The struggle against infections of cardiac implantable electrical devices: the burden of costs requires new personalized solutions

This editorial refers to ‘Infections and associated costs following cardiovascular implantable electronic device implantations: a nationwide cohort study’ by N. Clémenty et al., on pages 1974–1980. The number of cardiac implantable devices (CIEDs) has enormously increased in the past decades, with estimates of approximately 3.5 million pacemaker (PM)/implantable cardioverter-defibrillator (ICD) leads currently implanted in patients all around the world, and approximately 500 000–1 000 000 new leads implanted worldwide every year.1 In this issue of Europace, Clémenty et al.2 report the results of an economic analysis based on the French National Hospital Database estimating the costs associated with CIED-related infections (CIEDI) in new and replacement interventions from the perspective of the French social security system. Beyond being the first large European economic evaluation on this topic their results promote at least three important considerations. The first striking message is that management of CIEDI is associated with similar costs whenever they occur after de novo or replacement procedure, ranging at 12 month between €20 211 and €23 237. This is noteworthy, since a transvenous lead extraction (TLE) procedure should be easier in a de novo implant compared with the same procedure in patients who underwent several previous CIED procedures (at least for the shorter indwelling time of the leads) with less use of powered sheaths and a lower complication rate.3 Notably, overall costs were not significantly increased by extending the follow-up from 12–24 months, independently of the implanted CIED. Moreover, the DRG payment comprised on average 63–71% of the total mean cost, being drugs and/or implants reimbursed on top of this expenses. Taken together these considerations confirm the findings by Sohail et al.4 The authors analysed a cohort of 200 219 Medicare fee-for-service patients admitted for CIED generator implantation, replacement, or revision comparing the outcomes and costs associated with admission for patients with and without CIEDI. They found that CIEDIs were associated with an adjusted admission mortality rate ratio of 4.8–7.7 and an incremental admission costs of $14 360–16 498. The largest incremental cost was not associated with TLE or CIED reimplantation, while it derived from intensive care unit admission, drugs, and diagnostic process accounting for >40% of the difference. While at first look this could sound strange, it is perfectly in line with the issues connected with management of patients with CIEDI. The entire process is complex from CIEDI suspect to post-TLE management, requiring a team of experienced operators from the very beginning to appropriately interpret the clinical and instrumental findings and choose the best treatment for the patient. Notably, a similar scenario should be managed with an ‘hub-and-spoke’ organisation including referral centres and a centralized unit in charge of TLE interventions (Figure 1). For this reason, it is of paramount importance that the reimbursement system would properly cover the costs for the management of the patients occurring in all the involved centres. If this condition is not respected there could be some risk that the reimbursement policy could ‘interfere’ with physician choices in most complex patients for example: (i) by delaying TLE after several revisions, (ii) by performing TLE before appropriate diagnostic workup, (iii) by modifying reimplantation approach, (iv) by shortening in-hospital stay. These theoretical risks should require proper analysis of daily practice, taking into account all the complexity of managing this type of patients, for appropriately re-evaluate and interpret the evidence that more than one-fifth of the patients with CIEDI do not undergo TLE (28% in the study by Clémenty et al.2) while one-fourth to one-third are not reimplanted after TLE (35% in the same study).5 In other words, the reimbursement policy should always follow the most appropriate clinical choices and not being a hurdle influencing what is the best care, at any step of the chain of care, even if a great amount of resources is needed. The common observation that reimbursement practices can be inadequate in many complex cases, as well as the geographical heterogeneity in reimbursement policies are all factors that need to be addressed in a European perspective, overcoming the current situation.6 Health technology assessment is the specific tool for a comprehensive approach to this complex issue, involving all the stakeholders.7 Management of patient with a CIED infection in a Hub and Spoke system. CIEDI, cardiac implantable electrical device infection; TLE, transvenous lead extraction; TOE, transoesophageal echocardiography; TTE, transthoracic echocardiography. The second finding is that the authors did not find a significant difference in costs associated with CIEDI involving PM or ICD carriers. This is relevant since the average incidence of CIEDI is