Placental release of corticotrophin-releasing hormone across the umbilical circulation of the human newborn

This study attempted to determine the placental release of corticotrophin-releasing hormone (CRH) into the umbilical circulation, and the factors which affect it, by measuring venous and arterial levels for CRH across the umbilical circulation in labouring as well as non-labouring elective caesarean section patients. The relationship with measures of fetal oxygenation and acid-base status at birth was investigated also. Forty-eight patients were studied (term labour n = 30, term elective caesarean section n = 12, and preterm labour n = 6) with blood sampling from a clamped segment of cord after delivery of the fetus and from the cord at its insertion into the placenta after delivery of the placenta, with subsequent measurement of blood gases, pH, base excess, and CRH. For all patients, mean plasma CRH levels in the umbilical and placental vein (115+/-13 and 145+/-18 pg/ml) were higher than those from the corresponding artery (85+/-7 and 102+/-8 pg/ml, P<0.01 and P<0.05, respectively), indicating placental release of CRH into the fetal compartment. In addition, placental venous and arterial cord CRH levels were higher than those from the corresponding umbilical levels (P<0.01 and P<0.02, respectively) indicating continued placental release of CRH into blood within the placenta after clamping of the umbilical circulation and delivery of the fetus. While plasma CRH levels from respective cord vessels were all significantly higher in labouring patients at term versus elective caesarean section patients, there were no differences compared with preterm labouring patients. For all patients, CRH as measured in both the umbilical and placental vein showed a modest inverse correlation to base excess as measured in the umbilical artery, -0.31 and -0.33, respectively, both P<0.05. It is concluded that CRH is released by the placenta into the fetal compartment and is increased with both term and preterm labour, and with metabolic acidosis during labour, supporting a role in the endocrine events of labour and/or compensatory changes in uteroplacental blood flow.     

Interaction of placental hormones in mother-embryo system (author's transl)

Described in this paper is the relationship between chorionic gonadotrophin, placental lactogen, progesterone, oestradiol and oestriol in the blood of mother and embryo. The concentrations of human chorionic gonadotrophin (HCG) and human placental lactogen (HPL) in the mother's blood toward the full term of pregnancy were found to be between 300 and 800 times as high as those in foetal blood. No arteriovenous difference in concentration between the same hormones was recorded from the vessels of the umbilical cord. While the concentrations of progesterone, unconjugated oestriol, and total oestriol in the mother's blood were lower than those in foetal blood, the opposite was true for oestradiol which was lower in foetal blood. Clearly reduced levels of progesterone, unconjugated oestriol, and oestradiol but increased levels of total oestriol were recorded from the umbilical artery, as compared to the umbilical vein. The above findings are intended to encourage thinking about selective penetrability of the placenta, whenever hormone preparations are prescribed in advanced pregnancy.     

Influence of interleukin-1 and interleukin-6 on modulation of hypothalamo-pituitary-adrenal axis in pregnancy at term and in spontaneous and oxytocin-induced delivery

OBJECTIVES: During the pregnancy the placenta and hypothalamus produce trophic hormones for hypothalamo-pituitary-adrenal axis (HPA), i.e. corticotropin releasing hormone (CRH) and adrenocorticotropin (ACTH). The HPA axis of pregnant women is differentially modulated in comparison to non-pregnant ones. Beside steroids, the influence on CRH release may be modulated by cytokines, especially interleukin 1 (IL-1) and interleukin 6 (IL-6). DESIGN: To evaluate the effects of IL-1 and IL-6 on modulation of HPA axis in pregnancy, we have examined the group of women with spontaneous delivery, and second group consists of women delivered after intravenous oxytocin infusion. All women were at term and in the same pregnancy age. MATERIAL AND METHODS: Blood was sampled from a maternal peripheral vein days before labour, during the second stage of labour and on the second postnatal day, the levels of IL-1, IL-6, CRH, ACTH and cortisol were measured. The concentrations of hormones were measured using RIA method. RESULTS: The level of IL-1 before the delivery was significantly higher in the group with oxytocin-induction. CRH concentration before the labour was much higher in the group with spontaneous contractions. The levels of IL-1 and CRH in both groups decreased during the labour and were lowest after the delivery. Concentration of IL-6 did not changed dependently of group and time of blood sampling. Changes in CRH in time concentration did not correlate with changes in ACTH levels. ACTH concentrations were similar in both groups, low before delivery raised during the delivery and low again after labour. Cortisol concentration in spontaneous labour was much lower before delivery in comparison with second examined group, then lowered during and after the delivery. In group with oxytocin induction, cortisol levels raised during the delivery and maintained almost the same level after the labour. The time-changes in IL-1 concentration and ACTH and cortisol levels were similar in shape in group with spontaneous delivery. CONCLUSIONS: These results suggest that IL-6 is not involved in modulation of HPA axis in pregnancy at term. IL-1 may modulate mother's HPA axis, influencing the release of ACTH and cortisol, probably via stimulation of hypothalamic CRH.     

