Background
With an increase in the understanding of short-term and long-term health-influencing factors that affect SGA, the perinatal medical community has focused on the prevention and management of nutrition of SGA infants. Regarding the incidence of SGA, China (6.5% incidence) ranks fifth globally (16% incidence) [
1]. In 2016, the WHO defined SGA as a newborn whose birth weight is below the 10
th percentile of the birth weight for infants of the same sex of the same gestational age or whose Z-value of birth weight is < –1.28. The Fenton growth curve (2013) [
2] is used for the diagnosis of SGA. SGA can be divided into premature SGA, full-term SGA, and overdue SGA, among which premature SGA is affected by intrauterine growth retardation and immature gestational age. The risk of early complications after birth and perinatal death increases, and it can also lead to many long-term complications such as adult cardiovascular diseases, insulin resistance, and neurocognitive dysfunction, which increases the burden on society and families.
Guellec et al. [
3] established a correlation between postnatal growth impairment in infants with SGA and cognitive deficits and learning difficulties. This finding has been supported by additional studies. For example, in their publication in the Journal of Pediatrics, Kerstjens et al. [
4] discovered a connection between postnatal growth impairment in SGA infants and delayed intellectual development and learning difficulties. Euser et al. [
5] also identified an association between postnatal growth impairment in SGA infants and behavioral and emotional problems. These research outcomes emphasize the significance of monitoring and intervening in the postnatal growth of SGA infants to mitigate the occurrence of extrauterine growth restriction (EUGR) and enhance their neurodevelopment and growth. Currently, there is no international consensus regarding the optimal postnatal growth pattern for preterm SGA infants. It is imperative to closely monitor the growth pattern of preterm newborns to detect any deviations from the norm. Early and appropriate catch-up growth plays a beneficial role in the physical growth and neurodevelopment of SGA children. Therefore, it is essential to develop reliable methods for accurately identifying infants with genuine extrauterine growth restriction, comprehending the factors influencing the occurrence of EUGR, and providing adequate and appropriate nutrition. These measures are crucial for ensuring successful catch-up growth [
6,
7].
However, as a consequence of intrauterine growth retardation, SGA infants exhibit slow growth and development. Consequently, it becomes challenging for the growth and development parameters of SGA infants to reach the 10
th percentile value for the corresponding gestational age upon discharge. Thus, it takes a long time to complete the catch-up growth [
8]. Therefore, the incidence of extrauterine growth retardation (EUGR) in SGA infants is significantly higher than the incidence of EUGR in non-SGA infants. Many studies have reported that SGA is an independent risk factor for EUGR [
9,
10].
EUGR is related to intrauterine growth retardation (IUGR). Studies generally refer to the Fenton growth curve (2013) and define the 10
th percentile of the weight, height, and head circumference at the corrected gestational age of 36 weeks or at discharge as EUGR and the 3
rd percentile below the growth curve as severe EUGR. By this cross-sectional definition, the incidence of EUGR in SGA is 87.6% ~ 98.5% [
9,
11], which is significantly higher than 44.44% in non-SGA [
9]. Some researchers have suggested that the occurrence of EUGR in SGA is a continuation of intrauterine growth retardation but not “real EUGR” [
12]. Therefore, the percentile (
P-value) of the Fenton growth curve cannot reflect the growth pattern of SGA after birth. To better reflect the growth status of premature infants after birth, Simon et al. [
13] suggested that the change in the Z scores between the weight at discharge and birth weight (ΔZ value) should be be used as part of the longitudinal definition to evaluate EUGR. The Z-score indicates how far the infant’s weight and height are from the 50
th percentile or the median of the reference growth charts for infants of the same age and sex, i.e., Z value = (measured value-average value of the same gestational age and gender)/standard deviation of this gestational age and gender). Studies have shown that dynamic longitudinal definition is more effective than cross-sectional definition in predicting adverse neurodevelopmental outcomes at a 2-year follow-up [
14]. Furthermore, longitudinally defined EUGR is associated with weight and head circumference deficits at 24–30 months of age [
15]. Therefore, the longitudinal definition is superior to the cross-sectional definition in predicting long-term outcomes in preterm infants, and whenever feasible, it should be the preferred method for diagnosing EUGR. Therefore, the ΔZ value might be more suitable for analyzing the extrauterine growth of individuals after birth [
13]. We conducted a national prospective multicenter study in China to analyze the real-world incidence of EUGR and risk factors that affect very premature infants (VPI) in SGA, based on the ΔZ value of weight.
