Main

Acute episodes of intrauterine hypoxia or asphyxia stimulate an increase in the plasma concentrations of noradrenaline and adrenaline in fetal sheep(1–8). The increase in circulating catecholamines is important in the initiation and coordination of a range of fetal physiologic responses to hypoxic stress(9–12). These responses include a gradual increase in fetal arterial blood pressure associated with a redistribution of the fetal cardiac output to the brain, heart, and adrenal at the expense of the gastrointestinal tract, lungs, kidney, skeletal muscle, and skin(9–12). The relative contributions of the neural, paraganglia, and adrenomedullary sources to the circulating pool of fetal catecholamines during basal and hypoxic conditions are not entirely clear(13–16). There is evidence, however, that during the basal state, circulating catecholamines are predominantly derived from sympathetic neurons, whereas the fetal adrenal is the major source of catecholamines during acute hypoxia(13, 14).

Although cordocentesis studies in human fetuses have demonstrated that intrauterine hypoxemia and malnutrition are associated with an increase in plasma catecholamines(17), there have been comparatively few experimental studies on the impact of prolonged restriction of placental function on fetal sympathoadrenal function. Gagnon et al.(18) found that plasma noradrenaline concentrations were increased in fetal sheep during a 10-d period of chronic placental embolization in late gestation. Similarly, Jones and Robinson(19) found in preliminary studies that plasma catecholamines were elevated between 125 and 130 d of gestation in three chronically hypoxemic, growth-restricted fetal sheep. In the present study, we have used a well characterized model of restriction of placental growth and function(20) and investigated the impact of chronic PR on the fetal plasma concentrations of noradrenaline and adrenaline between 115 and 140 d of gestation. We have also used covariate analysis to examine the relationship between fetal blood gas status and catecholamine concentrations in animals with normal and restricted placental function. Finally, the catecholamine responses to intrafetal infusion of the indirectly acting sympathomimetic amine, tyramine, were measured after PR, to determine whether chronic PR alters the capacity of the developing sympathetic neurons to store and secrete catecholamines.

METHODS

Animals and surgery. All surgery and experiments were approved by The University of Adelaide Animal Ethics Committee. Eighteen pregnant Border-Leicester Merino cross ewes were used in this study. In 10 ewes, the majority of endometrial caruncles or placental placentation sites were removed from the uterus before conception to reduce the number of placental cotyledons subsequently formed and hence to restrict placental and fetal growth. Endometrial caruncles were removed aseptically as described previously(19, 20). Anesthesia was induced with an i.v. injection of sodium thiopentone (1.25 g, Boehringer Ingelheim, NSW, Australia) and maintained with 3-4% halothane in oxygen. The uterus was exposed through a low midline abdominal incision and opened along the antimesometrial border from the cervix to the tip of each horn. The majority of the visible caruncles were excised (64-80 caruncles) from both horns. After removal of the endometrial caruncles, the incisions were sutured with cat gut (2/0; Ethicon; Johnson & Johnson, Sydney, Australia) through the serosa and myometrium. The uterus was replaced and the abdomen closed in two layers with coated Vicryl (4/0; Ethicon). The ewes were kept under observation for 4-7 d postsurgery before being returned to the farm. After a minimum of 10 wk of recovery postsurgery, the ewes entered a mating program, and pregnancies were confirmed by ultrasound at approximately 60 d of gestation (PR group).

In all pregnant ewes (PR, n = 10; control, n = 8), vascular catheters were implanted between 102 and 112 d of gestation in a fetal and maternal carotid artery and jugular vein and in the amniotic cavity. All catheters were filled with heparinized saline, and the fetal catheters were exteriorized through an incision in the ewe's flank. All sheep received a 2-mL intra-muscular injection of antibiotics (procaine penicillin, 250 mg/mL; dihydrostreptomycin, 250 mg/mL; procaine hydrochloride, 20 mg/mL) after the induction of anesthesia. All fetuses also received an intramuscular injection of antibiotics (2 mL) at the time of surgery and intraamniotic injections of ampicillin sodium (500 mg; Austrapen, Commonwealth Serum Laboratories, Melbourne, Australia) over four consecutive postoperative days. The ewes were housed in individual pens in animal holding rooms with a 12-h light/dark lighting regimen. The ewes were fed Lucerne chaff and oats once daily between 0900 and 1300 h with water ad libitum. The ewe and fetus were killed between 140 and 141 d of gestation with an overdose of pentobarbitone.

