|
|
| |||||||||||||||||||||||||||||||||||||||||||||
Background Preterm birth is a major cause of perinatal morbidity and mortality. Whether the rate of preterm birth can be reduced by frequent contact between nurses and pregnant women or home monitoring of uterine activity is not known.
Methods We randomly assigned 2422 pregnant women with known risk factors for preterm labor (including 844 women who were pregnant with twins) to receive education and to have one of the following: weekly contact with a nurse, daily contact with a nurse, or daily contact with a nurse and home monitoring of uterine activity. The nurses elicited the women's own assessments of their symptoms and signs of preterm labor. The primary end point was the incidence of birth at less than 35 weeks' gestation. Secondary end points included cervical status at the time preterm labor was diagnosed and birth weight.
Results There were no significant differences among the groups in the incidence of birth at less than 35 weeks (14 percent in the weekly-contact group, 13 percent in the daily-contact group, and 14 percent in the home-monitoring group), in the mean amount of cervical dilatation at the time preterm labor was diagnosed (1.8 cm, 1.5 cm, and 1.4 cm, respectively), or in such neonatal outcomes as birth weights of less than 1500 g or less than 2500 g. However, daily contact with a nurse increased the mean number of unscheduled visits to obstetricians (1.2 in the weekly-contact group, 1.8 in the daily-contact group, and 2.3 in the home-monitoring group) and the proportion of women who received prophylactic tocolytic drugs (12 percent, 14 percent, and 19 percent, respectively).
Conclusions Women who have daily contact with a nurse, with or without home monitoring of uterine activity, have no better pregnancy outcomes than women who have weekly contact with a nurse.
Home monitoring of uterine activity may increase the early detection of preterm labor, but the results of randomized, controlled trials of its efficacy have been inconsistent. In some small studies, it facilitated earlier detection of preterm labor, resulting in lower rates of preterm birth,3,4,5,6 whereas in other studies there was no benefit.7,8 Recently, concern has arisen about the design of these studies, including the way withdrawals were handled, the restriction of analysis to subgroups, and the appropriateness of the end points.9,10,11
We undertook this study to determine whether adding home monitoring of uterine activity to daily contact with a nurse would improve clinical outcomes and whether daily contact with or without the use of a monitor would be more effective than weekly contact for pregnant women at increased risk for preterm labor. Our primary end point was the incidence of preterm birth at less than 35 weeks' gestation.
Methods
Study Subjects
All women receiving prenatal care at 30 Kaiser Permanente clinics in northern California were routinely screened for 14 risk factors for preterm labor, and those with at least 1 risk factor were eligible for the study. In addition, women were eligible if they had no preterm labor or premature rupture of membranes during the current pregnancy before enrollment, the gestational age was confirmed by ultrasonography by 24 weeks' gestation, they had access to a telephone, they were willing to comply with the study protocol, and they were at least 14 years old. The women with preexisting risk factors for preterm labor were enrolled between 24 and 30 weeks' gestation, and the women with risk factors that developed during the current pregnancy were enrolled before 33 weeks' gestation. The study protocol was approved by the Central Research Committee of the Kaiser Permanente Medical Care Program, Northern California, and by the Institutional Review Board of Kaiser Foundation Research Institute. Informed consent was obtained from all the participating women.
Randomization
After enrollment, the women were assigned to one of three treatment groups in a ratio of 1:1:1 with the use of a computer-generated randomization sequence. Randomization was stratified according to whether the women were pregnant with twins or at-risk singletons. The women who had preterm labor received care at eight tertiary centers. To control for possible differences in treatment philosophy, the randomization was also stratified according to treatment center.
Study Design
The treatment groups were those who had weekly contact with a nurse (the weekly-contact group), those who had daily contact with a nurse (the daily-contact group), and those who had daily contact plus home monitoring of uterine activity (the home-monitoring group). All the women received education about the symptoms and signs of preterm labor, such as changes in vaginal discharge, cramping, and backache; about the importance of early diagnosis and treatment; and about the way to palpate themselves for uterine contractions. An excessive number of contractions was defined for the women with singleton pregnancies as four or more contractions per hour and for the women with twin pregnancies as six or more contractions per hour. The women were asked to record their symptoms and signs as well as the number of contractions in a daily log. These 30-to-45-minute educational sessions followed a standard outline and used a teaching checklist.
The women in the weekly-contact group were asked to assess themselves for symptoms and signs of preterm labor and to palpate themselves for uterine contractions for one hour twice daily. The women who reached or exceeded their contraction thresholds or had symptoms or signs of preterm labor were asked to lie down, keep themselves hydrated, and palpate for another hour. If the symptoms or signs persisted or the contractions continued at or above the threshold, the women were told to call their obstetricians or labor-and-delivery units for evaluation. A nurse from a perinatal-service center called the women weekly to review their daily logs and to reinforce the importance of assessing their symptoms and palpating themselves for uterine contractions.
