Main

The metabolic responses which occur during the transition from fetal to neonatal life represent a change from a quiescent state in which inhibitory stimuli dominate to one of near maximal rates of metabolic activity that are rarely matched again during postnatal or adult life. These responses are necessary to establish breathing and maintain body temperature after birth, both of which are important in determining not only survival but also an individual's ability to thrive. In the ovine fetus minimal metabolic activity is the result of limited placental transfer of oxygen(1), the presence of placental inhibitory factors(2), low concentrations of stimulatory factors such as T3(3), and the immature state of BAT(4). Prematurely delivered lambs are therefore unable to generate heat by nonshivering thermogenesis(5), a deficiency which has been attributed to low plasma concentrations of T3(6). The ability of BAT to rapidly generate large amounts of heat is the result of the electron transport chain becoming uncoupled from ATP synthesis, an effect mediated by a unique mitochondrial uncoupling protein(7). Changes in ontogeny of BAT in utero prevent the occurrence of nonshivering thermogenesis, because mRNA for uncoupling protein and its thermogenic activity (GDP binding to mitochondrial protein) remain low during fetal life, but rise rapidly after birth(4, 8).

Studies which have adopted the model of fetal cooling and oxygenation have demonstrated that clamping the umbilical cord results in significant increases in fetal oxygen consumption and plasma concentrations of FFA and glycerol(1). These responses have all been interpreted as demonstrating the initiation of nonshivering thermogenesis in BAT(1), but no direct measurements of BAT function were made in these investigations. Moreover, this has led to the hypothesis that the onset of BAT thermogenesis after birth is due to the removal of placental inhibitory factors, such as adenosine and prostaglandin E2(2, 9). The extent to which stimulatory factors may contribute to changes in BAT function after birth has not been fully determined, although in vitro and in vivo studies in fetal and adult rats have established that norepinephrine and T3 can act to regulate uncoupling protein activity(10–12).

One process that appears to have a critical role in determining metabolic adaptation after birth is the route of delivery, as vaginally delivered lambs are able to maintain body temperature over the first few minutes and hours of life(6, 13), but this is not the case in lambs delivered near-term by cesarean section(13, 14). Similarly, cesarean section-delivered infants exhibit lower axillar, interscapular, and skin temperatures than those delivered per vaginum(15), a response which has been linked to a reduced postpartum surge in catecholamines(16). An increase in the rate of release of norepinephrine after birth is likely to stimulate heat production in BAT by its action on β3-adrenergic receptors(17). The discovery of theseβ3-adrenergic receptors has led to the development of several agonists, including Zeneca D7114 which is known to restore mRNA for uncoupling protein and its expression in BAT in adult dogs, as well as increasing heat production(18). Administration of Zeneca D7114 to newborn lambs can increase colonic temperature by more than 1.0°C(19), but it is not known what effect this compound may have on BAT function in cesarean section-delivered lambs. In addition, the time course over which colonic temperature may be restored in cesarean section-delivered lambs has yet to be fully elucidated, because body temperature declines over the first 30 min postpartum(13) but is within the normal range at 1 d of neonatal life(20).

Up to 24% of infants may be born by cesarean section delivery in the United States. As there is increasing evidence indicating an impairment of thermoregulatory control in these infants(15), it is important to gain an increased understanding of the mechanisms by which body temperature can be controlled after cesarean section birth. The present study was undertaken to determine whether the control of thermoregulation can be manipulated in near-term lambs delivered by cesarean section into a cool ambient temperature of 15°C. This was assessed by a combination ofin vivo recordings, measurements of the catecholamine content and thermogenic activity of BAT, and circulating levels of thyroid hormones, cortisol, and metabolites during the first few hours of neonatal life.

METHODS

Animals and diet. Seven single-bearing ewes, eight twinbearing ewes, and one triplet-bearing Bluefaced Leicester cross Swaledale ewe of known mating date and diagnosed as being pregnant by using a real-time ultrasound echograph were group housed 6 wk before their predicted lambing date. The mean daily minimum and maximum temperatures recorded at 0900 h were 3.2 ± 2.7 (SD) and 12.6 ± 2.1°C, respectively. Ewes were fed daily a diet comprised of barley based concentrate (0.2-0.8 kg) and chopped hay ad libitum, which contained sufficient energy and nitrogen to meet the requirements for maintenance and pregnancy for the final 4 wk of gestation. The mean ewe body weight was 91 ± 2 kg (SEM; n = 16), body condition score was 3.2 ± 0.1 (n = 16), and their mean colonic temperature, when maintained at an ambient temperature of 15°C immediately before cesarean section, was 39.4 ± 0.1°C (n= 16).