Blood sampling and experimental protocol. After surgery, fetal(3.5 mL) and maternal (5 mL) arterial blood samples were collected every 2nd d between 104 and 140 d of gestation and assayed for metabolites and catecholamines. Blood samples were collected into heparinized tubes on ice, then centrifuged at 4°C at 1500 × g for 10 min. An additional arterial blood sample (0.5 mL) was collected from the fetus and ewe, and whole blood pH, Po2, Pco2, oxygen saturation, and hemoglobin content were measured using an ABL 330 acid-base analyzer and an OSM2 Hemoximeter (Radiometer, Copenhagan, Denmark).

Tyramine protocol. Tyramine hydrochloride (10 mg/10 mL saline for 10 min, Sigma Chemical Co., St. Louis, MO) was administered i.v. to 10 fetal sheep (PR group, n = 6; control group, n = 4) between 118 and 125 d of gestation(14, 16). Tyramine was also administered between 132 and 140 d of gestation in 13 fetal sheep (PR group, n = 6; control group, n = 7). The vehicle(saline, 10 mL for 10 min) was also administered i.v. to 6 fetal sheep in the PR group (118-125 d, n = 6 experiments; 132-140 d, n = 5 experiments) and to 7 fetal sheep in the control group (118-125 d, n= 4 experiments; 132-140 d, n = 5 experiments). Fetal arterial blood samples (5 mL) were collected at -30, -5, and 0 min before and at 5, 10, 30, and 70 min after the start of the infusion of tyramine or saline. Fetal blood samples were collected into heparinized tubes on ice and centrifuged at 1500× g for 10 min at 4°C. Fetal plasma aliquots were frozen and stored for catecholamine determination.

Extraction and measurement of plasma catecholamines. Plasma catecholamines were extracted and measured as previously described(16). Plasma samples were thawed on ice, and a 1-mL aliquot was placed into a glass tube containing acid-washed alumina (10 mg; Hart Analytical, Melbourne, Australia) and sodium metabisulfite (30 μL, 10 mM). The volume was then increased to 2 mL with distilled deionized water. Recovery of catecholamines during each extraction was monitored using the internal standard, 3,4-dihydroxybenzlamine hydrobromide (20 pmol/100 μL), which was diluted with 0.01 M sodium metabisulfite and added to all tubes. Tris-EDTA buffer (400 μL, 1 M, pH 8.6, 2% EDTA) was added to all tubes, which were then vortexed for 40 min. The supernatant was discarded, and the alumina washed four times with distilled deionized water (0.5 mL). The catecholamines were desorbed from the alumina using perchloric acid (100μL, 0.1 M) containing sodium metabisulfite (0.01 M). Samples were stored at-70°C after extraction until they were assayed using HPLC combined with electrochemical detection (BAS 200A, Liquid Chromatography, Bioanalytical Systems Inc., West Lafayette, IN). Catecholamines were separated using a phase II catecholamine cartridge column (3 μm, 100 × 3.2 mm; Bioanalytical Systems Inc.) and an aqueous mobile phase (sodium acetate, 50 mM; citric acid, 20 mM; octanesulfonic acid, 3.75 mM; EDTA, 2 mM; methanol 5% vol/vol; pH 4.3). Integration of the area under the catecholamine peaks from each chromatogram was carried out automatically using a Barspec Data System chromatographic integration program (Barspec Ltd., Israel). The sensitivity of the catecholamine assay was 0.07 pmol/mL and the intra- and interassay COVs were less than 10%.

