Lead extraction outcomes in patients with congenital heart disease

Introduction

Patients with congenital heart disease (CHD) frequently require cardiac implantable electronic devices (CIEDs) including transvenous pacing and defibrillator systems. These devices are often placed at a young age, and with improved survival, the lifespan of the patient often exceeds that of the device. Patients may undergo multiple generator changes and lead revisions and many have abandoned leads. Patients with CHD experience higher rates of lead malfunction and infection when compared with individuals with structurally normal hearts.^1–3 The safety of advanced extraction techniques, including transvenous laser lead extraction (LLE), in patients with structurally normal hearts has been well established.^4,5 However, the safety of LLE in adults with CHD is largely unknown. Given the challenging anatomy, these patients are often referred to high-volume centres for extraction.⁶ Published data on CHD lead extraction come primarily from retrospective case studies.^7–11 These reports include a heterogeneous group of paediatric and young adult patients, many of whom did not have structural CHD but rather underwent device placement for a primary rhythm disturbance or channelopathy.^7–9,11,12 Two small retrospective case series, including a combined total of 38 patients, examined LLE in adults with structural CHD.^10,13 There was variation in the procedural success reported (74 vs. 91%); however, only one major complication was reported in the combined cohorts. Given the limited data available, we sought to review our centres' experiences with LLE in CHD.

Methods

We performed a retrospective, case-control study of all CHD patients undergoing extraction of chronically implanted transvenous CIED leads at two academic medical centres (Mayo Clinic, Rochester, MN and Oregon Health and Science University [OHSU], Portland, OR) between January 2001 and November 2014. All patients with a cardiac congenital malformation were included for analysis. Twenty-nine patients were identified from Mayo Clinic and 11 from OHSU. Data were compared with control patients to determine if there were differences in clinical parameters or outcomes. Two controls were selected for each CHD case to improve statistical powering. Controls were selected from the Mayo database of 652 patients (mean age 64 ± 17 years, 68% male), who underwent 702 procedures to remove a total of 1378 leads. Patients were matched by gender and age at the time of the procedure accepting a 10-year margin due to constraints of sample size. Because of sample size, controls could not be matched for duration of lead implantation or the presence of co-morbidities. In accordance with consensus definitions of extraction, only patients with leads in place >12 months at extraction were included.¹⁴ Epicardial leads were excluded. This study was approved by the institutional review boards of both institutions.

Successful extraction and complications were defined according to consensus statements.¹⁴ Success was achieved with complete extraction, or in cases with retained fragments, if the desired clinical outcome was obtained. Retained fragments were not considered a success if the indication for extraction was endocarditis. Major complications were life-threatening, required surgery, or resulted in death or disability. Minor complications included non-life-threatening events, which required a medical intervention to remedy and did not result in disability.

Procedural methods did not differ significantly between the two institutions. Extractions were performed according to established best practices by experienced operators with training in advanced extraction techniques.^14,15 Briefly, procedures were performed in either a hybrid operating room with fluoroscopy or the electrophysiology laboratory. Procedures were coordinated with a cardiothoracic surgeon, who was available for urgent intervention. Prior to the procedure, the patient's chest and groin were prepped for emergent sternotomy and cardiopulmonary bypass. Arterial blood pressure was followed via a radial arterial line throughout the procedure. A transoesophageal echocardiogram was monitored for rapid assessment of pericardial effusions. Following the procedure, all patients were admitted overnight to ensure appropriate monitoring.

After appropriate patient preparation, the pre-pectoral pocket was opened and the generator explanted. A regular stylet was introduced via the lead lumen, and if possible/applicable, the helix was retracted. Manual traction was applied. If the lead did not free with traction, it was clipped and a locking stylet introduced. If necessary, a laser sheath (LS) was advanced to release the lead from endovascular adhesions. Brief pulses of laser energy were applied to adhesions. If the lead failed to free with an LS, a mechanical sheath was used. Snares were employed from the femoral vein as needed to assist with lead removal or retrieve lead fragments.

Statistical analysis

Both patient- and lead-specific variables were analysed. Continuous variables were compared between the groups using a two-sample t-test. A χ² test was used to compare categorical variables. For the comparisons in Table 4, all leads for the individuals were compared. A Generalized Estimating Equations (GEE) model was utilized to account for the potential correlation among leads from the same individual.

