Effects of Epidural Steroid Injections on Blood Glucose Levels in Patients With Diabetes Mellitus

Abstract

A prospective cohort study. To evaluate the effects of epidural steroid injections (ESIs) on blood glucose levels in patients with diabetes mellitus. ESIs are commonly used in the treatment of multiple spinal disorders. Corticosteroid injections have been evaluated in the total joints and hand literature showing systemic effects to diabetics. Diabetic patients who were scheduled for an ESI were given an opportunity to enroll in our IRB-approved study. We collected the patient's most recent hemoglobin A(1c) (hA(1c)) and then asked them to track their blood glucose numbers at least twice per day for 2 weeks prior to and after their ESIs. We noted a statistically significant increase in blood glucose levels in diabetic patients (n = 30) after ESI. The mean blood glucose level prior to ESI was 160.18 ± 47.46, and, after ESI, it was 286.13 ± 111.11. This represents an average 125.96 ± 100.97 increase in blood glucose levels after injection. Using a nonlinear mixed effect model, the estimated half-life of this increase was 1.06 days (95% CI 0.80, 1.58), meaning that the patients were back within their normal standard deviation mean glucose levels within 2 days of injection. There was no association between observed glucose level change and preinjection hA(1c) levels or age (Spearman = 0.0326 and -0.1091 separately), indicating that there is no correlation between preinjection hA(1c) levels and systemic response to ESI. ESIs were noted to cause a significant increase in the blood glucose levels in diabetics. There was no correlation between preinjection diabetic control, represented by hA(1c) levels, and postinjection response. Diabetics who are candidates for ESI should be counseled that a blood glucose increase may be apparent post intervention, but effects should not last longer than approximately 2 days.