Conference Coverage

Patients with higher HbA1c levels face greater risk for diabetic ketoacidosis



Patients with poor glycemic control, especially those with a hemoglobin A1c (HbA1c) level of more than 9%, face an increased risk for developing diabetic ketoacidosis (DKA), results from a registry study have demonstrated.

Dr. Carol Wysham an endocrinologist in Spokane, Wash.

Dr. Carol Wysham

The findings come from an analysis of the 2016-2017 Type 1 Diabetes Exchange Clinic Registry dataset that researchers, led by Carol Wysham, MD, presented at the annual scientific and clinical congress of the American Association of Clinical Endocrinologists.

“This study is unique in that we have stratified patients based on A1c, and identified factors and patient characteristics associated with a greater risk of DKA as glycemic control diminishes,” Dr. Wysham, an endocrinologist at MultiCare Rockwood Clinic Diabetes & Endocrinology Center in Spokane, Wash., said in advance of the meeting. “Multiple interrelated factors, such as lower levels of education and household income, relationship status, access to private health insurance, being younger, and smoking, all affect a patient’s ability to properly manage insulin dosing and are associated with a higher risk of DKA. Health care providers should continue to monitor and educate all patients on the risk factors associated with developing DKA, and be vigilant in patients with an A1c of more than 9%.”

Dr. Wysham and her colleagues conducted a cross-sectional analysis of 6,242 patients in the registry. They examined associations between patient characteristics and treatment patterns with the occurrence of a DKA event in three categories of HbA1c: 7% to less than 8%, 8% to less than 9%, and 9% or greater. The researchers used chi-square or Fisher exact tests to compare DKA and non-DKA groups for categorical variables and t tests to compare continuous variables.

Of the 6,242 patients, 43% had an HbA1c from 7% to less than 8% (cohort 1), 31% had an HbA1c of 8% to less than 9% (cohort 2), and 30% had an HbA1c of 9% or greater (cohort 3). In all, 269 patients reported a DKA event. In addition, 1.7% of those in cohort 1 had a DKA episode versus 2.3% of those in cohort 2 and 9.5% of those in cohort 3.

In patients in cohort 1, the researchers observed no significant associations between individual patient demographic, socioeconomic, or treatment patterns and DKA incidence. In patients in cohort 2, race, marital status, insurance coverage, and annual household income were significantly associated with DKA incidence (P less than .01). In patients in cohort 3, DKA incidence was significantly associated with the same patterns as those in cohort 2, with the addition of age, type 1 diabetes duration, sex, education level, body mass index, and insulin delivery method (P less than .01).

On adjusted multivariate analysis, the researchers observed no significant associations between the factors studied and DKA in patients in cohort 2. In patients in cohort 3, only household income, smoking status, body mass index, and insulin delivery method (injection) were associated with DKA.

Dr. Wysham said she was surprised to learn that the patient characteristics and socioeconomic factors associated with DKA in patients with an HbA1c of more than 9% start to become less significant risk factors as patients achieved better glycemic control.

“Also, in the past, insulin pump users tended to have higher rates of DKA, compared with patients taking multiple daily insulin injections,” she said. “That phenomenon no longer seems to apply, and in this dataset, patients with an HbA1c of more than 9% and who were taking multiple daily injections had significantly higher risk of DKA than did pump users. Insulin delivery method was not a contributing factor to DKA risk at all in patients with an HbA1c of less than 9%.”

Dr. Wysham acknowledged certain limitations of the analysis, including the fact that the registry data “are collected from patients treated at diabetes centers of excellence and may not reflect the population as a whole. Data could have been collected from medical records and/or self-reported questionnaires from patients. Self-reported data are subjective and are limited to the patient’s own recollections.”

Dr. Wysham disclosed that she has received honoraria for advising, consulting, and/or speaking from Abbott, AstraZeneca, Boehringer Ingelheim, Eli Lilly, Dexcom, Novo Nordisk, and Sanofi. She also disclosed having received research funding from Mylan and Novo Nordisk that went to her institution.

Next Article: