Preeclampsia: the sFlt-1/PlGF ratio in diagnosis

A pregnant woman at 31 weeks comes to the visit with a blood pressure of 142 over 92 mmHg, some leg swelling and a mild headache. The urine dipstick shows traces of protein. She has no signs of severity, but the team wonders what to do: admit her for observation, discharge her with a close follow-up or order more tests?

This is the territory of suspected pre-eclampsia. The classic criteria, high blood pressure and protein in the urine, are useful but nonspecific, and the disease can worsen quickly. A few years ago, a blood test began to help with this decision: the ratio between two proteins linked to the placenta, sFlt-1 and PlGF.

This text shows how this biomarker changes the probability of pre-eclampsia in the pregnant woman with clinical suspicion, what the cutoffs mean, why it is better for ruling out than for confirming and what its limits are.

Key points

• The clinical diagnosis of pre-eclampsia is imprecise for predicting who will have complications.
• The sFlt-1/PlGF ratio reflects the placental imbalance that precedes the clinical disease.
• A low value rules out pre-eclampsia in the following week with high safety.
• A high value raises the probability, but does not confirm the diagnosis on its own.
• The test changes admission decisions, and a normal result does not last forever.

Why clinical suspicion is so uncertain

Pre-eclampsia is a disease of pregnancy that manifests, usually after 20 weeks, with hypertension and signs of organ damage in the mother. A 2023 review describes an incidence of 2 to 8 per 100 pregnancies and places it among the main causes of maternal death, with about 70 thousand direct maternal deaths per year worldwide (Velegrakis and colleagues, 2023).

The classic criteria and their limits

The clinical diagnosis relies on nonspecific signs, mainly high blood pressure and proteinuria (Stepan and colleagues, 2015). A European consensus document notes that the measurement of proteinuria is imprecise and that complications can appear before proteinuria becomes significant, which led recent guidelines to accept other signs of organ dysfunction.

The same document brings a figure that sums up the problem: the diagnosis based on blood pressure and proteinuria has a positive predictive value of about 30 per 100 for predicting adverse outcomes linked to pre-eclampsia (Stepan and colleagues, 2015). That is, most pregnant women who meet the classic criteria will not have the complications that are feared.

The mimics

Chronic hypertension, gestational hypertension, chronic kidney disease and autoimmune diseases such as lupus can look like pre-eclampsia (Lou and colleagues, 2019). These conditions are also risk factors for the disease, which makes the separation even harder.

In this gray zone, the clinical decision oscillates between two errors: admitting pregnant women who do not need it, with strain and anxiety, or discharging pregnant women who will worsen. This is where a test able to change the probability has value. To understand how, it is worth recalling the role of pretest probability in the interpretation of any test.

What sFlt-1 and PlGF measure

Pre-eclampsia is increasingly understood as a state of imbalance of the factors that regulate vessel formation in the placenta.

The biology in a few lines

PlGF, placental growth factor, is a protein produced by the placenta that helps remodel the uterine arteries and form the network of vessels in the placenta (Creswell and colleagues, 2023). A 2023 review describes that its levels are lower in pregnant women with preterm pre-eclampsia and that this drop precedes the clinical onset of the disease.

sFlt-1 is a soluble form of a receptor that binds to PlGF and neutralizes it. In pre-eclampsia, sFlt-1 rises and PlGF falls (Creswell and colleagues, 2023). The ratio between the two amplifies the difference and works as a thermometer of the placenta's distress.

The two-stage model

The 2023 review of the biomarker revisits the classic two-stage model (Velegrakis and colleagues, 2023). In the first, still without symptoms, the placenta implants poorly and suffers from low oxygenation. In the second, the distressed placenta releases factors into the maternal circulation, and hypertension and organ damage appear.

The sFlt-1/PlGF ratio captures the first stage before the second becomes evident. This explains why it can change in the blood weeks before the disease becomes clear in blood pressure and urine.

The ratio separates similar diseases

A retrospective Chinese study of 118 pregnant women with suspected or diagnosed pre-eclampsia compared the ratio across several conditions (Lou and colleagues, 2019). The ratio was higher in pre-eclampsia than in pregnant women with autoimmune disease and in those with isolated proteinuria, and in early pre-eclampsia it was higher than in gestational and chronic hypertension.

