Abstract
Objective To investigate the effect of sex on the clinical characteristics, prognoses, and therapeutic selection of eosinophilic granulomatosis with polyangiitis (EGPA).
Methods We retrospectively enrolled 170 hospitalized patients with EGPA who were managed at our hospital between 2007 and 2020. Detailed clinical data were reviewed. Manifestations, prognoses, treatments, and outcomes were compared between female and male patients. Cumulative survival rates were calculated using Kaplan-Meier curves.
Results In this cohort, the male to female ratio was 1.4:1. Renal involvement was more frequent in male patients, including serum creatinine elevation, and proteinuria > 1 g/24 h. Severe gastrointestinal (GI) involvement occurred more commonly in male patients. Female patients had longer allergy duration and higher ratios of allergic rhinitis and asthma. Sex differences in proteinuria > 1 g/24 h, serum creatinine > 150 mmol/L, severe GI involvement, and weight loss were more significant in patients aged ≤ 55 years than those in patients aged > 55 years. Overall, male patients had a higher Birmingham Vasculitis Activity Score and a worse prognosis assessed at diagnosis, with a lower proportion of 1996 Five Factor Score = 0 than females. Regarding treatment selection, methylprednisolone pulse and cyclophosphamide were administered more frequently to male patients. All-cause mortality and cumulative survival rates were comparable between the sexes.
Conclusion In this Chinese EGPA cohort, male and female patients showed distinct disease phenotypes. Male patients with EGPA had a higher disease activity at diagnosis and required more aggressive treatment for remission induction.
Eosinophilic granulomatosis with polyangiitis (EGPA), formerly called Churg-Strauss syndrome, is a rare type of antineutrophil cytoplasmic antibody (ANCA)–associated systemic vasculitis (AAV). Hallmarks of EGPA include asthma, blood and tissue eosinophilia, necrotizing vasculitis, and granulomatous formation.1,2 The annual incidence of EGPA was 1.1 to 2.7 per million person-years and the prevalence was 12.1 to 15.6 cases per million individuals in Europe between 1988 and 2017, and the rates have been increasing in recent decades.3-5 The age of EGPA onset is primarily 40 to 60 years.6 Unlike other rheumatic diseases with predominant sex differences, such as systemic lupus erythematosus (SLE), primary Sjögren syndrome, and rheumatoid arthritis, EGPA has no significant sex predominance.7-9 In EGPA, the clinical manifestations are diverse, including respiratory system involvement, peripheral neuropathy, cardiac lesions, renal involvement, digestive system lesions, and ear, nose, and throat involvement.1 In particular, severe cardiac, renal, or digestive system manifestations are associated with a poor prognosis in EGPA.10
Sex differences are widely analyzed in many other diseases, and this topic is getting more and more attention for predicting disease phenotype and developing precision medicine.11,12 However, the effect of sex on the clinical manifestations, prognoses, and treatment selection of EGPA has not been systematically reported to date. Therefore, this study aimed to analyze the clinical features of male and female patients with EGPA over recent decades, using a retrospective large Chinese EGPA cohort.
METHODS
Patients. Hospitalized patients with a diagnosis of EGPA were recruited from January 2007 to December 2020. All patients fulfilled the 1990 American College of Rheumatology classification criteria for EGPA, and the diagnosis was confirmed by 2 rheumatologists certified by Peking Union Medical College Hospital (PUMCH).13 Detailed clinical data were reviewed, including demographics, laboratory test results, manifestations, treatments, and outcomes. The data on the laboratory examinations, including ANCA testing, were collected at the time of diagnosis. All the patients signed informed consent forms. Our study was approved by the ethical committee of PUMCH (approval no. S-K1385).
Clinical evaluation. Different organ involvements at diagnosis were carefully assessed according to our previous study, with minor modifications.14 Severe gastrointestinal (GI) lesions included ulcers, bleeding, perforation, obstruction, and necrosis of the digestive tract that could not be explained by other diseases. The disease activity of EGPA at diagnosis was analyzed using the 1994 Birmingham Vasculitis Activity Score (BVAS) system,15 and the prognosis at baseline was calculated using the 1996 version of the Five Factor Score (FFS) and the 2011 revised version of the FFS.10,16
Treatments and outcomes. Treatments at diagnosis consist of 2 stages. In the stage of remission induction, high doses of glucocorticoids (GCs) combined with immunosuppressants were the primary therapy. Methylprednisolone (MP) pulse regime was defined as 500-1000 mg/day for 3-5 days, and high- and medium-dose prednisone was administered at 1-2 mg/kg/day and 0.5-0.8 mg/kg/day, respectively. Refractory EGPA in this study was determined based on a progressive condition unresponsive to GCs and cyclophosphamide (CYC) according to previous literature.17 In the maintenance stage, GCs were tapered to a low dose of 5-15 mg/day, and immunosuppressants were frequently selected among methotrexate, Tripterygium wilfordii Hook. F, and azathioprine based on the specific state of the patients. In terms of the outcomes, death referred to all-cause death, partial relief was defined as a decrease in BVAS of more than half, and complete remission was defined as a reduction in BVAS to 0 for at least 6 months.
