Early Surge in Circulatory Adiponectin Is Associated With Improved Growth at Near Term in Very Preterm Infants.

ORIGINAL

ARTICLE

Early Surge in Circulatory Adiponectin Is Associated With Improved Growth at Near Term in Very Preterm Infants Ingrid Hansen-Pupp1, Gunnel Hellgren2, Anna-Lena Hård3, Lois Smith4, Ann Hellström3 *, Chatarina Löfqvist3* 1 Department of Pediatrics, Institute of Clinical Sciences Lund, Lund University and Skane University Hospital, Lund, Sweden; 2Department of Pediatrics, Institute of Clinical Sciences, The Sahlgrenska Academy at University of Gothenburg, Gothenburg, Sweden; 3Department of Ophthalmology, Institute of Neuroscience and Physiology, The Sahlgrenska Academy at University of Gothenburg, Gothenburg, Sweden; 4Department of Ophthalmology, Boston Children’s Hospital, Harvard Medical School, Boston, MA.

Context: Adiponectin enhances insulin sensitivity and may play a role in fetal and postnatal growth. Objective: To determine if early postnatal adiponectin concentration change is related to postnatal growth in very preterm infants. Setting and Design: In-hospital, prospective, longitudinal cohort study. Patients: 52 preterm infants with a gestational age (GA) of 26.0 ⫾ 1.9 (SD) weeks and birth weight (BW) of 889 ⫾ 284 g. Interventions: Analysis of adiponectin was performed on cord blood at birth and peripheral blood at 72 h, day 7, and then weekly until postmenstrual age (PMA) 40 weeks. Weight, length, and head circumference (HC) measurement was performed weekly and standard deviation scores (SDS) calculated. Energy and protein intake was calculated daily from birth until PMA 35 weeks. Results: Mean adiponectin concentration increased from 6.8 ⫾ 4.4 ␮g/mL at 72 h to 37.4 ⫾ 22.2 ␮g/mL at 3 weeks; during days 3–21, it was 21.4 ⫾ 12 ␮g/mL and correlated with GA at birth (r ⫽ 0.46, P ⫽ .001; BW: r ⫽ 0.71, P ⬍ .001; BWSDS: r ⫽ 0.42, P ⫽ .003). Furthermore, mean adiponectin during days 3–21 correlated with weightSDS, lengthSDS, and HCSDS (r ⫽ 0.62, 0.65, and 0.62, respectively; all P ⬍ .001) at PMA 35 weeks). Energy intake (kcal/kg/d) correlated with mean adiponectin during days 3–21 (r ⫽ 0.35, P ⬍ .013). Conclusions: In very preterm infants, adiponectin concentrations increased markedly in the first 3 weeks, and a greater increase was associated with improved postnatal growth.

nfants born before the third trimester of gestation experience a high risk of malnutrition and growth restriction after birth (1). The abnormal growth pattern of preterm infants has been associated with early morbidity (2) and with impaired neurodevelopmental outcome, (3) and implicated in mechanisms for later metabolic consequences such as obesity, hypertension, cardiovascular disease, and type 2 diabetes (4).

I

Compared with infants born at term, preterm infants are lighter and shorter and have smaller head circumference (HC) at term age (5). Their body composition also differs, with less lean tissue and a higher percentage of total body fat (6). In addition, preterm infants have different fat tissue distribution at term with decreased subcutaneous and increased intra-abdominal adipose tissue (7). The optimal postnatal growth velocity for preterm

ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2015 by the Endocrine Society Received January 9, 2015. Accepted March 26, 2015.

Abbreviations:

doi: 10.1210/jc.2015-1081

J Clin Endocrinol Metab

jcem.endojournals.org

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 May 2015. at 22:24 For personal use only. No other uses without permission. . All rights reserved.

1

2

Adiponectin and postnatal growth in very preterm infants

infants is not known, but accelerated postnatal growth has been associated with increased adiposity (7). Adipose tissue is an energy-storing tissue but is also recognized as an endocrine organ that secretes different metabolic signaling adipokines with diverse regulatory functions in metabolic homeostasis (8). Adiponectin is the most abundant adipokine, mainly produced by adipocytes in brown and white adipose tissue. It is secreted by mature adipocytes and is therefore proposed to be a marker of adipose tissue maturation (9). Adiponectin exerts a variety of actions, including insulin-sensitizing, anti-inflammatory, and antiapoptotic effects, and plays a central role in regulating energy homeostasis (10, 11). Expressed in several fetal tissues early in gestation (12), its concentration increases markedly during gestation and correlates with birth weight (BW) and BW standard deviation score (SDS) (13). Circulatory concentrations of adiponectin obtained at discharge and change in concentrations between birth and term age have been associated with postnatal weight gain in preterm infants (14, 15). The early temporal postnatal change in adiponectin concentrations from birth and onwards in very preterm infants has not yet been evaluated. Our aim was to prospectively assess weekly concentrations of adiponectin and its interaction with other postnatal events such as nutritional intake and steroid treatment, where we hypothesized that early postnatal adiponectin concentrations would be related to postnatal growth in very preterm infants.

