Journal of Perinatology (2015), 1–6 © 2015 Nature America, Inc. All rights reserved 0743-8346/15 www.nature.com/jp
ORIGINAL ARTICLE
Standardization of amniotic fluid leptin levels and utility in maternal overweight and fetal undergrowth M Scott-Finley1,6, JG Woo2,3,6, M Habli1,2, O Ramos-Gonzales2, JF Cnota2, Y Wang3, BD Kamath-Rayne4, AC Hinton1, WJ Polzin1,5, TM Crombleholme5 and RB Hinton2 OBJECTIVE: Leptin is an adipokine that regulates energy homeostasis. The objective of this study was to establish a gestational age-specific standard for amniotic fluid leptin (AFL) levels and examine the relationship between AFL, maternal overweight and fetal growth restriction. STUDY DESIGN: Amniotic fluid was obtained at mid-gestation from singleton gravidas, and leptin was quantified using enzymelinked immunosorbent assay. Amniotic fluid samples from 321 term pregnancies were analyzed. Clinical data, including fetal ultrasound measurements and maternal and infant characteristics, were available for a subset of patients (n = 45). RESULTS: The median interquartile range AFL level was significantly higher at 14 weeks’ gestation (2133 pg ml − 1 (1703 to 4347)) than after 33 weeks’ gestation (519 pg ml − 1 (380 to 761), P trend o0.0001), an average difference of 102 pg ml − 1 per week. AFL levels were positively correlated with maternal pre-pregnancy body mass index (BMI) (r = 0.36, P = 0.03) adjusting for gestational age at measurement, but were not associated with fetal growth. CONCLUSIONS: AFL levels are higher at mid-gestation than at late gestation, and are associated with maternal pre-pregnancy BMI. Journal of Perinatology advance online publication, 30 April 2015; doi:10.1038/jp.2015.39
INTRODUCTION Fetal growth restriction (FGR) is among the most significant problems in perinatal medicine. It is the second leading cause of perinatal morbidity and mortality, exceeded only by prematurity.1 FGR is defined as an estimated fetal weight less than the tenth percentile for gestational age and corresponds with small for gestational age (SGA), which is defined as a birth weight less than the tenth percentile for gestational age.2 Perinatal morbidity and mortality is significantly increased in the presence of low birth weight for gestational age.3,4 In addition, FGR infants are at increased risk for a variety of long-term sequelae, including adult-onset metabolic diseases such as type 2 diabetes and hypertension.5,6 Despite the immediate and long-term importance of FGR, there are presently no predictive biomarkers or preventive treatment strategies. Growth abnormalities in the fetus may be caused by numerous insults that are generally classified as maternal, fetal or placental in origin and include maternal preeclampsia, fetal genetic anomalies and primary placental vasculopathies. However, the shared pathogenesis of FGR is poorly understood. Placental insufficiency may be a common final pathway for FGR, but predicting placental function is challenging.7 Several growth factors and signaling molecules have been implicated in FGR, including vascular endothelial growth factor, placental growth factor and leptin.8–10 Leptin is an adipokine that regulates energy homeostasis and is increased in obesity (anabolism) and decreased in cachexia (catabolism),11,12 and leptin is present in fetal tissue, including the placenta, suggesting a direct role for leptin in fetal growth and
development.13–15 However, amniotic fluid leptin (AFL) has not been standardized or examined systematically during pregnancy. Little is known about how AFL levels change during pregnancy and relate to fetal growth. Understanding leptin dysregulation may elucidate mechanisms of FGR at the intersection of maternal health and fetal growth. In this study, we examined AFL in a crosssectional manner across gestational ages with the goal of establishing a gestational age-specific standard. Because maternal weight has been associated with leptin regulation, we hypothesized that increased pre-pregnancy maternal body mass index (BMI) is associated with increased AFL levels at mid-gestation. These observations have important clinical implications and may lead to applications that improve the diagnostic and prognostic assessment of fetal growth. METHODS Study population This study was approved by the Institutional Review Boards of Cincinnati Children’s Hospital Medical Center and Good Samaritan Hospital (Cincinnati, OH. USA). Amniotic fluid was obtained from pregnancies at both community and referral hospitals that underwent a genetic amniocentesis at any time during gestation; however, the specific indications for testing are unknown because the study used discarded tissue that was anonymized. Inclusion criteria were a normal karyotype analysis and the absence of fetal anomalies by ultrasound anatomic survey. Exclusion criteria included prematurity ( o37 weeks’ gestation), multiple gestation pregnancy, and fetal or neonatal demise. Patients were enrolled from May 2009 to October 2010. Detailed medical record information was reviewed in a small subset of high-risk cases from a single center owing to
1 Division of Maternal Fetal Medicine, Obstetrics and Gynecology, Good Samaritan Hospital, Cincinnati, OH, USA; 2Division of Cardiology, The Heart Institute, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA; 3Division of Biostatistics and Epidemiology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA; 4Perinatal Institute, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA and 5Fetal Care Center, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA. Correspondence: Dr RB Hinton, Division of Cardiology, The Heart Institute, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Ave, MLC 2003, Cincinnati, OH 45229, USA. E-mail: [email protected] 6 These authors contributed equally to this work. Received 3 December 2014; Received 23 March 2015; accepted 24 March 2015
Standardization of amniotic fluid leptin M Scott-Finley et al
2 the fact that the majority of cases were derived from referring hospitals where information was not accessible. Informed consent was obtained from this subset of patients. Archived placental tissue was identified in a subset of these cases.
Maternal health
again and streptavidin–HRP was added for 20 min, after which a 1:1 mixture of colors A and B was added. Plates were stopped with 2 N H2SO4 and read at 450 nm minus 595 nm background. Values are expressed in pg ml − 1. The Bradford assay was employed by adding 200 μl 1X solution (R&D Systems) to a plate, followed by 5 μl per sample. Plates were read at 595 nm. Values were expressed as μg ml − 1.
Clinical records were reviewed for a subset (n = 45) with available data, with the following data collected: maternal age, calculated pre-pregnancy BMI from self-reported weight and measured height, BMI at delivery, and weight gain during pregnancy. In addition, medical history of preexisting diabetes mellitus or gestational diabetes (GD) and preeclampsia was noted. Gestational diabetes was identified by an abnormal diagnostic 100-g 3-hour oral glucose test defined as two elevated values. Finally, estimated fetal weight by ultrasound within 2 weeks of amniocentesis, gestational age at the time of amniocentesis, gestational age at the time of delivery, birth weight, birth head circumference and infant gender were recorded. For the purposes of analysis, GD was defined as either an abnormal glucose-tolerance test or confirmed clinical diagnosis of GD. Maternal pre-pregnancy BMI (in kg m − 2) was classified as normal weight (BMI o 25) or overweight/obese (BMI ⩾ 25).16
Histochemistry and immunohistochemistry were used for exploratory studies to assess placental tissue, which was obtained from 5/45 (11%) pregnancies from the subset of patients with medical record information in the course of routine pathological processes. Antibodies directed against leptin (1:500, Lifespan Biosciences, Seattle, WA, USA), and leptin receptor (1:50, Abcam, Cambridge, MA, USA) were used to assess the location and relative amount of protein. The leptin antibodies required additional citrate pretreatment. A semiquantitative scale was used to assess the expression.18 Briefly, staining was graded on a scale of 0 to 3 based on the percentage of positive cells as follows: 0 = immunoreactivity in o10% of cells; 1 = 10 to 35%; 2 = 35 to 70%; and 3 = 470%. Quantification was performed by a single blinded reader.
Fetal growth
Statistical analyses
Neonatal anthropometrics included birth weight and head circumference. Growth z-scores were derived from the Olsen standard, a contemporary set of growth curves derived from a large, racially diverse USA sample, resulting in gestational age- and gender-specific z-scores.17 SGA was defined as a birth weight less than or equal to the tenth percentile (z-score ⩽ − 1.28), and for the purposes of this study was considered a surrogate measure of FGR.
Statistical analyses were conducted using SAS v.9.3 (SAS Institute, Cary, NC, USA). Because wide variability in AFL levels has been reported widely, AFL levels were analyzed both as concentration (pg ml − 1) and standardized to amniotic fluid protein levels (pg μg − 1). There were fewer samples per week after 26 weeks’ gestation, and thus later gestational ages required grouping for analysis. AFL levels were first grouped by attained full weeks of gestation at the time of measurement, except for the later weeks, which were grouped into 25 to 27, 28 to 32 and 33+ weeks. The nonparametric Jonckheere–Terptsra trend test was then used to test for differences in AFL levels across gestational age categories. Linear regression was used to estimate average weekly differences in AFL. To construct reference curves from the 321 AFL data points, the LMS method implemented in LMS chartmaker Light v. 2.54 (Harlow Healthcare, Tyne and Wear, UK) was used to summarize the skewness (L), median (M) and coefficient of variation (S) of AFL and leptin/protein by gestational age, which were then fitted as cubic splines using penalized maximum likelihood methods.19 Estimated smoothing degrees of freedom for L, M and S parameters were set at 3, 7.9 and 4, respectively, for AFL and at 4, 6 and 4, respectively, for leptin/protein. For the subset of individuals with medical record data available (n = 45), median and interquartile range (IQR) or numbers (percent) are reported to
AFL level Leptin was quantified by the enzyme-linked immunosorbent assay and the Bradford assay as percent total protein. A sandwich enzyme-linked immunosorbent assay specific for leptin was used as per manufacturer’s instructions (R&D Systems, Minneapolis, MN, USA). Plates were coated with 4 μg ml − 1 mouse anti-human leptin overnight at room temperature, then washed three times with 175 μl 0.05% tween/phosphate buffer saline and blocked with 1% bovine serum albumin/phosphate buffered saline for 1 h RT. Samples were added for 2 h at room temperature. Plates were then washed again and the secondary antibody (biotin mouse anti-human leptin) was added at 12.5 ng ml − 1. Plates were incubated for 1 h, washed
Table 1.
Placenta assessment
Description of all amniotic fluid samples and cohort subset with clinical data All amniotic fluid samples N with data
Sex (% M) Fetal leptin GA at amnio (weeks) AFL(pg ml − 1) AF protein (μg ml − 1) AFL/AF protein (pg μg − 1)
Median (IQR)
Patients with clinical data N with data 45
321 321 321 321
18.3 1567 4.8 323
(16.6–21.6) (1031–2210) (3.9–6.2) (183–468)
Median (IQR) or n (%) 23 (51%)
41 41 41 41
17.4 1571 4.5 344
Maternal pregnancy Maternal pre-pregnancy BMI Maternal pre-pregnancy overweight or obese (%) Maternal weight gain (lbs) Maternal BMI at delivery Maternal diabetes (% suspected or confirmed) Maternal pre-eclampsia (%)
39 39 39 42 42 43
26.8 21 31 30.9 5 1
Infant anthropometrics GA at delivery (weeks) Birth weight (g) Birth weight z-score Birth head circumference (cm) Birth head circumference z-score
45 45 45 26 26
39 3200 − 0.28 34.5 0.16
(16.4–20.7) (1156–1914) (3.6–5.4) (259–482) (21.6–33.2) (54%) (18, 37) (27.1–37.2) (12%) (2%) (38.4–39.6) (2840–3560) (−0.87, 0.45) (33–35.5) (−0.46, 0.63)
Abbreviations: AF, amniotic fluid; AFL, amniotic fluid leptin; BMI, body mass index; GA, gestational age at delivery; IQR, interquartile range; M, male.
Journal of Perinatology (2015), 1 – 6
© 2015 Nature America, Inc.
Standardization of amniotic fluid leptin M Scott-Finley et al
3
Figure 1. Distribution of AFL levels by gestational age. (a) Standards for AFL are presented as absolute concentration expressed as pg ml − 1 and (b) adjusting for total protein expressed as pg ng − 1. For both panels, median (solid line), 3rd, 10th, 25th, 75th, 90th and 97th percentiles are shown based on LMS modeling, allowing estimation of percentiles (see percentile data in Supplementary Tables 1 and 2).
describe the cohort. Spearman’s rank correlation, full or partially adjusted for gestational age at amniocentesis, was evaluated for pairs of continuous variables. Differences between categories of clinical characteristics were assessed using the nonparametric Wilcoxon rank sum test, or from general linear models including gestational age at amniocentesis as a covariate. Statistical significance was determined if Po0.05.
RESULTS Study population demographics A total of 348 amniotic fluid samples were obtained. A total of 6 patients were excluded from the study because of nondetectable AFL levels and another 21 were excluded because of clinical characteristics, including 13 because of prematurity, 4 because of twin gestation, 2 because of abnormal karyotype and 2 because of intrauterine demise, leaving a study cohort size of 321. The median gestational age at the time of amniocentesis was 18 weeks (IQR: 16 to 21 weeks; range: 14 to 36 weeks). Importantly, there were no statistically significant differences in AFL concentrations between the small subset with clinical data and the whole cohort. The clinical description of the cohort is presented in Table 1. AFL levels are lower at higher gestational ages The median (IQR) leptin level was 2133 pg ml − 1 (1703 to 4347) at 14 weeks’ gestation, but only 519 pg ml − 1 (380 to 761) by 33+ weeks (Figure 1). AFL levels normalized to total protein controlling for AF dilution showed a similar decreasing trend, with values of 743 pg per μg protein (612 to 893) at 14 weeks’ gestation and 132 pg per μg protein (77 to 336) by 33+ weeks. Regression analysis demonstrated that the mean AFL level differed by 103 ± 12 pg ml − 1, or 26.4 ± 3.2 pg μg − 1, correcting for total protein, each week of gestation. The nonparametric Jonckheere– Terptsra trend test showed that both decreasing trends were statistically significant (P o 0.0001) (Table 2). Because of this decreasing trend, we developed a standard reference for clinical evaluation of AFL concentrations by gestational age of measurement. Figure 1 and Supplementary Tables 1 and 2 present these reference curves with appropriate consideration of the differing distribution of AFL and leptin/ protein concentrations by gestational age, to allow for estimation of the AFL percentile relative to a normal population. Maternal overweight and obesity affect a majority of pregnancies Among those cases with medical record information (n = 26 to 45), five (12%) had confirmed diabetes and one (2%) had preeclampsia. The median (IQR) pre-pregnancy BMI was 26.8 kg m − 2 (21.6 to © 2015 Nature America, Inc.
33.2), and 21 (54%) were overweight or obese (Table 1). The median (IQR) BMI at the time of delivery was 30.9 kg m − 2 (27.1 to 37.2), with a median (IQR) weight gain of 31 pounds (18 to 37). Higher pre-pregnancy BMI was correlated with higher AFL levels after adjusting for timing of amniocentesis (partial r = 0.36, P = 0.03, Table 2), but weight gain during pregnancy was not significantly correlated with AFL. In addition, in this small sample, a history of GD, history of preeclampsia and maternal prepregnancy BMI category (normal vs overweight/obese) were not significantly associated with AFL levels (Table 3). In this small sample size with clinical data, we were unable to demonstrate a difference in AFL with maternal pre-pregnancy obesity. Fetal and infant characteristics in relation to AFL Among the 45 patients with clinical information available, the vast majority demonstrated normal fetal growth (Table 1). The median birth weight was 3200 g (2840 to 3560), and median birth weight z-score was − 0.28 (−0.87, 0.45), with 5 (12%) classified as SGA. Similarly, birth head circumference was in the normal range, with a median z-score of 0.16 (−0.46, 0.63). AFL was not significantly correlated with infant gestational age at delivery, infant birth weight (g or z-score) or infant head circumference (cm or z-score, Table 2) and did not differ by infant gender (Table 3). However, SGA infants had marginally higher AFL levels compared with nonSGA infants in this small sample (P = 0.07), adjusting for the timing of amniocentesis (Table 3). Leptin and leptin receptor content appear changed in placental tissue in maternal overweight and SGA Among the 45 patients with medical record information, 5 (11%) had placental tissue available for examination. Of these, three were from pregnancies with preexisting maternal obesity (BMI ⩾ 30), and two were associated with SGA (Table 4). In this exploratory view, placental leptin appeared to be related to AFL, but not to maternal pre-pregnancy BMI, although the subset was too limited for formal statistics. Leptin was expressed in the membranes, parenchyma and cord regions of the placenta, including the parenchymal fetal endothelial cells and, to a lesser extent, in syncytial cytotrophoblasts (Supplementary Figure 1). Leptin receptor protein localization was similar but qualitatively less abundant. In the context of maternal obesity, the presence of leptin increased significantly in the membranes and parenchyma, but when both maternal obesity and SGA were present leptin decreased significantly (Supplementary Figure 1), suggesting distinct patterns of leptin dysregulation that require further investigation. Journal of Perinatology (2015), 1 – 6
Standardization of amniotic fluid leptin M Scott-Finley et al
4 Table 2.
Correlations between amniotic fluid measures and clinical variables N with data
Leptin (pg ml − 1)
Leptin (pg ml − 1), partial correlationa
Protein (μg ml − 1)
Leptin/protein (pg μg − 1)
321 321 321 36 36 41 41 41 24 24
− 0.51** — 0.12* 0.23 0.03 0.15 − 0.07 − 0.10 0.11 0.12
— — 0.32** 0.36* − 0.04 0.12 − 0.14 − 0.12 0.13 0.16
0.27** 0.12* — 0.12 − 0.12 − 0.01 0.05 0.07 0.09 0.16
− 0.63** 0.78** − 0.45** 0.08 0.04 0.09 − 0.05 − 0.06 0.003 − 0.04
GA at amniocentesis AFL Amniotic fluid protein Pre-pregnancy BMI Gestational weight gain GA at delivery Birth weight Birth weight z-score Head circumference Head circumference z-score
Abbreviations: AFL, amniotic fluid leptin; BMI, body mass index; GA, gestational age at delivery. *P ⩽ 0.05; **Po0.001 aPartial correlations, adjusting for GA at amniocentesis.
