Assessment of serum levels and placental bed tissue expression of IGF-1, bFGF, and PLGF in patients with placenta previa complicated with placenta accreta spectrum disorders

Abstract

Objective

This study aims to detect the serum levels of IGF-1, bFGF, and PLGF and their expressions in placental bed tissues of patients with placenta previa complicated with PAS disorders.

Methods

This case and control study included 40 multiparous pregnant women with complete placenta previa between 34 weeks and 38 weeks of gestation and they were divided into two groups: 25 patients with PAS (case group) and 15 patients without PAS (control group). The venous blood samples were collected 2 h before the cesarean section, and the placental bed tissues were taken intraoperatively at the placental implantation site and then were histologically examined to evaluate the gravity of the myometrial invasion of the placenta. According to FIGO PAS increasing grading, the 25 patients were also divided into three groups: PAS grade I group, PAS grade II group, and PAS grade III group. The concentrations of IGF-1, bFGF, and PLGF in serum were measured using ELISA, and the mean ratio of the relative mRNA expression of each biomarker in placental bed tissues was calculated using qRT-PCR. The staining intensity and the positive cells were quantitatively measured and expressed as means by using Image J software for IHC analysis.

Results

IGF-1 had low serum levels and high placental bed expression in placenta previa patients with PAS disorders compared to those without PAS (all p < 0.0001). PLGF had high serum levels (p = 0.0200) and high placental bed expression (p < 0.0001) in placenta previa patients with PAS disorders compared to those without PAS. IGF-1 serum levels decreased up to PAS grade II (means were 24.3 ± 4.03, 21.98 ± 3.29, and 22.03 ± 7.31, respectively for PAS grade I, PAS grade II, PAS grade III groups, p = 0.0006). PLGF serum levels increased up to PAS grade II (means were 12.96 ± 2.74, 14.97 ± 2.56, and 14.89 ± 2.14, respectively for the three groups, p = 0.0392). However, IGF-1 and PLGF mRNA placental bed expression increased up to PAS grade III. The relative expression of mRNA means for the three groups was 3.194 ± 1.40, 3.509 ± 0.63, and 3.872 ± 0.70, respectively for IGF-1; and 2.784 ± 1.14, 2.810 ± 0.71, and 2.869 ± 0.48, respectively for PLGF (all p < 0.0001). Their IHC (immunohistochemical) staining also had increasing trends, but p > 0.05. bFGF was not significantly expressed in placenta previa with PAS disorders in most of the analysis sections (p > 0.05).

Conclusions

Low serum levels and high expression in placental bed tissues of IGF-1, or high serum levels and high expression in placental bed tissues of PLGF, may differentiate placenta previa patients with FIGO PAS grade I and PAS grade II from those without PAS disorders. However, they could not significantly predict the degree of placental invasiveness in FIGO PAS grades II and III.

Keywords:

• IGF-1
• bFGF
• PLGF
• placenta previa with accreta
• PAS

Introduction

Placenta previa with accreta is one of the most severe pregnancy complications associated with high maternal morbidity rising to 60% and mortality up to 7% of women with PAS (placenta accreta spectrum) disorders [1], which are currently linked with the upward rate of cesarean sections, placenta previa, IVF (in vitro-fertilization) method, and smoking [2–5]. Placenta previa with placenta accreta arising respectively from abnormal implantation and placentation [6,7], is often related to massive and life-threatening perinatal bleeding complications because of the aberrant positioning of the placenta into the uterine cavity [7,8]. Under normal circumstances, the normal implantation and placentation [9], together with the normal development of the placental villous blood vessels and normal trophoblast invasion contribute to the maintenance of the pregnancy [10,11]. This process is regulated by both the balance of angiogenic growth factors (VEGF and PLGF) and anti-angiogenic factors (sFlt-1and Eng), and the oxygen partial pressure [12]. In some pathological cases, the equilibrium mechanism of angiogenesis is off-slope, and new blood vessel growth is out of control [9,13]. The breakdown of this balance leads to pathological pregnancies, such as PAS disorders [14], preeclampsia, and IUGR (Intrauterine growth restriction) pregnancy, etc. [9,13,15]. PLGF (Placental growth factor) as a pro-angiogenic factor and sFlt-1 as an anti-angiogenic factor may contribute effectively to the development and the maturing of placental vessels throughout the PAS neovasculature system [16]. In addition, in contrast to VEGF, PLGF, and sFlt-1 the main actors of angiogenesis [16,17], there are many other growth factors, such as IGF-1 (Insulin-like growth factor 1) that play an important role in the process of angiogenesis by affecting the equilibrium of new blood vessels originating from endothelial cells [18]. When blood vessels grow up to a certain period, the endothelial cells are in a static state and no new blood vessels grow [19,20]. This growth and development of new placental blood vessels are of great significance to the vigorous growth of the embryo and the successful continuation of the pregnancy [10,21]. IGF-1 as one member of the insulin-like growth factors family (IGF-1, IGF-2, and IGFBP-1), plays an important role in fetal and placental growth and development throughout pregnancy [15], promoting placental and fetal tissue metabolism, mitosis, anti-apoptotic, and cell differentiation [22]. IGF-1 with its structure comparable to insulin and with its metabolic properties like insulin is situated on the short arm of chromosome 12 and has 70 amino acids with a comparative molecular mass of 7.6 kD [23]. It was previously found to be significantly abnormal in women with placenta previa with PAS disorders [24,25]. bFGF (basic fibroblast growth factor) also known as FGF-2, is another strong mitogenic and angiogenic factor that controls the mechanical paths of the uterus and the placental vascular bed in pregnancy [26]. It has been also stated that bFGF is a stronger angiogenic factor than VEGF in enhancing angiogenesis and endothelial cell proliferation [27]. However, the global influence of IGF-1, bFGF, and PLGF biomarkers in patients with placenta previa with PAS disorders has not been fully clarified; hence, it needs further investigation in this manuscript.

