Open Collections

UBC Faculty Research and Publications

Clinical practice variation and need for pediatric-specific treatment guidelines among rheumatologists… Westwell-Roper, Clara; Lubieniecka, Joanna M; Brown, Kelly L; Morishita, Kimberly A; Mammen, Cherry; Wagner-Weiner, Linda; Yen, Eric; Li, Suzanne C; O’Neil, Kathleen M; Lapidus, Sivia K; Brogan, Paul; Cimaz, Rolando; Cabral, David A Aug 7, 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12969_2017_Article_191.pdf [ 491.11kB ]
JSON: 52383-1.0351995.json
JSON-LD: 52383-1.0351995-ld.json
RDF/XML (Pretty): 52383-1.0351995-rdf.xml
RDF/JSON: 52383-1.0351995-rdf.json
Turtle: 52383-1.0351995-turtle.txt
N-Triples: 52383-1.0351995-rdf-ntriples.txt
Original Record: 52383-1.0351995-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessClinical practice variation and need forpediatric-specific treatment guidelinesamong rheumatologists caring for childrenwith ANCA-associated vasculitis: aninternational clinician surveyClara Westwell-Roper1, Joanna M. Lubieniecka2, Kelly L. Brown1, Kimberly A. Morishita1, Cherry Mammen1,Linda Wagner-Weiner3, Eric Yen4, Suzanne C. Li5, Kathleen M. O’Neil6, Sivia K. Lapidus7, Paul Brogan8,Rolando Cimaz9, David A. Cabral1* and for ARChiVe Investigators Network within the PedVas initiativeAbstractBackground: Because pediatric antineutrophil cytoplasmic antibody-associated vasculitis is rare, management generallyrelies on adult data. We assessed treatment practices, uptake of existing clinical assessment tools, and interest inpediatric treatment protocols among rheumatologists caring for children with granulomatosis with polyangiitis (GPA)and microscopic polyangiitis (MPA).Methods: A needs-assessment survey developed by an international working group of pediatric rheumatologists andtwo nephrologists was circulated internationally. Data were summarized with descriptive statistics. Pearson’s chi-squaretests were used in inferential univariate analyses.Results: The 209 respondents from 36 countries had collectively seen ~1600 children with GPA/MPA; 144 had seenmore than two in the preceding 5 years. Standardized and validated clinical assessment tools to score disease severity,activity, and damage were used by 59, 63, and 36%, respectively; barriers to use included lack of knowledge and limitedperceived utility. Therapy varied significantly: use of rituximab rather than cyclophosphamide was more common amongrespondents from the USA (OR = 2.7 [1.3-5.5], p = 0.0190, n = 139), those with >5 years of independent practice experience(OR = 3.8 [1.3-12.5], p = 0.0279, n = 137), and those who had seen >10 children with GPA/MPA in their careers (OR = 4.39[2.1-9.1], p = 0.0011, n = 133). Respondents who had treated >10 patients were also more likely to continue maintenancetherapy for at least 24 months (OR = 3.0 [1.4-6.4], p = 0.0161, n = 127). Ninety six percent of respondents believed in a needfor pediatric-specific treatment guidelines; 46% supported adaptation of adult guidelines while 69% favoured guidelinesproviding a limited range of treatment options to allow comparison of effectiveness through a registry.Conclusions: These data provide a rationale for developing pediatric-specific consensus treatment guidelines for GPA/MPA.While pediatric rheumatologist uptake of existing clinical tools has been limited, guideline uptake may be enhanced ifoutcomes of consensus-derived treatment options are evaluated within the framework of an international registry.Keywords: Pediatric rheumatology, Anti-neutrophil cytoplasmic antibody-associated vasculitis, Granulomatosis withpolyangiitis, Microscopic polyangiitis, Physician practice patterns, Clinical practice guidelines, Disease classification, Vasculitistreatment, Outcome assessment* Correspondence: dcabral@cw.bc.ca1Clinical Professor, Division of Rheumatology, Department of Pediatrics,University of British Columbia, BC Children’s Hospital, Room K4-119 4480 OakStreet Vancouver, Vancouver, BC V6H 3V4, CanadaFull list of author information is available at the end of the article© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (, which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver( applies to the data made available in this article, unless otherwise stated.Westwell-Roper et al. Pediatric Rheumatology  (2017) 15:61 DOI 10.1186/s12969-017-0191-zBackgroundAntineutrophil cytoplasmic antibody- (ANCA-) associ-ated vasculitis (AAV) describes the subset of vasculitidesprimarily involving small vessels: granulomatosis withpolyangiitis (GPA, formerly Wegener’s granulomatosis),microscopic polyangiitis (MPA), eosinophilic granuloma-tosis with polyangiitis (EGPA, formerly Churg-Strausssyndrome), and renal-limited pauci-immune glomerulo-nephritis [1]. Although rare in childhood, AAV carries ahigh burden of morbidity. With the use of high-dosecorticosteroids and cyclophosphamide, AAV is no longerrapidly fatal in the majority of affected children; how-ever, these treatments have significant adverse effects.Balancing the risks associated with existing therapiesagainst the damage associated with under- or over-treatment remains a challenge. In adult populations, thisbalance has been incrementally fine-tuned through clin-ical trials that rely on accurate disease sub-classificationand scoring tools to stage disease severity, activity, anddamage (Table 1). The rarity of pediatric AAV has lim-ited opportunities for pediatric clinical trials such thatmanagement decisions are largely informed by adult data[2], with potential for significant practice variation.A 2005 survey of Childhood Arthritis and Rheumatol-ogy Research Alliance (CARRA) members elicited over-whelming consensus on the need to study childhood-onset vasculitis independently from adult disease [3].ARChiVe (A Registry for Childhood Vasculitis) wasestablished in 2007 based on this imperative forpediatric-specific data collection. A concurrent initiativeof the Pediatric Rheumatology European Society (PRES)also helped generate time-of-diagnosis data to supportthe development of pediatric-specific classification cri-teria and assessment tools. More recently, PedVas(Pediatric Vasculitis Initiative [4]; https://clinicaltrials.