Differences in umbilical venous and arterial leptin levels by mode of delivery

OBJECTIVE: To investigate the differences in umbilical venous and arterial leptin levels by mode of delivery. METHODS: Subjects were 30 mothers who had elective cesarean deliveries and 34 mothers who had vaginal deliveries. Umbilical venous and arterial leptin levels were measured immediately after delivery. Maternal age, neonatal gender, neonatal birth weight, placental weight, and gestational duration were recorded. Inter- and intragroup comparisons were made in umbilical venous and arterial leptin levels and obstetric variables. Significant determinants of differences in umbilical venous and arterial leptin levels were investigated. RESULTS: Umbilical venous and arterial leptin levels were higher in the vaginal delivery group (n = 34) than in the cesarean group (n = 30) (P <.01). In the vaginal delivery group, umbilical venous leptin levels were significantly higher than arterial leptin levels (P <.001). These differences were still significant after adjustment for neonatal gender, neonatal birth weight, and placental weight. However, in the cesarean group, leptin levels did not differ between umbilical vein and artery. CONCLUSION: Placental leptin release is augumented during advanced labor.     

Maternal and fetal plasma levels of free corticosteroids in pathological deliveries.

Does intrauterine acidosis induce increased steroid secretion? The concentration of free steroids (CS) increases in both fetal and maternal plasma during labor and delivery. Fetal levels are higher after vaginal than after cesarean section. These differences may indicate an important role of the fetal adrenal gland in the induction of labor or they may reflect merely the fetal response to the stress of delivery. During incrased intrauterine stress steroid secretion is increased as shown here. We examined 41 mothers and their infants during pathological labor. Pathology was assessed from fetal acidosis and/or a clinically obstetric disease of the mother or fetus. The 41 cases included 9 cesarean sections, 8 forceps deliveries; 24 spontaneous deliveries of which 7 were premature. At the time of delivery the pH and CS level were determined in maternal and umbilical vessels in all cases. During spontaneous labor blood samples were also taken during the different stages of labour. A competetive protein binding assay with transcortin without fractionation of the steroids was used. Progesteron was determined by the same assay. The level of this hormone, however, remains unchanged and hence any changes reflect changes in CS. The levels of CS were correlated with the pH values and compared to previously obtained normal values. During pathological deliveries CS levels in both mother and fetus are normal as long as there is no acidosis (Fig. 1). If acidosis is present the CS level in the umbilical artery is usually higher than normal. In 13 out of 18 vaginal deliveries the CS level was above normal, in the other 5 at the upper limit of normal (Fig. 1 and 2). At the same time the a--v difference becomes smaller and sometimes even negative. No changes were noted in maternal and umbilical venous blood (Tab. I and II). Similar dependence on the pH was found for cesarean sections (Tab. III). In premature deliveries without acidosis in the umbilical artery the CS levels were lower in both mother and fetus (Tab. I). These results indicate that the fetal adrenal gland reacts to acidosis, i.e., intrauterine stress, with increased corticosteroid secretion. This rise depends on the pH of fetal blood and not on the type of delivery (Fig. 3).     

Fetal sympatho-adrenal mediated metabolic responses to parturition

Summary. Plasma cortisol and noradrenaline were measured in 15 mothers and babies following vaginal delivery and in 17 following elective caesarean section and related to blood concentration of glucose and glycerol. Significant elevations in umbilical artery cortisol and noradrenaline concentration were found in the vaginal delivery group. Blood glucose and glycerol concentrations were significantly higher in the vaginal delivery group for maternal venous, umbilical artery and vein samples. However, when allowances were made for placental transfer of these substances, little evidence was found to support sympatho-adrenal mediated mobilization of glycerol and glucose in the fetus. The possible explanation for the apparent dissociation between marked sympatho-adrenal activity and metabolic responsiveness during parturition in the fetus is discussed.     

Fetal sympatho-adrenal mediated metabolic responses to parturition

Summary. Plasma cortisol and noradrenaline were measured in 15 mothers and babies following vaginal delivery and in 17 following elective caesarean section and related to blood concentration of glucose and glycerol. Significant elevations in umbilical artery cortisol and noradrenaline concentration were found in the vaginal delivery group. Blood glucose and glycerol concentrations were significantly higher in the vaginal delivery group for maternal venous, umbilical artery and vein samples. However, when allowances were made for placental transfer of these substances, little evidence was found to support sympatho-adrenal mediated mobilization of glycerol and glucose in the fetus. The possible explanation for the apparent dissociation between marked sympatho-adrenal activity and metabolic responsiveness during parturition in the fetus is discussed.     

Fetal sympatho-adrenal mediated metabolic responses to parturition

Plasma cortisol and noradrenaline were measured in 15 mothers and babies following vaginal delivery and in 17 following elective caesarean section and related to blood concentration of glucose and glycerol. Significant elevations in umbilical artery cortisol and noradrenaline concentration were found in the vaginal delivery group. Blood glucose and glycerol concentrations were significantly higher in the vaginal delivery group for maternal venous, umbilical artery and vein samples. However, when allowances were made for placental transfer of these substances, little evidence was found to support sympatho-adrenal mediated mobilization of glycerol and glucose in the fetus. The possible explanation for the apparent dissociation between marked sympatho-adrenal activity and metabolic responsiveness during parturition in the fetus is discussed.     

Lumbar epidural analgesia with bupivacaine in labor. Drug concentration in maternal and neonatal blood at birth and during the first day of life.

Lumbar epidural analgesia with bupivacaine was administered to 33 women for relief of pain during labor. At delivery blood samples were drawn from the umbilical cord vessels and from a maternal peripheral vein. Blood samples were also collected from the noenate and its mother, 1, 4, and 20 hours after delivery. Analyses of bupivaccaine concentrations were carried out in all samples with a gas-chromatographic technique. The drug concentration in the umbilical vein (UV) was usually higher than the corresponding umbilical artery value (UA), but with increasing time interval between the last bupivaccaine infection and delivery the UA/UV drug concentration ratio rose. After delivery the rate of drug decline in blood was similar in mother and newborn infant. The clinical condition of the infant was unrelated to the drug concentration in the umbilical cord vessels.     

Plasma oxytocin in human pregnancy and parturition

Oxytocin concentrations were determined in serial peripheral plasma samples collected from clinically normal women during pregnancy and labor. Measurable concentrations of this hormone were detected in all maternal plasma samples during pregnancy, but there were wide differences in values between patients. Serial samples from individual patients revealed a pattern of gradual rise of oxytocin levels with advancing gestation and the increase in concentration was statistically significant. There were no significant differences in oxytocin levels at any stage of labor, with or without epidural analgesia. Oxytocin levels at the onset of the second stage did not differ statistically from those at crowning. Comparison of cross-sectional data showed no significant difference between the mean oxytocin concentration in early labor and in late pregnancy. Oxytocin surges occurred, but not in a regular pattern. Plasma oxytocin concentration did not increase after pelvic examination, sweeping of the membranes, low amniotomy or after cervical vibration. After spontaneous vaginal delivery, umbilical arterial plasma levels of oxytocin were consistently higher than plasma concentrations from the umbilical vein. The fetal arterio-venous difference was less pronounced at elective cesarean section. At spontaneous vaginal delivery, with and without epidural anesthesia, plasma levels from the umbilical artery were significantly higher than the maternal levels. After vaginal delivery, oxytocin levels in cord plasma were significantly higher than at elective abdominal delivery. Some methodological aspects with regard to blood sampling and to plasma oxytocin radioimmunoassay procedures are discussed. From the results presented it is concluded that the human fetus can be an important source of oxytocin and that neurohumoral birth reflexes described in animals do not occur systematically in man.     

Assessment of maternal-fetal status of some essential trace elements in pregnant women in late gestation: relationship with birth weight and placental weight.

OBJECTIVE: The status of the essential trace elements copper (Cu), iron (Fe), zinc (Zn), selenium (Se) and molybdenum (Mo) has been investigated in maternal and umbilical cord blood in control, uncomplicated pregnancies at term, and the possibility assessed of a relationship between blood levels of these trace elements and newborn weight and placental weight. Fetal-maternal ratios of the elements were also computed to establish baseline values for the Kuwaiti obstetric population. METHODS: Blood samples were collected from a maternal vein, the umbilical artery and umbilical vein of normal pregnant women at the time of spontaneous delivery or Cesarean section, and the concentrations of various trace elements determined by atomic absorption spectrophotometry. RESULTS: The concentration of Cu, Fe, Mo, Se and Zn averaged 2406.1, 3252.1, 11.6, 107.3 and 696.2 microg/l, respectively, in maternal venous blood in the pregnant women (n=39) at term. Umbilical venous/maternal venous ratios of Cu, Fe, Mo, Se and Zn averaged 0.32, 1.96, 1.03, 0.83 and 1.55, respectively. Neonatal birth weight did not correlate with maternal blood levels of the trace elements (p>0.05) in the mother-child pairs studied. However, neonatal weight correlated negatively (p<0.05) with umbilical venous Cu level. Placental weight correlated positively (p<0.05) with Fe and Mo levels and negatively with Zn level in umbilical venous blood. CONCLUSIONS: Our results indicate an active placental transport for Fe and Zn, while Cu, Mo and Se appear to be exchanged passively between mother and fetus. Evaluation of Fe, Mo, Se and Zn levels in maternal and umbilical cord blood does not appear to be useful in the assessment of fetal growth.     

Biochemical changes in the mother and the fetus during labor and its significance for the management of the second stage

In 69 patients with uneventful pregnancies, term labor was studied prospectively with respect to length of second stage, number of bearing down efforts, maternal and fetal levels of lactate, epinephrine and norepinephrine. Maternal venous blood concentrations were measured in early labor and at the time of delivery while samples from umbilical artery and vein provided fetal blood. There was a significant rise of lactate and catecholamines in maternal blood during labor and at delivery fetal lactate concentration was lower than the maternal level while for epinephrine and norepinephrine fetal levels were higher. For all three compounds umbilical artery concentrations were higher than umbilical venous levels. While there was no correlation between the biochemical parameters in maternal blood and length of second stage maternal lactate and norepinephrine concentration at the time of delivery significantly correlated with the number of bearing down efforts. Umbilical artery lactate correlated with both, length of second stage and number of bearing down efforts.     

Contribution of maternal circulation to blood-borne progesterone in the fetus. I. Studies on human subjects

Tritiated progesterone was infused intravenously at a constant rate into three pregnant volunteers in labor, for at least two hours before delivery. Blood samples from a maternal peripheral vein and from unbilical vessels were taken at birth. Comparison of the specific activities of progesterone in these samples indicated that 10 per cent or less of the hormone in fetal circulation is derived from transfer of maternally circulating progesterone. After consideration of reported values of umbilical vein blood flow at term and measured arteriovenous differences in concentrations of progesterone in umbilical vessels, the secretion rate of the placental hormone toward the fetus was estimated to be about 1/10 of the rate of secretion of progesterone toward the maternal circulation. About 1 per cent of the maternally circulating hormone was found to cross the placenta.     

