Corticosteroids for treating hypotension in preterm infants
Subhedar NV, Duffy K, Ibrahim H
Dates
Date edited: 15/11/2006
Date of last substantive update: 31/08/2006
Date of last minor update: / /
Date next stage expected 30/11/2008
Protocol first published: Issue 2, 2005
Review first published: Issue 1, 2007
Contact reviewer
Nimish V Subhedar
Consultant Neonatal Paediatrician
Neonatal Intensive Care Unit
Liverpool Women's Hospital
Crown Street
Liverpool
UK
L 8 7SS
Telephone 1: +44 151 702 4272
Facsimile: +44 151 702 4272
E-mail: nvsubhedar_lwh@yahoo.com
Contribution of reviewers
NVS was responsible for initiating the review, developing the protocol, writing the review and overseeing the project.KD was responsible for devising and conducting the literature search, identifying the studies eligible for inclusion and extracting data from the included studies.
HI was jointly responsible for identifying eligible studies and extracting data from included studies.
Internal sources of support
Liverpool Women's Hospital NHS Trust, UK
External sources of support
NoneWhat's new
Dates
Date review re-formatted: / /
Date new studies sought but none found: / /
Date new studies found but not yet included/excluded: / /
Date new studies found and included/excluded: / /
Date reviewers' conclusions section amended: / /
Date comment/criticism added: / /
Date response to comment/criticisms added: / /
Text of review
Synopsis
It is unclear whether giving steroids to premature newborn babies who have hypotension (low blood pressure) is safe and effective. More studies are needed.
Low blood pressure is a relatively common problem in premature newborn babies and has been linked with serious short and long term problems such as death and disability. Various treatments are used to support the circulation and boost blood pressure. One such treatment is the use of steroid drugs. This review found only two small studies that evaluate the effect of steroids on low blood pressure in premature infants. At present, there is insufficient information on which to base recommendations about the value of giving steroids to babies born before term who have low blood pressure.
Abstract
Background
Systemic hypotension is a relatively common complication of preterm birth and is associated with periventricular haemmorhage, periventricular white matter injury and adverse neurodevelopmental outcome. Corticosteroid treatment has been used as an alternative, or an adjunct, to conventional treatment with volume expansion and vasopressor/inotropic therapy.
Objectives
To determine the effectiveness and safety of corticosteroids used either as primary treatment of hypotension or for the treatment of refractory hypotension in preterm infants.
Search strategy
Randomized or quasi-randomized controlled trials were identified by searching the Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library, Issue 1, 2005), MEDLINE (1996 - June 2005), EMBASE (1974 - June 2005), reference lists of published papers and abstracts from the Pediatric Academic Societies and the European Society for Pediatric Research meetings published in Pediatric Research (1995 - 2004).
Selection criteria
We included all randomised or quasi-randomised controlled trials investigating the effect of corticosteroid therapy in the treatment of hypotension in preterm infants (< 37 weeks gestation) less than 28 days old. Studies using corticosteroids as primary treatment were included as well as studies using corticosteroids in babies with hypotension resistant to inotropes/pressors and volume therapy. We included studies comparing oral/intravenous corticosteroids with placebo, other drugs used for providing cardiovascular support or no therapy in this review.
Data collection & analysis
Methodological quality of eligible studies was assessed according to the methods used for minimising selection bias, performance bias, attrition bias and detection bias. Studies that evaluated corticosteroids (1) as primary treatment for hypotension or (2) for refractory hypotension unresponsive to prior use of inotropes/pressors and volume therapy, were analysed using separate comparisons. Data were analysed using the standard methods of the Neonatal Review Group using Rev Man 4.2.7. Treatment effect was analysed using relative risk, risk reduction, number needed to treat for categorical outcomes and weighted mean difference for outcomes measured on a continuous scale, with 95% confidence intervals.
Main results
Two studies were included in this review enrolling a total of 57 babies. In the first study, persistent hypotension was more common in hydrocortisone treated infants as compared to those who received dopamine as primary treatment for hypotension (RR 8.2, 95% CI 0.47 to 142.6; RD 0.19, 95% CI 0.01 to 0.37). In the second study, persistent hypotension (defined as a continuing need for epinephrine infusion) was less common in dexamethasone treated infants as compared to controls who received placebo for refractory hypotension (RR 0.42 , 95% CI 0.17 to 1.06; RD -0.51, 95% CI - 0.91 to - 0.12). There were no statistically significant effects on any other short or long-term outcome. It was not considered appropriate to perform a meta-analysis. A further two studies that have only been published in abstract form to date, may be eligible for inclusion in a future update of this review.
Reviewers' conclusions
There is insufficient evidence to support the routine use of steroids in the treatment of primary or refractory neonatal hypotension.
Hydrocortisone may be as effective as dopamine in treating primary hypotension, but there are no data regarding the long-term safety of steroids used for this indication.
A single dose of dexamethasone may be effective in treating preterm infants with refractory hypotension receiving epinephrine. However, given the lack of data on long-term safety dexamethasone cannot be recommended for routine use in preterm hypotension.
Background
Systemic hypotension is a relatively common complication of prematurity, affecting approximately one-third of very low birth weight infants. It is associated with an increased risk of intraventricular haemorrhage, periventricular white matter injury and adverse long-term neurodevelopmental outcome (Cunningham 1999; Miall-Allen 1987; Goldstein 1995). Current treatment of hypotension in the premature infant includes the use of volume expansion, inotropes and vasopressor agents. Corticosteroids are generally reserved for infants with refractory hypotension.
There is no widely accepted definition of hypotension, or appropriate organ perfusion pressure in the preterm infant. Hypotension is often defined in terms of a mean blood pressure (BP) below the 5th or 10th centile of a birth weight and age-specific reference range created from a sample of stable, 'healthy' preterm neonates (Cunningham 1999; Lee 1999). A frequently used alternative 'rule of thumb' defines hypotension as mean BP below the gestation (in completed weeks) of an infant, although the origin of this definition is unclear. In other neonatal units, mean BP below 30 mm Hg would be considered sub-optimal (Miall-Allen 1987).
Blood pressure is the product of cardiac output and systemic vascular resistance. The majority of hypotensive preterm babies have normal or high cardiac output, suggesting that in these circumstances hypotension is the result of low systemic vascular resistance due to either a haemodynamically significant ductal shunt or abnormal regulation of vasomotor tone (Pladys 1999; Kluckow 1996).
Several lines of evidence support a role for corticosteroids in the treatment of hypotension in preterm infants. Relative or absolute adrenocortical insufficiency is increasingly recognised as a cause of hypotension in the preterm infant (Watterberg 2002). Sick preterm infants have lower cord blood cortisol concentrations and a limited ability to increase cortisol production in response to stressful conditions. Cortisol concentrations are inversely related to gestational age and are particularly low in hypotensive infants receiving inotropic support (Scott 1995).
Glucocorticoids increase beta-adrenergic receptor expression in the cardiovascular system, increase responsiveness to circulating catecholamines and may, therefore, increase vascular tone and/or myocardial contractility (Sasidharan 1998). Exposure to antenatal corticosteroids is associated with a reduction in the need for blood pressure support in extremely low birth weight infants (Moise 1995). There are also several reports from uncontrolled case series supporting the efficacy of postnatal corticosteroids for pressor-resistant hypotension (Seri 2001).
However, a direct toxic effect of corticosteroids on the developing central nervous system is of particular concern. Current evidence suggests that postnatal corticosteroid treatment for prevention of preterm chronic lung disease is associated with an increase in neurodevelopmental impairment (Barrington 2001). Other potential adverse effects include increased rate of sepsis (mainly fungal), growth failure, gastrointestinal haemorrhage/perforation and hyperglycaemia (Sasidharan 1998).
Objectives
Primary:
In preterm infants with hypotension, does the use of corticosteroids (1) as primary treatment or (2) for refractory hypotension, raise blood pressure and reduce mortality and morbidity?
Secondary:
Are there any other adverse effects or benefits to the preterm infant when corticosteroids are used to treat preterm hypotension? Are there certain sub-groups of infants in whom corticosteroid therapy for hypotension is particularly effective or harmful?
Criteria for considering studies for this review
Types of studies
Randomised or quasi-randomised trials comparing oral/intravenous corticosteroid therapy with placebo, other drug or no treatment in hypotensive preterm infants.
Types of participants
Participants will be preterm infants (< 37 weeks gestation) and less than 28 days old, who have hypotension. No birthweight or lower gestational age limits.
No specific definition of hypotension required for inclusion; this is as defined in individual studies. Studies using corticosteroids as primary treatment were included as well as studies using corticosteroids in babies with hypotension resistant to inotropes/pressors and volume therapy.
Types of interventions
Oral or intravenous corticosteroid therapy versus placebo, other drug used for providing cardiovascular support (e.g. inotrope) or no therapy. Age range at initiation of corticosteroid therapy < 28 days. Trials not limited in terms of dose, duration or type corticosteroid used.
Types of outcome measures
Primary outcome measures:
1. Mortality (at 28 days of age, hospital discharge and long term mortality at two years of age)
2. Long term neurodevelopmental outcome (cerebral palsy, developmental delay, sensorineural impairment, abnormal neurological examination)
3. Adverse neuroradiological sequelae (all intraventricular haemorrhage [Grade 1 - 4, Papile 1978], severe intraventricular haemorrhage [Grade 3 - 4], periventricular leukomalacia)
4. Short term haemodynamic changes (treatment failure i.e. failure to increase BP to a predetermined threshold, increase in BP, increase in cardiac output)
Secondary outcome measures:
1. Other morbidities: Chronic lung disease (oxygen requirement at 28 days of age; oxygen requirement at 36 weeks postmenstrual age), retinopathy of prematurity (stage 1 - 4; requiring cryo/laser therapy), necrotising enterocolitis.
2. Adverse effects of steroid therapy (hyperglycaemia, sepsis (bacterial or fungal), gastrointestinal haemorrhage, gastrointestinal perforation, hypertrophic cardiomyopathy)
Search strategy for identification of studies
See: Neonatal Group search strategy
The standard methods of the Cochrane Neonatal Review Group were used.
The following electronic databases were searched:
The Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library, Issue 1, 2005)
MEDLINE/PubMed, 1966 - June 2005
EMBASE 1974 - June 2005
Search strategies for The Cochrane Central Register of Controlled Trials, MEDLINE and EMBASE were developed using index terms "infant, premature", "infant low birth weight", "blood pressure", corticosteroid", "steroid", "hydrocortisone", "prednisolone", "dexamethasone", "beclomethasone", "inotropes", "dopamine". The abstracts of the annual Pediatric Academic Societies meetings from 1995 to present were searched. Original papers were retrieved and read as required. Papers with an English abstract, written in any language, that looked relevant, were retrieved and translated. Cited references from retrieved articles were searched for additional studies. Abstracts and letters to the editor were reviewed to identify randomised controlled trials which had not been published. If a randomised controlled trial was identified, the primary investigator was contacted directly to obtain further data. Editorials, indicating expert opinion, were reviewed to identify and ensure that no key studies were missed for inclusion in this review.
Methods of the review
Two review authors (HI and KD) independently assessed and selected the studies to be included in the review. The methodological quality of each trial was assessed by each review author using the criteria of the Cochrane Collaboration, focusing on concealment of allocation, blinding of the intervention, completeness of follow-up and blinding of the outcome assessors. Data was independently extracted by each review author. If disagreement arose on the suitability of a trial for inclusion in the review or its quality, a consensus was to be reached between all three review authors by discussion.
Analysis
Studies which evaluated corticosteroids (1) as primary treatment for hypotension or (2) for hypotension unresponsive to prior use of inotropes/pressors and volume therapy, were analysed using separate comparisons. Separate analyses were conducted for each outcome. Analysis was performed on the basis of intention to treat. The data was analysed using the standard methods of the Neonatal Review Group. Treatment effect was analysed using relative risk, risk reduction, number needed to treat (NNT) for categorical outcomes and weighted mean difference for outcomes measured on a continuous scale, with 95% confidence intervals.
Meta-analysis, if appropriate, was to be carried out using a fixed effect model. Heterogeneity between studies was to be formally examined using the I2 statistic.
Subgroup analysis based on birth weight (< 1000 g and > 1000 g, gestational age (< 28 weeks, > 28 weeks) type of corticosteroid, dose, route of administration and duration of treatment was to be carried out if appropriate.
All analyses were performed using Rev Man 4.2.7 software.
Description of studies
Included Studies
Two studies were identified as meeting the criteria for inclusion in the review (Bourchier 1997; Gaissmaier 1999). Details of these studies are given in the table 'Characteristics of included studies'.
1. Bourchier 1997
This single centre, randomised trial of 40 infants investigated the effectiveness of hydrocortisone versus dopamine in the primary treatment of hypotension. Hypotension was defined as a mean arterial pressure (MAP) of less than 25 mmHg (for babies with a birth weight of 500 - 749 g), less than 30 mmHg (babies with a birth weight of 750 - 999 g), less than 35 mmHg (babies with a birth weight of 1000 - 1499 g) on two occasions, 30 minutes apart. The method of blood pressure measurement was not specified. The intervention groups were hydrocortisone (2.5 mg/kg, four - six hourly for 48 hours, followed by 1.25 mg/kg six hourly for 48 hours, and then 0.625 mg/kg for a further 48 hours before stopping treatment) or dopamine (5 - 20 micrograms/kg/minute). Concurrent treatment with volume expansion was permitted in both groups. Approximately one-third of the babies had been exposed to antenatal steroids and all had received prior treatment with a volume expander. Babies with a clinically significant patent ductus arteriosus were excluded.
Twenty-one babies received hydrocortisone and 19 received dopamine. The primary outcome measure was persisting hypotension despite treatment; other outcomes included survival and a range of neonatal complications (e.g. sepsis, intraventricular haemorrhage and bronchopulmonary dysplasia). Baseline plasma cortisol levels, their relationship with BP response and change with hydrocortisone/dopamine therapy were also examined.
2. Gaissmaier 1999
This small study of 17 infants assessed the effectiveness of dexamethasone given to preterm infants with refractory hypotension. The design was a single centre, double-blind, placebo-controlled randomised trial. Babies who remained hypotensive despite treatment with volume expansion and dopamine (maximum dose 15 micrograms/kg/minute) were eligible. Babies who had been treated postnatally with glucocorticoids for longer than three days with a maximum equivalent dose of 0.5 mg/kg/day, and the last dose administered within seven days of study enrollment were excluded. Approximately 70% of babies had been exposed to antenatal steroids. Babies with a patent ductus arteriosus were not excluded. Hypotension was defined by identifying an 'individual minimum blood pressure' for each baby. This was determined by a complex algorithm including urine output, capillary filling time and target MAP ranges [MAP < 23 mmHg (babies with a birth weight < 750 g), MAP < 25 mmHg (babies with a birth weight 750 - 999 g), MAP < 28 to 30 mmHg (babies with a birth weight 1000 - 2000 g), MAP < 35 mmHg (babies with a birth weight 2000 - 3000 g) and MAP < 40 mmHg (babies with a birth weight > 3000 g)]. Blood pressure was measured invasively using radial or umbilical arterial catheters. Infants received a single intravenous dose of dexamethasone 0.25 mg/kg or placebo (same volume of normal saline solution) concurrently with the start of the epinephrine infusion. The intervention compared, therefore, was epinephrine plus dexamethasone versus epinephrine plus placebo. Concurrent treatment with volume expansion was permitted.
Eight babies were randomised to receive dexamethasone and nine received placebo. The primary outcome was the duration of epinephrine therapy after administration of dexamethasone or placebo, and the end-point of the study was 12 hours after drug administration. Management of hypotension in general, and epinephrine therapy in particular, was standardised and protocol-based.
Two further studies (Krediet 1998; Osiovich 2000) may prove to be eligible for inclusion, but as yet have only been published in abstract form and are awaiting further information and assessment prior to inclusion.
Excluded studies
Seventeen studies were considered for inclusion but subsequently excluded for a variety of reasons. Details of these studies are given in the table 'Characteristics of excluded studies'.
Most of these studies represented case series or case-control studies of preterm babies who had received steroids for hypotension (Emery 1992; Fauser 1993; Fernandez 2005; Helbock 1993; Juren 2003; Ng 2001; Noori 2002; Noori 2004; Seri 2001; Tantivit 1999; Visveshwara 1996). Others were randomised controlled trials of inhaled or systemic steroids used as prophylaxis to prevent hypotension (Efird 2005; Kopelman 1999; Ng 2004; Vanhole 2002). Two studies studied the effects of steroids in term neonates with refractory hypotension (Lespinasse 2001; Tantivit 1999). One study compared the incidence of fungal infections in hypotensive preterm babies treated with hydrocortisone versus dexamethasone (Ramanathan 1996).
Methodological quality of included studies
Methodological quality was assessed using the standard method for conducting a systematic review described in the Cochrane Collaboration Handbook. Also see Table, Characteristics of Included Studies.
1. Minimisation of selection bias
In the study by Bourchier 1997, allocation concealment was performed using a method of sealed envelopes containing a numerical code generated from a random number table. Gaissmaier 1999 also randomised infants, but the process of allocation concealment was not stated explicitly and was, therefore, classified as 'unclear'.
2. Minimisation of performance bias
Bourchier 1997 did not attempt to mask caregivers with respect to the assigned treatment that an infant received. The two treatments were both administered intravenously, but dopamine was given as a continuous infusion and infants allocated to receive hydrocortisone were given intermittent bolus injections every four to six hours. The assignment group would, therefore, have been clear to the attending caregivers. The only way of effectively blinding caregivers would have been to give every study participant both a continuous infusion and an intermittent bolus injection, one of which would have been real and the other a placebo (a 'double-dummy' strategy).
In contrast, Gaissmaier 1999 used a placebo control (isotonic saline solution) to mask caregivers to the group assignment. Ampoules of dexamethasone, or a corresponding volume of the placebo solution, were prepared by hospital pharmacy staff not directly involved in routine clinical management.
3. Minimisation of attrition bias
Bourchier 1997 assessed all randomised infants for the primary outcome and secondary outcomes included in this review. Gaissmaier 1999 randomised 20 infants, but three 'were later excluded from the analysis'. The reasons for exclusion were not stated. Two more infants were subsequently randomised, one of whom was again excluded because of hypertrophic obstructive cardiomyopathy (not a stated exclusion criterion). The authors report outcomes for 17 babies, leaving one further baby unaccounted for. Clearly, an intention-to-treat analysis was not performed in this study.
4. Minimisation of detection bias
In both of the included studies, the method of masking for outcome assessors is not explicitly stated. Nevertheless, since the primary outcome was a short term outcome (that would have been assessed while the infant was still receiving the assigned treatment), one may assume that the outcome assessment was performed in a blinded manner in the Gaissmaier 1999 study, but not in the study performed by Bourchier 1997. Secondary outcomes would also have been assessed 'blind' in the Gaissmaier 1999 study, since the assigned treatment was masked until study completion. It remains unclear whether any attempt was made to minimise detection bias in the assessment of secondary outcomes in the study by Bourchier 1997.
Results
Two studies were included in this review; one related to primary treatment of hypotension with hydrocortisone versus dopamine (Bourchier 1997) and the other investigated the effect of treatment of refractory hypotension with dexamethasone versus placebo (Gaissmaier 1999).
See: Table of comparisons
PRIMARY TREATMENT OF HYPOTENSION
Steroid versus placebo or nothing for the primary treatment of hypotension:
There were no eligible trials to address the above comparison.
Comparison 01: Steroid versus other drug for the primary treatment of hypotension
Bourchier 1997 and colleagues compared hydrocortisone to dopamine for the primary treatment of hypotensive preterm infants.
Mortality to discharge (Outcome 01.01)
In the single included trial (Bourchier 1997), there was no evidence of an effect of hydrocortisone on mortality versus dopamine (RR 1.81, 95%CI 0.18, 18.39; RD 0.04, 95% CI -0.12, 0.20).
IVH all grades (Outcome 01.02)
Bourchier 1997 reported data on IVH grades 2 to 4. There was no evidence of an effect of hydrocortisone on IVH versus dopamine (RR 1.51, 95% CI 0.42, 5.48; RD 0.08, 95%CI -0.16, 0.33).
Retinopathy of prematurity in surviving infants (Outcome 01.03)
In the single included trial (Bourchier 1997), there was no evidence of an effect of hydrocortisone versus dopamine on retinopathy of prematurity (ROP) stages 2 to 4 (RR 1.26, 95% CI 0.33, 4.88; RD 0.04, 95%CI -0.21, 0.30).
Chronic lung disease in surviving infants (at 36 weeks post-menstrual age) (Outcome 01.04)
Bourchier 1997 reported rates of chronic lung disease (CLD) in surviving infants defined as oxygen dependency at 36 weeks post-menstrual age. There was no evidence of an effect of hydrocortisone versus dopamine on the incidence of CLD (RR 2.37, 95% CI 0.52, 10.7; RD 0.15, 95%CI -0.09, 0.40).
Necrotising enterocolitis (Outcome 01.05)
There was no evidence of an effect of hydrocortisone versus dopamine on necrotising enterocolitis (NEC) (RR 3.62, 95% CI 0.44, 29.6; RD 0.14, 95%CI -0.06, 0.33) in the trial of Bourchier 1997.
Hyperglycaemia (Outcome 01.06)
Bourchier 1997 reported data on the incidence of hyperglycaemia with hydrocortisone, defined as the need for an insulin infusion. There was no evidence of an effect of hydrocortisone versus dopamine on hyperglycaemia (RR 1.27, 95% CI 0.48, 3.33; RD 0.07, 95%CI -0.21, 0.35).
Any sepsis (Outcome 01.07)
In the trial by Bourchier 1997, there was no evidence of an effect of hydrocortisone versus dopamine on any sepsis (RR 0.60, 95% CI 0.20, 1.82; RD -0.13, 95%CI -0.39, 0.14).
Bacterial sepsis (Outcome 01.08)
There was no statistically significant effect on the incidence of bacterial sepsis (RR 0.60, 95% CI 0.20, 1.82; RD -0.13, 95%CI -0.39, 0.14) in the trial by Bourchier 1997.
Fungal sepsis (Outcome 01.09)
Bourchier 1997 reported no cases of fungal sepsis in infants treated with hydrocortisone or dopamine (RD 0, 95%CI -0.09, 0.09).
Treatment failure (Outcome 01.10)
In the trial by Bourchier 1997, treatment failure (persistent hypotension) was more common in hydrocortisone treated infants as compared to those who received dopamine, although this was of borderline statistical significance (RR 8.2, 95% CI 0.47, 142.6; RD 0.19, 95% CI 0.01, 0.37; NNT = 5.3, 95% CI 2.7,100).
Other outcomes
There are currently no data from included trials for the following outcomes:
Mortality (< 28 days), mortality (long term), cerebral palsy, developmental delay, sensorineural impairment, abnormal neurological examination, IVH grades 3/4, periventricular leukomalacia, chronic lung disease (at 28 days), gastrointestinal haemorrhage, gastrointestinal perforation, increase in mean blood pressure, increase in cardiac output or hypertrophic cardiomyopathy.
TREATMENT OF REFRACTORY HYPOTENSION
Steroid versus other drug (treatment of refractory hypotension)
There were no eligible trials to address this comparison.
Comparison 02: Steroid versus placebo or nothing
Gaissmaier 1999 and colleagues compared dexamethasone to placebo in preterm infants with refractory hypotension.
Mortality to discharge (Outcome 02.01)
The single included trial (Gaissmaier 1999) reported mortality to discharge from hospital. There was no evidence of an effect of dexamethasone versus placebo on mortality (RR 1.69, 95% CI 0.37, 7.67; RD 0.15, 95%CI -0.28, 0.58).
IVH grades 3/4 (Outcome 02.02)
In the trial by Gaissmaier 1999, there was no evidence of an effect of dexamethasone versus placebo on severe IVH (RR 1.13, 95% CI 0.20, 6.24; RD 0.03, 95%CI -0.38, 0.43).
IVH all grades (Outcome 02.03)
In the trial by Gaissmaier 1999, there was no evidence of an effect of dexamethasone versus placebo on all grades of IVH (RR 1.13, 95% CI 0.41, 3.08; RD 0.06, 95%CI -0.42, 0.53).
Periventricular leukomalacia (Outcome 02.04)
Gaissmaier 1999 found no evidence of an effect of dexamethasone versus placebo on periventricular leukomalacia (RR 1.13, 95% CI 0.20, 6.24; RD 0.03, 95%CI -0.38, 0.43).
Necrotising enterocolitis (Outcome 02.05)
Gaissmaier 1999 reported no cases of NEC in infants treated with dexamethasone or placebo (RD 0, 95% CI -0.20, 0.20).
Bacterial sepsis (Outcome 02.06)
In Gaissmaier 1999, there was no evidence of an effect of dexamethasone versus placebo on the incidence of bacterial sepsis (RR 0.75, 95% CI 0.16, 3.41; RD -0.08, 95%CI -0.51, 0.35)
Treatment failure (Outcome 02.07)
In the trial by Gaissmaier 1999, treatment failure (persistent hypotension), defined as a continuing need for epinephrine infusion, was less common in dexamethasone treated infants as compared to controls, although this was of borderline statistical significance (RR 0.42 , 95% CI 0.17, 1.06; RD -0.51, 95% CI - 0.91, - 0.12; NNT = 2, 95% CI 1.1, 8.3).
Other outcomes
There are currently no data from included trials for the following outcomes:
Mortality (< 28 days), mortality (long term), cerebral palsy, developmental delay, sensorineural impairment, abnormal neurological examination, retinopathy of prematurity, chronic lung disease (at 36 weeks post-menstrual age), chronic lung disease (at 28 days), hyperglycaemia, any sepsis, fungal sepsis, gastrointestinal haemorrhage, gastrointestinal perforation, increase in mean blood pressure, increase in cardiac output or hypertrophic cardiomyopathy.
Sub-group analyses were not possible because this data was not available from the individual study publications.
Discussion
We found two small studies that met our inclusion criteria for this review, with a further two studies that have only been published in abstract form to date, but may be eligible for inclusion in future updates. One included study addressed the efficacy of corticosteroids as primary treatment in preterm hypotension; the other examined the efficacy of corticosteroids in the treatment of refractory preterm hypotension. We were, therefore, unable to perform a meta-analysis because it would have been inappropriate to combine the data from the two studies. Both studies contained deficiencies in methodological quality, but these were not considered to be significant enough to exclude them from inclusion in this review.
1. Corticosteroids for primary treatment of hypotension
Hydrocortisone may not be as effective as dopamine as primary treatment in preterm hypotension. There was no evidence that treatment with hydrocortisone was associated with an effect (either beneficial or detrimental) on other neonatal morbidities or on neonatal mortality. It is important to emphasise that only one-third of babies in the study by Bourchier 1997 had been exposed to antenatal steroids. Since the use of antenatal steroids is associated with a decreased need for BP support in extremely low birth weight infants (perhaps through increasing neonatal circulating steroid concentrations), it could be argued that hydrocortisone therapy may be less effective in treating hypotension in preterm babies who have already been exposed to antenatal steroids.
There was no evidence of a statistically significant effect of hydrocortisone on potential adverse effects of steroid therapy such as infection or hyperglycaemia. The effect of hydrocortisone compared to dopamine with respect to long term neurodevelopmental outcome remains unknown.
2. Corticosteroids for treatment of refractory hypotension
Dexamethasone therapy for preterm infants with hypotension refractory to volume expansion and dopamine treatment, is associated with a statistically significant reduction in persisting hypotension. This benefit was demonstrated as a decrease in the continuing need for epinephrine infusion in the study by Gaissmaier 1999. The size of the observed effect (NNT, number needed to treat = 2) means that approximately two babies with refractory hypotension would have to be treated to enable cessation of epinephrine therapy 12 hours later in one baby, although it must be noted that there is a wide confidence interval around this estimate of effect. There was no evidence of a statistically significant beneficial effect of steroid therapy on short or medium term neonatal morbidity or mortality. Reassuringly, there was no evidence of a statistically significant effect of dexamethasone on potential adverse effects of steroid therapy such as infection or hyperglycaemia. However, the effect of dexamethasone on long term neurodevelopmental outcome remains unknown.
3. Limitations of the results of this review
The results of this review must be interpreted with caution. Only two studies were identified for inclusion enrolling a total of 57 babies. Given the small individual study sizes, there is substantial uncertainty (wide confidence intervals) around any estimate of treatment effect. Therefore, even a relatively large, potentially clinically significant effect cannot reliably be excluded on the basis of these studies.
The studies used different definitions of hypotension and only Gaissmaier 1999 specifically stated that all measurements of BP were made using an indwelling arterial line. Both studies concentrated on relatively short-term effects such as successful treatment of hypotension, but the issue of long term safety and effectiveness was not addressed. Neither study attempted to investigate changes in the wider range of haemodynamic parameters such as cardiac output, cardiac contractility or systemic vascular resistance and hence provided no information on the mechanism of action of steroids in increasing BP. Similarly, neither study provides information about any particular subgroups of patients who may benefit more from steroid therapy rather than continued 'conventional' treatment strategies.
Reviewers' conclusions
Implications for practice
1. There is insufficient evidence to support the routine use of hydrocortisone in the primary treatment of hypotension. Other corticosteroids have not been studied in randomized trials. There are insufficient data from randomized trials regarding the safety of steroids used for this indication.
2. A single dose of dexamethasone (0.25 mg/kg) may be effective in treating preterm infants with refractory hypotension receiving epinephrine. However, the confidence interval around the estimate of the effect on treatment failure is very wide, because it is based on the results of a single, small study. Given the lack of data on long term safety (and the concerns about the association between post-natal dexamethasone and subsequent cerebral palsy), dexamethasone cannot be recommended for routine use in preterm hypotension.
Implications for research
A number of research questions need to be addressed including the following:
1. What is the relationship between preterm hypotension, cerebral blood flow, neurological injury in the neonatal period and long term neurodevelopmental outcome? What is the underlying pathophysiology? What other factors influence this relationship?
2. What is the relationship between adrenocortical function, cortisol levels and the development of hypotension in preterm infants? Are there particular groups of babies who may respond better to early steroid therapy?
3. Is there an identifiable threshold level of BP at which cardiovascular support (including steroid therapy) should be initiated?
4. Which agents (and at what doses and by which routes) are the most effective steroids for treating preterm hypotension? Are some drugs preferable to others in certain conditions (e.g. sepsis, lack of previous exposure to antenatal steroids etc.)?
5. What are the short and long term effects of steroid therapy? Do any long term benefits outweigh the potential long term risks of therapy?
6. Is prophylactic therapy with steroids safer and more effective than treatment of 'established' or refractory hypotension?
Acknowledgements
None.
Potential conflict of interest
None known.
Characteristics of included
studies
| Study | Methods | Participants | Interventions | Outcomes | Notes | Allocation concealment |
| Bourchier 1997 | Randomised
Single centre trial
Blinding of randomization: yes
Blinding of intervention: no
Blinding of outcome assessment: unclear
Complete follow-up: yes
| Preterm hypotensive neonates
Mean (SD) gestational age: 26.6 (2.1) in hydrocortisone-treated group versus 27.5 (1.6) in dopamine treated group.
Mean (SD) birth weight: 923 (188) g in hydrocortisone-treated group versus 1043 (184) g in dopamine-treated group.
Mean (SD) age: 11.4 (13.0) hours in hydrocortisone-treated group versus 15.1 (10.1) hours in dopamine-treated group.
Treatment with antenatal steroids: 32.5%
Pre-treatment with volume expansion: yes
Pre-treatment with dopamine: no
Concurrent treatment: volume expansion
Major exclusions: major congenital abnormalities, shock requiring immediate inotropic support or treatment with blood products, clinically significant PDA, age > 7 days, birth weight > 1499 g.
| Hydrocortisone (N=21) versus dopamine (N=19)
Route: IV
Dose of hydrocortisone: 6 day course as follows: 2.5 mg/kg 4 hourly initially for 48 hours, 1.25 mg/kg 6 hourly for 48 hours, 0.625 mg/kg 6 hourly for 48 hours.
Dose of dopamine: 5 micrograms/kg/minute initially to a maximum of 20 micrograms/kg/minute
| Persistent hypotension despite treatment.
Bronchopulmonary dysplasia (oxygen dependency at 36 weeks corrected gestational age)
Intraventricular haemorrhage
Sepsis | Definition of hypotension: BP < target defined according to birth weight.
Method of blood pressure monitoring unclear.
Study period 168 hours.
| A |
| Gaissmaier 1999 | Randomised
Single centre trial
Blinding of randomization: unclear
Blinding of intervention: yes
Blinding of outcome assessment: yes
Complete follow-up: no | Preterm hypotensive neonates
Gestational age: 25-36 weeks
Birth weight: 450-2650g
Age: 1-20 days
Treatment with antenatal steroids: some
Pre-treatment with volume expansion: yes
Pre-treatment with dopamine: yes
Concurrent treatment: epinephrine, volume expansion
Major exclusions: congenital heart defect, previous administration of glucocorticoids < 7 days before study enrollment
| Dexamethasone (N=8) versus placebo (N=9)
Route: IV
Dose: 0.25mg/kg (dexamethasone) or equivalent volume of normal saline solution (placebo).
| Duration of epinephrine infusion during first 12 hours post intervention.
| Invasive BP monitoring
Definition of hypotension: BP < individually determined minimum BP
Rescue dose of dexamethasone given to babies who failed to respond to epinephrine and dexamethasone/placebo intervention.
Three infants excluded post-randomisation. | B |
Characteristics of excluded studies
| Study | Reason for exclusion |
| Efird 2005 | Randomised trial of prophylactic hydrocortisone versus placebo. |
| Emery 1992 | Case study investigating the relationship between change in blood pressure after treatment with dexamethasone to postnatal age. |
| Fauser 1993 | Retrospective case study of the effect on blood pressure of infants treated with dexamethasone. |
| Fernandez 2005 | Retrospective cohort study evauating correlation between low serum cortisol levels and response to hydrocortisone therapy. |
| Helbock 1993 | Case study of the effects of glucocorticoid treatment on the blood pressure of six preterm neonates. |
| Juren 2003 | Retrospective analysis of the response to early hydrocortisone in extremely low birth weight infants. |
| Kopelman 1999 | Randomised trial of prophylactic dexamethasone versus placebo in preterm infants. |
| Lespinasse 2001 | Retrospective case study of term infants treated with dexamethasone. |
| Ng 2001 | Cases series of five infants with corticosteroid-responsive hypotension. |
| Ng 2004 | Randomised trial of prophylactic inhaled steroid versus placebo. |
| Noori 2002 | Case study of infants treated with dexamethasone for refractory hypotension. |
| Noori 2004 | Case study of infants with pressor-resistant hypotension treated with hydrocortisone. |
| Ramanathan 1996 | Study of hydrocortisone versus dexamethasone and incidence of candida infection. |
| Seri 2001 | Retrospective review of the cardiovascular response to hydrocortisone of preterm infants with refractory hypotension. |
| Tantivit 1999 | Trial observing baseline serum cortisol values and response to glucocorticoid therapy in hypotensive infants. |
| Vanhole 2002 | Randomised placebo controlled trial of prophylactic hydrocortisone in preterm infants. |
| Visveshwara 1996 | Case study of the effects on blood pressure of hydrocortisone. |
References to studies
References to included studies
Bourchier 1997 {published data only}Bourchier D, Weston PJ. Randomised trial of dopamine compared with hydrocortisone for the treatment of hypotensive very low birth weight infants. Archives of Disease in Childhood: Fetal and Neonatal Edition 1997;76:F174-8.
Gaissmaier 1999 {published data only}
Gaissmaier RE, Pohlandt F. Single dose dexamethasone treatment of hypotension in preterm infants. Journal of Pediatrics 1999;134:701-5.
References to excluded studies
Efird 2005 {published data only}Efird MM, Heerens AT, Gordon PV, Bose CL, Young DA. A randomized controlled trial of prophylactic hydrocortisone supplementation for the prevention of hypotension in extremely low birth weight infants. Journal of Perinatology 2005;25:119-24.
Emery 1992 {published data only}
Emery EF, Greenough A. Effect of dexamethasone on blood pressure - Relationship to postnatal age. European Journal of Pediatrics 1992;151:364-6.
Fauser 1993 {published data only}
Fauser A, Pohlandt F, Bartmann P, Gortner L. Rapid increase of blood pressure in extremely low birth weight infants after a single dose of dexamethasone. European Journal of Pediatrics 1993;152:354-6.
Fernandez 2005 {published data only}
Fernandez E, Schrader R, Watterberg K. Prevalence of low cortisol values in term and near-term infants with vasopressor-resistant hypotension. Journal of Perinatology 2005;25:114-8.
Helbock 1993 {published data only}
Helbock HJ, Insoft RM, Conte FA. Glucocorticoid-responsive hypotension in extremely low birth weight newborns. Pediatrics 1993;92:715-7.
Juren 2003 {published data only}
Juren T. The effect of the early hydrocortisone administration on the blood pressure in extremely low birth weight infants. Cesko-Slovenska Pediatrie 2003;58:546-51.
Kopelman 1999 {published data only}
Kopelman AE, Moise AA, Holbert D, Hegemier SE. A single very early dexamethasone dose improves respiratory and cardiovascular adaptation in preterm infants. Journal of Pediatrics 1999;135:345-50.
Lespinasse 2001 {published data only}
Lespinasse AA, Kamat M, Pildes R, Wilks A, Pyati S. Dexamethasone in critically ill term newborns with labile blood pressure. Pediatric Research 2001;49:268A.
Ng 2001 {published data only}
Ng PC, Fok TF, Liu F, Lee CH, Ma KC, Wong E. Refractory hypotension in preterm infants with adrenocortical insufficiency. Archives of Disease in Childhood Fetal and Neonatal Edition 2001;84:122-4.
Ng 2004 {published data only}
Ng PC, Fok TF, Liu F, Lee CH, Ma KC, Wong E. Effects of inhaled corticosteroids on systemic blood pressure in preterm infants. Biology of the Neonate 2004;86:201-6.
Noori 2002 {published data only}
Noori S, Siassi B, Acherman RA, Sardesai SR, Ramanathan R. Cardiovascular responses to very low doses of dexamethasone in very low birth weight (VLBW) infants with refractory hypotension. Pediatric Research 2002;51:385A.
Noori 2004 {published data only}
Noori S, Friedlich P, Ebrahimi M, Wong P, Siassi B, Seri I. Cardiovascular effects of low-dose hydrocortisone in preterm and term infants with pressor-resistant hypotension. Pediatric Research 2004;55:550A.
Ramanathan 1996 {published data only}
Ramanathan R, Siassi B, Sardesai S, de-Lamos R. Dexamethasone versus hydrocortisone for hypotension refractory to high dose inotropic agents and incidence of candida infection in extremely low birth weight infants. Pediatric Research 1996;39:240A.
Seri 2001 {published data only}
Seri I, Tan R, Evans J. Cardiovascular effects of hydrocortisone in preterm infants with pressor-resistant hypotension. Pediatrics 2001;107:1070-4.
Tantivit 1999 {published data only}
Tantivit P, Subramanian N, Garg M, Ramanathan R, deLemos RA. Low serum cortisol in term newborns with refractory hypotension. Journal of Perinatology 1999;19:352-7.
Vanhole 2002 {published data only}
Vanhole C, Naullers G, H Devilger, Van den Berghe, de Zegher F. Early low dose hydrocortisone treatment of preterm newborns. Pediatric Research 2002;52:783A.
Visveshwara 1996 {published data only}
Visveshwara N, Peck M, Wells R, Bansal V, Chopra D, Rajani K. Efficacy of hydrocortisone in restoring blood pressure in infants on dopamine therapy. Pediatric Research 1996;39:251A.
References to studies awaiting assessment
Krediet 1998 {published data only}Krediet TG, van der Ent K, Rademaker KMA, van Bel F. Rapid increase of blood pressure after low dose hydrocortisone (HC) in low birth weight neonates with hypotension refractory to high doses of cardio-inotropics. Pediatric Research 1998;43:38A.
Osiovich 2000 {published data only}
Osiovich H, Phillipos E, Lemke RP. A short course of hydrocortisone in hypotensive neonates < 1250 g in the first 24 hours of life: A randomized, couble blind controlled trial. Pediatric Research 2000;47:422A.
* indicates the primary reference for the study
Other references
Additional references
Barrington 2001Barrington KJ. The adverse neuro-developmental effects of postnatal steroids in the preterm infant: a systematic review of RCTs. BMC Pediatrics 2001;1:1. Epub 2001 Feb 27.
Cunningham 1999
Cunningham S, Symon AG, Elton RA, Zhu C, McIntosh N. Intra-arterial blood pressure reference ranges, death and morbidity in very low birthweight infants during the first seven days of life. Early Human Development 1999;56:151-65.
Goldstein 1995
Goldstein RF, Thompson RJ Jr, Oehler J M, Brazy JE. Influence of acidosis, hypoxaemia, and hypotension on neurodevelopmental outcome in very low birth weight infants. Pediatrics 1995;95:238-43.
Kluckow 1996
Kluckow M, Evans N. Relationship between blood pressure and cardiac output in preterm infants requiring mechanical ventilation. Relationship between blood pressure and cardiac output in preterm infants requiring mechanical ventilation. Journal of Pediatrics 1996;129:506-12.
Lee 1999
Lee J, Rajadurai VS, Tan KW. Blood pressure standards for very low birthweight infants during the first day of life. Archives of Disease in Childhood Fetal and Neonatal Ed 1999;81:F168-70.
Miall-Allen 1987
Miall-Allen VM, de Vries LS, Whitelaw AG. Mean arterial blood pressure and neonatal cerebral lesions. Archives of Disease in Childhood 1987;62:1068-9.
Moise 1995
Moise AA, Wearden ME, Kozinetz CA, Gest AL, Welty SE, Hansen TN. Antenatal steroids are associated with less need for blood pressure support in extremely premature infants. Pediatrics 1995;95:845-50.
Papile 1978
Papile LA, Burstein J, Burstein R, Koffler H. PPapile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. Journal of Pediatrics 1978;92:529-34.
Pladys 1999
Pladys P, Wodey E, Beuchee A, Branger B, Betremieux P. Left ventricle output and mean arterial blood pressure in preterm infants during the 1st day of life. European Journal of Pediatrics 1999;158:817-24.
Sasidharan 1998
Sasidharan P. Role of corticosteroids in neonatal blood pressure homeostasis. Clinics in Perinatology 1998;25:723-40, xi. Review.
Scott 1995
Scott SM, Watterberg KL. Effect of gestational age, postnatal age, and illness on plasma cortisol concentrations in premature infants. Pediatric Research 1995;37:112-6.
Seri 2001
Seri I, Tan R, Evans J. Cardiovascular effects of hydrocortisone in preterm infants with pressor-resistant hypotension. Pediatrics 2001;107:1070-4.
Watterberg 2002
Watterberg KL. Adrenal insufficiency and cardiac dysfunction in the preterm infant. Pediatric Research 2002;51:422-4.
Comparisons and data
01 Steroid versus other drug (primary treatment of hypotension)
01.01 Mortality to discharge
01.02 IVH all grades
01.03 Retinopathy of prematurity in surviving
infants
01.04 Chronic lung disease in surviving infants
(at 36 weeks post-menstrual age)
01.05 Necrotising enterocolitis
01.06 Hyperglycaemia
01.07 Any sepsis
01.08 Bacterial sepsis
01.09 Fungal sepsis
01.10 Treatment failure
02 Steroid versus placebo or nothing (treatment of refractory hypotension)
02.01 Mortality to discharge
02.02 IVH grade 3 or 4
02.03 IVH all grades
02.04 Periventricular leukomalacia
02.05 Necrotising enterocolitis
02.06 Bacterial sepsis
02.07 Treatment failure
| Comparison or outcome |
Studies |
Participants |
Statistical method |
Effect size |
| 01 Steroid versus other drug (primary treatment
of hypotension) |
| 01 Mortality to discharge |
1 |
40 |
RR (fixed), 95% CI |
1.81 [0.18, 18.39] |
| 02 IVH all grades |
1 |
40 |
RR (fixed), 95% CI |
1.51 [0.42, 5.48] |
| 03 Retinopathy of prematurity in surviving infants |
1 |
37 |
RR (fixed), 95% CI |
1.26 [0.33, 4.88] |
| 04 Chronic lung disease in surviving infants (at 36 weeks
post-menstrual age) |
1 |
37 |
RR (fixed), 95% CI |
2.37 [0.52, 10.70] |
| 05 Necrotising enterocolitis |
1 |
40 |
RR (fixed), 95% CI |
3.62 [0.44, 29.60] |
| 06 Hyperglycaemia |
1 |
40 |
RR (fixed), 95% CI |
1.27 [0.48, 3.33] |
| 07 Any sepsis |
1 |
40 |
RR (fixed), 95% CI |
0.60 [0.20, 1.82] |
| 08 Bacterial sepsis |
1 |
40 |
RR (fixed), 95% CI |
0.60 [0.20, 1.82] |
| 09 Fungal sepsis |
1 |
40 |
RR (fixed), 95% CI |
Not estimable |
| 10 Treatment failure |
1 |
40 |
RR (fixed), 95% CI |
8.18 [0.47, 142.62] |
| 02 Steroid versus placebo or nothing (treatment
of refractory hypotension) |
| 01 Mortality to discharge |
1 |
17 |
RR (fixed), 95% CI |
1.69 [0.37, 7.67] |
| 02 IVH grade 3 or 4 |
1 |
17 |
RR (fixed), 95% CI |
1.13 [0.20, 6.24] |
| 03 IVH all grades |
1 |
17 |
RR (fixed), 95% CI |
1.13 [0.41, 3.08] |
| 04 Periventricular leukomalacia |
1 |
17 |
RR (fixed), 95% CI |
1.13 [0.20, 6.24] |
| 05 Necrotising enterocolitis |
1 |
17 |
RR (fixed), 95% CI |
Not estimable |
| 06 Bacterial sepsis |
1 |
17 |
RR (fixed), 95% CI |
0.75 [0.16, 3.41] |
| 07 Treatment failure |
1 |
17 |
RR (fixed), 95% CI |
0.42 [0.17, 1.06] |
Contact details for co-reviewers
Kathryn Duffy
Medical Student
University of Liverpool
89 Langdale Road
Liverpool
UK
L15 3LA
E-mail: MdOu2144@liv.ac.uk
Dr Hafis Ibrahim
Specialist Registrar in Paediatrics
Neonatal Unit
Liverpool Women's Hospital
Liverpool
UK
| The review is published as a Cochrane review in The Cochrane Library,
Issue 1, 2007 (see http://www.thecochranelibrary.com for information).
Cochrane reviews are regularly updated as new evidence emerges and in response
to feedback, and The Cochrane Library should be consulted for the most recent
version of the Review. |
