Construction of a predictive model for allogeneic red blood cell transfusion volume in adult liver transplantation with preoperative variables
-
摘要: 目的 通过回顾性分析研究成人肝移植患者术前变量对术中异体红细胞输注量的影响，应用多元线性回归分析构建肝移植术中异体红细胞输注量的预测模型并评估分析，为肝移植术前备血提供参考依据。方法 回顾性收集2018年1月至2020年6月同一肝移植团队进行的185例肝移植患者临床资料，通过SPSS生成随机数后，按4：1比例分为训练样本和测试样本。以训练样本分析肝移植术中异体红细胞输注量的影响因素，再以术中异体红细胞输注量为因变量，以单因素分析结果中差异有统计学意义的影响因素为自变量，进行多元线性回归分析，构建肝移植术中异体红细胞输注量预测模型，并用测试样本验证模型。结果 ① 单因素分析结果显示，肝脏移植患者的性别、诊断、术前变量(Hb、HCT、PLT、INR、PT、APTT、TT、Fib、TBIL、DBIL、TP)和终末期肝病模型(MELD)评分与肝移植术中异体红细胞输注量存在相关性(P < 0.05)。②经多元线性回归分析拟合预测模型：肝移植术中异体红细胞输注量Y=23.920-0.497×HCT+0.262×MELD评分。③预测值与真实值存在相关性(P < 0.05)，预测值与真实值之间绝对误差小于术前常规备血量(20 U)与真实值之间绝对误差(P < 0.05)。结论 肝移植术中异体红细胞输注量的独立危险因素是术前高MELD评分，HCT是其保护性因素。基于成人肝移植术中异体红细胞输注影响因素构建的肝移植术中异体红细胞输注预测模型，可为临床医生评估备血量提供参考依据，降低备血量与实际用血量的误差。Abstract: Objective To analyze the influence of preoperative variables on intraoperative allogeneic red blood cell transfusion in adult liver transplant patients by retrospective analysis, and construct a prediction model of red blood cell transfusion volume in liver transplantation by multiple linear regression analysis, then evaluate and analysis the model, so as to provide reference for blood preparation before liver transplantation.Methods Clinical data were retrospectively analyzed for 185 patients undergoing liver transplantation performed by the same surgical team of our hospital from Affiliated Drum Tower Hospital Nanjing University Medical School from January 2018 to June 2020. After generating the random number through SPSS, they were divided into a training group and a testing group according to the ratio of 4 : 1. Training samples were utilized to analyze the influencing factors of allogeneic red blood cell infusion volume during liver transplantation. A multivariate linear regression analysis was conducted to construct a prediction model for the amount of allogeneic red blood cell transfusion during liver transplantation, with the amount of allogeneic red blood cell transfusion during liver transplantation as the dependent variable and the factors with statistically significant differences in the results of univariate analysis as the independent variable. Verify the model with test samples.Results ① Univariate analysis showed that gender, diagnosis, preoperative Hb, HCT, PLT, INR, PT, APTT, TT, Fib, TBIL, DBIL, TP and MELD score were correlated with allogeneic red blood cell transfusions during liver transplantation(P < 0.05). ②By multiple linear regression analysis, the prediction model of allogeneic red blood cell transfusion volume during liver transplantation was Y=23.920-0.497×HCT+0.262×MELD score. ③There was a correlation between the predicted value and the true value(P < 0.05), and the absolute error between the predicted value and the true value was smaller than the absolute error between the preoperative routine blood volume(20 U) and the true value(P < 0.05).Conclusion The independent risk factor for the amount of allogeneic red blood cell transfusion during liver transplantation was high MELD score, while HCT was its protective factors. The prediction model of allogeneic red blood cell transfusion volume during liver transplantation based on the influencing factors of allogeneic red blood cell transfusion during adult liver transplantation can provide a reference for clinicians to evaluate the prepared blood volume and reduce the error between the blood preparation volume and the actual blood used volume.
