The use of breast stimulation to prevent postdate pregnancy

The use of breast stimulation to prevent postdate pregnancy John P. Elliott, M.D., Lieutenant Colonel, MC, USA, and James F. Flaherty, D.O., Captain, MC, USA San Francisco, California Postdate pregnancy is estimated to occur in 3% to 12% of all gestations. Morbidity and mortality rates associated with this common obstetric problem are higher than those with term gestation. The incidence of fetal distress, birth injury, meconium aspiration, congenital malformations, macrosomia, and oligohydramnios is also greater in postdate pregnancy. We prospectively evaluated breast self-stimulation to determine its effect on the incidence of postdate pregnancy. Two hundred low-risk patients at 39 weeks' gestation were randomly assigned to either a control group or a breast-stimulation group. Results showed that breast stimulation reduced the number of pregnancies managed as postdates from 17 per 100 (17%) to five per 100 (5%) (p < 0.01 ), a 70% reduction. It is concluded that breast stimulation in postdates pregnancies can decrease significantly the number of patients that must be monitored by biochemical or biophysical means. (AM. J. 0BSTET. GYNECOL. 149:628, 1984.)

The most common reason for a pregnancy to become high risk is its extension beyond 42 weeks (294 days). The incidence of postdate pregnancy determined retrospectively varies from 2.9% to 12%,1-5 because in many cases, it is difficult to determine the due date accurately. Overdiagnosis of postdate pregnancy is a frequent problem in clinical practice. Rates of morbidity and mortality in the fetus associated with true postdate pregnancy are higher than those with delivery at term. 1• 2 • 4 - 6 Zwerdling2 reported a significant increase in mortality among postterm infants through 2 years of age. The importance of postdate pregnancy is also highlighted by the medicolegal consequences. Freeman (personal communication), in reviewing his experience of testifying for the defense in malpractice suits, reported that approximately 40% of obstetric malpractice cases involve postterm pregnanCies. Various management protocols have been evaluated in an attempt to select those pregnancies truly at risk for fetal mortality. Antepartum testing methods include the nonstress test (NST), 3 • 6 - 8 the contraction stress test (CST), 3 • 9 - 11 estrogens, 3 • 7 · 11 human chorionic somatomammotropin, 12 and amniocentesis. 10 • 13 None of these methods has been shown to be completely successful in preventing mortality and morbidity.

From the Department of Obstetrics and Gynecology, Letterman Army Medical Center. Presented at the meeting of the Armed Forces District of the American College of Obstetricians and Gynecologists, Las Vegas, Nevada, October 9-13, 1983. The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of the Army or the Department of Defense. Received for publication July 21, 1983; revised December 28, 1983; accepted january 31, 1984. Reprint requests: Technical Publications Editor, Letterman Army Medical Center, Presidio of San Francisco, CA 94129.

628

Routine induction of labor as an alternative in postdates pregnancies has not improved outcome and is associated with an increased rate of cesarean section. I. 4 ' 5 We 14 recently reported the successful use of breast self-stimulation to ripen the cervix in term pregnancies. We have performed a prospective trial to determine the effect of breast stimulation on the incidence of postdate pregnancy.

Material and methods Two hundred obstetric patients at approximately 39 weeks' gestation participated in this study conducted at Letterman Army Medical Center, San Francisco, California. All patients had uncomplicated prenatal courses and were at low risk for uteroplacental insufficiency. The study design was reviewed and approved by the Research and Human Use Committees at Letterman Army Medical Center. The nature of the study was explained and informed consent was obtained from each patient. Gestational age was determined by the best obstetric estimate of date of confinement, with the use of a reliable menstrual history (normal, predictable, cyclic, spontaneous menses), no recent use of oral contraceptives, an early pregnancy test, an early vaginal estimation of uterine size, fetal heart auscultation at 20 weeks, and obstetric sonograms. All 200 patients met the minimum criteria for 39 weeks' gestation suggested by the American College of Obstetricians and Gynecologists under its standards for performing a repeat cesarean section without evidence of fetal pulmonary maturity. Patients were randomly assigned to treatment or control groups based on a table of random numbers. One hundred patients were assigned to each group. Patients in the control group had a pelvic examination, and a Bishop score was assigned. These patients were asked to abstain from sexual intercourse and to avoid

