Early blood glucose profile and neurodevelopmental outcome at two years in neonatal hypoxic-ischaemic encephalopathy

Hypoxic-ischaemic encephalopathy (HIE) remains an important cause of neonatal death and long-term neurodisability [1]. Goals of management have been to maintain normoxaemia, normocapnia, normoglycaemia and normal blood pressure to avoid or ameliorate secondary cerebral injuries [2].

Neonatal hypoglycaemia, independent of HIE, has been associated with adverse outcome in both term and preterm infants [3‐5]. However, no conclusive evidence on the severity and duration of hypoglycaemia causing brain damage has been reported [6, 7]. Basu et al showed that the degree of hypoglycaemia was correlated to the severity of HIE in term asphyxiated newborns [8]. In term infants with severe fetal acidemia, an association between early adverse outcome and hypoglycaemia on the first blood sample was reported by Salhab et al [9]. These studies did not address long-term neurodevelopmental outcome. The mechanism of hypoglycaemic brain injury has been examined in animal models. Hypoglycaemia decreases the cerebrovascular response to hypoxia and increases cerebral superoxide production and aspartate levels into the brain extracellular space resulting in neuronal necrosis [10‐12].

Hyperglycaemia is associated with adverse outcome in premature infants, in critically ill children and in adult patients with stroke [13‐18]. In extremely low birth weight infants and in patients in paediatric intensive care unit, controlling elevated blood glucose levels by controlled insulin infusion was associated with good short-term outcome [19, 20]. However, no data on long-term neurodevelopmental outcome relating to hyperglycaemia in neonatal HIE has been reported [19, 20]. The mechanism of hyperglycaemic brain damage is thought to be related to neuronal cell apoptosis following reperfusion with high level of substrate (glucose) in an ATP- deleted cell. In an animal model, hyperglycaemia following hypoxic-ischaemic insult decreases fetal brain ATP and oxygen consumption and increases thickness of vascular endothelium with foci of infarction [21‐23].

In the absence of long-term neurodevelopmental outcomes on the effects of early hypoglycaemia and hyperglycaemia in HIE, the aim of this study was to describe the early blood glucose profile and to determine whether hypoglycaemia and hyperglycaemia in the 72 hours of birth were associated with adverse neurodevelopmental outcome at 24 months of age in infants with HIE.

This study was a retrospective analysis of a prospective cohort of babies with HIE, none of whom received therapeutic hypothermia. It was conducted in a large maternity hospital with an annual delivery rate of approximately 5000 babies. Ethical approval was obtained from Cork University Hospital Research Ethics Committee. Informed consent was obtained from all parents of the infants who were enrolled in the study. We examined the medical notes of a cohort of 55 term infants with HIE, recruited prospectively at birth between May 2003 and December 2005. Term infants were recruited to the cohort if they fulfilled 2 or more of the following criteria:

Demographics and data related to neonatal course, ventilation variables and the arterial, capillary and venous blood glucose values of the first 3 days of life were retrieved from the medical notes. All infants with HIE had blood glucose levels checked within the first 30 minutes after delivery. Blood glucose values were collected from either arterial, venous, or capillary samples. An on-site analyser (Radiometer) was used to measure blood glucose levels. Hypoglycaemia and hyperglycaemia were defined as blood glucose levels < 46.8 mg/dL (2.6 mmol/L) and > 150 mg/dL (8.3 mmol) respectively [5, 13].

Hypoglycaemia was treated with an initial parenteral bolus of 2 mL/kg of a 10% dextrose solution over one minute. Blood glucose levels were rechecked 30 minutes to 1 hour later according to the severity of hypoglycaemia. Dextrose infusion rate or concentration was adjusted to maintain blood glucose level > 50 mg/dL (2.8 mmol/L). In infants with hyperglycaemia, the rate or concentration of glucose infusion was adjusted to maintain blood glucose within normal levels.

Demographic details and clinical data were recorded in each case. Clinical grade of encephalopathy was assigned using a Sarnat score at 24 hours of age. Neurodevelopmental outcome was assessed at 24 months using the Revised Griffith's scales of Mental Development [24]. Adverse outcome was defined as death, a Griffith's Quotient (GQ) less than 87, or significant motor disability. Statistical analysis was performed using SPSS version 14.0 for Windows. Summary measures were calculated and are reported as mean and standard deviation (SD) or median and (range). Spearman correlation was used to explore the differences in categorical variables. Univariate and multivariate logistic regression models were used to estimate odds ratios and 95% confidence intervals. A p value <0.05 was considered statistically significant.

This study is the first to describe the blood glucose profile during the first 72 hours of life in infants with HIE. Despite a protocolised-driven approach to detecting and treating abnormal blood sugars following perinatal asphyxia in our unit, sustained normoglycaemia in the first 72 hours of birth was observed in only half of our infants with HIE.

What does the glucose profile in HIE tell us? Abnormal blood glucose values were observed in 1 out of 6 samples. Less than a third of hypoglycaemia and hyperglycaemia episodes were observed on the first blood sample taken within a half-hour of birth. However, half of hypoglycaemic samples and 90% of hyperglycaemic samples occurred within 6 hours of birth. These findings suggest that many asphyxiated babies, in addition to facing an hypoxic-ischaemic insult, are concurrently experiencing significant variations in blood sugars in the early newborn period. The hypoglycaemic episodes may be due to perinatal depletion of glycogen stores (many asphyxiated babies are post mature), whilst the hyperglycaemic episodes may reflect the release of stress-related hormones. In addition, since variations in blood glucose levels persist throughout the first 72 hours, this study indicates that regular measurements of the blood glucose should continue throughout this period.

In our study, early hypoglycaemia was associated with an increased risk of adverse outcome at 24 months of age in infants with HIE. However, when corrected for grade of HIE this was no longer significant. Our results are consistent with the findings of Salhab et al. In that study, hypoglycaemia on the initial blood sample after birth was associated with abnormal short-term outcomes (death as a consequence of severe encephalopathy and evidence of moderate to severe encephalopathy with or without seizures) in term infants with severe fetal acidaemia [9]. However no long term outcome has been reported from that group. In preterm infants, it has been shown that hypoglycaemia was associated with mental and motor development scores at age of 18 months corrected. However at age of 8 years, only arithmetic and motor scores were affected [5, 25]. It has been shown that intracranial MRI abnormalities in full term infants with neonatal hypoglycaemia are resolved two months later [26]. Hypoglycaemia has been associated with abnormal neurological outcomes but not with abnormal psychomotor development [3, 27, 28].

Available data on the relationship between hyperglycaemia and adverse outcome is inconclusive [13, 15]. In infant rats, it has been shown that hyperglycaemia during an hypoxic-ischaemic insult can have a beneficial effect against brain injury [29]. LeBlanc et al showed that, in newborn piglets, hyperglycaemia after hypoxic-ischaemic injury does not worsen the brain injury [30]. In newborn piglets, Park et al showed that brain energy metabolism was affected by hyperglycaemia during the immediate reperfusion period after hypoxic-ischaemic brain insult [21]. In fetal sheep, Blomstrand S et al showed that hyperglycemia during asphyxia reduces cerebral oxygen consumption and increases acidosis [22]. Hyperglycaemia following hypoxia-ischaemia insult was shown to be harmful, in adult rats [23].

However the glucose values studied were beyond those we see in clinical situation in humans. Despite the data on ELBW infants [13‐15] and experience with older children [16, 17], the ideal management of hyperglycaemia in neonatal encephalopathy remains unclear [19, 20].

We conclude that it is difficult to avoid hypoglycaemia and/or hyperglycaemia in infants with HIE, since such variations in blood glucose levels often occur soon after birth, and may be related to the asphyxial process. In term infants with HIE, glucose profiles vary widely during the first 72 hours of life. Early hypoglycaemia occurs more frequently in infants with severe HIE, and is therefore associated with abnormal outcome. The further exploration of the relationship between glycaemic control and neurological outcome will require larger numbers of patients and continuous blood glucose monitoring.