Author + information
- Received April 28, 2017
- Revision received August 4, 2017
- Accepted August 23, 2017
- Published online November 15, 2017.
- Maria Angela Guzzardi, PhDa,
- Tiziana Liistro, PhDa,
- Luna Gargani, MD, PhDa,
- Lamia Ait Ali, MD, PhDa,
- Gennaro D’Angelo, RNa,
- Silvia Rocchiccioli, PhDa,
- Federica La Rosa, BSca,
- Alessandra Kemeny, MDb,
- Elena Sanguinetti, PhDa,c,
- Nadia Ucciferri, PhDa,
- Mariarosaria De Simone, PhDa,
- Antonietta Bartoli, PhDa,
- Pierluigi Festa, MDd,
- Piero A. Salvadori, MSca,
- Silvia Burchielli, DVMd,
- Rosa Sicari, MD, PhDa and
- Patricia Iozzo, MD, PhDa,∗ ()
- aInstitute of Clinical Physiology, National Research Council, Pisa, Italy
- bGynaecology and Obstetrics Department, Azienda USL Toscana Nord Ovest, Massa e Carrara, Italy
- cScuola Superiore di Studi Universitari e di Perfezionamento S. Anna, Pisa, Italy
- dFondazione Toscana Gabriele Monasterio, Pisa, Italy
- ↵∗Address for correspondence:
Dr. Patricia Iozzo, Institute of Clinical Physiology, National Research Council, Via Moruzzi 1, 56124 Pisa, Italy.
Objectives The aim of this study was to investigate the consequences of maternal overweight on cardiac development in offspring in infants (short term) and minipigs (short and longer term).
Background The epidemic of overweight involves pregnant women. The uterine environment affects organ development, modulating disease susceptibility. Offspring of obese mothers have higher rates of cardiovascular events and mortality.
Methods Echocardiography was performed in infants born to lean and overweight mothers at birth and at 3, 6, and 12 months of age. In minipigs born to mothers fed a high-fat diet or a normal diet, cardiac development (echocardiography, histology), glucose metabolism and perfusion (positron emission tomography), triglyceride and glycogen content, and myocardial enzymes regulating metabolism (mass spectrometry) were determined from birth to adulthood.
Results In neonates, maternal overweight, especially in the last trimester, predicted a thicker left ventricular posterior wall at birth (4.1 ± 0.3 vs. 3.3 ± 0.2 mm; p < 0.05) and larger end-diastolic and stroke volumes at 1 year. Minipigs born to mothers fed a high-fat diet showed greater left ventricular mass (p = 0.0001), chambers (+100%; p < 0.001), stroke volume (+75%; p = 0.001), cardiomyocyte nuclei (+28%; p = 0.02), glucose uptake, and glycogen accumulation at birth (+100%; p < 0.005), with lower levels of oxidative enzymes, compared with those born to mothers fed a normal diet. Subsequently, they developed myocardial insulin resistance and glycogen depletion. Late adulthood showed elevated heart rate (111 ± 5 vs. 84 ± 8 beats/min; p < 0.05) and ejection fraction and deficient fatty acid oxidative enzymes.
Conclusions Neonatal changes in cardiac morphology were explained by late-trimester maternal body mass index; myocardial glucose overexposure seen in minipigs can justify early human findings. Longer term effects in minipigs consisted of myocardial insulin resistance, enzymatic alterations, and hyperdynamic systolic function.
- cardiac triglycerides
- developmental programming
- glucose metabolism
- positron emission tomography
The current epidemic of overweight involves women of childbearing age, and the prevalence of gestational obesity is estimated to reach 40% in 2020 (1). Overweight is a risk factor for cardiovascular disease and cardiomyopathies. Pregnancy is associated with metabolic adaptations, including increases in body weight, circulating lipids, glucose, and inflammatory markers. These changes are more pronounced in obese than normal-weight women (2). The uterine environment affects fetal organ development, modulating disease susceptibility during post-natal life, adulthood, and aging. Epidemiological studies suggest that maternal obesity results in a greater risk for cardiovascular disease and premature death in adult and elderly offspring (3,4). Cardiac growth occurs mostly during early life, but little is known concerning cardiac development and function in human infants born to overweight mothers. In 1 study, left ventricular (LV) mass in 6-month-old neonates was increased in proportion to maternal gestational weight gain (5).
Animal studies have shown that offspring of obese mothers are more vulnerable to myocardial ischemia-reperfusion injury (6), showing alterations in myocardial insulin signaling and cardiac function in fetal life (7,8), and cardiac hypertrophy during early life, followed by abnormal cardiac contractility during adulthood (8,9).
Although still limited, the evidence suggests that pre-gravid or gestational overweight is associated with adverse cardiological features in offspring. Understanding and recognizing this pathophysiological link would translate to possibilities for primary prevention. This study was undertaken to examine cardiac development in the offspring of overweight mothers in human infants and to gain mechanistic and long-term insight in a swine model.
Additional details are provided in the Online Appendix.
Ninety-one pregnant women were involved during early (n = 41) or late (n = 50) pregnancy. In neonates, we measured the metabolic profile in cord plasma and monitored body growth and cardiac development (by echocardiography) in the first year of life (Online Figure 1). To examine the effects of pre-gravid versus gravid maternal overweight, infants were stratified on the basis of respective maternal body mass index (BMI) median values of 23.6 kg/m2 (first visit) and 28.8 kg/m2 (end of pregnancy). The study protocol was approved by the Ethics Committee of Massa and Carrara, Italy, and latest amendments by the Ethical Committee of the Area Vasta Nord Ovest (Pisa, Italy). Parents gave their written informed consent before inclusion.
Adult female minipigs received a high-calorie, high-fat diet (HFD) (n = 5) or a standard normal diet (ND) (n = 5) 3 months before mating until the end of lactation. After weaning, offspring were fed the ND and studied at different ages (sample sizes according to Online Figure 2, unless given in respective legends). Characterization included echocardiography, positron emission tomography (PET) (for myocardial perfusion and insulin-mediated glucose metabolism), histology, myocardial metabolic assays, and mass spectrometry profiling of myocardial enzymes. The protocol was conducted in accordance with the D.L. 116/92 implementation of European Economic Community directive 609/86 regarding the protection of animals used for experimental and other scientific purposes.
Imaging and tissue analyses
Additional details are reported in the Online Appendix.
Transthoracic echocardiography was performed in infants (GE Logic, with a 2.5- to 6.0-MHz transducer, M-mode for LV mass, B-mode for LV volumes, GE Vingmed Ultrasound, Horten, Norway) and minipigs (Vivid I, with a 2.5- to 3.5-MHz phased-array sector scan probe with second-harmonic technology, GE Vingmed Ultrasound) of different ages (Online Figures 1 and 2) to measure LV dimensions and function. Mitral valve Doppler was used to determine E/A ratio. Epicardial fat thickness was measured on the free wall of the right ventricle from the parasternal long-axis view at end-diastole in 3 cardiac cycles. Readers were blinded to study groups.
Dynamic PET studies were performed in minipigs after an overnight fast in an ECAT HR+ tomograph (Siemens Healthcare, Erlangen, Germany) to assess myocardial perfusion (milliliters per milliliter tissue per minute) at baseline and during adenosine stress (13N-ammonia PET) and insulin-mediated glucose uptake (18F-fluorodeoxyglucose PET) during isoglycemic hyperinsulinemia. Respective fractional tissue tracer extraction rate constants were converted into blood flow or multiplied by glycemia to obtain glucose uptake rates.
Cardiomyocyte nuclei numbers and size and glycogen content were assessed by hematoxylin and eosin and periodic acid–Schiff staining in biopsy samples collected from the LV free wall. Myocardial triglyceride content was measured by enzymatic assay using a commercial kit (Giesse Diagnostic, Guidonia, Italy). Mass spectrometry profiling of enzymes involved in glucose and lipid metabolism and oxidative phosphorylation was performed using TripleTOF 5600 mass spectrometer (AB SCIEX, Framingham, Massachusetts) with a DuoSpray ion source (AB SCIEX).
Data are presented as mean ± SEM. Groups were compared within each age time point using 2-tailed analysis of variance. In enzyme profiling, multiple comparisons were corrected for false discovery rate (Benjamini-Hochberg). Regression analyses were carried out using standard techniques. Differences between groups were regarded as statistically significant at p ≤ 0.05. A p value <0.1 was regarded as meaningful in small groups analyzed with correction for false discovery rate.
Maternal triglyceride, glucose, and insulin resistance levels are shown in Online Figure 3. Gestational diabetes was present in 25% of women (10 and 13 lean and overweight women; p = NS). Gestational age and neonatal metabolic profiles (Online Tables 1 and 2) were not different; only at birth, mild overweight was seen in neonates born to mothers with higher gravid BMIs. A 4- to 5-fold increase in LV stroke volume (SV) and a 2-fold increase in LV mass occurred during the post-natal year (Figure 1, sex adjusted in Online Table 3). Babies born to mothers with gravid overweight had heavier and thicker left ventricles at birth compared with control infants. In the second life semester, they underwent more pronounced increases in LV SV and end-diastolic volume (EDV), leading to significant group differences at 12 months of age. In babies stratified according to maternal pre-gravid BMI, only SV at 12 months was significantly different (Online Figure 4). Maternal BMI was correlated with the aforementioned cardiac parameters, whereas maternal glycemia and triglyceride levels were related primarily to findings observed at birth (Online Table 4). Systolic ejection fraction (EF), diastolic E/A ratio, and end-systolic volume (ESV) were similar between groups (Online Figure 5).
As expected, HFD mothers were about 33% overweight before (33.0 ± 2.8 kg vs. 24.8 ± 2.1 kg; p = 0.049 vs. ND) and during pregnancy (55.6 ± 4.0 kg vs. 41.6 ± 2.0 kg; p = 0.014 vs. ND). One mother was excluded after repeated abortion. Metabolic offspring characteristics (Online Table 5) showed no differences in metabolic markers, except for high cholesterol in the offspring of HFD mothers.
At birth, offspring of HFD mothers showed marked enlargement in LV ESV, EDV, mass, and SV (Figure 2) and larger cardiomyocyte nuclei (Online Figure 6), in agreement with observations in infants, and elevations in myocardial glucose uptake and glycogen content (Figure 3). Myocardial protein profiling (Online Figures 7 and 8, Online Table 6) revealed up-regulation of enzymes involved in cardiomyocyte glucose entry, mitochondrial pyruvate entry, and fatty acid oxidation and down-regulation of glycolytic, distal Krebs cycle, and electron transporting enzymes, and creatine kinase tended to be low in offspring of HFD sows.
From infancy to adulthood, offspring of HFD mothers showed LV wall thickening and smaller ESV, resulting in elevated EF during infancy (Figure 2), compared with offspring of ND mothers. In late adulthood, they had a marked reduction in ESV, with elevated EF and heart rate, and the histological appearance of the myocardium was more disorganized and cell morphology more heterogeneous (Online Figure 6). From infancy through adulthood, the offspring of HFD mothers were characterized by myocardial insulin resistance (Figure 3), with defective glycogen storage. In late adulthood, myocardial glucose metabolism declined in control animals, abolishing group differences. Cardiac triglycerides and epicardial fat tended to be low in adult offspring of HFD mothers (Online Figure 9). Late-adulthood offspring of HFD mothers mostly showed an enzymatic profile (Online Figures 7 and 8) suppressive of fatty acid oxidation and adenosine triphosphate synthesis.
The novelty of this study was that it examined myocardial morphological and metabolic development in relation to maternal overweight by combining evidence in infants and in a large animal model, which is well suited to reflect human cardiac (patho)physiology. We performed in vivo and ex vivo tissue measurements to formulate mechanistic hypothesis, although the study was not intended to validate mechanisms. Understanding whether, when, and how maternal obesity affects cardiac development provides the necessary baseline for preventive efforts and mechanistic studies. Although cardiac morphology and function in our children were within the normal range, infants born to overweight compared with lean mothers had significantly greater LV thickness, EDV, and SV during the first year of life, and these parameters were significantly correlated with maternal BMI and metabolic profile. Several studies indicate that metabolic outcomes of maternal obesity relate to pre-gravid BMI, and pre-conception interventions are challenging. One important finding in our study was that end-of-pregnancy BMI was a stronger predictor of offspring’s cardiac parameters than pre-gravid BMI, suggesting that weight control in later gestation (a more realistic target) may alleviate the impact of maternal overweight on offspring’s hearts.
Echocardiographic findings observed in infants were confirmed and more pronounced in our minipigs, in which histological observations were suggestive of larger cardiomyocytes in the offspring of HFD than ND sows. Metabolic data suggested that an elevation in myocardial glucose uptake during intrauterine growth down-regulates glycolysis, the Krebs cycle, electron transfer, and oxidative enzyme expression, leading to glycogen accumulation and cardiac thickening and enlargement. This putative sequence of events is supported by evidence that cardiac hypertrophy can be induced by stimulating glucose uptake and is characterized by high levels of insulin-dependent glucose transporters and glycogen (10–12). The alterations observed at birth in the offspring of HFD minipigs may contribute to explain the onset of myocardial insulin resistance and glycogen depletion seen during lactation and early adulthood. These are negative prognostic factors that increase myocardial susceptibility to ischemic damage in humans (13–16), and ischemia susceptibility occurs in mice born to obese dams (6). They may contribute to explain the epidemiological observation of shorter survival due to cardiovascular causes in humans born to obese mothers (3,4). In later adulthood, minipigs born to HFD compared with ND mothers showed higher EF% and heart rate. SV and cardiac output were similar, suggesting that a greater degree and frequency of contraction were required to preserve cardiac performance. Hyperdynamic LV systolic function may help preserve systolic performance, but a long-standing stimulation of cardiac contraction can predispose to heart failure. In fact, adult offspring of obese dams show cardiac sympathetic dominance (9,17) and were suggested to be on the pathway to premature heart failure. Heart failure is characterized by impaired fatty acid use (18), and we observed a down-regulation in enzymes catalyzing β-oxidation and adenosine triphosphate synthesis in adult offspring of HFD mothers.
