How well can a masked ultrasound sweep by minimally trained personnel substitute for reference standard scans for the diagnosis of several pregnancy complications?
Masked ultrasound sweeps obtained by personnel with minimal training, transmitted to an off-site interpreting MFM physician, can be used to diagnose abnormal placentation, multiple gestations, abnormal amniotic fluid volume, or noncephalic presentation with reasonable accuracy. This approach may be useful in environments where a registered ultrasound technologist is unavailable, but MFM specialists can be accessed remotely.
Plan de l'etude:
Self-funded or unfunded
This prospective cohort study included pregnant people (N = 168 with a total of 194 sonographic studies) who were referred for obstetric ultrasound during the second and third trimester of pregnancy. In addition to a reference standard diagnostic ultrasound by an ultrasound technician, the pregnant persons were also scanned by minimally trained personnel using the portable Butterfly iQ+ ultrasound system connected to Apple iPads. Reference standard sonography confirmed or excluded 1 of 4 conditions: multiple gestation (n = 52), noncephalic presentation (n = 81), placenta previa (n = 30), or amniotic fluid abnormality (oligo- or polyhydramnios; n = 36). There were 49 ultrasonograms with normal results (control group). Excluded were patients with fetal congenital abnormalities, fetal death, English not the primary language, severe SARS-CoV-2 infection, medical emergency, or who presented at a time of high clinical volume. The minimally trained personnel were medical students who received 3 hours of training, consisting of a 1-hour training video and 2 hours of hands-on training. They learned to perform 8 masked sweeps using the ultrasound probe guided by external body landmarks only. Maternal fetal medicine (MFM) specialists evaluated the images. They were masked to the images and reports from the reference standard diagnostic sonograms. Overall sensitivity for correct diagnosis was 92% (95% CI 87% - 96%). The results for multiple gestation were sensitivity 100%, specificity 95%; and for noncephalic presentation: sensitivity 92%, specificity 84%. Detection rates were lower for low-lying placenta or placenta previa (sensitivity 76%) and abnormal fluid volume (sensitivity 57%). Interrater agreement by 5 MFM specialists was moderate to high overall, but fair for low-lying placenta.
Linda Speer, MD
Professor and Chair, Department of Family Medicine
University of Toledo