Results

Patient characteristics

A total of 40 CHD patients underwent extraction of 77 chronically implanted transvenous leads. Characteristics of CHD and control patients are listed in Table 1. Patients with CHD had a mean age of 38 ± 16 years at the time of the extraction (range 10–79 years). This was not different from the mean age in the age-matched control group (40 ± 16 years, range 8–79, P = 0.58). Two of the CHD patients were 10 and 12 years old, and the remaining patients were ≥17 years of age. The average lead age in the CHD group was 83 ± 87 vs. 62 ± 65 months in the control group (P = 0.24). The number of leads per CHD patient ranged from 1 to 5 with an average of 1.9 ± 0.9, which was not different from the control group (1.8 ± 0.8, P = 0.52). The ejection fraction (EF) was reported qualitatively in five patients in whom the systemic ventricle was not the morphologic left ventricle (LV). Two of these patients had mild to moderate EF impairment, and three had moderate to severe EF impairment. In the remaining 36 patients, the mean EF was 47 ± 15% when compared with 50 ± 17% in the control group (P = 0.26). Fifty-three per cent of the CHD patients and 29% of the control group were chronically anti-coagulated, but these medications were held for the extraction procedure.

Four patients in the CHD group had surgically uncorrected CHD. The remaining 36 patients had undergone one or more cardiac surgical procedures (range 1–7). This was significantly different from the control group, where only 21% of the patients had a history of cardiac surgery (P < 0.001). The CHD patients' diagnoses are summarized in Table 2. Detailed descriptions of individual patients' congenital diagnoses and their surgical histories are described in Supplementary material online, Table S1.

Procedural characteristics

The indications for CIED implantation and extraction, as well as lead type and location, are listed in Table 3. Thirty-eight CHD patients underwent a single extraction procedure, and two patients underwent two separate extraction procedures with a total of 77 leads extracted. Fourteen CHD patients had a single lead extracted; the remaining 26 patients underwent simultaneous extraction of multiple leads including 17 patients with two leads extracted, 7 patients with three leads extracted, and 2 patients with four leads extracted. Twenty-eight CHD patients had no abandoned leads, 8 patients had one abandoned lead, and 4 patients had two or more abandoned leads. The average age of abandoned leads was 164 ± 113 months. There were 146 leads extracted in the control group. Twenty-nine control patients had a single lead extracted, 39 had two leads, 9 had three leads, and 3 patients had four leads extracted. Only three of the leads extracted in the control population were abandoned.

Procedural techniques are summarized in Table 4. Simple manual traction was sufficient to remove 31 CHD leads and 69 control leads. The average age of leads removed with manual traction was 19 ± 28 months in the CHD group and 23 ± 27 months in the controls (P = 0.70). Four leads extracted in CHD patients required use of a locking stylet and 33 also required an LS. Of the 33 leads extracted with an LS, 1 also required a rotating mechanical sheath, and 8 required use of a snare from the femoral vein. The average age of the leads removed with an LS was 107 ± 48 months, whereas the average age of leads requiring the use of adjunct tools including the rotating mechanical sheath and femoral snares was 124 ± 56 months. In the control group, 71 leads were removed with an LS and 5 also required a rotating mechanical sheath. The average age of these leads was 90 ± 62 months for LS and 196 ± 104 months for combined laser and mechanical sheath extraction. There was similar utilization of LS (49 vs. 43%, P = 0.28) and mechanical sheaths (1 vs. 3%, P = 0.60) between groups; however, femoral snares were more frequently utilized in CHD patients (10 vs. 0%, P < 0.001).

Procedure outcomes

Complete extraction was achieved in 94% for both groups. In the CHD group, a 25-year-old abandoned right ventricular (RV) pacing lead fractured and the distal portion could not be removed despite the use of multiple femoral snares. This was clinically unsuccessful as the extraction indication was endocarditis. One CHD patient undergoing upgrade of a dual-chamber pacemaker to an ICD had an unsuccessful attempt to remove two 13-year-old abandoned pacing leads and one 8-year-old active RV pacing lead. Only manual traction was employed as there was no strong indication for extraction and the perceived risk for complications with advanced tools was high. Eight years later, the patient underwent heart transplantation and the intracardiac portion of the leads was severed at surgery. A preoperative chest X-ray is provided in Figure 1 as an illustrative example. Post-operatively, the remaining transvenous portions of the leads were removed with simple manual traction.

Figure 1

Chest X-ray from a patient with surgically corrected tetralogy of Fallot including pulmonary valve replacement, anomalous drainage of the right pulmonary veins to the right atrium, interrupted inferior vena cava with azygous continuation, and right aortic arch. There are five transvenous leads, three of which are abandoned leads in the left pectoral pocket including one right atrial lead (A), one dislodged RV lead (B), and one RV lead (D). There are two active leads including a right atrial lead (C), which is tunnelled from the left to the active right pectoral can right pectoral can and a dual coil RV ICD lead (E) from the right pectoral pocket. The lead ages were 21,² 16,² and 8 years¹ at the time of extraction. Extraction of three leads was unsuccessfully attempted with simple manual traction in 2001 at the time of device upgrade to an ICD. Extraction was accomplished in 2009 when the intracardiac portions of the five leads were severed during heart transplantation. Post-operatively the remaining transvenous portions of the leads were removed with simple manual traction.