Objective and methods
Study population
This study encompassed a prospective survey conducted across multiple centers from September 2019 to December 2020. Data for the study were gathered from 28 tertiary hospitals located in seven regions of China, including the northeastern, northern, eastern, central, southern, northwestern, and southwestern regions. The protocol was approved by the Ethics Committee of Women and Children’s Hospital affiliated with Xiamen University/Xiamen Maternity and Child Health Care Hospital (KY-2019–016), and the study was registered in the Chinese Clinical Trials Registry (
http://www.chictr.org.cn) with the registration number ChiCTR1900023418. Prior to participating in the study, written informed consent was obtained from the parents, ensuring their full understanding and agreement. The methodology employed in this study adhered to the applicable guidelines and regulations, ensuring its compliance with ethical standards.
We collected the clinical data of VPI with SGA hospitalized in the above mentioned multicenters. Inclusion criteria: ① SGA; ② Birth gestational age < 32 weeks; ③ Hospitalization time > 2 weeks; ④ Admission within 24 h after birth. Exclusion criteria: ① Congenital malformation or genetic metabolic disease; ② Death, interruption of treatment, or automatic discharge during hospitalization; ③ Incomplete data.
The VPI with SGA were divided into the EUGR and non-EUGR groups
A change in the Z-score (△Z value) of weight by more than 1.28 between two points (discharge and birth) was considered to be EUGR, and a change in the Z-score (△Z value) of weight by less than 1.28 was considered to be non-EUGR [
16].
Methods
Using a unified questionnaire, perinatal data of VPI with SGA were collected (gestational age at birth, Z value of physical indices at birth, sex, delivery mode, multiple births, prenatal glucocorticoid administration, and the 5-min Apgar score), maternal and pregnancy complications (gestational hypertension and gestational diabetes), growth and nutritional status during hospitalization [maximum weight loss, the age of recovering birth weight, the average weight gain velocity (GV), the ΔZ-value of physical indices at discharge, start time of enteral feeding, the age of reaching total enteral nutrition, cumulative fasting days, breast milk volume after the addition of human milk fortifier (HMF) and days needed for full fortification, the age of reaching the standard of oral calorie, cumulative calorie intake in the first week of hospitalization, cumulative dose of amino acids and fat milk in the first week of hospitalization, the duration of parenteral nutrition (PN)],main treatment conditions (invasive mechanical ventilation time, total oxygen consumption time, the use rate of postnatal hormones, cumulative duration of antibiotics used, hospitalization time) and main complications during hospitalization [neonatal respiratory distress syndrome (NRDS), early-onset sepsis (EOS), feeding intolerance (FI), patent ductus arteriosus with hemodynamic changes (hsPDA), neonatal necrotizing enterocolitis (NEC) ≥ stage 2, bronchopulmonary dysplasia (BPD), late-onset sepsis (LOS), grade III-IV intraventricular hemorrhage (IVH), periventricular leukomalacia (PVL), parenteral nutrition-associated cholestasis (PNAC), retinopathy of prematurity (ROP) requiring intervention, metabolic bone disease of prematurity (MBDP), EUGR], and other clinical data were also collected.
