Characteristics and early clinical outcomes of patients undergoing totally subcutaneous vs. transvenous single chamber implantable cardioverter defibrillator placement

Abstract

Aims

In 2012, the first totally Subcutaneous Implantable Cardioverter-Defibrillator (S-ICD) was approved by the Food and Drug Administration (FDA) in the United States. A possible benefit of this device is that it does not involve placing leads ‘in’ or ‘on’ the heart, potentially reducing complications.

Methods amd results

Ninety-one S-ICD and 182 single chamber TV-ICD implants were performed between 10/22/2012 and 9/22/2015. During this period of time, 91 patients with S-ICD were matched to TV-ICD patients using single centre NCDR ICD Registry Data based on dialysis status, gender, and age. Intra- and post-operative complications and deaths were examined within the first 180 days following implantation. Patients with S-ICDs had higher creatinine (2.3 ± 2.5 vs. 1.1 ± 0.7, P < 0.001) and were more likely to be on chronic dialysis (20.9% vs. 5.5%, P < 0.001) than TV-ICD patients. Patients in the S-ICD group were more likely to have had prior device infections (14.3% vs. 3.3%, P = 0.021) as well as prior TIA/CVA (14.3% vs. 4.4%, P = 0.049) compared to patients in the TV-ICD group. Seven patients experienced 7 complications or death in TV-ICD group and 5 patients experienced 7 complications or death in SQ-ICD group, P = 0.774.

Conclusion

In this retrospective matched single centre cohort study, there was no significant difference in implantation complications or death in patients receiving single chamber TV-ICDs compared to S-ICDs within 6 months following implantation. This occurred despite more severe preexisting illness in the S-ICD group. Further investigation is needed to determine outcomes after longer-term follow-up.

Introduction

Randomized controlled trials have demonstrated the benefit of implantable cardioverter defibrillator (ICD) therapy in reducing arrhythmic death and total mortality in patients with a history of sustained ventricular arrhythmias¹ or those at high risk for developing sustained ventricular arrhythmias.^2–6 These clinical trials were performed using transvenous lead systems, with some subjects in earlier trials receiving epicardial systems. Therefore, the ICD has gained widespread usage for both primary and secondary prevention indications, with primary prevention indications representing 74% of ICD implants in the United States.⁷

The totally subcutaneous implantable cardioverter-defibrillator (S-ICD) was approved for commercial use in the United States in 2012. As an entirely subcutaneous system, it may prevent some peri-procedural and long-term complications associated with transvenous implantable cardioverter-defibrillator (TV-ICD) systems.^8–10 The S-ICD may essentially reduce or eliminate acute procedural complications such as pneumothorax, cardiac perforation, and tamponade, while avoiding the need for fluoroscopy.¹¹ It is also hoped that the S-ICD will reduce the long-term lead risks that have plagued transvenous systems, including lead failure. These benefits would be especially important for young patients, in whom leads may fail during the many decades where ICDs would be in place.

The aim of the current study was to describe peri-procedural and short-term outcomes of S-ICD patients compared to TV-ICD patients. S-ICD patients were matched to TV-ICD patients implanted during the same time period. It was hypothesized that the S-ICD cohort might demonstrate a lower rate of adverse events following device implantation than matched patients undergoing transvenous ICD placement during 6-month follow-up.

Methods

Data source

The study protocol was reviewed and approved by the Cooper Hospital Institutional Review Board (IRB). Using single centre NCDR ICD Registry data obtained from Cooper University Hospital, all patients who had an S-ICDs implanted between 10/22/2012 and 9/22/2015 at our centre were examined. Ninety-one patients who received S-ICDs were consecutively identified and they were then matched to single chamber TV-ICD patients during this time frame based on dialysis status, gender, and age. All electrophysiology reports and hospital charts (using the electronic health record, EPIC) were reviewed to confirm proper patient identification. Post-operative in-hospital records were reviewed to identify acute complications. Outpatient records were also reviewed using EPIC. All re-hospitalizations in these patients at our centre were reviewed for a period up to 180 days. In addition, data regarding complications were also reviewed from our hospital NCDR ICD database and confirmed via individual patient chart review.

