Diabetic ketoacidosis: no clear difference between route of insulin administration.

Clinical bottom line (level 1b-)

  1. The route used to administer insulin in patients with diabetic ketoacidosis had no clear effect on the time taken to return to biochemical normality or the amount of insulin required.
Fisher et al: New England Journal of Medicine 1977; 297 (5): 238-241
Expires October 2003

The study

Unblinded concealed randomised trial with intention-to-treat
Setting: university hospital, USA

45 patients (aged range 19 to 75 years; mean 40, ?% male) diabetic ketoacidosis (defined as plasma glucose >16.5 mmol/l, pH <7.3, bicarbonate <15 mmol/l and glycosuria with ketonuria)
Control Group: (n = 15, 15 analysed): intravenous insulin infusion (with albumin)
Experimental Group: (n = 15, 15 analysed): intramuscular insulin injections
Experimental Group: (n = 15, 15 analysed): subcutaneous insulin injections
All patients had insulin 0.33 U/kg administered as a bolus, then seven units per hour until plasma glucose <14 mmol/l. Insulin was continued at two units per hour if glycosuria or hyperglycaemia was present. No insulin was given if plasma glucose <8 mmol/l. All patients had fluid and electrolyte replacement, and bicarbonate if required.
100% followed for 24 hours

The evidence

intravenous vs intramuscular
Outcome Control Group
(SD)		 Experimental Group
(SD)		 Mean Difference
(95% CI)
iv vs im: time to glucose <14 mmol/l (hours) 6.0
(5.4) 		4.9
(4.3) 		1.1
(-2.6 to 4.8)
iv vs im: time to bicarbonate >15 mmol/l (hours) 13.0
(8.5) 		12.2
(5.4) 		0.80
(-4.5 to 6.1)
iv vs im: time to pH >7.3 (hours) 8.7
(5.8) 		7.5
(4.3) 		1.2
(-2.6 to 5.0)
iv vs im: insulin required for total control (units) 100
(43) 		94
(58) 		6
(-32 to 44)

intravenous vs subcutaneous
Outcome Control Group
(SD)		 Experimental Group
(SD)		 Mean Difference
(95% CI)
iv vs sc: time to glucose <14 mmol/l (hours) 6.0
(5.4) 		5.6
(3.5) 		0.40
(-3.0 to 3.8)
iv vs sc: time to bicarbonate >15 mmol/l (hours) 13.0
(8.5) 		10.8
(4.3) 		2.2
(-2.8 to 7.2)
iv vs sc: time to pH >7.3 (hours) 8.7
(5.8) 		6.0
(3.1) 		2.7
(-0.78 to 6.2)
iv vs sc: insulin required for total control (units) 100
(43) 		85
(31) 		15
(-13 to 43)

Comments

  1. The small size of the study and the short follow-up means that significant benefit or harm from using one particular route may have been missed.
  2. Many clinicians feel that until adequate volume replacement has occurred, the intramuscular route is more likely to provide more efficient uptake from the depot site.

Citation

  1. Fisher JN, Shahshahani MN, Kitabchi AE: Diabetic ketoacidosis: low-dose insulin therapy by various routes. New England Journal of Medicine 1977; 297 (5): 238-241
Contributor: Chris Ball and Clare Wotton, July 2000
Reviewer: Jon Levine

Clinical Question.
Patient DKA
Intervention or Exposure route of administration of low-dose insulin
Outcome return to biochemical normality