Renal denervation in hypertension patients: Proceedings from an expert consensus roundtable cosponsored by SCAI and NKF.

Abstract

Despite the availability of efficacious pharmacological treatments, hypertension remains the leading global cause of death and disability,1 and rates of blood pressure (BP) control are stagnant, if not modestly declining. In the United States alone, for example, more than one-half of individuals with hypertension do not meet societal or guideline-directed BP goals, representing more than 29 million people.2, 3 Although reasons are multifactorial, patient non-adherence, physician inertia, as well as barriers such as lack of social support, depression and complex polypharmacy regimens are major contributors to lack of BP goal achievement. Interventional strategies, such as catheter-based renal denervation (RDN) using radiofrequency energy, ultrasound, or perivascular injection of neurotoxic agents, are gaining increasing attention. The rationale for RDN is to interrupt the activity of afferent and efferent sympathetic nerves located in the perivascular space of the renal arteries, reducing the sympathetic nervous system's influence on renal vascular resistance and renin release and sodium reabsorption that contribute to the maintenance and progression of hypertension.4 Following a succession of promising exploratory trials, enthusiasm for the efficacy of RDN was tempered by the neutral results of the sham-controlled, randomized SYMPLICITY-3 HTN trial in 2014, which demonstrated the safety of RDN but not a significant BP reduction relative to control.5 Amidst persistent interest in the potential benefit of RDN and lessons learned regarding trial conduct, procedural technique and the study population6 a new generation of randomized, sham-controlled trials has been performed (SPYRAL HTN-OFF MED, SPYRAL HTN-ON MED, SPYRAL HTN-OFF MED PIVOTAL, RADIANCE-HTN SOLO and RADIANCE-HTN TRIO).7-11 Despite varying procedural techniques, these trials demonstrated consistent and clinically meaningful reductions in ambulatory and office BP with RDN in both the presence and absence of anti-hypertensive medications. Together, these studies have renewed interest in device-based therapies for hypertension, and additional trials are ongoing. While RDN is included as a therapeutic option in the latest ESC/ESH guidelines,12 it remains an investigational technology in the United States. Against this background, and given the potential for this procedure, there is a need to address questions regarding the possible role of RDN as a therapeutic option in addition to medical therapy and lifestyle interventions. The assessment needs to be multifactorial and take into account the current factors influencing poor hypertension control, healthcare stakeholders, and, importantly, patient preferences. The present communication represents considerations and consensus views from a roundtable discussion between leaders in hypertension and interventional cardiology. The consensus conference was supported by the National Kidney Foundation and the Society for Cardiovascular Angiography and Interventions; however, the content of this report represents solely the opinions of the consensus committee members. Hypertension guidelines vary in their precise definitions and terminology. The ACC/AHA 2017 recommendations consider a BP ≥ 120/80 mm Hg as “elevated” and ≥ 130/80 as “stage 1 hypertension”,13 whereas the ESC/ESH document classifies BP as normal up to 130/80 mm Hg and assigns the label “high-normal” to the segment SBP 130–139 mm Hg or DBP 80–89 mm Hg.12 The differences are more in degree than in kind, as there is almost universal agreement that reducing BP will lower the risk of CV disease and mortality, as demonstrated in several meta-analyses.14, 15 In practice, there is broad consensus that a range 130–140/80 mm Hg would be acceptable.12, 13 Nevertheless, targets have not been consistently defined across guidelines. For instance, the American College of Physicians and American Academy of Family Physicians Joint Guideline recommends an SBP target of < 150 mm Hg in lower risk hypertensive adults aged 60 years or older.16 Irrespective of the definition, ample data suggest these targets are underachieved in routine clinical practice. In the United States, hypertension control rates (BP < 140/90 mm Hg) are currently not achieved in nearly 60% of all individuals with hypertension,3 irrespective of treatment. Although control rates approached 70% in 2014 among those prescribed medications, there has been a continual decline thereafter (Figure 1).3 Several reasons likely account for persistently low control rates. Although the effectiveness of pharmacotherapy varies between and within drug classes,17, 18 the lack of efficacious drugs is not perceived as the main problem. Physician inertia has long been recognized as a contributing factor: a national survey of ambulatory-care data from 2005 to 2012 on adults with BP > 140/90 mm Hg found that only one out of six