Abstract
Background
Fetal hypoxaemia is often a feature of fetal growth impairment. It has been suggested that perinatal outcome after suspected impaired fetal growth might be improved by giving mothers continuous oxygen until delivery.
Objectives
The objective was to assess the effects of maternal oxygen therapy in suspected impaired fetal growth on fetal growth and perinatal outcome.
Search methods
We searched the Cochrane Pregnancy and Childbirth Group's Trials Register (June 2009).
Selection criteria
Acceptably controlled trials comparing maternal oxygen therapy with no oxygen therapy in suspected impaired fetal growth.
Data collection and analysis
Eligibility and trial quality was assessed.
Main results
Three studies involving 94 women were included. Oxygenation compared with no oxygenation was associated with a lower perinatal mortality rate (risk ratio 0.50, 95% confidence interval 0.32 to 0.81). However, higher gestational age in the oxygenation groups may have accounted for the difference in mortality rates.
Authors' conclusions
There is not enough evidence to evaluate the benefits and risks of maternal oxygen therapy for suspected impaired fetal growth. Further trials of maternal hyperoxygenation seem warranted.
Plain language summary
Maternal oxygen administration for suspected impaired fetal growth
Too little evidence to show whether continuous oxygen therapy for pregnant women benefits babies in the womb who are smaller than expected.
Babies who receive too little oxygen from their mother's blood can grow more slowly than expected before birth (impaired fetal growth). With extreme lack of oxygen, the baby can die in the womb. Sometimes, it may be suggested that the mother breathe extra oxygen through a face mask 24 hours daily (oxygen therapy) until the baby's birth. The review of trials found that there is too little evidence to show whether the baby's growth improves when women have continuous oxygen therapy from mid‐pregnancy until the baby's birth. There is some evidence that fewer babies may die, although further research is needed.
Background
Impaired fetal growth is the failure of a newborn to achieve its genetically determined growth potential, which may cause death as well as short or long‐term childhood morbidity. It has been reported that 3% to 10% of neonates are small for corresponding gestational age and an estimated 30% of this is due to impaired fetal growth. The remaining 70% is due to constitutional factors such as maternal ethnicity, parity, weight and height (Lin 1998). The condition occurs with limited flow of nutrients and/or oxygen from mother to fetus as a result of fetal causes (e.g. chromosomal abnormalities, congenital malformations), placental factors (e.g. small placenta), or maternal factors (e.g. malnutrition, vascular/renal disease, drugs or other metabolic conditions) (Resnik 2002).
Ultrasound evaluation of the fetus by measuring the abdominal circumference, head circumference, length of upper leg and interpreting these using standardised formulae allows the clinician to estimate the fetal weight, to relate this to the gestational age and to follow the growth progress. Ultrasound evaluation also allows to some extent to estimate the timing and the cause of the impairment. Symmetrical growth of the fetus is generally due to early problems such as chromosomal abnormalities, drugs, chemical agents or infection. Asymmetric growth usually results from inadequacy of substrates the fetus needs particularly later in pregnancy (Resnik 2002). In low‐income settings where early pregnancy ultrasound is not available, fetal growth can be monitored by serial symphysis fundus measurements. However, there is no proven effective treatment that can be applied once growth impairment is diagnosed. In general, when no apparent congenital abnormality exists, management is conservative, involving frequent growth measurements, recommending smoking cessation if the mother smokes, and early delivery when the fetus is thought be mature enough to survive outside the womb.
The outcomes of impaired growth are variable and usually related to the specific cause. For example, if the growth impairment is due to chromosomal anomalies or congenital abnormalities, the fetus is more at risk of a perinatal death. Other short‐term outcomes may be moderate to mild metabolic problems (hypoglycemia, polycythemia, meconium aspiration, etc.) due to the chronic oxygen and nutrient deprivation. Depending on the severity and the duration of the condition, long‐term outcomes may differ from normal to small decreases in IQ to an increased risk of cerebral palsy (Bernstein 2000).
