Clinical Characteristics and Outcome of ANCA-Associated Vasculitis; Experience of A Single Reference Vasculitis Center

Objective: The aim of the study is to describe the clinical characteristics of Antineutrophil cytoplasmic antibodies-associated vasculitis and to analyze the parameters affecting the outcome. Materials and Methods: The study is a retrospective cohort study. Totally 130 patients with Antineutrophil cytoplasmic antibodies-associated vasculitis (18 years and over) who were followed up between October 2014 and October 2019 were analyzed. Demographic data, laboratory values, clinical course, thorax computed tomography findings and treatment approaches were noted from the charts of patients. Patients were divided into two groups as those with pulmonary involvement and non-pulmonary involvement. Results: We retrospectively reviewed the medical records of 130 patients with Antineutrophil cytoplasmic antibodies-associated vasculitis; 111 with granulomatosis with polyangiitis, 15 with microscopic polyangiitis, 1 with eosinophilic granulomatosis with polyangiitis, and 3 with other types of vasculitis. The ratio of having the abnormality in thoracic computed tomography was 72.2%. There were 84 cases with pulmonary involvement and 46 cases with non-pulmonary involvement. The frequency of microscopic polyangiitis was significantly higher (p=0.034) in non-pulmonary involvement cases. There were 67 cases with proteinase 3 Antineutrophil cytoplasmic antibodies and 39 cases with myeloperoxidase Antineutrophil cytoplasmic antibodies positivity. Most of the cases with proteinase 3 Antineutrophil cytoplasmic antibodies positivity were classified as granulomatosis with polyangiitis, this was statistically significant. Recovery was referenced for the outcome. Any of the variables were found statistically significant effective on outcome. Conclusions: Cases with pulmonary involvement were more than the cases without pulmonary involvement in our study. microscopic polyangiitis was significantly higher in nonpulmonary involvement cases. We studied on a large group, and these significant findings may have important implications for the investigation, pathogenesis, and treatment of Antineutrophil cytoplasmic antibodies-associated vasculitis.


INRODUCTION
Antineutrophil cytoplasmic antibodies (ANCA) associated vasculitis (AAV) is a necrotizing vasculitis of small blood vessels. It can affect multiple organ systems, but mostly upper and lower respiratory tracts, eyes, skin, and kidneys [1].
AAV can be classified based on serology, as the presence of myeloperoxidase (MPO-ANCA) and proteinase 3 (PR3-ANCA). MPO-ANCA-and PR3-ANCA-associated vasculitis is shown as genetically different patterns. MPO-ANCA is mostly seen in MPA, PR3-ANCA is commonly associated with GPA.
The radiological findings in AAV are poorly defined. Computed tomography (CT) shows lung parenchymal lesions, such as consolidation, ground glass opacities, interlobular septal thickening, appearance of honeycomb, reticular shadowing and interstitial pneumonia [5,7]. There are few studies with small sample sizes, focuses on CT findings of GPA. Also, there are some cases; 18% of MPO-ANCA-positive patients are reported as having normal thoracic CT [8][9][10][11].
Pulmonary involvement has main effect on prognosis and relapse rate; as pulmonary disease and early lung involvement is increasingly recognized in cases with raised morbidity and mortality in AAV [7,[12][13][14][15]. Also, outcome of disease especially, relapse rate, severity and system involvements differ between ANCA specificities. Before the immunosuppressive therapy, the mortality rate was 80% at one year. Whereas usage of immunosuppressive therapy decreased the mortality rates, as the recent studies reported to a 55% survival in MPA and 75% GPA, at 10 years [16,17].
As the pulmonary involvement is main prognostic factor, it is essential to clarify the characteristics of pulmonary involvement in AAV to be effective on the morbidity and mortality. Therefore, in our study we examined the cases with pulmonary involvement in a large cohort and searched for the role of ANCA seropositivity in AAV cases with pulmonary disease. The aim of this study is to evaluate clinical features and prognostic factors of AAV cases.