Table 3.
Differences in amniotic leptin between categorical variables AFL (pg μl − 1)
N with data Infant sex Male Female
P-value Adjusted P-value
20 21
1457 (1089–1734) 0.20 1678 (1173–1992)
0.26
Maternal pre-pregnancy BMI Normal (BMI o25) 16 Overweight/obese 20 (BMI ⩾25)
1358 (1045–1734) 0.25 1620 (1182–2101)
0.15
GD No GD Suspected/ confirmed GD
34 5
1585 (1173–1959) 0.93 1794 (1140–1803)
0.58
36 5
1549 (1148, 1843) 0.32 1959 (1346, 1992)
0.07
Neonatal SGA No Yes
Abbreviations: AFL, amniotic fluid leptin; BMI, body mass index; GD, gestational diabetes; SGA, small for gestational age. Median (IQR) presented with P-values from Wilcoxon rank sum test. Adjusted P-values from general linear models including gestational age at amniocentesis as a covariate.
DISCUSSION In this study, we report significantly and progressively lower AFL levels at later gestational ages compared with earlier gestational ages, and provide a normal reference standard for this potentially important biomarker. Paradoxically, AFL levels are progressively lower with each week of gestational age throughout pregnancy and appear to plateau in the third trimester. Both pre-pregnancy BMI and delivery BMI (data not shown) were positively correlated with AFL levels after adjusting for the timing of amniocentesis. The prevalence of SGA was 12% at birth, and the AFL levels of SGA infants at the time of amniocentesis were marginally but not significantly higher than those of non-SGA infants (P = 0.07) in this small sample. One plausible explanation may be a higher likelihood of SGA in pregnancies complicated by maternal obesity. Although the results need to be interpreted with caution owing to the small sample size, our findings from the present study suggest that there are complex interactions between maternal and fetal physiology with regard to leptin in pregnancy tissues. Further studies are required to determine the relationship between leptin and SGA independent of maternal obesity. The functional role of leptin before birth, both as a regulator of fetal growth and as a potential biomarker of nutrition, is poorly understood. During pregnancy, leptin is produced by the placenta, Journal of Perinatology (2015), 1 – 6
amnion cells and fetal and maternal adipose tissues.20 Leptin cannot cross the placental barrier. Several studies report that maternal serum leptin levels are mainly placental in origin, fetal cord blood leptin levels are mainly from fetal adipose tissues, and AFL levels derive mainly from amnion cells.21,22 The majority of placental leptin production is released into the maternal circulation, whereas only a small proportion is delivered to the fetus due to its high molecular weight, resulting in ‘compartmentalization’ of leptin.14,21 Leptin in fetal serum is produced by fetal adipose, and cord blood leptin levels positively correlate with placental weight and incidence of FGR and negatively correlate with fetal obesity.15,22 In the present study, we show that AFL levels are progressively lower with advancing gestation. Chan et al. also reported that AFL levels decrease with advancing gestational age in uncomplicated pregnancies,23 and this decline contrasts with the gradual increases documented in maternal serum leptin levels with advancing gestational age.20 In addition, uncomplicated pregnancies are also marked by gradual increases in placental leptin protein and mRNA levels with advancing gestational age that plateau in the third trimester.20 Given the dynamic nature of AFL levels and the wide variability observed at any given point in time, using both an absolute and an indexed standard (Supplementary Tables 1 and 2) may provide added strength of interpretation. At this time, it is unclear which measure is the biologically relevant measure. Female fetuses are reported to have higher AFL levels compared with male fetuses,23 although this was not significant in the present study. This sexual dimorphism may suggest additional regulatory mechanisms of growth by sex that potentially modify studies that attempt to correlate AFL with serum leptin.24 Observed differences between the trajectory of AFL levels and maternal serum and placental protein leptin may be attributed to several potential factors, suggesting tissue-specific processes—for example, maternal nutritional adaptation and placental maturity. One of leptin’s main functions in human physiology is to help regulate satiety, and higher serum levels have been associated with greater adiposity.12 Several studies report that higher maternal and fetal serum leptin levels are associated with higher maternal weight gain, maternal BMI and fetal growth.25 Fifty percent of the mothers in the studied population had BMI 425, consistent with the growing obesity epidemic in the United States, and in this cohort maternal pre-pregnancy BMI was positively associated with AFL levels, after adjusting for gestational age at the time of amniocentesis. Oktem et al. investigated the relationship between fetal weight and leptin levels in maternal serum, amniotic fluid and the umbilical cord.26 The authors reported that, whereas cord leptin level correlated with birth weight and placental weight, AFL did not correlate with birth weight, consistent with our findings. However, both studies may have been underpowered to detect this relationship. © 2015 Nature America, Inc.
Standardization of amniotic fluid leptin M Scott-Finley et al
5 Table 4. Sample 1 2 3 4 5
Leptin in amniotic fluid and placenta tissue in relation to birth weight Pre-preg BMI
BWT (g)
BWT (percentile)
GA
Gender
AFL
Placental leptin
Placental WT (g)
44 32 26 24 33
1870 2300 3110 3515 3910
1 2 30 40 85
385 380 375 411 384
F F F F F
805 1349 1750 2089 1894
+ + ++ ++ +++
277 345 513 421 799
Abbreviations: AFL, amniotic fluid leptin; BMI, body mass index; BWT, birth weight; F, female; GA, gestational age at delivery; g, grams; WT, weight.
The present study does provide some evidence for an association of placental leptin in SGA, suggesting an active role in fetal growth. Similar to our findings, Lea et al.27 reported that FGR placentas had less immunostaining of leptin protein in placental syncytiotrophoblast and endothelial cells. Indeed, one study comparing release of leptin of explanted gestational tissues (placental, amnion and choriodecidua) and subcutaneous adipose and skeletal muscle tissues of pregnant women with and without GD revealed that leptin release from GD mellitus gestational tissues was lower than that from nondiabetic women, whereas GD mellitus mother’s adipose and muscle tissue release was higher than that of nondiabetic women.28 This suggests that leptin levels in maternal serum, fetal serum, amniotic fluid and placenta could represent different markers for interrelated but distinct developmental processes during pregnancy. Misra et al. studied the longitudinal effect of pre-pregnancy BMI on maternal serum leptin levels from the early first trimester to the third trimester in a cohort of 143 adult gravidas with singleton pregnancies.29 Pre-pregnancy overweight/obese women had significantly higher serum leptin concentrations compared with their normal BMI counterparts. This is supported by the work of Catalano et al., who demonstrated that fetuses from pregnant obese mothers had higher birth weight, higher percent body fat and elevated cord leptin levels as compared with lean pregnant women.30 Such factors could influence the fetal intrauterine environment, leading to adverse pregnancy outcomes, such as fetal growth abnormalities, as well as fetal adaption leading to increased rates of disease in offspring, such as obesity and diabetes. Further study examining the connection between fetal programming and specific mechanisms of growth perturbation is warranted.7,8,31 This study has several strengths. Primarily, this study shows that AFL levels are detectable and measurable using a simple assay, and appropriate samples are relatively easily obtained by amniocentesis. Although our duplicate measures suggest sensitivity and reliability, validation studies are needed. In addition, the large number of samples from many gestational ages allowed for the development of a standard for AFL levels by gestational age. This study also has weaknesses, including a relatively small sample size with medical record information or placental tissue. The subset cohort with medical record information may represent a highly select patient population; therefore, the generalizability of results may be limited. Although the small subset with clinical data limits the power of this study, AFL values and gestational age of amniocentesis were similar to the larger cohort, suggesting limited bias in results. We did attempt to obtain length information as a more accurate reflection of overall growth; however, we were unable to obtain birth length measurements from the infant records to add to the study. The interpretation of placental findings from the current study is exploratory, ultimately identifying the need for further investigation. Studies correlating amniotic fluid and placenta findings are important, in part because of the overall decrease in amniocentesis for any reason. Finally, this study involves any inherent biases resulting from our ability only to study women clinically requiring amniocentesis. © 2015 Nature America, Inc.
Further studies are necessary to determine the relationship between AFL levels, placental development and fetal growth. For example, gene expression studies may accelerate our understanding of mechanism.32 The role of leptin in placental and fetal growth and the potential utility of AFL in pregnancy management are promising and require further study. In summary, these results are important in assessing the impact of AFL levels to maternal and fetal health outcomes. As less invasive measures are developed, such as maternal serum leptin, the use of amniocentesis may continue to decrease, identifying a need for correlation between measures.33 Efforts to combine genetic information with clinical biomarkers, including potentially nutritional assessments, promises to improve the prediction and treatment of fetal growth abnormalities.34–36 Leptin has the potential of being a biomarker that can predict abnormal patterns of fetal growth, the potential to refine prognosis and better predict neonatal outcomes, potentially impacting morbidity and mortality. CONFLICT OF INTEREST The authors declare no conflict of interest.
ACKNOWLEDGMENTS We thank Michelle Faust (Heart Institute Research Core) and Peggy Walsh (Hatton Research Center) for their assistance. Funding. We disclosed receipt of the following financial support for the research, authorship and/or publication of this article: the Cincinnati Children’s Research Foundation (RBH).
REFERENCES 1 Bernstein I, Gabbe SG. Intrauterine growth restriction In: Gabbe SG, Niebyl JR, Simpson JL, Annas GJ et al. (eds). Obstetrics: Normal and Problem Pregnancies6th ednElsevier: New York, NY, USA, 2012 pp 857–865. 2 Creasy RK, Resnik R. Intrauterine growth restriction In: Creasy RK, Resnik R (eds). Maternal-Fetal Medicine: Principles and Practice 4th edn. Saunders: Philadelphia, PA, USA, 2004 pp 495–512. 3 Alkalay AL, Graham JM, Pomerance JJ. Evaluation of neonates born with intrauterine growth retardation: review and practice guidelines. J Perinatol 1998; 18: 142–151. 4 McIntire DD, Bloom SL, Casey BM, Leveno KJ. Birth weight in relation to morbidity and mortality among newborn infants. N Engl J Med 1999; 340: 1234–1238. 5 Godfrey KM, Barker DJ. Fetal programming and adult health. Public Health Nutr 2001; 4: 611–624. 6 Barker DJ, Hales CN, Fall CH, Osmond C, Phipps K, Clark PM. Type 2 (non-insulindependent) diabetes mellitus, hypertension and hyperlipidaemia (syndrome X): relation to reduced fetal growth. Diabetologia 1993; 36: 62–67. 7 Neerhof MG. Causes of intrauterine growth restriction. Clin Perinatol 1995; 22: 375–385. 8 Szentpéteri I, Rab A, Kornya L, Kovács P, Joó JG. Gene expression patterns of vascular endothelial growth factor (VEGF-A) in human placenta from pregnancies with intrauterine growth restriction. J Matern Fetal Neonatal Med 2013; 26: 984–989. 9 Vrachnis N, Kalampokas E, Sifakis S, Vitoratos N, Kalampokas T, Botsis D et al. Placental growth factor (PlGF): a key to optimizing fetal growth. J Matern Fetal Neonatal Med 2013; 26: 995–1002. 10 Briana DD, Malamitsi-Puchner A. Intrauterine growth restriction and adult disease: the role of adipocytokines. Eur J Endocrinol 2009; 160: 337–347.
Journal of Perinatology (2015), 1 – 6
Standardization of amniotic fluid leptin M Scott-Finley et al
6 11 Ornoy A. Biomarkers of maternal diabetes and its complication in pregnancy. Reprod Toxicol 2012; 34: 174–179. 12 Ahima RS, Flier JS. Leptin. Annu Rev Physiol 2000; 62: 413–437. 13 Popovic V, Casanueva FF. Leptin, nutrition and reproduction: new insights. Hormones 2002; 1: 204–217. 14 Senaris R, Garcia-Caballero T, Casabiell X, Gallego R, Castro R, Considine RV et al. Synthesis of leptin in human placenta. Endocrinology 1997; 138: 4501–4504. 15 Lepercq J, Challier JC, Guerre-Millo M, Cauzac M, Vidal H, Hauguel-de Mouzon S. Prenatal leptin production: evidence that fetal adipose tissue produces leptin. J Clin Endocrinol Metab 2001; 86: 2409–2413. 16 Rasmussen KM, Yaktine AL (eds). Institute of Medicine (US) and National Research Council (US) Committee to Reexamine IOM Pregnancy Weight Guidelines. National Academies Press (US): Washington (DC), CO, USA, 2009. 17 Olsen IE, Groveman SA, Lawson ML, Clark RH, Zemel BS. New intrauterine growth curves based on United States data. Pediatrics 2010; 125: e214–e224. 18 Alexopoulos A, Bravou V, Peroukides S, Kaklamanis L, Varakis J, Alexopoulos D et al. Bone regulatory factors NFATc1 and Osterix in human calcific aortic valves. Int J Cardiol 2010; 139: 142–149. 19 Cole TJ, Green PJ. Smoothing reference centile curves: the LMS method and penalized likelihood. Stat Med 1992; 11: 1305–1319. 20 Henson MC, Castracane VD. Leptin in pregnancy: an update. Biol Reprod 2006; 74: 218–229. 21 Masuzaki H, Ogawa Y, Sagawa N, Hosoda K, Matsumoto T, Mise H et al. Nonadipose tissue production of leptin: leptin as a novel placenta derived hormone in humans. Nat Med 1997; 3: 1029–1033. 22 Karakosta P, Chatzi L, Plana E, Margioris A, Castanas E, Kogevinas M. Leptin levels in cord blood and anthropometric measures at birth: a systematic review and meta-analysis. Paediatr Perinat Epidemiol 2011; 25: 150–163. 23 Chan TF, Su JH, Chung YF, Hsu YH, Yeh YT, Yuan SS. Elevated amniotic fluid leptin levels in pregnant women who are destined to develop preeclampsia. Acta Obstet Gynecol Scand 2006; 85: 171–174. 24 Cagnacci A, Arangino S, Caretto S, Mazza V, Volpe A. Sexual dimorphism in the levels of amniotic fluid leptin in pregnancies at 16 weeks of gestation: relation to fetal growth. Eur J Obstet Gynecol Reprod Biol 2006; 124: 53–57.
25 Tessier DR, Ferraro ZM, Gruslin A. Role of leptin in pregnancy: Consequences of maternal obesity. Placenta 2013; 34: 205–211. 26 Oktem O, Dedeoğlu N, Oymak Y, Sezen D, Köksal L, Pekin T et al. Maternal serum, amniotic fluid and cord leptin levels at term: their correlations with fetal weight. J Perinat Med 2004; 32: 266–271. 27 Lea RG, Howe D, Hannah LT, Bonneau O, Hunter L, Hoggard N. Placental leptin in normal, diabetic and fetal growth-retarded pregnancies. Mol Hum Reprod 2000; 6: 763–769. 28 Lappas M, Permezel M, Rice GE. Leptin and adiponectin stimulate the release of proinflammatory cytokines and prostaglandins from human placenta and maternal adipose tissue via nuclear factor-kappaB, peroxisomal proliferatoractivated receptor-gamma and extracellularly regulated kinase 1/2. Endocrinology 2005; 146: 3334–3342. 29 Misra VK, Trudeau S. The influence of overweight and obesity on longitudinal trends in maternal serum leptin levels during pregnancy. Obesity 2011; 19: 416e21. 30 Catalano PM, Presley L, Minium J, Hauguel-de Mouzon S. Fetuses of obese mothers develop insulin resistance in utero. Diabetes Care 2009; 32: 1076e80. 31 Alexe DM, Syridou G, Petridou ET. Determinants of early life leptin levels and later life degenerative outcomes. Clinl Med Res 2006; 4: 326–335. 32 Nagy GR, Gyõrffy B, Galamb O, Molnár B, Nagy B, Papp Z. Use of routinely collected amniotic fluid for whole-genome expression analysis of polygenic disorders. Clin Chem 2006; 52: 2013–2020. 33 Misra VK, Straughen JK, Trudeau S. Maternal serum leptin during pregnancy and infant birth weight: the influence of maternal overweight and obesity. Obesity (Silver Spring) 2013; 21: 1064–1069. 34 Molvarec A, Szarka A, Walentin S, Beko G, Karádi I, Prohászka Z et al. Serum leptin levels in relation to circulating cytokines, chemokines, adhesion molecules and angiogenic factors in normal pregnancy and preeclampsia. Reprod Biol Endocrinol 2011; 9: 124. 35 Bloomfield FH, Oliver MH, Harding JE. The late effects of fetal growth patterns. Arch Dis Child Fetal Neonatal Ed 2006; 91: F299–F304. 36 McCormack RT, Armstrong J, Leonard D. Codevelopment of genome-based therapeutics and companion diagnostics: insights from an Institute of Medicine roundtable. JAMA 2014; 311: 1395–1396; Erratum in: JAMA 2014; 311: 1396.
Supplementary Information accompanies the paper on the Journal of Perinatology website (http://www.nature.com/jp)
Journal of Perinatology (2015), 1 – 6
© 2015 Nature America, Inc.