In late pre-eclampsia, however, the ratio was similar to that of gestational hypertension (Lou and colleagues, 2019). This detail already suggests an important limit: the test discriminates early disease better.

The cutoffs and what each one means

In practice, the ratio is not read as a continuous number. It is interpreted in bands, and each band answers a different question.

Rule out, suspect, confirm

The European consensus document organizes the bands (Stepan and colleagues, 2015). Below 38, pre-eclampsia is ruled out for at least 1 week, at any gestational age. Above 85, between 20 and 33 weeks and 6 days, or above 110, from 34 weeks onward, pre-eclampsia or another form of placental insufficiency is very likely.

Between 38 and these high cutoffs lies an intermediate zone, in which the risk of developing the disease in the following weeks is increased (Stepan and colleagues, 2015). These pregnant women need closer follow-up, but the test does not give them the diagnosis.

sFlt-1/PlGF ratio band	Question it answers	Practical reading
Below 38	Will pre-eclampsia appear in the next week?	Very unlikely; routine follow-up may be enough
Between 38 and the high cutoff	Is the risk in the coming weeks increased?	Yes; closer follow-up and reassessment
Above 85 (before 34 weeks)	Is pre-eclampsia likely now?	Very likely; manage according to the local protocol
Above 110 (34 weeks onward)	Is pre-eclampsia likely now?	Very likely; manage according to the local protocol

Table source: Stepan and colleagues, 2015.

Why different cutoffs by gestational age

The ratio rises naturally at the end of pregnancy. Therefore, the same value has a different meaning at 28 and at 36 weeks, and the diagnostic cutoffs are higher after 34 weeks (Stepan and colleagues, 2016). A Spanish study of high-risk pregnant women showed how the median ratio changes over the weeks in those who did not develop the disease (Perales and colleagues, 2017).

Other cutoffs in the literature

Some studies propose their own cutoffs. The Chinese study suggested a value below 21.5 to rule out and above 97.2 to confirm (Lou and colleagues, 2019). A small Iranian study found the best cutoff at 24.96 (Nikuei and colleagues, 2020). This variation is a reminder that the cutoff depends on the assay, the population and the question, and that the report should be read with the assay used by the laboratory in mind.

What accuracy studies show

Ruling out is stronger than confirming

The central point for reasoning is that the test is much better at ruling out than at confirming. This pattern appears in several studies.

The negative predictive value

A multicenter Asian study included pregnant women with suspected pre-eclampsia between 20 and 36 weeks and evaluated 700 of them (Bian and colleagues, 2019). The prevalence of pre-eclampsia was 14.4 per 100. A ratio of up to 38 had a negative predictive value of 98.6 per 100 for ruling out the disease in the following week, with sensitivity of 76.5 per 100 and specificity of 82.1 per 100.

A Japanese subanalysis of the same study notes that the original European study found a negative predictive value of 99.3 per 100 (Ohkuchi et al., 2021). In the Japanese subgroup, with 180 evaluable pregnant women, the negative predictive value reached 100 per 100.

A French study at a referral center followed 67 high-risk pregnant women (Caillon et al., 2018). Of the 53 with a ratio below 38, none developed preeclampsia in the following week.

The positive predictive value

The other side is far more modest. In the Asian study, a ratio above 38 had a positive predictive value of 30.3 per 100 for pre-eclampsia in the following 4 weeks (Bian et al., 2019). In the Japanese subgroup, it was 32.4 per 100 (Ohkuchi et al., 2021).

In the French study, the positive predictive value was 21 out of 100 at 1 week and 18 out of 100 at 4 weeks (Caillon et al., 2018). In other words, out of every 10 pregnant women with a ratio above 38, most will not have the disease during the study period.

How to read this asymmetry

The asymmetry has a simple explanation. The predictive value depends on the base rate of the disease in the group tested. When prevalence is low, a test with good specificity still produces many false positives, and the negative predictive value stays high almost by definition.

In practice, this means that a low value allows the pregnant woman and the team to be reassured safely for about a week. A high value calls for attention, but does not authorize drastic decisions without the clinical picture.

Diagnosis in those who already have the disease

A different question is whether the ratio separates pregnant women with confirmed preeclampsia from those without it. Here the numbers are better, because the contrast is sharper.

