Amniotic fluid C-peptide and cortisol in normal and diabetic pregnancies and pregnancies accompanied by fetal growth retardation

The interrelationship between amniotic fluid (AF) concentration and content (AF X conc) of C-peptide and cortisol was studied in four groups of women comprising 10 gestational and 16 type-I diabetics, 11 women with intrauterine growth retarded fetuses (IUGR), and 17 healthy control women. Mean AF volume was significantly greater (P less than 0.05) in the type-I diabetic group than in the control group. Both concentration and content of AF C-peptide was significantly higher in the type-I diabetic group than in the control group (P less than 0.05 and P less than 0.01, respectively). The corresponding values were significantly lower in mothers with IUGR fetuses compared to controls (P less than 0.05). AF cortisol content was significantly higher (P less than 0.05) in the gestational diabetic group compared to the control group; there were no other significant differences between the groups regarding the cortisol concentration or content. Both cortisol and C-peptide contents were significantly interrelated in both control women (r = 0.68, P less than 0.01) and women with gestational (r = 0.68, P less than 0.05) and type-I diabetes (r = 0.63, P less than 0.01). The C-peptide/cortisol ratio was lowest in the IUGR group and highest in the type-I diabetic group. The same ratio was intermediate and almost equal in the control and gestational diabetic group. Both C-peptide and cortisol concentrations were unrelated to AF volume as well as infant birthweight. C-peptide content was significantly correlated to birthweight percentile in type-I diabetic women (r = 0.61, P less than 0.05). No such correlation was found in the three other groups.     

Amino acid concentrations in maternal plasma and amniotic fluid in relation to fetal insulin secretion during the last trimester of pregnancy in gestational and type I diabetic women and women with small-for-gestational-age infants

Free amino acid concentrations were determined in maternal plasma and amniotic fluid (AF) under standardized and unstressed conditions in four groups of women comprising 6 gestational and 13 type I diabetics, 10 women with small-for-gestational-age (SGA) infants, and 18 healthy control women between 36 and 39 weeks of gestation. Plasma values for branched chain amino acids (the sum of leucine, isoleucine and valine) did not differ significantly between the four groups. The corresponding values in AF were significantly higher (P less than 0.05) in the type I diabetic group and significantly lower (P less than 0.05) in the gestational diabetic group as compared to the control group. The mean AF C-peptide concentration was elevated but not significantly so in gestational (0.69 nmol/l) or type I diabetic (0.54 nmol/l) pregnancies and significantly lower (P less than 0.05) in women with SGA infants (0.28 nmol/l) as compared to the control group (0.38 nmol/l). There was a significant correlation between C-peptide in AF and branched chain amino acids in maternal plasma (r = 0.63; P less than 0.05) as well as to maternal blood glucose (r = 0.79; P less than 0.01) in the type I diabetic group, which merely suggests a greater beta cell reactivity to insulin secretagogues in offspring of diabetic mothers. The correlation between AF C-peptide and branched chain amino acids in maternal plasma was significantly inverse in women with SGA infants (r = -0.75; P less than 0.05). Both individual, branched chain, or total amino acid concentration in AF were unrelated to AF C-peptide.     

Amniotic fluid C-peptide and phosphatidyl glycerol in diabetic pregnancy

The concentrations of C-peptide and phosphatidylglycerol in the amniotic fluid were determined in 36 pregnant diabetic women. Twenty-one patients who were being treated with insulin for gestational diabetes as well as 15 patients who were insulin dependent were studied. All patients were subjected to a program of strict metabolic control, and amniocentesis was performed at gestational week 36-37. Phosphatidyl glycerol was present in the amniotic fluid in 15 cases and absent in 21. The mean concentration of C-peptide did not differ whether phosphatidyl glycerol was present or absent. (C-peptide: 0.56 +/- 0.06 and 0.43 +/- 0.05 nmol/l respectively). Although the mean value for amniotic fluid C-peptide in both groups was close to that in diabetic pregnancies with an uneventful neonatal outcome, it was significantly higher than that in non-diabetic pregnancies, indicating minor fetal hyperinsulinemia. The level of C-peptide in the amniotic fluid showed a correlation to the subsequent birthweight of the infant (r = 0.50; p less than 0.01). It is concluded that with rigorous metabolic control of the pregnant diabetic patient, the presence or absence of phosphatidyl glycerol, as an index of fetal lung maturity, is apparently not related to the level of C-peptide in the amniotic fluid.     

