Background and Purpose—Early recurrent ischemic lesions (ERILs) on diffusion-weighted imaging after acute ischemic stroke have been suggested as a potential marker of early recurrent stroke. We hypothesized that biomarkers of inflammation or coagulation may be associated with the pathogenesis of ERILs and sought to investigate whether these biomarkers provide prognostic information on the risk of development of ERILs independently of clinical and imaging variables.

Methods—This prospective study enrolled 153 consecutive patients with acute ischemic stroke who underwent diffusion-weighted imaging within 24 hours and subsequently at 5 days after onset and whose plasma or serum for biomarkers (C-reactive protein, fibrinogen, D-dimer, tissue plasminogen activator, and plasminogen activator inhibitor-1) were collected within 24 hours of onset. Those receiving thrombolysis or interventional therapy were excluded. ERILs were defined as new ischemic lesions on 5-day diffusion-weighted imaging separate from the index stroke lesions, which were not accompanied by subsequent recanalization.

Results—ERILs were observed in 37 patients (24.2%). In univariate analysis, shorter time from onset to initial MRI (P=0.013), initial acute multiple infarcts (P<0.001), initial larger infarct volume (P=0.005), stroke subtype (P<0.001), elevated D-dimer (P=0.028), and anticoagulation after admission (P=0.001) were associated with ERILs. In multivariate analysis, initial acute multiple infarcts (OR, 16.60; 95% CI, 5.73 to 48.08), large artery atherosclerosis (OR, 4.62; 95% CI, 1.51 to 14.11), and log D-dimer (OR, 3.20; 95% CI, 1.14 to 9.00) remained independent predictors of ERILs.

Conclusion—These data suggest that elevated D-dimer level reflecting increase of thrombin generation and fibrin turnover may be an independent factor predicting ERILs.