- Email: [email protected]
Association of increased cord blood soluble endoglin with the development of bronchopulmonary dysplasia in preterm infants with maternal preeclampsia
Pregnancy Hypertension 13 (2018) 148–153
Contents lists available at ScienceDirect
Pregnancy Hypertension journal homepage: www.elsevier.com/locate/preghy
Association of increased cord blood soluble endoglin with the development of bronchopulmonary dysplasia in preterm infants with maternal preeclampsia Do-Hyun Kima, Seung Han Shinb, Ee-Kyung Kimb, Han-Suk Kimb, a b
T
⁎
Department of Pediatrics, Dongguk University Ilsan Hospital, Goyang, Republic of Korea Department of Pediatrics, Seoul National University College of Medicine, Seoul, Republic of Korea
A R T I C LE I N FO
A B S T R A C T
Keywords: Bronchopulmonary dysplasia Preeclampsia sFlt-1 Soluble endoglin PlGF
Objectives: To investigate whether the levels of angiogenic factors in cord blood are associated with the development of bronchopulmonary dysplasia (BPD) in preterm infants with maternal preeclampsia. Study design: This retrospective cohort study included 199 singleton infants (gestational age < 32 weeks), including the preeclampsia group (59 infants) with severe/moderate BPD (24 infants) or no/mild BPD (35 infants) and the no preeclampsia group (140 infants). Main outcomes measures: The levels of soluble fms-like tyrosine kinase-1 (sFlt-1), soluble endoglin, and placental growth factor (PlGF) in cord blood were measured and compared among the study groups. Results: The soluble endoglin level and the ratio of (sFlt-1 + soluble endoglin) to PlGF were significantly higher in the preeclampsia group than in the no preeclampsia group (P < .05). Among preterm infants with maternal preeclampsia, both of these parameters were also significantly higher in the severe/moderate BPD group than the no/mild BPD group (P < .05). Receiver operator curve analysis revealed that increased cord blood soluble endoglin was predictive of severe or moderate BPD in preterm infants with maternal preeclampsia (area under the curve 0.73). Gestational age (adjusted odds ratio [OR] 0.25; P < .001) and high soluble endoglin level in cord blood (> 3420 pg/mL) (adjusted OR 11.9; P = .006) were significant risk factors for the development of severe or moderate BPD in the preeclampsia group according to multivariate logistic regression analysis. Conclusion: Increased cord blood soluble endoglin is associated with the development of severe or moderate BPD in preterm infants with maternal preeclampsia.
1. Introduction Preeclampsia is the most common medical complication of pregnancy worldwide, resulting in substantial perinatal and neonatal morbidity and mortality [1]. Alterations in angiogenic state appear to be involved in the pathogenesis of preeclampsia. Increased levels of antiangiogenic factors, such as soluble vascular endothelial growth factor (VEGF) receptor-1 (soluble fms-like tyrosine kinase-1, also known as sFlt-1) and soluble endoglin, and decreased levels of proangiogenic factors, including free VEGF and placental growth factor (PlGF), play a central role in the pathogenesis of preeclampsia [2–5]. This antiangiogenic state is shared by the fetus, resulting with increased cord blood sFlt-1 but decreased VEGF and PlGF levels in infants with maternal preeclampsia [6]. Bronchopulmonary dysplasia (BPD) is a chronic lung disease in preterm infants that is characterized by arrested lung development due
⁎
to early lung injury [7]. Similar to the pathogenesis of preeclampsia, the development of BPD is also associated with dysregulation of angiogenesis in the pulmonary vasculature of the developing lung (the ‘vascular hypothesis’) [8–12]. In other words, alterations in angiogenic state appear to be involved in the pathogenesis of both preeclampsia and BPD. The relationship between preeclampsia and increased risk of BPD has been increasingly examined. Exposure to excess sFlt-1 in amniotic fluid during late gestation caused sustained reductions in alveolarization and pulmonary vascular growth in rats, leading to the development of BPD [13]. In many epidemiologic studies, preeclampsia was shown to be independently associated with a high risk of BPD, although little is known about the underlying mechanism [14–18]. To investigate the underlying mechanism that may link preeclampsia with a high risk for BPD under conditions of impaired angiogenesis, we hypothesized that the levels of sFlt-1 and soluble endoglin (as antiangiogenic factors) and PlGF (as a proangiogenic factor)
Corresponding author at: Department of Pediatrics, Seoul National University Children’s Hospital, 101 Daehak-ro, Jongno-gu, Seoul 03080, Republic of Korea. E-mail address: [email protected] (H.-S. Kim).
https://doi.org/10.1016/j.preghy.2018.06.002 Received 16 November 2017; Received in revised form 6 March 2018; Accepted 9 June 2018 Available online 09 June 2018 2210-7789/ © 2018 International Society for the Study of Hypertension in Pregnancy. Published by Elsevier B.V. All rights reserved.
Pregnancy Hypertension 13 (2018) 148–153
D.-H. Kim et al.
Fig. 1. Flow diagram showing the study design involving the 330 infants screened in this study. SNUH, Seoul National University Hospital; PCA, postconceptional age.
fluid index < 5 cm by ultrasound performed just before delivery).
in cord blood are associated with the development of severe or moderate BPD in preterm infants with maternal preeclampsia.
2.2. Umbilical cord blood samples and assays 2. Methods Umbilical cord blood was collected in ethylenediaminetetraacetic acid-containing blood collection tubes by arteriopuncture of the umbilical artery at birth. The samples were then centrifuged at 3000g for 5 min at 4 °C, and the supernatants were stored in polypropylene tubes at −70 °C. The sFlt-1, soluble endoglin, and PlGF levels in cord blood were measured using human ELISA kits (MyBioSource, Inc., San Diego, CA., USA; Cusabio Biotech Co., Baltimore, MD., USA; R&D Systems, Minneapolis, Minn., USA, respectively) according to the manufacturer’s protocols. The ratio of (sFlt-1 + soluble endoglin) to PlGF for each patient was calculated to assess the balance between the anti- and proangiogenic factors in the infants’ serum.
2.1. Patients All singleton infants with a gestational age of < 32 weeks born in Seoul National University Hospital during a 10-year period from January 2005 to December 2014 were screened. Infants with major congenital malformations and infants without a cord blood sample were excluded. Infants who were transferred to other hospitals or died before reaching a postconceptional age (PCA) of 36 weeks were also excluded (Fig. 1). Data on the patients’ clinical characteristics were collected and analyzed retrospectively. Clinical characteristics were directly evaluated by a single reviewer using medical records. Maternal preeclampsia was defined as new-onset hypertension (systolic blood pressure ≥ 140 mm Hg or diastolic blood pressure ≥ 90 mm Hg on at least two occasions at least 4 h apart) after 20 weeks’ gestation in a formerly normotensive patient, accompanied by proteinuria ≥ 300 mg per 24-hour urine collection [19]. BPD and its severity were defined using the criteria of the National Institute of Child Health Workshop definition for BPD [20], i.e., treatment with oxygen for at least 28 days with division into the following 3 subgroups at 36-week PCA: (1) mild (breathing room air); (2) moderate (need for a < 30% fraction of inspired oxygen), and (3) severe (need for ≥30% fraction of inspired oxygen and/or positive pressure support). BPD-associated pulmonary hypertension (PH) was diagnosed based on echocardiograms demonstrating elevated right ventricle pressure using the following criteria in preterm infants with BPD: (1) velocity of tricuspid valve regurgitation of ≥3 m/s in the absence of pulmonary stenosis, and (2) flat or leftward deviated interventricular septal configuration and right ventricular hypertrophy with chamber dilation. Patients with one or both of these findings beyond 2 months of age were characterized as having PH. The diagnosis of respiratory distress syndrome required the presence of respiratory distress, increased oxygen requirement and a radiological finding consistent with respiratory distress syndrome in the absence of other causes of respiratory distress. Patent ductus arteriosus was diagnosed by echocardiography, and only cases treated with prostaglandin inhibitor or surgical ligation were included. The presence of proven sepsis was defined as at least a single blood culture and clinical signs of infection. Other clinical characteristics studied included preterm premature rupture of membrane, histologic chorioamnionitis (HCAM; presence of acute inflammatory changes on a membrane roll and the placental chorionic plate), and oligohydramnios (defined as amniotic
2.3. Statistical analyses First, we performed univariate analyses to investigate demographic and clinical characteristics in the study subjects by groups. Categorical variables were analyzed by the χ2 test and Fisher’s exact test. Differences in continuous variables were assessed by Student’s t test. The sFlt-1, soluble endoglin, PlGF levels and the ratio of (sFlt-1 + soluble endoglin) to PlGF in cord blood were adjusted for gestational age by analysis of covariance (ANCOVA). For cord blood soluble endoglin, we calculated receiver operating characteristic (ROC) curve and performed a sensitivity analysis to determine the cut-off value for predicting the development of severe or moderate BPD in preterm infants born to mothers with preeclampsia. Next, to determine independent risk factors for the development of severe or moderate BPD in the preeclampsia group, we performed a multivariate logistic regression analysis. Variables that were significant in the previous univariate analysis were included in the logistic regression model. We calculated the adjusted odds ratio (OR) for the development of severe or moderate BPD and the 95% confidence interval (CI) of the selected variables. Data are presented as the mean ± standard deviation or frequency, and a P-value of < .05 was considered statistically significant. The statistical analyses were performed using SPSS for Windows, version 12.0 (SPSS Inc., Chicago, IL, USA). 3. Results 3.1. Classification of the study subjects A total of 199 infants fulfilled the study criteria, and this group accounted for 60% of all singleton infants with a gestational age < 149
Pregnancy Hypertension 13 (2018) 148–153
D.-H. Kim et al.
