A prospective observational cohort study and systematic review of 40 patients with mouth and genital ulcers with inflamed cartilage (MAGIC) syndrome.

OBJECTIVE

Mouth and genital ulcers with inflamed cartilage (MAGIC) syndrome is characterized by overlapping features of relapsing polychondritis (RP) and Behcet's disease (BD). To date, no studies have defined the clinical spectrum of disease in a cohort of patients with MAGIC syndrome.

METHODS

Adult patients within an ongoing prospective, observational cohort study in RP were clinically assessed for MAGIC syndrome. A systematic review was conducted to identify additional cases of MAGIC syndrome by searching four databases: PubMed (US National Library of Medicine), Embase (Elsevier), Scopus (Elsevier) and Web of Science: Core Collection (Clarivate Analytics). The inclusion criteria used were: [1] patients of any age or gender who were diagnosed with MAGIC syndrome, or both RP and BD; [2] case report or case series study; [3] published from 1985 - July 2020; and [4] in English language. Risk of bias was assessed using a checklist developed by the authors and based on the Consensus-based Clinical Case Reporting (CARE) Guidelines. Search results screening, article inclusion, data extraction and risk of bais assessment was performed independently by two investigators. Clinical characteristics, particularly BD-related features, were compared between patients with MAGIC syndrome and cases of non-MAGIC RP. The performance characteristics of different criteria to classify MAGIC syndrome were also evaluated.

RESULTS

Out of 96 patients with RP, 13 (14%) patients were diagnosed with MAGIC syndrome. For the systematic review, 380 articles were retrieved of which 90 were screened at title and abstract levels. Of these screened, 60 were excluded and 30 proceeded to full text review where an additional 8 were excluded. Twenty-two articles were included in our review and from which 27 additional cases of MAGIC syndrome were identified. Pooling all 40 cases together and comparing them with non-MAGIC RP, there was a significantly higher prevalence of ocular involvement (28% vs 4%, p<0.01), cutaneous involvement (35% vs 1%, p<0.01), GI involvement (23% vs 4%, p<0.01), and CNS involvement (8% vs 0, p = 0.04) in MAGIC syndrome. A higher prevalence of aortitis (23% vs 1%, p<0.01), Raynaud's phenomenon (54% vs 11%, p<0.01), and elevated anti-collagen II antibodies (50% vs 9%, p = 0.04) were observed in MAGIC syndrome. Fulfillment of either McAdam's or Damiani's Criteria for RP plus the International Criteria for Behçet's Disease had excellent sensitivity (98%) to classify cases of MAGIC syndrome.

CONCLUSION

A substantial proportion of patients with RP can be clinically diagnosed with MAGIC syndrome. These patients have features of RP, BD, and other unique features including aortitis, Raynaud's phenomenon and elevated anti-collagen II antibodies.