Plasma glucose, lactate, and urea determination. Plasma concentrations of glucose, lactate, and urea were determined using a COBAS MIRA automated analyzing system (Cobas, Roche Diagnostics). Fetal plasma concentrations of glucose, lactate, and urea were determined by enzymatic analysis using the enzymes hexokinase and glucose-6-phosphate dehydrogenase for glucose, lactate dehydrogenase, and glutamate-pyruvate transaminase for lactate and urease and glutamate dehydrogenase for urea. Intraassay COVs were 1.4, 1.3, and 1.2% for glucose, lactate, and urea, respectively.

Statistical analysis. All data are expressed as the mean± SEM. The fetal and placental weights in the PR and control groups were compared using an unpaired t test.

There were 245 arterial blood samples (PR, n = 140; control,n = 105) collected from the 18 fetal sheep for the measurement of blood gas variables and 161 plasma samples collected for catecholamine determination (PR, n = 86; control, n = 75) between 100 and 140 d of gestation. Logarithmic transformation of the catecholamine data were performed, when necessary, to reduce heterogeneity of variance. Fetal arterial blood gas variables, metabolite, and catecholamine concentrations were compared in the two experimental groups using a multifactorial ANOVA with repeated measures and group (PR versus control) and gestational age(grouped into 10-d age blocks) as factors.

Correlations of plasma noradrenaline or adrenaline with arterial Po2, Pco2, or pH were investigated using multivariate analysis of repeated measures. Before covariate analysis, plasma catecholamine and blood gas data were grouped in 2-d age blocks between 108 and 140 d of gestation. The analysis (unbalanced repeated measures model with structural covariate matrices) was designed for repeated measures data, including those with unequal variance, with covariance matrices of specified patterns and incomplete data (5V program, BMDP Statistical Software, Los Angeles, CA).

In the tyramine study, 288 plasma samples were collected for plasma catecholamine determination (PR, n = 158; control, n = 130). The catecholamine data were analyzed using a multifactorial ANOVA with group (PR versus control), treatment (tyramine versus saline), gestational age (<125 d versus >135 d), time(i.e. -30, -5, 0, 5, 10, 30, 70 min), and animal as the specified factors. When a significant interaction between specified factors was identified by the ANOVA, the data were split on the basis of the interacting factor and reanalyzed. The Duncan's new multiple range test was used post ANOVA to identify significant differences between mean values, and a probability level of 5% (p < 0.05) was taken as significant.

RESULTS

Placental and fetal growth. Total placental and fetal body weights were significantly reduced (p < 0.01) in the PR group(208 ± 42 g and 3.02 ± 0.35 kg. respectively) when compared with the control group (573 ± 65 g and 4.30 ± 0.29 kg. respectively) at 140 d of gestation.

Blood gas status and metabolites. Mean fetal arterial Po2 and oxygen saturation were significantly reduced (p < 0.001), whereas arterial Pco2 and hemoglobin were significantly higher in the PR group, when compared with the control animals between 104 and 140 d of gestation (Table 1). Plasma lactate concentrations were also significantly higher (p < 0.05) in the PR when compared with the control group across this gestational age range(Table 1). There were no significant differences, however, in the arterial pH, plasma glucose, and urea concentrations between the PR and control groups during the study period (Table 1).

Table 1 Blood gas, glucose, lactate, and urea levels in PR and control fetal sheep between 102 and 140 d of gestation
Full size table

Gestational age profile of plasma catecholamines. The overall mean plasma concentration of noradrenaline was higher (p < 0.05) in the PR group (7.06 ± 1.02 pmol/mL) than in the control group (3.55± 0.18 pmol/mL) between 104 and 140 d of gestation. There was a significant (p < 0.05) increase in the plasma concentrations of noradrenaline between 110 and 140 d of gestation in both the PR and control groups (Fig. 1a). In the PR group, plasma concentrations of noradrenaline increased (p < 0.05) from 4.51 ± 0.89 pmol/mL (110-119 d) to a peak of 12.46 ± 5.54 pmol/mL at 130-140 d of gestation (Fig. 1a). In the control group, plasma concentrations of noradrenaline similarly increased (p < 0.05) from 2.83 ± 0.44 pmol/mL at 110-119 d to 4.53 ± 0.54 pmol/mL at 130-140 d of gestation (Fig. 1a).

