Summary
Background
The small diameter Sprint Fidelis defibrillation lead has not been implanted in patients since 2007 due to its unusually high rate of fracture. Predictors of lead fracture risk were identified in several studies, mainly in North American studies.
Aim
We established a multicentre registry to determine the lead fracture rate and predictors of fracture in a large cohort of French patients.
Methods
Nine hundred and eighty-six patients implanted with a Sprint Fidelis lead at six centres between December 2004 and November 2007 were included in this registry.
Results
Over a mean follow-up of 51.4 ± 20 months, the mean fracture rate was 11.2%, and increased over time: 1.2% at 1 year, 3.8% at 2 years, 7.4% at 3 years, 13.9% at 4 years, and 20.7% at 5 years. In multivariable analysis, younger age (< 40 years) was associated with a higher risk of fracture compared to patients < 40 years, patients aged between 40–60 years had a relative risk of 0.53 (95% confidence interval [CI] 0.29–0.98) and patients > 60 years had a relative risk of 0.45 (95% CI 0.24–0.84) and subpectoral implantation (at 3 years) with a relative risk of 2.35 (95% CI 1.29–4.28). Lead model 6930 (single-coil, passive-fixation) had a relative risk of 3.47 (95% CI 1.13–10.7) compared with the 6949 model (double coil, active-fixation). No other predictor of fracture was identified.
Conclusions
In a large multicentre cohort of French patients implanted with a Sprint Fidelis electrode, the fracture rate remains high, especially in young patients with submuscular implant and the 6930 electrode model.
Résumé
Contexte
L’électrode de défibrillation Sprint Fidelis de petit diamètre n’est plus implantée depuis 2007 compte tenu d’un taux anormalement élevé de fracture. Les facteurs prédictifs du risque de fracture de cette électrode ont été évalués dans plusieurs études, principalement chez des patients d’Amérique du Nord.
Objectif
Nous avons mis en place un registre multicentrique pour déterminer le taux de fracture et les facteurs prédictifs de fracture dans une grande cohorte de patients français.
Méthodes
Neuf cent quatre-vingt-six patients implantés d’une électrode Sprint Fidelis entre décembre 2004 et novembre 2007 ont été inclus.
Résultats
Avec un suivi moyen de 51,4 mois, le taux de fractures a été de 1,2 %, 3,8 %, 7,4 %, 13,9 %, 20,7 % à respectivement 1, 2, 3, 4 et 5 ans. En analyse multivariée, l’âge jeune (moins de 40 ans) était associé à un risque plus élevé de fracture (comparés aux patients âgés de moins de 40 ans), ceux âgés entre 40 et 60 ans ont un risque relatif de 0,53 (intervalle de confiance 95 % [IC] 0,29–0,98) et ceux âgés de plus de 60 ans ont un risque relatif de 0,53 (IC 0,24–0,84), l’implantation en rétro-pectoral à un risque relatif de 2,35 (95 % IC 1,29–4,28) à 3 ans. Le modèle d’électrode 6930 (simple coil à fixation passive) a un risque de 3,47 (95 % IC 1,13–10,7) en comparaison au modèle 6949 (double coil à fixation active). Aucun autre facteur prédictif n’a été identifié.
Conclusions
Dans une cohorte importante multicentrique française de patients implantés d’une électrode Sprint Fidelis, le taux de fracture reste élevé, surtout chez les patients jeunes, avec implantation rétro-pectorale et le modèle 6930 de cette électrode.
Background
Large randomized trials have shown the beneficial effects of the implantable cardioverter-defibrillator (ICD) on mortality in both primary and secondary prevention of sudden cardiac death. In addition, the advent of transvenous ICD leads more than 20 years ago is responsible for the 20-fold increase in annual implantations over the past 20 years , with more than 120,000 devices implanted each year worldwide.
