- Email: [email protected]
Results of conservative management of premature rupture of the membranes
Results of conservative management of premature rupture of the membranes Janice L. Andreyko, M.D., Chee Pyn Chen, M.D., F.R.C.P.(C), Andrew T. Shennan, M.B., M.R.C.P.(U.K.), F.R.C.P.(C), and John E. Milligan, M.D., F.R.C.S.(C), F.A.C.O.G. Toronto, Ontario, Canada The maternal and fetal outcome of a conservative management protocol, at a tertiary care center, for premature rupture of membranes between 25 and 34 weeks' gestation was reviewed for the 2·year period 1980 to 1981. There were 139 patients with premature rupture of the membranes prior to 37 weeks' gestation, 47 with premature rupture of the membranes less than 24 hours prior to delivery, and 92 in whom premature rupture of the membranes occurred 24 hours or more before delivery. There was a significant difference in the incidence of chorioamnionitis and endometritis between patients in whom premature rupture of the membranes occurred 24 hours or more before delivery and patients in whom delivery took place within 24 hours (p < 0.001). However, neither prolongation of pregnancy with premature rupture of the membranes beyond 24 hours nor use of betamethasone was associated with any increase in maternal or neonatal infectious morbidity. Neonatal mortality was 3.3% and was related only to lower gestational age. (AM. J. OesTET. GVNECOL. 148:600, 1984.)
As a tertiary care center, the Regional Perinatal Unit at Women's College Hospital has adopted a conservative approach to the problem of premature rupture of the membranes prior to 34 weeks' gestation. The current study was undertaken as a retrospective review of all such cases presenting to the Unit over the 2-year period 1980 to 1981. The purpose was to examine maternal and fetal outcomes, with specific reference to morbidity related to gestational age and length of time of the presence of ruptured membranes, to determine whether conservative management was a rational approach to premature rupture of the membranes at this tertiary care center. Specifically examined were the incidences of maternal infectious morbidity, neonatal mortality, and neonatal morbidity as related to bacterial colonization, bacterial sepsis, and respiratory distress syndrome.
Material and methods Patients admitted to the Perinatal Unit with premature rupture of the membranes prior to 34 weeks' gestation were treated conservatively. Premature rupture of the membranes was defined as rupture of the membranes at any time before the onset of labor. The presence of premature rupture of the membranes was
From the Regional Perinatal Unit, Women's College Hospital, University of Toronto Perinatal Complex. Presented at the Thirty-ninth Annual Meeting of The Society of Obstetricians and Gynaecologists of Canada, Vancouver, British Columbia, Canada, june 14-18,1983. Reprint requests: Dr. J. E. Milligan, Obstetrical Director, Regional Perinatal Unit, Women's College Hospital, 76 Grenville Street, Toronto, Ontario, Canada M5S 1B2.
600
based on the mother's history and confirmed by Nitrazine testing of vaginal fluid, usually at the introitus. Delivery was not undertaken until the patient reached 34 weeks' gestation or other complications became superimposed, such as chorioamnionitis, intrauterine growth retardation, or any suggestion of potential or actual fetal hypoxia. Two doses of betamethasone, 12 mg intramuscularly, were given 12 hours apart and then weekly, until 33 weeks' gestation, in an attempt to enhance fetal lung maturity. The patients were also placed on a tocolytic agent, isoxsuprine, for 36 hours to allow the steroid time to take effect. The dosage of isoxsuprine was then tapered and discontinued; if labor began after this, no attempt was made to stop it. The exception was at 25 to 26 weeks' gestation, when neonatal survival rates were still low enough (60% to 70%)1 that an attempt to attain 1 or 2 additional weeks of intrauterine gestation was thought to be justified. If there were no complications and labor did not ensue, labor was induced at 34 weeks' gestation. Specimens from the maternal introitus were cultured for group B {:l-hemolytic streptococci, and patients were given penicillin orally or intravenously (or erythromycin if they were allergic to penicillin), which was discontinued if the cultures were reported as being negative, usually in 24 to 36 hours. If cultures were positive, the antibiotics were continued until the patient had negative cultures. Results of this protocol were examined for the incidence of maternal morbidity, including chorioamnionitis and endometritis. Chorioamnionitis and endometritis were defined as being present if the maternal temperature was greater than 37.5° C with no other
Conservative management of PROM
Volume 148 Number 5
Table I. Comparison of pregnancies with premature rupture of the membranes <24 hours to pregnancies with premature ruptures of the membranes hours prior to delivery
Gestation at premature rupture of the membranes (wk) (mean ± SD) Gestation at delivery (wk) (mean ± SD) Primiparous (%) Introital swab positive for group B streptococci (%) Chorioamnionitis (%) Endometritis (%)
~24
Premature rupture of the membranes ;;.24 hr
(n = 47)
(n = 92)
33.4 ± 2.8
30.0 ± 3.5
33.4 ± 2.8
31.2 ± 3.2
46.8 21.0
41.3 13.0
2.0 2.0
Table II. Mode of delivery in 92 pregnancies <37 weeks' gestation with premature rupture of the membranes ~24 hours No.
Mode of delivery
Premature rupture of the membranes <24 hr
g:~}p < 0.001
Lower-segment cesarean section Previous cesarean section Fetal distress Breech presentation Chorioamnionitis Abruptio placentae Cord prolapse Transverse lie Total
8 8 5 I I I I
25 (27.1%)
Classic cesarean section Breech presentation Transverse lie Fetal distress
12
Total Vaginal delivery
19(20.7%) 48 (52.2%)
5 2
Table III. Maternal infections and organisms isolated with premature rupture of the membranes Infection
Endometritis
hours
Site
Nonhemolytic streptococci Group B hemolytic streptococci
Vaginal swab Vaginal swab Uterocutaneous fistula Uterine swab Uterine swab and blood Placental swab Uterine swab Cervical swab Uterine swab Midstream urine Midstream urine Wound swab
No growth (2 patients) Wound infection
~24
Organism
Bacteroides bivius E. coli E. coli (2 patients) Klebsiella pneumoniae Gonococcus Haemophilus injluenzae
Urinary tract infection
E. coli Klebsiella pneumoniae E. coli
source of fever and/or there were uterine tenderness and foul-smelling vaginal discharge. Neonatal charts were reviewed for the incidence of the respiratory distress syndrome, superficial bacterial colonization, and bacterial sepsis. Respiratory distress syndrome was defined as respiratory difficulty, with a chest x-ray film showing miliary atelectasis, and the need for ventilation for more than 24 hours. If ventilation was required for less than 24 hours, and there were no further respiratory complications, this was defined as transient tachypnea of the newborn. All infants of mothers with premature rupture of the membranes for more than 24 hours underwent a septic workup, which included umbilical and ear swab cultures and cultures of gastric aspirate, blood, and cerebrospinal fluid. Superficial colonization was defined as positive ear, umbilical, or gastric aspirate culture, without clinical illness. Sepsis was defined as bandltotal neutrophil
601
count greater than 0.2 or total leukocyte count <5,000/ mm:!, 2 clinical evidence of pneumonia, or the presence of bacteria in blood or c.erebrospinal fluid. Statistical analyses were made by X2 • Results
Between January 1, 1980, and December 31, 1981, the total number of patients who presented with premature rupture of the membranes under 37 weeks' gestation with singleton pregnancies was 139. Fortyseven had premature rupture of the membranes for less than 24 hours prior to delivery, and 92 had premature rupture of the membranes for 24 hours or more prior to delivery. Twenty-three of the 47 patients (49%) underwent delivery within 24 hours of premature ru pture of the membranes, at more than 34 weeks' gestation, and labor was not inhibited; 21 of the 47 (45%) were in active labor which could not be stopped with a tocolytic agent; two of the 47 (4%) presented
602 Andreyko et al.
March I, 1984 Am. J. Obstet. Gynecol.
Table IV. Comparison of outcomes by gestational age grouping, premature rupture of the membranes ;;;.24 hours
Gestation at delivery (wk)
s27 28-33 34-36
Cesarean section No.
Mean birth weight (gm ± SD)
No.
16 54 22
894 ± 149 1,651 ± 402 2,447 ± 489
9 30 4
I
Maternal infection
%
No.
56 56 18
8 17 4
I
% 50 31.5 18
Table V. Comparison of outcomes by duration of premature rupture of the membranes prior to delivery Duration of premature rupture of membranes (hr)
24-72 73-168 169+
No.
Mean birth weight (gm ± SD)
Mean gestation at delivery (wk ± SD)
40 27 25
1,743 ± 660 1,543 ± 501 1,717 ± 449
31.2 ± 3.2 30.4 ± 2.9 31.6 ± 3.1
with severe fetal distress, and one of the 47 (2%) presented with chorioamnionitis-all accounting for those delivered within 24 hours of presentation. Table I compares those pregnancies in which delivery occurred within 24 hours of premature rupture of the membranes to those in which there was rupture of the membranes for 24 hours or more prior to delivery. There were no maternal deaths in either group. There was a highly significant difference (p < 0.001) between the incidences of maternal morbidity (chorioamnionitis and endometritis) in the two groups. In the group with premature rupture of the membranes less than 24 hours, the incidence of chorioamnionitis and endometritis was 2% (1/47) for each, whereas in the group with premature rupture of the membranes for 24 hours or more, the incidence of chorioamnionitis was 22% (20/ 92) and that of endometritis was 12% (11/92). Since the incidence of maternal infection was so much more significant in the group with premature rupture of the membranes for 24 hours or more, it was this group which was examined in greater detail, including neonatal outcome, and all further data pertain to this group. Table II shows the mode of delivery for patients within this group and the indications for cesarean section. Table III details the maternal infections encountered post partum and the organisms isolated. The most prevalent organisms were gram negative, particularly Escherichia coli. The total number of patients with postpartum infectious morbidity was 14/92 (15.2%). Of the 11 patients with endometritis, six had preceding chorioamnionitis. Two patients had cystitis, and one
Cesarean section No.
20 13 II
I
Maternal infection
%
No.
50 48 44
12 6 9
I
% 30 22 36
had a wound infection. Therefore, the total number of patients with antepartum or postpartum infectious morbidity was 28/92 (30.4%). The use of betamethasone in the patients who developed infection (23/28, 82%) was similar to that in those who did not (56/64, 87.5%). The incidence of neonatal mortality was 3.3% (3/92 infants) with premature rupture of the membranes for 24 hours or more. One infant died of an intracranial hemorrhage at 27 weeks' gestation, one died of severe respiratory distress syndrome at 27 weeks' gestation, and one died of respiratory distress syndrome and sepsis at 27 weeks' gestation. The septic infant had E. coli growing in the blood. The incidence of neonatal morbidity (defined as colonization with bacteria, sepsis, respiratory distress syndrome, or intracranial hemorrhage) was 25.3% (26/ 92). The incidence of minor morbidity (i.e., colonization only) was 5.5% (5/92), and that of major morbidity was 22.8% (21/92). Sixty-two percent of the infants had only transient tachypnea of the newborn, and 12% were completely asymptomatic. The incidence of neonatal respiratory distress syndrome was 15.2% (14/92), and that of neonatal sepsis was 8.7% (8/92). There was no incidence of respiratory distress syndrome beyond 32 weeks' gestation. Tables IV and V document birth weight, cesarean section rate, and maternal and neonatal morbidity for three gestational intervals at delivery and for three intervals of duration of premature rupture of the membranes at delivery. It may be noted from Table IV that the maternal infection rate was highest in the younger gestational age groups. The cesarean section rate was also
Conservative management of PROM
Volume 148 Number 5
603
Neonatal Colonization No.
I
Sepsis
%
No.
5.6 9.1
I 5 2
o 3 2
I
Respiratory distress syndrome
%
No.
6.3 9.3 9.1
7 7
I
% 44 13
o
Mortality No.
T
% 19
3
o o
Neonatal Sepsis
Colonization No.
I
3 I I
%
No.
7.5 3.7 4.0
4 2 2
I
Respiratory distress syndrome
%
No.
10 7.4 8.0
8 2
higher in the 27- to 33-week gestational age range than at 34 to 36 weeks, and this may explain the higher incidence of maternal infection in the form of endometritis. The rate of neonatal sepsis did not vary with gestational age and, as would be expected, the incidence of respiratory distress syndrome was highest in the group which was less than 27 weeks' gestation. Neonatal mortality occurred at 27 weeks only. It is evident from Table V that the cesarean section rate and maternal infection rate did not appear to vary with increasing duration of premature rupture of the membranes beyond 24 hours. Neonatal colonization and sepsis rates, as well as neonatal mortality, did not increase with increased duration of premature rupture of the membranes. Also, the incidence of neonatal respiratory distress syndrome did not decrease with increased duration of premature rupture of the membranes. Comment
The management of patients with premature rupture of the membranes prior to term remains controversial. Immediate delivery entails the risks of immaturity in the infant, particularly the development of the respiratory distress syndrome, and conservative observation raises the concern of placing the mother and fetus at risk for sepsis. Some centers perform amniocentesis at 28 weeks' gestation and deliver any fetus with a mature lecithin/sphingomyelin ratio or with infected amniotic fiuid,3 others administer corticosteroids and deliver the fetus empirically after 48 hours,4 and still others favor nonintervention until evidence of infection becomes manifest or other complications arise. 5 • 6 From this retrospective study at a tertiary care cen-
4
I
%
Mortality No.
20
2
7.4 16
I
I
% 5.0 3.7
o
ter, we have drawn certain conclusions. First, smce there was a significant difference in the incidences of maternal infection between deliveries within 24 hours of premature rupture of the membranes and deliveries after 24 hours of rupture of the membranes, it is evident that maternal infectious morbidity increases if rupture of the membranes persists for up to 24 hours. However, additional prolongation of pregnancy with ruptured membranes beyond 24 hours was not associated with any additional increase in infectious morbidity in mother or infant. This has also been reported by Varner and Galask,' who, with a conservative management protocol, showed no significant increase in amnionitis, endometritis, or neonatal sepsis with increased duration of ruptured membranes. Gibbs and Blanco,S in a recent review of articles on the management of premature rupture of the membranes, quoted seven studies that noted that prolonged rupture of the membranes decreased the incidence of the respiratory distress syndrome, and nine studies that showed no significant effect of duration of premature rupture of the membranes on the respiratory distress syndrome. With the management protocol reported here, there was no evidence that additional duration of ruptured membranes per se lessened the incidence of the respiratory distress syndrome. There has been concern that the use of corticosteroids in patients with premature rupture of the membranes may increase the rate of maternal infection. 9 In this study, there was no evidence that betamethasone in the presence of premature rupture of the membranes increased maternal infectious morbidity. Finally, in this Center, neonatal mortality and mor-
Andreyko et al.
March 1, 1984 Am. J. Obstet. Gynecol.
bidity were related only to lower gestational age, not to the duration of premature rupture of the membranes. At this Unit, we have found that neonatal mortality and long-term major morbidity reach a minimum at 32 weeks' gestation, 1 with no further decrease in this figure beyond 32 weeks. Therefore, the induction of labor in patients who present with premature rupture of the membranes at or beyond 34 weeks' gestation is a logical approach in order to minimize maternal infectious morbidity, since neonatal mortality and morbidity have already reached minimum levels at that stage of gestation. However, neither maternal nor neonatal infectious morbidity increases beyond 24 hours of premature rupture of the membranes, and neonatal mortality and morbidity are related to lower gestational age. Therefore, at earlier stages of gestation, conservative management in a tertiary care unit, with careful surveillance for infection and fetal hypoxia, is a rational approach to the problem of premature rupture of the membranes, in the attempt to achieve further in utero fetal maturation.
REFERENCES 1. MiIligan,J. E., Shennan, A. T., and Hoskins, E. M.: Perinatal intensive care: Where and how to draw the line, AM. J. OBSTET. GYNECOL. In press. 2. Philip, A. G. S., and Hewitt, J. R: Early diagnosis of neonatal sepsis, Pediatrics 65: 1036, 1980. 3. Garite, T. J., Freeman, R. K., Linzey, E. M., et al.: The use of amniocentesis in patients with premature rupture of membranes, Obstet. Gynecol. 54:226, 1979. 4. Mead, P. B., and Clapp, J. E.: The use of betamethasone and timed delivery in management of premature rupture of the membranes in the preterm pregnancy, J. Reprod. Med. 19:3, 1977. 5. Berkowitz, R L., Hoder, E. L., Freeman, R. M., et al.: Results of a management protocol for premature rupture of the membranes, Obstet. Gynecol. 60:271, 1982. 6. Graham, R L., Gilstrap, L. C., Hauth, J. C., et al.: Conservative management of patients with premature rupture of fetal membranes, Obstet. Gynecol. 59:607, 1982. 7. Varner, M. W., and Galask, R P.: Conservative management of premature rupture of the membranes, AM. J. OBSTET. GYNECOL. 140:39, 1981. 8. Gibbs, R S., and Blanco, J. D.: Premature rupture of the membranes, Obstet. Gynecol. 60:671, 1982. 9. Mead, P. B.: Management of the patient with premature rupture of the membranes, Clin. Perinatol. 7:243, 1980.
The cervical cap: A retrospective study of an alternative contraceptive technique Joann M. Johnson, M.D.
Winnipeg, Manitoba, Canada A follow-up study on 130 women fitted with a cervical cap over a 12-month period drew a response rate of 43% (56 respondents). The group was young, well educated, and highly motivated. The failure rate was 16.9 per 100 women years (Pearl method), with inconsistent use and dislodgement being of major importance. The continuation rate was 75% (minimum 3 months of use), and 84% expressed satisfaction with the method. No significant side effects or risks to health were encountered. The conclusion is that there is a significant demand for alternative contraceptive methods among a select group of women. However, in the present state of knowledge, use of the cap should probably not be encouraged as a primary means of contraception, but should be reserved for those women with multiple contraceptive problems or for highly motivated women who seek out this means, understand its limitations, and can accept the relative uncertainty of its effectiveness. (AM. J. OBSTET. GVNECOl. 148:604, 1984.)
In recent years, the lay press and women's health groups have given much attention to the use of the
From the Department of Obstetrics, Gymcology and Reproductive Sciences, University of Manitoba. Presented at the Thirty-ninth Annual Meeting of The Society ofObstetricians and Gynaecologists of Canada, Vancouver, British Columbia, Canada, june 14-18,1983. Reprint requests: Joann M. Johnson, M.D., Department of Obstetrics, Gynecology and Reproductive Sciences, University of Manitoba, WinniPeg, Manitoba, Canada.
604
cervical cap as a contraceptive method. The cap has, however, continued to find only limited acceptance among medical professionals in this country. The reason for this reluctance is mainly a lack of data on its safety, efficacy, and clinical acceptability. To attempt to clarify this, a follow-up study was conducted by means of a questionnaire survey of 130 women fitted with a cervical cap over a 12-rnonth period. The study was designed to answer three questions. What is the effectiveness of the cap in a group of
As a tertiary care center, the Regional Perinatal Unit at Women's College Hospital has adopted a conservative approach to the problem of premature rupture of the membranes prior to 34 weeks' gestation. The current study was undertaken as a retrospective review of all such cases presenting to the Unit over the 2-year period 1980 to 1981. The purpose was to examine maternal and fetal outcomes, with specific reference to morbidity related to gestational age and length of time of the presence of ruptured membranes, to determine whether conservative management was a rational approach to premature rupture of the membranes at this tertiary care center. Specifically examined were the incidences of maternal infectious morbidity, neonatal mortality, and neonatal morbidity as related to bacterial colonization, bacterial sepsis, and respiratory distress syndrome.
Material and methods Patients admitted to the Perinatal Unit with premature rupture of the membranes prior to 34 weeks' gestation were treated conservatively. Premature rupture of the membranes was defined as rupture of the membranes at any time before the onset of labor. The presence of premature rupture of the membranes was
From the Regional Perinatal Unit, Women's College Hospital, University of Toronto Perinatal Complex. Presented at the Thirty-ninth Annual Meeting of The Society of Obstetricians and Gynaecologists of Canada, Vancouver, British Columbia, Canada, june 14-18,1983. Reprint requests: Dr. J. E. Milligan, Obstetrical Director, Regional Perinatal Unit, Women's College Hospital, 76 Grenville Street, Toronto, Ontario, Canada M5S 1B2.
600
based on the mother's history and confirmed by Nitrazine testing of vaginal fluid, usually at the introitus. Delivery was not undertaken until the patient reached 34 weeks' gestation or other complications became superimposed, such as chorioamnionitis, intrauterine growth retardation, or any suggestion of potential or actual fetal hypoxia. Two doses of betamethasone, 12 mg intramuscularly, were given 12 hours apart and then weekly, until 33 weeks' gestation, in an attempt to enhance fetal lung maturity. The patients were also placed on a tocolytic agent, isoxsuprine, for 36 hours to allow the steroid time to take effect. The dosage of isoxsuprine was then tapered and discontinued; if labor began after this, no attempt was made to stop it. The exception was at 25 to 26 weeks' gestation, when neonatal survival rates were still low enough (60% to 70%)1 that an attempt to attain 1 or 2 additional weeks of intrauterine gestation was thought to be justified. If there were no complications and labor did not ensue, labor was induced at 34 weeks' gestation. Specimens from the maternal introitus were cultured for group B {:l-hemolytic streptococci, and patients were given penicillin orally or intravenously (or erythromycin if they were allergic to penicillin), which was discontinued if the cultures were reported as being negative, usually in 24 to 36 hours. If cultures were positive, the antibiotics were continued until the patient had negative cultures. Results of this protocol were examined for the incidence of maternal morbidity, including chorioamnionitis and endometritis. Chorioamnionitis and endometritis were defined as being present if the maternal temperature was greater than 37.5° C with no other
Conservative management of PROM
Volume 148 Number 5
Table I. Comparison of pregnancies with premature rupture of the membranes <24 hours to pregnancies with premature ruptures of the membranes hours prior to delivery
Gestation at premature rupture of the membranes (wk) (mean ± SD) Gestation at delivery (wk) (mean ± SD) Primiparous (%) Introital swab positive for group B streptococci (%) Chorioamnionitis (%) Endometritis (%)
~24
Premature rupture of the membranes ;;.24 hr
(n = 47)
(n = 92)
33.4 ± 2.8
30.0 ± 3.5
33.4 ± 2.8
31.2 ± 3.2
46.8 21.0
41.3 13.0
2.0 2.0
Table II. Mode of delivery in 92 pregnancies <37 weeks' gestation with premature rupture of the membranes ~24 hours No.
Mode of delivery
Premature rupture of the membranes <24 hr
g:~}p < 0.001
Lower-segment cesarean section Previous cesarean section Fetal distress Breech presentation Chorioamnionitis Abruptio placentae Cord prolapse Transverse lie Total
8 8 5 I I I I
25 (27.1%)
Classic cesarean section Breech presentation Transverse lie Fetal distress
12
Total Vaginal delivery
19(20.7%) 48 (52.2%)
5 2
Table III. Maternal infections and organisms isolated with premature rupture of the membranes Infection
Endometritis
hours
Site
Nonhemolytic streptococci Group B hemolytic streptococci
Vaginal swab Vaginal swab Uterocutaneous fistula Uterine swab Uterine swab and blood Placental swab Uterine swab Cervical swab Uterine swab Midstream urine Midstream urine Wound swab
No growth (2 patients) Wound infection
~24
Organism
Bacteroides bivius E. coli E. coli (2 patients) Klebsiella pneumoniae Gonococcus Haemophilus injluenzae
Urinary tract infection
E. coli Klebsiella pneumoniae E. coli
source of fever and/or there were uterine tenderness and foul-smelling vaginal discharge. Neonatal charts were reviewed for the incidence of the respiratory distress syndrome, superficial bacterial colonization, and bacterial sepsis. Respiratory distress syndrome was defined as respiratory difficulty, with a chest x-ray film showing miliary atelectasis, and the need for ventilation for more than 24 hours. If ventilation was required for less than 24 hours, and there were no further respiratory complications, this was defined as transient tachypnea of the newborn. All infants of mothers with premature rupture of the membranes for more than 24 hours underwent a septic workup, which included umbilical and ear swab cultures and cultures of gastric aspirate, blood, and cerebrospinal fluid. Superficial colonization was defined as positive ear, umbilical, or gastric aspirate culture, without clinical illness. Sepsis was defined as bandltotal neutrophil
601
count greater than 0.2 or total leukocyte count <5,000/ mm:!, 2 clinical evidence of pneumonia, or the presence of bacteria in blood or c.erebrospinal fluid. Statistical analyses were made by X2 • Results
Between January 1, 1980, and December 31, 1981, the total number of patients who presented with premature rupture of the membranes under 37 weeks' gestation with singleton pregnancies was 139. Fortyseven had premature rupture of the membranes for less than 24 hours prior to delivery, and 92 had premature rupture of the membranes for 24 hours or more prior to delivery. Twenty-three of the 47 patients (49%) underwent delivery within 24 hours of premature ru pture of the membranes, at more than 34 weeks' gestation, and labor was not inhibited; 21 of the 47 (45%) were in active labor which could not be stopped with a tocolytic agent; two of the 47 (4%) presented
602 Andreyko et al.
March I, 1984 Am. J. Obstet. Gynecol.
Table IV. Comparison of outcomes by gestational age grouping, premature rupture of the membranes ;;;.24 hours
Gestation at delivery (wk)
s27 28-33 34-36
Cesarean section No.
Mean birth weight (gm ± SD)
No.
16 54 22
894 ± 149 1,651 ± 402 2,447 ± 489
9 30 4
I
Maternal infection
%
No.
56 56 18
8 17 4
I
% 50 31.5 18
Table V. Comparison of outcomes by duration of premature rupture of the membranes prior to delivery Duration of premature rupture of membranes (hr)
24-72 73-168 169+
No.
Mean birth weight (gm ± SD)
Mean gestation at delivery (wk ± SD)
40 27 25
1,743 ± 660 1,543 ± 501 1,717 ± 449
31.2 ± 3.2 30.4 ± 2.9 31.6 ± 3.1
with severe fetal distress, and one of the 47 (2%) presented with chorioamnionitis-all accounting for those delivered within 24 hours of presentation. Table I compares those pregnancies in which delivery occurred within 24 hours of premature rupture of the membranes to those in which there was rupture of the membranes for 24 hours or more prior to delivery. There were no maternal deaths in either group. There was a highly significant difference (p < 0.001) between the incidences of maternal morbidity (chorioamnionitis and endometritis) in the two groups. In the group with premature rupture of the membranes less than 24 hours, the incidence of chorioamnionitis and endometritis was 2% (1/47) for each, whereas in the group with premature rupture of the membranes for 24 hours or more, the incidence of chorioamnionitis was 22% (20/ 92) and that of endometritis was 12% (11/92). Since the incidence of maternal infection was so much more significant in the group with premature rupture of the membranes for 24 hours or more, it was this group which was examined in greater detail, including neonatal outcome, and all further data pertain to this group. Table II shows the mode of delivery for patients within this group and the indications for cesarean section. Table III details the maternal infections encountered post partum and the organisms isolated. The most prevalent organisms were gram negative, particularly Escherichia coli. The total number of patients with postpartum infectious morbidity was 14/92 (15.2%). Of the 11 patients with endometritis, six had preceding chorioamnionitis. Two patients had cystitis, and one
Cesarean section No.
