SAT0245 RENAL INVOLVEMENT AT ONSET IN ANTI-NEUTROPHIL CYTOPLASMIC ANTIBODY (ANCA)-ASSOCIATED VASCULITIS: A MAJOR INDEPENDENT RISK FACTOR FOR RENAL RELAPSE

Background:

In ANCA-associated vasculitis (AAV), renal relapses are cause for concern as they are unpredictable and predictors of end-stage renal disease (ESRD).

Objectives:

We aimed to assess the frequency of major renal (MR) relapses in AAV in our cohort and identify independent predictors of the first MR relapse at diagnosis.

Methods:

We performed a retrospective monocentric observational study in our Vasculitis clinic from January 2000 to August 2019. Inclusion criteria were: 1) granulomatosis with polyangiitis (GPA), microscopic polyangiitis (MPA) and limited kidney disease (LKD) diagnosis fulfilling EMA algorithm criteria; 2) achievement of a stable remission, defined as absence of vasculitis symptoms or signs and adherence to the prednisone taper during remission-induction treatment. We excluded patients who developed ESRD before remission and those with incomplete data during the follow-up. Major renal (MR) relapses were defined as occurrence of at least one major item of renal Birmingham Vasculitis Activity Score version 3 (BVASv3).

All remitted patients were allocated in two subgroups: patients without MR relapse and patients with MR relapse. Univariate and multivariable analysis of first MR relapse predictors was performed with Fine and Gray (F&G) sub distribution hazards model to assess all competitive risks (progression to ESRD without MR relapse and death before MR relapse). Due to the relatively low frequency of events and the risk of overfitting, we performed several multivariable models with three variables, as recommended by Peduzzi e al1. The best multivariable model was selected accordingly to the Akaike information criterion (AIC).

Results:

96 (53% females) patients met the inclusion criteria: 74 GPA, 21 MPA and 1 LKD. Median age at diagnosis was 54 (44-64) years. ANCA testing was present in 94 patients, 85 were ANCA positive: 56 c-ANCA/PR3, 28 p-ANCA/MPO and 1 double positivity.

During a median follow-up (FU) of 54.5 months (29.3-96.5), we observed 19 MR relapses in 17 patients while 2 patients progressed to ESRD, 3 died without events and 76 reported no MR relapse. Density-incidence of MR relapses since remission was 3.6/100 person-year (CI 95% 2.2-5.6). Median time to first MR relapse after remission was 33 months (14-67.5).

At first MR relapse, 8 (53.3%) patients were on steroids while 10 (66.7%) were on immunosuppressant (5 azathioprine, 5 mycophenolate). In 2 cases, data about remission-maintenance treatment was not available.

MR relapses were observed only in ANCA positive patients with a significantly higher frequency of skin, kidney and nerve involvement at diagnosis (41.2% vs 17.7%, p=0.034, 94.1% vs 57.0% p=0.004, and 52.9% vs 25.3% p=0.024, respectively); while Ear, Nose and Throat (ENT) involvement was significantly lower (35.3% vs 62.0% p=0.043). Mean BVASv3 at diagnosis scored significantly higher in MR relapse group (24.1±6.2 vs 18.1±8.1. p=0.007).

At multivariable analysis with F&G model, renal involvement and induction treatment without cyclophosphamide and/or Rituximab at diagnosis were independent predictors of MR relapse (sHR 20.4 (2.6-158.2), p=0.004 and sHR 4.2 (1.5-12.0), p=0.007, respectively). Moreover, there was a trend of higher risk of MR relapse in PR3-ANCA (sHR 2.5 (0.9-7.1), p=0.091).

Conclusion:

Renal involvement at diagnosis and milder remission-induction treatment regimens resulted in a significantly higher risk of MR relapse during the FU in our cohort. PR3-ANCA specificity was not an independent predictor of MR relapse, even if we observed a trend of higher MR relapse risk with this covariate.

References:

[1]Peduzzi P et al. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol. 1996;49(12):1373-9.