Patients
The study was designed as a retrospective cohort study. In this study, 130 AAV patients (aged of 18 and over) who were followed up between October 2014 and October 2019 in our center were analyzed.
Our center is one of the main referral vasculitis centers in the country. Executive committee of the Vasculitis center consists of physicians from many departments, including Rheumatology, Pulmonology, Nephrology, Radiology and Pathology. Multidisciplinary approach improves management of the patients. These patients are followed in the Rheumatology Department, then consulted with Chest Diseases Department in each follow up and are considered together. ANCA-Associated Vasculitis is defined as according to the 2012 revised International Chapel Hill Consensus Conference on the Nomenclature of Systemic Vasculitis [18]. For the classification of vasculitis, American College of Rheumatology (ACR) 1990 criteria were used [19]. The patients were classified into 2 groups according to pulmonary involvement as: cases with pulmonary involvement (PI) and the cases without pulmonary involvement (non-PI). The study was approved by the Ethics Committee (2-2017-GO 17/157). Birmingham Vasculitis Activity Score version 3 was used to assess the disease activity at the time of diagnosis [20]. Patients whose data are missing and do not come to regular checkups were excluded from the study.

Definitions
Remission was defined as the absence of clinical and laboratory evidence of vasculitis activity. Relapses were defined as recurrence of signs or new symptoms after a remission had been achieved. Uncontrolled vasculitis (worsening unresponsive to treatment) was defined as the occurrence of new manifestations or aggravation of manifestations already present despite treatment for the disease [21].

Data
The following data were collected: date/age at diagnosis; gender; comorbidities; type of AAV; initial respiratory symptoms, ANCA subtypes, organ systems involvement at disease onset and throughout the disease course; laboratory data (erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), serum creatinine, aspartate aminotransferase (AST), alanine aminotransferase (ALT), albumin, estimated glomerular filtration rate (eGFR), at diagnosis and during follow-up; radiological findings from Thorax CT, duration of illness; relapses, remission, worsening progression, treatment approaches, knowledge of deaths recorded from the hospital database. All reports of Thorax CT imaging were evaluated by radiology specialist.
Outcome was defined according to clinical findings (additional symptoms, any change of complaints), radiological data (any progress compared to former CT), and laboratory results (inflammatory markers such as sedimentation, CRP).

Statistical Methods
SPSS Windows version 21.0 package program was used for statistical analysis. Mean ± standard deviation was used for numerical variables as descriptive statistics. Number and% values were used for categorical variables. P<0.05 was considered statistically significant. Student t test and Mann Whitney U test were used to compare normally distributed features in two independent groups. Relationship analysis between variables at quantitative measurement level was analyzed by Pearson Chi-square and Fisher Exact tests. The univariate logistic regression models were conducted to specify candidate variables in multivariate logistic regression. Backward elimination was performed with those variables. The results of final logistic regression models were represented with odds ratio (OR), 95% of confidence interval and p-value.

RESULTS
We retrospectively reviewed the medical records of 130 AAV patients; 111 with GPA, 15 with MPA, 1 with EGPA, and 3 with other types of vasculitis ( Figure 1). We have excluded EGPA, because there was only 1 case with EGPA, and this would not have clear statistics. Also, 3 cases with other types of vasculitis were excluded. Analysis were made between 126 patients (111 cases with GPA and 15 with MPA).
Patients were divided into two groups as those with pulmonary involvement (PI) and nonpulmonary involvement (non-PI). There were 84 cases with PI and 46 cases with non-PI. Table 1 shows the demographic, clinical, and laboratory characteristics of the study groups. According to serologic classification, MPO positivity was found as 28.6% in PI, 32.6 % in non-PI. This was not statistically significant. The frequency of MPA was 7.1% in PI cases, 19.6% in non-PI cases, this was significantly difference (p=0.034).
The most common organ involvement was renal, in whole and in each group (Table 1). Eye involvement was seen in 11 patients in whole with the statistically higher rates in non-PI ones (p=0.011). Hearing loss Presence of initial symptoms, such as cough, dyspnea, chest pain were found statistically higher  were given in Table 3b. Reticular opacity was seen in cases with PR3 ANCA positivity with a ratio of 14.9% and 2.6% of the cases with MPO ANCA positivity, this was statistically significant (p=0.048). Also, cavitary lesion was seen in 20.9% of cases with PR3 ANCA positivity, whereas in 5.3% of the cases with MPO ANCA positivity, by the statistically significant difference (p=0.032).
According to the results of the study, there is a significant difference between cases with PR3 ANCA positivity and MPO ANCA positivity in terms of urea, creatinine, ALT, AST values. ALT and AST was higher in cases with PR3 ANCA positivity, whereas urea and creatinine were higher in cases with MPO ANCA positivity (Table 4).
There were 111 patients with GPA, 15 with MPA, 1 with EGPA. The rest of 3 cases were diagnosed as other vasculitis types. Between GPA and MPA cases presence of comorbidity, asthma, HT, DM, congestive heart failure was statistically different. In Table 5a, demographic and laboratory characteristics and in Table 5b radiologic findings and treatment modalities according to ANCA associated vasculitis subtypes were given. Table 6 shows the factors related to outcome. Before the analysis, the relationships between the variables in the study and the outcome variables were examined, and the variables that were found to be significant were included in the logistic regression model. Recovery was referenced for

