Springtime hypoglycaemia in children with diabetes

Bandolier, Dec, 2003

The suggestion that hypoglycaemia might be more frequent in diabetic children in the Spring has been around for some time [1]. But other than something occasionally discussed, it does not seem to have been much studied. Bandolier tried to find out more because of reader enquiry. Two studies [2, 3] have addressed the issue, linking adverse events in intensively treated diabetic children with timing of hypoglycaemia and seasonal variation in glycated haemoglobin levels.

Studies

Children aged 0-18 years with diabetes in a geographical area of Sweden constituted the population. They were treated with four or more doses of insulin a day, with self-control from an early age. Fast-acting insulin was combined with slow- or intermediate-acting insulin. Meals were regular in time and content. A multi-disciplinary diabetic team was involved with the care of individuals, with regular contact.

Children or parents were asked to register every severe hypoglycaemic event and hospital admission.

Results

The children had an average age of 13 years, with a range of 1-18 years. Most (95%) used at least four insulin injections a day. Average haemoglobin A 1c levels were 7%.

The yearly incidence of unconsciousness and severe hypoglycaemia without unconsciousness were recorded during 1994 and 1995 for 126 and 122 children respectively [1]. In each year the incidence of unconsciousness was 0.2 per patient per year, with 12% of children unconscious. The incidence of severe hypoglycaemia without unconsciousness was 1.3 per patient per year, with 34% of children experiencing an event.

There was no difference in glycated haemoglobin or other parameters between those children who experienced unconsciousness and those who did not. Unconsciousness occurred more frequently in Spring than at other times of the year (Figure 1).

A second study [2] examined haemoglobin A 1c levels throughout the year, based on 810 blood samples from 114 children aged 2-18 years. Lower values were found in Spring and Summer (Figure 2), despite no change in insulin.

Comment

Seasonality of blood glucose, glycated haemoglobin, and hormones have been noted several times over the past few decades, as has seasonality in hypoglycaemia in childhood diabetics [4]. The reasons for it are not clear. The Swedish authors [2,3] suggest that it could be associated with a change from lesser activity during the winter months, to greater activity during the spring and summer months. This seems to make sense, but the amount of information we have is slight. Perhaps the take-home message is that increased hypoglycaemic episodes in the Spring in diabetic children are to be expected, and that they might benefit from knowing this.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

Hypoglycaemia and hospital admission are not without consequences, as a large French study shows [5]. Extrapolating from a study in southern France, it estimated that there would be 10,800 admissions for hypoglycaemia in France every year (population 59 million in metropolitan France). One in 10 of these admissions would result in a hospital stay of a day or more, though the average would be 6.5 days costing an average of US$2,100. Death would result in 1.9% of patients being admitted (an older age group than the paediatric population).

References:

[1] JD Baum, AL Kinmonth. Spring hypoglycaemia in diabetic children. BMJ 1980 280: 1227.

[2] S Nordfeldt, J Ludvigsson. Adverse events in intensively treated children and adolescents with type 1 diabetes. Acta Paediatrica 1999 88: 1184-1193.

[3] S Nordfeldt, J Ludvigsson. Seasonal variation of HbAic in intensive treatment of children with type 1 diabetes. Journal of Pediatric Endocrinology & Metabolism. 2000 13: 529-535.

[4] D Daneman et al. Severe hypoglycaemia in children with insulin-dependent diabetes mellitus: frequency and predisposing factors. Journal of Pediatrics 1989 115: 681-685.

[5] MP Allicar et al. Frequence et coutes des sejours hospitaliers pour hypoglycemie en France en 1995. La Presse Medicale 2000 29: 657-661.