Outcomes of lead extraction without subsequent device reimplantation

Abstract

Aims

Outcomes among patients who do not receive device reimplantation after cardiovascular implantable electronic device (CIED) extraction have not been well studied. The present study aims to investigate the outcomes of patients without device reimplantation after lead extraction and device removal.

Methods and results

We retrospectively searched for consecutive patients who underwent CIED extraction at Mayo Clinic, Rochester, MN and University of California San Diego Medical Center from 2001 through 2012. Among the patients identified, we compared characteristics of those who did and did not have device reimplantation. The Kaplan–Meier survival was analysed. Among 678 patients, 97 patients had their device extracted without reimplantation during 1-year follow-up (‘no-reimplant group’). Median age was younger in the no-reimplant group (60.7 vs. 70.6 years; P < 0.001). The reasons for no reimplantation were as follows: no longer meeting criteria for CIED (48%), inappropriate device indication at initial implantation (23%), patient preference (17%), and unresolved device complications (12%). Three major arrhythmias were reported in the no-reimplant group. Overall survival in the no-reimplant group was significantly lower than in the reimplant group (60 vs. 93%; P < 0.001). Ongoing device-related complications [hazard ratio (HR), 3.91; 95% CI, 1.74–8.81; P = 0.001], infection (HR, 3.06; 95% CI, 1.24–7.52; P = 0.02), and concurrent dialysis (HR, 2.74; 95% CI, 1.12–6.71; P = 0.03) were associated with increased mortality. Of 31 deaths in the no-reimplant group, 1 was secondary to cardiac arrhythmia.

Conclusion

Fourteen per cent of patients who had device extraction did not undergo reimplantation mainly because they no longer met CIED indications. The high mortality in these patients is related to device complications and comorbid conditions, whereas mortality associated with arrhythmia is rare.

Introduction

The rates of cardiovascular implantable electronic device (CIED) implantation have increased substantially since their initial introduction in 1958.¹ The advancement of CIED technology and the expansion of implantation indications, such as the use of implantable cardioverter-defibrillators (ICDs)^2–4 and cardiac resynchronization therapy (CRT) in patients with heart failure and conduction abnormalities, led to this increase.^5–7 In parallel, the number of CIED infections that require device extraction to eradicate the source of infection has also increased.⁸ The vast majority of patients who have CIED lead removal undergo device reimplantation. Little is known, however, about how frequently patients undergo device extraction without subsequent reimplantation or the clinical characteristics and outcomes of these patients. To our knowledge, this study presents the first investigation of patients who underwent device extraction without reimplantation in two major centres in the USA, the decision-making involved in these cases, and the patients' clinical outcomes.

Methods

Study patients

We retrospectively searched for the records of consecutive patients who had their CIED extracted at Mayo Clinic, Rochester, MN and the University of California San Diego (UCSD) Medical Center between 2001 and 2012. Patients who did not survive to hospital discharge or did not have at least 1 year of follow-up were excluded from the study. Among all included patients, we compared those who did not undergo device reimplantation (no-reimplant group) with those who received device reimplantation within 1 year (reimplant group). Patients were further divided according to the device type: pacemaker, ICD, or CRT devices. The study protocol was approved by the institutional review boards of Mayo Clinic and UCSD.

Data collection

Medical records of included patients were reviewed for age, sex, presence of heart failure and other medical comorbid conditions, CIED type and model, reasons for lead extraction, and, if applicable, reasons for not undergoing device reimplantation. The primary reasons for initial device implantation were sinus node dysfunction, atrioventricular block (AVB), carotid sinus hypersensitivity, atrioventricular nodal ablation, primary or secondary prevention of sudden cardiac death, or advanced heart failure with conduction abnormality requiring pacemaker, ICD, or CRT. The reasons for not undergoing reimplantation were categorized as follows: (i) patient no longer meets indication for device implantation, (ii) patient never met device indication at initial CIED implantation, (iii) unresolved device-related complications, or (iv) patient preference. American College of Cardiology/American Heart Association/Heart Rhythm Society 2008 and 2012 guidelines for CIED use and 2013 appropriate use criteria were used to define these classifications.^5–7

Outcomes assessment

Follow-up data, including cardiac clinic visits, hospital admissions, and electrocardiography and Holter monitor results, were reviewed retrospectively to report the incidence of hospitalization for heart failure, presyncope or syncope, and recurrence of ventricular tachycardia or bradyarrhythmias. The need for subsequent device implantation was thoroughly collected in patients without reimplant. Death information was obtained using the Accurint system through our institutional survey research group. Death certificates were reviewed to determine the cause of death.

