Author + information
- Received March 24, 2016
- Revision received December 23, 2016
- Accepted December 30, 2016
- Published online August 21, 2017.
- Dominic A.M.J. Theuns, PhDa,∗ (, )
- Lieselot van Erven, MD, PhDb,
- Geert P. Kimman, MD, PhDc,
- Carel C. de Cock, MD, PhDd,
- Arif Elvan, MD, PhDe,
- Marco A. Alings, MD, PhDf,
- Jurren van Opstal, MD, PhDg and
- Mathias Meine, MD, PhDh
- aDepartment of Cardiology, Erasmus Medical Center, Rotterdam, the Netherlands
- bDepartment of Cardiology, Leiden University Medical Centre, Leiden, the Netherlands
- cDepartment of Cardiology, Medical Centre Alkmaar, Alkmaar, the Netherlands
- dDepartment of Cardiology, VU University Medical Centre, Amsterdam, the Netherlands
- eDepartment of Cardiology, Isala Klinieken, Zwolle, the Netherlands
- fDepartment of Cardiology, Amphia Hospital, Breda, the Netherlands
- gDepartment of Cardiology, Medisch Spect Twente, Enschede, the Netherlands
- hDepartment of Cardiology, University Medical Centre, Utrecht, the Netherlands
- ↵∗Address for correspondence:
Dr. Dominic A.M.J. Theuns, Erasmus Medical Center, Department of Cardiology, PO Box 2040, 3000 CA, Rotterdam, the Netherlands.
Objectives This study sought to determine prospectively the rate of conductor externalization (CE), and whether this was associated with electrical failure.
Background The Riata family of defibrillator leads was placed under U.S. Food and Drug Administration advisory as of November 28, 2011 because of high rates of CE.
Methods A nationwide cohort established in 2012 of 1,029 patients with recalled Riata leads with 147 CE were followed until death, lead discontinuation, or 3 annual screenings with fluoroscopy and device interrogation.
Results Follow-up of 882 patients with normal baseline fluoroscopy revealed incident overt CE in 95 leads (11%) after median risk time of 2.9 years, yielding an incidence rate of 4.9 (95% confidence interval [CI]: 3.9 to 5.9) per 100 patient-years. The incidence rate was significantly higher in 8-F Riata leads than in 7-F Riata ST leads (7.0 vs. 3.2 per 100 patient-years; p < 0.001). Electrical follow-up demonstrated electrical abnormality in 77 leads, resulting in an incidence rate of 4.0 (95% CI: 3.2 to 5.0) per 100 patient-years. The incidence rate of electrical abnormalities was not different between leads without CE and those with CE (3.9 vs. 5.2 per 100 patient-years; p = 0.39).
Conclusions The development of CE is progressive in nature with an incidence rate of new CE of 4.9 per 100 patient-years, with a higher rate for 8-F Riata leads than for 7-F Riata ST leads. Despite the high rate of structural failure, no association between development of CE and electrical failure was observed.
Insulation abrasion is the most common cause of implantable cardioverter-defibrillator (ICD) lead failure (1). The 8-F Riata and 7-F Riata ST family of ICD leads (St. Jude Medical, Sylmar, California) present with a unique failure mechanism. High-voltage and/or low-voltage conductor cables wear through the silicone insulation, that is, inside-out abrasion, and appear outside the lead body, resulting in conductor externalization (CE) (2–5). On December 4, 2011, a Class I recall was issued for the Riata and Riata ST silicone leads by the U.S. Food and Drug Administration (6). The U.S. Food and Drug Administration recommended imaging via fluoroscopy or 2-view chest x-ray to identify insulation failure with CE (7). Cross-sectional studies found a prevalence of CE ranging between 11% and 27%, including leads with normal electrical function but that exhibit CE (8–14). However, the natural history and clinical sequelae of CE are still unclear. Is electrical dysfunction a precursor of CE or vice versa? In addition, the annual rate of CE over service time after implantation is unknown. In order to address these questions, we report the longitudinal follow-up of advisory Riata and Riata ST high-voltage defibrillation leads from an independent group of investigators who represent all ICD implantation centers in the Netherlands.
Study population and data collection
On January 4, 2012, the Device Advisory Committee of the Netherlands Heart Rhythm Association issued a recommendation to identify all patients with an active Riata or Riata ST silicone high-voltage defibrillation lead and to perform annual fluoroscopic screening of the lead. All 31 Dutch ICD implantation centers were contacted to collect data in a standardized format, such as lead model, serial number, date of implantation, date of screening, presence of CE, the location of CE, and lead parameters.
The study included patients with an active Riata (model 1570, 1580, 1581, and 1582) or Riata ST (model 7000, 7001, 7002, and 7040) high-voltage defibrillation lead in 1 of the Dutch ICD implantation centers. The respective leads were implanted between July 2002 and November 2008.
Fluoroscopy and externalized conductors
Fluoroscopy in the electrophysiology or intervention laboratory was performed in all patients. Cine loops of the high-voltage defibrillation lead were obtained in anteroposterior, left-anterior-oblique, and right-anterior-oblique projections. Additional projections or magnification settings as needed for better identification of CE were left to the discretion of the investigators. The full length of the lead, from pocket region to the tip in the right ventricle, was screened. At each ICD implantation center, the cine loops were examined at the time of image acquisition and reviewed by a local investigator, a cardiologist certified in cardiac pacing with extensive experience in device implantation and lead cine fluoroscopy. In case of questionable CE, local investigators were advised to consult members of the Device Advisory Committee for adjudication of the cine loops. The presence of externalized conductors was defined as follows: 1) the appearance of conductor cable(s) outside the lead body; and 2) a change in the radius of curvature of the suspected externalized conductor as compared to the lead body. The location of CE was divided into 3 zones: 1) distal from tricuspid valve annulus; 2) between superior vena cava (single-coil leads) or proximal coil (dual-coil leads) and tricuspid valve annulus; 3) between pocket and superior vena cava (single-coil leads) or proximal coil (dual-coil leads).
ICD interrogation and electrical failure
Standard ICD interrogation was performed to obtain standard lead parameters, for example, R-wave sensing amplitude, ventricular capture threshold, low-voltage impedance (LVI), and high-voltage impedance (HVI). Ventricular electrograms were examined with and without pocket manipulation and isometric muscle contractions for the presence of nonphysiological noise. Electrical failure was defined by absolute limits and relative changes compared with previous screenings: 1) LVI outside the range 200 to 2.000 Ω or ≥50% change compared with previous screenings; 2) HVI outside the range of 20 to 200 Ω or ≥50% change compared with previous screenings; 3) pacing threshold ≥5 V or at least 100% increase, which results in threshold ≥2.5 V; or 4) presence of nonphysiological signals not due to external interference.
The data are presented using descriptive statistics. Normality of distribution was determined by the Shapiro-Wilk test. Normally distributed continuous variables are expressed as mean ± SD and compared by Student t test. Non-normally distributed variables are expressed as median with 25th and 75th percentiles and compared using Kruskal-Wallis H test or Friedman test, where applicable. Categorical data are expressed as percentages and compared with the chi-square test or Fisher exact test.
The incidence rate of CE was calculated using time at risk from baseline to the latest fluoroscopy in patients with normal baseline fluoroscopy. Incidence rates were expressed per 100 patient-years with a 2-sided 95% confidence interval (CI). Comparative analysis was performed by estimating the incidence rate ratio. In addition, event-free rates of CE were calculated using the Kaplan-Meier method. Curves were compared by the use of the log-rank test. Statistical analysis was performed using Stata (version 12 SE for Windows, StataCorp, College Station, Texas) and PASW (version 21, IBM Corp., Somers, New York). A p value of <0.05 was considered statistically significant.
