Prevalence and characteristics of bronchiectasis in ANCA-associated vasculitis: A systematic review and meta-analysis

Abstract

Objectives: This systematic review and meta-analysis aimed to investigate the prevalence of bronchiectasis (BR) in antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV), the association of BR with manifestations of AAV, and the features of BR in AAV.

Materials and methods: PubMed, EMBASE, Web of Science, and Cochrane Library were searched for publications related to AAV and BR in English from January 1, 1994, until December 7, 2022. The prevalence of BR was synthesized using random-effects models, and sources of heterogeneity were assessed by sensitivity and subgroup analyses. Odds ratios (ORs) were calculated using fixed-effects models to evaluate the association of BR with manifestations of AAV. Only qualitative synthesis was performed on the features of BR.

Results: Studies that reported on the prevalence (n=24), the association (n=6), and the features (n=8) of BR were identified. The pooled overall prevalence of BR among AAV was 19% (95% confidence interval [CI] 13-27%). The prevalence of patients with myeloperoxidase (MPO)-ANCA was significantly higher than those with proteinase 3-ANCA (28% vs. 13%, p=0.01). The female sex (OR=2.41), peripheral neuropathy (OR=4.58), MPO-ANCA (OR=4.73), and microscopic polyangiitis (OR=2.72) were associated with BR in AAV. Compared to individuals without BR, AAV-BR patients exhibited relatively lower levels of proteinuria. The diagnosis of BR could follow, be concomitant to, or precede that of AAV. However, BR usually did not respond to immunosuppressive therapy.

Conclusion: AAV with BR is a common condition with special manifestations. The association of BR with AAV may not be accidental; however, the underlying pathogenesis remains to be clarified.

Introduction

Antineutrophil cytoplasmic antibody (ANCA)- associated vasculitis (AAV) is a necrotizing vasculitis, predominantly affecting small vessels, including granulomatosis with polyangiitis (GPA), microscopic polyangiitis (MPA), and eosinophilic granulomatosis with polyangiitis (EGPA).[1] Airway involvement is common in AAV. For example, almost all patients with EGPA had asthma in one study.[2] Subglottic stenosis and endobronchial disease were identified in 10% and 6% of GPA patients, respectively.[3] Blood vessels and airways of the respiratory system closely accompany each other. When pulmonary small vessels are involved, the accompanying airways are likely to be involved as well. The chronic inflammatory cells, which are exudated from small vessels and infiltrated around their walls, also invade adjacent bronchioles on histopathology.[4,5] However, the involvement of medium-small airways in AAV is usually underrecognized.

The prevalence of bronchiectasis (BR) ranged from 0.053 to 0.566% according to epidemical data from Europe and America,[6,7] while multiple studies have shown the prevalence of BR in AAV ranged from 2 to 39.6%,[8-11] which is dramatically higher than that in the general population. In BR patients, dysregulation of neutrophil metabolic pathways led to increased neutrophil extracellular trap (NET)-mediated epithelial damage and decreased ability of bacterial clearance.[12] NETs were involved in the pathogenesis and correlated with the severity of BR.[13] In AAV patients, coculture of NETotic neutrophil with myeloid dendritic cells demonstrated induction of ANCA,[14] and ANCA induced NET release from primed neutrophil.[15] A vicious cycle of NET formation and ANCA production was regarded to be involved in the pathogenesis of AAV.[16] Therefore, it was speculated that BR was likely associated with AAV since the dysfunctional neutrophil played an important role in the pathogenesis of both.

Infection was the leading cause of death in AAV patients.[11,17,18] A previous study of patients with glomerulonephritis showed that patients with AAV and BR (AAV-BR) were four times more likely than AAV patients without BR to have pulmonary infections.[17] In a cohort of patients treated with rituximab, the presence of severe BR was associated with a six-fold increased risk of respiratory tract infections.[19] Infections decreased the quality of life and increased the risk of hospitalization and mortality. Therefore, it is important to gain a better understanding of the prevalence and characteristics of AAV-BR to develop an appropriate management strategy for AAV-BR and provide a theoretical basis for future studies. The current systematic review and meta-analysis aimed to detect the prevalence of BR among patients with AAV, manifestations of AAV associated with the occurrence of BR, and the features of BR in patients with AAV to address these gaps in knowledge.

Patients and Methods

In the systemic review and meta-analysis, the primary endpoints were the prevalence of BR in patients with AAV (the prevalence study) and the association of BR with manifestations of AAV (the association study). The secondary endpoint was the features of BR in patients with AAV (the features of BR study). The study was preregistered on PROSPERO (identifier: CRD42022380039) and conducted in accordance with the guidelines of PRISMA 2020 checklist.[20]

Search strategy and selection criteria

Search strategies were built with the keywords of “ANCA-associated vasculitis” or “granulomatosis with polyangiitis” or “microscopic polyangiitis” or “eosinophilic granulomatosis with polyangiitis” and “bronchiectasis.” PubMed, EMBASE, Web of Science and Cochrane Library were queried on December 7, 2022 (Supplemental Table 1). Only articles published on January 1, 1994, or later were included since a clear nomenclature of AAV was first proposed at the Chapel Hill Consensus Conference in 1994.[21] Manual searches were performed to identify relevant articles from the reference lists of selected studies as well.

