Exploration of arrhythmia burden in cardiac amyloidosis using implantable loop recorders (the EXACLIBUR trial)
A Ioannou, R Patel, Y Razvi, A Sheikh, J Mansell, A Martinez-Naharro, L Venneri, T Lane, H Lachmann, P Hawkins, A Wechalekar, J Gillmore, M FontanaAbstract
Background
Cardiac amyloidosis (CA) is a progressive infiltrative cardiomyopathy associated with conduction abnormalities, arrhythmias, and an increased risk of sudden cardiac death.
Purpose
To characterise the arrhythmia burden through the use implantable loop recorders (ILRs) and identify predictors of arrhythmia in CA.
Methods
In this prospective single-centre observational study, 110 treatment naïve patients with a new diagnosis of transthyretin amyloid cardiomyopathy (ATTR-CM:63) or light-chain CA (AL-CA:47) underwent comprehensive deep phenotyping inclusive of cardiac magnetic resonance with multiparametric mapping followed by ILR implantation.
Results
Bradyarrhythmias with a Class I indication for device implantation occurred in 17.3% of patients (ATTR-CM: 15/63 [23.8%], AL-CA: 4/47 [8.5%]). Abnormalities on the resting 12-lead electrocardiogram (ECG) (atrial fibrillation [AF] or PR interval >200ms: HR=6.06, 95%CI[1.40-26.02], P=0.016; QRS duration >120ms: HR=6.76, 95%CI[2.43-18.83], P<0.001) and a severe amyloid load (extracellular volume [ECV] ≥50%: HR=8.80, 95%CI[1.17-65.97], P=0.034) were predictive of bradyarrhythmias requiring cardiac device implantation. Combining these risk markers refined risk stratification, with higher rates of cardiac device implantation seen with each additional risk marker (implantation rate per 100 person-years: 0 vs. 1.3 vs. 7.6 vs. 31.6, P<0.001). The presence of all 3 markers carried a positive predictive value of 50.0% (95%CI:29.1%-70.9%) and a negative predictive value of 91.9% (95%CI:83.6%-96.7%). In patients with all 3 markers the number needed to implant to detect one clinically significant bradyarrhythmia was 2. New AF occurred in 28.2% of patients (ATTR-CM: 15/30 [50.0%], AL-CA: 5/41 [12.2%]) who did not have a background of AF (n=71) and a severe amyloid load (HR=4.83, 95%CI[1.41-16.49], P=0.012) predicted the onset of new AF. A severe amyloid load carried a positive predictive value of 40.5% (95%CI:25.6%-56.7%) and a negative predictive value of 89.7% (95%CI:72.7%-97.8%). In patients with a severe amyloid load the number needed to implant to detect new AF was 2.5. During follow-up 21(19.1%) patients died (ATTR-CM:10, AL-CA:11). The terminal cardiac rhythm was pulseless electrical activity in all patients with ATTR-CM, whereas 2(18.2%) patients with AL-CA had a terminal ventricular arrhythmia, while an additional patient with AL-CA experienced non-fatal polymorphic ventricular tachycardia, all of which had a moderate amyloid load (ECV<50%).
Conclusions
Arrhythmic burden differs between ATTR-CM and AL-CA. ECG abnormalities and myocardial ECV can predict arrhythmia and guide decisions regarding ILR implantation. Ventricular arrhythmias are not uncommon in AL-CA and occur with relatively early amyloid infiltration. Further studies are needed to asses if this subgroup would benefit from targeted arrhythmia prophylaxis.For image description, please refer to the figure legend and surrounding text.For image description, please refer to the figure legend and surrounding text.