用于延长呼出一氧化氮分析研究的最佳流速采样设计。

Optimal flow rate sampling designs for studies with extended exhaled nitric oxide analysis.

作者信息

Molshatski Noa, Eckel Sandrah P

机构信息

Department of Preventive Medicine, University of Southern California, Los Angeles, CA, United States of America.

出版信息

J Breath Res. 2017 Feb 22;11(1):016012. doi: 10.1088/1752-7163/aa5ad0.

DOI:10.1088/1752-7163/aa5ad0

PMID:28104897

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC5421556/

Abstract

INTRODUCTION

The fractional concentration of exhaled nitric oxide (FeNO) is a biomarker of airway inflammation. Repeat FeNO maneuvers at multiple fixed exhalation flow rates (extended NO analysis) can be used to estimate parameters quantifying proximal and distal sources of NO in mathematical models of lower respiratory tract NO. A growing number of studies use extended NO analysis, but there is no official standard flow rate sampling protocol. In this paper, we provide information for study planning by deriving theoretically optimal flow rate sampling designs.

METHODS

First, we reviewed previously published designs. Then, under a nonlinear regression framework for estimating NO parameters in the steady-state two compartment model of NO, we identified unbiased optimal four flow rate designs (within the range of 10-400 ml s) using theoretical derivations and simulation studies. Optimality criteria included NO parameter standard errors (SEs). A simulation study was used to estimate sample sizes required to detect associations with NO parameters estimated from studies with different designs.

RESULTS

Most designs (77%) were unbiased. NO parameter SEs were smaller for designs with: more target flows, more replicate maneuvers per target flow, and a larger range of target flows. High flows were most important for estimating alveolar NO concentration, while low flows were most important for the proximal NO parameters. The Southern California Children's Health Study design (30, 50, 100 and 300 ml s) had ≥1.8 fold larger SEs and required 1.1-3.2 fold more subjects to detect the association of a determinant with each NO parameter as compared to an optimal design of 10, 50, 100 and 400 ml s.

CONCLUSIONS

There is a class of reasonable flow rate sampling designs with good theoretical performance. In practice, designs should be selected to balance the tradeoffs between optimality and feasibility of the flow range and total number of maneuvers.

摘要

引言

呼出一氧化氮分数浓度（FeNO）是气道炎症的生物标志物。在多个固定呼气流量下重复进行FeNO操作（扩展NO分析）可用于在数学模型中估计量化下呼吸道NO近端和远端来源的参数。越来越多的研究使用扩展NO分析，但尚无官方标准流量采样方案。在本文中，我们通过推导理论上最优的流量采样设计为研究规划提供信息。

方法

首先，我们回顾了先前发表的设计。然后，在用于估计NO稳态两室模型中NO参数的非线性回归框架下，我们通过理论推导和模拟研究确定了无偏最优四流量设计（在10 - 400 ml/s范围内）。最优标准包括NO参数标准误差（SEs）。使用模拟研究来估计检测与不同设计研究中估计的NO参数的关联所需的样本量。

结果

大多数设计（77%）是无偏的。对于具有以下特征的设计，NO参数SEs较小：更多的目标流量、每个目标流量更多的重复操作以及更大的目标流量范围。高流量对于估计肺泡NO浓度最为重要，而低流量对于近端NO参数最为重要。与10、50、100和400 ml/s的最优设计相比，南加州儿童健康研究设计（30、50、100和300 ml/s）的SEs大≥1.8倍，并且检测决定因素与每个NO参数的关联需要多1.1 - 3.2倍的受试者。