Two reviewers screened titles and abstracts of all potentially eligible publications using the following criteria: (i) original studies, (ii) cases with >5 patients, (iii) published in English, (iv) related to AAV-BR. After screening, full texts of relevant articles were assessed for inclusion by the same reviewers. Articles were excluded if they had the following features: (i) only abstracts available, (ii) focusing on special populations (e.g., the older or children), (iii) all participants with traction BR associated with interstitial lung disease (ILD). Only the most comprehensive articles with the highest quality were included if the sources of data overlapped.

Data extraction and quality assessment

Two reviewers independently extracted and cross-checked data for included articles with support from a third reviewer. In addition to the year of publication and the first author, the following data were extracted according to different purposes. For the prevalence study, AAV sample size (denominator), number of AAV-BR cases (numerator), study designs (retrospective or prospective), races (Western or Asian), research periods, screening BR tools, clinical classifications of AAV, ANCA status, and chest involvement or abnormal imaging proportions, median/mean age and female proportions were extracted. In the association study, number of AAV-BR patients (cases), number of AAV patients without BR (controls), manifestations of AAV with significant differences (p<0.1) between cases and controls were assessed; dichotomous variables reported in no less than three articles were included in meta-analysis regardless of their differences. The features of BR study included the timeline between BR and AAV diagnoses (BR preceded, was concomitant to, or followed AAV) with interval time, morphology, and severity of BR, and responsiveness to immunosuppressive treatment of BR on imaging.

The Hoy risk of bias tool was used to assess the quality of articles in the prevalence study.[22] This tool contains 10 domains of bias, each given a score of 0 or 1 for the absence or presence of bias. A summary score of 0 to 2 indicated a low risk of bias, 3 to 4 a medium risk of bias, and ≥5 a high risk of bias. The Agency for Health Care Research and Quality was used to assess the quality of articles in the association study.[23] This assessment tool has 11 criteria. Each criterion was assigned “yes” or “no” based on the information reported in each study. The formal risk of bias tool was not used in the features of BR study since only qualitative analysis was conducted. All studies were independently assessed by two researchers and checked by another researcher to resolve any disagreements.

Data synthesis and statistical analysis

In the prevalence study, we calculated pooled prevalence of AAV-BR using the logit transformation to correct for nonnormally distributed raw proportions. The incidence rate of “0” and “100%” were replaced with “1/(4N)” and “1-1/(4N)” respectively, according to Bartlett’s suggestions (N=the number of patients with AAV). A series of sensitivity analyses were conducted to test the robustness of our statistical model. First, a leave-one-out analysis was conducted to assess whether a particular article was one of the sources of heterogeneity. Second, three studies were removed due to all participants with well-defined renal involvement. Third, we limited the analyses to studies that only used chest computed tomography (CT) or high-resolution CT (HRCT) scans as tools for screening BR. In addition, subgroup analyses were also conducted according to different study designs, races, sample sizes, Hoy risk of bias, research periods, median/mean age, sex ratio, ANCA status, and clinical classifications. Chisquare tests were used to evaluate the differences between subgroups.

In the association study, the nonnormally distributed quantitative variables were first summarized with significant differences between AAV patients with and without BR. Second, dichotomous variables reported in no less than three articles were included in the meta-analysis. Odds ratios (ORs) were calculated from odds of events (the number of patients with certain manifestations) in AAV-BR patients (total number of cases) divided by odds of events in AAV patients without BR (total number of controls). In the features of BR study, only qualitative synthesis was conducted due to the small number of studies and substantial variability of findings.

Meta-analysis was conducted by random-effects models with the DerSimonianLaird method (p<0.05, I ²>50%) or fixed-effects models with Mantel-Haenszel method (p≥0.05, I ²≤50%) using Stata version 17.0 (StataCorp, College Station, TX, USA). Heterogeneity between studies was assessed by the Q statistics and the I ². For all tests, p<0.05 was considered statistically significant.

Due to the high heterogeneity shown in the meta-analysis of the prevalence study, funnel plot asymmetry and Egger’s regression tests were not used to statistically examine publication bias as conventionally suggested. Moreover, in the association study, the number of included studies was between three and five, which was not enough to make funnel plots.

Results

Study selection

Search strategies identified 420 studies. After the removal of duplicates, we screened titles and abstracts of 345 articles for eligibility, of which 37 full-text articles were reviewed, and finally, 24 unique studies were included for systemic review (Figure 1). In total, all 24 articles were included in the prevalence study,[8-11,17,18,24-41] six articles were included in the association study,[8,11,17,18,31,41] and eight articles were included in the features of BR study.[8,10,11,17,29,31,40,41]

Figure 1. PRISMA flow diagram of studies assessed for AAV with BR.
AAV: Antineutrophil cytoplasmic antibody-associated vasculitis; BR: Bronchiectasis; PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Quality assessment

In the Hoy risk of bias, nine studies presented low risks, 13 studies presented medium risks, and two studies presented high risks (Supplemental Table 2). The majority of included articles were single-center and retrospective studies with small sample sizes. Due to the nature of the studies, the most common domains of risks were “studies not representing the national population,” “sampling frame not representing the target population,” and “nonresponse bias not minimal.”

