MRI & Timing of Brain Injury 

Disclaimer: Before discussing some of the important concepts involved with using MRI to help time hypoxic-ischemic brain injury, it first bears stating that there is an art to this science, and that while some of the scientific principles involved are widely agreed upon, how they are used in and adapted to an individual case may very much be a matter of subjective interpretation. There are also various permutations of the principles discussed upon which experts may not agree. 

The timing of a brain insult is often one of the most important and contested aspects of a birth injury case. Absent some obvious sentinel event, it is usually impossible to identify the exact moment that injury began or was completed. However, strong arguments to support the timing of injury in these cases can be achieved by using a combination of the commonly referred to data points of EFM, cord gases, the neonatal exam, and imaging. Each has its uses and limitations, but the one that is the least susceptible to attack or controversy is imaging. This is due in part to the plethora of studies that have looked at changes in the brain in response to intrapartum or neonatal asphyxia, studies which, for the most part, are quite consistent with each other. Please understand, however, that while the science behind using imaging to help time injury is fairly well-established, it does not do away with the mischief and debate that can ensue when experts disagree about what a patient’s studies actually show. 

Currently the best imaging to assist in the timing of injury is MRI. Before discussing its use to help distinguish an injury at birth from one that occurred weeks earlier, one should not forget that it is also very helpful in establishing that HIE exists at all. Evaluation of MR imaging by a pediatric neuroradiologist is often a good first step in evaluating any potential birth injury case. It can help rule out infection, stroke, or congenital anomaly, all things which, in the right context, can alter or even lay waste to a claim (no matter how bad the tracings look). Moreover, in cases involving a neonatal injury, a normal MRI at one week of age can help establish causation for some post-birth event that occurs after that study. 

To address the issue of when a hypoxic/ischemic insult occurred, there are two components to magnetic resonance (“MR”) that are generally most helpful to refer to. They are diffusion weighted imaging, or DWI, which qualitatively identifies the presence or absence of normal movement of water within the brain, and apparent diffusion coefficient, or ADC, which can actually quantify the extent of any diffusion abnormality. The basic principle behind both is that water in the normal neonatal brain is known to move at a certain velocity and in a certain manner. An asphyxia mediated brain injury produces swelling and debris that interferes with that normal movement, causing a lessening or restriction of diffusion of water within the brain that is able to be seen on MR. 

While MR is not able to time an injury to within a matter of hours, depending on when it is performed it can create a timeframe for injury plus/minus a couple of days. In that sense, it is best used to “bracket” the timeframe for injury, with more specific timing being elucidated by combining MR with the clinical and lab information referenced above. To utilize MR in this fashion, one should begin with the general concept that diffusion abnormalities are generally seen within a 7-day window beginning on the day of injury (this applies to non-cooled babies- cooled babies have a different timeframe and will be addressed later). Corline Parmentier et al., Magnetic Resonance Imaging in (Near-) Term Infants with Hypoxic-Ischemic Encephalopathy, 12 DIAGNOSTICS 645, 655 (2022); Yvonne Wu et al., How Well Does Neonatal Neuroimaging Correlate with Neurodevelopmental Outcomes in Infants with Hypoxic-Ischemic Encephalopathy? 94 PEDIATRIC RESEARCH 1018, 1019 2023). 

As one gets further outside of that window, it becomes less and less likely that you will see those diffusion changes. Most authors believe that the peak for these changes occurs in the timeframe around days 3 or 4 post-injury. See Parmentier et al., supra at 650. So, for instance, if an MRI is performed on day 3 of life and the diffusion appears completely normal, it would suggest that whatever injury is ultimately seen on later imaging did not occur on the day of delivery. Conversely, if there is diffusion restriction on day 3 of life, it falls squarely within the expected timeframe for MR changes related to an injury at the time of birth. 

