Classification of Neurological Abnormalities in Children With Congenital Melanocytic Naevus Syndrome Identifies Magnetic Resonance Imaging as the Best Predictor of Clinical Outcome

R. Waelchli; S.E. Aylett; D. Atherton; D.J. Thompson; W.K. Chong; V.A. Kinsler


The British Journal of Dermatology. 2015;173(3):739-750. 

In This Article


We propose the continued use of the term 'CMN syndrome' for CMN with extracutaneous features, which we deem more appropriate than NCM due to nonmelanotic CNS lesions and a causative single gene defect for the cutaneous and neurological findings. This term also brings CMN into line with terminology used for other congenital naevi, where, for example, an epidermal naevus associated with extracutaneous features is termed 'epidermal naevus syndrome'.

MRI of the CNS in children with CMN was originally started in 1988 at Great Ormond St Hospital as a research project in order to try to delineate the spectrum of disease in parallel with other centres, and to exploit the newly described sensitivity of MRI specific to melanin. This research project produced its first set of preliminary guidelines for the use of MRI in 2008, where the principal recommendation was to stop performing a routine scan for children with only a single CMN at birth; however, as is correct practice, these guidelines have now been audited to see if they continue to be fit for purpose. In the interim, we have continued to see new patients and have continued to expand our cohort of prospectively collected long-term outcome data. This cohort is now large enough that we can start to answer the question of whether a single screening CNS MRI is a valuable clinical test in patients with CMN; in other words, does it alter our management?

The results presented herein show that routine MRI of the CNS in children with two or more CMN at birth, independent of projected adult size or site of the largest CMN, and undertaken within the first year of life (ideally within the first 6 months, as myelination obscures the signal for melanin), is the best predictor of neurodevelopmental abnormalities, seizures and the requirement for neurosurgery in childhood. It is, of course, not impossible that a child with a single CMN could have a neurological abnormality on MRI; however, comparison of our data from before and after the publication of the 2008 guidelines has found that while the percentage of children scanned has significantly decreased, there has been no significant change in the percentage of abnormal scans. This suggests that the guidelines are currently fit for purpose in terms of not missing large numbers of children with serious intracranial pathology.

Clinical management was altered substantially by the radiological results. Firstly, brain and spinal tumours were resected when causing compression, and follow-up scans were used to monitor recurrence. Secondly, ventriculoperitoneal shunts were inserted where there was hydrocephalus, and follow-up scans used to monitor position and progression/resolution. Thirdly, children with leptomeningeal disease were monitored by MRI, and where it progressed a biopsy was taken to look for melanoma histologically and genetically. Lastly, where intraparenchymal melanosis alone was detected the children were seen by a paediatric neurologist at least once, and by a developmental neurologist on an annual basis, and where intervention was required (e.g. speech therapy, occupational therapy, behavioural intervention), this was instituted promptly as for any child with these clinical problems.

For many years we have recognized that the clinical outcome of children with 'symptomatic' neurological disease has not necessarily been poor, despite the majority of publications on CMN still professing this idea. Therefore, the secondary aim of this study was an attempt at subclassification of radiological abnormalities, and to correlate this with outcome measures. This has identified a high-risk group frequently requiring neurosurgery in early childhood, and possibly at higher risk of death from CNS melanoma (although this has not been modelled statistically as the numbers of deaths are small) (group 2), while providing reassurance that the most common finding of intraparenchymal melanosis alone is a nonmalignant condition, even when symptomatic (group 1).

We propose that those with normal MRI results do not require either routine repeat MRI or formal neurodevelopmental follow-up; that those in in group 1 do not require routine repeat MRI but do require neurodevelopmental follow-up on an annual basis; and that those in group 2 should have regular repeat MRI and regular neurological follow-up (Fig. 4). The periodicity of these interventions in group 2 would be decided by the multidisciplinary team, depending on the MRI findings; however, we would suggest that those with leptomeningeal disease need to monitored extremely closely initially as this disease can spread rapidly. However, if the findings are stable this intensity could be relaxed. Very importantly, any change in neurological status at any age should always trigger a repeat MRI, independent of the initial MRI findings.

Figure 4.

Great Ormond Street Hospital management guidelines for children with congenital melanocytic naevi (CMN). MRI, magnetic resonance imaging; CNS, central nervous system.