960-86 Implications of Alternative Classifications of Sudden Cardiac Death: A Prospective Analysis of 109 Deaths in Defibrillator Trials

In order to explore the implications of using varied definitions of sudden cardiac death (SCD), a classification (CL) committee (3 cardiologists) prospectively evaluated 109 deaths over a period of 19 months in patients with an implantable cardioverter defibriliator (ICD). The basis for CL was the CAST approach with additional assessments of the consequences of considering autopsy and ICD interrogation information. Concordance and/or discordance between committee members was recorded.ResultsOf the 834 patients followed for 19 months, there were 109 deaths: 17 were classified SCD, 51 non-SCD. and 40 non-cardiac. Of the deaths classified as SCD, 10/17 were unwitnessed as compared to 6/51 non-SCD and 3/40 non-cardiac deaths; p < 0.001. ICD detections occurred in 5/17 SCD <1 hour, 7/17 SCD <6 hours; therefore, 10/17 SCD had no ICD detection or information available. There was committee discordance in 5/17 SCD compared to 18/51 non-SCD and 16/40 non-cardiac. SCD rates as high as 3.6% (30/834) can be estimated if all SCD cases Cl by ≥1 member was counted as SCD. Likewise. a SCD rate as low as 0.8% (7/834) is possible if SCD is limited to witnessed SCD ≤1 hour; (a 4-fold difference). Autopsy information was available in 29/109 deaths. In 7 cases, autopsy findings resulted in changing a “SCD” CL (5 witnessed; 2 unwitnessed) to either non-SCD or non-cardiac [ruptured abdominal (N=21 or thoracic aortic (N=1) aneurysm, acute MI (N=1), cerebral infarction (N=1). pulmonary embolism (N=2)]. Thus, had autopsy information been unavailable or not considered, the SCD rate would have increased to 24/834 12.9%). ICD interrogation was unavailable in 51/109 (47%), most commonly due to being buried with the patient or programmed off prior to death.ConclusionA 4-fold spectrum of SCD rates is possible to report from the identical data-set. ICD interrogation has significant limitations for use in death CL, in contrast to its utility in clinical management. Autopsy results clarify cause-specific mortality in deaths that are temporally quite “sudden.” Total mortality is the most objective primary end point