The impact of guidelines on the quality of MRI diagnostics in adult patients referred to a tertiary Epilepsy Centre

Abstract

Magnetic resonance imaging (MRI) is a crucial diagnostic imaging tool in epileptology. International guidelines have been published by the International league against epilepsy in 1998 and 2000 [1, 2]. In 2008, the German Neurological Society published Guidelines for appropriate MRI scans in epilepsy patients [3]. These guidelines are easily accessible by the internet and also published in print. The last edition was published in 2012 [4]. The Kork Epilepsy Centre is a large tertiary referral centre in Germany. We suggested that in patients who had been referred for reasons of drug-resistant epilepsy, MRI scans according to the guidelines had been performed prior to admission in almost every case. This was the question we addressed with this cross-sectional study. From April to September of 2012, consecutive referrals were systematically investigated. For the patients who had been referred due to their difficult-to-treat epilepsy, we assessed for how many MRI scans had been performed during the previous 5 years and how many of those had been performed according to the guidelines. Since appropriate MRI scans may be more difficult in patients with severe cognitive or behavioural problems, patients with such co-morbidities were excluded. With these restrictions, 208 patients were identified. Out of these, 81 (39 %) had been hospitalized at our Centre previously and had been investigated with guideline-concordant MRI scans at that time. Among the remaining 127 patients, 114 (90 %) had undergone a prior external MRI during the previous 5 years. Thus, in 13 patients, or 10 %, no MRI had been performed at all. If MRI scans had been done, only 34 (30 % of 114) fulfilled the criteria of the guidelines. In the remaining 80 patients, or 70 %, MRI did not accomplish the requirements of the guidelines due to inappropriately thick slides, too few signal weightings or sequence parameters, a lack of temporal angularization in spite of the question of temporal lobe epilepsy, a field strength below 1.5 T, or various of these reasons. There are two major potential drawbacks of MRI quality: first, the quality of the responsible physicians may vary. It has been reported that unequivocal epileptogenic lesions were detected only by 39 % of standard MRI scans that had been performed outside a specialized epilepsy centre. This rate increased to 50 % if these external standard MRI scans were analyzed by expert epileptologists [5]. If scans were added that were performed at the specialized epilepsy centre and that addressed the specific requirements appropriate MRI diagnostics should meet in epileptological issues, lesions were detected by experts in 91 % of all cases. The Guidelines of the German Neurological Society were recently published as the 5th edition and are available both as a book as well as digitally [4]. Requirements for an appropriate MRI comprise sagittal and coronal T1, axial and coronal T2, axial and coronal fluid attenuation inversion recovery (FLAIR) sequences with temporal angularization, a thickness of maximally 4 mm, and the additional application of gadolinium in case of a lesion. The second potential drawback of quality is the insufficient application of such guidelines. If patients are transferred to a tertiary referral epilepsy Centre, one should expect that these patients already had appropriate MRI diagnostics. However, we found that in 10 %, not only insufficient, but no MRI imaging at all had been performed. If external MRI scans had been done, only B. J. Steinhoff (&) M. Bacher I. Bucurenciu R. Kornmeier C. Kurth P. Martin K. Mathews A. M. Staack Kork Epilepsy Centre, Landstrasse 1, 77694 Kehl, Germany e-mail: bsteinhoff@epilepsiezentrum.de