ORIGINAL ARTICLE
Standardization of amniotic fluid leptin levels and utility in maternal overweight and fetal undergrowth M Scott-Finley1,6, JG Woo2,3,6, M Habli1,2, O Ramos-Gonzales2, JF Cnota2, Y Wang3, BD Kamath-Rayne4, AC Hinton1, WJ Polzin1,5, TM Crombleholme5 and RB Hinton2 OBJECTIVE: Leptin is an adipokine that regulates energy homeostasis. The objective of this study was to establish a gestational age-specific standard for amniotic fluid leptin (AFL) levels and examine the relationship between AFL, maternal overweight and fetal growth restriction. STUDY DESIGN: Amniotic fluid was obtained at mid-gestation from singleton gravidas, and leptin was quantified using enzymelinked immunosorbent assay. Amniotic fluid samples from 321 term pregnancies were analyzed. Clinical data, including fetal ultrasound measurements and maternal and infant characteristics, were available for a subset of patients (n = 45). RESULTS: The median interquartile range AFL level was significantly higher at 14 weeks’ gestation (2133 pg ml − 1 (1703 to 4347)) than after 33 weeks’ gestation (519 pg ml − 1 (380 to 761), P trend o0.0001), an average difference of 102 pg ml − 1 per week. AFL levels were positively correlated with maternal pre-pregnancy body mass index (BMI) (r = 0.36, P = 0.03) adjusting for gestational age at measurement, but were not associated with fetal growth. CONCLUSIONS: AFL levels are higher at mid-gestation than at late gestation, and are associated with maternal pre-pregnancy BMI. Journal of Perinatology advance online publication, 30 April 2015; doi:10.1038/jp.2015.39
INTRODUCTION Fetal growth restriction (FGR) is among the most significant problems in perinatal medicine. It is the second leading cause of perinatal morbidity and mortality, exceeded only by prematurity.1 FGR is defined as an estimated fetal weight less than the tenth percentile for gestational age and corresponds with small for gestational age (SGA), which is defined as a birth weight less than the tenth percentile for gestational age.2 Perinatal morbidity and mortality is significantly increased in the presence of low birth weight for gestational age.3,4 In addition, FGR infants are at increased risk for a variety of long-term sequelae, including adult-onset metabolic diseases such as type 2 diabetes and hypertension.5,6 Despite the immediate and long-term importance of FGR, there are presently no predictive biomarkers or preventive treatment strategies. Growth abnormalities in the fetus may be caused by numerous insults that are generally classified as maternal, fetal or placental in origin and include maternal preeclampsia, fetal genetic anomalies and primary placental vasculopathies. However, the shared pathogenesis of FGR is poorly understood. Placental insufficiency may be a common final pathway for FGR, but predicting placental function is challenging.7 Several growth factors and signaling molecules have been implicated in FGR, including vascular endothelial growth factor, placental growth factor and leptin.8–10 Leptin is an adipokine that regulates energy homeostasis and is increased in obesity (anabolism) and decreased in cachexia (catabolism),11,12 and leptin is present in fetal tissue, including the placenta, suggesting a direct role for leptin in fetal growth and
development.13–15 However, amniotic fluid leptin (AFL) has not been standardized or examined systematically during pregnancy. Little is known about how AFL levels change during pregnancy and relate to fetal growth. Understanding leptin dysregulation may elucidate mechanisms of FGR at the intersection of maternal health and fetal growth. In this study, we examined AFL in a crosssectional manner across gestational ages with the goal of establishing a gestational age-specific standard. Because maternal weight has been associated with leptin regulation, we hypothesized that increased pre-pregnancy maternal body mass index (BMI) is associated with increased AFL levels at mid-gestation. These observations have important clinical implications and may lead to applications that improve the diagnostic and prognostic assessment of fetal growth. METHODS Study population This study was approved by the Institutional Review Boards of Cincinnati Children’s Hospital Medical Center and Good Samaritan Hospital (Cincinnati, OH. USA). Amniotic fluid was obtained from pregnancies at both community and referral hospitals that underwent a genetic amniocentesis at any time during gestation; however, the specific indications for testing are unknown because the study used discarded tissue that was anonymized. Inclusion criteria were a normal karyotype analysis and the absence of fetal anomalies by ultrasound anatomic survey. Exclusion criteria included prematurity ( o37 weeks’ gestation), multiple gestation pregnancy, and fetal or neonatal demise. Patients were enrolled from May 2009 to October 2010. Detailed medical record information was reviewed in a small subset of high-risk cases from a single center owing to
1 Division of Maternal Fetal Medicine, Obstetrics and Gynecology, Good Samaritan Hospital, Cincinnati, OH, USA; 2Division of Cardiology, The Heart Institute, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA; 3Division of Biostatistics and Epidemiology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA; 4Perinatal Institute, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA and 5Fetal Care Center, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA. Correspondence: Dr RB Hinton, Division of Cardiology, The Heart Institute, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Ave, MLC 2003, Cincinnati, OH 45229, USA. E-mail: [email protected] 6 These authors contributed equally to this work. Received 3 December 2014; Received 23 March 2015; accepted 24 March 2015
Standardization of amniotic fluid leptin M Scott-Finley et al
2 the fact that the majority of cases were derived from referring hospitals where information was not accessible. Informed consent was obtained from this subset of patients. Archived placental tissue was identified in a subset of these cases.
Maternal health
again and streptavidin–HRP was added for 20 min, after which a 1:1 mixture of colors A and B was added. Plates were stopped with 2 N H2SO4 and read at 450 nm minus 595 nm background. Values are expressed in pg ml − 1. The Bradford assay was employed by adding 200 μl 1X solution (R&D Systems) to a plate, followed by 5 μl per sample. Plates were read at 595 nm. Values were expressed as μg ml − 1.
Clinical records were reviewed for a subset (n = 45) with available data, with the following data collected: maternal age, calculated pre-pregnancy BMI from self-reported weight and measured height, BMI at delivery, and weight gain during pregnancy. In addition, medical history of preexisting diabetes mellitus or gestational diabetes (GD) and preeclampsia was noted. Gestational diabetes was identified by an abnormal diagnostic 100-g 3-hour oral glucose test defined as two elevated values. Finally, estimated fetal weight by ultrasound within 2 weeks of amniocentesis, gestational age at the time of amniocentesis, gestational age at the time of delivery, birth weight, birth head circumference and infant gender were recorded. For the purposes of analysis, GD was defined as either an abnormal glucose-tolerance test or confirmed clinical diagnosis of GD. Maternal pre-pregnancy BMI (in kg m − 2) was classified as normal weight (BMI o 25) or overweight/obese (BMI ⩾ 25).16
Histochemistry and immunohistochemistry were used for exploratory studies to assess placental tissue, which was obtained from 5/45 (11%) pregnancies from the subset of patients with medical record information in the course of routine pathological processes. Antibodies directed against leptin (1:500, Lifespan Biosciences, Seattle, WA, USA), and leptin receptor (1:50, Abcam, Cambridge, MA, USA) were used to assess the location and relative amount of protein. The leptin antibodies required additional citrate pretreatment. A semiquantitative scale was used to assess the expression.18 Briefly, staining was graded on a scale of 0 to 3 based on the percentage of positive cells as follows: 0 = immunoreactivity in o10% of cells; 1 = 10 to 35%; 2 = 35 to 70%; and 3 = 470%. Quantification was performed by a single blinded reader.
Fetal growth
Statistical analyses
Neonatal anthropometrics included birth weight and head circumference. Growth z-scores were derived from the Olsen standard, a contemporary set of growth curves derived from a large, racially diverse USA sample, resulting in gestational age- and gender-specific z-scores.17 SGA was defined as a birth weight less than or equal to the tenth percentile (z-score ⩽ − 1.28), and for the purposes of this study was considered a surrogate measure of FGR.
