| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Maternal and Child Health Sciences (L.J., A.B., R.H.), Community Health Sciences (F.L.R.W.), and Pathology and Neurosciences (M.W.H.C.), University of Dundee, Dundee DD1 9SY, Scotland, United Kingdom
Address all correspondence and requests for reprints to: Professor Robert Hume, Maternal and Child Health Sciences, University of Dundee, Ninewells Hospital and Medical School, Dundee DD1 9SY, Scotland, United Kingdom. E-mail: r.hume{at}dundee.ac.uk.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
The fetus in utero is supplied with a constant transplacental source of maternal glucose. At birth this supply is abruptly terminated, and the infant must adapt to a state of independent and appropriate blood glucose homeostasis. In the immediate newborn period, the maintenance of blood glucose levels is dependent on mobilization of liver glycogen stores, the establishment of adequate nutritional intakes, and the maturation of key liver enzyme systems controlling gluconeogenesis and glycogenolysis (1). Hepatic glucose-6-phosphatase (G-6-Pase) is the common terminal enzymatic step for the pathways of glycogenolysis and gluconeogenesis (2), and genetic deficiencies of hepatic G-6-Pase (type I glycogen storage diseases) are severe metabolic disorders characterized by fasting hypoglycemia (3). Hepatic G-6-Pase activity is expressed at only low levels in the human fetus in utero (4). Most term infants increase hepatic G-6-Pase activities to adult levels within a few days of birth (4). In many preterm infants this increment in enzyme activity fails to occur (5, 6). This delayed postnatal appearance of hepatic G-6-Pase and glucose production (7) in preterm infants makes them potentially susceptible to repeated hypoglycemic episodes. This may result in cerebral damage and puts them at risk of sudden and unexpected death (8, 9, 10, 11, 12).
We have shown previously that 18% of a consecutive series of preterm infants at the time of discharge home were unable to maintain normal concentrations of blood glucose after a single routine feed was omitted (13). Cortisol, corticotropin, and blood lactate levels were higher in infants with severe and persistent hypoglycemia but insulin, glucagon, and human GH (hGH) did not differ from infants who remained normoglycemic (13). These results appear to indicate that the counterregulatory hormonal response to hypoglycemia was limited to activation of the hypothalamic-pituitary-adrenocortical axis, but the contribution of catecholamines to glucose homeostasis in these infants was not explored.
Catecholamines have important roles in the regulation of blood glucose levels in adult humans (e.g. Ref.14) and to adaptations of glucose homeostasis to extrauterine life in animal studies (15, 16). Their role in the regulation of glucose homeostasis in infants has not been extensively defined (17, 18), and indeed plasma catecholamines levels and their metabolites have not previously been studied in infants after the first postnatal week (19, 20, 21, 22, 23).
The aim of our study was to further define the roles of epinephrine and norepinephrine in the counterregulatory responses to spontaneous hypoglycemia occurring in infants when a single routine feed was omitted.
| Subjects and Methods |
|---|
|
|
|---|
Neonatal blood glucose measurements, over the first 5 d, were taken on all infants in the study, and the number of episodes of blood glucose levels less than 2.6 mmol/liter was recorded and defined as neonatal hypoglycemic episodes.
Preterm infants were studied at the time of discharge home and term infants at 14 postnatal days. Extensive epidemiological data were collected on each infant; this included maternal health, both generally and in pregnancy; prescribed medication in pregnancy; disorders of pregnancy; details of delivery and resuscitation; neonatal morbidity; and any medications prescribed (24). Birth weight ratios (25) and study weight ratios (13) were calculated for each infant. These ratios are standardized to the Scottish mean birth weight for each gestational week and for males and females separately. Anthropometric measurements recorded included crown-heel length; crown-rump length; occipital-frontal circumference; midarm and leg circumferences; and waist, chest, and hip circumferences. Ponderal indices, midarm circumference, and occipitofrontal circumference ratios were calculated. Skin-fold thickness was measured using Holtain calipers, using a standardized technique (26) and only one operator (L.J.), at the following sites: biceps, triceps, anterior and posterior thigh, suprailiac crest ,and infrascapular; the z score of the sum of these measurements was used as a proxy for total body fat.
