An MR framework enabling simultaneous multiple parametric T1 and T2 proton density mapping—MR fingerprinting—can identify lesions indicative of a severe neurological condition in patients with a common form of epilepsy—all in under 150 seconds.
In this prospective study, published online June 19 in Radiology, investigators created T1 and T2 maps from MRIs of patients with drug-resistant mesial temporal lobe epilepsy (MTLE) with unilateral or bilateral hippocampal sclerosis (HS).
Participants were divided into two groups: 33 patients with MTLE and 30 healthy patients.
Results showed a diagnosis rate (ratio of HS diagnosed with a 2.5-minute MR fingerprint exam compared to standard MRI, electroencephalography and PET) of 32 of 33 that “indicates improved accuracy and sensitivity of diagnosis,” wrote first author Congyu Liao, with the College of Biomedical Engineering and Instrumental Science at Zhejiang University in China and colleagues.
In comparison, routine examinations had a diagnostic rate of 23 of 33 (69.7 percent).
MR fingerprinting also determined the final diagnosis of 33 MTLE patients. At final diagnosis, 28 patients were determined to have unilateral temporal lobe epilepsy HS and five were diagnosed with bilateral TLE-HS. Liao and colleagues wrote these findings “also demonstrate that T2 values are different between HS lesions and contralateral normal tissue, which suggests that MR fingerprinting is reliable and effective for MTLE diagnosis.”
In a related editorial, Daniela Prayer with Medical University in Vienna, believed MR fingerprinting outperformed routine examinations due to its collection of additional information obtained through tissue fraction segmentation and volumetric calculations.
While she did express concern over the shift in diagnostic ability from the individual radiologists toward the postprocessing of gathered data done by MR fingerprinting, Prayer noted the additional information could eventually personalize surgical procedures for TLE patients.
“Future presurgical workups in patients with drug-resistant epilepsy may consist of MR fingerprinting and an additional diffusion-tensor imaging sequence that depicts an individual-specific network,” Prayer wrote. “These personal networks might be used to model a tailored surgery protocol with respect to postoperative outcome.