Novel implantable cardioverter defibrillator (ICD) discrimination algorithms and programming strategies have significantly reduced the incidence of inappropriate shocks, but there are still gains to be made with respect to reducing appropriate but unnecessary antitachycardia pacing (ATP) and shocks. We examined whether programming a number of intervals to detect (NID) of 60/80 for ventricular tachyarrhythmia (VT)/ventricular fibrillation (VF) detection was safe and the impact of this strategy on (1) adverse events related to ICD shocks and syncopal events; (2) ATPs/shocks; and (3) patient-reported outcomes. The “ENHANCED Implantable Cardioverter Defibrillator programming to reduce therapies and improve quality of life” study (ENHANCED-ICD study) was a prospective, safety-monitoring study enrolling 60 primary and secondary prevention patients at the University Medical Center Utrecht. Patients implanted with any type of ICD with SmartShock technology and aged 18 to 80 years were eligible to participate. In all patients, a prolonged NID 60/80 was programmed. The cycle length for VT/fast VT/VF was 360/330/240 ms, respectively. Programming a NID 60/80 proved safe for ICD patients. Because of the new programming strategy, unnecessary ICD therapy was prevented in 10% of ENHANCED-ICD patients during a median follow-up period of 1.3 years. With respect to patient-reported outcomes, levels of distress were highest and perceived health status lowest at the time of implantation, which both gradually improved during follow-up. In conclusion, the ENHANCED-ICD study demonstrates that programming a NID 60/80 for VT/VF detection is safe for ICD patients and does not negatively impact their quality of life.
Novel implantable cardioverter defibrillator (ICD) discrimination algorithms and programming strategies have significantly reduced the incidence of inappropriate shocks, but there are still gains to be made with respect to reducing appropriate but unnecessary antitachycardia pacing (ATP) and shocks. The “ENHANCED Implantable Cardioverter Defibrillator programming to reduce therapies and improve quality of life” study (ENHANCED-ICD study) was designed to further optimize ICD programming, by means of increasing the number of intervals to detect (NID) to 60/80, to support spontaneous termination of ventricular tachyarrhythmias (VTs). The ENHANCED-ICD study should be seen as an extension of the Primary Prevention Parameters Evaluation (PREPARE) study, the Multicenter Automatic Defibrillator Implantation Trial-Reduce Inappropriate Therapy (MADIT-RIT) study, the PainFree SmartShock technology (SST) study, and the Avoid Delivering Therapies for Non-sustained Arrhythmias in ICD Patients (ADVANCE) III trial. The primary objective of the ENHANCED-ICD study was to investigate whether the new programming strategy is safe for patients receiving an ICD for primary or secondary prevention indication. Secondary objectives were to examine the impact of the new programming strategy on (1) adverse events (AEs) related to ICD shocks and syncopal events, (2) ATPs/shocks, and (3) patient-reported outcomes.
Methods
The ENHANCED-ICD study is a prospective, single-arm safety-monitoring study, recruiting patients from April 2013 to December 2013 from the Department of Cardiology at the University Medical Center Utrecht (UMCU), The Netherlands. The study was designed together with Tilburg University (Department of Medical and Clinical Psychology) that functioned as the core laboratory for the patient-reported outcomes part of the study. Patients implanted with a Medtronic Protecta ICD/cardiac resynchronization therapy with defibrillator (CRT-D) device or any succeeding CE-approved and market-released Medtronic ICD/CRT-D device with SmartShock technology, aged 18 to 80 years, and with a primary or secondary prevention indication for an ICD implant according to current European guidelines were eligible to participate. Patients on the waiting list for heart transplantation, with a history of psychiatric illness other than affective/anxiety disorders, unable to complete the questionnaire because of cognitive impairments, or with insufficient knowledge of the Dutch language were excluded. The study was conducted in accordance with the Declaration of Helsinki, and all patients provided written informed consent. The study protocol was approved by the Medical Ethics Committee of the UMCU.
ICDs were implanted under local anesthesia in the left or right pectoral region with a subcutaneous or submuscular pocket. If it was not possible to implant the ICD lead through the cephalic vein, puncture of the subclavian vein was performed. The decision to use a single- or dual-coil ICD lead was left up to the discretion of the implanting physician. Before closing the ICD pocket, defibrillation threshold testing with 60/80 interval detection was performed by inducing ventricular fibrillation (VF), whereas termination was attempted using an internal ICD shock (25 J). Defibrillation threshold testing was only performed in the absence of contraindications (e.g., atrial fibrillation together with an international normalized ratio <2) and under sedation.
