There is insufficient evidence to support an evidence-based recommendation for the safety of diagnostic brain MRI in pTBI patients with identified or suspected metallic intracranial foreign bodies

In the absence of direct scientific evidence, EXPERT CONSENSUS concluded that:

• MRI of the brain should not be obtained in pTBI patients with unknown or ferromagnetic intracranial foreign bodies. (100% consensus)
• Non-ferromagnetic metals can be contaminated with ferromagnetic metals so MR imaging represents a potential hazard and should be avoided for any patient with intracranial metal following a penetrating brain injury. (100% consensus)
• Injuries in which the foreign body is known to be non-metallic, MRI may be considered. (100% consensus)

No evidence or expert opinion supported distinct recommendations based on patient gender, age, wounding mechanism, or military vs. civilian context.

Magnetic Resonance Imaging (MRI) is distinguished for its superior resolution over Computed Tomography (CT) in imaging most soft tissues, particularly the brain, making it highly effective in scenarios involving penetrating trauma. It excels in delineating ballistic trajectories, assessing the extent of soft tissue damage, and identifying retained foreign objects, while also eliminating the risk of exposure to ionizing radiation - a notable advantage in situations requiring frequent imaging. MRI's multiplanar imaging capabilities further enhance its utility by obviating the need for patient repositioning, thereby facilitating prompt and accurate assessments crucial for determining the severity of injuries, prognostication, and developing surgical or therapeutic strategies. This is especially vital in neuroimaging, where the choice between MRI and CT can significantly influence patient outcomes due to their respective safety and efficacy profiles. However, despite the advantages of MRI, the practical challenges related to its availability and the time constraints often make CT a more feasible option for immediate assessment.

However, the operation of MRI through intense magnetic fields introduces potential risks in the presence of ferromagnetic materials within the body, such as bullet fragments or shrapnel in penetrating cranial injuries. These materials can move or generate heat, posing risks of further serious injury. The majority of evidence supporting these concerns comes from experiments with gel and post-mortem models, which may not fully replicate clinical scenarios. Nevertheless, MRI's ability to provide superior resolution for evaluating brain tissue damage and identifying non-metallic objects underscores its potential, necessitating a thorough literature review to assess the safety and efficacy of MRI in such critical conditions.

Prior guidelines indicate that MRI has a limited role in the acute management of pTBI upon assimilation of very low evidence from the 1990s. MRI is generally not recommended for use in the acute management of pTBI. The concern for retained ferromagnetic fragments was evaluated in studies by Oliver and Kabala,26 Smith et al.,27 and Teitelbaum et al.28 In addition to artifact and image distortion, ferromagnetic missiles can also rotate and deflect in response to magnetic torque. However, the literature has not reported any patient suffering additional injury caused by MRI. In certain situations when the presence of metal fragments is not a consideration, MRI can be valuable. For example, whereas penetrating wood may be poorly visualized on CT scan, MRI can better reveal the projectile and associated tissue damage. The role of other imaging studies, such as intraoperative ultrasound, magnetic resonance angiography, stereotactic imaging techniques, PET, and SPECT in the acute management of pTBI has not yet been defined.

Two studies met criteria for inclusion as evidence and which included humans with penetrating intracranial trauma undergoing magnetic resonance imaging (MRI).1,2 The studies had extremely low numbers of subjects and were categorized as moderate to high risk of bias. The first study was a forensic analysis of individuals with fatal gunshot wounds to the head (N=10), examined postmortem with both computed tomography (CT) and MRI;1 the second was of patients with retained bullets or shrapnel, regardless of location, undergoing MRI, only one of whom had intracranial shrapnel. Both studies reported image artifacts with both MRI and CT. The artifact was considered a safety issue because it could result in missed diagnosis. MRI images were less likely than CT images to have artifacts, categorized as severe (10% vs. 50%) or considerable (0% vs. 20%) in the study of 10 forensic cases (relative risk 0.14, 95% confidence interval 0.02 to 0.96 for severe and considerable combined). The one case with severe artifact on MRI was the only one with a ferromagnetic bullet (which was intact), and no movement or heating was detected. In the study reporting one patient with intracranial metal, the MRI showed artifact that obliterated the image around the shrapnel fragment, and suggested it was ferromagnetic. However, the MRI also detected additional metal fragments not seen with X-ray or CT, and the MRI procedure reportedly was not associated with adverse clinical effects. Together, the two studies provide very low strength evidence to base any recommendations.

