Ebonwu Emmanuel O, Nagel Susanna E, Repsold Lisa, Pillay Tahir S
Tshwane University of Technology (TUT), Staatsartillerie Rd, Pretoria-West, Pretoria 0183, South Africa; Department of Chemical Pathology, Faculty of Health Sciences, University of Pretoria and National Health Laboratory Service (NHLS) Tshwane Academic Division, Pretoria, South Africa.
Department of Chemical Pathology, Faculty of Health Sciences, University of Pretoria and National Health Laboratory Service (NHLS) Tshwane Academic Division, Pretoria, South Africa.
Clin Chim Acta. 2020 Nov;510:79-87. doi: 10.1016/j.cca.2020.06.043. Epub 2020 Jul 3.
Many studies have assessed the predictive accuracy of serum osmolality equations. Different approaches for selecting a usable equation were compared using thirty published equations and patient data from a regional hospital laboratory.
Laboratory records were extracted with same-sample results for measured serum osmolality, sodium, potassium, urea and glucose analysed in a regional hospital laboratory between 1/1/2017-31/12/2018. Differences were analysed using Passing-Bablok and difference (Bland-Altman) analysis. Three approaches were compared: the shotgun approach, adjusting for bias, and deriving a novel equation using multivariate analysis. The criteria for success included bias ≤0.7%, a 230 - 400 mOsm/kg range, and osmolal gap (OG) 95% reference limits within ±10 mOsm/kg.
The majority of equations produced proportionally negative-biased results. The shotgun approach identified two equations (EQ19, EQ6) with bias ≤0.7% but unworkable OG reference limits. The bias adjustment approach produced several equations with bias ≤ 0.7% and OG reference limits within or equivalent to ±10 mOsm/kg. A novel equation generated by us (1.89Na + 1.71 K + 1.08 Urea + 1.08 Glucose + 13.7) improved with the adjustment of bias and was not superior to the adjusted published equations.
Few published equations are immediately usable. Adjustment of bias derives several usable equations of which the best had OG ranges <20 mOsm/kg. We conclude that adjustment of bias can generate equations of equal or superior performance to that of novel equations.
许多研究评估了血清渗透压方程的预测准确性。使用30个已发表的方程和一家地区医院实验室的患者数据，比较了选择可用方程的不同方法。
提取了2017年1月1日至2018年12月31日期间在一家地区医院实验室分析的测量血清渗透压、钠、钾、尿素和葡萄糖的同一样本结果的实验室记录。使用Passing-Bablok和差异（Bland-Altman）分析来分析差异。比较了三种方法：散弹枪法、偏差调整法和使用多变量分析推导新方程。成功的标准包括偏差≤0.7%、230 - 400 mOsm/kg范围以及渗透压间隙（OG）95%参考限在±10 mOsm/kg以内。
大多数方程产生了成比例的负偏差结果。散弹枪法确定了两个偏差≤0.7%但OG参考限不可用的方程（EQ19、EQ6）。偏差调整法产生了几个偏差≤0.7%且OG参考限在或等效于±10 mOsm/kg以内的方程。我们生成的一个新方程（1.89钠 + 1.71钾 + 1.08尿素 + 1.08葡萄糖 + 13.7）在偏差调整后有所改善，但并不优于调整后的已发表方程。
很少有已发表的方程可立即使用。偏差调整得出了几个可用方程，其中最佳方程的OG范围<20 mOsm/kg。我们得出结论，偏差调整可以生成性能与新方程相当或更优的方程。
