Fetal cardiac and motor responses to octave-band ... - Science Direct

11 downloads 0 Views 924KB Size Report
noises as a function of central frequency, intensity and heart rate variability. J-P. ... deceleration; motor response; heart rate variability pattern. Introduction.
81

Early Human Development, 18 (1988) 81-93 Elsevier Scientific Publishers Ireland Ltd. EHD 00915

Fetal cardiac and motor responses to octave-band noises as a function of central frequency, intensity and heart rate variability J-P. Lecanueta, C. Granier-Deferre” and M-C. Busnelb aLaboratoire de Psycho-Biologie de I’Enfant* CNRS, UA315 41, rue Gay-Lussac, Paris 75005, and bGroupe G&nPtique,Neurogt%tique et Comportements, UER Biom6dicale. Universitk PARIS V, 45, rue des Saints-P&es, Paris 75006, France Accepted for publication 18 January 1988

Summary Accelerative and decelerative cardiac responses and motor responses (leg movements) of 37-40 weeks (G.A.) fetuses are analyzed as a function of the frequency of three octave-band noises (respectively centered at 500 Hz, 2000 Hz and 5000 Hz) and of their intensity level (100, 105, 110 dB SPL, ex utero), during high (HV) and low (LV) heart rate (HR) variability pattern states. In both states, increasing the frequency and/or the intensity of the acoustic stimulation: (i) increases the ratios and amplitudes of accelerations, and the motor response ratios, (ii) reduces deceleration ratios and motor response latencies. Cardiac and motor reactiveness are higher in HV than in LV with acceleration ratios always greater than motor ones. However, when a high intensity and/or frequency is used, the reactiveness differences between states disappears. Low intensity and/or frequency stimulation levels induce a majority of decelerations. fetus; acoustical stimulation; intensity; frequency; cardiac acceleration; deceleration; motor response; heart rate variability pattern

cardiac

Introduction Since 1925 [33], human fetal reactiveness to sound has been investigated with loud external stimulations (see Ref. 8). They induce mostly accelerative (or biphasic) heart rate (HR) modifications and fast motor responses, components of the acoustic startle response. 0 1988 Elsevier Scientific Publishers Ireland Ltd. 0378-3782/88/$03.50 Published and Printed in Ireland

82

From 22 to 24 weeks G.A. onwards, the occurrence of cardiac and motor responses [7,23] and the amplitude of HR acceleration [21,40] increase with gestational age. During the last gestational weeks, acoustic reactiveness depends on behavioural state at the time of the stimulation. Nijhuis et al. [31] have shown that four different states can be clearly defined at 36 weeks. When states are precisely evaluated with all motor and cardiac variables [38], or only estimated from the HR pattern [24], reactiveness is higher in “active” than in “quiet” sleep, it seems the greatest in the short and rare quiet wakefulness episodes [38]. Reactiveness is also controlled by stimulus characteristics. When intensity rises, motor and accelerative cardiac responses are more likely to occur and the mean amplitude of accelerations is larger [18,22,24,36]. A few studies, with contradictory results, have also suggested that responsiveness might be dependent on stimulus frequency as well [10,15,30,43]. In line with our previous work [24], a comparative study was performed to analyze the interactive effects of: (i) the frequency and (ii) the intensity of the sound stimulus when given during two different HR Patterns (HRP) fetal states: high variability (HV) and low variability (LV). Fetal cardiac and motor responses to “low”, “medium” and “mid-high” frequency octave-band noises emitted at three different intensity levels were recorded. The lowest intensity (100 dB SPL) was lower than most levels used in other fetal studies, because we hypothetized that, if poor reports of decelerations had been found in the fetus [10,16,19,34] it might be due to the characteristics of the sound stimulus - particularly to the high intensities which were used. Method The experiment was performed with the same apparatus and data treatment procedure previously reported where all technical details not given here may be found [24]. Subjects One hundred and seventy-four healthy volunteer women; between the 37th and 40th week of pregnancy, from the Birth Training Program of the Clinique Universitaire Baudelocque (Paris, France) participated in this study. Acoustical stimuli Nine stimulation conditions were studied with three octave-band noises (duration: 5 s.) centered respectively on 500 Hz, 2000 Hz and 5000 Hz, and emitted at either 100, 105 or 110 dB SPL ( f 2 dB SPL, Ref. 20 PPa) measured 1 cm over the maternal abdomen. Apparatus Fetal heart beats were continuously monitored with a wide-range Doppler ultrasound transducer (Hewlett-Packard 8030). They were transmitted to a computer (Apple IIe) which processed and stored beat-to-beat intervals.