The first edition3 guideline recommended CT as the primary imaging modality for suspected penetrating head trauma, due to concerns about potential movement of metallic objects in the strong magnetic field of an MRI machine. MRI remains a secondary option, especially if no metallic fragments were confirmed or if more detailed imaging of brain tissue was needed post-initial assessment.

Previous recommendations were based on evidence from two studies using non-human models. While these studies were excluded from the current guidelines because they did not meet the inclusion criteria, they must not be disregarded completely. These animal models demonstrated substantial heat and movement of the ferrous fragments in simulated brain tissue as well as difficulty in the prediction of which ballistic fragments contain ferrous material.

There is very low evidence from 2 retrospective studies of moderate-high risk of bias that included only 11 subjects total, 10 of which were in post mortem patients, and only 1 in living human subjects. More data is needed to inform the role and safety of MRI for pTBI.

CT is the workhorse cranial imaging modality in neurotrauma because of its speed and because it readily allows detection of surgically relevant findings such as blood and midline shift of the brain. While MRI can play an important role in delineating the anatomy and trajectomy of intracranial foreign bodies which are radiolucent of CT such as wood, there are few indications for considering an MRI following pTBI in patients with metallic fragments. Potential harms of MR imaging include inducing motion of the fragment, causing further injury to the brain, as well as heating of the fragment to injurious temperatures. pTBI patient can encounter include concern for acute stroke or in a more chronic setting to rule out a brain tumor. Our review of the literature found evidence that ferromagnetic contamination is common in metals believed to be non-ferromagnetic. On this basis we feel it is safest to view all metallic foreign bodies as high risk for MR imaging. In some cases need for MR imaging could be a reason to consider removal of a metallic foreign body. In discussions amongst our panelists we note that several reported successfully obtaining MR imaging in such patients when there was a strong indication for doing so. Ultimately, however, came to a unified view that MR imaging is ill advised for any patient with intracranial metallic foreign bodies.

Future research on the use of MRI in the setting of pTBI with potential metallic intracranial foreign bodies should address the safety, feasibility of imaging protocols, and the efficacy of diagnostic outcomes. Given the significant safety concerns, it is unlikely that in-human prospective studies directly comparing the safety and diagnostic outcomes of acute MRI versus those of CT in pTBI patients with known or suspected metallic intracranial foreign bodies would be ethically endeavored. Robust post-mortem models in the acute setting may provide further information as well as animal models. However, there is potential equipoise in some settings where movement and/or heat generation of the metal fragment during MRI might be more acceptable, or produce less harm (e.g. location in non-eloquent regions of the brain, fragments embedded in bone). Both of these scenarios would require knowledge of these fragments prior to MRI, presumably by CT or other modality. In these cases of study, there is a need for developing and validating imaging protocols that minimize the risk to patients with metallic foreign bodies. This includes the use of low-field MRI, sequences designed to limit heating and displacement of metal fragments, protocols for rapid identification of adverse events and studies to explore technological advancements in MRI that may reduce the risks associated with metallic fragments.

Leveraging existing data from trauma registries and retrospective analyses to better understand the prevalence of adverse events associated with MRI in this patient population could help in identifying specific risk factors associated with negative outcomes. Addressing ethical considerations, including the development of informed consent processes that clearly communicate the potential risks and benefits of MRI in this context is crucial for ensuring that patients can make informed decisions about their care.