A nurse from a perinatal-service center called the women in the daily-contact group each day to review their symptoms and signs and the results of their twice-daily self-palpations for uterine contractions. These women were instructed to call the center if their symptoms or signs persisted or if their contractions continued to exceed their thresholds. The perinatal nurse then helped the women contact their obstetricians or labor-and-delivery units for evaluation.
The women in the home-monitoring group were each given a device (Corometrics 600, Corometrics Medical Systems, Wallingford, Conn.) that monitors uterine activity, stores the monitored information, and transmits it to a central receiver through telephone lines. The women were asked to use the device for one hour each morning and evening and to transmit the information after each session. The tracings were reviewed immediately, and any woman whose number of contractions per hour exceeded her threshold was called immediately. All the women were called daily to review their tracings and any symptoms or signs. The women were asked to call the center any time they had persistent symptoms or signs or wished to send emergency tracings. The perinatal nurse then helped the women contact their obstetricians or labor-and-delivery units for evaluation.
The women's obstetricians or nurse practitioners were aware of and consented to the women's participation in the study but were not informed of the group to which the women were assigned. The women and the perinatal-service nurses were instructed not to inform the obstetricians of the group assignments. The women and the perinatal-service nurses were also instructed not to divulge the method of detecting uterine activity (i.e., by palpation or monitoring device) when they reported increased uterine activity to the obstetricians.
Obstetrical Care
The women received care from their primary obstetricians or nurse practitioners, who were asked to follow broad care guidelines. These guidelines encouraged examinations of the cervix every 2 weeks, suggested specific criteria for the diagnosis of preterm labor, and recommended aggressive use of tocolytic drugs for preterm labor at less than 35 weeks' gestation, but they discouraged the prophylactic administration of tocolytic drugs (defined as the administration of tocolytic drugs to women with or without symptoms who did not meet the criteria for preterm labor). The criteria for preterm labor were regular contractions (>6 per hour) and either progressive effacement or dilatation of the cervix or cervical dilatation of >2 cm. The use of tocolytic drugs and glucocorticoids was left to the discretion of the obstetrician.
The women were followed according to their group assignments until either preterm delivery or 36 weeks' gestation. The guidelines of the initial group assignments were used to continue observing those who had preterm labor, except for 20 women whose obstetricians requested a specific change in services, such as providing home monitoring of uterine activity or daily contact that had not been initially assigned.
Statistical Analysis
Birth at less than 35 weeks' gestation was the primary study end point. Other end points were cervical dilatation at the time preterm labor was diagnosed, the change in the amount of cervical dilatation from the time of the previous examination to the time preterm labor was diagnosed, the number of unscheduled visits to an obstetrician, and surrogate markers for neonatal morbidity, such as birth weight, admission to an intensive care nursery, the number of days of oxygen or ventilator therapy, and the length of time the infant remained in the hospital.
For data analysis, preterm labor was diagnosed if regular contractions persisted with progressive effacement or dilatation of the cervix or if the cervix was dilated >2 cm. Women given a tocolytic drug for symptoms or signs that did not meet the criteria for preterm labor were considered to have received a prophylactic tocolytic drug after symptoms appeared. Unscheduled visits were defined as visits other than those scheduled for prenatal care.
Statistical analyses were performed on both an intention-to-treat and a completion-of-protocol basis. We present only the former because the results of the two analyses were similar. We also analyzed the data on at-risk singleton pregnancies and twin pregnancies separately. Additional analyses were then performed for the at-risk singleton pregnancies according to risk factor. All the data were analyzed by personnel of the Kaiser Permanente Division of Research using the chi-square test and Student's t-test. Wilcoxon's rank-sum test was used for the variables of cervical dilatation at the time preterm labor was diagnosed and the change in cervical dilatation. Because each treatment group was compared with two other groups, a P value of 0.025 was considered to indicate statistical significance.
Twin pregnancies were analyzed as if they had a single outcome. For continuous variables, such as birth weights, the mean of the two twin-pregnancy values represented the pregnancy outcome. For discrete variables, such as admissions to an intensive care nursery, if either infant had the outcome the pregnancy was considered to have had that outcome. To ensure the patients' safety, the results of the study were analyzed in a blinded fashion after each quartile of the required sample of 2154 women had given birth.12
Results
Between July 1992 and August 1996, we identified 3455 women as being eligible for the study. Their obstetricians recommended that 348 women (10 percent) not participate, and an additional 627 women (18 percent) declined to participate. Thus, 2480 women (72 percent) were enrolled in the study. After enrollment but before randomization, 58 women (2 percent) either gave birth or withdrew their consent. Thus, 2422 women were randomly assigned to one of the three treatment groups. All completed the study, but 93 women (4 percent of those randomized) did not receive the surveillance to which they had been assigned.
