Symposium
Chair: Janet A. DiúPietro
Discussant: Eduard J.H. Mulder
'The existence of a profusion of myths and superstitions has probably inhibited until modern timesscientific thought and investigation into maternal-fetal relationships...' (Sontag, 1941). At the turn of this century, investigators have just begun to graze the surface of the complexand dynamic interface between the pregnant woman and her fetus. The influence of maternalpsychologic function and social interaction on domains of infant development and well-being havelong been subjects of scrutiny. However, the maternal influences exerted prior to birth have thepotential to be more profound and pervasive because the intrauterine milieu is subject directly toneuroendocrine influences of maternal emotions. This non-genetic 'maternal womb environment'has been recognized to account for a significant proportion of variance of someconstitutionally-based characteristics, such as intelligence. Other potential sources of effects arelikely to involvealterations to the homeostatic 'set-point' of neuroregulatory systems of the developing fetalbrain, notably the hypothalamic-pituitary-adrenal axis. The most convincing, although stilllimited, body of evidence to date involves effects of maternally perceived psychological stress onlater development of offspring. However, little is currently known about how alterations inmaternal physiology directly affect fetal neurobehavior during either conditions of low or highmaternal arousal. This link is critical in understanding the mechanism by which chronic orepisodic stress alters postnatal outcomes. Such understanding also has potential clinicalapplication in developing interventions to ameliorate deleterious antenatal conditions. The focus of this symposium is on the role of maternal emotions, particularly anxiety andperceived psychological stress, on both fetal functioning and subsequent infant development, aswell as the mechanisms which may mediate these effects. The panel is international in scope. Thefirst three presentations are from among the few research programs which currently study the roleof contemporaneous maternal emotions and psychophysiology on the fetus. The first will presentdata temporally linking a variety of dimensions of maternal functioning (e.g., heart rate) withmeasures of fetal functioning (e.g., motor activity) under both baseline and maternal arousalconditions. The second presentation will describe how experimentally manipulated maternalstress affects fetal functioning, and explore the moderating role of maternal personalitycharacteristics. The next presentation describes studies of the inverse manipulation: the effects ofdirected maternal relaxation on uterine blood flow, another indicator of the functioning of thefetoplacental unit. Finally, the results of one of the largest on-going and comprehensiveinvestigations of the associations between prenatal maternal stress and individual differences indevelopment during the first year of life will be presented. Discussion of the role of maternalfactors in directing fetal development will be presented by a discussant with a distinguishedhistory in documenting fetal neurobehavioral development.
Details of individual items:
paper
The effect of maternal emotional state on the fetus has long been a source of speculationand anecdote, dating back to literary references during biblical times. An emerging body ofliterature indicates that maternal appraisal of stress during pregnancy has deleteriousconsequences for pregnancy outcome and infant development. However, few empirical data existon the relation between contemporaneous maternal state and fetal functioning. If the maternalpsychologic milieu has long-term sequelae, such effects should be evident during the fetal periodwhen the physiologic mechanisms mediating these processes are operational. In two previousstudies, we have reported that recent stress appraisal was associated with a) reductions in bothfetal heart rate variability and the degree of coupling between fetal movement and heart rate, anindicator of neural integrity, and b) increases in fetal movement. However, self-reported measuresof stress or state are confounded by a variety of maternal attributes, including personalitycharacteristics, thus introducing significant noise into clarifying the relation between maternal andfetal function. In order to truly elucidate the role of maternal psychophysiologic functioning onthe fetus, such functioning must be directly measured. To this end, we have developed a computerized data collection system in which fetal (fetalheart rate and movement) and maternal (ECG, skin conductance and resistance, and respiration)measures are collected simultaneously. Data processing yields a variety of measures, which aretime synchronized between mother and fetus. In the current longitudinal study, 50 minutes ofundisturbed maternal-fetal data are collected on 120 maternal-fetal pairs at 6 points in gestation,beginning in the 20th week. Maternal self-reported data on a range of personality and perceivedstress dimensions are collected at each visit. Following the baseline recording, maternal arousal ismanipulated by exposure to one of three psychological challenges. This presentation will detail results of multivariate time series analyses which are used toempirically detect the nature of the serial dependency of the second by second relations betweenthe mother and fetus. Cross correlations are computed between pairs of indicators of function(e.g., maternal heart rate and fetal heart rate; maternal skin conductance response and fetalmovement). A single correlation estimate is obtained for each distinct value of the timeseparation, or lag, which represents the maximum correlation coefficient (r) betweenmaternal-fetal measures. Gestational age specific lag times, in seconds, are also computed. Wehypothesize that the a) existence of a maternal-fetal temporal relation will be conclusivelyestablished, such that increased maternal arousal will be associated with increased fetal heart rateand movement but reduced variability; b) the temporal relationship will increase over gestation,and c) there will be disassociation (reduced r's/increased lag times) during periods of inducedstress compared to baseline. At the time of conference submission, data collection on the 120subjects is nearly complete; final data analysis will ensue upon completion. These data willprovide the most comprehensive description of the association between maternal-fetal functioningto date, and provide a basis for understanding the effects of maternal emotions on fetal and infantdevelopment.