Breast stimulation to prevent postdate pregnancy 629

Volume 149 Number 6

Table I. Characteristics of treatment and control groups Characteristic

Maternal age (yr) Parity Gestational age :s42 wk at delivery Birth weight (gm) Initial Bishop score NS

=

Breast stimulation (mean ± SD)

Control (mean ± SD)

Level of significance

25.0 ± 4.75 0.79 ± l.04 40 wk, 5.68 days ± 4.81 days

24.4 ± 4.88 0.84 ± l.10 40 wk, 5.45 days ± 4.62 days

NS NS NS

3594 ± 441 4.67 ± 2.27

3649 ± 394 4.15 ± 2.34

NS NS

Not significant.

breast stimulation. They were then evaluated weekly until 42 weeks' gestation, at which time a Bishop score was again determined. Patients with a score of ;:::8 were considered to have a favorable cervix and underwent induction of labor. Patients with a Bishop score of <8 were managed with a CST. Patients with a reactive negative CST were randomly assigned a second time into treatment and control groups, and management was according to the original protocol, with weekly pelvic examinations and CSTs. A ripe cervix or an abnormal CST was a reason for inducing labor. Patients in the treatment group were also assigned an initial Bishop score. These patients were instructed in breast self-stimulation, to be performed at home, according to the technique we 14 described. This technique consists of manually stimulating the nipple, areola, and distal breast with the balls of the fingertips. Breasts are stimulated, one at a time, for 15 minutes, alternating breasts, for a total of 1 hour. Skin creams or lubricants may be used. Our patients were encouraged to stimulate the breasts for 3 hours a day, based on the clinical impression that this is a "dose-related" response. The patients recorded the number of hours they stimulated the breasts. They were then evaluated routinely until 42 weeks' gestation, when a Bishop score was again determined. If the Bishop score was ;:::8, labor was induced. Patients whose Bishop scores were <8 were managed with a CST. A reactive (negative) CST allowed another week of noninterference; a ripe cervix or abnormal CST was a reason for intervention. Treatment and control groups were compared with respect to incidence of postdate gestation, labor and delivery outcome for mother, neonatal morbidity and mortality, and length of first and second stage of labor. The relationship between the amount of breast stimulation performed and the length of pregnancy was evaluated. Statistical analysis was performed by means of x2 , Student's t test, the Mann-Whitney U test, and Fisher's exact test as appropriate. Results

No perinatal or neonatal death occurred in the study group; significantly, no patients were admitted with

Table II. Effect of breast stimulation on postdate pregnancy

Group

Control Breast stimulation

No. of patients

Spontaneous labor at :542 wk or medical induction

Induction of labor at 42 wk

Completed 42 wk of gestation with Bishop score <8

100 100

79 92

4 3

17* 5*

*Significant at p < O.Ol. fetal distress diagnosed immediately. A comparison of patient characteristics in both groups showed no difference in the initial Bishop scores between the two groups (Table 1). Breast stimulation resulted in a significant reduction in the number of patients who reached 42 weeks' gestation with an unripe cervix (Table II). Table III compares the outcome for mothers and infants with delivery at :S42 weeks with that of those delivered at >42 weeks' gestation. The incidence of neonatal dysmaturity, as defined by Clifford,l5 in patients reaching 42 weeks' gestation was eight of 22 (36.4%). The presence of meconium was significantly greater in the gestations which were felt to be >42 weeks (p < 0.05). The rate of cesarean section was 9% in patients delivered at <42 weeks' gestation compared to 22.7% in those delivered at >42 weeks (p < 0.02). Comparison of the length of labor in patients who performed breast self-stimulation with that in control subjects showed no significant difference between the two groups (Table IV). At 42 weeks' gestation, those patients who were in the initial control group were again randomly divided into breast stimulation and control groups. The outcome in these patients is summarized in Table V. In an attempt to determine whether the amount of time spent in stimulating the breasts was related to duration of pregnancy after entry into the study, we examined the patients in the breast stimulation group who were delivered at :S42 weeks. Using 3 hours of stimulation per day as a criterion, we found a significant difference in the mean days to delivery for patients who performed stimulation ;:::3 hours per day

630

Elliott and Flaherty

July 15, 1984 Am. J. Obstet. Gynecol.