Experimental design. Cesarean section delivery was performed as described by Clarke et al.(20) between 139-147 d of gestation (term = 147 d), while ewes were maintained in a purpose-built constant temperature control room (15 ± 1°C). Paravertebral anesthesia was administered by inserting spinal needles to allow blockage of the T13, L1, and L2 spinal nerves by surrounding them with 2% Xylocaine (Astra Pharmaceutical Products, Worcester, MA), as the dorsal and ventral branches of these nerves pass above and below the transverse processes of the lumbar vertebrae. This was followed by jugular venous injection of 4-6 mL of ketamine (100 mg·mL in saline) into the ewe. A flank incision was then made, and the fetuses were delivered. All lambs were monitored to ensure that continuous breathing was established, which normally occurred within 2-4 min of birth. Colonic temperature was then continuously monitored using an electronic thermometer (type 3GID, Light Laboratories, Brighton, UK), and the lambs were dried with a towel. At 25-30 min after birth a jugular vein catheter was inserted(20), and a 5-mL blood sample was taken. All lambs were placed in an indirect calorimeter at a cool ambient temperature of 15°C during which time continuous measurements of colonic temperature, as well as breathing frequency and pattern, using inductance plethysmography, were taken over a 30-40-min period, which included at least one 10-15-min epoch of NREM. Sleep state and incidence of shivering were determined from the respiratory pattern measurements(21). During any period of NREM sleep an individual lamb was designated as shivering if interference of respiratory patterns was observed for more than 10% of the period in which it remained asleep. Between 60 and 80 min of neonatal life a 10-mL blood sample was taken without arousing the lamb, and all lambs were fed orally a 20-mL volume of milk replacer (100 g·L-1: Volac Lamlac, Royston, Herts, UK). One lamb from each set of twins, plus four out of seven single lambs and one triplet lamb were randomly allocated to receive theβ3-agonist Zeneca D7114, in which case this was dissolved in the 20-mL volume of milk so that they were fed at a dose of 10 mg·kg body weight-1. A total of 14 lambs were treated with theβ3-agonist (i.e. four single lambs, nine twins, and one triplet), and nine lambs acted as controls (i.e. three single lambs, six twins, and two triplets). One twin failed to establish continuous breathing after cesarean section delivery, so this lamb was not included in the study. Six lambs (four single lambs and two twins) failed to restore colonic temperature to normothermic values. Two of these lambs were fed milk alone, and four had milk plus β3-agonist. This latter group ofβ3-agonist-treated lambs was therefore considered as a separate hypothermic group for all results presented.

It was possible to obtain measurements of oxygen consumption and carbon dioxide production for only seven control (i.e. one single lamb, five twins, and one triplet), six normothermic β3-agonist-treated lambs (i.e. five twins and one triplet), and four hypothermicβ3-agonist-treated lambs (i.e. two single lambs and two twins). The mean values presented represent values obtained during periods of NREM sleep to minimize variations due to animal movement(20), and were recorded using two identical indirect-calorimetry systems based on that described by Symonds et al.(21), with the modification that air flow was measured using a differential flow indicator (Perflow Instruments Ltd., Willesdon, UK). Mean measurements of oxygen consumption and carbon dioxide production were then used to calculate rates of heat production for each lamb(22). Blood samples were then taken at 25-30-min intervals postfeeding until 110-150 min after birth, at which time colonic temperature had plateaued. Lambs were then humanely killed by i.v. administration of barbiturate (100 mg·kg pentobarbital sodium; Euthatol). Both perirenal adipose tissue depots plus 20 g of liver were rapidly removed and stored at >70°C for measurement of GDP binding, protein, lipid, Hb, glycogen, catecholamine content, and I5′D activity as described by Symonds et al.(21) and Clarkeet al.(20).

All operative procedures and experimental protocols had the required Home Office approval as designated by the Animals (Scientific Procedures) Act of 1986.

Laboratory procedures. Mitochondria were prepared from frozen tissue as described by Symonds et al.(21). The protein contents of homogenates and mitochondria were measured by the method of Lowry et al.(23) and cytochromec oxidase activity measured to assess the recovery of mitochondrial protein(21). The Hb content of BAT homogenates were measured via the cyanmethemoglobin method using the Boehringer Mannheim test kit (no. 124 7429). BAT thermogenic activity was assessed from the in vitro activity of the mitochondrial conductance pathway using GDP at a concentration of 2 μM, with nonspecific binding measured using a 200 μM concentration of GDP. The amount of [3H]GDP trapped in extramitochondrial spaces was corrected for by measuring the trapping of[14C]sucrose(21). The lipid content of BAT was measured by ether extraction of a 5-g sample of tissue(21). Hepatic glycogen content was assayed using the method of Keppler and Decker(24). The activities of type I (BAT and liver) and II I5′D (BAT) were determined by measuring the release of 125I- from 125I-labeled reverse T3(Amersham International plc, Bucks, UK) based on the method of Wu et al.(25), as described by Clarke et al.(20).

The total catecholamine content of perirenal adipose tissue was determined using a HPLC method as described by Arkinstall and Jones(26). Tissue homogenates were prepared from 0.5 g of frozen perirenal adipose tissue placed in ice-cold 0.4 M perchloric acid, containing 2 mM sodium metabisulfite and 3,4-dihydroxybenzylamine which acted as an internal standard (i.e. 40 ng/mL), using an Ultra-Turrax T25 homogenizer at 20 000 rpm for 30 s, to give a final volume of 4 mL. The homogenate was centrifuged at 800 × g for 10 min, at 4°C, and potassium hydroxide (10%, wt/vol) was added to 2 mL of supernatant to obtain pH 2-3. After this second centrifugation, the catecholamines were extracted from 0.5 mL of supernatant by inversion for 10 min with 30 mg of neutral alumina (Sigma Chemical Co., UK: activity grade 1) in the presence of 0.05 mL of 10 mM sodium metabisulfite and 1.0 mL of 0.5 M Tris-HCl, pH 8.6, containing 2% (wt/vol) EDTA. The extraction process was performed at 4°C in darkness, after which the alumina was sedimented by centrifugation at 9980× g for 2 min at 4°C and then washed three times in 1.0 mL of 5 mM Tris-HCl buffer containing 1 mM sodium metabisulfite. The catecholamines were desorbed from the alumina by shaking for 1 min in the presence of 0.25 mL of 0.8 M perchloric acid. After [Illegible Text] further centrifugation at 9980 × g for 2 min, at 4°C, 0.1 mL of supernatant was injected directly onto the HPLC analytical column.

The HPLC, with an electrochemical detection system, consisted of a solvent pump and valve injector with 0.1 mL of constant volume injection loop, precolumn filter (Hichrom, Theale, UK) and an Altex Ultrasphere-ODS (5-μm particle size) stainless steel analytical column (25 cm × 4.6 mm inside diameter; Anachem Ltd., Luton, UK). Catecholamines were detected using a flow-through electrochemical cell with a glassy carbon working electrode set at +0.72 V and an electronic control module. The catecholamines were eluted from the column at 1.0 mL/min with an eluting buffer consisting of 0.05 M anhydrous sodium acetate, 0.01 M citric acid, pH 5.1, containing 100 mg/L octyl sodium sulfate, 50 mg/L EDTA, and 15% (vol/vol) methanol, and was degassed with helium. Catecholamine concentrations were then determined from chromatograms based on peak height for standards (i.e. norepinephrine, epinephrine, and dopamine). Recoveries were calculated and corrected for that of 3,4-dihydroxybenzylamine, which was 60-75%. Each tissue was assayed in duplicate and the limit of catecholamine detection was 10 ng/mL, corresponding to approximately 50 pmol/g of tissue.

Plasma concentrations of glucose, lactate, and FFA were measured enzymatically, and T3, T4, and cortisol were measured using RIA as described by Clarke et al.(20).

Statistical analysis. Statistical analysis with respect to differences between each group of lambs and response to feeding was assessed by analysis of variance. For all measurements made, this analysis indicated that, for each group of normothermic or hypothermicβ3-agonist-treated lambs, there were no significant influences between individuals. Significant differences between the number of lambs in each group that shivered or had epinephrine detectable in BAT were assessed using a χ2 test.

RESULTS

Colonic temperature and metabolic adaptation after birth. Over the first 30-55 min of life, colonic temperature decreased in all lambs(Fig. 1), but distinct differences between animals were apparent in the magnitude and time course of this response, irrespective of treatment. Four lambs exhibited a lower (NS) colonic temperature when first recorded 4 min after birth that continued to decrease until 72 ± 22(SEM) min of life. These animals therefore exhibited a significantly greater(p < 0.05) decrease in colonic temperature after birth [3.57± 1.08°C (n = 4)] compared with all other lambs. Colonic temperature failed to increase in this group of lambs during the remainder of the experiment even though these lambs were treated withβ3-agonist. This group clearly differed from other lambs which were able to restore colonic temperature to normothermic values and were considered as a separate group of hypothermic β3-agonist treated lambs.

Figure 1
figure 1

Mean colonic temperature over the first 110-150 min of neonatal life in near-term lambs delivered by cesarean section. Normothermic lambs were fed milk alone (○; n = 9) or milk plus Zeneca D7114 at a dose of 10 mg·kg body weight-1 (•; n = 10) at 60-80 min of life. All hypothermic lambs (▪; n = 4) received milk plus Zeneca D7114 at a dose of 10 mg·kg body weight-1. Values are means ± SE.

Full size image

Colonic temperature in the control group (n = 9) declined over the first 28 ± 5.0 min of life by an average of 1.57 ± 0.41°C, and then increased by a rate of 0.89 ± 0.28°C·h-1. Similarly, in lambs which were to be treated with the β3-agonist (n = 10) colonic temperature fell over the first 29 ± 3.2 min of life by an average of 2.09 ± 0.60°C, and then increased at a rate of 1.34 ± 0.29°C·h-1. The majority (18 out of 19 lambs,Table 1) of these lambs shivered up to the time of feeding, but after administration of β3-agonist, 6 out of 10 lambs rapidly stopped shivering, whereas all control lambs continued to shiver. Despite the observation of reduced shivering in β3-agonist-treated lambs, heat production was not increased, and mean plateau colonic temperature was significantly greater in this group [control 39.59 ± 0.13(n = 9), β3-agonist 40.39 ± 0.18 (n = 10)°C (p < 0.05)]. In the hypothermicβ3-agonist-treated group, all lambs stopped shivering after being fed the β3-agonist, but heat production and breathing frequency were significantly lower (p < 0.05) than the normothermicβ3-agonist group.

Table 1 Mean heat production, breathing frequency, and incidence of shivering
Full size table

Mean gestational age of hypothermic β3-agonist-treated lambs[141.8 ± 1.0 d (n = 4)] tended to be shorter (NS) than control [145.1 ± 0.8 d (n = 9)] andβ3-agonist-treated [143.9 ± 0.7 d (n = 10)] lambs. There was no difference in lamb birth weight between groups(hypothermic β3-agonist-treated: 5.7 ± 0.6 (n = 4), control: 4.6 ± 0.5 (n = 9), β3-agonist: 5.2± 0.2 (n = 10) kg), but these lambs possessed significantly heavier lungs when sampled 110-150 min after birth [hypothermicβ3-agonist-treated: 131 ± 1 (n = 4) (p< 0.01), control: 93 ± 12 (n = 9), β3-agonist: 109 ± 7 (n = 10) g), and took longer (NS) to established continuous breathing (hypothermic β3-agonist-treated: 4.7 ± 1.3 (n = 4), control: 2.9 ± 0.4 (n = 9),β3-agonist: 2.9 ± 0.5 (n = 10) min]. There was little difference in the number of male and female lambs between groups(hypothermic β3-agonist-treated: 2 female, 2 male, control: 5 female, 4 male, β3-agonist: 6 female, 4 male).

Plasma metabolite and hormone concentrations. There were no differences in plasma glucose or lactate concentrations between groups(Table 2) and lactate levels decreased with time in all lambs. Plasma FFA concentrations increased (p < 0.001) after feeding milk in control and β3-agonist-treated, but not in hypothermic β3-agonist-treated lambs for which FFA levels were significantly lower (p < 0.05) than in the β3-agonist group. Plasma T3 and cortisol concentrations were significantly lower(p < 0.05) in hypothermic β3-agonist-treated than control lambs irrespective of sampling time. A significant increase(p < 0.05) in T3 but not cortisol levels after feeding was observed in control and hypothermic β3-agonist-treated groups. After treatment, the rise in plasma T3 concentrations were less pronounced (p < 0.05) in the β3-agonist- and hypothermic β3-agonist-treated groups than in the controls. No significant differences in plasma concentrations of T4 were observed between groups.

Table 2 Mean jugular venous plasma concentrations of glucose, lactate, FFA, and thyroid hormones
Full size table

Perirenal adipose tissue and liver composition. There were no differences in total weight, lipid, protein, or mitochondrial protein content of perirenal adipose tissue between groups (Table 3), but Hb content was significantly higher (p < 0.05) inβ3-agonist-treated than hypothermic β3-agonist-treated lambs. GDP binding to mitochondrial protein was unaffected byβ3-agonist treatment, but was significantly lower (p< 0.05) in hypothermic β3-agonist-treated than control lambs. Norepinephrine content of perirenal adipose tissue was also significantly lower (p < 0.01) in hypothermic β3-agonist-treated lambs, compared with other groups (Table 4). There was considerable variation in the epinephrine content of perirenal adipose tissue but mean values tended to be lower (NS) in hypothermicβ3-agonist-treated lambs. In addition, epinephrine was undetectable in a significant (p < 0.01) number ofβ3-agonist-treated lambs, and remained detectable only in those lambs which continued to shiver after administration of this compound. I5′D activities in both perirenal adipose tissue and liver were similar between groups (Tables 4 and 5), as were liver weights and glycogen content.

Table 3 Mean weight and composition of perirenal adipose tissue
Full size table
Table 4 Mean catecholamine content and I5′D activity of perirenal adipose tissue
Full size table
Table 5 Mean liver weight, glycogen content, and I5′D activity
Full size table

DISCUSSION

The present study clearly demonstrates that near-term lambs delivered by cesarean section into a cool ambient temperature are able to restore colonic temperature, but the time course of this adaptation varies and can be manipulated by β3-agonist administration. In normothermic lambs a number of differences were apparent with respect to endocrine and metabolic responses to feeding that appeared to be directly linked to treatment withβ3-agonist. Within 5-10 min of administration of theβ3-agonist, the majority of these lambs stopped shivering but did not increase their mean rate of heat production. These responses are indicative of an enhancement of nonshivering thermogenesis, but there was no difference in the thermogenic activity of BAT between β3-agonist and control groups. The absence of any increase in oxygen consumption afterβ3-agonist treatment is not unexpected, as all lambs were maintained in a cool ambient temperature of 15°C which is well below thermoneutrality for newborn lambs (i.e. 30°C)(27), so rates of oxygen consumption were already high. In the present study, the dose of β3-agonist used was identical to that previously fed to neonatal lambs born per vaginum in which an increase in colonic temperature was observed without any change in oxygen consumption or thermogenic activity of BAT(19). There are, however, large differences in sensitivity to the thermogenic effects ofβ3-agonists in neonatal lambs, as in some cases this can result in hyperthermia with colonic temperature increasing to above 42°C (M. E. Symonds and L. Clarke, unpublished results). Altered sensitivity toβ3-agonist administration could explain why in the present study some cesarean section-delivered lambs failed to stop shivering afterβ3-agonist treatment.

The cessation of shivering in the majority ofβ3-agonist-treated lambs may be indicative of a response mediated via the central nervous system, although the possibility of effects on peripheral tissues which possess β3-adrenoreceptors(e.g. muscle)(28) cannot be ruled out. BAT sampled from β3-agonist-treated lambs did have a greater Hb content than control lambs, suggesting a higher rate of blood flow to BAT, as observed during summit metabolism in 1-d-old lambs born per vaginum(29). An increased blood flow to BAT may contribute to the greater rise in colonic temperature recorded after β3-agonist treatment, as this would enable a higher rate of substrate supply to BAT, as well as resulting in a greater rate of heat dissipation from BAT to other tissues. This adaptation in conjunction with a reduction in the incidence of shivering, which is an inefficient method of thermoregulation due to an increase in air movement around the animal thereby reducing external insulation, may explain why colonic temperature was higher inβ3-agonist-treated lambs. A failure to detect epinephrine in BAT, which is derived from the adrenal medulla via the circulation in response to stress(30), of normothermic β3-agonist lambs which stopped shivering is therefore indicative of an improvement in metabolic homeostasis in these lambs.

In the present study the thermogenic activity of BAT sampled from 2-3-h-old cesarean section-delivered normothermic lambs was between 280 and 380 pmol·mg mitochondrial protein-1, a value that is appreciably higher than previously recorded in 20-30-h-old cesarean section-delivered lambs (i.e. 115-126 pmol·mg mitochondrial protein-1)(20). In this earlier study, although all lambs maintained normal colonic temperatures, BAT was sampled 14-20 h after feeding, which contrasts with the current study in which BAT was sampled 60-80 min after feeding. Heat production is increased by feeding of neonatal piglets(31), a species that does not possess BAT(32), whereas in rat pups the presence of the dam can alter postnatal development of BAT, although it has yet to be determined if this effect is linked to alterations in feed intake(33). Plasma T3 concentrations also increase soon after feeding in young growing pigs(34), but not in postnatal lambs(35). Feeding appears to stimulate both plasma T3 concentrations and thermogenic activity of BAT in cesarean section-delivered lambs(36), but the extent to which this contributes to longer term changes in BAT function have yet to be determined.

