Investigations

1st investigations to order

blood glucose

Test
Result
Test

Blood glucose is high; ≥30 mmol/L ( ≥540 mg/dL).[6]

The Joint British Diabetes Societies (JBDS) guideline recommends hourly monitoring of blood glucose for the first 24 hours to measure the patient’s response to treatment.[6]

Result

elevated (≥30 mmol/L [ ≥540 mg/dL]).[6]

blood ketones

Test
Result
Test

Use the blood ketone concentration to help distinguish HHS from diabetic ketoacidosis (DKA). There is usually no significant ketonaemia (ketones are usually ≤3 mmol/L) in patients with HHS.[6]

Result

negative or low (usually ≤3 mmol/L)[6]

venous blood gas

Test
Result
Test

Check the lactate level and pH.[6]

  • A mild acidosis (pH >7.3, bicarbonate >15 mmol/L [>15 mEq/L]) may be present due to renal impairment secondary to dehydration.[6]

  • Lactic acidosis may be present due to sepsis.[6]

Use a venous blood gas to monitor the patient’s biochemical progress (urea, electrolytes, glucose, serum osmolality, bicarbonate).[6]

Only perform an arterial blood gas if an accurate measurement of oxygen is required.[6]

Result

mild acidosis (pH >7.3, bicarbonate >15 mmol/L [>15 mEq/L]); lactic acidosis[6]

serum osmolality

Test
Result
Test

Calculate the serum osmolality; this is usually ≥320 mOsm/kg (≥320 mmol/kg) in patients with HHS.[6] [ Osmolality Estimator (serum) Opens in new window ]

The Joint British Diabetes Societies (JBDS) guideline recommends monitoring serum osmolality to measure the patient’s response to treatment as follows:[6]

  • Hourly for the first 6 hours

  • Every 2 hours from 6 to 12 hours as long as serum osmolality is falling at 3 to 8 mOsm/kg/hour (3-8 mmol/kg/hour) [ Osmolality Estimator (serum) Opens in new window ]

  • Every 4 hours after 12 hours if serum osmolality continues to improve.

    • Bear in mind, however, that in practice measurement of serum osmolality to monitor the patient's response to treatment may not be commonly used; check your local protocol. As an alternative, reduce monitoring of sodium, potassium, and urea to every 2 hours after 6 hours if these are improving.

Result

≥320 mOsm/kg (≥320 mmol/kg)[6]

urea, electrolytes, and creatinine

Test
Result
Test

Co-existing renal failure is common.[6]

In practice, always measure serum, rather than urine, electrolytes.

Hypokalaemia is common. However, hyperkalaemia may be present if there is significant acute kidney injury.[6]

Hypernatraemia or hyponatraemia may be present.

  • Hypernatraemia indicates severe dehydration.[1]

  • Hyponatraemia is mostly dilutional if the blood glucose is ≥30 mmol/L (≥540 mg/dL) and will normalise as the blood glucose is corrected.[6]

The Joint British Diabetes Societies (JBDS) guideline recommends monitoring sodium, potassium, and urea to measure the patient’s response to treatment as follows:[6]

  • Hourly for the first 6 hours

  • Every 2 hours from 6 to 12 hours as long as serum osmolality is falling at 3 to 8 mOsm/kg/hour (3-8 mmol/kg/hour)

  • Every 4 hours after 12 hours if they continue to improve.

Practical tip

In practice, local recommendations for monitoring may vary and measurement of serum osmolality to monitor the patient's response to treatment may not be commonly used; check your local protocol.

  • As an alternative, reduce monitoring of sodium, potassium, and urea to every 2 hours after 6 hours if these are improving.

Hypophosphataemia and hypomagnesaemia are common.[6]

Result

may show:

  • renal impairment

  • hypo/hyperkalaemia

  • hypo/hypernatraemia

  • hypophosphataemia

  • hypomagnesaemia

full blood count

Test
Result
Test

Leukocytosis is common in HHS and correlates with blood ketone levels.[11]

However, leukocytosis more than 25 × 10⁹/L (25,000/microlitre) may indicate infection and requires further investigation.[11]

Result

leukocytosis

ECG

Test
Result
Test

Use to look for cardiac precipitants of HHS such as myocardial infarction (MI), or consequences of HHS such as hypovolaemia, which could precipitate MI, or electrolyte abnormalities, which could cause arrhythmias.

  • Findings may include abnormal T or Q waves or ST segment changes.[53] 

Look for cardiac effects of electrolyte abnormalities.

  • Evidence of hypokalaemia (U waves) or hyperkalaemia (tall ‘peaked’ T waves) may be present.[54][55]

Result

abnormal T or Q waves or ST segment changes in myocardial infarction; evidence of hypokalaemia (U waves) or hyperkalaemia (tall ‘peaked’ T waves)[53][54][55]

Investigations to consider

urinalysis

Test
Result
Test

Look for signs of a urinary tract infection.

There is usually no significant ketonuria (ketones <2+). Use the result to help distinguish HHS from DKA.[6]

Result

usually no significant ketonuria (ketones <2+); leukocytes and nitrites in the presence of infection[6][60]

cardiac enzymes

Test
Result
Test

Order troponin T or I if you suspect myocardial infarction as a precipitant.[56]

Result

elevated with myocardial infarction

chest x-ray

Test
Result
Test

Order if the patient has reduced oxygen saturations.

Signs of pulmonary oedema are pleural effusions, interstitial and alveolar oedema, prominent superior vena cava, Kerley B lines, and dilated upper lobe blood vessels.[57]

Consolidation occurs in pneumonia.

Result

signs of pulmonary oedema are pleural effusions, interstitial and alveolar oedema, prominent superior vena cava, Kerley B lines, and dilated upper lobe blood vessels;[57] consolidation occurs in pneumonia

liver function tests

Test
Result
Test

Use to screen for an underlying hepatic precipitant of HHS. Abnormal LFTs indicate underlying liver disease (e.g., non-alcoholic fatty liver disease or congestive heart failure).[58]

Result

elevated with liver disease

C-reactive protein

Test
Result
Test

Order if you suspect infection as a precipitant.

Result

elevated if infection present

blood, urine, and sputum cultures

Test
Result
Test

Order if there are signs of infection.

The most common causes are pneumonia and urinary tract infection.[3]​​[9]​​[17]​

Result

positive if infection present

Use of this content is subject to our disclaimer