20 13 II
I
Maternal infection
%
No.
50 48 44
12 6 9
I
% 30 22 36
had a wound infection. Therefore, the total number of patients with antepartum or postpartum infectious morbidity was 28/92 (30.4%). The use of betamethasone in the patients who developed infection (23/28, 82%) was similar to that in those who did not (56/64, 87.5%). The incidence of neonatal mortality was 3.3% (3/92 infants) with premature rupture of the membranes for 24 hours or more. One infant died of an intracranial hemorrhage at 27 weeks' gestation, one died of severe respiratory distress syndrome at 27 weeks' gestation, and one died of respiratory distress syndrome and sepsis at 27 weeks' gestation. The septic infant had E. coli growing in the blood. The incidence of neonatal morbidity (defined as colonization with bacteria, sepsis, respiratory distress syndrome, or intracranial hemorrhage) was 25.3% (26/ 92). The incidence of minor morbidity (i.e., colonization only) was 5.5% (5/92), and that of major morbidity was 22.8% (21/92). Sixty-two percent of the infants had only transient tachypnea of the newborn, and 12% were completely asymptomatic. The incidence of neonatal respiratory distress syndrome was 15.2% (14/92), and that of neonatal sepsis was 8.7% (8/92). There was no incidence of respiratory distress syndrome beyond 32 weeks' gestation. Tables IV and V document birth weight, cesarean section rate, and maternal and neonatal morbidity for three gestational intervals at delivery and for three intervals of duration of premature rupture of the membranes at delivery. It may be noted from Table IV that the maternal infection rate was highest in the younger gestational age groups. The cesarean section rate was also
Conservative management of PROM
Volume 148 Number 5
603
Neonatal Colonization No.
I
Sepsis
%
No.
5.6 9.1
I 5 2
o 3 2
I
Respiratory distress syndrome
%
No.
6.3 9.3 9.1
7 7
I
% 44 13
o
Mortality No.
T
% 19
3
o o
Neonatal Sepsis
Colonization No.
I
3 I I
%
No.
7.5 3.7 4.0
4 2 2
I
Respiratory distress syndrome
%
No.
10 7.4 8.0
8 2
higher in the 27- to 33-week gestational age range than at 34 to 36 weeks, and this may explain the higher incidence of maternal infection in the form of endometritis. The rate of neonatal sepsis did not vary with gestational age and, as would be expected, the incidence of respiratory distress syndrome was highest in the group which was less than 27 weeks' gestation. Neonatal mortality occurred at 27 weeks only. It is evident from Table V that the cesarean section rate and maternal infection rate did not appear to vary with increasing duration of premature rupture of the membranes beyond 24 hours. Neonatal colonization and sepsis rates, as well as neonatal mortality, did not increase with increased duration of premature rupture of the membranes. Also, the incidence of neonatal respiratory distress syndrome did not decrease with increased duration of premature rupture of the membranes. Comment
The management of patients with premature rupture of the membranes prior to term remains controversial. Immediate delivery entails the risks of immaturity in the infant, particularly the development of the respiratory distress syndrome, and conservative observation raises the concern of placing the mother and fetus at risk for sepsis. Some centers perform amniocentesis at 28 weeks' gestation and deliver any fetus with a mature lecithin/sphingomyelin ratio or with infected amniotic fiuid,3 others administer corticosteroids and deliver the fetus empirically after 48 hours,4 and still others favor nonintervention until evidence of infection becomes manifest or other complications arise. 5 • 6 From this retrospective study at a tertiary care cen-
4
I
%
Mortality No.
20
2
7.4 16
I
I
% 5.0 3.7
o
ter, we have drawn certain conclusions. First, smce there was a significant difference in the incidences of maternal infection between deliveries within 24 hours of premature rupture of the membranes and deliveries after 24 hours of rupture of the membranes, it is evident that maternal infectious morbidity increases if rupture of the membranes persists for up to 24 hours. However, additional prolongation of pregnancy with ruptured membranes beyond 24 hours was not associated with any additional increase in infectious morbidity in mother or infant. This has also been reported by Varner and Galask,' who, with a conservative management protocol, showed no significant increase in amnionitis, endometritis, or neonatal sepsis with increased duration of ruptured membranes. Gibbs and Blanco,S in a recent review of articles on the management of premature rupture of the membranes, quoted seven studies that noted that prolonged rupture of the membranes decreased the incidence of the respiratory distress syndrome, and nine studies that showed no significant effect of duration of premature rupture of the membranes on the respiratory distress syndrome. With the management protocol reported here, there was no evidence that additional duration of ruptured membranes per se lessened the incidence of the respiratory distress syndrome. There has been concern that the use of corticosteroids in patients with premature rupture of the membranes may increase the rate of maternal infection. 9 In this study, there was no evidence that betamethasone in the presence of premature rupture of the membranes increased maternal infectious morbidity. Finally, in this Center, neonatal mortality and mor-
Andreyko et al.
