Diabetic ketoacidosis: no clear benefit from high-dose insulin infusions.

Clinical bottom line (level 2b-)

  1. Patients with diabetic ketoacidosis who were put on low-dose insulin infusions compared with high-dose insulin injections were not clearly more likely to become hypokalaemic.
  2. There was no clear difference in time taken to return to biochemical normality.
Gonzalez Villalpando et al: Journal of the American Medical Association 1979; 241: 925-927
Expires October 2003

The study

Unblinded concealed quasi-randomised trial with intention-to-treat
Setting: emergency department, acute hospital , USA

12 patients (aged ?, ?% male) with 18 episodes of diabetic ketoacidosis (defined as serum glucose > 17mM, arterial pH < 7.25, bicarbonate < 15mM, and ketonaemia)
Control Group: (n = 9, 9 analysed): conventional dose regimen: insulin 100 units iv bolus then 25 to 50 units iv per hour (bolus) 'depending on response to treatment'
Experimental Group: (n = 9, 9 analysed): low-dose regimen: insulin 30 units iv bolus then 0.1 units/kg/hr iv
Saline, potassium and phosphate was given to both groups. Use of bicarbonate was kept to a minimum.
100% followed for 12 hours
Outcome notes:
  • hypokalaemia : <3.0 mmol

The evidence

Outcome Time to outcome CEREERRRR
(95% CI)		ARR
(95% CI)		NNT
(95% CI)
hypokalaemia 12 hours 4
(44.4%)		 1
(11.1%)		 75%
(-82% to 97%)		 33.3%
(-5.08% to 71.8%)		 3
(NNT = 1 to infinity;
NNH = 20 to infinity)

Outcome Control Group
(SD)		 Experimental Group
(SD)		 Mean Difference
(95% CI)
time for bicarbonate 15 mmol/l or more (hours) 9.56
(3.3) 		10.1
(0.87) 		-0.54
(-3.0 to 1.9)
time for arterial pH 7.30 or more (hours) 7.40
(2.0) 		7.80
(2.6) 		-0.40
(-2.70 to 1.90)

  • No episodes of hypoglycaemia occurred in either group.

    • Comments

    1. Tiny numbers and quasi-randomisation: patients were assigned alternately between the two treatment protocols.
    2. This report from 21 years ago was one of the first studies to document that continuous low dose insulin infusions were as efficacious as intermittent higher dose insulin regimes for treating diabetic ketoacidosis.

    Citation

    1. Gonzalez Villalpando C, Blachley JD, Vaughan GM, et al: Low- and high-dose intravenous insulin therapy for diabetic ketoacidosis. Journal of the American Medical Association 1979; 241: 925-927
    Contributor: Richard Hardem and Chris Ball, October 2000
    Reviewer: Sean F. Dinneen

    Clinical Question.
    Patient diabetic ketoacidosis
    Intervention or Exposure low-dose insulin infusions
    Comparison high-dose insulin injections
    Outcome hypokalaemia