Both ICD generators and leads are critical components of the system. Whereas generator failures are relatively easy to handle, lead failures may have more severe consequences, such as failure to pace or to defibrillate, inappropriate shock delivery and even death; furthermore, lead removal may be dangerous. Since the first introduction of endocardial ICD leads in 1990, there has been a progressive reduction in lead size to simplify implantation. Medtronic Sprint Fidelis leads (Medtronic Inc., Minneapolis, MN, USA) were released in 2004 and were immediately used extensively worldwide because of their small size (6.6 F). In April 2007, Hauser et al. reported a significant incidence of Sprint Fidelis lead fractures, and in November 2007, Medtronic stopped release of the Sprint Fidelis lead. Since then, Medtronic, the Food and Drug Administration and other health authorities have published specific recommendations for follow-up in patients with a Sprint Fidelis lead .
Since 2009, several independent registries have confirmed a high incidence of Sprint Fidelis lead fracture . However, these included single-centre studies reporting a higher rate of fracture in the Sprint Fidelis lead compared with that in other leads and a multicentre registry confirming the high rate of Sprint Fidelis lead fracture (mainly dual-coil leads) in North America .
We report the results of a the French Fidelis registry, launched at the end of 2011 and conducted under the auspices of the French Working Group on Cardiac Pacing and Arrhythmia. This registry not only aimed to report the incidence of lead fracture, but also to determine the potential risk factors for developing such a lead fracture.
Methods
All patients implanted with a Sprint Fidelis lead in six large French ICD centres (Grenoble, Lille, Nancy, Rennes, Saint-Denis and Saint-Etienne) were included in the registry.
Data collection
The following demographic data were collected: age; sex; left ventricular ejection fraction (LVEF; subsequently stratified into LVEF < 0.35, LEVF between 0.35 and 0.55 and LVEF > 0.55); indication for implantation (prophylactic or secondary indication); and type of heart disease (divided into ischaemic, dilated, hypertrophic, arrhythmogenic right ventricular dysplasia [ARVD], electrical disorders or none).
The following technical data were collected: venous access (cephalic or subclavian); lead position (apex or septal); type of Sprint Fidelis lead (single-coil models 6930 [passive] or 6931 [active] and dual-coil models 6948 [passive] or 6949 [active]); number of implanted leads; device location (right or left side); and prepectoral or subpectoral.
The date of implantation, the date of the last follow-up or patient’s death (with cause of death), the use of appropriate and inappropriate therapy during follow-up (antitachycardia pacing, shocks or both) and the date of the Sprint Fidelis lead failure were all recorded. Lead failure was defined as one or more of the following: asymptomatic increase in pacing or high-energy circuit impedance at follow-up or detected by the lead integrity alert (LIA) algorithm; asymptomatic abrupt decrease in right ventricular sensing; non-sustained ventricular fibrillation episodes corresponding to short intervals (asymptomatic); and inappropriate therapy with shocks (symptomatic).
Statistical analysis
Statistical analysis was performed using SAS version 9.2 (SAS Institute Inc., Cary, NC, USA). Descriptive statistics were used to analyse the data. For continuous variables, data are expressed as means (ranges) ± standard deviations. Time-to-event analyses were conducted using Kaplan–Meier rates, to estimate the percentage of patients with no events at specified time intervals. Cox models were used to identify predictors of lead fracture, first by univariate analysis, testing the factors one by one, and then in a model with an ascending stepwise selection of variables. Hazard ratios and their 95% confidence intervals are provided for both univariate and multivariable analyses. No replacements were made in the case of missing data. The level for statistical significance was defined as a P value < 0.05.
Results
From December 2004 to November 2007, 1022 patients received a Sprint Fidelis lead in the six participating centres, representing 20.3% (1022/5024) of all Sprint Fidelis leads implanted in France during the same period. Follow-up was censored at the end of March 2012, and 36 patients were excluded for missing data.
