Author + information
- Received November 11, 1996
- Revision received October 21, 1997
- Accepted November 19, 1997
- Published online March 1, 1998.
- Peter A Brady, MB, MRCPA,
- Paul A Friedman, MDA,
- Jane M Trusty, RNA,
- Suellen Grice, RNA,
- Stephen C Hammill, MD, FACCA and
- Marshall S Stanton, MD, FACCA,* ()
- ↵*Dr. Marshall S. Stanton, Division of Cardiovascular Diseases, Mayo Clinic, Rochester, Minnesota 55905.
Objectives. The purpose of this study was to determine the risk of epicardial lead failure during long-term follow-up and its mode of presentation.
Background. Despite the high prevalence of epicardial lead–based implantable cardioverter-defibrillators, their long-term performance is unknown, and appropriate follow-up has not been established.
Methods. The study group comprised all patients in whom an epicardial lead system was implanted at the Mayo Clinic between October 31, 1984 and November 3, 1994. The number of lead fractures and leads with fluid within the insulation and the mode of presentation were determined retrospectively by review of patient visits, radiographs of lead systems and data derived from formal lead testing.
Results. At 4 years, the survival rate free of lead malfunction, using formal lead testing, for 160 Medtronic epicardial patches (models 6897 and 6921) was 72% compared with 92.5% for the 179 Cardiac Pacemaker, Inc. (CPI) patches (models 0040 and 0041) (p = 0.01). In addition, five Medtronic patches in three patients had fluid within the lead insulation but no obvious fracture. No CPI patches had fluid identified within the leads. Of 330 Medtronic epicardial pace/sense leads (model 6917), the 4-year survival rate free of lead malfunction as assessed by lead testing was 96%. In all, 19 presentations of lead malfunction were found in 17 patients (2 patients had more than one lead fracture at different times). In 11 (58%) of these presentations, the patients were asymptomatic despite the presence of obvious lead fracture.
Conclusions. Epicardial lead malfunction is common on long-term follow-up, and some leads have a failure rate of 28% at 4 years. Many patients with fractured leads remain asymptomatic, despite involvement of multiple leads in some cases. Therefore, consideration should be given to regular periodic lead testing in addition to routine X-ray examination, as asymptomatic lead malfunction can present with normal chest X-ray findings.
Because of the effectiveness of the implantable cardioverter-defibrillator (ICD) in preventing sudden death [1, 2], as many as 25,000 persons each year in the United States are candidates for the devices under currently accepted criteria [3, 4]. Although consistent implantation of nonthoracotomy lead systems is now possible , many patients still have epicardial lead systems although little is known about the long-term durability of such systems or the appropriate follow-up of those patients. Therefore, in the present study we attempted to determine the long-term performance of epicardial lead systems, the clinical mode of presentation of epicardial lead malfunction and the utility of screening examinations, radiographs and formal lead testing in detecting epicardial lead system malfunction. On the basis of our findings, we propose guidelines for the long-term follow-up of patients with an ICD.
The study group comprised all 171 patients in whom an epicardial ICD lead system was implanted at the Mayo Clinic between October 31, 1984 and November 3, 1994. Characteristics of the study patients, including the number of previous and concomitant surgical procedures, are summarized in Table 1.
1.2 Lead Systems
Epicardial lead systems from two manufacturers, Medtronic, Inc. and Cardiac Pacemakers, Inc. (CPI), were used exclusively. Medtronic epicardial patch models 6897 (n = 64) and 6921 (n = 96) were analyzed. These are identical except for the connector pin, which is 6.5 mm in model 6897 and 3.2 mm in model 6921. Because of the similarity of these two models, data from each lead system were combined for the purposes of analysis. (Data supporting combining these two leads for analysis are presented under Results.) The CPI patches (n = 179) analyzed were models 0040 and 0041. Almost all of the epicardial pace/sense leads implanted during the study period were Medtronic model 6917. Only four CPI pace/sense leads (model 4312) were implanted; none of these had evidence of malfunction and all four were excluded from further analysis. All patches use silicone as the insulator, platinum iridium as the conductor material and stainless steel as the manifold (mandrel) connecting the lead body to the patch.
1.3 Device Implantation
Surgeons at our institution first implanted epicardial lead systems using CPI patches in 1984. Implantation of Medtronic patches began in 1989. Initially implanted pulse generators included AID B models (n = 16), CPI Ventak models 1500 through 1600 (n = 52), which did not have noninvasive testing capability (40%), and PRx I model 1705 (n = 16), Medtronic PCD models 7216 and 7217 (n = 86) and Ventritex model V-100C (n = 1), which did have noninvasive testing capability (60%). All devices were implanted with the patient in the postabsorptive state and under general anesthesia. In general, surgical implantation was performed with use of a left thoracotomy approach unless concomitant surgery, such as coronary artery bypass graft surgery (CABG) or valve surgery, was planned, when a median sternotomy approach was used.
1.4 Data Collection
Data reviewed included the patients’ medical records, our ICD data base, data obtained from elective lead testing during scheduled follow-up and radiographs obtained after implantation and during follow-up. Each X-ray film obtained after lead implantation in all patients was reviewed retrospectively by one of us (M.S.S.). All X-ray films of patients with known lead malfunction were reviewed separately, and a particular effort was made to attempt to identify any X-ray evidence of lead abnormality. For the patients without known lead abnormalities, the same investigator (M.S.S.) carefully reviewed each X-ray film in a systematic fashion, following each lead through its entire course. All X-ray films from each patient were subsequently rereviewed with experts from the lead divisions of the respective manufacturers in an attempt to identify any factors that may have contributed to lead failure. Thus, the films were all overread in an unblinded manner. In addition, X-ray films from all patients were reviewed separately, in a blinded manner, by an experienced electrophysiologist to determine the sensitivity and specificity of X-ray data when used alone in the diagnosis of lead problems.
After device implantation and before discharge, all patients underwent comprehensive lead testing that included pacing threshold, sensed electrogram measurement (in devices with this capability) and arrhythmia induction. Pacing and sensing function were assessed every 3 months. Additionally, our routine follow-up recommendation was to repeat arrhythmia induction testing at 3 and 12 months after implantation, yearly thereafter (in ICDs capable of programmed stimulation), at the time of pulse generator replacement or whenever clinically indicated. A posteroanterior (PA) and lateral chest radiograph was performed before discharge, after implantation and every 6 months thereafter according to our protocol. An abdominal radiograph was frequently, though not consistently, obtained if the entire lead system was not visible on the chest X-ray film. The mean number of X-ray films obtained in each patient during the follow-up period was 15 (range 1 to 113). At 3, 6 and 12 months after implantation, the proportion of patients who had follow-up X-ray data available was 91%, 85% and 93%, respectively. At 2 and 4 years the proportion of patients who had follow-up X-ray data available was 72% and 71%, respectively.
Lead malfunctionwas said to occur if one or more of the following was present: 1) lead conductor fracture evident on X-ray film; 2) lead conductor fracture evident on inspection of the lead at the time of operation; 3) evidence of lead malfunction at electrical testing including a) pacing threshold >10 V at 0.5-ms pulse width, b) impedance >2,000 ohms for pacing or >200 ohms for shock delivery, or c) documented random oversensing (i.e., not related to cardiac events) resulting in lead abandonment.
Fluid (i.e., serum or blood) was sometimes visualized within the lead body. Leads with such fluid were routinely abandoned, at times without being tested. The clinical implications of fluid within the lead body are unclear (see Discussion). If fluid was identified and none of the preceding observations defining lead malfunction were present, the finding was defined as a fluid event. Fluid events were analyzed separately and were not combined with lead malfunction data except where specified.
