Fetal acoustic stimulation in the early intrapartum period as a predictor of subsequent fetal condition

Kinoshita

nant tumor of the ovary. Current concepts with regard to the histogenesis and histologic criteria for diagnosis of endodermal sinus tumor of the ovary were largely determined on the basis of the work of Teilum.' I have been able to identify only one other reported case of endodermal sinus tumor after the age of 60 years in the literature. Ferracini et al. 2 described an endodermal sinus tumor in a 63-year-old female , but their case is of extraovarian origin. Thus I believe the

March 1990 Am J Obstet Gynecol

present case is the first one reported of endodermal sinus tumor of the ovary in a woman after the age of fiO ye;Jn .

REFERENCES I. Teilum G. Endodermal sinus tumors of the ovary and testis. Cancer 1959;12:1092-105. 2. Ferracini R, Gardini G, Lanzanova G, Lorenzini P. Endodermal sinus tumor in a 53-year-old female. Pathologica 1979;71 :885-7.

Fetal acoustic stimulation in the early intrapartum period as a predictor of subsequent fetal condition Albert P. Sarno, Jr., MAJ, USA, Myoung Ock Ahn, MD, MPH, Jeffrey P. Phelan, MD, and Richard H. Paul, MD Los Angeles, California Fetal acoustic stimulation has recently received much attention in the literature. This study evaluates fetal acoustic stimulation in the early intrapartum period as a predictor of subsequent fetal condition . The study group consisted of 201 patients, approximately 60% of whom had complicated pregnancies. All were in the latent phase of labor with singleton, vertex-presenting fetuses. Gestational age ranged from 37 to 43 weeks. Fourteen of the 201 fetuses (7%) showed a nonreactive response to fetal acoustic stimulation and those fetuses were at significantly greater risk of initial and subsequent abnormal fetal heart rate patterns, meconium staining, and cesarean delivery because of fetal distress and Apgar scores <7 at both 1 and 5 minutes. Transient fetal heart rate decelerations after a reactive response occurred in 25% of patients; however, fetal outcome was not worse in this group. A reactive response to fetal acoustic stimulation was associated with high specificity and negative predictive values. Therefore we conclude that fetal acoustic stimulation in the early intraprtum period may discriminate the compromised from the noncom promised fetus. (AM J QaSTET GVNECOL 1990;162:762-7.)

Key words: Fetal acoustic stimulation, labor, fetal distress Fetal acoustic stimulation as a tool for the assessment of fetal condition has recently received much attention in the literature. Contemporary evaluation of the technique dates to the study of Read and Miller' in which the authors compared the fetal response to acoustic From the DiVISIon of Maternal-Fetal Medzcme , Department of Obstetrics and Gynecology, Univemty of Southern Calz(omia School of Medzcme, and Women's Hospital, Los Angeles County I Unzversity of Southern Californta M edical Center. Computational assistance was provided by the CLINFO Project, funded by the DiVISion of R esearch Resources, Natronallnstdutes of Health, under Grant No . RR-0043. The opinilYYls expressed hel'ein are those of the authors and do n o/ necessanl), represent those of the United States Army or the Department of Defense. Received for publication May 5, 1989; revised October 19, 1989; accepted October 24, 1989. Reprint requests: Albert P. Sarno, Jr., MAJ , USA , Department of Obstetncs and Gynecology, Tnpler Army M edzcal Center, Honolulu, H awaII 96859-5000. 611 /17681

762

stimulation with the cono"action stress test. The subsequent contraction stress test was always negative when the fetus exhibited a heart rate acceleration in response to acoustic stimulation. Conversely, suspicious or positive contraction stress tests were frequently preceded by failure of acoustic stimulation to generate a heart rate acceleration. Since that time, several investigators have evaluated the usefulness of acoustic stimulation as an adjunct to antepartum testing"'· and to clarify fetal acid-base status in the presence of abnormal intrapartum fetal heart rate (FH R) patterns. ' The bulk of the data available thus far clearly suggests clinical value to the technique. However, more study is necessary to validate previous studies and to document the safety of the technique. Because electronic fetal monitoring has not been shown to be superior to properly performed intermittent fetal auscultation,6.' there is a controversy regard-

