|
|
Click on any of the headings or subheadings to rapidly navigate to the relevant section of the guidemap
History of the present illness
Medical decision-making/treatment
| Introduction |
- diabetic ketoacidosis is a state characterized by hyperglycemia (glucose > 250mg/dl), acidosis (ph < 7.35), low serum bicarb, high anion gap and positive serum ketones
- DKA is caused by an absolute or relative deficiency of insulin +/- an increase in the level of counter-regulatory hormones
- the pathophysiology of DKA is based on the inability of cells to take up and utilize glucose and an increased degree of lipolysis and proteolysis
- the metabolic acidosis is mainly due to the presence of increased ketoacids – acetoacetic acid and beta-hydroxybutyric acid => high anion gap acidosis
- beta-hydroxybutyric acid is the major ketoacid produced, and it does not react positively with the nitroprusside reaction => so the urine or blood test for ketones may only be slightly positive and serum ketone testing does not correlate with the degree of ketoacidosis
- during treatment of DKA, beta-hydroxybutyrate is converted to acetoacetate => the test for ketones may become paradoxically more strongly positive - despite an improvement in the state of DKA and a decrease in the calculated anion gap
- if the patient remains relatively well hydrated during the development of DKA => sodium is excreted with the ketoacids => relative excess of chloride => hyperchloremic acidosis (non-anion gap acidosis)
- the same phenomenon may occur during the early treatment phase of DKA (first 2 - 4 hours) when sodium chloride solution is administered => iatrogenic hyperchloremic acidosis => the patient will remain acidotic despite an improvement in the state of DKA and a decrease in the anion gap
- the hyperchloremic acidosis is benign and will resolve with time, but a persistent metabolic acidosis may give the treating physician a false impression of 'resistant' DKA if only the serum pH (and not the anion gap) is monitored
- a high anion gap metabolic acidosis may also be due to an accompanying lacticacidosis (secondary to sepsis and/or shock) and/or renal failure secondary to volume depletion
- a mixed metabolic acidosis and metabolic alkalosis may co-exist if the patient vomits excessively (=> loss of gastric acid) and the combination may mask the true extent of the underlying ketoacidosis by producing a 'falsely' high serum pH
- extreme elevations of the anion gap (from loss of chloride from vomiting) + pH that is higher than expected for the size of the anion gap are the major clues suggesting a combined metabolic acidosis and alkalosis
(* think of the differential diagnosis of alcoholic ketoacidosis if the patient has a metabolic alkalosis + positive serum ketones)
- fluid volume deficits in DKA vary widely and average ~ 10% of body weight
- sodium deficits average 5 - 10 meq/kg and the serum sodium may be normal despite large sodium deficits
- potassium deficits average 3 - 5 meq/kg and the serum potassium may be normal, low or high
- a high serum potassium (despite total body potassium deficiency) could be due to the insulin deficiency (insulin is required to move potassium from the blood => cells), the metabolic acidosis secondary to increased ketoacids, decreased renal perfusion, and any accompanying tissue catabolism
- a total body deficiency of magnesium and phosphate usually also accompanies DKA, but specific replacement therapy is rarely required
- the serum osmolality is often increased (proportional to the level of hyperglycemia) and there is a close correlation between the serum osmolality and the state of altered mental status
- patients with a serum osmolality > 340mosm/L are often comatose; conversely, coma is not due to DKA if the calculated serum osmolality is < 300 osm/L
- euglycemic ketoacidosis exists when the serum glucose is < 250mg/dl in a patient presenting with DKA
- most of the euglycemic DKA patients are young, well hydrated, on insulin and facing an intercurrent illness
- the exact cause of euglycemic ketoacidosis is unknown, but it is important to recognize => because insulin must still be administered to correct the ketoacidotic state and IV glucose must be given concurrently to prevent resultant hypoglycemia secondary to insulin administration
| History of the present illness |
- 20 - 30% of all cases of DKA occur in patients with no history of diabetes and they may present with symptoms suggestive of gastro-enteritis – nausea, vomiting and abdominal pain
- the abdominal pain may be severe and is frequently of unknown cause (especially in children - postulated to be due to liver capsule distension or gastroparesis) or the abdominal pain could be secondary to pancreatitis or another abdominal illness precipitating DKA (eg. bowel infarction)
