The significance of antepartum variable decelerations

The significance of antepartum variable decelerations Akolisa Anyaegbunam, M.D., Lois Brustman, M.D., Michael Divon, M.D., and Oded Langer, M.D. Bronx, New York A total of 4886 nonstress tests were reviewed to establish the relationship between antepartum variable decelerations and perinatal outcome. The association between various fetal heart rate components and variable decelerations was also studied. The incidence of variable decelerations, defined as three or more decelerations ;;, 15 bpm lasting at least 15 seconds in a 20-minute period, was 1 .3%. The results suggest that in the presence of variable decelerations: (1) there is a higher incidence of fetal distress in labor, low Apgar scores, neonatal intensive care unit admissions, and nuchal cord involvement; (2) the presence of accelerations and normal variability is associated with good neonatal outcome, whereas their absence is associated with adverse outcome; (3) the presence of accelerations or good variability is not independently correlated with neonatal outcome. (AM J OBSTET GYNECOL 1986;155:707-10.)

Key words: Nonstress test, antepartum variable decelerations, perinatal outcome In the nonstress test, which is the most common method of antepartum fetal surveillance, the presence of accelerations with fetal movement has been shown to be indicative of fetal well-being.'·' While some investigators4 have reported good fetal outcome in the presence of antepartum variable decelerations, others5·7 have suggested its association with fetal compromise. Therefore, the purpose of this study was to further examine the relationship between antepartum variable decelerations and perinatal outcome and to assess the association between various fetal heart rate components and variable decelerations. Material and methods A retrospective case-control study was conducted. Cases with variable decelerations and matched controls were selected from 3300 patients in whom 4886 consecutive nons tress tests were performed during a 2-year period (Aprill983 to May 1985). Nonstress tests were performed for a variety of indications. The antepartum testing scheme for high-risk patients in our institution is to conduct nonstress tests every 3 days with daily fetal movement counting. Abnormal tests were further evaluated with a biophysical profile and patients were managed as indicated by the score of the profile." For all tests, patients were placed in the left lateral position and either a Corometrics Model III, Hewlett-Packard Model 9020, or Litton Model AM 550 fetal monitor with an external tocodynamometer was used. Fetal From the Department of Obstetrics and Gynecology, The Albert Einstein College of Medicine. Presented at the Sixth Annual Meeting of The Society of Perinatal Obstetricians, San Antonio, Texas, january 30-February 1, 1986. Reprint requests: Dr. Akolisa A nyaegbunam, Department of Obstetrics and Gynecology, Albert Einstein Hospital, 1825 Eastchester Road, Bronx, NY 10461.

movements were documented by the mother using an event marker. Cases were patients whose tracings demonstrated variable decelerations, defined as three or more decelerations ;:=: 15 bpm lasting at least 15 seconds and associated with fetal movement unrelated to uterine activity in a 20-minute observation period. Control subjects were selected from all patients with reactive nonstress tests by means of a systematic stratified sample. Controls were matched to cases by maternal age, gestational age, and parity. A reactive nonstress test was defined as two fetal heart rate ( FHR) accelerations associated with fetal movement in a 20-minute observation period. 2 A tracing that did not meet the reactive criteria in an SO-minute period was considered nonreactive. Normal FHR variability is six or more FHR oscillations around a normal FHR baseline with an amplitude of at least 6 bpm in a !-minute period. 9 · 10 A total sample of 64 cases and 150 controls were selected according to the described criteria. All FHR tracings and antepartum, intrapartum, and neonatal medical records were examined for cases and controls. Maternal factors considered included age, parity, diagnosis during pregnancy, volume of amniotic fluid, and method of delivery. Fetal and infant characteristics included fetal heart patterns, fetal movements, gestational age at testing and delivery, meconium, cord involvement, birth weight, Apgar scores and neonatal intensive care unit admissions. FHR tracings were analyzed for multiple variables, such as FHR baseline, depth and duration of decelerations, shape of decelerations, and presence of accelerations. From medical records the amniotic fluid volume was ascertained as a reflection of either ultrasound measurements or observation of an obstetrician at the time of delivery. When measurement was done by ultrasound 707

708 Anyaegbunam et al.

October 1986 Am J Obstet Gynecol

Table II. Perinatal outcomes for live-born infants of cases and controls (%)

(%)

I p Value

Decreased amniotic fluid volume 12 11 Meconium 28 Nuchal cord Apgar scores <7 23 1 min 5 min 11 NICU admission 16 >24 hr 9 Small for gestational age 64 Total patients (No.)

