Title : The role of surveillance MRI in pediatric-onset demyelinating diseases NMOSD and MOGAD: A retrospective cohort study
Abstract:
The standard of care for multiple sclerosis is to perform surveillance MRI to monitor disease activity, given that subclinical or ‘silent’ lesions can often occur. In two rare demyelinating diseases, neuromyelitis optica spectrum disorder (NMOSD) and myelin oligodendrocyte glycoprotein associated disease (MOGAD), silent lesions are rare. The utility of surveillance MRI monitoring in NMOSD and MOGAD compared to MS is unknown, and application of the imaging surveillance standards for MS to other demyelinating conditions is variable, as data regarding the relevance of routine imaging for NMOSD and MOGAD in pediatric patients is scant. Our objectives were to determine if surveillance MRI is useful for management of NMOSD and MOGAD, to define the optimal time to repeat imaging to re-establish baseline after a flare, and to analyze the costs of MRI scans deemed unnecessary to the U.S. healthcare system. As a retrospective cohort study, chart reviews were conducted to assess clinical and MRI data of thirty-five patients at the Duke Autoimmune Brain Disease Clinic diagnosed with antibody-positive NMOSD or MOGAD by 18 years of age. Key variables included the date of each scan, the purpose (for symptoms or during remission), the type of scan (brain, optics, or spine), and findings of demyelination. Of the 11 NMOSD patients, 6 (54.5%) were female. The mean age at onset was 11.4 years (range, 3-16), and the mean time of follow-up since onset was 6.5 years (range, 1-15). Of the 24 MOGAD patients, 12 (50%) were female. The mean age at onset was 9.3 years (range, 1-17), and the mean time of follow-up since onset was 6 years (range, 1-16). The NMOSD patients had a total of 37 MRIs performed during remission, and the MOGAD patients had a total of 76 remission MRIs performed. None of these scans during remission yielded any lesions, indicating a silent lesion rate of 0% for both NMOSD and MOGAD groups. Of 12 MRIs of neuroaxis locations different from the symptomatic location performed for the NMOSD patients, two lesions (both transverse myelitis (TM)) considered unexpected based on clinical phenotype were found at presentation. Of 34 MRIs of asymptomatic neuraxis locations of MOGAD patients, four unexpected lesions (3 TM, 1 acute demyelinating encephalomyelitis (ADEM)) were found at presentation. Clinical follow-up timing for evaluation and management of demyelination events was often inconsistent, but most flares improved within 2-4 months. In NMOSD and MOGAD, performing MRI of the brain, optics, and spine at presentation and of affected areas 3-4 months after a flare to re-establish baseline have clinical relevance. There were 31 MRI flares for both groups. If there was one MRI performed after every flare to re-establish baseline, this would be 31 clinically useful scans. The remaining 82 scans are potentially unnecessary, costing the healthcare system $150-650,000 (for only one part of the neuraxis, more if brain/orbits and/or spine were routinely imaged). Given the rare incidence of subclinical disease activity in pediatric NMOSD and MOGAD, as well as the costs of MRI, we recommend that further surveillance imaging is irrelevant and impractical.
Audience Take Away
- Research on the utility of surveillance MRI for these diseases in pediatric cohorts has been lacking, making clinical protocol ambiguous
- The following are the imaging guidelines that we propose for monitoring pediatric NMOSD and MOGAD patients:
- Screen orbits, brain and spine in patients at presentation to provide complete radiographic diagnostic information
- Some initial MRIs found unexpected lesions (unrelated to clinical phenotype, on different neuraxis location from symptomatic location)
- Consider repeat imaging (of affected areas only) in 3-4 months to establish a “new baseline” for comparison, to determine if a new future flare and return of symptoms shows active lesions or old residual damage from prior flare
- Most flares improved within 2-4 months, but still unclear since there are ongoing improvements likely past 1 year
- Evidence-based and conservatively clinically-sound time to define patient status when asymptomatic
- After a new baseline has been established, further routine surveillance imaging is unnecessary, as it does not identify subclinical disease activity, inform treatment, or improve health-related outcomes, as opposed to MS
- Ethical and financially practical protocol
- A significant number of financial resources would have been saved by ceasing the surveillance imaging that we found to be clinically irrelevant, given that none of these scans yielded any demyelination
- In fact, in the later years of our cohort chart review, fewer routine MRIs were already being done as clinical reasoning supported this change
- The potential guideline of performing a scan after each flare does add cost, but re-establishing baseline would be beneficial to patient outcomes, making the expense warranted
Being a pediatric study proves as an advantage, because research of children with these diseases is significantly limited and less represented in currently available data