83

Fetal tibia and fibula were observed with a linear array real-time ultrasound scanner (CGR-USL 500) and the images were videotaped (Orion VHS) with a frameincorporated digital timer (Sony VTG 22) triggered by the acoustic stimulus onset. Procedure The loudspeaker was located 20 cm above the abdomen of the woman who was listening to a musical sequence through earphones to prevent her perception of the stimulus. In order to allow posthoc assessment of HR variability pattern (state determination: LV or HV) HR recording began with a 10 min baseline period without any stimulation. Every fetus received the three frequencies each of them emitted at a different intensity. Conditions were counterbalanced over subjects. A minimum interstimulus interval of 5 min was kept. It could be extended to 15 min if the HRP changed, either spontaneously or as a possible consequence of the stimulation. This procedure was designed to prevent potential habituation or sensitization effects. When HR tracing was stable and no gross fetal movement could be seen prestimulus period was started (computer storage) and a stimulation delivered 60 s. later. Data Treatment (Z) Heart rate pattern states (HRP): Low variability (LV) - corresponding to FHRP A (state 1F) - and high variability (HV) - corresponding to FHRP B (state 2F)- heart rate patterns [31] were visually identified by three independent observers. Testing periods were kept for analysis only when the three evaluations matched. (2) Cardiac responses: HR prestimulus and stimulus files kept for analysis and with no sustained signal losses were filtered to remove Doppler artifacts and converted into beats per min. Average (prestimulus level) and standard deviation were computed for each prestimulus file. Were only considered as responses HR accelerations or decelerations which amplitude reached a value greater than the prestimulus level standard deviation before the end of the stimulus presentation. The amplitude of each response was computed as the difference between the value of the closest maximum HR deviation read after the onset of the stimulus and the prestimulus level. Apparent responses which amplitudes were equal to or less than 5 bpm were not considered. Accelerative and decelerative response ratios (AR and DR) and mean response amplitude were computed for every subgroup. These ratios were determined out of the total number of trials kept for analysis in each subgroup. (3) Motor responses: A displacement of the lower leg was considered to be a motor response if it occurred before the end of the acoustic stimulation (< 5 s.). Motor response ratios (MR) and latencies were computed for every subgroup. MR were determined out of the total number of trials (with clearly readable video tapes) in each subgroup. (4) Statistical analysis: Cardiac and motor responses were analyzed in 18 subgroups: nine stimulation conditions x two HRP states (HV and LV), each subgroup included 45-56 readable trials. Data consisting of frequency counts were analyzed with x2 tests. Cardiac response