Materials and Methods This study was conducted as part of a prospective longitudinal cohort study between January 2005 and May 2007 in which circulatory IGF-I concentrations were re-


lated to postnatal growth (16). The Regional Ethical Review Board in Lund, Sweden, approved the study. Inclusion criteria were a gestational age (GA) ⬍31 weeks at birth, absence of major congenital anomalies, and written informed parental consent from both parents. All pregnancies were dated by ultrasound at 17–18 gestational weeks. In the original study, 52 infants remained in the study until term age. Due to limited serum volumes in three infants, longitudinal adiponectin concentrations could be analyzed in 49 infants Growth measurements The procedure for growth measurements has been described previously (16). Standardized measurements of weight, length, and HC were performed at birth and then weekly on the same weekday as blood sampling for adiponectin, continued until discharge. A standard deviation score (SDS) was calculated for all measurements of each respective growth variable. LengthSDS and HCSDS at birth and weightSDS, lengthSDS and HCSDS at PMA 35 weeks were computed from a gender-specific growth reference in a Swedish population (17). WeightSDS at birth was calculated from an intrauterine growth curve based on ultrasonically estimated fetal weights in Scandinavia. BW small for GA (SGA) was defined as BW more than 2 SD below the age-related mean of the population (18). The lowest SDS for each respective growth parameter was defined as the lowest postnatal SDS that was followed by gradually increasing SDS

Nutritional strategy and calculation of nutritional intake The nutritional strategy has been described previously (16). Briefly, it was based on enteral nutrition using maternal or donor breast milk, and additional parenteral nutrition was initiated as soon as possible after birth. Minimal enteral feeding was started within 3 hours of age and administered every 2 to 3 hours (1–2 mL/meal) with a gradual increase in volume. Administered breast milk was analyzed at day 7 and then weekly for protein (g/100 mL) and energy (kcal/100 mL) content in a 24-hour sample until a PMA of at least 35 weeks. Supplemental individualized fortification based on results from analyzed breast milk was performed using a commercial bovine milk fortifier. Enteral and parenteral daily intakes of protein (g/kg/d) and energy (kcal/ kg/d) were prospectively registered and calculated from birth until at least PMA 35 weeks.

Clinical data

Figure 1. Mean (95% CI) concentrations of adiponectin (␮g/mL) in relation to postnatal age (weeks).

In 12 infants, treatment with hydrocortisone (3– 6 mg/kg/d) was initiated because of resistant arterial hypotension. Betamethasone was administered to facilitate ventilator weaning at a minimum postnatal age of 10 –14 days (0.2 mg/kg/d) initially, with gradual dose tapering usually over 5–10 days. The daily administered dose (mg/kg) of both drugs was registered, and the accumulative dose until PMA 35 weeks was calculated. Total administered steroid dose was estimated by converting betamethasone dosage into hydrocortisone equivalents (1:40).


doi: 10.1210/jc.2015-1081

Blood sampling and quantitative analysis of adiponectin Blood samples were obtained from umbilical cord blood at birth and from neonatal blood prior to enteral feeding at 72 hours of age, at 7 days postnatally, weekly until at least a GA 35 weeks and again at a term age (PMA 40 weeks). Sampling was initially performed from an umbilical or peripheral arterial catheter and later by venous puncture. After centrifugation, serum samples were stored at – 80 C° until assayed. All serum samples were diluted 1:306 and adiponectin levels assayed using a human adiponectin ELISA kit (E091M, Mediagnost, Reutlingen, Germany). The intra-assay coefficients of variation were 3.8% at 3.9 ␮g/mL and 4.7% at 13.1 ␮g/mL; the interassay coefficient of variation was 16.3% at 9.9 ␮g/mL. For values ⬎ 80 ␮g/mL, samples were further diluted using assay diluent and the assay repeated so that the results fell within the range. Samples were corrected for respective dilution.

Statistical evaluation Statistical analysis was performed using SPSS 20 for Microsoft Windows (IBM, Armonk, NY). P values ⬍ 0.05 were considered significant. PMA 35 weeks was chosen as an endpoint instead of term age to allow evaluation of the impact of nutritional and steroid intake in relation to adiponectin concentrations and growth. According to the temporal profile of adiponectin concentrations, mean adiponectin values were calculated during the interval after birth when increased concentrations were observed in all infants: birth up to 3 weeks of age (adiponectin days 3–21). Cord blood samples were not included in the calculation of mean values after birth because of a significant number of missing cord blood samples (n ⫽ 18). Univariate analysis of continuous variables and categorical variables was performed using Pearson correlation or one-sample t-tests, as appropriate. For variables without normal distribution (adiponectin cord blood concentrations, administered steroid dose, and postnatal age (weeks) for maximal adiponectin


concentration), the Spearman rank correlation coefficient or Mann–Whitney U test was used. Adjustment for other variables was performed using multiple linear regression analysis. To evaluate the contribution of adiponectin to postnatal growth retardation (weightSDS, lengthSDS, and HCSDS at PMA 35 weeks) forward linear regression was used. Independent variables entered into the the model were mean adiponectin concentrations during days 3–21, GA at birth, BWSDS, mean protein (g/kg/d) or energy (kcal/kg/d) intake from birth until PMA 35 weeks, and estimated total steroid intake from birth until PMA 35 weeks. When any two candidate variables were highly correlated (energy and protein intake) only one variable was included in the multiple regression models to avoid problems with multicollinearity.