-
-
表 1 训练样本与测试样本比较
M(P25，P75)，X±S 变量 训练样本 测试样本 U/t/χ2 P 异体红细胞输注量/U 8.0(4.0，15.0) 10.0(4.3，19.0) -1.290 0.197 性别/例(%) 2.985 0.084 男 103(69.6) 31(83.8) 女 45(30.4) 6(27.6) 年龄/岁 49.40±11.56 52.11±10.39 -1.300 0.195 诊断/例(%) 4.757* 0.168 肝细胞癌 51(34.5) 17(36.8) 肝功能衰竭 31(20.9) 5(13.5) 肝硬化 62(41.9) 12(32.4) 其他肝病 4(2.7) 3(8.1) 血型/例(%) 0.087 0.993 A 56(37.8) 14(37.8) B 41(27.7) 11(29.7) O 29(19.6) 7(18.9) AB 22(14.9) 5(13.5) 上腹部手术史/例(%) 1.461 0.227 是 38(25.7) 6(16.2) 否 110(74.3) 31(83.8) 重度脾亢/例(%) /** 0.303 是 25(16.9) 3(8.1) 否 123(83.1) 34(91.9) 门静脉血栓病史/例(%) /** 1.000 是 11(7.4) 3(8.1) 否 137(92.6) 34(91.9) 肝性脑病/例(%) < 0.001 1.000 是 20(13.5) 5(13.5) 否 128(86.5) 32(86.5) 门静脉高压/例(%) < 0.001 1.000 是 20(13.5) 5(13.5) 否 128(86.5) 32(86.5) Hb/(g·L-1) 112.43±27.62 108.43±28.28 0.783 0.435 HCT/% 32.94±7.70 32.52±8.25 0.315 0.753 PLT/(×109·L-1) 78.0(43.0，133.5) 84.0(46.0，134.0) -0.721 0.471 INR 1.28(1.10，1.72) 1.32(1.12，1.76) -0.314 0.753 PT/s 14.7(12.5，19.4) 14.5(12.6，19.9) -0.294 0.769 APTT/s 32.9(28.2，41.2) 32.0(28.1，37.9) -0.551 0.582 TT/s 20.3(18.7，23.0) 20.1(18.8，21.9) -0.616 0.538 Fib/(g·L-1) 1.9(1.2，2.4) 1.7(1.3，2.9) -0.959 0.338 ALT/(U·L-1) 34.0(20.1，69.9) 32.(18.0，67.3) -0.436 0.663 AST/(U·L-1) 47.3(26.9，105.0) 41.0(27.8，77.7) -1.047 0.295 LDH/(U·L-1) 143.6(92.7，217.5) 148.5(86.5，247.8) -0.122 0.903 TBIL/(μmol·L-1) 44.2(17.3，232.7) 31.2(16.9，126.8) -1.243 0.214 DBIL/(μmol·L-1) 20.0(6.9，132.2) 15.6(7.1，107.2) -0.858 0.391 TP/(g·L-1) 65.1±10.8 65.2±8.8 -0.031 0.975 A/G 1.32(1.07，1.61) 1.23(0.98，1.45) -1.352 0.176 BUN/(mmol·L-1) 72.5(5.5，267.3) 137.0(5.9，308.0) -1.143 0.253 Cr/(μmol·L-1) 58.0(45.0，73.8) 61.0(53.5，83.0) -1.864 0.062 MELD评分 9.1(3.8，16.6) 9.6(4.5，16.6) -0.412 0.680 *Fisher-Freeman-Halton检验拟合近似卡方分布计算出来的值，**Fisher精确检验无χ2值。 
下载: 导出CSV
表 2 148例训练样本肝移植术中异体红细胞输注量的单因素分析
变量 样本/例(%) 红细胞/U U/H P 性别 -2.301 0.021 男 103(69.6) 6.8(3.0，14.0) 女 45(30.4) 10.5(6.0，16.4) 诊断 25.269 < 0.001 肝细胞癌 51(34.5) 4.5(3.0，9.0) 肝功能衰竭 31(20.9) 12.9±7.8 肝硬化 62(41.9) 8.0(4.9，13.8) 其他肝病 4(2.7) 20.1(13.8，39.8) 血型 1.319 0.725 A 56(37.8) 8.0(4.1，14.8) B 41(27.7) 6.5(4.0，12.8) O 29(19.6) 10.0(4.0，16.8) AB 22(14.9) 7.5(3.8，14.3) 上腹部手术史 -1.171 0.242 是 38(25.7) 6.0(3.0，16.1) 否 110(74.3) 8.2(4.5，15.0) 重度脾亢 -0.436 0.663 是 25(16.9) 9.4±5.5 否 123(83.1) 8.0(4.0，15.5) 门静脉血栓病史 -0.318 0.750 是 11(7.4) 11.4±9.7 否 137(92.6) 8.0(4.0，14.0) 肝性脑病 -1.907 0.057 是 20(13.5) 12.9±9.2 否 128(86.5) 7.8(4.0，14.0) 门静脉高压 -1.404 0.160 是 20(13.5) 6.0(4.0，9.5) 否 128(86.5) 8.0(4.0，15.4)
下载: 导出CSV
表 3 肝移植术中异体红细胞输注量预测模型
变量 系数 P 95.0%CI 容差 VIF 常量 23.920 < 0.001 17.873~29.967 / / HCT -0.497 < 0.001 -0.658~-0.337 0.928 1.078 MELD评分 0.262 < 0.001 0.123~0.401 0.928 1.078
下载: 导出CSV
-
[1] 张利姗, 叶海丹, 廖昌贵. 肝移植受者生存质量现状及影响因素分析[J]. 中华移植杂志(电子版), 2019, 13(2): 127-130. doi: 10.3877/cma.j.issn.1674-3903.2019.02.010
[2] 滕大洪, 杨怡萍, 郑虹. 肝移植术中输血管理的研究进展[J]. 实用器官移植电子杂志, 2019, 7(6): 493-496. doi: 10.3969/j.issn.2095-5332.2019.06.023
