Soo Jeong Choi1,2, Yoshitsugu Obi1, Gang Jee Ko1,3, Amy S You1, Rieko Eriguchi1, Mengjing Wang1,4, Connie M Rhee1, Kamyar Kalantar-Zadeh5,6,7. 1. Division of Nephrology and Hypertension and Kidney Transplantation, University of California Irvine School of Medicine, Orange, California, USA. 2. Division of Nephrology, Department of Internal Medicine, Soonchunhyang University College of Medicine, Bucheon, Republic of Korea. 3. Department of Internal Medicine, Korea University College of Medicine, Seoul, Republic of Korea. 4. Division of Nephrology, Department of Medicine, Huashan Hospital, Fudan University, Shanghai, China. 5. Division of Nephrology and Hypertension and Kidney Transplantation, University of California Irvine School of Medicine, Orange, California, USA, kkz@uci.edu. 6. Fielding School of Public Health at University of California-Los Angeles, Los Angeles, California, USA, kkz@uci.edu. 7. Section of Nephrology, Tibor Rubin Veterans Affairs Medical Center, Long Beach, California, USA, kkz@uci.edu.
Abstract
BACKGROUND: It is not clear whether peritoneal dialysis (PD) and home hemodialysis (HHD) have similar outcomes, and little is known about how mortality associated with HHD versus PD differs according to the duration of dialysis. METHODS: We examined a national cohort of incident end-stage renal disease patients that was comprised of 1,993 and 16,514 patients transitioning to HHD and PD, respectively, from 2007 to 2011. The HHD patients were matched with PD patients using propensity score (PS). Demographics, comorbidities, duration of dialysis, and body mass index were adjusted for in logistic regression models using PS matching. We matched 1,915 HHD patients with 1,915 PD patients based on the PS. The patients were categorized by their vintage (duration of dialysis) at the time of the transition to HHD or PD (<3, 3 to <12, and ≥12 months). RESULTS: In the matched cohort, 237 and 359 deaths occurred in the HHD and PD patients, respectively (cumulative incidence 9.6 vs. 12.9/100 patient-years, p < 0.001). PD patients who transitioned within 12 months of starting dialysis had similar mortality risks, while PD patients who transitioned >12 months after starting dialysis had an 83% higher risk for mortality (hazard ratio 1.83; 95% CI 1.33-2.52). CONCLUSIONS: Whereas there was no meaningful survival difference in the first 12 months between HHD and PD, patients who transitioned to PD after 12 months of dialysis had worse survival than their HHD counterparts. Additional studies are warranted to investigate clinical implications of these differences.
BACKGROUND: It is not clear whether peritoneal dialysis (PD) and home hemodialysis (HHD) have similar outcomes, and little is known about how mortality associated with HHD versus PD differs according to the duration of dialysis. METHODS: We examined a national cohort of incident end-stage renal diseasepatients that was comprised of 1,993 and 16,514 patients transitioning to HHD and PD, respectively, from 2007 to 2011. The HHD patients were matched with PD patients using propensity score (PS). Demographics, comorbidities, duration of dialysis, and body mass index were adjusted for in logistic regression models using PS matching. We matched 1,915 HHD patients with 1,915 PD patients based on the PS. The patients were categorized by their vintage (duration of dialysis) at the time of the transition to HHD or PD (<3, 3 to <12, and ≥12 months). RESULTS: In the matched cohort, 237 and 359 deaths occurred in the HHD and PD patients, respectively (cumulative incidence 9.6 vs. 12.9/100 patient-years, p < 0.001). PD patients who transitioned within 12 months of starting dialysis had similar mortality risks, while PD patients who transitioned >12 months after starting dialysis had an 83% higher risk for mortality (hazard ratio 1.83; 95% CI 1.33-2.52). CONCLUSIONS: Whereas there was no meaningful survival difference in the first 12 months between HHD and PD, patients who transitioned to PD after 12 months of dialysis had worse survival than their HHD counterparts. Additional studies are warranted to investigate clinical implications of these differences.