Diabetic ketoacidosis: no clear benefit from phosphate supplementation.

Clinical bottom line (level 1b-)

  1. There was no clear biochemical benefit with phosphate supplementation of diabetic ketoacidosis, and a potential for harm.
Fisher and Kitabchi: Journal of Clinical Endocrinological Metabolism 1983; 57: 177-180
Expires October 2003

The study

Unblinded concealed randomised trial without intention-to-treat
Setting: university hospital, USA

30 patients (aged mean 31 years, ?% male) diabetic ketoacidosis (defined as glucose >17 mmol/l, bicarbonate <15 mmol/l, pH <7.30, ketonaemia and glycosuria or ketonuria)
Control Group: (n = 15, 15 analysed): potassium chloride 12.5 mmol/h for 24 hours
Experimental Group: (n = 15, 15 analysed): potassium phosphate (8.5 mmol/h phosphate) for 24 hours
All patients had insulin (0.22 U/kg im and 0.22 U/kg iv as a bolus and repeated hourly until a fall of at least 10% from initial glucose, then 7 U im hourly), fluids and supportive care.
100% followed for 12 hours

The evidence

Outcome Control Group
(SD)		 Experimental Group
(SD)		 Mean Difference
(95% CI)
time to glucose 14 mmol/l or less (hours) 3.6
(3.1) 		5.4
(7.6) 		-1.8
(-6.1 to 2.5)
time to bicarbonate 15 mmol/l or more (hours) 10.5
(3.1) 		12.7
(7.0) 		-2.2
(-6.2 to 1.8)
time to pH 7.3 or more 11.3
(7.6) 		8.3
(4.6) 		3.0
(-1.7 to 7.7)
rate of glucose decline (mmol/l/hour) 5.2
() 		5.0
() 		0.2
( to )
rate of ketone body decline (mmol/l/hour) 0.64
() 		0.80
() 		-0.16
( to )

  • There were no statistically significant differences between the two groups in terms of resolution of hyperglycaemia, ketonaemia or acidosis.

    • Comments

    1. Small numbers, so limited power.
    2. This study showed that not only was there no clinical benefit of phosphate administration (faster rate of bicarbonate regeneration, faster correction of acidosis, quicker decline in hyperglycaemia) but there was a potential risk, namely the development of hypocalcaemia.
    3. Current recommendations for treatment of the electrolyte disturbance in diabetic ketoacidosis are limited to replacement of potassium in the form of potassium chloride.

    Citation

    1. Fisher JN, and Kitabchi AE: A randomized study of phosphate therapy in the treatment of diabetic ketoacidosis. Journal of Clinical Endocrinological Metabolism 1983; 57: 177-180
    Contributor: Richard Hardern and Chris Ball, July 2000
    Reviewer: Jon Levine

    Clinical Question.
    Patient DKA
    Intervention or Exposure phosphate supplementation
    Outcome biochemical benefit