Metabolic control in diabetic pregnancy

Human Placental Lactogen in Relation to Maternal Metabolic Health and Fetal Outcomes: A Systematic Review and Meta-Analysis

Human placental lactogen (hPL) is a placental hormone which appears to have key metabolic functions in pregnancy. Preclinical studies have putatively linked hPL to maternal and fetal outcomes, yet-despite human observational data spanning several decades-evidence on the role and importance of this hormone remains disparate and conflicting. We aimed to explore (via systematic review and meta-analysis) the relationship between hPL levels, maternal pre-existing and gestational metabolic conditions, and fetal growth. MEDLINE via OVID, CINAHL plus, and Embase were searched from inception through 9 May 2022. Eligible studies included women who were pregnant or up to 12 months post-partum, and reported at least one endogenous maternal serum hPL level during pregnancy in relation to pre-specified metabolic outcomes. Two independent reviewers extracted data. Meta-analysis was conducted where possible; for other outcomes narrative synthesis was performed. 35 studies met eligibility criteria. No relationship was noted between hPL and gestational diabetes status. In type 1 diabetes mellitus, hPL levels appeared lower in early pregnancy (possibly reflecting delayed placental development) and higher in late pregnancy (possibly reflecting increased placental mass). Limited data were found in other pre-existing metabolic conditions. Levels of hPL appear to be positively related to placental mass and infant birthweight in pregnancies affected by maternal diabetes. The relationship between hPL, a purported pregnancy metabolic hormone, and maternal metabolism in human pregnancy is complex and remains unclear. This antenatal biomarker may offer value, but future studies in well-defined contemporary populations are required.

Accelerated starvation in late pregnancy: a comparison between obese women with and without gestational diabetes mellitus

We compared the glucose, insulin, free fatty acid, and 3-hydroxybutyrate responses to a briefly extended overnight fast during the third trimester of pregnancy between two groups: obese women with normal glucose tolerance (n = 10) and age- and weight-matched women with gestational diabetes mellitus (n = 10). After a 12-hour overnight fast, plasma glucose (95 +/- 4 vs. 78 +/- 2 mg/dl; p less than 0.01), insulin (32 +/- 5 vs. 17 +/- 2 microU/ml; p less than 0.02), and free fatty acid (860 +/- 63 vs. 639 +/- 79 mmol/L; p less than 0.05) levels were higher in the patients with gestational diabetes mellitus. 3-Hydroxybutyrate levels were similar in the two groups at that time (0.23 +/- 0.04 vs. 0.18 +/- 0.03 mmol/L; p greater than 0.3). When the fast was extended to 18 hours by having the patients skip breakfast, glucose levels fell more rapidly in the group with gestational diabetes mellitus but remained elevated compared with the nondiabetic women. Insulin levels declined at a similar rate in the two groups. Free fatty acid levels did not increase significantly in the group with gestational diabetes mellitus during the extended fast. In contrast, free fatty acid levels increased by 44% in the normal pregnant women, reaching the level observed in the group with gestational diabetes mellitus after 18 hours. 3-Hydroxybutyrate levels remained virtually identical in the two groups throughout the brief fast. Thus, compared with that of normal pregnant women, the response of obese women with gestational diabetes mellitus to brief caloric deprivation during late pregnancy was characterized by a greater fall in plasma glucose values without a greater propensity to ketosis. Our findings may have important implications for the dietary management of obese patients with gestational diabetes mellitus.