1.7 Statistical Analysis
Three separate analyses of lead survival were performed. Method 1 assumed that implanted leads were functioning in surviving patients at follow-up unless a lead malfunction was specifically documented. Thus, lead survival was the interval from implantation to the last follow-up visit or contact (at which point the data were censored) or the interval from implantation to documented failure. Method 2 determined lead survival as the interval from implantation to the most recent radiograph demonstrating an intact system (for those leads without malfunction) or as the interval from implantation to the documented failure. Because some patients had not had a recent X-ray film at the time of this analysis, follow-up times were shorter when this method was used. In Method 3 the most rigorous method, lead survival was the interval from implantation to the most recent electrical test documenting normal lead functioning or the interval from implantation to the documented failure. In all analyses, data for all leads free of failure were censored at the last follow-up end point. From these three end points (last patient follow-up, intact system on X-ray film and normal lead function on electrical testing), lead survival was determined by using the Kaplan-Meier method . A log-rank test was used to determine whether a difference between survival of different patches existed. Results are expressed as mean value ± SE.
2.1 Patients and Implanted Leads
One hundred seventy-one patients underwent initial implantation of an epicardial lead system (Table 1). Seventy-five of these patients received 160 Medtronic patches (model 6897 in 64 and model 6921 in 96). The other 96 patients received 179 CPI model 0040 and 0041 patches. In most cases Medtronic 6917 pace/sense epicardial screw-in leads were implanted (330 leads in 163 patients). Only four CPI model 4312 pace/sense leads were implanted; because none of these were found to have fractured or to have fluid within the insulation, they were excluded from further analysis. The mean follow-up time was 4.57 ± 0.08 years (maximal 5.4) and 3.39 ± 0.09 years (maximal 8.9) for the Medtronic and the CPI patch, respectively, and 8.34 ± 0.11 years for the Medtronic 6917 pace/sense lead.
2.2 Lead Malfunction and Fluid Events
Among the 179 CPI patches, three fractures (2%) were found in two patients (2%). These fractures occurred at 3.5 and 3.4 years after implantation. No fluid was found within the insulation of any CPI lead during visual inspection at pulse generator replacement operations.
Among the 160 Medtronic patches, 12 lead malfunctions (7.5%) were found in nine patients (12%); 10 (16%) of the involved patches were model 6897 and 2 (2%) were model 6921. These malfunctions occurred a mean of 4.43 ± 0.12 years (range 2.52 to 4.85) after implantation. In addition, five (8%) of the Medtronic 6897 patch leads in three patients were found to have fluid within the lead insulation with no documented evidence of fracture of the lead. Two of these patients also had fractures in other patches and are, therefore, also included in the lead malfunction group. No fluid events were identified in the 6921 patches. Among the 330 Medtronic 6917 pace/sense leads, seven fractures (2%) were found in seven patients (4%). No fluid events were observed in these leads.
2.3 Lead Survival After Implantation
Figs. 1–3⇓⇓show for each lead survival curves based on end points derived from 1) patient follow-up data, 2) X-ray data, and 3) data obtained from electrical lead testing. Fig. 1A depicts lead survival free from lead malfunction as assessed by data from Medtronic 6897 and 6921 patch leads combined. The mean freedom from lead malfunction of these patches at 4 years, as assessed by using all three end points, was 80.2 ± 6.3%. Fig. 1B shows survival free of lead malfunction and fluid events for both Medtronic patches combined. At 4 years of follow-up, 23% of leads had evidence of malfunction (including fracture and fluid events) as assessed with use of patient follow-up data, and 30% had evidence of malfunction as assessed by lead testing. When data derived retrospectively from X-ray films were analyzed, the mean failure rate of the Medtronic patches at 4 years was 27%. Fig. 1C illustrates freedom from malfunction of each of the Medtronic 6897 and 6921 patches when analyzed separately. No significant difference between the lead survival rates of models 6897 and 6921 was found within the shorter follow-up period of model 6921.
Lead survival for CPI model 0040 and 0041 patches is shown in Fig. 2. At 4 years the fracture-free lead survival rate as assessed by patient follow-up data was 96.2 ± 2.1%. This rate was significantly higher than that found for the Medtronic patch at 4 years (p < 0.01). No fluid events were observed in the CPI 0040 or 0041 patches. Fig. 3illustrates freedom from malfunction of the Medtronic 6917 pace/sense lead. The marked decrease in survival at the 8-year follow-up period follows a single event occurring in a small group of patients at this late time after implantation. Overall, for this lead, mean freedom from malfunction at 4 years as assessed by all three data end points was 96.7 ± 1.6%.
