Clinical value of biomarkers in relation to artery size in eosinophilic granulomatosis with polyangiitis: findings from an inception cohort at a Japanese City Hospital
摘要
Eosinophilic granulomatosis with polyangiitis (EGPA) is a necrotizing vasculitis that frequently affects small arteries in the skin and the nerve but can also involve the medium-sized arteries, resembling polyarteritis nodosa (PAN). This study aimed to evaluate the clinical value of biomarkers in EGPA in relation to the size of the affected arteries.
MethodIn an inception cohort of EGPA, we examined the relationships between eosinophil counts, myeloperoxidase-antineutrophil cytoplasmic antibody (MPO-ANCA) positivity, levels of rheumatoid factor (RF) and Immunoglobulin E (IgE) and pre-defined outcomes based on artery size. These outcomes included histologically confirmed cutaneous vasculitis, glomerulonephritis, PAN-like arteritis in the cranial and abdominal arteries, and cardiomyopathy.
ResultsIgE levels were significantly elevated in patients with cutaneous vasculitis and cardiomyopathy compared to those without the respective conditions. The median IgE values (range) were 3620 IU/mL (1160–6520) vs. 1000 IU/mL (229–5490) for cutaneous vasculitis (p = 0.0077), and 4065 IU/mL (1320–6520) vs. 1205 IU/mL (229–4030) for cardiomyopathy (p = 0.032). A numerically higher frequency of MPO-ANCA was observed in cases with PAN-like arteritis compared to those without (4/5 vs 8/25), similar to the findings for MPO-ANCA in glomerulonephritis. The crude odds ratio of MPO-ANCA positivity for both glomerulonephritis and PAN-like arteritis was 8.50 (95% CI: 1.05–181.45). The optimal IgE thresholds for detecting cutaneous vasculitis and cardiomyopathy were 1860 IU/mL and 2465 IU/mL, respectively, with areas under the curve of 0.82 and 0.84, indicating favorable diagnostic performance.
ConclusionsIgE levels were associated with cutaneous vasculitis and cardiomyopathy in an inception cohort of EGPA in Japan. Further studies with larger cohorts are needed to confirm these findings, as our results are based on a retrospective study with a small sample size.