104/The effect of chronic kidney disease and other determinants of prognosis in patients undergoing device therapy for heart failure

Introduction: Outcome data on patients undergoing implantable cardioverter defibrillator (ICD) and cardiac resynchronisation therapy (CRT) implantation with higher stages of chronic kidney disease (CKD) is lacking as patients with significant renal impairment are under-represented in the major clinical trials.¹ Registry data suggests that there is a significant cohort of CKD patients with indications for these devices who are not offered them.² In this single centre observational study of a device cohort comprising 280 patients, we present real world data on mortality at yearly time points between 1 and 4 years after device implantation. The cohort was stratified by patient characteristics – in particular, the stage of chronic kidney disease but also by aetiology of cardiac disease, baseline NYHA class, severity of LV dysfunction, QRS duration and underlying rhythm. The secondary outcome measure was the NYHA class at 2 years for these various patient groups.
Methods: Retrospective analysis of 280 patients undergoing device implantation for all indications between 2008–14 giving 4 years of follow up for all patients in the dataset. Where feasible, the relative risk and odds ratio of certain patient characteristics for primary and secondary outcomes was calculated.
Results: The total mortality of all patients was 9.6% at 1 year, 18.2% at 2 years, 23.2% at 3 years and 28.6% at 4 years. The majority of patients (58.6%) had stable or improved symptoms at 2 yearly follow up. 8.6% had deteriorating symptoms at 2 yearly follow up with the remainder lost to follow up.
The stage of chronic kidney disease, regardless of other variables was strongly associated with adverse outcomes. The total mortality at 4 years for patients with an eGFR ≥45 was 17.1% (48 out of 208 patients) while the total mortality at 4 years for patients with an eGFR ≤45 was 44.4% (32 out of 72 patients). The relative risk of mortality at 4 years for patients with an eGFR ≤45 compared with an eGFR ≥45 was 2.66. Increasing CKD stage correlated with increased mortality at each yearly time point from 1–4 years. For patients with CKD stage 4 (eGFR ≤30), total mortality at 4 years was 66.67%. For patients with CKD stage 5 (eGFR ≤15 or dialysis dependent), there was 100% mortality by the 3-year time point.
Patients undergoing device implantation as primary prevention of arrhythmia for ischaemic cardiomyopathy made up 50.4% of the dataset. Total mortality was 29.79% at 4 years. The second commonest indication for device implantation was primary prevention of arrhythmia for dilated cardiomyopathy making up 15.36%. Mortality at 4 years in this group was 16.3% suggesting better outcomes compared with ischaemic cardiomyopathy. NYHA class IV patients had a 4-year mortality of 50% and patients with a left ventricular ejection fraction <35% had a 4-year mortality of 30.2%.
In terms of the secondary outcome measure of NYHA status at 2-years follow up, AF was associated with worse symptoms after device implantation compared with sinus rhythm (odds ratio = 3.97, 95% confidence interval 1.59–9.93, p=0.003), as was a rise in the creatinine level at 2 years compared with patients whose renal function remained stable (odds ratio = 5.32, 95% confidence interval 1.96–14.45, p=0.001).
Conclusions: Patients with advanced stages of CKD undergoing device implantation have significantly higher mortality rates than those with preserved renal function. Worsening creatinine trends over time and AF are associated with less symptomatic benefit from device implantation.