Implementing a Standardized Workflow for Early Detection of Steroid-Induced Hyperglycemia: A Quality Improvement Project

Background Steroid-induced hyperglycemia (SIH) is associated with an increased risk of non-relapse mortality and worsened outcomes in the stem cell transplant (SCT) population. Early detection and intervention of SIH may improve outcomes and reduce healthcare-associated costs. Objective The aim of this quality improvement (QI) initiative was to decrease the median time to identification of SIH and the initiation of treatment in SCT recipients on glucocorticoids (GCs) for graft-versus-host disease (GVHD). Study Design Patients diagnosed with acute or chronic GVHD and started on 0.5 mg/kg/day or more of prednisone equivalent (PE) steroids were requested to monitor their blood sugar 1-2 hours after their largest meal of the day (preferably lunch or dinner) for 14 days. A control group (retrospective chart review) with the same inclusion criteria was used for comparison. Time to identification of SIH was compared between the two groups, as well as time to treatment of hyperglycemia, and secondary outcomes of infections and hospitalizations were observed. Results Over 9 weeks, 19 patients enrolled in the QI study. The median age was 64 years. 53% were male and 89% were white. The control group consisted of 21 patients with a median age of 60 years. Sixty-two percent were male, and 76% were white. The median steroid dose was 1 mg/kg/day of PE steroids in the QI group and the control group (p=0.8100). Eighteen of the 19 patients (95%) had at least 1 blood sugar (BS) > 180 mg/dL and only 6 of 21 patients (29%) had at least 1 BS > 180 mg/dL (p<0.0001). The median time to a BS > 180 mg/dL was 1.5 days in the QI group and 7 days in the control group (p=0.0232). The median time to insulin was 2 days in the QI group and 10 days in the control group (p=0.0355). Conclusion Currently, there is no standardized workflow for monitoring SIH in the SCT population. While HbA1C levels are used to monitor diabetes in the general population, they are not accurate in the SCT population due to anemia and blood transfusion use. Alternatively, HbA1C is not helpful in monitoring SIH because it is a retrospective value. This study showed that daily postprandial blood sugar monitoring is superior for the earlier identification and treatment of SIH when compared to monitoring at routine clinic visits alone. Now that the gold standard for measuring diabetes has shifted to monitoring the “time-in-range”, future studies are needed to further investigate SIH in the post-transplant population by means of CGMs, and how the monitor impacts patient outcomes.