Results The clinical characteristics of the study population are shown in Table 1. Adiponectin concentrations in cord blood and during days 3–21 The median (range) adiponectin concentration in cord blood (n ⫽ 31) was 2.2 (0.13–12.4) ␮g/mL and correlated positively with adiponectin concentrations at PMA 35 weeks (rs ⫽ 0.58, P ⫽ .001). Mean adiponectin during days 3–21 was 21.4 (12.0) ␮g/mL and correlated positively with adiponectin concentrations at PMA 35 weeks (r ⫽ 0.71, P ⬍ .001). All correlations remained significant after adjustment for GA at birth and BW. Table 1. Clinical characteristics of the study population (n ⫽ 49). Gestational age (wks), mean (SD)

Figure 2. Mean (95% CI) concentrations of adiponectin (␮g/mL) in relation to postmenstrual age (weeks). Solid line denotes infants appropriate for gestational age (AGA), and dashed line denotes infants small for gestational age (SGA).

3

Birth weight (g), mean (SD) Antenatal steroid treatment, n (%) Birth weight small for gestational age, n (%) Male, n (%) Treatment with systemic steroid, n (%) Mean energy intakea (kcal/kg/d), mean (SD) Mean protein intakea (g/kg/d), mean (SD) At birth Weight SDS, mean (SD) Length SDS, mean (SD) SDS head circumference at birth, mean (SD) At 35 weeks postmenstrual age Weight SDS, mean (SD) Length SDS, mean (SD) Head circumference SDS, mean (SD)

25.8 (1.9) 864 (262) 48 (98) 14 (29) 25 (49) 20 (41) 120 (10) 3.1 (0.25) ⫺1.0 (1.25) ⫺1.8 (1.5) ⫺0.5 (0.9)

⫺1.7 (1.4) ⫺2.8 (1.5) ⫺0.6 (1.3)

SDS, standard deviation score a

From birth until a postmenstrual age of 35 weeks.


4


Temporal changes in adiponectin concentrations Temporal postnatal change in mean adiponectin concentrations from birth onward is shown in Figure 1. Adiponectin concentrations increased steeply during the first 3 weeks after birth, from 6.8 ⫾ 4.4 (mean ⫾ SD) ␮g/mL at 72 hours to 37.4 ⫾ 22.2 ␮g/mL at 3 weeks (P ⬍ .001), followed by a gradual decrease. A peak in early postnatal adiponectin concentrations occurred at a median (range) postnatal age of 3 (27) weeks. The postnatal age in weeks for the adiponectin concentration peak inversely correlated with GA at birth (rs ⫽ – 0.57, P ⬍ .001); ie, the most immature infants had peak adiponectin concentrations at a later postnatal age. Absolute levels of the highest adiponectin concentration and GA at birth did not correlate. Mean adiponectin concentrations relative to PMA displayed a similar early increase and second later increase from PMA 34 weeks until PMA 40 weeks. Adiponectin increased from 19.9 ⫾ 10.6 (mean ⫾ SD) to 33.7 ⫾ 13.6 ␮g/mL (P ⬍ .001) (Figure 2). Adiponectin concentrations and perinatal variables Adiponectin concentrations in cord blood correlated positively with GA at birth (rs ⫽ 0.64, P ⫽ .001) and with BW (rs ⫽ 0.69, P ⬍ .001) whereas no correlation was seen with BWSDS. Mean adiponectin during days 3–21 correlated positively with GA at birth, BW, and BWSDS (r ⫽ 0.46, P ⫽ .001; r ⫽ 0.71, P ⬍ .001; r ⫽ 0.42, P ⫽ .003, respectively). In infants born SGA, mean adiponectin during days 3–21 was 12.6 ⫾ 7.9 ␮g/mL; it was 24.9 ⫾ 11.7 ␮g/mL in infants born appropriate for GA (AGA) (P ⫽ .001). This difference was independent of GA at birth. Longitudinal concentrations of adiponectin according to PMA remained significantly lower (P ⬍ .05) from birth until term age in infants born SGA, except at PMA 28 and 29 weeks, as compared to infants born AGA (Figure 2). Adiponectin concentration and nutritional intake Protein and energy intake during days 3–21 were 2.8 ⫾ 0.44 g/kg/d and 101 ⫾ 13 kcal/kg/d. Mean adiponectin concentrations correlated positively with mean energy intake (kcal/kg/d) during days 3–21 (r ⫽ 0.35, P ⫽ .013). The correlation did not remain significant after adjustment for GA at birth. No correlation was seen between mean adiponectin concentrations and mean protein intake (g/kg/d) during days 3–21. Adiponectin concentration and systemic steroid treatment Mean adiponectin concentrations during days 3–21 correlated negatively with total steroid intake (estimated