Factors influencing neonatal morbidity in gestational diabetic pregnancy

The influence of obstetric factors and indices of maternal blood glucose control on neonatal morbidity was examined in 261 women with gestational diabetes. A reference group of 218 women, matched for age and day of delivery, within 1 week, was used for comparison. Perinatal morbidity was significantly more frequent in the gestational diabetic pregnancies (23%) than in the reference group (13%), whereas the occurrence of large-for-gestational-age infants was not different between the groups. Infants born to women with gestational diabetes were categorized to a no-morbidity group (n = 200) and a morbidity group (n = 61). The group with morbidity had significantly shorter gestational age at delivery, higher frequency of caesarean section, higher maternal pre-pregnancy weight and higher area under the glucose tolerance curve. There was no significant difference in third-trimester blood glucose between the groups. Discriminant analysis revealed that the most significant influence on neonatal morbidity was gestational age at delivery. After correction for this factor the only factor with added significance for neonatal morbidity was maternal pre-pregnancy weight. The present study clearly illustrates that other factors beside blood glucose control are of importance for neonatal outcome in gestational diabetic pregnancy.

Gestational diabetes reverses the circadian variation of plasma insulin response to intravenous glucose

Both healthy third-trimester pregnant women and a group of women with gestational diabetes failed to show a difference in glucose clearance rates when given an intravenous glucose bolus at 8 AM compared with 4 PM. The plasma insulin response in the healthy pregnant women was greater at 8 AM. In the diabetic group, the peak insulin response was greater at 4 PM, but it was more prolonged after the 8 AM tests. These alterations in plasma insulin response were especially striking in the subgroup of obese women with gestational diabetes, who demonstrated metabolic differences compared with their nonobese counterparts.

Diagnosis and treatment of gestational diabetes according to amniotic fluid insulin levels

In spite of dietary treatment, the infants of pregnant patients with abnormal glucose tolerance have hyperinsulinism and diabetogenic fetopathy in 10 to 36% of cases. Those patients, who require insulin to prevent from fetopathy cannot be reliably selected by maternal parameters such as blood glucose and glycosylated hemoglobin values. We recommend the measurement of amniotic fluid insulin between the 28 and 32 weeks of pregnancy to differentiate whether the fetus is compromised or not. Subjects with values above the 97th centile require insulin therapy. Inadequate insulin dosage or delayed fetal hyperinsulinism can be discovered by checking the amniotic fluid insulin level at 33 to 36 weeks. In a total of 88 gestational diabetic patients 19 had raised amniotic fluid insulin levels indicating the onset of diabetic fetopathy at an early stage. Diabetic patients with raised amniotic fluid insulin levels needed large doses of insulin, namely 64.6 +/- 29.5 (Mean +/- SD) U/24 h. This treatment reduced mean blood glucose levels from 98 +/- 9 (Mean +/- SD) mg/dl to 82 +/- 10 mg/dl and was sufficient to prevent from diabetic fetopathy.

Gastroenteropancreatic hormones in normal and gestational-diabetic pregnancy: response to oral lipid

The effect of normal and gestational-diabetic pregnancy on the gastroenteropancreatic (GEP) hormone response to lipid ingestion was studied in 17 women, 8 normal and 9 with gestational diabetes, by determination of the plasma concentrations of gastric inhibitory polypeptide (GIP), gut glucagon-like immunoreactivity (gut GLI), insulin, glucagon, and pancreatic polypeptide (PP) following the ingestion of 67 g of triglyceride in late pregnancy and postpartum. Also, the plasma concentrations of free fatty acids (FFA), triglyceride, and glucose were determined. In both groups fasting plasma triglyceride and insulin were increased and PP was decreased. Fasting plasma glucose and gut GLI were decreased in normal pregnancy, but were unaltered in gestational-diabetic pregnancy. Fasting plasma glucagon were unaltered in normal but increased in gestational diabetic pregnancy. In both groups of women the GIP response to triglycerides was impaired in pregnancy. Postpartum, the GIP response was greater in the gestational diabetics than in normal women whereas no difference was found in pregnancy. The other hormones and metabolites responded similarly in pregnancy and postpartum and no difference between normal women and women with gestational diabetes was found. It is concluded that the GIP response to triglycerides is impaired in pregnancy.

