International Consortium of Vascular Registries Consensus

Eur J Vasc Endovasc Surg (2018) 56, 217e237

International Consortium of Vascular Registries ConsensusRecommendations for Peripheral Revascularisation Registry Data Collection

Christian-Alexander Behrendt a,*, Daniel Bertges b, Nikolaj Eldrup c, Adam W. Beck d, Kevin Mani e, Maarit Venermo f,Zoltán Szeberin g, Gabor Menyhei h, Ian Thomson i, Georg Heller j, Pius Wigger k, Gudmundur Danielsson l, Giuseppe Galzerano m,Cristina Lopez n, Martin Altreuther o, Birgitta Sigvant e, Henrik C. Rieß a, Art Sedrakyan p, Barry Beiles q, Martin Björck e,Jonathan R. Boyle r, E. Sebastian Debus a,y, Jack Cronenwett s,y

a Department of Vascular Medicine, University Heart Centre Hamburg, University Medical Centre Hamburg-Eppendorf, Hamburg, Germanyb Division of Vascular Surgery, University of Vermont Medical Center, Burlington, VT, USAc Department of Cardio-Thoracic and Vascular Surgery, Aarhus University Hospital, Aarhus, Denmarkd Division of Vascular Surgery and Endovascular Therapy, University of Alabama at Birmingham, AL, USAe Department of Surgical Sciences, Uppsala University, Uppsala, Swedenf Department of Vascular Surgery, Helsinki University Hospital and University of Helsinki, Helsinki, Finlandg Department of Vascular Surgery, Semmelweis University, Budapest, Hungaryh Department of Vascular Surgery, Pecs University Medical Centre, Pecs, Hungaryi Department of Vascular Surgery, Dunedin School of Medicine, Dunedin Hospital, Dunedin, New Zealandj Department of Vascular Surgery, Kantonsspital St. Gallen, Switzerlandk Department of Surgery, Kantonspital, Winterthur, Switzerlandl Reykjavik Venous Centre, Domus Medica, Reykjavík, Icelandm Vascular Surgery, Misericordia Hospital of Grosseto, Usl Toscana Sud-Est, Grosseto, Italyn Department of Vascular Surgery, University Hospital of Granada, Spaino Department of Vascular Surgery, St. Olavs Hospital, Trondheim, Norwayp Healthcare Policy and Research, Weill Cornell Medical College, New York, NY, USAq Australian and New Zealand Society for Vascular Surgery, Melbourne, Australiar Department of Vascular Surgery, Cambridge University Hospitals NHS Trust, Cambridge, UKs Department of Surgery Dartmouth-Hitchcock Medical Centre, Lebanon, NH, USA

y E.S* CoE-ma1078https

WHAT THIS PAPER ADDS
This paper presents the first international consensus on creation of a minimum and optimum core data set forregistries devoted to peripheral arterial revascularisation. A modified Delphi approach with online interactionwas used to achieve consensus among international experts from multiple countries. The concept of simple tomore complex levels of data capture allows harmonisation at all levels, despite variation among registries.Adoption of a standard variable set by the national registries within the International Consortium of VascularRegistries will provide opportunities for more advanced collaborations, including amalgamation of large scaleinternational data for assessment of outcomes after the introduction of new techniques and devices.
Objective/Background: To achieve consensus on the minimum core data set for evaluation of peripheral arterialrevascularisation outcomes and enable collaboration among international registries.Methods: A modified Delphi approach was used to achieve consensus among international vascular surgeons andregistry members of the International Consortium of Vascular Registries (ICVR). Variables, including definitions,from registries covering open and endovascular surgery, representing 14 countries in ICVR, were collected andanalysed to define a minimum core data set and to develop an optimum data set for registries. Up to threedifferent levels of variable specification were suggested to allow inclusion of registries with simpler versus morecomplex data capture, while still allowing for data aggregation based on harmonised core definitions.Results: Among 31 invited experts, 25 completed five Delphi rounds via internet exchange and face to facediscussions. In total, 187 different items from the various registry data forms were identified for potentialinclusion in the recommended data set. Ultimately, 79 items were recommended for inclusion in minimum coredata sets, including 65 items in the level 1 data set, and an additional 14 items in the more specific level 2 and 3recommended data sets. Data elements were broadly divided into (i) patient characteristics; (ii) comorbidities;(iii) current medications; (iv) lesion treated; (v) procedure; (vi) bypass; (vii) endarterectomy (viii) catheter basedintervention; (ix) complications; and (x) follow up.

.D. and J.C. share last authorship.rresponding author.il address: [email protected] (Christian-Alexander Behrendt).-5884/� 2018 European Society for Vascular Surgery. Published by Elsevier B.V. All rights reserved.://doi.org/10.1016/j.ejvs.2018.04.006

mailto:[email protected]

https://doi.org/10.1016/j.ejvs.2018.04.006

218 Christian-Alexander Behrendt et al.

Conclusion: A modified Delphi study allowed 25 international vascular registry experts to achieve a consensusrecommendation for a minimum core data set and an optimum data set for peripheral arterial revascularisationregistries. Continued global harmonisation of registry infrastructure and definition of items will overcomelimitations related to single country investigations and enhance the development of real world evidence.� 2018 European Society for Vascular Surgery. Published by Elsevier B.V. All rights reserved.Article history: Received 26 January 2018, Accepted 9 April 2018, Available online 30 May 2018Keywords: Consensus development, Delphi technique, Health services research, Peripheral arterial disease,

Registries

INTRODUCTION

Although peripheral arterial disease (PAD) remains anincreasing burden for national healthcare systems with>200 million people affected worldwide,1 many questionsregarding treatment of this disease cannot be answeredusing evidence from trials. Thus, in the absence of suchevidence, many recommendations in international practiceguidelines are built on expert consensus.2e4 As there areonly a few randomised controlled trials (RCTs) with wellknown problems of selection bias and limited external val-idity, with reasonable efforts registries and registry basedcohort studies can help to fill the gaps. Registries allowevaluation of treatment practice patterns, medical deviceevaluation, and can assess convergence of real world andRCT evidence.5 Although multiple national vascular regis-tries exist, lack of consensus around variables (and theirdefinitions) makes aggregation and comparison of findingsdifficult.

International collaborations such as the InternationalConsortium of Vascular Registries (ICVR; www.icvr-initiative.org) can help harmonise cross border research. The ICVR iscomprised of countries with vascular surgery registries,including the Vascular Quality Initiative (VQI; www.vqi.org)in the USA and the Vascunet Collaboration, consisting ofvascular registries from 12 countries in Europe and Aus-tralasia (www.vascunet.org). The ICVR was launched in 2014with the goal of establishing a collaborative platform acrossregistries to share data in order to improve the quality ofvascular health care.6 Contributions regarding abdominalaortic aneurysms (AAA) and carotid artery stenosis wererecently published by this collaboration.6e9 For this project,ICVR members aimed to apply a modified Delphi approachto achieve agreement on a minimum core data set and tocreate an optimum data set for registries capturing surgicaland interventional PAD treatments.