Fetal growth impairment is characterised by hypoxaemia, hypoglycaemia and various other metabolic derangements. Findings from uncontrolled studies using cordocentesis suggested that fetal hypoxaemia could be corrected with maternal oxygen therapy and led to the hypothesis that perinatal outcome might also be improved (Nicolaides 1987). The suggested technique is to give the mother oxygen by mask, continuously (24 hours/day) until delivery. Besides being an uncomfortable treatment with potential psychosocial adverse effects for women, there is concern about possible toxic effects (pulmonary dysfunction) of prolonged maternal hyperoxygenation for the mother (Johanson 1995).
Objectives
To assess the effects on fetal growth and perinatal outcome of oxygen therapy for suspected impaired fetal growth.
Methods
Criteria for considering studies for this review
Types of studies
All acceptably controlled evaluations of the effects of oxygen therapy for suspected impairment of fetal growth on relevant clinical outcomes.
Types of participants
Women with suspected impaired fetal growth.
Types of interventions
Oxygen therapy by mask to the mother during pregnancy.
Types of outcome measures
Fetal growth, perinatal mortality, neonatal morbidity, adverse effects on the mother and neonate, women's experiences with the therapy.
Search methods for identification of studies
Electronic searches
We searched the Cochrane Pregnancy and Childbirth Group’s Trials Register by contacting the Trials Search Co‐ordinator (June 2009).
The Cochrane Pregnancy and Childbirth Group’s Trials Register is maintained by the Trials Search Co‐ordinator and contains trials identified from:
quarterly searches of the Cochrane Central Register of Controlled Trials (CENTRAL);
weekly searches of MEDLINE;
handsearches of 30 journals and the proceedings of major conferences;
weekly current awareness alerts for a further 44 journals plus monthly BioMed Central email alerts.
Details of the search strategies for CENTRAL and MEDLINE, the list of handsearched journals and conference proceedings, and the list of journals reviewed via the current awareness service can be found in the ‘Specialized Register’ section within the editorial information about the Cochrane Pregnancy and Childbirth Group.
Trials identified through the searching activities described above are each assigned to a review topic (or topics). The Trials Search Co‐ordinator searches the register for each review using the topic list rather than keywords.
We did not apply any language restrictions.
Data collection and analysis
Trials under consideration were evaluated for methodological quality and appropriateness for inclusion, without consideration of their results. The methodological quality was assessed taking into consideration the allocation concealment, blinding, exclusion after randomization and loss to follow up.
There were no language restrictions in the search and application of inclusion criteria.
Risk ratios with a fixed‐effect model were used to generate summary estimates. Heterogeneity between trials was checked visually and by the chi2 test given by MetaView.
Results
Description of studies
Three trials conducted in Italy (Battaglia 1992), South Africa (Johanson 1995) and UK and South Africa (Lindow 2002) have been included in this review. Ninety‐four women were included in these trials.
There are two reports by Battaglia and colleagues published in 1992 (Battaglia 1992) and 1994 (Battaglia 1994). It is unclear whether the two reports refer to two studies or separate publication of subgroups from one study. In neither report was any detailed description of random allocation and power calculation given. The first report (Battaglia 1992) includes 36 women with 'intrauterine growth retardation' and the second (Battaglia 1994) 38 women with 'mild intrauterine growth retardation'. Both studies have also been conducted at similar periods. Attempts have been made to get clarification from the first author but no contact could be made. The second trial is therefore excluded until further information becomes available (Battaglia 1994).
In Battaglia et al (Battaglia 1992) the control group received bedrest and antihypertensive therapy if indicated. Bedrest was the standard therapy in the control group in Johanson 1995. Lindow 2002 was double‐blind and control group received humidified air under the same conditions as the treatment group (8 litres per minute giving a 40% concentration via a face mask 24 hours per day). Oxygen groups received 55% (Battaglia 1992) or 40% (Johanson 1995; Lindow 2002) humidified oxygen at 8 litres per minute by face mask, 24 hours per day.
