Batista et al.
16. Terasawa E, Rodriguez]S, Bridson WE, Wiegand Sj. Factors influencing the positive feedback action of estrogen upon the luteinizing hormone surge in the ovariectomized guinea pig. Endocrinology 1979; 104:680-6. 17. Dempsey EW. Follicular growth rate and ovulation after various experimental procedures in the guinea pig. Am ] Physiol 1937;120:126-32. 18. Brown T], Blaustein ]D. Inhibition of sexual behavior in female guinea pigs by a progestin receptor antagonist. Brain Res 1984;301:343-9. 19. Dempsey EW, Hertz R, Young we. The experimental induction of oestrus (sexual receptivity) in the normal and ovariectomized guinea pig. Am ] Physiol 1936;1l6: 201-9. 20. Croix D, Franchimont P. Changes in the serum levels of gonadotropins, progesterone and estradiol during the es-
July 1991 Am J Obstet Gynecol
21. 22.
23. 24.
trous cycle of the guinea pig. Neuroendocrinology 1975;19:1-11. Blatchley FR, Donovan BT, Ter Haar MB. Plasma progesterone and gonadotropin levels during the estrous cycle of the guinea pig. Bioi Reprod 1976;15:29-38. Shoupe D, Mishell DR, Lahteenmaki P, et al. Effects of the anti progesterone RU 486 in normal women. I. Singledose administration in the midluteal phase. AM] OBS1'ET GYNECOL 1987;157:1415-20. Nieman LK, Choate TM, Chrousos GP, et al. The progesterone antagonist RU 486: a potential new contraceptive agent. N Engl] Med 1986;316:187-91. Dubois C, Ulmann A, Baulieu EE. Contragestion with late luteal administration ofRU 486 (Mifepristone). Ferti! Steril 1988;50:593-6.
Fetal heart rate response to vibroacoustic stimulation during low and high heart rate variability episodes in late pregnancy John A.D. Spencer, MB, BS, Anne Deans, MB, BS, Peter Nicolaidis, MD, and Sabaratnam Arulkumaran, MD London, England In 52 women in late pregnancy, the mean durations of transient fetal tachycardia after vibroacoustic stimulation during low fetal heart rate variability (4.8 minutes) and high fetal heart rate variability (6.3 minutes) were similar. The fetal heart rate continued with high variability in all cases, suggesting that the fetus did not return to its prestimulation state after vibroacoustic stimulation during quiescence. In 10 women, the duration of high variability after vibroacoustic stimulation during low fetal heart rate variability was significantly shorter (mean, 22 minutes) than the preceding (mean, 36 minutes) or subsequent (mean, 43 minutes) high-variability components of complete rest activity cycles. In another 10 women, the duration of high variability after vibroacoustic stimulation during high fetal heart rate variability was similar to preceding and subsequent high-variability episodes. In these 20 women, the next complete rest-activity cycle after vibroacoustic stimulation was not different in duration to the complete cycle recorded on the previous day. (AM J OBSTET GVNECOL 1991 ;165:86-90.)
Key words: Vibroacoustic stimulation, fetal heart rate variability, fetal behavioral cycles The normal baseline fetal heart rate (FHR) in late pregnancy is characterized by alternating episodes of low and high FHR variability!. 2 that accompany quiet and active behavioral states in the fetus. 3 Such vari-
ability changes have been used to study fetal behavior in late pregnancy," 5 during labor,6 and in the newborn. 7 . 8 In the term fetus, vibroacoustic stimulation during fetal quiescence resulted in a transient increase g in baseline FHR (tachycardia) for up to 15 minutes. - 11 with an increase in the frequency and size of acceler-
From the Institute of Obstetrics and Gynaecology, Royal Postgraduate Medical School, University of London, Queen Charlotte's and Chelsea Hospital. P.N. was supported by the British Council. Presented to the Blair Bell Research Society, Glasgow, Scotland, September 8-9,1990. Received for publication October 22, 1990; revised December 28, 1990; accepted December 31, 1990. Reprint requests: J.A.D. Spencer, MBBS, Department of Obstetrics and Gynaecology, University College and Middlesex School of Medicine, University College London, 86-96 Chenies Mews, London, England. WC1E 6HX. 611 /27738
ations, associated with an increase in fetal movements, for up to 1 hour. g , 12, 13 The extent to which this abrupt change in the level of arousal,14 likened to a startle response,15 interrupts the natural cycle of fetal behavioral states is unknown. To assess the degree of fetal disturbance after vibroacoustic stimulation we compared the FHR response to stimulation during low FHR variability with the response obtained to stimulation during high FHR variability. We also compared complete cycles of low and high FHR variability episodes
Volume 165 Number I
before vibroacoustic stimulation with the type and duration of poststimulation variability as well as the next complete cycle of low and high FHR variability episodes after vibroacoustic stimulation. Material and methods
