Shock Testing to Unmask RV Lead and GEnerator Malfunction in ICD Patients: the SURGE-ICD Registry

Testing the ICD system during initial implantation was the standard of care for a long time. Since the publication of the SIMPLE trial, most centers have abandoned this practice for all uncomplicated, left-sided implantations of ICD. Remarkably, a high percentage of arrhythmia centers, including academic hospitals and high volume centers, routinely omit ICD testing at the time of ICD generator replacement. There is no data available to support the safety of this practice, and findings from the SIMPLE trial may not be extrapolated to the ICD generator replacement (ICD-GR) setting.

Data from case reports and own data from an ongoing multicenter randomized trial on internal vs. external cardioversion of ICD patients suggest that a relevant number of ICD patients carry a defective defibrillation lead with normal electrical parameters. Such "silent" lead failure may only be uncovered through a high voltage (HV) ICD shock during ICD testing or internal cardioversion. An in vitro trial suggests the common low-voltage measurement of shock impedance is not able to detect smaller insulation breaches leading to short circuits. Identification of these leads and patients is crucial to ensure the continuous adequate function of ICD therapy throughout the patients' lifespan. Omitting ICD testing during all subsequent generator exchanges many years after initial implantation, may thus put patients' safety at risk.

Investigators aim to assess the magnitude of silent ICD lead failure prevalence and to identify associated risk factors.

The objectives of this study are:

1. Main objective: To determine the prevalence of silent lead failure in contemporary ICD patients with chronically implanted ICD leads, presenting for elective ICD generator replacement (ICD-GR) or electrical cardioversion of atrial arrhythmia in Germany.
2. To identify risk factors predisposing to silent lead failure.
3. To identify possible differences in the prevalence of silent lead failure between ICD lead models and manufacturers. Primary Hypothesis H0: ICD leads, with unremarkable parameters during ICD interrogation, reliably perform high voltage ICD shocks at the time of ICD generator replacement or internal cardioversion. HA: ICD leads, with unremarkable parameters during ICD interrogation, are rarely not able to perform high voltage ICD shocks at the time of ICD generator replacement or internal cardioversion, due to silent lead failure. Lacking previous data, this trial is designed as a prospective registry to describe the prevalence of silent lead failure in a large cohort of patients. The basis for an estimate of the prevalence of silent lead failure are mostly case reports or small cases series. In a previous trial in 3 out of 112 (2,7%) patients receiving internal cardioversion silent lead failure was detected {Lüker 2019}. No larger trial or registry has systematically assessed ICD leads regarding possible silent lead failure in a larger population. The University Hospital Cologne will act as the lead center, six additional German high volume centers will be recruited. The investigators aim to include ICD patients presenting for generator exchange or cardioversion for atrial arrhythmia in a multicenter registry. We address an unselected cohort of ICD patients (single chamber ICD, dual chamber ICD, CRT-D). All generator and lead manufacturers will be included. To avoid a manufacturer bias, an upper limit of patients per manufacturer is imposed at 100 leads/patients per manufacturer. Patients who received adequate and successful ICD shock therapy within the past 12 months will be excluded, as they have a very low likelihood of defective RV lead prevalence. Also patients with parameters indicating lead failure during device control and patients with contraindications for high voltage shocks (e.g. thrombus) will be excluded. Eligible participants are all patients with an implanted cardioverter defibrillator (primary and secondary prevention) with an indication for generator replacement or electrical cardioversion for atrial arrhythmia. All patients presenting at the participating center with an indication for generator replacement/electrical cardioversion or identified in the outpatient clinic will be screened. A total number of 300 patients is to be included in this registry. Depending on the recruitment rate, the study phase is expected to be completed within 24-36 months. The registry phase will end after the cohort of 300 patients is complete (expected after 3 years). Screening of potential subjects will include:

• History, including anticoagulant medication to establish the necessity of a TOE (transoesophageal echocardiography)
• Physical examination
• Device interrogation (including thresholds, impedance, sensing of all leads, battery voltage, charge time, shock impedance, revision of all programmed parameters)
• 12-lead-resting ECG
• Trans-thoracic echocardiography Patients meeting the inclusion criteria will be included after written consent is obtained. All patients require administration of a maximum energy shock within 2 days of the generator exchange or alternatively during the ICD generator replacement surgery. Patients in the cardioversion arm require internal cardioversion following the manufacturers recommendations. ICD test during or before generator replacement In centers that routinely perform ICD shock testing during ICD-GR surgery, this will be carried out according to routine practice and in the presence of adequately trained personnel. In centers that do not routinely perform ICD shock testing at ICD-GR, a single synchronized maximum energy shock (RV -> SVC+can shock orientation in pts with SVC leads) must be performed prior to ICD-GR surgery or during ICD-GR surgery, using the old ICD generator, to save battery life of the new replacement system. Sedation, TOE if applicable and monitoring during and after ICD testing will not differ from standard clinical practice cardioversion. A TOE must be performed in case of atrial arrhythmia and insufficient anticoagulation to exclude left atrial thrombi. Intraprocedural monitoring will include ECG, SpO2 and non-invasive blood pressure (NIBP), as well as clinical monitoring of respiration and sedation depth. After shock delivery, all patients will be monitored for 3 hours according to current recommendations. Internal cardioversion Internal cardioversion will be performed delivering a commanded biphasic shock (28 to 40 J, depending on manufacturer). TOE to exclude left atrial thrombi will be done in case of insufficient anticoagulation. If the shock is not able to terminate atrial arrhythmia, external cardioversion will be performed. In both groups, ICD interrogation will be performed before the shock testing/cardioversion and within 2 minutes after shock testing. Baseline data is collected from all available medical records and charts, which includes all available ICD interrogation data, discharge letters as well as all meaningful paper-based or electronic medical documentation and patient history. Device Interrogation will be performed prior to (max. 48h prior to testing) and after (2-5min) shock delivery. The following parameters be measured for all leads where applicable: Impedance, sensing, pacing thresholds, charge time, the presence of lead noise in the pace/sense-EGM and the shock-EGM. All measurements will be performed by a trained cardiologist/ electrophysiologist, with the assistance of trained study nurse/research assistant staff. Lead noise tracings must be submitted for adjudication by the AE committee. Furthermore, abnormal parameters will be measured twice for confirmation. Shock testing will be performed according to either A) routine practice of electrical cardioversion or B) during or before the ICD generator replacement surgery. Sedation or anesthesia may only be performed by an anesthesiologist or an experienced (>50 electrocardioversions) and sufficiently trained cardiologist. During sedation, a nurse with experience in electrocardioversions and sedation (>50 sedation procedures, e.g. colonoscopy, electrocardioversion, sedation for EP study, gastroscopy and similar) must be present for assistance. Monitoring of vital signs (O2 saturation, NIBP, ECG) is mandatory. Adverse events (AE) and unanticipated events must be reported to the sponsor according to GCP standards. All adverse events will be recorded in the hospital notes and electronic case report form (eCRF). A sponsor nominated adjudication committee of 2 physicians who are not directly involved with the trial will assess and adjudicate all AE to ensure timely detection of AE that occurred as a consequence of the study protocol. Severe adverse events will be reported to the research ethics committee (REC), according to GCP standards. The sponsor and/or the investigator may take appropriate urgent safety measures in order to protect the subjects of the study against any immediate hazard to their health or safety. If safety measures are taken, the main REC approval is not required before the measure is taken. In addition, device and lead malfunctions (including lead failures after shock) or complaints will be reported to the respective manufacturer per usual post-market channels. Defective leads will be sent to the respective manufacturer, should a lead extraction be deemed necessary. The indication for lead extraction will be at the discretion of the treating physician and should be made independently of this trial, and in accordance with current guidelines and local clinical practice. The decision should weigh the risks and benefit of lead removal of a chronically implanted lead. The study protocol does not mandate or make any recommendations concerning lead removal. All data is reported via eCRF to an online database (RedCap Vanderbild University). Only the sponsor (University Hospital