Risk of bias in included studies
Battaglia et al (Battaglia 1992) do not specify the method of 'randomly assigning' women with impaired fetal growth to an oxygenation and to a control group. They do not state whether outcomes were assessed blind to group allocation. The rates of perinatal mortality may have been influenced by the fact that four of the control group fetuses who died were 25 to 26 weeks at enrolment, whereas the most immature fetus in the oxygenation group was 27.5 weeks.
Johanson 1995 used sealed opaque envelopes to conceal allocation but outcome assessments were not blinded. Results were not analysed on an intention‐to‐treat basis. One baby was dead in the bedrest group after birth. This case was excluded from the analysis.
Lindow 2002 is reported as a double‐blind trial. Randomization was made using opaque sealed envelopes and stratified into two groups according to gestational weeks. The two groups received the interventions from uniform coloured gas cylinders marked as either 'a' or 'b' although it is not clear whether they were identical. In order not to reveal the allocated group, arterial blood gas tensions were not measured.
Effects of interventions
Perinatal mortality was statistically significantly lower in the oxygen group (n = 46) (risk ratio 0.50, 95% confidence interval 0.32 to 0.81) compared to control group (n = 48) which was consistent across all three trials.
The time from diagnosis to delivery was (mean ± SE) 10.1 ± 3.5 versus 9 ± 3.1 days (Battaglia 1992), 17 versus 22 days (SE of the means are not available) (Johanson 1995) and 12.8 ± 2.2 versus 10.4 ± 2.0 days (Lindow 2002) in the oxygen and control groups respectively. In all studies birthweights were higher in the oxygen group.
No significant side‐effects or adverse outcomes have been reported.
Discussion
Maternal hyperoxygenation was suggested initially as a treatment for impaired fetal growth on the basis of uncontrolled trials. The three randomized trials reviewed here are consistent with respect to reduction in perinatal mortality which is a substantive outcome. However, both the Johanson (Johanson 1995) and Battaglia (Battaglia 1992; Battaglia 1994) studies have methodological problems. Not using a placebo may influence the timing of interventions and introduce bias. Both studies had greater gestational ages of fetuses in the oxygen groups which can explain the differences in perinatal mortality. The reason for this may be selection bias (unlikely in the Johanson 1995 study) or an alpha error due to small sample size. The Lindow 2002 study is the only double‐blind study, but has limited power to detect a significant benefit due to the small sample size.
Bearing in mind the methodological problems, the small number of women included, and the potential for harm against the possibility of reduction in perinatal mortality indicated in all three trials, there is good justification for investigating oxygen therapy further in the context of well‐designed controlled trials.
Authors' conclusions
Implications for practice.
In view of the methodological limitations and the small sample sizes of the trials identified so far, the use of maternal oxygenation should await the results of further trials, particularly in view of the suggestion from some studies that oxygen therapy may reduce uterine blood flow (Tervila 1973a; Tervila 1973b).
Implications for research.
Further trials on maternal oxygen therapy seem to be justified. These will need to have adequate sample sizes with realistic estimates of effects on substantive outcomes and therefore, will most probably require multicentre collaboration.
What's new
| Date | Event | Description |
|---|---|---|
| 24 June 2009 | New search has been performed | Search updated. No new trials identified. |
History
Protocol first published: Issue 1, 1995 Review first published: Issue 1, 1995
| Date | Event | Description |
|---|---|---|
| 18 September 2008 | Amended | Converted to new review format. |
| 24 November 2006 | New search has been performed | Search updated but no new trials identified. |
| 30 June 2004 | New search has been performed | Search updated but no new trials identified. |
| 11 November 2002 | New search has been performed | Search updated and one new trial included. |
Acknowledgements
None.