Effect of maternal ketoacidemia on the pregnant ewe and the fetus

Poorly controlled diabetic pregnancies are associated with hyperglycemia and elevated ketones. While glucose is known to cross the placenta, there is limited information regarding the placental transfer of ketones and their associated cardiovascular and metabolic effects in the fetus. Thus, the present study was undertaken to evaluate the transfer of the ketoacid beta hydroxybutyrate across the ovine placenta and to determine the effects of this ketoacid on maternal and fetal physiologic and metabolic parameters. Pregnant ewes (110 to 120 days' gestation) were instrumented with catheters in the lateral branch of both uterine arteries, uterine veins, femoral artery, and femoral vein, and electromagnetic flow probes were placed on both middle uterine arteries. Catheters were placed in the fetal carotid artery and jugular vein, and a catheter and balloon were placed in the amniotic fluid. Beta hydroxybutyrate (0.39 mmole/100 ml of uterine blood flow) and antipyrine (00.27 mmole/100 ml of uterine blood flow) as a second reference marker, were infused simultaneously into the uterine arteries for a period of 2 hours. The beta hydroxybutyrate concentrations in the uterine vein increased to 5.93 +/- 1.32 mmoles/L, and were associated with a significant increase in maternal heart rate and a slight but significant reduction in uterine blood flow. No changes in maternal arterial blood gas values were noted. The concentration of beta hydroxybutyrate in the fetal carotid arteries increased from 0.01 +/- 0.01 mmole/l to 0.15 +/- 0.03 mmole/L, and were associated with a significant reduction in fetal PaO2 (24.2 +/- 0.9 to 17.9 +/- 1.9 mm Hg) and an elevation of fetal lactate levels (1.86 +/- 0.17 to 5.07 +/- 1.56 mmoles/L).

Uteroplacental blood flow in diabetic pregnancy: measurements with indium 113m and a computer-linked gamma camera

The uteroplacental blood flow index in the last trimester of pregnancy in 26 women with diabetes mellitus was compared to that in 41 healthy control subjects. After an intravenous injection of 1 mCi of indium 113m, the radiation over the placenta was recorded with a computer-linked gamma camera. From time-activity analysis of the isotope accumulation curve, a uteroplacental blood flow index could be calculated. In the diabetic pregnant women, the maternal-placental blood flow index was reduced 35% to 45% compared to that in healthy women. The blood flow index tended to be further impaired in those diabetic women who had higher blood glucose values.

Insulin-induced hypoglycaemia in diabetic women during late pregnancy and one year post partum

Insulin (0.1 IW/kg) later followed by glucose was injected intravenously in nine diabetic women in the supine position both during pregnancy and one year post partum. C-peptide was present in five subjects, indicating some residual beta-cell function. Their mean basal C-peptide level, before insulin, was twice as high in the pregnant as inthe non-pregnant state. C-peptide decreased progressively after insulin. The mean basal plasma glucose level was lower during pregnancy (4.8 mmol/l) than after it (9.6 mmol/l), but decreased to the same level (2.2 mmol/l) after insulin. The rate of fall in glucose was thus lower during pregnancy (kt = 2.54) than after (kt = 4.08), but was unrelated to the basal glucose levels. Basal levels of free fatty acids (FFA), 3-hydroxybutyrate (3-HB), cyclic AMP, and lactate were similar, while glycerol was lower during pregnancy. Insulin-induced changes in FFA, glycerol, 3-HB, cyclic AMP, and lactate were similar during and after pregnancy. Plasma amino acid concentrations were generally lower in pregnancy, significantly so only for arginine and glycine. Amino acid levels were unaffected by insulin in pregnancy, whereas leucine, isoleucine and tyrosine decreased significantly in the non-pregnancy, whereas leucine, isoleucine and tyrosine decreased significantly in the non-pregnancy, whereas leucine, isoleucine and tyrosine decreased significantly in the non-pregnancy state. We conclude that there are differences in metabolic responses to insulin in diabetic women during and after pregnancy, indicating a decreased sensitivity to insulin during pregnancy in some tissues.