Human corticotropin-releasing hormone during pregnancy

Elevated irCRH levels up to 14 ng/ml were measured in 176 females in the last trimester. The highest maternal CRH levels were found in those females in whom the period from the onset of labour to full dilatation of the cervix and the time span of delivery were shortest. irCRH in amniotic fluid (120 +/- 180 pg/ml; n = 14) was in the same range as in umbilical cord plasma (233 +/- 188 pg/ml; n = 66) and 20-fold lower than in prepartal maternal plasma (5.38 +/- 4.49 ng/ml; n = 66). irCRH in maternal plasma correlated highly to irCRH in umbilical cord plasma (p less than 0.001; n = 66). After delivery irCRH disappeared from maternal plasma with a half-life of 50 minutes (n = 14). One day postpartum irCRH levels (n = 22) were undetectable. The height of the irCRH levels in the various biological fluids did not correlate to the mode or the pathological events of delivery (n = 43). Maternal ACTH levels above the normal range were encountered only in women immediately prepartal and did not correlate to the CRH levels (253 +/- 229 pg/ml; n = 66). Cortisol levels were higher in maternal plasma than in umbilical cord plasma due to elevated CBG (n = 78). Free cortisol levels were higher in the 3rd trimester than in the 1st (2.18 +/- 0.16 vs 1.16 +/- 0.73 ng/ml; n = 42). irCRH in maternal and umbilical cord plasma correlated to the hPl and estriol levels (p less than 0.001 and p less than 0.05; n = 66). We conclude that irCRH is secreted by the placenta into both maternal and fetal circulation. Though placental CRH is undistinguishable from hypothalamic CRH, the biological significance of placental CRH remains open. Our data show that placental CRH might be responsible for the changed function of the adrenal gland during pregnancy, with higher free cortisol levels in the last trimester. The extremely elevated ACTH levels during labour and delivery indicate that CRH is not the only mediator of stress-induced ACTH secretion in the regulation of the maternal hypothalamo-pituitary-adrenal axis.     

Transfer von Immunglobulinen durch die menschliche Plazenta

Immunoglobulin G (IgG) measurements in maternal blood and samples taken from the umbilical artery and umbilical vein at delivery during the third trimester have yielded evidence of transplacental transfer of IgG and its four subclasses. In all three patient subgroups (28–33, 34–36, 37–42 weeks) the concentration in the umbilical vein was consistently higher than that in the umbilical artery. There was an increase in transfer capacity with advancing gestational age during the third trimester. At term the level in the umbilical artery was higher than the maternal IgG, concentration, indicating a transfer against a concentration gradient across the placenta. The comparison of the IgG1 to IgG2 ratio in the maternal blood sample with that in samples from the umbilical artery and vein showed that the ratio was three times as high in the fetal blood, suggesting a preferential transfer of IgG1 over IgG2. In in vitro perfusion experiments in an isolated cotyledon of the human placenta, direct evidence for transfer of IgG was obtained. The in vitro data also suggested a more efficient transport system for IgG1 than for IgG2. In vivo and in vitro, the placental transfer of IgG is a slow process, and following changes in maternal IgG concentration a period of several weeks is required before equilibration with the levels in the fetal circulation is achieved.     

Relation between leptin and cortisol values in umbilical vessels at normal vaginal delivery.

To study the role of various hormones in the control of fetal leptin secretion during labour, 33 pregnant women with normal singleton term pregnancy were recruited. At the time of spontaneous vaginal delivery, a venous blood sample was taken from the women together with a venous and an arterial cord blood sample. In all blood samples, leptin, cortisol, prolactin and progesterone were measured. Serum leptin and cortisol values were significantly higher, while those of prolactin and progesterone were significantly lower in the mother than in the two umbilical vessels (p < 0.01). Cortisol levels were significantly higher in the umbilical artery than in the umbilical vein (p < 0.01). Serum leptin values in the umbilical artery and vein correlated significantly with the corresponding values of cortisol (r = 0.523 and r = 0.580 respectively, p < 0.01), but not with those of prolactin and progesterone. A weak but significant correlation was found between leptin values in the two umbilical vessels and birth weight (r = 0.385 and r = 0.401 respectively, p < 0.05). In multiple regression analysis, cortisol values but not birth weight was the most important determinant of leptin values. Birth weight, however, correlated significantly with placental weight (r = 0.776, p < 0.001). These results demonstrate for the first time that leptin concentrations in the umbilical vessels at normal vaginal delivery correlate significantly with cortisol values, thus providing evidence that cortisol mediates a labour stimulating effect on fetal leptin secretion. It is suggested that cord blood leptin values at delivery are not a good predictor of neonatal weight.     

Adrenocorticotropic hormone in human fetal blood at delivery.

Adrenocorticotropic hormone (ACTH) concentrations in umbilical arterial and venous plasma were measured in newborn infants delivered both vaginally and by cesarean section. Mean ACTH was significantly higher in the arterial (602.0 pg. per milliliter) than venous (261.5 pg. per milliliter) plasma in vaginal deliveries (pless than 0.01). Arterial plasma of babies born by elective cesarean section before initiation of labor contained less ACTH (384.7 pg. per milliliter) when compared to those delivered after labor had started (698.7 pg. per milliliter). Serial determinations of ACTH in pooled fetal blood obtained during labor showed that the average hormone titers increased significantly from first stage (295.2 pg. per milliliter) to second stage of labor (440.5 pg. per milliliter), and reached a peak at delivery (645.4 pg. per milliliter). The pituitary gland of the fetus at midtrimester contained ACTH. Possible mechanisms for fetal secretion of ACTH at delivery are discussed.     

