA and were similar to those among patients receiving tPA in those trials. There was no difference among the four neurologist groups on sICH (P = .286) or death (P = .682).

In adjusted analyses, Dr Tuffal said, they confirmed the previous finding, "that the experience of the neurologist is an independent predictor of absence of handicap or dependency at 3 months, but it does not influence symptomatic intracerebral hemorrhage or death at 3 months."

Table. Adjusted Odds of Independence or Absence of Disability at 3 Months (for Every Additional 10 Cases Treated by the Neurologist)

Outcome Adjusted Odds Ratio (95% Confidence Interval)
mRS score 0 - 1 1.062 (1.008 - 1.120)
mRS score 0 - 2 1.076 (1.016 - 1.140)


"The message is therefore that to increase the number of centers able to give rt-PA, one should not be afraid to involve neurologists with a low level of experience in thrombolysis because they provide a benefit," senior author Didier Leys, MD, also from the University of Lille, told Medscape Medical News.

"Of course, if we have the choice, being treated by an experienced neurologist is better," he said. "However, being treated immediately by a less experienced neurologist is probably better than being treated 1 half-hour later by an experienced one."

The experience is from a single center, however, and requires confirmation in other datasets, Dr Tuffal said. Although they should not be excluded from treating with thrombolysis, the authors recommend that the practice of less experienced neurologists be monitored and educational programs be instituted.

"These results are valid in a center with a continuous audit of all cases and strategies to improve quality," Dr Leys noted, but this may not be the case in other centers.

Session moderator Michael Brainin, MD, professor of clinical neurology at the Danube University in Krems, Austria, asked whether time to treatment differed between the groups. Dr Tuffal noted that treatment times were similar.

The results prompt the question of why these differences were found, Dr Brainin added, and suggest that if faced with the need for tPA treatment, "experience is better."

Finally, he congratulated Dr Tuffal on the study. "This is a very interesting paper because rarely do we question ourselves and test our own capabilities," he noted. "Usually we do that for others, but not for ourselves."

The authors have disclosed no relevant financial relationships.

Congress of the European Academy of Neurology (EAN). Abstract O2105. Presented June 21, 2015.

J Neurol. 2015;262:1209-1215. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.