METHODS

The Delphi approach is widely accepted and used to gainconsensus among a panel of experts,10 and has previouslybeen used in various specialties, including vascular sur-gery.11e15 Representatives of 14 national vascular registriesparticipating in the ICVR from Australia (AustralasianVascular Audit), Denmark (Karbase), Finland (HUSvasc),Germany (GermanVasc and Aortic Registry of the GermanVascular Society), Hungary (Hungarian Vascular Registry),Iceland (Isvasc), Italy (Italian Vascular and EndovascularRegistry), New Zealand (Australasian Vascular Audit), Nor-way (NORKAR), Spain, Sweden (Swedvasc), Switzerland

(Swissvasc), and the USA (VQI) submitted their registries’current data sheets and definitions of data elements. Anextensive narrative review of the literature was conductedto identify additional items in registry based studies onPAD. All participants in this study agreed to the scope ofitems identified through the abovementioned process.Members of the ICVR were then invited to participate inweb based anonymised electronic questionnaires. Opensource software (www.limesurvey.org) was used togenerate the questionnaires. The participants could onlysubmit one set of answers in each Delphi round. Followingeach round, a structured report, including anonymisedgroup responses, mean results with SDs, as well as com-ments, were forwarded to the participants by email beforethey were invited to the next round. Each participant wasasked to indicate whether they agreed that individual var-iables should be included in the consensus data set, andeach item was scored on a five point Likert scale comprising“strongly agree”, “agree”, “neutral”, “disagree”, and “stronglydisagree”. Additionally, a free text comment could be sub-mitted for each item. Items repeatedly rated with “stronglyagree” or “agree” were recommended for the minimumdata set. Items repeatedly rated with “strongly disagree” or“disagree” were eliminated from consideration. If consensuswas not achieved after three rounds, the remaining itemswere discussed by the experts in two face to face ICVRmeetings and added to the minimum data set if 80% of theexperts supported the variable.

During this evaluation, it became apparent that it wasimportant to determine not only which variables to include,but also what level of detail was needed for each variableincluded. By analysing each current national registry, it wasdetermined that considerable variation existed in the levelof detail collected, and in some cases the definition of thevariables. In order to allow different levels of detail to becollected by different registries, but still allow harmo-nisation, three “levels” of variable recording detail but withcommon core definitions were created. Thus, reportinglevels were stratified for data elements as level 1, 2, and 3,ranging from minimum to optimum. Reporting level 1 forvariables were considered the minimum informationnecessary and typically have a simple input (yes, no) orsimple numeric range. Level 2 and 3 variables have addi-tional increasing specificity and granularity. For example,reporting the comorbidity of diabetes includes yes/no inreporting level 1. The more specific reporting level 2 in-cludes the type of medical treatment (insulin, oral antidia-betic, etc.), whereas reporting level 3 includes HbA1c level

http://www.icvr-initiative.org

http://www.icvr-initiative.org

http://www.vqi.org

http://www.vascunet.org

http://www.limesurvey.org

Table 1. Seventy-nine items in the minimal core and optimal data set for registries evaluating peripheral arterial revascularisation.

Category Variable Reporting Level 1 Reporting Level 2 Reporting Level 3 Comments Reference1) PatientCharacteristics

Birth Date Day (dd), month (mm), andyear (yyyy) of birth

Used to calculate age at timeof procedure for subsequentanalysis and de-identifieddata sharing. If regulations donot allow date collection, age(years) can be substituted.

Sex Female, Male Female, Male, Transfemale, Trans male

Sex at birth. Trans female hastransitioned from female sexto male gender; Trans malehas transitioned from malesex to female gender.

Weight Body Weight in kgHeight Body Height in cmFunctional Status Full activity, Light Work,

Self Care, Assisted Care,Bedbound

Add disease specificquality of life survey,such as Vascu-Qol-6

A person’s level offunctioning in terms of theirability to care for them self,daily activity, and physicalability (walking, workingetc.).

Oken MM, Creech RH, TormeyDC, Horton J, Davis TE,McFadden ET, et al. Toxicity andresponse criteria of the EasternCooperative Oncology Group.Am J Clin Oncol. 1982;5(6):649e55.Nordanstig J, Wann-Hansson C,Karlsson J, Lundström M,Petterson M, Morgan MBF.Vascular Quality of LifeQuestionnaire-6 faciliates healthrelated quality of life assessmentin peripheral arterial disease. JVascSurg 2014;59:700-7.

Ambulation Fully Ambulatory,Ambulate with Prosthesis,Ambulate with AssistiveDevice, Wheelchair,Bedbound

Add walking distancesurvey such as WalkingImprovementQuestionnaire or GPSmonitored walking

ASA Grade Normal healthy patient (1),mild systemic disease (2),severe systemic disease (3),severe systemic disease thatis a constant threat to life (4),moribund patient who is notexpected to survive withoutthe operation (5), declaredbrain dead patient

Owens WD, Felts JA, SpitznagelEL. ASA Physical StatusClassifications: A Study ofConsistency ofRatings.Anesthesiology1978;49:239-243. Recentupdate: https://www.asahq.org/resources/clinical-information/asa-physical-status-classification-system

Continued

Intern

ational

Consortiu

mofVascu

larRegistries

219

https://www.asahq.org/resources/clinical-information/asa-physical-status-classification-system




Table 1-continued

Category Variable Reporting Level 1 Reporting Level 2 Reporting Level 3 Comments Reference2) Patient Co-morbidities

Diabetes No, Yes If Yes: Treated with:Insulin, Oral Antidiabetic,Both, Diet alone, NoTreatment

HbA1c in % Yes ¼ clinical diagnosisdocumented in medicalrecord based onFasting � 7 mmol/L or postglucose � 11.1 mmol/L and/or HbA1c � 6.5% and/orantidiabetic medication.

Stoner MC, Calligaro KD, ChaerRA, Dietzek AM, Farber A,Guzman RJ, et al. Reportingstandards of the Society forVascular Surgery forendovascular treatment ofchronic lower extremityperipheral artery disease. J VascSurg. 2016;64(1):e1-e21.Kumar R, Nandhini LP,Kamalanathan S, Sahoo J,Vivekanadan M. Evidence forcurrent diagnostic criteria ofdiabetes mellitus. World JDiabetes. 2016;7(17):396e405.

Current RenalFunction

Normal, Abnormal Serum Creatinine (inmmol/l or mg/dl)

Glomerular filtrationrate (GFR) in ml/min(Cockcroft-Gaultcalculation)

As defined by the NationalKidney Foundation::eGFR<60 mL/min/1.73 m2 on twooccasions separated by 3months and that is notassociated with a transient,reversible condition such asvolume depletion.

National Kidney F. K/DOQIclinical practice guidelines forchronic kidney disease:evaluation, classification, andstratification. American journalof kidney diseases: the officialjournal of the National KidneyFoundation. 2002;39(2 Suppl1):S1-266.

Current Dialysis No, Yes Duration of dialysisdependence (years)

Tobacco Use Current Smoker, FormerSmoker, Never Smoked

If Former Smoker: QuitDate

If Current or FormerSmoker: Pack Years(py) Smoked

Stoner MC, Calligaro KD, ChaerRA, Dietzek AM, Farber A,Guzman RJ, et al. Reportingstandards of the Society forVascular Surgery forendovascular treatment ofchronic lower extremityperipheral artery disease. J VascSurg. 2016;64(1):e1-e21.

220Christian

-Alexan

der

Beh

rendtet

al.

Current IschaemicHeart Disease

No, Yes If Yes: Asymptomatic,Angina only duringStrenuous or ProlongedPhysical Activity,Symptoms with EverydayLiving Activities, Inabilityto Perform any ActivityWithout Angina or Anginaat Rest

Yes ¼ current angina orpositive stress test indicatingischaemic heart disease

Lee TH, Marcantonio ER,Mangione CM, Thomas EJ,Polanczyk CA, Cook EF, et al.Derivation and prospectivevalidation of a simple index forprediction of cardiac risk ofmajor noncardiac surgery.Circulation. 1999;100(10):1043e9.

Prior MyocardialInfarction

No, Yes Add PriorRevascularisation: CABG,PCI

Add timing of MI andCABG/PCI:MI � 6 Months, >6 Months; CABG/PCI < 5 years, CABG/PCI > 5 years

Yes ¼ clinical historydocumented in medicalrecord


Congestive HeartFailure

Never, Former, Current If Current: New York HeartAssociation (NYHA)-Classification for heartfailure (Class I to IV)

Current EjectionFraction in %

Current/former ¼ clinicaldiagnosis/historydocumented in medicalrecord, based onPonikowskiet al. (2016) andLee et al.(1999): Patient has a historyof or current symptoms ofcongestive heart failure,pulmonary oedema, orparoxysmal nocturnaldyspnea, physicalexamination showingbilateral rales or S3 gallop, orchest radiograph showingpulmonary vascularredistribution.


Continued

Intern

ational

Consortiu

mofVascu

larRegistries

221

ble 1-continued

ategory Variable Reporting Level 1 Reporting Level 2 Reporting Level 3 Comments ReferencePonikowski P, Voors AA, AnkerSD, Bueno H, Cleland JG,CoatsAJ, et al. 2016 ESCGuidelines for the diagnosis andtreatment of acute and chronicheart failure: The Task Force forthe diagnosis and treatment ofacute and chronic heart failureof the European Society ofCardiology (ESC). Developed withthe special contribution of theHeart Failure Association (HFA)of the ESC. European journalof heart failure. 2016;18(8):891e975.

CardiacArrhythmia

Never, Former, Current If Current: Atrial,Ventricular, PreviousAblation, AV-Block withPacemaker, Defibrillator(ICD), Previous Ablationand Pacemaker or ICD,Other

Yes ¼ clinical diagnosisdocumented in medicalrecord, including atrialfibrillation, bradycardia,conduction disorders,premature contraction,tachycardia, ventricularfibrillation, other rhythmdisorders.

ChronicObstructivePulmonaryDisease

Never, Former, Current If Current: No Treatment,Medical Treatment, HomeOxygen

Yes ¼ clinical diagnosisdocumented in medicalrecord of chronic obstructivepulmonary disease orpulmonary medicaltreatment. As definedbyStoner et al. (2016).

Stoner MC, Calligaro KD, ChaerRA, Dietzek AM, Farber A,Guzman RJ, et al. Reportingstandards of the Society forVascular Surgery forendovascular treatment ofchronic lower extremityperipheral artery disease. J VascSurg. 2016;64(1):e1-e21.

Hypertension Never, Former, Current If Current: Treatedcontrolled, treateduncontrolled (bloodpressure > 140/90despite treatment)

Blood pressure > 140/90 ormedical treatment.

Mancia G, Fagard R, NarkiewiczK, Redon J, Zanchetti A, Böhm M,et al. 2013 ESH/ESC Guidelinesfor the management of arterialhypertension. Eur Heart J.2013;34:2159-2219.

222Christian

-Alexan

der

Beh

rendtet

al.

Ta

C

Prior PADRevascularisation

No, Yes If Yes: Interventional,Surgical, Both

Location: Inflow,Outflow; Type:Intervention, Bypass,Endarterectomy; Leg:R/L/Aorta

Prior Amputation Minor, Major If Yes: Level: Toe(s),transmetatarsal, belowknee, through knee,above knee, higher of R/Lleg

Conte MS, Geraghty PJ, BradburyAW, Hevelone ND, Lipsitz SR,Moneta GL, et al. Suggestedobjective performance goals andclinical trial design for evaluatingcatheter-based treatment ofcritical limb ischemia. J VascSurg. 2009;50(6):1462e73 e1-3.

3) CurrentMedication

Aspirin No, Yes If Yes: � 100 mg/d, 101e320 mg/d, >320 mg/d

Add medicationprescribed atdischarge.

Medication in effect at timeof procedure.

Other PlateletInhibitor

No, Yes if Yes, medication name Add medicationprescribed atdischarge.


Statin No, Yes If Yes: Low Dose, HighDose



Anticoagulant No, Yes If Yes: Vitamin KAntagonist, ThrombinInhibitor, Factor XaInhibitor, Other


Medication taken chronicallybefore procedure (even ifstopped in preparation forprocedure).

4) Lesiontreated

Symptoms forRight/Left Leg

Modified Rutherford-Classification: Asymptomatic,Mild Claudication, ModerateClaudication (>200m),Severe Claudication(<200m), Ischaemic RestPain, Ulcer/Necrosis, Non-healing Amputation, BothUlcer/Non-healingAmputation, Acute Ischaemia

will include ref RAPID for themodified Rutherford whenpublished

Continued

Intern

ational

Consortiu

mofVascu

larRegistries

223

Table 1-continued

Category Variable Reporting Level 1 Reporting Level 2 Reporting Level 3 Comments ReferenceFoot Infection forRight/Left Leg

No, Yes Grade 0 None, Grade 1Mild, Grade 2 Moderate,Grade 3 Severe

Mills, Joseph L. Sr., MD, et al.(2014). The Society for VascularSurgery Lower ExtremityThreatened Limb ClassificationSystem: Risk Stratification Basedon Wound, Ischaemia, and FootInfection (WIfI). Journal ofVascular Surgery, Volume 59(Issue 1) pp.220-pp.234.e2,https://doi.org/10.1016/j.jvs.2013.08.003