Figure 1
figure 1

The mean plasma concentrations of noradrenaline(a) and adrenaline (b) and the ratio of noradrenaline:adrenaline concentrations (c) in the PR (black histograms) and control (white histograms) groups. The different superscripts in a indicate significant differences (p < 0.05) between mean plasma noradrenaline values at different gestational age ranges in both the PR and control groups.

Full size image

The overall mean plasma concentration of adrenaline was significantly higher (p < 0.05) in the PR than in the control fetal sheep throughout late gestation (PR, 2.89 ± 0.65 pmol/mL; control, 1.30± 0.16 pmol/mL) (Fig. 1b). Plasma concentrations of adrenaline did not change significantly with increasing gestational age, however, in either the PR or control fetal sheep (Fig. 1b).

There were no significant differences between either the PR and control groups or across gestational age ranges in the ratios of plasma noradrenaline:adrenaline concentrations (Fig. 1c). The mean noradrenaline:adrenaline ratio was 5.1 ± 1.8 in the control group and 4.3 ± 1.3 in the PR group between 104 and 140 d of gestation.

Covariate analyses of plasma catecholamine and blood gas variables. Plasma noradrenaline concentrations were significantly correlated with arterial Po2 and pH, but not Pco2, in both the PR and control groups. The relationship between plasma noradrenaline and arterial Po2 was the same in the PR and control fetal sheep and described by the equation, plasma noradrenaline = 12.34 - 0.40 Po2(Fig. 2). In contrast, there was a small but significant difference in the relationship between plasma noradrenaline concentrations and arterial pH in the PR (plasma noradrenaline = 408 - 54.4 pH) and control groups (plasma noradrenaline = 408 - 54.8 pH). The relationships between plasma noradrenaline and arterial Po2 and between plasma noradrenaline and arterial pH were not independent.

Figure 2
figure 2

A three-dimensional plot showing the relationship between individual plasma noradrenaline concentrations, gestational age, and arterial Po2 in the PR (closed circles) and control(open circles) fetal sheep.

Full size image

There was a significant correlation between the fetal plasma concentrations of adrenaline and arterial Po2 but not between adrenaline and arterial Pco2 or pH in both the PR and control groups. The relationship between plasma adrenaline and arterial Po2 was different in the PR group(plasma adrenaline = 2.15 + 0.04 Po2) when compared with the control group (plasma adrenaline = 2.15 - 0.04 Po2).

Effect of tyramine infusion on fetal plasma concentration of catecholamines. The fetal noradrenaline response after tyramine infusion was significantly greater than after saline administration. There were no differences, however, in the noradrenaline responses to tyramine infusion at 115-125 d and at 130-140 d of gestation in either the PR or control groups. The noradrenaline responses to tyramine in these two age groups were therefore pooled for each of the PR and control groups. Plasma noradrenaline concentrations were higher (p < 0.05) before, during, and after the infusion of tyramine in the PR group when compared with the control group. In the PR group, plasma noradrenaline increased (p < 0.05) from 4.51 ± 1.28 pmol/mL (0 min) to a peak of 19.40 ± 3.56 pmol/mL at 10 min after tyramine, before returning to 4.54 ± 0.85 pmol/mL at 70 min (Fig. 3a). Plasma noradrenaline also increased after tyramine administration in the control group from 2.08 ± 0.30 pmol/mL(0 min) to a peak concentration of 12.23 ± 1.67 pmol/mL at 10 min before returning to 2.47 ± 0.36 pmol/mL at 70 min. When the noradrenaline responses to tyramine were expressed as proportional changes from baseline values, there was no significant difference between the tyramine responses in the PR and control groups (Fig. 3c). The maximal proportional changes in plasma noradrenaline were similar in the two experimental groups (10 min; PR, 319 ± 55%; control, 449 ± 100%).

