Introduction

Anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV) is a necrotizing vasculitis, mainly affecting small vessels, with few or no immune deposits, including microscopic polyangiitis (MPA), granulomatosis with polyangiitis (GPA), eosinophilic granulomatosis with polyangiitis (EGPA) [1]. ANCA specific for myeloperoxidase (MPO) is common in MPA, while proteinase 3 (PR3) is common in GPA [2]. The proportion of MPA patients is substantially higher in Asian countries compared with Northern Europe and the USA [3]. The incidence of AAV in China is mainly MPA (about 80%), GPA (about 20%) and EGPA (relatively rare), as reported [4].

AAV is a common disease with high mortality. In European countries, AAV usually occurs in patients aged 65–74 years, with 10–20 per million in the general population[5]. The 1- and 2-year mortality were 82% and 90%, respectively, in untreated populations, and the leading cause of death was uremia as a result of rapidly progressive glomerulonephritis (RPGN) [6]. Renal involvement in AAV commonly manifests as ANCA-associated glomerulonephritis (AAGN) [7], which has a great impact on the prognosis, especially on patient’s survival and renal survival [8]. According to Improving Global Outcomes (KDIGO), cyclophosphamide and glucocorticoids were the first-line induction treatments for AAV patients, and the high-dose methylprednisolone pulses were commonly used at the initiation of therapy [9]. As we reported earlier, intravenous methylprednisolone pulses could lead to the improvement of long-term prognosis and a trend to reduce mortality in AAGN patients [10]. Although the prognosis for AAV has greatly improved, the mortality rate remains high compared with the general population [11]. Most deaths occurred within the first year after diagnosis, especially in six months, particularly for patients with severe renal failure and age over 65 [11, 12]. It has been reported that an older age at diagnosis was associated with high frequency of renal involvement and worse prognosis [13, 14].

Although several studies have been reported [12, 15], the clinical manifestations and prognosis of AAGN patients may vary with different geographic distribution and ethnicity. In this retrospective study, we aimed to identify the risk factors for early all-cause mortality and ESRD within six months since the diagnosis of AAV in Chinese AAGN patients, to improve the understanding of the disease and strengthen the early diagnosis and treatment.

Materials and methods

Patient selection

We retrospectively investigated 350 AAGN patients who were admitted in the First Affiliated Hospital of Zhejiang University School of Medicine from January 2004 to June 2017. All patients were ethic Chinese. The vast majority of patients came from Zhejiang Province and surrounding provinces. The present research was in compliance with the Helsinki Declaration and approved by the ethics committee of our hospital. Disease diagnosis complied with the Chapel Hill Consensus Conference criteria for AAV [1]. Inclusion criteria for our study were: (1) positive serology for MPO‑ANCA or PR3‑ANCA; (2) renal involvement according to renal pathological or clinical presentation as acute progressive glomerulonephritis with proteinuria (> 300 mg/day) and/or haematuria (> 3/HPF or > 25/ul); (3) for negative serology of MPO‑ANCA or PR3‑ANCA, there is histological evidence of vasculitis. We excluded the patients with secondary vasculitis, including lupus nephritis, or propylthiouracil-induced AAV or other connective tissue diseases. All the patients were informed consents.

Clinical and laboratory data

All clinical and laboratory data were collected retrospectively at diagnosis and during the follow-up period, including age, gender, smoking history, routine blood analysis, serum albumin and globulin, liver enzymes, initial serum creatinine, estimated glomerular filtration rate (eGFR), 24-h urine protein excretion, urinary sediment red blood cell count, erythrocyte sedimentation rate (ESR), C‑reactive protein (CRP), serum ANCAs. Disease activity was scored using the Birmingham Vasculitis Activity Score (BVAS) [16]. Renal pathological data of 153 cases were evaluated using direct immunofluorescence, light and electron microscopy.

Comorbidities

Comorbidities were identified at the time of admission and included hypertension, diabetes mellitus, heart disease, malignancy, infection, anemia. Hypertension was defined as BP ≥ 140/90 mmHg, or taking antihypertensive medication. Diabetes mellitus was defined as having a fasting glucose ≥ 7.0 mmol/L or currently using diabetes medication. Heart disease was defined as having a history of at least one of the following diseases: congestive heart failure, coronary heart disease, cardiac arrhythmia. Malignancy were defined as those with claims information indicating C00–C97 malignant neoplasm ICD-10 codes. Infection was defined as any culture-positive infection from blood, urine or cerebrospinal fluid specimens, or as an imaging indication of infection. Pulmonary infection was diagnosed by clinical manifestations as well as lung imaging, sputum culture and other additional tests. Each lung imaging such as CT image was diagnosed by at least two radiologists. We analyzed all the clinical data before we concluded if the patient was infected, including fever, hemoptysis and other symptoms, the relationship between the time of using immunosuppressants and the occurrence of pulmonary symptoms, the results after anti-infection treatment, as well as the results of lung imaging, sputum culture and other additional tests. According to the World Health Organization (WHO), anemia was defined as hemoglobin ≤ 130 (male) or 120 (female) g/L.

Treatment

The immunosuppressive therapies included prednisone (1 mg/kg per day, n = 124) or prednisone (0.6–0.8 mg/kg per day) combined with intravenous cyclophosphamide (IVC, 0.75–1.0 g/m2 in monthly pulses, n = 98) or mycophenolate mofetil (MMF, 1.0–1.5 g per day, n = 97) or azathioprine (75–100 mg per day, n = 2). One hundred and twenty-three patients received methylprednisolone pulses (500 mg/d, for 3 days) and four patients received plasma exchange.

Study endpoints

The primary endpoint of this study was all-cause mortality in a six-month follow-up, and the second endpoint was end-stage renal disease (ESRD) in a six-month follow-up. ESRD was defined as estimated glomerular filtration rate (eGFR) < 15 ml/min/1.73m2 or receiving dialysis with no recovery signs of renal function. It should be noted that if the patient died within 6 months, but did not reach ESRD, his/her renal survival time was the time from diagnosis to death, which was equal to his/her survival time, but the patient belonged to the non-ESRD group. The last follow-up point was in December 2019.

Statistical analyses

We used SPSS statistical software version 24.0 (SPSS Inc., Chicago, IL, USA) for statistical analysis, setting statistical significance at a two-sided P value < 0.05. The continuous data of normal distribution were expressed as mean ± standard and compared by t test, while the data of non-normal distribution were expressed as median with interquartile interval (IQR) and compared by nonparametric test (Mann Whitney U test). Categorical data were expressed as number and percent, and the Chi‑square test was used for comparisons. Risk factors for death or ESRD were investigated with univariate and multivariate Cox regression models, and the results were expressed as hazard ratios (HRs) (with 95% confidence interval [CI]). Patient-years at risk were calculated for each patient as time from the date of treatment initiation through the earlier of either their date of death or December 2019. Standardized mortality ratios (SMR) were calculated according to the 2010 census population in China.

Results

Demographic and clinical features

A total of 350 patients (including 156 males) with AAGN were included in the study. MPO-ANCAs were detected in 269(76.9%) patients, and PR3-ANCAs were detected in 22(6.3%) patients. Fifty-three (15.1%) patients had non-specific ANCAs. Six (1.7%) patients were ANCA double-positive. Their median follow-up time was 60.8(31.2, 84.5) months, 141(40.3%) patients died during follow-up and 40(11.4%) patients died within first six months. According to the survival in 6 months, the patients were grouped into survival group (n = 310) and death group (n = 40). Table 1 shows the demographic and clinical features of the patients in survival group and death group at the time of diagnosis. The average age of death group was older than that of survival group (65.9 ± 14.3 years vs. 60.6 ± 13.4 years, p = 0.020). Compared with the survival group, the death group showed significantly higher levels of leukocyte counts, serum globulin, alanine aminotransferase, serum creatinine, ESR, CRP and BVAS and significantly lower levels of serum albumin and eGFR. There were no statistical differences between treatments in the two groups. One hundred and fifty-three patients had kidney biopsy results (143 biopsies in survival group and 10 biopsies in death group). There were no significant differences in renal pathology including the proportion of glomerular sclerosis, crescent, tubular atrophy or interstitial fibrosis.

