Suwasin Udomkarnjananun1,2,3, Natavudh Townamchai1,2,3, Stephen J Kerr4, Adis Tasanarong5, Kajohnsak Noppakun6, Adisorn Lumpaopong7, Surazee Prommool8, Thanom Supaporn9, Yingyos Avihingsanon1,2,3, Kearkiat Praditpornsilpa1, Somchai Eiam-Ong1. 1. Division of Nephrology, Department of Medicine, Faculty of Medicine, Chulalongkorn University and King Chulalongkorn Memorial Hospital, Bangkok, Thailand. 2. Excellence Center for Organ Transplantation (ECOT), King Chulalongkorn Memorial Hospital, Thai Red Cross Society, Bangkok, Thailand. 3. Renal Immunology and Transplantation Research Unit, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand. 4. Research Affairs, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand. 5. Division of Nephrology, Department of Medicine, Faculty of Medicine, Thammasat University, Pathum Thani, Thailand. 6. Division of Nephrology, Department of Medicine, Faculty of Medicine, Chiang Mai University, Chiang Mai, Thailand. 7. Division of Nephrology, Department of Pediatric, Faculty of Medicine, Phramongkutklao Hospital and College of Medicine, Bangkok, Thailand. 8. Division of Nephrology, Department of Medicine, Faculty of Medicine, Vajira Hospital, Navamindradhiraj University, Bangkok, Thailand. 9. Division of Nephrology, Department of Medicine, Faculty of Medicine, Phramongkutklao Hospital and College of Medicine, Bangkok, Thailand.
Abstract
BACKGROUND: Several kidney transplantation (KT) prediction models for patient and graft outcomes have been developed based on Caucasian populations. However, KT in Asian countries differs due to patient characteristics and practices. To date, there has been no equation developed for predicting outcomes among Asian KT recipients. METHODS: We developed equations for predicting 5- and 10-year patient survival (PS) and death-censored graft survival (DCGS) based on 6662 patients in the Thai Transplant Registry. The cohort was divided into training and validation data sets. We identified factors significantly associated with outcomes by Cox regression. In the validation data set, we also compared our models with another model based on KT in the United States. RESULTS: Variables included for developing the DCGS and PS models were recipient and donor age, background kidney disease, dialysis vintage, donor hepatitis C virus status, cardiovascular diseases, panel reactive antibody, donor types, donor creatinine, ischemic time, and immunosuppression regimens. The C statistics of our model in the validation data set were 0.69 (0.66-0.71) and 0.64 (0.59-0.68) for DCGS and PS. Our model performed better when compared with a model based on US patients. Compared with tacrolimus, KT recipients aged ≤44 years receiving cyclosporine A had a higher risk of graft loss (adjusted hazard ratio = 1.26; P = 0.046). The risk of death was higher in recipients aged >44 years and taking cyclosporine A (adjusted hazard ratio = 1.44; P = 0.011). CONCLUSIONS: Our prediction model is the first based on an Asian population, can be used immediately after transplantation. The model can be accessed at www.nephrochula.com/ktmodels.
BACKGROUND: Several kidney transplantation (KT) prediction models for patient and graft outcomes have been developed based on Caucasian populations. However, KT in Asian countries differs due to patient characteristics and practices. To date, there has been no equation developed for predicting outcomes among Asian KT recipients. METHODS: We developed equations for predicting 5- and 10-year patient survival (PS) and death-censored graft survival (DCGS) based on 6662 patients in the Thai Transplant Registry. The cohort was divided into training and validation data sets. We identified factors significantly associated with outcomes by Cox regression. In the validation data set, we also compared our models with another model based on KT in the United States. RESULTS: Variables included for developing the DCGS and PS models were recipient and donor age, background kidney disease, dialysis vintage, donorhepatitis C virus status, cardiovascular diseases, panel reactive antibody, donor types, donorcreatinine, ischemic time, and immunosuppression regimens. The C statistics of our model in the validation data set were 0.69 (0.66-0.71) and 0.64 (0.59-0.68) for DCGS and PS. Our model performed better when compared with a model based on US patients. Compared with tacrolimus, KT recipients aged ≤44 years receiving cyclosporine A had a higher risk of graft loss (adjusted hazard ratio = 1.26; P = 0.046). The risk of death was higher in recipients aged >44 years and taking cyclosporine A (adjusted hazard ratio = 1.44; P = 0.011). CONCLUSIONS: Our prediction model is the first based on an Asian population, can be used immediately after transplantation. The model can be accessed at www.nephrochula.com/ktmodels.
Authors: Jong Cheol Jeong; Tai Yeon Koo; Han Ro; Dong Ryeol Lee; Dong Won Lee; Jieun Oh; Jayoun Kim; Dong-Wan Chae; Young Hoon Kim; Kyu Ha Huh; Jae Berm Park; Yeong Hoon Kim; Seungyeup Han; Soo Jin Na Choi; Sik Lee; Sang-Il Min; Jongwon Ha; Myoung Soo Kim; Curie Ahn; Jaeseok Yang Journal: Sci Rep Date: 2022-05-24 Impact factor: 4.996