Cardiovascular disease accounts for 40% to 50% of deaths in dialysis populations. Overall, the risk of cardiac mortality is 10-fold to 20-fold greater in dialysis patients than in age and sex-matched controls without chronic kidney disease. The aim of this paper is to review critically the evidence that cardiac outcomes in dialysis patients are modified by cardiovascular risk factor interventions. There is limited, but as yet inconclusive controlled trial evidence that cardiovascular outcomes in dialysis populations may be improved by antioxidants (vitamin E or acetylcysteine), ensuring that hemoglobin levels do not exceed 120 g/L (especially in the setting of known cardiovascular disease), prescribing carvedilol in the setting of dilated cardiomyopathy, and by using cinacalcet in uncontrolled secondary hyperparathyroidism. Similarly, there are a number of negative controlled trials, which have demonstrated that statins, high-dose folic acid, angiotensin-converting enzyme inhibitors, multiple risk factor intervention via multidisciplinary clinics, and high-dose or high-flux dialysis are ineffective in preventing cardiovascular disease. Although none of these studies could be considered conclusive, the negative trials to date should raise significant concerns about the heavy reliance of current clinical practice guidelines on extrapolation of findings from cardiovascular intervention trials in the general population. It may be that cardiovascular disease in dialysis populations is less amenable to intervention, either because of the advanced stage of chronic kidney disease or because the pathogenesis of cardiovascular disease in dialysis patients is different from that in the general population. Large, well-conducted, multicenter randomized-controlled trials in this area are urgently required.

Objectives The aim of this study was to investigate the factors affecting recovery and durability of dialysis-independent renal function following commencement of peritoneal dialysis (PD). Design Retrospective, observational cohort study of the Australian and New Zealand PD patient population. Setting Australia and New Zealand Dialysis and Transplant (ANZDATA) Registry. Participants The study reviewed all patients in Australia and New Zealand who commenced PD for treatment of end-stage renal failure between 15 May 1963 and 31 December 2004. Main Outcome Measures The primary outcomes examined were recovery of dialysis-independent renal function and time from PD commencement to recovery of renal function. A secondary outcome measure was time to renal death (patient death or recommencement of renal replacement therapy) following recovery of dialysis-independent renal function. Results 24663 patients commenced PD during the study period. Of these, 253 (1%) recovered dialysis-independent renal function. An increased likelihood of recovery was predicted by autoimmune renal disease, hemolytic-uremic syndrome, paraproteinemia, cortical necrosis, renovascular disease, and treatment in New Zealand. A reduced likelihood of recovery was associated with polycystic kidney disease and indigenous race. Analysis of a contemporary subset of 14743 patients in whom complete data were available for body mass index, smoking, and comorbidities yielded comparable results, except that increasing age was additionally associated with a decreased likelihood of recovery. Of the 253 patients who recovered renal function, 151 (60%) recommenced renal replacement therapy and 49 (19%) died within a median period of 226 days (interquartile range 110 – 581 days). The only significant predictors of continued renal survival after renal recovery were autoimmune renal disease and cortical necrosis. Conclusions Recovery of renal function in patients treated with PD is rare and determined mainly by renal disease type and race. In the majority of cases, recovery is short term. The apparently high rate of early patient death or return to dialysis after recovery of renal function on PD raises questions about the appropriateness of discontinuing PD therapy under such circumstances.