Thus, we aimed to detect the serum levels of IGF-1, bFGF, and PLGF, to explore their expressions in placental bed tissues of patients with complete placenta previa complicated with PAS disorders; and compare the serum levels differences and their placental bed tissue expression differences in patients with complete placenta previa without PAS disorders. This may provide new insights into the diagnosis of PAS disorders, which is still challenging in current obstetrics. This study may also provide data beneficial to biomarkers screening for PAS disorders prediction in obstetrics clinics in the future.

Methods

1. Study design and population

This case and control study included 40 multiparous Chinese pregnant women with complete placenta previa between 34 weeks and 38 weeks of gestation. We made the diagnosis of complete placenta previa by color Doppler ultrasound and clinical diagnosis between September 2021 and December 2022 (according to complete placenta previa guidelines from ACOG (American Congress of Obstetricians and Gynecologists)). We collected the venous blood samples 2 h before the cesarean section, and the placental bed tissue sampling was taken intraoperatively cautiously at the placental implantation site and was then histologically examined directly by a proficient pathologist to evaluate the gravity of the myometrial invasion of the placenta in different grades of PAS disorders. According to the results of placental implantation site tissue biopsy and clinical diagnosis during surgery, all 40 pregnant women were divided into two groups: 25 patients with PAS disorders (case group) and 15 patients without PAS disorders (control group). According to FIGO PAS classification [5,16,28,29], the 25 patients with PAS disorders were also divided into three groups: 7 patients with FIGO PAS grade I, 12 patients with FIGO PAS grade II, and 6 patients with FIGO PAS grade III.

The patients were between 20 and 45 years old after excluding early pregnancy, multiple pregnancies gestational hypertension, preeclampsia, gestational diabetes, IUGR, placental dysfunction, oligohydramnios, fetal malformations, pregnancies with blood, immune and cardiovascular diseases, maternal infections, and any other chronic diseases (diabetes, hypertension, kidney diseases, thyroid diseases, etc.). All pregnant women signed the informed consent before being included in the study.

This study was approved by the Ethical Committee of Union Hospital, Tongji Medical College, Huazhong University of Science and Technology.

2. Experimental methods

(1) ELISA (Enzyme-linked immunosorbent assay)

A. Blood sample collection

The maternal blood samples were collected from the Department of Obstetrics and Gynecology, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology. 5 ml fasting venous blood from women with complete placenta previa in the last trimester of pregnancy was collected 2 h before cesarean section in an EDTA tube, and it was centrifuged and stored in a −80 ° C refrigerator before the experiment.

B. ELISA test protocol

The blood samples were taken from the refrigerator and we waited for the natural coagulation at room temperature for 10–20 min and centrifuged for about 20 min (2000–3000 RPM). After collecting the supernatant carefully, we set standard and sample holes, standard holes with different concentrations of the standard according to the experiment protocol (Wuhan HYcezmbio Co. LTD). After, we set three repeat wells for each sample, and measured the absorbance (OD value) of each hole in sequence at 450 nm wavelength. The concentrations of IGF-1, bFGF, and PLGF were recorded three times, and then the mean of each biomarker was calculated.

