Author + information
- Received June 25, 2013
- Revision received August 21, 2013
- Accepted September 9, 2013
- Published online January 21, 2014.
- Christopher M. Janson, MD∗ (, )
- Akash R. Patel, MD,
- William J. Bonney, MD,
- Karen Smoots, RN and
- Maully J. Shah, MBBS
- Division of Cardiology, The Children's Hospital of Philadelphia, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania
- ↵∗Reprint requests and correspondence:
Dr. Christopher M. Janson, Division of Cardiology, 8NW-25, Main Building, The Children's Hospital of Philadelphia, 34th Street and Civic Center Boulevard, Philadelphia, Pennsylvania 19104.
Objectives This study aimed to investigate the impact of lead diameter and design on implantable cardioverter-defibrillator (ICD) lead survival in children and young adults.
Background Recent reports have called attention to high rates of lead failure in adults with small-diameter ICD leads, but data in the pediatric population is limited.
Methods We reviewed lead performance in consecutive subjects ≤30 years with transvenous right ventricular ICD leads implanted at our center between January 1995 and October 2011. Lead failure was defined as fracture, perforation, or sensing failure necessitating revision.
Results A total of 120 ICD leads were implanted in 101 patients at a mean age of 15.5 ± 4.9 years. There were 47 small-diameter (≤8-F) and 73 standard-diameter (>8-F) leads. During a median follow-up of 28.7 months (interquartile range: 14.4 to 59.2 months), there were 25 lead failures (21% prevalence), with an incidence of 5.6%/year (95% confidence interval: 3.4 to 7.8). Sprint Fidelis (SF) (Medtronic, Inc., Minneapolis, Minnesota) leads had lower 3-year (69% vs. 92%, p < 0.01) and 5-year (44% vs. 86%, p < 0.01) survival probabilities than standard-diameter leads. In multivariate Cox regression, SF design conferred the greatest hazard ratio for lead failure (hazard ratio: 4.42, 95% confidence interval: 1.73 to 11.29, p < 0.01). Age and linear growth were not significantly associated with lead failure.
Conclusions In this single-center pediatric study that evaluated lead diameter, lead design, and patient factors, the SF design conferred the highest risk of lead failure, suggesting that design rather than diameter is the critical issue in ICD lead performance.
The population of children and young adults with implantable cardioverter-defibrillators (ICDs) has grown over the last decade, in part due to advances in technology that have enabled smaller leads and devices (1). Although the ICD can be a life-saving device, it carries a significant risk of morbidity, including lead-related complications (2–4).
Small-diameter ICD leads gained popularity because of their relative ease of insertion and because they are less likely to cause venous obstruction and tricuspid valve distortion (5). Reduced lead diameter is a particularly useful feature for children, whose small size might preclude transvenous placement of standard-diameter leads. However, recent reports have demonstrated an increased rate of complications in adults with small-diameter leads (6–8).
The Sprint Fidelis (SF) high-voltage, small-diameter lead (Medtronic, Inc., Minneapolis, Minnesota) was recalled in 2007 due to its high incidence of premature conductor fracture, with a failure rate of 2.6% to 4.8%/year (9–12). More recently, in November 2011, the St. Jude Riata and Riata ST leads (St. Jude Medical, Sylmar, California) were recalled due to premature insulation failure, resulting in externalization of conductor cables (13,14). Riata lead failure rates have been reported to range from 0.7 to 2.8%/year (12,13,15,16). The active fixation Riata models have also been associated with a significantly increased incidence of cardiac perforations and lead revisions (7,17,18).
The failure of 2 small-diameter leads from different manufacturers has raised the question of whether small-diameter leads are robust enough to endure over time (12). However, the real issue might be the recent alterations in engineering and design rather than absolute lead diameter.
There is a significant cohort of pediatric patients with small-diameter leads in place, and few data exist on their performance in this population (19). Children have previously been shown to have substantially higher rates of transvenous lead-related complications than adults (19,20). Prevalence of lead failure in the pediatric population has ranged from 14% to 21% (2,3). This observation has been attributed to higher activity levels and growth-related lead stress (3,20,21).
This study sought to examine the impact of lead diameter, lead design, and patient factors on ICD lead survival in children and young adults.
Data were collected in accordance with hospital institutional review board guidelines. This was a retrospective chart review of subjects who underwent transvenous right ventricular (RV) ICD lead implantation at the Children's Hospital of Philadelphia between January 1, 1995, and October 1, 2011. All consecutive subjects who were 30 years of age or younger at the time of lead implantation and who had a minimum of 1 month of follow-up were evaluated. All implantation procedures were performed by experienced electrophysiologists. Lead implantation was performed via the subclavian vein in all cases, which is the standard practice in our institution. Baseline patient, lead, and device characteristics were recorded.
Small-diameter ICD leads were defined as having lead diameter ≤8-F, and standard-diameter ICD leads were of diameter >8-F (7).
Lead status was reviewed from the time of implant to the most recent patient encounter. Lead functionality was assessed by device interrogation, including analysis of lead impedance and sensed electrograms. The primary outcome was lead failure. This included lead fracture, defined as a sudden increase in long-term pacing and high-voltage impedance (≥50% as compared with chronic values) and/or electrical noise artifact from sensed nonphysiologic, make-break potentials (12). In addition, we included in the analysis instances of lead perforation as well as sensing failure necessitating lead revision (3). Oversensing of noncardiac potentials, such as electromagnetic interference, was not considered lead failure for the purpose of this analysis. Patients without follow-up for more than 1 year were considered lost to follow-up, and lead status was documented as of the last encounter. Similarly, in the instance of death or transplant, lead status was censored at the time of the last evaluation before the event. For subjects with multiple leads implanted during the study period, each lead was analyzed separately.
The presence or absence of inappropriate shocks secondary to lead failure as well as any associated pro-arrhythmic events were documented. Strategies to address failed leads were analyzed, including use of extraction tools, outcomes, and complications.
Descriptive statistics were expressed as mean ± SD for normally distributed continuous variables and as medians with interquartile ranges (IQRs) for skewed distributions. Categorical variables were reported as frequency counts and percentages. Student t tests were used to compare continuous variables, and chi-square tests were used to compare categorical variables. Rates of lead failure/100 person-years and 95% confidence intervals (CIs) were calculated. Kaplan-Meier survival analysis was used to estimate survival probability for small and standard-diameter leads, censoring cases at the time of death or transplant and when lost to follow-up. Simple Cox regression was used to identify patient and lead-related variables associated with lead survival. A multivariate Cox regression model was then created to compare the survival between small and standard-diameter leads, controlling for variables that were significant at a p value ≤0.2 from the simple Cox regression.
During the study period, 120 RV ICD leads were inserted in 101 patients. The patient population included 60% male and 40% female subjects. There was a primary prevention indication in 66% and a secondary prevention indication in 34% of patients. Underlying disease substrate included primary electrical disease in 46%, cardiomyopathy in 34%, and congenital heart disease (CHD) in 20%. Mean age at time of lead implantation was 15.5 ± 4.9 years.
Of the 120 leads, 94% (n = 113) were implanted in the left subclavian vein (6% right subclavian vein), 58% (n = 70) had a single-coil (42% dual-coil), and 72% (n = 86) were implanted in the context of single-chamber devices (28% dual-chamber). Characteristics of lead models represented in this cohort are outlined in Table 1. Small-diameter leads represented 39% of the cohort (n = 47). Medtronic SF accounted for 53% (n = 25) of small-diameter leads. St. Jude models, including Riata (n = 6), Riata ST (n = 7), and Durata (n = 8), accounted for 45% of small-diameter leads. There was 1 Biotronik (Biotronik SE and Company, Berlin, Germany) small-diameter model. Medtronic Sprint Quattro and Sprint series were the most commonly used standard-diameter leads, accounting for 53% (n = 39) and 33% (n = 24) of standard-diameter leads, respectively. The remaining 14% (n = 10) of standard-diameter leads comprised Endotak leads from Boston Scientific (Boston Scientific, Inc., Natick, Massachusetts). All implanted leads had an active fixation mechanism, regardless of diameter. All leads demonstrated acceptable performance at initial implant.
In comparing patients who received standard and small-diameter leads, there were no significant differences in baseline characteristics, including age at implant, height and weight at implant, and indication (Table 2). There were no significant differences in patients receiving SF leads and St. Jude small-diameter models.
Outcomes: rates of lead failure
Over a median follow-up of 28.7 months (IQR: 14.4 to 59.2 months), there were 25 documented lead failures in the cohort, yielding an overall failure prevalence of 21%, with a median time to failure of 29.9 months (IQR: 19.5 to 54.3 months). Failure prevalence for individual lead families in this cohort is outlined in Table 3. Lead fractures represented 80% of failures (n = 20). Other failure mechanisms included 12% (n = 3) sensing failure and 8% (n = 2) perforation.
The overall rate of lead failure was 5.6%/year (95% CI: 3.4% to 7.8%/year). The rate of lead failure in the standard-diameter group was 2.9%/year (95% CI: 1.0% to 4.7%/year). The rate of lead failure in the small-diameter group was significantly higher at 12.1%/year (95% CI: 6.1 to 18.0%/year). Within the small-diameter cohort, SF leads and St. Jude models did not have statistically significantly different failure rates (SF: 14.9%/year, 95% CI: 6.5% to 23.4%/year; and St. Jude: 7.9%/year, 95% CI: 0.2% to 15.6%/year).
In 79% of cases (n = 95), leads remained functional as of the last evaluation. This included 48% (n = 58) with current follow-up (median follow-up 43 months, IQR: 17.8 to 92.9 months), 26% (n = 31) that were lost to follow-up (median follow-up 24.8 months, IQR: 11.9 to 44.5 months), 4% (n = 5) with leads in situ at the time of death or transplant (range 5.7 to 58.4 months), and 1% (n = 1) that underwent elective extraction of a functional recalled lead after 8.4 months.
Clinical presentation of lead failures
Eight patients received at least 1 inappropriate shock due to lead failure, including 4 patients with multiple shocks. There were no pro-arrhythmic events. In 6 patients, a device alarm prompted medical evaluation; response to the alarm was immediate in 3, but occurred after a delay ranging from 5 to 30 days in the other 3 patients. Two of these 3 patients experienced an inappropriate shock during the period of delay. In 5 cases, a physician was notified of lead failure via remote monitoring (Table 4).
Lead failure management
Of the 25 lead failures, 22 leads (88%) were extracted, and 3 (12%) were abandoned. Most extractions (n = 21) used a transvenous approach; 1 was a hybrid transvenous/open surgical procedure. Extractions were performed at a mean patient age of 16.5 years, after a median implant duration of 29.7 months (IQR: 17.9 to 51.1 months). Extraction was achieved by simple manual traction in 4 cases, with the use of mechanical sheaths in 5, and with powered sheaths in 13 cases (radiofrequency-powered in 1, and laser-powered in 12) (Table 4). Complete procedural success was achieved in all cases, and all extracted leads were removed without residual fragments in the vasculature or myocardium. New ICD lead insertion was performed after lead extraction during the same procedure in all patients. There were no procedural complications.
Failed lead product analysis
Manufacturer product analysis was available for 15 of 22 (68%) extracted ICD leads, including 9 of 12 (75%) SF leads. Sites of SF lead fracture included the distal pace/sense conductor in 5, the proximal pace/sense conductor in 1, and both distal and proximal conductors in 2 leads. In the 1 remaining analyzed SF lead, the root cause of failure was not identified. Manufacturer analysis was available for 2 of 3 (67%) extracted St. Jude leads. In 1 lead, there was a proximal conductor fracture, and in the second, the root cause of failure was not identified (Table 4).