Disclosure of Interests:

Mara Felicetti: None declared, Augusta Ortolan: None declared, Anna Chiara Frigo: None declared, Roberto Padoan: None declared, Michela Gasparotto: None declared, Mariagrazia Lorenzin: None declared, Andrea Doria Consultant of: GSK, Pfizer, Abbvie, Novartis, Ely Lilly, Speakers bureau: UCB pharma, GSK, Pfizer, Janssen, Abbvie, Novartis, Ely Lilly, BMS, Roberta Ramonda Speakers bureau: Novartis, Celgene, Janssen, Pfizer, Abbvie, Lilly, Franco Schiavon: None declared

SAT0243 SUBPHENOTYPES IN POLYARTERITIS NODOSA (PAN): TARGET ORGAN ASSOCIATIONS OF A WORLDWIDE COLLABORATION STUDY

Background:

There is a paucity of information on the current phenotypes, ethnic and geographic differences of PAN. A global PAN study group has been working for clinical subphenotype and GWAS studies.

Objectives:

This study is aimed to look for target organ associations in PAN.

Methods:

PAN patients fulfilling the EMA vasculitis classification algorithm were recruited. In addition to baseline characteristics, treatment and outcome data, occurrence of any of the clinical manifestations related to PAN during disease course was recorded.

Factor analysis was used to analyse target organ associations of 306 patients. Five factors were identified by factor analysis of variables sex, paediatric-onset, HBV, monogenic disease relationship, cutaneous features, musculoskeletal symptoms, constitutional symptoms and involved areas (abdominal, renal, neurologic, ENT, cardiac, pulmonary).

Results:

PAN cohort from 7 countries were used (Italy: n=59, Japan: n=39, Mexico: n=29, Slovenia: n=14, Sweden:11, TUR: n=106, UK: n=48). 306 (M/F: 171/135 and Caucasian 77.1%, Asian 13.4%, and Hispanic 9.5%) patients were included. 8 were HBV-related, and 22 of TUR patients had a monogenic form of disease (FMF n=15, DADA2 n=7). 21.8% of patients were cutaneous-only PAN patients. 48.4% of patients had radiologic, 64% had biopsy-proven PAN. Median age at disease onset was 40 (IQR 27.0-57.5) years. During a median 57 (16-120) months follow-up, 39 (13%) patients died.

Factor analysis revealed 5 factors that explained 54.1% of the original information on the matrix as follows:Factor 1,represented the association between gastrointestinal and renal involvement, male gender and negatively associated with cutaneous features;Factor 2,the association between monogenic relationship with paediatric onset disease;Factor 3,any of musculoskeletal findings with positive constitutional symptoms;Factor 4any neurologic involvement was associated with ENT and pulmonary findings;Factor 5cardiac involvement in non-HBV patients (Table).

The eigenvalues of the 5 factors were 2.034, 1.470, 1.427, 1.079 and 1.030, in decreasing order, i.e., the highest contribution to the overall variance in the matrix came from the togetherness of the 4 clinical and demographic characteristics that made up Factor 1.

Conclusion:

Target organ associations could support distinctive subphenotypes in PAN. Factor 1 seems the most severe form. Patients with FMF or DADA2 have distinct target organ associations. The jury is out to decide whether these patients should be classified as ‘vasculitis associated with probable etiology’ just as HBV-related-PAN. Factor 4 might define a different subphenotype (ANCA- medium vessel vasculitis?).

