If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Robert D. and Patricia E. Kern Center for the Science of Health Care Delivery, Mayo Clinic, Rochester, MNDivision of Health Care Policy and Research, Department of Health Sciences Research, Mayo Clinic, Rochester, MN
Pleural and pericardial involvements are well recognized in eosinophilic granulomatosis with polyangiitis (EGPA) but considered rare manifestations of the other forms of antineutrophil cytoplasmic autoantibody (ANCA)-associated vasculitis (AAV).
What are the frequency and clinical characteristics of pleuritis and pericarditis in AAV?
and Methods: Using an institutional database of 1,830 patients with AAV, we analyzed clinical notes and diagnosis codes for key words related to pleuritis and pericarditis. Chart review to confirm these findings was performed.
Eighty-eight of 1,058 patients (8.3%) with granulomatosis with polyangiitis (GPA), 27 of 267 (10.1%) with microscopic polyangiitis (MPA), and 35 of 201 (17.4%) with EGPA had a manifestation of pleuritis and/or pericarditis attributable to vasculitis. There was a higher frequency of pericarditis in EGPA compared with that in the other AAVs (P < .01). There was no difference in the frequency of pleuritis in GPA, MPA, or EGPA. In the 156 patients with AAV with pleuritis and/or pericarditis, this was a presenting feature in 127 (81.4%). Overall, it was a presenting feature in 6.9% of all patients with AAV, including 6.5% with GPA, 8.6% with MPA, and 15.9% with EGPA.
Pleuritis and pericarditis occur across all the AAVs and, when present, are commonly presenting features of these diseases. Patients with EGPA have a higher proportion of pericardial involvement compared with pleural involvement, whereas this distribution is more equal in patients with GPA and MPA. Pleuritis and pericarditis are underrecognized features of AAV. All forms of AAV should be considered in the differential diagnosis when evaluating a patient with pleuritis or pericarditis.
Study Question: What are the frequency and clinical characteristics of pleuritis and pericarditis in antineutrophil cytoplasmic autoantibody (ANCA)-associated vasculitis (AAV)?
Results: Pleuritis and pericarditis occur across all the AAVs and, when present, are commonly presenting features of these diseases. Patients with eosinophilic granulomatosis with polyangiitis have a higher proportion of pericardial involvement compared with pleural involvement, whereas this distribution is more equal in patients with granulomatosis with polyangiitis and microscopic polyangiitis.
Interpretation: All forms of AAV should be considered in the differential diagnosis when evaluating a patient with pleuritis or pericarditis.
Antineutrophil cytoplasmic autoantibody (ANCA)-associated vasculitis (AAV) syndromes include granulomatosis with polyangiitis (GPA), microscopic polyangiitis (MPA), and eosinophilic granulomatosis with polyangiitis (EGPA). They are rare small-vessel vasculitides involving multiple organs.
Depending on how quickly therapy can be initiated, patients may experience clinically significant organ damage from the disease. The variability in clinical manifestations of the AAV syndromes makes the diagnosis difficult. Familiarity with the possible manifestations of these disease processes can aid in a timely diagnosis and identification of relapse so that treatment can be started as early as possible to minimize end-organ damage.
Pleural and pericardial involvement is well recognized in EGPA.
suggesting these manifestations may not be uncommon in GPA and MPA. Data on the presence of these manifestations across all AAV phenotypes is limited because previous studies were based on a small number of subjects.
The current study aims to determine the frequency and clinical characteristics of pleuritis and pericarditis in a large cohort of patients with AAV.
Materials and Methods
A database of patients with AAV evaluated at a single referral institution from January 1996 to November 2015 was created. Two complementary search strategies were applied: a prospectively collected clinical database of patients with AAV who were consecutively evaluated and a search of the electronic medical record by using terms and diagnosis codes related to AAV. The Mayo Clinic Life Sciences System and Data Discovery and Query Builder was used to identify patients from January 1996 through November 2015 who had query terms and International Classification of Diseases, Ninth Revision codes related to AAV. A total of 1,953 potential patients with AAV were identified through a search of the electronic medical record for terms related to AAV, as well as from a prospectively collected clinical database. Of these, 32 patients were excluded owing to a lack of consent for research, and 91 were eliminated after chart review because the diagnosis was not consistent with the Chapel Hill Consensus Conference 2012 definition of GPA, MPA, or EGPA.
A total of 1,830 patients with confirmed AAV seen between January 1996 and November 2015 were included in the institutional database. This database of patients with AAV was then searched to identify clinical notes and diagnosis codes with the following key words: pleuritis, pleurisy, pleural, effusion, pericarditis, thoracentesis, and pericardiocentesis. Direct chart review to confirm these findings was performed. This study was approved by the institutional review board (17-006608). Need for informed consent was waived. Each patient included in the study had consented for their medical record to be used for research.
Pleuritis was defined as identification of pathologic inflammation at pleural biopsy or a pleural effusion seen at chest radiography or CT scanning of the chest. Pericarditis was defined as pericardial inflammation diagnosed by a physician on the basis of consistent clinical features or a pericardial effusion identified at echocardiography or CT scanning of the chest. Other potential causes of pleuritis and pericarditis, including heart failure, malignancy, and infection, were sought; if found, these patients were excluded. Presenting feature was defined as being present at the time of, or before, diagnosis. Clinically relevant was defined as causing symptoms and/or requiring treatment. Severe disease was defined as disease that threatens the function of an organ, leading to potentially irreversible organ damage or life-threatening complications. Disease manifestations that would qualify a patient as having severe disease include scleritis, nervous system involvement, alveolar hemorrhage, GI involvement, heart involvement, and glomerulonephritis.
Unspecified AAV indicated a diagnosis of AAV established by serologic, clinical, and/or renal or skin biopsy support without further characterization because of limited follow-up.
Categorical variables are reported as percentages. Continuous variables are reported as mean with SD if normally distributed and median with interquartile range if not normally distributed. χ2 and Kruskal-Wallis tests were used for comparisons between groups. A P value less than .05 was considered statistically significant. To represent the degree of difference and determine which group or groups accounted for the statistical significance, we reported 95% CIs for proportions where applicable. Analyses were performed using JMP, version 14 (SAS Institute). To explore the associations of pleuritis and pericarditis with disease phenotypes further, we also performed multiple correspondence analysis and hierarchical clustering on principal components. Analyses were performed in R (R project for Statistical Computing) by using the FactoMineR and Factoshiny software packages (e-Appendix 1).
Of the 1,830 patients with AAV in the database, 156 (8.5%) were identified through search of the medical records as having a manifestation of pleuritis and/or pericarditis attributable to vasculitis (Fig 1). Demographic characteristics of this cohort were consistent with previously reported demographic characteristics of patients with AAV, including the association of ANCA type with disease phenotype (Table 1, e-Fig 1, e-Tables 1, 2).
The median follow-up after pleuritis or pericarditis manifestations was 5 years. Thirty-one patients (19.9%) had less than 1 year of follow-up after pleuritis or pericarditis manifestations. Two of these patients died of pleuritis or pericarditis manifestation within the year, with neither death attributable to pleuritis, pericarditis, or AAV. Follow-up data were not available in the remaining 29 patients. All patients with unspecified AAV had > 1 year of follow-up, contributing to the lack of further delineation of their disease. In the remaining 125 patients, the median follow-up was 6 years (interquartile range, 3.0-11.0).
Table 1Demographic Characteristics of Patients With ANCA-Associated Vasculitis With a Manifestation of Pleuritis and/or Pericarditis
Total (N = 156)
GPA(n = 88)
MPA (n = 27)
EGPA (n = 35)
Unspecified AAV (n = 6)
Age at pleuritis or pericarditis diagnosis, mean (SD), y
Age at AAV diagnosis, mean (SD), y
Female, No. (%)
Ethnicity, No. (%)
Chose not to disclose
ANCA serology, No. (%)
Positive IF with negative PR3- and MPO-ANCA testing
Positive IF without concurrent PR3- and MPO-ANCA testing
There was no significant difference in median BVAS/WG score at pleuritis and/or pericarditis presentation between the different AAV syndromes (P = .14). Five patients had an incidental pericardial effusion identified at imaging without evidence of other disease activity and, therefore, had a BVAS/WG score of 0. Nine patients who had a previous diagnosis of AAV had a BVAS/WG score of 1, with pericarditis or pleurisy as their sole manifestation of relapse. The most common concurrent organ involvement at the time of pleuritis was general manifestations (49.4% [42 of 85]). Renal involvement was present in 45.9% (39 of 85) of patients with pleuritis. Of the 85 patients with pleuritis, 36 (42.4%) had no other concurrent pulmonary manifestations of AAV. The remaining patients had pulmonary nodules (24.7% [21 of 85]), pulmonary infiltrates (18.8% [16 of 85]), diffuse alveolar hemorrhage (9.4% [eight of 85]), or pulmonary nodules with diffuse alveolar hemorrhage (4.7% [four of 85]).