Amniotic Fluid Insulin Values in Women with Gestational Diabetes as a Predictor of Emerging Diabetes Mellitus

Summary: Amniotic fluid insulin levels were estimated in 30 women with insulin-dependent diabetes, 216 with gestational diabetes and 27 with normal glucose tolerance. Results were correlated with birth-weight, incidences of fetal macrosomia and neonatal hypoglycaemia, and the risk of the mothers with gestational diabetes developing diabetes mellitus on follow-up.
The women with prepregnaney diabetes had significantly higher amniotic fluid insulin values and showed a significant correlation between raised liquor insulin values (>97th percentile) and hypoglycaemia in the infant (p = 0.039).
In the gestational diabetic pregnancies there were highly significant associations between elevated liquor insulin values and macrosomia (p <0.0045) and birth-weight (p <0.00004), and a weak correlation with neonatal blood glucose levels (p = 0.042).
Women with gestational diabetes who later developed permanent diabetes mellitus had higher mean amniotic fluid insulin levels than those whose glucose tolerance remained normal on follow-up (p ≤0.0072) and more of them had a level greater than the 97th percentile than those whose glucose tolerance remained normal (odds ratio 6.48, 95% confidence interval 1.51–27.8, p = 0.0094). However a high amniotic fluid insulin level was of less clinical value for detection of women destined to develop diabetes (7 of 25, 28%) than was the need for insulin therapy during pregnancy (18 of 39, 46%) .     

Fetal beta cell function in diabetic pregnancy. Amniotic fluid concentrations of proinsulin, insulin, and C-peptide during the last trimester of pregnancy

Proinsulin, insulin C-peptide, insulin-binding antibody, and glucose concentrations were measured in amniotic fluid samples from 43 insulin-treated diabetic patients and 17 nondiabetic control patients between the thirty-six and thirty-ninth weeks of gestation. Insulin-binding antibodies in amniotic fluid were present in only three diabetic patients, although antibodies in maternal serum were found in 22 of the diabetic subjects. In the diabetic group, maternal serum insulin-binding antibodies were statistically unrelated to levels of C-peptide in amniotic fluid. The mean amniotic fluid concentrations of proinsulin (0.07 nmole/L), insulin (0.08 nmole/L), C-peptide (1.17 nmoles/L), and glucose (2.09 mmoles/L) were markedly elevated (p less than 0.001) in diabetic patients, as compared to nondiabetic control patients, thus suggesting exaggerated fetal beta cell function. C-peptide was correlated to both insulin (r = 0.69) and proinsulin (r = 0.35) in the diabetic group only. Infant birth weight and amniotic fluid C-peptide was significantly correlated in both the control group (r = 0.54) and the diabetic group (r = 0.38). Diabetic pregnancies associated with neonatal morbidity (n = 25) had significantly higher mean amniotic fluid concentrations of both insulin and C-peptide than did pregnancies without neonatal morbidity (n = 18). The amniotic fluid values of C-peptide and insulin in these latter two subgroups were overlapping and, therefore, could not serve to predict neonatal outcome in the individual case.     

What affects fetal pulmonary maturation during diabetic pregnancy?

Previous studies differ as to whether the fetus of the woman with diabetes mellitus has altered formation of lung surfactant. The factors responsible for these differences remain unclear. In this study, measures of blood glucose control, including birth weight percentile and the presence of factors potentially influencing fetal pulmonary maturation, such as diabetic class, maternal chronic hypertension, and preeclampsia, were compared with the amniotic fluid lecithin/sphingomyelin ratio, phosphatidylglycerol, and phosphatidylinositol within a group of 90 diabetic pregnancies. The factors were evaluated in combination with the techniques of canonical correlation and multiple regression analysis. Gestational age had the strongest effect in determining levels of amniotic fluid phospholipids, and hypertension was found to significantly accelerate the appearance of phosphatidylgycerol (p less than 0.05). The effect of hypertension was one third as important as that of gestational age. Neither diabetic blood glucose control, White classification, nor the remaining explanatory variables were found to play a significant role in determining the amniotic fluid phospholipid levels. This study suggests that, in the clinical management of diabetes, gestational duration remains the single most important determinant of amniotic fluid phospholipid levels.     

Decreased amniotic fluid magnesium concentration in diabetic pregnancy

Amniotic fluid samples from 21 diabetic pregnant women were pair-matched for gestational age with 21 samples obtained from nondiabetic women, and analyzed for magnesium concentration. Mean +/- standard deviation amniotic fluid magnesium concentration (mg/dL) was 0.86 +/- 0.21 in the diabetic group and 1.06 +/- 0.22 in the control group (P less than .001). It is concluded that in the diabetic pregnancy, a state of fetal magnesium deficiency exists. This deficient state may contribute to neonatal hypocalcemia in infants of diabetic mothers.     

Alterations of maternal metabolism in normal and diabetic pregnancies: differences in insulin-dependent, non-insulin-dependent, and gestational diabetes

In normal and diabetic pregnancies, the placenta functions as a complex endocrine gland that modulates all classes of maternal nutrients to the fetus. The metabolic alterations of normal pregnancy are diabetogenic and associated with modest resistance to endogenous insulin. Pregnant women with carbohydrate intolerance represent three metabolically heterogeneous groups: type I (insulin-dependent), type II (non-insulin-dependent), and gestational diabetes. Patients with type I diabetes are at risk for ketosis and require replacement therapy because of a deficient production of insulin. They have decreased 24-hour, around-the-clock levels of C-peptide and glucagon, and lower nocturnal cortisol values and higher 24-hour prolactin levels than those of women with type II diabetes. Type II pregnant diabetic patients are not prone to ketosis and are more resistant to endogenous and exogenous insulin. They have higher fasting and meal-stimulated levels of C-peptide, accentuated fasting hypertriglyceridemia, and significantly lower high-density lipoprotein cholesterol levels than those of normal or type I women. In gestational diabetes, the metabolic stress of pregnancy evokes reversible hyperglycemia which may be associated with either a surfeit or a deficiency of insulin. These metabolic differences among diabetic pregnant women could have implications for placental structure and function that might influence fetal growth.     

Good diabetic control early in pregnancy and favorable fetal outcome

This study of 74 diabetic pregnant women shows that tight maternal blood glucose control before the 32nd week of gestation significantly reduces the incidence of fetal macrosomia (11%) when compared with that of patients with fair to poor control before the 32nd week of gestation (44%, P less than .05) or with those whose good diabetic control was not achieved until after the 32nd week of gestation (34%, P less than .05). The macrosomic infant produced by a diabetic mother is associated frequently with an elevated amniotic fluid C-peptide level, which shows the evidence of intrauterine fetal hyperinsulinism. The use of tight diabetic control early in pregnancy to reduce the risk of fetal macrosomia and/or neonatal complications is of clinical importance in the management of diabetes in pregnancy.     

Correlation between newborn gastric aspirate microviscosity and gestational age

Fluorescence polarization (FP) values were determined in 59 samples of amniotic fluid and in 56 samples of newborn gastric aspirate, collected from the 28th to 41st week of gestation. Both amniotic fluid and gastric aspirate FP values showed a significant correlation respectively with gestational age (r = 0.77; p less than 0.01 and r = 0.50; p less than 0.01). The two regression lines were parallel (F = 0.052; p = 0.8046). Mean FP of gastric aspirate were about 0.05 lower than those of amniotic fluid. One infant whose gastric aspirate FP value was 0.342 developed RDS.     

Catecholamine responses in fetal lambs subjected to hemorrhage

We monitored plasma epinephrine and norepinephrine responses to hemorrhage of 20% of estimated blood volume in 11 chronically instrumented, unanesthetized fetal lambs. In addition, we performed control experiments--blood sampling but no hemorrhage--in five fetuses. Arterial blood gases, pH, mean arterial pressures, heart rates, and plasma norepinephrine and epinephrine levels were similar in both groups in the resting state. Arterial blood gases and pH did not change significantly during the experimental period in either group. Mean arterial pressure, heart rate, and plasma norepinephrine and epinephrine levels did not change in the control group during the experimental period. Hemorrhage was associated with a significant decrease in fetal mean arterial pressure, 39.8 +/- 1.2 to 29.6 +/- 2.1 mm Hg, and heart rate, 186 +/- 8 to 146 +/- 6 bpm (p less than 0.01 in both cases). There was a significant increase in plasma norepinephrine, 664.9 +/- 91.6 to 1384.8 +/- 216.7 pg/ml (p less than 0.02) and epinephrine, 224.6 +/- 43.6 to 681.7 +/- 199.0 pg/ml (p less than 0.01) with hemorrhage. These results demonstrate significant catecholamine responses to hypovolemia in the fetal lamb.     

Serum fructosamine and amniotic fluid insulin levels in patients with gestational diabetes and healthy control subjects

Gestational diabetic pregnancies with fetal hyperinsulinism should be identified because these cases require insulin therapy. To determine the relationship between the serum fructosamine and amniotic fluid insulin concentrations, these substances were measured in 87 pregnant women with impaired glucose tolerance. Fructosamine was also measured in 678 healthy pregnant control subjects, in 113 of whom amniotic fluid insulin levels were available. Fetal hyperinsulinism was rare at serum fructosamine levels of less than 2.6 mmol/L. These results suggest that when both the oral glucose tolerance test and fructosamine level are used, only 30% of women with gestational diabetes need to undergo amniocentesis to assess fetal insulin homeostasis.     

A comparison of amniotic fluid fetal pulmonary phospholipids in normal and diabetic pregnancy

OBJECTIVE: Our purpose was to determine whether there are differences in the timing of the appearance of various amniotic fluid fetal pulmonary phospholipids in normal and diabetic pregnancy. STUDY DESIGN: A case-control study of 295 subjects with diabetes and 590 control subjects was performed by use of gestational age-matched amniocentesis specimens analyzed for lecithin/sphingomyelin (L/S) ratio, phosphatidylinositol (PI), and phosphatidylglycerol (PG) composition. Diabetic subjects were stratified according to type of diabetes, degree of blood glucose control, and birth percentile of the neonate. RESULTS: There was no difference in L/S ratios over gestational age by type of diabetes or quality of glycemic control. Women with preexisting diabetes had significantly higher PI levels at 33 to 35 weeks' gestation, which became similar to levels of control subjects after 36 weeks, whereas patients with gestational diabetes mellitus and control subjects had similar PI levels throughout. In diabetic subjects, the onset of production of PG was delayed from 35.9 +/- 1.1 weeks (controls) to 38.7 +/- 0.9 weeks (overt diabetics) and 37.3 +/- 1.0 weeks for gestational diabetes mellitus (P <.001). The delay in PG synthesis was not related to infant sex, level of maternal glucose control, or fetal macrosomia. CONCLUSIONS: Fetal pulmonary maturation, as evidenced by the onset of PG production in the amniotic fluid, is delayed in diabetic pregnancy by 1 to 1.5 weeks. This delay appears to be associated with an early and sustained elevation in amniotic fluid PI levels at 32 to 34 weeks.     

Predominance of L-dopa in fetal plasma and the amniotic fluid during late gestation in the rat

The purpose of this study was to reevaluate catecholamine distribution in fetal and maternal compartments during late gestation in the rat. Fetal and maternal plasma and amniotic fluid were collected from anesthetized rats on consecutive days from day 17 to day 22, the day of parturition. The fluid was analyzed for dihydroxyphenylalanine (L-dopa), dopamine, norepinephrine, and epinephrine by radioenzymatic assays. Amniotic fluid volume was determined by a direct weighing method. L-Dopa concentrations constituted approximately 50% of total fetal plasma catecholamines and were significantly higher in fetal than in maternal circulation. Dopamine concentrations in fetal plasma were tenfold lower than those of L-dopa but were also significantly higher in fetal than in maternal plasma; norepinephrine levels were similar in both. Maternal plasma epinephrine levels remained relatively constant, whereas fetal epinephrine levels increased fiftyfold from day 17 to day 22. L-Dopa concentrations in the amniotic fluid were tenfold higher than those of dopamine, and the concentrations of both increased markedly during the last 2 days of gestation. However, this apparent rise could be attributed to the concomitant fivefold reduction in the amniotic fluid volume observed at this time. It is concluded that L-dopa is the predominant catecholamine in both the fetal plasma and the amniotic fluid during late gestation in the rat. At the present time, neither the source nor the possible physiologic functions of L-dopa during fetal life are known.     

Surfactant associated protein (SAP-35) in amniotic fluid from diabetic and nondiabetic pregnancies

We have recently shown that, with the current management of insulin-dependent diabetes during pregnancy, infants of diabetic mothers are at no greater risk of respiratory distress syndrome (RDS) than an appropriately matched control population. A previous study suggested a selective inhibition of surfactant associated protein of 35,000 daltons (SAP-35) in the amniotic fluid of diabetic pregnancies. In order to determine whether a selective inhibition of SAP-35 occurs in well controlled, insulin-dependent diabetic pregnancies, we compared SAP-35 concentration and lecithin/sphingomyelin (L/S) ratios in amniotic fluid from 30 well controlled, insulin-dependent women with 30 nondiabetic pregnant women pair-matched for gestational age, race, and indication for amniocentesis. Gestational ages ranged from 30-43 weeks, with a mean of 36.5 +/- 2.5 weeks, in both groups. Surfactant associated protein-35 was measured by an enzyme-linked capture immunoassay specific for SAP-35 and its oligomers. Mean +/- SEM SAP-35 was 3.7 +/- 0.4 micrograms/mL (N = 30) in the diabetic group, not significantly different from 5.0 +/- 1.1 micrograms/mL (N = 30) in the control group (P greater than .05). Mean L/S ratios were also not different: 2.4 +/- 0.1 (diabetic) compared with 2.3 +/- 0.1 (control); P greater than .05. The rate of RDS was similar in both groups. We conclude that in well controlled diabetic pregnancies, fetal lung maturation, as assessed by the L/S ratio, SAP-35 concentration, and outcome, is not adversely affected.     

Amniotic fluid insulin,C peptide concentrations,and fetal morbidity in infants of diabetic mothers

Glucose, insulin, C peptide, and insulin antibody concentrations were measured in amniotic fluid collected under basal conditions and 2 hours after an arginine challenge from 61 insulin-treated diabetic women (12 basal and 49 after arginine challenge) and 31 nondiabetic pregnant women in late gestation (23 basal and eight after arginine challenge). The insulin, C peptide, and glucose concentrations were significantly higher in diabetic pregnant women than in nondiabetic pregnant women in each case. In the amniotic fluid obtained after arginine challenge in diabetic pregnant women, C peptide concentration was correlated with both insulin concentration (r = 0.61) and birth weight (r = 0.53). The insulin and C peptide concentrations were significantly higher (p < 0.025) in samples from diabetic pregnancies associated with fetal morbidity than from diabetic pregnancies without fetal morbidity. Basal amniotic fluid insulin and C peptide concentrations were slightly greater in overweight infants of diabetic mothers compared to those of normal weight, whereas the differences for insulin and C peptide concentrations in the amniotic fluid obtained after arginine challenge were highly significant (p < 0.0125 and p < 0.0005, respectively). Finally insulin and C peptide concentrations in the amniotic fluid obtained after arginine challenge in diabetic pregnant women showed a correlation with maternal metabolic control but not with the degree (White classification) of maternal diabetes. No or negligible interference of insulin antibody in the radioimmunoassay of insulin in amniotic fluid was observed.     

Lung maturation in diabetes in pregnancy: if and when to test

Poorly controlled maternal diabetes in pregnancy may delay fetal pulmonary maturation. However, diabetic women with good glycemic control have fetal lung maturation at the same gestational age as nondiabetic women. With modern ultrasound technology, gestational dates can be accurately assessed in the first or early second trimester. Respiratory distress syndrome is rare in nondiabetic and well-controlled diabetic pregnancies confirmed by early ultrasound to be at or beyond 37 weeks. Early confirmation of dates thus eliminates the need for amniotic fluid assessment of fetal lung maturity prior to elective delivery at or beyond 38 weeks in well-controlled diabetic and nondiabetic women. Poorly controlled diabetic women and pregnancies without early ultrasound verified dates may require amniotic fluid analysis for lung maturity prior to elective delivery.     