Kaplan-Meier survival curves for all leads, standard-diameter leads, SF, and St. Jude leads are shown in Figure 1. The SF leads had a 3-year survival probability of 69% and a 5-year survival probability of 44%, compared with 92% and 86%, respectively, for standard-diameter leads (p < 0.01). Beyond 3 years, only 2 St. Jude leads were available for analysis.
Predictors of lead failure
Simple Cox regression was used to assess for the association of patient and lead-related variables with lead survival. The SF design (p = 0.0003), small lead diameter (p = 0.0007), younger age at implant (13.4 ± 4.1 years vs. 16 ± 5 years, p = 0.006), history of primary electrical disease (p = 0.003), and a single-chamber device (p = 0.02) were significantly associated with lead failure (Table 5). Linear growth, dual coil leads, and history of previous lead failure or a previous ipsilateral lead were not associated with lead failure.
In a multivariate Cox regression model, small lead diameter conferred a hazard ratio (HR) of 3.99 (95% CI: 1.66 to 9.61; p = 0.002) for lead failure. In the analysis of small-diameter leads subdivided by design, SF design conferred a HR of 4.42 (95% CI: 1.73 to 11.29; p = 0.002) for lead failure. The HR for St. Jude small-diameter design was 3.05 (95% CI: 0.99 to 9.37; p = 0.052), although this did not reach statistical significance. A history of CHD was protective (HR: 0.15, 95% CI: 0.03 to 0.77; p = 0.023), when compared with a history of primary electrical disease.
The ICD is a lifelong therapy in children, and its efficacy depends upon lead and device performance over several decades. In light of the premature failure of 2 current-generation small-diameter leads, this study sought to evaluate lead performance in children in the present era. This is the first pediatric study to analyze the interaction of patient-specific variables, such as growth, with ICD lead diameter and design.
At the outset, it is important to recognize that size-dependent susceptibility to lead failure might not be directly related to absolute lead diameter but rather to the engineering and design process used to downsize leads. Individual manufacturers used different strategies to create small-diameter leads. For example, in designing the SF lead, engineers achieved a 23% reduction in diameter by replacing the multiple isolated compression lumens used in their standard ICD leads with an integrated compression lumen around conductor wires (22).
Riata leads were designed to be true bipolar leads, with downsized components and a silicone core structure with 3 or 4 lumens (23). With body diameters of 6.7- to 7.6-F, these were the first ICD models capable of introduction through an 8-F sheath (23,24). The Riata ST lead represented further downsizing to a 6.3-F body, by reducing the central lumen and moving the conductor cables closer to the lead central axis, without changing the diameter of the outer silicone insulation. In addition, Riata ST models incorporated flat wire shocking electrodes, backfilled with silicone (23).
After reports of premature lead failures in the Riata and Riata ST leads secondary to inside-out abrasion of the silicone inner core of the lead, the Riata ST Optim lead was created. Similar in design to its predecessor, the most notable difference was the addition of an outer coating made of a silicone and polyurethane co polymer (Optim), designed to improve abrasion resistance (23). The Durata lead builds further on the Riata ST Optim design, with the addition of a soft silicone tip and a slightly curved RV coil, both efforts to improve lead tip-endocardium interface and prevent myocardial perforations (7,12,23).
Since recognition of premature failure of SF and Riata leads, these models have been recalled, and the only remaining small-diameter lead on the market is the St. Jude Durata. Nevertheless, there remains a significant population of patients, including children and young adults, that continue to harbor SF, Riata, and Riata ST leads, because prophylactic lead extraction has not been universally recommended. The majority of prior reports on ICD lead failures related to small diameter have not focused on the pediatric population (6,7,12,25).
