A hemodynamically responsive antitachycardia system. Development and basis for design in humans

Current automatic implantable cardioverter-defibrillators detect tachyarrhythmias primarily by rate-only algorithms and cannot adequately distinguish hemodynamically stable from unstable tachyarrhythmias. The responses of right atrial (mean) and right ventricular pressures (mean, systolic, diastolic, and pulse) to 64 induced and paced supraventricular and ventricular tachyarrhythmias were studied in 10 patients (left ventricular ejection fraction of 32 +/- 6%) to develop an algorithm capable of differentiating stable from unstable rhythms. Tachyarrhythmias were defined as hemodynamically unstable when mean arterial pressure decreased by 25 mm Hg or more during 15 seconds. Mean right atrial, right ventricular systolic, and right ventricular pulse pressures were found to be useful in distinguishing the hemodynamic significance of a tachyarrhythmia. A combined detection algorithm was developed that identified a hemodynamically unstable rhythm when the heart rate was 150 beats/min or more and mean right atrial pressure increased by 4 mm Hg or more and right ventricular systolic pressure decreased by 5 mm Hg or more during 15 seconds. This algorithm was then applied to the next 20 consecutive patients (left ventricular ejection fraction of 34 +/- 4%) and compared with the current rate-only algorithm (heart rate of 150 beats/min or more) in 143 tachyarrhythmias, and the sensitivity and specificity for detection of hemodynamically unstable tachyarrhythmias were determined. The rate-only detection algorithm had 100% sensitivity but only 68% specificity for detection of unstable tachyarrhythmias, whereas the combined rate-mean right atrial pressure-right ventricular systolic pressure detection algorithm had sensitivity and specificity of 100%. Therefore, the performance of an antitachycardia system may be significantly improved by detection algorithms that integrate hemodynamic and rate criteria.