Volume 162 Number 3

ing its use. Indeed, the American College of Obstetricians and Gynecologists recently stated that intermittent fetal auscultation" is as effective as electronic fetal monitoring in monitoring high risk patients during labor.'" Although patient acceptance of intermittent auscultation may be high, physician preference, nursing staff requirements, and litigation factors may mitigate against the routine use of intermittent fetal auscultation. Therefore, the obstetrician may be greatly aided by an evaluation performed early in the course of labor that can differentiate the fetus likely to tolerate the stress of labor from one that cannot. This information could then be used to rationally triage patients to a lowor high-risk category with regard to the fetus. This study evaluates the usefulness of fetal acoustic stimulation in the early intrapartum period as a predictor of subsequent fetal condition. Material and methods This prospective study included 20 I patients seen at the labor and delivery unit of Women's Hospital, Los Angeles County/University of Southern California Medical Center. Inclusion criteria were as follows: (1) gestational age ~37 weeks; (2) singleton fetus; (3) vertex presentation; and (4) latent phase of labor (cervical dilatation $4 cm). Consecutive patients who met inclusion criteria were studied during the period of August 1,1987, to November 1, 1987, during periods of availability of the first author. After admission the patiem was placed in bed and electronic fetal monitoring was instituted. A 40-minute baseline FHR monitor tracing was obtained, then fetal acoustic stimulation was performed with a fetal acoustic stimulator (Corometrics model 146, Wallingford, Conn.) (Fig. I), sound level 82 dB at 1 m in air. The acoustic stimulator was placed on the maternal abdomen over the fetal vertex and a 3-second pulse of sound stimulation was applied. I f no acceleration of the FHR was noted within I minute, an additional pulse was administered to a maximum of three pulses, each 1 minute apart. A reactive response was defined as one or more accelerations of the FHR IS beats/min from baseline, persisting for 15 seconds. A nonreactive response was defined as failure to elicit a qualifying acceleration after any of three separate stimuli and for 10 minutes after the last stimulus (Fig. 2). Care was taken not to perform acoustic stimulation during or immediately after a uterine contraction to avoid periods of transient fetal hypoxia and for standardization of the technique. The result of acoustic stimulation was blinded from the physicians who managed the patient's labor. The FHR tracing in the initial and last 40 minutes of monitoring were then evaluated and analyzed ac-

Intrapartum fetal acoustic stimulation

763

---1/1.

...

.

-- ,.-.

Fig. 1. Fetal acoustic stimulator (model 146, Corometrics, Wallingford, Conn.).

cording to the result of acoustic stimulation. The tracings were evaluated for baseline rate, spontaneous accelerations, variability, and the presence of variable decelerations, late decelerations, or prolonged decelerations. Variable decelerations were graded as mild , moderate , or severe by the criteria of Kubli et al. 9 Late decelerations were classified with respect to the presence or absence of average beat-to-beat variability. Prolonged decelerations were defined as a FHR of 90 beats/min or a decline in heart rate of 40 beats/min from baseline, persisting for ~ 1 minute. Spontaneous FHR accelerations were considered present if one or more accelerations of the FHR of 15 beats/min from baseline lasting IS seconds were noted. 10 All FHR tracings wel'e read by a single examiner without knowledge of the prior fetal acoustic stimulation result. Pregnancy outcome was assessed with respect to the incidences of meconium staining, fetal distress requiring cesarean delivery, Apgar scores <7 at 1 and 5 minutes, subsequent abnormal FHR patterns, and perinatal mortality. Fetal distress was defined as a persistently ominous FHR pattern or a fetal acid-base assessment indicative of acidosis (scalp pH <7.20). Maternal-fetal infection was defined as chorioamnionitis, endometritis, or neonatal sepsis proven by blood culture. Statistical analysis was carried out with one-tailed Fisher's exact test or X2 analysis, and significance was considered at p < 0.05. Results There were 201 patients included in the study (74 primigravid, 127 multigravid) . Mean gravidity and parity were 2.7 and 1.4, respectively. Maternal age (mean ± SD) were 25.9 ± 5.5 years and mean gestational age was 40.1 ± 2.2 weeks. Duration of ruptured membranes and labor (mean ± SD) were 14.2 ± 17.0 hours and 17.4 ± 8.5 hours, respectively. A total of 108 patients (53.8%) came to the hospital with ruptured membranes. Overall, 118 of20 I patients (59%) had one or more complications of pregnancy.

764

Sarno et al.

March 1990 Am J Obstct Gync(ol

4 305AAO

C " « (II'I O\l1IRlf, ,,,,,r 'u \1 ... \ ... ,1\'-'

,-

"1

68866 I , ' ~" '.0so· ----r---I

I

:"!:'

i i, .L-- ----. - - -

:-i ''1'1,[ I~ il~_

"

~jl l l UAREf 100

-

! i 1r

- - - - - 75--

J

!-------;:--+-'----I-

'

,'

i "

'.

