3D view improves cerebral microbleed diagnosis

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon

Accurately identifying patients who have cerebral microbleeds is of clinical importance because their presence has been found to be related to a risk of hemorrhagic transformation after ischemic stroke and to recurrence of spontaneous intracerebral bleeding. A group of Dutch researchers have found that accelerated 3D T2-weighted gradient-recalled-echo (GRE) MR images depict more cerebral microbleeds than do conventional 2D T2-weighted GRE MR images.

A research team from the departments of radiology and epidemiology and biostatistics at Erasmus Medical Center in Rotterdam, the Netherlands, recently conducted a prospective study comparing 3D and 2D T2-weighted GRE MR exams for the depiction of cerebral microbleeds. The results of their research were published in this month’s Radiology.

Axial MR images obtained with conventional 2D T2-weighted GRE (top) (repetition time msec/echo time msec, 775/20; flip angle, 25°; section thickness, 5 mm) and accelerated 3D T2-weighted GRE (bottom) (45/31; flip angle, 13°; section thickness, 1.6 mm zero-padded to 0.8 mm) sequences in the same participant. More cerebral microbleeds (arrows) seen bilaterally in the thalamus region, are visible on the 3D T2-weighted GRE image than the 2D T2-weighted GRE image. Image and caption courtesy of the Radiological Society of North America.  

“Two-dimensional T2-weighted GRE sequences have previously been shown to better depict cerebral microbleeds than conventional spin-echo and fast spin-echo T2-weighted sequences,” the authors wrote.

The team conducted MR exams on 200 patients, with a mean age of 79.2 years, using a 1.5-Tesla MR system (GE Healthcare) and an 8-channel head coil. The cohort underwent both 2D and 3D imaging sequences using standard protocols at the institution.

The images from the studies were randomly allocated to one of two reviewers that were blinded to the other sequence and to all clinical information. The 2D images were interpreted first, in random order, and three weeks later, the 3D images were interpreted by the same reviewers. All studies with a potential microbleed were reviewed for confirmation by a neuroradiologist with 7 years experience interpreting neurologic MR images.

“Microbleeds were defined as focal areas of very low signal intensity that were smaller than 10 mm in size,” the researchers noted.

Cerebral microbleeds were detected on both the 2D and 3D sequences; however, the difference in prevalence of the microbleeds between the sequences was highly significant.

“By using a 3D T2-weighted GRE sequence, we found cerebral microbleeds in even more participants than by using a 2D T2-weighted GRE sequence,” according to the researchers.

They reported that there were no microbleeds visualized on the 2D images that were not detected on the 3D images. They noted that a median of 2.5 to 1.0 microbleeds were seen on 3D visualization compared with the 2D sequence.

“Cerebral microbleeds were detected in more persons by using accelerated 3D T2-weighted GRE images than by using conventional 2D T2-weighted GRE images,” they wrote. “Furthermore, in participants who had microbleeds detected on images from both sequences, we identified significantly more microbleeds on the 3D T2-weighted GRE images than on the 2D T2-weighted GRE images.”

Although patients presenting with stroke who have a small number of cerebral microbleeds can be safely treated by thrombolysis, the presence of more microbleeds may indicate a widespread vasculopathic condition warranting a different treatment approach.

“In clinical practice, accurate identification of those patients who have cerebral microbleeds will thus become increasingly important for clinical decision making,” the authors stated.