uncontrolled patients experienced intensification of their drug regimen during this period.19 Analyses of prescription records paint a similarly discouraging story.20 Complete or partial non-adherence to treatment regimens also contributes significantly to low rates of BP control. Analyses of prescription refills from national claims databases in the US suggest that around one-third of treated hypertensive individuals are nonadherent to their medication regimens.21 Many other countries show similar nonadherence rates.22 Nonadherence rates seem highest among young adults, and adherence improves with increasing age.3 Adherence rates also vary among drug classes, and it has long been known that escalating medication number, especially when combined with multiple daily doses, are major barriers to adherence.23-25 Furthermore, adherence to medications is dynamic, and patients may often be partially adherent (not taking all medications every day) rather than entirely non-adherent. For example, even among individuals who are 80% adherent to their medication regimens, the pattern of pill-taking can vary widely (Figure 2), with possibly relevant differences in clinical impact.22 Finally, aside from non-adherence to medical therapy, avoidance of lifestyle and dietary modifications (e.g., exercise, weight loss, modifying salt, and alcohol intake) also contribute to poor long-term BP control. The largely asymptomatic nature of hypertension adds to the difficulty of controlling BP. Disparities in care are particularly relevant for socio-economically disadvantaged groups. Among factors predictive of higher control rates are access to medical insurance and regular visits to healthcare centers.3 Not coincidentally, hypertension rates among African-American individuals are among the highest in the world, also perhaps partly due to genetic factors.13 Community outreach activities, combining trusted sites for BP interventions, physician-pharmacist collaboration, and the availability of effective and low-cost antihypertensive drug regimens have demonstrated promising results.26, 27 Still, a greater need is assuring access to medical coverage among these disadvantaged groups. When considering BP targets and reductions, the panel recognized that increasing understanding of the multifactorial nature, treatment and behavioral factors in hypertension has led to more sophisticated views on the topic. Visit-to-visit variability over time in systolic or diastolic BP has been shown to be associated with worse CV and renal outcomes, as well with a greater mortality risk. The associations hold for variations in the range of as little as 5–10 mm Hg.28-30 Furthermore, the long-term sustainability of BP reductions imparts significant clinical implications. It is becoming clear that time within or very close to BP target has benefits on cardiovascular (CV) risk reduction. Time spent in an SBP target range of 110–130 mm Hg (replacing the target value of 120 mm Hg) was recently shown to be associated with a lower risk of major adverse CV events.31 Such findings have led to the suggested introduction of time in target range (TTR) to take into account BP variability over time.31, 32 TTR was recognized by the panel as a possible endpoint in hypertension clinical trials. However, the consensus was that even though TTR seems to matter for outcomes, it is beyond the resources of most primary care physicians to measure and adjust this parameter in clinical practice. Further evaluation and a standardized method for evaluation are needed. As both TTR and BP variability are dependent, at least in part, on patient adherence to treatment,31, 33 these measurements deserve close scrutiny in trials of RDN or other device therapies. However, considering that the effects of these device-based interventions are independent of patient adherence to drug therapy, they may provide a unique and potentially meaningful advantage. A second, related question addressed by the panel constitutes a “meaningful reduction in BP” with any antihypertensive treatment. This has been similarly discussed in European consensus documents for RDN, in addition to an FDA Panel meeting.34 The panel agreed that the term is not easy to define; it is not a matter of setting an absolute threshold of systolic or diastolic BP reduction, but a “meaningful” reduction should also consider patient risk. In patients at higher baseline CV risk or with higher baseline BP, the absolute reduction in global CV and renal risk can be significant even with only modest reductions in BP (5–10 mm Hg SBP). It is also likely that a clinically meaningful BP reduction will vary between individuals depending on starting BP, risk profile and the specific therapy being administered, giving rise to complex evaluations of benefit versus risk. Accounting for individual patient risk will be very relevant when assessing the suitability of RDN candidates for the procedure. The panel considered it unhelpful to specify an absolute parameter that is highly dependent on