Definition or diagnostic criteria of related diseases
(1) SGA is a newborn whose birth weight is lower than the 10
th percentile of the birth weight of a newborn of the same sex, and gestational age or whose birth weight Z value is < –1.28; (2) The EUGR evaluation criteria refer to the Fenton growth curve [
2] published in 2013. ① The evaluation criteria for percentile (
P value) were as follows: VPI with a weight below the 10
th percentile, based on the 2013 Fenton growth curve, at 36 weeks of corrected gestational age or upon discharge;② ΔZ value evaluation criteria: △Z value of weight = (Z value of weight at 36 weeks of corrected gestational age or during discharge-Z value of birth weight); EUGR is defined as weight ΔZ value < –1.28 [
16]; (3) BPD is defined as a newborn with persistent oxygen dependence for ≥ 28 days after birth [
17]; (4) EOS and LOS diagnostic criteria [
18] refer to the consensus of experts on the diagnosis and treatment of neonatal sepsis (2019 edition); (5) FI diagnostic criteria [
19]: the stomach residue exceeds 50% of the previous feeding amount, accompanied by vomiting and/or abdominal distension; the feeding plan fails, including reduced, delayed, or interrupted enteral feeding; (6) Diagnostic criteria of MBDP: refers to the consensus of clinical management experts of metabolic bone disease in premature infants (2021) [
20]; (7) NEC ≥ stage 2: was defined as Bell stage≥2 [
21]; (8) Diagnostic criteria of hsPDA: PDA catheter diameter > 1.5 mm, accompanied by heart murmur, tachycardia, rapid respiration, increased pulse pressure, hypotension; (9) The complications such as NRDS, IVH ≥ stage 3, PVL, PNAC, and ROP need intervention; refer to the diagnostic criteria [
22] in
Practical Neonatology (5th Edition).
Definition of enteral nutrition
(1) Start time of enteral feeding (h): the time to start oral feeding/nasal feeding of breast milk or formula milk after birth (excluding colostrum oral care); (2) Total enteral feeding time (d): the time required for oral milk intake to reach 150 mL/kg/d; (3) Time for total and oral calorie intake to reach the target: the recommended calorie intake standard was 110 kcal/(kg·d). (4) Mean GV [g/(kg·d)]: [1,000 × ln (Wn/W1)]/(Dn-D1) after regaining birth weight. In this formula, Wn indicates weight (g) at discharge, W1 indicates birth weight (g), Dn indicates the length of hospital stay (day), and D1 indicates the time to regain birth weight (day) [
23].
Statistical analysis
Statistical analysis was conducted using the SPSS 22.0 software. Measurement data that exhibited a normal distribution were reported as mean ± SD, and a comparison between groups was performed using independent-samples t-tests. Non-normally distributed quantitative data were presented as the median and interquartile ranges, and the Mann–Whitney U test was conducted for comparison between groups. The count data were presented as the number and rate of cases, and the Chi-squared test or Fisher’s exact test was conducted for comparison between groups. Variables that demonstrated a significance level of P < 0.05 in the single-factor analysis were selected for inclusion in the multivariate analysis. A stepwise approach was employed to screen these variables by constructing a multivariate logistic regression model, with a significance level (α) set at 0.05. All differences among and between groups were considered to be statistically significant at P < 0.05.
Discussion
Clark [
24] first proposed the concept of EUGR in 2003. He plotted a growth curve to evaluate the incidence of EUGR. However, there are still many controversies about the timing and standard of EUGR evaluation, leading to differences in clinical recommendations and practice [
25]. The Fenton curve, which is the revised growth curve for different sexes published in 2013, was established using data from four million premature infants. This comprehensive dataset included information from developed countries such as Germany, Italy, the United States, Australia, Canada, and Scotland, spanning the years 1991 to 2007. The Fenton curve serves as a valuable tool for monitoring and assessing the growth and development of premature infants. According to the data on the gestational age, weight, height, and head circumference of newborns, the accurate
p-value and the standardized Z value [
2] associated with the growth curve of the current growth of newborns can be calculated. This is the most commonly used method to evaluate the intrauterine and extrauterine growth of premature infants. Birth weight serves as a widely adopted indicator for the clinical assessment of newborn growth and nutritional status due to its simplicity, accurate measurement, and reliable repeatability. In clinical practice, the presence of EUGR is typically evaluated based on the weight of premature infants at 36 weeks of corrected gestational age or at hospital discharge. For the same study population, a big difference in the evaluation was found depending on whether the
p-value or the △Z value on the curve was considered as the criterion. Griffin et al. [
26] used two methods to evaluate the incidence of EUGR in 25,899 VPI with a birth weight of 500 ~ 1500 g and gestational age of 22 ~ 32 weeks in California, USA. The incidence of EUGR was 53.3% with the