- the patient may have a history of increased thirst and polyuria and nocturia
- fatigue and weakness are common presenting symptoms
- the patient may have altered mental status, which can vary in severity from lethargy => coma
- the patient may present with hyperventilation and peri-oral/limb paresthesias (often incorrectly diagnosed as hyperventilation syndrome)
- a known diabetic patient may give a history of deliberate or accidental omission of insulin therapy
- inquire about symptoms of illnesses that may have precipitated the DKA (infections, AMI, CVA, mesenteric ischemia)
- a rare cause of DKA is type 1 fulminant DKA that occurs in patients who have no previous history of diabetes => DKA develops rapidly over a few days (cause is unknown, and it may be due to a viral illness affecting the islets cells of Langerhans in the pancreas)
| Risk factors for developing DKA |
- inadequate insulin use
- inadequate fluid intake
- excess of counter-regulatory hormones
- steroid excess eg. Cushings syndrome or steroid therapy
- hyperthyroidism, pheochromocytoma, somatostatinoma, glucagonoma
- stress secondary to intercurrent illnesses or infection
- drugs (sympathomimetics, pentamidine, thiazides, dilantin, calcium channel blockers)
| Examination |
- hypotension could be secondary to massive volume depletion or sepsis or a precipitating cause of DKA eg. cardiogenic shock or massive pulmonary embolism
- hyperventilation (Kussmaul breathing) may be subtle or clinically obvious
- the fruity odor of acetone on the breath may only be detectable by ~ 50% of people
- clinical signs of dehydration depend on the degree of dehydration (5 - 20%)
- fever is not due to DKA and infection should be strongly suspected
- abdominal tenderness may be present and suggestive of an acute abdomen (pseudoperitonitis diabetica)
(* persistent abdominal tenderness after effective treatment of DKA => suggests associated intra-abdominal pathology)
- any altered mental status is variable in degree and does not correlate with the degree of acidosis
- focal neurological signs may rarely occur - they are more common in hyperosmolar non-ketotic hyperglycemia
- search diligently for signs of occult infection eg. perineum and peri-rectal area for incipient cellulitis or deep abscesses
| Diagnostic testing |
- routine blood work includes a serum glucose, serum electrolytes, CBC, serum ketones, +/- serum magnesium and serum phophate
- an increased WBC (10,000 - 40,000 cells/ml) is common in DKA and only suggests infection if significant left shift with bandemia is present
- blood cultures are indicated in the febrile DKA patient
- DKA causes an increased serum amylase (salivary), therefore a serum lipase should be measured if pancreatitis is a diagnostic possibility
(* fulminant type 1 DKA can cause an elevated serum lipase in the absence of pancreatitis)
- serum ketones may cause a factitious increase in the serum creatinine
- hypertriglyceridemia may produce factitiously low serum sodium and serum glucose readings (recognized by the presence of milky plasma)
- the serum betahydroxybutyrate is the best measure of the degree of ketoacidosis, because it is the predominant serum ketone present, and changes in the serum betahydroxybutyrate can be used to monitor successful treatment of DKA
- the serum acetoacetate may be low in early DKA and the standard serum ketone test (based on the nitroprusside reaction) may be negative or minimally positive in early DKA; betahydroxybutyrate gets converted to acetoacetate following insulin therapy and the serum ketone test may become paradoxically more positive during the course of DKA therapy
- the urine ketone test may be more sensitive than the serum ketone test in early DKA, because acetoacetate is concentrated in the urine; however, the urine ketone test cannot be used to monitor the successful treatment of DKA
- an arterial or venous blood gas is required to determine the degree of acidosis and the degree of compensatory hypocarbia
- venous blood gases can be used to follow the course of the acidosis during therapy of DKA, as it closely parallels the arterial blood gas reading
- to look for signs of cardiac ischemia and hyperkalemia/hypokalemia
- to look for evidence of pneumonia or CHF, which may have precipitated the DKA
- is indicated:-
- if there is specific clinical evidence of intra-cranial pathology
- if there are any focal neuro signs
- in a comatose patient with a calculated serum osmolality < 340 mosm/L
- if the patient does not awaken promptly following effective treatment of the hyperglycemia with insulin and hydration
| Medical decision-making and treatment |
- first focus your attention on stabilizing the vital signs
