Page 110 - 南京医科大学自然版
P. 110

第45卷第6期
               ·848 ·                            南 京    医 科 大 学 学         报                        2025年6月


             (续表1)
                         Characteristic          Total(n=230)   Training set(n=146)  Testing set(n=84)  t/Z/χ 2  P
                 TC[mmol/L,M(P25,P75)]        0 004.2(3.5,5.0)  004.3(3.6,5.0)   0 003.9(3.3,5.0)  -1.473 b  0.141
                 LDL⁃C(mmol/L,x ± s)          002.62 ± 0.79     2.65 ± 0.72       2.56 ± 0.91      -0.718 a  0.474
                 HDL⁃C[mmol/L,M(P25,P75)]     0 001.1(0.9,1.2)  001.1(0.9,1.2)   0 001.0(0.9,1.2)  -0.792 b  0.428
                 TG[mmol/L,M(P25,P75)]        0 001.2(0.9,1.5)  001.2(0.9,1.5)   0 001.2(0.8,1.6)  -0.082 b  0.934
                 Creatinine[μmol/L,M(P25,P75)]  0 070.5(59.9,85.8)  068.1(59.0,86.0)  0 072.5(61.0,85.7)  -0.858 b  0.391
                 eGFR[mL/(min·1.73 m ),M(P25,P75)]0 075.0(59.9,99.5)  076.4(58.7,103.1)  0 074.5(61.2,93.1)  -0.515 b  0.607
                                  2
                 Hemoglobin[g/L,M(P25,P75)]   0 136.0(123.3,147.0) 138.0(125.0,148.0) 0 134.0(121.5,144.2) -1.625 b  0.104
                 WBC[×10 /L,M(P25,P75)]       0 009.2(7.1,10.8)  009.2(7.2,10.9)  0 009.1(6.8,10.8)  -0.275 b  0.784
                        9
                 CRP[mg/L,M(P25,P75)]         0 005.0(5.0,11.1)  005.0(5.0,9.8)  0 005.0(5.0,18.2)  -1.155 b  0.248
                 NT⁃proBNP[pg/mL,M(P25,P75)]  1 026.0(341.6,2 261.7)999.6(341.6,2 240.5)1 059.2(379.1,2 267.7)-0.013 b  0.986
                 hs⁃cTnT[ng/L,M(P25,P75)]     0 013.7(8.2,24.7)  012.8(7.8,22.1)  0 014.1(9.1,28.0)  -1.263 b  0.215
                 PT[s,M(P25,P75)]             0 011.7(11.2,12.4)  011.7(11.2,12.3)  0 011.8(11.3,12.5)  -0.875 b  0.380
                 INR                          0 01.02(0.97,1.08)  01.02(0.97,1.07)  0 01.03(0.98,1.09)  -0.872 b  0.384
                 APTT[s,M(P25,P75)]           0 026.8(25.1,28.1)  026.6(24.8,28.3)  0 026.8(25.4,28.0)  -0.353 b  0.719
                 D⁃dimer[mg/L,M(P25,P75)]     0 00.98(0.41,2.76)  01.15(0.48,3.04)  0 00.76(0.39,2.36)  -1.646 b  0.099
               Imaging[mL,M(P25,P75)]
                 Ischemic volume              0 134.7(78.2,191.7)  140.5(76.7,196.0)  0 127.4(82.5,179.5)  -0.692 b  0.483
                 Core infarct volume          0 007.0(0.0,34.7)  009.0(0.0,34.7)  0 003.5(0.0,34.5)  -1.204 b  0.227
               ASPECTS[n(%)]                                                                        NA d  0.833
                 0                            0 000.4(1.7)      000.3(2.1)       0 000.1(1.2)
                 1                            0 000.1(0.4)      000.1(0.7)       000.00(0)
                 2                            0 000.7(3.0)      000.5(3.4)       000.02(2.4)
                 3                            0 00.14(6.1)      00.10(6.8)       000.04(4.8)
                 4                            0 00.29(12.6)     00.22(15.1)      000.07(8.3)
                 5                            0 00.28(12.2)     00.16(11.0)      000.12(14.3)
                 6                            0 00.31(13.5)     00.21(14.4)      000.10(11.9)
                 7                            0 00.26(11.3)     00.14(9.6)       000.12(14.3)
                 8                            0 00.40(17.4)     00.25(17.1)      000.15(17.9)
                 9                            0 00.30(13.0)     00.17(11.6)      000.13(15.5)
                 10                           0 00.20(8.7)      00.12(8.2)       000.08(9.5)
                 a:Continuous variables with a normal distribution were compared using Student’s t⁃test,and the test statistic reported was the t⁃value. b:Continuous
              variables with a non⁃normal distribution were compared using the Mann⁃Whitney U test,and the test statistic reported was the Z⁃value. c:Categorical
                                                                            2
              variables were compared using Pearson’s chi⁃square test,and the test statistic reported was the χ ⁃value. d:For categorical variables with expected cell
              counts less than 5,Fisher’s exact test was used,and no test statistic was reported. BMI:body mass index;mRS:the modified Rankin scale;SBP:systolic
              blood pressure;DBP:diastolic blood pressure;NIHSS:national institutes of health stroke scale;AST:aspartate aminotransferase;ALT:alanine amino⁃
              transferase;TC:total cholesterol;LDL⁃C:low⁃density lipoprotein cholesterol;HDL⁃C:high⁃density lipoprotein cholesterol;TG:triglyceride;eGFR:esti⁃
              mated glomerular filtration rate;WBC:white blood count;CRP:C⁃reactive protein;NT⁃proBNP:N⁃terminal pro⁃brain natriuretic peptide;hs⁃cTnT:high⁃
              sensitivity cardiac troponin T;PT:prothrombin time;INR:international normalized ratio;APTT:activated partial thromboplastin time;ASPECTS:alberta
              stroke program early CT score.


              性。图2A和2B分别展示训练集和测试集2年全因                           2.4  预测模型的应用
              死亡风险预测模型的校准曲线。训练集中基础模                                  NT⁃proBNP 预测 2 年全因死亡风险的 AUC 为
              型、模型 1、模型 2 的 Brier 评分分别为 0.153、0.137、            0.743,最佳截断值为 1 158.64 pg/mL(灵敏度 0.750,
              0.161;测 试 集 中 Brier 评 分 分 别 为 0.176、0.170、        特异度 0.732)。按 NT⁃proBNP 最佳截断值分组,高
              0.173。模型 1 在训练集与测试集中均显示更佳的                        NT⁃proBNP 组的死亡风险是低 NT⁃proBNP 组的 4 倍
              校准性。                                             (HR=4.18,95%CI:2.05~8.55,P < 0.001)。图3为基
   105   106   107   108   109   110   111   112   113   114   115