In 2012, at the first nationwide screening (baseline), 1,029 patients with an active Riata or Riata ST high-voltage defibrillation lead were invited for fluoroscopy and ICD interrogation at the Dutch ICD implantation centers. The results of this baseline screening have been published previously (9). In brief, externalized conductors were observed in 147 leads, yielding a prevalence of 14.3%. At baseline screening, the median dwell time of Riata/Riata ST leads was 5.0 years (4.2 to 5.9 years). The median dwell time of leads with CE was longer than for those without CE (5.5 vs. 4.9 years; p < 0.001). Figure 1 presents the longitudinal follow-up of these patients. At the 2015 screening, the number of patients with an active Riata or Riata ST lead was reduced to 530 (52%). During 3 years follow-up, 184 patients died (including 32 patients who died after lead revision), 6 patients underwent cardiac transplantation, and ICD therapy was disabled in 15 patients. A total of 279 leads were discontinued (64 extractions and 215 abandonments).
In 882 patients with normal baseline screening, 3 consecutive annual fluoroscopic screenings were performed. During a median risk time of 2.9 years (1.1 to 3.0 years), 95 leads (11%) exhibited overt CE, resulting in an incidence rate of 4.9 (95% CI: 3.9 to 5.9) per 100 patient-years. The models and respective numbers of leads with externalized conductors are presented in Table 1. The incidence rate was significantly higher in 8-F Riata leads than in 7-F Riata ST leads (7.0 vs. 3.2 per 100 patient-years; p < 0.001). No significant difference in incidence rate was observed with regard to the number of defibrillation coils (single vs. dual, 4.2 vs. 5.3 per 100 patient-years; p = 0.27). The majority of CE (78%) was located in the distal part of the affected leads, that is, below the tricuspid valve annulus.
The event-free rates of CE for the 7-F Riata ST and 8-F Riata leads are displayed in Figure 2. Time 0 presents the prevalence of CE as observed at baseline screening, 7.9% for 7-F Riata ST and 21.6% for 8-F Riata leads, respectively. At the 3-year follow-up, the event-free rate of externalization for 8-F Riata leads was significantly lower than for 7-F Riata ST leads (59.9% vs. 82.4%; p < 0.001).
When taking baseline and 3 annual consecutive fluoroscopic screenings into account, the total proportion of Riata/Riata ST leads presenting with overt CE was 23.5% (147 at baseline and 95 after longitudinal follow-up).
Table 2 shows the electrical parameters of the 8-F Riata and 7-F Riata ST leads obtained at the annual screenings. Of these parameters, only a significant difference in HVI was observed over the different annual screenings, 50 Ω (42 to 62 Ω) at baseline versus 57 Ω (47 to 72 Ω) at screening in 2015 (p < 0.001). In order to assess whether changes in electrical parameters are associated with the development of CE, electrical parameters obtained at the detection of CE were compared with electrical parameters obtained at the previous screening. No significant differences in R-wave amplitude, stimulation threshold, LVI, and HVI were observed.
Predefined electrical abnormalities
During a median risk time of 2.9 years (1.1 to 3.0 years), electrical abnormality was observed in 77 leads, resulting in an incidence rate of 4.0 (95% CI: 3.2 to 5.0) per 100 patient-years. No significant difference in incidence rate was observed with regard to lead diameter (p = 0.78) or the number of defibrillation coils (p = 0.89). The most common pre-defined electrical abnormality was a change in impedance (54%). Further pre-defined electrical abnormalities are listed in Table 3.
The majority of electrical abnormalities were observed in leads without CE (87%). The incidence rate of electrical abnormalities was not different between leads without CE and those with CE (3.9 vs. 5.2 per 100 patient-years; p = 0.39). After 3 annual screenings, the total proportion of Riata leads with electrical abnormalities was 9.0% (16 at baseline and 77 during follow-up).
