1285 Intradevice Interaction in a Dual Chamber Implantable Cardioverter Defibrillator Preventing Ventricular Tacbyarrhythmia Detection KALYANAM SHIVKUMAR. M.D., PH.D., ZENAIDA FELICIANO, M.D., NOEL G. BOYLE, M.D., PH.D., and ISAAC WIENER, M.D. From ihe Division of Cardiology. UCLA Medical Center. UCLA School of Medicine. Los Angeles. California Intradevice Interaction in Dual Chamber ICD. Failure to detect ventricular tachycardia and/or ventriiular fihrillation bv impluntahle cardioverter defibrillators (ICDs) is a rare but serious problem. We report a case of failure to detect an episode of indueed ventricular tacbycardia by a dual cbamber ICD, due to abbreviation of ventrieular detection window secondary to programmed pacing parameters and a rate-smoothing algorithm. In tbis patient, tbe intradevice interaction was corrected by programming rate-sm(K>thing ofl. Tbis report bigbligbts the potentially ietbal conse- quences of critical timing relationships among the pacing function, arrhythmia detection, and tbe characteristics of the arrhythmia when using a modern dual chamber ICD. Physicians responsible for patients with ICDs must be aware of such interactions. (J Cardiovasc Electrophysiol. Vol. II. pp. 1285-1288, November 2000) implantable cardioverter defibrillator, arrhythmia detection Introduction Since the introduction of implantable cardioverter de- fibriiialors (ICDs) into clinical practice, iheir usage has increased steadily, in the past, when pacing was needed in addition to an ICD. a separate pacemaker wa.s im- planted. Arrhythmia detection was carefully tested to avoid device-device interaction. Such interactions could result in nondetection of serious ventricular arrhythmias when the ICD misinterpreted pacing spikes as ventricular activity and missed lower amplitude ventricular signals. With the advent of dual chamber ICDs with sophisticated pacing, such concerns have lessened. However, modern dual chamber ICDs are subject to other interactions that can result in failure to detect serious ventricular tachy- arrhythmias. We report a case of failure to detect an episode of induced ventricular tachycardia (VT) due to interaction of the arrhythmia with pacing parameters, including rate smoothing. Case Report The patient was a 60-year-old woman with a history of myocurdial infarction, systolic ventricular dysfunction Dr. Shivkumar's eurreni address is Division of Ca'diology, University of Iowa Hospitals and Clinics. Address for corTCsponiJcnce: Isaac Wicticr. M.D,. IOO UCLA Medical Plaza. Suite 770. Los Angeles. CA 90024. Fax: 319-.WI-6686: E-mail; igiiizl@3ol.com Manuscript received 27 June 2000; Accepted for publication 23 Augusi 2000. (ejection fraction 30% to 35%), and nonsu.stained VT (rale t60 beats/min). After ischemia was excluded, the pittivnt underui>n( an electrophysiologic study, which revt-aled in- ducible sustained monomorphic VT (rate 1511 hculs/min). A dual chiimhiT ICD (Vcntak Prizm. (iuldant Corp.. St. Paul. MN. USA) was implanted using a Guidant Triad lead in the ventricle {bipolar integrated ventricular sensing) and an Oscor bipolar atriai lead (interelectrode distance 11 mm). Lead characteristics were as follows: ventricular. R wave = 9 mV. impedance = 8IK) 11. threshold 1.5 V. 0.8 mA: atrial, P wave - 2 m\ . impedance = 45» i\. threshold = 1.5 V. 3.5 mA. .\i ICD placement, defibrillation threshold was <2(» J. Because of Hrst-dcgree AV conduction delay, the PR inter- val of the ICD in the I>DD mode was programmed to 30(t msec. Due to persistence of ventricular fusion, in an attempt to decrea.se the use of the pacemaker and to conserve battery life, the de%icc was programmed to DDI 50 without a change in AV delay. Rate smoothing (12% do«n with an upper rate limit of 120 beats/mint also was activiited. The postatrial pace ventricular blanking interval was 65 msec, and the ventricular pace refractory period was 250 m.sec. With triple extrastimuU, VT (cycle length 3(MI msec) was induced. This episode of VT was not sensed by the device (Fig. I) and required an external rescue shock. When the device detected clitsely eoupled ventricular activity, rate smoothing was initiated. The resultant interaction among the tachycardia rate, the programmed AV delay, and rate smoothing interfered with arrhythmia detection due to blanking periods imposed hy postatrial pacing ventricular blanking and post ventricular pacing refrsictorj peri4>ds (Fig. 2). During this episode, which lasted IK.400 msec, there were 24 atrial spikes that intrtnluced l.56tl msec of blanking to the ventricular detection window. In addition, there were 15 ventricular pacing spikes, which introduced refractory periods (a total of 3.750 msec) in the ventricular