Tissue LossSeverity for Right/Left Leg

None, Grade 1, Shallow,Grade 2, Deep, Grade 3,Extensive

Mills, Joseph L. Sr., MD, et al.(2014). The Society for VascularSurgery Lower ExtremityThreatened Limb ClassificationSystem: Risk Stratification Basedon Wound, Ischaemia, and FootInfection (WIfI). Journal ofVascular Surgery, Volume 59(Issue 1) pp.220-pp.234.e2,https://doi.org/10.1016/j.jvs.2013.08.003

Ankle brachialindex for Right/Left Leg

Highest ABI� 1.3, ABI < 1.3and � 0.9, ABI < 0.9and � 0.7, ABI < 0.4, treatedleg

ABI exact measurementR/L leg

ABI and TBI exactmeasurement R/L leg

Aboyans V, Criqui MH, AbrahamP, Allison MA, Creager MA,Diehm C, et al. Measurementand interpretation of the ankle-brachial index: a scientificstatement from the AmericanHeart Association. Circulation.2012;126(24):2890e909.

Artery vs. Graft Native Artery, Bypass Graft Graft Conduit: Vein,Prosthetic; Location:Supra-Inguinal, Infra-Inguinal

Add Graft Origin andInsertion

224Christian

-Alexan

der

Beh

rendtet

al.

https://doi.org/10.1016/j.jvs.2013.08.003




Artery Treated orBypassed

Aorta, iliac, femoral,popliteal, tibial

Aorta, common Iliac,external iliac,common + external iliac,internal iliac, commonfemoral, superficialfemoral, profundafemoral, popliteal,SFA + popliteal, anteriortibial, posterior tibial,peroneal, tibioperonealtrunk, dorsal pedal,plantar

Add Artery Segment:Prox, Mid, Distal, as inPopliteal P1, P2, P3segments

Side for eachLesion Treated

Right, Left, Aorta

Lesion Length Short (<5cm for aorto-iliacand <25cm forfemoropopliteal lesions),Long (�5cm for aorto-iliacand � 25cm forfemoropopliteal lesions)

Length (cm) Acin F, de Haro J, Bleda S, VarelaC, Esparza L. Primary nitinolstenting in femoropoplitealocclusive disease: a meta-analysis of randomizedcontrolled trials. Journal ofendovascular therapy: an officialjournal of the InternationalSociety of EndovascularSpecialists. 2012;19(5):585e95.Aboyans V, Ricco J, BartelinkMEL, Björck M, Brodmann M,Cohnert T et al., 2017 ESCGuidelines on the Diagnosis andTreatment of Peripheral ArterialDiseases, in collaboration withthe European Society forVascular Surgery (ESVS),European Journal of Vascularand EndovascularSurgery (2017),http://dx.doi.org/10.1016/j.ejvs.2017.07.018

Most SevereStenosis

Stenosis, Occlusion Bollinger Score:Complete occlusion,Stenosis >50% oflumen, Stenosis of 25e50% of lumen, �25%stenosis of lumen

Bollinger A, Breddin K, Hess H,Heystraten FM, Kollath J,Konttila A, et al.Semiquantitative assessment oflower limb atherosclerosis fromroutine angiographic images.Atherosclerosis. 1981;38(3e4):339e46.

Continued

Intern

ational

Consortiu

mofVascu

larRegistries

225

http://dx.doi.org/10.1016/j.ejvs.2017.07.018

http://dx.doi.org/10.1016/j.ejvs.2017.07.018

Table 1-continued

Category Variable Reporting Level 1 Reporting Level 2 Reporting Level 3 Comments ReferenceOutflow fromTreated Artery

No outflow, one vessel,two vessels, three vessels

Modified Society forVascular Surgery (SVS)runoff score: from 0 to19 points

Davies MG, Saad WE, Peden EK,Mohiuddin IT, Naoum JJ,Lumsden AB. Impact of runoff onsuperficial femoral arteryendoluminal interventions forrest pain and tissue loss. J VascSurg. 2008;48(3):619e25;discussion 25e6.

5) Procedure Procedure Date Day (dd), month (mm), andyear (yyyy) of the procedure

Used to calculate length ofhospital stay for subsequentanalysis and de-identifieddata sharing. If regulations donot allow date collection,length of stay (days) can besubstituted.

Living Location Home, Nursing CareFacility, Homeless

Admission Date Day (dd), month (mm), andyear (yyyy) of admission ifhospitalised


Discharge Date Day (dd), month (mm), andyear (yyyy) of discharge ifhospitalised


DischargeDestination

Home, Nursing CareFacility, RehabilitationFacility, Homeless, Other

Performance Site Hospital Outpatient,Hospital Inpatient,Ambulatory Centre, Office

226Christian

-Alexan

der

Beh

rendtet

al.

Urgency of theProcedure

Elective, Urgent, Emergent Elective ¼ planned/scheduled procedure;urgent ¼ surgery within24 hrs of admission orpatient can’t be discharged;emergency ¼ surgery within6 hrs of admission

Provider Specialty Angiologist/VascularMedicine, Cardiologist,Radiologist, Surgeon,Other

Multidisciplinary TeamDecision (MTD)achieved prior to theProcedure?

Type of Procedure Surgical Bypass,Endarterectomy, Catheterbased Intervention, Hybrid

6) Bypass Surgical BypassLocation forRight/Left Leg

Ax-fem, Ax-bifem, Aorto-fem,Aorto-bifem, Fem-fem, fem-ATK popliteal, Fem-BTKpopliteal, Fem-tibial, Pop-tibial, Fem/Pop-DP/plantar

Proximal and Distalanastomosis location:Supra-aortic, ThoracicAorta, Abdominal Aorta,Common Iliac Artery,External Iliac Artery,Internal Iliac Artery,Common Femoral Artery,Deep Femoral Artery,Superficial Femoral Artery,Popliteal Artery AboveKnee, Popliteal ArteryBelowKnee, TibioperonealTrunk, Anterior TibialArtery, Posterior TibialArtery, Peroneal Artery,Pedal Artery, Plantar,Other

Conduit Prosthetic, Vein Prosthetic: Polyester,PTFE, heparin bonded;Cryopreserved allograft;Vein: great saphenousreversed, in situ,translocated; femoralvein; small saphenous;arm vein

Add number of veinsegements used or UDIidentification ofprosthetic graft

7)Endarterectomy

EndarterectomyLocation forRight/Left Leg

Aorta, common iliac, externaliliac, common femoral,profunda femoral, superficialfemoral, Popliteal