Pericarditis frequently occurred concomitantly with general manifestations (29.8% [28 of 94]) and with pulmonary (45.7% [43 of 94]) and nervous system (23.4% [22 of 94]) involvement. One patient had concurrent myocarditis and one patient had heart block at the time of pericarditis manifestation. Concurrent renal involvement occurred in 23.4% (22 of 94) of patients with pericarditis and consisted of hematuria (5.3% [five of 94]) and acute kidney injury (18.1% [17 of 94]). When pleuritis and pericarditis occurred together, multisystem organ involvement was common.
Pleuritis and/or pericarditis were more likely to occur in patients with EGPA compared with those with GPA or unspecified AAV. This difference was attributable to a higher frequency of pericarditis in EGPA compared with the other AAV subtypes (Table 2). Pleuritis was least likely to occur in patients with unspecified AAV. There were no other differences between groups in the frequency of pleuritis (Table 2). Patients with EGPA had a higher proportion of pericardial involvement compared with pleural involvement. The proportion of pericarditis and pleuritis was more equally distributed in patients with GPA and MPA. The relationships between pleuritis and/or pericarditis presentation and main clinicopathologic diagnosis (GPA, MPA, and EGPA) were also confirmed with the multiple correspondence analysis and hierarchical clustering on principal components analyses performed (e-Figs 1-3, e-Tables 1, 2).
Table 2Characteristics of Pleuritis and Pericarditis in ANCA-Associated Vasculitis
Pleuritis and/or pericarditis were clinically relevant in most cases (71.2% [111 of 156]). When pleuritis and/or pericarditis were present in AAV, they frequently were presenting features of the disease. Pleuritis and/or pericarditis were more likely to be a presenting feature in EGPA than in GPA and unspecified AAV (Table 2). Pleuritis and/or pericarditis occurred greater than 1 month before AAV diagnosis in 69 patients, with a median time to diagnosis of 8 months (Table 2). Most (87.4% [111 of 127]) of the patients in whom pleuritis and/or pericarditis was a presenting feature had a single episode of pleuritis and/or pericarditis before the AAV diagnosis. Seven patients had two episodes, two patients had three episodes, and five patients had four or more episodes of pleuritis and/or pericarditis before the AAV diagnosis. There were 13 patients in whom pleuritis or pericarditis manifestations were present in isolation before AAV diagnosis. Other symptoms later developed that prompted the AAV diagnosis. The clinical course for these patients is described in Table 3.
Table 3Patients With ANCA-Associated Vasculitis With Initial Manifestation of Pleuritis and/or Pericarditis
Demographic Characteristic at Pleuritis and/or Pericarditis Presentation
Pleuritis or Pericarditis Manifestation
Diagnosis and Serology
Time From Manifestation to AAV Diagnosis, d
How AAV Diagnosis Established
Treatment of Pleuritis and/or Pericarditis Before AAV Diagnosis
Treatment of AAV
Clinical Course After AAV Treatment Initiated
Recurrent pleural effusion for 1 y, then inflammatory arthritis, constitutional symptoms, rash, subglottic stenosis, and DAH, prompting AAV diagnosis
No AAV or pleuritis relapse
Recurrent pericarditis and pleuritis
Recurrent pericarditis prompted further investigation that revealed elevated creatinine levels. AAV diagnosis was established with renal biopsy.
No AAV or pleuritis or pericarditis relapse Long-term use of RTX because of CLL
Pericarditis, then 2 y later he experienced a seizure. Brain imaging demonstrated a mass lesion and lung nodules. VATS lung biopsy was consistent with GPA.
Limited follow-up after diagnosis
Two episodes of pericarditis, then 8 y after initial pericarditis he developed joint pain, prompting evaluation that revealed microscopic hematuria. AAV diagnosis was established with renal biopsy.
No AAV or pericarditis relapse
Pericarditis, then years later he developed constitutional symptoms, arthralgias, and mononeuritis multiplex, prompting investigation and AAV diagnosis.
No AAV or pericarditis relapse
Pericarditis, then years later she developed polyarthritis, microscopic hematuria, panniculitis, blepharitis, sinusitis, and pulmonary nodules. VATS lung biopsy was consistent with GPA.
Indomethacin, GCs, MTX
GCs, RTX, CYC
No AAV or pericarditis relapse
Recurrent pleural effusions were followed by meningoencephalitis 7 mo later. Two years after that, he experienced acute kidney injury. AAV diagnosis was established with renal biopsy.
Thoracentesis, intrapleural catheter
No AAV or pleuritis relapse
Persistent pleural effusion, then she developed acute renal failure. AAV diagnosis was established with renal biopsy.
No AAV or pericarditis relapse
Pericarditis episode, then years later she developed swelling and erythema of ears that led to evaluation that demonstrated elevated creatinine levels. AAV diagnosis was established with renal biopsy.
One AAV relapse No relapse of pericarditis
Episode of pericarditis, then many years he later developed renal failure. AAV diagnosis was established with renal biopsy.
No AAV or pericarditis relapse
Pleuritic chest pain, then 5 y later she developed a rash. Ten years after initial rash, epistaxis and recurrent rash developed, leading to AAV diagnosis.
No AAV or pleuritic relapse
Two episodes of pericarditis were followed by joint pains and epistaxis that prompted evaluation that revealed elevated creatinine levels. AAV diagnosis was established with renal biopsy.
Pericarditis, then 5 y later he developed iritis. One year after iritis, joint pains and constitutional symptoms developed that prompted evaluation that revealed hematuria. AAV diagnosis was established with renal biopsy.
Pleuritis and/or pericarditis occurred after the diagnosis of AAV in 29 patients, with a median time to pleuritis and/or pericarditis manifestation of 31 months. In these patients, pleuritis and/or pericarditis occurred later in patients with GPA compared with those with MPA or unspecified AAV (Table 2). There were no other significant differences between groups.
Treatment and Prognosis
Treatment of pleuritis included immunosuppressive therapy (43.5% [37 of 85]), a combination of symptom and immunosuppressive therapy (28.2% [24 of 85]), or symptom management only (21.2% [18 of 85]). Of the patients with pleuritis, 43.5% (37 of 85) underwent thoracentesis, and 7.1% (six of 85) underwent pleural biopsy. Pleural fluid analysis was documented in 31 patients, and most of it was exudative (77.4% [24 of 31]). The cell count was available for 12 of the 43 procedures and was most frequently lymphocyte predominant (41.7% [five of 12]) or monocyte or macrophage predominant (41.7% [five of 12]). Treatment of pericarditis included immunosuppressive therapy (52.1% [49 of 94]), a combination of symptom and immunosuppressive therapy (14.9% [14 of 94]), or symptom management only (18.1% [17 of 94]). Procedural management of pericarditis occurred in 13.8% (13 of 94) of patients with pericarditis. Those who received no treatment for pleuritis or pericarditis either had no other signs of active disease (three patients) or had the pleuritis or pericarditis manifestation before the diagnosis of AAV (six patients). Most patients who underwent only symptom management of pleuritis or pericarditis had the manifestation before the diagnosis of AAV (82.9% [29 of 35]).
