González RG1.
Key points:
- Explain the approach to imaging the patient with acute stroke used at the Massachusetts General Hospital (MGH)
It is useful to categorize patients into hemorrhagic and ischemic stroke and the latter into those with mild, moderate, and severe symptoms.
NIHSS | % Stroke population | Treatment | Key factors | Imaging |
<5 | 55 | None | None | None |
5-10 | 10 | IV tPA | Time, hemorrhage | CT |
>10 | 35 | IV tPA | Time. hemorrhage | CT |
IA | – | – | Time, hemorrhage, core, Penumbra | CT+MR or MR alone |
Most patients with a severe stroke syndrome (NIHSS, >10) have a major anterior circulation occlusion, most commonly of a middle cerebral artery. Yoo AJ et al., in a study of 107 patients who underwent IAT at the MGH for anterior circulation occlusions, demonstrated that nearly half of the patients with final infarct volumes of ≤60 mL had good outcomes (no one with a final infarct volume of >120 mL had good outcome).

Identification of infarct core is mandatory. If the core volume threshold is 70 mL, imaging must provide a measurement that is accurate within 10 to 20 mL.
Diffusion MRI
- The best method for the early detection of the infarct core (The high contrast/noise ratio of DWI makes it accurate);
- DWI abnormalities sometimes reverse (but it is rare);
- A DWI abnormality volume of >70 mL is highly specific for a poor outcome;
- The use of early infarct core identification for triage decisions is supported by the observations that the final infarct volume is the single best predictor of good outcome at 90 days.
Noncontrast CT
- Reliable for detecting hemorrhage;
- Unreliable for detecting the early infarct core;
- Highly specific for infarction when a hypodensity is clearly visible.
CT Perfusion
- It has less sensitivity and specificity than MR Perfusion (it has low signal/noise and contrast/noise ratios (CNR) and produces noisy images with high measurement error);
- The CNR of infarct cores on CTP-derived CBF images is very low (<1).
It is needed a less precision to identify the Ischemic Penumbra. All that is needed is the assessment of clinical importance of the penumbra with clinical/imaging correlation.
Clinical/DWI mismatch could be used (instead of PWI/DWI mismatch) in patients with large vessel occlusion, a DWI lesion and an NIHSS >10
A patient with a proximal MCA and distal ICA occlusion and a DWI lesion of ≤70 mL always had a diffusion/perfusion mismatch of ≥100%.
Time is brain, but each patient has his own time.
- Many patients have a small cores and large penumbras even many hours after ictus;
- Sorensen et al. and Neumann-Haefelin et al. found significant diffusion/perfusion mismatches after10 to 24 hours after stroke onset;
- Ribo et al. reported that 43 of 56 patients with stroke (77%) presenting 3 to 6 hours after stroke onset had a DWI/perfusion-weighted imaging mismatch of ≥50%;
- Copen et al., in a study of 109 patients, reported that more than 50% had a DWI/mean transit time mismatch volume of ≥160%.
Reference