Oral dextrose gel to prevent hypoglycaemia in at‐risk neonates
Abstract
Background
Neonatal hypoglycaemia is a common condition that can be associated with brain injury. Current practice usually includes early identification of at‐risk infants (e.g. infants of diabetic mothers; preterm, small‐ or large‐for‐gestational‐age infants), and prophylactic measures are advised. However, these measures usually involve use of formula milk or admission to the neonatal unit. Dextrose gel is non‐invasive, inexpensive and effective for treatment of neonatal hypoglycaemia. Prophylactic dextrose gel can reduce the incidence of neonatal hypoglycaemia, thus potentially reducing separation of mother and baby and supporting breastfeeding, as well as preventing brain injury.
This is an update of a previous Cochrane Review published in 2017.
Objectives
To assess the effectiveness and safety of oral dextrose gel given to newborn infants at risk of hypoglycaemia in preventing hypoglycaemia and reducing long‐term neurodevelopmental impairment.
Search methods
We used the standard search strategy of Cochrane Neonatal to search the Cochrane Central Register of Controlled Trials (CENTRAL 2020, Issue 10) in the Cochrane Library; and Ovid MEDLINE(R) and Epub Ahead of Print, In‐Process & Other Non‐Indexed Citations, Daily and Versions(R) on 19 October 2020. We also searched clinical trials databases and the reference lists of retrieved articles for randomised controlled trials and quasi‐randomised trials.
Selection criteria
We included randomised controlled trials (RCTs) and quasi‐RCTs comparing oral dextrose gel versus placebo, no intervention, or other therapies for the prevention of neonatal hypoglycaemia.
Data collection and analysis
Two review authors independently extracted data and assessed risk of bias. We contacted investigators to obtain additional information. We used fixed‐effect meta‐analyses. We used the GRADE approach to assess the certainty of evidence.
Main results
We included two studies conducted in high‐income countries comparing oral dextrose gel versus placebo in 2548 infants at risk of neonatal hypoglycaemia. Of these, one study was included in the previous version of this review. We judged these two studies to be at low risk of bias, and that the evidence for most outcomes was of moderate certainty.
Meta‐analysis of the two studies showed that oral dextrose gel reduces the risk of hypoglycaemia (risk ratio (RR) 0.87, 95% confidence interval (CI) 0.79 to 0.95; risk difference (RD) ‐0.06, 95% CI ‐0.10 to ‐0.02; 2548 infants; high certainty evidence). One study reported that oral dextrose gel probably reduces the risk of major neurological disability at two years' corrected age (RR 0.21, 95% CI 0.05 to 0.78; RD ‐0.05, 95% CI ‐0.09 to 0.00; 360 infants; moderate certainty evidence).
Meta‐analysis of the two studies showed that oral dextrose gel probably reduces the risk of receipt of treatment for hypoglycaemia during initial hospital stay (RR 0.89, 95% CI 0.79 to 1.00; 2548 infants; moderate certainty evidence) but makes little or no difference to the risk of receipt of intravenous treatment for hypoglycaemia (RR 1.01, 0.68 to 1.49; 2548 infants; moderate certainty evidence). Oral dextrose gel may have little or no effect on the risk of separation from the mother for treatment of hypoglycaemia (RR 1.12, 95% CI 0.81 to 1.55; two studies, 2548 infants; low certainty evidence).
There is probably little or no difference in the risk of adverse events in infants who receive oral dextrose gel compared to placebo gel (RR 1.22, 95% CI 0.64 to 2.33; two studies, 2510 infants; moderate certainty evidence), but there are no studies comparing oral dextrose with other comparators such as no treatment, standard care or other therapies.
No data were available on exclusive breastfeeding after discharge.
Authors' conclusions
Oral dextrose gel reduces the risk of neonatal hypoglycaemia in at‐risk infants and probably reduces the risk of major neurological disability at two years of age or greater without increasing the risk of adverse events compared to placebo gel. Additional large follow‐up studies at two years of age or older are required. Future research should also be undertaken in low‐ and middle‐income countries, preterm infants, using other dextrose gel preparations, and using comparators other than placebo gel. There are three studies awaiting classification and one ongoing study which may alter the conclusions of the review when published.
PICOs
Plain language summary
Oral dextrose gel for prevention of low blood glucose levels in newborn babies
Review question
Is oral dextrose gel effective and safe in preventing low blood glucose levels and reducing long‐term disability in newborn babies at risk of low blood glucose levels?
Background
Low blood sugar (glucose) levels are important because they are common and are associated with brain injury in newborn babies. Up to 15 of every 100 babies will have low blood glucose levels over the first few days after they are born, and as many as half of babies who are at higher risk (those born preterm, or smaller or larger than usual, or whose mothers have diabetes).
Low blood glucose levels can cause problems with academic achievement and development during childhood. Some evidence suggests that even one episode of low blood glucose or episodes that are undetected can contribute to these problems. Therefore, it would be useful to prevent low glucose levels from occurring. Additionally, treatments for low glucose levels often include formula milk or admission to the neonatal unit, leading to the separation of mother and baby. The treatments and the separation may both impair breastfeeding.
Dextrose (sugar) gel can be rubbed on the inside of a baby's mouth, where the sugar can be absorbed and help raise blood glucose levels, thus potentially helping prevent low glucose levels.
Study characteristics
We identified two studies in high‐income countries that compared oral dextrose gel with placebo (inactive) gel for preventing low blood glucose levels in 2548 at‐risk infants. The evidence in this Cochrane Review is current to October 2020.
Key results
Two studies showed that preventative oral dextrose gel reduces the risk of low blood glucose levels in newborn infants at risk. A single study reporting two‐year outcomes in 360 infants found that oral dextrose gel given to at‐risk infants to prevent low blood glucose levels probably reduces the risk of major disability at two years of age, but additional follow‐up studies are needed.
Two studies showed that oral dextrose gel probably reduces the chance of receiving any treatment for low blood glucose during initial hospital stay but makes little or no difference to the chance of receiving intravenous treatment for low glucose, or separation from the mother for treatment of low glucose levels.
Evidence from two studies suggests that infants given oral dextrose gel are not at a higher risk of adverse events (harms) such as choking or vomiting compared with infants given placebo gel, but there was no information to assess whether oral dextrose gel is safer than no treatment or other therapies. No data were available on exclusive breastfeeding after discharge.
We assessed the risk of bias in the included studies as low, meaning it is unlikely that there is a systematic error in estimating the effect of the intervention.
Future research should be undertaken in low‐ and middle‐income countries, preterm infants, using other dextrose gel preparations and comparators other than placebo gel. There are three studies awaiting classification and one ongoing study which may alter the conclusions of the review when they are published.
Certainty of evidence
We graded the certainty of the evidence as moderate for all outcomes except risk of low blood glucose levels (assessed as high certainty) and separation from mother (assessed as low certainty).
Authors' conclusions
Summary of findings
| Dextrose gel compared with placebo for prevention of hypoglycaemia in newborn infants | ||||||
| Patient or population: newborn infants at risk of neonatal hypoglycaemia | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | No. of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with placebo | Risk with dextrose gel | |||||
| Hypoglycaemia (investigator‐defined) | 433 per 1000 | 377 per 1000 | RR 0.87 (0.79 to 0.95) | 2548 (2 RCTs) | ⊕⊕⊕⊕ | |
| Major neurological disability at 2 years of age or older (Defined as any of the following: legal blindness, sensorineural deafness requiring hearing aids, moderate or severe cerebral palsy or developmental delay/intellectual impairment (developmental quotient or intelligence quotient lower than 2 SD below the mean)) | 60 per 1000 | 47 per 1000 | RR 0.21 | 360 (1 RCT) | ⊕⊕⊕⊝ | Low event rates: 3/277 in dextrose gel group and 7/138 in placebo group. |
| Receipt of treatment for hypoglycaemia during initial hospital stay (investigator‐defined, any treatment ‐ oral dextrose gel, intravenous dextrose, or other drug therapy) during initial hospital stay (yes/no) | 316 per 1000 | 281 per 1000 | RR 0.89 | 2548 | ⊕⊕⊕⊝ | |
| Receipt of intravenous treatment for hypoglycaemia (yes/no) | 37 per 1000 | 37 per 1000 | RR 1.01 | 2548 | ⊕⊕⊕⊝ | |
| Adverse events# (e.g. choking or vomiting at time of administration) | 10 per 1000 | 12 per 1000 (6 to 24) | RR 1.22 (0.64 to 2.33) | 2510 (2 RCTs) | ⊕⊕⊕⊝ | Low event rates: 27/1322 in dextrose gel group and 12/1188 in placebo group. |
| Separation from mother for treatment of hypoglycaemia (infant nursed in an environment that is not in the same room as the mother, e.g. for NICU admission or the like) (yes/no)) | 50 per 1000 | 56 per 1000 | RR 1.12 (0.81 to 1.55) | 2548 (2 RCTs) | ⊕⊕⊝⊝ LOWc,d | |
| Breastfeeding (exclusive after discharge) ‐ not reported (WHO 2008 definition (yes/no)) | ‐ | ‐ | ‐ | ‐ | ‐ | No data were reported for this outcome |
| *The risk in the intervention group (and its 95% confidence interval) is based on assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI); #Placebo gel was the comparator. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level for serious imprecision (due to low event rates) | ||||||
Background
Description of the condition
Hypoglycaemia is the most common metabolic disorder of the newborn (Hay 2009), and is potentially preventable (Chertok 2009; Singhal 1991; Singhal 1992). Neonatal hypoglycaemia can cause both brain damage (Burns 2008; Duvanel 1999; Kerstjens 2012; Koh 1988), and death (Achoki 2010; Cornblath 1965; Nadjm 2013; Willcox 2010). Some authors have reported brain abnormalities associated with neonatal hypoglycaemia on magnetic resonance imaging (MRI). Early studies reported that the most common site of damage is the occipital cortex (Alkalay 2005; Spar 1994). However, more recent studies have reported widespread MRI changes in the temporoparietal region, cerebral cortex and basal ganglia/thalamus (Burns 2008).
Up to 15% of newborn infants will have low blood glucose concentrations (Hay 2009). This rate is much higher among infants with additional risk factors: up to 50% in infants of diabetic mothers (Maayan‐Metzger 2009), and 66% in preterm infants (Lucas 1988). Those at highest risk of hypoglycaemia are infants of diabetic mothers (Agrawal 2000; Maayan‐Metzger 2009), born large for gestation (Weissmann‐Brenner 2012), small for gestation (Hawdon 1993), or preterm (Kerstjens 2012). Fifty per cent of these infants will develop at least one episode of hypoglycaemia, and 20% will have more than one episode (Harris 2012). Additional risk factors for neonatal hypoglycaemia include perinatal asphyxia (Salhab 2004), prolonged labour, hypothermia, sepsis, and maternal medications such as b‐agonists (Kurtoglu 2005), and beta‐blockers (Daskas 2013).
The definition of neonatal hypoglycaemia remains controversial, and various definitions have been proposed (Agrawal 2000; Burns 2008; Kerstjens 2012; Maayan‐Metzger 2009), or suggested as thresholds for intervention (Adamkin 2011; Cornblath 2000). A blood glucose concentration of less than 2.6 mmol/L has been widely accepted as a definition of hypoglycaemia (Harris 2014). This was heavily influenced by two studies. One described abnormal sensory evoked potentials among infants with blood glucose concentrations lower than 2.6 mmol/L (Koh 1988). The second reported a relationship between the number of days on which blood glucose measurements lower than 2.6 mmol/L were recorded in preterm infants and neurodevelopmental impairment at 18 months (Lucas 1988), and at seven to eight years (Lucas 1999). Infants who are ‘asymptomatic’ during periods of hypoglycaemia were previously considered to have a better outcome than those who exhibit signs (Hawdon 1993; Kalhan 2000; Koivisto 1972). However, Stenninger and colleagues found that longer‐term neurodevelopmental outcomes did not differ between infants who exhibited signs and those who did not (Stenninger 1998).
The effects of transient neonatal hypoglycaemia on longer‐term outcomes are not yet fully defined. However, some evidence suggests that even transient and undetected episodes may be associated with adverse outcomes (McKinlay 2017). For example, a retrospective population study demonstrated an association between a single initial blood glucose concentration of less than 2.6 mmol/L, followed by a repeat result above this, and lower academic test scores at 10 years of age (Kaiser 2015). Further, a prospective cohort study of infants at risk of neonatal hypoglycaemia, half of whom became hypoglycaemic and were treated to maintain blood glucose concentrations of 2.6 mmol/L or higher, found that clinically undetected low interstitial glucose concentrations were associated with an increased risk of executive dysfunction at four and a half years of age (McKinlay 2017). These findings suggest that even an effective treatment for neonatal hypoglycaemia may not be enough to optimise outcomes of all babies, and that prophylaxis should be considered.
Treatment of neonatal hypoglycaemia commonly requires admission to a newborn intensive care unit (NICU) or special care baby unit (SCBU), separating mothers and infants and interfering with the establishment of breastfeeding, thus incurring a high social and financial cost. The World Health Organization (WHO) states “… an approach aimed first at the prevention of hypoglycaemia, second at its reliable detection in infants at risk and third at appropriate treatment which will not be deleterious to breastfeeding is … of global importance” (WHO 1997). The American Academy of Pediatrics recommends “… early identification of the at‐risk infant and institution of prophylactic measures to prevent neonatal hypoglycaemia” (Adamkin 2011).
Widely accepted clinical monitoring and management of infants at risk of neonatal hypoglycaemia involves:
-
early identification of pertinent risk factors;
-
early feeding, skin‐to‐skin contact and ensuring that newborns are kept warm and dry;
-
pre‐feed blood glucose concentration measurement to determine blood glucose concentrations at the time when it is most at risk of being low; and
-
monitoring during the period of highest risk until blood glucose concentration is demonstrated to remain above the chosen threshold for intervention (Adamkin 2011; CPSFNC 2004; NICE 2008; UNICEF 2013; WHO 1997).
However, despite recommendations to prevent neonatal hypoglycaemia, there is little evidence of effective interventions to achieve this.
Early skin‐to‐skin contact may reduce the incidence of neonatal hypoglycaemia because it has a protective influence on thermo‐regulation (UNICEF 2013). A meta‐analysis of two randomised trials of low birthweight infants found that compared with standard of care, skin‐to‐skin contact was associated with a decreased risk of neonatal hypoglycaemia (Boundy 2016). Further, a Cochrane Review of three randomised trials of 144 healthy full‐term, low birthweight infants reported that skin‐to‐skin contact during the establishment of breastfeeding was associated with a 10 mg/dL (0.6 mmol/L) higher blood glucose concentration at 75 to 90 minutes after birth compared with standard contact. However, the evidence was graded as low certainty (Moore 2016). A retrospective pre‐ and post‐intervention study of 561 infants at risk of neonatal hypoglycaemia reported that prolonged skin‐to‐skin contact for at least 12 hours after birth during blood glucose monitoring was associated with a 4.6% reduction in admission rates to the NICU for neonatal hypoglycaemia and a 3.8% reduction in the number of infants receiving intravenous dextrose (Chiruvolu 2017). However, the small sample sizes and poor quality of some of these studies limit confidence in the findings.
Early feeding is also recommended for prevention of neonatal hypoglycaemia. In infants of diabetic mothers, early feeding (within 30 minutes of birth) was reported to decrease the incidence of subsequent neonatal hypoglycaemia, and increase mean blood glucose concentration (Chertok 2009). However, two other studies showed that the early initiation of breastfeeding within 30 minutes to one hour after birth does not affect blood glucose concentration at one and two hours in infants without risk factors for neonatal hypoglycaemia (Sweet 1999; Zhou 2017).
Antenatal expression of colostrum may potentially increase the available colostrum supply for infants soon after birth, while reducing the time taken to establish breastfeeding and decreasing the use of formula (Cox 2006). However, the Diabetes and Antenatal Milk Expressing (DAME) study that randomised women with diabetes to antenatal expression of breastmilk or standard care reported no differences between groups for mean blood glucose concentrations (Forster 2011), incidence of neonatal hypoglycaemia, admission to NICU for hypoglycaemia or time until euglycaemic status was achieved (defined as three consecutive blood glucose measurements of 2.6 mmol/L or higher) (Forster 2017). This is consistent with a post hoc analysis of a randomised controlled trial which showed that breastmilk expressed either before or after birth and given to hypoglycaemic infants did not increase glucose concentrations (Harris 2017).
Supplementation or substitution of breastfeeding with fluid or foods other than expressed breast milk may reduce the duration of breastfeeding (Blomquist 1994; Demir 2020; Smith 2016). Therefore, the commonly accepted practice is to advise exclusive breastfeeding (Eidelman 2012; UNICEF 2013). Healthy newborn infants will usually maintain their blood glucose concentration despite the small‐volume, low‐energy food source provided by colostrum. However, colostrum alone cannot be relied upon to meet the essential energy needs of infants with additional risk factors for neonatal hypoglycaemia. Thus, infants at risk of hypoglycaemia frequently receive supplemental or complementary feeding during establishment of feeding (Blomquist 1994; Harris 2013).
Powdered sugar has been used as an addition to formula in an attempt to prevent neonatal hypoglycaemia. Two randomised trials in India compared formula with formula plus added powdered sugar for prevention of subsequent hypoglycaemia in infants at risk of hypoglycaemia (small or large for gestational age) (Singhal 1991; Singhal 1992). Both studies reported a significant reduction in the incidence of subsequent hypoglycaemia among infants who received formula plus powdered sugar. In contrast, a randomised trial in Thailand of 425 infants at risk of neonatal hypoglycaemia reported no difference in the blood glucose concentrations measured at one, three and six hours after birth of infants who received 200 mg/kg prophylactic oral sucrose solution plus feeding versus infants who received feeding alone (Surachaidungtavil 2020).
The ideal intervention would be effective in preventing hypoglycaemia while reducing the need for artificial formula, improving breastfeeding rates, reducing costs, as well as potentially reducing the risk of later adverse outcomes. Oral dextrose gel is an effective first line treatment for neonatal hypoglycaemia (Harris 2013), that also improves breastfeeding (Weston 2016), reduces costs (Glasgow 2018), and appears safe (Harris 2016). This is an updated review of the previously published Cochrane Review, “Oral dextrose gel to prevent hypoglycaemia in at‐risk neonates,” which found that oral dextrose gel was reported to be effective and safe in reducing the incidence of neonatal hypoglycaemia in one randomised trial (Hegarty 2017). No longer‐term outcomes were available.
Description of the intervention
Dextrose gel is a non‐proprietary, low‐cost, simple carbohydrate in concentrated thickened aqueous solution, which can be administered by direct application to the oral mucosa ‐ buccal or sublingual. Administration via these highly vascularised, thin mucous membranes allows rapid access to the circulation. Some of the administered gel may be swallowed and absorbed from the gastrointestinal tract.
Commercially manufactured gel costs approximately USD (US dollars) 70 per 100 mL. Alternatively, gel can be prepared in hospital pharmacies (Harris 2013). Ingredients vary by pharmaceutical manufacturer but commonly include water, glucose, a gelling agent and preservative(s). Some preparations include flavourings and colourings. Suitability of the gel for use in neonates should be assessed on an individual basis. The difference in effectiveness of various formulations is unknown.
How the intervention might work
Dextrose gel administered to the oral mucosa will enter the systemic circulation via the lingual vein and the internal jugular vein. This contrasts with oral‐gastrointestinal administration, whereby the first pass effect of the portal circulation may diminish the systemic blood glucose concentration achieved. Prevention of neonatal hypoglycaemia achieved by providing additional glucose during the neonatal metabolic transition period may reduce the medical prescription of artificial formula feeds, reduce admission to the NICU for intravenous dextrose, and prevent the neurodevelopmental impairment associated with neonatal hypoglycaemia.
Why it is important to do this review
Neonatal hypoglycaemia is important because it is common and is associated with brain injury in newborn infants. Risk factors for neonatal hypoglycaemia are known, so specific groups of newborn infants are routinely targeted for screening (i.e. infants of diabetic mothers, those of high or low birth weight, preterm infants, and those with poor feeding). These infants are frequently treated prophylactically with supplemental formula milk or admission to the neonatal unit for intravenous dextrose, or both. Supplemental formula may impair establishment of breastfeeding; intravenous treatment is expensive, is not always available in resource‐poor settings, and usually requires separation of mother and infant.
Oral dextrose gel is simple to administer and inexpensive. Therefore, if it were found to be effective and safe in preventing neonatal hypoglycaemia, its use would provide many advantages, particularly in low‐resource settings. Results of this review may help to inform those preparing clinical practice guidelines, such as those currently available to guide the care of babies at risk of neonatal hypoglycaemia (Adamkin 2011; NICE 2008; UNICEF 2013).
Objectives
To assess the effectiveness and safety of oral dextrose gel given to newborn infants at risk of hypoglycaemia in preventing hypoglycaemia and reducing long‐term neurodevelopmental impairment.
Methods
Criteria for considering studies for this review
Types of studies
We included randomised controlled trials (RCTs) and quasi‐RCTs, including cluster‐randomised trials but not cross‐over trials. We included both published and unpublished studies. We included unpublished studies and studies published only as abstracts when assessment of study quality was possible and if other criteria for inclusion were fulfilled.
Types of participants
Newborn infants at risk of hypoglycaemia, including infants of diabetic mothers (all types), large for dates, small for dates, and those born preterm (< 37 weeks) or with other risk factors as determined by investigators (e.g. maternal medication such as beta‐blockers from birth to 24 hours of age), who had not yet received a diagnosis of hypoglycaemia (blood glucose concentration below normal range, investigator‐defined) and had not received treatment for hypoglycaemia.
Types of interventions
Dextrose gel, of any concentration and at any dose or number of doses, given orally compared with placebo, no treatment/standard care, or other therapies (such as antenatal expression of colostrum, early initiation of breastfeeding, supplementation or substitution of breastfeeding with formula milk), for prevention of hypoglycaemia at any gestational age and commenced within the first 24 hours following birth.
Types of outcome measures
Primary outcomes
-
Hypoglycaemia (investigator‐defined)
-
Major neurological disability at two years of age or older, defined as any of the following: legal blindness, sensorineural deafness requiring hearing aids, moderate or severe cerebral palsy, or developmental delay/intellectual impairment (developmental quotient or intelligence quotient lower than two standard deviations below the mean)
Secondary outcomes
-
Hypoglycaemia (any blood glucose concentration < 2.6 mmol/L) during initial hospital stay (yes/no)
-
Receipt of treatment for hypoglycaemia (investigator‐defined, any treatment ‐ oral dextrose gel, intravenous dextrose, or other drug therapy) during initial hospital stay (yes/no)
-
Receipt of intravenous treatment for hypoglycaemia (yes/no)
-
Receipt of oral dextrose gel treatment for hypoglycaemia (yes/no)
-
Receipt of any medication for hypoglycaemia, such as glucagon or corticosteroids (yes/no)
-
Number of episodes of hypoglycaemia (investigator‐defined) (total number per infant)
-
Adverse events (e.g. choking or vomiting at time of administration) (yes/no)
-
Separation from mother for treatment of hypoglycaemia (infant nursed in an environment that is not in the same room as the mother, e.g. for NICU admission or the like) (yes/no)
-
Neonatal seizures (yes/no)
-
Abnormal MRI of the brain in the neonatal period (yes/no)
-
Duration of initial hospital stay (days)
-
Breastfeeding (any) after discharge (yes/no)
-
Exclusive breastfeeding after discharge ‐ WHO 2008 definition (yes/no)
-
Exclusive breastfeeding at six months of age ‐ WHO 2008 definition (yes/no)
-
Developmental disability at two years of age or older ‐ investigator‐defined (yes/no)
-
Visual impairment and severity of impairment at two years of age or older
-
Hearing impairment and severity of impairment at two years of age or older
-
Cerebral palsy and severity of disorder at two years of age or older
-
Developmental delay/intellectual impairment and severity of impairment at two years of age or older
-
Executive dysfunction and severity of dysfunction at two years of age or older
-
Behavioural problems and severity of problems at two years of age or older
-
Abnormal MRI of the brain at two years of age or older
Search methods for identification of studies
Electronic searches
We conducted a comprehensive search including: Cochrane Central Register of Controlled Trials (CENTRAL 2020, Issue 10) in the Cochrane Library; and Ovid MEDLINE(R) and EpubAhead of Print, In‐Process & Other Non‐Indexed Citations, Daily and Versions(R) (1946 to 19 October 2020). We have included the search strategies for each database in Appendix 1. We did not apply language restrictions.
We searched clinical trial registries for ongoing or recently completed trials. We searched the World Health Organization’s International Clinical Trials Registry Platform (ICTRP) (www.who.int/ictrp/search/en/) and the US National Library of Medicine’s ClinicalTrials.gov (clinicaltrials.gov) via Cochrane CENTRAL. Additionally, we searched the ISRCTN Registry (www.isrctn.com) and the ANZCTR Registry (www.anzctr.org.au) for any unique trials not found through the Cochrane CENTRAL search.
For the 2020 update, we developed a new search strategy. The previous search methods are available in Appendix 2.
Searching other resources
We searched the reference lists of included studies and approached well‐known researchers in this clinical area to identify any unpublished or ongoing research.
Data collection and analysis
We used standard methods of the Cochrane Neonatal and GRADE to assess the certainty of evidence.
Selection of studies
Two review authors (TE, GL) independently assessed the eligibility of each study for inclusion. We did not encounter any disagreements.
Data extraction and management
Two review authors (TE, GL) independently extracted data from the eligible studies. We contacted investigators of two studies included in the meta‐analysis for additional data (Harding 2020; Hegarty 2016a). We did not encounter any disagreements.
Assessment of risk of bias in included studies
Two review authors (TE, GL) independently assessed the risk of bias (low, high, or unclear) of all included studies using the Cochrane ‘Risk of bias’ tool (Higgins 2011), for these domains:
-
sequence generation (selection bias);
-
allocation concealment (selection bias);
-
blinding of participants and personnel (performance bias);
-
blinding of outcome assessment (detection bias);
-
incomplete outcome data (attrition bias);
-
selective reporting (reporting bias);
-
any other bias.
We resolved any disagreements encountered by discussion or consultation with a third review author. See Appendix 3 for a more detailed description of each domain.
Measures of treatment effect
We used the numbers of events in control and intervention groups of each study to calculate risk ratios (RRs) for dichotomous data. We calculated mean differences (MDs) between treatment groups when outcomes were measured in the same way for continuous data. We reported risk differences (RDs), and when a significant effect was found, we calculated numbers needed to treat for an additional beneficial outcome (NNTB), or numbers needed to treat for an additional harmful outcome (NNTH). We reported 95% confidence intervals (CIs) for all outcomes.
Unit of analysis issues
We planned to include cluster‐randomised trials in analyses along with individually randomised trials, but we did not identify any cluster‐randomised trials for inclusion.
Dealing with missing data
We carried out analyses on an intention‐to‐treat basis for all outcomes. We contacted the original investigators to request missing data when possible.
Assessment of heterogeneity
We considered whether clinical and methodological characteristics of included studies were sufficiently similar for meta‐analysis to provide a clinically meaningful summary by assessing statistical heterogeneity using the Chi2 test and the I2 statistic, considering an I2 value of less than 25% to be none, 25% to 49% low, 50% to 74% moderate, and 75% or greater to be high heterogeneity. We took an I2 value greater than 50% and a low P value (< 0.10) from the Chi2 test for heterogeneity to indicate substantial heterogeneity (Higgins 2019). If we detected substantial heterogeneity, we planned to explore possible explanations by performing sensitivity or subgroup analyses but we were unable to do this given the small number of eligible trials. We took statistical heterogeneity into account when interpreting results, especially when we noted variation in the direction of effect.
Assessment of reporting biases
Reporting biases arise when dissemination of research findings is influenced by the nature and direction of results. Some types of reporting bias (e.g. publication bias, multiple publication bias, language bias) reduce the likelihood that all studies eligible for a review will be retrieved. If all eligible studies are not retrieved, the review may be biased. We conducted a comprehensive search for eligible studies and were alert for duplication of data. We were unable to assess publication bias by visually inspecting a funnel plot because we did not identify 10 or more trials to make such an inspection valid. Two review authors (TE, GL) examined the methods of each study for prespecified outcomes and if all prespecified outcomes were reported in the results, we considered the study carried low risk of bias. If any prespecified outcome was not reported in the results, we considered the study to carry higher risk of bias.
Data synthesis
We evaluated studies for potential clinical diversity and planned to restrict meta‐analysis to situations in which clinical consistency was apparent. We evaluated studies for bias, as above, and restricted meta‐analysis if bias would be compounded. We used a fixed‐effect model to combine data when it was reasonable to assume that studies were estimating the same underlying treatment effect. If we found evidence of clinical heterogeneity, we planned to try to explain this on the basis of different study characteristics and subgroup analyses.
Subgroup analysis and investigation of heterogeneity
We planned to carry out the following subgroup analyses using a fixed‐effect model.
-
Reason for risk of hypoglycaemia (infant of diabetic mother, preterm, small, large, other).
-
Gestation at birth (term and post‐term versus late preterm 35 to 36 weeks versus moderately preterm 30 to 34 weeks versus extremely preterm < 30 weeks).
-
Actual mode of feeding (formula versus breast versus mixed).
-
Method of administration of gel (rubbed into buccal mucosa versus sublingual versus other).
-
Dose of dextrose gel per administration (≤ 200 mg/kg versus > 200 mg/kg).
-
Number of dextrose gel doses administered (one versus > one dose).
-
Time of administration of first dose of gel (≤ one hour of age versus after one hour of age versus after two hours of age).
Sensitivity analysis
We planned to conduct sensitivity analysis by examining only trials considered to have a low risk of bias across all domains as determined by the Cochrane ‘Risk of bias’ tool (Higgins 2011). We planned to report results of sensitivity analyses for primary outcomes only. However, we did not conduct the planned sensitivity analysis because included trials had a low risk of bias.
Summary of findings and assessment of the certainty of the evidence
We used the GRADE approach, as outlined in the GRADE Handbook (Schünemann 2013), to assess the certainty of evidence for the following (clinically relevant) outcomes: hypoglycaemia; major neurological disability at two years of age or older; receipt of treatment for hypoglycaemia during initial hospital stay; receipt of intravenous treatment for hypoglycaemia; adverse events; separation from the mother for treatment of hypoglycaemia; and exclusive breastfeeding after discharge.
Two review authors (TE, GL) independently assessed the certainty of evidence for each of the outcomes. We considered evidence from randomised controlled trials as high certainty but downgraded the evidence by one level for serious (or two levels for very serious) limitations on the basis of the following: design (risk of bias), consistency across studies, directness of evidence, precision of estimates, and presence of publication bias. We used the (GRADEpro GDT) Guideline Development Tool to create ‘summary of findings Table 1’ to report the certainty of evidence.
The GRADE approach results in an assessment of the certainty of a body of evidence as one of four grades.
-
High certainty: further research is very unlikely to change our confidence in the estimate of effect.
-
Moderate certainty: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
-
Low certainty: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
-
Very low certainty: we are very uncertain about the estimate.
Results
Description of studies
We included two studies in this update: one new study (Harding 2020), that was identified as an ongoing study in the previous Cochrane Review (Hegarty 2017); and one study that was included in the previous review (Hegarty 2016a), that had new follow‐up data.
Results of the search
See Characteristics of included studies, Characteristics of excluded studies, Characteristics of studies awaiting classification, and Characteristics of ongoing studies.
Our search, in October 2020, identified 78 records, of which 18 were duplicates. After screening the titles and abstracts of 60 records, we excluded 48 because they were not relevant, and assessed the full text of 12 records (see Figure 1). We excluded four primary and two secondary references (Bourchier 1992; Coors 2018; Retbi 2013; Van Loghum 2014). One study is ongoing (NCT04353713); and three are awaiting classification (CTRI/2017/11/010645; NCT04185766; TCTR20190805003).
Study flow diagram
Included studies
We identified two eligible randomised, blinded, placebo‐controlled trials (Harding 2020; Hegarty 2016a), that enrolled a total of 2548 late preterm and term infants born at risk of neonatal hypoglycaemia (see Characteristics of included studies).
The study by Hegarty 2016a took place in two centres in New Zealand. They randomised 416 infants (including one infant randomised in error) in a 2:1 ratio to one of the following dose regimens of 40% dextrose gel or placebo: 0.5 mL/kg once, 1 mL/kg once, 0.5 mL/kg for four doses, and 1 mL/kg once followed by 0.5 mL/kg for three additional doses. Gel was massaged into the buccal mucosa at one hour after birth, followed by a breastfeed. Blood glucose concentration was checked using the glucose oxidase method at two hours after birth, and subsequent measurements were performed according to local hospital protocol (two to four hours pre‐feed for at least the first 12 hours). Of those randomised, 401 infants were eligible for follow‐up at two years' corrected age (13 withdrawals, one death) and a total of 360 were assessed.
The study by Harding 2020 recruited from 18 maternity hospitals in New Zealand and Australia. A total of 2149 infants born at risk of neonatal hypoglycaemia but unlikely to require neonatal unit admission for other reasons were randomised to receive a single dose of 40% dextrose gel or placebo gel massaged into the buccal mucosa at one hour after birth followed by a breastfeed. Blood glucose concentration was measured at two hours after birth and then according to hospital standard practice.
Excluded studies
We excluded four studies (six records) for the following reasons:
-
Three studies were commentaries which evaluated dextrose gel as a treatment for neonatal hypoglycaemia (Bourchier 1992; Retbi 2013; Van Loghum 2014).
-
One study (three records) was a non‐randomised study which investigated oral dextrose gel as prophylaxis against transient neonatal hypoglycaemia in at‐risk infants (Coors 2018).
See Characteristics of excluded studies for more details.
Risk of bias in included studies
Refer to Figure 2 and Figure 3 for a summary of the ‘Risk of bias’ assessment and Characteristics of included studies for more details.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies
Risk of bias summary: review authors' judgements about each risk of bias item for each included study
Allocation
Both studies reported using computer‐generated block randomisation with variable block sizes (Harding 2020; Hegarty 2016a). Allocation was concealed by a central randomisation system which assigned infants to a numbered trial pack containing identical‐looking syringes of either dextrose or placebo gel. We judged these studies to be at a low risk of selection bias.
Blinding
Both studies reported that clinicians, families and all study personnel were masked to treatment allocation (Harding 2020; Hegarty 2016a). The two‐year follow‐up of one study reported that follow‐up assessors were unaware of the infant’s treatment allocation (Hegarty 2016a). We judged these studies to be at low risk of performance and detection bias.
Incomplete outcome data
Both studies included all randomised infants in the intention‐to‐treat analysis of neonatal outcomes (Harding 2020; Hegarty 2016a). Hegarty 2016a assessed 90% of those eligible at two years' corrected age (dextrose 243/271 and placebo 117/130) and 87% of those randomised (dextrose 243/277 and placebo 117/138). Maternal and child characteristics were similar in those who were and were not assessed at two years. We judged these studies to be at low risk of attrition bias.
Selective reporting
Both studies reported data for all outcomes prespecified in the trial registration documentation (Harding 2020; Hegarty 2016a). We judged these studies to be at low risk of reporting bias.
Other potential sources of bias
Both studies reported that the demographic and prognostic characteristics of infants in each trial arm were balanced at baseline (Harding 2020; Hegarty 2016a). We did not identify any other potential sources of bias. We judged these studies to be at low risk of other bias.
Effects of interventions
Primary outcomes
1.1 Hypoglycaemia (investigator‐defined)
Meta‐analysis showed that oral dextrose gel reduces the risk of neonatal hypoglycaemia (defined as a blood glucose concentration < 2.6 mmol/L) compared with placebo gel (RR 0.87, 95% CI 0.79 to 0.95; two studies, 2548 infants; high certainty evidence; Analysis 1.1). The RD was ‐0.06 (95% CI ‐0.10 to ‐0.02), and on average, 17 infants would have to receive prophylactic dextrose gel to prevent one additional case of neonatal hypoglycaemia.
1.2 Major neurological disability at two years or older
Additional data from one study (Hegarty 2016a) showed that oral dextrose gel probably reduces the risk of major neurological disability at two years’ corrected age compared with placebo gel (RR 0.21, 95% CI 0.05 to 0.78; one study, 360 infants; moderate certainty evidence, downgraded for serious imprecision due to low event rates; Analysis 1.2). The RD was ‐0.05 (95% CI ‐0.09 to 0.00), and on average, 20 infants would have to receive prophylactic dextrose gel to prevent one additional case of major neurological disability.
Secondary outcomes
1.3 Hypoglycaemia (any blood glucose concentration less than 2.6 mmol/L) during initial hospital stay
No data were available for this outcome.
1.4 Receipt of treatment for hypoglycaemia during initial hospital stay
Meta‐analysis showed that oral dextrose gel probably reduces the risk of receipt of treatment for neonatal hypoglycaemia compared with placebo gel (RR 0.89, 95% CI 0.79 to 1.00; two studies, 2548 infants; moderate certainty evidence, downgraded for serious imprecision; Analysis 1.3).
1.5 Receipt of intravenous treatment for hypoglycaemia
Meta‐analysis showed that there is probably little or no difference in the risk of receipt of intravenous treatment for hypoglycaemia after oral dextrose gel compared to placebo gel (RR 1.01, 95% CI 0.68 to 1.49; two studies, 2548 infants; moderate certainty evidence, downgraded for serious imprecision; Analysis 1.4). We received additional data from the authors of one study (Harding 2020).
1.6 Receipt of oral dextrose gel treatment for hypoglycaemia
Meta‐analysis showed that there is probably little or no difference in receipt of oral dextrose gel for treatment of hypoglycaemia after oral dextrose gel compared to placebo gel (RR 0.90, 95% CI 0.79 to 1.01; two studies, 2548 infants; Analysis 1.5).
1.7 Receipt of any medication for hypoglycaemia
Additional data received from the authors of both studies showed that no infants in either the dextrose gel or placebo groups received any medication for hypoglycaemia, such as glucagon or corticosteroids.
1.8 Number of episodes of hypoglycaemia per infant
Additional data from one study (Hegarty 2016a) showed that there is probably little or no difference in the number of episodes of hypoglycaemia per infant after oral dextrose gel compared to placebo gel (MD ‐0.18, 95% CI ‐0.55 to 0.19; one study, 186 infants; Analysis 1.6).
1.9 Adverse events
Meta‐analysis showed that there is probably little or no difference in the risk of adverse events from use of oral dextrose gel compared to placebo gel (RR 1.22, 95% CI 0.64 to 2.33; two studies, 2510 infants; moderate certainty evidence, downgraded for serious imprecision; Analysis 1.7). We received additional data from the authors of one study (Harding 2020).
1.10 Separation from mother for treatment of hypoglycaemia
Meta‐analysis showed that oral dextrose gel may have little or no effect on the risk of separation of the infant from the mother for treatment of neonatal hypoglycaemia compared to placebo gel (RR 1.12, 95% CI 0.81 to 1.55; two studies, 2548 infants; low certainty evidence, downgraded for imprecision and substantial heterogeneity (I2 = 83%); Analysis 1.8). We were unable to perform subgroup analysis to further explore sources of heterogeneity as there were insufficient studies.
1.11 Neonatal seizures
Only one event of neonatal seizures was reported in the dextrose gel group (1/1347; 0.1%) in (Hegarty 2016a), and one event of neonatal seizures was reported in the placebo group (1/1201; 0.1%) in (Harding 2020), (RR 0.69, 95% CI 0.08 to 5.69; two studies, 2548 infants; Analysis 1.9). Both studies reported that the seizures were unrelated to neonatal hypoglycaemia.
1.12 Abnormal MRI of the brain in the neonatal period
No data were available for this outcome.
1.13 Duration of initial hospital stay
We received additional data from the authors of both studies which showed that oral dextrose gel makes little or no difference to the mean duration of initial hospital stay (days) (MD 0.06, 95% CI ‐0.13 to 0.24; two studies, 2537 infants; Analysis 1.10).
1.14 Breastfeeding (any) after discharge
We received additional data from the authors of both studies which showed that there is little or no difference in breastfeeding (any) after discharge after oral dextrose gel compared to placebo gel (RR 1.01, 95% CI 0.98 to 1.05; two studies, 2323 infants; Analysis 1.11). Oral dextrose gel also made no difference to full or exclusive breastfeeding after discharge (RR 1.00, 95% CI 0.98 to 1.03); two studies, 2528 infants).
1.15 Exclusive breastfeeding after discharge – WHO 2008 definition
No data were available for this outcome.
1.16 Exclusive breastfeeding at six months of age – WHO 2008 definition
No data were available for this outcome.
1.17 Developmental disability at two years of age or greater
One study (Hegarty 2016a) reported that oral dextrose gel may have little or no effect on the risk of any developmental disability at two years’ corrected age (RR, 0.75, 95% CI 0.49 to 1.17; one study, 359 infants; Analysis 1.12)
1.18 Visual impairment and severity at two years of age or greater
No infants were blind at two years’ corrected age (one study, 359 infants).
1.19 Hearing impairment at two years of age or greater
One infant in the placebo group was deaf at two years’ corrected age (1/116; 0.01%, RR 0.16, 95% CI 0.01 to 3.89; one study; 359 infants; Analysis 1.12).
1.20 Cerebral palsy at two years of age or greater
No infant had cerebral palsy at two years’ corrected age (one study, 359 infants).
1.21 Developmental delay/intellectual impairment and severity at two years of age or greater
Additional data from one study (Hegarty 2016a) reported that oral dextrose gel may have little to no effect on the risk of developmental delay/intellectual impairment at two years’ corrected age (RR 0.75, 95% CI 0.49 to 1.17; one study, 359 infants; Analysis 1.12).
1.22 Executive dysfunction and severity at two years of age or greater
Additional data from one study (Hegarty 2016a) reported that oral dextrose gel probably reduces the risk of executive dysfunction at two years’ corrected age compared with placebo gel (RR 0.48, 95% CI 0.23 to 0.99; one study, 357 infants; Analysis 1.12). The risk difference (RD) was ‐0.06 (95% CI ‐0.12 to 0.01), and on average, 17 infants would have to receive prophylactic dextrose gel to prevent one additional case of executive dysfunction. No severity data were available.
1.23 Behavioural problems and severity at two years of age or greater
No data were available for this outcome.
1.24 Abnormal MRI of the brain at two years of age or greater
No data were available for this outcome.
Discussion
Summary of main results
This updated review included two studies that evaluated oral dextrose gel compared with placebo gel in preventing neonatal hypoglycaemia in 2548 at‐risk infants (Harding 2020; Hegarty 2016a).
High certainty evidence from the two studies showed that oral dextrose gel reduces the risk of neonatal hypoglycaemia. On average, 17 at‐risk infants would need to be treated with oral dextrose gel to prevent one additional case of neonatal hypoglycaemia (95% CI 10 to 50). There was moderate certainty evidence from one study in 360 infants that oral dextrose gel compared with placebo gel probably reduces the risk of major neurological disability at two years of age or older (NNTB 20, 95% 11 to ∞) (see summary of findings Table 1).
Oral dextrose gel probably reduces the risk of receipt of treatment for hypoglycaemia during the initial hospital stay but makes little or no difference to intravenous treatment, may have little or no effect on separation of the mother from the infant for treatment of hypoglycaemia, and probably does not result in increased adverse events compared with placebo gel.
Oral dextrose gel probably also has little or no effect on receipt of oral dextrose gel for the treatment of hypoglycaemia, the number of hypoglycaemic episodes, neonatal seizures, duration of initial hospital stay and whether an infant is breastfed after discharge. The findings from a single study (360 infants) suggest that oral dextrose gel compared with placebo gel may have little or no effect on the risks of any developmental disability, developmental delay/intellectual impairment or hearing impairment, but probably reduces executive dysfunction (NNTB 17, 95% CI 8 to 100). No infants were blind or had cerebral palsy at two years of age or greater.
Overall completeness and applicability of evidence
More data have become available on the long‐term neurodevelopmental and disability outcomes since the first version of this review, although this comes from a single small study (Hegarty 2017). No data were available for the following prespecified secondary outcomes: abnormal MRI of the brain in the neonatal period and at two years of age or greater; exclusive breastfeeding after discharge or at six months of age; and behavioural problems and severity at two years of age or greater. Both included studies compared a single preparation of oral dextrose gel with placebo in late preterm and term infants in two high‐income countries (Australia and New Zealand) and therefore, the applicability of these findings to other preparations of gel, extremely and moderately preterm infants and other healthcare settings remains unknown. There were no randomised trials that compared oral dextrose gel with no treatment, standard care or other therapies (such as antenatal expression of colostrum, early initiation of breastfeeding and the supplementation of breastfeeding with formula milk).
Quality of the evidence
We assessed both studies as low risk of bias across all domains. The certainty of evidence assessed by GRADE was moderate for most outcomes, except for the co‐primary outcome of hypoglycaemia, which was graded as high certainty, and separation of mother and infant, which was graded as low certainty (see summary of findings Table 1). The most common reason for downgrading the evidence was imprecision due to wide confidence intervals and low event rates.
For the co‐primary outcome of major neurological disability at two years of age or older, data were available from only one small study with low event rates, so we graded the certainty of this evidence as moderate due to a lack of precision. Additional studies are needed to clarify the certainty of this finding.
We downgraded one outcome (separation from the mother for treatment of hypoglycaemia) to low certainty because of a lack of precision and substantial unexplained heterogeneity. We were unable to further explore this with planned subgroup analyses because there were only two studies. However, it is possible that this heterogeneity may be due to variations between hospitals in policies for admission to the NICU.
We downgraded the outcome of adverse events to moderate certainty because of imprecision due to low event rates. Thus, there is moderate certainty evidence that dextrose gel does not increase the risk of an infant choking or vomiting during administration compared to those who received placebo gel, and that the risk is low (< 1% in both groups), but there is no evidence about whether receipt of oral dextrose gel may increase the risk of these events or other harms compared with no treatment or other therapies such as formula feeding.
Potential biases in the review process
We believe that we have made every effort to minimise bias in the review process. We conducted a systematic search of the literature for randomised controlled trial evidence, not restricted by language or date of publication. When necessary, we contacted authors of primary studies to obtain additional outcome data. We adhered to the Cochrane methods of searching and performing data extraction and analysis.
Agreements and disagreements with other studies or reviews
This update agrees with the first version of this review (Hegarty 2017), which reported that oral dextrose gel compared with placebo was associated with a reduced risk of neonatal hypoglycaemia in at‐risk infants with no evidence of adverse events. The first version included a single study of 415 at‐risk infants. That study was included in this updated version and contributed data to the meta‐analysis.
Moreover, the Cochrane Review, “Oral dextrose gel for the treatment of hypoglycaemia in newborn infants,” (Weston 2016), reported that treatment of hypoglycaemia with oral dextrose gel was associated with a reduction in separation of the mother and infant and increased likelihood of full breastfeeding after discharge. Weston 2016 also found no evidence of adverse events of dextrose gel during the neonatal period or at two years' corrected age.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies
Risk of bias summary: review authors' judgements about each risk of bias item for each included study
Comparison 1: Dextrose gel versus control, Outcome 1: Hypoglycaemia
Comparison 1: Dextrose gel versus control, Outcome 2: Major neurological disability at two years of age or older
Comparison 1: Dextrose gel versus control, Outcome 3: Receipt of treatment for hypoglycaemia during initial hospital stay
Comparison 1: Dextrose gel versus control, Outcome 4: Receipt of intravenous treatment for hypoglycaemia
Comparison 1: Dextrose gel versus control, Outcome 5: Receipt of oral dextrose gel treatment for hypoglycaemia
Comparison 1: Dextrose gel versus control, Outcome 6: Number of episodes of hypoglycaemia (glucose oxidase method) (total number per infant)
Comparison 1: Dextrose gel versus control, Outcome 7: Adverse effects (e.g. choking or vomiting at time of administration)
Comparison 1: Dextrose gel versus control, Outcome 8: Separation from mother for treatment of hypoglycaemia (admission to NICU for hypoglycaemia)
Comparison 1: Dextrose gel versus control, Outcome 9: Neonatal seizures
Comparison 1: Dextrose gel versus control, Outcome 10: Duration of initial hospital stay (days)
Comparison 1: Dextrose gel versus control, Outcome 11: Breastfeeding (any) after discharge
Comparison 1: Dextrose gel versus control, Outcome 12: Neurodevelopmental and disability outcomes at two years of age or greater
| Dextrose gel compared with placebo for prevention of hypoglycaemia in newborn infants | ||||||
| Patient or population: newborn infants at risk of neonatal hypoglycaemia | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | No. of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with placebo | Risk with dextrose gel | |||||
| Hypoglycaemia (investigator‐defined) | 433 per 1000 | 377 per 1000 | RR 0.87 (0.79 to 0.95) | 2548 (2 RCTs) | ⊕⊕⊕⊕ | |
| Major neurological disability at 2 years of age or older (Defined as any of the following: legal blindness, sensorineural deafness requiring hearing aids, moderate or severe cerebral palsy or developmental delay/intellectual impairment (developmental quotient or intelligence quotient lower than 2 SD below the mean)) | 60 per 1000 | 47 per 1000 | RR 0.21 | 360 (1 RCT) | ⊕⊕⊕⊝ | Low event rates: 3/277 in dextrose gel group and 7/138 in placebo group. |
| Receipt of treatment for hypoglycaemia during initial hospital stay (investigator‐defined, any treatment ‐ oral dextrose gel, intravenous dextrose, or other drug therapy) during initial hospital stay (yes/no) | 316 per 1000 | 281 per 1000 | RR 0.89 | 2548 | ⊕⊕⊕⊝ | |
| Receipt of intravenous treatment for hypoglycaemia (yes/no) | 37 per 1000 | 37 per 1000 | RR 1.01 | 2548 | ⊕⊕⊕⊝ | |
| Adverse events# (e.g. choking or vomiting at time of administration) | 10 per 1000 | 12 per 1000 (6 to 24) | RR 1.22 (0.64 to 2.33) | 2510 (2 RCTs) | ⊕⊕⊕⊝ | Low event rates: 27/1322 in dextrose gel group and 12/1188 in placebo group. |
| Separation from mother for treatment of hypoglycaemia (infant nursed in an environment that is not in the same room as the mother, e.g. for NICU admission or the like) (yes/no)) | 50 per 1000 | 56 per 1000 | RR 1.12 (0.81 to 1.55) | 2548 (2 RCTs) | ⊕⊕⊝⊝ LOWc,d | |
| Breastfeeding (exclusive after discharge) ‐ not reported (WHO 2008 definition (yes/no)) | ‐ | ‐ | ‐ | ‐ | ‐ | No data were reported for this outcome |
| *The risk in the intervention group (and its 95% confidence interval) is based on assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI); #Placebo gel was the comparator. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level for serious imprecision (due to low event rates) | ||||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1.1 Hypoglycaemia Show forest plot | 2 | 2548 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.87 [0.79, 0.95] |
| 1.2 Major neurological disability at two years of age or older Show forest plot | 1 | 360 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.21 [0.05, 0.78] |
| 1.3 Receipt of treatment for hypoglycaemia during initial hospital stay Show forest plot | 2 | 2548 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.89 [0.79, 1.00] |
| 1.4 Receipt of intravenous treatment for hypoglycaemia Show forest plot | 2 | 2548 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.01 [0.68, 1.49] |
| 1.5 Receipt of oral dextrose gel treatment for hypoglycaemia Show forest plot | 2 | 2548 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.90 [0.79, 1.01] |
| 1.6 Number of episodes of hypoglycaemia (glucose oxidase method) (total number per infant) Show forest plot | 1 | 186 | Mean Difference (IV, Fixed, 95% CI) | ‐0.18 [‐0.55, 0.19] |
| 1.7 Adverse effects (e.g. choking or vomiting at time of administration) Show forest plot | 2 | 2510 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.22 [0.64, 2.33] |
| 1.8 Separation from mother for treatment of hypoglycaemia (admission to NICU for hypoglycaemia) Show forest plot | 2 | 2548 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.12 [0.81, 1.55] |
| 1.9 Neonatal seizures Show forest plot | 2 | 2548 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.69 [0.08, 5.69] |
| 1.10 Duration of initial hospital stay (days) Show forest plot | 2 | 2537 | Mean Difference (IV, Fixed, 95% CI) | 0.06 [‐0.13, 0.24] |
| 1.11 Breastfeeding (any) after discharge Show forest plot | 2 | 2323 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.01 [0.98, 1.05] |
| 1.12 Neurodevelopmental and disability outcomes at two years of age or greater Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 1.12.1 Developmental disability at two years of age or greater | 1 | 359 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.75 [0.49, 1.17] |
| 1.12.2 Hearing impairment at two years of age or greater | 1 | 359 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.16 [0.01, 3.89] |
| 1.12.3 Developmental delay/intellectual impairment at two years of age or greater | 1 | 359 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.75 [0.49, 1.17] |
| 1.12.4 Executive dysfunction at two years of age or greater | 1 | 357 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.48 [0.23, 0.99] |
