Zoppi F, De Gasperi A, Guagnellini E, Marocchi A, Mineo E, Pazzucconi F, Rossi C, Turrini D
Clinical Biochemistry and Haematology Laboratory, Niguarda Cà Granda Hospital, Milan, Italy.
Scand J Clin Lab Invest Suppl. 1996;224:259-74. doi: 10.3109/00365519609088646.
Only free magnesium has biological activity: technology for measuring the ionized fraction of magnesium is now available via ion-selective electrodes. We have evaluated an instrument (AVL 988/4) which determines ionized magnesium (cMg2+) with an ion-selective electrode based on the ionophore ETH 7025. The selectivity of the electrode is adequate for the ions normally present in plasma, except for calcium: the interference is automatically corrected by simultaneous measurements of calcium with compensation for the calcium interference to the magnesium signal. We have first verified possible interference caused by sampling procedures: known silicon interference has been avoided by use of glass tubes (BD Vacutainer with no additive, code 7626); heparin interference has been measured and found significant above 20 UI.mL-1 of plasma. Instrument evaluation according to NCCLS protocol gives the following imprecision results on 20 replicated analyses: cMg2+ (mmol.L-1) 1.29, 0.76, 0.23, CVs% (within-run) 0.67, 0.67, 3.00 and CVs% (between-run) 4.06, 3.91, 5.89 respectively. Linearity (in the range 0.23-1.60 mmol.L-1) was: measured cMg2+ = 0.981.(calculated cMg2+) + 0.009 mmol/L; r = 0.999. In healthy adults (n = 103) cMg2+ was in the range 0.46-0.74 mmol.L-1 (with a mean of 0.60 mmol/L and normal distribution). These values represent 57% to 84% of serum total magnesium concentration (TMg) (mean 71%). pH dependence of cMg2+ is present, usually to a lower extent with respect to cCa2+, but it seems different in patients with real or in vitro provoked acidosis and in hemodialyzed patients. Citrate interference on ionized magnesium measurements was found both in vitro and in vivo, whilst that due to lactate was demonstrated only in vitro. On a wide range of cMg2+ (n = 100), a good correlation is obtained both with TMg and ultrafiltrable Mg (UFMg): cMg2+ = 0.723.TMg + 0.008 mmol.L-1, r = 0.978; cMg2+ = 0.912.UFMg + 0.10 mmol.L-1, r = 0.968, respectively. The ionized magnesium in ultrafiltrate was found 25% lower than that in serum. The lifespan of the electrode, evaluated on the basis of both time from installation and on number of measured samples, was estimated longer than 4 months and able to analyze more than 1500 samples, whichever comes first. The four electrodes we used during 18 months behaved all the same way. The correlation between measurements performed in whole blood (WB-cMg2+) and in the corresponding serum (S-cMg2+) was excellent: WB-cMg2+ = 0.954.S-cMg2+ +0.02 mmol.L-1; r = 0.998; n = 60.
现在可通过离子选择电极测量镁的离子化部分。我们评估了一种仪器(AVL 988/4),它基于离子载体ETH 7025的离子选择电极测定离子化镁(cMg2+)。除钙外,该电极对血浆中通常存在的离子选择性足够:通过同时测量钙并补偿钙对镁信号的干扰自动校正干扰。我们首先验证了采样程序可能造成的干扰:使用玻璃管(无添加剂的BD Vacutainer,代码7626)避免了已知的硅干扰;已测量肝素干扰,发现在血浆浓度高于20 UI.mL-1时具有显著干扰。根据NCCLS方案进行仪器评估,在20次重复分析中得出以下不精密度结果:cMg2+(mmol.L-1)1.29、0.76、0.23,批内CVs%分别为0.67、0.67、3.00,批间CVs%分别为4.06、3.91、5.89。线性(在0.23 - 1.60 mmol.L-1范围内)为:测量的cMg2+ = 0.981×(计算的cMg2+)+ 0.009 mmol/L;r = 0.999。在健康成年人(n = 103)中,cMg2+在0.46 - 0.74 mmol.L-1范围内(平均值为0.60 mmol/L,呈正态分布)。这些值占血清总镁浓度(TMg)的57%至84%(平均71%)。cMg2+存在pH依赖性,通常相对于cCa2+程度较低,但在实际或体外诱发酸中毒的患者以及血液透析患者中似乎有所不同。在体外和体内均发现柠檬酸盐对离子化镁测量有干扰,而乳酸盐引起的干扰仅在体外得到证实。在广泛的cMg2+范围(n = 100)内,与TMg和超滤镁(UFMg)均具有良好的相关性:cMg2+ = 0.723×TMg + 0.008 mmol.L-1,r = 0.978;cMg2+ = 0.912×UFMg + 0.10 mmol.L-1,r = 0.968。发现超滤物中的离子化镁比血清中的低25%。根据安装后的时间和测量样本数量评估,电极寿命估计超过4个月,能够分析超过1500个样本,以先到者为准。我们在18个月内使用的四个电极表现均相同。全血(WB - cMg2+)和相应血清(S - cMg2+)中测量值之间的相关性极佳:WB - cMg2+ = 0.954×S - cMg2+ + 0.02 mmol.L-1;r = 0.998;n = 60。