Fifty-two women with singleton pregnancies between 36 and 42 weeks' gestation, all with a cephalic presentation, were studied after antenatal admission to Queen Charlotte's and Chelsea Hospital. The reasons for admission were postmaturity (8), suspected intrauterine growth retardation (6), mild pregnancy-induced hypertension (10), social problems (3), placenta previa (6), planned elective lower segment cesarean section (11), and other reasons (8). Only women who reported normal fetal movements were included in the study. No study participants were taking medication or were smokers. Informed consent was obtained from each woman and the study was approved by the hospital ethics committee. Studies were performed at the same time each morning to exclude the effects of diurnal rhythms. The FHR was recorded with a pulsed Doppler ultrasonographic transducer and an FM7 fetal monitor (Oxford Sonicaid, Chichester, United Kingdom). The mother was semirecumbent and in a quiet room. After 10 minutes of FHR record a vibroacoustic stimulator (Corometrics) was placed on the maternal abdomen over the fetal vertex and activated for 5 seconds. 9 , 12, 13 In 25 cases vibroacoustic stimulation was applied after 10 minutes of low FHR variability and in 27 cases it was applied after 10 minutes of high FHR variability. The FHR record was continued until the poststimulation episode of FHR variability changed. In a subgroup of 20 women a long FHR record was made without vibroacoustic stimulation on the first morning to obtain a complete behavioral cycle, as indicated by consecutive low and high FHR variability episodes. The next morning, at the same time, vibroacoustic stimulation was applied after 10 minutes of low FHR variability in 10 women and after 10 minutes of high FHR variability in the other 10 women. These FHR records were then continued until the next complete cycle of low and high FHR variability episodes was obtained after the poststimulation episode. Low FHR variability episodes were defined as periods of at least 5 minutes' duration with a stable FHR baseline, decreased baseline (long-term) variability (amplitude less than 10 beats/min) and few or no FHR accelerations (FHR pattern A associated with fetal state IF, quiescence 3 ). High FHR variability episodes were identified when the baseline FHR variability was greater than 10 beats/min amplitude with accelerations (FHR pattern B associated with fetal state 2F, activity'). Episodes included in the study were complete in that they were preceded and followed by a change from one
Vibroacoustic stimulation and FHR
87
level of variability to another. All durations were determined from the FHR trace and measured to the nearest whole minute. An acceleration was defined as an abrupt rise in FHR of more than 15 beats/min, returning toward the baseline again after reaching a peak. Tachycardia was defined as an increase in baseline FHR of more than 15 beats/min for longer than 1 'minute. The amplitude and duration ofthe initial tachycardia in response to vibroacoustic stimulation and the nature and duration of the variability episode immediately after vibroacoustic stimulation during low FHR variability were compared with the effects of vibroacoustic stimulation during high FHR variability with the unpaired Student t test. In the subgroup of 20 women the mean durations of the prestimulation episodes of low and high FHR variability in a complete cycle were compared with the duration of poststimulation variability after vibroacoustic stimulation and with the episodes in the next complete cycle of FHR variability episodes with the Student paired t test and the Wilcoxon matchedpair, signed-rank test. Significance was accepted when p < 0.05. Results
All fetuses stimulated during low FHR variability responded with a transient baseline tachycardia that continued in high variability after returning to the baseline (Fig. 1). Those stimulated during high FHR variability also showed a similar transient tachycardia and then continued in high variability. The amplitude of the tachycardia rarely exceeded the size of the largest acceleration noted during preceding or subsequent high FHR variability and was similar irrespective of the variability at the time of stimulation (Table I). The mean duration of the tachycardia and the mean duration of high variability heart rate immediately after the tachycardia were similar whether vibroacoustic stimulation was applied during low or high FHR variability (Table I). The baseline FHR after the tachycardia was not significantly different from the baseline rate before stimulation, despite the increased variability and number of accelerations (Fig. 2). Of the 20 women with long FHR records before vibroacoustic stimulation, the mean durations of high and low FHR variability episodes were not significantly different between the 10 stimulated in low FHR variability and the 10 stimulated in high FHR variability (Table II). The mean duration of high variability (22.8; SD, 11.6; range, 9 to 46 minutes) that followed vibroacoustic stimulation during low FHR variability (n = 10) was significantly shorter than the high variability episodes that contributed to the complete cycles in the same women before (36.4; SD, 14.9 minutes; p < 0.01) and after (42.8; SD, 20.1 minutes; p < 0.05) stimulation. However, the mean duration of high vari-
88
Spencer et al.
HI
''If tOt, "I" lO)Crun
July 1991 Am J Obstet Gynecol
~o
" ", '1 ~ob
, 200 .
!
I
"
:
," , I
200 ..
,l~~
1
!
1------~--------+_------~120 I I
----I ':0
•
25
80
I
'00 75 SO
60
75
I-
I
I
100
I,
80
'00 75
100
25
f
~
50
fC)')O S{Wl
Jl3S0SOO1
Fig. 1. Tachycardia of 5 minutes' duration associated with multiple fetal movements (state 4F, awake) followed by continuation of high variability and movements (state 2F, active) in response to vibroacoustic stimulation (large triangle below the record) during low FHR variability (state I F, quiet). Maternally perceived fetal movements are indicated by small trianglts along the top of the record. Vertiml bars, 3 cms. Table I. FHR response to vibroacoustic stimulation
Amplitude of tachycardia (beats/min) VAS in low FHRV VAS in high FHRV Duration of tachycardia (min) VAS in low FHRV VAS in high FHRV Subsequent high FHRV (min) VAS in low FHRV VAS in high FHRV
No.