In December 2011, a class I recall was issued for the 8-F Riata and 7-F Riata ST leads because of a unique insulation defect with inside-out abrasion that can lead to CE. Data of longitudinal studies are scarce and a clear scope of CE and the association with electrical performance has not been set. The main finding of the present longitudinal study is the progressive nature of externalization with an incidence rate of new CE of 4.9 per 100 patient-years. No association between electrical abnormalities and the development of externalization was found and electrical abnormalities were more prevalent among leads without CE.
The presence of externalized conductors is a common finding in cross-sectional studies using fluoroscopy. In these studies, the reported prevalence of CE ranged from 11% to 27% and was more common in 8-F Riata leads (8–14). These results were confirmed in a recent meta-analysis demonstrating an overall prevalence of CE of 23.1% (95% CI: 19.0% to 27.6%) with a 3-fold higher prevalence of externalization for 8-F Riata leads than for 7-F Riata ST leads (15). However, the cross-sectional design of observational studies prohibits in determining the rate of CE over time. Data on longitudinal follow-up are scarce (16–18). Among 239 patients with normal baseline fluoroscopy and repeated fluoroscopy after 1.1 years, Larsen et al. (17) found an incidence rate of new CE of 3.7 per 100 patient-years. They observed no significant differences in incidence rate with regard to lead diameter, number of defibrillation coils, and dwell time. McKeag et al. (18) prospectively followed 140 patients with normal baseline fluoroscopy for 3 years. During this follow-up, CE was observed in 11 leads, resulting in an incidence rate of 3.6 per 100 patient-years. In our study, we found an incidence rate of 4.9 per 100 patient-years over 3 years’ follow-up among 882 patients with normal baseline fluoroscopy. No differences in incidence rates were observed with regard to the number of coils and dwell time, but the incidence rate was significantly higher for 8-F Riata leads (7.0 per 100 patient-years). The higher rate of CE in 8-F Riata leads can be explained by longer dwell time for 8-F Riata leads due to earlier market introduction and lead design improvements in 7-F Riata ST leads.
The association between CE and electrical abnormalities is unsettled. In a recent systematic review on cross-sectional studies, there appears to be a 6-fold increased risk of electrical failure in the presence of conductor externalization (15). In the Danish longitudinal study, Larsen et al. (17) found a 4-fold higher incidence rate of electrical abnormalities in leads with externalization compared with those without. In contrast, the preliminary results of the multicenter prospective Riata Lead Evaluation Study did not demonstrate any difference in the rate of electrical failure in leads with or without CE over a follow-up of 10 months (11). In our study with a follow-up of 3 years, no association between electrical failure and CE was found. McKeag et al. (18) found no electrical abnormalities in any patient during 3 years of follow-up. A possible explanation for the absence of electrical abnormalities in leads with externalization might be survivor bias. Patients who presented with CE underwent lead replacement and are therefore lost from the study.
Despite there being no association between electrical failure and CE, the absolute risk of electrical failure is not 0 in leads without CE. In our study, the rate of electrical failure in leads without externalization is 3.9 per 100 patient-years. On this basis, a normal-appearing Riata lead is still not without risk. Of note, malfunction of Riata leads may not be reflected by routine device interrogation. Failure to deliver a shock despite normal electrical parameters has been observed at defibrillation threshold testing (19,20). A possible explanation is that the delivery of low current during noninvasive measurement of HVI might not be enough to reveal a short circuit, which can only be unmasked by the delivery of shocks. The extent of electrical failure may be underestimated by performing noninvasive device interrogation.