The strategy to further reduce inappropriate and unnecessary ICD therapy comprised (1) enhanced VT/supraventricular tachycardia discrimination, (2) prolonged VT/VF detection to support spontaneous termination of VTs, and (3) not treating hemodynamic stable VTs where ATP may accelerate the VT into a life-threatening VT or VF. The exact ICD programming for the ENHANCED-ICD study is listed in Table 1 . In summary, 1 monitoring zone (between 360 and 330 ms) and 2 therapy zones were programmed, treating all rhythms of cycle length <330 ms that met the duration criterion of 60/80 intervals and were discriminated as VT.
| ICD parameter | Programmed value |
|---|---|
| Ventricular fibrillation detection | ON ∗ |
| Ventricular fibrillation detection interval | 330 ms |
| Ventricular fibrillation initial number of intervals to detect | 60/80 |
| Ventricular fibrillation redetect number of intervals to detect | 30/40 |
| Fast ventricular tachycardia | Via ventricular fibrillation |
| Fast ventricular tachycardia detection interval | 240 ms |
| Ventricular tachycardia detection | Monitor |
| Ventricular tachycardia detection interval | 360 ms |
| Ventricular fibrillation therapy | 1x ATP (8 pulses, 88%) during charging followed by max 6 output shocks (35 J) |
| Fast ventricular tachycardia therapy | 3x ATP (8 pulses, 88%, scan 20 ms) followed by max 5 output shocks (35 J) |
| Supraventricular tachycardia/ventricular tachycardia discrimination | ON; limit 270 ms, high rate time out OFF |
| Wavelet | ON (70%) |
| Stability | OFF |
| Onset | Monitor (81%) |
| T-wave | ON |
| Lead noise | ON (timeout 0.75 min) |
| PR Logic (only in dual/triple-chamber ICDs) | ST, AF: ON, other 1:1 SVT: OFF |
∗ A probabilistic counter was used. Every interval that enters the window and meets the VF threshold will increment the counter. The counter does not automatically decrement if an interval falls within the VT zone or sinus rate. However, as the window slides beat by beat in a continuous fashion, the counter will decrement if a VF interval exists while a VT interval or sinus beat enters the window.
The outpatient clinic device follow-up assessments took place at 2, 6, and 12 months after implantation (standard procedure) and every 6 months afterward until the last included patient had completed the 12-month follow-up. Moreover, patients were connected to the Medtronic CareLink Remote Monitoring Network to increase safety and retrieve episode data from the online database that otherwise might not have been available because of cleared ICD memory. According to clinical practice in the UMCU, remote follow-up was performed twice a year, and interim CareAlert notifications informed the cardiologist/ICD technician in case of “VF detection” (red alert) and “Delivery of number of shocks” (yellow alert). To gather as much episode data as possible, an extra remote patient-monitoring transmission through CareLink was planned for all patients at the time the last included patient was scheduled for the 12-month follow-up (end of study). In addition, at baseline, 3, 6, and 12 months after implantation, patients were asked to complete a questionnaire package assessing several patient-reported outcomes. Information on demographic variables was obtained through purpose-designed questions at baseline (1 day before implantation), whereas information on clinical variables was extracted from patients’ medical records.
The primary end point was safety of the new ICD programming ( Enhanced programming) within 12 months after implantation. Safety was defined as a combined end point, including all arrhythmic syncopes (complete loss of consciousness with loss of postural tone) and other intervention-related safety events (i.e., hospitalizations, deaths, or other serious adverse events [SAEs] due to Enhanced programming ). All patients were asked to contact their cardiologist or ICD technician in case of a safety event. Information on safety events was also checked during outpatient clinic follow-up assessments.
On the basis of earlier research, an arrhythmic syncope rate of approximately 3.1% in the first 12 months after ICD implantation was chosen as baseline value. A doubling of this rate to 6.2% due to Enhanced programming was considered to be unacceptable. To detect this increase, if present, at least 52 patients had to be included in the study to be able to decide on whether Enhanced programming was safe based on an a priori power analysis. To closely monitor the safety of patients, the trial statistician performed an interim analysis after each intervention-related safety event using an established, sequential safety-monitoring model. The safety monitoring was carried out with a 1-sided type I error of 10% and a power of 50%. This enabled the investigators to be warned as soon as an unacceptable rate would be reached. Enhanced programming was considered safe if the number of arrhythmic syncopes and other intervention-related safety events would not exceed the monitoring boundary of 6.2%. This boundary guaranteed the type I error wherever it would be crossed. An independent data and safety monitoring board (DSMB), consisting of 1 clinical epidemiologist and 2 cardiologists, was established to accomplish ongoing safety monitoring by evaluating impending interim analyses. Furthermore, during regular meetings, they checked the progression of inclusion and discussed all SAEs. After each meeting, the DSMB reported on the safety of the study participants and advised the principal investigator on the continuation of the study. In addition, an endpoint adjudication committee, comprising 3 experienced electrophysiologists, was appointed to adjudicate all safety events that were possibly related to the Enhanced programming strategy.
