Comparison of peritoneal dialysis and hemodialysis as first renal replacement therapy in patients with end-stage renal disease and diabetes: a systematic review

Background Diabetes has become the most common cause of end-stage renal disease (ESRD) requiring renal replacement therapy (RRT) in most countries around the world. Peritoneal dialysis (PD) is valuable for patients newly requiring RRT because of the preservation of residual renal function (RRF), higher quality of life, and hemodynamic stability in comparison with hemodialysis (HD). A previous systematic review produced conflicting results regarding patient survival. As several advances have been made in therapy for diabetic patients receiving PD, we conducted a systematic review of studies published after 2014 to determine whether incident PD or HD is advantageous for the survival of patients with diabetes. Methods For this systematic review, the MEDLINE, EMBASE, and CENTRAL databases were searched to identify articles published between February 2014 and August 2017. The quality of studies was assessed using the GRADE approach. Outcomes of interest were all-cause mortality; RRF; major morbid events, including cardiovascular disease (CVD) and infectious disease; and glycemic control. This review was performed using a predefined protocol published in PROSPERO (CRD42018104258). Results Sixteen studies were included in this review. All were retrospective observational studies, and the risk of bias, especially failure to adequately control confounding factors, was high. Among them, 15 studies investigated all-cause mortality in diabetic patients initiating PD and HD. Differences favoring HD were observed in nine studies, whereas those favoring PD were observed in two studies. Two studies investigated effects on CVD, and both demonstrated the superiority of incident HD. No study investigated the effect of any other outcome. Conclusions In the present systematic review, the risk of death tended to be higher among diabetic patients with ESRD newly initiating RRT with incident PD in comparison with incident HD. However, we could not obtain definitive results reflecting the superiority of PD or HD with regard to patient outcomes because of the severe risk of bias and the heterogeneity of management strategies for diabetic patients receiving dialysis. Further studies are needed to clarify the advantages of PD and HD as RRT for diabetic patients with ESRD.