What are the key features that characterize DKA?
DKA is characterised by the following:
Uncontrolled hyperglycaemia (Blood glucose >13.9mmol/L i.e. 250mg/dL)
Metabolic acidosis (pH <7.3; Bicarb <18mmol/L)
Increased total body ketones (urine ketone 2+ or BHOB > 3.0mmol/L)
ADA guidelines also include the criteria of a high anion gap acidosis (HAGMA)
This occurs due to a few factors, namely: insulin deficiency, coupled with increased counter-regulatory hormones (glucagon, cortisol, GH, catecholamines) that promote gluconeogenesis, glycogenolysis, lipolysis and impaired glucose utilisation by tissues. Increased free fatty acids not only induces insulin resistance but is also then converted through oxidation in the liver to ketone bodies (beta-hydroxybutyrate and acetoacetate). Insulin usually inhibits lipolysis and hepatic gluconeogenesis but in DKA, these brakes are released, allowing lipolysis and gluconeogenesis to proceed at an accelerated rate.
In children DKA mortality is up to 50% and is the most common cause of mortality in children with DM. In adults, DKA mortality is about 1% but can increase in the elderly population to about 5%
Precipitating factors:
Commonly infection
Discontinuation of insulin therapy (explore if patients have fear of hypoglycaemia, fear of weight gain, or other stresses that might stop them from injecting insulin)
Pancreatitis
MI/CVA
Operations
Drugs may affect carbohydrate metabolism and volume status
diuretics
beta-blockers
corticosteroids
second-generation anti-psychotics
anti-convulsants
sodium-glucose cotransporter-2 (SGLT-2) inhibitors
immune checkpoint inhibitors
Physical examination:
Fluid status: mucous membranes, JVP, skin turgor, check for crepitations at the base of the lungs and pedal oedema
Look out for Kussmaul's breathing
Examine for any focal symptoms to suggest a source of infection:
Examine lungs for crepitations
Palpate the abdomen looking for evidence of acute abdomen (peritonism/rigidity/guarding)
Renal punch to check for pyelonephritis
Check for focal neurology, neck stiffness for meningism, check for any photophobia
Evidence of insulin injection marks/lipohypertrophied sites
Features of autoimmune disease such as vitiligo/buccal or palmar crease hyperpigmentation to suggest underlying autoimmune conditions
Investigations:
Urine pregnancy test (Pregnancy is a ketotic state and puts patients at a higher risk of developing DKA)
Electrocardiogram
CXR
Urinalysis
Blood cultures if there is fever
Renal panel - urea and electrolytes, and bicarbonate
Betahydroxybutyrate
Venous blood glucose
Arterial blood gas to check the pH
iSTAT electrolytes
Lactate, amylase/lipase (if suspecting sepsis, dehydration, pancreatitis)
FBC, CRP
RP, LFT, random glucose
cortisol and TFT (can be considered if other autoimmune conditions are suspected)
HbA1c, anti-GAD, anti-islet cell Ab (antibodies may be sent if the diabetes diagnosis or subtype is yet to be confirmed)
Treatment:
Key principles:
Keep patient NBM
Hydration/fluid therapy
Insulin
Electrolyte replacement
Monitoring
Treat the underlying cause
Resolution
Investigation of underlying aetiology
Fluid therapy:
Isotonic normal 0.9% saline at 15-20ml/kg/hr (non shock) 20-30ml/kg/hr (shock) for the first hour
Subsequently assess haemodynamics, hydration status, serum electrolytes and urine output
If sodium is normal or elevated, use 0.45% NaCl at 250-500ml/hour
If sodium is low use normal saline at 250-500ml/hour
Be careful in patients with heart failure or CKD/ESRF to avoid fluid overload
Once serum glucose is < 11.1mmol/L (in my institution a higher cut off of 14mmol/L is used), a dextrose drip 5% should be added to allow continued insulin administration until ketosis resolves without causing hypoglycaemia
Hyperglycaemia generally takes 6 hours to resolve while ketosis takes 12hours to resolve
Electrolytes:
Potassium:
Because volume expansion, insulin therapy and correction of acidaemia will lower serum potassium, it is important to replace potassium once the serum potassium is in the normal range
Infuse 20-30mmol/L of potassium per litre of IV fluids administered
If there is hypokalaemia, delay insulin administration until K > 3.3mmol/L
Continuing IV insulin could worsen the hypokalaemia due to intracellular shifts and result in arrhythmias
Consider insertion of central line for rapid replacement of IV 20mmol/L per hour for 2 hours or IV with oral supplementation
If there is concurrent hyperkalaemia then consider serving 10 units of actrapid STAT + calcium gluconate to stabilise the cardiac membranes - then continue on with the IV insulin infusion.