-gov/ct2/show/NCT02006134) has allowed collection ofclinical data and biological samples to ARChiVe. Whilepediatric clinical trials remain difficult, an internationalTable 1 Definitions of clinical tools included in survey questionsClinical tool DescriptionBirmingham Vasculitis Activity Score (BVAS) Scoring tool designed to document new orworseningfeatures of clinicallyactive AAV. Items are categorized into 9 groupsby organ system [11].BVAS Version 3 2009 update to BVAS [11, 39].BVAS for Wegener’s Granulomatosis (WG) Modification of BVAS that removes features unlikelyto occur in GPA [13, 40].Pediatric Vasculitis Activity Score (PVAS) Pediatric version of BVAS preliminarily validated inchildren [12].EUVAS Severity Score Classification system for AAV based on disease extentand severity. Subgroupsinclude localized, early systemic, generalized, severe renal,and refractory disease [41].Five Factor Score Scoring tool based on factors associated with poorprognosis: cardiac symptoms,gastrointestinal involvement, renal insufficiency, proteinuria,and central nervoussystem involvement [42].Wegener’s Granulomatosis EtanerceptTrial (WGET) Severity ScoreSub-classification based on modified ACR criteria for limitedversus extensive disease [43].Disease Extent Index (DEI) Tool for scoring disease activity based on organ system, withdistinct domains from thoseincluded in BVAS [44].Physician’s Global Assessment (PGA) Physician’s global assessment of disease activity on a 10 cmvisual analogue scale.Vasculitis Damage Index (VDI) Scoring tool used to record damage due to disease or treatment.Items are categorizedinto 11 groups by organ system with binary scoring [14].Pediatric VDI (PVDI) Pediatric modification of the adult VDI; not yet formally validatedin children [14, 15].AAV Index of Damage (AVID) Tool for grading disease- or treatment-associated damage, withweighting of items basedon potential morbidity and mortality [45].Combined Damage Assessment Index Combination of items from AVID and VDI, divided into 17organ-based categories [46].Westwell-Roper et al. Pediatric Rheumatology  (2017) 15:61 Page 2 of 10registry such as ARChiVe might provide the opportunityto compare effectiveness of a limited range of standard-ized treatment options and ultimately generate evidence-based guidelines. This strategy has been pursued foruvenile dermatomyositis, another rare pediatric rheum-atic disease [5].As a first step toward developing treatment guidelinesfor pediatric AAV, we sought to better understand thediversity of beliefs and practices associated with currentcare. We conducted an international needs assessmentsurvey of physicians involved in the diagnosis and man-agement of children with GPA and MPA, with the fol-lowing aims: (1) to assess the level of communityexperience with pediatric AAV; (2) to assess uptake ofexisting classification criteria, clinical scoring tools, andtreatment guidelines; (3) to assess the extent of variationin current treatment practices; and (4) to determineinterest in and capacity for use of pediatric treatmentprotocols.MethodsA survey draft developed by the Pediatric Rheumatologygroup and a nephrologist at British Columbia Children’sHospital was finalized with input from the VasculitisWorking Groups of CARRA and PRES. The survey com-prised 47 predominantly categorical multi-choice ques-tions in three sections (Additional file 1). The firstsection addressed practice type and experience withpediatric GPA and MPA (EGPA was excluded given itsrarity in pediatrics). Respondents caring for fewer thantwo children with GPA and/or MPA over the past 5years were then given the option to exit. The secondsection queried use and knowledge of classification cri-teria and formal scoring tools for disease severity, activ-ity and damage (Table 1). Practitioners were also askedabout their use of adult treatment guidelines and per-ceived need for pediatric-specific guidelines. The thirdsection queried motivations for and barriers to participa-tion in clinical registries. Levels of assigned value wererequested using a five-point Likert scale from 1 (not im-portant) to 5 (extremely important).Members of CARRA, the Canadian Alliance ofPediatric Rheumatology Investigators (CAPRI), the Aus-tralian Pediatric Rheumatology Group, the PRES-CARRA Vasculitis Working Group, and the PediatricRheumatology Bulletin Board ( were invited by email to complete the web-basedsurvey (estimated 500 personal email invitations). Thesurvey was administered using REDCap electronic datacapture tools [6] hosted at the University of BritishColumbia (UBC) from July 24 to September 29, 2015.The survey was conducted as a quality assurance initia-tive for PedVas, which was approved by the Children’sand Women’s Research Ethics Board of UBC (H12-00894). Under Article 2.5 of the Tri Council PolicyStatement, quality assurance/improvement activities arenot subject to further institutional review.Descriptive statistics were used to quantify responsefrequencies and means. Unless otherwise specified, fre-quencies are reported relative to the total number of re-spondents completing each question. Odds ratios (OR)were determined as measures of association. Limits of95% confidence intervals of OR are reported in squarebrackets. Inferential univariate analysis with two-sidedPearson’s chi-square tests and Benjamini-Hochberg cor-rection for multiple comparisons was used to test 11 hy-potheses with a false discovery rate of 5% [7]. Correctedp-values are reported. Analysis was performed usingMicrosoft Excel Version 14.6.7 and Prism Version 5.0a(GraphPad Software Inc., La Jolla, CA).ResultsRespondent characteristicsOf 216 respondents opening the survey, 209 completed it,yielding an estimated response rate of approximately 40%.Sixty-five (31%) chose to exit as they had not cared for twoor more patients with GPA/MPA in the preceding 5 years.The 144 respondents completing the full survey practicedin 36 countries, predominantly the USA (48%), with othersfrom Canada (13%), Italy (4%), Australia (3%), Germany(4%), Turkey (4%), United Kingdom (4%), Brazil (2%), andSweden (2%). Most respondents (92%) belonged to one ormore national or international rheumatology organizations,including CARRA (56%), PRES (32%), CAPRI (10%), thePRES-CARRAVasculitis Working Group (10%), and others(18%). Approximately half of respondents had practiced forless than 10 years after formal training (21% for less than 5years and 26% for 5-10 years); among the rest, 20% eachhad practiced for 10-20 and 20-30 years and 10% for 30-40 years. The combined lifetime experience of respondentswas at least 1600 patients with GPA/MPA (some shared):47% had seen fewer than ten patients, 39% had seen 10-20,and only 14% had seen more than 20. The majority of re-spondents (61%) had seen five or fewer patients with a newdiagnosis of GPA or MPA within the past 5 years (mode 3,median 5, IQR 3-8 patients).Most respondents belonged to group practices, sharingdiagnostic and treatment decisions for all patients (46%)or managing patients independently while sharing on-call responsibilities (41%). A minority worked in a solopractice (7%) or in another practice arrangement (6%),predominantly hospital-based. For patients who had alsobeen assessed by a nephrologist, the majority of respon-dents reported making collaborative treatment decisions,30% in a combined nephrology/rheumatology clinic and37% in separate clinic settings. Some reported variationin the responsible subspecialty depending on factorssuch as who had seen the patient first (18%), while 13Westwell-Roper et al. Pediatric Rheumatology  (2017) 15:61 Page 3 of 10and 2% of respondents reported independent manage-ment by rheumatology or nephrology, respectively.Classification criteriaThere were between 129 and 144 responses (89-100%)to subsequent individual questions regarding diagnosisand management. While the majority (73%) reportedthat they always sub-classify AAV as either GPA orMPA, 26% endorsed only sometimes sub-classifying.Table 2 shows the number of respondents familiar with– and using – existing criteria for defining or classifyingAAV. Most were familiar with ACR1990 [8] andEULAR/PRINTO/PRES 2008 [9] classification criteriafor GPA, and with CHCC 1994 disease definitions [10].Serology (cANCA/PR3 versus pANCA/MPO) andEULAR/PRINTO/PRES 2008 criteria [9] were most fre-quently used for distinguishing between GPA and MPAin clinical practice. Notably, 66% of all respondentsreported using more than one set of criteria, and thoseusing informal methods considered histopathology, anti-body status, and specific organ involvement. The rationalefor sub-classification was predominantly prognostication(73%), influence on treatment choice (48%), involvementin clinical studies/trials (39%), and access to treatments(29%). All respondents felt that distinguishing betweenGPA and MPA was at least somewhat important forresearch (mode 4, median 4, IQR 3-4).Disease severity, activity, and damageThe proportions of respondents using formal tools to as-sess disease severity, activity, and damage are shown inTable 3. While most respondents used formal assess-ment tools for severity (59%) and activity (63%), only36% had ever formally assessed disease damage. The Bir-mingham Vasculitis Activity Score (BVAS) [11], thePediatric Vasculitis Activity Score (PVAS) [12], andBVAS for Wegener’s Granulomatosis [13] – toolsdesigned primarily for assessment of disease activity –were frequently used to stage severity, but tools such asthe Five Factor Score and the EUVAS and WGET sever-ity scores were used by a minority. PVAS, BVAS, andphysician’s global assessment on a 10 cm visual analoguescale (PGA) were the primary tools used by respondentsfor scoring disease activity, typically at the time of diag-nosis: 39% at only some follow-up visits, 33% at all visits,16% at prescribed times according to research protocols,and 7% at the time of diagnosis only. Most respondentsfelt formal assessment of disease activity was somewhatimportant to clinical management (mode 3, median 5,IQR 3-4), and more important to research (mode 5, me-dian 5, IQR 4-5). Most commonly used tools for scoringdisease damage include the Vasculitis Damage Index(VDI) [14] or its pediatric adaptation, PVDI [15]. Similarto scoring of disease activity, respondents assessed dis-ease damage primarily at the time of diagnosis: 35% atonly some follow-up visits, 23% at prescribed times only,and 31% at every visit. Respondent rationales for notTable 2 Familiarity with and use of vasculitis classification criteriaand disease definitions to differentiate GPA from MPA in childrenaCriteria Familiar (%) Using (%)EULAR/PRINTO/PRES2008 criteria [9]78 57ANCA (PR3/MPO) NDb 57ACR 1990 criteria [8, 19] 72 29CHCC 1994 [10] 62 32CHCC 2012 [1] 47EMA classificationalgorithm (2007) [22]18 5Pediatric Modification of EMAclassification algorithm (2012) [21]Informal criteria NDb 19Other formal criteria NDb 1aData represent percentage of respondents out of n = 144 (familiar) orn = 143 (using)bNo dataTable 3 Use of formal disease severity, activity, and damagescoring tools by survey respondents in the assessment of childrenwith GPA or MPAaParameter Scoring tool Respondents (%)Severityb PVAS [12] 41BVAS [11] 20BVAS for WG [13] 19EUVAS Severity Score [41] 11Five Factor Score [42] 8WGET Severity Score [43] 7Disease Extent Index [44] 1Other 1Never formally assess 41Activityc PVAS [12] 38Physician’s Global Assessment 29BVAS [11] 19BVAS for WG [13] 13BVAS version 3 [39] 6Other tool 0Never formally assess 37Damagec Pediatric VDI [15] 31VDI [14] 10AAV Index of Damage [45] 3Combined Damage AssessmentIndex [46]1Other 0Never formally assess 64aData represent percentage of respondents out of bn = 142, cn = 141Westwell-Roper et al. Pediatric Rheumatology  (2017) 15:61 Page 4 of 10using formal assessment tools are shown in Table 4 andincluded use of histopathology, lack of effect on manage-ment, lack of familiarity, and inconvenience.Treatment guidelinesA majority of respondents (53%) reported using a com-bination of resources to guide treatment decisions, mostcommonly EULAR/EUVAS recommendations (24%). Aminority used site-specific