the starting point and patient conditions. In recent meta-analysis of 51 randomized trials involving pharmaceutical therapies, an absolute systolic BP reduction of 5 mm Hg was achieved with a single antihypertensive drug,35 yet this seemingly modest difference translates into an approximately 5% relative reduction in CV death and 10%–15% reductions in CV events and stroke. Similarly, a 10 mm Hg reduction in systolic BP is associated with a 13% decline in CV mortality and 20%–30% reductions in CV and stroke events, independent of initial BP level or coexisting conditions.36 At least similar reductions in office systolic BP have been observed with RDN, and the panel agreed that there is no reason to suggest the BP reductions achieved with RDN would not convey similar benefit in outcome to what has been observed with medical therapies. In addition to the magnitude of BP reduction, it is also essential to take into account stabilization or decrease in BP medications and changes in patient reported quality of life (QoL). Specific therapies for treatment of hypertension require a thoughtful work-up of the patient with uncontrolled BP. The panel discussed how multidisciplinary hypertension centers might play a role in a treatment landscape that includes the option of an RDN procedure. Accurate BP measurements are crucial for risk assessment and to evaluate the effectiveness of antihypertensive therapies. Clinic BP interpretation is limited by the white coat effect, inter-measurement variability, and the inability to uncover masked or nocturnal hypertension. These deficiencies may lead to suboptimal accuracy, with an impact on treatment decisions.37-39 Home BP measurements and ambulatory BP monitoring are more reliable, but patients need careful instruction and periodic reinforcement regarding how to perform measurements. Moreover, despite initial adherence, performance of home BP monitoring often wanes when patients lose interest in continuing to measure BP. There is also a relevant financial impact of self-measurement devices among patients with lower socioeconomic status, who often are those who represent the greatest need for BP control. The rapid development of wearable devices is expected to contribute to more complete and patient-friendly monitoring in the future. The panel emphasized that most patients with hypertension encountered in routine clinical care will not have difficult-to-treat, let alone resistant, hypertension. In the large majority of patients, diet and life-style modifications and a two-drug regimen, ideally in the form of a single-pill as recommended in ACC guidelines,13 can be expected to reduce BP. The need to add a third medication, or in the future to refer for RDN, may not apply to these patients. For this reason, the panel recommended that the term “uncontrolled hypertension” should be preferred over “resistant hypertension” until treatment options have been exhausted. Evaluation for secondary causes need not be performed on every hypertensive patient. However, when evaluating patients with true resistant hypertension (defined as office SBP/DBP ≥ 130/80 mm Hg despite prescription of ≥3 antihypertensive medications at optimal doses including a diuretic if possible, or a requirement of ≥4 medications to control BP),13 testing for autonomous aldosterone secretion should be a priority. The prevalence of primary aldosteronism in patients with resistant hypertension is above 20% but largely unrecognized.40 Even among academic centers, <10% of patients with resistant hypertension are screened for primary aldosteronism, and early diagnosis remains uncommon.41, 42 Beyond patients with true resistant hypertension, it is also reasonable to evaluate secondary causes in young hypertensive individuals, patients with accelerated or abrupt-onset hypertension, and in patients with signs, symptoms or laboratory findings indicating a discrete etiology. The presence of hypertension-mediated organ damage or severe comorbidities will intensify the need for BP lowering. Most hypertensive patients are evaluated by a primary care provider and are not referred further for specialist care. When referrals are made, evidence suggests that referral patterns are ad hoc and rely on anecdote and familiarity among practitioners and patients. Referrals are often made to general cardiologists or nephrologists, who may have different priorities and areas of expertise. A number of patients self-refer to hypertension centers. Many of the factors predicting uncontrolled hypertension (age, lack of healthcare insurance, not having seen a provider in the past year) also forecast the low likelihood of consulting a specialist.3 Appropriate structures will need to be set up to improve access to care. Capacity also remains an issue. There are approximately 1200 AHSCP (American Hypertension Specialist Certification Program) Certified Clinical Hypertension Specialists and 22 AHA Certified Hypertensive Center programs to meet needs of a population of more than 30 million