Bicarbonate:
Use is controversial. 9 studies looking at 434 patients failed to show any difference in cardiac/neurologic function or rate of recovery from hyperglycaemia or ketonaemia when bicarbonate was used (Viallon A, Zeni F, Lafond P, Venet C, Tardy B, Page Y, Bertrand JC. Does bicarbonate therapy improve the management of severe diabetic ketoacidosis? Crit Care Med. 1999;27:2690–2693.)
There is also an increased risk of hypokalaemia, decreased tissue oxygen uptake, cerebral oedema and paradoxical CNS acidosis with bicarbonate use
ADA recommends the use of 100mmol of 8.45% IV sodium bicarbonate in 400ml of sterile water at 200ml/hour over 2 hours along with 20mmol of IV KCL ONLY IF the pH is <6.9 as severe acidosis can have adverse vascular effects.
This should be given until the pH is >7.0 and can be repeated every 2H until pH > 7.0
Phosphate:
Regular phosphate replacement has not been shown to have significant benefits
Aim for normal phosphate, slow replacement if required
Insulin therapy:
Either continuous IV insulin OR regular frequent SC/IM insulin injections
IV insulin is preferred due to the short half life and easy titration whereas SC and IM insulin has a delayed onset of action with a prolonged half life
Options include 0.1unit/kg bolus followed by 0.1unit/kg/hr infusion OR
0.14units/kg/hr continuous infusion WITHOUT initial bolus
Aim for a 3-4mmol/hour reduction in plasma glucose
When glucose levels reach 11.1mmol/L (in my institution 14mmol/L is used instead), consider decreasing the insulin infusion rate to 0.02-0.05units/kg/hr or halving the insulin infusion rate and add a dextrose drip component as outlined above
Maintain the plasma glucose between 8-11mmol/L for DKA
For euglycaemic DKA: (From JBDS March 2023 Guidelines)
1) Initiate dextrose 10% straight away at 125 ml/hr (3L/24h) because the glucose is <14 mmol/L
2) Begin with 0.1units/kg/hr insulin rate
3) If glucose continues to fall despite the 10% dextrose drip reduce IV insulin to 0.05 units/kg/hr to avoid hypoglycaemia
Criteria for resolution:
Blood glucose< 11.1mmol/L AND 2 of the following:
Bicarbonate >15mmol/L
Venous pH > 7.3
Calculated anion gap < 12
Ketones <0.6mmol/L
Transition to SC insulin:
Allow an overlap of 1-2 h between administration of SC insulin and discontinuation of IV insulin to prevent recurrence of hyperglycaemia or ketoacidosis
If the patient is kept NBM/fasting then continue IV insulin
If the patient is allowed diet, initiate insulin at 0.5-0.8units/kg/day MDII
Transition to basal glargine + rapid acting insulin had lower rates of hypoglycaemia (15%) as opposed to 41% with NPH and actrapid (Umpierrez GE, Jones S, Smiley D, et al. Insulin analogs versus human insulin in the treatment of patients with diabetic ketoacidosis: a randomized controlled trial. Diabetes Care. 2009;32(7):1164-1169. doi:10.2337/dc09-0169)
Common exam questions to consider:
What is the difference between urinary ketones and blood ketones:
In urinary ketone measurement, acetoacetate is measured. Nitroprusside reacts with acetoacetate to produce a purple-coloured complex.
In diabetic ketoacidosis (DKA) the 3-β-OHB:acetoacetate ratio increases from 1:1 to 5:1. With treatment of DKA, 3-β-OHB is oxidised back to acetoacetate. As a result, 3-β-OHB will decrease, but acetoacetate will increase. Measuring acetoacetate with urine sticks may therefore initially underestimate the severity of DKA and then continue to yield positive readings after the resolution of DKA.
Kidneys freely filter ketone bodies and reabsorb >99% of ketone bodies. Reabsorption is further increased in starvation states. Reabsorption is limited by the maximal transport capacity and can get inundated in ketoacidosis states, leading to ketonuria. Hence, urinary ketones are only detected above a certain threshold of ketosis and cannot accurately identify early ketosis.
Laffel L. Ketone bodies: a review of physiology, pathophysiology and application of monitoring to diabetes. Diabetes Metab Res Rev 1999;15:412–26.
Resources:
Laffel L. Ketone bodies: a review of physiology, pathophysiology and application of monitoring to diabetes. Diabetes Metab Res Rev 1999;15:412–26.
https://abcd.care/joint-british-diabetes-societies-jbds-inpatient-care-group
Comments