Statistical analysis

The patients' clinical characteristics, device characteristics, and reasons for extraction in the reimplant and no-reimplant groups were compared using t-tests for continuous variables. Categorical variables were compared using the χ² test for independence. The Kaplan–Meier method was used to estimate survival in both groups. Propensity score matching was performed on 97 patients from the reimplant group with the 97 patients from the no-reimplant group to eliminate the effects of confounding variables (see Supplementary material online, Table A1). Potential risk factors for mortality in the no-reimplant group were evaluated using the Cox proportional hazards models. The rates of heart failure hospitalization, presyncope or syncope hospitalization, recurrence of arrhythmia, and device reimplantation were further studied in the no-reimplant group using descriptive statistics. A P-value of <0.05 was considered statistically significant.

Results

Baseline characteristics

Among the 828 consecutive patients identified at the 2 institutions during the study period, 678 survived to hospital discharge and had at least 1 year of follow-up. Of these patients, 581 received device reimplantation within 1 year (reimplant group), and 97 did not undergo device reimplantation (no-reimplant group). Among these 97 patients, the original device type was a pacemaker in 48, an ICD in 41, and CRT in 8 (Figure 1). The device was not reimplanted in 48 of 300 patients (16%) with a pacemaker, 41 of 221 patients (19%) with an ICD, and 8 of 157 patients (5%) with CRT devices.

Patients without CIED reimplantation were younger than patients with device reimplantation [mean (SD) age, 60.7 (18.9) vs. 70.6 (16.4) years; P < 0.001] (Table 1). Overall, sex distribution was similar between the groups (P = 0.37). More patients in the reimplant group had cardiomyopathy during initial implantation (60.6 vs. 33.0%; P < 0.001) and were New York Heart Association Class III and IV (13.6 vs. 5.2%; P = 0.02). The rates of ventricular tachycardia and atrial flutter or atrial fibrillation were also significantly higher in the reimplant group than in the no-reimplant group (both P < 0.05), as were the rates of use of antiarrhythmic drugs (P = 0.02), β-blockers (P = 0.001), and digoxin (P < 0.001). Twelve patients (12.4%) in the no-reimplant group were on dialysis, compared with 24 patients (4.2%) in the reimplant group (P < 0.001). The rates of other medical comorbid conditions, including coronary artery disease, diabetes mellitus, and peripheral vascular disease, were similar in both groups (Table 1).

The initial indications for CIED implantation were different between the two groups. Sinus node dysfunction, carotid sinus hypersensitivity, and primary sudden death prevention were more common indications for the initial CIED in the no-reimplant group, whereas high-grade AVB, atrioventricular nodal ablation, and resynchronization therapy for heart failure were more common in the reimplant group (Table 1).

Indications for lead extraction

Lead extraction for device infection was 52.7% for the reimplant group and 59.8% for the no-reimplant group (P = 0.58). The no-reimplant group had a higher rate of systemic infection (30 vs. 14.4%; P < 0.001) and a lower rate of pocket infection (29.9 vs. 38.3%; P < 0.001) (Figure 2). The reimplant patients more often underwent lead extraction for lead malfunction or lead recall than did the no-reimplant group (31.2 vs. 4.1%; P < 0.001). Few patients underwent lead extraction for venous thrombosis or obstruction, tricuspid valve incompetence, or lead cardiac perforation in both groups. Nearly a quarter of the no-reimplant group underwent elective device removal because the device therapy was no longer needed.

Reasons for no cardiovascular implantable electronic device reimplantation and outcomes

Patient no longer meets indication for cardiovascular implantable electronic device

The most common reason that patients did not have their CIED reimplanted was that they no longer met criteria for device implantation (n = 47, 48%). The reason for prior device implantation in these patients was sinus node dysfunction (resolved) in 18, history of high-grade AVB (resolved) in 8, and primary sudden death prevention in 15 with subsequent improvement or normalization of left ventricular ejection fraction (Figure 3). Three patients had an ICD device for secondary prevention of sudden death, all of whom had a heart transplant at the time of explantation. Three patients who had CRT devices initially had tachyarrhythmia-induced cardiomyopathy, which resolved after subsequent rhythm treatment. During follow-up, 13 of these patients died. Ten died of noncardiac causes, one with prior pacemaker died of end-stage cardiomyopathy, one with prior ICD died of acute myocardial infarction, and one died of unknown cause. The noncardiac causes of death were cancer (n = 3), respiratory failure due to pneumonia or pulmonary embolism (n = 2), end-stage renal disease (n = 2), sepsis (n = 2), and cerebrovascular accident (CVA) (n = 1).