84

mean amplitudes were compared with one or two-way analysis of variance and, because of the asymmetrical distribution of the data, median motor response latenties with a Mann-Whitney U-test. For these two variables, statistical comparisons were run only between subgroups where a sufficient number of responses had been elicited, and in some cases, analysis of intensity and frequency effects were run on pooled data. Statistically significant differences which do not appear in the text are presented below the figures. Results (1) Cardiac responses ratios: (Fig. I) Accelerative response ratios (ARs) are higher than decelerative response ratios (DRs) in 819 stimulation conditions in HV and only 4/9 in LV. In both states, when the intensity and/or the frequency are enhanced ARs increase while DRs decrease. Accelerative ratios: In all stimulation conditions, ARs are higher in HV than in LV, but the difference becomes weaker as intensity and frequency rise from low to high levels (AR is 28% greater in HV than in LV for a 500 Hz/100 dB stimulus and only 8% higher for a 5000 Hz/l 10 dB stimulus). In fact, when intensity and/or frequency rise ARs show different increments according to state: in LV, ARs rise from Oqo at 500 Hz/100 dB to 75% at 5000 Hz/l 10 dB, while, in HV, they rise from 28% to 83%. Decelerative ratios: In both states, DRs are moderate (from 14% to 32% in LV and from 8% to 36% in HV); highest ones are found at the lowest levels of stimulation. They show a small decrease, when either intensity and/or frequency are enhanced. This decrease is greater in HV than in LV. In LV, DRs are equal to or higher than ARs at 100 dB for the three frequencies tested and at all intensities for the 500-Hz stimulus; moreover, only decelerations are recorded at 500 Hz/100 dB. In HV, this is also the condition under which DRs are greater, although non-significantly, than ARs. (2) Motor responses ratios: (Fig. 2) In both states MRs, like ARs, increase with the intensity and the frequency and are greater in HV than in LV. MRs are always weaker than ARs, significantly in only two conditions in LV (100 dB/500 Hz and 110 dB/2000 Hz), but in six conditions - intermediate levels of stimulation - in HV (105 and 110 dB at 500 Hz, for the three intensities at 2000 Hz and 100 dB at 5000 HZ); at higher levels, in this state, accelerations are mostly accompanied by a motor response and at low levels both types of responses are poor. (3) Mean cardiac responses amplitude: (Table I) Accelerative cardiac responses: At 2000 Hz and 5000 Hz (sample size uneveness at 500 Hz did not allow statistical comparison) acceleration amplitudes are modulated by state: in LV they are significantly lower than in HV. They are also modulated by intensity but only in LV where they gradually rise with intensity; in HV amplitudes are high whatever the intensity tested. The interactive effect of state and intensity

AR

DR

5000

100

HZ

IlOci

105

Fig. 1. Accelerative (AR) and decelerative (DR) cardiac response ratios (o/o) per stimulation condition during high (I-IV) and low variability (Lv) heart rate pattern states. Subgroups Comparisons ($ tests):

(a) Accelerative mponses LV

HV

NS 500 Hz/5000 Hz, P = 0.01 500 Hz/2000 Hz, P = 0.02 500 Hz/5000 Hz, P< 0.001

NS NS 500 Hz/2000 Hz, P = 0.05 500 Hz/5000 Hz, P = 0.05

Frequency effect 100 dB 105 dB 1lOdB

Intensity effect 500 Hz

NS lOOdB/lOS dB, P = 0.05 1 1OOdB/110dB, P< 0.001 lOOdB/105 dB, P = 0.05

2OOOHz 5000 Hz

(lOOdB/llOdB,P = 0.10) lOOdB/llOdB, P = 0.02 1 105 dB/llOdB, P = 0.05

I 100 dB/llO dB, P = 0.01

100 dB/llO dB, P = 0.05

HRPState effect (HV/LV): 500 Hz, 105 dB, P = 0.05; 500 Hz, 110 dB, P = 0.05; 2000 Hz, 100 dB, P = 0.01.

(b) hceierative responses: no significant difference (c)Accelerative responses/Decelerative responses: in HV

500 Hz

2000 Hz

5000 Hz

100 dB

NS

NS

NS

105 dB

NS

P = 0.02

P = 0.02

1lOdB

P = 0.05

P = 0.001

P = 0.001

--

100

105

5000 Hz

110dB

Fii.2. Motor response ratios (%) per stimulation condition during high (HV) and low variability (LV) heart rate pattern states. Subgroups comparisons ($ tests):