Of the patients who had pleuritis and/or pericarditis as a presenting feature and had 1 or more years of follow-up after their initial pleuritis and/or pericarditis manifestation (105 patients), most (92.4% [97 of 105]) did not have a recurrence of pleuritis or pericarditis after treatment of AAV was initiated. Six patients had one recurrence and the other two patients had chronic relapsing pericarditis. All 14 patients who had multiple episodes of pleuritis or pericarditis before their diagnosis were treated with symptom management only or symptom management with a short course of prednisone therapy. Only two of these 14 patients had a recurrence after AAV treatment was initiated. A combination of glucocorticoids and cyclophosphamide was the most common induction regimen for patients with GPA (52.3% [46 of 88]), MPA (55.6% [15 of 27]), and unspecified AAV (66.7% [four of six]) in this cohort. The most common therapy for EGPA was glucocorticoids alone (45.7% [16 of 35]). Twenty patients had the first manifestation of pleuritis and/or pericarditis after AAV diagnosis and had 1 or more years of follow-up. Most (85.0% [17 of 20]) of these patients had a single episode of pleuritis or pericarditis throughout the disease course.
The current study demonstrated a lower frequency of pleuritis and pericarditis among patients with AAV compared with that seen in previous studies. In the current study, 14.4% of patients with EGPA had a manifestation of pericarditis and 3% had pleuritis. Pleural and pericardial involvement in EGPA
The current study showed a higher prevalence of pericarditis in EGPA than in the other vasculitis subtypes, whereas pleuritis was similar between groups. This is consistent with other studies in which pericarditis is reported in 3.4% to 17.7% of patients with GPA
In the current study, we found pleuritis in 4.9% of patients with GPA and 6.4% of patients with MPA.
When present, pleuritis and/or pericarditis were a presenting feature of AAV in most cases (81.4%). When unrecognized as a manifestation of AAV, pleuritis and pericarditis frequently were treated in terms of symptoms and led to recurrences or further organ involvement of AAV. This morbidity potentially may be avoided if AAV is considered in the differential diagnosis of pleuritis and pericarditis and treated with immunosuppression. After treatment of AAV with immunosuppression was initiated, pleuritis and pericarditis rarely recurred.
The lower frequency of pleuritis and pericarditis in this study may be partially attributable to the large cohort of patients with AAV and, therefore, a large denominator. This was a retrospective study with limitations of documentation that may omit pertinent manifestations, including pleuritis or pericarditis. Complete follow-up data were not available in all patients. Chest imaging and echocardiograms were not completed in all patients with AAV in the present study; hence, subclinical manifestations may not have been detected, resulting in lower frequencies of pleuritis and pericarditis in this cohort compared with that in other studies in which imaging was completed in all patients with AAV.
Strengths of this study include this being, to our knowledge, the largest known cohort of patients with AAV. Use of the database allowed identification of manifestations of pleuritis or pericarditis before and after the diagnosis of AAV, permitting the determination of the temporal relationship of this manifestation within the overall disease course of individual patients. Use of the database also allowed characterization of the clinical course of pleuritis and pericarditis in AAV and the effect of treatment on these manifestations. Most pleuritis and pericarditis manifestations in the current study were clinically relevant and discovered owing to symptoms. In previous studies, most pleuritis and pericarditis manifestations were found incidentally at screening imaging.
The frequencies reported in the current study likely represent a more accurate estimation of the clinically relevant manifestations of pleuritis and pericarditis in the population with AAV.
Pleuritis and pericarditis are detected in all forms of AAV. Patients with EGPA have a higher proportion of pericardial involvement compared with pleural involvement, whereas this distribution is more equal in patients with GPA and MPA. All forms of AAV should be considered in the differential diagnosis when evaluating a patient with pleuritis or pericarditis of unknown cause. These manifestations can occasionally be the sole presenting manifestations of such systemic diseases, and alone they are not enough to affiliate the patient presentation correctly with a specific clinicopathologic entity. Recognition of this disease manifestation may result in decreased morbidity, with earlier AAV identification and treatment.
Author contributions: G. E. T. takes responsibility for the integrity of the data and the accuracy of the data analysis. G. E. T. completed data organization and analysis and writing of manuscript. M. H. B. completed data organization. G. E. T., M. C. M., B. J. B., and U. S. analyzed the data. M. H. B., M. C. M., M. B., R. C. C., A. M., M. J. K., K. J. W., and U. S. contributed substantially to the study design, data interpretation and writing of the manuscript. All authors were involved with editing the manuscript and approved the final version.