Mean
SD
25 27
33.4 37.4
7.87 9.24
15 20
50 60
25 27
4.8 6.3
3.02 3.24
2
14 15
25 27
37.6 51.6
9 10
120 109
25.7 30.9
Minimum
Maximum
VAS, Vibroacoustic stimulation; FHRV, fetal heart rate variability.
ability (42.4; SD, 28.2; range, 16 to 109 minutes) after vibroacoustic stimulation during high FHR variability (n = 10) did not differ from the prestimulation (50.5; SD, 27.3 minutes) and poststimulation (40.4; SD, 18.2 minutes) high variability episodes in these women. In all 20 cases, the mean durations of low and high FHR variability episodes before stimulation were not significantly different from the mean durations of those in the next complete FHR variability cycle that occurred subsequent to the response episode (Table II). The results of analyses with the Student t test and the Wilcoxon matched-pair test were similar. There was no significant difference in the amplitude of accelerations during the spontaneous high FHR variability episodes before and after stimulation. Comment
The immediate response to vibroacoustic stimulation by all cases in this study was an abrupt rise in baseline FHR. We consider this a transient tachycardia because it lasted longer than the duration of an acceleration although it always returned to the preceding baseline level within 15 minutes. Previous reports have described this initial response as a significant rise in baseline FHR,9. 10. 13 a significant increase in mean duration
of accelerations,1(; and a tachycardia. II We found that the mean amplitude of this response often took the rate above 150 beats/min, the recently proposed upper limit of normal for the FHR,17 for between 1 and 15 minutes and so we believe that the description of tachycardia is more appropriate. However, contrary to one previous report l " we found that the mean amplitude of this tachycardia was not greater than the largest of spontaneous accelerations that occurred either before after stimulation. FHR accelerations are known to be associated with spontaneous fetal movements I" and this supports the idea that the transient tachycardia after vibroacoustic stimulation is associated with an increase in fetal movements, consistent with the awake fetal state." However, because this may be brief (Fig. 2), some methods of analysis have failed to show an increase in fetal movements. 10.12 One study with ultrasonography found a fetal startle response each time after vibroacoustic stimulation. 15 All fetuses in our study responded to vibroacoustic stimulation. We found the duration of transient tachycardia after vibroacoustic stimulation during low FHR variability to be similar to that reported previously. 'l. '" It is of interest that both amplitude and duration of the tachycardia after stimulation during high FHR variability were sim-
Vibroacoustic stimulation and FHR
Volume 165 Number I
~',dNd!Atf)
0
89
~()
~ "'.~200~::'" ~
'01
ULT oeO?o ••
09;1"-~·-···)~· --~:o lI\.'
-----'---j 200
-~,~- ,. ~..,--__ .J
,;..--L.80~~--r -~ 1
1e0,0"
180,·
O_-----+---------r--------+4~O------+_------_4~--~--_H2o_----_+--------4_--------~1~----100-
-,
60
Fig. 2. Contrast between FHR variability, number and size of accelerations, and number of fetal movements during fetal quiet state before vibroacoustic stimulation (large triangle below the record) with the poststimulation appearance. Maternally perceived fetal movements are indicated by small triangles along the top of the record. Vertical bars, 3 ems.
Table II. Cycles of FHR variability episodes before and after vibroacoustic stimulation Before VAS Mean
VAS in low variability Low FHRV episodes High FHRV episodes Complete cycle VAS in high variability Low FHRV episodes High FHRV episodes Complete cycle
I
SD
I
Minimum
After VAS
I
Maximum
Mean
I
SD
I
Minimum
I
Maximum
11.6 36.4 48.0
6.5 14.9 16.1
5 16 21
25 54 72
10.7 42.8 53.5
8.8 20.1 18.9
5 14 30
28 90 96
12.8 50.5 63.3
5.1 27.3 25.6
8 22 36
24 120 128
10.7 40.0 50.7
4.0 18.2 17.9
5 15 23
18 70 80
All durations in minutes. VAS, Vibroacoustic stimulation; FHRV, fetal heart rate variability.
ilar to that after vibroacoustic stimulation during low FHR variability, indicating that this aspect of the fetal response tei vibroacoustic stimulation did not depend on the preexisting FHR variability. Devoe et al. l\l also reported that all fetuses responded to vibroacoustic stimulation with a brief tachycardia irrespective of the preexisting state. The sudden and abrupt change in FHR has led to the suggestion that the fetus actually wakes up (state 4P)19 and this is supported by the fact that the mother feels vigorous fetal movements"' 12. " associated with the sustained but transient tachycardia (FHR pattern D according to Nijhuis et al.'). The initial FHR response to vibroacoustic stimulation can therefore be explained by the abrupt' transition into a transient period of wakefulness; our findings confirm that this response also follows vibroacoustic stimulation when applied during high FHR variability. It is generally accepted that episodes of high FHR variability represent the active sleep state in the fetus (state 2F) and these alternate with episodes of low FHR variability, representing periods of quiet sleep (state 1F) to characterize most of the continuous FHR record after 36 weeks' gestation.1.4.2o The duration of the second component of the fetal response to vibroacoustic stim-
ulation applied in low FHR variability, namely the continuation in high variability, has not been previously reported. Persistance of body movements and a raised computer-derived baseline FHR for up to 1 hour have been described. 12. '3 We found that all fetuses, whether stimulated during low or high variability, continued in high variability after vibroacoustic stimulation. Thus after the initial response to stimulation, all fetuses remained in state 2F and the duration was similar irrespective of the variability at the time of vibroacoustic stimulation. Therefore it would seem that when stimulated into wakefulness from active sleep the fetus returned to this state but when stimulated into wakefulness from quiet sleep the fetus did not return to the quiet sleep state but remained in active sleep. Thus alteration of fetal behavioral states is greater after vibroacoustic stimulation during low FHR variability. The 20 women in the subgroup studied on 2 consecutive days were selected because they had a complete cycle of FHR episodes within the 120-minute record made on the first occasion. Therefore the mean durations of consecutive low and high FHR variability episodes shown in Fig. 2 are shorter than some previous reports 4.6 but are similar to others derived from shorter
90
Spencer et al.