mg/kg) during the corresponding time period (r ⫽ – 0.54, P ⬍ .001). Clinical variables and growth at 35 weeks PMA GA at birth correlated positively with lengthSDS and HCSDS at PMA 35 weeks (r ⫽ 0.37, P ⫽ .008; r ⫽ 0.4, P ⫽ .004, respectively), whereas GA at birth and weightSDS at PMA 35 weeks showed no correlation (P ⫽ .083) (.WeightSDS,lengthSDS, and HCSDS at birth correlated with corresponding SDS for each growth parameter at PMA 35 weeks (r ⫽ 0.64, P ⬍ .001; r ⫽ 0.83, P ⬍ .0001; r ⫽ 0.65, P ⬍ .001, respectively). Accumulated protein intake from birth up to PMA 35 weeks correlated positively with weightSDS at PMA 35 weeks (r ⫽ 0.32, P ⫽ .028), and accumulated energy intake from birth up to PMA 35 weeks correlated positively with weightSDS,lengthSDS, and HCSDS at PMA 35 weeks (r ⫽ 0.39, P ⫽ .007; r ⫽ 0.35, P ⫽ .014; r ⫽ 0.36, P ⫽ .012, respectively). Total steroid intake from birth until PMA 35 weeks (estimated mg/kg) correlated negatively with weightSDS,lengthSDS, and HCSDS at PMA 35 weeks ( r ⫽ – 0.43, P ⫽ .002; r ⫽ – 0.6, P ⬍ .001; r ⫽ – 0.53, P ⬍ .001, respectively). Adiponectin concentration and growth at PMA 35 weeks Temporal change in adiponectin concentrations and weightSDS according to postmenstrual age are shown in Figure 3. Postnatal age (weeks) for maximal adiponectin concentrations correlated with the time point for lowest weightSDS (rs ⫽ 0.42, P ⫽ .003). Cord blood adiponectin correlated with HCSDS, rs⫽0.38, P ⫽ .037but not with weightSDS or lengthSDS at PMA 35 weeks. Correlations between mean adiponectin during days 3–21 and weightSDS at PMA 35 weeks in relation to infants born SGA/AGA are shown in Figure 4. Mean adiponectin during days 3–21 correlated positively with weightSDS, lengthSDS, and HCSDS at PMA 35 weeks (r ⫽ 0.63, P ⬍ .001; r ⫽ 0.65, P ⬍ .001; r ⫽ 0.62, P ⬍ .001, respectively and all correlations remained significant after adjustment for GA at birth, total steroid intake (estimated mg/kg), and mean protein (g/kg/d) or energy (kcal/kg/d) intake from birth until PMA 35 weeks. After inclusion of BWSDS as an additional independent variable, mean adiponectin during days 3–21 was not independently associated with growth variables at PMA 35 weeks. Predictive variables for weight SDS at PMA 35 weeks Results of multivariate analysis (forward regression) assessing the combination of variables with highest predictivity for weightSDS at PMA 35 weeks are presented in Table 2. The combination of mean adiponectin concen-