The characteristics of the 2422 women are shown in Table 1, and the distribution of risk factors for preterm labor among the three groups is shown in Table 2. There were no significant differences in demographic characteristics or risk factors among the three groups.
|
|
|
Preterm labor at less than 35 weeks' gestation was diagnosed slightly more often in the home-monitoring group (P = 0.06). Among the women who had preterm labor at less than 35 weeks' gestation, there were no intergroup differences in mean cervical dilatation or in the change in cervical dilatation from that at the previous examination (Table 4). There also were no differences in the percentages of women with cervical dilatation of <2 or <3 cm.
|
In the women with singleton pregnancies, there were no differences in the incidence of preterm birth at less than 35 weeks' gestation or in cervical status at the time preterm labor was diagnosed among the three groups. In addition, no differences in outcomes were evident in the largest subgroup of women with singleton pregnancies — i.e., those with previous preterm deliveries — despite their having made more unscheduled visits and used more tocolytic drugs prophylactically after symptoms appeared. This study had a power of more than 95 percent to detect a 1-cm difference between groups in cervical dilatation at the time of preterm-labor diagnosis for all study participants, for women with twins, and for women with at-risk singletons.
The women in the home-monitoring and daily-contact groups complied with the requirement of at least one daily session of monitoring uterine contractions 86 percent of the time, as compared with 79 percent in the weekly-contact group. No woman reported adverse effects of participating in this study; however, 24 (1 percent) had complications resulting from tocolytic-drug therapy, including pulmonary edema, chest pain, and hyperglycemia.
Discussion
Much of the controversy about the efficacy of home monitoring of uterine activity results from the inconsistent levels of care received by the control groups with which the home-monitoring groups in other studies have been compared. Most studies reporting a benefit of home monitoring4,5,6,13 have compared women receiving this intervention with women who attended one educational session on the symptoms and signs of preterm labor and received some instruction on self-palpation of the uterus. Although these studies have suggested that the monitoring device can facilitate early diagnosis of preterm labor, they have not addressed whether more frequent contact between nurses and pregnant women could accomplish the same goal.
We found that adding the monitoring device to daily contact with a nurse did not improve either the early detection of preterm labor or the outcome of pregnancy, a finding consistent with the results of all previous studies comparing home monitoring with frequent contact with a nurse (five to seven days per week).7,8,14,15 We therefore ask, how much education and contact with a provider is necessary to achieve results similar to those of home monitoring? In our study, we found no difference in cervical status at the time preterm labor was diagnosed or in pregnancy outcome between the women contacted weekly and those contacted daily.
We think the similarity in outcome between our weekly-contact group and the other two groups resulted from three factors. First, educational consistency was ensured by the use of a standard outline and a teaching checklist. Second, women were asked to palpate their uteruses for contractions routinely and keep a daily log of contractions and the symptoms and signs of preterm labor. Third, each woman was scheduled for weekly contact with a designated nurse who reviewed her daily logs and reminded her to seek help if she suspected preterm labor. Health-education studies suggest that frequent reinforcement of desired behavior alters patients' behavior more effectively than a single educational session.16,17
The incidence of preterm labor diagnosed at less than 35 weeks' gestation was slightly but not significantly higher in the home-monitoring group than in the weekly-contact group. However, the significant, progressive intergroup increase in the number of unscheduled visits to obstetricians suggests that more intensive surveillance promoted more unscheduled visits. Increased surveillance was also accompanied by greater prophylactic use of tocolytic drugs in women who had some symptoms of preterm labor and contractions but who did not meet the criteria for a diagnosis of preterm labor. The additional unscheduled visits and use of tocolytic drugs did not improve clinical outcomes.
This study had a power of more than 95 percent to detect a 1-cm difference in cervical dilatation at the time preterm labor was diagnosed in the three major analysis groups (all women, women carrying twins, and women carrying at-risk singletons). Without at least a 1-cm difference in mean cervical dilatation at the time of preterm-labor diagnosis, improvement in the perinatal outcome would be highly unlikely.
In conclusion, for both at-risk singleton and twin pregnancies, we found that neither daily contact with a nurse nor home monitoring of uterine activity provided additional benefit in terms of preventing preterm delivery, as compared with education, daily uterine self-palpation, and weekly contact with a nurse, among culturally diverse, well-educated, and predominantly middle-class women. In fact, adding daily contact or home monitoring increased the number of unscheduled visits and the prophylactic administration of tocolytic drugs without improving outcomes.
Supported in part by a grant (01 41 9032) from the Sidney Garfield Memorial Fund.
We are indebted to the obstetricians and nurse practitioners who referred the women to our study, to the perinatal-service-center nurses who maintained contact with the women, and to the Medical Editing Department of the Kaiser Foundation Research Institute.
Source Information
From the Divisions of Maternal–Fetal Medicine, Departments of Obstetrics and Gynecology, Kaiser Permanente Medical Care Program, Northern California. Presented at the 17th Annual Meeting of the Society of Perinatal Obstetricians, Anaheim, Calif., January 20–25, 1997, and published in abstract form (American Journal of Obstetrics and Gynecology 1997;176:S30).
Address reprint requests to Dr. Dyson at the Department of Obstetrics and Gynecology, Kaiser Permanente Medical Center, 900 Kiely Blvd., Bldg. J, Suite 2, Santa Clara, CA 95051-5386.
References
| |||||||||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
