Documente Academic
Documente Profesional
Documente Cultură
www.cochranelibrary.com
Contact address: Catalin M Stan, Private Practice, Avenue Vinet 16, Lausanne, 1004, Switzerland. catalin.stan@hin.ch.
Citation: Stan CM, Boulvain M, Pfister R, Hirsbrunner-Almagbaly P. Hydration for treatment of preterm labour. Cochrane Database
of Systematic Reviews 2013, Issue 11. Art. No.: CD003096. DOI: 10.1002/14651858.CD003096.pub2.
Copyright 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
ABSTRACT
Background
Hydration has been proposed as a treatment for women with preterm labour. Theoretically, hydration may reduce uterine contractility
by increasing uterine blood flow and by decreasing pituitary secretion of antidiuretic hormone and oxytocin.
Objectives
To evaluate the effectiveness of intravenous or oral hydration to avoid preterm birth and its consequences in women with preterm
labour.
Search methods
We searched the Cochrane Pregnancy and Childbirth Groups Trials Register (30 September 2013) and bibliographies of relevant
papers.
Selection criteria
Randomised controlled trials, including women with a viable pregnancy less than 37 completed weeks gestation and presenting with
preterm labour, comparing intravenous or oral hydration with no treatment. The intervention might or might not be associated with
bed rest. Studies comparing tocolytic drugs with intravenous fluids used in the control group as a placebo were not included in this
review.
Data collection and analysis
Two review authors independently assessed the reports, to determine if the study met the inclusion criteria and to evaluate the
methodological quality. Data were extracted independently by two of the review authors. The results were expressed as risk ratios (RR)
for dichotomous outcomes and mean difference (MD) for continuous outcomes.
Main results
Two studies, including a total of 228 women with preterm labour and intact membranes, compared intravenous hydration with bed
rest alone. Risk of preterm delivery, before 37 weeks (RR) 1.09; 95% confidence interval (CI) 0.71 to 1.68), before 34 weeks (RR
0.72; 95% CI 0.20 to 2.56) or before 32 weeks (RR 0.76; 95% CI 0.29 to 1.97), was similar between groups. Admission to neonatal
intensive care unit occurred with similar frequency in both groups (RR 0.99; 95% CI 0.46 to 2.16). Cost of treatment was slightly
Hydration for treatment of preterm labour (Review) 1
Copyright 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
higher (US$39) in the hydration group. This difference was not statistically significant and only includes hospital costs during a visit
of less than 24 hours. No studies evaluated oral hydration.
Authors conclusions
The data are too few to support the use of hydration as a specific treatment for women presenting with preterm labour. The two
small studies available do not show any advantage of hydration compared with bed rest alone. Intravenous hydration does not seem
to be beneficial, even during the period of evaluation soon after admission, in women with preterm labour. Women with evidence of
dehydration may, however, benefit from the intervention.
Unless they are dehydrated, there seems to be no benefit from additional intravenous fluids for women in preterm labour.
Preterm birth (before 37 weeks) can cause health problems and be life-threatening for babies. As women in preterm labour often
have lower amounts of fluid in their circulation, using an intravenous drip to increase the womans blood volume is sometimes tried
(hydration). It has been hoped that the extra fluid might somehow slow down contractions. However, from the limited information
available (two studies involving 228 women), the review found that there is no evidence of a benefit in the use of hydration to help
prevent preterm labour, although it may be helpful for women who are dehydrated.
BACKGROUND women are often evaluated for a period of time after admission,
to differentiate between true and false labour.
Preterm birth occurs in 9.6% to 12.2% of pregnancies (Beck 2010;
Martin 2011). Prematurity is the leading cause of infant deaths Neonatal mortality and morbidity decrease as gestational age
and is responsible for 35% of the perinatal mortality (Callaghan or birthweight, or both, increases (Bird 2010; Callaghan 2006;
2006). An evaluation of a cohort of neonates admitted to an in- Costeloe 2012; Moutquin 2003). Thus, prolongation of preg-
tensive care unit after delivery at 20 to 26 weeks showed an over- nancy would be expected to decrease neonatal mortality and im-
all survival of 57% (Costeloe 2012). Infants born preterm often prove subsequent child development by reducing the effects of pre-
require prolonged and expensive care in specialised units (Russel maturity. However, prolonging the pregnancy is not a guarantee
2007). Among infants with birthweight less than 750 g, 23% of that the outcome for the infant will be improved (Hackney 2013).
the survivors assessed at school age will suffer from neurodevelop- There are conditions, such as intra-amniotic infection, haemolytic
mental impairment and 5% from cerebral palsy (Wilson-Costello disease or impaired fetal growth, for which preterm delivery may
2007). Infants weighting between 1000 g and 1500 g at birth are be beneficial. In these cases, prolonging the pregnancy may have
also at risk of severe mental disability (8%) and of cerebral palsy adverse consequences for the health and the development of the
(6%) (Hack 1994; Oskoui 2013). infant.
Preterm birth usually follows either preterm labour, preterm Despite the development of various therapeutic interventions in
prelabour rupture of membranes or a medical decision to termi- recent decades, little progress has been made in reducing the inci-
nate the pregnancy because of maternal or fetal disease. About dence of preterm birth (Haas 2012; Martin 2011). Pharmacolog-
one-third to one-half of preterm births are the consequence of ical agents currently used to inhibit uterine contractions include
preterm labour without preterm prelabour rupture of the mem- betamimetics, magnesium sulphate, prostaglandin inhibitors, cal-
branes (Henderson 2012). Identification of women who will give cium channel blockers and oxytocin receptor antagonists (Haas
birth preterm among those presenting with symptoms of preterm 2012; Mackeen 2011). Some authors, based on uncontrolled ob-
labour is difficult. As yet, risk factors, biological markers and ul- servations, have suggested that intravenous hydration might de-
trasonography are limited predictors of preterm delivery (Chang crease contractions or delay the delivery in women presenting
2013; Faron 1998; Goldenberg 2008; Owen 2010). Therefore, with preterm contractions (Bieniarz 1971; Goodlin 1981). Stud-
Hydration for treatment of preterm labour (Review) 2
Copyright 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
ies on plasma volume reported that women with preterm labour women with a multiple pregnancy was planned, but insufficient
had lower plasma volumes than control women with a normal data were available.
pregnancy (Goodlin 1981). An uncontrolled case series showed
that 48% of women admitted with preterm labour did not de-
liver within 10 days, and one-third delivered after 37 weeks, when
Types of interventions
treated with bed rest and plasma volume expansion (Valenzuela
1983). The above studies suffer from the limitations of not pro- Intravenous or oral hydration compared with no treatment. The
viding a comparison with controls, and, as diagnosis of preterm intervention might or might not be associated with bed rest. We
labour that will end in preterm birth is difficult, these results may have not considered in this review studies comparing tocolytic
have been achieved without any intervention. drugs with intravenous fluids used in the control group as a
placebo.
How might plasma volume expansion reduce the risk of preterm
birth? A possible mechanism is that volume expansion inhibits
contractions by increasing uterine blood flow, thus stabilising de-
cidual lysosomes and decreasing prostaglandin production (Guinn Types of outcome measures
1997). Volume expansion, through left atrial distension, decreases
the secretion of antidiuretic hormone from the posterior pituitary
through the Henry-Gauer reflex (Bieniarz 1971; Guinn 1997).
The hypothesis is that oxytocin secretion will decrease simulta- Primary outcomes
neously (Bieniarz 1971); however, this mechanism has only been
described in animal models (Gauer 1976). Perinatal death or severe neonatal morbidity (defined as
respiratory distress syndrome, intracranial haemorrhage,
Preterm birth is related to significant neonatal mortality and infant necrotising enterocolitis, neonatal sepsis or seizures).
morbidity. Despite important research, prediction and treatment
of preterm labour has little impact in decreasing the incidence of
preterm births. Hydration of women in preterm labour may de-
crease uterine contractions by decreasing prostaglandin produc- Secondary outcomes
tion or oxytocin secretion. This procedure may have a place in the Prolongation of pregnancy more than 48 hours and seven
management of preterm labour. days.