Table 1 Demographic and clinical characteristics in the study subjects by groups. Study groups
Preeclampsia
Maternal characteristics for preeclampsia “Highest” systolic BP prior to delivery (mmHg) “Highest” diastolic BP prior to delivery (mmHg) Proteinuria Liver involvement Hematologic complications Male Cesarean section Gestational age, weeks Birth weight, g Birth weight < 3rd percentile for age Prenatal steroids 5-min Apgar scores ≤ 6 PPROM HCAM Oligohydramnios Severe/moderate BPD BPD associated PH RDS PDA IVH (grade ≥3) NEC (stage ≥2) Proven sepsis Duration of O2 therapy, days Duration of CV or HFV, days
No Preeclampsia (n = 140)
Total (n = 59)
Severe/Moderate BPD (n = 24)
Mild/No BPD (n = 35)
163 ± 16.2 106 ± 10.9 59 (1 0 0) 10 (17) 11 (19) 27 (46) 56 (95)* 29.1 ± 1.6 915 ± 249* 14 (24)* 45 (76) 32 (54) 2 (3)* 10 (17)* 9 (15) 24 (41)* 4 (7) 27 (46) 41 (69)* 2 (3) 2 (3) 5 (8) 45.4 ± 52.9 11.6 ± 20.0
167 ± 17.9 106 ± 11.7 24 (1 0 0) 6 (25) 6 (25) 14 (58) 23 (96) 28.0 ± 1.4† 747 ± 207† 9 (38)† 21 (88) 14 (58) 1 (4) 4 (17) 6 (25) – 4 (17)† 12 (50) 20 (83) 1 (4) 1 (4) 4 (17) 86.3 ± 60.1† 25.3 ± 25.8†
159 ± 14.0 105 ± 10.5 35 (1 0 0) 4 (11) 5 (14) 13 (37) 33 (94) 29.8 ± 1.2 1,030 ± 207 5 (14) 23 (66) 18 (51) 1 (3) 6 (17) 3 (9) – 0 (0) 15 (43) 21 (60) 1 (3) 1 (3) 1 (3) 17.4 ± 18.6 2.2 ± 3.3
N/A N/A N/A N/A N/A 80 (57) 70 (50) 28.6 ± 2.4 1,209 ± 408 12 (9) 103 (74) 55 (39) 84 (60) 82 (59) 28 (20) 37 (26) 12 (9) 48 (34) 72 (51) 7 (5) 6 (4) 15 (11) 45.3 ± 71.3 19.3 ± 47.3
BPD, bronchopulmonary dysplasia; BP, blood pressure; PPROM, preterm premature rupture of membrane; HCAM, histologic chorioamnionitis; PH, pulmonary hypertension; IVH, intraventricular hemorrhage; NEC, necrotizing enterocolitis; CV, conventional ventilation; HFV, high-frequency ventilation; N/A, not applicable. Values are expressed as the mean ± standard deviation or numbers of individuals (%). IVH and NEC were classified using the Volpe grading system and modified Bell’s staging criteria, respectively. * P < .05 vs. No Preeclampsia. † P < .05 vs. Mild/No BPD.
the no preeclampsia group (41% vs. 26%, P = .035). Among infants with preeclampsia, the severe/moderate BPD group had a lower gestational age and birth weight, a higher rate of IUGR and BPD-associated PH, and longer durations of O2 or invasive ventilator therapies than the no/mild BPD group (Table 1).
32 weeks during the study period. Among the 199 infants, 59 (30%) were diagnosed as being exposed to preeclampsia during pregnancy (the preeclampsia group). Among these 59 infants, 24 (41%) were subclassified as severe or moderate BPD (the severe/moderate BPD group) and 35 (59%) as no or mild BPD (the no/mild BPD group) according to the National Institute of Child Health Workshop definition [20]. A flow diagram showing the study design involving the 330 infants screened in this study is presented in Figure. Population descriptions are presented in Table 1.
3.3. Levels of cord blood angiogenic factors in the study subjects by groups The soluble endoglin level and the ratio of (sFlt-1 + soluble endoglin) to PlGF in cord blood were significantly higher in the preeclampsia group than in the no preeclampsia group (mean 3815 vs. 2686 pg/mL and 45.8 vs. 38.4, P < .05, respectively). The sFlt-1 and PlGF levels were not significantly different between the two groups. Among preterm infants with preeclampsia, the soluble endoglin level and the ratio of (sFlt-1 + soluble endoglin) to PlGF in cord blood were also significantly higher in the severe/moderate BPD group than in the no/mild BPD group (mean 4811 vs. 3132 pg/mL and 52.7 vs. 41.0, P < .05, respectively). The sFlt-1 and PlGF levels were not significantly
3.2. Characteristics of the study subjects by groups Infants with preeclampsia had lower birth weight, lower rates of preterm premature rupture of membrane and HCAM and higher rates of Cesarean section, birth weight less than the 3rd percentile for age (intrauterine growth retardation, IUGR), and significant patent ductus arteriosus than those without preeclampsia. Severe or moderate BPD occurred significantly more frequently in the preeclampsia group than
Table 2 Levels of sFlt-1, soluble endoglin, and PlGF in cord blood in the study subjects by groups. Study groups
sFlt-1 (pg/mL) Soluble endoglin (pg/mL) PlGF (pg/mL) Ratio of (sFlt-1 + soluble endoglin):PlGF
Preeclampsia
No Preeclampsia (n = 140)
Total (n = 59)
Severe/Moderate BPD (n = 24)
Mild/No BPD (n = 35)
1245 ± 434 3815 ± 2407* 119 ± 79 45.8 ± 20.6*
1324 ± 469 4811 ± 2717† 132 ± 107 52.7 ± 20.3†
1191 ± 408 3132 ± 1.925 110 ± 48 41.0 ± 19.7
sFlt-1, soluble fms-like tyrosine kinase-1; PlGF, placental growth factor. Values are expressed as the mean ± standard deviation. * P < .05 vs. No Preeclampsia. † P < .05 vs. Mild/No BPD. 150
1.272 ± 587 2686 ± 2014 108 ± 37 38.4 ± 23.1
Pregnancy Hypertension 13 (2018) 148–153
D.-H. Kim et al.
Table 3 sFlt-1, soluble endoglin, and PlGF in cord blood adjusted for gestational age.
sFlt-1 (pg/mL) Soluble endoglin (pg/mL) PlGF (pg/mL) Ratio of (sFlt-1 + soluble endoglin):PlGF
Severe/moderate BPD (n = 24) Mean ± SD
Mild/No BPD (n = 35) Mean ± SD
difference
95% CI
P values
1361 ± 100 4954 ± 525 131 ± 18 54.6 ± 4.6
1166 ± 80 3033 ± 420 110 ± 15 39.7 ± 3.7
195 1921 21 14.9
−81.4 to472 464–3380 −29.2 to71.7 2.3–27.7
0.163 0.011 0.403 0.022
sFlt-1, soluble fms-like tyrosine kinase-1; PlGF, placental growth factor; BPD, bronchopulmonary dysplasia; SD, standard deviation; CI, confidence interval. P values were derived from analysis of covariance (ANCOVA).
transforming growth factor (TGF)-ß, is highly expressed on cell membranes of vascular endothelium and syncytiotrophoblasts. Placental endoglin is up-regulated in preeclampsia, releasing soluble endoglin into the maternal circulation. Soluble endoglin has an antiangiogenic effect that may inhibit TGF-ß signaling in the vasculature [5,21]. Circulating soluble endoglin and sFlt-1, which both cause endothelial dysfunction by a different mechanism, may contribute to the development of preeclampsia. Notably, Levine et al. reported that the ratio of (sFlt-1 + soluble endoglin):PlGF was more strongly predictive of preeclampsia than were individual biomarkers [5]. Maternal serum levels of soluble endoglin increase during the last two months of normal pregnancy, and they increase more steeply in women in whom preeclampsia developed, reaching a peak at the onset of clinical disease. Before the onset of preeclampsia in women who subsequently developed preterm preeclampsia, maternal serum soluble endoglin began to increase at 17 weeks through 20 weeks of gestation compared with controls and had a steep increase at 33 through 36 weeks [5]. Thus, we assumed that differences in cord blood soluble endoglin between the preeclampsia and no preeclampsia groups would be increased to a greater extent if preterm infants above 31 weeks of gestation were included in this study. Levin et al. [5] and Tsao et al. [6] reported similar findings that cord blood or maternal serum sFlt-1 levels were correlated with small for gestational age (SGA) preterm infants with preeclampsia, suggesting that excess antiangiogenic factors, such as sFlt-1, may play an important role in maternal preeclampsia-induced SGA. However, in our study, cord blood sFlt-1 levels were not associated with IUGR in preterm infants with preeclampsia, in contrast to previous studies. Instead, infants with preeclampsia and IUGR had higher levels of cord blood soluble endoglin than those with preeclampsia and no IUGR (mean 5003 vs. 3445, P = .030) (Table not shown). Further large studies are required to demonstrate the association of antiangiogenic factors, such as soluble endoglin and sFlt-1, with preeclampsia-induced IUGR. In the present study, infants with preeclampsia had higher rates of severe or moderate BPD compared to those without preeclampsia, consistent with previous epidemiologic studies [14–18]. In addition, preeclampsia was found to be a significant independent risk factor for the development of severe or moderate BPD (adjusted OR 2.82; 95% CI 1.13–7.02; P = .026) (Table not shown). The major finding of the present study was that among preterm infants with preeclampsia, the soluble endoglin level and the ratio of (sFlt-1 + soluble endoglin) to PlGF in cord blood were significantly higher in the severe or moderate BPD group than the no or mild BPD group, even after adjusting for gestational age. Gestational age has a potent influence on the levels of soluble endoglin during gestation as mentioned above. In this study, adjustment of gestational age further accentuated the difference in cord blood soluble endoglin levels between the two groups. This finding indicates that increased level of cord blood soluble endoglin may be associated with the development of severe or moderate BPD in infants with preeclampsia. Currently, little is known about the association between antiangiogenic factors, such as soluble endoglin and BPD. Pulmonary microvasculature of short-term ventilated preterm infants showed significant upregulation of endoglin mRNA and protein expression [22]. In
different between these two groups (Table 2). Among preterm infants with preeclampsia, after adjustment for gestational age using ANCOVA, the soluble endoglin level and the ratio of (sFlt-1 + soluble endoglin) to PlGF in cord blood were still significantly higher in the severe/moderate BPD group than in the no/mild BPD group (mean 4954 vs. 3033 pg/mL, P = .011 and 54.6 vs. 39.7, P = .022, respectively) (Table 3). 3.4. Performance of cord blood soluble endoglin The ROC curve, constructed to assess whether the level of cord blood soluble endoglin could predict the development of severe or moderate BPD in preterm infants with preeclampsia, yielded an area under the curve (AUC) of 0.73 (95% CI 0.60–0.86, P = .003). A cut-off value of 3420 pg/mL had a sensitivity of 67% and a specificity of 66% (data not shown). 3.5. Multivariate logistic regression analysis The relationship between the development of severe or moderate BPD in the preeclampsia group and increased soluble endoglin level was further examined by a multivariate logistic regression analysis. Gestational age (OR 0.25; 95% CI 0.12–0.52; P < .001) and high soluble endoglin level in cord blood (> 3420 pg/mL) (OR 11.9; 95% CI 2.02–69.6; P = .006) were found to be significant independent risk factors for the development of severe or moderate BPD in the preeclampsia group after adjusting for gestational age, IUGR, and high soluble endoglin level in cord blood (> 3420 pg/mL) (Table 4). 4. Discussion The present study demonstrated that an antiangiogenic factor, soluble endoglin in cord blood, is increased in preterm infants with maternal preeclampsia, and increased cord blood soluble endoglin may be associated with the development of severe or moderate BPD among preterm infants with preeclampsia. Endoglin, a co-receptor of Table 4 Factors associated with the development of severe or moderate bronchopulmonary dysplasia in preterm infants with maternal preeclampsia determined by multivariate logistic regression analysis.
Gestational age Birth weight < 3rd percentile for age High soluble endoglin level in cord blood (> 3420 pg/mL)
Crude OR [95% CI]
P values
Adjusted* OR [95% CI]
P values
0.38 (0.23–0.64) 3.60 (1.03–12.7) 3.83 (1.28–11.5)
< 0.001
0.25 (0.12–0.52) 3.96 (0.67–23.6) 11.9 (2.02–69.6)
< 0.001
0.046 0.017
0.130 0.006
OR, odds ratio; CI, confidence interval. * Adjusted for gestational age, birth weight < 3rd percentile for age, and high soluble endoglin level in cord blood (> 3420 pg/mL) in preterm infants with maternal preeclampsia. 151
Pregnancy Hypertension 13 (2018) 148–153
D.-H. Kim et al.
a cross-sectional human study, an increase in the amniotic fluid levels of soluble endoglin in infants with intra-amniotic infection and/or inflammation was associated with subsequent development of BPD [23]. Most recently, cord blood PlGF and granulocyte-colony stimulating factor that were decreased with placental maternal vascular underperfusion were predictive of BPD-associated PH [24]. In contrast to these previous results, we did not find an association between cord blood levels of three angiogenic factors and the development of severe or moderate BPD among all study infants (data not shown). However, among infants with preeclampsia, the association between cord blood level of soluble endoglin and the development of severe or moderate BPD reached statistical significance. The etiology of BPD is multifactorial and involves exposure to various antenatal and/or postnatal factors. In addition to preeclampsia, recent studies showed that HCAM is significantly associated with the increased incidence of BPD, although the relationship between HCAM and BPD is still controversial [25,26]. In the present study, the rate of HCAM was significantly higher in infants without preeclampsia than those with preeclampsia. These findings suggest that HCAM may be a risk factor for the development of BPD in infants without preeclampsia, and therefore, both preeclampsia and HCAM are likely to independently contribute to the development of BPD. We propose that investigating multifactorial etiologies of BPD individually can be useful for elucidation of the pathogenesis of BPD. To our knowledge, the present study is the first report identifying an association between alteration in antiangiogenic factors, such as soluble endoglin, and the subsequent development of severe or moderate BPD in infants with preeclampsia. Based on previous studies, we inferred that increased soluble endoglin during pregnancy involving preeclampsia and the subsequent increase in the ratio of (sFlt-1 + soluble endoglin) to PlGF might be shared by the fetus and cause a shift in the balance toward antiangiogenesis, which influenced the development of BPD. Although our findings do not indicate that high level of cord blood soluble endoglin directly causes the development of BPD, the association demonstrates that cord blood level of soluble endoglin may be a useful predictor of the development of severe or moderate BPD in preterm infant with preeclampsia. For development as a biomarker to predict subsequent severe or moderate BPD, further large studies of cord blood soluble endoglin are needed. In summary, soluble endoglin and the ratio of (sFlt-1 + soluble endoglin):PlGF in cord blood were increased in preterm infants with maternal preeclampsia, and, among infants with maternal preeclampsia, this parameter was increased in infants that developed severe or moderate BPD relative to infants who developed no or mild BPD. We therefore suggest the possibility that alterations in antiangiogenic factors, such as increased soluble endoglin and the ratio of (sFlt-1 + soluble endoglin):PlGF in maternal circulation, with preeclampsia are shared by the fetus and reflect early lung injury in angiogenesis that leads to the subsequent development of severe or moderate BPD. This may explain the underlying mechanism that links preeclampsia with a high risk for BPD under conditions of impaired angiogenesis.
5.2. Ethics approval The study was approved by the Institutional Review Board (IRB) of Seoul National University Hospital. Umbilical blood collection and data analyses were performed after obtaining informed consent from the subjects’ parents. 5.3. Disclosure of interests The authors declare no conflict of interest. 5.4. Source of funding This study was supported by a grant from the Department of Pediatrics, Seoul National University College of Medicine (No. 2016002) and by the research project of Seoul National University College of Medicine (800-20160034) in 2016. Acknowledgments The authors would like to thank Professor Bo Hyun Yoon for the kind assistance and advice. We are also grateful to Kyeung Yeup Lee for his technical support. The manuscript was revised for English by a scientific language editing service (American Journal Experts). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.preghy.2018.06.002. References [1] S.E. Maynard, S.A. Karumanchi, Angiogenic factors and preeclampsia, Semin. Nephrol. 31 (1) (2011) 33–46. [2] S.E. Maynard, J.Y. Min, J. Merchan, K.H. Lim, J. Li, S. Mondal, et al., Excess placental soluble fms-like tyrosine kinase 1 (sFlt1) may contribute to endothelial dysfunction, hypertension, and proteinuria in preeclampsia, J. Clin. Invest. 111 (5) (2003) 649–658. [3] R.J. Levine, S.E. Maynard, C. Qian, K.H. Lim, L.J. England, K.F. Yu, et al., Circulating angiogenic factors and the risk of preeclampsia, N. Engl. J. Med. 350 (7) (2004) 672–683. [4] C. Lam, K.H. Lim, S.A. Karumanchi, Circulating angiogenic factors in the pathogenesis and prediction of preeclampsia, Hypertension 46 (5) (2005) 1077–1085. [5] R.J. Levine, C. Lam, C. Qian, K.F. Yu, S.E. Maynard, B.P. Sachs, et al., Soluble endoglin and other circulating antiangiogenic factors in preeclampsia, N. Engl. J. Med. 355 (10) (2006) 992–1005. [6] P.N. Tsao, S.C. Wei, Y.N. Su, H.C. Chou, C.Y. Chen, W.S. Hsieh, Excess soluble fmslike tyrosine kinase 1 and low platelet counts in premature neonates of preeclamptic mothers, Pediatrics 116 (2) (2005) 468–472. [7] A.J. Jobe, The new BPD: an arrest of lung development, Pediatr. Res. 46 (6) (1999) 641–643. [8] S.H. Abman, Bronchopulmonary dysplasia: a vascular hypothesis, Am. J. Respir. Crit. Care. Med. 164 (10 Pt 1) (2001) 1755–1756. [9] A.J. Bhatt, G.S. Pryhuber, H. Huyck, R.H. Watkins, L.A. Metlay, W.M. Maniscalco, Disrupted pulmonary vasculature and decreased vascular endothelial growth factor, Flt-1, and TIE-2 in human infants dying with bronchopulmonary dysplasia, Am. J. Respir. Crit. Care Med. 164 (10 Pt 1) (2001) 1971–1980. [10] P. Lassus, M. Turanlahti, P. Heikkilä, L.C. Andersson, I. Nupponen, A. Sarnesto, et al., Pulmonary vascular endothelial growth factor and Flt-1 in fetuses, in acute and chronic lung disease, and in persistent pulmonary hypertension of the newborn, Am. J. Respir. Crit. Care Med. 164 (10 Pt 1) (2001) 1981–1987. [11] B. Thébaud, Angiogenesis in lung development, injury and repair: implications for chronic lung disease of prematurity, Neonatology 91 (4) (2007) 291–297. [12] J. Janér, S. Andersson, E. Kajantie, P. Lassus, Endostatin concentration in cord plasma predicts the development of bronchopulmonary dysplasia in very low birth weight infants, Pediatrics 123 (4) (2009) 1142–1146. [13] J.R. Tang, S.A. Karumanchi, G. Seedorf, N. Markham, S.H. Abman, Excess soluble vascular endothelial growth factor receptor-1 in amniotic fluid impairs lung growth in rats: linking preeclampsia with bronchopulmonary dysplasia, Am. J. Physiol. Lung. Cell. Mol. Physiol. 302 (1) (2012) L36–L46. [14] C. Bose, L.J. Van Marter, M. Laughon, T.M. O'Shea, E.N. Allred, P. Karna, R.A. Ehrenkranz, et al., Fetal growth restriction and chronic lung disease among infants born before the 28th week of gestation, Pediatrics 124 (3) (2009) e450–e458. [15] A.R. Hansen, C.M. Barnés, J. Folkman, T.F. McElrath, Maternal preeclampsia predicts the development of bronchopulmonary dysplasia, J. Pediatr. 156 (4) (2010)
5. Declarations 5.1. Contributorship D.H. Kim conceptualized and designed the study, collected data, carried out the analyses, drafted the initial manuscript, and reviewed and revised the manuscript. S.H. Shin and E.K. Kim coordinated and supervised data collection, and critically reviewed the manuscript. H.S. Kim conceptualized and designed the study, coordinated and supervised data collection, and critically reviewed and revised the manuscript. All authors contributed to the writing and/or editing the manuscript and approved its content.
152
Pregnancy Hypertension 13 (2018) 148–153
D.-H. Kim et al.
[22] M.E. De Paepe, C. Patel, A. Tsai, S. Gundavarapu, Q. Mao, Endoglin (CD105) upregulation in pulmonary microvasculature of ventilated preterm infants, Am. J. Respir. Crit. Care Med. 178 (2) (2008) 180–187. [23] S.K. Kim, R. Romero, Z.A. Savasan, Y. Xu, Z. Dong, D.C. Lee, et al., Endoglin in amniotic fluid as a risk factor for the subsequent development of bronchopulmonary dysplasia, Am. J. Reprod. Immunol. 69 (2) (2013) 105–123. [24] K.K. Mestan, N. Gotteiner, N. Porta, W. Grobman, E.J. Su, L.M. Ernst, Cord blood biomarkers of placental maternal vascular underperfusion predict bronchopulmonary dysplasia-associated pulmonary hypertension, J. Pediatr. 185 (2017) 33–41. [25] L. Hartling, Y. Liang, T. Lacaze-Masmonteil, Chorioamnionitis as a risk factor for bronchopulmonary dysplasia: a systematic review and meta-analysis, Arch. Dis. Child. Fetal Neonatal 97 (1) (2012) F8–F17. [26] S.Y. Kim, C.W. Choi, E. Jung, J. Lee, J.A. Lee, H. Kim, et al., Neonatal morbidities associated with histologic chorioamnionitis defined based on the site and extent of inflammation in very low birth weight infants, J. Korean Med. Sci. 30 (10) (2015) 1476–1482.
532–536. [16] H. Ozkan, M. Cetinkaya, N. Koksal, Increased incidence of bronchopulmonary dysplasia in preterm infants exposed to preeclampsia, J. Matern. Fetal. Neonatal. Med. 25 (12) (2012) 2681–2685. [17] L. Eriksson, B. Haglund, V. Odlind, M. Altman, H. Kieler, Prenatal inflammatory risk factors for development of bronchopulmonary dysplasia, Pediatr. Pulmonol. 49 (7) (2014) 665–672. [18] L. Gemmell, L. Martin, K.E. Murphy, N. Modi, S. Håkansson, B. Reichman, et al., Hypertensive disorders of pregnancy and outcomes of preterm infants of 24 to 28 weeks' gestation, J. Perinatol. 36 (12) (2016) 1067–1072. [19] A. Jeyabalan, Hypertensive disorders of pregnancy, in: R.J. Martin, A.A. Fanaroff, M.C. Walsh (Eds.), Neonatal-Perinatal Medicine, Elsevier, St Louis, 2015, pp. 250–252. [20] A.H. Jobe, E. Bancalari, Bronchopulmonary dysplasia, Am. J. Respir. Crit. Care Med. 163 (7) (2001) 1723–1729. [21] S. Venkatesha, M. Toporsian, C. Lam, J. Hanai, T. Mammoto, Y.M. Kim, et al., Soluble endoglin contributes to the pathogenesis of preeclampsia, Nat. Med. 12 (6) (2006) 642–649.