paper
The effect of pregnant women's emotions on the fetus has long been a source ofspeculation. Recent studies suggest that stress during pregnancy is associated with preterm birthand lower birthweight, and other work indicates that maternal anxiety is related to more activeand irritable newborn behavior. However, few studies have directly measured the impact ofwomen's emotions on the fetus. In two separate studies, we used concurrent maternal and fetalphysiology monitoring while exposing the mother to an acute stressor to examine the influence of maternal stress and anxiety on fetal heart rate (FHR) and to explore possible pathways by whichmaternal emotions affect the fetus. In the first study, 17 women in the third trimester of pregnancy completed Spielberger'sState-Trait Anxiety Inventory and then underwent a 5-minute baseline period and performedeither a 5-minute arithmetic or Stroop color-word test while continuously monitored for EKG,blood pressure (BP), respiration rate and FHR. The stressors led to significant increases inmaternal systolic BP and respiratory rate but changes in maternal HR, diastolic BP, and FHRwere not significant. However, when subjects were dichotomized into groups that had above(ANX(+)) or below (ANX(-)) average anxiety scores, both groups had similar respiration rateincreases to the stressors, but the BP and FHR responses were significantly different. Womenin the ANX(-) versus women in the ANX(+) had significantly greater BP responses while thefetuses of ANX (+) showed significant HR increases and the fetuses of women ANX(-) exhibitednon-significant decreases. These findings suggest that women's emotional reactivity duringpregnancy can influence fetal HR patterns and that a stress-induced increase in maternalBP is not the primary signal by which a women's stress response is transduced to her fetus. Theyraise the intriguing possibility that fetuses of ANX (+) compared to ANX (-) women may showhyper-reactive responsiveness to the same maternal input, i.e., respiratory rate increase. These findings were replicated in a larger study of 40 third-trimester pregnant women withone important difference. For this second study, subjects completed both Stroop and arithmetictasks. In the context of two tasks, a difference in the tasks' effectiveness as stressors emerged.During Stroop, both BP and respiratory rate increased while during arithmetic only respiratoryrate increased. When subjects were divided into above and below average anxiety groups, fetalHR increase was significantly greater in the ANX (+) group during the Stroop but notarithmetic task. There were no simple associations between the magnitude of maternal and fetalstress reactivity. Similar to the first study, these data suggest that by the third trimester individualdifferences in FHR reactivity are associated with maternal anxiety status and not directlyrelated to the magnitude of coincident maternal cardiovascular responses. In order to observe theputative cumulative effects of different prenatal environments and genetics on FHR reactivity,subjects must experience a significant stressor. Taken together, these findings point to thepotential for maternal mood to mediate the effects of stress on fetal behavior and development.
paper
Anxiety and stress in pregnancy have been linked in most but not all studies with early deliveryand small for gestational age babies (SGA). We have shown in a study of 136 mothers at 30weeks that State anxiety was associated with SGA babies. Fifteen of 45 babies born to mothersscoring >40 on the Spielberger had babies below the 10th percentile, adjusted for sex andgestational age, compared with 11/91 in the non anxious group (X2 8.8;p0.003).Little is known about the mechanisms by which the antenatal emotional state of the mother mayaffect outcome. In a previous study we found that women who are anxious at 32 weeks gestationhave abnormal blood flow patterns in the uterine arteries, with increased resistance index (RI)indicative of impaired blood flow. We have recently confirmed these findings in a further study.This could be one mechanism by which maternal antenatal mood leads to adverse outcome,including smaller babies. In a new study we have investigated the effect of directed relaxation on uterine blood flow.Fifty one women were studied at mean of 32 weeks gestation; 29 had a 30 min session of directedrelaxation and 22 controls sat quietly for the same period. Uterine blood flow was measured bycolour Doppler ultrasound before and after each session. There was a significant difference inchange in RI in the two groups (p<0.02), with the directed relaxation group showing better bloodflow. This was not linked with a change in plasma cortisol which fell equally in both groups. These results suggest that uterine blood flow can be sensitive acutely to the emotional state ofthe mother, and may be improved by directed relaxation. Whether the degree of change that weobserved would be of clinical benefit will need to be determined in future studies.
paper
Maternal stress during pregnancy has been described to reduce birth weight, gestationalage at birth and head circumference in humans. Furthermore, animal studies have showndeleterious effects of stress during pregnancy on behavior and development of the offspring. Forinstance, stress in pregnant rhesus macaques has been found to be associated with abnormalmotor development in the offspring. In addition, in animals, prenatal stress was related tobehavioral changes in the offspring, such as less exploration of a novel environment and increasedstress responses. In human retrospective studies, it has been found that stress during pregnancywas associated with more difficult behavior at childhood, while others have found increasedincidence rates of schizophrenia at adult age. Thus, prenatal stress could have an effect on temperament, behavior and development of the child. To overcome methodological problems wehave used a a prospective longitudinal design and have studied the effects of prenatal stressduring early, mid and late pregnancy on infant motor and mental development at the age of 3 and8 months (Bayley Scales of Infant Development). Also, infant temperament was examined bymeans of questionnaires, by the examiner and psychophyiological reactivity (e.g., ICQ, Bates;vagal tone). Prenatal stress was carefully described in a multidimensional model and consisted oflife events, daily hassles, coping style, social support, pregnancy related anxiety and perceivedstress. By means of a LISREL path analyses model, the concept prenatal stress was formulatedand used as an independent variable in the analysis. In addition, maternal levels of cortisol werecollected during pregnancy and were used as a separate independent variable in the analysis. Temperament characteristics of the child as rated by the parent and the examiner,the vagal tonemeasures, and PDI and MDI were the dependent variables. Assocations between maternal stesslevels and infant outcome were tested with multivariate statistical analysis Our results of a first sample of 99 women and their children indicate significant main effectsof the amount of daily hassles during mid pregnancy, after correction for possible confounders,on motor development (F 11.53, p < .001). Daily Hassles were associated with acceleratedmotor development. Perceived stress of the pregnant women during early pregnancy wasassociated with increased difficult behavior of the child (F 11.55, p < .001) and maladaptation(F 8.84, p < .005). The final sample for this study will include 180 women and their children.Final results relating pregnancy stress and child development and behavior will be based onanalysis of the full sample.