(2) qRT-PCR (quantitative real-time PCR) and IHC (Immunohistochemistry)

A. Placental tissue specimens’ collection

The maternal placental bed samples were collected from the Department of Obstetrics and Gynecology, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology. After delivery of the placenta, during cesarean section, the discarded placental bed tissues of 1.5 cm × 1.5 cm were taken from the superficial myometrial layer at the placental implantation site. Specimens were rinsed repeatedly with PBS (phosphate-buffered saline) or 0.9% normal saline and each specimen was divided into two parts: one part was fixed with 4% formaldehyde and then embedded in paraffin for IHC staining, and the other part was stored in −80 °C refrigerator or frozen with liquid nitrogen for mRNA extraction.

B. qRT-PCR test protocol

We added placental bed tissues with the size of 0.5 cm × 0.5 cm in a 2 ml EP tube, and then the total mRNA extraction and the cDNA synthesis by reverse transcription were performed systematically according to the experiment protocol (Wuhan Servicebio Co. LTD). The collected cDNA products were stored in the refrigerator at −20 °C for qRT-PCR reaction use. The Primer sequence used for the qRT-PCR reaction process is shown in the following table:

Table

Target gene	Primer sequence
IGF-1	F 5′-ACATTGCTCTCAACATCTCCCA-3′; R 5′-TGGTGTGCATCTTCACCTTCA-3′
bFGF	F 5′-GAAGAGCGACCCTCACATCAA −3′; R5′-ACGGTTAGCACACACTCCTTT -3′
PLGF	F 5′-GCTCGTCAGAGGTGGAAGTG-3′; R 5′-CAGCAGGGAGACACAGGATG-3′
BA	F 5′-CCTTCCTGGGCATGGAGTC-3′; R 5′-TGATCTTCATTGTGCTGGGTG-3′

Abbreviations: IGF-1: insulin-like growth factor-1; bFGF: basic fibroblast growth factor; PLGF: placental growth factor; BA: β-actin; F: forward (upstream) primer; R: reverse (downstream) primer.

On the ice, the reaction system was prepared according to the same experiment protocol (Wuhan Servicebio Co. LTD). After, we set it at 95 °C × 10 s, 60 °C × 30 s, and 72 °C × 30 s, a total of 40 cycles. Three repeat wells were set up for each sample, and the 2-ΔΔCt value was selected to represent the mRNA expression level of each target gene by BA (β-actin).

C. IHC test protocol

We used the kits from Wuhan Servicebio Co. LTD. Tissue sections were dewaxed to water, immersed in citrate buffer (pH 6.0), heated for 15 min, then cooled naturally, and rinsed sections with PBS. Endogenous peroxidase was removed by soaking the sections in a 3% hydrogen peroxide solution for about 20 min. Sections were incubated with 10% goat serum at room temperature for about 1h. Antibodies (Anti-IGF-1 Rabbit pAb for IGF-1, Anti-bFGF Rabbit pAb for bFGF, Anti-PLGF Rabbit pAb for PLGF) were incubated primarily and placed at room temperature for about 1h. The same antibodies were incubated secondly and placed at room temperature for about 30 min. We wiped off the excess PBS on the section, then added DAB (diaminobenzidine) color development agent and placed it under the microscope for observation. After the sections were re-dyed with hematoxylin for about 5 min, and then washed under running water; 1% hydrochloric alcohol. The sections were used for color separation for a few seconds and then washed under running water. The sections were soaked in blue return solution for about 5 min, and then in flow underwater rinse. The sections were dehydrated with anhydrous ethanol, then sealed with neutral resin, and then under a microscope observed and collected images. In the end, the IHC images were quantitatively measured and expressed as means by using Image J software. The staining intensity was obtained by counting the total blue granules, while the positive cells total count/field was obtained by counting the total brown-yellow granules of each image.

3. Statistical analysis

All data presented in this manuscript were expressed as means ± SD. GraphPad Prism 9.0 and SPSS 26.0 statistical software tools were used. For normally distributed variables, the significance of the differences between the two sets of data was compared using unpaired T-test, and the comparison between three or more sets of data was compared using one-way analysis of variance (ANOVA). p < 0.05 indicated statistically significant differences between groups. The IHC images were quantitatively measured and expressed as means by using Image J software. The relative mRNA expression and the staining intensity in placental bed tissues of the three biomarkers were also statistically set as fold-change, control-1.

Results

1. Baseline characteristics of the two groups

In our study, we recruited 40 patients with complete placenta previa in the last trimester of pregnancy. According to the results of placental tissue histology and clinical diagnosis at the time of surgery, all 40 pregnant women were divided into two groups: 25 patients with PAS disorders (case group) and 15 patients without PAS disorders (control group) (Table 1).

