, Graphical abstract
Introduction
Over the past 3 decades, major advances in neonatology have significantly improved the mortality rate of very preterm infants (VPIs), defined as those born before 32 weeks' gestation. This evolution has shifted the focus of the neonatologist from simply increasing the survival rate of VPIs to improving their nutrition, growth, and development [1,2]. In fact, the aggressive nutritional management of VPIs is recommended by the American Academy of Pediatrics (AAP) and European Society of Pediatric Gastroenterology, Hepatology and Nutrition with the goal of achieving postnatal growth analogous to fetal rates [3,4].
VPIs are expected to have fetal growth rates like their term-born counterparts with comparable body compositions and functional outcomes. A reduction in the negative weight z score—defined as the number of standard deviations by which an individual value is above or below the population mean—was demonstrated in VPIs who had not developed a cumulative nutritional deficit following improved nutritional practices [5]. As previously detailed above, this suggests that it is possible to mitigate early-life weight growth failure with appropriate nutritional intervention. However, at term-corrected age, VPIs are generally shorter, which suggests stunting, and have greater fat mass and lower lean and bone mass than term-born infants; this difference in body composition reflects the preferential increase in adipose tissue that occurs during catch-up growth and has been observed in VPIs at age-appropriate weight [6,7]. In terms of body composition, length gains represent lean body mass and protein gains and indicate organ growth, while weight gains represent total body mass, including fat mass, and indicate balance between energy intake and expenditure [8].
During the first few months of life, a VPI undergoes a rapid period of fat deposition, with more than half of the energy income prioritized for adipose tissue growth [6,9]. This adaptive physiology protects the growing and energy-demanding brain from periodic energy deficits. These findings are concerning given the concept of the developmental origins of health and disease, which connects an unfavorable early-life body composition with negative long-term metabolic and cardiovascular health outcomes [10,11].
VPIs frequently present with postnatal length growth failure—defined as a length at more than minus 2 standard deviations for age and sex—that remains lower than the standard weight until at least 2 years corrected age [12-14]. This condition differs from the classical pattern of child undernutrition, defined as a significantly lower weight for a given length, and suggests that nonnutritional factors play a key role in length growth restriction [15]. The nonnutritional factors underlying poor length growth in VPIs are currently not completely understood [16]. While weight faltering has been extensively studied, the significance of length growth failure in VPIs has only recently emerged.
In this review, we comment on recent advances and knowledge gaps in understanding the pathophysiology of length growth restriction in VPIs, which has been identified as a major predictor of adverse neurodevelopmental and cognitive outcomes, and highlight the effects of nonnutritional factors on postnatal linear growth, focusing primarily on the effects of sustained neonatal inflammation. Finally, we discuss hypotheses and future lines of research aiming to understand the complex inflammatory-endocrine interactions and pathophysiological changes that occur during early postnatal life, appropriately guide and apply clinical strategies aimed at optimizing length growth of VPIs, and identify evidence of associations between sustained neonatal inflammation, stunting, and long-term health risks and the potential implications.
Short- and long-term consequences of poor length growth in VPIs
Length growth failure in VPIs should be considered an adverse outcome, not only because it is associated with reduced height throughout childhood and into adolescence with no evidence of catch-up growth and a direct cause of adult short stature but also because stunting represents reduced lean mass accretion and restricted growth of all organ systems and serves as a key marker of the underlying processes in early life that lead to negative long-term health outcomes [15,17].
Although young very preterm-born adults with short stature are identified by comparison of their height to that of a reference population, short stature alone is not typically considered a cause for concern. However, in our opinion, it should be conceptualized as a stunted phenotype in which a constellation of pathological changes, characterized by length and organ growth restriction, increased morbidity and mortality, and reduced physical, neurodevelopmental and cognitive capacities, is present [18].
There is a growing body of evidence suggesting that length growth restriction may contribute to altered growth of the tissues and organs [6,8-11,16,18]. However, evidence of a causal relationship is often lacking, and the pathophysiological mechanisms remain to be determined.
The interactions between growth hormone (GH), insulin-like growth factor-1 (IGF-1), and the immune system are complex. A recent prospective observational study conducted by our team reported that sustained neonatal inflammation may induce postnatal stunting related to lower bone mass accrual via GH/IGF-1 axis inhibition in VPIs [19]. Most data of short- and long-term consequences of length growth restriction were derived from retrospective studies. Therefore, a meticulous interpretation is essential, and the potential impact of sustained neonatal inflammation must be adjusted for other confounding factors that may influence postnatal growth and adverse outcomes later in life [13].
It is important to note that a significant number of VPIs are already growth restricted at birth. This is attributable to in utero conditions that impede their future growth irrespective of nutritional status. Therefore, it is imperative that we gain an understanding of the in utero growth rates of infants born very preterm to enable the identification of the extent of future growth restriction, particularly in terms of length and lean body mass [20].
The detrimental impact of postnatal length growth restriction in VPIs on neurodevelopment and cognitive capacities were well documented by Belfort et al. [8] and Ramel et al. [12] Very preterm birth and sustained neonatal inflammation have been linked to short- and long-term outcomes such as cerebral palsy, neurodevelopmental delay, behavioral problems, chronic lung disease (CLD), airflow reduction, reduced bone mineral density, increased left and right ventricular mass, hypertension, altered systolic and diastolic function, increased risk of inflammatory bowel disease, reduced renal volume and function, chronic renal disease, significant reductions in muscle thickness and muscle power among young adults [17,21-23]. However, to the best of our knowledge, the specific effects of length growth restriction have not yet been studied in any organ other than the brain. Thus, is important that clinicians become aware of the potential long-term adverse consequences associated with length growth restriction in VPIs, particularly in terms of their metabolic and cardiovascular health.
Metabolic risks
Barker's groundbreaking hypothesis about the developmental origins of health and disease was based on the observation that patients with coronary artery disease often had neonatal weight growth restriction [24]. The infant with a weight growth restriction is programmed to have a "thrifty" metabolism to such an extent that even a normal substrate intake can simulate an excessive intake later in life, leading to obesity, hyperglycemia, hyperlipoproteinemia, atherosclerosis, and its sequelae [25]. For example, in a Finnish study, former weight-restricted VPIs at 18–27 years of age had 6.7% higher mean blood glucose levels and 40.0% higher mean insulin levels than full-term infants [26]. Latent insulin resistance is a precursor to diabetes mellitus. Although it was rare overall in a Swedish VPIs cohort of patients at age 26–37 years, diabetes was significantly more likely to develop, with an adjusted odds ratio of 1 [5.27].
Cardiovascular risks
According to a large meta-analysis of 9 cohorts, former growth-restricted VPIs have an average systolic/diastolic arterial blood pressure that is 3.4/2.1 mmHg higher than those born at term, another risk factor for cardiometabolic disease [28]. The predisposition to arterial hypertension, which is of interest for its regularity rather than severity, is explained by, among other things, increased large-vessel wall stiffness and capillary bed thinning [25]. In addition, cardiac magnetic resonance imaging revealed an altered geometry of the left ventricular myocardium in adults born very preterm compared to those born at term and followed prospectively [29]. All of these risk factors probably contribute to the increased risk of coronary heart disease found in a Swedish cohort study of adults aged 30–43 years who were born before 34 weeks' gestation [30]. Fig. 1 illustrates the complex risk profile of VPIs with length growth restriction due to sustained neonatal inflammation as well as their potential short- and long-term adverse outcomes.
Nutritional factors impacting growth among VPIs
Vulnerability to nutritional deficits is a characteristic of VPIs that is most pronounced at the time of their life cycle. The risk of malnutrition is related to several factors, including reduced nutrient stores at birth, immature nutrient absorption and use, organ immaturity, delayed advancement of parenteral and enteral feeds due to cautious health management, and dependence on health care providers to accurately identify and effectively provide needed nutrients during this period of rapid growth and development [31]. It is widely accepted that malnutrition results in poor growth. Growth less than expected, as well as caloric and protein intakes below that recommended amounts for age, are endorsed indicators of infant undernutrition [31].
Postnatal nutritional growth restriction has the potential to affect the development of VPIs from birth. It is widely recognized that an infant's nutritional intake is a major determinant of their growth. Wasting is defined as the tendency to be too thin for one's length, sometimes referred to as weight for length for age and sex. Stunting is defined as the condition of VPIs whose length for age and sex is less than minus 2 standard deviations from the median of the VPIs growth standards. While wasting is the result of acute significant caloric nutrient deficiency, stunting represents chronic protein deficiency [32-35]. Acute postnatal wasting is the most common cause of weight growth restriction among VPIs. The etiology, pathophysiology, clinical features, and short- and long-term outcomes of nutritional wasting versus nonnutritional stunting among VPIs are presented in Table 1.
Inflammation-related morbidities in VPIs
Neonatal care has significantly improved VPI survival. However, this improvement has come at a cost, as there has been a steady increase in the incidence of several morbidities, most of which involve inflammation as a crucial pathophysiological factor. Inflammation-related morbidities such as neonatal sepsis, bronchopulmonary dysplasia (BPD), and necrotizing enterocolitis (NEC), are common examples [13].
Preterm birth results in organ and system immaturity, which often requires respiratory ventilation. However, these interventions can have potential side effects, including ventilator-associated pneumonia and BPD. BPD is a CLD that can have serious consequences, including impaired pulmonary and cardiovascular health as well as adverse neurodevelopmental outcomes [12,15]. BPD has emerged as the most prevalent CLD among infants. The balance between the inflammatory response and IGF-1 is a pivotal factor in the development of immature lung cells, including alveolar, epithelial, and endothelial cells. Evidence links alterations in systemic IGF-1 patterns with pulmonary dysfunction [30]. These changes occur simultaneously at the systemic and local levels, resulting in a growth restriction of all body organs [38].
Recognizing that BPD is an inflammatory disease, the AAP has urgently called for new anti-inflammatory therapies for it since 2006. BPD remains a significant challenge in neonatology today since no safe and effective therapy is currently available [39,40]. Early enteral feeding can be challenging due to intestinal immaturity, and prolonged parenteral nutrition required to advance nutrition increases the risk of catheter-related sepsis [41,42]. Preterm birth commonly leads to NEC, a serious gastrointestinal complication among VPIs [43].
NEC typically develops after the start of enteral nutrition. Its pathogenesis is multifactorial and involves numerous inflammatory mediators [43-45]. VPIs are at a higher risk of requiring invasive treatments and procedures, further increasing their susceptibility to infection. The systemic inflammatory response that occurs during neonatal sepsis is a major contributor to morbidity and mortality among VPIs [46,47].
Neonatal sepsis, BPD, NEC, and sustained neonatal inflammation are linked to decreased length at 12 months' corrected age. Ramel et al. and Belfort et al. demonstrated that length was more stunted and remained lower than weight in very preterm children during the first 2 years of life compared with their term counterparts [12,14]. This pattern differs from the classic pattern of child undernutrition, defined as a significantly lower weight for a given length, and suggests that nonnutritional factors unrelated to the insulin-glucose axis play an important role in this suppression. Thus, we propose that epigenetic changes in metabolic pathways may not be limited to undernutrition and to the prenatal period; this window should be extended at least to the first trimester of postnatal life [13].
Height growth depends primarily on the GH/IGF-1 axis, whereas weight gain depends primarily on the insulin-glucose axis. Fat mass rather than lean and bone mass gains are mediated by an adequate caloric intake and insulin secretion [19,48,49]. Height growth, the most reliable indicator of children's well-being, provides an accurate marker of differences in human development. Unfortunately, this is reflected in the millions of children worldwide who not only fail to achieve their height potential due to suboptimal health conditions and inadequate nutrition and care but also suffer the severe irreversible physical and cognitive damage that accompanies stunted growth. Stunting refers to poor height growth due to undernutrition, infection, or sustained inflammation, with height-for-age z scores that are more than 2 standard deviations below normal [19].
Nonnutritional factors underlying stunting in VPIs are poorly understood [19]. There is evidence that severe neonatal illness, mainly neonatal sepsis, BPD, NEC, and sustained neonatal inflammation – defined as persistently elevated concentrations of inflammation-related proteins (e.g., C-reactive protein or proinflammatory cytokines) in the circulation – are associated with reduced length at 12 months' corrected age [12-14].
Our quantification of this association for BPD, neonatal sepsis, and NEC resulted in an independent height decrease at 12 months' corrected age of 3.9%, 2.1%, and 1.1%, respectively, compared with VPI controls without these pathologic conditions [13]. During our investigations to understand the nonnutritional causes of postnatal length growth restriction among VPIs, we found it plausible that they had developed adaptive strategies to maximize fitness under extremely adverse conditions during the neonatal period.
The GH/IGF-1 axis represents an attractive candidate to aid our understanding of the underlying mechanisms of the nonnutritional stunted phenotype in VPIs. In this context, it is important to recognize stunting according to the biological hypothesis that focuses on energetic trade-offs. Slowed length growth can be compensatory for the high energetic needs caused by sustained inflammation. Demonstrating life history trade-offs has frequently proved challenging, with empirical studies indicating positive correlations between life history traits and survival rather than the expected inverse relationships.
The case of VPI survivors with sustained neonatal inflammation represents a particular example of a positive correlation between stunting as an adaptive compensation in one's early-life history and having suffered severe proinflammatory conditions such as neonatal sepsis, BPD, and NEC during the neonatal period [18].
Sustained inflammation and stunting in children
DeBoer et al. [50] demonstrated a direct relationship between sustained inflammation, GH signaling, and restricted length growth in growing mice with infectious colitis. Sustained inflammation is related to GH resistance in the liver as evidenced by higher systemic levels of GH, lower hepatic production of IGF-1, lower systemic levels of IGF-1, and slowed length growth.
DeBoer et al. [51] later demonstrated that children from impoverished communities who suffer from recurrent infections may experience stunted growth. The study found temporal links between slowed height growth and repeated clinically relevant diarrheal and respiratory infections. Sustained inflammation plays a crucial role in suppressing length growth during infection.
Sustained neonatal inflammation and stunting in VPIs
We recently demonstrated a pathophysiological mechanism explaining the association between sustained neonatal inflammation and stunting among VPIs with adequate birth weight for gestational age and a comparable nutritional status to VPIs without sustained neonatal inflammation at term-corrected age and subsequently demonstrated that an even low sustained neonatal inflammation of the GH/IGF-1 axis at systemic and local levels inhibits bone mass formation in VPIs [13,19]. The mechanism we proposed is as follows: IGF-1 is produced by the liver in response to GH; and sustained neonatal inflammation induces resistance to GH signaling in the liver and growth plate during early postnatal development.
The development of hepatic GH resistance in response to sustained neonatal inflammation can be attributed to 2 primary mechanisms. The initial mechanism is the downregulation of GH receptors (GHR), while the subsequent mechanism involves upregulation of suppressors of the cytokine signaling (SOCS) family members, particularly SOCS1 and SOCS3, contributing to negative regulation of the growth-promoting actions of GH. Tumor necrosis factor-α (TNF-α) and interleukin (IL)-1β primarily inhibit hepatic GHR expression, whereas IL-6 inhibits hepatic GH signaling by inducing SOCS3 expression and has no effect on GHR expression. While proinflammatory cytokines induce hepatic GH resistance, they may also exert a systemic effect on IGF-1 action by influencing the metabolism of IGF-1-binding proteins (IGFBPs), which in turn affects IGF-1 clearance. A notable decrease was also noted in serum IGF-1 levels [19].
The epiphyseal growth plates, which are situated in the proximal and distal regions of the long bones, represent growth factor destinations. Proinflammatory cytokines reportedly exert a negative direct local effect on the growth plate. Furthermore, research has shown that a combination of IL-1β, IL-6, and TNF-α has a synergistic effect that amplifies their inhibitory impact on growth. It is also important to note that both TNF-α and IL-1β, which are produced endogenously in growth-plate chondrocytes, appear to play a role in normal growth. However, negative effects of these cytokines on chondrocytes have been observed at the supraphysiological level. For these reasons, the immune system activity – specifically sustained neonatal inflammation – exerts a significant effect on the GH/IGF-1 axis and longitudinal growth (Fig. 2) [19].
Importance of characterizing different endotypes of stunted phenotype in VPIs
The biological contributions to growth restriction in VPIs are complex. A comprehensive pathophysiological investigation will help elucidate the etiopathology of this neonatal disorder and identify linear growth restriction earlier and guide future clinical research tailored to the specific disease phenotype.
The direct relationship between sustained neonatal inflammation, GH signaling, and height growth restriction in VPIs provides an ideal example of pathophysiology-based approaches to defining their phenotypes as well as critical insights into the mechanisms that contribute to disease development and severity.
Phenotypes are artificial constructs; in the case of stunting, it is useful for the neonatologist to characterize a particular subset of VPIs that define a particular risk factor, respond to a particular treatment, or have a particular prognosis. This ability would define phenotypes based primarily on clinical and physiological measures. Another way to phenotype stunting in VPIs would be to include a disease pathogenic mechanism that may be different. Indeed, the stunted phenotype must be divided into distinct disease entities with specific causal mechanisms such as genetic, nutritional, or inflammatory, which should be more specifically called endotypes [54].
Given the significant number of individuals affected and the societal costs of caring for adult patients born very preterm with adverse outcomes related to stunting, substantial efforts must be made to improve their clinical care. Future improvements in both clinical care and outcomes can be achieved by: (1) improving our ability to diagnose linear growth failure in a timely manner, (2) supporting studies of specific stunting endotypes to develop tailored and effective therapies, and (3) advocating for multidisciplinary integrated care models to support stunting in very preterm survivors as a chronic condition that should be addressed throughout the lifespan.
Hypothesis and future research lines
Here we hypothesized that sustained neonatal inflammation during critical periods (i.e., early postnatal life) leads to persistent peripheral GH resistance resulting from permanent GHR downregulation due to epigenetic GHR reprogramming that occurs during the neonatal period in VPIs, which in turn results in an inflammatory endotype within the stunted phenotype (Fig. 3). This approach is based on the following well-documented observations reported in clinical studies: (1) sustained neonatal inflammation induces lower bone mass accrual, which is associated with higher GH and lower IGF-1 levels among VPIs at birth-corrected age [19,55], (2) very preterm birth is associated with lower bone mineral density in young adulthood [56], (3) height remains consistently lower in the very preterm group versus term controls throughout childhood and adolescence, with no evidence of catch-up [57], and (4) pubertal status is not related to adult height among VPIs [58].
If our hypothesis is true, it remains to be demonstrated in future studies whether very preterm-born adults compared with their term counterparts will present peripheral resistance to GH and lower IGF-1 secretion with normal GH concentrations after provocation tests due to a lower GHR gene expression because of GHR epigenetic reprograming as a sustained low-grade inflammatory response during the neonatal period [18].
Future research directions may involve the fact that underlying pathophysiological mechanisms in prespecified VPIs stunted subgroups remain to be elucidated. This can be achieved using mechanistic neonatal animal models and in silico modeling. The development of very preterm animal models is critical for studying the multisystem mechanisms of extrauterine growth restriction, a condition associated with adult disease. Nonlitter animal models, such as sheep, monkeys, and lambs, are particularly useful due to their similarities in embryology, anatomy, and physiology to humans. The use of these species has allowed significant advances in our understanding the pathophysiology and consequences of growth restriction following very preterm birth with observations comparable to those in humans.
Experimental manipulations in these models include caloric or protein restriction, micronutrient deficiency, stress, hypoxemia, ischemia, inflammation, infection, genetic and hormonal alterations, or specific organ dysfunction. However, a key challenge of in vivo research is that animal models often reproduce only some of the symptoms observed in humans, possibly due to interspecies differences in critical programming windows. This approach underscores the importance of understanding and addressing these limitations to improve model relevance and clinical applicability.
To ascertain the influence of postnatal factors, it is essential to correlate neonatal data sets with follow-up data from cohort studies to identify critical periods for targeted neonatal intervention (e.g., time, type, and severity of inflammatory processes and different phenotypes of adult health outcomes).