Figure 3
figure 3

Mean plasma noradrenaline (a) and adrenaline(b) concentrations, before, during, and after intrafetal infusion of tyramine (indicated by the dark horizontal bar) in PR (closed circles) and control (open circles) groups. The percentage changes in plasma noradrenaline (c) and adrenaline (d) during and after tyramine are also shown. The asterisks denote mean values which are significantly different from baseline values within a treatment group.

Full size image

There was no significant change in plasma noradrenaline concentrations after saline administration in the PR and control groups at either gestational age. Plasma concentrations of noradrenaline were 3.31 ± 0.58 pmol/mL(-30 min; PR group) and 2.70 ± 0.54 pmol/mL, (-30 min; control group) and 3.02 ± 0.45 pmol/mL (10 min; PR group) and 2.78 ± 0.40 pmol/mL (10 min; control group).

There was a significant increase in the plasma concentrations of adrenaline in response to tyramine in both the PR and control fetal sheep(Fig. 3b). The adrenaline responses to tyramine were not different in the two age groups, and the responses were therefore pooled. Plasma concentrations of adrenaline were significantly higher (p< 0.05) in the PR group, before, during, and after the infusion of tyramine than in the control group. Plasma concentrations of adrenaline increased in the PR group from 1.05 ± 0.29 pmol/mL (0 min) to a peak of 5.89± 1.59 pmol/mL at +10 min before returning to 1.58 ± 0.42 pmol/mL at +70 min (Fig. 3b). Similarly, in the control group, plasma adrenaline increased from 0.57 ± 0.16 pmol/mL (0 min) to a peak of 1.64 ± 0.30 pmol/mL (10 min) before returning to 0.73± 0.09 pmol/mL at 70 min (Fig. 3b). There was no difference between the PR and control groups in the adrenaline responses to tyramine when these responses were expressed as a percentage change from baseline (10 min: PR, 316 ± 52%; control, 238 ± 55%)(Fig. 3d). There was no significant effect of saline infusion on the plasma concentrations of adrenaline in either the PR or control fetal sheep at either gestational age range.

DISCUSSION

We have used a well characterized model of restriction(20) of placental growth and function in the sheep and demonstrated that plasma catecholamine concentrations are increased throughout late gestation in the growth-restricted fetal lamb. Covariate analysis revealed that there was a similar and inverse relationship between arterial Po2 and noradrenaline in the control and PR groups, whereas the relationship between fetal arterial Po2 and adrenaline was different in the two groups. We have also demonstrated that fetal catecholamine responses to tyramine were maintained after PR.

Fetal catecholamine concentrations were elevated in the immediate postoperative period (104-109 d of gestation), which presumably reflects the impact of anesthesia and surgery on the fetal sympathoadrenal system in both the control and PR groups. Plasma noradrenaline concentrations were also consistently higher in the growth-restricted fetal sheep than in their control counterparts between 110 and 140 d of gestation. These higher values appeared to be a consequence of the relationship between plasma noradrenaline, arterial pH, and Po2. Covariate analysis demonstrated that, at any given arterial pH value, plasma noradrenaline concentrations were around 2.92 pmol/mL higher in the PR than in the control group. Further, for every 1 mm Hg decrease in arterial Po2, noradrenaline increased by 0.4 pmol/mL during basal conditions in both the PR and control fetal sheep. Throughout late gestation, the prevailing mean arterial Po2 was approximately 8 mm Hg lower in the PR than in the control group, and this would therefore account for a difference of around 3.2 pmol/mL noradrenaline between the two groups. Given that the relationships of plasma noradrenaline with arterial pH and Po2 were not independent, it appears that chronic hypoxemia is a major factor that contributes to the increased circulating noradrenaline concentrations in the PR group between 110 and 140 d of gestation. Interestingly, the relationship between plasma adrenaline and arterial Po2 was different in the control and PR groups. There was an inverse relationship between plasma adrenaline and arterial Po2 in the control group, such that a fall in arterial Po2 would result in an increase in basal circulating adrenaline concentrations. In contrast, in the PR group, a fall in arterial Po2 was associated with a relative suppression of plasma adrenaline concentrations. The plasma adrenaline concentrations were consistently higher, however, in the PR than in the control group between 110 and 140 d of gestation. This would suggest that, in contrast to noradrenaline, factors other than arterial blood gas variables, perhaps glucose or lactate, are important in the maintenance of high basal adrenaline concentrations in the circulation of the PR fetal sheep.