Table 1 Comparisons of clinical characteristics between survival group and death group
Full size table

Comorbidities

As shown in Table 2, comorbidities were more frequent in the death group, especially infection (52.5% vs. 22.3%, p < 0.001) and heart disease (20% vs. 7.1%, p = 0.013). All infections were pulmonary infection, except one skin infection in the survival group. All 89 patients with pulmonary infection had abnormal chest CT, and 26 of them had positive sputum culture. Sputum culture was mainly Candida albicans and mixed bacteria, including Acinetobacter baumannii, Staphylococcus aureus, Pseudomonas aeruginosa, Klebsiella pneumoniae and so on. However, there was no significant difference in hypertension, diabetes mellitus, malignancy and anemia between the survival group and death group.

Table 2 Comorbidities between survival group and death group
Full size table

Patient survival

Within the first 6-month follow-up, 40 patients died and the 6-month patient survival was 88.6%. The univariate analysis for risk factors affecting all-cause death showed that the older age (age ≥ 65 years, defined according to the maximum of Youden Index), leukocyte counts, serum albumin, serum globulin, ESR, CRP, BVAS, infection and heart disease were associated with the risk of death.

All of the above factors with P < 0.05 entered into the multivariate analysis. As shown in Table 3, in the multivariate analysis, age ≥ 65 years (HR = 2.245, 95%CI 1.085–4.645, P = 0.029), high leukocyte counts (HR = 1.089, 95%CI 1.015–1.168, P = 0.018), high BVAS (HR = 1.089, 95%CI 1.017–1.165, P = 0.014), infection (HR = 2.023, 95%CI 1.013–4.042, P = 0.046) and low serum albumin (HR = 0.916, 95%CI 0.845–0.992, P = 0.030) were independent risk factors for mortality.

Table 3 Multivariable Cox regression analysis for patient survival
Full size table

Table 4 shows the numbers of deaths and mortality rates in different age groups. The mortality in AAV was significantly higher than that in Chinese general population. During 1805 patient-years (PY) of follow-up, 141 patients died and SMR of 4.345 (95%CI, 4.016–4.691). The death rates increased with age, especially in patients with AAGN over 65 years of age.

Table 4 Relationship between age and death rates
Full size table

Renal survival

A total of 95 patients (43 males and 52 females) reached ESRD during the first 6 months. The renal survival rate was 72.9% at 6 months. Table 5 shows the clinical features at diagnosis between ESRD group and non-ESRD group. Patients in ESRD group had higher levels of leukocyte counts, serum creatinine, ESR, CRP, BVAS, hematuria, daily proteinuria, alanine aminotransferase and lower levels of hemoglobin, platelet, serum albumin, eGFR than those in non-ESRD group. Notably, the patients with ESRD had higher percentage of glomerular sclerosis (p = 0.040) and higher grades of interstitial fibrosis (p = 0.001) than those of non-ESRD patients.

Table 5 Comparisons of clinical characteristics between non-ESRD and ESRD groups
Full size table

The univariate analysis for risk factors of ESRD showed that high levels of leukocyte counts, hemoglobin, platelets counts, albumin, eGFR, ESR, CRP, BVAS, daily urine protein, glomerular sclerosis, interstitial fibrosis and the state of infection were associated with the risk of ESRD. These factors entered into the multivariate analysis (Table 6), which showed that high BVAS (HR = 1.198, 95%CI 1.043–1.376, P = 0.011), high daily urine protein (HR = 1.316, 95%CI 1.046–1.656, P = 0.019) and low eGFR (HR = 0.877, 95%CI 0.804–0.957, P = 0.003) were independent risk factors for ESRD.

Table 6 Multivariable Cox regression analysis for renal survival
Full size table

Discussion

Consistent with the literature [12, 17, 18], we found that most of the deaths occurred within the first six months for AAV patients. The 6-month mortality rate of 16.1% in AAV patients was reported in the France [12], and 15.2% was reported in Japan [17], whereas the rate was 11.4% in our cohort. In another Chinese study, the cumulative mortality rates at 6 months, 1 year and 2 years were 30.1%, 34.1% and 37.4%, respectively [18]. In our study, the mortality of AAGN patients was higher than the general population of the same age, the risk of mortality increased with age, particularly in patients who were ≥ 65 years old. Previous evidence suggested that older age was an independent risk factor for early death within 6 months from diagnosis [19]. Patients over 65 years of age with renal impairment are at particular risk of early death due to their higher risk of infection and reduced tolerance to immunosuppressants [20]. Therefore, it has been proposed in the literature that for people over 65, an induction regimen with less corticosteroid exposure and fixed low-dose intravenous cyclophosphamide pulses (500 mg per time, up to 6 times) reduced the incidence of serious adverse events and did not affect the remission rate compared with conventional immunosuppressive therapy [21].

Inflammation related injury are obvious in AAV patients. WBC, CRP and ESR are widely used noninvasive markers of inflammation. Neutrophils play a major role in the pathogenesis of AAV. Circulating monocytes and neutrophils are activated by ANCA [22], inducing active disease in certain animal models [23, 24]. A recent study has suggested that the neutrophil-to-lymphocyte ratio (NLR) was positively correlated with CRP in patients with MPO-AAV and it may independently predict the risk of all-cause mortality [25]. Some investigators have suggested that leukopenia due to the use of immunosuppressant increased the risk of infectious and was closely related to mortality in AAV patients [15, 26]. In our study, the leukocyte counts, ESR, CRP and BVAS in death group were significantly higher than those in the survival group, indicating the high disease activity. The incidence of infection in the death group was also significantly higher than that in the survival group. Albumin level decreases significantly in many inflammatory diseases, and inflammation may reduce albumin synthesis and increase the rate of partial breakdown of albumin, providing both mechanisms for hypoalbuminemia. It is reported that lower albumin level was associated with more severe renal impairment and higher inflammation levels in AAV patients [27]. Hypoalbuminemia has been reported to be associated with poor prognosis in many diseases including vasculitis [28,29,30]. Birmingham vasculitis activity score (BVAS) is a reliable tool for assessing disease activity and severity, showing a good correlation with disease activity [31]. However, studies on the predictive value of BVAS for survival have reported conflicting results. Some studies showed that patients with high BVAS had worse early mortality and had more frequent relapses [11, 32]. On the contrary, predictive value of BVAS was not found in several other investigations. In a Swedish study, BVAS did not predict either survival or ESRD in their cohort [33]. Among patients with MPO-ANCA disease in Japan, no association between BVAS and mortality was found during a two-year follow up [34]. In our study, BVAS was a risk factor for both early death and ESRD. The reason may be that we only investigate the patients’ survival and ESRD in 6 months. In our cohort, AAGN patients had low baseline renal function at the time of diagnosis and high BVAS indicating high disease activity and poor prognosis in short term. In addition, in terms of long-term survival, the literature has showed that cardiovascular diseases (CVD) were major causes of death [11]. In a cohort study with up to 20 years follow-up, they found that the risk for CVD was >3 fold higher and the risk for deep vein thrombosis was 6-fold higher in AAV patients than those in matched comparator subjects [35]. This may be related to the hypercoagulable state of blood in AAV patients [36]. Meanwhile, endothelial dysfunction due to inflammation may accelerate atherosclerosis and thrombosis [37, 38].

Our study had several limitations. First, this was a retrospective study with a relatively large time span of the enrolled patients, which may cause bias in the management. Second, the patients enrolled in this study were all from a single center and were limited to native Chinese individuals. Third, although no significant differences were observed between the survival group and the death group in terms of the initial therapy at diagnosis, the detailed records of the cumulative doses of corticosteroids and immunosuppressive drugs used during follow-up were not available. Forth, only 44% of kidney biopsies are performed, which may impair the power to show a correlation between kidney pathological changes with outcome.

In conclusion, we found that factors including over 65 years of age, high vasculitis activity (including leukocytosis, hypoalbuminemia, high BVAS) and infection were risk factors for mortality in the first 6 months in AAGN patients, while high BVAS, low eGFR and massive proteinuria were independent predictors for poor renal outcome. To improve the short-term prognosis of AAGN patients, early identification of the risk factors for mortality and ESRD is important.