Apparent diffusion coefficient thresholds and diffusion lesion volume in acute stroke.

Area of Science:

• Neurology
• Radiology
• Medical Imaging

Background:

• Apparent diffusion coefficient (ADC) thresholds are utilized for acute stroke lesion volume determination.
• The reliability and comparability of ADC thresholds to magnetic resonance diffusion-weighted imaging (MR-DWI) hyperintense lesion volumes remain unclear.

Purpose of the Study:

• To evaluate the accuracy of commonly used ADC thresholds in determining acute ischemic stroke lesion volume.
• To compare ADC-derived lesion volumes with manually outlined MR-DWI hyperintense lesion volumes.

Main Methods:

• Prospective recruitment and clinical assessment of acute ischemic stroke patients.
• Acquisition of MR-DWI within 24 hours and 3-7 days post-stroke.
• Comparison of manually outlined DWI lesion volume (reference) with volumes derived from three ADC thresholds (.55, .65, .75 × 10⁻³ mm²/s), with and without manual editing.

Main Results:

• Significant discrepancies were observed between unedited/edited ADC volumes and the reference DWI lesion volume.
• Median unedited ADC volumes (e.g., 52,972 mm³ at .55 threshold) greatly overestimated the acute DWI lesion volume (15,284 mm³).
• Edited ADC volumes showed improvement but still required substantial manual correction, with some thresholds yielding unmeasurable volumes.

Conclusions:

• Threshold-derived ADC volumes necessitate significant manual editing to accurately reflect visible DWI lesions.
• Caution is advised when using unedited ADC thresholds for acute stroke lesion volume quantification due to over/underestimation risks.