| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Department of Obstetrics and Gynaecology, Sunnybrook and Women’s College Health Sciences Centre, the Faculty of Nursing, Maternal Infant, and Reproductive Health Research Unit at the Centre for Research in Women’s Health, University of Toronto, Toronto, Ontario, Canada; Departments of Obstetrics and Gynaecology, St Joseph’s Health Centre, University of Western Ontario, London, Ontario, Canada; and St. Boniface Hospital, University of Manitoba, Winnipeg, Manitoba, Canada.
Address reprint requests to: Mary Elizabeth Hannah, MDCM University of Toronto Centre for Research in Women’s Health Maternal, Infant, and Reproductive Health Research Unit 790 Bay Street, Suite 751 Toronto, ON M5G 1N8 Canada E-mail: mary.hannah{at}utoronto.ca
 |
Abstract |
|---|
Top
 Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Methods: We undertook a secondary analysis of data from the International TermPROM Study for women managed expectantly at home or in a hospital. Using multiple logistic regression analyses, we determined the effect of home and hospital management and controlled for differences in baseline characteristics, in measures of maternal and neonatal infections and rates of cesarean.
Results: Six hundred fifty-three women (39.1%) were managed at home, and 1017 (60.9%) in a hospital. Management at home, compared with in a hospital, increased risk of nulliparas needing antibiotics before delivery (odds ratio [OR] 1.52 95% confidence interval [CI] 1.04, 2.24, P = .03), those not colonized with group B streptococcus having cesareans (OR 1.48 95% CI 1.03, 2.14, P = .04), and neonatal infections (OR 1.97 95% CI 1.00, 3.90, P = .05). More multiparas managed at home said they would participate in the study again (OR 1.80 95% CI 1.27, 2.54, P < .001).
Conclusion: Expectant management at home, rather than in a hospital, might increase the likelihood of some adverse outcomes.
Approximately 5–10% of all pregnancies are complicated by premature rupture of membranes (PROM), 60% of which occur at term.1 In preterm pregnancies with PROM, management is usually expectant (ie, waiting for spontaneous onset of labor as long as there is no evidence of maternal or fetal compromise) because risks of preterm birth usually outweigh risks of infection. In term pregnancies, induction of labor with intravenous (IV) oxytocin usually is recommended if women are not in labor within reasonable time, which reduces risks of maternal infection2,3 and neonatal infection if women are colonized with group B streptococcus.4 Expectant management is a reasonable option for women with PROM at term if they are not colonized with group B streptococcus and wish to avoid labor induction. When women choose expectant management, the question is whether she and her infant will have better outcomes if management is at home rather than in a hospital.
Women were managed expectantly in hospitals in almost all published randomized controlled trials that compared expectant management with induction of labor for women with PROM at term.3,5,6 Information on the effects of expectant management at home for women with PROM at term is limited. We wished to compare maternal and neonatal outcomes of expectant management for those managed at home with those managed in a hospital.
|
Materials and Methods |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Among 3333 women enrolled between January 15, 1994 and May 31, 1995, additional information was collected on whether they were managed partially or completely at home or in hospital. Among those women, 1670 were assigned to expectant management. Women who were managed partially or completely at home after entry were considered managed at home. Partial management at home included women admitted initially to hospitals, subsequently discharged, and later readmitted when labor was established, or initially referred home and later admitted for continuing expectant management. Others were managed completely in hospital. We compared baseline characteristics at entry (maternal age at least 35 years, gestational age at least 41 weeks, prior cesarean, cervical dilatation less than 3 cm, cervical effacement less than 80%, unripe cervix [<3 cm dilated and <80% effaced], previous digital or speculum vaginal examination after membrane rupture, nulliparity, maternal colonization with group B streptococcus, smoking during pregnancy, and duration of membrane rupture) for women in the home and hospital groups using 
2 test.
Principal outcomes were clinical chorioamnionitis, postpartum fever (temperature above 38C), neonatal infection, cesarean rates, and women’s evaluations of the care they received. Clinical chorioamnionitis was defined as maternal temperature before or during labor above 37.5C on two occasions at least 1 hour apart, or above 38C; maternal white blood cell (WBC) count above 20,000/mm3; or foul-smelling AF. Neonatal infection was defined as clinical signs of infection and a positive central culture or antigen detection test, chest radiograph compatible with pneumonia, positive Gram stain of cerebrospinal fluid (CSF), or histologic diagnosis of pneumonia (definite infection); or a high or low neutrophil count, a high ratio of immature neutrophils to total neutrophils, a high immature neutrophil count, or abnormal CSF findings (probable infection).2 Infants had blood collected routinely after birth for cultures and complete blood cell counts, and diagnoses of neonatal infection were made by an adjudication committee masked to randomization group and management site. Other outcomes were use of maternal and neonatal antibiotics and neonatal intensive care longer than 24 hours.