153
Contents lists available at ScienceDirect
Pregnancy Hypertension journal homepage: www.elsevier.com/locate/preghy
Association of increased cord blood soluble endoglin with the development of bronchopulmonary dysplasia in preterm infants with maternal preeclampsia Do-Hyun Kima, Seung Han Shinb, Ee-Kyung Kimb, Han-Suk Kimb, a b
T
⁎
Department of Pediatrics, Dongguk University Ilsan Hospital, Goyang, Republic of Korea Department of Pediatrics, Seoul National University College of Medicine, Seoul, Republic of Korea
A R T I C LE I N FO
A B S T R A C T
Keywords: Bronchopulmonary dysplasia Preeclampsia sFlt-1 Soluble endoglin PlGF
Objectives: To investigate whether the levels of angiogenic factors in cord blood are associated with the development of bronchopulmonary dysplasia (BPD) in preterm infants with maternal preeclampsia. Study design: This retrospective cohort study included 199 singleton infants (gestational age < 32 weeks), including the preeclampsia group (59 infants) with severe/moderate BPD (24 infants) or no/mild BPD (35 infants) and the no preeclampsia group (140 infants). Main outcomes measures: The levels of soluble fms-like tyrosine kinase-1 (sFlt-1), soluble endoglin, and placental growth factor (PlGF) in cord blood were measured and compared among the study groups. Results: The soluble endoglin level and the ratio of (sFlt-1 + soluble endoglin) to PlGF were significantly higher in the preeclampsia group than in the no preeclampsia group (P < .05). Among preterm infants with maternal preeclampsia, both of these parameters were also significantly higher in the severe/moderate BPD group than the no/mild BPD group (P < .05). Receiver operator curve analysis revealed that increased cord blood soluble endoglin was predictive of severe or moderate BPD in preterm infants with maternal preeclampsia (area under the curve 0.73). Gestational age (adjusted odds ratio [OR] 0.25; P < .001) and high soluble endoglin level in cord blood (> 3420 pg/mL) (adjusted OR 11.9; P = .006) were significant risk factors for the development of severe or moderate BPD in the preeclampsia group according to multivariate logistic regression analysis. Conclusion: Increased cord blood soluble endoglin is associated with the development of severe or moderate BPD in preterm infants with maternal preeclampsia.
1. Introduction Preeclampsia is the most common medical complication of pregnancy worldwide, resulting in substantial perinatal and neonatal morbidity and mortality [1]. Alterations in angiogenic state appear to be involved in the pathogenesis of preeclampsia. Increased levels of antiangiogenic factors, such as soluble vascular endothelial growth factor (VEGF) receptor-1 (soluble fms-like tyrosine kinase-1, also known as sFlt-1) and soluble endoglin, and decreased levels of proangiogenic factors, including free VEGF and placental growth factor (PlGF), play a central role in the pathogenesis of preeclampsia [2–5]. This antiangiogenic state is shared by the fetus, resulting with increased cord blood sFlt-1 but decreased VEGF and PlGF levels in infants with maternal preeclampsia [6]. Bronchopulmonary dysplasia (BPD) is a chronic lung disease in preterm infants that is characterized by arrested lung development due
⁎
to early lung injury [7]. Similar to the pathogenesis of preeclampsia, the development of BPD is also associated with dysregulation of angiogenesis in the pulmonary vasculature of the developing lung (the ‘vascular hypothesis’) [8–12]. In other words, alterations in angiogenic state appear to be involved in the pathogenesis of both preeclampsia and BPD. The relationship between preeclampsia and increased risk of BPD has been increasingly examined. Exposure to excess sFlt-1 in amniotic fluid during late gestation caused sustained reductions in alveolarization and pulmonary vascular growth in rats, leading to the development of BPD [13]. In many epidemiologic studies, preeclampsia was shown to be independently associated with a high risk of BPD, although little is known about the underlying mechanism [14–18]. To investigate the underlying mechanism that may link preeclampsia with a high risk for BPD under conditions of impaired angiogenesis, we hypothesized that the levels of sFlt-1 and soluble endoglin (as antiangiogenic factors) and PlGF (as a proangiogenic factor)
Corresponding author at: Department of Pediatrics, Seoul National University Children’s Hospital, 101 Daehak-ro, Jongno-gu, Seoul 03080, Republic of Korea. E-mail address: [email protected] (H.-S. Kim).
https://doi.org/10.1016/j.preghy.2018.06.002 Received 16 November 2017; Received in revised form 6 March 2018; Accepted 9 June 2018 Available online 09 June 2018 2210-7789/ © 2018 International Society for the Study of Hypertension in Pregnancy. Published by Elsevier B.V. All rights reserved.
Pregnancy Hypertension 13 (2018) 148–153
D.-H. Kim et al.
Fig. 1. Flow diagram showing the study design involving the 330 infants screened in this study. SNUH, Seoul National University Hospital; PCA, postconceptional age.
fluid index < 5 cm by ultrasound performed just before delivery).
in cord blood are associated with the development of severe or moderate BPD in preterm infants with maternal preeclampsia.
2.2. Umbilical cord blood samples and assays 2. Methods Umbilical cord blood was collected in ethylenediaminetetraacetic acid-containing blood collection tubes by arteriopuncture of the umbilical artery at birth. The samples were then centrifuged at 3000g for 5 min at 4 °C, and the supernatants were stored in polypropylene tubes at −70 °C. The sFlt-1, soluble endoglin, and PlGF levels in cord blood were measured using human ELISA kits (MyBioSource, Inc., San Diego, CA., USA; Cusabio Biotech Co., Baltimore, MD., USA; R&D Systems, Minneapolis, Minn., USA, respectively) according to the manufacturer’s protocols. The ratio of (sFlt-1 + soluble endoglin) to PlGF for each patient was calculated to assess the balance between the anti- and proangiogenic factors in the infants’ serum.
2.1. Patients All singleton infants with a gestational age of < 32 weeks born in Seoul National University Hospital during a 10-year period from January 2005 to December 2014 were screened. Infants with major congenital malformations and infants without a cord blood sample were excluded. Infants who were transferred to other hospitals or died before reaching a postconceptional age (PCA) of 36 weeks were also excluded (Fig. 1). Data on the patients’ clinical characteristics were collected and analyzed retrospectively. Clinical characteristics were directly evaluated by a single reviewer using medical records. Maternal preeclampsia was defined as new-onset hypertension (systolic blood pressure ≥ 140 mm Hg or diastolic blood pressure ≥ 90 mm Hg on at least two occasions at least 4 h apart) after 20 weeks’ gestation in a formerly normotensive patient, accompanied by proteinuria ≥ 300 mg per 24-hour urine collection [19]. BPD and its severity were defined using the criteria of the National Institute of Child Health Workshop definition for BPD [20], i.e., treatment with oxygen for at least 28 days with division into the following 3 subgroups at 36-week PCA: (1) mild (breathing room air); (2) moderate (need for a < 30% fraction of inspired oxygen), and (3) severe (need for ≥30% fraction of inspired oxygen and/or positive pressure support). BPD-associated pulmonary hypertension (PH) was diagnosed based on echocardiograms demonstrating elevated right ventricle pressure using the following criteria in preterm infants with BPD: (1) velocity of tricuspid valve regurgitation of ≥3 m/s in the absence of pulmonary stenosis, and (2) flat or leftward deviated interventricular septal configuration and right ventricular hypertrophy with chamber dilation. Patients with one or both of these findings beyond 2 months of age were characterized as having PH. The diagnosis of respiratory distress syndrome required the presence of respiratory distress, increased oxygen requirement and a radiological finding consistent with respiratory distress syndrome in the absence of other causes of respiratory distress. Patent ductus arteriosus was diagnosed by echocardiography, and only cases treated with prostaglandin inhibitor or surgical ligation were included. The presence of proven sepsis was defined as at least a single blood culture and clinical signs of infection. Other clinical characteristics studied included preterm premature rupture of membrane, histologic chorioamnionitis (HCAM; presence of acute inflammatory changes on a membrane roll and the placental chorionic plate), and oligohydramnios (defined as amniotic
2.3. Statistical analyses First, we performed univariate analyses to investigate demographic and clinical characteristics in the study subjects by groups. Categorical variables were analyzed by the χ2 test and Fisher’s exact test. Differences in continuous variables were assessed by Student’s t test. The sFlt-1, soluble endoglin, PlGF levels and the ratio of (sFlt-1 + soluble endoglin) to PlGF in cord blood were adjusted for gestational age by analysis of covariance (ANCOVA). For cord blood soluble endoglin, we calculated receiver operating characteristic (ROC) curve and performed a sensitivity analysis to determine the cut-off value for predicting the development of severe or moderate BPD in preterm infants born to mothers with preeclampsia. Next, to determine independent risk factors for the development of severe or moderate BPD in the preeclampsia group, we performed a multivariate logistic regression analysis. Variables that were significant in the previous univariate analysis were included in the logistic regression model. We calculated the adjusted odds ratio (OR) for the development of severe or moderate BPD and the 95% confidence interval (CI) of the selected variables. Data are presented as the mean ± standard deviation or frequency, and a P-value of < .05 was considered statistically significant. The statistical analyses were performed using SPSS for Windows, version 12.0 (SPSS Inc., Chicago, IL, USA). 3. Results 3.1. Classification of the study subjects A total of 199 infants fulfilled the study criteria, and this group accounted for 60% of all singleton infants with a gestational age < 149
Pregnancy Hypertension 13 (2018) 148–153
D.-H. Kim et al.
