Oxytocin challenge test for antepartum fetal assessment

FETUS, PLACENTA, AND NEWBORN

Oxytocin challenge test for antepartum fetal assessment Report of a clinical experience JOHN F. HUDDLESTON, M,D. GAIL SUTLIFF, B.S.N. F. EDWARD CARNEY, JR., M.D. CHARLES E. FLOWERS, JR., M.D. Birmingham, Alabama OCT's were performed _1 ,209 times on 533_ fetuses at risk for UPI. OCT interpretation was negative for 72% of tests, and there were no fetal deaths within a week follOwing a negative test. Our data thus support the concept that the use of th·e OCT for fetal surveillance most frequently justiftes a course of nonintervention. Of 69 fetuses with a positive test, 47 were subjected to the stresses of labor with maternal hyperoxygenatiori and lateral positioning. Twenty of these 47 (43%) tolerated labor Without biophysical evidence of distress. Reactivity of the FHA, when present during a positive test, significantly increased the likelihood that a fetus would tolerate labor. The fetal and perinatal mortality rates in the patients identified as at risk for UPI and studied with OCT's were no greater than in a comparable group of pregnancies without identifiable risk for UPI and not studied with OCT's. (AM. J. OesTET. GYNECOL. 135:609, 1979.)

DATA FROM SEVERAL centers 1- 3 have established the principles that (I) normal uterine contractions exert repetitive hypoxic stresses upon the fetus and (2) a fetus with marginal oxygenation will respond to these hypoxic stresses with myocardial depression evidenced by late decelerations of the fetal heart rate (FHR). Hammacher 4 was the first to propose that the response of the FHR to induced uterine contractions be studied From the Diviswn of Maternal and Fetal Medicine, Department of Obstetrics and Gynecology, University of Au1bama Medical Center. Supported in part by a grant from Corometrics Medical Systems, Inc. Rereivedfor publication November 29, 1978. Arrepted December 28, 1978. Reprint requests: Dr. John F. Huddleston, Department of Obstetrics and Gynecology, University of Alabama School of Medicine, Birmingham, Alabama 35294. 0002-9~nH/i912!0609+06$00.60/0

©

1979 The C.\'. Mosby Co.

in fetuses at risk for uteroplacental insufficiency (UPI) and that the results be used as a test of antepartum fetal status. The small series reported by Hammacher and co-workers 5 was one of several which suggested that the response of the FHR was predictive of fetal outcome. Ray and associates 6 utilized a standardized version of this test, which they called the oxytocin challenge test (OCT), during a blinded study of 66 patients at risk for UPI. No fetal deaths were noted within a week of a test showing a normal FHR response (negative OCT), but there were 20% fetal deaths and 53% 5-minute Apgar scores less than 7 among those fetuses demonstrating persistent late decelerations (positive OCT). The OCT has subsequently been studied by several groups, and it has been repeatedly shown that the negative test is highly predictive of fetal well-being. A recent review 7 of these reports notes that, although there have been individual case reports of fetal death

609

1111

610

Huddleston et al. Am.

When: 1)2:38 weeks 2) L/S ratio 2:2.0 3):!: Favorable Cervix

Fig. 1. Clinical use of the oxytocin challenge test. LIS = lecithin/sphingomyelin; £ 3 =estriol. (From Huddleston,]. F., and Freeman, R. K.: In Bolognese, R. ]., and Schwarz, R. H., editors: Perinatal Medicine: Management of the High Risk Fetus and Neonate, Baltimore, 1977, The Williams & Wilkins Company.)

occurring within a week of a negative OCT, the combined data from 12 large series reveal such a "falsenegative" rate of only 2.2 per 1,000 tested patients. The authors of these reports have generally considered the positive OCT as ominous and have followed various management protocols aimed toward early delivery. However, the common finding that positive-OCT fetuses frequently tolerate labor without distress (socalled "false-positive" tests) has been disturbing. This report details a large experience with the OCT, supports claims made concerning the safety of the test and the reliability of a negative test, and suggests which of the fetuses with positive tests are more likely to tolerate labor. Materials and methods