records. 18 However, each of these 20 women acted as her own control for the comparison between complete cycles before and after vibroacoustic stimulation and for the comparison between high variability episodes before and after with the duration of high variability that followed the transient tachycardia. Our results show that vibroacoustic stimulation during high variability resulted in a duration of poststimulation high variability similar to the durations of high variability episodes during behavioral cycles before and after stimulation. However, the duration of high variability after vibroacoustic stimulation during low variability was significantly shorter than the prestimulation and posts timulation high variability components of complete restactivity cycles, suggesting that the active fetal state after the abrupt transition from quiescence to wakefulness may not be the same as the spontaneous episodes of fetal activity that contribute to rest-activity cycles. The next complete spontaneous behavioral cycle was of similar duration to the cycle recorded on the previous day, indicating that the disruption of fetal behavior did not extend beyond the stimulated cycle. In conclusion, our findings agree with previous studies that indicate that vibroacoustic stimulation during low FHR variability produces an abrupt change from quiet to active fetal state through a transient period of wakefulness. In addition, we have shown that the fetal response to vibroacoustic stimulation during high FHR variability is a similar transient period of wakefulness before resumption of the preexisting active state. Although we have shown that disruption of fetal restactivity cycles does not extend beyond the stimulated cycle, we do not believe that this implies safety. The value of vibroacoustic stimulation as a means of antenatal fetal investigation needs to be investigated in the form of a randomized trial relating to neonatal outcome and not solely to the production of a reactive FHR record, which it does quite effectively.21
REFERENCES 1. Wheeler T, Guerard P. Fetal heart rate during late pregnancy. Br] Obstet Gynaecol 1974;84:348-56. 2. Dawes GS, Houghton CRS, Redman CWG, Visser GHA. Pattern of the normal human fetal heart rate. Br] Obstet Gynaecol 1982;89:276-84. 3. Nijhuis ]G, Prechtl HFR, Martin CB, Bots RSGM. Are there behavioral states in the human fetus? Early Hum Dev 1982;6:177-95. 4. Timor-Tritsch IF, Dierer L], Hertz RH, Deagan NC, Ro-
July 1991 Am J Obstet Gynecol
5.
6. 7. 8.
9.
10.
II.
12.
13.
14. 15.
16. 17. 18. 19.
20. 21.
sen MG. Studies of antepartum behavioral state in the human fetus at term. AM] OBSTET GYl'\ECOL 1978;132: 524-8. Visser GHA, Carse EA, Goodman]DS,]ohnson P. A comparison of episodic heart-rate patterns in the fetus and newborn. Br] Obstet Gynaecol 1982;89:50-5. Spencer ]AD, Johnson P. Fetal heart rate variability changes and fetal behavioural cycles during labour. Br] Obstet Gynaecol 1986;93:314-21. de Haan R, Patrick], Chess GF, ]aco NT. Definition of sleep state in the newborn infant by heart rate analysis. AM] OBSTET GYNECOL 1977;127:753-8. van Geijn HP,]ongsma HW, de Haan], Eses TAB, Prechtl HFR. Heart rate as an indicator of the behavioural state. AM] OBSTET GYl'\ECOL 1980;136:1061-6. Ohel G, Birkenfeld A, Rabinowitz R, Sadovsy E. Fetal response to vibratory acoustic stimulation in periods of low heart rate reactivity and low activity. AM ] OBSTET GYNECOL 1986;154:619-21. Devoe LD, Seale NA, Ruedrich DA, Castillo RA, Metheny WP. The effects of vibroacoustic stimulation on baseline heart rate, breathing activity, and body movements of normal term fetuses. AM] OBSTET GYNECOL 1989;161: 524-9. Thomas RL, Johnson TRB, Besinger RE, Rafin D, Treanor C, Strbino D. Preterm and term fetal cardiac and movement responses to vibratory acoustic stimulation. AM ] OBSTET GY:\IECOL 1989;161:141-5. Gagnon R, Hunse C, Carmichael L, Fellows F, Patrick J. Effects of vibratory acoustic stimulation on human fetal breathing and gross fetal body movements near term. AM ] OBSTET GYNECOL 1986;155:1227-30. Gagnon R, Hunse C, Carmichael L, Fellows F, Patrick]. External vibratory acoustic stimulation near term: fetal heart rate and heart variability responses. AM ] OBSTET GYNECOL 1987;156:323-7. Goodlin RC, Schmidt W. Human fetal arousal levels as indicated by heart rate recordings. AM] OB5TET GY:\IECOL 1972;141:613-21. Divon MY, Platt LD, Cantrell C], Smith CV, Yeh S-Y, Paul RH. Evoked fetal startle response: a possible intrauterine neurological examination. AM ] OB5TET GY:\IECOL 1985; 153 :454-6. Gagnon R, Patrick], Foreman], West R. Stimulation of human fetuses with sound and vibration. AM ] OBSTET GYNECOL 1986;155:848-51. International Federation of Gynecology and Obstetrics. Guidelines for the use of fetal monitoring. 1m] Gynaecol Obstet 1987;25: 159-67. Wheeler T, Gennser G, Lindvall R, Murrills AJ. Changes in the fetal heart rate associated with fetal breathing and fetal movement. Br] Obstet Gynaecol 1980;87: 1068-79. Devoe LD, Murray C, Faircloth D, Ramos E. Vibroacoustic stimulation and fetal behavioral state in normal term human pregnancy. AM] OB5TET GYNECOL 1990;163:115661. Wheeler T, Murrills A. Patterns of fetal heart rate during normal pregnancy. Br] Obstet Gynaecol 1978;85: 18-27. Smith CV, Phelan ]P, Broussard P, Paul RH. Fetal acoustic stimulation testing: a retrospective analysis of the fetal acoustic stimulation test. AM] OB5TET GY:-'-ECOL 1985; 153:567-8.