It is standard practice in the United Kingdom to measure blood glucose levels in the early newborn period in all infants admitted to neonatal units and to feed preterm infants initially by the parenteral route or by continuous or frequent bolus nasogastric milk feeds, usually hourly. Thereafter when blood glucose levels are apparently stable, it is not normal practice to routinely measure blood glucose concentrations, and most infants will advance to an increased reliance on enteral nutrition and a progression from hourly to 4-h feeds. Blood glucose levels are measured in term infants with identifiable risk factors for hypoglycemia, and they are discharged when normoglycemic. On the study day, all infants were on regular 3- to 4-h feeds. An in-dwelling venous cannula for blood sampling was sited at around 3 h from the start of the last feed. Blood glucose and lactate concentrations were measured 4 h from the start of the last feed. The next scheduled feed was omitted, and blood glucose and lactate concentrations were measured 2 and 4 h later. The infants were studied immediately before their intended discharge home. This study design was selected because it mimics the natural progression in feeding patterns toward longer intervals and omission of a feed, which would be expected for many infants post discharge home. Hypoglycemia was predefined as a blood glucose concentration less than 2.6 mmol/liter. The study was discontinued if two consecutive blood glucose levels were less than 2.6 mmol/liter. Insulin, glucagon, cortisol, hGH, and a range of catecholamines were measured at the end of the infants fast (prefeed); the state of arousal of the infant at the time of blood sampling was recorded (sleeping, fractious, or crying). Transient hypoglycemia was defined as one measurement recorded less than 2.6 mmol/liter with spontaneous recovery. Severe and persistent hypoglycemia was defined as at least two episodes recorded less than 2.6 mmol/liter.
All blood glucose and lactate samples were analyzed on-site using the 2300 Stat plus glucose analyzer (Yellowsprings Inc., Yellow Springs, OH). Three plasma hormones levels were measured by the Department of Biochemical Medicine, Ninewells Hospital and Medical School: cortisol using ADVIA Centaur assay (Bayer Diagnostic, Leuverkusen, Germany) and hGH using IMMULITE assay (Olympus Diagnostic System, Southall, UK). Insulin was measured using the IMx system (Abbott Laboratories, Diagnostic Division, Tokyo, Japan). N-terminal glucagon levels were determined by RIA at the Metabolic Laboratory, Royal Victoria Hospital, Belfast, Northern Ireland.
Norepinephrine (NE), norepinephrine sulfate (NE sulfate), epinephrine (EPI), epinephrine sulfate (EPI sulfate), dihydroxyphenylglycol (DHPG), and DHPG sulfate were measured after extraction from plasma using an alumina extraction procedure and then separated and quantified by liquid chromatography with electrochemical detection (27). The interassay coefficients of variation within and above the normal range were for EPI 10.418.5% and 3.66.5% and for NE 3.04.3% and 1.84.6%, respectively. Interassay coefficient of variation for DHPG was 8.6%. Intraassay coefficients of variation were EPI 13.4%, NE 6.8%, and DHPG 4.8%.
Plasma amino acids, free fatty acids, 3-hydroxybutyrate, urinary amino acids, and organic acids were also measured (by the Department of Biochemical Medicine, Ninewells Hospital and Medical School using routine procedures) in hypoglycemic infants so that recognized metabolic disorders could be excluded.
Statistical analysis
Means and SDs were calculated for all the variables; one extreme measurement for glucagon of 2000 ng/liter in one infant was excluded from statistical analysis. Differences between and within groups were found by ANOVA and by using the t test for unequal variance for determining the specific differences. The association among: EPI, NE, and metabolites; cortisol; glucagon; and hGH was quantified in a correlation matrix (Spearmans rho).