uncontrolled hypertensive patents in the United States.43, 44 Community-based specialists are rare, and much of the expertise is currently concentrated at academic and large tertiary-care centers. The paucity of hypertension experts should not be a problem for the majority of patients with hypertension, and the panel noted that management of poorly controlled BP often does not require hypertension specialists. But serious educational efforts directed at primary care are necessary to improve treatment patterns, reduce physician inertia, and enable referrals for RDN, should the option for such therapy become available. Medical training programs should teach appropriate tests and patient characterization. Greater support for continuing medical education activities focusing on hypertension is necessary. A highly important objective is to generate a feedback loop with patients, for example, by enabling them to submit BP readings on a regular basis to inform treatment decisions.45 A protocolized approach to subsequent medication intensification is advisable to achieve treatment goals efficiently and at scale.46, 47 For an interventional treatment such as RDN, peer support networks for patients may be a helpful tool allowing patients to benefit from adequate information and expectations. Understanding the impact of TTR also underscores the individual responsibility for BP management and the overall importance of patient involvement. Although currently understudied in hypertension, the panel expects that patient preferences and patient-reported outcomes measures (PROM) will become of major importance in future treatment decisions. Establishing and communicating individual risk profiles would aid discussions as not all patients will share the same motivation or treatment expectations. PRO measurements, which include self-reported quality of life measures, are increasingly being included in clinical trials.48 For instance, some RDN studies have employed the Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36), Beck Depression Inventory (BDI), and Spielberger's state and trait anxiety questionnaires.49, 50 The group recommended that the inclusion of such measurements in RDN trials should be explored. At present, however, standardized PRO measures are lacking for uncontrolled hypertension treatment, whether with medications alone, or following an intervention like RDN. The primary outcome in hypertension trials is generally BP reduction but it is not clear if this represents the highest priority to patients compared with, for example, stroke avoidance or pill reduction. While many patients are not prescribed antihypertensive medications, cannot tolerate side effects or otherwise refuse to take medications, it is important to understand procedure avoidance as well as medication nonadherence from the patient perspective. This will need dedicated efforts to develop hypertension-specific questionnaires and to ensure that PRO measures are validated and sensitive for specific geographies, cultures and ethnicities. The FDA has long recommended that substantial patient input should be included in PROM development. 51 After the lack of demonstrable efficacy of RDN compared with antihypertensive therapy alone in the SYMPLICITY HTN 3 trial, identified confounders related to procedural technique, medication variability, and selected patient subgroups have been addressed in contemporary trial design.6, 52 A new generation of randomized, sham-controlled trials with revised design and conduct have demonstrated significant BP reductions following RDN in hypertensive patients, both in the presence and absence of antihypertensive drug therapy (Figure 3), and notably with two different procedural methods.7-11 Results of additional studies using radiofrequency and ultrasound ablation are awaited (SPYRAL HTN-ON MED Study [NCT02439775] and REQUIRE [NCT02918305]), in addition to ongoing study with alcohol-mediated RDN (TARGET BP 1 [NCT02910414]). A consistent observation across these trials has been the “always on” effect with RDN to provide a consistent reduction in BP over all time points within a 24-h interval in comparison with a control group.7, 9, 10, 53, 54 These findings not only eliminate issues of patient nonadherence and limitations associated with pharmacokinetic profiles and dosing regimens of drug therapies, but may have special benefit to individuals whose BP phenotype conveys especially high risk (e.g., nocturnal and/or early morning hypertension).55 A more consistent reduction in BP may also translate into improved TTR and stability of BP control. In parallel with reductions in BP, several analyses have also supported an association between RDN and a reduced need for antihypertensive drugs. In the RADIANCE-SOLO trial, fewer medications were administered in the RDN group to achieve the same BP-lowering effect at 6 months.56 Significantly more patients in the RDN group required no medications at 6 months than in the sham-controlled population (p = 0.008). Similar reduction