Background Children require reference intervals distinct from adults and, based on their unique developing physiology, several intervals are often required to accurately reflect the distribution of expected results in a healthy pediatric population. The establishment of pediatric reference intervals has proven challenging. While consenting adults volunteer to contribute their biological material towards the development of a reference interval, well children usually do not have the occasion to get blood drawn and the smaller the child, the more significant the impact of phlebotomy. For these reasons, scientists have sought out other methods for determining pediatric reference intervals. As an alternative, indirect statistical methods may be applied to large data sets of laboratory test results to ascertain a reference interval. The Multi-Modal Decomposition (MMD) is an iterative indirect method that decomposes a mixture of multiple normal distributions into separate components using the expectation-maximization (EM) algorithm The objective of the current study, was to apply MMD to ascertain reference intervals for free T4 in infants. Methods The study population included infants ranging in age from 1 day to 60 days who had specimens submitted for free T4 testing involving equilibrium dialysis followed by LC-MS/MS in a commercial reference laboratory (Quest Diagnostics Nichols Institute, San Juan Capistrano, CA). MMD was performed on 25,271 de-identified free T4 results to establish the reference intervals that were validated by comparison against free T4 values obtained on 238 de-identified specimens submitted for acylcarnitine testing, the specimen were analyzed for TSH and only the in-range specimens were include for Free T4 reference interval analysis. Results MMD analysis demonstrated distinct reference intervals for the following ages: 0 to 6 days (1.8 - 6.1 ng/dL), 7 days to <2 weeks (1 - 4.4 ng/dL), 2 to <3 weeks (0.8 - 3.5 ng/dL), 3 to <4 weeks (0.8 - 3 ng/dL), and 4 to <8 weeks (0.7 - 2.8 ng/dL). The data did not support the use of separate intervals for male and female children. Conclusions MMD analysis demonstrated distinct reference intervals for the following ages: 0 to 6 days (1.8 - 6.1 ng/dL), 7 days to <2 weeks (1 - 4.4 ng/dL), 2 to <3 weeks (0.8 - 3.5 ng/dL), 3 to <4 weeks (0.8 - 3 ng/dL), and 4 to <8 weeks (0.7 - 2.8 ng/dL). The data did not support the use of separate intervals for male and female children.

中文翻译：

---

A-082 通过间接方法建立婴儿游离 T4 参考区间


背景 儿童需要与成人不同的参考区间，并且根据他们独特的发育生理机能，通常需要几个区间来准确反映健康儿科人群中预期结果的分布。事实证明，建立儿科参考区间具有挑战性。虽然同意的成年人自愿为参考区间的发展贡献他们的生物材料，但健康儿童通常没有机会抽血，而且孩子越小，放血的影响就越大。由于这些原因，科学家们一直在寻找其他方法来确定儿科参考区间。作为替代方案，间接统计方法可以应用于实验室测试结果的大型数据集，以确定参考区间。多模态分解 （MMD） 是一种迭代间接方法，它使用期望最大化 （EM） 算法将多个正态分布的混合物分解为单独的分量。当前研究的目的是应用 MMD 来确定婴儿游离 T4 的参考区间。方法 研究人群包括 1 天至 60 天不等的婴儿，他们在商业参考实验室（Quest Diagnostics Nichols Institute，San Juan Capistrano，CA）提交标本进行免费 T4 检测，包括平衡透析，然后进行 LC-MS/MS。对 25,271 个去标识化的游离 T4 结果进行 MMD，以确定参考区间，这些参考区间通过与 238 个提交酰基肉碱检测的去标识化标本获得的游离 T4 值进行比较来验证，对标本进行 TSH 分析，并且仅包括范围内的标本进行游离 T4 参考区间分析。 结果 MMD 分析显示以下年龄的不同参考区间：0 至 6 天 （1.8 - 6.1 ng/dL）、7 天至 <2 周 （1 - 4.4 ng/dL）、2 至 <3 周 （0.8 - 3.5 ng/dL）、3 至 <4 周 （0.8 - 3 ng/dL） 和 4 至 <8 周 （0.7 - 2.8 ng/dL）。数据不支持对男孩和女孩儿童使用单独的间隔。结论 MMD 分析显示以下年龄存在不同的参考区间： 0 至 6 天 （1.8 - 6.1 ng/dL）、7 天至 <2 周 （1 - 4.4 ng/dL）、2 至 <3 周 （0.8 - 3.5 ng/dL）、3 至 <4 周 （0.8 - 3 ng/dL） 和 4 至 <8 周 （0.7 - 2.8 ng/dL）。数据不支持对男孩和女孩儿童使用单独的间隔。