Table 1. Maternal and fetal demographic and clinical features/outcomes in patients with complete placenta previa with PAS disorders and without PAS disorders.

Demographic and clinical variables	PAS	Non-PAS	P-value
	N = 25 (62.5%)	N = 15 (37.5%)
Age (years)	34.92 ± 4.02	31.27 ± 4.67	0.014
<35 years	9 (36%)	11 (73.33%)	0.048
≥35 years	16 (64%)	4 (26.67%)	0.048
Weight (kg)	71.08 ± 6.52	64.31 ± 6.04	0.023
Height (cm)	161.48 ± 4.66	161.47 ± 5.25	0.993
BMI (kg/m^2)	27.56 ± 2.77	24.64 ± 1.75	0.0007
GA at admission (weeks)	34.55 ± 1.53	35.54 ± 1.56	0.055
Gravida (G)	3.96 ± 1.77	2.07 ± 0.96	0.0004
Parita (P)	1.16 ± 0.85	0.47 ± 0.52	0.007
Prior cesarean section (CS)	1.12 ± 0.78	0.2 ± 0.41	0.0001
0 CS	4 (16%)	13 (86.67%)	0.000
1 CS	16 (64%)	2 (13.33%)	0.000
≥ 2 CS	5 (20%)	0 (0%)	0.000
Prior diagnostic/curettage (D/C)	1.72 ± 1.31	0.6 ± 0.74	0.004
0 D/C	5 (20%)	8 (53.33%)	0.058
1 D/C	10 (40%)	5 (33.33%)	0.058
≥2 D/C	10 (40%)	2 (13.33%)	0.058
Conceived through IVF	3 (12%)	0 (0%)	0.279
History of smoking	5 (20%)	1 (6.67%)	0.381
Prior placenta previa	1 (4%)	0 (%)	0.625
Prior uterine adhesions	3 (12%)	1 (6.67%)	0.516
Vaginal bleeding	11 (44%)	10 (66.67%)	0.204
Cervical length (cm)	3.73 ± 0.92	3.84 ± 0.69	0.696
Placental thickness above the cervix (cm)	3.752 ± 1.78	2.713 ± 1.08	0.048
Placental length from the cervix (cm)	4.232 ± 1.96	2.987 ± 1.57	0.043
GA at the termination of pregnancy (weeks)	35.25 ± 1.33	36.17 ± 1.41	0.046
Bleeding amount during CS (ml)	1100 ± 959.06	433.33 ± 238.05	0.012
< 100 ml    (non-PPH)	15 (60%)	14 (93.33%)	0.03
≥ 100 ml (PPH)	10 (40%)	1 (6.67%)	0.03
Blood transfusion	12 (48%)	2 (13.33%)	0.04
Autotransfusion	14 (56%)	1 (6.67%)	0.002
Transfusion products
RBC (u)	1.2 ± 2.29	0.27 ± 0.70	0.134
Plasma (ml)	108 ± 293.92	0 ± 0.00	0.165
Autotransfusion (ml)	266.2 ± 310.58	33.33 ± 129.10	0.008
Duration of the surgery (minutes)	125.81 ± 71.64	73.93 ± 14.80	0.008
Post-surgery hemoglobin (g/l)	95.68 ± 13.20	101.67 ± 16.80	0.217
Post-surgery hospitalization days	7.4 ± 1.73	6.06 ± 0.80	0.008
Newborn weight (g)	2353 ± 568.06	2742 ± 381.30	0.024
Newborn APGAR score
At 0 min	7.6 ± 1.89	8.47 ± 0.64	0.095
At 5 min	8.56 ± 2.04	9.54 ± 0.52	0.079
Admission to NICU	14 (56%)	3 (20%)	0.046

Abbreviations: PAS: Placenta accreta spectrum; IVF: in-vitro fertilization; GA: gestational age; PPH: post-partum hemorrhage; RBC: red blood cells; NICU: neonatal intensive care unit.

Note: All data presented in this table were expressed as means ± SD.