Frequencydf=t 100 dB 105 dB 1lOdB

LV

HV

NS 500 Hz/5000 Hz, P = 0.02 500 Hz/5ooO Hz, P = 0.01 2ooO Hz/5000 Hz, P = 0.05

NS

500 Hz/5000 Hz, P = 0.05 500 Hz/2000 Hz, P = 0.01 500 Hz/5ooO Hz, P< 0.001

Intensity qfet 500 Hz 2OCKtHz 5OoOHz

NS NS 1 100 dBAO5 dB, P = 0.05 100 dB/llO dB, P = 0.01

HRPStute(HV/LV):2OOOHz,

NS lOOdB/llOdB, 105 dB/llO dB, 1 lOOdB/llOdB, 105 dB/llO dB,

P< 0.001 P = 0.05 P = 0.01 P = 0.05

llOdB, P = 0.01; 5OOOHz. 100dB. P = 0.05

(significant at 5000 Hz, only a trend at 2000 Hz) may be due to the fact that increasing the intensity enhances the ratio of concomitant leg movements which, in turn, increase the amplitude of the acceleration. This effect is more salient in LV than in HV because in LV low-intensity or low-frequency levels induce no or poor motor responses. The somato-cardiac effect of leg movements is demonstrated in both states when considering pooled data for all subgroups (Fig. 3): amplitudes of accelerations with movement are greater than the ones without; and they are not dependent on state, only accelerations without movement are smaller in LV than in HV. Similar results are obtained when an analysis is conducted on the only subgroup where the number of accelerations with and without movement is even (2000 Hz/l 10 dB): amplitude of accelerations without movement is weaker (9.7 bpm in LV and 15.7 bpm in HV) than the ones accompanied by a leg response (23.4 bpm in LV and 20.5 bpm in HV) in both states (F(l,31) = 9.9; P = 0.093). State has a modulatory

81

1OOdB

105 dB

1lOdB

HV

HV

HV

LV

2amHz n: SIandS.E.:

9 17.8 f 1.6

2 -

14 19.4 f 2.2

5 12.6 f 2.7

28 19.1 f 1.5

11 15.9 f 2.9

5oooHzx n: I and S.E.:

16 18.9 f 1.3

6 8.7 f 1.4

15 17.7 f 1.4

9 15 f 1.8

19 22.5 f 2.1

12 21.7 f 1.5

Anova at 5ooOHz F(1.71) = 6.1; P = 0.02 F(2.71) = 8.3;P< 0.001 F(1.2) = 3.2; P = 0.05

- HRP state effect (HV/LV): -1ntcuaity(100dB/105dB/110dB): - Interaction (Intc!naity/statc):

Anowat2tWHz F(1.54) = 3.6; P = 0.05 NS NS

effect only on accelerations without movement: amplitude is higher in HV (15.7 bpm) than in LV (9.7 bpm) (P < 0.08). The effect of frequency on accelerations amplitude is weak and seems also dependent on state: between 2000 Hz and 5OCKI He there is no amplitude difference in HV at all intensitiestested; on the other hand, the difference is statisticallysignificant in LV at 110 dB (P < 0.04) and may also reflect the impact of the higher proportion of leg movements at 5000 Hz.

bpm

0 P1.02 00 Pt.002

MR

I

HV

0

L I

00

1

I

Fii. 3. Man amplitudes of accelerative raponaea in LV and in HV heart rate pattern 8mte11 when accompa&d by a motor response (MR) or without any motor response (WMR), all condltiotw pooled.

88

These results suggest that acceleration amplitude is dependent on: fatal state, intensity and frequency levels of the stimulation: (i) through a linear modulation when acceleration is the only response to the stimulation, and (ii) through a general enhancement of motor reactiveness which is known to induce accelerations and/or increase their amplitude. Decelerative cardiac responses: Pooled data show no amplitude difference between decelerations recorded in LV (- 10.4 bpm) and those recorded in HV (- 10.2 bpm). These values are comparable to the acceleration amplitudes without movement in LV (11 bpm) and in HV (13 bpm); this is about half the values of the acceleration amplitudes with movements recorded either in LV or in HV (Table I). In both states there were neither intensity nor frequency effect between the subgroups in which decelerations were mostly recorded: 500-Hz subgroups, lOO-dB subgroups. (4) Motor response median latencies In HV, the only state where sufficient numbers of leg responses were recorded to allow statistical comparisons between the subgroups, median latencies decrease with a IO-dB intensity rise at 5000 Hz (1.3 s. at 100 dB; 0.73 s. at 110 dB; P = 0.02). They also decrease with frequency increase at 110 dB (1.1 s. at 2000 Hz; 0.73 s. at 5000 Hz; P = 0.04). No latency difference has been found between states in the conditions where enough MRs were recorded in both states (5000 Hz at 105 and 110 dB). Thus, analysis of motor responses shows that intense and mid-high frequency band noises induce more leg movements and reduce their latency compared to less intense and/or lower frequency band noises. Discussion

This experiment extends previous results [24] obtained with a high-pass filtered (> 800 Hz) pink noise on fetuses of the same gestational age (see (I) to (3) below), and reveals two new findings: (i) fetal cardiac and motor reactiveness is modulated by the acoustic frequency of the stimulus; (ii) fetuses can show decelerative cardiac responses when stimulated with appropriate auditory stimulations (see (4) and (5) below). (1) Effect of intensity and HRP state on response ratios As with the pink noise [24], increasing the intensity of the octave band noises increases both cardiac acceleration and leg movement response ratios; furthermore, accelerative responses are more easily elicited than leg movements. Response ratios recorded here are within the range of those reported in other prenatal experiments with pure tones (500 Hz--4000 Hz) of similar intensities [10,12,19,30,43]. Fetal responsiveness, in particular the motor one, is significantly higher during a high variability HRP than during a low variability one; this is in agreement with other prenatal experiments using different procedures [16,38&t] and can be related to neonatal studies [2,20,28,35].

89

(2) Combined effects of intensity and state on response ratios and mean cardiac responses amplitudes Our data show that when intensities are high responsiveness is important and does not depend on fetal state; this can also be drawn from experiments using over 1IO-dB vibro-acoustic stimulations directly applied on the mothers’ abdomen, therefore with slight intensity attenuation [7,13,14,23,32]. But when lower intensities are used responsiveness is dependent on state. Accelerative cardiac response amplitudes are higher in HV than in LV and an interactive effect was demonstrated between state and intensity: in LV amplitude rises with intensity, not in HV where even the lowest intensity (100 dB) induces high amplitude accelerations. Mean amplitude is also dependent on the presence of a concomitant leg movement: strong motor responses - like spontaneous movements [1,26] - induce important heart rate accelerations. Thus, exposing fetuses to sound in a highly reactive state (HV) and/or using high intensities, increases the probability of eliciting a movement which, in turn, enhances the amplitude of the accelerative response. (3) Leg movement latencies They are close to the global motor response latencies recorded in the neonate (42, 48,49) and are modulated in the same way by the intensity (42): in HV they shorten when increasing intensity from 100 dB to 110 dB. (4) Effect of frequency Cardiac accelerations are more frequent and their amplitude greater, motor reactiveness is higher and its latency shorter when the central frequency of the stimulation rises. Frequency exerts the same modulatory effect as intensity on all dependent variables. The data also show that decelerative response ratios might be dependent on the frequency of the stimulation (see (5) below). The effect of increasing the frequency level seems independent of the effect of increasing the intensity: a 5-dB enhancement has the same effect on both accelerative and decelerative ratios for the 200-Hz and for the NO-Hz stimulus, there is a rough curvilinear additive effect between the two factors, Different frequency pure tone stimulations were used in other prenatal experiments [10,15,30,43], but it is difficult to compare our results with those since fetal states were not controlled, age groups and characteristics of the stimulations were also different. The direction of this frequency effect, i.e. the stronger reactivity to the mid-frequency band noises (2000 Hz and 5000 Hz), can be found rather puzzling considering that analysis of intra-amniotic recordings of aerial externally generated sound stimulations [3,36,37,47] have all evaluated that the higher the frequency of the stimulus, the higher its attenuation. The attenuation of our 2000-Hz and 5000Hz stimulus can be estimated to be greater (between 20 and 30 dB) than that of our 500 Hz (15 dB) [36,37]. Four factors may nevertheless explain our results. First, the 2000-Hz and 5000-Hz noises probably have a better emergence than the 500 Hz from the intra-uterine

90

noise, which is mostly composed of frequencies between a few Hz and 700 Hz [36,37]; thus, the 500-Hz stimulus could be partly masked. Second, it can be hypothetized that fetuses are less reactive to the 500-Hz stimulation because it is part of their continuous back ground noise. Third, and fourth, our data also suggest that near term fetuses, like newborns and young infants, (i) may have a lower auditory absolute threshold to the mid-high frequencies octave-band noises than to the low frequency one [27,39,41,45] and/or (ii) may display poor overt responses to the 500Hz noise [48,29]. (5) Cardiac deceierative responses This experiment is the first prenatal study to demonstrate that the nature of the cardiac response varies with the characteristics of the acoustic stimulation: decelerative response ratios lessen when either intensity and/or frequency of the stimulation are enhanced while accelerative response ratios increase. A majority of decelerative responses were recorded in LV for all frequencies at the lowest intensity tested (100 dB) and for all intensities at the lowest frequency tested (500 Hz): they represented 50%-80% of all cardiac responses and 100% at 500 Hz/100 dB. In HV, recorded decelerative responses only represented lo%56% of all cardiac responses. Thus, there is a majority of decelerations in LV, the least reactive state, to the low intensity/frequency stimulus. In the newborn, decelerations are usually elicited during quiet wakefulness [17] but they also occur during quiet sleep, especially when stimulated with speech sounds [l 11. It can also be noted that, in agreement with neonatal data [17,29], a deceleration is never accompanied by a leg movement and its amplitude, lower (from 6 to 18 bpm) than that of the acceleration, is within the range of the values reported in newborn and fetal studies [10,11,16,17]. Thus, our results suggest an effect of the central frequency of the noise: the 500-Hz noise induced the least movements and the more decelerations in the fetus. This can be related to the newborn reactiveness which includes less movements and more HR decelerations to low frequencies than to medium or high ones 15,29&l. We recently demonstrated 1251that, at a lower intensity than the ones classically used, another acoustic stimulus (speech syllables at 95 dB SPL ex utero) elicited only cardiac decelerations in near term fetuses in LV. This is similar to the neonates response to speech stimuli [9,11]. Fetal decelerations in response to sound seem therefore dependent on a variety of features of the stimulus. Human and animal post natal studies have demonstrated that cardiac decelerations can be induced, at low levels, with continuous noises, complex rhythmic acoustic stimuli as well as species-specific sounds [6,9,11,17,25,46]. It still remains to be shown that similar mechanisms underly these decelerations and the ones evidenced in this study; can the latter also be seen as orienting responses in the context of psychophysiological continuity between late fetal and early postnatal life emphasized by the results of this study? Acknowledgements

This research has been partially supported by two grants: CNAMTS (No. 2166)

91

and an INSERM (No. 130006). We are deeply grateful to the mothers and to the babies involved in this study. We thank Pr C. Sureau, head of the Clinique Universitaire Baudelocque, Mrs C. Contrepas, midwife, Mr G. Alcuri, A-J. Andrieu, R Le Houezec, acoustical engineers, Drs F. Herve, J. Poli, P. Dubois, L. Pa&au; M. Tostain and A. Rodrigues for their valuable help during the course of the experiment, and J . Bertoncini for her helpful comments on the manuscript. References 1 2 3 4

5 6 7 8 9 10 11 12 13

14 15 16 17 18 19

20

Aladjem, !I.. Rest, J. and Stojanovic, J. (1977): Fetal heart rate responses to fetal movements. Br. J. Obstet. Gynecol., 84,487491. Ashton, R. (1971): State and the auditory reactivity of the human neonate. J. Exp. Child. Psychol., 12.339-346. Bench, R.J. (1968): Sound transmission to the human fetus through the maternal abdominal wall. J. Genet. Psychol., 113,1172-1174. Bench. R.J. and Mentz. D.L. (1975): On the measurement of human fetal auditory responses, In: Sound Reception in Mammals, pp. 23-40. Editors: R.J. Bench, A. Pye, J.D. Pye. Academic Press. New York. Berg, W.K. and Smith, M. (1983): Behavioural thresholds for tones during infancy. J. Exp. Child Psychol.. 35409-425. Bemtson. G.G. and Boysen, S.T. (1984): Cardiac startle and orienting responses in the great apes. Behav. Neurosci., 98.914-918. Bimholz, J.C. and Benacerraf, B.B. (1983): The development of the human fetal hearing. Science, 222,516-518. Busnel. M-C. and Granier-Deferre, C. (1983): And what of fetal audition? In: The Behavior of Human Infants. Editors: A. Oliveirio and M. Zappella. New York Life Science toll., Plenum Press. Clarkson, M.G. and Berg, W.K. (1983): Cardiac orienting and wove1 discrimination in newborns: crucial stimulus parameters. Child Dev., 54,162-171. Dwomicka, B., Jasienska, A., Smolarz, W. and Wawryk, R. (1964): Attempt of determining the fetal reaction to acoustic stimulation. Acta-Oto-Laryngol, (Stock.), 57.571-574. Eiiberg, R.B., Marmarou, A. and Giovachino. P. (1974): Infant heart rate changes to a synthetic speech sound. J. Aud. Res., 14,21-28. Fleischer, K. (1955): Untersuchungen zur Entwicklung der hmenohrfunktion (Intra-uterine Kinderbewegungen nach Schallreizen). Z. Laryngol. Rhinol., 34,733-740. Gagnon, R., Hunse, C., Carmichael, L., Fellows, F. and Patrick. J. (1987): External vibratory acoustic stimulation near term: Fetal heart rate and heart rate variability response. Am. J. Obstet. Gynecol.. 156,323-327. Gagnon, R.. Patrick, J., Foreman, J. and West, R. (1986): Stimulation of human fetuses with soundand vibration. Am. J. Obstet. Gynecol., 155.848-851. Gelman, S.R., Wood, S.. Spellacy, W.N. and Abrams, R.M. (1982): Fetal movements in response to sound stimulation. Am. J. Obstet. Gynecol., 143,484-485. Goodlin. R.C. and Schmidt, W. (1972): Human fetal arousal levels as indicated by heart rate recording. Am. J. Obstet. Gynecol.. 114.613-621. Graham, F.K.. Anthony, B.J. and Zeigler, B.L. (1983): The orienting response and developmental processes. In: The Orienting Response, Editor: D. Siddle. Granier-Deferre. C., Lecanuet, J-P., Cohen, H. and Busnel, M-C. (1985): Feasability of prenatal hearing test. Acta-Oto-Laryngol. (Stock.), suppl., 421,93-101. Grimwade, J.C., Walker, D.W., Bartlett, M.. Gordon, S. and Wood, C. (1971): Human fetal heart rate change and movement in response to sound and vibration. Am. J. Obstet. Gynecol., 109. lOl103. Hutt, S.J., Hutt, C., Lenard, H.G., Bernuth, H.V. and Muntjerwerff, W.J. (1%8): Auditory responsivity in the human neonate. Nature, 218.888-890.