Case-control study

A European study compared 178 women with confirmed pre-eclampsia or HELLP syndrome with 391 controls with normal pregnancy (Stepan et al., 2016). The ratio had a sensitivity of 94.0 per 100 and a specificity of 99.4 per 100 in early disease, and a sensitivity of 89.5 per 100 and a specificity of 95.4 per 100 in late disease.

The same study compared the ratio with isolated PlGF dosing, which had much lower specificity in late disease (Stepan et al., 2016). The ratio combines the two pieces of information and therefore gains in specificity.

Watch out for the study design

Case-control studies, which compare clear patients with healthy pregnant women, tend to overestimate accuracy. An Iranian study with 38 pregnant women with pre-eclampsia and 20 controls found an area under the curve of 0.90, sensitivity of 84.2 per 100 and specificity of 85.0 per 100 (Nikuei et al., 2020).

In real life, the test is ordered precisely for pregnant women in whom doubt exists, not for those with an obvious diagnosis. That is why prospective studies with pregnant women under suspicion, such as those described in the previous section, best represent clinical use. This difference between study and practice is a way to avoid overdiagnosis.

Impact on practice and limits of the test

The test changes decisions

A biomarker is only worthwhile if it changes what the team does. Some studies looked directly at this.

Decision to admit

A prospective European study asked physicians to record the intended management before knowing the ratio result and then to confirm or change it (Klein et al., 2016). In 20 of 118 pregnant women, 16.9 per 100, the decision to admit changed after the result.

In 11.0 out of every 100, the initial decision to admit became a decision not to admit, and in 5.9 out of every 100 the opposite happened (Klein et al., 2016). An independent panel considered all the revised decisions appropriate.

Length of stay

A German real-world data analysis compared a hospital that used the ratio to decide with the cohort of a study in which the result was concealed (Dathan-Stumpf and colleagues, 2022). Admissions for confirmed pre-eclampsia were shorter in the hospital that used the test, with a median of 1 day versus 4 days.

The authors note that, in the German cohort of the original study, using the cutoff of 38 to rule out the disease could reduce admissions of pregnant women with suspected preeclampsia from 44.6 to 24.0 per 100 (Dathan-Stumpf et al., 2022). The gains depend, of course, on the service having access to the test and on the team trusting the result.

The limits of the test

Like every biomarker, the sFlt-1/PlGF ratio has limits that must enter the reasoning.

A normal result has an expiration date

The exclusion holds for about a week (Stepan et al., 2015). A 2023 review describes that, in one study, 10.9 per 100 pregnant women with a normal PlGF were diagnosed with preeclampsia later in pregnancy (Creswell et al., 2023). In another series, almost one third of pregnant women with suspected preeclampsia and a normal initial result had an abnormal result when the test was repeated.

The practical consequence is simple: if symptoms persist or change, the test may need to be repeated. An old result does not protect against a new disease.

What the test does not do

• It does not replace blood pressure measurement or the assessment of signs of severity.
• It does not decide on its own when delivery should happen.
• It does not perform the same at every gestational age.
• It is not available in every service, and the cutoffs depend on the assay.

The clinical sign remains sovereign

A pregnant woman with very high blood pressure, severe headache, visual changes, upper abdominal pain or altered blood tests needs immediate assessment, regardless of the ratio. The test helps in the gray zone, not in the pregnant woman who already has signs of severity. Clinical decision-making combines both types of information.

Screening is not diagnosing

It is worth separating two questions that are often mixed up. One is screening, early in pregnancy, for who is more likely to develop preeclampsia. The other is deciding, in the pregnant woman who has already arrived with a suspicion, whether the disease is there or will appear in the coming weeks. PlGF enters both conversations, but with different roles. In screening between 11 and 13 weeks, it forms part of a competing risks model that combines the a priori risk from maternal history with the uterine artery pulsatility index, mean arterial pressure and PlGF, and the result is an individual risk, not a diagnosis (Creswell et al., 2023). In the pregnant woman with symptoms after 20 weeks, PlGF-based tests serve to speed up diagnosis and estimate the time to delivery, as the same review shows (Creswell et al., 2023).

