Role of soluble fms-like tyrosine kinase-1/placental growth factor ratio as an early predictor of preeclampsia in pregnant women

INTRODUCTION

Preeclampsia (PE), a multisystem disorder unique to pregnancy, complicates 2–8% of gestations worldwide and is a major driver of maternal mortality (~60,000 deaths/year) and perinatal loss (~500,000 deaths/year).^[1,2] The only definitive treatment is delivery of the ischemic placenta, as untreated PE can lead to multiorgan failure (e.g., hepatic, renal, and cerebral) and adverse outcomes such as placental abruption, eclampsia, hemolysis, elevated liver enzymes, low platelet count syndrome, and preterm birth.^[1,2] Fetal complications include growth restriction, hypoxia, and intrauterine demise.^[3] PE is the second leading cause of maternal mortality (14%), according to the World Health Organization.^[4]

Diagnostic criteria for PE include new-onset hypertension (systolic/diastolic blood pressure ≥140/90 mmHg) manifesting after 20 weeks’ gestation, accompanied by either proteinuria (≥300 mg/24 h) or end-organ dysfunction in the absence of proteinuria.^[5] Traditional biomarkers lack sensitivity, especially in the presence of chronic hypertension or renal disease.^[6] Incidence is higher in developing nations (2.8% vs. 0.4%).^[7]

Risk factors include genetic predisposition, maternal age >40, obesity, chronic hypertension, diabetes, multifetal gestation, in vitro fertilization, and a family history of PE.^[7,8]

PE is characterized by an imbalance between pro-angiogenic factors (vascular endothelial growth factor [VEGF] and placental growth factor [PlGF]) and the anti-angiogenic factor soluble fms-like tyrosine kinase-1 (sFlt-1).^[9,10] PlGF, produced by trophoblasts, peaks at around 30 weeks in normal pregnancies but declines early in PE.^[11,12] sFlt-1, a VEGF/PlGF antagonist, rises in PE and contributes to endothelial dysfunction.^[13]

This study aims to comprehensively evaluate the role of the sFlt-1 to PlGF ratio in predicting the early onset and severity of PE in pregnant women.

MATERIALS AND METHODS

A prospective hospital-based cohort study was conducted over 20 months in the Department of Biochemistry in collaboration with Obstetrics and Gynecology at North Eastern Indira Gandhi Regional Institute of Health and Medical Sciences (NEIGRIHMS) hospital, involving 150 primigravida women aged 18–49 years (gestational age 20–36 weeks) with clinical suspicion of PE. Exclusion criteria included confirmed PE/HELLP syndrome (defined as proteinuria ≥2+ or BP ≥140/90 mmHg), multiple pregnancies, diabetes mellitus, or autoimmune disorders. Using consecutive sampling, participants underwent detailed clinical evaluation and blood collection (5 mL venous blood at initial visit and second visit follow-up) after informed consent being taken for analysis of sFlt-1/PlGF ratio (measured through Roche Cobas Pro [e801+c503], Roche Diagnostics, Mannheim, Germany), liver/kidney function, lipid profile, uric acid, coagulation parameters, and platelet count (analyzed using Beckman Coulter AU5800, Beckman Coulter Inc., United States of America), with quality control maintained through Roche internal quality controls, Randox internal controls and association of clinical biochemist of India, Christian Medical College external controls.

Using nMaster 2.0 software, we determined a minimum sample size of 150 participants to achieve 89% specificity for the sFlt-1/PlGF ratio, with a 5% margin of error at 95% confidence intervals (CI). The study was conducted after approval of the Institution Ethics Committee of NEIGRIHMS (NEIGR/IEC/M12/T1/2023).

RESULTS

A total of 150 pregnant women were enrolled, of whom 25 developed PE – 16 with mild and 9 with severe disease. The preeclamptic group (PE) was significantly older (30.68 ± 2.61 years vs. 25 ± 3.46 years; P < 0.001) and had a higher period of gestation (32 ± 2.1 weeks vs. 30.5 ± 2.33 weeks; P = 0.027) when compared to non-preeclamptic controls (NPE). Body mass index (BMI) did not differ significantly between groups (P = 0.112). However, both systolic and diastolic blood pressures were significantly elevated in the PE group (133.44 ± 11.07 mmHg and 90.56 ± 6.21 mmHg) compared to the NPE group (112.77 ± 10.2 mmHg and 81.82 ± 5.13 mmHg; P < 0.001) [Table 1].

Receiver operating characteristic (ROC) analysis identified optimal sFlt-1/PlGF cut-off ratios of 21.35 (visit 1) and 35 (visit 2), with excellent diagnostic accuracy, for the first cut-off: Sensitivity = 96%, specificity = 97.6%, positive predictive value (PPV) = 88.89%, negative predictive value (NPV) = 99.19%, and area under the curve (AUC) = 0.994 and for the second: Sensitivity = 92%, specificity = 99.2%, PPV = 95.83%, NPV = 98.41%, and AUC = 0.983 [Figure 1]. These findings support the use of serial sFlt-1/PlGF ratio measurements in early detection and risk stratification of PE.