Neonatal morbidity in pregnancy complicated by diabetes mellitus: predictive value of maternal glycemic profiles

The relationship between glycemic control and perinatal outcome was assessed in a relatively uniform population of 75 White Class B through D pregnant diabetic women. All patients used glucose reflectance meter self-monitoring and performed a minimum of four determinations daily. Mean capillary blood glucose was calculated from a minimum of 16 weeks of determinations. Regression analysis confirmed a correlation between these values and third-trimester hemoglobin A1 (p less than 0.001). The study population was divided into two groups on the basis of mean capillary blood glucose values: group I, mean capillary blood glucose less than 110 mg/dl (43 patients) (mean = 96.8 +/- 7.1); group II, mean capillary blood glucose greater than 110 mg/dl (32 patients) (mean = 126 +/- 9.0). Of the 32 patients in group II, eight had mean capillary blood glucose greater than or equal to 130 mg/dl. The degree of maternal glycemic control appeared to affect perinatal outcome. At least one form of infant morbidity was present in 33% of group I infants compared with 53% of group II. Significant differences were observed for the incidence of hypoglycemia (p less than 0.05), macrosomia (p less than 0.05), and respiratory distress syndrome (p less than 0.01). One of six group I infants delivered at 35 to 36 weeks developed respiratory distress syndrome, compared with four of seven group II patients. The appearance of phosphatidylglycerol in amniotic fluid appeared delayed in group II patients at term. These data suggest that maintaining mean capillary blood glucose values less than 110 mg/dl may serve to reduce several major forms of morbidity in the infant of the diabetic mother. This information is helpful in establishing objectives for glycemic control in pregnant women using self-monitoring techniques.     

Fetal pancreatic function in infants of diabetic and rhesus-isoimmunized women

OBJECTIVE: To measure insulin and glucagon concentrations in amniotic fluid (AF) collected near term in basal conditions and after an arginine test in diabetic, rhesus-isoimmunized, and control pregnant women. METHODS: At baseline, AF was collected from 44 diabetic, 32 rhesus-isoimmunized, and 27 control pregnant women in late pregnancy. Fifty-two diabetic, six rhesus-isoimmunized, and nine control pregnant women had amniocentesis 2 hours after arginine infusion (30 g intravenous/30 minutes) at 33-36 weeks. RESULTS: Baseline AF glucose concentrations were significantly greater in diabetic women than the other conditions, and they related to the gestational age in the women with hemolytic disease of the newborn. Insulin and glucagon AF content of isoimmunized pregnancies overlapped controls, whereas insulin and insulin/glucagon molar ratios were significantly higher, and glucagon values lower, in diabetic pregnancies compared with isoimmunized and control pregnancies. In isoimmunized pregnancies, the AF concentrations of glucose, insulin, and glucagon were correlated with gestational age (less than 34, 34 weeks or more). The samples collected after arginine infusion, compared with those collected at baseline, showed significantly greater insulin and insulin/glucagon molar ratio values in diabetic (28 +/- 5 versus 11 +/- 1 microU/mL, P = .001; 29.4 +/- 1.7 versus 12.0 +/- 2.8, P = .001) and in Rh pregnant women (18 +/- 6 versus 7.7 +/- 0.7 microU/mL, P = .001; 30 +/- 9 versus 3.4 +/- 0.4 I/G, P = .001), whereas no significant difference was observed in the controls. CONCLUSION: Basal islet hormone concentrations in AF are modified by maternal diabetes and further influenced by arginine administration. Arginine produces an AF response that is similar in pregnancies complicated by diabetes mellitus and rhesus-isoimmunization, despite different (hyperglycemia and euglycemia) maternal blood glucose levels.     

Amniotic fluid prolactin and fetal lung maturation

Concentrations of prolactin in amniotic fluid, fetal plasma, and maternal plasma were determined in 34 rhesus monkeys delivered by hysterotomy under general anesthesia at gestational ages of 110 to 160 days (term, 165 days). Included were 15 cases (gestational ages 110 to 143 days) in which the mothers received 2 mg of betamethasone intramuscularly daily for 3 days prior to delivery. Fetal lung maximum volumes were determined in addition to the following indices of fetal lung surfactant: lung alveolar stability, lung phosphatidylcholine concentrations, lung extract surface tensions, and amniotic fluid lecithin to sphingomyelin ratios. Amniotic fluid prolactin was found to correlate significantly with lung alveolar stability (r = 0.51; p less than 0.01), lung phosphatidylcholine (r = 0.51; p less than 0.01), lung extract surface tension (r = -0.39, p less than 0.05) and amniotic fluid lecithin/sphingomyelin ratio (r = 0.50; p less than 0.01). These correlations remained statistically significant even when the effects of gestational age were taken into account. These findings suggest that amniotic fluid may modulate fetal production of surfactant via its prolactin content.