Secondary end points were (1) number of AEs ( related to ICD shock or syncopal events) , (2) number of ATPs/shocks ( appropriate/inappropriate ), and (3) patient-reported outcomes ( anxiety, depression, Type D personality, ICD concerns, ICD expectations, and health status ). Information on AEs was obtained during outpatient clinic device follow-up assessments. Information on ATPs/shocks was obtained during outpatient clinic and remote device interrogations. Patient-reported outcomes were measured by standardized and validated questionnaires ( Appendix 1 ). Because of the study design and the lack of a control group, patient-reported outcomes of ENHANCED-ICD patients were compared with those of patients enrolled in 3 other ICD studies (i.e., the Web-based distress management program for patients with an implantable cardioverter defibrillator [WEBCARE], the Mood and personality as precipitants of arrhythmia in patients with an implantable cardioverter defibrillator: a prospective study [MIDAS], and the ADVANCE III trial ). The proportion of patients with a primary prevention indication in the WEBCARE, MIDAS, and ADVANCE III study was 69%, 66%, and 75%, respectively.
Categorical variables were expressed as counts and proportions; continuous variables were expressed as mean with SD when normally distributed, and as median with interquartile range (IQR) in case of a non-normal distribution. The hypothetical cumulative event rates (i.e., delivered ICD therapy) stratified by type of ICD programming (i.e., NID 18/24 vs 30/40 vs 60/80) were compared using the Kaplan–Meier method. Statistical analyses were performed using SPSS 21.0 for Windows (SPSS Inc, Chicago, Illinois).
Results
Seventy-nine patients were assessed for eligibility, of which 19 patients were excluded. Six patients did not meet the inclusion criteria, 11 patients declined to participate, and 2 patients were excluded for other reasons. Hence, a total of 60 patients consented to participate and were enrolled in the ENHANCED-ICD study. Participants and nonparticipants showed comparable values on baseline characteristics. However, a smaller proportion of participants than nonparticipants had renal insufficiency (6.7% vs 31.3%; other results not shown). Forty-three (71.7%) ENHANCED-ICD patients received a single-chamber ICD, 6 (10.0%) a dual-chamber ICD, and 11 (18.3%) a CRT-D. Baseline characteristics are provided in Table 2 . During a median follow-up of 1.3 years (IQR 1.1 to 1.5), there were 2 deaths (3.3%) due to progressive heart failure, whereas 2 other patients (3.3%) underwent left ventricular assist device implantation. Furthermore, there was 1 patient “lost to follow-up” as he did not show up for follow-up visits in the hospital. However, the patient stayed connected to CareLink, and therefore data collection on arrhythmias was still possible. In 5 of 60 patients, the Enhanced programming was changed during follow-up. In 2 patients, the detection duration was further prolonged because of left ventricular assist device implantation, and in the 3 remaining patients, the fast VT zone was changed from >182 beats/min into 170/194/200 beats/min, respectively. However, these deviations from the Enhanced programming did not influence the results.
| Characteristics | Total (n = 60) |
|---|---|
| Age (years), mean ± SD | 60.1±12.4 |
| Men | 47 (78%) |
| Having a partner | 46 (77%) |
| Lower education ∗ | 6 (10%) |
| Currently employed | 26 (43%) |
| Primary prevention indication | 32 (53%) |
| Cardiac resynchronization therapy | 11 (18%) |
| New York Heart Association class III/IV | 11 (18%) |
| Left ventricular ejection fraction (%), mean ± SD | 36.7±15.5 |
| QRS (ms), mean ± SD | 117±28 |
| Atrial fibrillation in the past | 17 (28%) |
| Left bundle branch block | 11 (18%) |
| Ischemic etiology | 30 (50%) |
| Previous myocardial infarction | 29 (48%) |
| Previous percutaneous coronary intervention | 20 (33%) |
| Previous coronary artery bypass grafting | 13 (22%) |
| Diabetes mellitus | 12 (20%) |
| Chronic obstructive pulmonary disease | 7 (12%) |
| Renal insufficiency | 4 (7%) |
| Smoker | 4 (7%) |
| Cardiac rehabilitation † | 10 (20%) |
| ACE-inhibitors | 36 (60%) |
| Beta-blockers | 50 (83%) |
| Diuretics | 37 (62%) |
| Statins | 37 (62%) |
| Psychotropic medication † | 4 (7%) |
† Missing values: cardiac rehabilitation, 10/60; psychotropic medication, 1/60.
