Guardiola Jordi, Baliellas Carme, Xiol Xavier, Fernandez Esparrach Glòria, Ginès Pere, Ventura Pere, Vazquez Santiago
Gastroenterology Unit, Hospital de L'Alt Penedès, Vilafranca del Penedès, Barcelona, Spain.
Am J Gastroenterol. 2002 Sep;97(9):2374-8. doi: 10.1111/j.1572-0241.2002.05928.x.
Cirrhotic patients with refractory ascites (RA) have a poor prognosis, although individual survival varies greatly. A model that could predict survival for patients with RA would be helpful in planning treatment. Moreover, in cases of potential liver transplantation, a model of these characteristics would provide the bases for establishing priorities of organ allocation and the selection of patients for a living donor graft. Recently, we developed a model to predict survival of patients with RA. The aim of this study was to establish its generalizability for predicting the survival of patients with RA.
The model was validated by assessing its performance in an external cohort of patients with RA included in a multicenter, randomized, controlled trial that compared large-volume paracentesis and peritoneovenous shunt. The values for actual and model-predicted survival of three risk groups of patients, established according to the model, were compared graphically and by means of the one-sample log-rank test.
The model provided a very good fit to the survival data of the three risk groups in the validation cohort. We also found good agreement between the survival predicted from the model and the observed survival when patients treated with peritoneovenous shunt and with paracentesis were considered separately.
Our survival model can be used to predict the survival of patients with RA and may be a useful tool in clinical decision making, especially in deciding priority for liver transplantation.
尽管肝硬化难治性腹水（RA）患者的个体生存差异很大，但总体预后较差。一个能够预测RA患者生存情况的模型将有助于制定治疗方案。此外，在潜在肝移植的情况下，具有这些特征的模型将为确定器官分配优先级和选择活体供肝移植患者提供依据。最近，我们开发了一个预测RA患者生存的模型。本研究的目的是确定该模型预测RA患者生存的普遍适用性。
通过评估该模型在一项多中心、随机、对照试验中纳入的RA患者外部队列中的表现来验证该模型，该试验比较了大量腹腔穿刺放液术和腹腔静脉分流术。根据该模型确定的三组风险患者的实际生存值和模型预测生存值，通过图形比较和单样本对数秩检验进行比较。
该模型与验证队列中三组风险患者的生存数据拟合良好。当分别考虑接受腹腔静脉分流术和腹腔穿刺放液术治疗的患者时，我们还发现模型预测的生存情况与观察到的生存情况之间具有良好的一致性。
我们的生存模型可用于预测RA患者的生存情况，可能是临床决策中的一个有用工具，特别是在确定肝移植优先级方面。