Comparison of human and sheep chorion laeve permeability to glucose, beta-hydroxybutyrate, and glycerol

Significant permeability of in vitro human chorion laeve to glucose, glycerol, and beta-hydroxybutyrate (betaOH butyrate) and in vitro sheep chorion leave to glycerol and betaOH butyrate was measured. Diffusion across these tissues of glycerol and betaOH butyrate corresponded to the relative molecular size, charge, and lipid solubility of these compounds. The simple diffusion of glucose across human chorion leave was somewhat reduced by metabolic conversion of this compound during the transfer process. Demonstration of in vitro human placental tissue permeability to ketoacids and glycerol suggests that these compounds cross to the fetus when elevated in the maternal diabetic state.

Care of pregnant diabetics: medical aspects

A method has been worked out for the intensive care of pregnant diabetics with the object of preventing damage both to the mother and the fetus. The method requires close cooperation between a team of doctors including obstetrician, internist and pediatrician. The present paper reports the experience of the first two years of application of the method and refers to a total of 75 women. The main point consists in an attempt to maintain a normal blood glucose level by administering increasing doses of insulin. Periods of inpatient management alternated with close outpatient control. Of the 75 insulin-dependent diabetics 45% had severe diabetes (White classes D-F); nevertheless, perinatal mortality was only 5.78 per cent. It is worth stressing that the 25 diabetics who came under intensive care before conception or in the early stages of pregnancy all gave birth to live healthy babies.

Changes in amniotic fluid glucose, beta-hydroxy-butyrate, glycerol, and lactate concentration in diabetic pregnancy

Amniotic fluid glucose, beta OH butyrate, glycerol, and lactate concentrations were measured in 75 samples collected in the third trimester of pregnancy from 50 diabetic patients, all but four of whom required insulin. Increases in maternal fasting plasma sugar were accompanied by corresponding increases in amniotic fluid glucose and on occasion increases in amniotic fluid beta OH butyrate. These data correspond to previous reports of placental glucose transfer and in addition, provide statistically significant evidence of placental betaOH butyrate transfer since the hyperglycemic, hyperinsulinemic fetus of a diabetic mother would be a poor primary source for ketogenesis. Relatively poor correlation of elevated fluid levels of these solutes to fetal outcome probably reflects a low incidence of maternal hyperglycemia, ketogenesis. Relatively poor correlation of elevated fluid levels of these solutes to fetal outcome probably reflects a low incidence of maternal hyperglycemia, ketoacidosis, and over-all reduced neonatal morbidity-mortality rates in this group of metabolically well-controlled, predominantly insulin-requiring diabetic patients managed in a regional high-risk perinatal center.

Simplified management of pregnancy complicated by diabetes

Simple methods for the management of pregnancy complicated by diabetes are described. Emphasis was placed on keeping patients out of hospital, good control of diabetes and vaginal delivery at 38 weeks: using these methods, there were nine perinatal deaths in a consecutive series of 101 pregnancies complicated by diabetes. Four of the perinatal deaths were due to the respiratory distress syndrome. The patients whose diabetes was diagnosed during pregnancy had significantly heavier babies (18 pregnancies, mean birth weight 3337 g) than the established diabetics (83 pregnancies, mean weight 3011 g) despite significantly lower fasting blood glucose levels in the former and similar mean gestational ages at delivery. Mean fasting blood glucose levels for the whole series during the first, second and third trimesters were 9.0, 6.7 and 5.6 mmol/l respectively. The mean duration of antenatal stay in hospital for complications related to diabetes was 29 days. Diabetic retinopathy did not seem to be adversely affected by pregnancy.

Metabolism during normal and diabetic pregnancy and its effect on neonatal outcome

Diurnal profile studies have been used to define the fetal carbohydrate and lipid substrate environment in normal and diabetic women during late pregnancy. In women with normal glucose tolerance the diurnal plasma glucose concentration was maintained within close limits (mean +/- S.D., 4.70 +/- 0.38 mmol/l) but in chemical and insulin-dependent diabetics there was a marked increase in both the mean diurnal glucose value and in the variability of the plasma glucose levels observed through the day (mean +/- S.D., 5.61 +/- 5.61 +/- 1.03 and 6.02 +/- 1.26 mmol/l respectively, P less than 0.01). No difference was observed between the peripheral insulin activity of the normal and chemical diabetic women, and the impaired glucose tolerance of the latter group was due to a deficient insulin response to goucose. The diurnal glucose variability, expressed as the standard deviation of the mean, was found to be inversely correlated with the residual C-peptide response in insulin-requiring diabetics. The mean diurnal plasma free fatty acid (FFA) concentration was slightly raised in chemical diabetic subjects compared to normal women (mean +/- S.D., 0.77 +/- 0.34 and 0.68 +/- 0.20 mmol/l respectively) but this difference was not significant. Insulin treatment produced a marked reduction in circulating FFA concentration, with a mean value in the insulin-dependent diabetic group of 0.45 +/- 0.11 mmol/l (P less than 0.001). Neonatal glucose assimilation during the first two hours of life correlated strongly with several functions of maternal carbohydrate tolerance. This was associated with higher plasma insulin concentrations at birth, and a marked tendency to hypoglycaemia in the infants of untreated chemical diabetic women. Impaired mobilization of triglyceride stores was also observed during the two hours after birth in the infants of diabetic women. This, however, appears to be due not to impaired lipolysis but to rapid re-esterification of FFA. These findings all indicate a state of functional hyperinsulinism in the infant of the diabetic women secondary to maternal hyperglycaemia.

Evidence for fatty acid transfer across the human placenta

Lipid analysis of blood from umbilical artery and vein, experiments on artificially perfused human placentas, measurements of fetal blood triglyceride concentrations and the relative percentage of essential fatty acids in fetal adipose tissue are all consistent with the view that fatty acids cross the human placenta and that the flow to the fetus is influenced by maternal blood concentrations of free fatty acids and triglycerides.

Maternal plasma glucose and free fatty acid concentrations related to infant birth weight

A new and specific method is used for the measurement of plasma free fatty acids (FFA) in pregnancy. Fasting plasma FFA and glucose concentrations were measured serially in twenty normal and twelve overweight women in pregnancy and related to infant birth weight. Large variation between individuals was noted for FFA and no change was found with advancing gestation. Studies of the day-to-day variation in the same women at 20 and 36 weeks gestation showed wide variations in concentrations. No significant change in glucose concentration was noted with advancing gestation, but postnatal values were significantly higher. Small variability of glucose concentrations was noted in both the serial and day-to-day studies. In normal-weight women a positive correlation between the observed birth weight of their infants and the fasting plasma glucose levels was found at 20 and 40 weeks and also with the mean of 5 levels measured in pregnancy. No similar correlation was found for the group of overweight women.

Gestational and pregestational diabetes: an approach to therapy

The objective of management in the pregnant diabetic patient is to achieve physiologic glucose homeostasis through the use of diet and insulin. As outlined, the numerous ancillary tests developed during the past 15 years to assist the clinician in determining impending fetal death have left much to be desired, especially where metabolic homeostasis has not been achieved prior to the thirty-sixth week of gestation. The statistics from this institution indicate that the maintenance of the plasma glucose concentration below 100 mg. per cent throughout gestation, regardless of the severity of the diabetes, all but removes the risk of maternal-fetal complications due to diabetes. The management is uniform for all patients exhibiting an abnormality of carbohydrate metabolism, and, although it is rather difficult to accept, there have been minimal neonatal complications when the protocol outlined in this presentation has been followed.