Division of Neonatology, Department of Pediatrics, The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania, USA,
Division of Neonatology, Department of Pediatrics, The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania, USA.
Neonatology. 2019;116(3):260-268. doi: 10.1159/000501005. Epub 2019 Jul 19.
We aimed to determine the accuracy and validity of the Ultrasonic Cardiac Output Monitor (USCOM) measurements of cardiac output (CO) compared to echocardiography in newborn infants, and the inter-rater agreement of USCOM measurements.
In a single-center study we prospectively evaluated neonates undergoing an echocardiographic evaluation. USCOM measurements of CO were obtained at the pulmonary and aortic valve by 2 physicians blinded to the echocardiographic results. All echocardiographic measurements were performed blinded to USCOM measurements. We first enrolled an ascertainment cohort which was subsequently validated in an independent new cohort. Agreement between echocardiography and USCOM methods was assessed by Bland-Altman analysis. Intra-class correlation coefficients (ICC) assessed the agreement between the 2 operators. The ascertainment cohort correction factors were applied in a second validation cohort and agreement of the calibrated measures evaluated with repeat Bland-Altman comparisons.
A total of 50 infants were enrolled in the initial cohort and 15 in the validation cohort. There was a high degree of correlation between the USCOM operators (ICC = 0.975). USCOM measurements of CO were significantly higher compared to echocardiography (left ventricular output bias 95 ± 52 mL/kg/min and right ventricular output bias 64 ± 30 mL/kg/min). There was no difference in the subgroup of infants with and without a ductus arteriosus. After the correction was applied to the validation cohort, there was no longer a significant difference between the measures.
CO measured by USCOM consistently overestimated the results obtained from echocardiography. USCOM is not adequate to provide absolute estimates of CO. However, it may allow longitudinal hemodynamic assessment of sick neonates.
本研究旨在比较超声心输出量监测仪（USCOM）与超声心动图测量新生儿心输出量（CO）的准确性和有效性，并评估 USCOM 测量的组内一致性。
在一项单中心前瞻性研究中，我们对接受超声心动图评估的新生儿进行了评估。两名对超声心动图结果不知情的医生在肺动脉瓣和主动脉瓣处使用 USCOM 测量 CO。所有超声心动图测量均在不知道 USCOM 测量值的情况下进行。我们首先纳入了一个确定队列，随后在一个独立的新队列中进行了验证。通过 Bland-Altman 分析评估超声心动图和 USCOM 方法之间的一致性。采用组内相关系数（ICC）评估两名操作者之间的一致性。将确定队列的校正因子应用于第二个验证队列，并通过重复 Bland-Altman 比较评估校准测量值的一致性。
共有 50 名婴儿纳入初始队列，15 名婴儿纳入验证队列。两名 USCOM 操作者之间具有高度相关性（ICC = 0.975）。与超声心动图相比，USCOM 测量的 CO 明显更高（左心室输出偏差 95 ± 52 ml/kg/min，右心室输出偏差 64 ± 30 ml/kg/min）。在有和没有动脉导管未闭的婴儿亚组中，没有差异。在验证队列中应用校正后，测量值之间不再存在显著差异。
USCOM 测量的 CO 始终高估了超声心动图获得的结果。USCOM 不能提供 CO 的绝对估计值。然而，它可能允许对患病新生儿进行纵向血流动力学评估。