2.4 Mode of Presentation of Lead Malfunction (Table 2)
In total, 17 patients with 22 lead malfunctions and 5 fluid events presented on 19 occasions (2 patients had lead problems on more than one occasion). Seven patients with fractured pace/sense electrodes presented with oversensing, which was the most common mode of presentation. An isolated fluid event without evidence of concomitant lead fracture occurred in one patient and was discovered because of failure of pacing at the highest voltage during pace/sense lead testing. One patient, with a fracture in each of two patches, presented with unsuccessful defibrillation of spontaneous in-hospital ventricular fibrillation and required external cardioversion. In 11 (58%) of the 19 presentations, the patients were asymptomatic. These 11 patients comprised three patients whose condition was diagnosed after failed defibrillation; three whose diagnosis was made after lack of pacing output during routine electrophysiologic lead testing and follow-up, including one patient with an isolated fluid event; 4 patients whose lead malfunction was diagnosed during routine pulse generator replacement for battery depletion; and one patient in whom a previously unsuspected diagnosis of lead malfunction was made on X-ray evidence alone. Careful retrospective review of all relevant radiographs demonstrated that evidence of lead fracture was apparent in only 68% of the 19 presentations. No false positive “fractures” were detected. In the blinded review, sensitivity of X-ray data when used alone to diagnose lead fracture was 71%. When all lead malfunction events were included in the analysis (lead fracture and fluid events causing lead problems), the sensitivity of X-ray data was 62%. The specificity of X-ray diagnosis was 100% as there were no false positive findings.
Six patients died suddenly during the follow-up period. No lead fractures were identified in any of these patients. In all except one of these six patients, the ICD system was implanted before 1990, and interrogations are not available to indicate whether a failed shock might have been the cause of death. The maximal interval between the occurrence of sudden death and follow-up in these six patients was 3 months for X-ray examination and 6 months for lead testing.
2.5 Site of Fracture
All leads were intact without marked bending at the site of eventual fracture, on review of the first X-ray film after implantation. The fracture site of Medtronic patch and pace/sense leads is shown in Fig. 4. The exact fracture site was not visible in two epicardial pace/sense leads (model 6917), and was unavailable for one lead. No specific fracture site was found predominantly in either lead, suggesting the importance of complete visualization of the entire lead system on all follow-up X-ray films.
This study provides information on the long-term performance of two epicardial defibrillation lead systems that account for the vast majority of epicardial leads already implanted worldwide. We found that epicardial lead malfunction during long-term follow-up is more common than previously reported . Specifically, lead malfunction was identified in 22 leads (12 Medtronic model 6897 and 6921 patches [7.5%], 3 CPI patches [2%] and 7 Medtronic pace/sense leads [2%]). In addition, 5 Medtronic 6897 patch leads (8%) had fluid within the lead insulation on direct inspection. However, no fluid was observed within the lead insulation of the CPI patches or the Medtronic 6921 patches or 6917 pace/sense leads. Of particular importance was the finding that most lead failures occurred >2 years after implantation (mean 4.43 ± 0.12 years [Medtronic patches] and 3.39 ± 0.02 years [CPI patches] after implantation as assessed from data derived from lead testing [Figs. 1–3]). This finding argues for continued close late follow-up of all patients with an epicardial lead system.
Most patients with lead malfunction were asymptomatic, despite multiple lead fractures in some cases, (Fig. 5). We could not identify a patient characteristic predictive of multiple lead fractures. One patient had been involved in a motor vehicle accident 9 months before presentation with fractures of two CPI patches. Although in many cases fractures could be identified after careful retrospective review of the chest or abdominal radiograph, in which the entire lead system was visible, evidence of a fracture could be quite subtle (Fig. 5). In almost one third of presentations no fracture was visible. Fluid within the lead insulation could be detected only by visual inspection of the lead at operation.