p-value of weight at discharge < 10% and 41.4% with △Z value < –1. Premature infants with gestational age ≤ 32 weeks at Mount Sinai Medical Center in the United States were evaluated by Lin et al. [
16]. The incidence of EUGR at discharge was found to be 35.3% when using the diagnosis criterion of a discharge weight Z score < –1.28 (equivalent to a
p-value < 10
th percentile). For a △Z (change in Z score) of less than –1.28, the EUGR incidence was 25.5%, and for a △Z of less than –2, the EUGR incidence was 4.5%. There were considerable differences among the three evaluation methods. The incidence of SGA in this cohort was 5.30%, which was slightly lower than the national average [
1] and slightly higher than that reported in an American study (4.12%) [
27]. In our evaluation of 133 VPI with SGA cases, the incidence of EUGR was 98.50% following the
p-value criterion and 36.84% following the criterion of △Z < –1.28; there was a discrepancy of 61.66% in this study due to the difference between the evaluated population and the △Z value. The incidence of EUGR differed considerably with different evaluation methods. The
p-value evaluation method was based on the horizontal evaluation of group data, while the △Z value was based on the vertical evaluation and objective analysis of individual data. Longitudinal evaluation offers a more accurate depiction of the actual growth pattern of neonates [
28,
29]. Fenton et al. [
30] highlighted shortcomings in the cross-sectional definition itself, emphasizing its limited ability to accurately predict adverse outcomes. The utilization of the 10th percentile as a subjective threshold may result in an overdiagnosis of EUGR, potentially causing parental distress and increasing the risks of overfeeding and obesity. In contrast, the longitudinal definition considers crucial factors such as birth weight and gestational age. It not only helps mitigate the issue of overdiagnosis of EUGR to some extent but also provides a more precise prognosis for preterm infants. Furthermore, in comparison to the cross-sectional definition, the dynamic delta value-based definition demonstrates superior effectiveness in predicting adverse neurodevelopmental outcomes over a 2-year follow-up period [
14,
31]. Hence, the delta value-based definition proves to be superior in predicting the long-term outcomes of preterm infants. In our study, we employed the ΔZ value to assess the true incidence of EUGR in VPI with SGA, with the aim of establishing scientific standards for optimizing nutritional strategies for this specific population. Table
1 demonstrates the variations in EUGR diagnosis when different definitions are used, and the application of the longitudinal definition partially mitigated the influence of IUGR. Recently, some researchers have proposed using the lowest postnatal weight age as the reference point for calculating ΔZ value changes. This approach not only offers partial prediction of long-term adverse outcomes but also avoids the impact of physiological postnatal weight loss [
32]. Building on this concept, Maiocco et al. [
15] conducted a study and revealed that a ΔZ value decrease for head circumference exceeding one standard deviation between discharge and recovery of birth weight within 14 to 21 days after birth is a significant risk factor for neurodevelopmental delays. Unfortunately, this aspect was not considered in the design of our study, and precise evaluation data for EUGR within the 14 to 21-day period were not included in the paper. This limitation provides a direction for future research endeavors.
The results of the univariate analysis showed that the non-EUGR group had a higher birth weight (
P < 0.001) and a larger Z-value of birth weight (
P = 0.017). The results of the multivariate analysis showed that high birth weight was a protective factor related to the occurrence of EUGR in VPI with SGA (OR = 0.997, 95% CI: 0.994 ~ 0.999,
P = 0.024). Our results were similar to those of previous studies [
33]. The results showed that the birth weight of infants in the EUGR group was lower, the intrauterine growth was more restricted, and the organs and tissues were relatively underdeveloped. EUGR is caused by scarcity of nutrients in the uterus, greater nutritional demand, and higher energy metabolism, which is more likely to lead to nutritional deficiency and premature infant-related complications after birth [
34]. The postnatal nutritional status of VPI with SGA is closely associated with the occurrence of EUGR. The findings from the multivariate analysis indicated that a prolonged duration for breast milk fortification and the slow recovery of birth weight were identified as independent risk factors for EUGR in VPI with SGA, while high GV was found to be a protective factor against EUGR. Breast milk is the best source of nutrition for babies, especially premature infants. However, the energy and nutrients in breast milk cannot meet the growth-related needs of premature infants at the early stages after birth, especially of premature SGA infants. Therefore, HMF containing multiple nutrients is commonly added to breast milk [
35].