We found that there was a clear correlation between PAS disorders and the following patients’ variables: age, weight, BMI, gravida, parity, prior cesarean section, prior diagnostic/curettage, placental thickness above the cervix, placental length from the inner os of the cervix, GA (gestational age) at the termination of pregnancy, bleeding amount during cesarean section, blood transfusion, autotransfusion, duration of the surgery, post-surgery hospitalization days, newborn weight, and admission to NICU. There were significant statistical differences in all above-mentioned variables (p < 0.05), while there were no statistical significances for the rest of the variables (Table 1). Our experiment results have also found that IGF-1serum levels in the non-PAS and PAS groups were 29.4 ± 4.43 and 22.65 ± 4.61, respectively, p < 0.0001; PLGF serum levels were 12.38 ± 2.44 and 14.39 ± 2.58, respectively for the two groups, p = 0.0200; while bFGF serum levels were not significantly different between the two groups (p = 0.063 > 0.05), their means were 13.71 ± 1.96 and 12.38 ± 2.20, respectively for the two groups (Table 2). The mRNA relative expression levels of IGF-1/βA, bFGF/βA, and PLGF/βA were significantly different between the two groups, p < 0.0001 for all the three biomarkers, their means were 1.000 ± 0.64 vs 3.527 ± 0.70, 1.000 ± 0.42 vs 2.120 ± 0.36, and 1.000 ± 0.38 vs 2.817 ± 0.79, respectively for the non-PAS and PAS groups (Table 2). Concerning the IHC staining, the staining intensity fold-change of IGF-1, bFGF, and PLGF was also significantly different between non-PAS group and PAS group, p = 0.018, p = 0.033, and p = 0.026, respectively; their means were 1.000 ± 0.10 vs 1.070 ± 0.08, 1.000 ± 0.11 vs 1.072 ± 0.10, and 1.000 ± .010 vs 1.060 ± 0.07, respectively for the two groups; while the positive cells total count/field was not significantly different between the two groups, p > 0.05 for all of the three biomarkers (Table 2). These results imply that the staining intensity fold-change obtained by counting the total blue granules of each image was significantly expressed in PAS group compared to non-PAS group, while the positive cells count/field obtained by counting the total brown-yellow granules was not significantly expressed in PAS group compared to non-PAS group (Figure 1).

Figure 1. IGF-1, bFGF, and PLGF placental bed tissue expression at 400x magnification in complete placenta previa patients without PAS disorders (control group) and with PAS disorders (case group). (A) IGF-1 expression in placental bed tissues of patients without PAS disorders (CPP). (B) IGF-1 expression in placental bed tissues of patients with PAS disorders. (C) bFGF expression in placental bed tissues of patients without PAS disorders. (D) bFGF expression in placental bed tissues of patients with PAS disorders. (E) PLGF expression in placental bed tissues of patients without PAS disorders. (F) PLGF expression in placental bed tissues of patients with PAS disorders. (M) represents the myometrium. (V) represents the villous trophoblastic cells. (Black arrows) indicate the blue-colored granules, representing the staining intensity of the nucleus. (Red arrows) indicate the brown-yellow granules, representing the positive cells.

Table 2. Serum levels and placental bed tissue expression of IGF-1, bFGF, and PLGF in patients with complete placenta previa with PAS disorders and without PAS disorders.

Variables	PAS	Non-PAS	P-value
	N = 25 (62.5%)	N = 15 (37.5%)
Serum level (ELISA)
IGF-1 (5–160 ng/mL)	22.65 ± 4.61	29.4 ± 4.43	<0.0001
bFGF (2.5–80 pg/mL)	12.38 ± 2.20	13.71 ± 1.96	0.063
PLGF (2.5–80 pg/mL)	14.39 ± 2.58	12.38 ± 2.44	0.0200
Relative mRNA expression (qRT-PCR)
IGF-1/BA   (fold-change)	3.527 ± 0.70	1.000 ± 0.64	<0.0001
bFGF/BA   (fold-change)	2.120 ± 0.36	1.000 ± 0.42	<0.0001
PLGF/BA   (fold-change)	2.817 ± 0.79	1.000 ± 0.38	<0.0001
Staining intensity (IHC)
IGF-1(fold-change)	1.070 ± 0.08	1.000 ± 0.10	0.018
bFGF (fold-change)	1.072 ± 0.10	1.000 ± 0.11	0.033
PLGF (fold-change)	1.060 ± 0.07	1.000 ± 0.10	0.026
Positive cells count/field (IHC)
IGF-1 (count/field)	177.5 ± 20.02	167.8 ± 14.57	0.110
bFGF (count/field)	186.4 ± 19.01	180.6 ± 14.51	0.317
PLGF (count/field)	172.9 ± 17.82	167.4 ± 15.57	0.335

Abbreviations: PAS: Placenta accreta spectrum; qRT-PCR: quantitative real-time polymerase chain reaction; IHC: immunohistochemistry; IGF-1: insulin-like growth factor-1; bFGF: basic fibroblast growth factor; PLGF: placental growth factor; BA: β-actin.