Disclosure of Interests:

Omer Karadag: None declared, Ertugrul Cagri Bolek: None declared, Shunsuke Furuta: None declared, Giacomo Emmi: None declared, ALOJZIJA HOCEVAR: None declared, Andrea Hinojosa-Azaola: None declared, Aladdin J Mohammad Speakers bureau: lecture fees from Roche and Elli Lilly Sweden, PI (GiACTA study), Serdal Ugurlu: None declared, Fatma Alibaz-Oner: None declared, Ayten Yazici: None declared, Luca Quartuccio: None declared, Enrica Bozzolo: None declared, Lorenzo Dagna Grant/research support from: Abbvie, BMS, Celgene, Janssen, MSD, Mundipharma Pharmaceuticals, Novartis, Pfizer, Roche, SG, SOBI, Consultant of: Abbvie, Amgen, Biogen, BMS, Celltrion, Novartis, Pfizer, Roche, SG, and SOBI, Giuseppe Alvise Ramirez: None declared, Luca Cantarini: None declared, Gina Gregorini: None declared, Jeannin Guido: None declared, Sara Monti: None declared, Eduardo Martin-Nares: None declared, Franco Schiavon: None declared, Roberto Padoan: None declared, Hajime Kono: None declared, Augusto Vaglio: None declared, Saadettin Kiliçkap: None declared, Ali İhsan Ertenli: None declared, Haner Direskeneli: None declared, Seza Özen Consultant of: Novartis, Pfizer, Speakers bureau: SOBI, Novartis, David Jayne Grant/research support from: ChemoCentryx, GSK, Roche/Genentech, Sanofi-Genzyme, Consultant of: Astra-Zeneca, ChemoCentryx, GSK, InflaRx, Takeda, Insmed, Chugai, Boehringer-Ingelheim

Renal involvement at baseline can predict major renal relapse in anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis

OBJECTIVES

In ANCA-associated vasculitis (AAV), renal relapses are cause of concern as they are unpredictable and predictors of end-stage renal disease (ESRD). We aimed to assess the frequency of major renal (MR) relapses in AAV and to identify independent base-line predictors.

METHODS

We performed a retrospective monocentric observational cohort study of patients affected by granulomatosis with polyangiitis (GPA), microscopic polyangiitis (MPA) and renal limited vasculitis (RLV), diagnosed from 2000 to 2019, and who achieved clinical remission defined as Birmingham Vasculitis Activity Index version 3 (BVASv3)=0 and/or clinical judgment. MR relapse was defined as the occurrence of major items of renal BVASv3. Univariate and multivariable analysis was performed with competitive risk analysis.

RESULTS

We included 96 patients: 73 GPA, 21 MPA and 2 RLV. Eighty-five (90%) patients were ANCA-positive: 56 c-ANCA/PR3, 28 p-ANCA/MPO and 1 double positive. During the follow-up, 17/96 patients developed at least one MR relapse, 2/96 progressed to ESRD and 3/96 died without events; 74 did not develop MR relapse. Patients with MR relapse were all ANCA positive and had higher frequency of skin (p=0.034), kidney (p=0.004) and nervous system (p=0.024) involvement and lower fre¬quency of ear, nose and throat (ENT) manifestations (p=0.043). At multivariable analysis, renal involvement at baseline (sHR 20.4, 95% confidence interval (95% CI) 2.6-158.2, p=0.004) and remission-induction treatment without cyclophosphamide and/or rituximab (sHR 4.2, 95% CI 1.5-12.0, p=0.007) were independent predictors of MR relapses.

CONCLUSIONS

Baseline renal involvement predicts MR relapse in AAV while intense initial treatment seems to be protective.

Rituximab-associated hypogammaglobulinaemia in ANCA-associated vasculitis and connective tissue diseases: a longitudinal observational study

OBJECTIVES

The burden of hypogammaglobulinaemia following rituximab (RTX) treatment in rheumatic diseases has not been fully elucidated yet. Our aim was to evaluate the frequency and predictors of hypogammaglobulinaemia in patients affected by ANCA-associated vasculitis (AAV) and connective tissue diseases (CTD).

METHODS

We retrospectively reviewed prospectively collected data of patients receiving RTX. Immunoglobulins (Ig) levels and lymphocyte subsets were recorded at RTX administration and 3-6 months later. We assessed frequency of hypogammaglobulinaemia (serum IgG<6 g/L) and its related events. Univariate and multivariable analysis were performed using SPSS 20.0 package.

