Author + information
- Received December 5, 2017
- Revision received March 7, 2018
- Accepted March 13, 2018
- Published online May 2, 2018.
- Ernest Spitzer, MDa,b,
- Eugène McFadden, MDa,c,
- Pascal Vranckx, MD, PhDd,
- Ton de Vries, MSca,
- Ben Ren, MD, PhDa,b,
- Carlos Collet, MDe,
- Yoshinobu Onuma, MD, PhDa,
- Hector M. Garcia-Garcia, MD, PhDf,
- Renato D. Lopes, MD, PhDg,
- Gregg W. Stone, MDh,
- Donald E. Cutlip, MDi,j and
- Patrick W. Serruys, MD, PhDk,∗ ()
- aCardialysis Core Laboratories and Clinical Trial Management, Rotterdam, the Netherlands
- bDepartment of Cardiology, Thoraxcenter, Erasmus University Medical Center, Rotterdam, the Netherlands
- cDepartment of Cardiology, Cork University Hospital, Cork, Ireland
- dDepartment of Cardiology and Critical Care Medicine, Hartcentrum Hasselt, Jessa Ziekenhuis, Faculty of Medicine and Life Sciences Hasselt University, Hasselt, Belgium
- eDepartment of Cardiology, Academic Medical Center, Amsterdam, the Netherlands
- fDepartment of Cardiology, MedStar Washington Hospital Center, Washington, DC
- gDivision of Cardiology, Duke University Medical Center/Duke Clinical Research Institute, Durham, North Carolina
- hClinical Trials Center, Cardiovascular Research Foundation and Division of Cardiology, Columbia University Medical Center, New York, New York
- iBaim Institute for Clinical Research, Boston, Massachusetts
- jBeth Israel Deaconess Medical Center, Boston, Massachusetts
- kInternational Centre for Circulatory Health, NHLI, Imperial College London, London, United Kingdom
- ↵∗Address for correspondence:
Prof. Patrick W. Serruys, International Centre for Circulatory Health, National Heart and Lung Institute, Imperial College London, 59 North Wharf Road, London W2 1LA, United Kingdom.
Patients in coronary intervention trials may require more than 1 procedure to complete the intended revascularization strategy. However, these staged interventions are not consistently defined. Standardized definitions are needed to allow meaningful comparisons of this outcome among trials. This document provides guidance on relevant parameters involving staged procedures, including minimum data collection and consistent classification of coronary procedures initially identified as staged; the aim is to achieve consistency among clinical trialists, sponsors, health authorities, and regulators. Definitions were developed jointly among representatives of academic institutions and clinical research organizations based on clinical trial experience and published literature. Reasons for staged procedures were identified and include baseline kidney function, contrast load and radiation exposure, lesion complexity, and patient or operator fatigue. Moreover, nonclinical reasons include procedure scheduling and reimbursement. Management of staged procedures should be a standalone section in clinical trial protocols and clinical events committee charters. These documents should clearly define a time window for staged procedures that allows latitude for local policies, while respecting accepted clinical guidelines, and consistency with study objectives. Investigators should document in the case report form the intent to stage a procedure, the lesions to be treated, and the reasons for staging, preferably before randomization. Ideally, all reinterventions, or at least all procedures performed after the recommended time window, those in which data suggest an anticipated procedure due to a worsening condition and those where a revascularization is attempted in the target vessel, should be reviewed by an independent clinical events committee.
Up to one-tenth of patients enrolled in coronary intervention trials require more than a single procedure to complete an intended percutaneous revascularization strategy due to multivessel disease (1,2). It would be ideal, both from patient and societal (health care economic) perspectives that all lesions requiring intervention could be treated in a single session. However, there are legitimate clinical and nonclinical reasons that may justify a staged procedure (3).
Coronary intervention trials aim to provide an unbiased comparison between a novel device and a predicate device, or to compare a percutaneous strategy with a surgical strategy. In trials designed to compare stent/scaffold platforms, consistency in the definition of reintervention is critical to ensure comparability among trials and to allow meaningful conclusions from pooled data and meta-analyses. This may be extended to trials comparing different percutaneous strategies (e.g., in the setting of ST-segment elevation myocardial infarction [STEMI]) or to those investigating combined percutaneous therapies (e.g., transcatheter aortic valve replacement and percutaneous coronary intervention [PCI]). The current absence of standardized definitions for staged procedures poses challenges for the interpretation of data among trials that involve staged interventions. Specifically, for 2 different studies, the same post-index intervention may be adjudicated as a staged procedure in one trial and as a re-intervention in another trial; with direct impact on the number of revascularizations that will be considered as an endpoint.
In the present document, we propose a standardized definition around data collection, and provide guidance for adjudicating staged procedures by clinical events committees (CECs). Clear guidance will also facilitate incorporation of trial data into clinical practice guidelines.
Representatives of academic institutions and clinical research organizations jointly elaborated this guidance document based on clinical trial experience and published reports. A systematic review was performed independently by 2 authors to identify time windows and definitions used in selected coronary intervention randomized trials published between 2007 and 2017 in 5 major clinical journals that randomized at least 1,000 patients and for which information on staged procedures was available online. Details are provided in Online Tables 1 and 2, Online Figure 1.
Circumstances associated with staged procedures
In clinical trials, as well as in routine practice, staged procedures allow completion of an optimal coronary revascularization strategy, when this is not possible or preferred in a single coronary intervention procedure. Such circumstances include the clinical presentation (e.g., STEMI as compared with non-STEMI or stable angina) (4); the baseline angiographic characteristics (e.g., unanticipated procedural complexity with the need for ancillary techniques such as rotational atherectomy) (4); patient-related factors (e.g., renal impairment, contrast and radiation exposure); and in the setting of chronic total occlusions (CTOs), where a strategy that involves a second attempt may be prospectively defined by protocol (3). Furthermore, nonclinical reasons such as logistic issues (e.g., an on-call setting where another STEMI patient is en route) and rare scenarios such as equipment failure may also play a role. Finally, reimbursement practices may influence the decision to “stage” a procedure or, more likely, may affect the timing of “staged” procedures. More specifically, if a second procedure is scheduled within a specific time window, the procedure may or may not be reimbursed in some jurisdictions. This practice is not scientifically justified, and consideration should be given to only including centers/countries that can comply with protocol requirements. An overview of the situations that may result in an additional, planned procedure is provided in Table 1.
Definition of a staged procedure
Both the patient-orientated composite endpoint and the device-orientated composite endpoint, proposed by the Academic Research Consortium (ARC) have become widely accepted in coronary intervention trials (5) (Central Illustration). However, for both definitions, the consistent classification of any repeat intervention, after the index procedure, is necessary for the inherent validity of cross-trial comparisons (5). We propose to define a staged procedure as a planned intervention performed after the first catheterization when it fulfills the following requirements: 1) the intent to stage is documented, provisionally or definitely, before or within 24 h after completion of the first procedure (Figure 1); 2) the lesion(s) to be treated during the staged procedure should be defined upfront and should not involve the index vessel, except in specific study designs, such as trials for left main disease or CTOs; 3) the procedure should be performed within the protocol-defined time frame; and 4) stability of symptoms is required between the first and the subsequent procedure(s), because acute ischemia (including worsening of angina) would disqualify the intervention as a staged procedure. Confirmed staged interventions are counted as “index” interventions, thus described in the baseline characteristics of the study population. Provisional staging refers to the intent to stage pending the results of subsequent ischemia (e.g., through fractional flow reserve or instantaneous wave-free ratio) or viability testing that may validate the requirement for staging (6).
Reinterventions not fulfilling these requirements should be considered potential revascularizations and should be adjudicated by an independent CEC, which will classify the procedure either as a staged procedure not fulfilling the protocol requirements (i.e., protocol violation but described as part of the index procedure), or as a revascularization and counted as a repeat intervention. Each study protocol should describe how staged procedures will be approached (Table 2). Guidance around basic parameters to capture this information is provided in the Online Table 3, including reporting the intent to stage, procedural data of the staged procedure, and if applicable, reason(s) for cancellation of a staged procedure when a planned procedure is not performed.
As shown in Figure 2 and Table 3, a systematic review including 15 PCI trials revealed that the time frames vary substantially among trials and are sometimes expressed in the protocol in terms of a “recommended” rather than a “prescriptive” timeline. Not all protocols mandate an independent CEC to review interventions reported by the investigator as staged, and to determine whether they comply with protocol requirements. Furthermore, clear, pre-defined rules to classify planned procedures performed outside the protocol-specified deadline are not specified.
This document does not aim to prescribe a universally applicable time frame for clinical trials. Preferred timing for performing staged procedures in clinical practice is addressed (to the extent that current evidence allows) in clinical practice guidelines, where clinical presentation and persistence of symptoms have a pivotal role in the decision-making process (4). We, however, consider it reasonable to suggest that all staged procedures in patients with stable angina as initial clinical presentation should be undertaken within 2 to 3 months, whereas a shorter timeframe may be more appropriate where the index was performed in the setting of acute coronary syndromes. Updated meta-analyses show better outcomes in patients presenting with STEMI who undergo complete revascularization either in a single procedure or in staged procedures when compared with patients with culprit-only PCI (7,8), although this benefit remains unproven when complicated with cardiogenic shock (9). Conversely, registry data suggest no difference in long-term outcomes among stable or non–ST-segment elevation acute coronary syndrome patients undergoing either single-session multivessel PCI or staged multivessel PCI (10). Recent trials in patients with left main or 3-vessel disease in which event adjudication was performed, such as the SYNTAX (SYNergy between percutaneous coronary intervention with TAXus and cardiac surgery) II trial and the EXCEL (Evaluation of XIENCE versus Coronary Artery Bypass Surgery for Effectiveness of Left Main Revascularization) trial (11), used 4 weeks as the upper time window (12). An ongoing largescale antithrombotic therapy trial in patients undergoing PCI, in which adjudication of events is not planned, uses 3 months as a sharp cutoff to differentiate among protocol-defined staged procedures and revascularizations, taking in consideration the potential variability among 100 interventional cardiology sites (13).
Background on lesion definitions
Index lesions are those intervened at the first catheterization and during subsequent staged procedures. When stents are used, the treated segment, including 5 mm proximal and 5 mm distal to the implanted device, is also referred to as the “target lesion.” Moreover, the target vessel refers to the entire major intervened coronary vessel (Figure 3). Reinterventions of the target lesion are considered target lesion revascularizations. When reinterventions are performed outside the target lesion, but in the same major coronary vessel, these are referred to as “target vessel non-target lesion revascularizations.” Moreover, reinterventions in a different vessel are defined as non-target vessel revascularizations.
Some scenarios require additional technical considerations, and any deviation from these would require a detailed description in a study protocol. Firstly, the inclusion of the proximal and distal 5 mm coronary portions from the edge of a device in the target lesion extends to the side branches. When a side branch originates outside the stented area but within the target lesion, the proximal portion of the side branch included in the target lesion should sum up to 5 mm from the edge of the stent (Figure 3B). When a side branch emerges from the stented area, the proximal 5 mm of the side branch are part of the target lesion (Figure 3C). Secondly, when the left main is treated at baseline, the entire left coronary system becomes the target vessel (both circumflex and left anterior descending [LAD] coronary arteries) (Figure 4A). Thirdly, if the intermediate branch is treated at index, the target vessel is formed by the intermediate branch and the left main (Figure 4A). Moreover, if the treated lesion is within 5 mm of the left main, then the left main also becomes a target lesion. Fourthly, when assessing arterial or venous grafts, the target vessel is defined by the insertion of the distal anastomosis. For example, if an aortocoronary bypass with a distal anastomosis in the right posterior descending coronary artery is treated during a first procedure, the native right coronary artery, as well as the bypass, are components of the target vessel (Figure 4B). Finally, when the proximal portions of the LAD, intermediate branch, or left circumflex are treated, if the implanted device is within 5 mm of the ostium of the treated artery, the proximal portion of the other vessel(s) up to 5 mm will be considered component(s) of the target lesion (Figure 3D). More specifically, if a proximal LAD has an implanted device 2 mm from its ostium, the proximal 3 mm of the left circumflex are considered part of the target lesion. However, exceptionally, the rest of the left circumflex is not analyzed as a target vessel. Thus, in this scenario, treatment of a distal circumflex in a subsequent procedure would qualify as a non-target vessel revascularization. Some of the preceding considerations are arbitrary; however, their use allows consistency when analyzing and reporting the data.
Staged coronary intervention in the index vessel
By convention, a staged procedure generally excludes further procedures at the lesions or in the vessels treated at the first procedure. There are valid reasons for such an approach. For example, inadvertent disruption of a metallic stent in a vulnerable position such as the right coronary ostium where the intended target lesion during a staged procedure is in the distal right coronary artery, or disruption of a bioresorbable scaffold implanted in the first procedure, where the intended target is located in a branch originating from the index scaffold. In both situations, iatrogenic complications are predictable. However, in all-comers trials where complex lesions and multivessel disease are observed in up to two-thirds of participants, there may be legitimate reasons to reconsider such a firm recommendation. Interestingly, in the EXCEL trial, the authors allowed staged procedures in the left coronary system after stenting the left main (11). Consequently, this needs to be taken into consideration when comparing results with other trials or when pooling data. Occasionally, when a second staged procedure is necessary, the vessels treated during the initial procedure and the first staged procedure are considered target vessels.
Chronic total occlusions
In contemporary CTO treatment, the concept of an “investment procedure” and the potential need for a second attempt (where no significant progress is made during the first procedure) is accepted practice. This is because despite advances in antegrade and retrograde techniques, some procedures will inevitably require a second attempt. Moreover, even when the initial procedure is successful, some experienced operators have advocated planned “relook” angiography with intravascular imaging in order to optimize previously implanted stents/scaffolds (e.g., where intravascular imaging demonstrates acquired malapposition). On the basis of the ARC criteria, any further procedure at the index lesion would be systematically classified as a repeat intervention (target lesion revascularization) and not as a staged procedure. Furthermore, where quantitative coronary angiography documents a lesion>70% diameter stenosis (thus, all failed initial attempts to open a CTO or the incidental documentation of a >70% stenosis on any repeat angiogram), any repeat intervention would be classified, based on ARC criteria, as clinically justified, irrespective of symptoms or evidence of inducible ischemia. Both situations illustrate the need to reconsider whether such procedures should be systematically classified as reinterventions or classified as “staged procedures” and thus considered as part of the “index procedure.”
ST-segment elevation myocardial infarction
An example of the impact of the heart team approach on the definition of a “staged” procedure is provided by a device study where the index procedure in performed on the “culprit lesion” in the setting of primary PCI for STEMI with the proviso that other lesions should be treated in a staged procedure with the assigned study device. Additional information may emerge after intervention in an acute setting, such as more precise detail on compliance or the detection of newly diagnosed diabetes that may override prior “conditional” decisions to pursue a percutaneous strategy or indeed to perform any further revascularization.
Furthermore, repeat diagnostic angiography in patients in stable condition at planned reintervention may demonstrate differences from the acute setting in, for example, reference vessel diameter or percent diameter stenosis, or could provide additional information through functional tests (e.g., fractional flow reserve, instantaneous wave-free ratio) that may legitimately preclude use of the assigned device or alternatively demonstrate that no intervention is clinically indicated. Thus, rather than a definitive statement at the index procedure, it may be more appropriate to adopt the “provisional staging” or “intent to stage” approach, where the decision made at the end of the index procedure is subject to modification based on a heart team discussion or in other protocol pre-defined circumstances.
Guidance for clinical events committees
An independent CEC is critical to ensure that revascularizations after the first intervention are correctly classified as staged procedures (these considered as index lesions) or reinterventions (common component of composite endpoints). Generally, any attempt (successful or not) to treat a lesion (i.e., implantation of a device, balloon angioplasty, thrombectomy, or even failed attempts to cross the lesion with a wire or device) should be counted as a PCI at any time point (i.e., index procedure including staged and any subsequent procedures). Planned invasive diagnostic procedures (e.g., follow-up optical coherence tomography) are not classified as interventions. The CEC charter of a clinical trial, which describes all adjudication procedures, should contain a prescribed algorithm for consistent triggering and precise definitions for adjudication of events. The CEC charter should define whether all staged procedures or only those that are triggered due to noncompliance with the protocol definitions (e.g., outside the time window) will be reviewed. Notwithstanding, if possible, the review of all cases is preferred, including a pre-adjudication review of the angiograms by an independent angiographic core laboratory. The following scenarios should be adjudicated as revascularizations and not as staged procedures: 1) any reintervention at the index lesion(s), which should be further classified either as clinically indicated or not. Second interventions after an investment procedure for CTOs or in the target vessel territory for left main disease, may be considered staged procedures in the target lesion or target vessel, respectively, if defined in the study protocol; 2) when a procedure reported by the investigator as staged was not previously planned; and 3) when the procedure occurred before the planned date due to documented objective unexpected worsening of ischemic signs and/or symptoms, and/or electrocardiographic or biomarker changes compatible with an acute coronary syndrome leading to an anticipated intervention. Interventions reported as staged procedures that are performed outside the time frame dictated by the protocol will be listed within protocol violations and further classified by the CEC as confirmed staged procedures or reinterventions. In the rare scenario in which a planned coronary artery bypass graft is performed after a first PCI (staged coronary artery bypass graft), the same conditions apply for classification of the event.
Reintervention during a planned catheterization in a patient in stable condition
Patients are sometimes exposed to a “relook and retouch strategy,” which seems to be the new “oculostenotic reflex” (14). This scenario refers to patients brought back to the catheterization laboratory for a planned staged procedure in a vessel different from the one already treated, or for intracoronary imaging tests (e.g., intravascular ultrasound or optical coherence tomography), which can be mandated or not by the protocol. In this scenario, the operator decides to reintervene at a previously treated lesion based on a perceived angiographic suboptimal result or more commonly on the basis of “drive-by” intravascular imaging findings. In this guidance document, we suggest that any intervention at an index lesion should be adjudicated as a reintervention, and further defined as clinically indicated or nonclinically indicated according to ancillary criteria.
Staged procedures are common in contemporaneous PCI trials. A standardized approach for defining staged procedures is needed in order to allow comparability among studies. This document provides guidance about relevant aspects around staged procedures aiming to achieve consistency among clinical trialists, sponsors, health authorities, and regulators. Management of staged procedures should be a standalone section in clinical trial protocols and independent CEC charters.
The authors would like to thank Hans Jonker at Cardialysis B.V. for helping in preparing the figures.
Dr. Onuma has been a member of an advisory board for Abbott Vascular. Dr. Garcia-Garcia has served as a consultant for Tryton; has received speaker fees from Volcano/Philips; and has received institutional grants from Biotronik, Medtronic, Infraredx, and Neovasc. Dr. Lopes has received personal fees for consulting from Bayer, Bristol-Myers Squibb, Daiichi Sankyo, GlaxoSmithKline, Merck & Co., Inc., Pfizer, Portola Pharmaceuticals, and Boehringer Ingelheim; and has received institutional grant support from Bristol-Myers Squibb, GlaxoSmithKline, Medtronic, and Pfizer. Prof. Serruys has served as a consultant to Abbott, AstraZeneca, Biotronik, Cardialysis, GLG Research, Medtronic, Sinomedical, Soc. Europa Dig. Publishing, Stentys, Svelte, Phillips/Volcano, St. Jude Medical, Qualimed, and Xeltis. Dr. Cutlip received research support from Celonova. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- Academic Research Consortium
- clinical events committee
- chronic total occlusion
- left anterior descending coronary artery
- percutaneous coronary intervention
- ST-segment elevation myocardial infarction
- Received December 5, 2017.
- Revision received March 7, 2018.
- Accepted March 13, 2018.
- 2018 American College of Cardiology Foundation
- Wykrzykowska J.J.,
- Kraak R.P.,
- Hofma S.H.,
- et al.
- Stefanini G.G.,
- Serruys P.W.,
- Silber S.,
- et al.
- Windecker S.,
- Kolh P.,
- Alfonso F.,
- et al.
- Cutlip D.E.,
- Windecker S.,
- Mehran R.,
- et al.
- Gotberg M.,
- Christiansen E.H.,
- Gudmundsdottir I.J.,
- et al.
- Bangalore S.,
- Toklu B.,
- Stone G.W.
- Elgendy I.Y.,
- Mahmoud A.N.,
- Kumbhani D.J.,
- Bhatt D.L.,
- Bavry A.A.
- Kolte D.,
- Sardar P.,
- Khera S.,
- et al.
- Toyota T.,
- Morimoto T.,
- Shiomi H.,
- for the CREDO-Kyoto PCI/CABG Registry Cohort-2 Investigators,
- et al.
- Stone G.W.,
- Sabik J.F.,
- Serruys P.W.,
- et al.
- Escaned J.,
- Collet C.,
- Ryan N.,
- et al.
- Vranckx P.,
- Valgimigli M.,
- Windecker S.,
- et al.
- Topol E.J.,
- Nissen S.E.
- Bonaa K.H.,
- Mannsverk J.,
- Wiseth R.,
- et al.
- von Birgelen C.,
- Kok M.M.,
- van der Heijden L.C.,
- et al.
- Xu B.,
- Gao R.,
- Yang Y.,
- et al.
- Kaiser C.,
- Galatius S.,
- Jeger R.,
- et al.
- Pilgrim T.,
- Heg D.,
- Roffi M.,
- et al.
- von Birgelen C.,
- Sen H.,
- Lam M.K.,
- et al.
- Park K.W.,
- Chae I.H.,
- Lim D.S.,
- et al.