Logistic regression analysis was used to determine the influence of various factors (explanatory variables) on glycemic status (Table 1
). The explanatory variables were derived from the significant results of the bivariate analysis. The Wald
2 and the odds ratios (with 95% confidence intervals) were used to assess the importance of each explanatory variable, adjusted for the other factors in the model.
|
| Results |
|---|
|
|
|---|
NE levels did not differ among the groups. DHPG levels were marginally raised in the severe and persistent hypoglycemic group, compared with the normoglycemic group (Table 2
). EPI levels were significantly higher in the severe and persistent hypoglycemic group, compared with the normoglycemic group (Table 2
).
|
|
|
| Discussion |
|---|
|
|
|---|
The blood glucose level of less than 2.6 mmol/liter taken as the definition of hypoglycemia was based on the neuroglycemic responses to induced hypoglycemia (29) and on a study of preterm infants in which values of blood glucose less than 2.6 mmol/liter were associated with abnormal neuromotor and intellectual performance (8). The definition of what constitutes hypoglycemic blood glucose levels in infants remains contentious (e.g. Refs.30, 31), but evidence to support a level higher than 2.6 mmol/liter is lacking.
In adults a blood glucose level of 3.2 mmol/liter is the threshold for the cortisol response (32). A similar threshold is suggested in infants and children older than 6 months who are spontaneously hypoglycemic, but in infants younger than 3 months, there is no difference in cortisol levels between those who are hypoglycemic (<2.5 mmol/liter) and those who are normoglycemic (33). It is possible that with recurrent spontaneous hypoglycemia in these young infants (<3 months) cortisol reserves are depleted (33); in contrast, our hypoglycemic infants have elevated cortisol levels in response to a comparatively short period of hypoglycemia (maximum 2 h). Cortisol depletion cannot be the complete explanation because preterm infants can maintain very high cortisol levels over a prolonged period in response to the stress of illness (34).
hGH levels in our infants are similar to previously fasted normoglycemic and hypoglycemic infants before discharge from the hospital (13) and also to infants younger than 3 months with spontaneous hypoglycemia whether or not hypoglycemic at sampling (33). Although hGH responses to spontaneous hypoglycemia in older infants (>6 months) and children are reduced in magnitude, indirect evidence suggests that the glycemic threshold of 3.7 mmol/liter is similar to that of adults (32, 33).
In adults EPI constitutes more than 80% of adrenal catecholamine secretion (35) and increases more than NE levels in response to stressful conditions such as hypoglycemia, reflecting relatively greater adrenomedullary hormonal than sympathetic neuronal activation (36). Our infants with severe and persistent hypoglycemia show a similar pattern; EPI levels are significantly higher, but NE levels are the same. In a recent large and well-controlled study of nonstressed children (517 yr), basal levels of EPI were higher (particularly in boys) than those of adults, but NE levels were similar (37). A similar study is required with comparable data from children who are stressed by illness or hypoglycemia.
In normal adults EPI transiently increases hepatic glycogenolysis followed by a sustained increase in hepatic and renal gluconeogenesis (38, 39). Lactate is the predominant gluconeogenic substrate of EPI-stimulated gluconeogenesis in both liver and kidney (40), with glycerol, alanine, and glutamine of lesser importance (40, 41). In our study there was a significant trend toward higher lactate levels in infants with severe and persistent hypoglycemia, as in our previous study (13). This suggests increased EPI levels are affecting a compensatory metabolic response but with insufficient glucose production to correct the hypoglycemia, but other gluconeogenic substrates or hepatic glycogen levels were not determined to assess sufficiency for glucose synthesis.
Glucagon and EPI are important in limiting the glucose nadir and promoting rapid recovery from transient hypoglycemia of rapid onset in adults (42) and also in countering slow-onset but progressive hypoglycemia (43). This study confirmed our previous observations (13) that increases in plasma glucagon are not a feature of our infants with severe and persistent hypoglycemia. Glucagon release by pancreatic
-cells is stimulated directly by low blood glucose levels (44) or indirectly by autonomic inputs from sympathetic (NE) and parasympathetic (acetylcholine) innervations as well as adrenal medulla (EPI) (45). Glucagon release is inhibited by local insulin production (45). Regulation of glucagon release is complex, but the nature of the failed response to low blood glucose levels in our infants is not obvious. There is also a very limited glycemic response to exogenous glucagon stimulation in such infants (46), which is also present in the early neonatal period in preterm infants (van Kempen, A., personal communication), but not later in infancy (47). These data suggest that, during the developmental phase of our study, EPI rather than glucagon is the major acute regulator of glycogenolysis and gluconeogenesis.
