One of the major trends in patient safety in a variety of venues has been a push toward early mobilization of patients. Early mobilization has become part of numerous guidelines for a variety of medical conditions (for example see our December 2010 What’s New in the Patient Safety World column “The ABCDE Bundle”).
Early mobilization in acute stroke patients theoretically makes a lot of sense and several studies have demonstrated positive outcomes from early mobilization in stroke patients. Patients with acute stroke and significant limb weakness are obviously at risk for venous thromboembolism and pressure ulcers among other complications. And the “deconditioning” that occurs with bedrest is likely detrimental to a number of physiologic systems. But the great unknown is how much activity is best and when should it be started. The logical assumption is that the earlier it is started, the better the outcomes are likely to be.
AVERT (A Very Early Rehabilitation Trial for Stroke) was undertaken to answer those questions. AVERT was a randomized (single blinded) controlled trial involving stroke patients (including both ischemic stroke and intracerebral hemorrhages) in 56 acute stroke units in 5 countries (Bernhardt 2015). The very early intervention group began therapy and mobilization within the first 24 hours after stroke (median time to first mobilization 18.5 hours), almost 5 hours on average earlier than those patients in the “usual care” group.
Much to the surprise of the investigators, the primary outcome of the study (a good functional outcome based on a modified Rankin Scale score of 0-2 at three months) was actually worse in the group receiving very early mobilization. And there was no evidence of accelerated walking recovery.
Interestingly, there was no difference in complications between the very early intervention group and the usual care group in AVERT. That includes no difference in the incidence of venous thromboembolism. There was also no statistically significant difference in mortality between the two groups.
Subgroup analysis did not reveal a better outcome in any subgroups in the very early intervention group. But there was some evidence that patients with more severe ischemic strokes and intracerebral hemorrhages might be especially prone to worse outcomes with the very early intervention. The author posit a number of physiologically plausible possible explanations for the worse outcomes in the very early intervention period.
Overall, the outcomes for all patients in the study were relatively good. Complication rates were low in both groups. And the “usual care” group actually was getting mobilization quite early (median time to first mobilization was 22 hours in the “usual care” group).
The investigators plan to further analyze the data to determine whether any dose-response relationship exists.
Will the study results change the approach to early mobilization in stroke patients? Unfortunately, it may be years before more definitively is known about the optimal time to begin and intensity of such mobilization. But the AVERT results would certainly suggest that very early, intensive mobilization may be detrimental in stroke patients.
The study is a reminder that sometimes things that make a lot of sense may not always translate into clinical benefit and may, in fact, have unexpected consequences. The investigators should be commended for undertaking this important study.
One very useful outcome of AVERT is that it demonstrates that randomized controlled trials (RCT’s) can be used to evaluate rehabilitation interventions in a manner similar to how we evaluate medication, surgical or procedural interventions. However, it did take the investigators almost 8 years to accumulate enough patients to have a study powered adequately enough to address the questions asked.
Bernhardt J, Langhorne P, Lindley RI, et al. for the AVERT Trial Collaboration group. Efficacy and safety of very early mobilisation within 24 h of stroke onset (AVERT): a randomised controlled trial. Lancet 2015; Published Online April 17, 2015
Print “PDF version”