Gagnon et al.(18) produced fetal hypoxemia between 125 and 135 d of gestation by fetal placental embolization using repeated injections of nonradioactive microspheres. There was a 2-fold increase in fetal plasma noradrenaline concentrations in the embolized group, which was maintained throughout the 10-d study(18). There was no change, however, in circulating adrenaline concentrations throughout this period. The precise relationship between fetal arterial Po2 and circulating catecholamines was not, however, investigated in either of these studies. The source of the increase in circulating noradrenaline and adrenaline during PR may be either increased secretion from the fetal adrenal medulla, extra adrenal chromaffin tissue, or sympathetic neurons. Alternatively the increase may result, in part, from a decreased placental clearance of the catecholamines.

Cohen et al.(2) have previously found that there was an inverse exponential relationship between plasma catecholamine concentrations and Po2 during varying degrees of acute hypoxemia in fetal lambs. These authors also made direct measurements of increased catecholamine concentrations in the venous effluent from fetal adrenal glands during acute episodes of hypoxia in unanesthetized fetal sheep(3). They also noted that the qualitative relationship between fetal arterial Po2 and adrenal venous catecholamines was very similar to the relationship between arterial Po2 and peripheral plasma concentrations of noradrenaline and adrenaline during acute hypoxemia(2). Furthermore, Jones et al.(7) found that destruction of the adrenal medulla abolished 90% of the increase in circulating catecholamines during fetal hypoxemia(13). These data suggest that the fetal adrenal medulla is the major source of circulating noradrenaline and adrenaline during acute hypoxemia. In the present study, however, although there was a clear inverse relationship between arterial Po2 and noradrenaline in the chronically hypoxemic fetal sheep, there was not a similar relationship between arterial Po2 and adrenaline concentrations in the PR group. This might imply that chronic hypoxemia exerts different actions on the noradrenaline- and adrenaline-containing cells of the fetal adrenal or alternatively that prolonged hypoxemia results in an increase in the noradrenaline synthetic and secretory capacity of developing symphathetic neurons. We have used intrafetal infusion of tyramine to investigate the impact of PR on the catecholamine storage and secretory capacity of the fetal sympathetic neurons.

Jones et al.(14) have previously demonstrated in the late gestation sheep fetus that adrenal demedullation did not reduce the noradrenaline response to intrafetal infusion of tyramine. In contrast, treatment of intact fetal sheep with either guanethidine sulfate for 4-6 d(15) or immunosympathectomy with anti-nerve growth factor(16) abolished the fetal noradrenaline response to tyramine, confirming that tyramine acts to displace noradrenaline from catecholamine containing vesicles within postganglionic sympathetic neurons. In the present study, the noradrenaline responses to tyramine were the same in the PR and control groups when expressed in relation to the basal circulating concentrations of noradrenaline. The proportional relationship between basal and stimulated noradrenaline concentrations in the two groups suggests that it is likely that noradrenaline is derived from the same source during these conditions, i.e. from sympathetic nerve terminals. The data suggest, therefore, that the catecholamine synthetic and secretory capacity of the developing sympathetic nervous system is not impaired as a consequence of restriction of placental function. One possibility is that PR and the presence of chronic hypoxemia throughout late gestation is a stimulus for hyperinnervation of fetal vessels and tissues by sympathetic, postganglionic neurons or alternatively low Po2 or other factors associated with PR reflexly stimulates catecholamine synthesis and secretion in developing sympathetic neurons.