standardized protocols (7%);others used pediatric textbook recommendations (4%),individualized approaches according to personal inter-pretation (4%), or advice from colleagues (6%). The ma-jority of respondents (96%) believed in a need forpediatric treatment guidelines for GPA/MPA. Over half(58%) were interested in being involved in the process ofconsensus guideline development – 43% through an it-erative survey – while others felt they did not have thetime (13%) or relevant expertise (19%), or were unsure(9%). There was no association between group member-ship and the desire to participate in guideline develop-ment (OR = 1.4 [0.4, 5.1], p = 0.7206, n = 131), althoughrespondents practicing outside the USA and Canadawere more likely to want to participate than those within(OR = 4.2 [1.9, 9.0], p = 0.0011, n = 135). Most respon-dents supported consensus guidelines drafted by anexpert group (69%) and believed in the need for a lim-ited range of options to allow for comparative outcomeassessment through a clinical registry (63%). Slightly lessthan half (46%) felt modification of recommendationsfor adult disease was an acceptable method for generat-ing pediatric treatment guidelines.Treatment practicesAll respondents followed a remission-induction andremission-maintenance model, switching from inductionto maintenance therapy within 3-6 months. In adultstudies, choice of induction agent may be adjusted basedon measures of disease severity, primarily to limit use ofaggressive life-saving treatment – specifically cyclophos-phamide – that also has significant toxicity and may beunwarranted in milder disease. Two-thirds (67%) of re-spondents endorsed always using more aggressive treat-ment for patients with severe disease, while 32%reported only sometimes choosing induction therapybased on disease severity. Table 5 shows first-line induc-tion and maintenance therapies preferred by respon-dents for treatment of AAV; respondents were not askedto distinguish between GPA- and MPA-specific regi-mens. In patients with severe disease not requiring in-tensive care, cyclophosphamide (CYC) was the firstchoice of remission-induction agent by 66% of respon-dents, while 31% stated they would choose rituximab(RTX). Use of RTX rather than CYC as an inductionagent was more frequently reported by respondents fromthe USA (OR = 2.7 [1.3, 5.5], p = 0.0190, n = 139), thosewith greater than 5 years of experience (OR = 3.8 [1.3,12.5], p = 0.0279, n = 137), and those who had seenmore than 10 patients with GPA/MPA in their careers(OR = 4.39 [2.1, 9.1], p = 0.0011, n = 133).Figure 1 shows the heterogeneity among prescribingpractices for both CYC and RTX. Most respondentsused CYC for induction until clinically inactive diseaseup to a maximum of 6 months (69%), while 16% treatedTable 4 Reported rationales against use of formal assessmenttools for scoring disease severity, activity, and damage in theassessment of children with GPA or MPAaParameter Rationale against use of tool Respondents (%)Severityb Use of histopathological findings 53No effect on management 34Lack of familiarity with tools 26No value added beyond clinical judgment 24Inconvenience 16Other 5.3Activityc Lack of familiarity with tools 47Inconvenience 43No effect on management 22Lack of applicability of tools for adults 20No value added beyond clinical judgment 18Damaged Lack of familiarity with tools 58Inconvenience 32Lack of applicability of tools for adults 20No effect on management 18No value added beyond clinical judgment 12aData represent percentage of respondents out of bn = 38, cn = 51, dn = 91Table 5 Induction and maintenance agents used byrespondentsaIndication Agent Respondents(%)Induction agent in severe diseaseb Cyclophosphamide 66Rituximab 31Other 2Induction agent if not using CYCor RTX (e.g. for less severe disease)cMethotrexate 40Azathioprine 30Mycophenolate 25Other 4Maintenance therapyd Azathioprine 45Methotrexate 23Mycophenolate 18Rituximab 11Other 3aData represent percentage of respondents out of bn = 140, cn = 138, dn = 141Westwell-Roper et al. Pediatric Rheumatology  (2017) 15:61 Page 5 of 10for a defined duration between 3 and 7 months, 7% gaveonly 1-2 IV doses in conjunction with RTX, 5% contin-ued induction therapy until clinical inactivity regardlessof duration, and 2% employed some other regimen. Themost common first choice of induction therapy in chil-dren not receiving CYC or RTX was methotrexate(MTX), followed by azathioprine (AZA) and mycophe-nolate mofetil (MMF) (Table 5).The most commonly used remission-maintenancetreatment was AZA, followed by MTX, MMF, and RTX(Table 5). There was no association between mainten-ance therapy and practice location (Canada vs. USA vs.other; p = 0.2009, n = 139), years of practice (<5 vs. 5 ormore; p = 0.3042, n = 141), or total number of patientsseen with GPA/MPA (<10 vs. 10 patients or more;p = 0.8328, n = 141). The provisional choice for main-tenance therapy duration of 24 (46%), 18 (19%), 36(14%), or 12 (13%) months was associated with total pa-tient experience: 75% of respondents who had seen morethan 10 GPA/MPA patients in their careers planned forat least 24 months of therapy, compared to 52% of thosewho had seen fewer than 10 patients (OR = 3.0 [1.4,6.4], p = 0.0161, n = 127). Duration of maintenance ther-apy was not associated with practice location (Canadavs. USA vs. other; p = 0.3110, n = 130) or years of prac-tice (<5 vs. 5 or more; OR = 1.6 [0.70, 3.89], p = 0.3109,n = 127). Concurrent corticosteroid duration initiated atinduction was typically six (39%) or 12 (40%) months,with some respondents continuing for 18 (6%), 24 (4%),or 36 (0.7%) months and 11% chose variable periods ac-cording to disease severity. 