in a medication index has been reported applying win-ratio methods to the SPYRAL ON-MED pilot study, comparing RDN versus sham control in the presence of uncontrolled hypertension despite prescribed medications.57 The win ratio allows the different groups in a clinical trial to be compared in a fashion which gives appropriate priority to the more clinically important events in composite endpoints.58 In this analysis, RDN scored more wins than losses on ambulatory SBP, office SBP, and change in medication index, in decreasing order of prioritization, leading to a total win ratio of 2.78 (95% confidence interval 1.58–5.48; Figure 4). The long-term impact of medication changes following RDN was a common interest among panel members. Assessment of durability in BP reduction is recognized as a difficult to ascertain measure given changes in pharmacotherapy, patient lifestyle, and co-existing health conditions. To date, the largest evidence is derived from the Global Symplicity Registry, a survey of RDN among a broad, unselected patient population with uncontrolled hypertension. Through 3 years of follow-up, RDN was associated with sustained and significant reductions in both office and ambulatory BP independent of an escalation in medication burden.59 Whether the reduction in BP is consistent across patients with varied risk and comorbidities is a focus of ongoing investigation. In the Global Symplicity Registry, a similar magnitude of BP reduction was observed following RDN among patients with varying CV risk and comorbidities.60, 61 Although RDN studies have not been designed to compare long-term clinical outcomes, a recent meta-analysis of existing randomized trials demonstrated that RDN is associated with a BP reduction at least similar to that of a single-pill dual-agent combination62 and may translate to an estimated 10% relative reduction in lifetime risk of CV events and 7.5% relative reduction in all-cause mortality.63 Regarding safety following RDN, studies to date have not identified any unexpected device- or procedure-related risks. In particular, randomized trials have not demonstrated an increased risk of vascular complications, decline in renal function, or later identification of renal artery stenosis. In a meta-analysis of 50 trials that included a net population of 5769 patients and 10,249 patient-years of data, the annual incidence of renal artery stenting following RF renal denervation was 0.20%, a rate comparable to the reported natural incidence of events in an untreated hypertensive population.64 Specifically, no adverse events were reported in the distal arteries in the subgroup of 396 patients from 9 reports treated with radiofrequency RDN beyond the main bifurcation. In a separate analysis of 14 studies and 511 subjects using computed x-ray tomography after a median of 11 months post procedure (range: 1–36 months), only one significant stenosis was identified that required stenting (0.2%). Registry data over 3 years similarly indicate no changes in age-related renal function60 and a 0.3% rate of renal artery stenosis.59 Safety data in patients with CKD are sparse, but small studies have provided no reason for concern in this population as recently reviewed.65 When discussing renal safety, the panel noted that although a 10 mm Hg reduction in BP may be expected to increase creatinine levels, there is no negative effect on renal function apparent in the trials and registry data. Understandably, uncertainties regarding long-term safety and effectiveness remain given the current state of evidence. Most of the currently available information is derived from the Global Symplicity Registry that, by design, has limitations related to absence of a comparator group, unmeasured confounders, and potential for selection bias. Finally, as with pharmacotherapy, individual responses to RDN vary and the questions regarding how to measure response and what markers may predict response are being intensely investigated. Ambulatory BP has been an efficient way to measure response: in the SPYRAL HTN-OFF MED trial, reductions in ambulatory BP and office BP correlated relatively well for the changes with RDN, but there was no significant correlation in the sham group.7 However, in RDN trials, in general, about one third of the population did not experience at least a 5 mm Hg SBP reduction over 24-h.66 Variability in BP, reductions in medications and variable adherence complicate the assessment of whether these patients represent true non-responders. Several possible predictors of response to RDN have been proposed34 but progress has been limited. In SPYRAL HTN-OFF MED, ambulatory heart rate (HR) above the median (>73.5 bpm) was predictive of reduction in average daytime SBP, daytime DBP, and office SBP.67 Recently, both plasma renin activity and aldosterone levels were significantly reduced following RDN compared with sham control; higher baseline plasma renin activity was associated with a significantly greater reduction in both office and 24-h