Magosso Elisa, Ursino Mauro, Colì Luigi, Baraldi Olga, Bolondi Luigi, Stefoni Sergio
Department of Electronics, Computer Science and Systems, University of Bologna, Bologna, Italy.
Artif Organs. 2006 Apr;30(4):285-300. doi: 10.1111/j.1525-1594.2006.00216.x.
This work presents a quantitative description, by means of a mathematical model, of bilirubin removal during Molecular Adsorbent Recirculating System sessions. The model includes four compartments: two for the patient, and two for the albumin circuit. Equations in each compartment express mass preservation, mass exchange between compartments, and bilirubin-albumin binding kinetics. Model development and validation are based on in vivo data of bilirubin concentration acquired in eight sessions at different times during the session. The accuracy of the model in reproducing real data is high (error in blood = -0.3 +/- 0.93 mg/dL), if three parameters, representing the depurative efficacy of the system (the dialysance of the blood filter and the initial and final clearance of the depurative elements in the albumin circuit), are estimated on each single session. However, model accuracy is only slightly deteriorated (error in blood = -0.4 +/- 0.99 mg/dL) if a single set of parameters (fixing the three parameters at their mean values) is adopted. These results suggest that the model may be used a priori (i.e., using a single set of parameters) to achieve a satisfactory prediction of the overall bilirubin removal, as well as a posteriori for the estimation of device parameters. The latter use may allow the investigation of the dependence of these parameters on the operative and clinical conditions, in the effort to arrive at a rationalization and optimization of the treatment.
本研究通过数学模型对分子吸附循环系统治疗期间胆红素清除情况进行了定量描述。该模型包括四个隔室:两个用于患者,两个用于白蛋白循环回路。每个隔室中的方程表示质量守恒、隔室间的质量交换以及胆红素与白蛋白的结合动力学。模型的开发和验证基于在治疗期间不同时间的八次治疗中获取的胆红素浓度的体内数据。如果根据每次治疗估计代表系统净化效率的三个参数(血液滤过器的透析率以及白蛋白循环回路中净化元件的初始和最终清除率),则该模型在重现实际数据方面的准确性较高(血液中的误差=-0.3±0.93mg/dL)。然而,如果采用一组单一参数(将这三个参数固定在其平均值),模型准确性仅略有下降(血液中的误差=-0.4±0.99mg/dL)。这些结果表明,该模型可先验使用(即使用一组单一参数)以实现对总体胆红素清除的满意预测,也可后验用于估计设备参数。后者的用途可能有助于研究这些参数对操作和临床条件的依赖性,从而努力实现治疗的合理化和优化。