Main strengths include the first evidence of cardiac development in neonates born to overweight and lean mothers and the development of an animal model reproducing human findings, making it possible to formulate mechanistic hypotheses and gain longer-term insights. Limitations included the small sample size in some substudies and risk for type I error, which we partially addressed using correction for false discovery rate in multiple comparison analyses, and the cross-sectional nature of comparisons, as the number of subjects at each time point varied. In humans, potential confounders in post-natal life were not controlled for, given the sample size, and therefore findings at birth are considered more directly related to maternal obesity than findings at 12 months. In animals: 1) the lack of epigenetic, catecholamine, and vascular resistance measurements reduces our mechanistic insights; 2) myocardial enzyme expression does not equal enzyme activity, so our molecular findings are descriptive, although supported by coherent functional (in vivo) changes; 3) by design, our minipigs were fed the ND after lactation to selectively characterize the impact of early-life nutrient overexposure, excluding the confounding effect of eating behavior during life (however, offspring of overweight mothers may be prone to overeating, which would aggravate their phenotype); and 4) the translational value of our adulthood findings awaits testing in adult humans.
In humans, neonatal changes were explained by late-trimester maternal BMI. The animal model demonstrated that the elevation in myocardial glucose uptake, leading to glycogen accumulation and inhibition of glucose oxidative pathways, may explain cardiac findings in the newborn human. Life span data in minipigs showed that these cardiac changes subside, and later ages were characterized by myocardial insulin resistance, enzymatic alterations, and hyperdynamic systolic function.
COMPETENCY IN MEDICAL KNOWLEDGE: The effects of maternal overweight on cardiac development were examined in infants, showing that end-of-pregnancy (more so than pre-gravid) BMI is a predictor of LV thickness, EDV, and SV in the first year of life. Two important aspects are that BMI range in our women is representative of the great majority of childbearing women, and BMI in later pregnancy may be an effective and realistic prevention target.
TRANSLATIONAL OUTLOOK: Although our minipig data reproduced infants’ results and we could follow minipigs through adulthood, which showed vulnerabilities, findings of myocardial insulin resistance and hyperdynamic function should be investigated and confirmed in adult humans. More research is required to validate molecular mechanisms and targets. In addition, studies preventing or reverting maternal obesity to demonstrate prevention potential are warranted.
The authors acknowledge the families participating in the study and the technical, nursing, and animal care facility staff, supporting data collection at the Institute of Clinical Physiology, National Research Council, and ASL1-Massa, Italy.
This research was funded by the European Commission under the Seventh Framework Programme (FP7-HEALTH-DORIAN Project: Developmental Origins of Healthy and Unhealthy Ageing: The Role of Maternal Obesity; grant 278603) and the European Foundation for the Study of Diabetes (EFSD/Roche Educational Grant 2009). The authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Guzzardi and Liistro contributed equally to this work and are joint first authors.
- Abbreviations and Acronyms
- body mass index
- end-diastolic volume
- ejection fraction
- end-systolic volume
- high-fat diet
- interventricular septum
- left ventricular
- normal diet
- positron emission tomography
- posterior wall diameter
- stroke volume
- Received April 28, 2017.
- Revision received August 4, 2017.
- Accepted August 23, 2017.
- 2017 American College of Cardiology Foundation
- Catalano P.M.,
- Presley L.,
- Minium J.,
- Hauguel-de Mouzon S.
- Reynolds R.M.,
- Allan K.M.,
- Raja E.A.,
- et al.
- Wang J.,
- Ma H.,
- Tong C.,
- et al.
- Heather L.C.,
- Howell N.J.,
- Emmanuel Y.,
- et al.
- Iozzo P.,
- Chareonthaitawee P.,
- Dutka D.,
- Betteridge D.J.,
- Ferrannini E.,
- Camici P.G.
- Dutka D.P.,
- Pitt M.,
- Pagano D.,
- et al.
- Marinho N.V.,
- Keogh B.E.,
- Costa D.C.,
- Lammerstma A.A.,
- Ell P.J.,
- Camici P.G.
- Kaye D.M.,
- Lefkovits J.,
- Jennings G.L.,
- Bergin P.,
- Broughton A EM