Just How Common Is Readmission for Diabetic Ketoacidosis?

Readmission isn't uncommon for patients with type 1 diabetes, according to a new study.

In a review of over 90,000 hospitalizations involving an episode of diabetic ketoacidosis (DKA), about 20.2% of adults were readmitted to the hospital within 30 days, Hafeez Shaka, MD, of Cook County Hospital in Chicago, reported at the Endocrine Society's virtual ENDO 2021 meeting.

And this fifth of patients with type 1 diabetes who were readmitted within 30 days had a more than double higher mortality rate while rehospitalized (risk ratio 2.06, 95% CI 1.74-2.43, P<0.001), the study showed.

This finding reflects poorly on the glucose control of these patients, Shaka explained during a press conference, calling the high readmission rate "quite alarming."

"And this [data] is only looking at the window of 30 days," he added, noting that readmission rates for this patient population is likely much higher if the window is extended beyond a month.

Readmission for DKA or similar hyperglycemic emergency was also associated with an average 1-day longer length of hospital stay compared with the index admission (95% CI 0.9-1.2, P<0.001), he reported.

Also compared with the index admission, readmission for DKA was tied to an average of $8,217 (95% CI $6,940-$9,492, P<0.001) more healthcare utilization costs.

Many factors were also predicative of certain patients being readmitted within the month for DKA, Shaka noted. This included being a female patient (HR 1.14, 95% CI 1.74-1.20, P<0.001), and not surprisingly being previously discharged against medical advice (HR 1.54, 95% CI 1.43-1.66, P<0.001).

Having certain comorbidities were also predictors of readmission for DKA, including hypertension (HR 1.28, 95% CI 1.20-1.35), chronic kidney disease (HR 1.13, 95% CI 1.04-1.22), and anemia (HR 1.42, 95% CI 1.34-1.51).

Surprisingly however, these type 1 diabetes patients with comorbid hyperlipidemia (HR 0.92, 95% CI 0.87-0.98) or obesity (HR 0.70, 95% CI 0.62-0.79) had a lower risk of being readmitted for DKA.

Shaka explained that in this population of patients with type 1 diabetes, obesity likely served as a surrogate marker for greater insulin compliance.

Additional efforts should be placed on identifying more predictors of DKA readmission, as well as on proper discharge planning in order to decrease this burden of readmission, he stated.

The analysis drew upon patient data from the National Readmission Database, including adult hospitalizations between January to November 2017.

A total of 91,625 patients age 18 and older with type 1 diabetes principally admitted for DKA were included in the analysis, and the majority were then admitted to intensive care. Of these patients, 91,401 were discharged alive. Patients were excluded if they were electively or traumatically admitted.

Study limitations, Shaka said, included a lack of data on whether or not these patients were using multiple daily injections or continuous glucose monitoring plus insulin pumps prior to hospitalization. This was due to the fact that the database included only inpatient information, Shaka explained to MedPage Today.

But as for how these patients' blood glucose was monitored in the hospital, he noted that although The Endocrine Society has its own guidelines on best management practices of hyperglycemia in non-critical care hospitalized patients, most hospitals have their own protocols.

Comment