6 5 15

NS NS <0.05

9

<0.01 <0.01

II Group I p< 01

l"l

~Group

jcasesl Controls

II

o<01

11

2

2 3 150

<0.001 NS (p-0.07)

NICU = Neonatal intensive care unit. Accelerat ion Variability

Present Normal

Absent Normal

Absent Decrease

Present Decrease

Fig. 1. Perinatal outcome in patients with variable decelera· tions and different combinations of accelerations and FHR

variability.

Table I. Selected characteristics of cases and controls Cases (n = 64) Group I

(n

Matching factors Maternal age

= 15)

ll I Group = (n

49)

26 ± 7

(mean ± SD) 24 (37%) Nulliparous (No.) 38 ± 3 Gestational age (wk, mean± SD) Interval from test to delivery (No.) 0 (0%) 5 (33%) 24 hr 6 (40%) 30 (61 %) 1-5 days 4 (27%) 19 (39%) >5 days

Controls (n = 150)

27 ± 5 62 (41 %) 38 ± 2 2 (1 %)

97 (65 %) 51 (34%)

the volume was considered reduced if fluid pockets were < 1 em in vertical diameter. 11 Intrauterine growth retardation was defined as the tenth percentile of birth weight for a given gestational age. 12 To further examine the relationship between variable decelerations and neonatal outcomes cases were stratified into two groups. Infants were classified in group I if their !-minute Apgar score was < 7 and/or the y were admitted to the neonatal intensive care unit for ;;:24 hours. Cases in group I represent adverse neonatal outcomes. Group II infants had !-minute Apgar scores ;;:7 with a well-baby nursery admission or a neonatal intensive care unit admission of <24 hours. Univariate and bivariate statistical analysis was performed with the use of X2 and Fisher's exact tests and continuous data by Student's t test.

Results

The identification of 64 cases in the 4886 tracings reviewed represents a 1.3% incidence of variable decelerations. Indications for testing in cases and controls were similar and include postdates, 44 % and 49%; hypertension, 13% and 15%; and diabetes mellitus, 17% and 15 %. The successful matching of cases and controls is reflected by the comparability of maternal age, parity, and gestational age at delivery as presented in Table I. The majority of all patients were delivered within 5 days, 64% (41) of cases and 66% (99) of controls. Furthermore, when stratified by perinatal outcome, 33% of adverse outcome cases in group I were delivered in the first 24 hours compared with no patients in group II and only 1% of controls. The perinatal outcomes for live-born infants of cases with variable decelerations and matched controls are presented in Table II. Although the prevalence of meconium and decreased amniotic fluid was higher for cases than for controls, the differences were not statistically significant. There was, however, a significant excess of instances of nuchal cord in cases compared with controls (p < 0.05). Significantly more neonates of cases than controls had low (<7) !-minute Apgar scores (23% compared with 7%) as well as low (< 7) 5-minute Apgar scores (11% versus 2%). Neonatal intensive care unit admissions >24 hours were also significantly higher for cases than for controls. The higher incidence of small for gestational age infants in cases (9%) compared with controls (3 %) approached statistical significance. There were few fetal deaths with a rate of 16 per 1000 for cases and 6 per 1000 for controls. Additional analyses were performed on the FHR tracings from cases to examine FHR characteristics as they relate to neonatal outcome. The a priori subgroups, group I (adverse outcomes) and group II (good outcomes), were used to stratify the cases for these com-

Antepartum variable decelerations 709

Volume 155 Number 4

I

I I

c

I

-~ '

I""'

·-·

I

I I

. I • I I : I. II I III ,L I I ' II I

.....

..

'

. I

·'

.

' I.L

·.

-

I I' I

-

J

I

.:j

j :j

: Vt

!~[" ; : _;:-~ ~

-I

1,!. 1 I

• -f \i' ""'f,r lit,L

!'-"" i ~

I I I I II I I ..l..L. .l .Li· l if..- ..=t.l

li I -

I l

I II

V ·SHAPED VARIABLE DECELERATION

I I 'I

.,

I 1 ' I I ~ I I I 1 l,..j_ .J. I .I I I

. ' ~- : \.: ':V - • j_

•

:

•

~ -~

I I I I I I

'

I

.

E~-+ ,_

I

j

~

- --~. :J!i-:::;

J

I

~ I .l

.Y, o'--r;-'"- ~

I I

._ - ::Lf-l:::i= ::J::f-l:::h-l~r .:I H 1- 1=1l=h ·:t

l-- ;. ::r 1+-+

I

''

I I

j

I ,

J.

""

:.!

"!""

1::

- .1

!= • :.1~f-+-+-+~ .:. . -= 1_J::

_;

1-.:J ::j::

:~

EUI.I:I[!IIII!lJillll,ll Fig. 2. Nonstress test tracings showing the two shapes of variable decelerations.

parisons. The FHR characteristics associated with variable decelerations revealed a normal variability in 26% of group I compared with 80% in group II (p < 0.001). Accelerations were present in 53% of group I versus 85% of group II (p < 0.05). Different combinations of accelerations and variability were analyzed in relation to adverse (group I) and good (group II) outcomes for the 64 case infants. These results are displayed in Fig. 1. The combination of accelerations and normal variability was significantly (p < 0.01) associated with good outcomes (group II). In contrast, there was a significant association (p < 0.01) with adverse outcomes (group I) when acceleration and variability were absent. No association was seen for the combinations of either acceleration in the presence of decreased variability or normal variability in the absence of accelerations. Other components of the FHR such as baseline, depth, and duration of deceleration were carefully examined. No significant differences were identified for any of these characteristics when analyzed with respect to neonatal outcomes (group I versus group II). Two shapes of variable decelerations were identified (Fig. 2). All cases had V-shaped decelerations and 22 (34%) had both V-shaped and W-shaped (biphasic) decelerations. Examination of the relationship between type of variable deceleration and nuchal cord revealed that the risk was 12 times greater for cases with Wshaped deceleration compared with cases with only Vshaped decelerations (relative risk of 12.5). The method and indications for delivery for cases

Table III. Method of delivery and indication for cases and controls Cases Group I !Group II Controls

Method of delivery Spontaneous vaginal delivery Cesarean section Indication for cesarean section Repeat Fetal distress Dystocia Total patients (No.)

(%)

(%)

(%)

60 40

76 24

74 26

33

0

7

8 6* 10

6 4t 16

15

49

150

*p<0.05 (group I versus group II). tp<0.001 (group I versus control).

and controls are presented in Table III. Although the cesarean section rates for cases and controls were virtually the same, being 28% and 26%, respectively; indications for fetal distress were significantly different between subgroups I and II of cases and between group I and controls.

Comment The findings from this study confirm previous reports 5 -7 that variable decelerations are significantly associated with adverse outcomes. However, in the present study we demonstrated that, in the presence of variable decelerations, accelerations and good variabil-

71 0 Anyaegbunam et al.