Table 1 Demographic and clinical characteristics in the study subjects by groups. Study groups
Preeclampsia
Maternal characteristics for preeclampsia “Highest” systolic BP prior to delivery (mmHg) “Highest” diastolic BP prior to delivery (mmHg) Proteinuria Liver involvement Hematologic complications Male Cesarean section Gestational age, weeks Birth weight, g Birth weight < 3rd percentile for age Prenatal steroids 5-min Apgar scores ≤ 6 PPROM HCAM Oligohydramnios Severe/moderate BPD BPD associated PH RDS PDA IVH (grade ≥3) NEC (stage ≥2) Proven sepsis Duration of O2 therapy, days Duration of CV or HFV, days
No Preeclampsia (n = 140)
Total (n = 59)
Severe/Moderate BPD (n = 24)
Mild/No BPD (n = 35)
163 ± 16.2 106 ± 10.9 59 (1 0 0) 10 (17) 11 (19) 27 (46) 56 (95)* 29.1 ± 1.6 915 ± 249* 14 (24)* 45 (76) 32 (54) 2 (3)* 10 (17)* 9 (15) 24 (41)* 4 (7) 27 (46) 41 (69)* 2 (3) 2 (3) 5 (8) 45.4 ± 52.9 11.6 ± 20.0
167 ± 17.9 106 ± 11.7 24 (1 0 0) 6 (25) 6 (25) 14 (58) 23 (96) 28.0 ± 1.4† 747 ± 207† 9 (38)† 21 (88) 14 (58) 1 (4) 4 (17) 6 (25) – 4 (17)† 12 (50) 20 (83) 1 (4) 1 (4) 4 (17) 86.3 ± 60.1† 25.3 ± 25.8†
159 ± 14.0 105 ± 10.5 35 (1 0 0) 4 (11) 5 (14) 13 (37) 33 (94) 29.8 ± 1.2 1,030 ± 207 5 (14) 23 (66) 18 (51) 1 (3) 6 (17) 3 (9) – 0 (0) 15 (43) 21 (60) 1 (3) 1 (3) 1 (3) 17.4 ± 18.6 2.2 ± 3.3
N/A N/A N/A N/A N/A 80 (57) 70 (50) 28.6 ± 2.4 1,209 ± 408 12 (9) 103 (74) 55 (39) 84 (60) 82 (59) 28 (20) 37 (26) 12 (9) 48 (34) 72 (51) 7 (5) 6 (4) 15 (11) 45.3 ± 71.3 19.3 ± 47.3
BPD, bronchopulmonary dysplasia; BP, blood pressure; PPROM, preterm premature rupture of membrane; HCAM, histologic chorioamnionitis; PH, pulmonary hypertension; IVH, intraventricular hemorrhage; NEC, necrotizing enterocolitis; CV, conventional ventilation; HFV, high-frequency ventilation; N/A, not applicable. Values are expressed as the mean ± standard deviation or numbers of individuals (%). IVH and NEC were classified using the Volpe grading system and modified Bell’s staging criteria, respectively. * P < .05 vs. No Preeclampsia. † P < .05 vs. Mild/No BPD.
the no preeclampsia group (41% vs. 26%, P = .035). Among infants with preeclampsia, the severe/moderate BPD group had a lower gestational age and birth weight, a higher rate of IUGR and BPD-associated PH, and longer durations of O2 or invasive ventilator therapies than the no/mild BPD group (Table 1).
32 weeks during the study period. Among the 199 infants, 59 (30%) were diagnosed as being exposed to preeclampsia during pregnancy (the preeclampsia group). Among these 59 infants, 24 (41%) were subclassified as severe or moderate BPD (the severe/moderate BPD group) and 35 (59%) as no or mild BPD (the no/mild BPD group) according to the National Institute of Child Health Workshop definition [20]. A flow diagram showing the study design involving the 330 infants screened in this study is presented in Figure. Population descriptions are presented in Table 1.
3.3. Levels of cord blood angiogenic factors in the study subjects by groups The soluble endoglin level and the ratio of (sFlt-1 + soluble endoglin) to PlGF in cord blood were significantly higher in the preeclampsia group than in the no preeclampsia group (mean 3815 vs. 2686 pg/mL and 45.8 vs. 38.4, P < .05, respectively). The sFlt-1 and PlGF levels were not significantly different between the two groups. Among preterm infants with preeclampsia, the soluble endoglin level and the ratio of (sFlt-1 + soluble endoglin) to PlGF in cord blood were also significantly higher in the severe/moderate BPD group than in the no/mild BPD group (mean 4811 vs. 3132 pg/mL and 52.7 vs. 41.0, P < .05, respectively). The sFlt-1 and PlGF levels were not significantly
3.2. Characteristics of the study subjects by groups Infants with preeclampsia had lower birth weight, lower rates of preterm premature rupture of membrane and HCAM and higher rates of Cesarean section, birth weight less than the 3rd percentile for age (intrauterine growth retardation, IUGR), and significant patent ductus arteriosus than those without preeclampsia. Severe or moderate BPD occurred significantly more frequently in the preeclampsia group than
Table 2 Levels of sFlt-1, soluble endoglin, and PlGF in cord blood in the study subjects by groups. Study groups
sFlt-1 (pg/mL) Soluble endoglin (pg/mL) PlGF (pg/mL) Ratio of (sFlt-1 + soluble endoglin):PlGF
Preeclampsia
No Preeclampsia (n = 140)
Total (n = 59)
Severe/Moderate BPD (n = 24)
Mild/No BPD (n = 35)
1245 ± 434 3815 ± 2407* 119 ± 79 45.8 ± 20.6*
1324 ± 469 4811 ± 2717† 132 ± 107 52.7 ± 20.3†
1191 ± 408 3132 ± 1.925 110 ± 48 41.0 ± 19.7
sFlt-1, soluble fms-like tyrosine kinase-1; PlGF, placental growth factor. Values are expressed as the mean ± standard deviation. * P < .05 vs. No Preeclampsia. † P < .05 vs. Mild/No BPD. 150
1.272 ± 587 2686 ± 2014 108 ± 37 38.4 ± 23.1
Pregnancy Hypertension 13 (2018) 148–153
D.-H. Kim et al.