Table III. Labor and delivery data: Patients delivered at :s42 weeks' gestation compared with patients delivered at >42 weeks' gestation Delivery :542 wk Breast stimulation Outcome

No.

Spontaneous labor Induction of labor Ripe cervix Pregnancy induced hypertension Fetal distress Cesarean section Failure to progress/cephalopelvic disproportion Fetal distress Other Large for gestational age (<4000 gm) Fetal distress in labor Meconium in labor Meconium aspiration Apgar score <7 1 min 5 min Dysmature infant Death

89 6 3 2 1 9 6 1 2 18 13 25 0 6 1 0 0

I

%

Delivery >42 wk Breast stimulation

Control No.

I

%

9

18.9 13.6 26 6 1 0 0

1 0 0 0

I

%

5 0

77

6 4 1 1 5 3 1 1 18 9 22 0

No.

6

0

0

21.7 10.8 26

1 0 0 0

25 0 0

1 0 0 0

1 0 3 0

20 0 60 0

Control No.

15 2 1 1 0 5 2 1 2 1 3 11 0 2 0 5 0

I

%

29

5.8 16.7 64* 12 0 29 0

*Significant at p < O.Ol.

Table IV. Comparison of length of labor in study patients delivered vaginally at :s42 weeks

Patient

Multiparous First stage (hr) Second stage (min) Nulliparous First stage (hr) Second stage (min) NS

=

Breast stimulation (mean± SD)

Control (mean± SD)

Level of significance

5.30 ± 2.72 20.3 ± 28.2

5.43 ± 3.01 16.1 ± 13.5

NS NS

8.13 ± 3.98 74.2 ± 63.0

8.77 ± 4.1 84.5 ± 69.0

NS NS

Not significant.

when compared with those who performed stimulation <3 hours per day (Table VI).

Comment Postdate pregnancy represents a common but serious problem in obstetrics. Zwerdling, 2 in studying a large obstetric population (400,000 births), found increased rates of perinatal mortality, fetal distress, birth injury, meconium aspiration, and congenital malformations in the postdate pregnancy group. Follow-up studies showed a continued increase in the perinatal mortality rate among postdates infants up to 2 years of age; however, at 5 years of age, no differences in mortality rate between babies born at term and those with postdate delivery were noted. Vorherr 1 added an in-

creased risk of macrosomia and oligohydramnios to the problems of postdate delivery. Postdates pregnancies are at higher risk than term pregnancies for several reasons. First, in some patients, the placenta continues to function normally beyond term, and the fetus continues to grow. This situation may lead to an increased incidence of macrosomia, as documented by Freeman et al., 9 who reported a 25% incidence of birth weight >4,000 gm in postdates pregnancies. In their series, 10 cases (2%) of shoulder dystocia occurred in the postdates infants delivered vaginally. Thus, neonates that are large for gestational age contribute to the increased morbidity of postdates pregnancies. Second, some placentas deteriorate functionally with resultant placental insufficiency that may contribute to increased antepartum and intrapartum fetal loss and morbidity. Third, increasing placental dysfunction leads to a decrease in volume of amniotic fluid, leaving the cord vulnerable to accidents. The finding of meconium in 25% to 43% of postdates pregnancies2· 7 • 11 • 16 is consistent with our finding of 50% but disagrees with the low incidence of 13% reported by Green and Paul. 13 This difference may be related to the success of identifying the true postdate pregnancy. In the 22 patients in our study who were considered as having postdates pregnancies, meconium was present in 50%, and the diagnosis of dysmaturity as defined by Clifford 15 was made in 36% and is consistent with that quoted in the literature for dysmaturity (20% to 40% ). 1 Neonates of postdates pregnancies are at risk before,

Breast stimulation to prevent postdate pregnancy

Volume 149 Number 6

631

Table V. Patients with postdates pregnancies randomized to breast stimulation and control Spontaneous labor S.43 wk Group

No. of patients*

No.

Breast stimulation Control

6 8

4 5

I

Induction of labor s.43 wk with ripe cervix

%

No.

67 63

2 1

I

43 wk with unripe cervix

Abnormal

CST

%

No.

33 13

0 I

l

%

No.

0

0 2

13

I

% 0

25t

*Numbers too small for statistical comparison. tOne patient with ripe cervix had labor induced at 43 weeks, 4 days. One patient with suspicious CST had labor induced at 44 weeks.

during, and after delivery. Depp 17 stated that 85% of fetal deaths related to postmaturity were intrauterine rather than neonatal. Several authors 16 • 18 have documented no increased incidence of fetal distress during labor; others 9 • 11 have noted a significant increase in fetal distress. Miller and Read 16 emphasized the dangerous combination of meconium and fetal distress, which leads to a particularly poor outcome. Immediate suctioning of the neonate's oral pharynx with a De Lee suction trap has improved perinatal outcome by reducing the incidence of meconium aspiration syndrome. 19 Management efforts to date have focused on identifying the fetus at risk for antepartum or intrapartum death with selective intervention. The best result with postdate management protocols has been reported by Freeman et al., 9 with no perinatal mortality when the CST was used for surveillance. Others3 • 10 • 11 using the CST have reported a perinatal mortality rate of 0.6% to 1.1 %. The NST has been reported to be associated with a perinatal mortality rate of 1.0% to 2.4% in postdate pregnancy. Estriol or the estrogen/creatinine ratio has a reported perinatal mortality rate of 0.23% to 0.9%. 3 • 7 Although the perinatal mortality rate is low with most of these techniques, fetal losses still occur, and intervention is necessary in some pregnancies for fetal distress. Routine induction oflabor in the postdate pregnancy to prevent potential fetal loss has not proved successful.4· 5 While there is no change in perinatal mortality rate, the cesarean section rate increases for several reasons. Neonates that are large for gestational .age may develop cephalopelvic disproportion, and fetal distress may be increased. The most significant factor, however, is probably that the frequency of an unripe cervix approaches 70% in patients with postdate pregnancy,' leading to a higher cesarean section rate for failure to progress. We 14 previously documented a significant increase in the number of patients in labor (45%) and a significant change in the mean Bishop score (2.4 points) in patients who performed breast self-stimulation over a 3-day period when compared with control subjects. These qualities were believed to be useful in reducing

Table VI. Relationship between amount of breast stimulation and onset of labor at :S42 weeks Breast stimulation (hrlday)

No. of patients

Days to delivery (mean± SD)

>3 <3

18 77

4.61 ± 3.53* 8.54 ± 5.03*

*Student's t test: p < 0.0004. the number of patients who are considered to have postdates pregnancies and, therefore, in reducing the number at risk for complications. Our data in the present study are consistent with a significantly decreased incidence of pregnancies considered postdates. Our finding that 5% of breast-stimulated patients compared with 17% of control subjects reached 42 weeks' gestation with an unripe cervix supports the findings in our initial paper. This reduction in postdates pregnancies was achieved with no perinatal deaths. The data suggest that the technique is also useful when a patient reaches postdate gestation; however, the number of patients is too small to suggest using this technique once postdate pregnancy has occurred (Table V). Our clinical impression that labor was shorter in patients who performed breast self-stimulation was not confirmed by this study, which documented only a slight trend to shorter labor in both nulliparous and multiparous patients (Table IV). We found it difficult to motivate our patients to stimulate the breasts as much as 3 hours per day. Only 18 of 95 (19%) patients dehvered at :s.42 weeks stimulated the breasts for that amount of time, and none of the five with failure of the technique stimulated the breasts 2::3 hours per day. Table IV shows that patients who stimulated the breasts for 2::3 hours per day had a mean days-to-delivery time of 4.61 compared with 8.54 for the group with less frequent stimulation. This is highly significant at the p < 0.0004 level. These data support a direct positive correlation between breast stimulation and time of delivery and suggest that the length of 2::3 hours per day is important.

632

Elliott and Flaherty

July 15, 1984

Am.]. Obstet. Gynecol.

The use of breast stimulation at home presumes a normal, healthy, term pregnancy. Because breast stimulation performed at home represents an unmonitored situation, the stress of the contractions might worsen preexisting fetal distress in high-risk pregnancies. Furthermore, breast stimulation can cause uterine hypertonus which might be detrimental to fetuses with preexisting uteroplaceiual insufficiency. Patients must be carefully selected for this treatment. We recommend daily fetal movement counts for patients who are stimulating the breasts at term. In questionable cases, a breast stimulation stress test should be performed before the patient is instructed to use this technique at home. Postdate pregnancy presents a problem to all obstetricians and pediatricians. Instructing low-risk patients at term in the technique of breast self-stimulation leads to approximately a 70% reduction in the incidence of postdates pregnancies that must be considered "high risk" and monitored by management protocol. The excellent outcome of normal term pregnancies as documented in this study supports the concept of safe delivery at term rather than surveillance by any method after term. On the basis of our experience in more than 250 pregnancies in which patients have stimulated the breasts with no known adverse effects, we feel that the safety of the technique is excellent. Adverse fetal effects appear to be minimal. We believe that the benefits for the fetus in preventing postdate delivery outweigh unknown potential risks. For consistent results, breast stimulation should be performed ~3 hours per day. Since obstetric and perinatal complications do occur in prolonged pregnancy not accompanied by meconium or a positive CST, research into alternative methods of fetal surveillance and obstetric management is necessary. REFERENCES 1. Vorherr H. Placental insufficiency in relation to postterm pregnancy and fetal postmaturity. AM J 0BSTET GYNECOL 1975;123:67.

2. Zwerdling MA. Factors pertammg to prolonged pregnancy and its outcome. Pediatrics 1967;40:202. 3. Khouzami VA, Johnson JWC, Daikoku NH, Rotmensch ], Hernandez E. Comparison of urinary estrogens, contraction stress tests and nons tress tests in the management of post-term pregnancies. J Reprod Med 1983;28: 189. 4. Gibb DMF, Cardozo LD, Studd JWW, Cooper DJ. Prolonged pregnancy: is induction of labor indicated? Br J Obstet Gynaecol 1982;89:292. 5. Nakano R. Post-term pregnancy. Acta Obstet Gynecol Scand 1972;51:217. 6. Miyazaki FS, Miyazaki BA. False reactive nons tress tests in postterm pregnancies. AM J 0BSTET GYNECOL 1981; 140:269. 7. Yeh S, Read JA. Management of post-term pregnancy in a large obstetrical population. Obstet Gynecol 1982; 60:282. 8. Eden RD, Gergely RZ, Schifrin BS, Wade ME. Comparison of antenatal testing schemes for the management of postdate pregnancy. AM J 0BSTET GYNECOL 1982;144: 683. 9. Freeman RK, Garite TJ, Modanlou H, Dorchester W, Rommal C, DeVaney M. Postdate pregnancy: utilization of contraction stress testing for primary fetal surveillance. AM J 0BSTET GYNECOL 1981; 140: 128. 10. Knox GE, HuddlestonJF, Flowers CEJr. Management of prolonged pregnancy: results of a prospective randomized trial. AMJ 0BSTET GYNECOL 1979;134:376. 11. Schneider JM, Olson RW, Curet LB. Screening for fetal and neonatal risk in the postdate pregnancy. AM J 0BSTET GYNECOL 1978;131:473. 12. Berkowitz RL, Hobbins JC. A reevaluation of the value of hCS determination in the management of prolonged pregnancy. Obstet Gynecol 1977;49: 156. 13. Green JN, Paul RH. The value of amniocentesis in prolonged pregnancy. Obstet Gynecol 1978;51:293. 14. Elliott JP, Flaherty JF. The use of breast stimulation to ripen the cervix in term pregnancies. AM J 0BSTET GYNECOL 1983;145:553. 15. Clifford SH. Postmaturity-with placental dysfunction: clinical syndrome and pathological findings. J Pediatr 1954;44: 1. 16. Miller FC, Read JA. Intrapartum assessment of the postdate fetus. AM J 0BSTET GYNECOL 1981; 141:516. 17. Depp R. The puzzle of postmaturity. Contemp. Ob/Gyn 1974;3:109. 18. Klapholz H, Friedman EA. The incidence of intrapartum fetal distress with advancing gestational age. AM J 0BSTET GYNECOL 1977;127:405. 19. Carson BS, Losey RW, Bowes WA Jr, Simmons MA. Combined obstetric and pediatric approach to prevent meconium aspiration syndrome. AM J 0BSTET GYNECOL 1976;126:712.