This series comprises 1,209 OCT's performed upon 533 consecutive fetuses (529 gravidas, four sets of twins) studied at the University of Alabama Medical Center during the 18-month period from April 1, 1974, through September 30, 1975. Seventeen singleton fetuses were referred only for testing and were delivered in area private hospitals. The other 516 fetuses (96.8%) were of gravidas who attended our obstetric complications clinics, were delivered on our inpatient services, arid· were managed by the faculty and residents of the Department of Obstetrics and Gynecology. These 516 represented 10.0% of the 5,154 infants delivered on our services during that period. The indications for testing were those which placed the fetus at risk for UPI. Contraindications included previous classical cesarean section, known placenta previa, and concern about precipitating premature labor. Testing was usually begun at 34 weeks of gestation or at such time thereafter when an appropriate indication developed. An informed consent was obtained by the requesting physician.

Novembt·t I. 1'17'• Obstrt. (;yttc
J

All tests were performed by a nurse specialist (G. S.) in a facility located separate from the delivery unit, but close enough for rapid transfer. Physician backup was immediately available. No test was part of an induction oflabor. All tests were interpreted by the senior author and results were made immediately available to the clinicians. The test was conducted as described by Freeman,H except that elevation of the right hip, rather than the semi-Fowler position, was used in an attempt to prevent aortocaval compression. A Corometrics FMS101B (Corometrics Medical Systems, Inc., Wallingford, Connecticut) external monitor was used. Baseline uterine and FHR activities were observed carefully for I 0 to 20 minutes prior to any decision to begin low-dose oxvtocin via infusion pump. Adequate uterine activity was defined as three contractions, each lasting at least 40 seconds, within a 10-minute period. If this level of uterine activity was achieved, either spontaneously or with oxytocin, and no late decelerations were noted, the test was interpreted as negative and was repeated only after 7 days. In these negativeOCT patients, spontaneous labor was awaited or delivery was effected at term if the amniotic fluid lecithin/ sphingomyelin (LIS) ratio indicated fetal pulmonary maturity (2.0 or more). A test was interpreted as positive if, in the absence of uterine hyperstimulation, late decelerations were associated with most (at least half) of the contractions. Suspicious, hyperstimulation, and unsatisfactory tests, as defined by Freeman, 8 were considered to impart no useful information, prompted no clinical intervention, and were repeated the following day whenever possible. The pregnancy with a positive OCT was managed according to the LIS ratio. If the LIS ratio was 2.0 or more, the patient was delivered. If the LIS ratio was less than 2.0, the patient was hospitalized and serial urinary estriol measurements were made. If these measurements were normal, a favorable LIS ratio was awaited. In the absence of a favorable LIS ratio, delivery for fetal indications was effected only if low or deteriorating estriol values occurred when the gravida was hospitalized at bed rest. A diagrammatic summary of this management method is given in Fig. I. When a decision to deliver was made, the delivery route was predicated on the Bishop 9 score of the cervix. In the absence of contraindications to vaginal delivery, induction of labor (amniotomy and/or oxytocin infusion) was attempted if the Bishop score was 5 or more. Prior to attempted induction, efforts were made to enhance fetal oxygenation (nasal oxygen administration, maintenance of the lateral decubitus position, and intravenous hydration). All induction attempts were monitored by internal techniques.

..

..

·--

.

-

'!. y

y"'

' ., -,.

OCT for antepartum fetal assessment 611

Volume 135 Number!">

For all laboring patients, continuous electronic fetal monitoring was used unless delivery was imminent on admission. Fetal distress was defined as persistent late decelerations, persistent severe variable decelerations, or acute, prolonged decelerations. 10 Transient distress was defined if these ominous FHR patterns resolved with maternal hyperoxygenation, lateral decubitus positioning, cessation of any oxytocin administration, and correction of any hypotension. Persistent distress was defined if these efforts at intrauterine resuscitation failed and if, as a consequence, immediate delivery was effected. FHR reactivity in response to fetal movement 5 was not considered in OCT classification or patient management during the period of this study, but the occmTence of such movement was noted on the monitor record. Without reference to subsequent clinical events, we later reviewed the positive-OCT records for FHR reactivity (accelerations of 15 beats per minute above baseline FHR) 11 coincident with fetal movement and classified them as normal (5 or more accelerations per 20 minutes) 11 • marginal (I to 4 accelerations per 20 minutes), or absent. The hospital charts of every mother and infant were reviewed. Data were analyzed with standard chi-square techniques employing Yates' correction; P < 0.05 was considered significant. Results