Table 3 sFlt-1, soluble endoglin, and PlGF in cord blood adjusted for gestational age.
sFlt-1 (pg/mL) Soluble endoglin (pg/mL) PlGF (pg/mL) Ratio of (sFlt-1 + soluble endoglin):PlGF
Severe/moderate BPD (n = 24) Mean ± SD
Mild/No BPD (n = 35) Mean ± SD
difference
95% CI
P values
1361 ± 100 4954 ± 525 131 ± 18 54.6 ± 4.6
1166 ± 80 3033 ± 420 110 ± 15 39.7 ± 3.7
195 1921 21 14.9
−81.4 to472 464–3380 −29.2 to71.7 2.3–27.7
0.163 0.011 0.403 0.022
sFlt-1, soluble fms-like tyrosine kinase-1; PlGF, placental growth factor; BPD, bronchopulmonary dysplasia; SD, standard deviation; CI, confidence interval. P values were derived from analysis of covariance (ANCOVA).
transforming growth factor (TGF)-ß, is highly expressed on cell membranes of vascular endothelium and syncytiotrophoblasts. Placental endoglin is up-regulated in preeclampsia, releasing soluble endoglin into the maternal circulation. Soluble endoglin has an antiangiogenic effect that may inhibit TGF-ß signaling in the vasculature [5,21]. Circulating soluble endoglin and sFlt-1, which both cause endothelial dysfunction by a different mechanism, may contribute to the development of preeclampsia. Notably, Levine et al. reported that the ratio of (sFlt-1 + soluble endoglin):PlGF was more strongly predictive of preeclampsia than were individual biomarkers [5]. Maternal serum levels of soluble endoglin increase during the last two months of normal pregnancy, and they increase more steeply in women in whom preeclampsia developed, reaching a peak at the onset of clinical disease. Before the onset of preeclampsia in women who subsequently developed preterm preeclampsia, maternal serum soluble endoglin began to increase at 17 weeks through 20 weeks of gestation compared with controls and had a steep increase at 33 through 36 weeks [5]. Thus, we assumed that differences in cord blood soluble endoglin between the preeclampsia and no preeclampsia groups would be increased to a greater extent if preterm infants above 31 weeks of gestation were included in this study. Levin et al. [5] and Tsao et al. [6] reported similar findings that cord blood or maternal serum sFlt-1 levels were correlated with small for gestational age (SGA) preterm infants with preeclampsia, suggesting that excess antiangiogenic factors, such as sFlt-1, may play an important role in maternal preeclampsia-induced SGA. However, in our study, cord blood sFlt-1 levels were not associated with IUGR in preterm infants with preeclampsia, in contrast to previous studies. Instead, infants with preeclampsia and IUGR had higher levels of cord blood soluble endoglin than those with preeclampsia and no IUGR (mean 5003 vs. 3445, P = .030) (Table not shown). Further large studies are required to demonstrate the association of antiangiogenic factors, such as soluble endoglin and sFlt-1, with preeclampsia-induced IUGR. In the present study, infants with preeclampsia had higher rates of severe or moderate BPD compared to those without preeclampsia, consistent with previous epidemiologic studies [14–18]. In addition, preeclampsia was found to be a significant independent risk factor for the development of severe or moderate BPD (adjusted OR 2.82; 95% CI 1.13–7.02; P = .026) (Table not shown). The major finding of the present study was that among preterm infants with preeclampsia, the soluble endoglin level and the ratio of (sFlt-1 + soluble endoglin) to PlGF in cord blood were significantly higher in the severe or moderate BPD group than the no or mild BPD group, even after adjusting for gestational age. Gestational age has a potent influence on the levels of soluble endoglin during gestation as mentioned above. In this study, adjustment of gestational age further accentuated the difference in cord blood soluble endoglin levels between the two groups. This finding indicates that increased level of cord blood soluble endoglin may be associated with the development of severe or moderate BPD in infants with preeclampsia. Currently, little is known about the association between antiangiogenic factors, such as soluble endoglin and BPD. Pulmonary microvasculature of short-term ventilated preterm infants showed significant upregulation of endoglin mRNA and protein expression [22]. In
different between these two groups (Table 2). Among preterm infants with preeclampsia, after adjustment for gestational age using ANCOVA, the soluble endoglin level and the ratio of (sFlt-1 + soluble endoglin) to PlGF in cord blood were still significantly higher in the severe/moderate BPD group than in the no/mild BPD group (mean 4954 vs. 3033 pg/mL, P = .011 and 54.6 vs. 39.7, P = .022, respectively) (Table 3). 3.4. Performance of cord blood soluble endoglin The ROC curve, constructed to assess whether the level of cord blood soluble endoglin could predict the development of severe or moderate BPD in preterm infants with preeclampsia, yielded an area under the curve (AUC) of 0.73 (95% CI 0.60–0.86, P = .003). A cut-off value of 3420 pg/mL had a sensitivity of 67% and a specificity of 66% (data not shown). 3.5. Multivariate logistic regression analysis The relationship between the development of severe or moderate BPD in the preeclampsia group and increased soluble endoglin level was further examined by a multivariate logistic regression analysis. Gestational age (OR 0.25; 95% CI 0.12–0.52; P < .001) and high soluble endoglin level in cord blood (> 3420 pg/mL) (OR 11.9; 95% CI 2.02–69.6; P = .006) were found to be significant independent risk factors for the development of severe or moderate BPD in the preeclampsia group after adjusting for gestational age, IUGR, and high soluble endoglin level in cord blood (> 3420 pg/mL) (Table 4). 4. Discussion The present study demonstrated that an antiangiogenic factor, soluble endoglin in cord blood, is increased in preterm infants with maternal preeclampsia, and increased cord blood soluble endoglin may be associated with the development of severe or moderate BPD among preterm infants with preeclampsia. Endoglin, a co-receptor of Table 4 Factors associated with the development of severe or moderate bronchopulmonary dysplasia in preterm infants with maternal preeclampsia determined by multivariate logistic regression analysis.
Gestational age Birth weight < 3rd percentile for age High soluble endoglin level in cord blood (> 3420 pg/mL)
Crude OR [95% CI]
P values
Adjusted* OR [95% CI]
P values
0.38 (0.23–0.64) 3.60 (1.03–12.7) 3.83 (1.28–11.5)
< 0.001
0.25 (0.12–0.52) 3.96 (0.67–23.6) 11.9 (2.02–69.6)
< 0.001
0.046 0.017
0.130 0.006
OR, odds ratio; CI, confidence interval. * Adjusted for gestational age, birth weight < 3rd percentile for age, and high soluble endoglin level in cord blood (> 3420 pg/mL) in preterm infants with maternal preeclampsia. 151
Pregnancy Hypertension 13 (2018) 148–153
D.-H. Kim et al.
a cross-sectional human study, an increase in the amniotic fluid levels of soluble endoglin in infants with intra-amniotic infection and/or inflammation was associated with subsequent development of BPD [23]. Most recently, cord blood PlGF and granulocyte-colony stimulating factor that were decreased with placental maternal vascular underperfusion were predictive of BPD-associated PH [24]. In contrast to these previous results, we did not find an association between cord blood levels of three angiogenic factors and the development of severe or moderate BPD among all study infants (data not shown). However, among infants with preeclampsia, the association between cord blood level of soluble endoglin and the development of severe or moderate BPD reached statistical significance. The etiology of BPD is multifactorial and involves exposure to various antenatal and/or postnatal factors. In addition to preeclampsia, recent studies showed that HCAM is significantly associated with the increased incidence of BPD, although the relationship between HCAM and BPD is still controversial [25,26]. In the present study, the rate of HCAM was significantly higher in infants without preeclampsia than those with preeclampsia. These findings suggest that HCAM may be a risk factor for the development of BPD in infants without preeclampsia, and therefore, both preeclampsia and HCAM are likely to independently contribute to the development of BPD. We propose that investigating multifactorial etiologies of BPD individually can be useful for elucidation of the pathogenesis of BPD. To our knowledge, the present study is the first report identifying an association between alteration in antiangiogenic factors, such as soluble endoglin, and the subsequent development of severe or moderate BPD in infants with preeclampsia. Based on previous studies, we inferred that increased soluble endoglin during pregnancy involving preeclampsia and the subsequent increase in the ratio of (sFlt-1 + soluble endoglin) to PlGF might be shared by the fetus and cause a shift in the balance toward antiangiogenesis, which influenced the development of BPD. Although our findings do not indicate that high level of cord blood soluble endoglin directly causes the development of BPD, the association demonstrates that cord blood level of soluble endoglin may be a useful predictor of the development of severe or moderate BPD in preterm infant with preeclampsia. For development as a biomarker to predict subsequent severe or moderate BPD, further large studies of cord blood soluble endoglin are needed. In summary, soluble endoglin and the ratio of (sFlt-1 + soluble endoglin):PlGF in cord blood were increased in preterm infants with maternal preeclampsia, and, among infants with maternal preeclampsia, this parameter was increased in infants that developed severe or moderate BPD relative to infants who developed no or mild BPD. We therefore suggest the possibility that alterations in antiangiogenic factors, such as increased soluble endoglin and the ratio of (sFlt-1 + soluble endoglin):PlGF in maternal circulation, with preeclampsia are shared by the fetus and reflect early lung injury in angiogenesis that leads to the subsequent development of severe or moderate BPD. This may explain the underlying mechanism that links preeclampsia with a high risk for BPD under conditions of impaired angiogenesis.
