BACKGROUND: The question of which modality, either peritoneal dialysis (PD) or hemodialysis (HD), confers the survival advantage for incident ESRD patients with pre-existing cardiovascular disease (CVD) remains unanswered. METHODS: Data used in this study were extracted from the National Health Insurance Research Database in Taiwan. From 1997 to 2007, incident ESRD patients who underwent dialysis longer than three months were selected. The established dialysis modality at day 90 was used to analyze the impact of dialysis modality on survival. For each PD patient indentified, five HD patients matched for age, sex, and year in which the patients received their first dialysis treatment were randomly selected. Finally, a total of 35 664 patients including 29 720 HD patients and 5944 PD patients were selected. The primary outcome was death after commencing dialysis. RESULTS: For diabetic ESRD patients with or without coronary artery disease (CAD) or congestive heart failure (CHF), patients receiving PD had inferior survival compared with those receiving HD (P<.001, adjusted HR=1.34 to 1.43). For nondiabetic patients with CAD or CHF, patients receiving PD also had inferior survival compared with those receiving HD (adjusted HR=1.30, CI: 1.08 to 1.57; adjusted HR=1.31, CI: 1.11 to 1.55). For nondiabetic ESRD patients without CAD or CHF, there was no statistically significant difference in survival between PD and HD (adjusted HR=1.00, CI: 0.92 to 1.09; adjusted HR=0.98, CI: 0.90 to 1.07). CONCLUSIONS: PD was associated with poorer survival among ESRD patients with CVD or diabetes mellitus compared with HD.
BACKGROUND: The question of which modality, either peritoneal dialysis (PD) or hemodialysis (HD), confers the survival advantage for incident ESRDpatients with pre-existing cardiovascular disease (CVD) remains unanswered. METHODS: Data used in this study were extracted from the National Health Insurance Research Database in Taiwan. From 1997 to 2007, incident ESRDpatients who underwent dialysis longer than three months were selected. The established dialysis modality at day 90 was used to analyze the impact of dialysis modality on survival. For each PDpatient indentified, five HDpatients matched for age, sex, and year in which the patients received their first dialysis treatment were randomly selected. Finally, a total of 35 664 patients including 29 720 HDpatients and 5944 PDpatients were selected. The primary outcome was death after commencing dialysis. RESULTS: For diabetic ESRDpatients with or without coronary artery disease (CAD) or congestive heart failure (CHF), patients receiving PD had inferior survival compared with those receiving HD (P<.001, adjusted HR=1.34 to 1.43). For nondiabeticpatients with CAD or CHF, patients receiving PD also had inferior survival compared with those receiving HD (adjusted HR=1.30, CI: 1.08 to 1.57; adjusted HR=1.31, CI: 1.11 to 1.55). For nondiabetic ESRDpatients without CAD or CHF, there was no statistically significant difference in survival between PD and HD (adjusted HR=1.00, CI: 0.92 to 1.09; adjusted HR=0.98, CI: 0.90 to 1.07). CONCLUSIONS:PD was associated with poorer survival among ESRDpatients with CVD or diabetes mellitus compared with HD.