Data and analyses
Comparison 1. Maternal oxygen therapy versus no treatment.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 1 Perinatal mortality | 3 | 94 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.50 [0.32, 0.81] |
| 2 Birthweight | 2 | 57 | Mean Difference (IV, Fixed, 95% CI) | 85.11 [‐53.75, 223.97] |
1.1. Analysis.
Comparison 1 Maternal oxygen therapy versus no treatment, Outcome 1 Perinatal mortality.
1.2. Analysis.
Comparison 1 Maternal oxygen therapy versus no treatment, Outcome 2 Birthweight.
Characteristics of studies
Characteristics of included studies [ordered by study ID]
Battaglia 1992.
| Methods | Women 'randomly assigned' to 2 groups. | |
| Participants | 36 women at 26‐34 weeks' gestation, known dates, and intrauterine growth retardation defined by decreased amniotic fluid, abnormal doppler flow velocity waveform and decreased abdominal circumference participated. | |
| Interventions | Traditional management (bed rest, antihypertensives where necessary) versus traditional management + 55% humidified oxygen via a face mask 24 hours a day). | |
| Outcomes | Doppler velocimetry. Blood gases. Perinatal mortality and morbidity. | |
| Notes | Italy. | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Allocation concealment? | Unclear risk | B ‐ Unclear |
Johanson 1995.
| Methods | Randomized, allocation by sealed opaque envelopes. No blinding used. Analysis reported as intention‐to‐treat. | |
| Participants | 26 women with singleton pregnancy between 24‐30 weeks' gestation and with absent end‐diastolic flow in the umbilical artery. Labour, imminent eclampsia, pre‐eclampsia with organ failure and possible chromosomal abnormality were reasons for exclusion. 1 is excluded from the analysis after the birth of a baby with Down syndrome. | |
| Interventions | Oxygen group: bed rest and continuous 40% oxygen by a face mask until delivery or fetal demise. Control group: bed rest only which was the standard management. | |
| Outcomes | Perinatal mortality. | |
| Notes | South Africa. All women received low‐dose aspirin. | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Allocation concealment? | Low risk | A ‐ Adequate |
Lindow 2002.
| Methods | Randomized, allocation was made by sealed opaque envelopes. Double‐blind, using uniform gas cylinders marked either 'a' or 'b'. Stratified into 2 groups according to the gestational weeks. | |
| Participants | 32 women with impaired fetal growth diagnosed by doppler studies between 24‐30 weeks of gestation in the United Kingdom and South Africa. | |
| Interventions | Oxygen group: continuous 40% humidified oxygen by a face mask at 8 L/min until delivery. Control group: continuous humidified air by a face mask at 8 L/min giving a 40% of gas. |
|
| Outcomes | Doppler velocimetry, perinatal mortality. | |
| Notes | South Africa and the UK. All women received 81 mg aspirin per day and 12 mg betamethasone, daily for 2 days repeated weekly. | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Allocation concealment? | Low risk | A ‐ Adequate |
min: minutes
Characteristics of excluded studies [ordered by study ID]
| Study | Reason for exclusion |
|---|---|
| Battaglia 1994 | It is not clear whether this study is different from Battaglia 1992 which is included in the review. Attempts to get clarification have failed. |
| Nicolaides 1988 | This study was planned and funded but no participants were recruited and subsequently abandoned. |
| Rasanen 1998 | This is a randomized controlled trial of maternal hyperoxygenation in women with healthy pregnancies and healthy fetuses. The objective was to assess the reactivity of the fetal pulmonary circulation to 5 minutes of 60% oxygen at 20 to 26 weeks and 31 to 36 weeks of pregnancy. 20 women in each period were randomly allocated to oxygen or room air. Doppler measurements were taken at baseline and after oxygenation (or room air administration) in the 2 groups. The authors found that the fetal pulmonary circulation is more responsive to oxygenation later in pregnancy. All changes returned to baseline when oxygenation was discontinued. |
| Tervila 1973a | No clinically useful outcomes given. Oxygen therapy with or without vasodilator therapy in healthy women at term increased fetal oxygenation (pO2) and there was a trend towards decreased uterine perfusion (cervical blood flow measured by thermistor probe). |
