Rapid-sequence MRI for evaluation of pediatric traumatic brain injury: a systematic review

Abstract

OBJECTIVE Rapid-sequence MRI (RSMRI) of the brain is a limited-sequence MRI protocol that eliminates ionizing radiation exposure and reduces imaging time. This systematic review sought to examine studies of clinical RSMRI use for pediatric traumatic brain injury (TBI) and to evaluate various RSMRI protocols used, including their reported accuracy as well as clinical and systems-based limitations to implementation. METHODS PubMed, EMBASE, and Web of Science databases were searched, and clinical articles reporting the use of a limited brain MRI protocol in the setting of pediatric head trauma were identified. RESULTS Of the 1639 articles initially identified and reviewed, 13 studies were included. An additional article that was in press at the time was provided by its authors. The average RSMRI study completion time was variable, spanning from 1 minute to 16 minutes. RSMRI with “blood-sensitive” sequences was more sensitive for detection of hemorrhage compared with head CT (HCT), but less sensitive for detection of skull fractures. Compared with standard MRI, RSMRI had decreased sensitivity for all evidence of trauma. CONCLUSIONS Protocols and uses of RSMRI for pediatric TBI were variable among the included studies. While traumatic pathology missed by RSMRI, such as small hemorrhages and linear, nondisplaced skull fractures, was frequently described as clinically insignificant, in some cases these findings may be prognostically and/or forensically significant. Institutions should integrate RSMRI into pediatric TBI management judiciously, relying on clinical context and institutional capabilities.

Non-ionizing Imaging for the Emergency Department Assessment of Pediatric Minor Head Trauma

“…A recent systematic review, without meta-analysis, included 13 studies on the use of rapid-sequence MRI in children. Of these, seven included children with head trauma as a sole inclusion criterion, three included patients meeting various indications to undergo neuroimaging, and three focused exclusively on abusive head trauma ( 58 ). An additional relevant study was published concurrently to the systematic review ( 59 ).…”

“…Of the eight studies considered for this review ( 59 – 66 ), samples ranged between 23 and 233 patients and only two were prospective ( 60 , 61 ). The average reported time to imaging completion varied between 1 and 16 min, depending on the MRI protocols used ( 58 ). Rapid MRI was successfully completed in 99% of children with head trauma, as reported by the largest prospective study by Lindberg et al where the median time to completion was 6 min ( 60 ).…”

“…These data show that rapid MRI is a feasible option for stable children with MHT in the ED. With respect to its diagnostic accuracy, although difficult to compare across studies due to the variable rapid MRI protocols used, this technique resulted overall comparable to CT for the detection of TBIs (sensitivity between 85 and 100%, and specificity between 83 and 100%), while less sensitive for linear non-displaced skull fractures (with wide variability in sensitivity across studies, between 34 and 100%) ( 58 , 59 ). The accuracy in detecting linear skull fractures was reported to remain limited, even when complemented by a black bone sequence ( 62 ), although missed fractures never required neurosurgical intervention ( 58 , 59 , 63 , 67 ).…”

“…With respect to its diagnostic accuracy, although difficult to compare across studies due to the variable rapid MRI protocols used, this technique resulted overall comparable to CT for the detection of TBIs (sensitivity between 85 and 100%, and specificity between 83 and 100%), while less sensitive for linear non-displaced skull fractures (with wide variability in sensitivity across studies, between 34 and 100%) ( 58 , 59 ). The accuracy in detecting linear skull fractures was reported to remain limited, even when complemented by a black bone sequence ( 62 ), although missed fractures never required neurosurgical intervention ( 58 , 59 , 63 , 67 ). CT scans and rapid-sequence MRIs, in the relevant studies ( 59 – 66 ) were mostly performed within a time interval of 48 h. It is important to note that the variable time interval between CT scans and rapid-sequence MRIs may have affected the sensitivity of the latter for minor intracranial hemorrhages.…”

“…Based on the feasibility and accuracy data of rapid MRI, this imaging modality seems to fit the ED needs for a timely diagnosis and a rapid throughput. However, institutions should integrate rapid MRI into acute pediatric TBI management judiciously, relying on the clinical context and institutional capabilities, especially for the prognostic and legal implications ( 58 ). Unfortunately, the infrastructural accessibility of rapid MRI is limited compared to CT ( 71 ) and it requires experienced and qualified staff for proper exam reading and interpretation, in order to avoid pitfalls.…”