doi: 10.1210/jc.2015-1081


5

trations in dried blood spot samples of preterm infants identified increased adiponectin in relation to postnatal age at sampling (days 3–12) in preterm but not term infants (19). In term infants, reports on changes in adiponectin concentrations from birth and during the first postnatal week have discrepant results, with no postnatal change (20) or a moderate increase (21) Figure 3. Individual values for adiponectin (␮g/mL) and postnatal SD score for weight in relation to postmenstrual age (weeks). A: Circles denote infant concentrations of adiponectin (␮g/mL) identified. and solid line denotes mean concentrations of adiponectin (␮g/mL). B: Circles denote infant Immediately after birth, profound postnatal SD score for weight, and solid line denotes mean postnatal SD score. metabolic changes ensure sufficient glucose production during the trantrations during days 3–21, BWSDS and total steroid intake had the highest predictivity for weightSDS at PMA 35 sition from continuous placental supply of nutrients to external nutritional intake. For very preterm infants, this weeks. period is accompanied by an inevitable nutritional deficit (22). Adiponectin is one of several hormones regulating nutritional intake. Experimental studies suggest that adiDiscussion ponectin serves as a starvation signal and as an appetite This study is the first to describe weekly longitudinal stimulator and seems to decrease energy expenditure by changes in adiponectin concentrations in preterm infants decreasing oxygen consumption (10). We speculate that from birth until term age. We found that very preterm the observed early steep rise in adiponectin concentrations infants experience a profound immediate circulatory in- might be an effect of nutritional depletion and state of crease in adiponectin during the first three postnatal starvation present during the early postnatal period in very weeks. Furthermore, early postnatal adiponectin concen- preterm infants. trations correlated with weightSDS, lengthSDS, and HCSDS The postnatal growth pattern after very preterm birth at PMA 35 weeks. is characterized by an initial phase of growth restriction, Adiponectin concentrations increased more than 10 where lowest weightSDS is the time point for the transition fold from birth until 3 weeks of postnatal age, with the between the growth restriction and catch-up phases (16). most profound increase occurring between 7–14 postnatal The temporal coincidence between adiponectin concendays. A previous study of postnatal adiponectin concentration peak and lowest weightSDS might explain why a decline in adiponectin concentrations starts when the period of catch-up growth begins, ie, when infants enter from a catabolic into an anabolic state. The early postnatal increase in adiponectin levels was not related to GA at birth (data not shown), indicating that the most immature infants can increase their adiponectin levels in the same manner as more mature preterm infants. On the other hand, peak adiponectin concentrations occurred at a greater postnatal age with decreasing GA at birth. This result is in line with our previous findings of a similar relationship between GA at birth and lowest weight, length, and HCSDS in which the time point for lowest SDS occurred latest in the most immature infants (16). We observed a less pronounced increase in adiponectin Figure 4. Individual mean values of adiponectin (␮g/mL) during concentrations from PMA 34 weeks up to term age, corpostnatal days 3–21 in relation to SD score for weight at a responding to the period when catch-up growth usually postmenstrual age of 35 weeks. Filled circles denote infants with birth has been established after the initial period of growth reweight SGA, and open circles denote infants with birth weight tardation. Extensive accumulation of fat occurs during the appropriate for gestational age (AGA).


6


Table 2. Best model of variables contributing to weight SDS at 35 weeks postmenstrual age (forward linear regression analysis). Independent variables included in the model are mean adiponectin (APN) concentrations during postnatal days 3–21, gestational age at birth (days), weightSDS at birth, total steroid exposure from birth until a postmenstrual age of 35 weeks, and mean protein (g/kg/d) or energy (kcal/kg/d) intake from birth until a postmenstrual age of 35 weeks. Variable Adiponectin days 3– 21 SDS weight at birth Total steroid exposure

␤

95% CI

P

0.021

0.001– 0.041

0.04

0.554 ⫺0.008

0.38 – 0.73 ⫺0.014 – ⫺0.002

⬍0.001 0.009

R2 ⫽ 0.70; adj R2 ⫽ 0.68.

third trimester (23), and leptin, another adipokine secreted in proportion to adipose tissue quantity, increases substantially after PMA 34 weeks (24). In a recent study in preterm infants evaluated at term age, circulatory adiponectin concentrations correlated with subcutaneous fat but not with visceral fat amounts (25). Thus, the observed second increase in adiponectin concentrations during the late preterm period may plausibly be related to physiological mechanisms other than the early increase in adiponectin concentrations immediately after birth. We identified a strong positive correlation between postnatal adiponectin concentrations and weightSDS, lengthSDS, and HCSDS at 35 weeks, independent of GA, nutritional intake, and administered steroids during the corresponding time period. These findings are supported by other results for preterm infants, where weight gain rate was associated with adiponectin concentrations (14, 15). However, previous as well as our findings does not support any evidence for adiponectin as a mechanistic link promoting longitudinal growth, rather that adiponectin could be a marker of nutritional and metabolic status. The correlation between early postnatal adiponectin concentrations and growth parameters at PMA 35 weeks displayed a strong interaction with weightSDS at birth. This observation is expected because of the strong covariation between weightSDS at birth and weightSDS observed with advancing PMA. Mean adiponectin concentration during days 3–21 also displayed a strong positive correlation with BWSDS, but we found no univariate correlation between cord blood adiponectin concentrations and BWSDS. However, when GA at birth was added as a contributing factor in a multivariate model, a strong interaction between BWSDS and cord blood adiponectin emerged (data not shown), suggesting that the correlation between BWSDS and cord blood adiponectin in our study was GA dependent. In a previous


study of a larger sample, a strong independent relationship between BWSDS and cord blood adiponectin was observed in preterm but not in term infants (13). Lower cord blood adiponectin concentrations have been described in term infants with BW SGA as compared to infants with BW AGA (26). There is no evidence that adiponectin is produced in the placenta or that maternal adiponectin crosses the placental barrier; thus, adiponectin in cord blood is considered to be exclusively of fetal origin (27). Infants in this study who were born SGA also displayed consistently lower mean adiponectin concentrations from postnatal day 3 until term age, in line with other studies (14, 19). The lower adiponectin concentrations in infants born SGA may result from lower mature fat tissue mass, the primary source for adiponectin production. Less subcutaneous fat is identified in growth-restricted fetuses and in newborn infants with BW SGA (28). In an experimental mouse model, fetal adiponectin influences lipogenic genes and enhances fetal fat deposition (29). Thus, in contrast to adults where adiposity is associated with decreased adiponectin concentrations (30), adiponectin appears to play a different role in the fetus and newborn infant in terms of growth and fat deposition. Regarding metabolic syndrome, adiponectin is considered a key hormone. In children born SGA, lower adiponectin concentrations as well as insulin resistance have been described, and intrauterine growth restriction as well as postnatal growth catch-up seem to be important (31, 32). An abnormal growth pattern may thus contribute to later metabolic dysfunction. Adiponectin is essential for lipid and glucose homeostasis. It enhances hepatic insulin sensitivity by reducing hepatic gluconeogenesis and also promotes survival of beta cells (11). Hypoadiponectinemia in preterm infants has been associated with early hyperglycemia, which has been considered a risk factor both in early and later neonatal morbidity (33). In term infants, an inverse correlation between insulin and adiponectin cord blood concentrations has been described (21). In our study, energy intake during the first 3 postnatal weeks correlated univariately with adiponectin concentrations during the corresponding time period, but not with protein intake. However, the correlation between adiponectin concentration and energy intake disappeared after adjustment for GA at birth In one study of preterm infants, energy intake was not related to adiponectin concentrations at term age (34). Adiponectin is present in human breast milk, and in breast milk levels of adiponectin are related to energy and fat composition (35). Thus, our findings do not preclude that nutritional intake, with enteral feeding initiated soon after birth, influenced circulatory adiponectin concentrations. The achieved nutritional intake in our study was much


doi: 10.1210/jc.2015-1081

lower than present recommendations and an optimized nutritional regime could have a different effect on adiponectin levels. Preterm infants receiving supplementation with long-chain polyunsaturated fatty acids in formula have higher adiponectin concentrations than infants receiving standard nutrition (36). Thus, a change of lipid composition in enteral or parenteral nutrition may be one possible intervention to modify circulatory adiponectin levels. Accumulated steroid intake during the first 3 weeks after birth displayed a univariate association with lower adiponectin concentrations during the corresponding time period. Adiponectin and steroid hormones participate together with insulin in the regulation of energy metabolism, but the interaction between adiponectin and steroid administration has not been fully elucidated (37). In children born SGA, increased endogenous glucocorticoid bioactivity is associated with decreased adiponectin concentrations (38). In conclusion, this study showed a profound postnatal increase of the adipocytokine adiponectin during the first weeks of life after very preterm birth. Furthermore, higher adiponectin concentrations were associated with improved growth at PMA 35 weeks. Because early growth has been linked to later metabolic derangement and adiponectin is a metabolically active hormone, we stress the importance of short- and long-term studies of early changes in adiponectin after very preterm birth in relation to defined metabolic events.

Acknowledgments We thank Aimon Niklasson for helpful advice. This study has been supported by the Swedish Medical Research Council (# 2011–2432), the European Commission FP7 project 305 485 PREVENT-ROP, Swedish government grants (#ALFGB2770), VINNOVA (2009 – 01 152), the Skåne Council Foundation for Research and Development, the Linnéa and Josef Carlsson Foundation, the Helsingborg and the Skane University Hospital foundation and donations. Address all correspondence and requests for reprints to: Chatarina Löfqvist, Sahlgrenska Academy at Univeristy of Gothenburg, The Queen Silvia Children´s Hospital, 416 85 Göteborg, SWEDEN, E-mail: www.rop.gu.se. This work was supported by . * Shared last authorship. Disclosure Summary: The authors have nothing to disclose.

References 1. Marks KA, Reichman B, Lusky A, Zmora E, Israel Neonatal N. Fetal growth and postnatal growth failure in very-low-birthweight infants. Acta Paediatr. 2006;95:236 –242.


7

2. Kaempf JW, Kaempf AJ, Wu Y, Stawarz M, Niemeyer J, Grunkemeier G. Hyperglycemia, insulin and slower growth velocity may increase the risk of retinopathy of prematurity. J Perinatol. 2011; 31:251–257. 3. Belfort MB, Rifas-Shiman SL, Sullivan T, Collins CT, McPhee AJ, Ryan P, Kleinman KP, Gillman MW, Gibson RA, Makrides M. Infant growth before and after term: effects on neurodevelopment in preterm infants. Pediatrics. 2011;128:e899 –906. 4. Okada T, Takahashi S, Nagano N, Yoshikawa K, Usukura Y, Hosono S. Early postnatal alteration of body composition in preterm and small-for-gestational-age infants: implications of catch-up fat. Pediatr Res. 2014; 5. Westerberg AC, Henriksen C, Ellingvag A, Veierod MB, Juliusson PB, Nakstad B, Aurvag AK, Ronnestad A, Gronn M, Iversen PO, Drevon CA. First year growth among very low birth weight infants. Acta Paediatr. 2010;99:556 –562. 6. Johnson MJ, Wootton SA, Leaf AA, Jackson AA. Preterm birth and body composition at term equivalent age: a systematic review and meta-analysis. Pediatrics. 2012;130:e640 – 649. 7. Uthaya S, Thomas EL, Hamilton G, Dore CJ, Bell J, Modi N. Altered adiposity after extremely preterm birth. Pediatr Res. 2005;57:211– 215. 8. Rosen ED, Spiegelman BM. Adipocytes as regulators of energy balance and glucose homeostasis. Nature. 2006;444:847– 853. 9. Korner A, Wabitsch M, Seidel B, Fischer-Posovszky P, Berthold A, Stumvoll M, Bluher M, Kratzsch J, Kiess W. Adiponectin expression in humans is dependent on differentiation of adipocytes and downregulated by humoral serum components of high molecular weight. Biochem Biophys Res Commun. 2005;337:540 –550. 10. Yamauchi T, Kadowaki T. Physiological and pathophysiological roles of adiponectin and adiponectin receptors in the integrated regulation of metabolic and cardiovascular diseases. Int J Obes (Lond). 2008;32 Suppl 7:S13–18. 11. Tao C, Sifuentes A, Holland WL. Regulation of glucose and lipid homeostasis by adiponectin: effects on hepatocytes, pancreatic beta cells and adipocytes. Best Pract Res Clin Endocrinol Metab. 2014; 28:43–58. 12. Corbetta S, Bulfamante G, Cortelazzi D, Barresi V, Cetin I, Mantovani G, Bondioni S, Beck-Peccoz P, Spada A. Adiponectin expression in human fetal tissues during mid- and late gestation. J Clin Endocrinol Metab. 2005;90:2397–2402. 13. Kajantie E, Hytinantti T, Hovi P, Andersson S. Cord plasma adiponectin: a 20-fold rise between 24 weeks gestation and term. J Clin Endocrinol Metab. 2004;89:4031– 4036. 14. Saito M, Nishimura K, Nozue H, Miyazono Y, Kamoda T. Changes in Serum Adiponectin Levels from Birth to Term-Equivalent Age Are Associated with Postnatal Weight Gain in Preterm Infants. Neonatology. 2011;100:93–98. 15. Siahanidou T, Mandyla H, Papassotiriou GP, Papassotiriou I, Chrousos G. Circulating levels of adiponectin in preterm infants. Arch Dis Child Fetal Neonatal Ed. 2007;92:F286 –290. 16. Hansen-Pupp I, Lofqvist C, Polberger S, Niklasson A, Fellman V, Hellstrom A, Ley D. Influence of insulin-like growth factor I and nutrition during phases of postnatal growth in very preterm infants. Pediatr Res. 2011;69:448 – 453. 17. Niklasson A, Albertsson-Wikland K. Continuous growth reference from 24th week of gestation to 24 months by gender. BMC pediatrics. 2008;8:8. 18. Marsal K, Persson PH, Larsen T, Lilja H, Selbing A, Sultan B. Intrauterine growth curves based on ultrasonically estimated foetal weights. Acta Paediatr. 1996;85:843– 848. 19. Klamer A, Skogstrand K, Hougaard DM, Norgaard-Petersen B, Juul A, Greisen G. Adiponectin levels measured in dried blood spot samples from neonates born small and appropriate for gestational age. Eur J Endocrinol. 2007;157:189 –194. 20. Sivan E, Mazaki-Tovi S, Pariente C, Efraty Y, Schiff E, Hemi R, Kanety H. Adiponectin in human cord blood: relation to fetal birth weight and gender. J Clin Endocrinol Metab. 2003;88:5656 –5660.


8


21. Mami C, Marseglia L, Manganaro R, Saitta G, Martino F, Gargano R, Gemelli M. Serum levels of resistin and its correlation with adiponectin and insulin in healthy full term neonates. Early Hum Dev. 2009;85:37– 40. 22. Embleton NE, Pang N, Cooke RJ. Postnatal malnutrition and growth retardation: an inevitable consequence of current recommendations in preterm infants? Pediatrics. 2001;107:270 –273. 23. Ziegler EE, O’Donnell AM, Nelson SE, Fomon SJ. Body composition of the reference fetus. Growth. 1976;40:329 –341. 24. Jaquet D, Leger J, Levy-Marchal C, Oury JF, Czernichow P. Ontogeny of leptin in human fetuses and newborns: effect of intrauterine growth retardation on serum leptin concentrations. J Clin Endocrinol Metab. 1998;83:1243–1246. 25. Nakano Y, Itabashi K, Sakurai M, Aizawa M, Dobashi K, Mizuno K. Accumulation of subcutaneous fat, but not visceral fat, is a predictor of adiponectin levels in preterm infants at term-equivalent age. Early Hum Dev. 2014;90:213–217. 26. Mazaki-Tovi S, Kanety H, Pariente C, Hemi R, Kuint J, Yinon Y, Schiff E, Sivan E. Cord blood adiponectin and infant growth at one year. J Pediatr Endocrinol Metab. 2011;24:411– 418. 27. Aye IL, Powell TL, Jansson T. Review: Adiponectin–the missing link between maternal adiposity, placental transport and fetal growth? Placenta. 2013;34 Suppl:S40 – 45. 28. Rodriguez G, Collado MP, Samper MP, Biosca M, Bueno O, Valle S, Ventura P, Garagorri JM. Subcutaneous fat distribution in small for gestational age newborns. J Perinat Med. 2011;39:355–357. 29. Qiao L, Yoo HS, Madon A, Kinney B, Hay WW, Jr., Shao J. Adiponectin enhances mouse fetal fat deposition. Diabetes. 2012;61: 3199 –3207. 30. Kishida K, Funahashi T, Shimomura I. Adiponectin as a routine