- then focus your attention on IV hydration and electrolyte replacement therapy
- insulin therapy should commence after IV hydration has been instituted
- constantly cardiac monitor the patient for early electrocardiographic evidence of hyperkalemia or hypokalemia
- a flow chart should be created and the serum glucose and potassium should be measured every 1 - 2 hours initially => then 2 - 4 hourly when effective therapeutic control of the DKA is well under way
- a 20 cc/kg bolus of normal saline should be given IV to patients in overt hypovolemic shock and repeat boluses of 10 - 20 cc/kg of normal saline should be given IV until the systolic blood pressure is > 80 mmHg
- if the adult patient is not in hypovolemic shock, one litre of fluid can be given IV over the first hour => then 500 - 1,000cc of fluid given over the second hour => and then ~ 500cc of fluid given every hour for the next few hours for the average-sized adult (this is the simplified approach)
- a more accurate approach is to base fluid therapy on the body weight and the degree of clinical dehydration
- calculate the replacement fluid requirement as 10cc/kg of IV replacement fluid/every 1% of estimated dehydration => and give 50% of the replacement fluid over the first 8 hours and the rest evenly over the rest of the 24-hour time period
- maintenance fluid requirements should be calculated and be given evenly over the 24-hour time period
(* see the appendix for details of fluid administration – works out to about ~ 500cc of IV fluid/hour during the first few hours in the average sized adult)
- a patient with a calculated serum osmolality > 340 mosm/L should be given fluid more slowly, and fluid deficits should be replaced over 36 - 48 hours
- obese patients should have fluid requirements calculated on the basis of ideal body weight to prevent possible over-hydration
- IV boluses of hypotonic fluid should be avoided in children because of the risk of cerebral edema occuring as a result of the too rapid reduction of the serum osmolality => use 7.5-10 ml/kg of normal saline during the first hour of treatment => 3.5-5ml/kg/hour of normal saline (or 0.45 normal saline) thereafter
- sodium is replaced as normal saline for the first 1 - 2 hours and then switched to 0.45 normal saline for the rest of the replacement period - unless the patient remains hyponatremic (serum sodium < 135 meq/L)
- using isotonic sodium chloride solution as the standard replacement fluid results in a 'relative' excess of chloride => hyperchloremic acidosis (which is a temporary phenomenon that is not necessarily harmful)
Correction of potassium deficit
- potassium replacement should be started as soon as it is determined that the patient is urinating + not hyperkalemic (serum potassium >5.0 mEq/L)
(* use the EKG to look for signs of hyperkalemia if stat serum potassium levels are not readily available)
- expect a precipitous drop in the serum potassium when starting insulin therapy (mainly due to insulin-mediated re-entry of potassium into cells and due to correction of the metabolic acidosis)
- monitor the serum potassium every ~ 2 hours and add potassium to all saline solutions during the first 4 - 6 hours if the patient is not hyperkalemic (serum potassium > 5.0 mEq/L)
(* some experts recommend that potassium be omitted from the first litre of IV fluid replacement - while the serum potassium level and urine output are being determined during the first hour of therapy, so as to avoid inadvertently administering potassium to patients with unknown hyperkalemia)
- the amount of potassium given per hour depends on the serum potassium
level
| serum potassium | potassium replacement |
| < 3.0 meq/L | 40 meq/hour |
| 3 - 4 meq/L | 30 meq/hour |
| 4 - 5 meq/L | 20 meq/hour |
| 5 - 5.5 meq/L | 10 meq/hour |
- standard formulae recommendations include giving 2/3's of the potassium as potassium chloride and 1/3 as potassium phophate, but this may result in excess phosphate replacement => iatrogenic hypocalcemia and tetany
(* it is probably safer not to give any phosphate IV, or to limit the amount of administered phosphate to the minimum amount required to treat any clinically 'apparent' signs of hypophosphatemia)
- immediate insulin administration should be avoided if the initial serum potassium is <3.0 mEq/L (to prevent inducing life-threatening hypokalemia) => administer potassium at 40 mEq/hour and add insulin therapy when the serum potassium is >3.5 mEq/L
- fluid therapy alone will result in a significant decrease in the serum glucose and immediate insulin therapy is not mandatory
- regular insulin is given by constant IV infusion (0.1 unit/kg/hour) +/- an initial IV bolus of regular insulin (0.1 unit/kg)
