自体动静脉内瘘血管超声测量对血液透析尿素清除指数评估的临床价值

The clinical value of ultrasonography measurement for autologous arteriovenous fistula on the estimation of urea clearance index in hemodialysis patients

目的探讨超声测量自体动静脉内瘘(arteriovenous fistula,AVF)血管对AVF成熟及血液透析(hemodialysis,HD)治疗充分性的评价作用。方法对8个HD中心符合条件的患者进行横断面研究,超声测量前臂AVF血管,包括AVF侧上肢肱动脉末段(a1)、桡动脉(a2)和尺动脉起始段(a3)、AVF吻合口上桡动脉(a4)、AVF流出道静脉穿刺引血段(a5)等血管的内径(diameter,D)、收缩峰流速(systolic peak flow velocity,SPV)、舒张末流速(end-diastolic flow velocity,EDV)、血流量(blood flow volume,BFV),选取其中稳定HD测算单次尿素溶质清除指数Kt/V,以Da5≥0.5cm且BFVa5>500ml/min为AVF成熟,Kt/V≥1.2为HD治疗达标,对AVF成熟和Kt/V达标进行Logistic回归分析,用受试者工作特征曲线(receiver operating characteristic curve,ROC)根据Logistic回归公式评估Kt/V达标情况;根据AVF成熟和Kt/V达标情况,分为双达标组(组1)、AVF成熟但Kt/V不达标组(组2)、AVF不成熟但Kt/V达标组(组3),对各组间各参数进行对比分析。结果 8家HD中心入组患者324例,AVF成熟223例(68.83%),HD治疗达标310例(95.68%),AVF成熟的相关参数为Da2(相关系数为5.779,OR为0.003,P=0.003),Kt/V达标的相关参数为体质量(相关系数为-0.057,OR为1.059,P=0.000)、有无糖尿病(相关系数为0.983,OR为0.374,P=0.035)、和Da5(相关系数为6.137,OR为0.002,P=0.000),Logistic公式估算值对Kt/V进行ROC计算得到的曲线下面积=0.849(P=0.000,95%CI 0.784~0.914);组1有209例(64.5%)、组2有14例(4.3%)、组3有101例(31.2%),组2与另两组比糖尿病比例(78.6%对组1的40.2%,χ2=15.234,P=0.000;对组2的35.6%,χ2=84.236,P=0.000)、身高[(173.64±8.35)cm对组1的(167.37±7.86)cm,P=0.003,95%CI:2.15~10.39;对组3的(166.10±6.89)cm,P=0.001,95%CI:3.29~11.80]、体质量[(85.04±22.19)Kg对组1的(66.63±10.72)Kg,P=0.000,95%CI:-24.88~11.93;对组3的(65.31±12.37)Kg,P=0.000,95%CI:-26.41~13.03]更大、Kt/V[(1.04±0.13)对组1的(1.31±0.28),P=0.000,95%CI:0.12~0.42;对组3的(1.29±0.27)cm,P=0.001,95%CI:0.101~0.406]更小,组3的Da1~Da4[Da1为(0.54±0.11)cm对(0.61±0.13)cm,P=0.000,95%CI:-0.063~0.072;Da2为(0.41±0.10)cm对(0.48±0.13)cm,P=0.000,95%CI:-0.062~0.016;Da3为(0.41±0.09)cm对(0.46±0.11)cm,P=0.000,95%CI:-0.071^-0.021;Da4为(0.31±0.08)cm对(0.35±0.10)cm,P=0.001,95%CI:-0.062^-0.016]、SPVa1[(112.01±48.70)cm/s对(129.45±48.12)cm/s,P=0.003,95%CI:-28.780^-6.097]和EDVa1[(55.34±31.20)cm/s对(69.08±43.19)cm/s,P=0.004,95%CI:-23.050^-4.415]明显小于组1,而Da5[(0.46±0.15)cm对组1的(0.75±0.31)cm,P=0.000,95%CI:-0.349^-0.221;对组2的(0.63±0.26)cm,P=0.037,95%CI:-0.311^-0.010]和BFVa5[(587.67±413.08)ml/min对组1的(2039.28±1550.72)ml/min,P=0.000,95%CI:-1762.72~1140.502;对组2的(1695.37±401.58)ml/min,P=0.003,95%CI:-1839.843^-375.571]明显小于另两组。结论超声测量血管可以反映AVF成熟情况,Da2是AVF成熟与否的独立影响因素,包含Da5、体质量和患DM情况的回归公式可以估测单次HD的Kt/V水平,AVF成熟与Kt/V达标存在不匹配。