Clinical implications of the study findings
Conductor externalization in Riata leads has quickly emerged as the latest large-scale medical device malfunction. The clinical management of patients with a Riata/Riata ST high-voltage lead is still unsettled. Is a clear intervention or treatment available for in situ Riata leads at the time of diagnosis of CE? Consensus on prophylactic extraction of Riata leads with conductor externalization and a normal electrical function is lacking. To date, only a few studies reported on the performance of extraction of Riata leads (21–24). Some studies showed no difference in procedural outcomes of extracting Riata leads compared with Sprint Fidelis leads (Medtronic, Minneapolis, Minnesota) (21,23). In contrast, the multicenter study by Maytin et al. (22) demonstrated that the extraction of Riata/Riata ST leads is feasible but can be challenging, especially in those with externalized conductors. The study by Bongiorni et al. (24) had similar findings on the feasibility and complexity of extraction of Riata leads. Both studies concluded that extraction of Riata leads should be performed in specialized centers with extensive experience in lead extraction. The clinical management of externalized but functional Riata/Riata ST leads is complex. The decision to extract must be individualized by considering multiple variables, such as age, the institution’s experience in lead extraction, and clinical characteristics such as diabetes, renal insufficiency, and low body weight, which are associated with increased risk for complications. Another option is comprehensive monitoring of electrical parameters (25). However, noninvasive lead measurements may not detect all electrical abnormalities in Riata/Riata ST leads that can be unmasked only by the delivery of shocks. The role of defibrillation testing in clinical practice has been questioned but testing at elective generator change may identify electrical failure of Riata/Riata ST leads and thus the need for lead revision.
Several limitations of the present study warrant consideration. For patients who died between annual screenings, the actual status of the lead—whether externalized conductors and or electrical abnormalities were present—is unknown. Therefore, our result may underestimate the actual rate of externalized conductors and/or electrical abnormalities. Another limitation is the lack of data on inappropriate shocks triggered by electrical dysfunction of the Riata/Riata ST leads, such as noise. Furthermore, per protocol, defibrillation testing was not performed.
The development of externalized conductors is progressive in nature with an incidence rate of new CE of 4.9% per 100 patient-years, with a higher rate for 8-F Riata leads than for 7-F Riata ST leads. Despite the high rate of structural failure, the rate of electrical failure during follow-up was low. In addition, no association between development of CE and electrical failure was observed. The majority of electrical abnormalities were observed in leads without externalization. The management of patients with Riata leads with overt externalization in the absence of electrical abnormalities must be individualized on a case-by-case basis by weighing the risks of lead revision (extraction or abandon) or a meticulous clinical follow-up.
COMPETENCY IN MEDICAL KNOWLEDGE: The development of conductor externalization in Riata leads is progressive in nature. No association between development of structural failure and electrical failure was observed.
TRANSLATIONAL OUTLOOK: Future research should address the clinical management of patients with Riata leads presenting with structural failure and a normal electrical function.
Dr. Alings has received honoraria for advisory board service or presentations from Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, and Pfizer. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- conductor externalization
- confidence interval
- high-voltage impedance
- implantable cardioverter-defibrillator
- low-voltage impedance
- Received March 24, 2016.
- Revision received December 23, 2016.
- Accepted December 30, 2016.
- 2017 American College of Cardiology Foundation
- Kleemann T.,
- Becker T.,
- Doenges K.,
- et al.
- ↵St Jude Medical. Medical Device Advisory. Important Product Information Update. Available at: http://www.sjmprofessional.com/Resources/product-notices-advisories/riata-improtant-information-us.aspx 2011. Accessed March 2, 2016.
- ↵U.S. Food and Drug Administration. FDA Safety Communication: Premature Insulation Failure in Recalled Riata Implantable Cardioverter Defibrillator (ICD) Leads Manufactured by St. Jude Medical, Inc. November 20, 2012. Available at: http://www.fda.gov/MedicalDevices/Safety/AlertsandNotices/ucm314930.htm. Accessed March 2, 2016.
- Theuns D.A.,
- Elvan A.,
- de Voogt W.,
- de Cock C.C.,
- van Erven L.,
- Meine M.
- Larsen J.M.,
- Nielsen J.C.,
- Johansen J.B.,
- et al.
- McKeag N.A.,
- Hodkinson E.C.,
- Noad R.L.,
- et al.
- Richardson T.D.,
- Kolek M.J.,
- Goyal S.K.,
- et al.