Open in new tabDownload slide

Procedure complications

There were four major complications in the controls and none in the CHD group (P = 0.30). There were no minor complications in either group. The control complications included three superior vena cava lacerations requiring sternotomy (one was fatal) and one ventricular perforation requiring sternotomy. Review of the four control complications shows that a total of six leads were extracted. The average lead age was 15.3 years, and in all cases, an LS was utilized. The indications for removal included elective removal of an abandoned lead,¹ device malfunction with inappropriate ICD therapy,¹ and endocarditis.² Two were ICD leads, both of which had SVC coils. None of the control patients with a major complication had a history of cardiac surgery.

Discussion

It has been shown that implantation of CIEDs can be safely accomplished in adults with CHD;¹⁶ however, there are little data on device extraction in this population. There is a high rate of lead failure and complications which require lead extraction in CHD patients. This is due in part to multiple device revisions which increase the risk of infection, and multiple indwelling leads, which can cause venous occlusion and lead malfunction.^1–3 These leads may be difficult to extract due to long duration of lead implantation and subsequent endovascular fibrosis. Further complicating matters are anatomic abnormalities from the congenital malformation and subsequent surgical repairs. Previously, studies of lead extraction in CHD patients have focused on paediatric patients, included epicardial leads, and combined structural CHD patients with primary arrhythmia and channelopathy patients with anatomically normal hearts and no history of cardiac surgery.^7–11 To our knowledge, this is the largest case series exclusively examining CHD patients with structural heart disease and chronically implanted leads undergoing transvenous lead extraction, 95% of whom are adults.

Encouragingly, our study found a 94% success rate in both groups. Refinements in extraction techniques, including utilization of an internal jugular vein approach, are yet to be described in the CHD population and may further improve the efficacy of extraction.¹⁷ Simple traction was sufficient to extract 40% of CHD leads and 47% of control leads. This is comparable with what has been reported in non-CHD adults undergoing lead extraction.¹⁸ Augmented extraction with laser and mechanical sheaths was successful in 52%. There was no statistically significant difference in complications between the groups.

Understanding extraction risk is particularly relevant in managing abandoned leads. Thirty per cent of the CHD patients had one or more abandoned leads. The decision to remove abandoned leads is dictated by the indication for extraction and by the perceived difficulty and risk. It is impossible in a retrospective study to quantify the long-term benefits of removing abandoned leads, but early extraction of abandoned leads may minimize the risk of future complications such as venous obstruction, tricuspid regurgitation, lead malfunction, and lead erosions.^19,20 Certainly, it decreases the risks of extracting those leads at a later point when the dwell time will be even longer.

It is notable that the control group complication rate was higher than our combined centres' total experience, which includes 834 procedures with an overall complication rate of 1.7%. The large number of complications in the control group may reflect sampling bias, which occurred as a result of age matching within a limited population. A closer look at the control patient complications shows that the average lead age was 12.9 years, which is significantly older than the mean lead age seen in both the CHD and control populations (6.7 and 4.2 years, respectively, P = 0.05 for the combined CHD and controls without complications). This is unsurprising as lead age is a major predictor of complications during extraction.²¹ Patients who receive CIEDs at a young age may be pre-disposed to more extensive fibrosis than their older counterparts and therefore may experience more complications. While sampling bias likely over-represented complications in the controls, the absence of complications within the CHD group is consistent with what has previously been reported.^10,13 Furthermore, there may be a relationship between complications and prior cardiac surgery. Pericardial adhesion to the epicardial surface is potentially protective by reducing the risk of massive extravasation when endocardial trauma occurs.

This study is limited by its retrospective nature and by sample size. The sample size precluded correcting for confounders such as variation in lead implantation duration, and the number of ICD leads vs. pacemaker leads. The inability to match by lead age accounts in part for the higher complication rate observed in the controls. Furthermore, correcting for the impact of these confounders on the outcome by using a linear regression model was not possible as there were zero complications within the CHD group. Given the small number of complications, our analysis is limited to only individual comparisons. However, our study suggests that although technically challenging, extraction of chronic CIED leads in CHD patients can be accomplished at high-volume centres with safety and efficacy comparable with non-CHD adults.

Conclusions

Successful extraction of chronically implanted CIED leads in the CHD population can be accomplished with a safety profile similar to the general lead extraction population. Over a third of our centres' extractions were accomplished by simple manual traction. Advanced techniques, including LSs, were highly successful in removing leads, which could not be freed with traction alone. In spite of the challenges of advanced lead age and anatomic abnormalities, lead extraction can be carried out safely by an experienced operator.

Supplementary material

Supplementary material is available at Europace online.

Conflict of interest: none declared.

References