Note: All data presented in this table were expressed as means ± SD.

2. Baseline characteristics of the four groups

According to FIGO PAS classification [5,16,28,29], the patients were also divided into four groups: 15 patients without PAS disorders, 7 patients with FIGO PAS grade I, 12 patients with FIGO PAS grade II, and 6 patients with FIGO PAS grade III (Table 3). Our experiment results have also found that the serum levels of IGF-1 means were 29.43 ± 4.43, 24.3 ± 4.03, 21.98 ± 3.29, and 22.03 ± 7.31, respectively for the four groups, p = 0.0006; the serum levels of PLGF means were 12.38 ± 2.44, 12.96 ± 2.74, 14.97 ± 2.56, and 14.89 ± 2.14, respectively for the four groups, p = 0.0392; while the serum levels of bFGF were not significantly different between the four groups, p = 0.1927 > 0.05 (Table 3). The mRNA relative expression of IGF-1/βA, bFGF/βA, and PLGF/βA in placental bed tissues was also significantly different between the four groups, p < 0.0001 for the three biomarkers (Table 3). The relative expression of mRNA IGF-1/βA mean for non-PAS group, PAS grade I group, PAS grade II group, and PAS grade III group was 1.000 ± 0.64, 3.194 ± 1.40, 3.509 ± 0.63, and 3.872 ± 0.70, respectively, p < 0.0001; Meanwhile, the relative expression of mRNA PLGF/βA mean for the four groups was 1.000 ± 0.38, 2.784 ± 1.14, 2.810 ± 0.71, and 2.869 ± 0.48, respectively, p < 0.0001 (Table 3). Surprisingly, the relative expression of mRNA bFGF/βA mean for the four groups was 1.000 ± 0.42, 2.240 ± 0.31, 2.140 ± 0.30, and 1.940 ± 0.51, respectively, p < 0.0001. Concerning the IHC staining, the staining intensity fold-change and the positive cells count/field of IGF-1, bFGF, and PLGF were not significantly different between the four groups, p > 0.05 for all of the three biomarkers (Table 3).

Table 3. Serum levels and placental bed tissue expression of IGF-1, bFGF, and PLGF in non-PAS, PAS grade I, PAS grade II, and PAS grade III.

Variables	Non-PAS	PAS grade I	PAS grade II	PAS grade III	P-value
	N = 15 (37.5%)	N = 7 (17.5%)	N = 12 (30%)	N = 6 (15%)
Serum level (ELISA)
IGF-1 (5–160 ng/mL)	29.43 ± 4.43	24.30 ± 4.03	21.98 ± 3.29	22.03 ± 7.31	0.0006
bFGF (2.5–80 pg/mL)	13.70 ± 1.96	13.08 ± 2.26	11.90 ± 1.98	12.52 ± 2.67	0.1927
PLGF (2.5–80 pg/mL)	12.38 ± 2.44	12.96 ± 2.74	14.97 ± 2.56	14.89 ± 2.14	0.0392
Relative mRNA expression (qRT-PCR)
IGF-1/BA (fold-change)	1.000 ± 0.64	3.194 ± 1.40	3.509 ± 0.63	3.872 ± 0.70	<0.0001
bFGF/BA (fold-change)	1.000 ± 0.42	2.240 ± 0.31	2.140 ± 0.30	1.940 ± 0.51	<0.0001
PLGF/BA (fold-change)	1.000 ± 0.38	2.784 ± 1.14	2.810 ± 0.71	2.869 ± 0.48	<0.0001
Staining intensity count (IHC)
IGF-1 (fold-change)	1.000 ± 0.10	1.059 ± 0.10	1.063 ± 0.05	1.100 ± 0.09	0.1044
bFGF (fold-change)	1.000 ± 0.11	1.084 ± 0.10	1.047 ± 0.10	1.109 ± 0.07	0.1062
PLGF (fold-change)	1.000 ± 0.10	1.047 ± 0.04	1.053 ± 0.07	1.071 ± 0.07	0.2071
Positive cells count (IHC)
IGF-1 (count/field)	167.75 ± 14.57	167.93 ± 20.22	177.80 ± 21.17	187.91 ± 13.71	0.0924
bFGF (count/field)	180.56 ± 14.51	184.99 ± 17.26	183.05 ± 21.49	194.52 ± 15.87	0.4367
PLGF (count/field)	167.44 ± 15.57	170.81 ± 21.73	173.23 ± 17.40	174.51 ± 16.75	0.7871

Abbreviations: PAS: Placenta accreta spectrum; ELISA: enzyme-linked immunosorbent assay; qRT-PCR: quantitative real-time polymerase chain reaction; IHC: immunohistochemistry; IGF-1: insulin-like growth factor-1; bFGF: basic fibroblast growth factor; PLGF: placental growth factor; BA: β-actin.

