Are We Implanting Too Many Primary-Prevention Implantable Cardioverter-Defibrillators?

Canadian Cardiovascular Society/Canadian Heart Failure Society guidelines recommend optimizing patients with heart failure with reduced ejection fraction (HFrEF) on guideline directed medical therapy (GDMT) prior to implanting primary prevention implantable cardioverter defibrillators (ICDs). However, often GDMT is suboptimal at the time of ICD implantation, and is associated with reduced survival. Furthermore, with evidence for declining sudden cardiac death (SCD) rates in patients with cardiomyopathy, there is a need for better risk stratification to determine the indication for ICD.1 This raises concern for potentially unnecessary ICD implantation in this population, and efforts are needed to address this issue in clinical practice. Canadian Cardiovascular Society/Canadian Heart Failure Society guidelines recommend optimizing patients with heart failure with reduced ejection fraction (HFrEF) on guideline directed medical therapy (GDMT) prior to implanting primary prevention implantable cardioverter defibrillators (ICDs). However, often GDMT is suboptimal at the time of ICD implantation, and is associated with reduced survival. Furthermore, with evidence for declining sudden cardiac death (SCD) rates in patients with cardiomyopathy, there is a need for better risk stratification to determine the indication for ICD.1 This raises concern for potentially unnecessary ICD implantation in this population, and efforts are needed to address this issue in clinical practice. The rate of ICD implantation in Canada has increased from approximately 1.0 per 10,000 in 2005, to 1.3 per 10,000 in 2018-2019 (based on data from the Canadian Institute for Health Information; https://www.cihi.ca). This trend likely reflects guideline recommendations on primary and secondary prevention ICD implantation, based on several large trials in the late 1990's to 2010's demonstrating significant reduction in all-cause mortality, SCD, and ventricular tachyarrhythmias (VTA) with primary prevention ICDs for patients with HFrEF.1Butler J. Talha K.M. Aktas M.K. et al.Role of implantable cardioverter defibrillator in heart failure with contemporary medical therapy.Circ Heart Fail. 2022; 15e009634https://doi.org/10.1161/CIRCHEARTFAILURE.122.009634Crossref Scopus (1) Google Scholar The majority of these trials were conducted in ischemic cardiomyopathy (ICM). However, the most contemporary trial in this area, the DANISH study, did not demonstrate a mortality benefit with primary prevention ICD in patients with non-ischemic cardiomyopathy.1Butler J. Talha K.M. Aktas M.K. et al.Role of implantable cardioverter defibrillator in heart failure with contemporary medical therapy.Circ Heart Fail. 2022; 15e009634https://doi.org/10.1161/CIRCHEARTFAILURE.122.009634Crossref Scopus (1) Google Scholar Part of the reason for this lack of benefit may have been the lower SCD rate. In fact, studies have demonstrated that the baseline absolute risk for SCD has declined from 1999 (publication of MUSTT trial) to 2016 (publication of DANISH trial). One systematic review reported the baseline risk for SCD, assessed by the annualized event rate (AER) of SCD in the control arms of these randomized controlled trials (RCTs), declined from 5.7–7.8% in the MADIT and MUSTT trials (start of randomization 1990), to 2.5–3.0% in the DEFINITE and SCD-HeFT trials (start of randomization 1997–1998), to 1.5% in the DANISH trial (start of randomization 2008).2Disertori M. Mase M. Rigoni M. et al.Declining clinical benefit of ICD in heart failure patients: Temporary trend of mortality outcomes from randomized controlled trials.J of Card. 2020; 75: 148-154Abstract Full Text Full Text PDF PubMed Google Scholar Much of this reduction in the risk of SCD could initially be attributed to the use of angiotensin converting enzyme inhibitor (ACEi), beta blocker (βB), and mineralocorticoid receptor antagonist (MRA) therapies. Subsequent to the publication of the primary prevention ICD trials, the use of an angiotensin receptor neprilysin inhibitor (ARNI) and sodium glucose cotransporter 2 inhibitors (SGLT2i), in combination with other GDMT, have significantly further reduced the risk of SCD.1Butler J. Talha K.M. Aktas M.K. et al.Role of implantable cardioverter defibrillator in heart failure with