Clinical course and long-term follow-up of patients receiving implantable cardioverter-defibrillators Harikrishna Tandri, MD, Lawrence S. Griffith, MD, Tania Tang, MD, Khurram Nasir, MD, Omeed Zardkoohi, MD, Chandrasekhar Vasam Reddy, MD, Melissa Capps, RN, Hugh Calkins, MD, J. Kevin Donahue, MD From the Division of Cardiology, Johns Hopkins University School of Medicine, Baltimore, Maryland. BACKGROUND Implantable cardioverter-defibrillators (ICDs) are increasingly used for primary and secondary prevention of sudden cardiac death. Defibrillators were introduced into clinical practice in 1980. Since that time, factors affecting long-term survival and the natural history of defibrillator patients have not been de- scribed. OBJECTIVES The purpose of this study was to identify clinical predictors of long-term survival in patients receiving ICDs. METHODS The prognostic value of several clinical variables on the likelihood of survival or appropriate ICD therapy in 1,382 consec- utive patients receiving ICDs from 1980 to 2003 were evaluated. Data were collected at the time of device implantation, and fol- low-up was completed through March 2005. RESULTS In 70  51 months of follow-up (range 0 –282 months), 792 patients died and 421 patients received appropriate ICD ther- apy at least once. Age, left ventricular ejection fraction, New York Heart Association (NYHA) functional class, Charlson comorbidity index, and antiarrhythmic drug use correlated with mortality. -Blocker and angiotensin-converting enzyme inhibitor use was associated with improved survival. Only NYHA functional class correlated with ICD therapy. Patients free of shocks for the first 5 years after ICD implantation had continued risk of arrhythmia recurrence. CONCLUSION The heart failure characteristics of patients pre- dicted ICD shock probability and survival better than the ar- rhythmia characteristics or the underlying heart disease. Anti- arrhythmic drug use was associated with increased mortality. Beta-blocker and angiotensin-converting enzyme inhibitor use was associated with improved survival. A measurable arrhyth- mic risk even after prolonged shock-free intervals indicates the need for continued ICD therapy in all patients with appropriate ICD indications. KEYWORDS Implantable cardioverter-defibrillator; Long-term fol- low-up; Ventricular arrhythmia; Sudden death; Outcomes; Natural history (Heart Rhythm 2006;3:762–768) © 2006 Heart Rhythm Society. All rights reserved. Ventricular arrhythmias account for a majority of cardiac deaths in the developed world. 1 The introduction of the implantable cardioverter-defibrillator (ICD) by Michel Mi- rowski in 1980 was a major milestone in sudden death management. 2 Since that time, numerous clinical trials have shown improved survival after defibrillator implantation in patients experiencing malignant ventricular arrhythmias. 3-5 As such, ICDs have become the standard of care for sec- ondary prevention of sudden death. More recently, the re- sults of several primary prevention trials, including the Multicenter Unsustained Tachycardia Trial (MUSTT) and the two Multicenter Automatic Defibrillator Implantation Trials (MADIT I and MADIT II), have significantly ex- panded the indications for ICD implantation. 6-8 Studies evaluating the use of ICDs in the idiopathic cardiomyopathy population and the pairing of ICDs with biventricular pac- ing have further broadened the indications for implanta- tion. 9-11 With increases in the duration of patient survival and the number of implantation indications comes a need for long- term follow-up data to evaluate problems unique to the ICD population. Previous multivariate studies have been limited in size or duration of follow-up. 12-17 Here we report a long-term follow-up study of 1,382 patients who received ICDs. We evaluated the prognostic value of several clinical variables on survival or appropriate ICD therapy in this patient cohort. Methods Patient population and data collection We constructed a database at the time of the first ICD implantation. At the conclusion of the study period, 1,382 consecutive patients who had undergone ICD placement by physicians at the Johns Hopkins University (JHU) had been enrolled into the database, from the first human ICD implant in February 1980 2 through the end of March 2003. Baseline data for all the patients were collected on standardized implant forms and prospectively entered into the database at the time of device implantation. Variables obtained in- cluded patient demographics, implant indication, primary Dr. Donahue has a consultant agreement with Medtronic, Inc., and a research agreement with St. Jude. Dr. Calkins receives research support from Guidant, Medtronic, Inc., and St. Jude. Address reprint requests and correspondence: Dr. Kevin Donahue, Case Western Reserve University Med- icine, 2500 MetroHealth Drive, Hamann 322, Cleveland, OH 44109-1998. E-mail address: kdonahue@metrohealth.org. (Received June 23, 2005; accepted March 18, 2006.) 1547-5271/$ -see front matter © 2006 Heart Rhythm Society. All rights reserved. doi:10.1016/j.hrthm.2006.03.027