patients receiving CRT-ICD to those receiving ICD during the course of RAFT. METHODS: 1798 pts with NYHA class II or III CHF, LVEF less than or equal to 30%, QRS greater than or equal to 120ms were randomized to CRT-ICD or ICD and were followed for 40  20 mos. All device-related adverse events occurring dur- ing implant or follow-up were adjudicated by an events com- mittee. Only those events which were related to the device and led to an intervention or prolonged hospitalization were con- sidered. Impulse generator replacements for expected end-of- life, upgrades from ICD to CRT-ICD, and inappropriate shocks were excluded from the analysis. RESULTS: Eleven pts did not undergo device implantation. Overall, 485 events occurred in 371 patients (20%) during follow-up. Pts randomized and receiving CRT-ICD (n = 888) had a higher risk of at least one adverse event than those ran- domized and receiving an ICD (n = 899). No deaths were attributable to device complications. Specific complications are detailed in the table. CONCLUSION: This is the first report identifying the long- term risks associated with CRT-ICD. CRT-ICD is life sav- ing and prevents CHF hospitalization but is associated with an increased risk of device-related complications. Improve- ments in CRT technology are needed to reduce the risks associated with this treatment so as to increase CRT-ICD use in eligible pts. Canadian Institutes of Health Research (CIHR) 502 CLINICAL PREDICTORS OF UPGRADES FROM IMPLANTABLE CARDIAC DEFIBRILLATOR TO CARDIAC RESYNCHRONIZATION THERAPY A Chelvanathan, M Golian, J Tam, S Zieroth, C Seifer Winnipeg, Manitoba BACKGROUND: Implantable cardiac defibrillators (ICD’s) with or without cardiac resynchronization therapy (CRT), improves survival in the heart failure population. Patients who have received ICD’s, may subsequently meet criteria suggesting benefit from upgrading to a CRT device. Up- grading can be technically difficult; have associated morbid- ity and cost implications. The purpose of this study was to determine if there are any baseline clinical variables in pa- tients who have ICD’s which may predict subsequent up- grading to CRT. METHODS: We identified patients at a single centre, tertiary facility who were upgraded from ICD therapy to CRT between January 2008 and December 2010. We then compared the clinical characteristic of this group (CRT-upgrade group) to patients who received ICD’s (ICD group) during the same time period. RESULTS: Sixteen patients received upgrades from ICD to CRT. The mean age was 55.7  SD 8.9 years; 93.7% male. Two hundred and seventy four patients received an ICD dur- ing the same period, mean age 60.4  13.3 years; 79.3% male. The comparison of baseline characteristics is presented in Table 1. CONCLUSION: QRS width and left ventricular systolic dysfunc- tion, may predict which ICD eligible patients may subse- quently meet criteria for CRT. 503 INCIDENCE OF DEVICE UPGRADES TO CARDIAC RESYNCHRONIZATION THERAPY IN HEART FAILURE POPULATION A Chelvanthan, M Golian, S Zieroth, C Seifer Winnipeg, Manitoba INTRODUCTION: The use of implantable cardiac defibrillators (ICDs) has be shown to improve survival among patients who have New York Heart Association (NYHA) class II to IV heart failure with left ventricular systolic dysfunction (ejection frac- tion [EF] 35%) and has become the standard of care. Over the last decade, cardiac resynchronization therapy (CRT) has emerged as an important treatment modality in patients with heart failure. Upgrading from pacemakers (PPMs) or implant- able cardiac defibrillators to cardiac resynchronization devices (CRTs) is an increasing occurrence. Our objectives were to S244 Canadian Journal of Cardiology Volume 27 2011