Gestational age at delivery: more than 28 weeks, 32 weeks,
34 weeks and 37 weeks.
OBJECTIVES Perinatal outcomes: low birthweight (less than 2500 g),
To evaluate the effectiveness of intravenous or oral hydration to very low birthweight (less than 1500 g), Apgar score less than
avoid preterm birth and its consequences in women with preterm seven at five minutes, perinatal death, neonatal morbidity as
labour. separate outcomes (respiratory distress syndrome, intracranial
haemorrhage, necrotising enterocolitis, neonatal sepsis or
seizures, patent ductus arteriosus, hypoglycaemia), admission to
neonatal unit and need for mechanical ventilation.
METHODS
Long-term sequelae: neurologic impairment and chronic
lung disease.
Serious maternal outcomes: death, cardiac arrest,
Criteria for considering studies for this review
respiratory arrest, pulmonary oedema, cardiac arrhythmias.
Other maternal outcomes: hypotension, chest pain,
dyspnoea, nausea, vomiting, headaches, endometritis,
Types of studies
chorioamnionitis, hyperglycaemia, hypokalaemia, womens
Randomised controlled trials, with or without blinding, with loss
assessment of their treatment.
to follow-up of less than 20% of women originally randomised.
Percentage of caesarean section, instrumental delivery or other out-
Types of participants comes that the authors of the original trials have reported would
Women with a viable pregnancy less than 37 completed weeks have been included in the analysis, if available. Economic evalua-
gestation presenting with preterm labour, as defined by the au- tions of this therapy were included as an outcome. Mean interval
thors. A separate analysis was performed for women included be- between randomisation and delivery and the use of tocolytic drugs
fore 34 completed weeks gestation. A subgroup comparison of are reported as additional outcome measures (not prespecified).
Searching other resources (2) Allocation concealment (checking for possible selection
We examined bibliographies of identified studies for references to bias)
other trials. For each included study, we have described the method used to
We did not apply any language restrictions. conceal allocation to interventions prior to assignment, and assess
whether intervention allocation could have been foreseen in ad-
vance of, or during recruitment, or changed after assignment.
Data collection and analysis We assessed the methods as:
low risk of bias (e.g. telephone or central randomisation;
Methods used in previous versions of the review are set out in consecutively numbered sealed opaque envelopes);
Appendix 1. For this update, although no new trials were added, high risk of bias (open random allocation; unsealed or non-
we re-examined the trials already included in the review and carried opaque envelopes, alternation; date of birth);
out assessments of risk of bias and data analysis using up-to-date unclear risk of bias.
methods as set out in the Cochrane Handbook for Systematic Reviews
of Interventions (Higgins 2011).
(3.1) Blinding of participants and personnel (checking for
possible performance bias)
Selection of studies For each included study, we have described the methods used, if
Two review authors independently assessed eligibility for inclusion any, to blind study participants and staff from knowledge of which
of studies identified as a result of the search strategy. We resolved intervention a participant received. We considered studies to be at
any disagreement through discussion or, if required, we consulted low risk of bias if they were blinded, or if we judged that the lack
a third person. of blinding would be unlikely to affect results.
(5) Selective reporting (checking for reporting bias) Unit of analysis issues
For each included study, we have described how we investigated
the possibility of selective outcome reporting bias and what we
found. Cluster-randomised trials
We assessed the methods as: We did not identify any cluster-randomised trials in this version
low risk of bias (where it is clear that all of the studys pre- of the review. In future updates if such trials are identified we will
specified outcomes and all expected outcomes of interest to the include them in the analyses along with individually-randomised
review have been reported); trials. We will adjust their sample sizes using the methods described
high risk of bias (where not all the studys pre-specified in the Cochrane Handbook (Higgins 2011), using an estimate of
outcomes have been reported; one or more reported primary the intracluster correlation co-efficient (ICC) derived from the
outcomes were not pre-specified; outcomes of interest are trial (if possible), from a similar trial or from a study of a similar
reported incompletely and so cannot be used; or study failed to population. If we use ICCs from other sources, we will report this
include results of a key outcome that would have been expected and conduct sensitivity analyses to investigate the effect of varia-
to have been reported); tion in the ICC. If we identify both cluster-randomised trials and
unclear risk of bias. individually-randomised trials, we plan to synthesise the relevant
Data synthesis
Cross-over trials We carried out statistical analysis using Review Manager software
Cross-over studies were not included; this is not a suitable design (RevMan 2012). We used fixed-effect meta-analysis for combining
for this type of intervention. data where it was reasonable to assume that studies were estimating
the same underlying treatment effect: i.e. where trials examined
the same intervention, and the trials populations and methods
Other unit of analysis issues were judged to be sufficiently similar. If we had suspected clin-
ical heterogeneity sufficient to expect that the underlying treat-
If we had identified any trials with more than two arms (multi-
ment effects differed between trials, or if substantial statistical het-
arm trials), we planned to include all arms relevant to the scope erogeneity had been detected, we planned to use random-effects
of the review. If such trials are identified when we update the meta-analysis to produce an overall summary, provided an average
review, where appropriate, we will combine arms (using methods
treatment effect across trials was considered clinically meaningful.
described in the Cochrane Handbook (Higgins 2011)) to create a
In this version of the review, only two studies were included; we
single pair-wise comparison. If it is not appropriate to combine
carried out meta-analysis for very few outcomes and no important
them, we will present results separately for each arm, sharing results
statistical heterogeneity was identified.
for the control arm between each to avoid double counting (for
dichotomous outcomes we will divide the number of events and
total sample by two, for continuous outcomes we will assume the Subgroup analysis and investigation of heterogeneity
same mean and standard deviation but halve the control sample The following comparisons were performed: any type of hydration
size for each comparison). versus bed rest alone.
Subgroup analysis would have included:
oral versus intravenous hydration, and
Dealing with missing data prelabour preterm rupture of membranes versus intact
For included studies, we noted levels of attrition. We planned to membranes
explore the impact of including studies with high levels of missing
However, separate data were not available. We have set out data
data in the overall assessment of treatment effect by using sensi-
separately for a study that included women at less than 34 weeks
tivity analysis; in this version of the review only two trials were
of gestation but did not perform any formal subgroup analysis.
included so we did not carry out this planned additional analysis.
If data for subgroups become available for future updates, we will
For all outcomes, we carried out analyses, as far as possible, on
assess subgroup differences by interaction tests available within
an intention-to-treat basis, i.e. we attempted to include all partic-
RevMan (RevMan 2012). We will report the results of subgroup
ipants randomised to each group in the analyses, and all partici-
analyses quoting the Chi statistic and P value, and the interaction
pants are analysed in the group to which they were allocated, re-
test I value.
gardless of whether or not they received the allocated intervention.
The denominator for each outcome in each trial is the number
randomised minus any participants whose outcomes are known Sensitivity analysis
to be missing.
If we had included cluster-randomised trials in the review, we
planned to carry out sensitivity analysis. We also planned sensitiv-
ity analysis according to trial quality; temporarily excluding trials
Assessment of heterogeneity at high risk of bias due to inadequate allocation concealment to
We assessed statistical heterogeneity using the Tau, I and Chi explore whether this has any impact on the direction or size of the
statistics. We would have regarded heterogeneity as substantial if effect estimate. In this version of the review we did not carry out
an I was greater than 30% and either the Tau greater than zero, any sensitivity analysis as insufficient data were available. If more
or there was a low P value (less than 0.10) in the Chi test for data are included in updates, we will carry out planned sensitivity
heterogeneity. analysis using our primary outcomes only.
Secondary outcomes
Neither study reported on pregnancy prolongation of more than
Risk of bias in included studies
48 hours or 7 days.
The two studies included are of generally good methodological Preterm delivery before 37 weeks was similar (risk ratio (RR) 1.09;
quality. 95% confidence interval (CI) 0.71 to 1.68) in the hydration and
control groups (Analysis 1.5). Delivery before 34 weeks (RR 0.72;
95% CI 0.20 to 2.56) or before 32 weeks (RR 0.76; 95% CI 0.29
Allocation to 1.97) were also similar between groups (Analysis 1.6; Analysis
Both studies used computer-generated randomisation sequences 1.7). Admission to neonatal intensive care unit occurred with the
and allocations were concealed in opaque sealed envelopes opened same frequency in both groups (RR 0.99; 95% CI 0.46 to 2.16)
sequentially. (Analysis 1.8).
No other prespecified perinatal or maternal morbidity outcomes
were reported.
Blinding Cost of treatment was slightly higher (US$39) in the hydration
Blinding of women and clinical staff was not attempted in either group (Guinn 1997). This difference was not statistically signif-
study. The lack of blinding of staff may have affected clinical de- icant and only included hospital costs during a visit of less than
cisions that may have had an impact on outcomes introducing a 24 hours. The interval between randomisation and delivery (not
serious risk of bias. prespecified) was similar in the two groups. Slightly less use of
Blinding of outcome assessment was not mentioned, except that tocolytic drugs (not prespecified) was reported in the hydration
in the Helfgott 1994 trial monitor tracings of uterine contractions group, but this difference was not statistically significant (RR 0.83;
were reported to have been examined by blinded assessors. 95% CI 0.57 to 1.20) (Analysis 1.14).
Limited data are available for the subgroup of women included
before 34 weeks. Results are similar to those reported in all women
Incomplete outcome data (Comparison 2). No trials evaluated oral hydration therapy.
REFERENCES
References to studies included in this review Pircon 1989 {published data only}
Pircon RA, Strasser HT, Kirz DS, Towers CV. Controlled
trial of hydration and bedrest vs bedrest alone in the
Guinn 1997 {published data only}
Guinn DA, Goepfert AR, Owen J, Brumfield C, Hauth evaluation of preterm uterine contractions. Proceedings
of 9th Annual Meeting of the Society of Perinatal
JC. Management options in women with preterm uterine
contractions: a randomized clinical trial. American Journal Obstetricians; 1989 Feb 1-4; New Orleans, Louisiana, USA.
1989:396.
of Obstetrics and Gynecology 1997;177:8148.
Guinn DA, Goepfert AR, Owen J, Brumfield CG, Hauth Pircon RA, Strassner HT, Kirz DS, Towers CV. Controlled
trial of hydration and bed rest versus bed rest alone in
JC. Management options in women with preterm uterine
contractions: a randomized clinical trial. American Journal the evaluation of preterm uterine contractions. American
Journal of Obstetrics and Gynecology 1989;161:7759.
of Obstetrics and Gynecology 1997;176(1 Pt 2):S44.
Guinn 1997
Outcomes Interval from randomisation to delivery and to discharge, gestational age at delivery,
delivery before 34 and 37 weeks, costs
Notes A third group treated with terbutaline is not included in this review
Risk of bias
Allocation concealment (selection bias) Low risk Consenting women were assigned to one
of three treatment groups by opening the
next sealed, opaque envelope
Blinding of participants and personnel High risk The treatment regimens were different, and
(performance bias) clinical decisions (which may have had an
All outcomes impact on outcomes) were made by staff
who knew which treatment women were
receiving
Blinding of outcome assessment (detection High risk Blinding of outcome assessors was not
bias) mentioned and treatment was likely to
All outcomes have been recorded in patient records out-
come data were abstracted from the medi-
cal record and the computerized perinatal
database
Incomplete outcome data (attrition bias) Low risk All participants seem to be accounted for.
All outcomes
Selective reporting (reporting bias) Unclear risk Assessment from published study report.
Helfgott 1994
Outcomes Interval from randomisation to delivery, delivery before 32 and 37 weeks, need for
additional treatment
Notes Hydration only and hydration + morphine groups were combined for the purpose of
this review
Risk of bias
Random sequence generation (selection Low risk the assignments were computer gener-
bias) ated.
Allocation concealment (selection bias) Low risk Allocations were placed in sealed opaque
envelopes, and opened at the time of entry
Blinding of participants and personnel High risk Women and clinicians were not blinded,
(performance bias) which may have affected clinical decisions
All outcomes about their care
Blinding of outcome assessment (detection Unclear risk It was stated that some outcome assessment
bias) was blinded, but for other outcomes it was
All outcomes no clear that assessors were blind to treat-
ment group. The monitor tracings were
analysed by a blinded observer so as not to
bias the information regarding uterine con-
tractions upon entry and completion of the
study
Incomplete outcome data (attrition bias) Unclear risk 5 out of 124 women were enrolled into the
All outcomes study but then excluded. Another 3 sub-
sequently left the study. Delivery informa-
tion was available for 110 women, (11.3%
attrition)
Selective reporting (reporting bias) Unclear risk Assessment from published study report.
IM: intramuscular
IV: intravenous
Pircon 1989 Inadequate method of concealment of the allocation (alternate date). Women in the control group were treated with
hydration after 2 hours if contractions persisted. 3 women were readmitted after the initial treatment and included
again in the study, with a cross-over of the intervention. Report of the results is unclear
No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size
1 Perinatal death 0 0 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
2 Severe neonatal morbidity 0 0 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
3 Prolongation of pregnancy more 0 0 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
than 48 hours
4 Prolongation of pregnancy more 0 0 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
than seven days
5 Delivery before 37 weeks 2 228 Risk Ratio (M-H, Fixed, 95% CI) 1.09 [0.71, 1.68]
6 Delivery before 34 weeks 1 118 Risk Ratio (M-H, Fixed, 95% CI) 0.72 [0.20, 2.56]
7 Delivery before 32 weeks 1 110 Risk Ratio (M-H, Fixed, 95% CI) 0.76 [0.29, 1.97]
8 Admission to neonatal intensive 1 118 Risk Ratio (M-H, Fixed, 95% CI) 0.99 [0.46, 2.16]
care unit
9 Low birthweight (less than 2500 0 0 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
g)
10 Very low birthweight (less than 0 0 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
1500 g)
11 Maternal death 0 0 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
12 Womens assessment of their 0 0 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
treatment
13 Cost of treament (first 24 1 103 Mean Difference (IV, Fixed, 95% CI) 39.0 [-26.11, 104.
hours, in US$) 11]
14 Use of tocolytic drugs (not 2 234 Risk Ratio (M-H, Fixed, 95% CI) 0.83 [0.57, 1.20]
prespecified)
15 Time to delivery (days, not 2 228 Mean Difference (IV, Fixed, 95% CI) -0.99 [-7.85, 5.87]
prespecified)
Comparison 2. Hydration versus no treatment/bed rest alone (women included before 34 weeks)
No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size
1 Prolongation of pregnancy more 0 0 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
than 48 hours
2 Prolongation of pregnancy more 0 0 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
than seven days
3 Delivery before 32 weeks 0 0 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
4 Delivery before 34 weeks 1 118 Risk Ratio (M-H, Fixed, 95% CI) 0.72 [0.20, 2.56]
5 Delivery before 37 weeks 1 118 Risk Ratio (M-H, Fixed, 95% CI) 1.32 [0.72, 2.42]
6 Low birthweight (less than 2500 0 0 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
g)
Hydration for treatment of preterm labour (Review) 14
Copyright 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
7 Very low birthweight (less than 0 0 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
1500 g)
8 Perinatal death 0 0 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
9 Severe neonatal morbidity 0 0 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
10 Admission to neonatal intensive 1 118 Risk Ratio (M-H, Fixed, 95% CI) 0.99 [0.46, 2.16]
care unit
11 Maternal death 0 0 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
12 Womens assessment of their 0 0 Risk Ratio (M-H, Fixed, 95% CI) 0.0 [0.0, 0.0]
treatment
13 Cost of treament (first 24 1 103 Mean Difference (IV, Fixed, 95% CI) 39.0 [-26.11, 104.
hours, in US$) 11]
14 Use of tocolytic drugs (not 1 118 Risk Ratio (M-H, Fixed, 95% CI) 0.72 [0.31, 1.70]
prespecified)
15 Time to delivery (days, not 1 118 Mean Difference (IV, Fixed, 95% CI) -4.30 [-13.61, 5.01]
prespecified)
Analysis 1.5. Comparison 1 Hydration versus no treatment/bed rest alone (all women), Outcome 5
Delivery before 37 weeks.
Analysis 1.7. Comparison 1 Hydration versus no treatment/bed rest alone (all women), Outcome 7
Delivery before 32 weeks.
Analysis 1.13. Comparison 1 Hydration versus no treatment/bed rest alone (all women), Outcome 13 Cost
of treament (first 24 hours, in US$).
Mean Mean
Study or subgroup Treatment Control Difference Weight Difference
N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI
Guinn 1997 54 246 (172) 49 207 (165) 100.0 % 39.00 [ -26.11, 104.11 ]
Analysis 1.15. Comparison 1 Hydration versus no treatment/bed rest alone (all women), Outcome 15 Time
to delivery (days, not prespecified).
Mean Mean
Study or subgroup Treatment Control Difference Weight Difference
N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI
Helfgott 1994 73 49.91 (25.63) 37 46.96 (25.68) 45.7 % 2.95 [ -7.20, 13.10 ]
-10 -5 0 5 10
Favours control Favours treatment
Comparison: 2 Hydration versus no treatment/bed rest alone (women included before 34 weeks)
Comparison: 2 Hydration versus no treatment/bed rest alone (women included before 34 weeks)
Analysis 2.10. Comparison 2 Hydration versus no treatment/bed rest alone (women included before 34
weeks), Outcome 10 Admission to neonatal intensive care unit.
Comparison: 2 Hydration versus no treatment/bed rest alone (women included before 34 weeks)
Comparison: 2 Hydration versus no treatment/bed rest alone (women included before 34 weeks)
Mean Mean
Study or subgroup Treatment Control Difference Weight Difference
N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI
Guinn 1997 54 246 (172) 49 207 (165) 100.0 % 39.00 [ -26.11, 104.11 ]
Analysis 2.14. Comparison 2 Hydration versus no treatment/bed rest alone (women included before 34
weeks), Outcome 14 Use of tocolytic drugs (not prespecified).
Comparison: 2 Hydration versus no treatment/bed rest alone (women included before 34 weeks)
Comparison: 2 Hydration versus no treatment/bed rest alone (women included before 34 weeks)
Mean Mean
Study or subgroup Treatment Control Difference Weight Difference
N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI
-10 -5 0 5 10
Favours treatment Favours control
APPENDICES
17 December 2013 Amended The graph labels for outcome 1.15 (time to delivery in days) were the wrong way around. We have
therefore corrected them. There are no implications for the text of the review
HISTORY
Protocol first published: Issue 2, 2001
Review first published: Issue 2, 2002
3 October 2013 New citation required but conclusions have not Review updated.
changed
30 September 2013 New search has been performed Search updated. No new trials identified.
31 May 2009 New search has been performed Search updated. No new trials identified.
27 June 2007 New search has been performed Search updated. One addtional trial report added to
Guinn 1997 and one additional trial report added to
Pircon 1989.
CONTRIBUTIONS OF AUTHORS
The 2013 update was prepared by Catalin Stan with contributions from the other authors.
Catalin Stan and Michel Boulvain wrote the protocol, extracted the data and drafted the text of the first version of this review (Stan
2002). Pascale Hirsbrunner-Almagbaly and Riccardo Pfister contributed to the protocol and to the final text of the review. Pascale
Hirsbrunner-Almagbaly did the initial search of the studies.
SOURCES OF SUPPORT
Internal sources
University of Geneva, Switzerland.
External sources
UNDP-UNFPA-UNICEF-WHO-World Bank Special Programme of Research, Development and Research Training in Human
Reproduction (HRP), Department of Reproductive Health and Research (RHR), World Health Organization, Switzerland.
INDEX TERMS