(* the initial IV bolus of regular insulin can be safely omitted in a patient with DKA because there is no EBM evidence to support its use; insulin therapy should be temporarily delayed if hypokalemia is present; bolus IV insulin therapy may cause a too rapid reduction in the serum glucose in children and may it may precipitate cerebral edema due to a too rapid reduction in the serum osmolality if concurrent IV hypotonic fluids are also given too rapidly)
- the IV tubing should be 'primed' by first running ~ 10% of the prepared insulin solution through the IV tubing because polvinyl chloride adsorbs insulin => then albumin will not have to be added to the IV bag to prevent insulin adsorption by the plastic tubing
- IV insulin should be given by constant infusion until the acidosis and ketosis is corrected
- the serum glucose should decrease by ~ 75 - 100 mg/dl/hour
- double the dose of IV insulin if the decrease in serum glucose is < 50 mg/dl/hour
- glucose should be administered IV to prevent iatrogenic hypoglycemia when the serum glucose reaches 250 - 300 mg/dl -- administered as a 5 - 10% glucose solution at ~100 - 200 cc/hour
- the glucose can either be administered as a D5-1/2NS (or D5-water solution) as deemed appropriate
- a patient with euglycemic DKA may require glucose administration from the onset of therapy, which may necessitate a separate IV for glucose administration at an infusion rate (~50 - 200cc of 5 - 10% glucose solution) that will keep the serum glucose in the range of 100 - 300 mg/dl
(* remember that the end-point of therapy is not the correction of hyperglycemia, but correction of the ketoacidotic state)
- the dose of insulin can be decreased to 0.05 units/kg/hour when euglycemia and significant correction of the ketoacidosis (pH > 7.3 and serum bicarb > 15) has been achieved => switch over to sc insulin therapy
- the IV infusion should be continued for 1 - 4 hours after sc insulin is given to ensure that a smooth overlap occurs
- there is no definite evidence that bicarbonate administration is therapeutic (even in patients with a severe metbolic acidosis)
- therefore, bicarbonate administration should only be given to patients with a profound metabolic acidosis (pH <7.0) if the acidosis does not correct within 1 hour of fluid and insulin therapy
- bicarbonate administration is fraught with many potential complications:- i) precipitous hypokalemia, ii) shift of the oxy-hemoglobin saturation curve to the left and decreased tissue unloading of oxygen, iii) induction of paradoxical CNS acidosis, iv) sodium overload, vi) increased serum osmolality and vii) late alkalemia when bicarbonate is regenerated from ketoacids
- the amount of bicarbonate that can safely be given depends on the
serum pH
| Serum pH | Acceptable amount of
sodium bicarbonate |
| pH < 6.8 | 2 meq/kg over 30 - 60 minutes IV |
| pH 6.8 - 6.9 | 1 - 1.5 meq/kg over 60 minutes IV |
| pH 6.9 - 7.0 | 0.5 meq/kg over 1 - 2 hours IV |
| pH > 7.0 | No bicarbonate should be given |
- if a decision is made to give bicarbonate => it should only be given to achieve a serum pH of 7.0 - 7.1
Correction of magnesium deficit
- there is no evidence that routine magnesium therapy is therapeutic in DKA
- most patients with DKA are total body magnesium depleted, although the serum magnesium may be normal
- 2 - 4 g of magnesium can be given over a few hours IV if the serum magnesium is very low or if the patient is definitely symptomatic from hypomagnesemia
Correction of phosphate deficit
- most patients with DKA are phosphate depleted, but there is no evidence to show that phosphate replacement is actually needed in DKA patients
- phosphate can be given to the severely hyphosphatemic patient (serum phophate < 1 mg/dl), or if definite evidence of clinical hyphophatemia (respiratory depression, skeletal muscle weakness, rhabdomyolysis, cardiac dysfunction, confusion and hemolytic anemia) exists, or if the patient has severe underlying hypoxia - which may be worsened by low 2,3-DPG levels
- only small amounts of phosphate need to be given IV over 6 hours [ 0.08 - 0.16mmol/kg of potassium phosphate ( 2.5 - 5.0 mg of elemental phosphorus) in 500 cc of 0.45%NS ] if phosphate therapy is indicated
- large amounts of phosphate IV => may cause hypocalcemia and secondary tetany
- more common in children (~0.3% of cases)
- often becomes clinically apparent ~ 6 hours after commencement of therapy, although some degree of sub-clinical cerebral edema is probably present in all patients with DKA
- unknown etiology, although there is some evidence to suggest that it is causally related to the over-rapid correction of the elevated serum osmolality secondary to IV insulin bolus therapy and too rapid administration of hypotonic IV fluids.