Continued

Intern

ational

Consortiu

mofVascu

larRegistries

227

Table 1-continued

Category Variable Reporting Level 1 Reporting Level 2 Reporting Level 3 Comments ReferencePatch Used No, Yes None, Polyester, PTFE,

heparin bonded, Bovinepericardium, Autogenousvein

UDI identification ofprosthetic patch

8) CatheterbasedIntervention

Access Site(s) Femoral, Popliteal, Pedal,Arm

Femoral Retrograde,Femoral Antegrade,SFA, Popliteal, DorsalPedal, Posterior Tibial,Brachial, Radial,Axillary, Graft, FemoralRetro to Antegrade,Femoral Ante toRetrograde

Largest SheathSize

French size number

Intervention Type Balloon angioplasty, drugcoated balloon, bare metalstent, drug eluting stent,covered stent, atherectomy,other

Balloon angioplasty, drugcoated balloon, baremetal stent, drug elutingstent, mechanicalthrombectomy, coveredstent, brachytherapy,atherectomy, laserassisted angioplasty,aspiration, scoringballoon, cutting balloon,cryoplasty, other

GUDID identificationof device

Planned AdjunctProcedure

No, Yes None, ThrombolysisPharmacological,Thrombolysis Mechanical,Suction Thrombectomy,Embolic ProtectionDevice, IVUS, CTO-Device,Bypass, Endarterectomy


Closure Device No, Yes If Yes: List Device Name GUDID identificationof device

UnplannedProcedure forComplication

Bailout Stent, Bailout Stentgraft

None, ThrombolysisPharmacological,Thrombolysis Mechanical,Suction Thrombectomy,Embolic ProtectionDevice, IVUS, CTO-Device,Bypass, Endarterectomy


228Christian

-Alexan

der

Beh

rendtet

al.

Final TechnicalResult

Successful (Stenosis � 30%),Stenosis > 30% or 10mmGradient, Target LesionOcclusion, Failure (unable tocross or deploy device)

Bollinger Score: Completeocclusion (1), Stenosis >50% of lumen (2), Stenosisof 25e50% of lumen (3),� 25% stenosis of lumen(4), Failure (unable tocross or deploy device)

Including only treated region Bollinger A, Breddin K, Hess H,Heystraten FM, Kollath J,Konttila A, et al.Semiquantitative assessment oflower limb atherosclerosis fromroutine angiographic images.Atherosclerosis. 1981;38(3e4):339e46.

9)Complications

UnplannedAmputation forRight/Left Leg

No, Minor Amputation,Major Amputation

Level: Toe(s),transmetatarsal, belowknee, through knee,above knee, higher of R/Lleg

Amputation that resultedfrom a complication orabrupt change in diseaseseverity that was notanticipated at the time of theprocedure

Site or GraftThrombosis forRight/Left Leg

No, Yes If Yes: No Treatment,Medical Treatment,Interventional Treatment,Surgical Treatment

Site or GraftStenosis forRight/Left Leg

No, Yes If Yes: No Treatment,Medical Treatment,InterventionalTreatment, SurgicalTreatment

Graft Stenosis is definedas � 70% diameter reducingstenosis by DopplerUltrasonography, CTAngiography, MRAngiography or Fluoroscopy

DistalEmbolisation forRight/Left Leg


Target LesionDissection forRight/Left Leg


Device Failure No, Yes If Yes: Failure to Deploy,Fracture, Rupture, Other

Bleeding,Hematoma,Pseudoaneurysm

No, Yes If Yes: Minor, Transfusion,Thrombin Injection,Surgical Treatment

CompartmentSyndrome

No, Yes If Yes: Medical, SurgicalTreatment

Wound Infection No, Yes If Yes: Medical, SurgicalTreatment

MyocardialInfarction

No, Yes If Yes: Troponin only(NSTEMI), ECG (STEMI) orClinical Symptoms

Continued

Intern

ational

Consortiu

mofVascu

larRegistries

229

Table 1-continued

Category Variable Reporting Level 1 Reporting Level 2 Reporting Level 3 Comments ReferenceStroke No, Yes If Yes: Minor, Major Modified Rankin level As defined byEaston et al.

(2009).Easton JD, Saver JL, Albers GW,Alberts MJ, Chaturvedi S,Feldmann E, et al. Definition andevaluation of transient ischemicattack: a scientific statement forhealthcare professionals fromthe American Heart Association/American Stroke AssociationStroke Council; Council onCardiovascular Surgery andAnesthesia; Council onCardiovascular Radiology andIntervention; Council onCardiovascular Nursing; and theInterdisciplinary Council onPeripheral Vascular Disease. TheAmerican Academy of Neurologyaffirms the value of thisstatement as an educational toolfor neurologists. Stroke; a journalof cerebral circulation.2009;40(6):2276e93.

New DialysisRequired

No, Yes No, Acute In hospital only,Chronic Dialysis

Any acute renal replacementtherapy for acute kidneyinjury (intermittent/continuous,haemodiafiltration,haemofiltration,haemodialysis etc.).

Death No, Yes If Yes: Death caused byprocedure?

10) Follow up Follow up Date Day (dd), month (mm), andyear (yyyy) of the procedure

Used to calculate duration offollow up for subsequentanalysis and de-identifieddata sharing. If regulations donot allow date collection,follow up duration (months)can be substituted.

Death No, Yes If Yes: Date of Death If Yes: Procedurerelated, Not Procedurerelated

230Christian

-Alexan

der

Beh

rendtet

al.

Functional Status Full activity, Light Work,Self Care, Assisted Care,Bedbound

Add disease specificquality of life survey,such as Vascu-Qol-6

A person’s level offunctioning in terms of theirability to care for themself,daily activity, and physicalability (walking, workingetc.).

Oken MM, Creech RH, TormeyDC, Horton J, Davis TE,McFadden ET, et al. Toxicity andresponse criteria of the EasternCooperative Oncology Group.Am J Clin Oncol. 1982;5(6):649e55.Nordanstig J, Wann-Hansson C,Karlsson J, Lundström M,Petterson M, Morgan MBF.Vascular Quality of LifeQuestionnaire-6 faciliateshealth-related quality of lifeassessment in peripheral arterialdisease. J VascSurg 2014;59:700-7.

Ambulation Fully ambulatory,Ambulate with Prosthesis,Ambulate with Assistivedevice, Wheelchair,Bedbound

Add walking distancesurvey such as WalkingImprovementQuestionnaire or GPSmonitored walking

Symptoms forRight/Left Leg

Modified Rutherford-Classification: Asymptomatic,Mild Claudication, ModerateClaudication (> 200m),Severe Claudication (<200m), Ischaemic Rest Pain,Ulcer/Necrosis, Non-healingAmputation, Both Ulcer/Non-healing Amputation, AcuteIschaemia

Level 1 Modified Rutherford !need to ref RAPID for themodified Rutherford!

