Fluid and electrolyte balance are important considerations during the perioperative care of small children. Fasting guidelines have been developed to minimise the likelihood of dehydration and electrolyte imbalance in the preoperative period. The consequences of undetected perioperative hypoglycaemia are so catastrophic that predisposing factors should be actively sought during the preoperative assessment. A review of glucose management highlighted several areas for caution.1 These include children less than 48 h old,2 children below the third centile for weight3 and children receiving concurrent medical therapy. Propranolol is used to treat a number of conditions in childhood, including paroxysmal supraventricular tachycardia, migraine and hypercyanotic spells in children with Tetralogy of Fallot. Perioperative hypoglycaemia has been previously reported to occur following prolonged fasting in children treated with propranolol.4,5 We report a case of intraoperative hypoglycaemia in a child receiving propranolol for symptomatic management of Tetralogy of Fallot after a short preoperative fast.
A 19-month-old, 5.2 kg child with unrepaired Tetralogy of Fallot was admitted electively for microlaryngoscopy, rigid bronchoscopy, oesophogastroduodenoscopy, Nissen's fundoplication and open gastrostomy. The child was born at 35 weeks postconceptual age and underwent repair of tracheoesophageal fistula and oesophageal atresia on day 2 of life. He was discharged home, but was slow to gain weight owing to difficulty with feeding and regurgitation. Six weeks prior to this elective admission, he began to develop cyanotic episodes whilst feeding. After treatment for a viral chest infection, it was decided that he should be fed via the nasogastric route and propranolol therapy was started. It was suspecting that these episodes may be hypercyanotic spells related to his cardiac condition.
He was transferred to our tertiary hospital from a district general hospital on the morning of surgery. His fluid balance chart indicated that he had been receiving Enfamil nasogastric formula feed at 26 ml h−1 until midnight prior to the morning of surgery. He received Dioralyte isotonic liquid (3.65 g glucose per 200 ml) at 26 ml h−1 via the nasogastric route until 0600 h. No significant nasogastric aspirate or vomits were recorded on the chart over the preceding 24 h. A routine set of observations were performed by the nursing staff, heart rate 120, SaO2 85% and apyrexial. A blood glucose measurement is not routinely performed preoperatively or on admission to the ward.
At 0930 h, the child underwent a gaseous induction with oxygen and sevoflurane in theatre beneath an overhead heater, lying on a warming blanket. Intravenous access was established; the upper airway was anaesthetised with 1.5 ml 1% lidocaine under direct vision and surgery commenced with microlaryngoscopy and rigid bronchoscopy. The trachea was then intubated before oesophogastroduodenoscopy was performed. The child received two 10 ml kg−1 boluses of Volplex during this period. At 1045 h, the child was repositioned for surgery, and maintenance fluid administered. Owing to the complexity of the child's cardiac condition, a 24 gauge radial arterial line was inserted prior to laparotomy. An initial arterial blood gas sample was sent for analysis and intravenous maintenance fluids were started.
The initial arterial blood gas analysis was as follows: pH 7.32, PCO2 5.07 kPa, PO2 8.35 kPa, base −6.4, glucose 0.8 mmol l−1. We immediately sought to confirm the result with bedside blood glucose monitoring, which recorded 1.0 mmol l−1. We administered 5 ml kg−1 10% dextrose via the intravenous route. When we rechecked the blood glucose after 5 min, it had risen to 12.7 mmol l−1. A background infusion of NaCl 0.45% with Dextrose 5% was started at 15 ml h−1. Blood glucose was monitored hourly, and there were no further episodes of hypoglycaemia. The child was extubated in theatre fully awake and transferred to paediatric intensive care unit (PICU), where further close attention was paid to his glucose levels and his neurological state. He went on to make an uneventful recovery and was discharged from PICU after 48 h.
We believe that this case highlights several important points. Previous case reports that document perioperative hypoglycaemia in children receiving propranolol describe the event occurring after a prolonged fast, such as 16 h.4 The period of the documented fast in this case was very brief, receiving glucose-containing solution up until 3 h prior to the scheduled surgery start time. However, whilst there was no documented history of difficult with nasogastric feeding over the last 24 h, it was later elicited that he had experienced difficulty establishing nasogastric feeding over the week prior to surgery, with several episodes of regurgitation. It was clear that the child was significantly underweight, and it is likely that this and the difficulty establishing nasogastric feeding had resulted in reduced glycogen reserves. Glycogenolysis and gluconeogenesis are mediated by β-2 receptors. Because propranolol is a non-selective β adrenergic receptor antagonist, it is accepted that a side effect of treatment is impaired glucose metabolism.6 Under normal circumstances, in healthy children, there is little risk of compromise. However, in chronically undernourished or fasted children there is a significant risk of hypoglycaemia. Signs of hypoglycaemia include tachycardia, hypertension, cool skin, sweating and irritability. These warning indicators can be masked by β-adrenergic blockade and anaesthesia. If routine blood gas analysis had not been performed, hypoglycaemia may well have progressed undetected, resulting in critical substrate deprivation in the brain and permanent neurological damage.
On reflection, we would make the following amendments to our practice. Any child being treated with propranolol, listed for elective surgery, will undergo preoperative blood glucose monitoring, regardless of the brevity of the fast. Children receiving propranolol and at particular risk of hypoglycaemia, such as those which are underweight or experiencing feeding difficulties, may require starting an intravenous fluid contain glucose during the fasting period or immediately after induction of anaesthesia.
Even very short fasting times may provoke asymptomatic hypoglycaemia in severely underweight infants treated with propranolol. Under these circumstances, provision of preoperative intravenous glucose is recommended as part of a balanced isotonic fluid replacement regime.
There are no conflicts of interest.
1 Leelanukrom R, Cunliffe M. Intraoperative fluid and glucose management in children. Paediatrc Anaesth
2 Larsson LE, Nilsson K, Niklasson A, et al
. Influence of fluid regimens on perioperative blood glucose concentrations in the neonate. Br J Anaesth 1990; 64:419–424.
3 Payne K, Ireland P. Plasma glucose level in the perioperative period in children. Anaesthesia 1986; 39:868–872.
4 Baines DB, Murrell D. Preoperative hypoglycaemia, propranolol and the Jervell and Lange-Nielsen syndrome. Paediatr Anaesth 1999; 9:156–158.
5 Zeligs MA, Lockhart CH. Perioperative hypoglycaemia in a child treated with propranolol. Anaesth Analg 1983; 62:1035–1037.
6 Bush GH, Steward DJ. Severe hypoglycaemia associated with perioperative fasting and intraoperative propranolol. Paediatr Anaesth 1996; 6:415–417.