Article Text

Download PDFPDF

Lactate, rather than ketones, may provide alternative cerebral fuel in hypoglycaemic newborns
  1. Deborah L Harris1,2,
  2. Philip J Weston1,
  3. Jane E Harding2
  1. 1Newborn Intensive Care Unit, Waikato District Health Board, Hamilton, New Zealand
  2. 2Liggins Institute, University of Auckland, Auckland, New Zealand
  1. Correspondence to Professor Jane E Harding, Liggins Institute, University of Auckland, Private Bag 92019, Auckland 1142, New Zealand; j.harding{at}


Objective Alternative cerebral fuels are reputed to provide neuroprotection during hypoglycaemia, particularly in breastfed babies. We measured concentrations of alternative cerebral fuels in hypoglycaemic babies in the first 48 h.

Patient and methods Babies were ≥35 weeks, ≤48 h old and at risk of hypoglycaemia (infant of diabetic, preterm, small or large). Plasma glucose, β-hydroxybutyrate, lactate and insulin concentrations were measured in babies who had been hypoglycaemic (<2.6 mM) for >1 h.

Results Samples were taken from 35 hypoglycaemic babies at 3.7; 1.8–39.6 (median; range) hours after birth. Concentrations of glucose and β-hydroxybutyrate were low (2.03; 0.19–3.39 mM and 0.06; 0.00–1.20 mM), but lactate concentrations varied widely (3.06; 0.02–7.96 mM). Infants of diabetics had lower β-hydroxybutyrate and higher insulin concentrations, but mode of feeding did not influence plasma concentrations of alternative cerebral fuels.

Conclusions Hypoglycaemic babies within the first 48 h after birth are unlikely to receive neuroprotection from ketones. However, lactate may provide an alternative cerebral fuel for many. Lactate, rather than ketones, may provide alternative cerebral fuel in hypoglycaemic newborns.

Trial registration number ACTRN12608000623392.

  • Neonatology
  • Endocrinology

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Neonatal hypoglycaemia is common and linked with poor neurological outcome. Some babies compensate for low cerebral fuel availability during hypoglycaemia by increasing cerebral blood flow and using non-glucose cerebral fuels such as ketones and lactate, which can be derived from glycerol, fatty acids and some amino acids.1 ,2

It has been speculated that breastfed babies are protected from brain injury during hypoglycaemia because synthesis of alternative cerebral fuels, particularly ketones, is specifically facilitated by breast feeding.3 However, these speculations are based on studies of small numbers of babies up to a week old, few of whom were hypoglycaemic, and few of whom were sampled in the first 48 h when the risk of hypoglycaemia is highest.1

Lactate is continually produced by many tissues, including within the brain parenchyma by astrocytes, and can be a critical fuel for brain metabolism when oxygenation is adequate.4 ,5 It protects cognitive function and reduces counter-regulatory responses during insulin-induced hypoglycaemia in adults.6 However, there are few reports regarding the plasma concentrations of lactate in very young hypoglycaemic babies.7 ,8

Therefore, we sought to measure the plasma concentrations of alternative cerebral fuels in babies who had been hypoglycaemic for at least 1 h, during the first 48 h after birth.

Patient and methods

Babies were enrolled in a randomised controlled trial (The Sugar Babies Study).9 In brief, we studied babies receiving routine clinical care who were ≥35 weeks, ≤48 h old and at risk of hypoglycaemia because they were small (birth weight <10th centile or <2.5 kg), large (birth weight >90th centile or >4.5 kg), infants of diabetic mothers or late preterm (35 or 36 weeks).

Mothers were encouraged to breast feed as soon as possible after birth. Babies were fed on demand but not less than three hourly. Mothers who wished to breast feed were encouraged to express breast milk, and babies who did not suckle were given expressed breast milk or formula via a syringe or cup. Formula fed babies were offered up to 60 mL/kg in the first 24 h, and 90 mL/kg in the following 24 h, with feeds offered within the first hour after birth and then 2–4 hourly.

Hypoglycaemia was defined as a blood glucose concentration <2.6 mM. Blood glucose concentrations were measured on heel-prick samples using the glucose oxidase method (Radiometer, ABL800Flex, Copenhagen, Denmark). Samples were taken 1 h after birth and then before feeds 3–4 hourly.

Babies who became hypoglycaemic received 200 mg/kg of dextrose gel, or an identical appearing placebo gel, massaged into the buccal mucosa, and were encouraged to feed. Blood glucose concentrations were measured again 30 min after treatment and the same gel was repeated once if hypoglycaemia persisted. Babies who remained hypoglycaemic 30 min after the second dose of gel were admitted to the newborn intensive care unit, and blood samples were taken for measurement of glucose, insulin, lactate, β-hydroxybutyrate and glycerol before any further treatment. Samples were immediately placed on ice, centrifuged and the plasma frozen at −80°C until analysis using a Hitachi 902 automatic analyser (Hitachi Australia, North Ryde, New South Wales, Australia) and commercial reagents kits (Randox Laboratories, Antrim, UK).

Statistical analysis

Data were analysed using JMP V.10.0, 2013 (SAS Institute Inc, Cary, North Carolina, USA) and are presented as number (per cent) or median (range). Groups were compared using χ2 or Kruskal–Wallis with Dunn All Pairs post hoc correction for multiple comparisons.


Of the 514 babies enrolled in the trial, 242 (47%) became hypoglycaemic and 45 (8%) remained hypoglycaemic after two doses of gel. 39 babies (87% of those who remained hypoglycaemic) had blood samples taken for measurement of alternative cerebral fuels. However, four were excluded as the samples were taken after treatment, leaving samples for analysis from 35 (78%) babies taken at a median age of 3.7 h (range 1.8–39.7 h). 12 of these babies had been randomised to dextrose and 23 to placebo gel. Eight babies showed clinical signs potentially related to hypoglycaemia (two were jittery and six were too sleepy to feed). The most common risk factor for hypoglycaemia was preterm birth (table 1).

Table 1

Characteristics of babies studied (n=35)

Plasma concentrations of glucose and β-hydroxybutyrate were low (table 2). Infants of diabetic mothers had lower plasma concentrations of β-hydroxybutyrate (p=0.03), but there were no other differences between risk groups, and no differences between breastfed babies and those who received combination or no feeds (table 2). There were no differences between babies randomised to dextrose or placebo gel (data not shown). Plasma concentrations of glycerol were also low (0.28; 0.00–1.03 mM) and not different between risk groups, suggesting minimal lipolysis in these babies.

Table 2

Plasma concentrations of glucose, lactate, β-hydroxybutyrate and insulin in hypoglycaemic babies with different risk factors and feeding

Plasma lactate concentrations varied widely (0.02–7.96 mM). Six babies (17%) had lactate concentrations ≤1.8 mM (0.50; 0.02–1.38 mM), which was the mean concentration shown to protect cognitive function in hypoglycaemic adults6 (figure 1). These babies also had low plasma concentrations of β-hydroxybutyrate (0.001; 0.001–0.102 mM), detectable plasma insulin concentrations (3.0; 0.01–7.2 μM), and were from all risk groups (two late preterm, three infants of diabetic mothers, one small) and both feeding groups (three breast fed, three combination breast and formula).

Figure 1

Plasma concentrations of glucose, lactate and ß-hydroxybutyrate in hypoglycaemic babies within the first 48 hours. Each bar represents the sum of measured cerebral fuels (glucose, lactate, β-hydroxybutyrate) in one baby (n=35). Numbers below the bars represent the postnatal age in hours that the blood sample was taken. *Babies with lactate concentrations ≤1.8 mM.

Detectable concentrations of insulin were found in 25 babies from all risk groups (table 2). However, more infants of diabetics (7/9, 78%) had insulin concentrations above the median of 1.2 μU/mL than did preterm (4/15, 26%) or small babies (4/10, 40%); p=0.04.


We sought to measure the plasma concentrations of alternative cerebral fuels in at-risk babies who had been hypoglycaemic for at least 1 h during the first 48 h after birth, and relate these to mode of feeding. We found that very young hypoglycaemic babies have low plasma concentrations of β-hydroxybutyrate. However, many have plasma lactate concentrations high enough to potentially provide an alternative cerebral fuel source. Alternative fuel concentrations were not related to feed type.

Between 2 and 6 days after birth, breastfed babies have higher plasma concentrations of ketones than formula fed babies,1 and it is reported that suckling breast milk may contribute to the production of ketones.3 These findings, coupled with the strong encouragement for exclusive breast feeding, have led to speculation that breastfed babies are protected from the risk of neurological injury from hypoglycaemia by the availability of ketones. However, the cerebral utilisation of both ketones and lactate is related to plasma concentrations,10 ,11 and we found that hypoglycaemic babies have low plasma concentrations of β-hydroxybutyrate in the first 48 h, regardless of type of feeding. Thus, it is unlikely that ketones provide substantial neuroprotection during early hypoglycaemia in at-risk breastfed babies.

Furthermore, insulin increases glucose utilisation and inhibits ketogenesis. As reported elsewhere,12 we found that infants of diabetics have higher plasma insulin concentrations, but that some small and late preterm babies also have detectable plasma insulin concentrations during early hypoglycaemia, and thus are likely to have limited access to alternative cerebral fuels via ketogenesis over this period. Low plasma glycerol concentrations in all risk groups also suggest low rates of lipolysis at this time.

Lactate is an important cerebral fuel, and plasma concentrations peak within the first few hours after birth before decreasing with increasing postnatal age.1 Hypoglycaemic adults given an intravenous lactate infusion to a reach a mean plasma lactate concentration of 1.8±0.1 mM showed improved cognitive function and reduced counter-regulatory responses, suggesting that this lactate was providing an alternative cerebral fuel.6 In our cohort, 83% of babies had plasma concentrations ≥1.8 mM, suggesting that lactate may provide a useful source of cerebral fuel in most hypoglycaemic babies.

However, six babies (17%) with plasma concentrations of lactate ≤1.8 mM were also hypoketonaemic and had measurable insulin concentrations as well as hypoglycaemia, suggesting that this group may be at particular risk for neurological injury. Since many analysers now measure both glucose and lactate concentrations in the same small blood sample, it should be possible in future studies to investigate the relationship between plasma concentrations of both glucose and lactate and neurological outcome.

We conclude that in at-risk hypoglycaemic babies within the first 48 h after birth, ketones are unlikely to provide protection from the risk of neurological injury posed by hypoglycaemia, even in breastfed babies. However, lactate may provide an alternative cerebral fuel in many hypoglycaemic babies, and those with low plasma lactate concentrations may be at particular risk.


We would like to acknowledge research nurses Catherine McBride and Paula Middlemiss for their assistance with the Sugar Babies Study and Eric Thorstensen for his help with the analysis of the samples.



  • Contributors DLH contributed to the literature search, study design, data collection, analysis and interpretation. DLH wrote the first draft of the manuscript and contributed to the subsequent revisions. PJW contributed to the study design, data collection and the final manuscript. JEH contributed to the study design, data analysis, interpretation and writing the manuscript, in addition to having overall responsibility for the Sugar Babies Study.

  • Funding The Sugar Babies Study was funded from the Waikato Medical Research Foundation, the Auckland Medical Research Foundation, the Maurice and Phyllis Paykel Trust, the Health Research Council of New Zealand, and the Rebecca Roberts Scholarship.

  • Competing interests None.

  • Ethics approval Northern Y Ethics Committee of New Zealand.

  • Patient consent Obtained from the parents.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Linked Articles