Estriol

If you are pregnant and want to know whether your body is on track or drifting toward early labor, this test offers a window using only spit. It measures a hormone your baby's own development produces, so it reflects how the pregnancy is progressing.

The catch is that this is a research-grade marker, best read as a trend across weeks rather than a single verdict. Used that way, it can flag a rising risk of preterm birth before symptoms appear. Major obstetric societies, including the American College of Obstetricians and Gynecologists (ACOG), consider salivary estriol investigational because of its high false-positive rate, and none currently recommend it for routine prenatal care.

What This Test Actually Measures

The hormone here is E3 (estriol), one of the main estrogens and, in pregnancy, the dominant one. During pregnancy it is made almost entirely by the baby and placenta working together, using building blocks from the fetal adrenal gland and liver, rather than by the salivary glands themselves. So a saliva reading is really a readout of how the pregnancy's shared hormone factory is running.

Saliva captures only the free, unbound form of the hormone, the fraction that is biologically active. In pregnancy, most estriol travels in blood in a bound, inactive form, and the free portion is under 15% of the total. That is why this test reflects the active hormone rather than the whole circulating pool.

This free saliva fraction tracks the free hormone in blood reasonably well, with matched saliva and blood samples in pregnant women lining up at correlations between about 0.79 and 0.95 (where 1.0 would be a perfect match). The test measures salivary free estriol concentration directly; what it signals is fetoplacental activity, which it reflects but does not measure outright.

Preterm Birth Risk

The clearest use of this marker is gauging the risk of spontaneous preterm labor and birth. In late pregnancy, estriol tends to surge, and in women who deliver early that surge arrives roughly four weeks sooner than in those who go to term.

Sources: Heine et al. 2000 (Obstetrics & Gynecology); Heine et al. 1999 (AJOG); Karunakaran et al. 2023 (Biomedicine).

What this means for you: this is a marker that shines when you watch it move, not when you read it once. A single number is easy to misread, but a repeated pattern that climbs faster than expected carries real signal. One early trial found that crossing a study threshold caught about 71 out of 100 women who later had preterm labor while correctly clearing about 77 out of 100 who did not.

There is an apparent contradiction worth resolving here. Studies report cutoffs that vary widely from one lab to the next, and single readings alone often lack useful accuracy, yet serial testing performs better. The resolution is that this is not a simple high-equals-bad number. It is a trajectory marker, most trustworthy as a trend and most useful for ruling out near-term risk, where its ability to correctly reassure runs high. Even so, its false-positive rate is high enough that ACOG classifies the test as investigational, so it belongs alongside other tools rather than as a stand-alone screen. The test was FDA-cleared only for single-baby pregnancies.

Preeclampsia

Estriol can also move in the opposite direction when the placenta is struggling. In severe preeclampsia, a serious blood-pressure disorder of pregnancy, circulating estriol runs lower than in healthy pregnancy, alongside broader disruption of how the body makes and clears estrogens. This evidence comes from blood measurements rather than saliva, so it is a related but different sample type.

This is why estriol is not a marker where high is simply bad or low is simply good. It reads out the activity of the pregnancy's hormone system, and different problems push it in different directions. A premature surge can precede early labor, while a drop can reflect a placenta that is failing to keep up, as in severe preeclampsia.

Early Pregnancy and Fetal Wellbeing

Estriol appears in saliva remarkably early. One study detected it from the sixth week of pregnancy, with sharp rises between weeks 7 and 8 and again between weeks 10 and 11. From the second trimester onward, it climbs steadily with gestational age, mirroring the well-established pattern seen in blood.

Because the hormone comes from the baby and placenta, it has long been used as an index of fetal wellbeing. In at least one reported case, a saliva-based assay charted a fetal demise the same way a blood-based assay did, suggesting the two can track major changes comparably.

Estriol Beyond Pregnancy

Outside pregnancy, estriol sits inside broader estrogen-metabolism networks that have been tied to disease, but this evidence is preliminary and mostly not based on saliva. A large follow-up study found that higher pregnancy estrogens in the 16-hydroxylation pathway, including estriol, were modestly linked to later hormone-receptor-negative breast cancer, with roughly 11% higher odds per doubling of the metabolite. Those measurements were made in blood, not saliva, so they describe a related molecule pattern rather than this specific test.

Researchers have also proposed that estrogens and the mouth's bacterial community interact in ways that could matter for oral cancer, but this remains a hypothesis rather than an established human association. For now, the practical, evidenced use of salivary estriol stays firmly in pregnancy.

Why One Reading Is Not Enough

Salivary estriol naturally bounces around, which is the strongest reason to track it rather than trust a single value. In one classic study, repeated samples varied by roughly 17% from day to day, about 23% from hour to hour, and about 13% within ten-minute windows. Some of that is real biology, and some is how the sample was collected.

Timing adds another layer. While one study found stable levels across the waking day, another found a pronounced nighttime rise peaking around 4:00 am and settling back by early morning. Sampling at the same time of day, during daytime hours, keeps readings comparable.

The takeaway is to build a trajectory. In pregnancy research, weekly sampling from around 22 weeks, or looking at the rise across a 2 to 4 week interval, was far more informative than any one-off measurement. A sensible cadence is a baseline, then repeated readings on a regular schedule so you can see the direction, not just a dot.

When Results Can Be Misleading

Because this is spit, what is in your mouth matters. Eating or drinking right before collection can depress the measured value sharply, with immediate drops of about 19% after food and about 32% after water in one study. Collecting on an empty mouth, rinsing beforehand, and waiting after meals reduce this.

• Time of day and posture: samples taken overnight or first thing on waking while lying down are more variable, so consistent daytime, seated collection is more reliable.
• Oral contamination: blood from gum bleeding or gingivitis, plus food, nicotine, toothpaste, and mouthwash, can distort the reading; visibly discolored samples can produce outlier values.
• Corticosteroids: the steroid betamethasone, often given to help fetal lungs mature, suppresses maternal salivary estriol by about 23% after each weekly dose before rebounding. This is a temporary measurement effect, not a sign of fetoplacental disease.
• Assay differences: older immunoassays can drift at low concentrations, so results from different labs or methods may not be directly comparable.

What an Unexpected Result Should Prompt

A surprising reading is a reason to look wider, not to act alone. If your estriol is rising faster than expected, the next steps are to repeat the measurement under standardized conditions and to bring it to your obstetric clinician, who can pair it with other preterm-risk tools such as cervical length ultrasound, fetal fibronectin, or a confirmatory serum unbound estriol.

Patterns matter more than any single number. A rising trend combined with symptoms of preterm labor and a short cervix is a very different picture than an isolated high reading in a woman who feels well, which usually warrants watchful monitoring rather than intervention. This marker earns its place as one input in a larger workup, guided by a licensed obstetric clinician, not as a stand-alone decision-maker.