March 1, 1984 Am. J. Obstet. Gynecol.
bidity were related only to lower gestational age, not to the duration of premature rupture of the membranes. At this Unit, we have found that neonatal mortality and long-term major morbidity reach a minimum at 32 weeks' gestation, 1 with no further decrease in this figure beyond 32 weeks. Therefore, the induction of labor in patients who present with premature rupture of the membranes at or beyond 34 weeks' gestation is a logical approach in order to minimize maternal infectious morbidity, since neonatal mortality and morbidity have already reached minimum levels at that stage of gestation. However, neither maternal nor neonatal infectious morbidity increases beyond 24 hours of premature rupture of the membranes, and neonatal mortality and morbidity are related to lower gestational age. Therefore, at earlier stages of gestation, conservative management in a tertiary care unit, with careful surveillance for infection and fetal hypoxia, is a rational approach to the problem of premature rupture of the membranes, in the attempt to achieve further in utero fetal maturation.
REFERENCES 1. MiIligan,J. E., Shennan, A. T., and Hoskins, E. M.: Perinatal intensive care: Where and how to draw the line, AM. J. OBSTET. GYNECOL. In press. 2. Philip, A. G. S., and Hewitt, J. R: Early diagnosis of neonatal sepsis, Pediatrics 65: 1036, 1980. 3. Garite, T. J., Freeman, R. K., Linzey, E. M., et al.: The use of amniocentesis in patients with premature rupture of membranes, Obstet. Gynecol. 54:226, 1979. 4. Mead, P. B., and Clapp, J. E.: The use of betamethasone and timed delivery in management of premature rupture of the membranes in the preterm pregnancy, J. Reprod. Med. 19:3, 1977. 5. Berkowitz, R L., Hoder, E. L., Freeman, R. M., et al.: Results of a management protocol for premature rupture of the membranes, Obstet. Gynecol. 60:271, 1982. 6. Graham, R L., Gilstrap, L. C., Hauth, J. C., et al.: Conservative management of patients with premature rupture of fetal membranes, Obstet. Gynecol. 59:607, 1982. 7. Varner, M. W., and Galask, R P.: Conservative management of premature rupture of the membranes, AM. J. OBSTET. GYNECOL. 140:39, 1981. 8. Gibbs, R S., and Blanco, J. D.: Premature rupture of the membranes, Obstet. Gynecol. 60:671, 1982. 9. Mead, P. B.: Management of the patient with premature rupture of the membranes, Clin. Perinatol. 7:243, 1980.
The cervical cap: A retrospective study of an alternative contraceptive technique Joann M. Johnson, M.D.
Winnipeg, Manitoba, Canada A follow-up study on 130 women fitted with a cervical cap over a 12-month period drew a response rate of 43% (56 respondents). The group was young, well educated, and highly motivated. The failure rate was 16.9 per 100 women years (Pearl method), with inconsistent use and dislodgement being of major importance. The continuation rate was 75% (minimum 3 months of use), and 84% expressed satisfaction with the method. No significant side effects or risks to health were encountered. The conclusion is that there is a significant demand for alternative contraceptive methods among a select group of women. However, in the present state of knowledge, use of the cap should probably not be encouraged as a primary means of contraception, but should be reserved for those women with multiple contraceptive problems or for highly motivated women who seek out this means, understand its limitations, and can accept the relative uncertainty of its effectiveness. (AM. J. OBSTET. GVNECOl. 148:604, 1984.)
In recent years, the lay press and women's health groups have given much attention to the use of the
From the Department of Obstetrics, Gymcology and Reproductive Sciences, University of Manitoba. Presented at the Thirty-ninth Annual Meeting of The Society ofObstetricians and Gynaecologists of Canada, Vancouver, British Columbia, Canada, june 14-18,1983. Reprint requests: Joann M. Johnson, M.D., Department of Obstetrics, Gynecology and Reproductive Sciences, University of Manitoba, WinniPeg, Manitoba, Canada.
604
cervical cap as a contraceptive method. The cap has, however, continued to find only limited acceptance among medical professionals in this country. The reason for this reluctance is mainly a lack of data on its safety, efficacy, and clinical acceptability. To attempt to clarify this, a follow-up study was conducted by means of a questionnaire survey of 130 women fitted with a cervical cap over a 12-rnonth period. The study was designed to answer three questions. What is the effectiveness of the cap in a group of