Patient population and matching

During this time period, 91 consecutive patients were identified in the S-ICD group and matched to the TV-ICD group. There were 566 patients who had TV-ICD implants during the same period of time. Of the 566 patients, only 182 patients (28%) of all TV-ICD implants had single chamber TV-ICDs that were eligible for matching to the S-ICD group (Figure 1). Thus, the matching included all 91 S-ICD implants compared to 182 single chamber TV-ICD implants during the same time frame. Although matching was attempted based on dialysis status, this was not possible due to small number of patients available within the TV-ICD group.

The matching process was completed by sorting data by dialysis status, gender, and age and numbering the pairs for analysis based on the matching of the criteria. There were 19 patients on dialysis in the S-ICD group and 5 patients on dialysis in the TV-ICD group. Five patients on dialysis in the TV-ICD group were matched to the 5 patients on dialysis in the S-ICD group. The other 14 S-ICD patients on dialysis could not be matched based on dialysis status due to the lack of additional patients on dialysis in TV-ICD group. We were able to match 90 of the 91 subjects by gender while we were able to match 82 of the 91 subjects within 5 years of age. Attempts were made to match age within 5 years; when this was not possible, the closest age was chosen.

Outcomes

Outcomes were examined within the first 180 days following implantation. Adverse events were defined as procedural complications in addition to inappropriate shocks occurring within the first 180 days following device implantation. All in-hospital complications were reviewed in detail using the electronic health record (EPIC). As defined in the NCDR ICD Registry, haematoma was considered a procedural complication if the patient required a blood transfusion or reoperation for evacuation and/or explant. System revision was defined as any patient who returned to the laboratory for a surgical procedure or revision within this time frame. Results of defibrillation testing were examined in both groups. All stored intracardiac electrograms were reviewed by two of the investigators (AR and AM). If there was disagreement, a third reviewer was planned, although concordance between investigators was achieved 100% of time.

Statistical analysis

Standard descriptive statistics reporting proportions means (with standard deviations) and medians (with their interquartile ranges) were used to describe baseline characteristics and clinical factors. Variables included in the study were patient demographics, co-morbidities, cardiac risk factors, medications on discharge, ECG findings, and laboratory studies. The baseline measures and the differences in clinical characteristics and medications between men and women were assessed using Wilcoxon signed-rank tests and paired t-tests for continuous variables and McNemar test for dichotomous variables. The Shapiro Wilk (SW) test was used to determine the distribution of the continuous data. Those variables that were determined to have a normal distribution (by having the SW yield a P > 0.05) had analysis run by using the paired t-test while those that were determined to be non-normally distributed (non-parametric) had the Wilcoxon signed rank test run. All tests of significance were two-tailed. The complications and all-cause mortality were compared in S-ICD and TV-ICD patients using the McNemar Test. Times to adverse events or death were examined using Kaplan–Meier curves with log rank tests. Statistical analyses were performed using SPSS v22 (Armonk, NY).

Results

Baseline characteristics

Baseline clinical characteristics and comorbidities of the S-ICD and TV-ICD patients are described in Table 1. Groups were similar with respect to gender, race, ICD indication, type of heart disease, and heart failure class. Patients with S-ICDs had higher creatinine (2.3 ± 2.5 vs. 1.1 ± 0.7, P < 0.001), lower haemoglobin (12.2 ± 1.8 vs. 12.9 ± 2.0, P = 0.008), and were more likely to be on chronic dialysis (20.9% vs. 5.5%, P < 0.001) than TV-ICD patients. Mean age was 54.9 ± 13.6 for S-ICD group compared to 56.3 ± 12.7 in the TV-ICD group (P = 0.017). Patients in the S-ICD group were more likely to have had prior device infections (14.3% vs. 3.3%, P = 0.021) as well as prior TIA/CVA (14.3% vs. 4.4%, P = 0.049) than patients in the TV-ICD group.

Defibrillation testing

There were 79/91 (86.8%) patients who underwent defibrillation testing (DT) or attempted DT in the S-ICD group. DT was not performed in the other 12 patients due to clinical status related to multiple comorbidities, electrolyte abnormalities, LV thrombus, or atrial fibrillation without adequate anticoagulation. One patient had spontaneous termination of VT/VF and the defibrillation energy requirement could not be determined. Of the patients in whom sustained VT/VF could be induced, 69/78 (88.5%) had successful conversion after first attempt with 65J. Four additional patients had successful defibrillation testing with higher energy requirements (75-80J). Of the remaining 5 patients, 3 required repositioning of the pulse generator and/or leads using fluoroscopy, with successful DT at 65-70J. For the remaining 2 patients, 1 patient was not retested after failed defibrillation at 65J due to difficulty with sedation, and the other patient had failed repeat DT with 80J after which the pulse generator was re-positioned but not re-tested due to comorbidities. Overall, the median number of inductions per patient in the S-ICD group was 1 (range 1–3).