Foot Infection forRight/Left Leg

No, Yes If Yes: Grade 0 None,Grade 1 Mild, Grade 2Moderate, Grade 3 Severe

Mills, Joseph L. Sr., MD, et al.(2014). The Society for VascularSurgery Lower ExtremityThreatened Limb ClassificationSystem: Risk Stratification Basedon Wound, Ischemia, and FootInfection (WIfI). Journal ofVascular Surgery, Volume 59(Issue 1) pp.220-pp.234.e2,https://doi.org/10.1016/j.jvs.2013.08.003

Continued

Intern

ational

Consortiu

mofVascu

larRegistries

231



Table 1-continued

Category Variable Reporting Level 1 Reporting Level 2 Reporting Level 3 Comments ReferenceTissue LossSeverity for Right/Left Leg

If Yes: None, Grade 1,Shallow, Grade 2, Deep,Grade 3, Extensive

Mills, Joseph L. Sr., MD, et al.(2014). The Society for VascularSurgery Lower ExtremityThreatened Limb ClassificationSystem: Risk Stratification Basedon Wound, Ischemia, and FootInfection (WIfI). Journal ofVascular Surgery, Volume 59(Issue 1) pp.220-pp.234.e2,https://doi.org/10.1016/j.jvs.2013.08.003

Amputation forRight/Left Leg

No, Minor, Major If Yes: Level: Toe(s),Transmetatarsal, BelowKnee, Through Knee,Above Knee, Higher ifRight/Left Leg

If Yes: Date ofAmputation

Graft/Site Patencyfor Right/Left Leg

No, Yes Method of Assessment(Clinical Examination, ABI,Ultrasound, MRA, CTA,Fluoroscopy, Other)

Patency of the treated indexlesion.

Graft/Site Re-Intervention forRight/Left Leg

No, Yes If Yes: Interventional,Surgical, Both

If Yes: Date of Re-Intervention

Major AdverseCardiac Events(MACE, MACCE)

No, Yes If Yes: Stroke, MI, CABG,PCI

If Yes: Date ofCoronary Procedure orDeath

Composite endpoint: Anymajor adverse cardiac orcerebrovascular event(MACE, MACCE) such as allcause death, myocardialinfarction, stroke, or coronaryvessel revascularisation.

Conte MS, Geraghty PJ, BradburyAW, Hevelone ND, Lipsitz SR,Moneta GL, et al. Suggestedobjective performance goals andclinical trial design for evaluatingcatheter-based treatment ofcritical limb ischemia. J VascSurg. 2009;50(6):1462e73 e1-3.Kip KE, Hollabaugh K, MarroquinOC, Williams DO. The problemwith composite end points incardiovascular studies: the storyof major adverse cardiac eventsand percutaneous coronaryintervention. Journal of theAmerican College of Cardiology.2008;51(7):701e7.

Procedure relatedRe-Admission

No, Yes If Yes: For infection,Treatment Failure,Systemic Complication

If Yes: Date of Re-Hospitalisation

Any hospital readmission orrehospitalisation afterdischarge.

232Christian

-Alexan

der

Beh

rendtet

al.



Anklebrachial

index

forRight/

LeftLeg

HighestABI�

1.3,

ABI<

1.3

and�

0.9,

ABI<

0.9

and�

0.7,

ABI<

0.4,treated

leg

ABIexactmeasuremen

tR/L

leg

ABIandTB

Iexact

measuremen

tR/L

leg

Abo

yans

V,Criqu

iMH,Abrah

amP,AllisonMA,Creag

erMA,

Diehm

C,et

al.Measurement

andinterpretation

ofthean

kle-

brachial

index:

ascientific

statem

entfrom

theAmerican

HeartAssociation

.Circulation

.20

12;126

(24):289

0e90

9.

International Consortium of Vascular Registries 233

(see Table 1). Some data elements, such as urgency oftreatment and type of procedure were judged sufficientlyimportant to always be required (level 1).

Statistical analyses were performed with SPSS Statisticssoftware version 23.0 (IBM, Armonk, NY, USA).

RESULTS

Thirty-one experts were contacted and 25 accepted andcompleted the first online survey. In total, 187 items weresubmitted by them and were included in the panel dis-cussion (Table 2). The items were reviewed by the expertpanel and subsequently sorted into 11 main topics: (i) pa-tient characteristics; (ii) comorbidities; (iii) current medica-tions; (iv) lesion treated; (v) procedure; (vi) bypass; (vii)endarterectomy; (viii) catheter based intervention; (ix)complications; (x) follow up. The panel comprised vascularsurgeons representing their national registry in ICVR, fromthree continents, 14 countries, and 18 institutions. The finalnumber of data elements was not specified a priori. Allpanel experts (100%) completed rounds 1e4 and 18 (90%)completed round 5. The Delphi process (Fig. 1) resulted inthe ICVR suggested data set for PAD revascularisation reg-istries with different levels of potential detail for each var-iable (Table 1). After two Delphi rounds, 68 items weredesignated twice as “agree” or “strongly agree”, whereas 60items were designated twice as “disagree” or “stronglydisagree”. After five Delphi rounds, a total of 79 items wereincluded in the recommended data set, of which 65 wereincluded in the level 1 version, with an additional 14included with more data specificity at the level 2 and 3versions (Table 1). For example interventional device datacan be recorded as level 1, which reports devices by class(plain angioplasty, drug coated balloon angioplasty, athe-rectomy, stent, etc.); level 2, which includes adjuncts suchas embolic protection; and level 3, which records the GlobalUnique Device Identification Database.

It was recommended that all registries create an optionto indicate that the state of the variable is “unknown”, inorder to differentiate omitted from unknown data, and notforce users to choose an option when it is unclear. Forsimplicity, the “unknown” options for each variable have notbeen included in Table 1.

DISCUSSION

In this modified Delphi study with international experts,consensus was achieved on items to be collected in regis-tries on peripheral arterial revascularisation. Sixty-five itemswere recommended for a minimum core (level 1) data setwith an additional 14 variables with increased specificationrecommended for the optimum dataset (level 2e3). Ac-cording to the existing literature and methodological rec-ommendations, the minimum and optimum number forDelphi studies is two rounds. Regarding the panel size, atleast 6e11 members are usually recommended to workefficiently. In this Delphi study more experts were included(n ¼ 25) and there were more rounds (n ¼ 5), emphasizingthe rigor of the approach to address the complexity of the

