Skip to main content

02-27-2020 | Ketoacidosis | News

Hyperosmolarity warns of poor outcomes in DKA patients

Author: Eleanor McDermid


medwireNews: A large study backs hyperosmolarity as a key risk stratification factor for patients admitted to hospital with diabetic ketoacidosis (DKA).

“Whether or not it is directly causal, this may have practical applications to improve risk stratification for illness severity,” say Sebastiaan Blank (Tamworth Base Hospital, New South Wales, Australia) and study co-authors.

The team looked at 17,379 intensive care unit admissions for DKA across Australia and New Zealand. People with hyperosmolarity (serum osmolarity ≥320 mosm/L), who comprised 19.6% of the population, were older on average than those without, at 51 versus 33 years, had more comorbidities, and were more severely ill at admission.

The database used did not record whether the patients had type 1 or type 2 diabetes, but the researchers note that prior insulin use was significantly less common in the hyperosmolar group, at 70% versus 75% for the other patients.

The analysis “confirms that adults with ketoacidosis have a good prognosis overall,” write Blank at team in Diabetic Medicine.

Hospital mortality was just 1%, the median length of stay 3.6 days, and 92% of patients were discharged direct to home.

However, the situation was significantly worse for patients with hyperosmolarity, who had a hospital mortality rate of 3.0% versus 0.7% for patients without this risk marker. The same was true of the need for mechanical ventilation (18 vs 4%) and for acute kidney injury (13 vs 4%), in addition to which their hospital stay was longer (median 6.4 vs 3.2 days) and they were less likely to be discharged home (84 vs 94%).

On dividing patients into quintiles of osmolarity, the team found a significantly increased mortality risk starting at the third quintile (300–319 mosm/L) and rising to a 9.69-fold increased risk for those in the fifth quintile (≥350 mosm/L) relative to those in the second (280–299 mosm/L). This was independent of confounders including serum bicarbonate, hospital type, and the ANZROD mortality risk score.

Further analysis showed that the increased mortality risk associated with hyperosmolarity was present irrespective of patients’ blood pH.

The researchers say that a prospective study is now needed “to more reliably record peak values and better define a threshold for increased risk.”

And they add: “In addition to the need to optimize risk stratification, a critical research area will also be clarifying which interventions truly improve outcomes in those at risk of increased mortality. In particular, the ideal fluid resuscitation strategy and rate of correction in ketoacidosis with hyperosmolarity needs to be determined and incorporated into clinical guidelines.”

medwireNews is an independent medical news service provided by Springer Healthcare. © 2020 Springer Healthcare part of the Springer Nature group

Diabet Med 2020; doi:10.1111/dme.14277