92 21 22 23 24

25

26 27

28 29 30 31 32

33 34 35

36 31 38 39 40 41 42 43 44 45

Jensen, O.H. (1984b): Fetal heart rate response to a controlled sound stimulus as a measure of fetal well-being. Acta Obstet. Gynecol. &and., 64,97-101. Jensen, O.H. and Flottorp, G. (1982): A method for controlled sound stimulation of the human fetus. &and Audiol., 11,145-150. Leader, L.R., Baillie, P., Martin, B. and Vermeulen, E. (1982): The assessment and significance of habituation to a repeated stimulus by the human fetus. Early Hum. Dev., 7,21 l-283. Lecanuet, J-P., Granier-Deferre, C., Cohen, H., Le Houezec, R. and Busnel, M-C. (1986): Fetal responses to acoustic stimulation depend on heart rate variability pattern, stimulus intensity and repetition. Early Hum. Dev., 13,269-283. Lecanuet, J-P., Granier-Deferre, C., DeCasper, A.J., Maugeais, R., Andrieu, A.J. and Busnel, M-C. (1987): Perception et discrimination foetale de stimuli langagiers, mise en evidence a partir de la reactivitt cardiaque. RCsultats pr(5liminaires. C.R. AC. Sci. Paris, t. 305, SQie Ill, 161-164. Lee, C.Y., Di Loreto, P.C. and O’Lane, J.M. (1975): A study of fetal heart rate acceleration patterns. Obstet. Gynecol., 45, 142-146. Mochizuki, Y., Go, T., Ohkubo, H. and Motomura, T. (1983): Development of human brainstem auditory potentials and gender differences from infants to young adults. Proc. Neurobiol., 20, 273 -285. Monod, N. and Garma, L. (1971): Auditory responsivity in the human premature. Biol. Neonate, 17,292-316. Morrongiello, B.A. and Clifton, R.K. (1984): Effect of sound frequency on behavioural and cardiac orienting in newborn and five month old infants. J. Exp. Child Psychol., 38,429-446. Murphy, K.P. and Smyth, C.N. (1962): Response of fetus to auditory stimulation. Lance& 1,972973. Nijhuis, J.G., Prechtl, H.F.R., Martin, C.B. and Bots, R.S.G.M. (1982): Are there behavioural states in the human fetus? Early Hum. Dev.. 6,177-195. Ohel, G., Birkenfeld, A., Rabinowitz, R. and Sadovsky, E. (1986): Fetal response to vibratory acoustic stimulation in periods of low heart rate activity and low activity. Am. J. Obstet. Gynecol., 154,619-621. Peiper, A. (1925): Sinnesempfindugen des Kindes vor seiner Geburt. Msr., Kinderheik, 29,236. Pereira-Luz. N., Pereira-Lima, C., Hecker-Luz, S. and Feldens, V.L. (1980): Auditory evoked responses of the human fetus. Acta Obstet. Gynecol. Stand., 59,395~404. Prechtl, H.F.R. and O’Brien, M.J. (1982): Behavioural states of the full-term newborn. The emergence of a concept, In: Psychobiology of the Human Newborn, pp. 53-73. Editor: P. Stratton. Wiley and Sons. Querleu, D., Renard, X. and Versyp, F. (1981): Les perceptions auditives du foetus humain. Med. et Hyg., 39,2101-2110. Querleu, D., Renard, X. and Versyp, F. (1985): Vie sensorielle du foetus. In: L’environnement de la naissance, pp. 15-41. Editors: M. Tournaire and 0. Levy. Vigot, Paris. Schmidt, W., Boos, R., Gniers, J., Auer, L. and Schulze, S. (1985): Fetal behavioural states and controlled sound stimulation. Early Hum. Dev., 12, 145-153. Schulman-Galambos, C. and Galambos, R.(l979): Brainstem evoked response audiometry on newborn hearing screening. Arch. Otolaryngol., 105,86-90. Sontag, L.W. and Wallace, R.F. (1936): Changes in the rate of the human fetal heart in response to vibratory stimuli, Am. J. Dis. Child., 51,583-589. Starr, A., Amlie, R.N., Martin, W.H. and Sanders, S. (1977): Development of auditory function in newborn infants revealed by auditory brainstem potentials. Pediatrics, 60,831-839. Steinschneider, A., Lipton, E.L. and Richmond, J.B. (1966): Auditory sensitivity in the infant: effect of intensity on cardiac and motor responsivity. Child Dev., 37,233-252. Tanaka, Y. and Arayama, T. (1969): Fetal responses to acoustic stimuli. Pratt. Oto-Rhinolaryngol., 31,269-273. Trudhrger, B.J. and Boylan. P. (1980): Antepartum fetal heart rate monitoring: value of sound stimulation. Obstet. Gynecol., 55,265-268. Uziel, T., Marot, M. and Gerrnain, M. (1980): Les potentiels evoques du nerf auditif et du tronc cerebral chez le nouveau-n& Revue de Laryngologie, 101,55-71.

93 46 47 48 49

Vmce, MA., Arm&age, S.E., Shillito-Walser, E., Reader, M. (1982): Postnatal consequences of prenatal sound stimulation in the sheep. Behaviour, 81,128-139. Walker, D.W., Grimwade, J.C. and Wood, C. (1971): Intrauterine noise: a component of the fetal environment. Am. J. Obstet., Gynecol., 109,91-95. Weir, C. (1979): Auditory frequency sensitivity of human newborn: some data with improved acoustic and behavioural controls. Percep. Psychophysics, X,282-293. Yamada, A. (1984): Blink reflex elicited by auditory stimulation: clinical study in newborn infants. Brain Dev.. 1,45-53.