- can be heralded by a headache and vomiting, or a sudden deterioration of the patient's mental status, or a sudden respiratory arrest
- mannitol (0.5 g/kg) should immediately be given IV over 15 minutes if cerebral edema is clinically suspected (prior to ordering a CT scan)
- associated with a high mortality rate and there is no known effective therapy
- mannitol (0.25-0.5mg/kg q hourly), 3% hypertonic saline, intubation +/- hyperventilation, dexamethasone have been used empirically to treat cerebral edema
Venous and arterial thrombosis
- an infrequent complication
- a cerebral venous sinus thrombosis should be considered in a patient with a headache and altered mental status
- prophylactic heparinization should be considered for DKA patients, who remain comatose for prolonged periods
- cardiogenic pulmonary edema may occur in elderly patients with underlying cardiac disease
- may require cautious fluid administration (under the guidance of pulmonary artery catheter wedge pressure readings) in these vulnerable patients to decrease the risk of this complication
- non-cardiogenic pulmonary edema occurs rarely => typical ARDS presentation
- thought to be secondary to the associated hypertriglyceridemia
- diagnosed by an elevation of the serum lipase
- discharge and outpatient treatment is appropriate for patients with mild DKA if all of the the following criteria are met:-
- the initial glucose is < 500mg/dl + initial serum pH > 7.2 (or a bicarb > 10meq/L)
- the ph is > 7.35 + serum bicarb >20 + the patient becomes euglycemic with no ketonuria after a few hours of ED therapy
- there is resolution of all significant symptoms
- normal vital signs
- there are no co-morbid or precipitating illnesses that warrant hospital admission
| Appendix |
Calculated serum osmolality
Calculated serum osmolality = 2 x [serum sodium + potassium] + serum glucose/18 + BUN/2.8
or a more simplified (but slightly less accurate) calculation
Calculated serum osmolality = 2 x [serum sodium + potassium] + serum glucose/20
Correction of serum sodium
- add 1.6 meq/L to the serum sodium level for every 100mg/dl increase in the serum glucose > 100mg/dl
Replacement fluid calculation
- 10cc/kg/1% dehydration
- 100 cc/kg for the first 10 kgs of body weight, + 50 cc/kg for the second 10 kgs of body weight and 20 cc/kg for each kg greater than 20 Kg of body weight
Sample calculation:- 70kg patient, who is 10% dehydrated
Maintenance fluid = 1,000 + 500 + 1,000 = 2,500 over 24 hours = ~ 100 cc/hour
Replacement fluid = 100cc/kg = 7,000 cc over 24 hours with ~ 50% of the fluid (3,500 cc) given over the first 8 hours => ~ 400 cc/hour
Total amount fluid per hour => ~ 500cc/hour for the first 8 hours
(a calculated correction should be made for any fluid given in the first recuscitative hour and for any ongoing losses from vomiting etc.)
Disclaimer:
My EM guidemaps reflect my personal approach to problem-solving/managing
clinical cases in an ED setting and they should not be regarded as the
standard of care. They merely represent the personal opinions of the author
and they should only be used in clinical practice if the reader-user has
substantial reason to believe that the clinical advice contained in the
guidemaps is valid and accurate. The guidemaps are not meant to be "authoritative"
and the reader-user should consult standard medical textbooks and expert
opinion articles/guidelines for more authoritative advice. The reader-user
should particularly confirm all drug doses, their indications and contra-indications,
prior to their use.