There were 46/91 (50.5%) patients who received DT testing in the TV-ICD group. Performing DT in TV-ICD patients is based on implanting physician preference and has decreased over time at our centre. Successful termination of VT/VF occurred in 39/46 (84.8%) patients with 25J on the first attempt. Four patients had higher energy requirements (30J). One patient had unsuccessful initial testing at implant, but subsequent successful defibrillation with 25 J after initiation of dofetilide during the same hospitalization. Another patient required a subcutaneous array and had successful DT at 25J. The last patient had unsuccessful testing at maximum output during initial implantation, but successful DT at follow-up testing. However, repeat DT months later after starting amiodarone was unsuccessful, requiring the addition of a subcutaneous array with subsequent successful defibrillation at 31J. Overall, the median number of inductions per patient during implantation testing in the TV-ICD group was 1 (range 1–5).

Patients receiving S-ICDs were more likely to undergo DT at the time of initial implantation than those receiving TV-ICDs (P = <0.0001). However, there was no significant difference between first shock success for induced VT/VF between S-ICD (88.5%) and TV-ICD (84.8%) patients (P = 0.555).

Adverse events

Adverse events are described in Table 2. These include peri-procedural complications, as well adverse events and mortality during 6-month follow-up. Overall, 5.5% of subjects in the S-ICD group and 7.7% in the TV-ICD group had adverse events or death (P = 0.774) during follow-up.

Infection requiring explant occurred in 3 patients in the S-ICD group and 1 patient in the TV-ICD group. Haematoma requiring intervention occurred in 1 patient in the S-ICD group and did not occur in the TV-ICD group. Fifteen (16.5%) patients in the S-ICD group were taking oral anticoagulation and it was held in 13 (86.7%) patients prior to implant. In the TV-ICD group, 10 (11.0%) patients were taking oral anticoagulation and it was held in 6 (60.0%) patients, representing no significant difference between S-ICD and TV-ICD groups (86.7% vs. 60%, P = 0.175).

Inappropriate shocks: Two patients in the TV-ICD group experienced inappropriate shocks for atrial tachycardia while one patient in the S-ICD experienced inappropriate shock therapy for noise. The inappropriate shocks for atrial tachycardia occurred at 86 and 142 days following device implantation. Both patients required a modification in device programming and medication adjustments. It should be noted that no patients in the transvenous group had implanted leads on advisory, including no Fidelis or Riata leads.

The shock in the S-ICD patient occurred 17 days after device implantation due to ‘noise’ in a patient with history of coronary artery bypass grafts and previous TV-ICD device that was explanted due to prior infection after generator replacement. The superior electrode of the S-ICD was located in close proximity to the sternal wire from CABG surgery, and this contact resulted in ‘noise,’ leading to inappropriate shock therapy. To correct this problem, sensing was changed from the secondary vector to the primary vector, resolving the problem. No further inappropriate shocks occurred during follow-up.

Device-related infections: Three patients (3.3%) in the TV-ICD and 13 patients (14.3%) in the S-ICD group had prior TV device infection requiring explantation or laser lead extraction (P = 0.021). In our study, 1 patient from the TV-ICD group and 3 patients from S-ICD group required explantation of their devices due to infection. The 1 patient in the TV group had bacteraemia from Group B streptococcus that required explantation within 20 days. This patient had no prior history of device implantation or infection. Interestingly, 2 of the 3 patients in the S-ICD group who required explantation for infection had previously infected TV devices that required explantation as well. Furthermore, all 3 patients requiring S-ICD explantation only had local pocket infections without evidence for bacteraemia. The 1 patient in TV group underwent explantation within 20 days, while the median time to explant for the 3 patients in the S-ICD was 18 days.

System Revision: Two patients in the TV-ICD group required system revision while no patients in the S-ICD group required revision. One patient in the TV-ICD had persistent pain after implant which did not resolve with conservative observation and required exploration of the pocket with change in the generator position. The second patient was started on amiodarone as an outpatient after discharge and brought back to the laboratory for defibrillation testing, which was initially unsuccessful, requiring system revision with addition of a subcutaneous array for successful defibrillation.