Our results showed that the quantity of HMF added to milk was more in the EUGR group than that in the non-EUGR group (100 mL vs. 88 mL), and it took longer (9 d vs. 3 d) to reach full fortification in the EUGR group. In China, experts recommend initiating the use of HMF for premature infants when their breastfeeding volume reaches 50–80 mL/(kg·d). It is advised to achieve standard adequate fortification within 3–5 days. A study demonstrated that adding HMF when the breastfeeding volume reaches the recommended threshold was the most effective approach in reducing the incidence of EUGR [
36].
In a prospective randomized controlled study conducted by Bozkurt et al. [
37], it was observed that achieving full-dose intensive breastfeeding at an earlier stage resulted in higher GV in VPI. This, in turn, contributed to a shorter duration of birth weight recovery. The GV was higher during hospitalization, which was a significant independent protective factor to avoid EUGR and promote the development of the nervous system [
38]. Consistent with the findings of this study, Jeffrey et al [
39] documented an increase in GV from 11.8 to 12.9 g/kg/day, accompanied by a decrease in the incidence of EUGR in very low birth weight infants (VLBWI) from 64.5% to 50.3%. These results suggested that more attention should be paid to enteral nutrition support for VPI with SGA. By following the recommendations of HMF experts, full breast milk fortification can be achieved at the earliest, the growth rate can be increased, and the recovery time of birth weight can be shortened. These factors play an important role in reducing the incidence of EUGR.
Early postnatal complications directly affect the nutritional supply and extrauterine growth and development of VPI with SGA. The findings from the univariate analysis revealed that the 5-min Apgar score was lower (
P = 0.012), and the duration of invasive ventilation was longer (
P = 0.003) in the EUGR group compared to the non-EUGR group. The severity of illness after birth hindered the effective implementation of recommended early enteral nutrition measures, consequently leading to delayed initiation of enteral feeding. The average starting time of enteral feeding of the EUGR group in this study was later than that in the non-EUGR group (36.00 h vs. 21.75 h). A delay in enteral feeding might cause gastrointestinal mucosa atrophy and delayed functional maturity and also increase the incidence of FI (
P = 0.031) and NEC [
40,
41]. The incidence of LOS among infants in the EUGR group was higher than that among infants in the non-EUGR group (
P = 0.022), which led to longer administration of antibiotics (
P = 0.011), greater extent of intestinal microecology disorder and a higher incidence of NEC among infants in the EUGR group [
42]. The incidence of hsPDA in the EUGR group was higher (
P = 0.018), the proportion of blood transfusion was higher (
P = 0.01), and the frequency of blood transfusion was higher (
P < 0.001) than that in the non-EUGR group. These factors might increase the risk of NEC [
43]. In a study, the incidence rate of NEC in premature infants was 2% ~ 5%, among which the incidence rate of very low birth weight infants was 4.5% ~ 8.7% [
44]. Our study observed that the incidence of NEC ≥ stage 2 in the EUGR group was 20.4%. However, no significant difference was found in the occurrence of NEC requiring surgery between the EUGR and non-EUGR groups (
P = 0.625). The results of the multivariate analysis confirmed that NEC ≥ stage 2 was an independent risk factor for EUGR (OR = 5.835, 95% CI: 1.051–32.384,
P = 0.044), which showed that the risk of EUGR increased by 5.8 times after NEC occurred in VPI with SGA. These results were similar to those of previous studies [
45]. In this study, most infants with NEC ≥ stage 2 were treated conservatively in internal medicine, and clinicians were often very cautious about the fasting time and the indications for re-starting milk, which might lead to a decrease in the nutrient intake [
11]. A comprehensive assessment of the risk balance between FI and NEC should be performed to avoid unnecessary fasting and prevent NEC from worsening.
The results of the multivariate analysis also showed that the male sex was a protective factor of EUGR in VPI with SGA. Male infants with premature SGA were reported to have a faster physical catch-up growth in the early postnatal period than female infants [
46]. This might be related to the differences in the effects of gender on the physical growth of premature SGA, although it needs to be confirmed in future studies.
