Department of Laboratory Medicine, Changi General Hospital, Singapore.
Clin Chim Acta. 2013 Jun 25;422:26-8. doi: 10.1016/j.cca.2013.03.034. Epub 2013 Apr 6.
As new high-sensitivity (hs) cardiac troponin (cTn) assays are introduced decision limits, based on the 99th percentile upper reference limit (99 percentile URL), for each method must be determined and from a sufficiently large cohort to mitigate against distortionary effects of high-end outliers. There is a paucity of studies with large multi-ethnic cohorts.
We determined the 99 percentile URL in 1120 (mean age: 50.4±8.2 y) apparently healthy (no history of diabetes, hypertension, heart, lung, or renal disease) Asians (597 men aged 35-65 y, 523 women aged 40-65 y) in a pre-market hs-cTnI assay (Abbott Diagnostics).
Hs-cTnI performance was: limit of blank - 0.6 ng/l, limit of detection (LoD) - 1.5 ng/l; cTnI inter-assay coefficient of variation of 20% and 10% were 1.5 and 6.0 ng/l respectively. Hs-cTnI concentrations (range: 0-49.3 ng/l) were detectable (>assay LoD) in 92.3% of participants, and higher in men and individuals >50 y. All-subject, male, and female 99 percentile URL (90% CI) were 25.6 (19.6-32.6), 32.7 (21.1-47.9) and 17.9 (10.7-26.3) ng/l respectively.
This hs-cTnI assay exhibits hs performance besides gender and age differences. The all-subject 99 percentile URL values are similar to those reported from some groups but not in others. Users need to establish their own decision limits.
随着新型高敏(hs)心肌肌钙蛋白(cTn)检测试剂的问世,每种方法都必须基于第 99 百分位上限(99 percentile URL)确定参考限值,且需基于足够大的队列以减轻高端离群值的扭曲效应。目前,具有多民族大样本队列的研究较为匮乏。
我们在Abbott Diagnostics 的 hs-cTnI 预市场检测中,确定了 1120 例(平均年龄:50.4±8.2 岁)貌似健康(无糖尿病、高血压、心脏、肺部或肾脏疾病史)亚洲人(597 名 35-65 岁男性,523 名 40-65 岁女性)的 99 percentile URL。
hs-cTnI 的检测性能为:空白限-0.6 ng/L,检测下限(LoD)-1.5 ng/L;20%和 10%的批内变异系数分别为 1.5 ng/L 和 6.0 ng/L。92.3%的参与者 hs-cTnI 浓度(范围:0-49.3 ng/L)可被检出(>检测 LoD),且男性和>50 岁的个体浓度更高。所有受试者、男性和女性的 99 percentile URL(90%CI)分别为 25.6(19.6-32.6)、32.7(21.1-47.9)和 17.9(10.7-26.3)ng/L。
该 hs-cTnI 检测试剂除性别和年龄差异外,还具有 hs 性能。所有受试者的 99 percentile URL 值与一些研究组报告的相似,但与其他组的不同。使用者需要建立自己的决策限值。