Note: All data presented in this table were expressed as means ± SD.

Discussion

Placenta previa with accreta is among the most serious obstetric placental placentation and implantation abnormalities that can lead to severe bleeding affecting about 50% of the cases [2,8,30]. Placenta previa alone does, not cause severe dangerous bleeding [30,31]. However, it is frequently associated with PAS disorders, leading to massive postpartum hemorrhage, and commonly leads to emergency hysterectomy [5,31,32]. This concept was correlated with our results, in which the PAS group had a significant post-partum hemorrhage of 1100 ± 959.06 ml during cesarean section (Table 1). Hence, to minimize the serious bleeding harm of PAS, the precise and timely prediction of PAS disorders before the surgery is imperative, specifically to discriminate placenta previa cases with PAS disorders from those without PAS [4,31,32]. Besides the ultrasound color Doppler and magnetic resonance imaging (MRI) as the primary and auxiliary modalities for the antenatal diagnosis of placenta previa with PAS [5,31,33], the serological examination of some growth factors (biomarkers) may also be important in the potential prediction of placenta previa with PAS disorders despite its controversial specificity [3–5]. The results of this study with 25 patients with PAS disorders and 15 patients without PAS disorders found low serum levels and high placental bed tissue expression (for both mRNA relative expression and for IHC intensity of staining) of IGF-1 in women with PAS disorders compared to those without PAS (Table 2). Some of our results are consistent with previous studies performed in the first trimester of pregnancy. Previously, Lyell et al. [25] have reported that the low levels of PAPP-A as a constituent of the IGF system [34] and as a marker of aberrant trophoblastic invasion and placental development [25] are often related to elevated levels of IGFBP protein and then with low levels of free IGF in the initial trimester of pregnancy with PAS disorders [24]. Some other previous studies have revealed that low levels of IGF-1 in the maternal blood and high expression of IGFBP-1 in the placenta may result in placental dysplasia (PAS), preeclampsia, and IUGR pregnancies [23]. However, the role of IGF-1 in pregnancies with PAS is still not fully understood in current literature, it may play an important regulatory role in pregnancies with placenta previa with accreta by influencing angiogenesis and trophoblastic invasion. As it is known, in the process of placental angiogenesis, wide and invasive neovascularization commonly leads to the occurrence of PAS disorders [24]. In addition, IGF-1 may improve VEGF-dependent signaling actions and facilitate its proangiogenic properties [18]. Regarding the trophoblastic invasion, IGF-1 is an angiogenic growth factor produced by the human decidual cells differentiated from endometrial stromal cells during implantation in response to progesterone hormone secreted by the corpus luteum during ovulation [35,36]. It may also regulate the ingestion and transportation of glucose and amino acids in trophoblasts and play a key part in the autocrine and paracrine invasion of trophoblasts into the decidual cells [24,37]. Our results also found significant high placental bed tissue expression of mRNA bFGF in women with PAS disorders compared to those without PAS, and there were even considerable significances for staining intensity (p = 0.033). However, surprisingly, there were no significant differences between the two groups for serum levels (p > 0.05) (Table 2). Hence, these results are contradictory to previous studies. Previously it has even been identified the implication of bFGF in driving the actions of VEGF, the strongest angiogenic [38,39]. Martinez- Fierro et al. [39] have pointed out that in improving angiogenesis, bFGF is known to be blocked by VEGF inhibition, which indicates that bFGF regulates angiogenesis upstream of VEGF by controlling VEGF action. The other results also found high serum levels and high placental bed expression of PLGF in women with PAS compared to placenta previa (Table 2). These results are conflicting with previous data. Wehrum et al. [40], in 2011 and Biberglu et al. [12], later in 2016 in their prospective studies did not find any statistical differences in the maternal serum levels of PLGF in the third trimester of pregnancy, while in 2018 prospective studies conducted by Uyanıkoğlu et al. [41], low maternal serum levels of PLGF revealed a good prediction for PAS in the last trimester of pregnancy. In addition, Wang et al. [2], recently conducted retrospective studies in 2021 and reported that serum PIGF was significantly increased in the first trimester of pregnancy with PAS disorders. These contradictory results may exist due to diverse factors such as various sample origins, testing kits, testing procedures, and gestational weeks [31]. Our study findings globally suggested that mRNA expression of PLGF, IGF-1, and bFGF all increased in placental bed tissues, but unfortunately, IGF-1 and bFGF levels decreased in maternal serum. This opposite correlation of IGF-1 and bFGF levels in serum and placental tissues may be explained. It is reasonable that the expression of factors in the serum is not consistent with that in the placenta, Firstly, serum protein and mRNA level are not the same, mRNA to protein process requires translation, modification, cutting, even folding, and so on [42]. Secondly, the source of the specimen is different; the serum is more active in the blood vessels and has a strong effect on the blood vessels, while the placental tissues are richer in the cellular components inside. The expression of these factors can be understood as the enrichment of factors, which means that there is inflammation in a certain place and then it is enriched in that place. This point is a very interesting topic that needs to be documented more and more.