With respect to the primary endpoint, 1 (1.7%) patient experienced an arrhythmic syncope during the first 12 months after implantation. No other intervention-related safety events (i.e., hospitalizations, deaths, or other SAEs due to Enhanced programming ) occurred during the first 12 months after implantation. Because the number of safety events did not exceed the safety threshold in the sequential safety-monitoring model, the Enhanced programming was considered safe ( Figure 1 ) and left unchanged. In case of crossing the monitoring boundary, ICD programming would have been changed into a NID 30/40 in all study patients.
With respect to the secondary endpoints, a total of 7 syncopal events were reported by 6 patients, of which 5 events (i.e., 3 presyncopes and 2 syncopes) were classified as related to arrhythmia during a median follow-up of 1.3 years (IQR 1.1 to 1.5). Moreover, a total of 3 ICD shocks were delivered in 3 patients during follow-up. Two of the patients were not aware of the delivered therapy, whereas 1 patient (receiving an inappropriate ICD shock) consciously experienced the shock. The latter patient endorsed the shock intensity as a 9 on a visual analog scale (range 0 to 10), and a 9 on a visual analog scale evaluating the degree of fear at the time of the shock (range 0 to 10). During the same follow-up period, 6 ENHANCED-ICD patients (10%) received ICD therapy, with 5 (8.3%) receiving appropriate therapy and 1 (1.7%) receiving inappropriate therapy. In total, 13 episodes with a cycle length <330 ms (≥182 beats/min) and a duration of >60/80 intervals were detected. An overview of all episodes treated by ICD therapy is provided in Table 3 .
| Pat. ID | Type of indication | Type of rhythm disturbance | Delivered ICD therapy | Median cycle length (ms) | Duration of episode (s) |
|---|---|---|---|---|---|
| Appropriate therapy | |||||
| ENH03 | Secondary | VF | 1x shock | 190 | 24 |
| ENH14 | Secondary | VT | 2x ATP | 250 | 29 |
| ENH14 | VT | 1x ATP | 260 | 18 | |
| ENH14 | VT | 1x ATP | 250 | 18 | |
| ENH14 | VT | 1x ATP | 260 | 18 | |
| ENH22 | Secondary | VT | 1x ATP | 290 | 20 |
| ENH22 | VT | 1x ATP | 300 | 21 | |
| ENH22 | VT | 1x ATP | 320 | 21 | |
| ENH22 | VT | 1x ATP | 310 | 21 | |
| ENH22 | VT | 1x ATP | 320 | 25 | |
| ENH31 | Secondary | VT | 1x ATP | 310 | 22 |
| ENH53 | Primary | VT | 1x shock | 190 | 22 |
| Inappropriate therapy | |||||
| ENH59 | Primary | AF ∗ | 3x ATP → 1x shock | 310 | 56 |
| ENH59 | AF ∗ | 2x ATP | 290 | 39 | |
∗ 1 AF episode that was labeled as 2 separate episodes by the ICD.
Because of Enhanced programming , unnecessary therapy could be prevented in 6 ENHANCED-ICD patients (10%) experiencing nonsustained episodes with a cycle length <330 ms and a duration of >30/40 (standard detection) but <60/80 intervals, which would have been treated in the long-detection group of the ADVANCE III trial. Additional information on these episodes is presented in Table 4 , with an electrogram of a self-terminating episode in Figure 2 . The hypothetical cumulative event rate (i.e., delivered ICD therapy) stratified by type of ICD programming (i.e., NID 18/24 vs 30/40 vs 60/80) for the ENHANCED-ICD patients is shown in Figure 3 . The all-cause therapy-free rate and the all-cause shock-free rate for ENHANCED-ICD patients at 1 year was 93.1% and 96.6%, respectively (2 deaths were excluded). Possibly, more unnecessary therapies have been prevented than currently presented. Because ICDs automatically delete episode data in case of excessive accumulation and not all episodes that were saved in the online CareLink database had stored electrograms, we might have missed a few episodes with a duration of >18/24 and <60/80 intervals.