16. Terasawa E, Rodriguez]S, Bridson WE, Wiegand Sj. Factors influencing the positive feedback action of estrogen upon the luteinizing hormone surge in the ovariectomized guinea pig. Endocrinology 1979; 104:680-6. 17. Dempsey EW. Follicular growth rate and ovulation after various experimental procedures in the guinea pig. Am ] Physiol 1937;120:126-32. 18. Brown T], Blaustein ]D. Inhibition of sexual behavior in female guinea pigs by a progestin receptor antagonist. Brain Res 1984;301:343-9. 19. Dempsey EW, Hertz R, Young we. The experimental induction of oestrus (sexual receptivity) in the normal and ovariectomized guinea pig. Am ] Physiol 1936;1l6: 201-9. 20. Croix D, Franchimont P. Changes in the serum levels of gonadotropins, progesterone and estradiol during the es-
July 1991 Am J Obstet Gynecol
21. 22.
23. 24.
trous cycle of the guinea pig. Neuroendocrinology 1975;19:1-11. Blatchley FR, Donovan BT, Ter Haar MB. Plasma progesterone and gonadotropin levels during the estrous cycle of the guinea pig. Bioi Reprod 1976;15:29-38. Shoupe D, Mishell DR, Lahteenmaki P, et al. Effects of the anti progesterone RU 486 in normal women. I. Singledose administration in the midluteal phase. AM] OBS1'ET GYNECOL 1987;157:1415-20. Nieman LK, Choate TM, Chrousos GP, et al. The progesterone antagonist RU 486: a potential new contraceptive agent. N Engl] Med 1986;316:187-91. Dubois C, Ulmann A, Baulieu EE. Contragestion with late luteal administration ofRU 486 (Mifepristone). Ferti! Steril 1988;50:593-6.
Fetal heart rate response to vibroacoustic stimulation during low and high heart rate variability episodes in late pregnancy John A.D. Spencer, MB, BS, Anne Deans, MB, BS, Peter Nicolaidis, MD, and Sabaratnam Arulkumaran, MD London, England In 52 women in late pregnancy, the mean durations of transient fetal tachycardia after vibroacoustic stimulation during low fetal heart rate variability (4.8 minutes) and high fetal heart rate variability (6.3 minutes) were similar. The fetal heart rate continued with high variability in all cases, suggesting that the fetus did not return to its prestimulation state after vibroacoustic stimulation during quiescence. In 10 women, the duration of high variability after vibroacoustic stimulation during low fetal heart rate variability was significantly shorter (mean, 22 minutes) than the preceding (mean, 36 minutes) or subsequent (mean, 43 minutes) high-variability components of complete rest activity cycles. In another 10 women, the duration of high variability after vibroacoustic stimulation during high fetal heart rate variability was similar to preceding and subsequent high-variability episodes. In these 20 women, the next complete rest-activity cycle after vibroacoustic stimulation was not different in duration to the complete cycle recorded on the previous day. (AM J OBSTET GVNECOL 1991 ;165:86-90.)
Key words: Vibroacoustic stimulation, fetal heart rate variability, fetal behavioral cycles The normal baseline fetal heart rate (FHR) in late pregnancy is characterized by alternating episodes of low and high FHR variability!. 2 that accompany quiet and active behavioral states in the fetus. 3 Such vari-
ability changes have been used to study fetal behavior in late pregnancy," 5 during labor,6 and in the newborn. 7 . 8 In the term fetus, vibroacoustic stimulation during fetal quiescence resulted in a transient increase g in baseline FHR (tachycardia) for up to 15 minutes. - 11 with an increase in the frequency and size of acceler-
From the Institute of Obstetrics and Gynaecology, Royal Postgraduate Medical School, University of London, Queen Charlotte's and Chelsea Hospital. P.N. was supported by the British Council. Presented to the Blair Bell Research Society, Glasgow, Scotland, September 8-9,1990. Received for publication October 22, 1990; revised December 28, 1990; accepted December 31, 1990. Reprint requests: J.A.D. Spencer, MBBS, Department of Obstetrics and Gynaecology, University College and Middlesex School of Medicine, University College London, 86-96 Chenies Mews, London, England. WC1E 6HX. 611 /27738
ations, associated with an increase in fetal movements, for up to 1 hour. g , 12, 13 The extent to which this abrupt change in the level of arousal,14 likened to a startle response,15 interrupts the natural cycle of fetal behavioral states is unknown. To assess the degree of fetal disturbance after vibroacoustic stimulation we compared the FHR response to stimulation during low FHR variability with the response obtained to stimulation during high FHR variability. We also compared complete cycles of low and high FHR variability episodes
Volume 165 Number I
before vibroacoustic stimulation with the type and duration of poststimulation variability as well as the next complete cycle of low and high FHR variability episodes after vibroacoustic stimulation. Material and methods
