Hyperglycemic crises in adults: A look at the 2024 consensus report

Diagnosis

The diagnosis of DKA requires the presence of 3 criteria: (1) diabetes or hyperglycemia (D criterion), with a glucose level of 200 mg/dL or greater or a prior history of diabetes; (2) ketosis (K criterion), with a beta-hydroxybutyrate level of 3.0 mmol/L or greater or urine ketones at 2+ or higher; and (3) metabolic acidosis (A criterion), with pH less than 7.3, a bicarbonate concentration less than 18 mmol/L, or both.

The diagnosis of HHS requires the presence of 4 criteria: (1) hyperglycemia, with a plasma glucose level of 600 mg/dL or greater; (2) hyperosmolarity, with a calculated effective serum osmolality greater than 300 mOsm/kg or total serum osmolality greater than 320 mOsm/kg; (3) absence of significant ketonemia, with beta-hydroxybutyrate less than 3.0 mmol/L or a urine ketone strip of 2+ or lower; (4) absence of acidosis, with pH of 7.3 or greater and bicarbonate concentration of 15 mmol/L or greater.

Direct measurement of beta-hydroxybutyrate is strongly recommended for diagnosing DKA and monitoring treatment, using either serum samples in a central laboratory or capillary blood with point-of-care testing devices. Although both are acceptable methods with comparable precision, point-of-care testing offers easier testing and quicker results, potentially reducing admission duration and DKA recovery time. If beta-hydroxybutyrate determination is not available, a urine ketone strip of 2+ or higher will meet this criterion.

Anion gap is not recommended as a first-line diagnostic criterion, but it may have some value if ketone measurement is unavailable.

Management

Categorizing DKA severity helps guide decisions on the required level of care:

• Individuals with mild DKA (beta-hydroxybutyrate ≤ 6 mmol/L, pH > 7.25, bicarbonate ≥ 15 mmol/L, normal mental status) can be managed in a regular or observation nursing unit

• For those with moderate DKA (beta-hydroxybutyrate ≤ 6 mmol/L, pH 7.0–7.25, bicarbonate 10 to < 15 mmol/L, normal or drowsy mental status), management in a step-down or intermediate care unit is suggested

• Those with severe DKA (beta-hydroxybutyrate > 6 mmol/L, pH < 7.0, bicarbonate < 10 mmol/L, stupor or coma), HHS, or a concomitant critical illness should be managed in an intensive care unit.

DKA and HHS management involves administering intravenous fluids, insulin, and electrolytes, along with treating the precipitating cause. During treatment of DKA, capillary blood glucose should be checked every 1 to 2 hours, and electrolytes, phosphate, creatinine, beta-hydroxybutyrate, and venous pH should be checked every 4 hours until DKA resolution. For HHS, blood glucose, creatinine, electrolytes, and serum osmolality should be measured every 4 hours.

Patients without cardiac or renal compromise should receive isotonic saline or balanced crystalloid solutions at 500 to 1,000 mL per hour for the first 2 to 4 hours. Once intravascular volume is restored, subsequent fluid replacement depends on hemodynamics, fluid balance, and sodium concentration.

Dextrose 5% to 10% should be added once glucose is less than 250 mg/dL to prevent hypoglycemia and permit insulin administration to continue until ketonemia resolves (Table 1).⁵ For HHS, glucose reduction should not exceed 90 to 120 mg/dL per hour to avoid cerebral edema; sodium decline should not exceed 10 mmol/L in 24 hours, and osmolality should fall no more than 3.0 to 8.0 mOsm/kg per hour to minimize neurologic risks. Smaller fluid boluses (eg, 250 mL) should be considered in older adults and individuals with heart or kidney failure.

Insulin therapy for severe DKA should begin as soon as possible, either through a fixed-rate intravenous insulin infusion started at 0.1 units/kg per hour or by a nurse-driven insulin infusion protocol with a variable rate. Insulin should be adjusted to maintain glucose levels around 200 mg/dL and continued until ketoacidosis resolves (Table 1).

Most individuals with uncomplicated mild or moderate DKA can be treated with subcutaneous rapid-acting insulin analogues every 1 to 2 hours, with close nursing supervision.

For HHS, a fixed-rate intravenous insulin infusion should be started at 0.05 units/kg per hour. If there are mixed features (hyperosmolality with significant ketonemia or acidosis), the condition should be treated as DKA and a fixed-rate intravenous insulin infusion should be started at 0.1 units/kg per hour.

For patients already taking basal insulin at the time of hospitalization for DKA or HHS, basal insulin can be continued at the usual dose and adjusted as needed during hospitalization, in addition to the continuous intravenous insulin infusion. This may reduce rebound hyperglycemia and prevent recurring DKA.

To transition from intravenous to subcutaneous insulin, an estimation of the total daily insulin requirement is needed, considering hypoglycemia risk and anticipated nutritional intake. Estimations can be based on weight (estimating a total daily dose of 0.3–0.6 units/kg per day), preadmission insulin dose, or in-hospital insulin requirements. A basal-bolus regimen is recommended, starting basal insulin at least 1 to 2 hours before stopping the insulin infusion.

Potassium should be measured at baseline, 2 hours after starting insulin, and every 4 hours thereafter until resolution of DKA. Potassium replacement should start after serum levels fall below 5.0 mmol/L to maintain levels between 4 and 5 mmol/L. If potassium levels are lower than 3.5 mmol/L at presentation, replacement should begin at a rate of 10 mmol per hour, and insulin therapy should be postponed until a potassium level higher than 3.5 mmol/L is reached.

Routine bicarbonate and phosphate administration is not recommended. Bicarbonate should be considered only in severe acidosis (pH < 7.0), and phosphate replacement should be considered if levels are under 1.0 mmol/L, particularly if muscle weakness or cardiac or respiratory impairment is present.

Patient education

Before discharge, all patients admitted with DKA or HHS should receive education focused on both the current event and overall diabetes management, including injection techniques, glucose monitoring, and urine or blood ketone testing.

Diagnosis of DKA

D criterion. Reducing the glucose cutoff to 200 mg/dL or greater is reasonable, as this level is typically diagnostic of diabetes in the general population, and there is no justification for it to be different in the context of DKA (Table 2). Adding a history of diabetes as an alternative to glucose values (and irrespective of them) allows for the inclusion of patients with euglycemic DKA. Mentioned in the 2009 consensus report, euglycemic DKA has become more common with the advent of sodium-glucose cotransporter 2 inhibitors. Use of these agents has been shown to increase the risk of euglycemic DKA in individuals with type 2 diabetes and in those with type 1 diabetes using them off-label.^6,7

K criterion. This is a major update, as it involves the key diagnostic feature of DKA. The recommendation to measure beta-hydroxybutyrate is largely based on the pathophysiology of ketosis in DKA, in which the ratio of beta-hydroxybutyrate to acetoacetate rises from a physiologic 1:1 to up to 10:1. Then, during resolution, beta-hydroxybutyrate is oxidized to acetoacetate, causing its levels to decrease long before those of acetoacetate. Since the nitroprusside reaction (used to measure ketones semiquantitatively in urine and blood) measures only acetoacetate, it can underestimate the degree of ketosis at presentation but overestimate it during resolution.^8,9 Both tests have similar sensitivity, but beta-hydroxybutyrate is more specific for DKA diagnosis. Additionally, drugs can interfere with urine ketone testing¹⁰; in particular, false positives can be seen with commonly used medications like captopril and valproic acid. Therefore, the preferred method for assessing ketosis, both at diagnosis and during treatment, is quantitative assessment of beta-hydroxybutyrate, when available.

A criterion. The 2024 consensus removed the anion gap criterion to better account for the various factors influencing acid-base status in individuals with DKA. While an increased anion gap indicates a net gain in ketoacid anions, the accumulation of ketoacids in the extracellular fluid results in bicarbonate loss, which may not be immediately apparent due to extracellular volume contraction. Hyperglycemia-induced diuresis and natriuresis cause marked volume contraction, affecting the determination of the severity of metabolic acidemia, as standard calculations are based on concentrations rather than total content.¹⁰ There is also associated hyperventilation due to acidosis, all of which contribute to the frequent occurrence of mixed acid-base disorders in individuals presenting with DKA.

Severity of DKA

Quantitative beta-hydroxybutyrate is now recommended for assessing DKA severity, with the introduction of quantitative cutoffs for mild and moderate (3–6 mmol/L) and severe DKA (> 6 mmol/L). Anion gap is no longer a severity criterion for DKA. The new consensus suggests assigning the level of hospital care based on DKA severity at presentation, including the possibility of managing mild DKA in the general ward. Not all criteria must be met to classify a patient as mild, moderate, or severe; clinical judgment and resource availability should ultimately determine severity and guide decisions on admission and level of care.

Diagnosis of HHS

The 2024 consensus report lowers the effective serum osmolality cutoff for diagnosing HHS and introduces total serum osmolality as a new criterion (Table 3). Including urea (ie, using total serum osmolality) in the diagnostic criteria, despite it not being an effective osmolyte, accounts for the severe dehydration commonly seen in these patients.¹¹

Mental status impairment is no longer a diagnostic criterion. Although past studies linked osmolality with mental status, many individuals, though very ill, do not necessarily have mental impairment, so this is no longer a requirement for diagnosis of HHS.^4,11

Quantitative cutoffs were added for the allowed ketonemia in the diagnosis of HHS, and the bicarbonate level was lowered from 18 to 15 mmol/L to allow for a degree of acidosis that can occur due to insulinopenia.

Treatment of DKA and HHS

One key change (Table 4) involves the choice of fluids for initial resuscitation, now suggesting balanced crystalloids (when available) because their use is associated with faster recovery, less hyperchloremic metabolic acidosis, and shorter hospital stay.^12,13 Additionally, suggested fluid replacement speed and time are more conservative in the new consensus report, likely due to an older, more comorbid patient profile.

The 2009 consensus report introduced the concept of managing mild DKA with subcutaneous insulin, but the strength of the evidence now supports a formal recommendation of using it as an alternative to intravenous infusion in mild and uncomplicated moderate DKA,^14,15 thus avoiding the need for an intensive care unit admission.

Resolution criteria for DKA and HHS are outlined in Table 1. Criteria for DKA were updated to incorporate quantitative beta-hydroxybutyrate, and HHS criteria were established for the first time. The report also offers guidance on treating DKA and HHS in special populations, including older adults, those on sodium-glucose cotransporter 2 inhibitors, patients undergoing dialysis, pregnant patients, and those with COVID-19.