Personalized risk assessment may predict ICD benefit
Click Here to Manage Email Alerts
The risk for ventricular tachyarrhythmic events weighted against the risk for nonarrhythmic mortality predicted the likely benefit of implantable cardioverter defibrillators, according to data presented at the virtual Heart Rhythm Society Annual Scientific Sessions.
Data from MADIT trials
Researchers analyzed data from 4,531 patients with an ICD from MADIT trials between 1997 and 2011. Ventricular tachyarrhythmic events were defined as ventricular tachycardia of 170 beats per minute or greater.
“We decide to put an ICD mainly looking at the [ejection fraction], if the EF is below 35% or below 30% in ischemic patients, but we know that not all patients with low EF derive consistent benefit from primary prevention of an ICD,” Arwa Younis, MD, cardiologist at University of Rochester Medical Center, said during a press conference. “This is due to the differences in the risk of ventricular tachyarrhythmic events and the nonarrhythmic mortality in ICD candidates because, in the end, many of these patients also have heart failure, and their risk of nonarrhythmic mortality varies a lot. Improved selection of patients at high risk is highly warranted.”
Best-subsets Fine-Gray regression analysis identified eight factors associated with an increased risk for ventricular tachyarrhythmic events without a significant increase in mortality risk: men (parameter estimate = 0.46; HR = 1.58; P < .001), digitalis (parameter estimate = 0.21; HR = 1.23; P = .008), younger than 75 years (parameter estimate = 0.43; HR = 1.54; P < .001), prior nonsustained ventricular tachycardia (parameter estimate = 0.57; HR = 1.78; P < .001), diuretics (parameter estimate = 0.24; HR = 1.27; P = .004), no aspirin (parameter estimate = 0.26; HR = 1.42; P < .001), left ventricular EF less than 20% (parameter estimate = 0.39; HR = 1.48; P < .001) and no beta-blocker (parameter estimate = 0.21; HR = 1.41; P = .04).
Researchers also identified five factors associated with an increased risk for death without prior ventricular tachyarrhythmic events: aged 75 years and older (parameter estimate = 0.57; HR = 1.74), amiodarone (parameter estimate = 0.66; HR = 1.92), diabetes (parameter estimate 0.45; HR = 1.58), CABG (parameter estimate = 0.37; HR = 1.42) and prior HF hospitalization (parameter estimate = 0.34; HR = 1.42; P for all < .001).
The ventricular tachyarrhythmic event risk score and the nonarrhythmic mortality score were combined to create the MADIT-ICD benefit score. Patients were then categorized by corresponding MADIT-ICD benefit groups: lowest benefit (n = 956), intermediate-low benefit (n = 1,179), intermediate-high benefit (n = 789) and highest benefit (n = 1,471).
Risk for events, mortality
Higher ICD benefit scores were directly correlated with a 4-year predicted risk for ventricular tachyarrhythmic events. These higher scores were also inversely correlated with the 4-year predicted risk for death without a prior arrhythmia.
Patients in the highest MADIT-ICD benefit group had the lowest predicted risk for death without a prior arrhythmia (7%) and the highest predicted risk for ventricular tachyarrhythmic events (29%; P < .001). This group also had the highest potential for life-years saved by an ICD.
The lowest MADIT-ICD benefit group had a similar risk for death without a prior arrhythmia (18%) and for ventricular tachyarrhythmic events (19%; P = .72).
“[This is] indicating that the life-years saved by an ICD is actually zero,” Younis said during the press conference.
Model stability was confirmed through internal validation with similar C-indices in the original cohort (C-index = 0.66; 95% CI, 0.63-0.68) and the validation cohort (C-index = 0.67; 95% CI, 0.65-0.69).
“The score may be used for shared decision-making in potential ICD candidates,” Younis said during the presentation. – by Darlene Dobkowski
Reference:
Younis A, et al. LBCT04-02. Presented at: Heart Rhythm Society Annual Scientific Sessions; May 6-9, 2020 (virtual meeting).
Disclosure: Younis reports no relevant financial disclosures.
Collapse
Younis A, et al. LBCT04-02. Presented at: Heart Rhythm Society Annual Scientific Sessions; May 6-9, 2020 (virtual meeting).
Click Here to Manage Email Alerts