Objectives To study the measurement of arteriovenous fistula（AVF） by ultrasonography for the estimation of AVF maturation and hemodialysis（HD） adequacy. Methods HD patients with forearm AVF in the 8 HD centers were enrolled into this cross-sectional study. AVFs were measured by ultrasonography to obtain the parameters of diameter（D）, systolic peak flow velocity（SPV）, end-diastolic flow velocity（EDV）and blood flow volume（BFV） at the sites of end segment of brachial artery（a1）, beginning segments of radial artery（a2） and ulnar artery（a3）, upper segment of radial artery before anastomotic stoma（a4） and segment of cephalic vein near the blood access suite（a5）. Urea clearance index（Kt/V） was measured in one stable HD session in the middle of a week. An AVF with D-a5 ≥0.5 cm and BFV-a5〉500 ml/min was regarded as a maturated AVF. The adequacy of HD was identified as Kt/V ≥1.2. Logistic regression was done to analyze the parameters affecting maturation of AVF and the rate of compliance with the target Kt/V value. Receiver operating characteristic curve（ROC） was performed to estimate the rate of compliance with the target Kt/V value based on the formula from logistic regression analyses. Patients were then divided into 3 groups according to maturation of AVF and compliance with the target Kt/V value, group one with maturated AVF and compliance with the target Kt/V value, group 2 with maturated AVF and without compliance with the target Kt/V value, and group 3 without maturated AVF but with compliance with the target Kt/V value. The parameters were compared among the three groups. Results A total of 324 patients were enrolled in this study, including 223 patients（68.83%） with maturated AVFs and 310 patients（95.68%） with Kt/V compliant with the target level.The independent correlation parameter for AVF maturation was D-a2（RR=5.779, OR=0.003, P=0.003）. The independent correlation parameters for Kt/V compliant with the target level were body weight（RR=-0.057,OR=1.059, P=0.000）, diabetes（RR=0.983, OR=0.374, P=0.035） and D-a5（RR=6.137, OR=0.002, P=0.000）.The area under the curve to estimate the rate of compliance with target Kt/V value based on the formula from logistic regression was 0.849（P=0.000, 95% CI 0.784-0.914）. There were 209 patients（64.5%） in group one, 14 patients（4.3%） in group 2 and 101 patients（31.2%） in group 3. Compared to the other two groups, patients in group 2 had more diabetes patients（78.6% vs. 40.2%, χ^2=15.234, P=0.000, compared to group one;78.6% vs. 35.6%, χ^2=84.236, P=0.000, compared to group 3）, higher height [（173.64 ± 8.35）cm vs.（167.37 ±7.86）cm, P=0.003, 95% CI 2.15-10.39, compared to group one;（173.64±8.35）cm vs.（166.10±6.89）cm, P=0.001, 95% CI 3.29-11.80, compared to group 3）, heavier weight [（85.04±22.19）kg vs.（66.63±10.72）kg, P=0.000, 95% CI-24.88-11.93, compared to group one;（85.04±22.19）kg vs.（65.31±12.37）kg, P=0.000, 95%CI-26.41-13.03, compared to group 3）, and lower Kt/V [（1.04 ± 0.13） vs.（1.31 ± 0.28）, P=0.000, 95% CI0.12-0.42, compare to group one;（1.04±0.13） vs.（1.29±0.27）, P=0.001, 95% CI 0.101-0.406, compared to group 3]. Compared to group one, patients in group 3 had less values of D-a1, D-s2, D-a3 and D-a4 [（0.54±0.11）cm vs.（0.61±0.13）cm, P=0.000, 95% CI-0.063-0.072 for D-a1;（0.41±0.10）cm vs.（0.48±0.13）cm, P=0.000, 95% CI-0.062-0.016 for D-a2;（0.41±0.09）cm vs.（0.46±0.11）cm, P=0.000, 95% CI-0.071--0.021 for D-a3;（0.31 ± 0.08）cm vs.（0.35 ± 0.10）cm, P=0.001, 95% CI-0.062--0.016 for D-a4], lower SPV-a1[（112.01 ± 48.70）cm/s vs.（129.45 ± 48.12）cm/s, P=0.003, 95% CI-28.78-6.097） and lower EDV-a1（55.34 ±31.20）cm/s vs.（69.08 ± 43.19）cm/s, P=0.004, 95% CI-23.050--4.415]. Compared to groups one and 2, patients in group 3 had less value of D-a5 [（0.46±0.15）cm vs.（0.75±0.31）cm, P=0.000, 95% CI-0.349--0.221,compared to group one; 0.46 ± 0.15 cm vs.（0.63 ± 0.26）cm, P=0.037, 95% CI-0.311--0.010, compared togroups 2] and less BFV-a5 [（587.67 ± 413.08）ml/min vs.（2,039.28 ± 1,550.72）ml/min, P=0.000, 95% CI-1,762.72-1,140.502, compared to group one; 587.67±413.08 ml/min vs. 1,695.37±401.58 ml/min, P=0.003, 95%CI-1,839.843--375.571, compared to group 2]. Conclusion Maturation of AVF can be assessed by ultrasonography measurement. The value of D-a2 is an independent factor for AVF maturation. Logistic formula including D-a5, body weight and diabetes can be used to estimate Kt/V in a HD ssession. There exists unmatched situation between AVF maturation and compliance with target Kt/V value.