This discussion of DWI- and ADC-based timing is unfortunately not as simple as the general rule stated above. By way of one example, if you see restricted diffusion on an MRI performed on day two, how do you know that is a restriction that is on the way up to its peak at day 3 or 4, or instead on its way down after having reached a peak 3 days earlier (and therefore inconsistent with an intrapartum event)? For that matter, how do you know it’s even at its peak as you look at it? The answer to the latter question lies in part in the subjective analysis referenced above. Many experts will testify that by looking at the degree of brightness on DWI, they can tell you if it is at its peak or something short of that. The answer to the question of whether the restricted diffusion is on its way up or down has a more objective-based approach. You may have heard experts often speak of Ti- and T2-weighted imaging in addition to the DWI portion of the MR study. Using it as an adjunct to DWI and ADC, T2 identifies fluid abnormalities and will begin to appear dark in certain parts of the brain somewhere around 6 days post insult. Alexandra O’Kane et al., Early Versus Late Brain Magnetic Resonance Imaging after Neonatal Hypoxic Ischemic Encephalopathy Treated with Therapeutic Hypothermia, 232 J. PEDIATRICS 73, 75-83 (2021). Therefore, restricted diffusion that is accompanied by T2 changes can be argued to represent changes that are falling from peak levels but not yet outside the window for diffusion abnormalities. 

In the age of frequent total body cooling of babies suspected of having suffered an hypoxic-ischemic insult, it is important to note that most experts believe that such cooling changes the timeframes identified above, lengthening all of the relevant parameters by 1-3 days. This understanding has also often changed the timing of when clinicians will order MRI, with initial imaging perhaps not occurring until day 5-7 of life, as opposed to days 3 or 4. Jeanie Cheong et al., Prognostic Utility of Magnetic Resonance Imaging in Neonatal Hypoxic-Ischemic Encephalopathy: Substudy of a Randomized Trial, 166 ARCH PEDIATR. ADLOSEC. MED. 634, 638 (2012). 

Unfortunately, early MRI does not always take place with the neonate suspected of having had an injury. Take, for example, a case where the first MRI is not done until 14 days of life and shows changes compatible with an hypoxic-ischemic injury. The diffusion changes indicative of an H/I injury will no longer be present, replaced by the normal or even increased diffusion findings of the post-asphyxia event, often referred to as pseudo-normalization. See Wu et al., supra at 1019. Therefore, using diffusion abnormalities to help create the time window for injury becomes very problematic, if not impossible to do. Nevertheless, MR in this setting remains a valuable tool in the diagnosis of HIE, as later changes in the brain involving infarction and volume loss are often identified and can be correlated with the clinical information provided in the labor, delivery, and neonatal records to create a compelling argument for perinatal injury. 

Discussion of these later findings of HIE primarily involves looking for one of two injury patterns, as each may be very relevant to any particular case. It is well known that the typical pattern of injury in cases of partial prolonged asphyxia, like the 6 hours of bad EFM that goes unacted upon, is in the cerebral cortex, particularly the watershed (peripheral) areas of the brain. Linda Vries & Floris Groenendaal, Patterns of Neonatal Hypoxic-Ischaemic Brain Injury, 52 NEURORADIOLOGY 555, 556 (2010). Conversely, H/I brain injury from sudden, severe acute events such as cord prolapse or massive abruption will typically injure the deep brain structures, such as the basal ganglia and thalamus. Correlating these MR findings to the circumstances surrounding a particular labor and delivery can often add significant weight to the argument about when the insult occurred, even if the absence of timing information that timely diffusion imaging can afford.

Time, space, and knowledge base do not allow for a deeper dive into this complex and ever-evolving area of medical and legal interest. It is of great import that you find yourself a good pediatric neuro-radiologist in every brain-injury birth case and do it earlier rather than later in the process. Identifying the type and timing of an injury will help you look for the right clinical data to support your claim. Reading about MR will undoubtedly help one understand the issues involved, but there is truly no substitute for spending some time with those who really know this stuff.

This article originally appeared in the Winter 2025 edition of the American Association for Justice’s BTLG Journal.