Description of the condition
Neonatal hypoglycaemia is a common condition affecting 5 to 15% of infants in the immediate postnatal period (Cornblath 2000; McGowan 2006; Hay 2009). Neonatal hypoglycaemia is important because it is associated with brain injury in newborn infants (Koh 1988; Lucas 1988; Kerstjens 2012). The incidence is likely to be increasing as factors that predispose infants to hypoglycaemia are increasing, including prematurity (Blencowe 2012), maternal diabetes (Wild 2004), and obesity (Doherty 2006). There are known risk factors for neonatal hypoglycaemia, and specific groups of infants are routinely targeted for screening (infants of diabetic mothers, large or small for postmenstrual age, low birth weight, preterm, and those with poor feeding). Less common causes include hyperinsulinism and disorders of beta-oxidation of fatty acids. Neonatal hypoglycaemia is reported to be more common in maternity hospitals in resource-poor settings (Anderson 1993). Screening is usually performed using capillary heel-lance blood samples, as the associated clinical signs are not diagnostically helpful. The accuracy of screening blood tests depends on the method of measurement, with point-of-care testing systems having a greater error range than laboratory systems using glucose oxidase methods (Beardsell 2010).
The definition of hypoglycaemia remains controversial (Hay 2009), with different publications using definition thresholds ranging from 1.7 to 2.6 mM (Pildes 1967; Lubchenco 1971; Sexson 1984; Holtrop 1993; Anderson 1993; Hume 1999; Agrawal 2000; Maayan-Metzger 2009). Several clinical thresholds for treatment have been suggested (Cornblath 2000), but currently a blood glucose concentration of < 2.6 mM is widely accepted as the definition of hypoglycaemia (Harris 2009), based on observations that concentrations below this may be associated with altered brain function and delayed development (Koh 1988; Lucas 1988).
Once diagnosed, infants are frequently managed with increased feeding, supplemental infant formula, or intravenous (IV) dextrose. Supplemental infant formula may disrupt the establishment of breast feeding (Blomquist 1994). IV treatment is expensive, requires separation of the mother and baby, and is not always available in resource-poor settings (Graz 2008), or in settings of lower levels of perinatal care.
Breast feeding is recommended by the World Health Organization (WHO) for all babies up to six months of age (WHO 2008), and the health benefits of breast feeding for both the mother and baby are well recognised. The WHO definitions of breast feeding are available. Human studies have shown breast milk volume in the first 24 hours post-partum is low and progressively increases by day 3 (Kulski 1981; Saint 1984; Le Huerou-Luron 2010). The concentration of lactose within breast milk is also low in the first 24 hours (Kulski 1981; Saint 1984) and steadily increases over the first three days. Therefore, at-risk babies may be at increased risk of severe and prolonged episodes of hypoglycaemia if they are solely breast fed.
One prospective study sought to reduce the incidence of neonatal hypoglycaemia by establishing breast feeding within 30 minutes of delivery (Chertok 2009). Blood glucose concentrations were measured at three hours of age in 84 term infants born to women with gestational diabetes, and showed less hypoglycaemia in babies who were breast fed early. Therefore, it is possible early feeding may reduce the incidence of neonatal hypoglycaemia.
Formula milk is often given to hypoglycaemic infants, although there have been no reported studies of this treatment. The carbohydrate content of breast milk on the first day, in terms of both volume and concentration, is low (Saint 1984). Therefore, formula milk may be more effective as a treatment for neonatal hypoglycaemia than breast milk.
Two studies from India aimed to prevent hypoglycaemia in both small and large for gestational age infants by feeding. Babies with a blood glucose concentration of ≤ 30 mg/dl (1.6 mmol/L) measured by Dextrostix® (Ames Dextrometer) at less than 30 minutes of age were randomised to be fed either formula alone or formula plus added powdered sugar (1.5 g per 30 ml). In both studies hypoglycaemia was reduced four-fold in the group who received additional sugar (Singhal 1991; Singhal 1992).
If feeding does not improve the blood glucose concentration the next step is commonly admission to the neonatal intensive care unit (NICU) for IV dextrose. A bolus of 200 mg/kg.min of 10% dextrose followed by an IV infusion of 8 mg/kg.min increases the blood glucose concentration within one minute, without causing hyperglycaemia (Lilien 1977; Lilien 1980).
IV dextrose and glucagon (200 mcg/kg) or intragastric medium chain triglycerides (5 ml/kg) have also been assessed in a randomised controlled trial (Hawdon 1993). Both treatments substantially increased the blood glucose concentration in babies already receiving 5 mg/kg.min IV dextrose for hypoglycaemia.
Brain injury as a result of hypoglycaemia may be evident on magnetic resonance imaging (MRI). Whereas early publications focused on the occipital cortex as the prime site of injury (Alkalay 2005), more recent reports have extended the possible sites of injury (Burns 2008). Injury is particularly evident with co-morbidities of hypoxic ischaemic encephalopathy and sepsis (Caksen 2011).
Description of the intervention
Dextrose gel contains dextrose, a simple carbohydrate, in concentrated aqueous solution, which can be administered by direct application to the mucosal surfaces of the mouth, including buccal and lingual surfaces. Specific absorption from these sites may allow rapid access to the circulation. Some proportion of the dose may also be swallowed and absorbed from the gastrointestinal tract.
In infants with hypoglycaemia, the simplicity of treatment with dextrose gel and the potential avoidance of more complex treatments, such as IV dextrose or complementary milks, would be an attractive option, if effective. Dextrose gel is typically available in a 40g/100ml form (40%) and is administered at doses of 200-400 mg/kg. There are two observational studies that suggest it may be helpful in treating hypoglycaemia and do not report any adverse effects (Ang 1990; Bourchier 1992).
How the intervention might work
Dextrose may be absorbed directly from the oral mucosa, bypassing the portal circulation and gaining more rapid access to the circulation. Dextrose is also rapidly absorbed by the gastrointestinal mucosa, since it does not require any digestion, but may then be taken up by the liver via the portal circulation and hence have a more delayed effect on blood glucose concentrations.
Why it is important to do this review
Treatment of hypoglycaemia in the neonate usually involves additional feeding, often with formula milk, with potential adverse impact on the quality and duration of breast feeding. If feeding is not effective, IV dextrose is usually administered, commonly requiring admission to NICU and resulting in separation of mother and baby, impaired initiation of breastfeeding and increased healthcare costs.
Dextrose gel is simple to administer and inexpensive. If it is effective in treating neonatal hypoglycaemia without adverse effects then these problems may be avoided. Further, dextrose gel could be used in resource-poor settings where higher levels of neonatal care are unavailable, and potentially prevent brain damage caused by untreated neonatal hypoglycaemia.