In our study, regarding the gravity of the myometrial invasion of the placenta, we have also progressively analyzed the serum levels and placental bed tissue expression of IGF-1, bFGF, and PLGF according to FIGO PAS classification [5,16,28,29] (Table 3). Interestingly, our results indicated that the serum levels and placental bed tissue expression of bFGF were not significantly different between the 15 patients without PAS disorders, 7 patients with PAS grade I, 12 patients with PAS grade II, and 6 patients with PAS grade III, p > 0.05 mostly in all sections, except in the mRNA placental bed tissue expression section (p < 0.0001), but without any trends with the increasing FIGO PAS grading. We found that IGF-1 serum levels revealed a trend to decrease with increasing FIGO grading up to FIGO grade II, whereas PLGF serum levels tended to increase with increasing FIGO grading up to FIGO grade II. The serum levels of IGF-1 were significantly decreased in the patients with grade II with placental invasion (21.98 ± 3.29) compared to grade I with placental adhesions (24.3 ± 4.03), whereas the serum levels of PLGF were significantly increased in the patients with grade II (14.97 ± 2.56) with placental invasion compared to grade I with placental adhesions (12.96 ± 2.74) (Table 3). We detected that IGF-1 and PLGF serum levels did not have significance in the most invasive FIGO PAS grade III (Table 3). In addition, contrary to serum levels, IGF-1, and PLGF mRNA placental bed tissue expression revealed a trend to increase with increasing FIGO grading up to FIGO grade III (Table 3); the IHC staining also showed the same increasing trends, but p > 0.05 in all sections (Table 3). Regarding PLGF biomarker, some of these histochemical findings are consistent with previously published related studies in 2021. Similar to our studies, Alessandrini et al. [16] recently pointed out that placental PLGF expression showed a tendency to rise with increasing FIGO grading up to grade III and the differences in placental bed tissue expression of PLGF did not correlate with the serum levels. Serum levels of PLGF revealed a trend to increase with increasing FIGO grading only up to FIGO grade II.

Overall, our study has numerous limitations. The first limitation of our study is the small number of cases, possibly leading to selection bias. The second limitation is the gestational age, we only focused on the third trimester of multiparous pregnancies, and this may lead to a possible selection bias too. Hence, the gestational age and parity’s influence on those biomarkers, which can change during pregnancy, needs further investigation in the future. The third limitation is that this study included only the angiogenic growth factors; we could not find the kit for anti-angiogenic factors such as sFlt-1 as a control to explore the balance between angiogenic and anti-angiogenic growth factors as the process of placentation is regulated by both the angiogenic growth factors and anti-angiogenic factors. The fourth limitation is that all the pregnant women are from China, which fully minimizes the confounding effects of ethnic background.

Conclusions

In conclusion, low serum levels and high expression in placental bed tissues of IGF-1, or high serum levels and high expression in placental bed tissues of PLGF, may differentiate placenta previa patients with adhesive FIGO PAS grade I and invasive FIGO PAS grade II from those without PAS disorders. However, they could not significantly predict the degree of placental invasiveness in the invasive FIGO PAS grades II and III.

Ethics approval and consent to participate

Ethical approval was obtained from Wuhan Union Hospital, Tongji Medical College, Huazhong University of Science and Technology. All methods were carried out following relevant guidelines and regulations.

Authors’ contributions

The first author developed the data collection tools, collected and analyzed the data, and composed and reviewed the initial draft and the final manuscript. Other coauthors contributed to the conception and design of the study, the data analysis, and the writing of the manuscript. All authors approved the final version of the manuscript.

Acknowledgements

We thank all the colleagues in the Central Laboratory of Wuhan Union Hospital for their guidance in our experiments and all the patients who took part in this study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

All raw data that have been processed are owned by the first author and the corresponding author.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.