RESULTS

Sixty-eight patients (30 AAV, 25 systemic lupus erythematosus, 9 systemic sclerosis and 4 idiopathic inflammatory myopathies) were treated with RTX (95 infusions, median 2 [2-6]). Following RTX, IgG<6 g/L were observed in 15/68 patients (15.8%), IgM<0.4 g/L in 28/68 (41%) and IgA<0.7 g/L in 7/68 (10.2%). Hypogammaglobulinaemia was more common in patients with AAV (p=0.008), short disease duration (p=0.001), low IgG levels at baseline (p=0.008), high cyclophosphamide exposure (p=0.018), high daily and cumulative prednisone dosage (p=0.001 and p=0.006). At multivariate analysis, cumulative cyclophosphamide dosage (OR 1.1 [1.0-1.3] p=0.045), daily prednisone intake >15mg (OR 9.5 [2.2-41.7] p=0.03) and IgG levels before RTX (OR 0.74 [0.59-0.93] p=0.009) were independent predictors of hypogammaglobulinaemia. Five patients experienced severe infections within 12 months, more frequently in those with IgG<6 g/L (26.7% vs 1.9%, p=0.007).

CONCLUSIONS

Hypogammaglobulinaemia following RTX is uncommon in AAV and CTD and is more likely in patients with high glucocorticoids and cyclophosphamide exposure and low IgG levels at baseline.

Alveolar haemorrhage in ANCA-associated vasculitis: Long-term outcome and mortality predictors

INTRODUCTION

Alveolar haemorrhage (AH) is considered an important cause of morbidity and early mortality in anti-neutrophil cytoplasmic antibodies (ANCA)-associated vasculitides (AAV).

OBJECTIVES

The aim of this study was to identify predictors of outcome in patients with AH-AAV and to evaluate outcome and causes of death in this subset.

MATERIALS AND METHODS

A multicenter retrospective study was conducted in 29 Italian Centers. Clinicians were asked to recruit all patients diagnosed with AAV-associated AH during the last 10 years, from 2007 to 2016. Univariate and multivariable analysis were performed.

RESULTS

One-hundred and six patients were included (median age at onset of 55 years [IQR 42-67]). The majority were ANCA-positive (PR3 57.1%, MPO 33.7%) and 72.6% had also renal involvement. At presentation, anaemia was shown in 97 (92.4%) patients, hemoptysis in 54 (51.9%), respiratory failure in 68 (66.7%), of whom 48 (70.6%), requiring respiratory support. At the end of the 37 months [IQR 13-77] follow-up, 19/106 (17.9%) patients were dead. The main causes of death were active disease and infections. By stepwise regression analysis, age >65 years (HR 3.66 [95% CI 1.4-9.51], p = 0.008) and the need for respiratory support (HR 4.58 [95% CI 1.51-13.87], p = 0.007) at AH onset were confirmed to be predictive of mortality.

CONCLUSIONS

Predictors of outcome in AAV-AH were determined. Factors related to the patient's performance status and the severity of the lung involvement strongly influenced the outcome. Balancing harms and benefits for the individual patient in induction and maintenance treatment strategies is crucial.

Fully integrated 18F-FDG PET/MR in large vessel vasculitis

BACKGROUND

To evaluate the usefulness of [18F] fluorodeoxyglucose (FDG) positron emission tomography (PET)/magnetic resonance (MR) in large vessels vasculitis (LVV) patients.

METHODS

We performed an observational retrospective study based on our records. Images were acquired on a PET/MR scanner using 18F-FDG-PET whole body imaging. For each PET scan, a qualitative analysis and a semi-quantitative measure using the maximum of the standardized uptake value (SUVMax) were performed. SUVMax measurements normalized to the liver uptake were categorized using a grading scale. Vessel's wall thickness (WT) was measured at five fixed points (inferior margin of T5, T9, T12, L3, thickest area-max WT).