more than two-fold in ICD patients (about 1.33% vs. 0.56%) the ratio in terms of overall costs of CIEDI management ratio results to be inverted (due to the larger amount of implanted PMs).2 For this reason, we cannot target our efforts for CIEDI prevention on the basis of the carried device (PM vs. ICD) but appropriately evaluate the clinical characteristics of the patients at higher risk. At this regard, it is relevant to highlight that the risk factors for CIEDI are at the same time predictors of survival after TLE,8 underlying the need to improve our patient evaluation by associating to the prediction of CIEDI risk the prediction of outcome after CIEDI. The third point is that replacement procedures are associated with a two-fold risk of CIEDI with respect to de novo implants. This finding confirms previous data and highlights a limitation of this study. In this analysis, the authors classified in the same group the ‘box exchange’ procedures and the procedures entailing insertion/revision of at least one lead. The last subgroup of procedures was traditionally classified as ‘upgrade procedures’ in registries aimed at estimating the incidence of CIEDI.9 In the same studies, the incidence of CIEDI was significantly higher with respect to ‘box exchange procedures’.4,8,9 Since CIED ‘upgrades’ to ‘box exchange’ ratio is between one-fourth to one-third,9 it could be speculated that the gap in terms of CIEDI incidence between replacements and de novo procedures that was recorded by Clemency et al.2 could be attributed to this specific subgroup of upgrade procedures. This consideration brings important consequences from both a clinical and scientific perspective. In the first decade of 2000, after the recognition of the gap between de novo procedures and CIED replacements in terms of CIEDI, all the efforts of researchers and CIED manufacturers were targeted at decreasing the number of CIED replacements.1 This has been done not only by improving battery technology but also by refining CIED programming (e.g. reducing pacing output, implementing automatic management of pacing output, decreasing unnecessary ICD interventions, activating only the algorithm useful for each specific patient, etc.). All these efforts have dramatically decreased the number of elective ‘box exchange procedures’10 but all these interventions cannot affect the need for ‘upgrade’ procedures. If we want to look inside the group of the CIED ‘upgrade’ procedures, we would find quite different procedures: (i) real upgrades procedures, by addition of a lead in a new chamber (i.e. from single to dual chamber device or from dual chamber to a cardiac resynchronization therapy device) or by implementation of defibrillator capability (e.g. from single chamber PM to single chamber ICD); (ii) ‘adding-a-lead’ procedure, consisting in implanting an additional PM/ICD lead in the same chamber where a former lead performance was found inadequate or clearly failing; (iii) TLE assisted lead replacements (for non-functional leads or unnecessary leads); (iv) CIED revision for decubitus/impending infection; (v) TLE removal and re-implantation for CIEDI. On the basis of the differences characterizing each specific subgroup of CIED ‘upgrade’ procedures it emerges the need to identify what potential changes in CIED systems technology could have a positive impact on the need for these procedures, which are characterized by costs and clinical risks. From a technological point of view the current approach is to avoid the need for intravenous leads which constitute a weak and fragile component of CIED systems, by developing and validating the clinical use of subcutaneous ICDs and leadless PMs. However, this approach could affect the types of procedures previously classified as (ii) and (iii) procedures, but not the others. While the (iv) and (v) groups will never be affected until the development of therapies not requiring implantable devices the (i) group can be, at least partially, diminished by accurately improving patient selection and management.11 A completely new approach in CIED provision that could help reducing all these subgroups of CIED ‘upgrades’ could be based on development and distribution of modular CIED systems, with capability to add new functions, via software. Leadless intracardiac devices for pacing that can be added (or removed) independently one from each other, actually represent an interesting evolution for cardiac pacing/defibrillation provided that they can interact with sensors to obtain arrhythmia discrimination and proper atrioventricular and interventricular timing and can be coupled with subcutaneous or even wearable ICDs in order to allow defibrillation when needed (Figure 2). Example of a possible leadless defibrillator with cardiac resynchronization therapy system with modular interactive components. (A) atrial pacing device; (B) right ventricle pacing device; (C) left ventricle pacing device; (D) extravascular defibrillator. Conflict of interest: G.B. reported speaker’s fees of small amount from Medtronic, Boston and Boehringer. C.E. and I.D. reported no conflicts to disclose.