Fifty-two women with singleton pregnancies between 36 and 42 weeks' gestation, all with a cephalic presentation, were studied after antenatal admission to Queen Charlotte's and Chelsea Hospital. The reasons for admission were postmaturity (8), suspected intrauterine growth retardation (6), mild pregnancy-induced hypertension (10), social problems (3), placenta previa (6), planned elective lower segment cesarean section (11), and other reasons (8). Only women who reported normal fetal movements were included in the study. No study participants were taking medication or were smokers. Informed consent was obtained from each woman and the study was approved by the hospital ethics committee. Studies were performed at the same time each morning to exclude the effects of diurnal rhythms. The FHR was recorded with a pulsed Doppler ultrasonographic transducer and an FM7 fetal monitor (Oxford Sonicaid, Chichester, United Kingdom). The mother was semirecumbent and in a quiet room. After 10 minutes of FHR record a vibroacoustic stimulator (Corometrics) was placed on the maternal abdomen over the fetal vertex and activated for 5 seconds. 9 , 12, 13 In 25 cases vibroacoustic stimulation was applied after 10 minutes of low FHR variability and in 27 cases it was applied after 10 minutes of high FHR variability. The FHR record was continued until the poststimulation episode of FHR variability changed. In a subgroup of 20 women a long FHR record was made without vibroacoustic stimulation on the first morning to obtain a complete behavioral cycle, as indicated by consecutive low and high FHR variability episodes. The next morning, at the same time, vibroacoustic stimulation was applied after 10 minutes of low FHR variability in 10 women and after 10 minutes of high FHR variability in the other 10 women. These FHR records were then continued until the next complete cycle of low and high FHR variability episodes was obtained after the poststimulation episode. Low FHR variability episodes were defined as periods of at least 5 minutes' duration with a stable FHR baseline, decreased baseline (long-term) variability (amplitude less than 10 beats/min) and few or no FHR accelerations (FHR pattern A associated with fetal state IF, quiescence 3 ). High FHR variability episodes were identified when the baseline FHR variability was greater than 10 beats/min amplitude with accelerations (FHR pattern B associated with fetal state 2F, activity'). Episodes included in the study were complete in that they were preceded and followed by a change from one
Vibroacoustic stimulation and FHR
87
level of variability to another. All durations were determined from the FHR trace and measured to the nearest whole minute. An acceleration was defined as an abrupt rise in FHR of more than 15 beats/min, returning toward the baseline again after reaching a peak. Tachycardia was defined as an increase in baseline FHR of more than 15 beats/min for longer than 1 'minute. The amplitude and duration ofthe initial tachycardia in response to vibroacoustic stimulation and the nature and duration of the variability episode immediately after vibroacoustic stimulation during low FHR variability were compared with the effects of vibroacoustic stimulation during high FHR variability with the unpaired Student t test. In the subgroup of 20 women the mean durations of the prestimulation episodes of low and high FHR variability in a complete cycle were compared with the duration of poststimulation variability after vibroacoustic stimulation and with the episodes in the next complete cycle of FHR variability episodes with the Student paired t test and the Wilcoxon matchedpair, signed-rank test. Significance was accepted when p < 0.05. Results
All fetuses stimulated during low FHR variability responded with a transient baseline tachycardia that continued in high variability after returning to the baseline (Fig. 1). Those stimulated during high FHR variability also showed a similar transient tachycardia and then continued in high variability. The amplitude of the tachycardia rarely exceeded the size of the largest acceleration noted during preceding or subsequent high FHR variability and was similar irrespective of the variability at the time of stimulation (Table I). The mean duration of the tachycardia and the mean duration of high variability heart rate immediately after the tachycardia were similar whether vibroacoustic stimulation was applied during low or high FHR variability (Table I). The baseline FHR after the tachycardia was not significantly different from the baseline rate before stimulation, despite the increased variability and number of accelerations (Fig. 2). Of the 20 women with long FHR records before vibroacoustic stimulation, the mean durations of high and low FHR variability episodes were not significantly different between the 10 stimulated in low FHR variability and the 10 stimulated in high FHR variability (Table II). The mean duration of high variability (22.8; SD, 11.6; range, 9 to 46 minutes) that followed vibroacoustic stimulation during low FHR variability (n = 10) was significantly shorter than the high variability episodes that contributed to the complete cycles in the same women before (36.4; SD, 14.9 minutes; p < 0.01) and after (42.8; SD, 20.1 minutes; p < 0.05) stimulation. However, the mean duration of high vari-
88
Spencer et al.
HI
''If tOt, "I" lO)Crun
July 1991 Am J Obstet Gynecol
~o
" ", '1 ~ob
, 200 .
!
I
"
:
," , I
200 ..
,l~~
1
!
1------~--------+_------~120 I I
----I ':0
•
25
80
I
'00 75 SO
60
75
I-
I
I
100
I,
80
'00 75
100
25
f
~
50
fC)')O S{Wl
Jl3S0SOO1
Fig. 1. Tachycardia of 5 minutes' duration associated with multiple fetal movements (state 4F, awake) followed by continuation of high variability and movements (state 2F, active) in response to vibroacoustic stimulation (large triangle below the record) during low FHR variability (state I F, quiet). Maternally perceived fetal movements are indicated by small trianglts along the top of the record. Vertiml bars, 3 cms. Table I. FHR response to vibroacoustic stimulation
Amplitude of tachycardia (beats/min) VAS in low FHRV VAS in high FHRV Duration of tachycardia (min) VAS in low FHRV VAS in high FHRV Subsequent high FHRV (min) VAS in low FHRV VAS in high FHRV
No.