Appropriate shock therapy

Only 1 patient in the S-ICD group received an appropriate shock for VT/VF and no patients in the TV-ICD group received appropriate shock therapy within 6-month follow-up period.

Mortality

Each group had 2 patients who died within 180 days of ICD implantation. In the TV-ICD group, one patient had prior lead extraction with subsequent device re-implantation, eventually dying with multi-organ failure, and witnessed PEA arrest in the emergency room. The other TV- patient developed cardiogenic shock and eventually died from pump failure after being made comfort care. In the S-ICD group, both patients had prior lead extraction with re-implantation of using the S-ICD device. One patient developed enterococcus sepsis one week after implantation and died from sepsis. The other patient developed a large haematoma after S-ICD implant with persistent pain that required device explantation. The patient died during the same hospitalization from worsening heart failure with multi-organ failure after being made comfort care.

Adverse events or death

The time to adverse events or death is illustrated in Figure 2, showing the Kaplan–Meier event-free survival. Although there were a small number of adverse events, most appeared to occur early within the first 30 days after S-ICD implantation, while adverse events in the TV-ICD group appeared more distributed over time. However, there was no significant difference in overall outcome between groups.

Comorbidity index to predict early mortality

The Charlson Comorbidity Index (CCI) was used to help classify comorbid conditions which might alter the risk of mortality.¹² S-ICD patients had a higher Charlson comorbidity index (5.1 ± 3.1 vs. 4.2 ± 1.9, P = 0.001) and higher relative risk of death (8.9 ± 7.1 vs. 6.0 ± 4.1, P < 0.001) than patients in the TV-ICD group.

Discussion

We report the largest case-matched experience with the S-ICD to date in the United States. In this retrospective nested case-control study, there was no significant difference in implantation adverse events in patients receiving single chamber TV-ICDs compared to S-ICDs within 6 months following implantation (P = 0.774). This occurred despite more severe preexisting illness including renal failure, previous device infection, and TIA/CVA in the S-ICD group. It is important to note that 4 out of the 5 S-ICD patients who had adverse events were not de novo implants, but had prior TV-ICDs that required explantation for infection, while only 1 of the 7 patients with TV-ICDs with adverse events had prior device implantation procedures. This suggests probable selection bias for implantation of the S-ICD in high risk patients, particularly those who may be at increased risk for recurrent infection given CCI and higher relative risk of death.

It is also important to note that the patients were matched with priority given to dialysis status, gender and age, respectively in that order. Dialysis was given priority since it is associated with high bleeding and infection rates following device implantation.¹³ During the time frame of the study, 19 patients in S-ICD group were on dialysis while only a total of 5 patients were on dialysis in the TV-ICD group. Again, this suggests probable selection bias, as implanters recognize the risk of bacteraemia with transvenous systems and, therefore, may preferentially select the S-ICD for patients on haemodialysis. All dialysis patients receiving S-ICDs or single chamber TV-ICDs during the time period of the study were included in this study for matching. Second priority was given to gender, as female gender has been demonstrated to be associated with higher rates of complications in prior studies using transvenous ICD systems.^14,15 Lastly, we tried to match for age. Given our small sample size of 182 single chamber ICD patients, options were limited once dialysis and gender were utilized; thus, the age difference was close but still statistically different between the S-ICD and TV-ICD groups (54.9 ± 13.6 vs. 56.3 ± 12.7, P <0.017).

TV-ICD experience

Along with the mortality benefit of transvenous ICD systems demonstrated in multiple previous clinical trials are a myriad of periprocedural, short- term, and long-term complications that are largely related to permanent indwelling transvenous ICD leads. Transvenous leads are considered the ‘weakest link’ of the ICD system. Lead related complications are often highly morbid and include cardiac perforation, lead fracture, lead-related endocarditis, and venous thrombosis.⁷ Prior studies demonstrate a progressive increase in lead failure over time.¹⁶ As longer follow-up is associated with an increase in transvenous lead failure, a relatively short duration of follow-up in the current study likely underestimates long-term failure of TV systems. Although the sample size is small, this is consistent with the K–M curve suggesting a higher ‘upfront’ adverse event rate with the S-ICD, while adverse events in the TV-ICD group appeared more distributed over time.

