Prediction of motor recovery using initial impairment and fMRI 48 h poststroke

Eric Zarahn, Leeor Alon, Sophia L. Ryan, Ronald M. Lazar, Magnus Sebastian Vry, Cornelius Weiller, Randolph S. Marshall, John W. Krakauer

Research output: Contribution to journalArticlepeer-review

78 Scopus citations


There is substantial interpatient variation in recovery from upper limb impairment after stroke in patients with severe initial impairment. Defining recovery as a change in the upper limb Fugl-Meyer score (ΔFM), we predicted ΔFM with its conditional expectation (i.e., posterior mean) given upper limb Fugl-Meyer initial impairment (FMii) and a putative functional magnetic resonance imaging (fMRI) recovery measure. Patients with first time, ischemic stroke were imaged at 2.5 ± 2.2 days poststroke with 1.5-T fMRI during a hand closure task alternating with rest (fundamental frequency = 0.025 Hz, scan duration = 172 s). Confirming a previous finding, we observed that the prediction of ΔFM by FMii alone is good in patients with nonsevere initial hemiparesis but is not good in patients with severe initial hemiparesis (96% and 16% of the total sum of squares of ΔFM explained, respectively). In patients with severe initial hemiparesis, prediction of ΔFM by the combination of FMii and the putative fMRI recovery measure nonsignificantly increased predictive explanation from 16% to 47% of the total sum of squares of ΔFM explained. The implications of this preliminary negative result are discussed.

Original languageEnglish (US)
Pages (from-to)2712-2721
Number of pages10
JournalCerebral Cortex
Issue number12
StatePublished - Dec 2011


  • fMRI
  • prediction
  • recovery
  • stroke

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Cellular and Molecular Neuroscience


Dive into the research topics of 'Prediction of motor recovery using initial impairment and fMRI 48 h poststroke'. Together they form a unique fingerprint.

Cite this