Incremental value of atrial sensing in the diagnosis of ICD recordings: findings from the THINGS registry. (14th October 2021)

• Record Type: Journal Article
• Authors: Carinci, V; Ziacchi, M; Iori, M; De Maria, E; Bolognesi, G; Zardini, M; Calvi, V; Allocca, G; Ammendola, E; Boggian, G; Saporito, D; Giorgi, D; Grassini, D; Giacopelli, D; Biffi, M
• Is Part Of: European heart journal. Volume 42(2021)Supplement 1
• Publication Date: 2021-10-14
• Subjects: Device Treatment; Cardiology -- Periodicals; Heart -- Diseases -- Periodicals; 616.12005
• DOI: 10.1093/eurheartj/ehab724.0672
• Languages: English
• ISSNs: 0195-668X

Abstract

Abstract: Background: Automatic diagnosis of implantable cardioverter defibrillator (ICD) requires confirmation through analysis of electrograms (EGM) recordings by an electrophysiologist (EP) specialist. Purpose: To investigate whether the atrial sensing capability in single-chamber ICD improves reliability in episodes assessment. Methods: The THINGS registry enrolled 526 consecutive patients without atrial pacing indication who underwent single-lead ICD implantation. Of them, 343 patients received a conventional single-chamber ICD (ICD VR) and 183 a single-lead ICD with atrial sensing capability (ICD DX, BIOTRONIK). Device-detected ventricular episodes were independently classified by two experienced electrophysiologists as ventricular (VT) or supraventricular tachycardia (SVT) by analyzing the EGMs. In case of disagreement, a third experienced EP specialist, blinded to previous assessment, was consulted. The inter-rate reliability was evaluated using the Cohen's kappa (k) coefficient in the ICD VR and ICD DX group. Results: A total 441 (383 in the ICD VR and 58 in the ICD DX group) device-detected events were analyzed. Using ventricular EGM alone for ICD VR patients, the proportion of observed agreement between the 1st and 2nd adjudicator was 86.4% consisting of 268 (70.0%) episodes confirmed as VT and 63 (16.4%) classified as SVT. The k coefficient was 0.61 and no agreement was reached for 52 (13.6%) tracings. For them, the third assessment had low agreement with both previous EP evaluations (the k coefficients 1st/3th and 2nd/3th EP specialist were 0.03 and −0.05, respectively). In ICD DX group, with available atrial EGM, the proportion of observed agreement in the first stage improved to 93.0%: 44 (75.8%) episodes were adjudicated as VT and 10 (17.2%) as SVT with a k=0.79. Only 4 (7.0%) recordings had a discordant classification. For them, the 2nd and 3rd EP specialist had perfect agreement (k=1). Conclusions: Atrial sensing capability improves EP specialist assessment of ICD episodes allowing dual EGMs analysis. Misinterpretation of ICD recordings may thus be less common in ICD DX than in conventional single-chamber ICD VR and may improve clinical decisions. Funding Acknowledgement: Type of funding sources: None.