In CKD-5D, clinicians are cautious about using aldosterone receptor

blocker for fear of hyperkalaemia. However, a systematic review of 7051 patients from six studies Nutlin-3a on spironolactone treatment in CKD-5D patients with heart failure reported that episodes of hyperkalaemia were rare; mean serum potassium was 4.9 mmol/L and no patients developed an adverse event as a result of hyperkalaemia.[26] In view of the potential benefit of aldosterone receptor blockers, it is not unreasonable to advocate their use in patients with CKD-5D, particularly with close monitoring in patients with stable serum potassium levels. RCTs of mineralocorticoid blockade in haemodialysis patients are needed, and at least one is currently in the design phase.[27] In the general population, the first-line therapy for primary and secondary prevention of SCD is insertion of

an ICD.[28] The indications for therapy are relatively narrow and target only specific high-risk groups (Table 1). The uptake of ICD in haemodialysis patients in line with current guidelines is proportionately lower than in general population patients with the same indication for device therapy. This is despite the guidelines specifying that these patients MAPK inhibitor should not be excluded. Greater than 4 weeks post myocardial infarction and either LVEF <35% AND Non-sustained ventricular tachycardia on 24 hour holter monitoring AND Ventricular tachycardia inducible on electrophysiological testing or LVEF <30% AND QRS duration ≥ 120ms Familial condition that predisposes to high risk of sudden cardiac death Mannose-binding protein-associated serine protease such as long QT syndrome This may be partly due to the increased complication rate following device insertion in CKD-5D patients, including infection, thrombosis, haematoma and lead dislodgement.[30] Furthermore, non-use is sometimes justified on the basis of cost-effectiveness as the absolute risk reduction in terms of additional life-years after ICD is lower for patients with non-dialysis CKD compared with those with

normal eGFR.[31] A recent meta-analysis of 15 observational studies reported that the presence of CKD (including CKD-5D) is still associated with a greater risk of death (HR = 2.86, 95% CI = 1.91–4.27, P < 0.05) despite ICD.[32] Another meta-analysis of seven studies including 89 dialysis patients, and 2417 non-dialysis CKD patients, found that the relative risk for mortality in dialysis patients with ICD compared with stage 3 or 4 CKD with ICD was 1.62 (95% CI 0.84–3.14, P = 0.15).[33] One explanation for the lower absolute risk reduction in CKD-5D may be a difference in defibrillation threshold.[34] Retrospective data from USRDS reported that the commonest cause of death in dialysis patients with ICD was still arrhythmia,[35] with 38.2% dying from an arrhythmic death, mostly ventricular arrhythmias, compared with 16% in an unselected cohort of 822 patients who had ICD inserted (65% for secondary prevention) over a 10 year period.