Statistical analysis. All analyses were conducted using SPSS 25.0 (IBM Corp.) and Prism 7 (GraphPad Software). Continuous, normally distributed data are presented as mean (SD). Continuous data that did not conform to the normal distribution are shown as median (IQR). Correspondingly, t tests and nonparametric tests were used to analyze the differences between the 2 groups. Qualitative variables are described as n (%) and were compared using Pearson chi-square test or Fisher exact test. Survival analysis was assessed using Kaplan-Meier curves, and the data were compared using log-rank tests. Statistical difference was determined at 2-sided P < 0.05. For multiple comparisons of 3 ratios, the test level was adjusted using Bonferroni correction, which was 0.017. We further performed the stratification analysis based on the age > 55 or ≤ 55 years.
RESULTS
Demographics. In total, 170 hospitalized patients with EGPA were consecutively enrolled in the study. The male to female ratio was 1.4:1 (98/72). The mean (SD) age was 45.4 (14.5) years, with a range of 13-80 years. The median duration from allergy to EGPA diagnosis was 24 (range 0-475) months. The median durations in the 2 groups were 12 months in male patients and 15 months in female patients, which was not statistically different. The patients in this cohort were from across China.
Clinical characteristics. We compared the differences in clinical characteristics at diagnosis between female and male patients with EGPA (Table 1). Regarding laboratory tests, we did not find differences in the eosinophil count or ratio, erythrocyte sedimentation rate, or rheumatoid factor level. However, male patients had relatively higher C-reactive protein (CRP) levels than female patients (20.9 [IQR 5.0-70.7] mg/L vs 10.6 [IQR 2.9-38.6] mg/L, P = 0.06). The ANCA-positive proportion was 22% without differences between the 2 groups.
Baseline clinical characteristics of patients with EGPA.
We reviewed the clinical manifestations in the 2 groups. Male patients had higher ratios of renal involvement (43% vs 19%; Figure 1A) and severe GI lesions (15% vs 6%; Figure 1B) than female patients. Urine test abnormalities (41% vs 18%) and serum creatinine elevation (15% vs 3%) were more prominent in male patients; in particular, the percentage of proteinuria > 1 g/24 h was higher than that in female patients (10% vs 1%; Table 1). Additionally, male patients experienced weight loss at a relatively higher rate, but there was no remarkable difference (45% vs 31%). We assessed the disease activity for all patients with EGPA at diagnosis using BVAS and found that the male patients had a higher BVAS than the female patients (median 15 vs 13, P = 0.047; Table 1).
Sex differences of clinical manifestations in EGPA. Comparisons regarding (A) renal involvement, (B) severe GI lesions, (C) allergic rhinitis, and (D) asthma between male and female patients. EGPA: eosinophilic granulomatosis with polyangiitis; GI: gastrointestinal.
In contrast, female patients displayed a longer duration from allergy to EGPA diagnosis than male patients (36 [IQR 11-120] vs 20 [IQR 0-48] months, P = 0.01; Table 1). Further, female patients presented higher rates of allergic rhinitis (49% vs 32%, P = 0.03; Figure 1C) and asthma (85% vs 59%, P < 0.001; Figure 1D).
To adjust for the potential effect of age, we stratified all patients into 2 groups based on age ≤ 55 years or age > 55 years (Table 2). Compared with the patients aged > 55 years, males had higher proportions in terms of proteinuria > 1 g/24 hour, serum creatinine elevation, severe GI lesions, weight loss, as well as abnormal urine test results and renal lesions than females in patients aged ≤ 55 years. Myocardial involvement was more common in males among patients aged > 55 years. In contrast, the female predominance in the allergic duration was significant and showed a tendency in allergic rhinitis in patients > 55 years of age, and asthma presented with female predominance in both age groups.
