CHICAGO – Treatment with antipsychotic agents in acute psychiatric inpatient units is not associated with increased blood glucose levels, according to a multicenter study.
"This finding is contrary to what many of us may have previously considered," Dr. Dawn D. Smiley observed in presenting the study findings at the annual scientific sessions of the American Diabetes Association.
Dr. Dawn Smiley
A well-recognized association exists between the use of antipsychotic agents and an increased prevalence of diabetes and hyperglycemia, although it remains unclear whether the relationship is causal. The association is especially strong for the atypical or second-generation antipsychotics, which account for more than 90% of all antipsychotic use and are among the top-selling prescription drugs in the United States. Indeed, the Food and Drug Administration has for a decade required the product labeling for all atypical antipsychotics to include a warning about the risks of hyperglycemia and diabetes.
But most data on antipsychotic-associated diabetes come from long-term studies, which show that diabetes, when it occurs, most often does so within the first 6 months on the medication. Little is known about the acute impact of antipsychotic agents on blood glucose levels and clinical outcomes in patients admitted to acute psychiatric units for stays averaging a few weeks. This information gap provided the rationale for Dr. Smiley’s study.
She presented a chart review of 1,212 patients admitted to three acute inpatient psychiatric units in inner-city Atlanta during 2008. The mean age of the patients was 57 years, they had a mean body mass index of 27 kg/m2, and the mean daily blood glucose concentration was 122 mg/dL during their stay. Twenty-one percent of the patients had a history of diabetes, and 6% had new hyperglycemia at admission. The most common primary diagnoses on admission were suicidal ideation in 28% of patients, a primary psychotic disorder in 27%, and depression in 16%.
Thirty-seven percent of patients had no prior experience with antipsychotic agents but were newly started on an antipsychotic during their acute inpatient stay. An additional 38% had been on antipsychotic therapy prior to admission and continued on it while hospitalized. Twenty-five percent of patients had never been treated with antipsychotic agents and were not during their hospitalization. Fifty-two percent of patients were treated with atypical antipsychotic agents during their inpatient stay, 17% received both atypical and typical antipsychotics, and 6% got typical antipsychotics only, explained Dr. Smiley, an endocrinologist at Emory University, Atlanta.
She sought answers to three research questions:
• Does prior exposure to antipsychotic therapy affect inpatient blood glucose levels or clinical outcome measures? Mean psychiatric inpatient length of stay was significantly shorter for patients never exposed to antipsychotics, at 9.5 days, compared with 14.2 days in patients who continued on their previous antipsychotic regimen and 13.7 days in antipsychotic-naive patients started on an antipsychotic for the first time. The most likely explanation for the longer stays in patients on antipsychotic therapy was the need to monitor them for side effects, according to Dr. Smiley.
Importantly, mean daily blood glucose concentrations during psychiatric hospitalization did not differ between patients never exposed to antipsychotic agents, those started on an antipsychotic for the first time during their hospitalization, and patients continued on their previous antipsychotics.
• Does the type of antipsychotic agent used during the acute inpatient admission affect blood glucose levels or clinical outcomes?
Mean daily blood glucose levels were similar regardless of whether patients were on atypical antipsychotics, typical antipsychotics, or both. The mean hemoglobin A1c level was 7.3% in all three groups of antipsychotic-treated patients. However, the rate of medical complications during psychiatric hospitalization was significantly higher among patients on atypical agents, at 23%, compared with 17% in patients on typical antipsychotics and 13% in those on both.
Three percent of patients on atypical antipsychotics-only required transfer to a medical unit for treatment of a complication that arose during psychiatric hospitalization. That was also true for 3% of patients on combined atypical/typical antipsychotic therapy. In contrast, none of the relatively few patients on typical antipsychotics-only required transfer to a medical unit. It’s worth noting that all atypical antipsychotics are not alike in terms of the strength of their association with diabetes. Most antipsychotic-treated patients in this study got olanzapine (Zyprexa) or clozapine (Clozaril). Newer, less diabetogenic atypical agents were less frequently used in this largely indigent population.
• Does a history of diabetes or new hyperglycemia at admission affect outcome measures? Not unexpectedly, patients with a history of diabetes prior to admission had a significantly higher rate of medical complications during their psychiatric inpatient stay: a 28% rate, compared with 22% in patients with no history of diabetes but with hyperglycemia at admission and 20% in patients with neither metabolic abnormality. The most common complication in patients with a history of diabetes was urinary tract infection.