There is no consensus that hyperglycaemia is an independent predictor of morbidity-mortality in children. This retrospective observational study aimed to assess the association between abnormal perioperative glucose levels and adverse outcomes in infants receiving open-heart surgery with cardiopulmonary bypass. The records of a total of 233 infants who underwent cardiopulmonary bypass for a variety of congenital cardiac procedures between January and December 2010 were reviewed. The blood glucose levels, demographic and perioperative information were recorded. Patients who experienced complications (n=91) were compared with those who did not (n=142). We found both intraoperative and postoperative glucose levels increased compared to the preoperative values (P<0.001). Thirty patients (12.8%) experienced hyperglycaemia and 15 patients (6.4%) experienced at least one episode of hypoglycaemia during surgery. Within the first two days after surgery, 12 (5.2%) patients experienced hyperglycaemia and 32 (13.7%) became hypoglycaemic in the paediatric intensive care unit. However, the abnormal perioperative glucose levels were not associated with increased adverse outcomes. After adjusting for other potential variables, lower weight at surgery, longer surgery time and hospital length-of-stay are the independent predictors of morbidity-mortality. Our findings suggest that perioperative hyperglycaemia and mild transient hypoglycaemia do not appear to be detrimental to infants with congenital heart disease, although we did not assess neurological outcomes. Nevertheless, due to the limitations of the retrospective design of this study and its limited power, more thorough clinical randomised controlled trials are needed.