Mean
SD
25 27
33.4 37.4
7.87 9.24
15 20
50 60
25 27
4.8 6.3
3.02 3.24
2
14 15
25 27
37.6 51.6
9 10
120 109
25.7 30.9
Minimum
Maximum
VAS, Vibroacoustic stimulation; FHRV, fetal heart rate variability.
ability (42.4; SD, 28.2; range, 16 to 109 minutes) after vibroacoustic stimulation during high FHR variability (n = 10) did not differ from the prestimulation (50.5; SD, 27.3 minutes) and poststimulation (40.4; SD, 18.2 minutes) high variability episodes in these women. In all 20 cases, the mean durations of low and high FHR variability episodes before stimulation were not significantly different from the mean durations of those in the next complete FHR variability cycle that occurred subsequent to the response episode (Table II). The results of analyses with the Student t test and the Wilcoxon matched-pair test were similar. There was no significant difference in the amplitude of accelerations during the spontaneous high FHR variability episodes before and after stimulation. Comment
The immediate response to vibroacoustic stimulation by all cases in this study was an abrupt rise in baseline FHR. We consider this a transient tachycardia because it lasted longer than the duration of an acceleration although it always returned to the preceding baseline level within 15 minutes. Previous reports have described this initial response as a significant rise in baseline FHR,9. 10. 13 a significant increase in mean duration
of accelerations,1(; and a tachycardia. II We found that the mean amplitude of this response often took the rate above 150 beats/min, the recently proposed upper limit of normal for the FHR,17 for between 1 and 15 minutes and so we believe that the description of tachycardia is more appropriate. However, contrary to one previous report l " we found that the mean amplitude of this tachycardia was not greater than the largest of spontaneous accelerations that occurred either before after stimulation. FHR accelerations are known to be associated with spontaneous fetal movements I" and this supports the idea that the transient tachycardia after vibroacoustic stimulation is associated with an increase in fetal movements, consistent with the awake fetal state." However, because this may be brief (Fig. 2), some methods of analysis have failed to show an increase in fetal movements. 10.12 One study with ultrasonography found a fetal startle response each time after vibroacoustic stimulation. 15 All fetuses in our study responded to vibroacoustic stimulation. We found the duration of transient tachycardia after vibroacoustic stimulation during low FHR variability to be similar to that reported previously. 'l. '" It is of interest that both amplitude and duration of the tachycardia after stimulation during high FHR variability were sim-
Vibroacoustic stimulation and FHR
Volume 165 Number I
~',dNd!Atf)
0
89
~()
~ "'.~200~::'" ~
'01
ULT oeO?o ••
09;1"-~·-···)~· --~:o lI\.'
-----'---j 200
-~,~- ,. ~..,--__ .J
,;..--L.80~~--r -~ 1
1e0,0"
180,·
O_-----+---------r--------+4~O------+_------_4~--~--_H2o_----_+--------4_--------~1~----100-
-,
60
Fig. 2. Contrast between FHR variability, number and size of accelerations, and number of fetal movements during fetal quiet state before vibroacoustic stimulation (large triangle below the record) with the poststimulation appearance. Maternally perceived fetal movements are indicated by small triangles along the top of the record. Vertical bars, 3 ems.
Table II. Cycles of FHR variability episodes before and after vibroacoustic stimulation Before VAS Mean
VAS in low variability Low FHRV episodes High FHRV episodes Complete cycle VAS in high variability Low FHRV episodes High FHRV episodes Complete cycle
I
SD
I
Minimum
After VAS
I
Maximum
Mean
I
SD
I
Minimum
I
Maximum
11.6 36.4 48.0
6.5 14.9 16.1
5 16 21
25 54 72
10.7 42.8 53.5
8.8 20.1 18.9
5 14 30
28 90 96
12.8 50.5 63.3
5.1 27.3 25.6
8 22 36
24 120 128
10.7 40.0 50.7
4.0 18.2 17.9
5 15 23
18 70 80
All durations in minutes. VAS, Vibroacoustic stimulation; FHRV, fetal heart rate variability.
ilar to that after vibroacoustic stimulation during low FHR variability, indicating that this aspect of the fetal response tei vibroacoustic stimulation did not depend on the preexisting FHR variability. Devoe et al. l\l also reported that all fetuses responded to vibroacoustic stimulation with a brief tachycardia irrespective of the preexisting state. The sudden and abrupt change in FHR has led to the suggestion that the fetus actually wakes up (state 4P)19 and this is supported by the fact that the mother feels vigorous fetal movements"' 12. " associated with the sustained but transient tachycardia (FHR pattern D according to Nijhuis et al.'). The initial FHR response to vibroacoustic stimulation can therefore be explained by the abrupt' transition into a transient period of wakefulness; our findings confirm that this response also follows vibroacoustic stimulation when applied during high FHR variability. It is generally accepted that episodes of high FHR variability represent the active sleep state in the fetus (state 2F) and these alternate with episodes of low FHR variability, representing periods of quiet sleep (state 1F) to characterize most of the continuous FHR record after 36 weeks' gestation.1.4.2o The duration of the second component of the fetal response to vibroacoustic stim-
ulation applied in low FHR variability, namely the continuation in high variability, has not been previously reported. Persistance of body movements and a raised computer-derived baseline FHR for up to 1 hour have been described. 12. '3 We found that all fetuses, whether stimulated during low or high variability, continued in high variability after vibroacoustic stimulation. Thus after the initial response to stimulation, all fetuses remained in state 2F and the duration was similar irrespective of the variability at the time of vibroacoustic stimulation. Therefore it would seem that when stimulated into wakefulness from active sleep the fetus returned to this state but when stimulated into wakefulness from quiet sleep the fetus did not return to the quiet sleep state but remained in active sleep. Thus alteration of fetal behavioral states is greater after vibroacoustic stimulation during low FHR variability. The 20 women in the subgroup studied on 2 consecutive days were selected because they had a complete cycle of FHR episodes within the 120-minute record made on the first occasion. Therefore the mean durations of consecutive low and high FHR variability episodes shown in Fig. 2 are shorter than some previous reports 4.6 but are similar to others derived from shorter
90
Spencer et al.
