Failure to detect life-threatening arrhythmias in ICDs using single-chamber detection criteria.

BACKGROUND: There are anecdotal reports of sudden death despite a functional implantable cardioverter defibrillator (ICD). We sought to describe scenarios leading to fatal or near-fatal outcome due to inappropriately inhibited ICD therapy in devices programmed with single-chamber detection criteria. METHODS: Programmed settings, episode lists, and intracardiac electrograms from 24 patients with a life-threatening event (n = 12) or fatal outcome (n = 12) related to failed ventricular arrhythmia detection were used to clarify the underlying scenario. RESULTS: Fifty episodes of failed ventricular arrhythmia detection were identified and categorized into six scenarios: (1) spontaneous ventricular tachycardia (VT) or ventricular fibrillation (VF) with a rate below the detection limits, (2) misclassification of polymorphic VT (PVT) or VF as supraventricular tachycardia (SVT), (3) misclassification of VT/VF as cluster of nonsustained VT episodes, (4) misclassification of monomorphic VT (MVT) as SVT, (5) inappropriate shock abortion, and (6) false termination detection. These scenarios occurred respectively 6, 9, 3, 9, 8, and 15 times. In 9/9 (100%) patients with PVT/VF classified as SVT, rate stability was active for rates ranging from 222 to 250 beats/min. MVT detected as SVT was due to the sudden onset criterion in 7/9 (78%) patients and twice a consequence of the rate stability criterion active for rates ranging from 200 to 250 beats/min. CONCLUSION: We describe six scenarios leading to failure of ventricular arrhythmia detection in a single-chamber detection setting withholding life-saving therapy. These scenarios are more likely to occur with high-rate programming and long detection times, especially if combined with rate stability and sudden onset.

Biventricular Paced QRS Area Predicts Acute Hemodynamic CRT Response Better Than QRS Duration or QRS Amplitudes.

INTRODUCTION: Vectorcardiographic (VCG) QRS area of left bundle branch block (LBBB) predicts acute hemodynamic response in cardiac resynchronization therapy (CRT) patients. We hypothesized that changes in QRS area occurring with biventricular pacing (BV) might predict acute hemodynamic CRT response (AHR). METHODS AND RESULTS: VCGs of 624 BV paced electrocardiograms (25 LBBB patients with 35 different pacing configurations) were calculated according to Frank's orthogonal lead system. Maximum QRS vector amplitudes (XAmpl , YAmpl , ZAmpl , and 3DAmp ) and QRS areas (XArea , YArea , ZArea , and 3DArea ) in the orthogonal leads (X, Y, and Z) and in 3-dimensional projection were measured. Volume of the 3D vector loop and global QRS duration (QRSD) on the surface electrocardiogram were assessed. Differences (Δ) in VCG parameters between BV paced and LBBB QRS complexes were calculated. An increase of 10% in dP/dt max was considered as AHR. LBBB conduction is characterized by a large ZArea (109 µVs, interquartile range [IQR]:75;135), significantly larger than XArea (22 µVs, IQR:10;57) and YArea (44 µVs, IQR:32;62, P < 0.001). Overall, QRS duration, amplitudes, and areas decrease significantly with BV pacing (P < 0.001). Of all VCG parameters, 3DAmpl , Δ3DAmpl , ZArea, ΔZArea , Δ3DArea , and ΔQRSD differentiate AHR response from nonresponse (P < 0.05). ΔZArea predicted best positive AHR (area under the curve = 0.813) and outperformed any other VCG parameter or QRSD measurement. CONCLUSION: Of all VCG parameters, reduction in QRS area, calculated in Frank's Z lead, predicts acute hemodynamic response best. This method might be an easy, noninvasive tool to guide CRT implantation and optimization.

Alternans of the Ventricular Electrogram in Patients with an Implanted Cardioverter-Defibrillator.

BACKGROUND: The occurrence and significance of alternans of the ventricular electrogram (VEGM) in patients with an implanted cardioverter-defibrillator (ICD) has been rarely reported. OBJECTIVES AND METHODS: This report describes our observations of VEGM alternans documented in nine patients with an ICD (seven new cases and two previously published cases for comparison). RESULTS: We found seven new cases of near-field VEGM alternans and added two of our previously reported examples. Catecholaminergic polymorphic ventricular tachycardia (CPVT) was diagnosed in one patient based on ICD recordings. Alternans occurred during ventricular tachycardia (VT) in eight patients. A fast sinus tachycardia could not be ruled out in one patient. Stable cycle length alternans was found in five patients. QRS alternans of the left ventricular (LV) electrogram (EGM) was recorded in all five patients who had a device for cardiac resynchronization therapy capable of sensing by the LV channel. These five cases exhibited corresponding alternans of the right ventricular (RV) EGM in three cases, none in one patient, and a questionable recording in another. Alternans of the far-field (FF) VEGM occurred simultaneously with RV EGM alternans in all four patients whose device provided an FF tracing. CONCLUSION: Ventricular alternans may be more common than realized in ICD patients with VT. The correlation of VEGM alternans with the surface electrocardiogram remains unknown. Although QRS alternans itself as an electrical pattern is generally benign, its cause may not be, as illustrated in our patient with CPVT. Furthermore, associated cycle length alternans or undersensing of the smaller alternans component may complicate ICD therapy.

