PracticeUpdate Diabetes June 2019

EDITOR’S PICKS 8

Early Intervention for Diabetes Inpatients Decreases Hyperglycemia and Hospital- Acquired Infections Diabetes Care Take-home message • This study of 1002 consecutive adult inpatients with diabetes or new hypergly- cemia was designed to assess whether early electronic identification and bedside management improves glycemic control. A cluster-randomized trial was conducted on eight wards, and individuals were randomized to a 12-week active period of early intervention or usual care. Significantly more patients who received early intervention received specialist diabetes management and new insulin treatment compared with those who received usual care. Furthermore, the number of adverse glycemic days, overt hyperglycemia, and hospital-acquired infections significantly decreased in patients who received early intervention. • These results demonstrate that early identification and management of inpatients with diabetes contributes to a decrease in hyperglycemia and hospital-acquired infections.

COMMENT By Mary Korytkowski MD and Esra Karslioglu-French MD I t is well-known that poor inpatient glycemic control is asso- ciated with adverse events. This is the first randomized trial to investigate the effect of early intervention by an inpatient diabetes team (IDT) on blood glucose (BG) control measured as adverse glucose days (AGDs; defined as patient days with BG values <72 or >270 mg/dL) and adverse clinical outcomes (hospital-acquired infections, AKI, MI, unplanned critical care admission, and mortality). IDT members received training in inpa- tient glycemic management by a senior diabetologist prior to the intervention. Patients with diabetes or new hyperglycemia hospitalized on one of eight hospital units (four medical, four surgical) were identified using electronic surveillance of capillary BG measures from net- worked meters. Cluster randomization was performed according to hospital unit. Patients (n=270) admitted to units in the control arm received usual diabetes care with consultation by the IDT only upon request of the primary team. Patients (n=291) admit- ted to units in the intervention arm were seen and managed by the IDT within 24 hours of hospital admission. Compared with AGDs measured on the same units during a 10-week baseline period, patients hospitalized on the interven- tion units decreased AGDs by 24% at the cluster level and 28% at the individual level (both, P < .001) without an increase in hypogly- cemia. There was no difference in AGDs in the control group. The reduction in AGDs was associated with a significant reduction in hospital-acquired infections, but not other adverse outcomes. These favorable results following early IDT intervention occurred despite similar mean BG in the control and intervention arms, suggesting that focusing on preventing extreme glycemic

excursions may be more important than targeting specific BG targets in hospitalized patients, where testing is often affected by timing of meals, glucose-containing IV fluids, or timing of insu- lin dosing. These results also call into question the practice of providing IDT consultation only by request, which often occurs when there are multiple AGDs resulting in delays in achieving glycemic stability. A barrier to changing to proactive glycemic interventions is the need for the IDT to justify their existence with adequate numbers of billing events, which, as Kyi et al demon- strate, may not be the best way of ensuring optimal outcomes for hospitalized patients with diabetes. events …may not be the best way of ensuring optimal outcomes for hospitalized patients with diabetes. " " A barrier to changing to proactive glycemic interventions is the need for the IDT to justify their existence with adequate numbers of billing

Dr. Korytkowski is Professor of Medicine at the University of Pittsburgh School of Medicine in Pittsburgh, Pennsylvania.

Dr. Karslioglu-French is Clinical Assistant Professor of Medicine and Clinical Lead of the Inpatient Diabetes Program at the University of Pittsburgh in Pittsburgh, Pennsylvania.

PRACTICEUPDATE DIABETES