DISCUSSION
In the present study, we evaluated demographic, clinic, laboratory findings, thoracic CT abnormalities and treatment modalities of 130 AAV cases. We classified them according to clinical sub-types, ANCA serology, and presence of pulmonary involvement or not.
The most common symptoms at presentation were nonspecific. The initial symptoms followed by cough, dyspnea, chest pain and hemoptysis in PI cases. Most cases had multisystem involvement; among them renal involvement was the most common one. Renal involvement was present in almost half of patients (48%). Gastrointestinal and neurologic involvements were the least ones. In comparison of the cases with PR3 ANCA positivity and MPO ANCA positivity; only difference for the other system involvements was the renal and nasal involvement. The Renal involvement was statistically higher in cases with MPO ANCA positivity. On the other hand, nasal involvement was more in cases with PR3 ANCA positivity.
The clinical pathogenesis of PR3-ANCA and MPO-ANCA is different, as it is associated more likely to involve granulomatous inflammation after an exposure to respiratory bacteria in PR3-ANCA whereas silial occupation occurs in MPO-ANCA  [22][23][24]. That's why, bronchiectasis as an indicator of chronic injury and alveolar obstruction, seen more in MPO-ANCA. Mohammad et al. reported 44% of bronchiectasis, another study as 20%, we had less than all with a rate of 7.6% [25,26]. Former studies found bronchiectasis more in patients with positive MPO-ANCA, we found similar rates of bronchiectasis in both cases; MPO and PR3 ANCA positivity [26,27]. Parallel to the reported data that nodular disease with cavitation was more common in cases with positive PR3-ANCA, our prevalence of cavitary lesion was more (almost 4 times) in cases with PR3 ANCA (20.9%) (5.3% in cases with MPO ANCA) [28].
Mohammad et al. reported 80% of pulmonary abnormalities on thoracic CT [25]. The frequency of abnormal thoracic CT findings was 72.2% in our study. This was similar with previous studies reported the abnormal CT findings in 69%-82% [11]. As in the general literature reported that, there were no major differences in CT findings between ANCA serotype, our findings were like that [25]. The most common abnormalities were nodular opacity (45.2%) and ground glass opacities (39.4%), without difference among the PR3-ANCA and MPO-ANCA positive cases. Researchers found the most common pulmonary finding as nodules with or without cavitation, with a ratio of 50%, like supported in our data [28].
Pleural effusion is a nonspecific finding that may be related to systemic inflammation, heart failure, or renal dysfunction [25]. Previous studies reported the frequency of pleural effusions as 19% in whole and 26% in cases with MPO-ANCA. In our study, pleural effusion was seen in 28.4% of cases with PR3 ANCA positivity, in 15.8% of the cases with MPO ANCA positivity. The difference may arise from the comorbidities found in previous studies.
On the contrary to literature, fibrosis was seen in almost 15-16% of our cases both with PR3 and MPO ANCA positivity. Whereas, in a multicenter report of 49 patients; fibrosis was found in an extremely high rate as 88%, in MPO-ANCA cases [29]. This marked difference can be explained by the follow up period, as it was known that pulmonary fibrosis usually was seen in late phase of disease.
As a result of basis classification of AAV, nodular opacity was related to GPA [30].  [31]. We found none of the immunosuppressive having statistically significant effects on outcome. Current studies report any difference between AAV serotypes. Mohammad et al. found better survival in cases with positive PR3-ANCA [25]. In our study, we found any survival difference between AAV serotypes.
Pulmonary involvement is one of the most frequently reported involvement as our finding [32]. We found the most common organ involvement as renal like previous reports revealed after pulmonary involvement [5,33]. In PI cases, there was no case with gastrointestinal system involvement. Ear-nose -throat involvement rate was reported less than 20% in recent studies [34]. Our results were similar.

Limitations
As the study retrospectively designed, we didn't have the chance of correlating results with invasive diagnostic methodologies such as lung biopsy or bronchoalveolar lavage.

CONCLUSION
Our study has many cases evaluated for demographic, clinical and outcome features in separated groups according to clinical and serologic classification. Cases with pulmonary involvement were more than the cases without pulmonary involvement in our study. MPA was significantly higher in non-PI cases. We studied on a large group, and these significant findings may have important implications for the investigation, pathogenesis, and treatment of AAV.