Baseline clinical characteristics of patients with EGPA stratified according to the age class.
Prognosis analysis. FFS was used to evaluate prognoses for all patients with EGPA at diagnosis. In the 1996 FFS (Figure 2A), the proportions of patients with FFS = 0, FFS = 1, and FFS ≥ 2 between the sexes were statistically different (P = 0.03). Specifically, more female patients had FFS of 0 than male patients (63% vs 44%, P = 0.02). Conversely, male patients presented with a relatively higher proportion of FFS ≥ 2 (15% vs 6%, P = 0.046). Further, in the secondary analysis (Table 2), only in those aged ≤ 55 years, a higher percentage of FFS = 0 was seen in female patients (63% vs 42%), and a higher proportion of FFS ≥ 2 occurred in male patients (17% vs 5%).
Prognosis assessment and crucial treatment regimens in EGPA. (A) The 1996 version of FFS was used to assess the prognosis of EGPA. (B-D) Comparisons of treatment selection between the sexes in EGPA. CYC: cyclophosphamide; EGPA: eosinophilic granulomatosis with polyangiitis; FFS: Five Factor Score; IVIG: intravenous immunoglobulin; MP: methylprednisolone.
Treatment and survival analysis. We compared the crucial therapeutics in the remission induction stage between male and female patients (Figure 2B-D). The male patients were more frequently administered MP pulse therapy (34% vs 18%, P = 0.02) and CYC (including oral and intravenous administration; 90% vs 75%, P = 0.01) than female patients. Intravenous immunoglobulin (IVIG) was prescribed for male patients relatively more commonly (20% vs 10%, P = 0.06). Additionally, biological agents were used in a subset of patients with EGPA and included rituximab (6 patients), omalizumab (3 patients), and tocilizumab (1 patient). There was no difference between the 2 groups in the proportion of refractory patients with EGPA (24% male vs 19% female, P = 0.53).
Through follow-up of a median of 23 (range 1-132) months, we calculated the death rates for the cohort to be 6% in male patients and 4% in female patients. Additionally, the cumulative survival rates were comparable between the 2 groups (log-rank test, P = 0.59; Figure 3).
Survival rates of patients with EGPA. Comparison of cumulative survival rates between the sexes in patients with EGPA. EGPA: eosinophilic granulomatosis with polyangiitis.
DISCUSSION
In this single-center cohort of EGPA, we found significant differences in the clinical presentations between male and female patients. Male patients were more prone to develop renal and severe GI involvement, especially patients ≤ 55 years of age, whereas female patients presented more frequently with allergic rhinitis, asthma, and longer allergy duration.
According to previous literature, the ratio of males to females with EGPA was 0.8-1.1:1, without sex predominance, which is consistent with the result in our study.9,18-20 Biological differences exist between the sexes that extend beyond the reproductive system and the underlying mechanisms remain elusive.21 Notably, most rheumatic diseases affect women disproportionately.22 For instance, in SLE and primary Sjögren syndrome, female patients account for more than 85% of all patients,23,24 indicating that sex is implicated in the pathogenesis of these diseases. However, to date, no comprehensive investigation has been performed to evaluate the effect of sex on EGPA clinical manifestations. EGPA is a complex, heterogeneous disease, the etiology of which is unclear. In this study, we show differences in disease presentation of EGPA among male and female patients, showing a higher frequency of renal involvement in men and a higher frequency of rhinitis and asthma in women. Our previous study revealed that male sex, myeloperoxidase-ANCA positivity, and ear lesions were independent factors associated with renal lesions in EGPA.14 Severe GI involvement is relatively rare but more prevalent in male patients, which might contribute to more frequent weight loss. Our data could suggest that sex plays a role in clinical presentation and disease phenotype. In a French Vasculitis Study Group cohort study, Comarmond et al reported that female patients had a longer duration of asthma and less frequent renal involvement compared with male patients, consistent with our findings.20
Our results indicated that male patients with EGPA had a relatively higher CRP level and a significantly higher BVAS at diagnosis compared to female patients, indicating that the baseline disease activity was higher in male patients. A similar phenotype was observed in male patients with IgG4–related disease and SLE.23,25 FFS in patients with EGPA reflected that male patients had a more aggressive presentation of EGPA at diagnosis than female patients. Correspondingly, the male patients tended to be prescribed more aggressive therapies involving MP pulse, CYC, and IVIG. Therefore, male sex might be a predictor of severe clinical manifestations in patients with EGPA.
