medwireNews: Results of a randomised trial suggest that low-dose insulin may be at least as good as standard-dose insulin for the treatment of children with diabetic ketoacidosis (DKA).

The open-label trial of insulin at a low dose (0.05 U/kg per hour) versus a standard dose (0.10 U/kg per hour) involved 50 children aged 12 years or younger.

It was based in India, where DKA is common and comorbidities such as malnutrition “carry a high risk of therapy-related hypokalemia and hypoglycemia”, according to study author Muralidharan Jayashree (Postgraduate Institute of Medical Education and Research, Chandigarh, India) and co-workers.

The primary outcome of the rate of reduction in blood glucose to a level of 250 mg/dL or lower was similar in the low-dose and standard-dose groups, at 45.1 and 52.2 mg/dL per hour, respectively.

The upper limit of the 95% confidence interval for the mean difference between low-dose and standard-dose was 19.0 mg/dL per hour, which just exceeded the researchers’ prespecified noninferiority margin of 18.0 mg/dL. However, they note that this was “less than the entire assumed treatment effect and hence is a statistically persuasive finding.”

They add: “In a clinical context too, an inferior [blood glucose] decrease when associated with continuing resolution of acidosis is acceptable.”

Indeed, the average time to resolution of acidosis was similar with low-dose and standard-dose insulin, at 16.5 and 17.2 hours, respectively. There were no significant between-group differences in serial measurements of pH, bicarbonate and anion gap.

There were fewer instances of hypokalaemia (five vs 12) and hypoglycaemia (one vs five) among children given the low versus the standard dose, but this was not statistically significant. Two children in the low-dose group and one in the standard-dose group required an insulin dose increase, and one child given the standard dose developed cerebral oedema.

During the first hour of treatment, the average blood glucose reduction was smaller with the low than the standard insulin dose, at 39 versus 61 mg/dL. Also, children in the low-dose group less often had blood glucose reductions that were more rapid than desired (>90 mg/dL per hour), at 4% versus 10%.

In an editorial accompanying the study in JAMA Pediatrics, Andrea Granados and Joyce Lee, from the University of Michigan in Ann Arbor, USA, highlight this point, noting the importance of avoiding “excessive decreases” in blood glucose, which could “increase the potential risk of cerebral edema due to rapid changes in serum osmolarity”.

This finding “would be a reason to favor low-dose therapy over standard-dose therapy”, they say.

The editorialists caution that the study findings may be less applicable to Western settings, but describe the study as “an important development in the field”, which opens the door to larger trials to determine the ideal insulin dose.

By Eleanor McDermid, Senior medwireNews Reporter

JAMA Pediatr 2014; Advance online publication

Free abstract