The numbers of patients tested for each indication are given in Table I. Multiple indications were present in R2 patients. In 895 instances (74%), the gravidas were outpatients at the time of the test. The distribution of OCT interpretations is given in Table I I. A mean of 2. 3 tests (range I to 10) was performed per fetus. Spontaneous uterine activity was adequate for interpretation in 86 of the entire I ,209 tests (7.1 o/c), but was adequate in 47 of those 262 tests ( 17.9%) performed following an amniocentesis done the same day. In only three instances did spontaneous, premature labor (less than 38 weeks) result in delivery within 48 hours of a test. In each instance, adequate spontaneous uterine activity had been present, and thus no oxytocin had been given. In two of the three, the birth weight exceeded 2,500 grams. For various maternal and/or fetal indications, a total of 64 gravidas underwent cesarean section without a trial of labor ( 12.1 %). The other 465 patients underwent labor, and 441 (94.8%) were monitored electronically. Of the 24 not monitored, 21 were delivered on our services: 11 were delivered promptly after arrival; for 10 there was no monitoring equipment available. One infant ofthis latter 10 was meconium stained,

Table I. Indications for OCT in 529 patients

% I. 2. 3. 4. 5. 6. 7. 8. 9.

Prolonged pregnancy Chronic hypertension Acute hypertension Suspected growth retardation Elderly gravida Previous stillbirth Diabetes, insulin-dependent Chronic renal disease Others Multiple indications

240 133 112 31 31 29 24 19 12 S2

45.4 25.1 21.2 5.9 5.9 5.5 4.5

3.6 2.3 15.S

Table II. OCT interpretations in 1,209 tests Interpretation

No.

Negative Positive Suspicious Unsatisfactory H yperstimulation

867 73 190 48 31

%

71.7 6.0

77.7

1.'1.7 40

22.3

2.6

Table III. Percentage chance that OCT will ultimately become positive Indication

I. Diabetes, insulin-

2. 3. 4. 5.

6. 7. 8.

dependent Chronic renal disease Suspected growth retardation Elderly gravida Prolonged pregnancy Chronic hypertension Previous stillbirth Acute hypertension

Prrcentage

24

6

2S.O

19 31

4

s

21.1 16.1

31 240 133 29 112

5 30

16.1 12.S

14 3

10.5 10.3 8.0

9

had endotracheal suction only after a tO-minute delay, and was the onlv one of these 24 to receive a ,;-;-minute Apgar score less than 7. There were 321 fetuses delivered within 7 days of a negative OCT. There were no fetal deaths in this group. Labor occurred in 296 cases and was monitored in 279 (94.3%). Fetal distress occurred in 51 of these monitored patients, but was persistent in only 22. In 17 of the 22 (6. 1% of those monitored within a week of a negative test), the FHR abnormality was persistent late decelerations. There were only two of these 279 fetuses monitored within a week of a negative OCT (0.7%) for whom the f'>-minute Apgar score was less than 7. Table III defines the chance, for a gravida with a given indication, that the OCT ultimately would become positive. Of the 69 patients with a positive OCT. 54 (78%) had a LIS ratio of 2.0 or greater and were delivered within 24 hours. Fourteen (20tYc) whose LIS

1111

612

Huddleston et al.

~o\Tmlwr

:\m.

Table IV. Delivery of patients with positive OCT %

Cesarean section without labor Allowed trial of monitored labor No distress Transient distress Persistent distress

22

32

47 20 9

68 43

I8*

38

19

*Four of I8 were delivered vaginally.