| Tervila 1973b | No clinically useful outcomes given. Oxygen therapy resulted in decreased pH both in uterine and fetal blood. There was also decreased uterine perfusion with oxygen therapy in hypertensive women (cervical blood flow measured by thermistor probe). |
| Young 1980 | No clinically useful endpoints were reported. Oxygen administration prior to elective caesarean section in healthy women was associated with higher cord blood pO2 levels, but no other blood gas differences. |
| Zhang 1997 | This study compares a group of randomly selected women receiving oxygen inhalation 3 times a day, an hour each time together with vitamins B and E, 10% glucose and 7% amino acid infusions daily to another group of randomly selected group receiving 500 ml of 10% intralipid emulsion (containing linoleic acid, linolenic acid and oleic acid) alternating with the amino acid infusion. The study was evaluated for the Cochrane Review 'Maternal nutrient supplementation for suspected impaired fetal growth' and excluded from the review. The design does not appear to be a randomized controlled design. |
Contributions of authors
Justus Hofmeyr wrote the original review for the Cochrane Pregnancy and Childbirth Database. Metin Gülmezoglu updated the review, and has been responsible for maintaining the review since 1995. Both review authors did the data extraction and contributed to the text of the review. Lale Say contributed to the recent update by checking the data entries and revising the text of the review.
Sources of support
Internal sources
Effective Care Research Unit, University of the Witwatersrand, East London/Johannesburg, South Africa.
UK Cochrane Centre, NHS R&D Programme, Oxford, UK.
HRP ‐ UNDP/UNFPA/WHO/World Bank Special Programme in Human Reproduction, Geneva, Switzerland.
External sources
No sources of support supplied
Declarations of interest
None known.
New search for studies and content updated (no change to conclusions)
References
References to studies included in this review
Battaglia 1992 {published data only}
- Battaglia C, Artini PG, D'Ambrogio G, Galli PA, Segre A, Genazzani AR. Maternal hyperoxygenation in the treatment of intrauterine growth retardation. American Journal of Obstetrics and Gynecology 1992;167:430‐5. [DOI] [PubMed] [Google Scholar]
Johanson 1995 {published data only}
- Johanson R, Lindow S, Elst C, Westhuizen S, Tucker A, Jaquire Z. RCT of maternal oxygen therapy for in‐utero asphyxia. Proceedings of 2nd International Scientific Meeting of the Royal College of Obstetricians and Gynaecologists; 1993 Sept 7‐10; Hong Kong. 1993:168.
- Johanson R, Lindow SW, Elst C, Jaquire Z, Westhuizen S, Tucker A. A prospective randomised comparison of the effect of continuous O2 therapy and bedrest on fetuses with absent end‐diastolic flow on umbilical artery Doppler waveform analysis. British Journal of Obstetrics and Gynaecology 1995;102:662‐5. [DOI] [PubMed] [Google Scholar]
Lindow 2002 {published data only}
- Lindow S, Mantel G, Anthony J, Coetzee E, Pharoah P. A double blind randomised controlled trial of continuous oxygen therapy for compromised fetuses. Journal of Perinatal Medicine 2001;29 Suppl 1:123. [DOI] [PubMed] [Google Scholar]
- Lindow S, Mantell G, Anthony J, Coetzee E. A double‐blind randomised controlled trial of continuous oxygen therapy for compromised fetuses [abstract]. Journal of Obstetrics and Gynaecology 2002;22(2 Suppl):S35. [DOI] [PubMed] [Google Scholar]
- Lindow SW, Mantel GD, Anthony J, Coetzee EJ. A double‐blind randomised controlled trial of continuous oxygen therapy for compromised fetuses. BJOG: an international journal of obstetrics and gynaecology 2002;109:509‐13. [DOI] [PubMed] [Google Scholar]
References to studies excluded from this review
Battaglia 1994 {published data only}
- Battaglia C, Artini PG, D'Ambrogio G, Bencini S, Galli PA, Genazzani AR. Maternal hyperoxygenation in the treatment of mild intrauterine growth retardation: a pilot study. Ultrasound in Obstetrics & Gynecology 1994;4:472‐5. [DOI] [PubMed] [Google Scholar]
Nicolaides 1988 {published data only}
- Nicolaides KH. Randomized prospective study on maternal hyperoxygenation vs bed rest in the treatment of second trimester hypoxic fetal growth retardation. Personal communication 1988.
Rasanen 1998 {published data only}
- Rasanen J, Wood DC, Debbs RH, Cohen J, Weiner S, Huhta JC. Reactivity of the human fetal pulmonary circulation to maternal hyperoxygenation increases during the second half of the pregnancy. Circulation 1998;97:257‐62. [DOI] [PubMed] [Google Scholar]
Tervila 1973a {published data only}
- Tervila L, Vartiainen E, Kivalo I, Finnila M‐J, Vayrynen M. The effect of oxygen ventilation and a vasodilator on uterine perfusion, foetal oxygen and acid‐base balance. I. A study in healthy gravidae. Acta Obstetricia et Gynecologica Scandinavica 1973;52:177‐81. [DOI] [PubMed] [Google Scholar]
Tervila 1973b {published data only}
- Tervila L, Vartiainen E, Kivalo I, Finnila M‐J, Vayrynen M. The effect of oxygen ventilation and a vasodilator on uterine perfusion, fetal oxygen and acid‐base balance. II. A study in abnormal gravidae. Acta Obstetricia et Gynecologica Scandinavica 1973;52:309‐15. [DOI] [PubMed] [Google Scholar]
Young 1980 {published data only}
- Young DC, Popat R, Luther ER, Scott KE, Writer WDR. Influence of maternal oxygen administration on the term fetus before labor. American Journal of Obstetrics and Gynecology 1980;136:321‐4. [DOI] [PubMed] [Google Scholar]
Zhang 1997 {published data only}
- Zhang L. The effects of essential fatty acids preparation in the treatment of intrauterine growth retardation. American Journal of Perinatology 1997;14(9):535‐7. [DOI] [PubMed] [Google Scholar]
Additional references
Bernstein 2000
- Bernstein IM, Horbar JD, Badger GJ, Ohlsson A, Golan A. Morbidity and mortality among very low birth weight infants with intrauterine growth restriction. The Vermont Oxford Network. American Journal of Obstetrics and Gynecology 2000;182:198. [DOI] [PubMed] [Google Scholar]
Lin 1998
- Lin C, Santolaya‐Forgas J. Current concepts of fetal growth restriction: Part 1. Causes, classification, and pathophysiology. Obstetrics & Gynecology 1998;92(6):1044‐55. [DOI] [PubMed] [Google Scholar]
Nicolaides 1987
- Nicolaides KH, Campbell S, Bradley RJ, Bilardo CM, Soothill PW, Gibb D. Maternal oxygen therapy for intrauterine growth retardation. Lancet 1987;1:942‐5. [DOI] [PubMed] [Google Scholar]
Resnik 2002
- Resnik R. Intrauterine growth restriction. Obstetrics & Gynecology 2002;99(3):490‐60. [DOI] [PubMed] [Google Scholar]
References to other published versions of this review
Gülmezoglu 2002
- Gülmezoglu AM, Hofmeyr GJ. Maternal oxygen administration for suspected impaired fetal growth. The Cochrane Library 2002, Issue 2. [DOI] [PubMed] [Google Scholar]