Confusing these two situations leads to predictable errors. A high risk in first-trimester screening does not mean the pregnant woman has preeclampsia; it means she belongs to a group that deserves closer follow-up. Likewise, a normal ratio at 30 weeks says nothing about what the screening was like at the start. Each test answers the question it was designed for, and clinical reasoning begins by knowing which question is being asked.

Serial measurements in high-risk pregnant women

A multicenter Spanish study followed pregnant women with a risk factor for preeclampsia, such as preeclampsia or growth restriction in a previous pregnancy, preexisting diabetes or altered uterine Doppler, and measured the ratio at 20, 24 and 28 weeks, before any symptoms (Perales et al., 2017). Of the 729 women analyzed, 78 developed preeclampsia, 24 of them with early onset (Perales et al., 2017). At 28 weeks, the median ratio was 51.9 in the women who would have early preeclampsia and 3.3 in those who did not develop the disease, while in the late forms the values were close to those of the group without the disease (Perales et al., 2017). The finding reinforces the biology of the two-stage model: in early forms, angiogenic imbalance comes before hypertension and proteinuria.

But history is not a diagnostic criterion. The authors themselves built a model that added mean arterial pressure, parity and previous pre-eclampsia to the ratio, and that model was better than the ratio alone (Perales et al., 2017). The review on PlGF notes that the value of repeating the test still needs to be explored in prospective studies (Creswell et al., 2023). For the student, the lesson is not to turn a single number into a diagnosis of a disease that has not yet manifested.

Different populations, different cutoffs

The most cited cutoffs came from European cohorts, and studies in other populations tested whether they hold. In a Japanese subanalysis of pregnant women between 18 and 36 weeks with suspected disease, a ratio of 38 or less had a negative predictive value of 100 per 100 for ruling out pre-eclampsia within one week, with a prevalence of 13.3 per 100 (Ohkuchi et al., 2021). In the same group, a ratio above 38 had a positive predictive value of 32.4 per 100 for the disease within four weeks (Ohkuchi et al., 2021). Even so, local validation matters, because prevalence changes predictive values, as discussed in the text on base rate. A test with the same technical performance has different predictive values in a low-risk outpatient clinic and in a referral maternity hospital.

Building the reasoning step by step

Suspicion of pre-eclampsia can be organized in a sequence.

1. Measure blood pressure correctly and look for proteinuria and other signs of organ damage.
2. Look for signs of severity, which call for immediate assessment without waiting for tests.
3. Without signs of severity and with uncertainty, consider the sFlt-1/PlGF ratio if available.
4. Ratio below 38: the disease in the following week is very unlikely; combine a return visit and warning signs.
5. Intermediate ratio: closer follow-up and reassessment.
6. Ratio above the high cutoff: pre-eclampsia very likely; manage according to the local protocol.
7. Repeat the assessment if symptoms persist or change.

Back to the pregnant woman in the case

The pregnant woman at 31 weeks has slightly high blood pressure, traces of protein and no sign of severity. The pre-test probability of pre-eclampsia is intermediate. If the ratio comes back below 38, the chance of the disease appearing in the following week is very low (Bian et al., 2019), and she can have close outpatient follow-up, with clear guidance on warning signs.

If the ratio comes back above 85, pre-eclampsia is very likely at this gestational age (Stepan et al., 2015), and management follows the service protocol. In any case, the reasoning should be recorded in the medical record, with the date of the test and the reassessment plan.

How NEXO helps train this reasoning

In NEXO, the case of the day is fictional, with a real clinical presentation and up to six progressive clues, and you propose the diagnosis by searching for the ICD-10 code.

The limits of using the game are described on the NEXO Economy page. If you research teaching of clinical reasoning, the page for Researchers explains how to get in touch. Professionals who write or review cases find information in Curator and in Creators, and institutions can learn about the proposal on the partnerships page. Training differential diagnosis with clues and searching for the right ICD-10 code complete this training.

Frequently asked questions

Does the sFlt-1/PlGF ratio diagnose pre-eclampsia?

Not on its own. It changes the probability, and the diagnosis remains clinical. The test is stronger for ruling out than for confirming (Stepan et al., 2015).

What does a ratio below 38 mean?

That pre-eclampsia in the following week is very unlikely, with a negative predictive value of 98.6 per 100 in an Asian multicenter study (Bian et al., 2019).