Table V. Tolerance of labor when OCT positive Persistent distress

FHR reactivity

Normal Marginal Absent Uncertain

28 28 10 3

20 20 7 0

II

Table VI. Perinatal death Fetal

OCT-study patients (N = 516) Deaths Rate/ I ,000 total births Non-study patients >2,000 grams (N = 4299) Deaths Rate/ I ,000 total births

3 6 0

9 0

I

Neonatal

6 5 7

I

Perinatal

1.9

3 5.8

4 7.8

32 7.4

21 4.9

53 12.3

I

ratio was less than 2.0 were hospitalized at bed rest, urinary estriol excretion was followed, and they were delivered (within 9 days) either following spontaneous labor or by intervention, usually when the LIS ratio became favorable. In two of the 14 hospitalized patients, the OCT reverted to negative. In another two of these 14 patients, estriol measurements were very low and delivery was effected despite the unfavorable LIS ratios; neither infant developed respiratory distress. One patient with two positive tests refused amniocentesis or admission, maintained strict bed rest at home, and was admitted in prodromal labor 12 days after her initial positive test. She had a Bishop score of 8, was given oxygen and maintained in lateral recumbency, and tolerated labor without fetal distress. In three of the 22 positive-OCT cases in which no labor was allowed (Table IV), obstetric indications dictated the delivery route: transverse lie in two and previous cesarean section in one. In 19, the operative indication was a Bishop score less than 5. Of the 4 7 patients (68%) who were allowed to labor, labor occured spontaneously in 12, was induced by amniotomy alone in eight, and was induced with oxytocin in 27. Twenty ( 43%) of these 4 7 tolerated labor without FHR patterns

J.

I, 1~17~1 Ohqet. ( ~\'JH'( ()L

of fetal distress; nine ( 19o/c) exhibited transient distres' patterns; 18 (38o/c) had persistent distress and H of these underwent cesarean section. All 47 neonates had ;)-minute Apgar scores of 7 or more. Eight of the 69 positive-OCT perinates ( IJ.6C)tj were small for gestational age, 12 whereas 39 of the 464 perinates (8.4%) without a positive test were so identified (X 2 = 0.41, N.S.). !'\umbers of infants wih :">-minute Apgar scores less than 7 did not differ significantly whether an OCT was positive or not (two of 69 versus 13 of 464, x2 = 0.14, N.S.). The tolerance of the fetus to labor as a function of FHR reactivity during a positive OCT is shown in Table V. Assessment of reactivity was possible on the tracings of 66 patients. For the reactivity assessments of normal and marginal, there was no discernible difference in the development of distress during labor. Twenty of the 40 (50%) with these assessments tolerated labor without distress and only 11 (28clc) manifested persistent distress. Overall, labor was tolerated without persistent distress in 29 of 40 fetuses with reactivity present (normal or marginal), whereas labor was not tolerated by any of the seven fetuses with absent reactivity (X 2 = I 0.4, P < 0.00 1). Two of the three patients with absent reactivity underwent cesarean section without labor and were delivered of infants with 0-minute Apgar scores less than 7. These two were the only perinates with low 5-minutc Apgar scores among the 69 with positive OCT's. Of the 69 positive-OCT patients, the first test was positive in 35 (51%). Positive tests were preceded directly by a suspicious OCT in 10 of the other 34 cases (29%). In 21 of these 34 (62%), however, at least one of the tests antecedent to the positive OCT was suspicious. Thus, a positive test was ultimately found in only this 21 (14%) of the 155 suspicious tests which were repeated. There were four perinatal deaths in this study: one fetal and three neonatal. The fetal loss occurred in a woman with chronic hypertension who noted cessation of fetal movement I 0 days after a negative OCT and 3 days after she failed to keep her weekly OCT appointment. Two of the neonatal deaths occurred in severely asphyxiated infants whose pathologic intrapartum FHR patterns were appreciated too late for their emergency cesarean sections to be of benefit. The other neonatal death was in the case of a severely hypertensive woman whose infant died of hyaline membrane disease and sepsis 4 days after the spontaneous labor and delivery of a I ,900 gram, premature infant with good Apgar scores. This infant was the only one in the study with severe respiratory distress. A mortality rate comparison made between the 516

y

..

....