systolic BP.68 In RADIANCE-HTN SOLO, baseline nighttime SBP and its variability predicted response, albeit with low sensitivity, and the HR criterion had insufficient predictive value to be of use.69 Finally, given that no reliable procedural measure exists to confirm effective nerve ablation, the extent to which incomplete denervation contributes to clinical non-responsiveness is uncertain. Peri-procedurally, renal nerve stimulation may help predict the size of the SBP reduction70 and has helped identify the most responsive ablation sites in animal models,71 but it is complex, time consuming, and labor intensive. The panel agreed that the availability of a reliable, validated method to predict treatment response would have a large impact on the recommendation of RDN to suitable populations. To date, investigation of RDN has largely been limited to a selected patient population with uncontrolled systolic and diastolic hypertension. Yet as more data become available, the clinical perspective of what characterizes the most suitable patient candidate is evolving rapidly. Instead of a niche procedure for the most intractable cases of elevated BP, there may be a larger role for RDN as part of the stepped-care toolkit for hypertension management. The meaningful criterion for the value of the intervention would be to provide greater benefit, not to placebo (existing trials have furnished sufficient evidence for the efficacy) but to a comparator medication or other intervention. When considering RDN as an alternative to adding another hypertensive medication, the benefit and risk must be compared. Currently available data indicate that the main potential advantages of an interventional therapy are tolerability and adherence (Table 1). Although current randomized trials have been limited to patients with combined hypertension, there was a lack of consensus among panel members whether the threshold to consider RDN should be based on both systolic and diastolic criteria (i.e., reflecting currently available RCTs) or systolic BP alone (i.e., reflecting registry data). The panel also agreed that attempts to identify “responders” to RDN may not be possible given the limited progress with establishing reliable predictors of response. Instead, a standardized definition of response would need to be developed. A more formal definition should account for method and timing of BP assessment, relative versus absolute BP reductions and variance between patient-level and population-based assessments. The panel also unanimously agreed that patient preference, which has been largely unaddressed in hypertension trials, deserves important consideration. Although investigation of the topic is limited in RDN studies, pooled surveys among hypertensive patients indicate that approximately one-third of patients prefer a catheter-based intervention over escalation of medications for uncontrolled BP.72 Moreover, patient preference for RDN appears independent of medication number or BP severity of hypertension; indeed, preference for RDN was highest among those with hypertension who were not taking medications. Patient preference for RDN was associated with a greater wish to lower BP and a greater appreciation and understanding of the benefit of BP-lowering. Personal experience with the effects of elevated BP and symptoms further influenced preferences. These results also indicate that physician endorsement influences patients' preferences as has been seen in other areas of medicine.73 In comparison, however, a disparity exists between patients and healthcare providers regarding referral for RDN; among healthcare providers, recommendation of RDN is greatest for those patients taking multiple medications and/or with severe hypertension. In addition to inadequately controlled BP and patient preference, evidence of hypertensive end-organ damage (albuminuria, LVH by echocardiography, declining eGFR) should be an important priority for consideration of RDN. The risk profile should also take into account atherosclerotic CV disease risk and established coronary artery disease. Patient age may also be an important consideration; adherence rates to medication are low in young individuals,3, 22 who may experience lifelong benefit from the RDN procedure, if studies confirm durability. Managing patient expectations of RDN will also be important. Most eligible patients on previous polypharmacy will not be able to discontinue all medications even after successful RDN. Instead, a common scenario may be performance of RDN versus an additional drug medication. A reduction in the need for pills59 would be a sign of therapeutic effectiveness, but it is unclear how large a reduction would be needed for patients to judge the procedure successful. The maximum acceptable levels of treatment-related risks that patients are willing to accept in exchange for specific treatment benefits need to be explored, and also the minimum level of treatment benefit required for patients to be willing to accept specific