ICD harm and benefit: risk scores applied to the Swedish ICD-treated LQTS population

Abstract Objectives. The use of implantable cardioverter defibrillators (ICDs) in long QT syndrome (LQTS) patients is essential in high-risk patients. However, it is sometimes used in patients without high-risk profiles for whom the expected benefit may be lower than the risk of ICD harm. Here, we evaluated ICD benefit and harm by assessing risk according to risk scores and pre-ICD clinical characteristics. Design. We studied 109 Swedish LQTS patients drawn from the Swedish ICD and Pacemaker Registry with data collected from medical records. In addition to clinical characteristics, we used two risk scores to assess pre-ICD risk, and evaluated ICD benefit and harm. Results. Twenty percent of all patients received ≥1 appropriate shock with a first appropriate shock incidence rate of 4.3 per 100 person-years. A long QTc (≥550 ms) and double mutations were significantly associated with appropriate shock. Low risk scores among patients without pre-ICD aborted cardiac arrest were not significantly associated with low risk of first appropriate shock. The incidence rates of a first inappropriate shock and first complication were 3.0 and 7.6 per 100 person-years, respectively. Conclusion. Our findings on ICD harm emphasize the importance of careful individual pre-ICD consideration. When we applied two risk scores to patients without pre-ICD aborted cardiac arrest, we could not validate their ability to identify patients with low risk of appropriate shocks and patients who were assessed as having a low risk still received appropriate shocks. This further supports the complexity of risk stratification and the difficulty of using risk scores.