Acknowledgements
The authors thank the neonatal units in the following hospitals and centres for providing data for this survey (Information of the Chinese Multicenter EUGR Collaborative Group).
Department of Neonatology, Women's and Children's Hospital Affiliated to Xiamen University/Xiamen maternal and Child Health Hospital, Xiamen, Fujian 361003, China (WS, ZZ, XL). Department of Neonatology, The Third Affiliated Hospital of Guangzhou Medical University, Guangzhou, Guangdong 510150, China (FW, Qianxin Tian, and Qiliang Cui). Department of Pediatrics, Shengjing Hospital of China Medical University, Shenyang, Liaoning 110000, China (JM, Yuan Yuan and Ling Ren). Department of Neonatology, Guiyang maternal and Child Health Hospital·Guiyang Children‘s Hospital, Guiyang, Guizhou 550002, China (LL, Bizhen Shi, and Yumei Wang). Department of Pediatrics, Peking University Third Hospital, Beijing 100191, China (YC, Jinghui Zhang, and XT). Department of Neonatology, Children’s Hospital of Fudan University, Shanghai 201102, China (Yan Zhu, WS, RZ and CC). Department of Neonatology, Guangdong Province Maternal and Children’s Hospital, Guangzhou, Guangdong 510030, China ((Jingjing Zou and XY). Department of Neonatology, General hospital of Ningxia Medical University, Yinchuan, Ningxia 750001, China (Yuhuai Li, Baoyin Zhao, and YQ). Department of Neonatology, Children’s Hospital of Hebei Province, Shijiazhuang, Hebei 050031, China (Shuhua Liu and LM). Department of Neonatology, Children' hospital of Nanjing Medical University, Nanjing, Jiangsu 210000, China (Ying Xu and RC). Department of neonatology, The first hospital of Jilin university, Changchun, Jilin 130000, China (Wenli Zhou and HW). Department of Neonatology, Quanzhou maternity and Children's Hospital, Quanzhou, Fujian 362000, China (Zhiyong Liu and DC). Department of Pediatrics, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei 430000, China (Jinzhi Gao, Jing Liu, and Ling Chen). Department of Neonatology, Liaocheng people's hospital, Liaocheng, Shandong 252000, China (Cong Li, Chunyan Yang, and Ping Xu). Department of Neonatology, the Affiliate Hospital of Inner Mongolia Medical University, Hohhot, Inner Mongolia 010010, China (Yayu Zhang, Sile Hu, and Hua Mei). Department of Neonatology, Suzhou Municipal Hospital, Suzhou, Jiangsu 215002, China (Zuming Yang and Zongtai Feng). Department of Neonatology, The Third Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan 450052, China ((Er-Yan Meng and Li-Hong Shang). Department of Neonatology, Chengdu Women' and Children's Central Hospital, School of Medicine, University of Electronic Science and Technology of China, Chengdu, Sichuan 611731, China (Shaoping Ou and Rong Ju). Department of Neonatology, Hunan children's Hospital, Changsha, Hunan 410000, China (Gui-Nan Li). Department of Neonatology, People's Hospital of Xinjiang Uygur Autonomous Region, Urumqi, Xinjiang 830001, China (Long Li). Department of Neonatology, Guangzhou Women and Children's Medical Center, Guangzhou, Guangdong 510150, China (Zhe Zhang). Department of Neonatology, Shanghai Children's Medical Center, Shanghai, 200120, China (Fei Bei). Department of Neonatology, Children’s Hospital of Chongqing Medical University, Chongqing, 400014, China (Chun Deng). Department of Neonatology, The First People's Hospital of Yulin, Yulin, Guangxi 537000, China (Ping Su). Department of Neonatology, the People’s Hospital of Baoji, Baoji, Shanxi 721000, China (Ling-Ying Luo). Department of Pediatrics, Affiliated Hospital of Qingdao University, Qingdao, Shandong 266000, China (Xiao-Hong Liu). Departments of Neonatology, Shandong Provincial Hospital Affiliated to Shandong First Medical University, Jinan, Shandong 250021, China (Li-Jun Wang). Departments of Neonatology, Xi'an Children's Hospital, Xi’an, Shanxi 710003, China (Shu-Qun Yu).
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.