RESULTS

23 LVV patients were included, 56.5% giant cells arteritis, 34.8% Takayasu's arteritis and 8.7% isolated aortitis, all Caucasian, mostly females (82%). We considered 32 PET scans for the LVV group (from min. 1 to max. 3 scans/patient) mainly during follow-up (29/32 scans), and 23 PET scans from a control group of non-metastatic malignancies patients. We found higher SUVMax compared to controls, in all sites, irrespective of clinical disease activity. Mean WT resulted higher in patients than in controls but was not correlated to SUVMax. Mean WT positively correlated with age in both cohorts, inversely correlated to disease duration, while no correlation with SUVMax was observed. The concordance between clinically active disease and PET hypermetabolism was poor (K Cohen 0.33).

CONCLUSIONS

PET/MR is a safe imaging technique capable of detecting inflammation in aortic wall. Low radiological exposure of PET/MR should be considered especially in young women receiving follow-up studies.

No Differences in Nasal Tissue Inflammatory Cells and Adhesion Molecules (iCAM-1 and vCAM-1) Based on the Comparison of EGPA With Eosinophilic Chronic Sinusitis With Polyposis

Background

An example of aggressive eosinophilic polyposis can be found in eosinophilic granulomatosis with polyangiitis (EGPA). Intercellular adhesion molecule-1 (iCAM-1) and vascular cell adhesion molecule-1 (vCAM-1) play a part in mediating the recruitment and adhesion of leukocytes to the vessel wall, and their blood-to-tissue migration under inflammatory conditions.

Objective

This prospective study compared 3 groups—patients with a definite diagnosis EGPA, non-EGPA patients with phenotypic features suggestive of EGPA, and patients with non-eosinophilic nasal polyposis (controls)—in terms of nasal tissue histology, iCAM-1 and vCAM-1 expression, and blood inflammatory cells.

Methods

A total of 58 adults underwent sinus surgery (13 patients with EGPA, 23 suspected of having EGPA, and 22 controls).

Results

Mean tissue eosinophil counts were significantly higher in EGPA patients and suspected cases of EGPA than in controls. Although iCAM-1 and vCAM-1 were diffusely expressed in sinonasal tissues, they did not differently stain EGPA, eosinophilic-type and non-eosinophilic polyposis. Blood basophil and eosinophil levels were high in both the EGPA and the suspected EGPA groups. Intergroup differences were found for eosinophils but not for basophils.

Conclusions

We do not have yet blood or tissue markers able to differentiate the early phase of EGPA from chronic rhinosinusitis with nasal polyps. Further investigations are mandatory considering EGPA patients at their initial diagnosis and before any treatment, in terms of nasal histology and blood inflammatory cells, to identify markers characterizing sinonasal mucosa inflammation and useful for an early diagnosis of EGPA.

Nailfold avascular score and coronary microvascular dysfunction in systemic sclerosis: A newsworthy association

BACKGROUND AND AIMS

We aimed to assess the relationship between nailfold videocapillaroscopy (NVC) abnormalities and coronary flow reserve (CFR), a marker of coronary microvascular dysfunction (CMD) in patients with systemic sclerosis (SSc).

METHODS

We studied 39 SSc patients (33 females, mean ± SD age 54 ± 12 years, median disease duration 11 years, range 6-22) and 22 controls (matched for age and sex) without any evidence of cardiovascular disease. Clinical assessment was performed by modified Rodnan skin score (mRss) and EUSTAR score. Coronary flow velocities in the left anterior descending coronary artery were measured by transthoracic echocardiography. Average peak flow velocities, CFR and microvascular resistance at baseline (BMR) and in hyperaemic (HMR) condition were assessed. CFR ≤2.5 was considered marker of CMD. Six NVC-abnormalities were evaluated by a semi quantitative scoring system: enlarged and giant capillaries (diameter > 20 μm and >50 μm, respectively), hemorrhages, disarray, capillary ramifications and loss of capillaries (avascular score). Statistic was performed using SPSS.