S-ICD experience

In a pooled analysis of the S-ICD Investigational Device Exemption (IDE) study and the Evaluation oF FactORs ImpacTing CLinical Outcome and Cost EffectiveneSS of the S-ICD (EFFORTLESS) registry that included 882 patients with 2 year follow-up demonstrated that the 6-month complication rate decreased by quartile of enrolment, suggesting a potential role of operator experience over time.^17,18 The EFFORTLESS registry shows a 6.4% rate of implant related complications requiring intervention with a mean follow up of 558 days.¹⁰ The device-related complication rate or death with TV-ICDs (single and dual chamber devices) at 3 months is 4.8% in men and 7.2% in women in the NCDR ICD Registry.¹⁵ The current study likely represents a group of patients with a greater severity of illness with multiple comorbidities and a high rate of prior infection and renal failure when compared to the S-ICD IDE study or overall NCDR Registry data, with similar adverse events.

Friedman et al. reported rapidly increasing early usage of the S-ICD in the U.S., describing differences in comorbidities between S-ICD and TV-ICD patients in the NCDR ICD registry.¹⁹ Twenty percent of S-ICD implants occurred in patients undergoing chronic dialysis compared with 2.9% of patients with single chamber and 2.4% of patients with dual chamber ICDs. Early adoption of the S-ICD was associated with low complication rates, despite frequent usage in highly comorbid patients.^19,20 This is similar to our study where 20.9% of patients who received an S-ICD were on dialysis. Furthermore, El–Chami et al. note that 27/79 (34%) of patients who received an S-ICD were on dialysis.²¹ Despite the large percentage of patients on dialysis and high rate of co-morbidities, there was no increased risk of implant related complications.

Our study adds to the literature as it compares outcomes of S-ICD patients to TV-ICD patients implanted during the same time period. Prior studies examining the S-ICD, including the IDE study in the U.S. were single arm studies.⁸ One of the prior reports directly comparing the S-ICD with single chamber TV-ICD is a 1:1 age and sex matched case control study including 69 S-ICD patients from three German centres.²² Similar to our study, the small sample size of 69 patients limited the power for detecting significant complications between the two groups. In the study by Köbe et al., 6 (8.7%) patients in the S-ICD group were on dialysis compared to our study which had 19 (20.9%) patients on dialysis. In both studies, most patients (77% in the German study and 86% in the current study) were de novo S-ICD implants. The study by Köbe et al. also showed a higher rate of adverse events (approximately 13% in the S-ICD and 10% in the TV-ICD group), including inappropriate shocks, during a mean follow-up of approximately 9 months (range 213–759 days) compared to a total adverse event rate of 5.5% in the S-ICD group and 7.7% in the TV-ICD group in our study during 6-month follow-up. These differences in outcome may be at least partially due to the longer follow-up duration in the German study. Although programming data is not available, programming on a conditional zone for S-ICD patients is typically performed at our centre, and this could reduce inappropriate therapy.²³

Brouwer et al. compared long term outcomes between S-ICD and TV-ICD therapies at two high volume centres in the Netherlands in a propensity matched cohort, with only one of the two hospitals implanting S-ICD devices.²⁴ A total of 148 S-ICD implants were placed at one hospital and matching was performed with patients at the second hospital, resulting in a comparison between 140 S-ICD and 140 TV-ICD patients. There were major differences between baseline characteristics of patients in our study compared to the Netherland’s study. The median age of the matched S-ICD cohort was 41 years in the Netherland’s study compared to a mean age of 55 years in our study. The S-ICD patients in their study also had fewer comorbidities and less frequent structural heart disease, including a lower frequency of diabetes (6% vs. 41%), hypertension (21% vs. 81%), nonischaemic cardiomyopathy (20% vs. 63%), and coronary artery bypass surgery (2% vs. 19%). The mean LVEF was 50% with only 5% of patients having NYHA class III-IV heart failure in the Netherland’s study compared to a mean LVEF of 27% and 30% of patients having NYHA class III-IV heart failure in our study. While only 1% of their patients had poor renal function (GFR <30 mL/min), 21% of our patients were on dialysis. In their propensity matched group, 88.6% of the TV-ICD patients had dual chamber devices and only 11.4% had single chamber devices, while our study examined only single chamber TV-ICD patients. Their complications rates were higher than our study, specifically 13.7% in the S-ICD group and 18.0% in the TV-ICD group compared with 7.7% and 5.5%, respectively, in our study. The difference in complication rates may be due the longer follow-up duration in their study (i.e. median follow-up 3 years in their S-ICD matched cohort vs. 6-month follow-up in our study). Based on marked differences in baseline characteristics, our study adds additional information to the S-ICD literature as it examines a different population of patients undergoing contemporary device implantation in the U.S., including ‘sicker’ patients who have multiple comorbidities and high frequency of structural heart disease.