Never met cardiovascular implantable electronic device indication

The second most common reason for not receiving a reimplant was that the patient did not meet the device indication at initial implantation (n = 22, 23%). Eleven of these patients had pacemaker implantation despite not meeting the criteria for pacemaker therapy: eight for vasovagal syncope, two for asymptomatic sinus bradycardia, and one for high-grade AVB in the setting of Lyme myocarditis (Figure 3). Eleven patients underwent unindicated intracardiac defibrillator placement: two for ventricular tachycardia within 48 h of acute myocardial infarction that was treated with revascularization and nine for long QT syndrome that either was misdiagnosed initially or could have been treated with alternative medical therapy. In this group, only one patient, with previously implanted pacemaker for sinus bradycardia, died (unknown cause). None of these patients required admission for heart failure or syncope.

Patient preference

Sixteen patients (17%) preferred not to undergo device reimplantation, contrary to medical advice (Figure 3). The majority, 11 patients, avoided another implantation despite meeting the criteria for ICD therapy. Five of these patients required ICD for secondary sudden death prevention. Four patients had symptomatic sinus node dysfunction, and one had carotid sinus hypersensitivity. Nine of these patients died during follow-up: seven deaths were due to noncardiac causes (four CVA, one cancer, one respiratory failure due to severe pulmonary hypertension, and one suicide), one with prior ICD died of end-stage cardiomyopathy, and one due to unknown cause.

Unresolved device complications

Unresolved device-related complications prevented 12 patients (12%) from receiving a new device (Figure 3). Of these, two had fungaemia, seven had persistent and recurrent Staphylococcus aureus bacteraemia, two had superior vena cava syndrome due to venous obstruction, and one had severe tricuspid valve incompetence. All had successful complete extraction of their intracardiac system. Four out of the nine patients with ongoing systemic infection had other sources of infection beside CIED, including, infected dialysis catheter, necrotizing pancreatitis, Peripherally inserted central catheter (PICC)-line-associated infection with secondary seeding of CIED, and fungal endocarditis. All of these patients had control of other sources of infection, including, PICC-line removal, dialysis catheter exchange, radiologic debridement of necrotizing pancreatitis, and tricuspid valve replacement for tricuspid valve endocarditis. Three patients with persistent bacteraemia had history of retained leads that were successfully extracted during the index procedure. Eight patients died during follow-up. Six were noncardiac related, including sepsis (n = 3), end-stage renal disease (n = 2), and cancer (n = 1). The other two patients with prior CRT-D died of heart failure and ventricular tachycardia, respectively.

Overall survival

During an average follow-up of 4.2 years (±3.1), the survival rate was significantly lower in the CIED no-reimplant group than in the reimplant group (60 vs. 93%; P < 0.001) (Figure 4). Among 31 deaths in the no-reimplant group, 16 patients had a prior pacemaker, 11 had a prior ICD, and 4 had prior CRT (Table 2). Most of the deaths in this group (23, 74%) were not related to cardiac causes. Only five deaths were cardiac in nature, one of which was directly related to cardiac arrhythmia (Table 2); the others were related to heart failure among which one had CRT device.

Mortality in the no-reimplant CIED group was associated with several risk factors by univariate analysis: unresolved device complications [hazard ratio (HR), 3.91; 95% CI, 1.74–8.81; P = 0.001], extraction for CIED infection (HR, 3.06; 95% CI, 1.24–7.52; P = 0.02), and current use of dialysis (HR, 2.74; 95% CI, 1.12–6.71; P = 0.03) (Table 3). Never having met device indication at initial implantation was associated with lower mortality (HR, 0.19; 95% CI, 0.05–0.80; P = 0.02). Furthermore, the association of death with unresolved device complications (HR, 4.79; 95% CI, 1.98–11.55; P < 0.001) and extraction for CIED infection (HR, 2.69; 95% CI, 1.09–6.46; P = 0.03) in the no-reimplant group remained significant on multivariate analysis after excluding confounding factors.

Clinical outcomes

Five patients in the CIED no-reimplant group had hospital admission for heart failure during follow-up. One of these patients had prior CRT device not reimplanted because of ongoing systemic infection. Two required admission for the diagnosis of cardiogenic syncope. Both patients had prior implanted pacemakers. One underwent pacemaker reimplantation after the 1-year follow-up period for reoccurrence of high-grade AVB. The other underwent ICD implantation after the 1-year follow-up period for cardiac arrest due to new onset of ventricular tachycardia (Table 2). One patient died of out-of-hospital cardiac arrest after CRT-defibrillator lead extraction for ongoing bacteraemia. The cardiac arrest was secondary to lethal ventricular tachycardia that degenerated into ventricular fibrillation not detected by his LifeVest (ZOLL).