In this single-center pediatric study, we observed substantially higher rates of lead failure (5.6%/year for all leads) than reported in the adult published data (2%/year for all leads, in the recent publication by Rordorf et al. ). There might be additive effects on fracture risk from pediatric-specific variables, such as linear growth and physical activity, which might create increased mechanical stress on vulnerable leads. In our analysis, patients with primary electrical disease were more likely to experience lead failure than patients with CHD. We speculate that this difference might reflect more stringent activity restrictions in patients with complex CHD. This finding is similar to adult data that have demonstrated higher lead failure rates in patients with channelopathies or cardiomyopathies, when compared to patients with ischemic heart disease, also presumably secondary to levels of physical activity (26,27). In our study, height at implant and change in height were not significantly associated with lead failure. Younger age at implant was associated with lead failure in the univariate analysis, as has been seen in previous reports; however, this factor was not significant in the multivariate model.
In multivariate Cox regression analysis, small lead diameter conferred a 4-fold higher risk of failure than standard diameter. When small-diameter leads were analyzed according to design, SF design was the strongest predictor of lead failure, with an HR of 4.42. The HR for St. Jude small-diameter design was 3.05, although this did not reach statistical significance. These data suggest that design specifics are more clinically relevant to lead performance than absolute lead diameter.
In this cohort, SF leads exhibited a significantly higher failure rate than standard leads, confirming previously published reports in the adult published data. We found an overall failure prevalence of 48% in 25 SF leads. The 3-year survival probability for SF leads was 69%, and 5-year survival probability was 44% in this study. The observed rate of SF failure (14.9%/year, 95% CI: 6.5% to 23.4%/year) is higher than that reported in published lead survival series of adult (2.6% to 4.8%/year) and pediatric patients (9.1%/year) (9–12,19). There were only 2 St. Jude leads available for analysis beyond 3 years of follow-up, thus limiting conclusions with regard to chronic performance of these models.
The technical issues implicated in Riata and SF lead failure are different. The SF lead failure is usually due to a conductor fracture, resulting in sudden loss of function and a significant change in pacing thresholds and impedance (8,10,11,26,28). With the Riata family, lead failure is due to insulation damage, resulting in cable externalization; this can occur slowly over time without affecting pacing thresholds (11,14,28,29). The Riata failure mechanism might have some similarities to the design flaw that led to the 1994 recall of the Telectronics Accufix J (Telectronics Pacing Systems, Englewood, Colorado) pacing lead (model numbers 033-812, 329-701, and 330-801), which was prone to external protrusion of its retention wire (30,31). A recent case report described a Durata lead failure due to internal conductor migration and abrasion in the absence of cable externalization (32). This has prompted 1 author to speculate that the unique flat ribbon design of the conductor spaces might allow for cable migration and internal abrasion (33). This proposed mechanism is 1 example of a potential link between downsized design features and susceptibility to failure.
In our study, returned product analysis was available for 68% of extracted leads, most of which were SF leads; in these cases, the failure mechanism was predominantly conductor fracture, consistent with published reports. Product analysis was available for only 1 failed Riata ST lead, which did not demonstrate the typical inside-out abrasion fracture related to insulation damage.
Improved lead integrity alarm algorithms have been introduced since the recognition of premature fracture of small-diameter leads (34–36). An important finding in our study was that only a small number of patients with lead failure recalled having heard a device alarm, and only one-half of those recognizing an alarm sought medical care in prompt fashion. Two patients experienced an inappropriate shock, despite having heard an alarm. In these cases, the child misinterpreted the audible alarm as electronic noise from a phone or video game and ignored it. This draws attention to the need for improving audible alarm tones as well as patient education on alarm recognition and significance, especially in the pediatric population. In this cohort, 8 patients (32% of lead failures) received at least 1 inappropriate shock, and 4 (16% of lead failures) received multiple shocks, a phenomenon that can be particularly devastating in a young child.