An additional finding from the present study is that the ability of some cesarean section-delivered lambs to restore colonic temperature and/or respond to β3-agonist treatment was associated with low plasma T3 concentrations. This proposal is based upon the observation that lambs, which demonstrated a more rapid decline in colonic temperature after birth, became hypothermic and had consistently lower plasma T3, but not T4 concentrations, than those lambs which became normothermic. Hypothermic lambs also failed to exhibit a lipolytic response to feeding. It has been established that the rate of decline in colonic temperature is closely correlated with serum T3 or FFA concentrations in near-term lambs delivered by cesarean section(14). The importance of thyroid hormones in preventing an accelerated fall in body temperature after birth has also been indicated from observations that premature lambs bornper vaginum(6) or delivered by cesarean section nearterm(13) have significantly lower plasma T3 concentrations than lambs born spontaneously at term and are unable to maintain a constant colonic temperature. Furthermore, fetal lambs thyroidectomized at approximately 130 d of gestation exhibit a greater decline in colonic temperature than shamoperated controls when delivered near-term by cesarean section into the warm(37) or per vaginum into a cool ambient temperature(36). Newborn lambs can exhibit very rapid thermogenic responses to exogenous thyroid hormone administration without the prolonged latency period observed in adults(38). For example, s.c. T3 injection into neonatal lambs maintained at thermoneutrality can increase oxygen consumption within 1 h of treatment(39). In addition, umbilical vein injection of a physiologic dose (2 nmol) of T3 into near-term lambs delivered by cesarean section reduces the incidence of shivering within 1 h after birth and results in a higher colonic temperature at 2 h postpartum(40).

Prematurity contributing to a low thermogenic activity and catecholamine content in BAT may have contributed to the poor thermoregulatory control of hypothermic lambs as cesarean section was performed on average 2-3 d earlier in this group, i.e. 141-142 d of gestation. Rapid development of both BAT and the hypothalamus occurs over the final days of gestation in ovine fetuses(8, 41). Immature development of the hypothalamic-pituitary-thyroid axis is likely to be a primary factor contributing to the reduced postpartum surge in T3 observed in hypothermic lambs, as no differences between groups were observed for I5′D activity in liver or BAT, indicating that T3 production from peripheral tissues was similar in all lambs. Hypothermic lambs also exhibited lower plasma cortisol concentrations than normothermic lambs throughout the study. This difference is likely to reflect their shorter gestational age as cortisol levels increase rapidly during late gestation in lambs(42). Cortisol is known to have a unique function in the fetus, inducing a wide range of enzymes before birth that have little or no function during fetal life but on which neonatal survival depends(42). This role of cortisol does not include the regulation of BAT function at term(13), but a combination of prematurity plus low circulating levels of T3 and cortisol may together contribute to poor thermoregulatory control.

Thyroid hormones are not only necessary for the functional development of BAT(36), but are essential for the maturation of both lungs(43) and skeletal muscle(44), which if reduced can contribute to metabolic deficiencies after birth. The observation that hypothermicβ3-agonist-treated lambs possessed heavier lungs and had lower breathing frequencies indicates lung function may have been compromised after birth. This effect is likely to be closely associated with the low plasma T3 concentrations found in these lambs, as T3 promotes lung maturation by stimulating pulmonary blood flow(45), as well as being necessary for lung liquid reabsorption(43). Despite an apparent failure to clear lung liquid fluid after birth, hypothermic lambs did not exhibit any significant differences in plasma lactate or glucose concentrations, indicating that reduced lung function plus poor thermoregulatory control were not contributing to any occurrence of metabolic acidosis.

It is concluded that administration of the β3-agonist Zeneca D7114 to euthyroid cesarean section-delivered lambs enhances their ability to thermoregulate and restore colonic temperature without altering the thermogenic activity of BAT. This response may be mediated by increasing blood flow to BAT and/or an improvement in the animals thermal efficiency(i.e. decreased rate of heat loss) due to a reduced reliance on shivering thermogenesis.