93% of respondents routinelyrecommended plasma exchange in certain situations:severe pulmonary hemorrhage and/or rapidly progres-sive renal disease (76%), rapidly progressive renal diseaseonly (8%), pulmonary hemorrhage only (5%), and withco-existent conditions (10%).Interest in clinical registriesAll respondents felt that an international collaborativeregistry was important for comparative outcome assess-ment of treatment strategies for GPA/MPA, with 64% ofrespondents selecting “very important” (mode 5, median5, IQR 4-5). Primary motivations for participation inclinical studies or collaborative registries included thepotential to improve outcomes for children with AAV(91%), access to available tools and resources (58%), en-dorsement by a formal network of investigators (49%),potential publication authorship (46%), and associationwith specific research objectives (40%). Only 15% of re-spondents felt that a monetary stipend was a major mo-tivation. Registry-associated resources believed to be ofmost value included an automated PVAS calculator(82%), an online algorithm to stage disease severity withlinks to corresponding treatment guidelines (68%), anautomated PVDI calculator (66%), a classification toolbased on entered patient data (54%), and a printabletable to track patient data over multiple visits (54%).Most clinicians had resources available to assist with theuse of registry data entry and clinical tools, includingcomputer and internet access in the clinic (80% each)and a trainee or fellow (58%). Fewer had a research as-sistant (47%), support for review board applications(37%), technology support (33%), or a research nurse(23%). Major barriers to registry participation includedFig. 1 Typical dosing regimens prescribed for cyclophosphamide (CYC) and rituximab (RTX) induction therapy. Data represent number ofrespondents out of n = 140 for both agents. aEUVAS protocolWestwell-Roper et al. Pediatric Rheumatology  (2017) 15:61 Page 6 of 10insufficient support for data entry (59%), lack of time(43%), burden of ethics review board approval (32%),and lack of patients (24%).DiscussionThis survey emphasizes the limited individual experienceamong clinicians caring for children with GPA and MPAand reiterates clinicians’ aspirations to study childhood-onset vasculitis independently from adult disease [3].International collaborations within the last decade havefacilitated the development of pediatric-specific classifi-cation criteria for GPA [9] and scoring tools for diseaseactivity and damage [12, 15]. They have also evaluatedthe pediatric utility of adult patient algorithms for classi-fying MPA [16] and staging disease severity [17]. More-over, the recent revision of the CHCC definitions isrelevant to both adults and children [1]. These newlydeveloped criteria, definitions, and clinical tools are es-sential for systematic evaluation of drug efficacy. How-ever, the present survey reveals limited uptake bypediatric caregivers and identifies several challenges inthe development, implementation, and evaluation ofpediatric treatment guidelines.Distinguishing GPA from MPAMost respondents believed that the distinction betweenGPA and MPA in patients with AAV was important forprognostication, consistent with meta-analyses of adultstudies showing lower 5-year survival in MPA versusGPA [18] and of pediatric studies showing less frequentrelapses [18] and more severe renal disease [16] in MPA.While respondents were familiar with both the ACR1990 criteria for GPA [9] and its EULAR/PRINTO/PRES2008 pediatric adaptation, neither organization includescriteria to define MPA [8, 19]. Not surprisingly, respon-dents often used multiple methods to differentiate (sub-classify) GPA and MPA phenotypes, and 25% reportedsometimes not differentiating. Frequently, ANCA speci-ficity to PR3 or MPO was used “erroneously” as theprimary differentiating criterion to define GPA andMPA, respectively. Adult studies suggest that genetics,pathophysiology, and clinical outcomes may be morestrongly associated with antibody status than with theclinical syndromes [20]; although the populations de-fined by PR3 or MPO antibodies overlap with GPA andMPA, respectively, they are not the same. Indeed, mostrespondents were not aware of the recently revisedCHCC 2012 definitions that do not incorporate ANCAspecificity but arguably provide the clearest discriminat-ing definitions of GPA and MPA [1]. Moreover, less than20% of respondents were familiar with the EuropeanMedicines Agency (EMA) classification algorithm or itspediatric modification [21, 22]. However, this is a com-plicated tool primarily used in the research setting.Clinical scoring toolsMultiple national and international rheumatology organiza-tions have published treatment guidelines for managementof AAV in adults [23–31]. Many require formal clinicalmeasurement tools to assess disease “severity” (to stratify ap-propriate therapy) and “activity” (to determine when to start,stop, or modify therapy). These standardized measures alsoguide entry of adult patients into research studies and willinevitably guide research eligibility in children. However,one third of respondents reported only sometimes tailoringinduction therapy to a formalized disease severity scale, sug-gesting that other factors – likely including informal assess-ment, histopathology [32], and specific organ involvement –influence management decisions. In children, PVAS [12]may show better correlation with treatment decisions thanthe BVAS v.3 [33]; nevertheless, equal proportions of re-spondents used BVAS or its variants. Furthermore, evalu-ation of permanent damage from disease or treatment isrequired for outcome assessment and may allow predictionof mortality [34]. The PVDI is a relatively recent pediatricmodification of the adult VDI [14] yet to be validated [15];however, few respondents used either tool. Lack of know-ledge or familiarity – as well as perceived lack of utility –may account for some lack of uptake. Successful implemen-tation of practice guidelines requiring standardized assess-ment tools will depend on improved clinician engagement,clear clinical utility, and feasibility of use.Treatment practicesConsistent with guidelines for treatment of adults withAAV, all survey respondents employed a remission-induction and remission-maintenance model. Although25% of respondents used EULAR/EUVAS recommenda-tions to guide their treatment decisions, many employeda combination of resources. There was significant vari-ation in medication and dosing