SHORT-TERM CORTICOSTEROID USE AND HYPERGLYCEMIA IN HOSPITALIZED PATIENTS

Program: Abstracts - Orals, Featured Poster Presentations, and Posters
Session: SUN 807-838-Diabetes - Diagnosis, Complications & Outcomes
Clinical
Sunday, June 16, 2013: 1:45 PM-3:45 PM
Expo Halls ABC (Moscone Center)

Poster Board SUN-819
Yu Kuang Lai1, Ha Nguyen*2, Davi sa Leitao3, Arthur Chernoff4 and Sherry Pomerantz5
1Albert Einstein Medical Center, philadelphia, 2Albert Einstein Medical Center, Philadelphia, PA, 3Albert Einstein Medical Center, philadlephia, PA, 4Albert Einstein Med Ctr, Rydal, PA, 5Albert Einstein Medical Center
Introduction

Hyperglycemia in non-diabetic patients receiving corticosteroids is well documented but there is little known about its occurrence or its pre-disposing risk factors in hospitalized patient.

Method

A retrospective study was conducted of hospitalized non-diabetics receiving corticosteroids at a tertiary care medical center. Records of patients who received corticosteroids during inpatient hospitalization from May 2010 to January 2011 were reviewed. Exclusion criteria included pre-existing diabetes or glucocorticoid use prior to admission. Steroid-induced diabetes mellitus was diagnosed if 2 fasting glucose readings were ≥126 mg/dl or a random glucose reading ≥ 200 mg/dl at any time during the hospital stay. Glucose intolerance was diagnosed when fasting glucose was ≥ 100mg/dl but < 126mg/dl. Patients without glucose determinations in the above ranges were designated as normoglycemic. Variables such as body mass index, age, race, co-morbidities, medications, steroid dose and route of administration were compared between the hyperglycemic group (glucose intolerance and steroid-induced diabetes) and normoglycemic group. Descriptive statistics were used to summarize the data.

Results

There were 250 eligible subjects. One hundred eighty four patients (73.6 %) developed hyperglycemia. Variables such as sex, race, hypertension, chronic kidney disease, diuretic use , inhaled steroid use and total dose of corticosteroids were not significant risk factors for hyperglycemia (p-value=0.67, 0.31, 0.885, 0.799, 0.879, 0.772, and 0.356, respectively). Duration of steroid therapy and its administration by both the intravenous and oral routes were found to be associated with hyperglycemia (p-value= <0.001, and 0.019 respectively). The use of beta blocker had protective effect on the development of hyperglycemia (p-value=0.028). Steroid induced diabetes was diagnosed in 94 patients (38%). Their mean glucose reading was 235mg/dl (SD 74.41 mg/dl). Only 46.8% of the patients in the diabetic group received ongoing glucose monitoring and intervention.

Conclusion

 In hospital use of corticosteroids is associated with a high occurrence of hyperglycemia. More than a third of patients fulfilled criteria for diabetes; only 46.8% receiving glucose monitoring and intervention. In viewing the negative impact of poor glucose control on hospitalized patient, clinicians should consider glucose monitoring and early taper of corticosteroid when appropriate.

Nothing to Disclose: YKL, HN, DS, AC, SP

*Please take note of The Endocrine Society's News Embargo Policy at http://www.endo-society.org/endo2013/media.cfm