We used the Charlson Comorbidity Index (CCI) to help classify comorbid conditions which might alter the risk of mortality.¹² Bhavnani et al. used the CCI to predict the risk of early mortality and appropriate shock therapy in patients undergoing ICD implantation for primary or secondary prevention.²⁵ In their study, patients experiencing early mortality demonstrated higher CCI scores (mean 2.8 ± 1.3 vs. 1.5 ± 1.2, P < 0.001) when compared to individuals without early mortality. The CCI was an independent predictor of early mortality, and the incidence of early mortality increased from 5% to 78% among patients with a CCI of 0, 1, 2, 3, 4, and ≥5. In our study, S-ICD patients had a higher Charlson comorbidity index and higher relative risk of death than patients in the TV-ICD group, confirming that the S-ICD group were ‘sicker’ with a higher mortality risk than patients receiving TV-ICDs. When compared to the study by Bhavnani et al., the CCI of our patients was notably much higher, suggesting a ‘sicker’ cohort.

Notably, the Prospective, RAndomizEd comparison of subcuTaneOus and tRansvenous ImplANtable cardioverter-defibrillator therapy (PRAETORIAN) study is underway and will randomize a total of 850 patients in a 1:1 manner to either a S-ICD or transvenous ICD.²⁶ This randomized clinical trial will help to reduce any impact of selection bias. The study is powered to investigate non-inferiority of the subcutaneous ICD with respect to the composite primary endpoint of inappropriate shocks and ICD-related complications.

Clinical implications

Our study has a number of key clinical implications. (1) The S-ICD is being implanted in high risk patients. (2) There was no significant difference in implantation complications or death in patients receiving single chamber TV-ICDs compared to S-ICDs within 6 months following implantation. This occurred despite more severe preexisting illness in the S-ICD group. Although we initially hypothesized that the S-ICD might demonstrate a lower rate of adverse events following device implantation than matched patients undergoing TV-ICD placement, we did not anticipate the extent of differences in severity of illness between S-ICD and TV-ICD patients and degree of selection bias. In addition, the study follow-up duration was limited, and it is certainly possible that a longer term follow-up might have yielded a difference between groups.

Future prospective trial results should help to determine if outcomes of high risk patients receiving S-ICDs might be better than those receiving TV devices after longer term follow-up, as the risk for lead failure increases over time in patients with transvenous systems. It should be noted that there were no electrode failures throughout the follow-up period or any S-ICD related bacteraemia in the pooled analysis of the IDE study and EFFORTLESS registry.¹⁷ Due to the high rate of failure of TV leads over time and implantation of devices in younger patients who may require ICD leads in place for many decades, longer term follow-up is necessary to better compare outcomes of the S-ICD and TV-ICD systems.²⁷

Limitations

Our study has a number of key limitations. Treatment was not randomized, and patients were primarily matched for dialysis status, gender, and age. Despite the intent to match patients by dialysis status, there were an insufficient number of dialysis patients undergoing TV-ICD placement during this time period, likely due to selection bias. Our study was not powered to detect a statistically significant difference in adverse events. In addition, follow-up was limited to only 6 months, making it unlikely that any potential advantages of a non-transvenous device would be detectable in this period of time. Therefore, longer follow-up might alter results.

Conclusions

Despite frequent use in highly comorbid patients, early adoption of the S-ICD appears to be associated with comparable rates of adverse events as the TV-ICD. Further investigation is needed to determine outcomes after longer-term follow-up.

Conflict of interest: Dr. Andriulli received honoraria or consulting fees from Medtronic and Spectranetics. Dr. Russo received research grants from Boston Scientific and Medtronic, honoraria or consulting fees from Boston Scientific, Medtronic, St. Jude and Biotronik. Ms. Hunter and Ms. Field along with Drs. Mithani, Kath, and Ortman have nothing to disclose.

References