regimens for childhoodGPA/MPA, with variable use of evidence-based recom-mendations from existing adult guidelines.Potential engagement in comparative effectivenessresearchAn overwhelming majority of respondents believed in theneed for pediatric-specific treatment guidelines and sup-ported guideline development by expert group consensus. Itis not simple to reconcile these aspirations with the limiteduptake of recent pediatric-specific research and consensusreports, although the described lack of knowledge or famil-iarity – as well as unclear added benefit beyond currentpractice – likely contribute. In view of the high frequency ofrenal disease (75%) among pediatric AAV patients [16] andthe existing high level of cross-specialty collaboration, futureefforts towards developing and implementing guidelines formanaging patients with AAV must further engage pediatricnephrologists.Westwell-Roper et al. Pediatric Rheumatology  (2017) 15:61 Page 7 of 10A recent European initiative to develop standards ofcare for pediatric rheumatic diseases, SHARE (Single Huband Access point for Paediatric Rheumatology in Europe)[35], will soon provide consensus-based guidelines formanagement of pediatric vasculitides, including GPA [36].While many of these recommendations will be based onadult data, most survey respondents favored longer-termcomparative outcome assessment of a range of possibleconsensus approaches. Given barriers to uptake of existingpediatric-specific tools – many of which will be part ofstandardized guidelines – incentives for implementationare critical. A registry that provides treatment guidelinesendorsed by national/international rheumatology organi-zations, access to clinical tools with automated scoringcalculators, and individual patient summary progress re-ports might further improve engagement and uptake.Limitations of this study include a sample populationthat likely underestimates true variability in practicegiven the selection bias associated with elective surveycompletion [37] and response rates below 50% [38]. Par-ticipants were reached through formal organizations andtherefore may represent a particularly informed group,leading to overestimation of the use of clinical tools andguidelines. Moreover, the survey included few respon-dents outside Europe and North America, and thereforeis limited in generalizability particularly with respect topractice variation in Asian and African countries.ConclusionsTaken together, these data suggest a need for and interestin consensus treatment guidelines for pediatric GPA andMPA. They also point to potential challenges associatedwith guideline development and implementation. Under-standing the barriers to uptake of existing classificationcriteria and formal assessment tools will inform efforts toimprove standardization of classification and assessmentto guide therapy. Physicians’ aspirations for pediatric-specific, evidence-based treatments may motivate uptakeof guidelines that evolve from this needs assessment sur-vey and, imminently, from SHARE. Consensus-derivedguidelines that include a range of specific treatment op-tions – if provided together with mechanisms for com-parative effectiveness research through an internationalregistry – may facilitate clinician engagement and ultim-ately lead to improved outcomes for children with AAV.Additional fileAdditional file 1: Copy of Needs-assessment Survey for Treatment ofpediatric ANCA-associated vasculitis. (PDF 110 kb)AbbreviationsAAV: Antineutrophil cytoplasmic antibody-associated vasculitis;ANCA: Antineutrophil cytoplasmic antibody; ARChiVe: A registry forchildhood vasculitis; AZA: Azathioprine; BVAS: Birmingham vasculitis activityscore; CAPRI: Canadian alliance of pediatric rheumatology investigators;CARRA: Childhood arthritis and rheumatology research alliance; CHCC: Chapel hillconsensus conference; CYC: Cyclophosphamide; EGPA: Eosinophilic granulomatosiswith polyangiitis; EMA: European medicines agency; EULAR: European leagueagainst rheumatism; EUVAS: European vasculitis study group; GPA: Granulomatosiswith polyangiitis; IQR: Interquartile range; MMF: Mycophenolate mofetil;MPA: Microscopic polyangiitis; MTX: Methotrexate; PedVas: Pediatric vasculitisinitiative; PRES: Pediatric Rheumatology European Society; PRINTO: Paediatricrheumatology international trials organisation; PVAS: Pediatric vasculitis activityscore; PVDI: Pediatric vasculitis damage index; RTX: Rituximab; SHARE: Singlehub and access point for Paediatric Rheumatology in Europe; VDI: Vasculitisdamage index; WGET: Wegener’s granulomatosis etanercept trialAcknowledgementsThe authors thank CARRA, CAPRI, the Vasculitis Working Group of PRES, andthe Australian Paediatric Rheumatology Group for input into survey designand/or distribution to their members. CARRA is supported by NIAMS, Friendsof CARRA, and the Arthritis Foundation. We also thank Angelyne SarmientoRivera and Jayson Tan (British Columbia Children’s Hospital) for assistancewith REDCap and preliminary data analysis, respectively.FundingThis work was supported by a Canadian Institutes of Health Research grantfor the PedVas Initiative [TR2-119188 to DAC].Availability of data and materialsThe survey is attached to this manuscript as a Additional file 1. All datagenerated or analyzed during this study are included in this published article.Authors’ contributionsDAC conceived of the study. CW-R, KLB, KAM, CM, and DAC developed the pilotsurvey. LW-W, EY, SCL, KMO, SKB, PB, RC reviewed, pilot-tested, and modifiedthe survey for the final draft. CW-R analyzed the data with input from JML andprepared the initial manuscript draft. All authors participated in the writing ofthe manuscript and also read and approved the final draft for submission.Ethics approval and consent to participateThe survey was conducted as a quality assurance initiative for PedVas, whichwas approved by the Children’s and Women’s Research Ethics Board of theUniversity of British Columbia (H12-00894). Under Article 2.5 of the Tri CouncilPolicy Statement, quality assurance/improvement activities are not subject tofurther institutional review.Consent for publicationNot applicableCompeting interestsThe authors declare that they have no competing interests.Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.Author details1Clinical Professor, Division of Rheumatology, Department of Pediatrics,University of British Columbia, BC Children’s Hospital, Room K4-119 4480 OakStreet Vancouver, Vancouver, BC V6H 3V4, Canada. 2Simon Fraser University,Burnaby, BC, Canada. 3University of Chicago Comer Children’s Hospital,Chicago, USA. 4University of California – Los Angeles, Los Angeles, CA, USA.5Joseph M. Sanzari Children’s Hospital, Hackensack, NJ, USA. 6Riley Hospitalfor Children at IU Health, Indianapolis, IN, USA. 7Morristown Medical Center,Morristown, NJ, USA. 8Great Ormond Street Hospital NHS Foundation Trust,London, UK. 9Ospedale Pediatrico Meyer Firenze, Florence, Italy.Received: 10 April 2017 Accepted: 1 August 2017References1. Jennette JC, Falk RJ, Bacon PA, Basu N, Cid MC, Ferrario F, Flores-SuarezLF, Gross WL, Guillevin L, Hagen EC, et al. Revised international ChapelWestwell-Roper et al. Pediatric Rheumatology  (2017) 15:61 Page 8 of 10Hill consensus conference nomenclature of Vasculitides. Arthritis Rheum.2012;2013(65):1–11.2. Morishita K, Brown K, Cabral D. Pediatric vasculitis: advances in treatment.Curr Opin Rheumatol. 2015;27:493–9.3. Wilkinson NM, Page J, Uribe AG, Espinosa V, Cabral DA. Establishmentof a pilot pediatric registry for chronic vasculitis is both essential andfeasible: a childhood arthritis and rheumatology alliance (CARRA) survey.J Rheumatol. 2007;34:224–6.4. Cabral D, Benseler S, Foell D, Hancock RE, Ross C, Graham J, LuqmaniR. Chronic primary vasculitis in children: a new research opportunity.CRAJ. 2012;22:10–1.5. Huber AM, Giannini EH, Bowyer SL, Kim S, Lang B, Lindsley CB, PachmanLM, Pilkington C, Reed AM, Rennebohm RM, et al. Protocols for the initialtreatment of moderately severe juvenile dermatomyositis: results of aChildren's arthritis and rheumatology research alliance consensusconference. Arthritis Care Res. 2010;62:219–25.6. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Researchelectronic data capture (REDCap) - a metadata-driven methodology andworkflow process for providing translational research informatics support. JBiomed Inform. 2009;42:377–81.7. Benjamini Y, Hochberg Y. Controlling the false discovery rate - a practicaland powerful approach to multiple testing. J R Stat Soc Series B StatMethodol. 1995;57:289–300.8. Leavitt RY, Fauci AS, Bloch DA, Michel BA, Hunder GG, Arend WP, CalabreseLH, Fries JF, Lie JT, Lightfoot RW Jr, et al. The American College ofRheumatology 1990 Criteria for the classification of Wegener’sgranulomatosis. Arthritis Rheum. 1990;33:1101–7.9. Ozen S, Pistorio A, Iusan SM, Bakkaloglu A, Herlin T, Brik R, Buoncompagni A,Lazar C, Bilge I, Uziel Y, et al. EULAR/PRINTO/PRES criteria for Henoch-Schonlein purpura, childhood polyarteritis nodosa, childhood Wegenergranulomatosis and childhood Takayasu arteritis: Ankara 2008. Part II: finalclassification criteria. Ann Rheum Dis. 2010;69:798–806.10. Jennette JC, Falk RJ, Andrassy K, Bacon PA, Churg J, Gross WL, Hagen EC,Hoffman GS, Hunder GG, Kallenberg CG, et al. Nomenclature of systemicvasculitides. Proposal of an international consensus conference. ArthritisRheum. 1994;37:187–92.11. Luqmani RA, Bacon PA, Moots RJ, Janssen BA, Pall A, Emery P, Savage C,Adu D. Birmingham Vasculitis activity score (BVAS) in systemic necrotizingvasculitis. QJM. 1994;87:671–8.12. Dolezalova P, Price-Kuehne FE, Ozen S, Benseler SM, Cabral DA, Anton J,Brunner J, Cimaz R, O'Neil KM, Wallace CA, et al. Disease activity assessmentin childhood Vasculitis: development and preliminary validation of thepaediatric Vasculitis activity score (PVAS). Ann Rheum Dis. 2013;72:1628–33.13. Stone JH, Hoffman GS, Merkel PA, Min YI, Uhlfelder ML, Hellmann DB,Specks U, Allen NB, Davis JC, Spiera RF, et al. A disease-specific activity indexfor Wegener’s granulomatosis: modification of the Birmingham Vasculitisactivity score. International network for the study of the systemicVasculitides (INSSYS). Arthritis Rheum. 2001;44:912–20.14. Exley AR, Bacon PA, Luqmani RA, Kitas GD, Gordon C, Savage CO, Adu D.Development and initial validation of the Vasculitis damage index for thestandardized clinical assessment of damage in the systemic vasculitides.Arthritis Rheum. 1997;40:371–80.15. Dolezalova P, Wilkinson NW, Brogan PA, Anton J, Benseler SM, BrunnerJ, Cabral DA, Cimaz R, O'Neil KM, Özen S, Luqmani R. PaediatricVasculitis damage index: a new tool for standardised diseaseassessment. Ann Rheum Dis. 2014;73:696–7.16. Cabral DA, Canter DL, Muscal E, Nanda K, Wahezi DM, Spalding SJ, Twilt M,Benseler SM, Campillo S, Charuvanij S, et al. Comparing presenting clinicalfeatures of 48 children with microscopic polyangiitis (MPA) against 183having granulomatosis with polyangiitis (GPA). An ARChiVe study. ArthritisRheumatol. 2016;68:2514–26.17. Morishita K, Guzman J, Chira P, Muscal E, Zeft A, Klein-Gitelman M, Uribe AG,Abramson L, Benseler SM, Eberhard A, et al. Do adult disease severitysubclassifications predict use of cyclophosphamide in children with ANCA-associated vasculitis? An analysis of ARChiVe study treatment decisions. JRheumatol. 2012;39:2012–20.18. Iudici M, Quartier P, Terrier B, Mouthon L, Guillevin L, Puechal X. Childhood-onset granulomatosis with polyangiitis and microscopic polyangiitis:systematic review and meta-analysis. Orphanet J Rare Dis. 2016;11:141.19. Fries JF, Hunder GG, Bloch DA, Michel BA, Arend WP, Calabrese LH, FauciAS, Leavitt RY, Lie JT, Lightfoot RW, Jr., et al: The American College ofRheumatology 1990 criteria for the classification of vasculitis. Summary.Arthritis Rheum 1990, 33:1135-1136.20. Pepper RJ, Salama AD. Classifying and predicting outcomes in ANCA-associated glomerulonephritis. Nephrol Dial Transplant. 2012;27:2135–7.21. Uribe AG, Huber AM, Kim S, O'Neil KM, Wahezi DM, Abramson L, Baszis K,Benseler SM, Bowyer SL, Campillo S, et al. Increased sensitivity of theEuropean medicines agency algorithm for classification of childhoodgranulomatosis with polyangiitis. J Rheumatol. 