Inpatient Glycaemic Control is Poor in Patients Receiving Insulin Therapy

Abstract

Patients with diabetes mellitus (DM) are admitted to hospital more frequently than the general population and have longer inpatient stays. Tight inpatient glycaemic control improves outcome but is difficult to achieve, especially in those on insulin. We prospectively reviewed the capillary blood glucose (CBG) control of hospital inpatients receiving insulin to determine their level of control and to examine healthcare providers’ responses to dysglycaemia. Between March and June 2020, all patients on insulin admitted to Bantry General Hospital for more than 24 hrs were reviewed. CBG recordings were individually categorised as either hypoglycaemic (CBG < 4mmol/L), euglycaemic (CBG = 4–9.9mmol/L), mildly hyperglycaemic (CBG = 10-14mmol/L) or severely hyperglycaemic (CBG >14mmol/L). Each patient’s individual insulin dose was reviewed and compared with their concurrent CBG level and categorised as either appropriate or inappropriate. Other factors that were reviewed included whether a HbA1C was performed, whether a referral to the diabetic clinical nurse specialist was made and whether the patient was on glucocorticoid therapy.16 patients were enrolled, average age 77. 497 CBG checks were recorded over 108 patient days. Average length of stay was 6.6 days. All patients had type 2 DM. 10 patients were on basal insulin therapy, 4 were on basal-bolus insulin, 2 were on mixed insulin. CBG was checked sub-optimally (i.e. fewer than 4 times per day) on 35/108 patient days. 355/497 CBG checks were in the euglycaemic range; but only 3 patients were entirely euglycaemic throughout admission.17 hypoglycaemic events occurred in 5 patients. These events were spread unevenly with 1 patient having 10 hypoglycaemic events while the other 7 events were spread between the remaining 4 patients.125 severely hyperglycaemic events occurred in 13 patients. Again, these events were spread unevenly with over half (67/125) events occurring in only 3 patients. Of these 13 patients, 5 were on glucocorticoid therapy. All 5 patients that had hypoglycaemic events also had severely hyperglycaemic events. Of the hypoglycaemic events recorded, only 12/17 were treated appropriately (i.e. with 15-20g of fast-acting carbohydrate and a reduction in subsequent insulin dose). Of the hyperglycaemic events recorded, 77/125 were managed appropriately (i.e. with either a stat dose of bolus insulin being administered or an increase in subsequent insulin doses being made or both). 6 prescribing errors were recorded in total. HbA1c was checked in 5 patients. An inpatient referral was made by the medical team to the diabetes clinical nurse specialist for only 4 patients. Inpatient glycaemic control was suboptimal in the majority of patients. Dysglycaemia, especially hyperglycaemia, remains problematic. Insulin doses were not adjusted often enough or dramatically enough. Further doctor and nurse education is required in this area.