Date of most recent substantive amendment: 14 May, 1997.
Objectives: To evaluate the effect of hospitalization for bed rest in women with a multiple pregnancy on the risk of preterm birth, fetal and neonatal mortality, neonatal morbidity, and women's satisfaction with their care.
Search strategy: This review has drawn on the search strategy developed for the Pregnancy and Childbirth Group as a whole. Relevant trials were identified in the Group's Specialized Register of Controlled Trials. See Review Group's details for more information.
Selection criteria: All published, unpublished, and ongoing randomized trials with reported data that compare outcomes in women with a multiple pregnancy and their babies who were offered hospitalization for bed rest during the antenatal period compared with women who did not receive routine hospitalization, but selective admission if complications occurred.
The primary outcomes are preterm birth, perinatal mortality, and fetal growth. Secondary outcomes include other neonatal morbidity, long-term neurodevelopment, maternal morbidity, and women's assessment of their care.
Data collection and analysis: Suitability for inclusion and methodological quality of the trials were assessed by the reviewer. Data were extracted by the reviewer and double entered. All eligible trials were included in the initial analysis. Prespecified sensitivity analyses have been carried out to evaluate the effect of trial quality, the effects of hospitalization for bed rest in women with an uncomplicated twin pregnancy, in women with a triplet pregnancy, and in women with a twin pregnancy complicated by cervical effacement and dilatation prior to labor.
1. Analyses of all trials. A policy of routine hospitalization for bed rest in multiple pregnancy did not reduce the risk of preterm birth, or perinatal mortality. There was a trend to a decreased number of low birthweight infants born to women in the routinely hospitalized group, which reached the conventional levels of statistical significance when the trial given a D rating for concealment of allocation was excluded (OR = 0.79; 95% CI 0.63–0.99). No differences were seen in the number of very low birthweight infants. No support for the policy was found in other neonatal outcomes. No information is available on developmental outcomes for infants in any of the trials. Women's views about the care they received were reported rarely.
2. Analyses of hospitalization for bed rest in women with an uncomplicated twin pregnancy. The risk of preterm birth was not reduced. Indeed significantly more women delivered very preterm (< 34 weeks gestation) (OR = 1.84; 95% CI 1.01–3.34). No differences were seen in perinatal mortality, or in other neonatal outcomes.
Women receiving hospitalization for bed rest had a decreased risk of developing hypertension (OR = 0.55; 95% CI 0.32–0.97), although this effect was no longer apparent when the trial given a D rating for concealment of allocation was excluded. This post hoc observation was open to observer bias and needs to be interpreted cautiously.
3. Analyses of hospitalization for bed rest in women with a triplet pregnancy. Most of the comparisons made between the hospitalized and control groups suggest beneficial treatment effects from routine hospitalization for bed rest. However all the differences observed between the experimental and control groups were compatible with chance variation.
4. Analyses of hospitalization for bed rest in women with a twin pregnancy complicated by cervical effacement and dilatation prior to labor. No differences were seen in the risk of preterm birth, perinatal mortality, fetal growth, or in other neonatal outcomes.
Conclusions: There is currently no sound evidence to support a policy of routine hospitalization for bed rest in multiple pregnancy. No reduction in the risk of preterm birth or perinatal death is evident, although there is a suggestion that fetal growth is improved. For women with an uncomplicated twin pregnancy the results of this review suggest that it may be harmful in that the risk of very preterm birth is increased. Until further evidence is available to the contrary, the policy cannot be recommended for clinical practice.
In women with triplets, although mainly beneficial effects of the policy are observed, it must be reiterated that these could all be ascribable to chance variation, and do not provide a basis for widespread adoption of the policy into clinical practice.
For women with a twin pregnancy at high risk of preterm birth because of signs of cervical effacement and dilatation prior to labor, there is so far no basis for the adoption of the policy into clinical practice.
Hospitalization for bed rest in multiple pregnancy has only been subjected to limited, well-controlled evaluation, and to clarify further the beneficial or adverse effects, additional, controlled evaluation is necessary, especially in triplet and higher order multiple pregnancies. Further assessment of the favorable effects on fetal growth observed is warranted. Any future trials should provide long-term developmental outcomes for the infants, assess women's views of the care received, and test the hypothesis that women receiving hospitalization for bed rest have a decreased risk of developing hypertension.
Citation: Crowther CA. Bed rest in hospital for multiple pregnancy (Cochrane Review). In: The Cochrane Library, Issue 1, 1999. Oxford: Update Software.
The preceding report is an abstract of a regularly updated, systematic review prepared and maintained by the Cochrane Collaboration. The full text of the review is available in The Cochrane Library (ISSN 1464–780X).
Update Software Ltd., Summertown Pavilion, Middle Way, Oxford OX2 7LG, United Kingdom (Tel.: +44 1865 513902; Fax: +44 1865 516918).