肝硬化患者血清钠特点及含钠终末期肝病模型对预后的判断价值

目的 分析终末期肝硬化患者的血清钠特点与患者生存状况和门静脉高压并发症之间的关系,比较终末期肝病模型(MELD)及其含钠模型对预后的判断价值.方法 选取我院2005年6月至2010年10月失代偿期肝硬化患者的住院资料进行登记和随访.将血清钠水平按≤125 mmol/L、＞ 125 ～＜135 mmol/L和≥135 mmol/L进行分级,分析不同血清钠水平肝硬化患者的生存情况及与肝硬化门静脉高压相关并发症的关系,并分析Child-Pugh分级与血清钠水平的相互关系.利用Kaplan-Meier方法分析不同血清钠水平患者的生存率变化,利用接受者工作特征(ROC)曲线下面积比较MELD与MELD-Na和iMELD判断患者生存不同时间的准确性.组间均数的比较用t检验或方差分析,率的比较用x2检验,ROC曲线下面积的比较采用正态性Z检验.结果 至随访期截止,共有467例患者被纳入本研究.总体低钠血症(血清钠＜ 135 mmol/L)发生率为50.54％ (236/467),其中死亡患者低钠血症发生率为66.81％ (155/232),生存患者为34.47％ (81/235),差异有统计学意义(x2=9.73,P＜0.01).血清钠≤125 mmol/L、＞125 ～＜135 mmol/L和≥135 mmol/L患者的病死率分别为86.00％ (43/50)、60.10％ (110/183)和33.76％ (79/234),差异有统计学意义(P＜ 0.01).Child-Pugh A、B、C级患者的血清钠水平分别为(138.80±4.42)mmol/L、(135.30±6.66) mmol/L和(131.18±7.53) mmol/L,各组间差异均有统计学意义(P值均＜0.05).肝性脑病、肝肾综合征和自发性腹膜炎的发病率因血清钠水平的下降而升高(r值分别为-0.213、-0.342和- 0.142,P值均＜0.05),腹水量也随血清钠水平的降低而增加(P＜0.01),而消化道出血的发生则与血清钠水平无明显关系(r=0.40,P＞0.05).MELD、MELD-Na和iMELD模型在判断患者3个月预后方面无明显差异(P＞ 0.05),而在判断患者6个月和1年预后方面,MELD-Na和iMELD优于MELD(P值均＜0.05).结论 低钠血症与终末期肝病患者的预后及相关并发症发生有一定的关系.MELD与钠相结合后,可以提高MELD判断患者预后的能力.

Serum sodium concentration profile for cirrhotic patients and its effect on the prognostic value of the MELD score

To analyze the characteristics of serum sodium in decompensated cirrhosis and evaluate the prognostic ability of the model for end-stage liver disease (MELD) in Na-containing models. Patients diagnosed with decompensated cirrhosis at our hospital were enrolled for study between June 2005 and October 2010. Patients were classified among three groups, according to serum sodium concentration: less than 125 mmol/L, 125 to 135 mmol/L, and more than 135 mmol/L. Mortality rates among the three groups were compared by Kaplan-Meier survival analysis. In addition, the different serum sodium concentrations were analyzed for correlations between Child-Pugh score and complication incidence rates of portal hypertension. The area under the receiver operating characteristic (ROC) curve (AUC) was used to compare the predictive abilities of MELD, MELD-Na, and the integrated (i) MELD scores for 3-month, 6-month and 1-year mortalities. A total of 467 patients were analyzed, and 50.54% had hyponatremia ( less than 135 mmol/L). Sodium concentration was correlated with mortality, and Kaplan-Meier survival analysis indicated that mortality was significantly higher in each subgroup with lower sodium concentration (all, P = 0.000). Likewise, sodium concentration decreased in conjunction with increased severity of decompensation, as classified by Child-Pugh scoring (sodium: A more than B more than C; mortality: A less than B less than C). With the exception of digestive tract bleeding, complication incidence rates of hepatic encephalopathy, ascites, spontaneous bacterial peritonitis, and hepatorenal syndrome increased when sodium concentration decreased. For predicting 3-month mortality, the AUC scores of MELD were not significantly different from the MELD-Na and iMELD scores (P more than 0.05). For predicting 6-month and 1-year mortality, the AUC scores of MELD-Na and iMELD were significantly higher than those of MELD (P less than 0.05). Hyponatremia is correlated with mortality and complications in decompensated cirrhosis patients. Incorporation of Na into the MELD may enhance it's prognostic ability.