Discussion

Main findings

The rate of lead extraction has been increasing in parallel with the expansion of CIED use in recent years. Specific guidelines and indications for lead extraction were implemented by the Heart Rhythm Society Expert Consensus to unify such practice.^9,10 Our study is, to our knowledge, the first to report outcomes of patients who underwent device extraction without reimplantation. Approximately 70% of the patients without reimplant either did not meet any indication for ongoing device therapy or their device was not indicated at the index implantation based on current guidelines.^5–7 Patient preference and unresolved device complications factored in reimplantation decision-making in the other cases. The survival rate was lower in the no-reimplant group than in the reimplant group. However, 74% of deaths were secondary to noncardiac causes and were unrelated to lack of CIED.

Patient no longer meets indication for CIED

Nearly half of the group did not undergo device reimplantation because they did not have ongoing need for the device. The initial implantation was appropriate because these patients met the criteria for symptomatic high-grade AVB, sinus node dysfunction, severe cardiomyopathy, and advanced heart failure at one point in their lifetime.¹¹ The decision to not reimplant for this group was also appropriate because the initial conditions in these patients were resolved or improved at the time of reimplant decision-making. Only one patient, with a prior pacemaker, died of end-stage cardiomyopathy in this group, which supports the appropriate decision to not reimplant the device in this group. Moreover, such a decision is most likely associated with forgoing the unnecessary cost of CIED implantation and eliminating procedure-related complications.¹²

Patient never met cardiovascular implantable electronic device indication

Approximately 20% of patients had initial inappropriate CIED implantation; most were related to vasovagal syncope without documented cardioinhibitory response or ICD overuse for presumed or confirmed long QT syndrome.¹³ Forty per cent of cases referred to Mayo Clinic for presumptive diagnosis of long QT syndrome have an incorrect diagnosis.⁶ The patients with confirmed long QT syndrome were subsequently treated successfully with the current recommended first-line therapy of a selective β-blocker.^6,14–16 The absence of arrhythmia-related complications and cardiac death in this group during the follow-up period confirms that these patients are at low risk for arrhythmic sudden death.

Unresolved device complications

Twelve per cent of the study cohort did not undergo device reimplantation because of ongoing device-related complications despite meeting the indication for CIED reimplantation. Most of these patients had uncontrolled systemic infection or had a high risk of recurrent infection. Tarakji et al.^17,18 previously demonstrated decreased survival of patients with CIED complicated by systemic infection or endocarditis compared with localized pocket infection.^17,18 In our study, extraction for infection and avoiding reimplantation because of ongoing device complications were associated with four-fold increases in mortality. The overall higher mortality in the no-reimplant cohort was mainly driven by ongoing device complications. However, cardiac death preventable by CIED was less common than noncardiac death. Severe noncardiac diseases with poor prognosis may prompt physicians and/or patients, appropriately, to defer considering device reimplant.

Patient preference

Despite meeting reimplant criteria, 17% of the study patients chose discontinuation of device therapy. Some of them made this decision because of severe comorbid conditions such as CVA or malignancy. Some of this group had ICD for primary sudden death prevention; these patients preferred stopping device therapy with the understanding of the risks of life-threatening arrhythmia. More than half of these patients died, mainly of noncardiac causes, which supports appropriate patient decision-making in the current study.

Study limitations

The data gathering of baseline clinical features and outcomes was conducted in a retrospective manner, which may be associated with bias in data collection and interpretation of results. An attempt was made to include all confounding factors while performing statistical analysis; however, other confounding factors may not have been addressed. Some of the patients with unresolved device complications were not offered alternative implantation strategies for unclear reasons such as transfemoral approach for venous obstruction or epicardial approach for severe tricuspid insufficiency. Moreover, the cause of death was unknown in three patients, which might explain the negligible rates of cardiac arrhythmia-related deaths. Finally, our study size may be underpowered to determine differences in clinical outcomes between groups. Larger studies in specific subsets with a control group may be able to provide further insight.

Conclusion

Fourteen per cent of patients did not undergo reimplantation after extraction of their CIED, most of which were pacemakers and ICDs. Half of the extractions were related to infection complications. The no-reimplant group had decreased survival compared with the reimplant patients. The higher mortality rates in the no-reimplant population were mainly driven by noncardiac medical comorbid conditions, ongoing device complications, and infection. Appropriate decision-making for not reimplanting a CIED does not appear to significantly affect heart failure admission, cardiac syncope, and cardiac-related death.

Supplementary material

Supplementary material is available at Europace online.

Conflict of interest: none declared.

Ethics: The research is approved by the locally appointed ethics committee.

References