The baseline characteristics of the remaining 986 patients are reported in Table 1 . Briefly, the mean age of the patients was 60.9 ± 10.6 years, 82% were men, and the indication was primary prevention in 58% of patients. The mean LVEF was 0.32 (range 0.15–0.70) and the most frequent underlying cause was ischaemic heart disease. Aetiologies were subsequently stratified into two groups:
- •
group 1, comprising patients with ischaemic heart disease or dilated cardiomyopathy;
- •
group 2, consisting of patients with other aetiologies (hypertrophic cardiomyopathy, ARVD, channelopathies or an apparently normal heart).
| Variable | |
|---|---|
| Age at implantation (years) | 60.9 ± 10.6 |
| Men | 809 (82) |
| Primary prevention | 576 (58) |
| Heart disease (group 1, n = 843; group 2, n = 143) | |
| Ischaemic | 517 (52) |
| Dilated | 326 (33) |
| Hypertrophic cardiomyopathy | 27 (3) |
| Electrical | 23 (2) |
| ARVD | 27 (3) |
| Others, none | 66 (7) |
| LVEF | 0.32 (0.15–0.70) |
| LVEF stratification | |
| < 0.35 | 642 (55) |
| 0.35–0.55 | 232 (24) |
| > 0.55 | 112 (12) |
| Cardiac resynchronization therapy | 369 (37) |
| Sprint Fidelis lead model | |
| 6930 (single-coil, passive-fixation lead) | 18 (2) |
| 6931 (single-coil, active-fixation lead) | 525 (53) |
| 6948 (dual-coil, passive-fixation lead) | 60 (6) |
| 6949 (dual-coil, active-fixation lead) | 383 (39) |
| Venous access | |
| Cephalic | 628 (64) |
| Subclavian | 358 (36) |
| Left side location | 763 (79) |
| Device location | |
| Prepectoral | 876 (89) |
| Subpectoral | 110 (11) |
| RV lead position | |
| Apex | 819 (83) |
| Septal | 167 (17) |
| Number of leads implanted | |
| 1 | 301 (31) |
| 2 | 374 (37) |
| 3 | 311 (32) |
Four different types of Sprint Fidelis leads were implanted, most commonly the single-coil, active-fixation (6931) lead (53%) and the dual-coil, active-fixation (6949) lead (39%).
The mean follow-up duration was 51.4 ± 20 months, with a mean rate of Sprint Fidelis lead fracture of 11.2% at the end of follow-up. The incidence of lead fracture increased with time, with a lead fracture incidence of 1.2% at 1 year, reaching 20.7% at 5 years In the group of 118 patients with Sprint Fidelis lead failure, the failure was detected in 70 patients (59%) after inappropriate shocks caused by lead failure, in 33 (28%) by LIA, and in 15 (13%) at the time of routine 3-month follow-up. Overall, 148 patients had Sprint Fidelis lead revision: 114 (76%) for lead failure; 16 (11%) for prophylactic lead change at the time of generator replacement; and 18 (13%) for non-lead-related reasons, mainly infection ( n = 10). Four patients with lead failure underwent ICD deactivation with no lead revision (patient’s choice, n = 2; terminal illness, n = 2).
The results of the univariate analysis of lead failure predictive factors are shown in Table 2 . Briefly, univariate analysis revealed a higher incidence of lead fracture with the single-coil, passive-fixation 6930 lead compared with that in other models ( P = 0.009), and an increased risk of lead fracture with passive-fixation leads (the 6930 and 6848 models) compared with active-fixation leads (the 6931 and 6949 models) ( P < 0.05); however, there was no difference between single-coil and dual-coil leads ( P = 0.82). Four other factors were identified as risk factors for lead fracture: patient age < 40 years ( P = 0.0001); group 2 patients ( P = 0.012); patients with LVEF > 55% ( P = 0.035); and subpectoral implantation ( P = 0.0006). For subpectoral implantation, the statistical significance appeared only after 3 years.