The associations between site of expectant management and clinical chorioamnionitis, maternal antibiotics, cesarean, postpartum fever, women’s evaluations of care, neonatal infection, neonatal antibiotics, and neonatal intensive care for longer than 24 hours, were determined initially using univariate analyses. Subsequently, multiple stepwise logistic regression analysis was used to determine whether site of management was associated with each outcome, controlling for country and baseline characteristics (maternal age at least 35 years, gestational age at least 41 weeks, unripe cervix, digital or speculum vaginal examination, nulliparity, maternal colonization with group B streptococcus, and duration of rupture of membranes 12–24 hours and at least 24 hours [versus under 12 hours]). Whether women had a previous cesarean was also considered as a variable for the outcome of cesarean. If maternal colonization with group B streptococcus, nulliparity, unripe cervix, or vaginal examination remained in the model, the interaction of management site and the baseline characteristic was considered. A statistical interaction would indicate that the effect of management site differed for women with different baseline characteristics. For example, an interaction between management site and parity for the response ‘would participate in the study again’ indicates that management site had a different effect on whether women would participate in the study again if they were nulliparas or multiparas. If unripe cervix remained in the model, the total number of women in the model was reduced from 1670 to 1163 because of missing data on cervical dilatation or effacement. Variables, including interactions, remained in the model if the corresponding level of significance was less than or equal to .1. For all other analyses P < .05 was used to indicate statistical significance.
|
Results |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
. Women managed at home were significantly more likely to be nulliparas, less likely to have had digital or speculum vaginal examinations at entry, and were more likely to have had longer duration of membrane rupture at entry compared with those managed in a hospital (Table 1).
|
The results of the univariate analyses are given in Table 2. Women managed at home were more likely to have clinical chorioamnionitis, receive antibiotics, and be delivered by cesarean than women managed in hospital, and their infants were more likely to receive antibiotics and be in the neonatal intensive care unit for longer than 24 hours. Women who were managed at home were more satisfied with their care than those managed in hospital (Table 2).
|
). Nulliparas were more likely to receive antibiotics if they were managed at home (odds ratio [OR] 1.52 95% confidence interval [CI], 1.04, 2.24, P = .03). There was a significant interaction between management site and maternal colonization with group B streptococcus, for cesarean (P = .09). Women who were not colonized were more likely to have cesareans if they were managed at home versus in hospital (OR 1.48 95% CI 1.03, 2.14, P = .04) (Table 3). There was significant interaction between management site and parity, for the response "would participate again in the study" (P = .08). Multiparas were more likely to say that they would participate in the study again if their management was at home rather than in hospital (OR 1.80 95% CI 1.27, 2.54, P < .001) (Table 3).
|
).
|
|
Discussion |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Our results suggest that it is generally safer for women with PROM at term to remain in hospital if they do not want labor induction. We found the likelihood of receiving antibiotics before or after delivery significantly higher for nulliparas if they were managed at home rather than in hospitals. We also found that infants were at a twofold higher risk of becoming infected if management was at home. There was an increased risk of cesarean for women not colonized with group B streptococcus (most women in the study), if they remained at home rather than in hospital. These findings do not differ from those of other reports. In a prospective study of 59 women with PROM for longer than 24 hours who were managed expectantly at home, Gilson et al9 found rates of chorioamnionitis, endometritis, and cesarean to be 8.5%, 6.1%, and 20.3%, respectively. There were no infants with culture-proved neonatal sepsis, although 5.1% had possible infections. The study was not controlled and the sample was small, so it is not certain whether there were any advantages or disadvantages to expectant management at home.9 A similar study was done by Hagskog et al10 in which outcomes were determined for 176 nulliparas with PROM at term who were managed expectantly at home. In that study, the rates of maternal infection, cesarean, and neonatal infection were 5%, 6%, and 2%, respectively. They compared those outcomes to a historical reference group and concluded that there was no advantage to expectant management at home for nulliparas with PROM at term.10 Carlan et al11 conducted a randomized controlled trial of home management compared with hospital management for women with preterm PROM and found no differences in adverse outcomes between groups. However, the sample was too small to rule out clinically important differences.