And a ratio above 38?

The risk in the following weeks is increased, but most pregnant women will not have the disease in the period. The positive predictive value was around 30 per 100 (Bian et al., 2019).

Does the test need to be repeated?

It may need to be. A normal result is valid for about one week, and some pregnant women with a normal result have an abnormal result when retested (Creswell et al., 2023).

Why do cutoffs change with gestational age?

Because the ratio rises naturally at the end of pregnancy. That is why the diagnostic cutoff is 85 before 34 weeks and 110 after (Stepan et al., 2015).

Take-home messages

• Blood pressure and proteinuria are useful criteria, but they are not very precise for predicting complications.
• The sFlt-1/PlGF ratio reflects placental distress before clinical disease.
• Below 38, preeclampsia in the following week is very unlikely.
• Above 38, the risk increases, but most women will not have the disease in that period.
• The test changes admission decisions and can reduce unnecessary admissions.
• Signs of severity call for immediate assessment, whatever the ratio.

References

1. Bian X, Biswas A, Huang X, et al. Short-Term Prediction of Adverse Outcomes Using the sFlt-1 (Soluble fms-Like Tyrosine Kinase 1)/PlGF (Placental Growth Factor) Ratio in Asian Women With Suspected Preeclampsia. Hypertension. 2019. PMID 31188674.
2. Caillon H, Tardif C, Dumontet E, et al. Evaluation of sFlt-1/PlGF Ratio for Predicting and Improving Clinical Management of Pre-eclampsia: Experience in a Specialized Perinatal Care Center. Ann Lab Med. 2018. PMID 29214752.
3. Creswell L, O'Gorman N, Palmer KR, et al. Perspectives on the Use of Placental Growth Factor (PlGF) in the Prediction and Diagnosis of Pre-Eclampsia: Recent Insights and Future Steps. Int J Womens Health. 2023. PMID 36816456.
4. Dathan-Stumpf A, Rieger A, Verlohren S, et al. sFlt-1/PlGF ratio for prediction of preeclampsia in clinical routine: A pragmatic real-world analysis of healthcare resource utilisation. PLoS One. 2022. PMID 35202416.
5. Klein E, Schlembach D, Ramoni A, et al. Influence of the sFlt-1/PlGF Ratio on Clinical Decision-Making in Women with Suspected Preeclampsia. PLoS One. 2016. PMID 27243815.
6. Lou WZ, Jiang F, Hu J, et al. Maternal Serum Angiogenic Factor sFlt-1 to PlGF Ratio in Preeclampsia: A Useful Marker for Differential Diagnosis and Prognosis Evaluation in Chinese Women. Dis Markers. 2019. PMID 31396294.
7. Nikuei P, Rajaei M, Roozbeh N, et al. Diagnostic accuracy of sFlt1/PlGF ratio as a marker for preeclampsia. BMC Pregnancy Childbirth. 2020. PMID 32033594.
8. Ohkuchi A, Saito S, Yamamoto T, et al. Short-term prediction of preeclampsia using the sFlt-1/PlGF ratio: a subanalysis of pregnant Japanese women from the PROGNOSIS Asia study. Hypertens Res. 2021. PMID 33727707.
9. Perales A, Delgado JL, de la Calle M, et al. sFlt-1/PlGF for prediction of early-onset pre-eclampsia: STEPS (Study of Early Pre-eclampsia in Spain). Ultrasound Obstet Gynecol. 2017. PMID 27883242.
10. Stepan H, Herraiz I, Schlembach D, et al. Implementation of the sFlt-1/PlGF ratio for prediction and diagnosis of pre-eclampsia in singleton pregnancy: implications for clinical practice. Ultrasound Obstet Gynecol. 2015. PMID 25736847.
11. Stepan H, Hund M, Gencay M, et al. A comparison of the diagnostic utility of the sFlt-1/PlGF ratio versus PlGF alone for the detection of preeclampsia/HELLP syndrome. Hypertens Pregnancy. 2016. PMID 27028698.
12. Velegrakis A, Kouvidi E, Fragkiadaki P, et al. Predictive value of the sFlt‑1/PlGF ratio in women with suspected preeclampsia: An update (Review). Int J Mol Med. 2023. PMID 37594116.