[3] Sachan D, Tiwari AK, Dara R, et al. Patient blood management in a patient with multiple red cell antibodies(anti-C, anti-e, and anti-K)undergoing liver transplant in South India: A team approach[J]. Asian J Transfus Sci, 2020, 14(1): 74-78. doi: 10.4103/ajts.AJTS_54_18
[4] Vega M, Zubair A, Perry D, et al. Liver Transplantation in a Patient With Anti-Coa Antibodies[J]. Prog Transplant, 2019, 29(3): 287-288. doi: 10.1177/1526924819855358
[5] Butler P, Israel L, Nusbacher J, et al. Blood transfusion in liver transplantation[J]. Transfusion, 1985, 25(2): 120-123. doi: 10.1046/j.1537-2995.1985.25285169201.x
[6] Bonnet A, Gilquin N, Steer N, et al. The use of a thromboelastometry-based algorithm reduces the need for blood product transfusion during orthotopic liver transplantation: A randomised controlled study[J]. Eur J Anaesthesiol, 2019, 36(11): 825-833. doi: 10.1097/EJA.0000000000001084
[7] 田大治, 滕大洪, 于杨, 等. 无输血肝移植的临床初步探索研究[J]. 中华外科杂志, 2021, 59(5): 348-352.
[8] Yoon JU, Byeon GJ, Park JY, et al. Bloodless living donor liver transplantation: Risk factors, outcomes, and diagnostic predictors[J]. Medicine(Baltimore), 2018, 97(50): e13581.
[9] Park B, Yoon J, Kim HJ, et al. Transfusion Status in Liver and Kidney Transplantation Recipients-Results from Nationwide Claims Database[J]. J Clin Med, 2020, 9(11): 3613. doi: 10.3390/jcm9113613
[10] Murphy WG. The sex difference in haemoglobin levels in adults-mechanisms, causes, and consequences[J]. Blood Rev, 2014, 28(2): 41-47. doi: 10.1016/j.blre.2013.12.003
[11] 陈波, 刘银, 刘业, 等. 肝移植围手术期输血的综合评估[J]. 中国输血杂志, 2017, 30(7): 711-713. https://www.cnki.com.cn/Article/CJFDTOTAL-BLOO201707025.htm
[12] 李莺, 宋志群, 熊婷, 等. 肝移植患者围手术期成分输血分析与评估[J]. 临床血液学杂志, 2016, 29(12): 935-939. https://www.cnki.com.cn/Article/CJFDTOTAL-LCXZ201612001.htm
[13] 陈杰, 李平, 刘婷婷, 等. 肝移植术中大量输血的影响因素分析及预测模型构建[J]. 中华器官移植杂志, 2021, 42(1): 29-33. doi: 10.3760/cma.j.cn421203-20190408-00190
[14] Pustavoitau A, Lesley M, Ariyo P, et al. Predictive Modeling of Massive Transfusion Requirements During Liver Transplantation and Its Potential to Reduce Utilization of Blood Bank Resources[J]. Anesth Analg, 2017, 124(5): 1644-1652. doi: 10.1213/ANE.0000000000001994
[15] Singh SA, Prakash K, Sharma S, et al. Predicting packed red blood cell transfusion in living donor liver transplantation: A retrospective analysis[J]. Indian J Anaesth, 2019, 63(2): 119-125. doi: 10.4103/ija.IJA_401_18
[16] Makroo RN, Walia RS, Aneja S, et al. Preoperative predictors of blood component transfusion in living donor liver transplantation[J]. Asian J Transfus Sci, 2013, 7(2): 140-146. doi: 10.4103/0973-6247.115586
[17] 许育兵, 刘广亚, 朱展鸿, 等. 建立多元回归分析方程指导肝移植术前合理预备血[J]. 中国输血杂志, 2018, 31(5): 474-477. https://www.cnki.com.cn/Article/CJFDTOTAL-BLOO201805011.htm
[18] 卢根杰, 陈巍, 卢长江, 等. DCD供肝肝移植术中输血预测及预后分析[J]. 临床血液学杂志, 2020, 33(6): 381-385. https://www.cnki.com.cn/Article/CJFDTOTAL-LCXZ202006002.htm
[19] 王建睿, 苗伶俐, 孙海潮, 等. TEG及传统凝血检测指标在孕妇血液不同稀释状态下的相关性探讨[J]. 临床血液学杂志, 2021, 34(4): 284-287. https://www.cnki.com.cn/Article/CJFDTOTAL-LCXZ202104017.htm
-