The Tests That Told Us Nothing: Estriol, L/S Ratios, X-Ray Pelvimetry, and Weekly Herpes Cultures
We spent millions on tests that didn’t predict what we needed to know. The evidence said so. We kept ordering them.
Every week in the third trimester, we drew estriol levels. We believed the numbers reflected placental function and fetal health. We plotted them on graphs. We worried when they fell. We intervened when they dropped.
The test had poor sensitivity and poor specificity. It did not predict fetal compromise reliably. It did not improve outcomes. It was abandoned when biophysical assessment and nonstress testing proved superior. But for a decade or more, it consumed significant laboratory resources and generated clinical decisions based on numbers that meant very little.
Estriol was not an isolated case. We performed X-ray pelvimetry on primigravidas, exposing mothers and fetuses to ionizing radiation to measure pelvic diameters that did not predict the need for cesarean delivery. We performed amniocentesis for lecithin/sphingomyelin ratios before any delivery to assess fetal lung maturity, a practice rendered unnecessary by universal corticosteroid administration and improved gestational age dating. We cultured for herpes simplex virus weekly in the third trimester, a practice shown to have no predictive value for viral shedding at delivery. We performed routine third-trimester ultrasound in low-risk pregnancies, evaluated in the landmark RADIUS trial, which found no improvement in perinatal outcomes.
Six testing practices. Each ordered routinely. Each expensive. Each unsupported by evidence of clinical benefit.
The common thread is not that these tests were technically flawed. Some measured what they claimed to measure. The problem was that accurate measurement of the wrong thing does not improve patient care.
🎯 Free Subscriber Bottom Line: Six routine testing practices in obstetrics, including serial estriol monitoring, X-ray pelvimetry, L/S ratio testing, weekly herpes cultures, routine third-trimester ultrasound, and routine amniocentesis for advanced maternal age, were all abandoned after failing to improve clinical outcomes. Each consumed resources and generated clinical decisions based on tests that did not predict what mattered.
Below, paid subscribers get: - The RADIUS trial: why routine ultrasound in low-risk pregnancy didn’t help - Arvin and Prober’s NEJM studies that ended weekly herpes cultures - The evolution from amniocentesis to cell-free DNA screening - A test-by-test analysis with sensitivity, specificity, replacement, and year abandoned - What today’s testing practices might look like on a future list - The cost analysis: what we spent on tests that didn’t work.