The high failure rate of epicardial leads in our patients may be a result of several factors, including surgical trauma at the time of implantation, mechanical forces related to cardiac motion or chemical factors causing lead degradation. The possibility that lead malfunction was related to surgical technique is unlikely, as surgeons at our institution first began implantation of epicardial defibrillators in 1984 by using CPI lead systems. Medtronic patches were first implanted in 1989. Therefore, any learning curve related to device implantation would most likely have coincided with implantation of the CPI lead systems, which was not the case in our experience. Similarly, it is possible that the increased tunneling required during a lateral thoracotomy approach, used most frequently by the surgeons performing implantation in this study, may have contributed to lead complications. If this was the case, similar rates of complications might be expected for CPI and Medtronic leads and would be expected to occur sooner after implantation than they did. Furthermore, review of the X-ray fracture sites on earlier radiographs with normal findings did not reveal any acute bend of the involved lead that might have led to undue strain and fracture. Such a mechanism is believed to be responsible for a least some fractures reported previously [7–11]. Fractures were not observed in this study before 2 years of follow-up, again supporting the hypothesis that the observed lead problems were unlikely to be due to surgical technique or to lead trauma during implantation .
The role played by other physiochemical factors is poorly defined. A definitely fractured lead will not function normally, and it should be replaced. Less clear is the management approach when fluid is discovered within the lead insulation (defined as a fluid event in this analysis) but the lead is functionally normal under full testing. The insulation of both CPI and Medtronic leads is manufactured from silicone which, under normal circumstances, is permeable to serous fluid. However, larger elements such as cells or solid material are not expected to permeate silicone, and the presence of such elements within the lead insulation suggests a breach of the insulation or of vulnerable parts of the lead system such as joints or connector ends. Therefore, the presence of blood within the lead insulation suggests an insulation breach. Moreover, although it is unclear whether a small breach can affect electrical performance, persistence of the causative factor or factors may lead to propagation of the defect and to subsequent lead dysfunction. This sequence of events has been described for both epicardial and endocardial [7, 13]leads. Differences in the design of the CPI and Medtronic leads may in part explain the absence of fluid in the CPI leads. Fig. 6, which presents a cross section of each lead, shows that the space surrounding the coil is larger in the Medtronic patch. In addition, the insulation of the Medtronic lead is thinner, and the inner portion of the CPI lead insulation is red, which could prevent visualization of fluid. The electrical performance of lead systems that have fluid within the lead insulation is uncertain. However, in the five instances in our study in which fluid was found within the lead insulation, two of the three affected patients had concomitant fractures of other leads, raising the concern that subclinical fractures may occur that cannot be identified on chest X-ray examination. Thus, the presence of an insulation breach in any lead system should alert the treating physician to the possibility of other system-related problems. Prophylactic abandonment of a lead with fluid within the insulation should be strongly considered.
In most cases, lead system malfunction is diagnosed because patients present with symptoms suggestive of lead problems. Our finding that epicardial pace/sense lead malfunction presents most commonly with inappropriate ICD discharges due to oversensing is in agreement with earlier studies [8, 14]. Of greater importance is our finding that 11 of 19 lead malfunction presentations were asymptomatic and detected only by routine surveillance with X-ray examinations and electrical lead testing; this observation suggests that the prevalence of clinically important epicardial lead system malfunction may be underestimated. Although no deaths were found to be directly attributable to lead malfunction, one patient with two patch fractures presented with unsuccessful defibrillation of spontaneous ventricular fibrillation and required external cardioversion. In addition, in three patients lead problems were diagnosed after failure to defibrillate induced ventricular fibrillation during lead testing. Such events further emphasize the importance of routine follow-up for early detection of lead malfunction.
3.1 Recommendations for Follow-Up
Formal recommendations for routine follow-up of patients with an ICD have not been published by cardiology organizations. Some manufacturers include in their product manuals the recommendation that chest X-ray examinations be performed every 6 months. However, they acknowledge, and our informal polling confirms, that most physicians do not follow this practice. We have previously published our approach to the follow-up of patients with an ICD. The findings of the present study argue for the use of periodic X-ray examinations and electrophysiologic assessment of lead integrity and function. Our data show that neither assessment alone detects all lead abnormalities. Newer ICDs may be able to assess defibrillation lead impedance without shocks. With such devices it should be possible to assess defibrillation lead integrity without arrhythmia induction. As lead fractures presented between 2 and >8 years after implantation, routine continued follow-up is essential. One approach would be to assess pacing and sensing thresholds, check pacing lead impedance and review a routine radiograph of the lead every 6 months and, in addition, perform arrhythmia induction to test defibrillator lead function and impedance annually. For patients with lead models demonstrating a high failure rate, such as those described here, some physicians may elect to offer prophylactic lead replacement with a nonthoracotomy system.
3.2 Limitation of the Study
Lead malfunction rates may be underestimated because some patients refuse electrophysiologic testing at follow-up and because data on the function of leads implanted in patients who died may be incomplete. In addition, the fact that early devices were not capable of providing enhanced diagnostic information may have reduced the detection rate of lead malfunction, resulting in further underestimation of the incidence of lead malfunction in our study. However, the number of leads at risk detailed in the Fig. 1aFig. 2shows that, at 4 years after implantation electrophysiologic testing of lead systems was actually more common in patients with CPI leads than in those with Medtronic leads. This difference was predominantly due to lead testing at ICD pulse generator replacement in the group with CPI leads. Our initial analysis was based on a retrospective review of X-ray films that reflects a best case scenario in the X-ray diagnosis of lead fracture. In routine clinical practice, however, the sensitivity of X-ray films may be lower, and the incidence of false positive findings higher. We tried to address these uncertainties by having an experienced electrophysiologist conduct a blinded review of all X-ray data from all patients in the study. This review found that the sensitivity of X-ray data was 71% when they alone were used to diagnose lead fracture, but it decreased to 62% when all malfunctioning leads were included in this analysis because leads with fluid within the insulation, but without a visible fracture, were reported as normal. The latter rate of 62% may more closely reflect the sensitivity of X-ray data in routine clinical practice. Whether unrecognized lead malfunction was a factor in the six sudden deaths that occurred during follow-up cannot be determined. Moreover, sudden deaths due to lead malfunction occurring in the early part of the study could have been underestimated if some deaths were misclassified as “nonsudden” and attributed to other causes. The vast majority of pace/sense leads used in this series were made by one manufacturer (Medtronic), and these results cannot necessarily be extrapolated to other leads. The implications for nonthoracotomy leads are unknown. However, it should be noted that the construction of the Medtronic subcutaneous patch is similar to that of the Medtronic epicardial patches.
Lead malfunction was common on long-term follow-up of epicardial systems and occurred in close to 30% of some leads at 4 years. In addition, fractures were most frequently observed >2 years after implantation and were identified in most cases in asymptomatic patients during routine surveillance utilizing X-ray examinations and formal lead testing. Taken together, these findings highlight the importance of regular lead testing in patients with an epicardial ICD.
Since acceptance of this report, we have discovered three additional patients with lead fractures. Two leads were Medtronic 6921 patches with fractures detected due to ventricular tachycardia nonconversion and incidently noted at pulse generator changeout occurring 4.4 and 5.6 years after implantation, respectively. The other lead was a Medtronic 6917 pace/sense electrode discovered due to lack of pacing capture occurring 3.2 years after implantation. All three lead fractues were visible on X-ray films.
- coronary artery bypass graft surgery
- implantable cardioverter-defibrillator
- Received November 11, 1996.
- Revision received October 21, 1997.
- Accepted November 19, 1997.
- The American College of Cardiology
- Zipes DP,
- Roberts D
- Wever EFD,
- Hauer RNW,
- Van Capelle FJL,
- et al.
- Dreifus LS,
- Fisch C,
- Griffin JC,
- Gillette PC,
- Mason JW,
- Parsonnet V
- Peters RW,
- Foster AH,
- Shorofsky SR,
- Nolan DA,
- Gold MR
- Winkle RA,
- Mead RH,
- Ruder MA,
- et al.
- Stambler BS,
- Wood MA,
- Damiano RJ,
- Greenway PS,
- Smutka ML,
- Ellenbogen KA
- Almeida HF,
- Buckingham TA
- Stanton MS