ity were associated with good fetal outcomes whereas their absence was correlated with adverse fetal outcomes. Adverse outcomes could not be explained by delayed delivery as 33% of group I were delivered within 24 hours of the abnormal fetal heart tracing compared with no cases in Group II and I% of controls. The cesarean section rate for fetal distress was also significantly higher in the adverse outcome group. The findings also support both the documented association between variable decelerations and cord involvemenf· 13 • 14 and the role of the shape of the variable decelerations. Hon 15 previously observed that Wshaped variable decelerations appeared to be produced by cord compression. We also demonstrated a strong association between W-shaped variable decelerations and nuchal cord. Furthermore, W-shaped variable decelerations have also been reported as significantly associated with greater length of the umbilical cord. 16 Thus a long cord may predispose the fetus to a greater risk of nuchal cord involvement and the possibility of cord compression, which is manifested by W-shaped variable decelerations on the nonstress test. Although many clinicians have hypothesized that variable decelerations during the antepartum period are associated with reduced amniotic fluid and intrauterine growth retardation, no previously reported study confirmed this association. From our work the association between variable decelerations and intrauterine growth retardation approached statistical significance (p < 0.07) requiring confirmation by larger sample sizes. In summary, our results suggest that in the presence of antepartum variable decelerations there is a higher incidence of fetal distress in labor, low Apgar scores, neonatal intensive care unit admissions, and nuchal cord involvement. We also found that the presence of accelerations and normal variability is associated with good neonatal outcome, whereas their absence is associated with adverse outcome. Finally, neither the presence of accelerations nor good variability is independently correlated with neonatal outcome.

October 1986 Am J Obstet Gynecol

REFERENCES 1. Lee CY, DiLoreto PC, Logrand B. Fetal activity acceleration determination for the evaluation of fetal reserve. Obstet Gynecol1976;48:19-26. 2. Evertson LR, Gauthier RJ, Schifrin BS, Paul RH. Antepartum fetal heart rate testing. Evolution of the non stress test. AM J 0BSTET GYNECOL 1979; 133:29-33. 3. Phelan JP. The nonstress test: a review of 3,000 tests. AM J 0BSTET GYNECOL 1981; 139:7-10. 4. Meis PJ, UredaJR, Swain M, Randall KT, Penny M, Penny S. Variable decelerations during nonstress tests are not a sign of fetal compromise. AM J 0BSTET GYNECOL 1986;154:586-90. 5. Pazos R, Vuolo K, Aladjem S, Lueck J, Anderson C. Association of spontaneous fetal heart rate decelerations during antepartum nonstress testing and intrauterine growth retardation. AM J 0BSTET GYNECOL 1982;144: 574-7. 6. Druzin ML, Gratacos J, Keegan KA, Paul RH. Antepartum fetal heart rate testing. VII. The significance of fetal bradycardia. AMJ 0BSTET GYNECOL 1981;139:194-8. 7. PhelanJP, Lewis PEJr. Fetal heart rate decelerations during a nonstress test. Obstet Gynecol1981;57:228-32. 8. Manning FA, Morrison I, Lange IR, Harman CR, Chamberlain PE. Fetal assessment based on fetal biophysical profile scoring: experience in 12,620 referred high-risk pregnancies. I. Perinatal mortality by frequency and etiology. AMJ 0BSTET GYNECOL 1985;151:343-50. 9. Fischer WM, Stude I, Brandt H. Ein Vorschlag zur Beureteilung des antepartualen Kardiotokogramms. Z. Geburtshilfe Perinatol1976;180:117-23. 10. Krebs HB, Petres RE. Clinical application of a scoring system for evaluation of antepartum fetal heart rate monitoring. AM J 0BSTET GYNECOL 1978; 130:765-71. 11. Manning FA, Hill LM, Platt LD. Qualitative amniotic fluid volume determinations by ultrasound: antepartum detection of intrauterine growth retardation. AM J OBSTET GYNECOL 1981;139:254-8. 12. Battaglia FC, Lubchenco LO. A practical classification of newborn infants by weight and gestational age. J Pediatr 1967;71: 159-63. 13. O'Leary JA, Andrinopoulos GC, Giordano PC. Variable decelerations and the nons tress test: an indicator of cord compromise. AMJ 0BSTET GYNECOL 1980;137:704-6. 14. Hon EH, Bradfield AH, Hess OW. The electronic evaluation of the fetal heart rate v. the vagal factor in fetal bradycardia. AMJ 0BSTET GYNECOL 1961;82:291-300. 15. Hon EH. A maneuver for the diagnosis of umbilical cord complications. Obstet Gynecol 1959; 14:154. 16. Welt SI. The fetal heart rate W-sign. Obstet Gynecol 1984;63:405.