Table 2. 187 items submitted by the expert panel or identified bythe literature review.Date of admissionDate of procedureDate of dischargeMode of admission (e.g. emergency vs. elective)/Urgency of the procedurePerformance siteIntensive care unit LOSSpecial discipline responsible for hospital treatmentHospital capacity/volume/teaching statusDischarge destinationAge (Birth Date)Sex/GenderIncomeOccupationHousingFunctional statusAmbulationNursing statusMigration backgroundRace and ethnicityEducationHealth insurance statusWalking distanceExistence of rest painExistence of ulcersExistence of gangreneExistence of infectionFontaine classification (symptoms of index leg)Rutherford classification (symptoms of index leg)Tissue loss severity (for index leg)WIfI ScoreTexas classificationWound depthAngiosomeAcute limb ischaemia (ALI)Ankle brachial index (ABI)Toe pressureOscillographicsTranscutaneous oxygen measurement (tcp O2)Duplex ultrasoundContrast enhanced CT angiography (SCTA)Contrast enhanced MR angiography (MRA)Invasive digital subtraction angiography (DSA)TASC classificationExact location of treated lesions/artery treated or bypassedSide for each lesion treatedExact location of all lesions (even untreated)Grade of stenosisLength of stenosisInflow quantificationOutflow quantification (from treated artery)Genuine vessel or graft/artery vs. graftPretreatment of lesionLaboratory findings (e.g. cholesterol, HDL, platelet count,haemoglobin)Ever smoked/Tobacco useCurrent smoking/Tobacco useDiabetes type 1Diabetes type 2Hypertension

HyperlipidemiaWeight (or Body Mass Index)Height (or Body Mass Index)ASA Grade (Risk score: American Society of Anesthesiologists)Renal functionDialysisChronic obstructive pulmonary disease (COPD)Myocardial infarction (MI)Congestive heart failure (CHF)Atrial fibrillation or flutter (AF)Cardiac arrhythmiaPacemaker, defibrillator, orthopedic endoprosthesis or otherartificial materialCoronary artery disease (CAD)/Ischaemic heart diseaseHistory of stroke or TIAHistory of PAD revascularisationHistory of acute limb ischaemiaHistory of amputationOpen wound or wound infectionHistory of vascular proceduresAspirinP2Y12/Clopidogrel/Other platelet inhibitorOther platelet inhibitorsVitamin K antagonistsNew oral anticoagulantsLow dose heparin (not procedural)High dose heparin (not procedural)Beta blockersACE inhibitors or sartansStatin/Lipid lowering agentsPGE1 infusionsCilostazolNaftidrofuryl oxalatePentoxifyllineInositol nicotinateSpecial discipline performing the procedureLevel of residency supervisionPre-treatment of lesionSide of interventionPrincipal anaesthesia techniqueIntra-procedural heparinAdditional anaesthesia techniqueOperation timeType of main endovascular procedureHybrid procedureType of devices usedInstructions for Use (IFU) followedDetailed device information (includes length and diameterof different devices)Atherectomy devicePTA deviceStent deviceDistal protection deviceEmbolic protection deviceChronic total occlusion (CTO) deviceAccess site and approachAccess guidanceLargest sheath sizeUse of duplex ultrasound (access)Information about sheath usedInformation about guidewire usedPlanned adjunctive procedure


ThrombolysisMechanical thrombectomyAdditional approach (e.g. Nitro)Completion angiography performedFinal technical result (includes technical success)Dose area product (DAP)/Radiation dosageTreatment aborted or incompleteAcute conversion to open surgeryPatencyClosure deviceType of main open procedureHybrid procedureAccess site and approachDiathermia or ligature used for cut downExistence of infectionConduitType of bypassType of prosthetic grafts usedType and location of vein graftInstructions for Use (IFU) followedType of surgical suture materialProximal anastomosisDistal anastomosisBlood lossRe-intervention of bypassCompletion angiography performedTreatment aborted or incompleteAcute conversion to endovascular procedurePatencyDissection(Pseudo)AneurysmArteriovenous fistulaDistal embolisationPerforationBleedingMyocardial infarction (MACE)Death (MACE)Stroke or TIA (MACE)Major amputation (MALE)Minor amputation (MALE)Re-operation or re-interventionTransfusionAcute limb ischaemia/Lower extremity ischaemiaDevice fracture or ruptureCompartment syndromeNerve injuryWound infection or graft infectionLymphoedemaLymph fistula or seromaRe-intervention open surgeryRe-intervention endovascularPneumoniaDeep venous thrombosis (DVT)Pulmonary embolismAcute renal replacement therapyDeliriumStent or graft thrombosisGastrointestinal complicationsMACEMALEPatencyLimb salvage

Quality of lifeWalking distanceRehospitalisationRe-interventionInfectionAnkle brachial index (ABI)Destination at discharge/discharge destination


research questions. For example, the additional threerounds resulted in 58 positive or negative recommenda-tions that influenced the final result. An existing and wellestablished international research collaboration of ICVR wasused in order to include a high proportion of vascular reg-istry experts. In addition to two face to face meetings, allcomments of the panel experts were shared electronically.This innovative approach made the processes rigorous andefficient, considering the fact that international expertsfrom different time zones participated in this Delphi study.

Importantly, the Delphi process resulted in a reductionfrom 187 items originally included in PAD registries acrossthe ICVR to 79 items in the optimum data set. This is still alarge number of variables after the consensus process andmay pose a hurdle owing to the burden of data collection.However, considering the complexity of the PAD, and theabundance of medical and surgical treatment alternatives, amore comprehensive data collection is necessary to allowmeaningful data analysis. Ultimately, the trade-off betweencomplexity and practicality was challenging and proved tobe even more difficult when designing a registry databasefor PAD than for AAA or carotid artery disease.

The key to collaboration and data sharing among regis-tries is harmonisation of data elements, definitions, andmethod for similarly recording each variable. While eachregistry would prefer to register the most detailed datapossible, real world practice and current lack of ability toeasily extract uniform data from all electronic medical re-cord (EMR) systems makes this impractical. For this reason,it is valuable to recommend not only a minimum core dataset (and uniform definitions) that can be used by all regis-tries, but also to recommend more detailed categories(levels) for data recording as registries mature and EMRextraction becomes more feasible. It is important that thesehigher levels of data collection harmonise with core levelsso that all data can be merged at some level. The authorsbelieve that the current proposal of core to optimum datacollection “levels” is a novel contribution that could bevaluable for other specialties.

It is recognised that the number and selection of dataelements is contingent on the intended uses of the registry.A PAD revascularisation registry designed for qualityimprovement would probably have different elements orfewer granular data than one established for clinicalresearch or device evaluation. In this work there was anattempt to balance the competing interests of inclusivenesswith the practicality of data entry. The concept of differentlevels of modern web based registries with contingentvariables already enables efficient data entry, but there ismuch work to be done. In the future, it is expected that

Figure 1. Flow chart of the modified Delphi study.


registries will integrate with EMR systems and claims datato allow more automated data capture to minimise thework of data entry. Clearly, such future developments facingbig-data applications will need to meticulously deal withdata privacy and safety concerns.16

Harmonisation of registries will allow for more meaningfulcomparisons of practice patterns, medical device perfor-mance, and outcomes across countries. Such collaborationwill improve our ability to generate real world evidence anddesign registry based studies of peripheral vascular in-terventions. In the future, registry based studies may sup-plement the evidence gained from RCTs and prospectivecohort studies. Similar work has been reported by the Reg-istry Assessment of Peripheral Arterial Devices (RAPID)group for multispecialty collaboration within USA.17,18 TheICVR-recommended data set has many agreements withdata elements and definitions in RAPID, which is focused ondevice evaluation for peripheral vascular intervention inUSA. The current ICVR database recommendations extendthose of RAPID to encompass both open and endovascularrevascularisation and for international studies, while stillallowing device evaluation. These efforts provide an impor-tant opportunity for global harmonisation of clinical data toimprove vascular health care. Existing ICVR registry mem-bers are committed to adopting these data elements as thenext stage of evolution for ICVR.

CONCLUSIONS

This large scale modified Delphi study among internationalvascular registry specialists achieved a consensus agree-ment on a minimum core and optimum data set for regis-tries evaluating peripheral arterial revascularisation. Itreduced the overall number of initially suggested variablesby nearly half. Global harmonisation of registry infrastruc-ture and definition of items will overcome limitationsrelated to single country investigations and has the poten-tial to speed up and enhance acquisition of real world ev-idence. National registries in the ICVR plan to incorporatethese core data elements into their PAD registries to in-crease the opportunity for future collaboration.

ACKNOWLEDGEMENTS

The authors thank Thea Schwaneberg (University MedicalCentre Hamburg-Eppendorf, Hamburg, Germany) and theIDOMENEO-Consortium (University of Hamburg, Hamburg,Germany) for the valuable comments and editing, whichgreatly improved the manuscript. The authors also thankAxel Larena-Avellaneda (University Heart Centre Hamburg,Hamburg, Germany), Thomas Troëng (Uppsala, Sweden),Andres Schanzer (University of Massachusetts, Massachu-setts, United States), and Reinhart T. Grundmann (Bur-ghausen, Germany) for participating in this survey inaddition to the authors of this manuscript.


REFERENCES

1 Fowkes FG, Rudan D, Rudan I, Aboyans V, Denenberg JO,McDermott MM, et al. Comparison of global estimates ofprevalence and risk factors for peripheral artery disease in2000 and 2010: a systematic review and analysis. Lancet2013;382:1329e40.

2 Gerhard-Herman MD, Gornik HL, Barrett C, Barshes NR,Corriere MA, Drachman DE, et al. 2016 AHA/ACC guideline onthe management of patients with lower extremity peripheralartery disease: a report of the American College of Cardiology/American Heart Association Task force on clinical practiceguidelines. J Am Coll Cardiol 2017;69:e71e126.

3 Society for Vascular Surgery Lower Extremity GuidelinesWriting Group, Conte MS, Pomposelli FB, Clair DG, Geraghty PJ,McKinsey JF, et al. Society for Vascular Surgery practiceguidelines for atherosclerotic occlusive disease of the lowerextremities: management of asymptomatic disease and clau-dication. J Vasc Surg 2015;61:2Se41S.

4 Aboyans V, Ricco JB, Bartelink MEL, Bjorck M, Brodmann M,Cohnert T, et al. Editor’s Choice - 2017 ESC guidelines on thediagnosis and treatment of peripheral arterial diseases, incollaboration with the European Society for Vascular Surgery(ESVS). Eur J Vasc Endovasc Surg 2018;55:305e68.

5 Behrendt CA, Heidemann F, Riess HC, Stoberock K, Debus SE.Registry and health insurance claims data in vascular researchand quality improvement. Vasa 2017;46:11e5.

6 Sedrakyan A, Cronenwett JL, Venermo M, Kraiss L, Marinac-Dabic D, Bjorck M. An international vascular registry infra-structure for medical device evaluation and surveillance. J VascSurg 2017;65:1220e2.

7 Beck AW, Sedrakyan A, Mao J, Venermo M, Faizer R, Debus S,et al. Variations in abdominal aortic aneurysm care: a reportfrom the International Consortium of Vascular Registries. Cir-culation 2016;134:1948e58.

8 Venermo M, Wang G, Sedrakyan A, Mao J, Eldrup N,DeMartino R, et al. Editor’s Choice - carotid stenosis treatment:variation in international practice patterns. Eur J Vasc EndovascSurg 2017;53:511e9.

9 Sedrakyan A, Cronenwett JL, Venermo M, Kraiss L, Marinac-Dabic D, Bjorck M. An International vascular registry infra-structure for medical device evaluation and surveillance. Eur JVasc Endovasc Surg 2017;53:600e2.

10 Waggoner J, Carline JD, Durning SJ. Is there a consensus onconsensus methodology? Descriptions and recommendationsfor future consensus research. Acad Med 2016;91:663e8.

11 Eleftheriadou V, Thomas K, van Geel N, Hamzavi I, Lim H,Suzuki T, et al. Developing core outcome set for vitiligo clinicaltrials: international e-Delphi consensus. Pigment Cell Mela-noma Res 2015;28:363e9.

12 Wallace SJ, Worrall L, Rose T, Le Dorze G. Core outcomes inaphasia treatment research: an e-Delphi consensus study ofinternational aphasia researchers. Am J Speech Lang Pathol2016;25:S729e42.

13 Al Wattar BH, Tamilselvan K, Khan R, Kelso A, Sinha A, Pirie AM,et al. Development of a core outcome set for epilepsy inpregnancy (E-CORE): a national multi-stakeholder modifiedDelphi consensus study. BJOG 2017;124:661e7.

14 INSITE Collaborators. Diagnosis and management of iliac arteryendofibrosis: results of a Delphi consensus study. Eur J VascEndovasc Surg 2016;52:90e8.

15 Golan R, Bernstein A, Sedrakyan A, Daskivich TJ, Du DT,Ehdaie B, et al. Development of a nationally-representativecoordinated registry network for prostate ablation technolo-gies. J Urol 2018;199:1488e93.

16 Behrendt CA, Pridöhl H, Schaar K, Federrath H, Debus ES.Clinical registers in the twenty-first century: balancing act be-tween data protection and feasibility? Chirurg 2017;88:944e9.

17 Jones WS, Krucoff MW, Morales P, Wilgus RW, Heath AH,Williams MF, et al. Registry Assessment of Peripheral Inter-ventional Devices (RAPID): registry assessment of peripheralinterventional devices core data elements. J Vasc Surg2018;67:637e44. e30.

18 Jones WS, Krucoff MW, Morales P, Wilgus RW, Heath AH,Williams MF, et al. Registry assessment of peripheral interven-tional devices (RAPID) e registry assessment of peripheral inter-ventional devices core data elements. Circ J 2018;82:316e22.

http://refhub.elsevier.com/S1078-5884(18)30213-2/sref1






























































































Documents

International Consortium of Vascular Registries Consensus