treatment-related risks. The win ratio analysis may be insightful for managing expectations and can be adapted to patients' priorities. To avoid physicians unduly influencing patients' choice and to respect the patient perspective, education and information should be developed and provided before a patient enters the consultation. In summary, when considering the possible place of RDN in the therapeutic landscape for hypertension, after ruling out secondary causes of hypertension, the procedure should prioritize hypertensive individuals with elevated CV risk and those with established end organ damage. Those with hypertension who are unable to take medications, for whatever reason, might be a further priority group. Although no confirmatory studies exist to date, potential effectiveness of RDN in indications beyond hypertension (e.g., diabetes, heart failure, and sleep apnea) may also inform patient selection. RDN should be a shared decision that includes perspectives from the treating physician, a discussion of risk and benefits with the patient and incorporating patient preferences. Ideally the procedure would be endorsed by more than one healthcare provider. The intervention should be performed at an experienced specialist interventional center with catheter-based equipment and appropriate imaging. A good triage system is important to increase access to care without overwhelming resources. A proposal for an Interventional Hypertension Center of Excellence (IHCE) network is shown in Figure 5. Finally, from an administrative and billing standpoint, the panel confirmed that an ICD-10 code specific for uncontrolled hypertension would represent an important contribution to clinical practice, research and reimbursement. There is widespread consensus on the effectiveness and safety of RDN. Should the interventional therapy gain approval in the United States, it has the potential for substantial public health impact to address an epidemic of uncontrolled hypertension. The challenge will be to identify appropriate patients and to build interdisciplinary networks for referrals. Although RDN remains an investigational therapy, it seems appropriate to strive towards a resolution of as many outstanding questions as possible in advance to best inform patient selection and the potential role in clinical practice. The Expert Consensus Roundtable was supported by an unrestricted grant from Medtronic. Writing assistance was provided by Pelle Stolt, PhD, Basel Switzerland. George Bakris: Consultant for Merck, Bayer, Vascular Dynamics, KBP Biosciences, Ionis, Alnylam, Astra Zeneca, Quantum Genomics, Horizon, Novo Nordisk; Steering committee of trials supported by Bayer, Vascular Dynamics, Quantum Genomics, Alnylam and Novo Nordisk. Jan Basile: Consultant for Medtronic; Research grants from Eli Lilly, ReCor, and Ablative Solutions; Royalties from UpToDate. Michael J. Bloch: Consultant/Honoraria from Recor, Medtronic, Amgen, Esperion, Janssen, Amarin, and Bristol Meyers Squibb; Research Support from Recor. Debbie L. Cohen: Research support from Medtronic and ReCor. Cara East: Research support from Medtronic. Keith C. Ferdinand: Consultant for Amgen, Novartis, and Medtronic. Naomi Fisher: Consultant for Aktiia; Consultant and research support from Recor Medical. David E. Kandzari: Consulting honoraria from Cardiovascular Systems, Inc, Medtronic. Institutional research/grant support from Ablative Solutions, Boston Scientific, Biotronik, Cardiovascular Systems, Medtronic, Orbus Neich, and Teleflex. Ajay Kirtane: Institutional funding to Columbia University and/or Cardiovascular Research Foundation from Medtronic, Boston Scientific, Abbott Vascular, Abiomed, CSI, CathWorks, Siemens, Philips, ReCor Medical, Neurotronic. In addition to research grants, institutional funding includes fees paid to Columbia University and/or Cardiovascular Research Foundation for consulting and/or speaking engagements in which Dr. Kirtane controlled the content. Personal: Consulting from IMDS; Travel Expenses/Meals from Medtronic, Boston Scientific, Abbott Vascular, Abiomed, CSI, CathWorks, Siemens, Philips, ReCor Medical, Chiesi, OpSens, Zoll, and Regeneron. David P. Lee: Consultant for Medtronic and DeepQure; Research grants from Medtronic and Ablative Solutions. Florian Rader: Consultant for Recor, Medtronic, and MyoKardia. Eric Secemsky: Consulting/Speaking: Abbott, Bayer, BD, Boston Scientific, Cook, CSI, Inari, Janssen, Medtronic, Philips, and VentureMed; Research grants from NIH/NHLBI K23HL150290, Harvard Medical School's Shore Faculty Development Award, AstraZeneca, BD, Boston Scientific, Cook, CSI, Laminate Medical, Medtronic, and Philips. Raymond R. Townsend: Consultant for Medtronic, Axio, Regeneron, Ionis, and IMEDIC; Royalties from UpToDate; Grants from NIH. Joseph A. Vassalotti: Consultant for RenalytixAI, plc. Michael A. Weber: Consultant to Medtronic, ReCor, Ablative Solutions, Johnson & Johnson, Urovant, and Omron. Kerry Willis and Gary Puckrein have no conflicts to disclose.