Close

Figure 1:

Cut off soluble fms-like tyrosine kinase-1/placental growth factor (sFlt-1/PlGF) ratio (First Visit)=21.35; area under the curve (AUC): 0.994; Sensitivity: 96%; Specificity: 97.6% Cut Off sFlt-1/PlGF ratio (Second visit)=35; AUC: 0.983; Sensitivity: 92%; Specificity: 99.2%. ROC: Receiver-operating characteristic curve.

Export to PPT

sFlt-1 levels were significantly higher in the PE group (7906.96 pg/mL) versus NPE (3906.06 pg/mL), while PlGF levels were markedly reduced (233.01 pg/mL vs. 752.31 pg/mL; P < 0.001). Consequently, the sFlt-1/PlGF ratio was substantially elevated in PE participants. At the first visit, the ratio was 41.42 in PE versus 8.4 in NPE, and at the second visit, it rose to 56.54 versus 12.69 (both P < 0.001), indicating disease progression and diagnostic value.

These findings were corroborated by significantly elevated 24-h urine protein levels in the PE group (3510.92 mg vs. 607.36 mg; P < 0.001), emphasizing renal involvement [Table 2].

Serum uric acid and total cholesterol were significantly higher in PE cases (8.7 mg/dL and 190.64 mg/dL) compared to NPE (4.52 mg/dL and 125.44 mg/dL; P <0.001). Liver enzymes (aspartate transaminase, alanine transaminase, and alkaline phosphatase) were also elevated in PE, supporting hepatic involvement. Platelet counts were significantly lower in PE (176 × 10³ vs. 209.67 × 10³; P = 0.017.

sFlt-1/PlGF ratio across 2 time point measurements shows that in the first ratio measurement, the PE group exhibited a mean ratio of 47.97, compared to just 3.52 in NPE, with a highly significant P< 0.01. The second ratio measurement showed an even more pronounced difference, with the PE group reaching 65.25 versus 5.41 in NPE, again with a highly significant P< 0.01.

The statistical analyses, including repeated measures analysis of variance (ANOVA) and mixed-methods ANOVA, demonstrated significant changes over time. The total sFlt-1/PlGF ratio increased from 11.85 to 16.63 across measurements, with PE groups consistently showing substantially higher ratios. The Bonferroni post hoc tests confirmed statistically significant differences between PE and NPE groups [Table 3].

These results strongly suggest that the sFlt-1/PlGF ratio is not only a sensitive marker for PE but also shows a progressive change that could potentially be used for early detection and monitoring of PE complications during pregnancy.

Logistic regression analysis revealed that sFlt-1/PlGF ratio 1 significantly predicted both mild (P = 0.021; odds ratio [OR] = 1.874) and severe PE (P = 0.008; OR = 2.353), whereas ratio 2 lacked predictive significance [Table 4a and b]. These findings highlight the diagnostic and prognostic value of early sFlt-1/PlGF ratio measurements.

Moderately positive correlations were observed between sFlt-1/PlGF ratios and clinical parameters: diastolic BP (r = 0.28; P = 0.001), systolic BP (r = 0.38; P < 0.001), and 24-h urine protein (r = 0.39; P < 0.001), further supporting the role of angiogenic imbalance in PE pathophysiology.

DISCUSSION

Preeclampsia (PE), a hypertensive disorder of pregnancy, remains a leading cause of maternal and fetal morbidity and mortality, highlighting the need for early and accurate prediction.^[14] Its complex pathophysiology involves endothelial dysfunction, systemic inflammatory responses, and dysregulation of angiogenic balance, particularly through elevated soluble fms-like tyrosine kinase-1 (sFlt-1)—an anti-angiogenic factor inhibiting VEGF and PlGF—and reduced placental growth factor (PlGF), a pro-angiogenic molecule essential for placental development. The sFlt-1/PlGF ratio has emerged as a validated biomarker that correlates with disease severity and adverse outcomes. Elevated ratios can precede the onset of clinical symptoms, enabling risk stratification, differentiation from other hypertensive disorders, and timely intervention.^[15-17] In our study, maternal age was significantly higher among women with PE, aligning with Duckitt and Harrington, who identified maternal age over 30 years as a significant risk factor, likely attributable to age-related vascular and placental changes.^[18] The slightly higher gestational age at presentation agrees with Bodnar et al.^[19] with evidence that the condition often arises in the late second or early third trimester as a result of progressive placental dysfunction.^[19] Elevated blood pressure in the PE group reinforces hypertension as a defining feature of PE, consistent with Roberts and Hubel’s findings linking it to endothelial dysfunction and increased vascular resistance.^[20] Body mass index showed no significant association with PE in this cohort, differing from observations in other populations, likely due to limited BMI variability and a relatively homogenous sample.^[21] sFlt-1 levels were significantly elevated in the PE group, while PlGF levels were substantially lower. These results are consistent with the findings of Zitouni et al.,^[22] who reported that PE is characterized by reduced PlGF, elevated sFlt-1/PlGF ratios, and additional hormonal and genetic alterations, further supporting the concept of a characteristic angiogenic imbalance in PE.^[22]