The use of prospective cohort studies is also important to improving our understanding of the long-term outcomes of VPIs. Such studies can help identify risk factors and aid the development of treatment and prevention options. Here are 3 examples of paradigmatic longitudinal cohort studies of VPIs: (1) Evaluation in a Longitudinal Cohort of Very Preterm Infants: This study analyzed different growth assessment methods in VPIs. Their nutritional intake and clinical development were recorded, and their growth velocity, extrauterine growth restriction, and postnatal growth failure were calculated. The study concluded that the z score analysis was better at identifying postnatal growth failure [59]. (2) Growth in Very Preterm Infants: a longitudinal study. This study characterized the longitudinal growth of a population-based group of preterm infants born at a gestational age of <29 weeks. The study found that these children had subnormal weight and height growths compared to children with a normal birth weight [60]; and (3) The Avon Longitudinal Study of Parents and Children, also known as the Children of the 90s study, was a long-term health research project based in Bristol, United Kingdom. It is one of the most detailed studies of its kind, having aimed to understand how genetic and environmental factors influence VPI health, growth, and development across generations [61].
Further clinical research is needed to address the long-term effects of perinatal exposure such as antibiotics; postnatal biomarkers of growth and inflammation such as S100A8/9 proteins, alpha-1 acid glycoprotein, and C-reactive protein; regulatory T cells such as CD3+, CD4+, and CD25+; Foxp3+ cells; cell surface markers of inflammation such as CD39, 5' nucleotidase/CD73, CTLA-4, GITR, LAG-3, LRRC32, and neuropilin-1; interleukins including immunosuppressive cytokines such as TGF-beta, IL-10, and IL-35; biomarkers of linear growth such as IGF-1 and IGF-1 binding protein of circulating concentration (IGFBP3); anti-inflammatory drugs such as anakinra; and stem cells in Phase I–III clinical trials and randomized placebo-controlled studies with long-term follow-up.
There is a dearth of valid outcome measures for important organ functions such as liver structure and function in very preterm-born stunted young adults. Thus, the development of new tools to enable early short-term growth and body composition assessments with the aid of machine learning and artificial intelligence is a crucial priority. There is also a need for infant followup with detailed determination of possible beneficial factors, such as human milk and vaccinations, and probable harmful exposures, such as transfusions of specific blood compounds, drugs, environmental exposures, and invasive medical interventions. Finally, potential interventions to promote long-term health must be studied in controlled clinical trials, including anti-inflammatory medications, hormones, and nutrients.
Conclusions
Optimizing length growth among VPIs remains challenging due to a lack of knowledge about optimal growth patterns. Preventing postnatal length growth failure requires a comprehensive nutritional regimen that provides adequate nutritional support as soon as possible after birth and is maintained throughout the infant’s hospital stay and later at home. However, optimal nutrition alone appears to be insufficient for preventing poor outcomes, as VPIs are deficient in critical hormones and growth factors whose levels normally increase in late gestation and undergo adaptive responses that influence postnatal length growth and the accumulation of lean, bone, and fat mass. The pathological conditions of VPIs, in which inflammation plays a critical role, can alter macronutrient supply and utilization as well as neonatal hormone production and action, thereby altering their postnatal length growth.
Research is needed to elucidate the pathophysiology of the nutritional, inflammatory, hormonal, and growth factors involved in VPI stunting and provide the basis for the development of preventive, diagnostic, and therapeutic approaches. Future studies should also investigate antiinflammatory therapies that can prevent inflammationdriven injury; specifically, the supplementation of growthpromoting hormones and factors may prevent length growth restriction among and would likely improve the short- and long-term outcomes of VPIs. However, until the efficacy and safety of these interventions have been clinically proven, inflammation-induced length growth restriction cannot be reduced other than by improving the health of VPIs. Finally, enhancing linear growth outcomes depends on our ability to understand the complex inflammatory-endocrine interactions and pathophysiologic changes that occur during the early clinical course. Moreover, we must accumulate evidence of neonatal determinants of the stunted phenotype to adequately guide the choice and application of clinical strategies aimed at optimizing the clinical care of VPIs.
PDF Links
PubReader
ePub Link
XML Download
PubMed
Download Citation