We found that multiparas were more likely to evaluate their care positively if expectant care had been partially or completely at home rather than in hospital. However, if women are informed that expectant management at home might increase the likelihood of their needing antibiotics if they are nulliparous, increase the risk of having a cesarean if they are not colonized with group B streptococcus, and increase the risk of their infants becoming infected, they might choose expectant management in a hospital or to proceed directly with labor induction. Other women, particularly multiparas, given the results of this study, might still prefer expectant management at home. Women’s choices, nevertheless, should be respected.8
In this study of 1670 women, the problem was not statistical power, but the study was limited because women were not randomly assigned to home versus hospital management. We attempted to control for differences in selection factors, present at baseline, to ensure that differences in outcomes were not caused by these factors but rather were a result of management at home rather than in a hospital. In addition, by controlling for country, we hoped to control for systematic differences in procedures (tests and treatments) that might have been associated with the outcomes of interest. Although we expect that higher-risk women might have been selected for care in a hospital rather than at home, thus any bias might have increased and not decreased the risk of adverse outcome in the hospital group, other factors for which we had no or only incomplete information might have been responsible for the effects we observed. Only a properly designed randomized controlled trial would be able to put that issue to rest.
According to the available evidence, women are most likely to choose to have labor induction with IV oxytocin if their membranes rupture before labor at term. Recruitment of women with PROM at term into a study of home versus hospital expectant management would therefore be difficult. We believe that the question of home rather than hospital management would be pursued best among women with preterm PROM because available information suggests that expectant management is best for that condition.
|
Footnotes |
|---|
Supported by grant MA-11392 from the Medical Research Council of Canada. Mary E. Hannah is a Medical Research Council of Canada Senior Scientist.
Received December 30, 1999. Received in revised form April 6, 2000. Accepted May 3, 2000.
|
References |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
2. Hannah ME, Ohlsson A, Farine D, Hewson SA, Hodnett ED, Myhr TL, et al. Induction of labor compared with expectant management for prelabor rupture of the membranes at term. N Engl J Med 1996;334:1005–10.
3. Mozurkewich EL, Wolf FM. Premature rupture of membranes at term: A meta-analysis of three management schemes. Obstet Gynecol 1997;89:1035–43.[Abstract]
4. Hannah M, Ohlsson A, Wang EEL, Matlow A, Foster G, Willan A, et al. Inducing labor with IV oxytocin may reduce the risk of neonatal infection in GBS positive women with PROM at term. Am J Obstet Gynecol 1997;177:780–5.[Medline]
5. Tan BP, Hannah ME. Oxytocin for prelabour rupture of the membranes at or near term. [Cochrane Review]. In: The Cochrane Library, Issue 2. Oxford: Update Software, 1999.
6. Tan BP, Hannah ME. Prostaglandins for prelabour rupture of the membranes at or near term. [Cochrane Review]. In: The Cochrane Library, Issue 2. Oxford: Update Software, 1999.
7. Gafni A, Goeree R, Myhr TL, Hannah ME, Blackhouse G, Willan AR, et al. Induction of labour versus expectant management for prelabour rupture of the membranes at term: An economic evaluation. Can Med Assoc J 1997;157:1519–25.[Abstract]
8. Hodnett ED, Hannah ME, Weston JA, Ohlsson A, Myhr TL, Wang EEL, et al. Women’s evaluations of induction of labor versus expectant management for prelabor rupture of the membranes at term. Birth 1997;24:214–20.[Medline]
9. Gilson GJ, O’Brien ME, Vera RW, Block A, Grubb PN. Expectant management of premature rupture of membranes at term in a birthing center setting. J Nurse-Midwifery 1988;33:134–9.[Medline]
10. Hagskog K, Nisell H, Sarman I, Westgren M. Conservative ambulatory management of prelabor rupture of the membranes at term in nulliparous women. Acta Obstet Gynecol Scand 1994;73:765–8.[Medline]
11. Carlan SJ, O’Brien WF, Parsons M, Lense JL. Preterm premature rupture of membranes: A randomized study of home versus hospital management. Obstet Gynecol 1993;81:61–4.
This article has been cited by other articles:
|
|
 |
|
  S. E. Lee, J. S. Park, E. R. Norwitz, K. W. Kim, H. S. Park, and J. K. Jun Measurement of Placental Alpha-Microglobulin-1 in Cervicovaginal Discharge to Diagnose Rupture of Membranes Obstet. Gynecol., March 1, 2007; 109(3): 634 - 640. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