In the Agency for Health Care Research and Quality, most of the risks were derived from lacking detailed descriptions of data processing (Supplemental Table 3). All the included studies failed to describe “how confounding was assessed and/or controlled,” “how missing data were handled,” or summarize “patient response rates and completeness of data collection.”

Prevalence of BR in AAV

The data collected in each study relating to the prevalence of AAV-BR are shown in Table 1. The AAV sample size ranged from 17 to 2,035 patients. The research periods spanned nearly 40 years. The majority of studies were retrospective in design except for three prospective studies. Fourteen studies were conducted in Asian populations, and 10 studies were conducted in Western populations. Most of the studies used CT or HRCT to confirm the BR diagnosis. While some patients of Takahashi et al.,[36] Greenan et al.,[24] and Aydın Tufan et al.[38] studies were diagnosed with BR based on chest X-ray, two cases in Ren et al.’s[8] study were diagnosed with BR by surgical pathology. Seven studies included multiple clinical classifications of AAV. Seven studies focused on MPA, five focused on GPA, and three focused on EGPA. Five studies only included patients with myeloperoxidase (MPO)-ANCA-associated AAV. The proportion of patients with CT or HRCT abnormalities was 38.5 to 100% (data from 11 studies). The proportion of patients with chest or lung involvement was 39.1 to 68.7% (data from nine studies). The mean/median age of the participants was 47 to 76 years old, with 27.1 to 67.7% female participants.

Table 1. Studies reporting the prevalence of BR among patients with AAV (n=24)

Among the 24 studies, 516 AAV-BR cases were identified from a total of 4,160 patients with AAV. The pooled overall prevalence of AAV-BR in the random-effects meta-analysis with DerSimonian-Laird method was 19% (95% confidence interval [CI]: 13 to 27%). Forest plots of the included studies are shown in Figure 2. In the leave-one-out analysis, the pooled prevalence fluctuated between 19% and 22% (Supplemental Figure 1). After removing three studies in which all patients had renal involvement,[17,24,37] the prevalence slightly increased (20%, 95% CI: 13 to 28%). When the analysis was restricted to studies that only conducted CT or HRCT scans, the prevalence was still 19% (95% CI: 13 to 27%).

However, substantial unexplained heterogeneity still remained (I ²=94.17%, p<0.01) after the sensitivity analyses. Subgroup analyses were performed to investigate the sources of heterogeneity (Supplemental Figure 2). The prevalence in retrospective studies was higher than that in prospective studies (20% vs. 14%, p=0.10). The prevalence in Asian populations was higher than that in Western populations (23% vs. 15%, p=0.29; after the study of Lhote et al.[10] was excluded: 23% vs. 19%, p=0.43). No significant difference was found between the studies with different Hoy risk of bias, sample sizes, research periods, median/mean age, and sex ratio.

Patients with different ANCA statuses or classifications of AAV had distinct manifestations. The prevalence in patients with MPO-ANCA was 28% (95% CI: 21 to 36%), which was significantly higher than that in patients with proteinase 3 (PR3)-ANCA (13%, 95% CI: 8 to 21%, p=0.01, Supplemental Figure 3a). The prevalence of BR was 25% (95% CI: 19 to 31%) among MPA patients and 17% (95% CI: 11 to 24%, Supplemental Figure 3b) among GPA patients (p=0.08). And the prevalence was 23% (95% CI: 10 to 47%) in EGPA patients.

Association of BR with Manifestations of AAV

The manifestations with significant differences between AAV patients with and without BR reported in six studies are shown in Table 2. The dichotomous variables reported in no less than three studies were summarized, and ORs were calculated to analyze the association of BR with manifestations of AAV.

Table 2. Studies reporting the association of BR with manifestations of AAV (n=6)

No significant difference was found in median/mean age between patients with AAVBR and controls except for Néel et al.’s[11] study (median age: 72 years in AAV-BR vs. 58 years in controls, p=0.002). Compared to males, females exhibited a significantly higher prevalence of BR (OR=2.41, 95% CI: 1.62 to 3.59, Figure 3a). Afterward, the association of BR with different classifications of AAV and ANCA statuses was analyzed. As shown in Figures 3b, c, patients diagnosed with MPA presented a significantly higher prevalence of BR than other clinical classifications (OR=2.72, 95% CI: 1.58 to 4.66), while patients diagnosed with GPA presented a significantly lower prevalence of BR than the others (OR=0.40, 95% CI: 0.22 to 0.72). The pooled analysis revealed a positive association between MPO-ANCA and BR (OR=4.73, 95% CI: 1.70 to 13.18, Figure 3d). Further analysis of the association between PR3-ANCA and BR was omitted because of a lack of data.

Figure 3. Forest plot and meta-analysis on the association of BR with specific manifestations of AAV.
AAV: Antineutrophil cytoplasmic antibody-associated vasculitis; BR: Bronchiectasis.

The association of BR with organ-specific manifestations of AAV was also analyzed. The meta-analysis revealed that patients with nervous system involvement exhibited a significantly higher prevalence of BR (OR=1.94, 95% CI: 1.25 to 3.01, Supplemental Figure 4a). Moreover, patients with peripheral neuropathy demonstrated a higher prevalence of BR (OR=4.58, 95% CI: 1.68 to 12.48) than those with central nervous system involvement (p=0.06, Supplemental Figure 4b). The results also indicated no significant difference in the pooled analyses of the associations of BR with cutaneous, mucous/eye, ear/nose/throat, cardiac, and renal involvement. Although the proportion of renal involvement was not associated with BR, the studies of Ren et al.,[8] Luo et al.,[17] and Néel et al.[11] indicated that AAV-BR patients exhibited relatively lower levels of proteinuria than patients without BR.

Features of BR in AAV

The features of BR in AAV patients reported in eight studies are summarized in Supplemental Table 4. Although the majority of the diagnosis of BR was concomitant to or followed that of AAV (56.3 to 88.5%), 13.6 to 43.8% of the diagnosis of BR still preceded AAV with a median interval time of 8.5 to 16 years. In addition, Ren et al.[8] and Lhote et al.[10] showed that the proportions of patients with nervous system involvement and MPO-ANCA were higher in those diagnosed with BR before the onset of AAV.

The included studies used various tools to assess BR severity on imaging, including FACED, Bhalla, Smith and modified Reiff score. Ren et al.[8] showed that patients diagnosed with BR before AAV presented an increased total Smith score, representing more severe BR. Luo et al.[17] reported that the modified Reiff score was positively correlated with MPO-ANCA titers.

Cylindrical type was the most common pattern of BR in AAV patients (55 to 92.6%), while varicoid and cystic types were rare. Four studies reevaluated CT or HRCT scans after immunosuppressive therapy, showing that most BR lesions remained stable or deteriorated.

Discussion

This systemic review summarized the current evidence on the prevalence and characteristics of AAV-BR in 24 studies. The results showed a pooled overall prevalence of BR among patients with AAV of 19%, indicating that BR was not a rare respiratory complication of AAV. The prevalence of BR was 25% in patients with MPA and 28% in patients with MPO-ANCA. The female sex, peripheral neuropathy, relatively lower levels of proteinuria, positive MPO-ANCA in serum, and diagnosis of MPA were associated with the occurrence of BR in patients with AAV. The diagnosis of BR could follow, be concomitant to, or precede that of AAV. However, BR of AAV patients usually did not respond to immunosuppressive therapy.

The most striking and clinically relevant finding was that BR was highly prevalent in the AAV population, emphasizing that the condition of AAV-BR should be recognized and fully studied. It is well-known that BR is a common extra-articular feature of rheumatoid arthritis. A meta-analysis has shown that the prevalence of BR among rheumatoid arthritis was 18.7%.[42] Our study also obtained the prevalence of BR among AAV (19%), which was similar to that among rheumatoid arthritis, indicating that AAV-BR was not a rare condition. Compared to more fatal complications (e.g., alveolar hemorrhage, ILD, and other vital organ damages) caused by AAV, BR is more likely to be overlooked. However, BR dramatically increased the risk of respiratory tract infection,[17,19] which should be taken into consideration during long-term immunosuppressive treatment. Given that ILD was common in AAV, the coexistence of ILD and BR should be considered. However, ILD-related BR was usually secondary to the surrounding architectural distortion. After excluding fibrosis-related BR, a few included studies showed low incidence rates of BR in AAV patients with ILD.[10,17,35]

The prevalence of AAV-BR ranged from 2.1 to 42.9% across all included studies.[8,10,11,30,31] The incidence rate of 2.1% of AAV-BR was likely to be underestimated in Lhote et al.’s[10] study since HRCT scans were not mandatory at enrollment in the French Vasculitis Study Group database of their study. Although different screening BR tools did not affect the robustness of the statistical model in sensitivity analysis, it is worth noting that the variable response rates (the percentage of AAV patients screened for BR in the real-world cohorts) of included studies were probably the main source of internal bias, as presented by the domain of Hoy risk of bias (“nonresponse bias not minimal”).[22] In the retrospective studies included in this study, AAV patients suspected of pulmonary involvement were more likely to be screened with chest imaging, while patients without chest imaging were excluded, resulting in the high prevalence of BR herein. In the prospective studies, BR was detected in all AAV patients using CT or HRCT scans without any exception. The number of participants with AAV (denominator) represented the real-world sample size. Nevertheless, the pooled prevalence in prospective studies (14%) still deserves to be considered.

Our meta-analysis showed that the female sex, peripheral neuropathy, and relatively lower levels of proteinuria were related to BR in patients with AAV. The female ratio was higher than the male ratio in 63.6% of studies on the prevalence of AAV-BR (Table 1), which was consistent with the epidemiological data of BR in the general populations.[6,7] Peripheral neuropathy and glomerulonephritis are typical organ-specific manifestations of small vessel vasculitis and are closely associated with positive ANCA.[2,43] There was no statistical significance in the proportion of renal involvement between AAV patients with and without BR. However, the severity of renal injury of AAV-BR patients was probably milder than that of AAV patients without BR, which was reflected by the differences in levels of proteinuria rather than serum creatine or glomerular filtration rate.[8,11,17] The underlying mechanism remains to be clarified.

