Orphanet Journal of Rare Diseases (Mar 2021)

Clinical features of central nervous system involvement in patients with eosinophilic granulomatosis with polyangiitis: a retrospective cohort study in China

  • Suying Liu,
  • Ling Guo,
  • Xiaoyuan Fan,
  • Zhaocui Zhang,
  • Jiaxin Zhou,
  • Xinping Tian,
  • Mengtao Li,
  • Xiaofeng Zeng,
  • Li Wang,
  • Fengchun Zhang

DOI
https://doi.org/10.1186/s13023-021-01780-x
Journal volume & issue
Vol. 16, no. 1
pp. 1 – 10

Abstract

Read online

Abstract Background Central nervous system (CNS) involvement is extremely rare in eosinophilic granulomatosis with polyangiitis (EGPA), but is associated with a poor prognosis in the five-factor score. This study aims to elucidate the clinical features and independently associated factors of EGPA with CNS involvement. Results CNS involvement was observed in 17.3% (19/110) of EGPA patients from Peking Union Medical College Hospital between 2007 and 2019. We retrospectively reviewed their clinical data and analyzed the independently associated factors. Their mean age was 51.7 ± 11.56 years with no male/female predominance. Ischemic lesions were the most common manifestations, accounting for 63.2% of the 19 cases, followed by posterior reversible encephalopathy syndrome (36.8%), spinal cord involvement (15.8%), medulla oblongata involvement (15.8%), and intracranial hemorrhages (15.8%). Compared to the control group, patients with CNS involvement were of older age (51.7 ± 11.56 vs. 43.7 ± 13.78 years, p = 0.019) and had a higher ratio in the digestive tract involvement (52.6% vs. 28.6%, p = 0.042). Further multivariate analysis revealed that age, disease duration, and fever were the potential independent risk factors for CNS involvement of EGPA. Glucocorticoids combined with cyclophosphamide were the strategic therapy (94.7%). Intrathecal injections of dexamethasone and methotrexate were administered to 21.1% of the patients. Although seven patients relapsed during glucocorticoid reduction, seventeen patients finally achieved clinical remission. One patient died of acute intracerebral hemorrhage within one month, and another died of gastrointestinal perforation. Outcomes and cumulative survival show no significant differences between the two groups. Conclusions CNS involvement is uncommon in EGPA with various manifestations, and ischemic lesions are the most frequent. Age, disease duration, and fever are independent factors associated with CNS involvement in EGPA. The therapy of glucocorticoids combined with cyclophosphamide and intrathecal injections yields favorable responses. Acute intracranial hemorrhage and gastrointestinal perforation may be the principal causes of death.

Keywords