sFlt-1/PlGF ratio was markedly higher in women who developed PE compared to those who did not, even at the first measurement, underscoring its potential utility as an early screening tool. Our data showed a clear elevation of the sFlt-1/PlGF ratio in PE cases between the first and second visits, whereas non-preeclamptic controls exhibited only a minor increase. This pattern aligns with studies by Verlohren et al.^[23] and Rana et al.,^[24] who reported higher ratios in PE and progressive changes over time, although the magnitude and timing of changes vary between studies.

The concurrent rise in blood pressure, proteinuria, and biochemical markers of renal and hepatic stress in our PE group reinforces this mechanistic link, consistent with earlier reports.^[25,28] The moderate correlations between the ratio and systolic/diastolic BP and proteinuria suggest that while related, the biomarker provides independent information that may precede over clinical manifestations.

Our findings also show agreement with prior literature on associated biochemical changes in PE, including elevated uric acid,^[26] dyslipidemia,^[27] and reduced platelet counts,^[29] supporting the concept of PE as a multisystem disorder.

Logistic regression revealed that a higher sFlt-1/PlGF ratio at the first visit was significantly associated with mild preeclampsia, indicating nearly double the risk per unit increase. This supports the role of angiogenic imbalance in preeclampsia pathogenesis, aligning with studies of Teixeira et al.^[30] and Stubert et al.^[31] who reported elevated sFlt-1 and reduced PlGF levels in affected individuals.^[30,31] However, the ratio measured at the second visit did not show statistically significant predictive power in either model, highlighting the superiority of the first measurement in predicting disease severity, as observed by Rolnik et al.^[32] By the later visit, the ratio may plateau or become influenced by interventions, reducing its additional predictive value.^[32] Clinically, the very high diagnostic accuracy at our first-visit cut-off (21.35) indicates that incorporating sFlt-1/PlGF testing between 20– 28 weeks in high-risk or symptomatic women could facilitate earlier risk stratification and more targeted monitoring. The slightly reduced predictive strength at the second visit suggests that early measurement, rather than repeated late-pregnancy testing, may be the most cost-effective strategy, particularly in resource-limited settings.

A positive correlation was observed between the sFlt-1/PlGF ratio and 24-hour urine protein levels, consistent with findings by Barton et al.,^[33] who noted that proteinuria increases with disease progression but typically manifests later. Similarly, Dong et al.^[34] reported greater PE severity in cases with proteinuria exceeding 0.3 g/L.^[33,34]

The sFlt-1/PlGF ratio also showed a moderate positive correlation with both systolic and diastolic blood pressure, further supporting its role as a predictive biomarker for preeclampsia. This finding is in agreement with earlier research that highlighted its diagnostic specificity and association with adverse outcomes.^[23,35]

ROC curve analysis demonstrated excellent diagnostic performance of the sFlt-1/PlGF ratio for predicting preeclampsia. At the initial visit, the ratio exhibited near-perfect discrimination, with very high sensitivity, specificity, and predictive values. These results are consistent with previous studies validating its utility for early pregnancy prediction. During follow-up, the ratio maintained high accuracy and reliability across gestation. The slightly lower performance at later stages may reflect evolving disease mechanisms, but overall, the diagnostic value remained robust.^{[23,24,36,37]}

Based on our findings, we recommend implementing sFlt-1/PlGF testing between 20 and 28-week gestation, as this early measurement can effectively identify women who would require more intensive monitoring. For women presenting with suspected PE, a ratio above 21 could serve as an objective criterion to guide clinical decisions regarding hospitalization and delivery timing, potentially improving maternal and fetal outcomes.

Despite strong associations, this study has limitations, including a relatively small sample size and limited generalizability.

CONCLUSIONS

In this cohort of pregnant women with suspected preeclampsia, the sFlt-1/PIGF ratio was markedly higher in those who developed the condition compared to those who did not, at both time points. However, the relative increase in the ratio between visits was proportionally greater in the non-preeclamptic group than in the preeclamptic group, indicating that progression patterns differ between populations. While absolute values remained much higher in preeclampsia, these findings suggest that a single elevated early ratio may be more clinically relevant for diagnosis than changes over time. A cut-off of 21.35 at the first visit provided excellent diagnostic accuracy and may be valuable for early risk stratification and targeted monitoring. Nevertheless, the small sample size, limited follow-up, and intergroup variability in ratio change limits generalizability. Larger, multi-center studies are required to refine cut-offs, clarify progression patterns, and confirm the prognostic value of serial measurements across diverse populations.