50-------'- _' - - -

Fig. 2. Fetal responses to acoustic stimulation: A, purely reactive response; B, reactive response followed by transient decelerations of FHR; and C, nonreactive response.

Table I. Pregnancy outcome on the basis of results of fetal acoustic stimulation ReactIVe (11

I

Nonreactive (n = 14)

I

'*

n

42

22.5 5,9

7 5

50* 35.7t

22 4

11.7 2.1 0.5 3.7

7 2 0 2

50+ 14.3* 0 14.3

11

Meconium Cesarean delivery because of fetal distress Apgar scores <7 I minute 5 minutes Perinatal mortality Maternal-fetal infection

187)

II

I

7

%

*P <0.05. tp <0.0005. +P <0.001. Table II. Pregnancy outcome for patients with a reactive fetal acoustic stimulation, with and without FHR decelerations Reactive (1/

Reactwe wllh FHR dereiemt/(Jns (n = 46)

I

%

n

32 9

22.7 6.4

10 2

21.7 4.3

15 4 0 40

10.6 2.8 0 28.4

7 0 I 14

15.2 0 2.2 30.0

It

Meconium Cesarean delivery because of fetal distress Apgar scores <7 I minute 5 minutes Perinatal mortality Nuchal co rd at delivery

I

141)

Table I documents pregnancy outcome according to the result of fetal acoustic stimulation early in the course of labor. Fourteen of 201 patients (7%) showed a nonreactive response to fetal acoustic stimulation. This group of patients had a statistically significant increase in the incidence of meconium staining of the amniotic fluid, cesarean delivery because of fetal distress, and Apgar scores <7 at both 1 and 5 minutes after birth

%

when compared with the group of patients with a reactive response to fetal acoustic stimulation. There was no significant difference in perinatal mortality. Of the 187 patients with a reactive response to fetal acoustic stimulation, 141 showed a FHR acceleration with return to baseline. whereas 46 patients (25% of the reactive group) showed a transient FHR deceleration of at least 15 beats/min for 15 seconds but <60

Intrapartum fetal acoustic stimulation

VOhllllC Hi;! 1\:1I1l1bCI

:1

765

Table III. Incidence of abnormal FHR patterns in early labor in relation to fetal acoustic stimulation Rl'flr/Il'1'

tachycardia Decreased vanability La<-k of acceleration, Variable de< eleration, MIld Moderate Severe Late decelerations A\ erage val iabIlity Decreased variabIlItv Prolonged deederat iO;l <;

= 187)

I

II

Ba~eline

(II

o

NOIIII'(U/IVl' (II

'X

0 0 7

0 0 50*

I :~.4

2 1 0

14.3 7.1 0

2

14.3t 0 0

1.1

4 ')"

I

II

o

2

-"5

= 14)

27 05

1 fi

:~.2

1 6

0.5

3.2

0 0

*p <0005.

tp «UJ5

Table IV. Incidence of abnormal FHR patterns in late labor in relation to fetal acoustic stimulation Rmr/nl(' II

Baselme tachvcanha Decreased variabilIty Lack of accelerations Vanable decelerations Mild Moderate Severe Late deceleratIons Average variability Decreased variabilIty Prolonged deceleratio;"s

(II

= 182)

Norll('(l(/lIIf

I

14 14

11

(n =

I

13)

%

£)

26

7.7 7.7 14.3

7

30.R* 3R.5t 5H.3t

40 19 37

22.0 10.4 20.3

6 0 6

44.I:j: 0 46.1:j:

12

6.6 16 9.3

1 0

7.7 0 15.4

:~

17

4

2

*p <0.01. ttJ <0 OOS. :j:jJ <0.05.

beats/ min for 60 seconds after the initial acceleration. Table II shows the outcome for these two subgroups. There are no significant differences in pregnancy outcome noted, including the incidence of nuchal cords at delivery. Table I I I correlates the result of acoustic stimulation to the incidence of abnormal FH R patterns in t he initial 40 minutes of the FHR tracing. Patients with a nonreactive respon~e to fetal acoustic stimulation had a significant increase in the incidence of lack of spontaneous FHR accelerations and late decelerations with average variability. Table IV correlates the result of acoustic stimulation in early labor to the risk of abnormal FHR patterns in the last 40 minutes of labor. Of the 20 I study patients, six FHR tracings were unavailable for review. Patients with a nonreactive fetal acoustic stimulation had a significant increase in the incidence of multiple abnormalities to include baseline tachycardia, decreased beatto-beat variability, lack of spontaneous accelerations, and mild and severe variable decelerations. The occurrence of fetal tachycardia appeared to be indepen-