Cortisol and EPI are interrelated in a number of ways. High intraadrenal steroid levels are required for production of catecholamines in the human adrenal medulla (e.g. Ref.48), and plasma EPI levels in many situations vary more closely with those of corticotropin than those of NE (49). Cortisol and EPI levels were both elevated in our infants with severe and persistent hypoglycemia, and cortisol levels were significantly correlated to not only EPI levels but also NE.
In diabetes mellitus antecedent hypoglycemia causes defective glucose counterregulation, in particular a reduction in the EPI response and an absent glucagon response, (e.g. Ref.50); these changes may be mediated by elevations in cortisol levels (51, 52, 53). In healthy adults, antecedent hypoglycemia reduces EPI, NE, and glucagon responses to subsequent hypoglycemia, with GH and cortisol responses reduced or unaltered (51, 54, 55). It is standard practice to monitor blood glucose levels in the early neonatal period in preterm and other vulnerable infants and, once these levels are apparently stable, to discontinue further monitoring and progress feeding to 4-h intervals. However, the number of hypoglycemic episodes during the first five postnatal days reflects the glycemic status at the time of study; suggesting an underlying developmental disorder in glucose homeostasis. It is possible, particularly in our group with severe and persistent hypoglycemia, that some infants had episodes of low blood glucose in the days before the test. If infants respond to antecedent hypoglycemia in the same manner as adults, this may have modified the hormonal responses to the controlled test fast.
Insulin inhibits gluconeogenesis and acts predominantly by suppressing the expression of the genes for the key gluconeogenic enzymes phosphoenolpyruvate carboxykinase and G-6-Pase. The expression of the genes for G-6-Pase and phosphoenolpyruvate carboxykinase is induced by glucagon, glucocorticoids, and catecholamines (see review in Ref.56). The molecular bases of delayed postnatal development of hepatic G-6-Pase in infants are not known, but dysfunctional postnatal endocrine regulation is likely to be a contributory factor. Our preterm infants were studied at between 1 and 13 wk of age. Severe and persistent hypoglycemia was associated with elevated plasma EPI concentrations. This is contrary to previous studies in the early neonatal period (17, 18). These differences suggest that ontogenic studies of catecholamine responsiveness to glycemic status are required in preterm infants and that these studies are related to gestational and postnatal age.
Infants with severe and persistent low blood glucose values are not sufficiently distinctive from normoglycemic infants to identify them accurately on clinical grounds. This is our second series of preterm infants we have described in which asymptomatic hypoglycemia is common at the time of discharge home; we now show the same phenomenon in growth-restricted term infants. Until this work is repeated by other researchers in other populations and it is determined whether this observed hypoglycemia is detrimental to neurodevelopmental outcome, we recommend that the feeding regimen in preterm and growth-restricted term infants be frequent and regular with a low threshold for readmission at times of metabolic stress.
| Acknowledgments |
|---|
| Footnotes |
|---|
Abbreviations: DHPG, Dihydroxyphenylglycol; EPI, epinephrine; G-6-Pase, glucose-6-phosphatase; hGH, human GH; NE, norepinephrine.
Received March 22, 2004.
Accepted September 21, 2004.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. J. Rozance, S. W. Limesand, J. S. Barry, L. D. Brown, S. R. Thorn, D. LoTurco, T. R. H. Regnault, J. E. Friedman, and W. W. Hay Jr. Chronic late-gestation hypoglycemia upregulates hepatic PEPCK associated with increased PGC1{alpha} mRNA and phosphorylated CREB in fetal sheep Am J Physiol Endocrinol Metab, February 1, 2008; 294(2): E365 - E370. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Hussain, J. Bryan, H. T. Christesen, K. Brusgaard, and L. Aguilar-Bryan Serum Glucagon Counterregulatory Hormonal Response to Hypoglycemia Is Blunted in Congenital Hyperinsulinism Diabetes, October 1, 2005; 54(10): 2946 - 2951. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |