Session Time: 6:00pm-7:00pm
Presentation Time: 6:00pm-7:00pm
Location: Hall D1
Background This is the era of big data and it may help manage clinical practice better. Decision tree analysis is a useful data mining tool for disorderly realistic world data. Results from former literature on factors influencing renal transplant outcomes are quite confusion and few include factors influencing outcomes of rejection cases. This study was intended to get the factors in order and to provide reliable predictive models for clinical practice with decision tree analysis. Moreover, it attempted to discuss about decision tree integrated with renal transplant database to fulfill the dynamic model growth and to serve personal medical care in big data era. Methods Renal transplant recipients registering between May 1988 and April 2014 in Kidney Center of the First Affiliated Hospital of Zhejiang University were included. Living state, rejection state, state post rejection and overall outcomes were analyzed by decision tree. Results Totally 3921 cases were included. The loss-to-follow rate is 9.7%. With survival rate as the object, we got the most important factor, steroids (P < 0.01). Patients in the following factor combination showed the highest survival rate: steroids, mycophenolate mofetil (MMF), stenting, warm ischemic time (WIT) ≤6.5min, cold ischemic time (CIT) ≤300min, and aspartate aminotransferase (AST) 6 months post operation≤34U/L (survival rate 100%, 305 cases). Patients had the lowest acute rejection rate when in the factor combination: steroids, antihuman thymocyte globulin or basiliximab induction, no blood transfusion history and female (0%, 83). Patients with acute rejection were regarded as a new database. The patients who took steroids and received stents shows similar outcomes with patients who had no rejection. Important factors that could be intervened for long-term outcomes included stenting, statins, steroids, etc. Conclusions Decision tree analysis is an outstanding choice for risk stratification, prognosis prediction and dynamic follow-up. Immunosuppression therapy was regarded as the most important factor for renal transplant recipients' survival, rejection and other outcomes. Factors should be considered in combination for each specific patient.
CITATION INFORMATION: Zhou J, Shen Y, Wang R, Huang H, Wu J, Chen J. Decision Tree Analysis of Renal Transplantation Recipients Outcomes: A Single Center Data Mining in the Big Data Era. Am J Transplant. 2017;17 (suppl 3).
To cite this abstract in AMA style:Zhou J, Shen Y, Wang R, Huang H, Wu J, Chen J. Decision Tree Analysis of Renal Transplantation Recipients Outcomes: A Single Center Data Mining in the Big Data Era. [abstract]. Am J Transplant. 2017; 17 (suppl 3). https://atcmeetingabstracts.com/abstract/decision-tree-analysis-of-renal-transplantation-recipients-outcomes-a-single-center-data-mining-in-the-big-data-era/. Accessed August 6, 2020.
« Back to 2017 American Transplant Congress