Patients with MPO-ANCA or MPA were more likely to have BR. Previous studies have shown that a few patients initially diagnosed with idiopathic pulmonary fibrosis emerged MPOANCA-positive conversion during long-term follow-up, and some of them were even diagnosed with MPA due to ultimately developing multiorgan involvement.[44-46] The duration of the evolution from idiopathic pulmonary fibrosis to positive MPO-ANCA and, finally, MPA could last decades.[44] Similarly, the features of BR study showed a few patients with an initial diagnosis of BR developed typical manifestations of vasculitis during long-term follow-up.[8,10,11,17,41] Therefore, it was speculated that a similar process of BR exhibiting positive MPO-ANCA and resulting in MPA existed in those patients.[8,10,17] The hypothesis was also supported by other results. Patients with MPO-ANCA were more susceptible to BR (Figure 3d), and the positive rate of MPO-ANCA was higher in patients diagnosed with BR preceding AAV.[8,10] Furthermore, a similar process of evolution might exist in patients with GPA. A large case-control study in the UK showed that people with GPA were five times more likely to have a previous diagnosis of BR than controls (mainly composed of nonvasculitis patients),[47] suggesting that preexisting BR was associated with GPA. ANCA was probably associated with the development of AAV in BR patients, while the biomarker associated with BR in AAV was still not identified. In the future, it would be nice to determine a biomarker for the occurrence of BR in AAV.

Bronchiectasis can both precede the onset of AAV or occur during AAV follow-up; each scenario seems to have different underlying pathophysiological processes. More than half of the patients developed BR during AAV follow-up or at the same time, indicating that ANCA was derived from other systems outside the respiratory tracts. In patients whose diagnosis of BR preceded AAV with long interval times, chronic bronchial suppuration might contribute to the development of autoimmunity and ANCA production.[10] Neutrophilic inflammation, which plays an important role in both the pathogenesis of BR and AAV, can be induced by neutrophil dysfunction[12] with degranulation of primary granule components.[13] These granule proteins accrued to an extent that prompted the loss of self-tolerance and production of ANCA,[48] which was regarded as a trigger of AAV onset. Two other findings supported the hypothesis above as well. Patients with BR preceding AAV had more frequent MPO-ANCA,[8,10] and the severity of BR was positively correlated with MPO-ANCA titers.[17] However, since BR has a variety of causes, one should be prudent to attribute BR to AAV in patients whose diagnoses of AAV and BR were not simultaneous. Further longitudinal studies with large sample sizes should be of paramount interest to better decipher the association between the pathogenesis of AAV and BR.

This study has several limitations. First, the findings were mainly derived from single-center and retrospective studies, which might not represent a true prevalence of the national or international population. The exact prevalence of AAV-BR should be detected by inclusive screening using CT or HRCT scans in an international AAV population. Second, publication bias was not assessed with funnel plots and Egger’s regression tests since no more than five studies were included in the association study. While the heterogeneity among included studies was minimal (I ² ≤50%), the pooled ORs and ORs calculated by every single study almost directed to the same results (Figure 3). Third, the quantitative analysis was not done in the “features of BR” study due to the small sample size and variable findings. Finally, AAV can be complicated with other airway diseases other than BR, such as bronchiolitis. However, only two of the included studies reported the incidence rate of bronchiolitis (55%[40] and 6.5%[17]). Further studies are needed to confirm the results.

In conclusion, to our knowledge, this is the first systematic review and meta-analysis related to AAV-BR, showing that the prevalence of BR among patients with AAV was nearly 20%. The high prevalence of AAV-BR implies that routine chest CT or HRCT screening should be an integral aspect of clinical management. Female AAV patients and AAV patients with peripheral neuropathy, relatively lower levels of proteinuria, MPO-ANCA, and MPA diagnosis were prone to have BR. The association of BR with AAV appears not accidental, and the underlying pathogenesis needs to be further studied.

Supplemental Table 1. Search strategies in four databases to identify studies reporting the prevalence and characteristics of ANCA-associated vasculitis with bronchiectasis

Supplemental Table 2. Quality assessment of the articles included in the “prevalence” study (Hoy Risk of Bias)

Supplemental Table 3. Quality assessment of the articles included in the “association” study (Agency for Health care Research and Quality)

Supplemental Table 4. Studies reporting the features of bronchiectasis in patients with AVV (n=8)

Supplemental Figure 1. Leave-one-out meta-analysis on the prevalence of BR among patients with AAV.
AAV: Antineutrophil cytoplasmic antibody-associated vasculitis; BR: Bronchiectasis; CI: Confidence interval.

Supplemental Figure 2. Subgroup analyses on the prevalence of BR among patients with AAV. The analyses were based on different study designs, races, Hoy risk of bias, sample sizes (a), median/mean age, and sex ratio (b). Two studies were excluded in Supplemental Figure 2B due to missing data.
AAV: Antineutrophil cytoplasmic antibody-associated vasculitis; BR: Bronchiectasis; CI: Confidence interval.

Supplemental Figure 3. Forest plot and random-effects metaanalysis on the prevalence of BR among patients with AAV based on different ANCA statuses (a) and clinical classifications (b).
AAV: Antineutrophil cytoplasmic antibody-associated vasculitis; BR: Bronchiectasis; CI: Confidence interval.

Supplemental Figure 4. Forest plot and fixed-effects meta-analysis on nerve system involvement for the association of BR with AAV. (a) Includes all nervous system involvement, (b) Presents peripheral and central nervous system involvement separately.
AAV: Antineutrophil cytoplasmic antibody-associated vasculitis; BR: Bronchiectasis; CI: Confidence interval.

Citation: Gu Y, Zhang T, Zhu W, Han Y, Shi J. Prevalence and characteristics of bronchiectasis in ANCA-associated vasculitis: A systematic review and meta-analysis. Arch Rheumatol 2024;39(3):488-509. doi: 10.46497/ ArchRheumatol.2024.10352.

All authors contributed to the study conception and design. Data collection and analysis were performed: Y.G., T.Z., W.Z., Y.H.; Made tables and figures: Y.G., T.Z.; The first draft of the manuscript was written by YG and JS, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

The authors declared no conflicts of interest with respect to the authorship and/or publication of this article.

The Chinese National High Level Hospital Clinical Research Funding (2022-PUMCH-A-010).

We appreciate the efforts of all the authors whose articles were included in this review.

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

1. Jennette JC, Falk RJ, Bacon PA, Basu N, Cid MC, Ferrario F, et al. 2012 revised International Chapel Hill Consensus Conference Nomenclature of Vasculitides. Arthritis Rheum 2013;65:1-11. doi: 10.1002/art.37715.
2. Chang HC, Chou PC, Lai CY, Tsai HH. Antineutrophil cytoplasmic antibodies and organ-specific manifestations in eosinophilic granulomatosis with polyangiitis: A systematic review and meta-analysis. J Allergy Clin Immunol Pract 2021;9:445-52.e6. doi: 10.1016/j.jaip.2020.07.038.
3. Quinn KA, Gelbard A, Sibley C, Sirajuddin A, Ferrada MA, Chen M, et al. Subglottic stenosis and endobronchial disease in granulomatosis with polyangiitis. Rheumatology (Oxford) 2019;58:2203- 11. doi: 10.1093/rheumatology/kez217.
4. Chino H, Hagiwara E, Kitamura H, Baba T, Yamakawa H, Takemura T, et al. Myeloperoxidase anti-neutrophil cytoplasmic antibody-positive interstitial pneumonia associated with granulomatosis with polyangiitis diagnosed by surgical lung biopsy. Respiration 2016;92:348-55. doi: 10.1159/000449529.
5. Sayad E, Vogel TP, Cortes-Santiago N, Patel KR, McNeill DM, Spielberg D, et al. Lung biopsy in the diagnosis of pediatric ANCA-associated vasculitis. Pediatr Pulmonol 2021;56:145-52. doi: 10.1002/ ppul.25151.
6. Quint JK, Millett ER, Joshi M, Navaratnam V, Thomas SL, Hurst JR, et al. Changes in the incidence, prevalence and mortality of bronchiectasis in the UK from 2004 to 2013: A population-based cohort study. Eur Respir J 2016;47:186-93. doi: 10.1183/13993003.01033-2015.
7. Weycker D, Hansen GL, Seifer FD. Prevalence and incidence of noncystic fibrosis bronchiectasis among US adults in 2013. Chron Respir Dis 2017;14:377-84. doi: 10.1177/1479972317709649.
8. Ren J, Ding Y, Zhao J, Sun Y. Bronchiectasis in patients with antineutrophil cytoplasmic antibody-associated vasculitis: A case control study on clinical features and prognosis. Expert Rev Respir Med 2022;16:697-705. doi: 10.1080/17476348.2022.2088512.
9. Lin X, Lin Y, Lai Z, Wei S, Qiu M, Li J, et al. Retrospective comparison of high-resolution computed tomography of eosinophilic granulomatosis with polyangiitis with severe asthma. Ann Transl Med 2021;9:983. doi: 10.21037/atm-21-2243.
10. Lhote R, Chilles M, Groh M, Puéchal X, Guilpain P, Ackermann F, et al. Spectrum and prognosis of antineutrophil cytoplasmic antibody-associated vasculitis-related bronchiectasis: Data from 61 patients. J Rheumatol 2020;47:1522-31. doi: 10.3899/jrheum.190313.
11. Néel A, Espitia-Thibault A, Arrigoni PP, Volteau C, Rimbert M, Masseau A, et al. Bronchiectasis is highly prevalent in anti-MPO ANCA-associated vasculitis and is associated with a distinct disease presentation. Semin Arthritis Rheum 2018;48:70-6. doi: 10.1016/j. semarthrit.2017.12.002.
12. Giam YH, Shoemark A, Chalmers JD. Neutrophil dysfunction in bronchiectasis: An emerging role for immunometabolism. Eur Respir J 2021;58:2003157. doi: 10.1183/13993003.03157-2020.
13. Keir HR, Shoemark A, Dicker AJ, Perea L, Pollock J, Giam YH, et al. Neutrophil extracellular traps, disease severity, and antibiotic response in bronchiectasis: An international, observational, multicohort study. Lancet Respir Med 2021;9:873-84. doi: 10.1016/S2213- 2600(20)30504-X.
14. Frangou E, Vassilopoulos D, Boletis J, Boumpas DT. An emerging role of neutrophils and NETosis in chronic inflammation and fibrosis in systemic lupus erythematosus (SLE) and ANCA-associated vasculitides (AAV): Implications for the pathogenesis and treatment. Autoimmun Rev 2019;18:751-60. doi: 10.1016/j.autrev.2019.06.011.
15. Kessenbrock K, Krumbholz M, Schönermarck U, Back W, Gross WL, Werb Z, et al. Netting neutrophils in autoimmune small-vessel vasculitis. Nat Med 2009;15:623-5. doi: 10.1038/nm.1959.
16. Nakazawa D, Masuda S, Tomaru U, Ishizu A. Pathogenesis and therapeutic interventions for ANCAassociated vasculitis. Nat Rev Rheumatol 2019;15:91- 101. doi: 10.1038/s41584-018-0145-y.
17. Luo L, Li H, Tang J, Liao Z, Wang F, Jin Y, et al. Clinical characteristics and outcomes of MPO-ANCAassociated glomerulonephritis with bronchiectasis: A retrospective case-control study. Semin Arthritis Rheum 2022;57:152082. doi: 10.1016/j. semarthrit.2022.152082.
18. Tashiro H, Takahashi K, Tanaka M, Komiya K, Nakamura T, Kimura S, et al. Characteristics and prognosis of microscopic polyangiitis with bronchiectasis. J Thorac Dis 2017;9:303-9. doi: 10.21037/jtd.2017.02.15.
19. Kronbichler A, Kerschbaum J, Gopaluni S, Tieu J, Alberici F, Jones RB, et al. Trimethoprimsulfamethoxazole prophylaxis prevents severe/ life-threatening infections following rituximab in antineutrophil cytoplasm antibody-associated vasculitis. Ann Rheum Dis 2018;77:1440-7. doi: 10.1136/annrheumdis-2017-212861.
20. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. Int J Surg 2021;88:105906. doi: 10.1016/j.ijsu.2021.105906.
21. Jennette JC, Falk RJ, Andrassy K, Bacon PA, Churg J, Gross WL, et al. Nomenclature of systemic vasculitides. Proposal of an international consensus conference. Arthritis Rheum 1994;37:187-92. doi: 10.1002/art.1780370206.
22. Hoy D, Brooks P, Woolf A, Blyth F, March L, Bain C, et al. Assessing risk of bias in prevalence studies: Modification of an existing tool and evidence of interrater agreement. J Clin Epidemiol 2012;65:934- 9. doi: 10.1016/j.jclinepi.2011.11.014.
23. Rostom A, Dubé C, Cranney A, Saloojee N, Sy R, Garritty C, et al. Celiac disease. Evid Rep Technol Assess (Summ) 2004;104:1-6.
24. Greenan K, Vassallo D, Chinnadurai R, Ritchie J, Shepherd K, Green D, et al. Respiratory manifestations of ANCA-associated vasculitis. Clin Respir J 2018;12:57-61. doi: 10.1111/crj.12478.
25. Guneyli S, Ceylan N, Bayraktaroglu S, Gucenmez S, Aksu K, Kocacelebi K, et al. Imaging findings of pulmonary granulomatosis with polyangiitis (Wegener's granulomatosis): Lesions invading the pulmonary fissure, pleura or diaphragm mimicking malignancy. Wien Klin Wochenschr 2016;128:809- 15. doi: 10.1007/s00508-015-0747-1.
26. Kida T, Tanaka T, Yokota I, Tamagaki K, Sagawa T, Kadoya M, et al. Association between preexisting lung involvements and the risk of diffuse alveolar hemorrhage in patients with microscopic polyangiitis: A multi-center retrospective cohort study. Mod Rheumatol 2020;30:338-44. doi: 10.1080/14397595.2019.1601855.
27. Lao M, Huang M, Li C, Li H, Qiu Q, Zhan Z, et al. Infectious profile in inpatients with ANCA-associated vasculitis: A single-center retrospective study from Southern China. Clin Rheumatol 2020;39:499-507. doi: 10.1007/s10067-019-04779-9.
28. Lee KS, Kim TS, Fujimoto K, Moriya H, Watanabe H, Tateishi U, et al. Thoracic manifestation of Wegener's granulomatosis: CT findings in 30 patients. Eur Radiol 2003;13:43-51. doi: 10.1007/s00330-002- 1422-2.
29. Lohrmann C, Uhl M, Schaefer O, Ghanem N, Kotter E, Langer M. Serial high-resolution computed tomography imaging in patients with Wegener granulomatosis: Differentiation between active inflammatory and chronic fibrotic lesions. Acta Radiol 2005;46:484-91. doi: 10.1080/02841850510021733.
30. Magkanas E, Detorakis E, Nikolakopoulos I, Gourtsoyianni S, Linardakis M, Sidiropoulos P, et al. Air trapping in Wegener's granulomatosis: An additional finding on expiratory chest HRCT. Radiol Med 2011;116:858-67. doi: 10.1007/s11547-011- 0675-8.
31. Mohammad AJ, Mortensen KH, Babar J, Smith R, Jones RB, Nakagomi D, et al. Pulmonary involvement in Antineutrophil Cytoplasmic Antibodies (ANCA)- associated vasculitis: The influence of ANCA subtype. J Rheumatol 2017;44:1458-67. doi: 10.3899/ jrheum.161224.
32. Nakamoto K, Saraya T, Ogawa Y, Ishii H, Takizawa H. Comparison of findings on thoracic computed tomography with the severity and duration of bronchial asthma in patients with eosinophilic granulomatosis with polyangiitis. Respir Med 2018;139:101-5. doi: 10.1016/j.rmed.2018.05.003.
33. Ono N, Inoue Y, Miyamura T, Ueda N, Nagano S, Inoue H, et al. The association of airway comorbidities with the clinical phenotypes and outcomes of patients with antineutrophil cytoplasmic autoantibodyassociated vasculitis. J Rheumatol 2021;48:417-25. doi: 10.3899/jrheum.190373.
34. Saraya T Sr, Ogawa Y, Nakamoto K, Fujiwara M, Ishii H. Pulmonary involvement in microscopic polyangiitis: Computed tomography findings in 55 patients with analysis of risk factors for recurrence. Cureus 2022;14:e21285. doi: 10.7759/cureus.21285.
35. Suzuki A, Sakamoto S, Kurosaki A, Kurihara Y, Satoh K, Usui Y, et al. Chest high-resolution CT findings of microscopic polyangiitis: A Japanese first nationwide prospective cohort study. AJR Am J Roentgenol 2019;213:104-14. doi: 10.2214/AJR.18.20967.
36. Takahashi K, Hayashi S, Ushiyama O, Sueoka N, Fukuoka M, Nagasawa K. Development of microscopic polyangiitis in patients with chronic airway disease. Lung 2005;183:273-81. doi: 10.1007/s00408-004- 2540-1.
37. Trivioli G, Gopaluni S, Urban ML, Gianfreda D, Cassia MA, Vercelloni PG, et al. Slowly progressive anti-neutrophil cytoplasmic antibody-associated renal vasculitis: Clinico-pathological characterization and outcome. Clin Kidney J 2020;14:332-40. doi: 10.1093/ckj/sfaa139.
38. Aydın Tufan M, Tekkarışmaz N. Predictive factors of mortality in granulomatosis with polyangiitis: A singlecenter study. Arch Rheumatol 2021;36:435-44. doi: 10.46497/ArchRheumatol.2021.8594.
39. Worthy SA, Müller NL, Hansell DM, Flower CD. Churg-Strauss syndrome: The spectrum of pulmonary CT findings in 17 patients. AJR Am J Roentgenol 1998;170:297-300. doi: 10.2214/ajr.170.2.9456932.
40. Yamagata M, Ikeda K, Tsushima K, Iesato K, Abe M, Ito T, et al. Prevalence and responsiveness to treatment of lung abnormalities on chest computed tomography in patients with microscopic polyangiitis: A multicenter, longitudinal, retrospective study of one hundred fifty consecutive hospital-based Japanese patients. Arthritis Rheumatol 2016;68:713-23. doi: 10.1002/art.39475.
41. Yang S, Chai D, Li Y, Wang Y, Zhan X, Zhang L, et al. Patterns of lung diseases predict survival in patients with MPO-ANCA-associated vasculitis: A single-center retrospective study. Clin Rheumatol 2022;41:783-93. doi: 10.1007/s10067-021-05964-5.
42. Martin LW, Prisco LC, Huang W, McDermott G, Shadick NA, Doyle TJ, et al. Prevalence and risk factors of bronchiectasis in rheumatoid arthritis: A systematic review and meta-analysis. Semin Arthritis Rheum 2021;51:1067-80. doi: 10.1016/j. semarthrit.2021.08.005.
43. Fagni F, Bello F, Emmi G. Eosinophilic granulomatosis with polyangiitis: Dissecting the pathophysiology. Front Med (Lausanne) 2021;8:627776. doi: 10.3389/ fmed.2021.627776.
44. Kagiyama N, Takayanagi N, Kanauchi T, Ishiguro T, Yanagisawa T, Sugita Y. Antineutrophil cytoplasmic antibody-positive conversion and microscopic polyangiitis development in patients with idiopathic pulmonary fibrosis. BMJ Open Respir Res 2015;2:e000058. doi: 10.1136/ bmjresp-2014-000058.
45. Sebastiani M, Luppi F, Sambataro G, Castillo Villegas D, Cerri S, Tomietto P, et al. Interstitial lung disease and anti-myeloperoxidase antibodies: Not a simple association. J Clin Med 2021;10:2548. doi: 10.3390/ jcm10122548.
46. Sun X, Peng M, Zhang T, Li Z, Song L, Li M, et al. Clinical features and long-term outcomes of interstitial lung disease with anti-neutrophil cytoplasmic antibody. BMC Pulm Med 2021;21:88. doi: 10.1186/s12890- 021-01451-4.
47. Pearce FA, Lanyon PC, Watts RA, Grainge MJ, Abhishek A, Hubbard RB. Novel insights into the aetiology of granulomatosis with polyangiitis-a casecontrol study using the Clinical Practice Research Datalink. Rheumatology (Oxford) 2018;57:1002-10. doi: 10.1093/rheumatology/kex512.
48. McQuillan K, Gargoum F, Murphy MP, McElvaney OJ, McElvaney NG, Reeves EP. Targeting IgG autoantibodies for improved cytotoxicity of bactericidal permeability increasing protein in cystic fibrosis. Front Pharmacol 2020;11:1098. doi: 10.3389/ fphar.2020.01098.