records. 18 However, each of these 20 women acted as her own control for the comparison between complete cycles before and after vibroacoustic stimulation and for the comparison between high variability episodes before and after with the duration of high variability that followed the transient tachycardia. Our results show that vibroacoustic stimulation during high variability resulted in a duration of poststimulation high variability similar to the durations of high variability episodes during behavioral cycles before and after stimulation. However, the duration of high variability after vibroacoustic stimulation during low variability was significantly shorter than the prestimulation and posts timulation high variability components of complete restactivity cycles, suggesting that the active fetal state after the abrupt transition from quiescence to wakefulness may not be the same as the spontaneous episodes of fetal activity that contribute to rest-activity cycles. The next complete spontaneous behavioral cycle was of similar duration to the cycle recorded on the previous day, indicating that the disruption of fetal behavior did not extend beyond the stimulated cycle. In conclusion, our findings agree with previous studies that indicate that vibroacoustic stimulation during low FHR variability produces an abrupt change from quiet to active fetal state through a transient period of wakefulness. In addition, we have shown that the fetal response to vibroacoustic stimulation during high FHR variability is a similar transient period of wakefulness before resumption of the preexisting active state. Although we have shown that disruption of fetal restactivity cycles does not extend beyond the stimulated cycle, we do not believe that this implies safety. The value of vibroacoustic stimulation as a means of antenatal fetal investigation needs to be investigated in the form of a randomized trial relating to neonatal outcome and not solely to the production of a reactive FHR record, which it does quite effectively.21
REFERENCES 1. Wheeler T, Guerard P. Fetal heart rate during late pregnancy. Br] Obstet Gynaecol 1974;84:348-56. 2. Dawes GS, Houghton CRS, Redman CWG, Visser GHA. Pattern of the normal human fetal heart rate. Br] Obstet Gynaecol 1982;89:276-84. 3. Nijhuis ]G, Prechtl HFR, Martin CB, Bots RSGM. Are there behavioral states in the human fetus? Early Hum Dev 1982;6:177-95. 4. Timor-Tritsch IF, Dierer L], Hertz RH, Deagan NC, Ro-
July 1991 Am J Obstet Gynecol
5.
6. 7. 8.
9.
10.
II.
12.
13.
14. 15.
16. 17. 18. 19.
20. 21.
sen MG. Studies of antepartum behavioral state in the human fetus at term. AM] OBSTET GYl'\ECOL 1978;132: 524-8. Visser GHA, Carse EA, Goodman]DS,]ohnson P. A comparison of episodic heart-rate patterns in the fetus and newborn. Br] Obstet Gynaecol 1982;89:50-5. Spencer ]AD, Johnson P. Fetal heart rate variability changes and fetal behavioural cycles during labour. Br] Obstet Gynaecol 1986;93:314-21. de Haan R, Patrick], Chess GF, ]aco NT. Definition of sleep state in the newborn infant by heart rate analysis. AM] OBSTET GYNECOL 1977;127:753-8. van Geijn HP,]ongsma HW, de Haan], Eses TAB, Prechtl HFR. Heart rate as an indicator of the behavioural state. AM] OBSTET GYl'\ECOL 1980;136:1061-6. Ohel G, Birkenfeld A, Rabinowitz R, Sadovsy E. Fetal response to vibratory acoustic stimulation in periods of low heart rate reactivity and low activity. AM ] OBSTET GYNECOL 1986;154:619-21. Devoe LD, Seale NA, Ruedrich DA, Castillo RA, Metheny WP. The effects of vibroacoustic stimulation on baseline heart rate, breathing activity, and body movements of normal term fetuses. AM] OBSTET GYNECOL 1989;161: 524-9. Thomas RL, Johnson TRB, Besinger RE, Rafin D, Treanor C, Strbino D. Preterm and term fetal cardiac and movement responses to vibratory acoustic stimulation. AM ] OBSTET GY:\IECOL 1989;161:141-5. Gagnon R, Hunse C, Carmichael L, Fellows F, Patrick J. Effects of vibratory acoustic stimulation on human fetal breathing and gross fetal body movements near term. AM ] OBSTET GYNECOL 1986;155:1227-30. Gagnon R, Hunse C, Carmichael L, Fellows F, Patrick]. External vibratory acoustic stimulation near term: fetal heart rate and heart variability responses. AM ] OBSTET GYNECOL 1987;156:323-7. Goodlin RC, Schmidt W. Human fetal arousal levels as indicated by heart rate recordings. AM] OB5TET GY:\IECOL 1972;141:613-21. Divon MY, Platt LD, Cantrell C], Smith CV, Yeh S-Y, Paul RH. Evoked fetal startle response: a possible intrauterine neurological examination. AM ] OB5TET GY:\IECOL 1985; 153 :454-6. Gagnon R, Patrick], Foreman], West R. Stimulation of human fetuses with sound and vibration. AM ] OBSTET GYNECOL 1986;155:848-51. International Federation of Gynecology and Obstetrics. Guidelines for the use of fetal monitoring. 1m] Gynaecol Obstet 1987;25: 159-67. Wheeler T, Gennser G, Lindvall R, Murrills AJ. Changes in the fetal heart rate associated with fetal breathing and fetal movement. Br] Obstet Gynaecol 1980;87: 1068-79. Devoe LD, Murray C, Faircloth D, Ramos E. Vibroacoustic stimulation and fetal behavioral state in normal term human pregnancy. AM] OB5TET GYNECOL 1990;163:115661. Wheeler T, Murrills A. Patterns of fetal heart rate during normal pregnancy. Br] Obstet Gynaecol 1978;85: 18-27. Smith CV, Phelan ]P, Broussard P, Paul RH. Fetal acoustic stimulation testing: a retrospective analysis of the fetal acoustic stimulation test. AM] OB5TET GY:-'-ECOL 1985; 153:567-8.