Atrioventricular block precipitated by isoproterenol.

An isoproterenol infusion was administered during an electrophysiologic study (EPS) in a patient with a history of near syncope, left bundle branch block, and no documented atrioventricular (AV) block. Isoproterenol precipitated classic 2:1 Infra-Hisian AV block most probably proximal to the site of recording a His-Purkinje potential consistent with right bundle branch activity. Paroxysmal AV block also occurred during isoproterenol washout at a different site located distal to the presumed right bundle branch potential. Isoproterenol may be valuable diagnostically in an occasional patient suspected of AV block in whom an EPS is unrevealing and a drug challenge is negative.

A review of the confirmation algorithm and related timing cycles of St. Jude ICDs.

The confirmation algorithm of implantable cardioverter-defibrillators (ICDs) functioning in the uncommitted mode verifies the presence of a continuing tachyarrhythmia during or following charging of the capacitor. Confirmation is usually applied in relation to the first shock. The details of the normal reconfirmation process of St. Jude ICDs and its role in delayed shock delivery or other abnormalities have received relatively little attention. This review describes the confirmation process of St. Jude ICDs and provides an illustrative clinical example of delayed shock as a result of a pacing or sensing lead fracture. The fundamental timing cycles of St. Jude ICDs are also reviewed to facilitate understanding of the problems related to the confirmation process.

Automatic mode switching of a dual chamber implantable cardioverter-defibrillator induced by a ventricular escape rhythm.

This report describes a form of group beating induced by a St Jude dual chamber ICD which interpreted a ventricular escape rhythm (with retrograde conduction) as premature ventricular complexes (PVC). These pacemaker-defined PVCs activated the atrial pace-PVC algorithm in 2 steps. 1. The postventricular atrial refractory period (PVARP) was terminated upon detecting a retrograde P wave within its unblanked portion, and 2. An atrial stimulus was released 330ms after the end of the PVARP. This response resulted in automatic mode switching because the 330ms interatrial interval was shorter than the atrial tachycardia detection interval. The arrhythmia may be considered to represent an unusual form of pacemaker escape-capture bigeminy.

Undersensing by an ICD due to alternans of the ventricular electrogram.

Alternans of the ventricular electrogram (VEGM) during ventricular tachycardia (VT) is a rare cause of ventricular undersensing by an implantable cardioverter-defibrillator (ICD). This report describes a patient with a St. Jude ICD who exhibited sustained monomorphic VT associated with surface QRS alternans, alternating cycle lengths, alternans of the VEGM causing intermittent undersensing of the smaller component, and intermittent 2:1 counting of ventricular intervals during 1:1 sensing in response to the ICD detection algorithm. VEGM undersensing was corrected noninvasively simply by programming the threshold start from 62.5% to 50% which increased the sensitivity based on the amplitude of the VEGM. This maneuver did not affect the satisfactory and stable defibrillation threshold.

Dissimilar ventricular rhythms: implications for ICD therapy.

Sensing of left ventricular (LV) activity in some devices used for cardiac resynchronization therapy (CRT) was designed primarily to prevent the delivery of an LV stimulus into the LV vulnerable period. Such a sensing function of the LV channel is not universally available in contemporary CRT devices. Recordings of LV electrograms may provide special diagnostic data unavailable solely from the standard right ventricular electrogram and corresponding marker channel. We used the LV sensing function of Biotronik CRT defibrillators to find 3 cases of dissimilar ventricular rhythms or tachyarrhythmias. Such arrhythmias are potentially important because concomitant slower right ventricular activity may prevent or delay implantable cardioverter-defibrillator therapy for a life-threatening situation involving a faster and more serious LV tachyarrhythmia. Dissimilar ventricular rhythms may not be rare and may account for cases of unexplained sudden death with a normally functioning implantable cardioverter-defibrillator and no recorded terminal arrhythmia.