With regard to lead management, lead extraction was the most common strategy, used for 88% of failed leads. Only 3 failed leads were abandoned in situ, and this was on the basis of physician discretion. All lead extraction attempts were successful, irrespective of post-implantation time. Of note, 100% of SF leads were safely extracted.
This is a retrospective study, but all consecutive patients receiving transvenous ICDs during the study period were included to minimize bias. The number of young patients meeting indications to receive ICDs is small, and therefore the number of patients in our study and in each subgroup is also small, which might impact study results. However, despite small numbers, this is an important subgroup of patients that warrants analysis and reporting, because there was a remarkable increase in transvenous ICD insertion in young patients in response to the marketing of small-diameter leads. The lead failure rates in this subgroup are much higher than in adult cohorts, conferring significant morbidity to young children. Silent lead failures, especially with the Riata leads, might have been under diagnosed, because it is only recently that patients have been undergoing systematic fluoroscopy to assess for cable externalization. A subclavian crush mechanism has been implicated in the SF lead failure mechanism. In this study, subclavian venous access was used in all cases, making it difficult to speculate whether an alternative access site would have influenced lead failure rates. However, most of the available returned product analysis pertaining to the SF leads reported fracture in the distal tip of the lead. Implanter technique and experience might affect lead survival, but owing to a large number of operators spanning different generations of lead design, it was not feasible to include implanting physician as a variable in the model.
In this single-center pediatric study, we observed higher ICD lead failure rates than have been published in adult series. In an analysis of lead diameter, lead design, and patient factors, the SF lead design conferred the highest risk of lead failure, suggesting that design rather than diameter is the critical issue in ICD lead survival. Small-diameter leads are optimal for transvenous ICD insertions in the pediatric population; therefore, research efforts should focus on redesigning a small-diameter ICD lead with enhanced durability and lead performance.
The authors thank Xuemei Zhang, MS, for statistical analysis.
Dr. Shah has served on the Speakers' Bureau for Medtronic. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- congenital heart disease
- confidence interval
- hazard ratio
- implantable cardioverter-defibrillator
- interquartile range
- right ventricle/ventricular
- Sprint Fidelis
- Received June 25, 2013.
- Revision received August 21, 2013.
- Accepted September 9, 2013.
- American College of Cardiology Foundation
- Berul C.I.,
- Van Hare G.F.,
- Kertesz N.J.,
- et al.
- Atallah J.,
- Erickson C.C.,
- Cecchin F.,
- et al.
- Cooper J.M.,
- Stephenson E.A.,
- Berul C.I.,
- Walsh E.P.,
- Epstein L.M.
- ↵Medtronic, Inc. Medtronic Technical Concept Paper: Insights on Sprint Design Enhancements. June 2004.
- ↵St. Jude Medical, Inc. St. Jude Medical ICD Lead Design and Long-Term Performance. May 2013. Available at: http://professional.sjm.com/professional/resources/product-performance/riata-important-info/overview. Accessed May 15, 2013.
- St. Jude Medical, Inc. St. Jude Medical Announces FDA Approval of the World's Thinnest, Most Advanced Defibrillation Lead Family. News Release, April 2002. Available at: http://investors.sjm.com/phoenix.zhtml?c=73836&p=irol-news. Accessed May 15, 2013.
- Hauser R.G.,
- Maisel W.H.,
- Friedman P.A.,
- et al.
- Kleemann T.,
- Becker T.,
- Doenges K.,
- et al.
- Dorman H.G.,
- van Opstal J.M.,
- Stevenhagen J.,
- Scholten M.F.
- Kay G.N.,
- Brinker J.A.,
- Kawanishi D.T.,
- et al.
- Swerdlow C.D.,
- Gunderson B.D.,
- Ousdigian K.T.,
- Abeyratne A.,
- Sachanandani H.,
- Ellenbogen K.A.
- Swerdlow C.D.,
- Gunderson B.D.,
- Ousdigian K.T.,
- et al.