Spinal Cord Leptomenigeal Enhancement as a Marker of Myelitis Severity in Children with MOGAD
Serenella Bartiromo1, Cesar Alves2, Arastoo Vossough3, Julia O'Mahony4, Ann Yeh5, Ann Marrie Ruth6, Sridar Narayanan7, Amit Bar-Or8, Alberto Gajofatto1, Brenda Banwell9, Giulia Fadda10
1University of Verona, 2Harvard University, 3University of Pennsylvania - Children'S Hospital of Philadelphia, 4Cleveland Clinic, 5The Hospital for Sick Children, 6University of Manitoba, 7Montreal Neurological Institute, 8University of Pennsylvania, 9Childrens Hospital of Philadelphia, 10Medicine, University of Ottawa
Objective:

To characterize the clinical, CSF and MRI features associated with spinal cord leptomeningeal enhancement (sLME) in children with myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD).

Background:

Leptomeningeal inflammation has been recently reported in children with MOGAD as evidenced by enhancement of the spinal leptomeninges on post-contrast MRI. Whether sLME is associated with disease severity in MOGAD myelitis is unknown.

Design/Methods:

We included participants in the prospective Canadian Pediatric Demyelinating disease study with evidence of spinal lesion(s) on MRI, and for whom both gadolinium-enhanced MRI and serum MOG-IgG testing results were available. We compared clinical and MRI findings between participants with MOGAD myelitis and seronegative monophasic myelitis with and without evidence of sLME.

Results:

We included 33 children with MOGAD myelitis and 45 with seronegative monophasic myelitis. Spinal cord LME was observed in 20/33 children with MOGAD and in 13/45 children with seronegative myelitis.  Among children with MOGAD, sLME associated with higher frequency of longitudinally extensive lesions (95 vs 62%), H sign (75% vs 38%), spinal cord tumefactive lesions (50% vs 8%), complete cross-sectional involvement (80% vs 38%), nodular enhancement (35% vs 0%) and more spinal lesions (median 2 vs 1), (all p<0.05). No significant differences were found in the frequency of CSF oligoclonal bands (21% vs 13%). Only 5 children with MOGAD, all with sLME, experienced clinical relapses, while no relapses were reported among the 33 MOGAD children without sLME (p=0.13). Among children with seronegative myelitis, sLME associated with more spinal cord tumefactive lesions (40% vs 13%) and complete cross-section involvement (79% vs 42%) (all p<0.05).

Conclusions:

Spinal cord LME, irrespective of MOG-IgG status, associates with more extensive spinal cord lesion severity (as defined by more extensive cross-sectional involvement and a tumefactive cord lesion appearance). The mechanism behind this association might provide insights into determinants of outcome and deserves further studies.

 

10.1212/WNL.0000000000205824