contemporary medical therapy.Circ Heart Fail. 2022; 15e009634https://doi.org/10.1161/CIRCHEARTFAILURE.122.009634Crossref Scopus (1) Google Scholar Therefore, prescription of GDMT would be expected to significantly improve the outcome of these patients, thus diminishing the benefit of primary prevention ICDs and potentially altering the indication for implanting ICDs.1Butler J. Talha K.M. Aktas M.K. et al.Role of implantable cardioverter defibrillator in heart failure with contemporary medical therapy.Circ Heart Fail. 2022; 15e009634https://doi.org/10.1161/CIRCHEARTFAILURE.122.009634Crossref Scopus (1) Google Scholar Importantly βBs, ACEis, ARNI and MRAs have all been demonstrated to increase left ventricular ejection fraction (LVEF) over time.1Butler J. Talha K.M. Aktas M.K. et al.Role of implantable cardioverter defibrillator in heart failure with contemporary medical therapy.Circ Heart Fail. 2022; 15e009634https://doi.org/10.1161/CIRCHEARTFAILURE.122.009634Crossref Scopus (1) Google Scholar This further supports that prescription of these GDMTs may reduce the need for ICD implantation. The role of coronary revascularization is important with regards to the management of ICM. This is a management strategy that should be considered before implanting a primary prevention ICD. In fact, consideration of the indication for coronary artery bypass graft (CABG) surgery should be performed in parallel with the institution of GDMT. Further detailed discussion about this management strategy is beyond the scope of this paper. Despite guideline recommendations about the importance of GDMT, studies have demonstrated patients are not optimized (Table 1). A large study demonstrated only 61.1% of patients with HFrEF filled any GDMT prescription prior to primary prevention ICD implant.3Roth G.A. Poole J.E. Zaha R. et al.Use of guideline-directed medications for heart failure before cardiac defibrillator.J Am Coll Cardiol. 2016; 67: 1062-1069Crossref PubMed Scopus (52) Google Scholar This suggests possible early or unnecessary primary prevention ICD implantation in a population that has the potential to improve their LVEF and thereby no longer meet indications for ICD implantation had they achieved sufficient GDMT optimization.Table 1Reasons for suboptimal optimization of GDMTCategoryFactorsPatient-related factors•Presence of medical comorbidities•Inability to tolerate GDMT (due to reduced renal function; hyperkalemia; hypotension; bradycardia; other side effects)•Medical adherence concerns (due to cognitive impairment; low health literacy; or inability to afford medications)Provider-related factors•Lack of provider knowledge or experience with GDMT optimization•Fear of adverse effects in patients•Lack of time•Therapeutic inertiaHospital-level factors•Lack of dedicated interdisciplinary HF team•Lack of dedicated HF outpatient clinicSystem-level factors•Low prioritization of GDMT optimization (due to lack of GDMT performance metrics, or lack of financial incentives) Open table in a new tab There are important implications of reducing unnecessary primary prevention ICD implantation ranging from decreasing healthcare costs, to reducing procedure- and device-related complications, and to avoiding adverse outcomes which diminish quality of life.1Butler J. Talha K.M. Aktas M.K. et al.Role of implantable cardioverter defibrillator in heart failure with contemporary medical therapy.Circ Heart Fail. 2022; 15e009634https://doi.org/10.1161/CIRCHEARTFAILURE.122.009634Crossref Scopus (1) Google Scholar With respect to device-related complications, this also includes device recalls for reasons ranging from minor issues unlikely to cause injury or illness to those that can cause severe injury or death. Other non-complication considerations of ICD implantation that impact quality of life include the added burden of device follow-up appointments and the requirement for repeat procedures related to battery depletion. The consequences of potentially unnecessary ICD implantation are therefore not insignificant. Considering the above factors, we propose several potential strategies to address this issue in contemporary clinical practice (Figure 1). 