Combs C Andrew, Lee Ryan C, Lee Sarah Y, Amara Sushma, Ashimi Balogun Olaide
Pediatrix Center for Research, Education, Quality & Safety, Sunrise, FL 33323, USA.
Obstetrix of California, Los Gatos, CA 95032, USA.
J Clin Med. 2024 Nov 16;13(22):6895. doi: 10.3390/jcm13226895.
: Systematic quality review of ultrasound exams is recommended to ensure accurate diagnosis. Our primary objectives were to develop a quantitative method for quality review of estimated fetal weight (EFW) and to assess the accuracy of EFW for an entire practice and for individual personnel. A secondary objective was to evaluate the accuracy of fetal sex determination. This is a retrospective cohort study. Eligible ultrasound exams included singleton pregnancies with live birth and known birth weight (BW). A published method was used to predict BW from EFW for exams with ultrasound-to-delivery intervals of up to 12 weeks. Mean error and median absolute error (AE) were compared between different personnel. Image audits were performed for exams with AE > 30% and exams with reported fetal sex different than newborn sex. We analyzed 1938 exams from 890 patients. In the last exam before birth, the median AE was 5.9%, and the predicted BW was within ±20% of the actual BW in 97.2% of patients. AE was >30% in 28 exams (1.4%); image audit found correct caliper placement in all 28. Only two patients (0.2%) had AE > 30% on the last exam before birth. One sonographer systematically over-measured head and abdominal circumferences, leading to EFWs that were overestimated. Reported fetal sex differed from newborn sex in seven exams (0.4%) and five patients (0.6%). Images in four of these patients were annotated with the correct fetal sex, but a clerical error was made in the report. In one patient, an unclear image was labeled "probably female", but the newborn was male. The accuracy of EFW in this practice was similar to literature reports. The quantitative analysis identified a sonographer with outlier measurements. Time-consuming image audits could be focused on a small number of exams with large errors. We suggest some enhancements to ultrasound reporting software that may help to reduce clerical errors. We provide tools to help other practices perform similar quality reviews.
建议对超声检查进行系统的质量评估，以确保准确诊断。我们的主要目标是开发一种对估计胎儿体重（EFW）进行质量评估的定量方法，并评估整个医疗机构以及个体工作人员的EFW准确性。次要目标是评估胎儿性别判定的准确性。这是一项回顾性队列研究。符合条件的超声检查包括单胎活产且已知出生体重（BW）的妊娠。对于超声检查至分娩间隔长达12周的检查，采用已发表的方法从EFW预测BW。比较了不同工作人员之间的平均误差和中位数绝对误差（AE）。对AE>30%的检查以及报告的胎儿性别与新生儿性别不同的检查进行图像审核。我们分析了来自890名患者的1938次检查。在出生前的最后一次检查中，中位数AE为5.9%，97.2%的患者预测BW在实际BW的±20%范围内。28次检查（1.4%）的AE>30%；图像审核发现这28次检查的卡尺放置均正确。只有两名患者（0.2%）在出生前的最后一次检查中AE>30%。一名超声检查人员系统性地测量头围和腹围过大，导致EFW被高估。7次检查（0.4%）和5名患者（0.6%）报告的胎儿性别与新生儿性别不同。其中4名患者的图像标注了正确的胎儿性别，但报告中出现笔误。在一名患者中，一张不清楚的图像被标记为“可能是女性”，但新生儿是男性。该医疗机构中EFW的准确性与文献报告相似。定量分析识别出一名测量值异常的超声检查人员。耗时的图像审核可以集中在少数误差较大的检查上。我们建议对超声报告软件进行一些改进，这可能有助于减少笔误。我们提供工具以帮助其他医疗机构进行类似的质量评估。
