Long-term Outcome of Implantable Cardioverter/Defibrillator Lead Failure

Abstract

This study compares outcomes of a recalled implantable cardioverter/defibrillator lead with a control lead in individuals in Iceland.

Implantable cardioverter/defibrillators (ICDs) are established for treatment and prevention of sudden death from ventricular arrhythmias.¹ An ICD system consists of a generator and a lead. The design of the ICD leads differs between manufacturers, some of which have had higher lead failure rates than others.²

The Riata defibrillator leads, manufactured by St Jude Medical Inc, were introduced in 2002. They had a novel silicone insulation design that was found to be prone to a specific insulation abrasion, characterized by externalization of the conductor (EC). These leads were designated to a class I recall by the US Food and Drug Administration in late 2011 owing to significant rates of lead failure and an insulation defect, posing significant management challenges.

After the recall, a specific protocol was initiated for heightened surveillance of these individuals at the National University Hospital of Iceland in Reykjavik, Iceland, where all ICD recipients in Iceland are followed up in a designated clinic, including fluoroscopy of the leads to evaluate for EC.

Methods

We identified all 52 individuals who had the recalled lead implanted from November 2002 to October 2009 in Iceland, along with 50 individuals who had ICDs from other manufacturers from February 2010 to November 2012 to serve as controls. We determined the occurrence of lead failure by (1) EC, defined as direct visualization of the conductor wires outside the lead body on fluoroscopy or being visualized on a chest radiograph; (2) electrical dysfunction, high-frequency, low-amplitude irregular signal (electrical noise) between ventricular signals on the intracardiac ICD electrogram; and (3) lead fracture. Additionally, we compared inappropriate shocks and deaths. The University Hospital Institutional Review Board approved the study. Because the data collection was retrospective, informed consent was waived by the institutional review board.

Hazard ratios for risk of lead failure and death between the groups were estimated with Cox regression models, adjusting the propensity score of having this lead, using demographic and device characteristics. Statistical analyses were performed using R statistical software, version 3.6.0 (R Foundation for Statistical Computing). A 2-sided P value less than .05 was statistically significant. Analysis began February 2019.

Results

Across the full study period, there were 52 individuals in the recalled lead group and 50 individuals in the control group. There were no significant differences with respect to demographic and device characteristics between the groups (Table). In the recalled lead group, 7 individuals (13.5%) had electrical dysfunction, 2 (3.8%) had EC, 7 (13.5%) had both, and 3 (5.7%) had lead fractures across the full study period. Therefore, in total, 19 individuals (36.5%) had lead failure, including 1 individual in whom therapy for ventricular tachycardia was not delivered, resulting in death. In the control group, there were 4 (8.0%) (P < .01; difference, 0.29; 95% CI, 0.12-0.46) with lead failure, 1 (2.0%) with electrical dysfunction, and 3 (6.0%) with lead fractures across the full study period. When follow-up was limited to 8.5 years for both groups, the number of cases of unexpected lead failure was 4.41 per 100 patient-years for patients with the recalled lead but 1.13 per 100 patient years for those with another type of ICD lead.

Table. Baseline Patient Characteristics.

Characteristic	Group, No. (%)		P Value
	Recalled Lead (n = 52)	Control (n = 50)
Age at implant, median (IQR), y	54.5 (16-80)	56 (12-77)	.73
Sex
Male	40 (76.9)	36 (72.0)	.73
Female	12 (23.1)	14 (28.0)	.73
Coronary artery disease	33 (63.5)	23 (46.0)	.12
Left ventricular ejection fraction, %
<30	9 (17.3)	12 (24.0)	.55
30-50	29 (55.8)	21 (42.0)	.23
>50	14 (26.9)	17 (34.0)	.57
Device type
Single chamber	27 (51.9)	23 (46.0)	.69
Dual chamber	23 (44.2)	22 (44.0)	>.99
Biventricular	2 (3.8)	5 (10.0)	.26
No. of defibrillation coils
Single	7 (13.5)	4 (8.0)	.57
Dual	45 (86.6)	46 (92.0)	.57
Follow-up, median (IQR), y	8.23 (2.25-15.98)	7.6 (0.27-9.26)	NA

Abbreviations: IQR, interquartile range; NA, not applicable.

After propensity score adjustment (with equally long follow-up times), individuals who had a recalled lead implanted were at higher risk of lead failure (25.0% [13 of 52] vs 6.0% [3 of 50]; hazard ratio, 4.67; 95% CI, 1.18-18.50; Figure) but lower risk for death (7.7% [4 of 52] vs 20.0% [10 of 50]; hazard ratio, 0.25; 95% CI, 0.07-0.87).

Figure. Risk of Lead Failure in the First 8.5 Years.

Lead failures included electrical dysfunction, externalized conductor, and lead fracture. The follow-up time presented in this table was limited to 8.5 years for both groups.

In total, there were 14 deaths (26.9%) (6 cardiovascular and 8 other) in the recalled lead group but 10 (20.0%) (3 cardiovascular and 7 other) among controls. In the recalled lead group, 10 (19.2%) experienced an inappropriate shock compared with 8 (16%) in the control group (P = .87; difference, 0.032; 95% CI, −0.135 to 0.200).

Discussion

While a strength of the study would be almost complete follow-up and exemplary record keeping in Iceland, limitations would include a relatively small number of study participants and retrospective design. Implantable cardioverter/defibrillators are intended to sustain and improve quality of life but inevitably failure of these products can occur, resulting in injury or even death. While patients with the recalled leads did not receive more inappropriate shocks, the cumulative incidence of serious lead problems was high with more than one-third experiencing significant consequence. The lead failures among patients who had the recalled leads occurred over the course of the follow-up time, indicating that leads should be monitored for the full dwell time. While there is still no clear consensus about the root cause of the Riata lead failure,³ we conclude that development of more durable ICD leads with regards to biomaterials and engineering design is warranted.