Microhemorrhages in Traumatic Brain Injury May be Missed if Imaged too Late

This week's featured research article continues the focus on cerebral microbleeds (CMBs). Since they are a relatively new medical finding, much information is still unknown about their characteristics over time. What is known is that susceptibility weighted imaging (SWI) is a more sensitive MRI sequence than the traditional gradient recalled echo (GRE) T2* sequence at detecting and characterizing bleeds. While the former is used often in the research setting, it's usage in the clinical setting has not gained much traction. Further, the raw magnitude and phase data from the SWI sequence can be post-processed to create a Quantitative Susceptibility Map (QSM, or Susceptibility Weighted Imaging Map (SWIM)). A QSM represents underlying susceptibility of the brain tissue. It is more accurate than the phase information, because it is an intrinsic tissue property which is independent of the data acquisition parameters, and devoid of phase artifacts. With a QSM or SWIM image, one can detect and quantify CMBs along with their location and volume. Magnetic susceptibility from blood products in lesions yields bright signal in QSM. The image can also discriminate between paramagnetic regions like iron-laden and diamagnetic regions like calcium, which are easily confused which true bleeds due to their similar hypo-intense appearance in SWI.

The following study out of Radiology longitudinally evaluates cerebral microhemorrhages in a cohort of US military service members who have chronic TBI using SWI. It has been estimated that 10-20% of military personnel who have been deployed in Iraq and Afghanistan have experienced TBI, with most cases (80%) being in the mild state. Even mild TBI cases have shown subtle venous damage and microbleeding. 

Imaging Cerebral Microhemorrhages in Military Service Members with Chronic Traumatic Brain Injury. Wei Liu, et al. Radiology , 2015.

Within the study, 603 service members were recruited who have had TBI in severe (1.5%), moderate (5.8%), and mild (92.7%) cases.

MRI data were collected on a 3T GE scanner with a protocol which included GRE, and flow-compensated multiecho  GRE imaging. Typically, SWI can be run from a Siemens scanner, however, in this case the data were constructed using SPIN Software (MR Innovations, Detroit, MI).  The QSM images were processed using morphology-enables dipole inversion, or MEDI approach.  Some of the patients also underwent follow-up scans at 6-month, 12-month, and 36-month periods after the injury date.  

Two radiologists reviewed the GRE, SWI, and QSM data for microbleeds and found 43 of the patients (7.1%) showed at least one CMB. The SWI data showed higher sensitivity at detecting bleeds compared to the GRE data (585 hemorrhages detected vs. 362) and generally appeared on SWI with more conspicuousness. This finding confirms work by Cheng et. al (Stroke 2013).

Figure 1: A, GRE, B, SWI, and, C, QSM images in a patient with TBI. The arrows indicate a CMH that is visible on SWI, GRE, and QSM images. The arrowheads indicate another CMH that is visible on SWI and QSM images, but not the GRE image. ppb = parts per billion.

The majority of the bleeds could be found in the frontal subcortical region (30%), parietal subcortical region (15%), and the temporal subcortical region (14%). 

A, SWI and, B, QSM images show the evolution of microhemorrhages (arrows) in a patient who underwent follow-up imaging. Each column from left-right represents 6-months after injury, 12-months after injury, and 36-months after injury,

Plot of the mean magnetic susceptibility of microhemorrhages in individual patients who underwent follow-up imaging.  The results show that mean susceptibility of  CMBs decreases over time. 

The longitudinal assessment also showed that bleed counts, volume, and mean susceptibility decreased over time, which for the first time shows the evolution of CMBs in the chronic stage. Volumes decreased at a rate of -0.85 mm³ ± 1.59 per day (P = .039) and -0.10 parts per billion per day ± 0.14 for mean magnetic susceptibility (P =.016).

The prevailing thought is that the brain is incapable of removing hemosiderin iron once it is deposited in a lesion, however, this new research shows lesion changes which implies that hemosiderin blood products undergo continued subtle changes over time. 

The authors concluded that:

For the identification of the presence or absence of cerebral microbleeds, SWI should be the method of choice in a clinical setting. It is critical to image a TBI patient as close to the injury date as possible, otherwise subtle damage may me missed.

The diagnostic sensitivity of SWI married with the quantitative information of QSM shows great power in tracking the progress of TBI in the chronic state.  

For a full link to the text: Click Here.

MR Innovations is on the cutting edge of quantitative MRI post-processing software and analysis including SWI data construction, and Quantitative Susceptibility Mapping, please visit www.mrinnovations.com.

References:

1.   Liu W, Soderlund K, Senseney JS, et al. Imaging Cerebral Microhemorrhages in Military Service Members with Chronic Traumatic Brain Injury. Radiology 2015:150160.