Volume ns ;\lumber 0

OCT-studv perinates delivered on our service and the 4299 non-study perinates delivered during the time period and weighing more than 2,000 grams is presented in Table VI. Non-study group infants weighing 2,000 grams or less were excluded from the comparison because the fetuses identified as at risk for UPI (OCT group) generally were not entered into that group until 34 weeks, at which time the average fetal weight is about 2,000 grams. 12 Despite the apparent numerical improvement in both fetal and perinatal mortality rates of those perinates in the OCT group, when compared to the non-OCT group, neither comparison proved statistically significant (X 2 = 1.32, x2 = 0.58, respectively, N.S.). Comment

Antepartum fetal assessment is an important aspect of the management of pregnancies at risk for UP I. Maternal urinary estriol excretion has been the most commonly used test for such assessment, although its interpretation frequently is confounded by a variety of clinical and methodologic problems. This test is frequently unavailable at irregular times and/or in remote areas. Moreove.r, considerable time is required to collect and analyze the samples. Thus, the value of urinary estriol determinations is reduced in those situations where rapid assessment is desirable. Serum or plasma estriol analysis, while overcoming collection problems, still requires constant availability of sophisticated laboratory facilities and time-consuming serial determinations for trend analysis. By contrast, the OCT has the advantages of widespread availability, speed of procedure, and ease of interpretation. It can be performed with standard fetal-monitoring equipment and by anyone trained in such techniques, provided that immediate medical backup is available. The data in this report suggest that a definitive OCT interpretation-negative or positive-is possible for more than three fourths of test~. In addition, the very low false-negative rate for tests performed only weekly compares the OCT favorablv to tests, such as estriol determinations, which must be performed more frequently. The absence of OCT-related premature labor in this study attests to the safety of the test and emphasizes the importance of observing baseline uterine activity before adding oxytocin. Since obtaining a safe and interpretable OCT requires constant attention by the testing person, we feel that this person should be free of other clinical responsibilities during this time. Therefore, it is preferable to separate the OCT facility from the activities of the labor and delivery unit. Such a separation was probably instrumental in our obtaining such a low percentage of unsatisfactory tests.

OCT for antepartum fetal assessment

613

Although the weekly test interval originallv was chosen arbitrarily, 6 our data support its continued use. However, the single fetal death occurring I 0 days after a negative OCT in our series suggests that a 7 -da~ interval is nearly maximal. Other recent case reports challenge the validity of even a 7 -dav interval. "1 Our finding persistent fetal distress in 7.r nearly one fifth of tests done following amniocentesis, an alternative means of stressed FHR testing might he available for those patients with contraindications to oxytocin administration, but for whom amnion·ntesis is indicated. Aubry and Pennington 16 have reported that it is possible in early pregnancy to identify the ~O'l of pregnancies which eventually will produce fieF~ of the perinatal deaths. In the present studv, \\{' performed OCT's on the 10% identified to be at highest risk for UPI and pursued a uniform policv of patient management based upon the test interpretations. Since this study was not controlled, we cannot be certain that this identified I 0% included most of those destined to result in perinatal losses. \-\'e have shown, however, that a system of surveillance relying principally upon the OCT is associated with fetal and perinatal mortalitv rates, in the group at risk for UPL which are at least no greater than rates found in the neonates weighing more than 2,000 grams who were not considered at such high risk. Since three of the four pninatal deaths in this study can be classified as preventable. it is possible that the system is capable of producing even more favorable results. It is unreasonable, however, to expect a test based upon an evaluation of fcta I oxygenation to be able to predict all perinatal death. Acute antepartum events such as abruptio placentae. intr·apartum catas-

Ill

614

Huddleston et al. Am.

trophes such as umbilical cord prolapse, and neonatal problems such as those associated with prematmity likely will not be predictable on the basis of OCT results. Finally, we confirm that the primary value of the OCT is in the negative test. When repeated weekly, negative tests allow confidence in a policy of nonintervention, even for those very high-risk pregnancies in which the fear of fetal jeopardy might provoke unnecessary intervention. Positive tests, at least when

FHR reactivity is present, do not contraindicate a trial . of Carefully monitored labor, provided that efforts are made to increase fetal oxygenation. The authors express appreciation to the resident staff of the Department of Obstetrics and Gynecology for their help in carrying out this study and to Ethel Coats, Deborah K. Schenk, M.S.N., Bruce A. Harris, Jr., M.D., and William F. Huggins, M.D., for their assistance in reviewing the manuscript.

