Hyperglycemia in Hospitalized Patients Receiving Corticosteroid Premedication before the Administration of Radiologic Contrast Medium

Authors: Matthew S. Davenport, MDaCorresponding Author Informationemail address, Richard H. Cohan, MDb, Shokoufeh Khalatbari, MSb, James Myles, PhDb, Elaine M. Caoili, MDb, James H. Ellis, MDb

Rationale and Objectives

To assess the effect of short-term corticosteroid premedication on serum glucose in hospitalized patients.

Materials and Methods

Serum glucose in adult inpatients receiving corticosteroid premedication before a radiology study was compared to serum glucose in nonpremedicated controls between May 1, 2008, and September 12, 2009. Patients were categorized by type of nonionic contrast medium (intravenous [IV] or none) and route of corticosteroid premedication (oral, IV, none). Diabetes mellitus (types I and II) was evaluated as an independent risk factor. Patients without glucose estimations before and after premedication were excluded. Results were analyzed with analysis of variance and a stepwise linear regression analysis.


There were 390 corticosteroid premedication episodes in 390 patients; 873 examinations in 844 patients served as controls. Cohorts receiving corticosteroid premedication experienced a brief (24–48 hour) increase in mean maximum postbaseline serum glucose (IV, +81 mg/dL; oral, +70 mg/dL) that was greater than the increase in nonpremedicated controls (+46 mg/dL). Type I (+144 mg/dL) and type II (+108 mg/dL) diabetics had a greater elevation than nondiabetics (+34 mg/dL). Both corticosteroid premedication (IV, P = .02; oral, P = .01) and diabetes mellitus (type I, P = .0002; type II, P < .0001) were significant independent risk factors of hyperglycemia. The use of nonionic intravascular contrast medium was not (P = .7). There was no significant difference between IV and oral premedication (P = .6), or type I and type II diabetes mellitus (P = .8).


Diabetes mellitus (type I and type II) and corticosteroid premedication (oral and IV) are significant independent risk factors for the development of brief hyperglycemia near the time of inpatient radiology studies.