It was also interesting that there was a similar and significant increase in plasma noradrenaline concentrations in both the control and growth-restricted fetal sheep after 130 d of gestation. It has previously been reported the late gestation increase in basal noradrenaline concentrations persists in normoxemic fetal sheep after bilateral adrenalectomy(21). It appears likely, therefore, that the gestational increase in circulating noradrenaline in control and placental-restricted groups is also derived at least in part from sympathetic neurons or the extra adrenal chromaffin tissue, which is a source of noradrenaline in fetal life(14).

There was a significant increase in plasma adrenaline concentrations in response to tyramine infusion in the control and PR groups. Jones et al.(14) demonstrated that the relatively small fetal adrenaline response to tyramine treatment at 128-138 d of gestation was abolished after guanethidine sulfate treatment for 4-6 d. The adrenaline response to tyramine may therefore represent the response of a subpopulation of adrenaline-containing sympathetic neurons. The present study does not allow us to conclude, however, whether the elevated basal adrenaline concentrations in the growth-restricted fetal sheep are derived from the sympathetic neurons or from the adrenaline-containing cells of the fetal adrenal medulla. The lack of an inverse relationship between arterial Po2 and adrenaline after PR suggests that acute and chronic hypoxemia have different actions on fetal adrenaline secretion.

Placental mass was predictably decreased in the PR group when compared with control animals. The placenta is a site of clearance of fetal catecholamines(22–24) and it is therefore possible that the higher basal circulating catecholamine concentrations in the PR group are, in part, a consequence of decreased placental clearance. It was noteworthy, however, that there was no evidence of slower clearance of catecholamines from the fetal circulation after tyramine infusion in the PR group when compared with the control animals. Plasma catecholamine concentrations decreased to baseline values by 60 min after the end of the infusion in both groups, which implies that the remaining functional placenta in the PR group can adequately metabolize the prevailing concentrations of the catecholamines.

It is clear that increased catecholamines play a major role in the physiologic adaptation of the fetus to acute hypoxemia(9–12). Although blood flow to the brain, heart, and adrenals is increased during acute hypoxemia, blood flow to the gastrointestinal, renal, and peripheral vascular beds decreases. The vasoconstrictor response to hypoxemia and asphyxia is reduced by sympathectomy and α-adrenergic blockade(9, 15, 16). It has also been shown that the redistribution of fetal cardiac output is also maintained with prolonged hypoxemia in pregnancy(25, 26). Although there is an increase in circulating catecholamines in the hypoxemic fetal lambs in this study and in hypoxemic, growth-restricted human fetuses during late gestation(17), it has not yet been demonstrated that the redistribution of fetal cardiac output during chronic hypoxemia is dependent on an increase in fetal sympathoadrenal activity.

In summary, we have demonstrated that there is an increase in fetal catecholamine concentrations after restriction of placental function. We have also found that the catecholamine responses to tyramine are enhanced in growth-restricted fetal lambs. There is a similar proportional relationship, however, between basal and tyramine-stimulated catecholamines in the control and growth-restricted fetal lambs. We have therefore argued that the most likely source of the increased plasma catecholamines is either“overspill” from an increased number of sympathetic neurons or a reflex stimulation of catecholamine synthesis and secretion from sympathetic neurons. Independently of their source, it is likely that the high basal catecholamines play a role in the generation of key metabolic and cardiovascular responses that enable the fetus to survive prolonged periods of chronic restriction of placental function.

The longer term consequences of exposure to high circulating catecholaminesin utero are unknown, although it is interesting that recent epidemiologic studies have clearly identified a relationship between growth restriction in utero and the onset of hypertension in adult life(27, 28). One possibility is that exposure to elevated catecholamines and/or other vasoactive neurohormones in utero may initiate morphologic and/or functional changes within the fetal cardiovascular system that are subsequently amplified and contribute to the increased incidence of hypertension in later life.