2012;39:1687–97.22. Watts R, Lane S, Hanslik T, Hauser T, Hellmich B, Koldingsnes W, Mahr A,Segelmark M, Cohen-Tervaert JW, Scott D. Development and validation of aconsensus methodology for the classification of the ANCA-associatedvasculitides and polyarteritis nodosa for epidemiological studies. AnnRheum Dis. 2007;66:222–7.23. McGeoch L, Twilt M, Famorca L, Bakowsky V, Barra L, Benseler S, Cabral DA,Carette S, Cox GP, Dhindsa N, et al. CanVasc recommendations for themanagement of antineutrophil cytoplasm antibody (ANCA)-associatedvasculitides - executive summary. Can J Kidney Health Dis. 2015;2:43.24. Lapraik C, Watts R, Bacon P, Carruthers D, Chakravarty K, D'Cruz D, Guillevin L,Harper L, Jayne D, Luqmani R, et al. BSR and BHPR guidelines for the managementof adults with ANCA associated vasculitis. Rheumatology. 2007;46:1615–6.25. Ntatsaki E, Carruthers D, Chakravarty K, D'Cruz D, Harper L, Jayne D,Luqmani R, Mills J, Mooney J, Venning M, et al. BSR and BHPR guideline forthe management of adults with ANCA-associated vasculitis. Rheumatology.2014;53:2306–9.26. French Vasculitis Study Group: Vascularites nécrosantes systémiques -Protocole national de diagnostic et de soins. L'Haute autorité de santé deFrance; 2009.27. Charles P, Bienvenu B, Bonnotte B, Gobert P, Godmer P, Hachulla E,Hamidou M, Harle JR, Karras A, Lega JC, et al. Rituximab: recommendationsof the French Vasculitis study group (FVSG) for induction and maintenancetreatments of adult, antineutrophil cytoplasm antibody-associatednecrotizing vasculitides. Presse Med. 2013;42:1317–30.28. KDIGO Work Group. Kidney Disease: Improving Global Outcomes (KDIGO)Glomerulonephritis Work Group: KDIGO Clinical Practice Guideline forGlomerulonephritis. Kidney Int. 2012;2:139–274.29. Mukhtyar C, Guillevin L, Cid MC, Dasgupta B, de Groot K, Gross W, Hauser T,Hellmich B, Jayne D, Kallenberg CG, et al. EULAR recommendations for themanagement of primary small and medium vessel vasculitis. Ann RheumDis. 2009;68:310–7.30. Menahem S, Hiremagalur B, Mudge D, Toussaint N, Walters G. Caring for Australianswith renal impairment: the CARI guidelines. Induction and maintenance therapy inANCA-associated systemic vasculitis. Nephrology. 2008;13(Suppl 2):S24–36.31. Japanese Circulation Society Joint Working Group: Guideline formanagement of vasculitis syndrome (JCS 2008). Japanese CirculationSociety. Circ J. 2011(75):474–503.32. Berden AE, Ferrario F, Hagen EC, Jayne DR, Jennette JC, Joh K, Neumann I,Noel LH, Pusey CD, Waldherr R, et al. Histopathologic classification of ANCA-associated glomerulonephritis. J Am Soc Nephrol. 2010;21:1628–36.33. Morishita K, Li SC, Muscal E, Spalding S, Guzman J, Uribe A, Abramson L,Baszis K, Benseler S, Bowyer S, et al. Assessing the performance of theBirmingham Vasculitis activity score at diagnosis for children withantineutrophil cytoplasmic antibody-associated Vasculitis in a registry forchildhood Vasculitis (ARChiVe). J Rheumatol. 2012;39:1088–94.34. Bhamra K, Luqmani R. Damage assessment in ANCA-associated vasculitis.Curr Rheumatol Rep. 2012;14:494–500.35. Wulffraat NM, Vastert B. SHARE consortium: time to share. PediatrRheumatol Online J. 2013;11:5.36. de Graeff N, Groot N, Kamphuis S, Beresford M. FRI0271 final evidence-based recommendations for diagnosis and treatment of paediatricVasculitides. Ann Rheum Dis. 2015;74:522–3.37. Khazaal Y, van Singer M, Chatton A, Achab S, Zullino D, Rothen S, Khan R,Billieux J, Thorens G. Does self-selection affect samples’ representativenessin online surveys? An investigation in online video game research. J MedInternet Res. 2014;16:e164.38. Edwards PJ, Roberts I, Clarke MJ, DiGuiseppi C, Wentz R, Kwan I, CooperR, Felix LM, Pratap S. Methods to increase response to postal and electronicquestionnaires. Cochrane Database of Systematic Reviews. 2009. Issue 3. Art.No.: MR000008. doi:10.1002/14651858.MR000008.pub4.39. Mukhtyar C, Lee R, Brown D, Carruthers D, Dasgupta B, Dubey S, Flossmann O,Hall C, Hollywood J, Jayne D, et al. Modification and validation of theBirmingham Vasculitis activity score (version 3). Ann Rheum Dis. 2009;68:1827–32.Westwell-Roper et al. Pediatric Rheumatology  (2017) 15:61 Page 9 of 1040. Mahr AD, Neogi T, Lavalley MP, Davis JC, Hoffman GS, McCune WJ, SpecksU, Spiera RF, St Clair EW, Stone JH, et al. Assessment of the item selectionand weighting in the Birmingham vasculitis activity score for Wegener'sgranulomatosis. Arthritis Rheum. 2008;59:884–91.41. European therapeutic trials in ANCA-associated systemic vasculitis: diseasescoring, consensus regimens and proposed clinical trials. EuropeanCommunity study group on clinical trials in systemic VasculitisECSYSVASTRIAL. Clin Exp Immunol. 1995;101(Suppl 1):29–34.42. Guillevin L, Pagnoux C, Seror R, Mahr A, Mouthon L, Le Toumelin P. FrenchVasculitis study G: the five-factor score revisited: assessment of prognoses ofsystemic necrotizing vasculitides based on the French Vasculitis studygroup (FVSG) cohort. Medicine. 2011;90:19–27.43. Stone JH. Wegener’s Granulomatosis Etanercept trial research group: limitedversus severe Wegener’s granulomatosis: baseline data on patients in theWegener’s granulomatosis etanercept trial. Arthritis Rheum. 2003;48:2299–309.44. de Groot K, Gross WL, Herlyn K, Reinhold-Keller E. Development andvalidation of a disease extent index for Wegener’s granulomatosis. ClinNephrol. 2001;55:31–8.45. Seo P, Luqmani RA, Flossmann O, Hellmich B, Herlyn K, Hoffman GS, JayneD, Kallenberg CG, Langford CA, Mahr A, et al. The future of damageassessment in vasculitis. J Rheumatol. 2007;34:1357–71.46. Suppiah R, Flossman O, Mukhtyar C, Alberici F, Baslund B, Brown D, HasanN, Holle J, Hruskova Z, Jayne D, et al. Measurement of damage in systemicVasculitis: a comparison of the Vasculitis damage index with the combineddamage assessment index. Ann Rheum Dis. 2011;70:80–5.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript your next manuscript to BioMed Central and we will help you at every step:Westwell-Roper et al. Pediatric Rheumatology  (2017) 15:61 Page 10 of 10


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items