31.

32.

33.

34.

35.

36.

37. 38.

clinical biomarker. Best Pract Res Clin Endocrinol Metab. 2014; 28:119 –130. Cianfarani S, Martinez C, Maiorana A, Scire G, Spadoni GL, Boemi S. Adiponectin levels are reduced in children born small for gestational age and are inversely related to postnatal catch-up growth. J Clin Endocrinol Metab. 2004;89:1346 –1351. Huang Y, Li Y, Chen Q, Chen H, Ma H, Su Z, Du M. Low serum adiponectin levels are associated with reduced insulin sensitivity and lipid disturbances in short children born small for gestational age. Clin Endocrinol (Oxf). 2014; Oberthuer A, Donmez F, Oberhauser F, Hahn M, Hoppenz M, Hoehn T, Roth B, Laudes M. Hypoadiponectinemia in extremely low gestational age newborns with severe hyperglycemia–a matched-paired analysis. PLoS One. 2012;7:e38481. Yoshida T, Nagasaki H, Asato Y, Ohta T. Early weight changes after birth and serum high-molecular-weight adiponectin level in preterm infants. Pediatr Int. 2011;53:926 –929. Ley SH, Hanley AJ, Stone D, O’Connor DL. Effects of pasteurization on adiponectin and insulin concentrations in donor human milk. Pediatr Res. 2011;70:278 –281. Siahanidou T, Margeli A, Lazaropoulou C, Karavitakis E, Papassotiriou I, Mandyla H. Circulating adiponectin in preterm infants fed long-chain polyunsaturated fatty acids (LCPUFA)-supplemented formula–a randomized controlled study. Pediatr Res. 2008; 63:428 – 432. Sukumaran S, Dubois DC, Jusko WJ, Almon RR. Glucocorticoid effects on adiponectin expression. Vitam Horm. 2012;90:163–186. Tenhola S, Todorova B, Jaaskelainen J, Janne OA, Raivio T, Voutilainen R. Serum glucocorticoids and adiponectin associate with insulin resistance in children born small for gestational age. Eur J Endocrinol. 2010;162:551–557.


Transient hypothyroxinaemia associated with developmental delay in very preterm infants.

Improvement in long-term breastfeeding for very preterm infants.

Early growth and neurodevelopmental outcome in very preterm infants: impact of gender.

Less invasive surfactant administration is associated with improved pulmonary outcomes in spontaneously breathing preterm infants.

Very early postnatal changes in splanchnic circulation in term infants.

Moderately preterm, late preterm and early term infants: research needs.

Reduced NK cell percentage at birth is associated with late onset infection in very preterm neonates.

Glucagon and insulin cord blood levels in very preterm, late preterm and full-term infants.

Preterm birth is associated with an increased fundamental frequency of spontaneous crying in human infants at term-equivalent age.

Perinatal factors associated with early neonatal deaths in very low birth weight preterm infants in Northeast Brazil.

Very long apnea events in preterm infants.

Population-based study shows improved postnatal growth in preterm very-low-birthweight infants between 1995 and 2010.

Is it correct to correct? Developmental milestones in 555 "normal" preterm infants compared with term infants.

Circulatory effects of fast ventilator rates in preterm infants.

Circulatory effects of fast ventilator rates in preterm infants.

Very preterm infants show earlier emergence of 24-hour sleep-wake rhythms compared to term infants.

Long-term outcomes of moderately preterm, late preterm, and early term infants.

Language ability in preterm children is associated with arcuate fasciculi microstructure at term.

Early Elevation in Interleukin-6 is Associated with Reduced Growth in Extremely Low Birth Weight Infants.

Respiratory disorders in moderately preterm, late preterm, and early term infants.

Funisitis is associated with adverse neonatal outcome in low-risk unselected deliveries at or near term.

Neurologic and metabolic issues in moderately preterm, late preterm, and early term infants.

Impact of Changing Clinical Practices on Early Blood Gas Analyses in Very Preterm Infants and Their Associated Inpatient Outcomes.

Neonatal pain in very preterm infants: long-term effects on brain, neurodevelopment and pain reactivity.