5.2. Ethics approval The study was approved by the Institutional Review Board (IRB) of Seoul National University Hospital. Umbilical blood collection and data analyses were performed after obtaining informed consent from the subjects’ parents. 5.3. Disclosure of interests The authors declare no conflict of interest. 5.4. Source of funding This study was supported by a grant from the Department of Pediatrics, Seoul National University College of Medicine (No. 2016002) and by the research project of Seoul National University College of Medicine (800-20160034) in 2016. Acknowledgments The authors would like to thank Professor Bo Hyun Yoon for the kind assistance and advice. We are also grateful to Kyeung Yeup Lee for his technical support. The manuscript was revised for English by a scientific language editing service (American Journal Experts). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.preghy.2018.06.002. References [1] S.E. Maynard, S.A. Karumanchi, Angiogenic factors and preeclampsia, Semin. Nephrol. 31 (1) (2011) 33–46. [2] S.E. Maynard, J.Y. Min, J. Merchan, K.H. Lim, J. Li, S. Mondal, et al., Excess placental soluble fms-like tyrosine kinase 1 (sFlt1) may contribute to endothelial dysfunction, hypertension, and proteinuria in preeclampsia, J. Clin. Invest. 111 (5) (2003) 649–658. [3] R.J. Levine, S.E. Maynard, C. Qian, K.H. Lim, L.J. England, K.F. Yu, et al., Circulating angiogenic factors and the risk of preeclampsia, N. Engl. J. Med. 350 (7) (2004) 672–683. [4] C. Lam, K.H. Lim, S.A. Karumanchi, Circulating angiogenic factors in the pathogenesis and prediction of preeclampsia, Hypertension 46 (5) (2005) 1077–1085. [5] R.J. Levine, C. Lam, C. Qian, K.F. Yu, S.E. Maynard, B.P. Sachs, et al., Soluble endoglin and other circulating antiangiogenic factors in preeclampsia, N. Engl. J. Med. 355 (10) (2006) 992–1005. [6] P.N. Tsao, S.C. Wei, Y.N. Su, H.C. Chou, C.Y. Chen, W.S. Hsieh, Excess soluble fmslike tyrosine kinase 1 and low platelet counts in premature neonates of preeclamptic mothers, Pediatrics 116 (2) (2005) 468–472. [7] A.J. Jobe, The new BPD: an arrest of lung development, Pediatr. Res. 46 (6) (1999) 641–643. [8] S.H. Abman, Bronchopulmonary dysplasia: a vascular hypothesis, Am. J. Respir. Crit. Care. Med. 164 (10 Pt 1) (2001) 1755–1756. [9] A.J. Bhatt, G.S. Pryhuber, H. Huyck, R.H. Watkins, L.A. Metlay, W.M. Maniscalco, Disrupted pulmonary vasculature and decreased vascular endothelial growth factor, Flt-1, and TIE-2 in human infants dying with bronchopulmonary dysplasia, Am. J. Respir. Crit. Care Med. 164 (10 Pt 1) (2001) 1971–1980. [10] P. Lassus, M. Turanlahti, P. Heikkilä, L.C. Andersson, I. Nupponen, A. Sarnesto, et al., Pulmonary vascular endothelial growth factor and Flt-1 in fetuses, in acute and chronic lung disease, and in persistent pulmonary hypertension of the newborn, Am. J. Respir. Crit. Care Med. 164 (10 Pt 1) (2001) 1981–1987. [11] B. Thébaud, Angiogenesis in lung development, injury and repair: implications for chronic lung disease of prematurity, Neonatology 91 (4) (2007) 291–297. [12] J. Janér, S. Andersson, E. Kajantie, P. Lassus, Endostatin concentration in cord plasma predicts the development of bronchopulmonary dysplasia in very low birth weight infants, Pediatrics 123 (4) (2009) 1142–1146. [13] J.R. Tang, S.A. Karumanchi, G. Seedorf, N. Markham, S.H. Abman, Excess soluble vascular endothelial growth factor receptor-1 in amniotic fluid impairs lung growth in rats: linking preeclampsia with bronchopulmonary dysplasia, Am. J. Physiol. Lung. Cell. Mol. Physiol. 302 (1) (2012) L36–L46. [14] C. Bose, L.J. Van Marter, M. Laughon, T.M. O'Shea, E.N. Allred, P. Karna, R.A. Ehrenkranz, et al., Fetal growth restriction and chronic lung disease among infants born before the 28th week of gestation, Pediatrics 124 (3) (2009) e450–e458. [15] A.R. Hansen, C.M. Barnés, J. Folkman, T.F. McElrath, Maternal preeclampsia predicts the development of bronchopulmonary dysplasia, J. Pediatr. 156 (4) (2010)
5. Declarations 5.1. Contributorship D.H. Kim conceptualized and designed the study, collected data, carried out the analyses, drafted the initial manuscript, and reviewed and revised the manuscript. S.H. Shin and E.K. Kim coordinated and supervised data collection, and critically reviewed the manuscript. H.S. Kim conceptualized and designed the study, coordinated and supervised data collection, and critically reviewed and revised the manuscript. All authors contributed to the writing and/or editing the manuscript and approved its content.
152
Pregnancy Hypertension 13 (2018) 148–153
D.-H. Kim et al.
[22] M.E. De Paepe, C. Patel, A. Tsai, S. Gundavarapu, Q. Mao, Endoglin (CD105) upregulation in pulmonary microvasculature of ventilated preterm infants, Am. J. Respir. Crit. Care Med. 178 (2) (2008) 180–187. [23] S.K. Kim, R. Romero, Z.A. Savasan, Y. Xu, Z. Dong, D.C. Lee, et al., Endoglin in amniotic fluid as a risk factor for the subsequent development of bronchopulmonary dysplasia, Am. J. Reprod. Immunol. 69 (2) (2013) 105–123. [24] K.K. Mestan, N. Gotteiner, N. Porta, W. Grobman, E.J. Su, L.M. Ernst, Cord blood biomarkers of placental maternal vascular underperfusion predict bronchopulmonary dysplasia-associated pulmonary hypertension, J. Pediatr. 185 (2017) 33–41. [25] L. Hartling, Y. Liang, T. Lacaze-Masmonteil, Chorioamnionitis as a risk factor for bronchopulmonary dysplasia: a systematic review and meta-analysis, Arch. Dis. Child. Fetal Neonatal 97 (1) (2012) F8–F17. [26] S.Y. Kim, C.W. Choi, E. Jung, J. Lee, J.A. Lee, H. Kim, et al., Neonatal morbidities associated with histologic chorioamnionitis defined based on the site and extent of inflammation in very low birth weight infants, J. Korean Med. Sci. 30 (10) (2015) 1476–1482.
532–536. [16] H. Ozkan, M. Cetinkaya, N. Koksal, Increased incidence of bronchopulmonary dysplasia in preterm infants exposed to preeclampsia, J. Matern. Fetal. Neonatal. Med. 25 (12) (2012) 2681–2685. [17] L. Eriksson, B. Haglund, V. Odlind, M. Altman, H. Kieler, Prenatal inflammatory risk factors for development of bronchopulmonary dysplasia, Pediatr. Pulmonol. 49 (7) (2014) 665–672. [18] L. Gemmell, L. Martin, K.E. Murphy, N. Modi, S. Håkansson, B. Reichman, et al., Hypertensive disorders of pregnancy and outcomes of preterm infants of 24 to 28 weeks' gestation, J. Perinatol. 36 (12) (2016) 1067–1072. [19] A. Jeyabalan, Hypertensive disorders of pregnancy, in: R.J. Martin, A.A. Fanaroff, M.C. Walsh (Eds.), Neonatal-Perinatal Medicine, Elsevier, St Louis, 2015, pp. 250–252. [20] A.H. Jobe, E. Bancalari, Bronchopulmonary dysplasia, Am. J. Respir. Crit. Care Med. 163 (7) (2001) 1723–1729. [21] S. Venkatesha, M. Toporsian, C. Lam, J. Hanai, T. Mammoto, Y.M. Kim, et al., Soluble endoglin contributes to the pathogenesis of preeclampsia, Nat. Med. 12 (6) (2006) 642–649.
153