RESULTS

CFR was lower in SSc patients than in controls (2.6 ± 0.5 vs 3.3 ± 0.5). CMD was detected in 24 patients (61.5%) vs 0 controls (p < .0001). CFR was inversely correlated with NVC-avascular score (rho =-0.750, p < .0001). Avascular and capillary ramifications scores (p = .001 and p = .03, respectively), mRss (p = .003) and EUSTAR score (p = .01) were higher in patients with CMD than in those without. At multivariable analysis, avascular score was independently associated with CMD (p = .01). HMR was directly correlated with avascular score (rho = 0.416, p = .008).

CONCLUSIONS

In our SSc patients NVC-avascular score was associated with CMD which seems to be the result of a structural microvascular remodeling.

Update on Vertigo in Autoimmune Disorders, from Diagnosis to Treatment

The prevalence of autoimmune diseases has been increasing over the last 20 years. The clinical presentation of this large and heterogeneous group of disorders depends on whether the involvement is organ-specific or non-organ-specific. Dizziness, vertigo, and disequilibrium are common symptoms reported by patients with vestibulocochlear involvement. The association of vertigo and autoimmune diseases has been largely documented, suggesting that autoimmune disorders could be overrepresented in patients with vertigo in comparison to the general population. The aim of this review is to present the recent literature findings in the field of autoimmune-mediated diseases with cochleovestibular involvement, focusing on the clinical presentation, diagnosis, and treatment of immune-mediated inner ear diseases including autoimmune inner ear disease (AIED), Meniere's disease, and bilateral vestibulopathy, as well as of systemic autoimmune diseases with audiovestibular disorders, namely, Behçet's disease, Cogan's syndrome, sarcoidosis, autoimmune thyroid disease, Vogt-Koyanagi-Harada syndrome, relapsing polychondritis, systemic lupus erythematosus, antiphospholipid syndrome, IgG4-related disease, and ANCA-associated vasculitides.

Unusual Findings in Synovial Fluid Analysis: A Review

Synovial fluid analysis is one of the most useful laboratory test in the diagnosis of joint diseases. It allows to determine the degree of synovial inflammation, the presence of pathogenic crystals and microorganisms, and to evaluate the effect of pharmacological treatments as well as the progression of the disease. Synovial fluid therefore represents a precious substrate able to give valuable information from both the clinical and the research points of view.In this educational review we present and discuss some unusual findings, at times associated with rare rheumatic conditions, observed while routine synovial fluid examination is performed. These findings can be highlighted under ordinary or polarized light using simple wet preparations and supravital staining.

The prognostic significance of the Birmingham Vasculitis Activity Score (BVAS) with systemic vasculitis patients transferred to the intensive care unit (ICU)

Systemic vasculitides represent a heterogeneous group of diseases that share clinical features including respiratory distress, renal dysfunction, and neurologic disorders. These diseases may often cause life-threatening complications requiring admission to an intensive care unit (ICU). The aim of the study was to evaluate the validity and responsiveness of Birmingham Vasculitis Activity Score (BVAS) score to predict survival in patients with systemic vasculitides admitted to ICU.A retrospective study was carried out from 2004 to 2014 in 18 patients with systemic vasculitis admitted to 2 different Rheumatology divisions and transferred to ICU due to clinical worsening, with a length of stay beyond 24 hours. We found that ICU mortality was significantly associated with higher BVAS scores performed in the ward (P = 0.01) and at the admission in ICU (P = 0.01), regardless of the value of Acute Physiology And Chronic Health Evaluation (APACHE II) scores (P = 0.50). We used receiver-operator characteristic (ROC) curve analysis to evaluate the possible cutoff value for the BVAS in the ward and in ICU and we found that a BVAS > 8 in the ward and that a BVAS > 10 in ICU might be a useful tool to predict in-ICU mortality.BVAS appears to be an excellent tool for assessing ICU mortality risk of systemic vasculitides patients admitted to specialty departments. Our experience has shown that performing the assessment at admission to the ward is more important than determining the evaluation before the clinical aggravation causing the transfer to ICU.