dent of maternal chorioamnionitis and fetal sepsis Cfable I). Table V shows the sensitivities, specificities, predictive values, and overall test efficiency for fetal acoustic stimulation with respect to the various indicators of pregnancy outcome. A single perinatal mortality occurred in the study group in a 38-year-old gravida I, para 0 woman at 38 weeks' gestation who was admitted in early labor with ruptured membranes and oligohydramnios by ultrasonography. The initial FHR tracing was reassuring and fetal acoustic stimulation elicited an acceleration of the FHR followed by a transient deceleration. Labor progressed slowly with oxytocin augmentation, complicated by chorioamnionitis, treated with intravenous antibiotics. After 16 hours the patient underwent cesarean delivery because of arrest of dilatation. Before delivery, the FHR tracing showed a baseline rate of 170 beats/min with accelerations and mild variable decelerations present. A 3010 gm male infant with Apgar scores of 5, 6, and 7 at 1, 5, and IO minutes was delivered. Umbilical arterial blood gases showed pH 7.21,

766

Sarno et al.

March 1990 Am J Obstet Gynecol

Table V. Sensitivity, specificity, positive and negative predictive values and overall efficiency of fetal acoustic stimulation for various indicators of pregnancy outcome P os/tlVe predIctIVe

Meconium Cesarean delivery due to fetal distress Apgar scores <7 I minute 5 minutes

value

NegatIVe predictIve value

Efficiency

14.3 31.2

95.4 95.1

50.0 35.7

77.5 94.1

75.6 90.0

24.1 33.3

95.9 93.8

50.0 14.3

88.2 97.9

85.6 92.0

Peo. SI mm Hg, Po. 21 mm Hg, and base excess - 6. The neonatal course was complicated by suspected sepsis, meconium aspiration, neonatal seizures, and cerebral infarction. The neonate died on day 15 after birth.

Comment The role of fetal acoustic stimulation in contemporary obstetrics is still in evolution. Application of the technique has been found in a variety of settings. Recent studies evaluating the use of fetal acoustic stimulation as an adjunct to antepartum testing have documented a decrease in falsely nonreactive tests and a decrease in testing time with comparable false-negative rates (i.e., fetal death within 7 days of a reactive test) when compared with the nonstress test. 3.' The use of acoustic stimulation has been extended to the intrapartum period to clarify fetal condition in the presence of abnromal FHR patterns or during periods of decreased variability. Smith et aI.' found that a reactive response to fetal acoustic stimulation in the presence of an abnormal FHR pattern was uniformly associated with a scalp pH 2:7.25. Of those fetuses that were nonreactive, 50% had a pH <7.2S. The authors concluded that intrapartum fetal acoustic stimulation was a reasonable alternative to fetal scalp sampling. Polzin et al. II showed that a FHR acceleration in response to acoustic stimulation of 10 beats/min, persisting for 10 seconds, provided the same prognostic value as a 15 beats/min X IS-second acceleration in the presence of an abnormal FHR pattern. Divon et al. 12 showed that in the presence of decreased variability the healthy fetus responds with a FHR acceleration in 100% of cases and fetal movement in approximately 75% of cases. Ingemarsson et al. 13 have studied the usefulness of fetal acoustic stimulation in the presence of an abnormal FHR tracing in the early intrapartum period, and have found the technique to improve the positive predictive value of the FHR tracing. They found a 7S% risk of fetal distress when the initial FHR tracing was nonreactive and acoustic stimulation was followed by no response or a FHR deceleration. Thus there clearly appears to be clinical application of this screening technique as an adjunct for the antepartum and intrapartum assessment of the fetus.

The purpose of this study is to evaluate fetal acoustic stimulation as an "admission test" in the early intrapartum period as a predictor of subsequent fetal condition. The study showed that 7% of fetuses will exhibit a nonreactive response to the acoustic stimulus and are at increased risk of adverse outcome. In addition, 2S% of fetuses with a reactive response will also exhibit a transient FHR deceleration, which does not seem to pose added risk when compared with a purely reactive response in this limited number of patients. Though there was no difference in nuchal cord incidence in the two groups, the cause is most likely transient cord compression associated with the fetal "startle response"" induced by the acoustic stimulus. However, it should be noted that at least one case has been reported in which the vigorous fetal movement associated with the stimulus resulted in a prolonged deceleration that was considered to be a result of tightening of a nuchal cord. 15 The association of a nonreactive response to acoustic stimulation with an initial FHR tracing that shows a lack of spontaneous reactivity and the presence of late decelerations would suggest that fetal acoustic stimulation in the early intrapartum period screens for the fetus entering labor with preexisting chronic or subacute hypoxia. The FHR patterns in subsequent labor that show baseline tachycardia, decreased variability, lack of reactivity, and mild and severe variable decelerations reflect the cumulative effects of the stress of labor on these already compromised fetuses. The clinical utility of a screening test lies in its predictive value for subsequent outcome. That is, given an abnormal test, what is the probability of an abnormal outcome (positive predictive value)? Conversely, given a normal test, what is the probability of a normal outcome (negative predictive value)? In this study a reactive FHR response to acoustic stimulation was associated with a high negative predictive value. Therefore a reactive response should reassure the clinician. Intermittent auscultation or intermittent electronic fetal monitoring in the active phase of labor, or ambulation in the latent phase of labor without telemetry after a reactive response may be reasonable options and deserve study. The positive predictive value of a nonreactive FHR

Volume 162 l'\umber 3

response to acoustic stimulation is ,omewhat less than the negative predictive value. However, it must be realized that the positive predictive value of a screening test varies with the prevalence of the disease in the population. For example, given a 5% disease prevalence and sensitivity and specificity fixed at 95%. the maximum positive predictive value obtainable is 50%. In the study population, the incidence of fetal distress that requires cesarean delivery was 5.5% (II of 201 patients). Therefore, a positive predictive value of 35.7% for fetal distress should not indicate immediate cesarean delivery but, rather, should prompt continuous electronic fetal monitoring and increased physician vigilance in the group of patients with a nonreactive response to fetal acoustic stimulation in early labor. Used in this manner, fetal acoustic stimulation could then be valuable as an "admission test" in early labor to triage fetuses to a low- or high-risk category for intrapartum fetal distress. The safety of fetal acoustic stimulation and standardization of the technique have been questioned. It, A report evaluating intrauterine sound levels in pregnant ewes documents an average sound pressure of 135 dB when an artificial larynx was used. J7 However. Smith et al. IH placed microphones in the uteri of nine term gravid volunteers after amniorrhexis. Transabdominal acoustic stimulation with an artificial larynx produced intrauterine sound levels of 91 to III dB, a level that should not pose major fetal risks. In addition, a study that compared 20 neonates exposed to acoustic stimulation during the intrapartum period with 20 controls l " showed there was no difference in auditory nerve testing or brain stem-evoked responses between the two groups. Therefore it would seem that the potential clinical benefit outweighs any theoretic risk currently associated with the technique. In conclusion, fetal acoustic stimulation in the early intrapartum period appears to be a reasonable technique for triage of the fetus to a high- or a low-risk category. Such information may be helpful to the clinician managing patients in labor. REFERENCES I. Read JA. Miller FC. Fetal heart rate acceleration 111 response to acoustic stimulation a; a measure of fetal wellbeing. A~IJ ()s,IE"i GY:\ECOL 1977;129:512-7.

Intrapartum fetal acoustic stImulation

767

2. Trudinger Bf. Boylan P. Antepartum fetal heart rate monitoring: value of sound stimulation. Ob5tet Gynecol 1980;55:265-8. 3. Serafini P. Lind5ay MBI. Nagev DA. et a!. Antepartum fetal heart rate response to sound stimulation: the acoustic stimulation test. A:-IJ OIlS 1I'. r GY:'\H.OL 1984: 148:41-5. 4. Smith CV, Phelan JP. Platt LD. et al Fetal acoustic stimulation tC51mg II. A randomized climcal comparison with the nomtress test. A \J .I OIl,T[ r (;Y'\U.Ol. 1986; 155: 131-4. 5. Smith CV. Ngu)en HN, Phelan./P. Paul RH. Intrapartum assessment of fetal well-being: a compariwn of fetal acoustic stimulation with acid-base determinations. AM J 011,'11'.'1 GY:,\EU1L 1986.155:726-8. 6. Thacker SB. The efficacy of intrapartum electronic fetal monitoring. A \1 J Ollsn, 1 GY:\H.OL 1987; 156:24-30. 7. Leveno KJ, Cunningham FG, ~elson S. Roark M. et a!. A prospective companson of selective and universal electromc fetal monitonng in :H.995 pregnancies. N Engl J Med 1986;31:615-9. 8. ACOG Newsletter. American College of Ob,tetncians and Gynecologists November 1988:32: I. 9. Kubli FW. Hon EH, Kh,lzin AF. Takemur