REFERENCES I. Pose, S. V., Escarcena, L., Althabe, 0., Schwarcz, R. L.,

2. 3.

4.

5.

6.

7.

lbarra-Polo, A. A., Quintanilla, R., and Caldeyro-Barcia, R.: The influence of uterine contractions on the partial pressure of oxygen of the human fetus, in CaldeyroBarcia, R., Mendez-Sauer, C., and Dawes, G. S., editors: Effects of Labour on the Foetus and Newborn, Oxford, 1967, Pergamon Press, p. 48. Greiss, F. C., and Anderson, S. G.: Uterine blood flow during labor, Clin. Obstet. Gynecol. 11:96, 1968. Myers, R. E., Mueller-Heubach, E., and Adamsons, K.: Predictability of the state of fetal oxygenation from the quantitative analysis of the components of late deceleration, AM. J. OssTET. GYNECOL. 115:1083, 1973. Hammacher, K.: Friiherkennung intrauteriner gefahrenzustande durch electrokardiographie und tokographie, in Elert, R., and Hiiter, K. A., editors: Prophyaxe Friihkindlicher Hirnshaden, Stuttgart, 1966, Georg Theime Verlag, p. 120. Hammacher, K., Hiiter, K. A., Bokelmann, J., and Werners, P. H.: Foetal heart frequency and perinatal condition of the foetus and newborn, Gynaecologia 166:349. 1968. Ray, M., Freeman, R., Pine, S., and Hesselgesser, R.: Clinical experience with the oxytocin challenge test, AM . .J. 0BSTET. GYNECOL. 114: I' 1972. Huddleston, J. F., and Freeman, R. K.: The use of the oxytocin challenge test for the management of pregnancies at risk for uteroplacental insufficiency, in Bolognese,

Non·mbtT I. J~n~~ Obstet. (:\TH'«•'-

J

8. 9. 10.

II. 12.

13. 14. 15. 16.

R. J., and Schwarz, R. H., editors: Perinatal Medicine: Management of the High Risk Fetus and Neonate, Baltimore, 1977, The Williams & Wilkins Company, p. 68. Freeman, R. K.: The use of the oxytocin challenge test for antepartum clinical evaluation of uteroplacental respiratory function, AM. j. 0BSTET. GYNECOL. 121:481, 1975. Bishop, E. H.: Pelvic scoring for elective induction, Obstet. Gynecol. 24:266, 1964. Freeman, R. K.: Management of acute intrapartum fetal distress, in Freeman, R. K., editor: A Clinical Approach to Fetal Monitoring, San Leandro, 1974, Berkeley Bio-Engineering, p. 69. Paul, R. H., and Miller, F. C.: Antepartum fetal heart rate monitoring, Clin. Obstet. Gynecol.21:375, 1968. Lubchenco, L. 0., Hansman, C., Dressler, M., and Boyd, E.: Intrauterine growth as estimated from live-born birth weight data at 24 to 42 weeks of gestation, Pediatrics 32:793, 1963. Salerno, N.J., and Kay, T. R.: A further challenge to the validity of the weekly interval between oxytocin challenge tests, AM. j. 0BSTET. GYNECOL. 130:849, 1978. Bruce, S. L., Petrie, R. H., and Yeh, S. Y.: The suspicious contraction stress test, Obstet. Gynecol. 51:415, 1978. Braly, P., and Freeman, R. K.: The significance of fetal heart rate reactivity with a positive oxytocin challenge test, Obstet. Gynecol. 50:689, 1977. Aubry, R. H., and Pennington, J. C.: Identification and evaluation of high-risk pregnancy: The perinatal concept. Clin. Obstet. Gynecol. 16:3, 1973.

... '

'

··,.