Statistical analyses were conducted using SAS v.9.3 (SAS Institute, Cary, NC, USA). Because wide variability in AFL levels has been reported widely, AFL levels were analyzed both as concentration (pg ml − 1) and standardized to amniotic fluid protein levels (pg μg − 1). There were fewer samples per week after 26 weeks’ gestation, and thus later gestational ages required grouping for analysis. AFL levels were first grouped by attained full weeks of gestation at the time of measurement, except for the later weeks, which were grouped into 25 to 27, 28 to 32 and 33+ weeks. The nonparametric Jonckheere–Terptsra trend test was then used to test for differences in AFL levels across gestational age categories. Linear regression was used to estimate average weekly differences in AFL. To construct reference curves from the 321 AFL data points, the LMS method implemented in LMS chartmaker Light v. 2.54 (Harlow Healthcare, Tyne and Wear, UK) was used to summarize the skewness (L), median (M) and coefficient of variation (S) of AFL and leptin/protein by gestational age, which were then fitted as cubic splines using penalized maximum likelihood methods.19 Estimated smoothing degrees of freedom for L, M and S parameters were set at 3, 7.9 and 4, respectively, for AFL and at 4, 6 and 4, respectively, for leptin/protein. For the subset of individuals with medical record data available (n = 45), median and interquartile range (IQR) or numbers (percent) are reported to
AFL level Leptin was quantified by the enzyme-linked immunosorbent assay and the Bradford assay as percent total protein. A sandwich enzyme-linked immunosorbent assay specific for leptin was used as per manufacturer’s instructions (R&D Systems, Minneapolis, MN, USA). Plates were coated with 4 μg ml − 1 mouse anti-human leptin overnight at room temperature, then washed three times with 175 μl 0.05% tween/phosphate buffer saline and blocked with 1% bovine serum albumin/phosphate buffered saline for 1 h RT. Samples were added for 2 h at room temperature. Plates were then washed again and the secondary antibody (biotin mouse anti-human leptin) was added at 12.5 ng ml − 1. Plates were incubated for 1 h, washed
Table 1.
Placenta assessment
Description of all amniotic fluid samples and cohort subset with clinical data All amniotic fluid samples N with data
Sex (% M) Fetal leptin GA at amnio (weeks) AFL(pg ml − 1) AF protein (μg ml − 1) AFL/AF protein (pg μg − 1)
Median (IQR)
Patients with clinical data N with data 45
321 321 321 321
18.3 1567 4.8 323
(16.6–21.6) (1031–2210) (3.9–6.2) (183–468)
Median (IQR) or n (%) 23 (51%)
41 41 41 41
17.4 1571 4.5 344
Maternal pregnancy Maternal pre-pregnancy BMI Maternal pre-pregnancy overweight or obese (%) Maternal weight gain (lbs) Maternal BMI at delivery Maternal diabetes (% suspected or confirmed) Maternal pre-eclampsia (%)
39 39 39 42 42 43
26.8 21 31 30.9 5 1
Infant anthropometrics GA at delivery (weeks) Birth weight (g) Birth weight z-score Birth head circumference (cm) Birth head circumference z-score
45 45 45 26 26
39 3200 − 0.28 34.5 0.16
(16.4–20.7) (1156–1914) (3.6–5.4) (259–482) (21.6–33.2) (54%) (18, 37) (27.1–37.2) (12%) (2%) (38.4–39.6) (2840–3560) (−0.87, 0.45) (33–35.5) (−0.46, 0.63)
Abbreviations: AF, amniotic fluid; AFL, amniotic fluid leptin; BMI, body mass index; GA, gestational age at delivery; IQR, interquartile range; M, male.
Journal of Perinatology (2015), 1 – 6
© 2015 Nature America, Inc.
Standardization of amniotic fluid leptin M Scott-Finley et al
3
Figure 1. Distribution of AFL levels by gestational age. (a) Standards for AFL are presented as absolute concentration expressed as pg ml − 1 and (b) adjusting for total protein expressed as pg ng − 1. For both panels, median (solid line), 3rd, 10th, 25th, 75th, 90th and 97th percentiles are shown based on LMS modeling, allowing estimation of percentiles (see percentile data in Supplementary Tables 1 and 2).
describe the cohort. Spearman’s rank correlation, full or partially adjusted for gestational age at amniocentesis, was evaluated for pairs of continuous variables. Differences between categories of clinical characteristics were assessed using the nonparametric Wilcoxon rank sum test, or from general linear models including gestational age at amniocentesis as a covariate. Statistical significance was determined if Po0.05.
RESULTS Study population demographics A total of 348 amniotic fluid samples were obtained. A total of 6 patients were excluded from the study because of nondetectable AFL levels and another 21 were excluded because of clinical characteristics, including 13 because of prematurity, 4 because of twin gestation, 2 because of abnormal karyotype and 2 because of intrauterine demise, leaving a study cohort size of 321. The median gestational age at the time of amniocentesis was 18 weeks (IQR: 16 to 21 weeks; range: 14 to 36 weeks). Importantly, there were no statistically significant differences in AFL concentrations between the small subset with clinical data and the whole cohort. The clinical description of the cohort is presented in Table 1. AFL levels are lower at higher gestational ages The median (IQR) leptin level was 2133 pg ml − 1 (1703 to 4347) at 14 weeks’ gestation, but only 519 pg ml − 1 (380 to 761) by 33+ weeks (Figure 1). AFL levels normalized to total protein controlling for AF dilution showed a similar decreasing trend, with values of 743 pg per μg protein (612 to 893) at 14 weeks’ gestation and 132 pg per μg protein (77 to 336) by 33+ weeks. Regression analysis demonstrated that the mean AFL level differed by 103 ± 12 pg ml − 1, or 26.4 ± 3.2 pg μg − 1, correcting for total protein, each week of gestation. The nonparametric Jonckheere– Terptsra trend test showed that both decreasing trends were statistically significant (P o 0.0001) (Table 2). Because of this decreasing trend, we developed a standard reference for clinical evaluation of AFL concentrations by gestational age of measurement. Figure 1 and Supplementary Tables 1 and 2 present these reference curves with appropriate consideration of the differing distribution of AFL and leptin/ protein concentrations by gestational age, to allow for estimation of the AFL percentile relative to a normal population. Maternal overweight and obesity affect a majority of pregnancies Among those cases with medical record information (n = 26 to 45), five (12%) had confirmed diabetes and one (2%) had preeclampsia. The median (IQR) pre-pregnancy BMI was 26.8 kg m − 2 (21.6 to © 2015 Nature America, Inc.
33.2), and 21 (54%) were overweight or obese (Table 1). The median (IQR) BMI at the time of delivery was 30.9 kg m − 2 (27.1 to 37.2), with a median (IQR) weight gain of 31 pounds (18 to 37). Higher pre-pregnancy BMI was correlated with higher AFL levels after adjusting for timing of amniocentesis (partial r = 0.36, P = 0.03, Table 2), but weight gain during pregnancy was not significantly correlated with AFL. In addition, in this small sample, a history of GD, history of preeclampsia and maternal prepregnancy BMI category (normal vs overweight/obese) were not significantly associated with AFL levels (Table 3). In this small sample size with clinical data, we were unable to demonstrate a difference in AFL with maternal pre-pregnancy obesity. Fetal and infant characteristics in relation to AFL Among the 45 patients with clinical information available, the vast majority demonstrated normal fetal growth (Table 1). The median birth weight was 3200 g (2840 to 3560), and median birth weight z-score was − 0.28 (−0.87, 0.45), with 5 (12%) classified as SGA. Similarly, birth head circumference was in the normal range, with a median z-score of 0.16 (−0.46, 0.63). AFL was not significantly correlated with infant gestational age at delivery, infant birth weight (g or z-score) or infant head circumference (cm or z-score, Table 2) and did not differ by infant gender (Table 3). However, SGA infants had marginally higher AFL levels compared with nonSGA infants in this small sample (P = 0.07), adjusting for the timing of amniocentesis (Table 3). Leptin and leptin receptor content appear changed in placental tissue in maternal overweight and SGA Among the 45 patients with medical record information, 5 (11%) had placental tissue available for examination. Of these, three were from pregnancies with preexisting maternal obesity (BMI ⩾ 30), and two were associated with SGA (Table 4). In this exploratory view, placental leptin appeared to be related to AFL, but not to maternal pre-pregnancy BMI, although the subset was too limited for formal statistics. Leptin was expressed in the membranes, parenchyma and cord regions of the placenta, including the parenchymal fetal endothelial cells and, to a lesser extent, in syncytial cytotrophoblasts (Supplementary Figure 1). Leptin receptor protein localization was similar but qualitatively less abundant. In the context of maternal obesity, the presence of leptin increased significantly in the membranes and parenchyma, but when both maternal obesity and SGA were present leptin decreased significantly (Supplementary Figure 1), suggesting distinct patterns of leptin dysregulation that require further investigation. Journal of Perinatology (2015), 1 – 6
Standardization of amniotic fluid leptin M Scott-Finley et al
4 Table 2.
Correlations between amniotic fluid measures and clinical variables N with data
Leptin (pg ml − 1)
Leptin (pg ml − 1), partial correlationa
Protein (μg ml − 1)
Leptin/protein (pg μg − 1)
321 321 321 36 36 41 41 41 24 24
− 0.51** — 0.12* 0.23 0.03 0.15 − 0.07 − 0.10 0.11 0.12
— — 0.32** 0.36* − 0.04 0.12 − 0.14 − 0.12 0.13 0.16
0.27** 0.12* — 0.12 − 0.12 − 0.01 0.05 0.07 0.09 0.16
− 0.63** 0.78** − 0.45** 0.08 0.04 0.09 − 0.05 − 0.06 0.003 − 0.04
GA at amniocentesis AFL Amniotic fluid protein Pre-pregnancy BMI Gestational weight gain GA at delivery Birth weight Birth weight z-score Head circumference Head circumference z-score
Abbreviations: AFL, amniotic fluid leptin; BMI, body mass index; GA, gestational age at delivery. *P ⩽ 0.05; **Po0.001 aPartial correlations, adjusting for GA at amniocentesis.
Table 3.
Differences in amniotic leptin between categorical variables AFL (pg μl − 1)
N with data Infant sex Male Female
P-value Adjusted P-value
20 21
1457 (1089–1734) 0.20 1678 (1173–1992)
0.26
Maternal pre-pregnancy BMI Normal (BMI o25) 16 Overweight/obese 20 (BMI ⩾25)
1358 (1045–1734) 0.25 1620 (1182–2101)
0.15
GD No GD Suspected/ confirmed GD
34 5
1585 (1173–1959) 0.93 1794 (1140–1803)
0.58
36 5
1549 (1148, 1843) 0.32 1959 (1346, 1992)
0.07
Neonatal SGA No Yes
Abbreviations: AFL, amniotic fluid leptin; BMI, body mass index; GD, gestational diabetes; SGA, small for gestational age. Median (IQR) presented with P-values from Wilcoxon rank sum test. Adjusted P-values from general linear models including gestational age at amniocentesis as a covariate.
DISCUSSION In this study, we report significantly and progressively lower AFL levels at later gestational ages compared with earlier gestational ages, and provide a normal reference standard for this potentially important biomarker. Paradoxically, AFL levels are progressively lower with each week of gestational age throughout pregnancy and appear to plateau in the third trimester. Both pre-pregnancy BMI and delivery BMI (data not shown) were positively correlated with AFL levels after adjusting for the timing of amniocentesis. The prevalence of SGA was 12% at birth, and the AFL levels of SGA infants at the time of amniocentesis were marginally but not significantly higher than those of non-SGA infants (P = 0.07) in this small sample. One plausible explanation may be a higher likelihood of SGA in pregnancies complicated by maternal obesity. Although the results need to be interpreted with caution owing to the small sample size, our findings from the present study suggest that there are complex interactions between maternal and fetal physiology with regard to leptin in pregnancy tissues. Further studies are required to determine the relationship between leptin and SGA independent of maternal obesity. The functional role of leptin before birth, both as a regulator of fetal growth and as a potential biomarker of nutrition, is poorly understood. During pregnancy, leptin is produced by the placenta, Journal of Perinatology (2015), 1 – 6
amnion cells and fetal and maternal adipose tissues.20 Leptin cannot cross the placental barrier. Several studies report that maternal serum leptin levels are mainly placental in origin, fetal cord blood leptin levels are mainly from fetal adipose tissues, and AFL levels derive mainly from amnion cells.21,22 The majority of placental leptin production is released into the maternal circulation, whereas only a small proportion is delivered to the fetus due to its high molecular weight, resulting in ‘compartmentalization’ of leptin.14,21 Leptin in fetal serum is produced by fetal adipose, and cord blood leptin levels positively correlate with placental weight and incidence of FGR and negatively correlate with fetal obesity.15,22 In the present study, we show that AFL levels are progressively lower with advancing gestation. Chan et al. also reported that AFL levels decrease with advancing gestational age in uncomplicated pregnancies,23 and this decline contrasts with the gradual increases documented in maternal serum leptin levels with advancing gestational age.20 In addition, uncomplicated pregnancies are also marked by gradual increases in placental leptin protein and mRNA levels with advancing gestational age that plateau in the third trimester.20 Given the dynamic nature of AFL levels and the wide variability observed at any given point in time, using both an absolute and an indexed standard (Supplementary Tables 1 and 2) may provide added strength of interpretation. At this time, it is unclear which measure is the biologically relevant measure. Female fetuses are reported to have higher AFL levels compared with male fetuses,23 although this was not significant in the present study. This sexual dimorphism may suggest additional regulatory mechanisms of growth by sex that potentially modify studies that attempt to correlate AFL with serum leptin.24 Observed differences between the trajectory of AFL levels and maternal serum and placental protein leptin may be attributed to several potential factors, suggesting tissue-specific processes—for example, maternal nutritional adaptation and placental maturity. One of leptin’s main functions in human physiology is to help regulate satiety, and higher serum levels have been associated with greater adiposity.12 Several studies report that higher maternal and fetal serum leptin levels are associated with higher maternal weight gain, maternal BMI and fetal growth.25 Fifty percent of the mothers in the studied population had BMI 425, consistent with the growing obesity epidemic in the United States, and in this cohort maternal pre-pregnancy BMI was positively associated with AFL levels, after adjusting for gestational age at the time of amniocentesis. Oktem et al. investigated the relationship between fetal weight and leptin levels in maternal serum, amniotic fluid and the umbilical cord.26 The authors reported that, whereas cord leptin level correlated with birth weight and placental weight, AFL did not correlate with birth weight, consistent with our findings. However, both studies may have been underpowered to detect this relationship. © 2015 Nature America, Inc.
Standardization of amniotic fluid leptin M Scott-Finley et al
5 Table 4. Sample 1 2 3 4 5
Leptin in amniotic fluid and placenta tissue in relation to birth weight Pre-preg BMI
BWT (g)
BWT (percentile)
GA
Gender
AFL
Placental leptin
Placental WT (g)
44 32 26 24 33
1870 2300 3110 3515 3910
1 2 30 40 85
385 380 375 411 384
F F F F F
805 1349 1750 2089 1894
+ + ++ ++ +++
277 345 513 421 799
Abbreviations: AFL, amniotic fluid leptin; BMI, body mass index; BWT, birth weight; F, female; GA, gestational age at delivery; g, grams; WT, weight.
The present study does provide some evidence for an association of placental leptin in SGA, suggesting an active role in fetal growth. Similar to our findings, Lea et al.27 reported that FGR placentas had less immunostaining of leptin protein in placental syncytiotrophoblast and endothelial cells. Indeed, one study comparing release of leptin of explanted gestational tissues (placental, amnion and choriodecidua) and subcutaneous adipose and skeletal muscle tissues of pregnant women with and without GD revealed that leptin release from GD mellitus gestational tissues was lower than that from nondiabetic women, whereas GD mellitus mother’s adipose and muscle tissue release was higher than that of nondiabetic women.28 This suggests that leptin levels in maternal serum, fetal serum, amniotic fluid and placenta could represent different markers for interrelated but distinct developmental processes during pregnancy. Misra et al. studied the longitudinal effect of pre-pregnancy BMI on maternal serum leptin levels from the early first trimester to the third trimester in a cohort of 143 adult gravidas with singleton pregnancies.29 Pre-pregnancy overweight/obese women had significantly higher serum leptin concentrations compared with their normal BMI counterparts. This is supported by the work of Catalano et al., who demonstrated that fetuses from pregnant obese mothers had higher birth weight, higher percent body fat and elevated cord leptin levels as compared with lean pregnant women.30 Such factors could influence the fetal intrauterine environment, leading to adverse pregnancy outcomes, such as fetal growth abnormalities, as well as fetal adaption leading to increased rates of disease in offspring, such as obesity and diabetes. Further study examining the connection between fetal programming and specific mechanisms of growth perturbation is warranted.7,8,31 This study has several strengths. Primarily, this study shows that AFL levels are detectable and measurable using a simple assay, and appropriate samples are relatively easily obtained by amniocentesis. Although our duplicate measures suggest sensitivity and reliability, validation studies are needed. In addition, the large number of samples from many gestational ages allowed for the development of a standard for AFL levels by gestational age. This study also has weaknesses, including a relatively small sample size with medical record information or placental tissue. The subset cohort with medical record information may represent a highly select patient population; therefore, the generalizability of results may be limited. Although the small subset with clinical data limits the power of this study, AFL values and gestational age of amniocentesis were similar to the larger cohort, suggesting limited bias in results. We did attempt to obtain length information as a more accurate reflection of overall growth; however, we were unable to obtain birth length measurements from the infant records to add to the study. The interpretation of placental findings from the current study is exploratory, ultimately identifying the need for further investigation. Studies correlating amniotic fluid and placenta findings are important, in part because of the overall decrease in amniocentesis for any reason. Finally, this study involves any inherent biases resulting from our ability only to study women clinically requiring amniocentesis. © 2015 Nature America, Inc.
