A Retrospective Observational Study to Describe Anaemia Management, Burden of Disease and Outcomes in Chinese Patients on Peritoneal Dialysis Using a Clinical Database

Real-world evidence for chronic kidney disease (CKD)-associated anemia in China is scarce. This study aims to provide supporting novel evidence for the real-world management and treatment of patients on PD with CKD-associated anemia. This is an observational cohort study using secondary data. The data source is the PD Telemedicine-based Management Platform (PDTAP) database, which is a clinical database that prospectively collects data in patients receiving PD and evaluates PD management and clinical practice in China. The PDTAP database has enrolled around 7,000 patients receiving PD from 27 hospitals in 14 provinces located in all seven geographical regions (northwest, northeast, north, central, southwest, southeast and south) in China.

The overall objective is to describe the epidemiology, patient characteristics, treatment patterns relevant to anemia and clinical outcomes in patients on PD in real-world practice in China. Specifically, the incidence and prevalence of anemia will be estimated. The patient characteristics, medication use, select laboratory values and clinical outcomes, and healthcare resource utilization (HCRU) will be described in PD patients with and without anemia. The primary outcomes are 1) the incidence and prevalence of anemia, including the factors that predict the incidence of anemia; 2) the number and percentage of patients who experienced all-cause hospitalizations; 3) adverse clinical outcomes, such as mortality MACE and modified MACE+. The secondary outcomes are 1) hemoglobin levels or anemia, iron variables, medications on anemia management and clinical outcomes in patients with or without inflammation; 2) ESA responsiveness; 3) Other clinical outcomes, such as transfer to hemodialysis and peritonitis.

Associations between anemia, anemia-relevant factors and above clinical outcomes can be analyzed in series of paper with specific aims. For example, 1) the relationship between hemoglobin levels or anemia and clinical outcomes (such as mortality, MACE, modified MACE+, hospitalization, transfer to hemodialysis and peritonitis); 2) prognostic factors that predict the incidence of anemia during the follow up; 3) hemoglobin levels or anemia, iron variables, medications on anemia management and clinical outcomes in patients with or without inflammation; 4) the relationship between hemoglobin levels or anemia, erythropoietin responsiveness and clinical outcomes; 5) prognostic factors that predict the hyporesponsiveness to erythropoietin; 6) determine cut-off values of the iron index (such as serum iron, ferritin, TIBC and TSAT) to predicting worse clinical outcomes. Each study could select hemoglobin relevant variables including clinical characteristics and treatment patterns at baseline and/or during the follow up, and explore their associations with specific clinical outcomes.

Confounding, interaction or mediation effects could be evaluated. Subgroups, such as with or without diabetes, with or without CVD, with or without inflammation status et al could be analyzed as needed. Cox, competing risk models, Poisson or NegBin regression models may be used as needed.