Other forms of AAV, including microscopic polyangiitis (MPA) and granulomatosis with polyangiitis (GPA) have been analyzed in terms of sex differences. An AAV cohort study from Korea that was published in 2021 reported that male patients received CYC more frequently than female patients, but there were no significant differences in the poor outcomes between males and females. However, the cumulative survival and multivariable Cox hazards model analysis based on all-cause mortality suggested that male sex is a significant and independent predictor of all-cause mortality in patients with AAV. This cohort is composed of MPA (55%), GPA (26%), and EGPA (19%).26 A Norwegian AAV cohort study published in 2018 revealed that male sex was associated with an increased risk of endstage renal disease in patients with ANCA-associated glomerulonephritis, but did not mention the specific subtype of AAV.27 A German AAV cohort study that included only MPA and GPA, published in 2021, indicated that female individuals developed AAV manifestations at an older age with more joint involvement and less interstitial inflammation and vasculitis in peritubular capillaries than male individuals.28 Moreover, in Behçet disease, a kind of systemic vasculitis affecting vessels of various sizes, vascular symptoms with severe complications or mortality were more frequent in males than in females.29 Collectively, male individuals had more severe clinical manifestations than female individuals, which might be a common characteristic in systemic vasculitis.
The mechanisms for the sex differences in clinical characteristics of EGPA remain unclear, and some essential factors might have a role. Sex hormones modulate immune system development and activity.30,31 Estrogen receptor signaling is required for thymus development in mice,32 and estrogen facilitates humoral responses and promotes B cell differentiation and Ig production.33 Of note, male sex hormones downregulated systemic eosinophil responses to infection with Brugia pahangi, whereas female mice showed significantly enhanced eosinophil responses.34 In our study, although the eosinophil count and ratio were comparable between the sexes, female patients had more prominent manifestations of asthma and allergic rhinitis, which is associated with a higher eosinophil response and might be enhanced by female hormones.35 Moreover, a previous study suggests that the complement system is a source of sexual dimorphism in vulnerability to different disorders.36
Our stratification analysis based on age suggests an age-sex interaction in EGPA presentations. The mean age of natural menopause ranges from 46 to 54 years in females according to the literature, and is associated with the risk of cardiovascular disease.37,38 To minimize the influence of age-associated estrogen before menopause, we selected 55 years as the cut-off value to divide all patients into 2 groups, as used previously.25 Notably, we found that the sex differences in proteinuria > 1 g/24 h, serum creatinine > 150 mmol/L, severe GI lesions, and weight loss became more significant in the younger patients aged ≤ 55 years, whereas the older group did not show significant differences. This phenomenon indicates that older patients have fewer clinical differences between the sexes in terms of EGPA.
The present results should be interpreted taking into consideration some limitations. First, this study was based on a retrospective cohort; thus, the data may be incomplete, and the statistical power is not strong enough. Second, although EGPA is rare, a single-center study for EGPA was conducted successfully because our hospital received a large number of complex and critically ill patients from across China. Moreover, only hospitalized patients were analyzed in this study. Therefore, conclusions from this EGPA cohort study might not be extrapolated to the general EGPA population. A prospective multicenter cohort study for EGPA is needed in China that should enroll inpatients and outpatients.
In summary, our study revealed significant differences in the clinical manifestations of EGPA between the sexes. Renal involvement and severe GI lesions were more common in male patients, whereas asthma and allergic rhinitis occurred more frequently in female patients. Male patients tended to require more aggressive treatment than female patients. More in-depth studies are warranted to reveal the contributions of sex in EGPA.
ACKNOWLEDGMENT
We would like to thank all the patients enrolled in our study and all the professional staff in our hospital.
Footnotes
This work was supported by the National Key Research and Development Program of China (2018YFE0207300, 2016YFA0101003, 2016YFC0903901), Nonprofit Central Research Institute Fund of Chinese Academy of Medical Sciences (2019XK320022), National Natural Science Foundation of China (81501414, 81771764, 81571594), and CAMS Innovation Fund for Medical Sciences (2017-I2M-3–008, 2016-I2M-1–003).
S. Liu and L. Han contributed equally to this work as co-first authors.
L. Wang and F. Zhang contributed equally to this work as co-senior authors.
The authors declare no conflicts of interest relevant to this article.
- Accepted for publication May 11, 2023.
- Copyright © 2023 by the Journal of Rheumatology