Changes in serum steroid concentrations at delivery and during neonatal period after DHA-S injection into mother before delivery

DHA-S loading has been widely used for the assessment of placental function and ripening of the cervix before delivery. However, changes in the steroidal milieu in both mother and fetus has not been clarified after DHA-S administration. Free (f) and conjugated (c) DHA, delta 5 androstenediol (Adiol), estradiol (E2), estriol (E3), f-testosterone (T) and delta 4 androstenedione (Adione) levels in maternal vein, cord blood and peripheral vein of new born infants after maternal DHA-S administration were measured by radioimmunoassay, and the levels were compared to those in the controls. DHA-S administration resulted in a marked and significant rise in all steroid levels except for E3 in maternal vein. Significant augmentation of T, Adione, c-Adiol, f-DHA, f.c-E2 in umbilical artery, and T, Adione, f.c-Adiol, f-DHA, f-E2 in umbilical vein was observed. Steroid levels during the neonatal period were also measured on the 1st, 3rd and 5th days in both male and female infants, and the levels were compared to those of the controls. Although three androgens in both male and female infants at delivery were significantly higher than the controls, no significant difference was observed on the 5th day of life. Adiol and Adione levels on the 1st day in females, and Adiol levels on the 3rd day in males were higher than those of the controls. E2 levels remained high till the 5th day of life. These results indicated that DHA-S administered to the mother was converted to androgen and estrogen promptly in the placenta and transferred to the fetus.(ABSTRACT TRUNCATED AT 250 WORDS)     

Metabolic and endocrine findings in appropriate and small for gestational age fetuses.

The oxygenation and metabolism in appropriate (AGA) and small for gestational age (SGA) fetuses has been investigated by cordocentesis. The umbilical venous and arterial pO2 and pH decrease with gestational age while pCO2 increases and blood lactate concentration does not change. The mean umbilical venous blood glucose concentration is higher than in the umbilical artery indicating that there is fetal glucose uptake from the placenta. Similarly, the maternal glucose concentration is higher than the fetal and the levels in the two compartments are significantly correlated. The plasma insulin concentration increases exponentially with gestation reflecting the progressive maturation of the fetal pancreas. The fetal plasma cortisol does not change but the fetal plasma ACTH increases with gestation. Fetal plasma triglyceride concentration decreases exponentially with gestation and this is likely to be the result of increased utilization by the fetus for deposition into adipose tissue. There is a high correlation between fetal and maternal levels for individual amino acids and the concentration in the fetus is higher than in the mother, supporting the active transport of amino acids by the placenta. Some SGA fetuses are compromised by hypoxemia, hypercapnia, hyperlacticemia and acidosis, are starved of glucose and amino acids, and are hypertriglyceridemic. Furthermore, some of these fetuses are hypoinsulinemic and the degree of hypoinsulinemia is disproportional to the degree of hypoglycemia suggesting pancreatic dysfunction. In SGA fetuses the plasma cortisol is increased and the plasma ACTH decreased. Knowledge of human fetal oxygenation and metabolism may help in deciding the optional timing of delivery but may also constitute a basis for future fetal therapy in the form of oxygen and nutrient supplementation.     

Fetal adrenocorticotropic hormone and prolactin at delivery

The concentrations of adrenocorticotropic hormone (ACTH) and prolactin in fetal scalp and umbilical cord blood were measured in 49 full-term fetuses, 25 of whom were delivered spontaneously, 10 of whom underwent induced vaginal delivery, and 14 of whom underwent elective cesarean section. Serial fetal scalp blood samples taken during the 35 vaginal deliveries showed no change in concentration of these hormones. The mean (+/- SD) plasma level of ACTH in cord artery blood after vaginal delivery (315 +/- 249 ng/liter) was significantly higher (P less than .05) than after cesarean section (184 +/- 201 ng/liter), indicating that the fetal anterior pituitary is capable of responding to partal stress.