1.Improve optimization of GDMT Opportunities exist at various timepoints in the care trajectory of patients with HFrEF, to optimize GDMT. In the ambulatory setting, there is evidence for heart failure (HF) management programs and dedicated multidisciplinary clinics to ensure GDMT optimization. Additional interventions, such as pharmacist-delivered reminders and targeted electronic health record alerts that provide specific guidance on HF medical management, have been shown to be effective methods of improving GDMT initiation compared to no alerts.4DiDomenico R.J. Vardeny O. Optimizing heart failure therapy requires all-hands-on-deck and a DASH of technology.Circ Heart Fail. 2023; 16e010886Crossref PubMed Scopus (0) Google Scholar In the inpatient setting, admission and discharge order sets, such as the Canadian Heart Failure Society order set (https://hfordersets.ca/), remind clinicians to initiate GDMT during hospitalization. The triaging of primary prevention ICD referrals also represents an opportunity to ensure adherence and uptitration of GDMT. Primary prevention ICD referrals should have standardized expectations for GDMT optimization at a system-wide level, ensuring the referring clinician lists all GDMT including name, dose, duration of therapy, and if not on target or maximally-tolerated dosing, an explanation for why. Furthermore, there may be benefit in expanding the role of allied health professionals (such as pharmacists or nurse practitioners) at the time of triaging the referral, to assist with GDMT optimization. Future clinic infrastructure should also focus on ensuring close collaboration between HF and ICD/Device clinics.2.Increase exposure to ambulatory HF practice in adult cardiology residency programs A recent survey-based study of 15 Canadian cardiology programs examined the degree of exposure to ambulatory HF patient management among cardiology trainees in Canada. The study found that the current state of HF training amongst Canadian cardiology residents was suboptimal. Only 3 programs (20%) required ≥ 2 ambulatory HF rotations; when HF rotations were provided, only 7 programs (47%) offered moderate or higher exposure to ambulatory nontransplant patients with HF (defined as ≥ 5 clinics/rotations).5Almufleh A. Turgeon R.D. Ducharme A. et al.Proposal for an ambulatory heart failure management curriculum for cardiology residency training programs.CJC Open. 2022; 4: 866-872Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar This is in spite of the fact that all institutions had a multidisciplinary HF clinic, and the majority (87%) had access to an inpatient HF service, a consultative HF service, and/or a specialist pharmacist. The authors concluded that greater exposure to ambulatory HF care is needed in cardiology residency, focusing on practical and experiential aspects of GDMT optimization. The authors of the survey are working to develop a program for cardiology training programs across Canada, to enable trainees to better manage this growing and increasingly complex patient population.3.Improve patient selection for primary prevention ICDs It has been demonstrated that between two-thirds to three-quarters of all implanted primary prevention devices have not delivered any shocks before battery replacement is required (between four and eight years). This emphasizes the importance of optimizing patient selection to determine who is at highest risk of arrhythmic events and therefore stands to benefit most from primary prevention ICD implantation. Presently, we rely solely on the LVEF to determine risk of VTA and SCD; however, this may not be adequate on its own as a predictor. More studies are needed to evaluate the role of clinical, biochemical, ECG, cardiac imaging and other variables to better predict SCD and VTA risk amongst potential primary prevention ICD recipients. Related to better risk stratification, is the need for updated RCTs to re-evaluate the degree of benefit of primary prevention ICD in patients with HFrEF who are on contemporary GDMT, stratified by ischemic and non-ischemic phenotypes. The reduction in SCD since the publication of the landmark primary prevention ICD trials, coupled with improved outcomes of contemporary GDMT, may point towards a diminishing benefit of primary prevention ICD implantation in the HFrEF population. Furthermore, evidence suggests that primary prevention ICDs are being implanted in patients with suboptimal GDMT. These underscore the fact that we may be implanting too many primary prevention ICDs in patients with HFrEF.