Cologne) has access to the database. All eCRF entry will undergo checks for plausibility and completeness by the sponsor nominated data monitor within 24h of entry. In case of implausible parameter values, a query is set, and the monitor will request to see the source data. Adverse events are reported through the online database and reviewed by the adjudication committee as soon as they become available. Stopping criteria The occurrence of >15 SAE, likely or proven to be caused by a study-specific procedure (e.g. induction of VF during shock testing, respiratory compromise during sedation, stroke, aspiration pneumonia, death) will prompt early stopping of the trial. All SAE associated with lead failure (e.g., unsuccessful shock, lead revision) are exempt because the unmasking of silent lead failure is deemed beneficial to the patient. The chief investigator will prepare and submit an annual progress report (APR) to the REC. The APR will include a list of SAE and an overview of the study recruitment progress. Sample size and statistical power This study is of explorative nature. Expecting an event rate of about 3% in our study population, a sample size of n=300 is sufficient to correctly detect a difference between observed event rate and a reference value of 7% with a power of 87.9%. (two-sided type I error 5%, one-sample binomial test, performed with G*Power 3.1.7). The estimates for the expected event rate are based on a previous trial with an incidence of 2,7% (3/112). Statistical methods The investigators will describe variables of interest using mean values +/- standard deviation, median (interquartile range), or frequencies and percentages. The proportion of patients with a prevalent silent lead will be given together with its 95% confidence interval. We will compare the observed proportion with a reference value of 7% using an exact binomial test to answer the question if the observed proportion is significantly different from 7%. Differences between patients with and without a prevalent silent lead failure will be evaluated using Fisher's exact tests for categorical variables and unpaired t-tests or Mann-Whitney U tests for continuous variables, depending on the assumption of normality holds or not. ETHICAL CONSIDERATIONS This study will be conducted in accordance with EN540, the declaration of Helsinki and ICH-GCP. Only certified and approved medical devices are utilized. An ethics committee vote will be obtained prior to study initiation. Additional local ethics application is required at each center. Patient consent Informed written patient consent is required for each patient. Informed written consent will be acquired in all patients prior to inclusion through a thorough and detailed consultation with one of the investigators. Including information about data protection, data de-identification (anonymization) and further data analysis and publication. All patients will be advised about their right to withdraw consent at any point during or after the study. Patient consent will be obtained in writing. Confidentiality Patient doctor confidentiality will be obeyed. All acquired data will be treated confidentially. Re-identifiable data will be stored electronically in eCRF on the RedCap Database. The secure RedCap database servers are provided by the Clinical trial Center Cologne ("Zentrum für klinische Studien", University of Cologne) and are located on the University Hospital premises. Only representatives of the sponsor (PI, Co-PI, data manager/study nurse) will have access to the database. For analysis and publication, all data will be de-identified (anonymized). Personal data will only be disclosed after consent is obtained. OUTCOMES AND SIGNIFICANCE In those centers which routinely perform ICD shock testing during ICD generator replacement, patients will be treated according to routine practice. Participation in the study will not be associated with a benefit or risk. In centers which do not routinely perform ICD shock testing during ICD-GR, study participants would undergo study specific ICD shock testing either before or during the generator replacement surgery. ICD-Patients attending for cardioversion for atrial arrhythmia will receive a commanded shock, if no shock was delivered during the past 12 months and if device battery status does not impede internal cardioversion. ICD shock testing at ICD-GR is recommended by current guidelines and is associated with a very low risk of complications. Based on the available data on complications in the SIMPLE trial, the risk of unplanned intubation (0.6% risk) or aspiration (0.1% risk) is very low. The risk of death or stroke is not elevated when compared with no ICD shock testing. {Healey 2015} Shock testing in these patients will uncover silent lead failure in approx. 1-3% of patients, which would prevent these patients from ineffective ICD therapy and essentially the risk of sudden cardiac death due to ineffective detection and shocks of ventricular arrhythmia in the future. The overall results would have major implications for future guideline recommendations on shock testing at the time of ICD replacement. The trial results will provide valuable evidence on the necessity of routine testing, which is currently recommended but not followed by 95% of centers.