Figure 1 . The risk of DKA . |
• Omission of insulin therapy , reduction in insulin dose or inadequate dosing .
• Insulin pump failure , occlusion or infusion site failure .
• Prolonged nausea and vomiting .
• Infections ( eg , pneumonia , COVID-19 , urinary tract infections , sepsis ).
• Eating disorders , diabulimia .
• Binge drinking .
• Pregnancy .
• Very low-carbohydrate / food intake or prolonged fasting .
• Extreme physiological stress ( eg , dehydration , blood loss , major surgery , trauma , myocardial infarction , pancreatitis ).
• Initiation of certain medications ( corticosteroids , pentamidine , check-point inhibitors , antipsychotic drugs , illicit drugs ).
|
|||||||||
( EKA ), and is a rare complication |
such as severe illness , extended fast- |
the characteristic ‘ rotting apple ’ or |
Treatment with SGLT2 inhibitors |
type 1 diabetes ( that is , before insu- |
||||||
of various different severe metabolic |
ing ( for example , gastrointestinal |
‘ nail polish ’ breath odour of a patient |
in individuals with type 2 diabetes |
lin is started ). Indeed , DKA can be |
||||||
stressors in susceptible individuals |
illness , acute abdomen ), may rarely |
with ketoacidosis . |
causes glycosuria ( glucose wasting ) |
a presenting feature in up to half |
||||||
( see box 1 ). |
precipitate ketoacidosis , especially in |
Most DKA events are sudden . |
that induces comparable metabolic |
of Australian children with type 1 |
||||||
People with diabetes can also |
women with gestational or pre-exist- |
Although a profound deficiency of |
changes to a low-carbohydrate diet , |
diabetes . 12 |
||||||
experience EKA . In fact , when put |
ing diabetes . |
insulin is the predisposing state , DKA |
including reduced insulin and ele- |
DKA can also occur in patients |
||||||
under the same metabolic stress ( for |
EKA has also been described fol- |
is invariably associated with a sig- |
vated glucagon concentrations . |
with type 1 diabetes because of |
||||||
example , pregnancy , starvation , sep- |
lowing excessive chronic alcohol |
nificant intercurrent illness or trig- |
On average , blood ketone levels |
non-adherence or omission of insu- |
||||||
sis ), EKA is more common in diabetic |
intake , often during withdrawal or |
ger event ( see box 2 ), that results in |
are approximately double in diabetic |
lin therapy ( for example , missed |
||||||
individuals . |
because of symptoms preventing |
increased release of counter-regu- |
patients treated with SGLT2 inhibi- |
insulin doses , insulin pump failure , |
||||||
EKA is also observed following the use of SGLT2 inhibitors in people with type 2 diabetes . 15 Indeed , SGLT2 inhibitors are contraindicated |
food intake such as nausea , vomiting or abdominal pain . 5
EKA may rarely occur in non-diabetic patients with prolonged and
|
latory ( or stress ) hormones such as glucagon , corticosteroids , catecholamines , leptin and growth hormones . These hormones further oppose the |
tors compared with other therapies . 18 These levels are similar to the magnitude of increase observed during pregnancy , where the metabolic |
or diabulimia [ an eating disorder where those with type 1 diabetes intentionally give themselves less insulin than they need or stop |
||||||
in patients with type 1 diabetes , spe- |
severely reduced carbohydrate / food |
actions of ( inadequate ) insulin and |
activity of the placenta and develop- |
using their insulin in order to lose |
||||||
cifically because of the increased risk of EKA in this setting . 16 Recent clinical trials in patients with type 2 diabetes suggest that treatment with SGLT2 inhibitors may double the risk of EKA . 15 Nonetheless , because so many patients with type 2 diabetes are now receiving SGLT2 inhibitors as part of their ongoing care , SGLT2 |
intake ( for example , prolonged fasting , starvation , illness , gastrointestinal disorders ) because of increased ketogenesis combined with dehydration . 5 For example , EKA has also been reported after bariatric surgery . 17
PATHOGENESIS
KETONES are naturally generated
|
markedly augment ketogenesis . Intercurrent illnesses are an unavoidable part of normal life . Even in adults with long-standing established type 1 diabetes , such events can sometimes sporadically trigger DKA , even if there is no history of DKA events , poor glucose control or risk-taking behaviour . |
ing baby , rather than the urine , provides the metabolic sink for glucose losses . 4 Again , pregnancy-related ketosis is distinct from pathophysiological ketoacidosis . However , both pregnancy and SGLT2 inhibition are ‘ ketosis-prone ’ states ( see box 1 ) and risk factors for EKA , given the right ( or more correctly , the wrong ) |
weight ]).
Most patients with established type 1 diabetes develop DKA during intercurrent infections that suddenly increase their insulin resistance , while at the same time hypermetabolism , volume depletion and vasodilation associated with infections can amplify ketogene-
|
||||||
inhibitors have become a leading |
from fatty acids by the liver as an |
Blood ketone levels are naturally |
trigger . |
sis in some patients . Concurrently , |
||||||
cause of EKA . There has been only one recorded event of ketoacidosis in participants without diabetes receiving an SGLT2 . 15 |
alternative mitochondrial fuel source for metabolically demanding tissues , such as the brain , the heart and skeletal muscle . Ketone production |
elevated when following a diet low in carbohydrate ( for example , a “ keto ” diet that is high in fat and very low in carbohydrate , see figure 3 ). This |
RISK FACTORS AND TRIGGERS
KETOACIDOSIS is almost always
|
patients often reduce their insulin dose because of the perceived risk of hypoglycaemia when they feel unwell and are not eating and |
||||||
EKA is a very rare complication of |
increases as insulin levels fall and |
is known as physiological ketosis |
the result of a trigger event ( see |
drinking . Finally , in some adults |
||||||
pregnancy . 4 Pregnancy naturally acts to amplify ketone production , and |
glucagon levels rise in the fasting / starved state as glycogenolysis by the |
or nutritional ketosis , which actually has the goal of increasing serum |
box 2 ) in susceptible individuals . The most common trigger event |
with type 1 diabetes , binge drinking can supress gluconeogenesis and |
||||||
ketone levels are usually higher in |
liver is exhausted ( see figure 2 ). |
ketone concentrations of above |
in patients with type 1 diabetes is |
increase the demand for an addi- |
||||||
pregnancy because of systemic insu- |
The major ketones , acetoacetate |
0.5mmol / L . This state is very distinct |
severe volume depletion / dehydra- |
tional ( ketone ) fuel substrate . |
||||||
lin resistance and the large metabolic |
and β-hydroxybutyrate , are organic |
from pathophysiological ketoacidosis |
tion due to severe hyperglycaemia , |
Importantly , these same trig- |
||||||
demands of the developing fetus and |
acids that in high enough levels cre- |
( see figure 4 ), where levels of ketone |
which drives a strong counterreg- |
ger events are also implicated in |
||||||
placenta , particularly in the third trimester . However , intercurrent dia- |
ate the ‘ high anion gap ’ metabolic acidosis characteristic of ketoacidosis . |
concentrations are much higher and fail to be offset by renal compensa- |
ulatory response and unimpeded ketogenesis . This is most likely at |
most cases of EKA in ketosis-prone settings ( see box 1 ). For example , |
||||||
betes and / or during trigger events |
Acetone is also generated , creating |
tion of acidosis . |
the first presentation of new-onset |
severe COVID-19 infection |
PAGE 30 |