Further studies are necessary to determine the relationship between AFL levels, placental development and fetal growth. For example, gene expression studies may accelerate our understanding of mechanism.32 The role of leptin in placental and fetal growth and the potential utility of AFL in pregnancy management are promising and require further study. In summary, these results are important in assessing the impact of AFL levels to maternal and fetal health outcomes. As less invasive measures are developed, such as maternal serum leptin, the use of amniocentesis may continue to decrease, identifying a need for correlation between measures.33 Efforts to combine genetic information with clinical biomarkers, including potentially nutritional assessments, promises to improve the prediction and treatment of fetal growth abnormalities.34–36 Leptin has the potential of being a biomarker that can predict abnormal patterns of fetal growth, the potential to refine prognosis and better predict neonatal outcomes, potentially impacting morbidity and mortality. CONFLICT OF INTEREST The authors declare no conflict of interest.
ACKNOWLEDGMENTS We thank Michelle Faust (Heart Institute Research Core) and Peggy Walsh (Hatton Research Center) for their assistance. Funding. We disclosed receipt of the following financial support for the research, authorship and/or publication of this article: the Cincinnati Children’s Research Foundation (RBH).
REFERENCES 1 Bernstein I, Gabbe SG. Intrauterine growth restriction In: Gabbe SG, Niebyl JR, Simpson JL, Annas GJ et al. (eds). Obstetrics: Normal and Problem Pregnancies6th ednElsevier: New York, NY, USA, 2012 pp 857–865. 2 Creasy RK, Resnik R. Intrauterine growth restriction In: Creasy RK, Resnik R (eds). Maternal-Fetal Medicine: Principles and Practice 4th edn. Saunders: Philadelphia, PA, USA, 2004 pp 495–512. 3 Alkalay AL, Graham JM, Pomerance JJ. Evaluation of neonates born with intrauterine growth retardation: review and practice guidelines. J Perinatol 1998; 18: 142–151. 4 McIntire DD, Bloom SL, Casey BM, Leveno KJ. Birth weight in relation to morbidity and mortality among newborn infants. N Engl J Med 1999; 340: 1234–1238. 5 Godfrey KM, Barker DJ. Fetal programming and adult health. Public Health Nutr 2001; 4: 611–624. 6 Barker DJ, Hales CN, Fall CH, Osmond C, Phipps K, Clark PM. Type 2 (non-insulindependent) diabetes mellitus, hypertension and hyperlipidaemia (syndrome X): relation to reduced fetal growth. Diabetologia 1993; 36: 62–67. 7 Neerhof MG. Causes of intrauterine growth restriction. Clin Perinatol 1995; 22: 375–385. 8 Szentpéteri I, Rab A, Kornya L, Kovács P, Joó JG. Gene expression patterns of vascular endothelial growth factor (VEGF-A) in human placenta from pregnancies with intrauterine growth restriction. J Matern Fetal Neonatal Med 2013; 26: 984–989. 9 Vrachnis N, Kalampokas E, Sifakis S, Vitoratos N, Kalampokas T, Botsis D et al. Placental growth factor (PlGF): a key to optimizing fetal growth. J Matern Fetal Neonatal Med 2013; 26: 995–1002. 10 Briana DD, Malamitsi-Puchner A. Intrauterine growth restriction and adult disease: the role of adipocytokines. Eur J Endocrinol 2009; 160: 337–347.
Journal of Perinatology (2015), 1 – 6
Standardization of amniotic fluid leptin M Scott-Finley et al
6 11 Ornoy A. Biomarkers of maternal diabetes and its complication in pregnancy. Reprod Toxicol 2012; 34: 174–179. 12 Ahima RS, Flier JS. Leptin. Annu Rev Physiol 2000; 62: 413–437. 13 Popovic V, Casanueva FF. Leptin, nutrition and reproduction: new insights. Hormones 2002; 1: 204–217. 14 Senaris R, Garcia-Caballero T, Casabiell X, Gallego R, Castro R, Considine RV et al. Synthesis of leptin in human placenta. Endocrinology 1997; 138: 4501–4504. 15 Lepercq J, Challier JC, Guerre-Millo M, Cauzac M, Vidal H, Hauguel-de Mouzon S. Prenatal leptin production: evidence that fetal adipose tissue produces leptin. J Clin Endocrinol Metab 2001; 86: 2409–2413. 16 Rasmussen KM, Yaktine AL (eds). Institute of Medicine (US) and National Research Council (US) Committee to Reexamine IOM Pregnancy Weight Guidelines. National Academies Press (US): Washington (DC), CO, USA, 2009. 17 Olsen IE, Groveman SA, Lawson ML, Clark RH, Zemel BS. New intrauterine growth curves based on United States data. Pediatrics 2010; 125: e214–e224. 18 Alexopoulos A, Bravou V, Peroukides S, Kaklamanis L, Varakis J, Alexopoulos D et al. Bone regulatory factors NFATc1 and Osterix in human calcific aortic valves. Int J Cardiol 2010; 139: 142–149. 19 Cole TJ, Green PJ. Smoothing reference centile curves: the LMS method and penalized likelihood. Stat Med 1992; 11: 1305–1319. 20 Henson MC, Castracane VD. Leptin in pregnancy: an update. Biol Reprod 2006; 74: 218–229. 21 Masuzaki H, Ogawa Y, Sagawa N, Hosoda K, Matsumoto T, Mise H et al. Nonadipose tissue production of leptin: leptin as a novel placenta derived hormone in humans. Nat Med 1997; 3: 1029–1033. 22 Karakosta P, Chatzi L, Plana E, Margioris A, Castanas E, Kogevinas M. Leptin levels in cord blood and anthropometric measures at birth: a systematic review and meta-analysis. Paediatr Perinat Epidemiol 2011; 25: 150–163. 23 Chan TF, Su JH, Chung YF, Hsu YH, Yeh YT, Yuan SS. Elevated amniotic fluid leptin levels in pregnant women who are destined to develop preeclampsia. Acta Obstet Gynecol Scand 2006; 85: 171–174. 24 Cagnacci A, Arangino S, Caretto S, Mazza V, Volpe A. Sexual dimorphism in the levels of amniotic fluid leptin in pregnancies at 16 weeks of gestation: relation to fetal growth. Eur J Obstet Gynecol Reprod Biol 2006; 124: 53–57.
25 Tessier DR, Ferraro ZM, Gruslin A. Role of leptin in pregnancy: Consequences of maternal obesity. Placenta 2013; 34: 205–211. 26 Oktem O, Dedeoğlu N, Oymak Y, Sezen D, Köksal L, Pekin T et al. Maternal serum, amniotic fluid and cord leptin levels at term: their correlations with fetal weight. J Perinat Med 2004; 32: 266–271. 27 Lea RG, Howe D, Hannah LT, Bonneau O, Hunter L, Hoggard N. Placental leptin in normal, diabetic and fetal growth-retarded pregnancies. Mol Hum Reprod 2000; 6: 763–769. 28 Lappas M, Permezel M, Rice GE. Leptin and adiponectin stimulate the release of proinflammatory cytokines and prostaglandins from human placenta and maternal adipose tissue via nuclear factor-kappaB, peroxisomal proliferatoractivated receptor-gamma and extracellularly regulated kinase 1/2. Endocrinology 2005; 146: 3334–3342. 29 Misra VK, Trudeau S. The influence of overweight and obesity on longitudinal trends in maternal serum leptin levels during pregnancy. Obesity 2011; 19: 416e21. 30 Catalano PM, Presley L, Minium J, Hauguel-de Mouzon S. Fetuses of obese mothers develop insulin resistance in utero. Diabetes Care 2009; 32: 1076e80. 31 Alexe DM, Syridou G, Petridou ET. Determinants of early life leptin levels and later life degenerative outcomes. Clinl Med Res 2006; 4: 326–335. 32 Nagy GR, Gyõrffy B, Galamb O, Molnár B, Nagy B, Papp Z. Use of routinely collected amniotic fluid for whole-genome expression analysis of polygenic disorders. Clin Chem 2006; 52: 2013–2020. 33 Misra VK, Straughen JK, Trudeau S. Maternal serum leptin during pregnancy and infant birth weight: the influence of maternal overweight and obesity. Obesity (Silver Spring) 2013; 21: 1064–1069. 34 Molvarec A, Szarka A, Walentin S, Beko G, Karádi I, Prohászka Z et al. Serum leptin levels in relation to circulating cytokines, chemokines, adhesion molecules and angiogenic factors in normal pregnancy and preeclampsia. Reprod Biol Endocrinol 2011; 9: 124. 35 Bloomfield FH, Oliver MH, Harding JE. The late effects of fetal growth patterns. Arch Dis Child Fetal Neonatal Ed 2006; 91: F299–F304. 36 McCormack RT, Armstrong J, Leonard D. Codevelopment of genome-based therapeutics and companion diagnostics: insights from an Institute of Medicine roundtable. JAMA 2014; 311: 1395–1396; Erratum in: JAMA 2014; 311: 1396.
Supplementary Information accompanies the paper on the Journal of Perinatology website (http://www.nature.com/jp)
Journal of Perinatology (2015), 1 – 6
© 2015 Nature America, Inc.
