Perioperative diabetes audit identifies inconsistent practice and suboptimal glycaemia — ASN Events

Perioperative diabetes audit identifies inconsistent practice and suboptimal glycaemia (#296)

Natalie Nanayakkara 1 , Peter G Colman 1 , Spiros Fourlanos 1
  1. Department of Diabetes and Endocrinology, Royal Melbourne Hospital, Melbourne, VIC, Australia

Background: Diabetes affects approximately 30% of hospitalised patients (1). Inpatient diabetes management is complex with suboptimal glucose control and medication errors being common (2). Perioperative diabetes management is complicated by reduced oral intake and increased metabolic demands. Adverse outcomes include life-threatening metabolic complications (diabetic ketoacidosis – DKA, Hyperosmolar nonketotic coma - HONK, hyper- and hypoglycaemia), infection (3), increased length of stay and increased morbidity and mortality (4).  

Aim: To assess perioperative diabetes management specifically in relation to diabetes medication errors and resultant glucose control.  

Methods: We retrospectively audited 104 consecutive inpatients with diabetes (mean age 69, 54% male, 95% type 2 diabetes, 42% insulin-requiring) who underwent procedures (majority orthopaedic, vascular or interventional cardiology) at the Royal Melbourne Hospital. We examined patients’ medical records (progress notes, glycaemic records and medication charts) in relation to the immediate preoperative diabetes management (12 hours pre and post procedure) including the adequacy of glycaemic monitoring (2 hourly preop and 4 hourly postop), oral hypoglycaemic medication and insulin management and frequency of adverse events (hypoglycaemia: capillary blood glucose level [BGL] <4mmol/l, hyperglycaemia:  BGL>11mmol/l).

Results: 26% of inpatients had written perioperative diabetes management plans. 80% of patients had oral hypoglycaemic medications appropriately withheld. Preoperative BGLs were monitored 2 hourly in 39% of inpatients and postoperative BGLs monitored 4 hourly in 38% of patients. Hypoglycaemia (BGL<4) was identified in 10% of patients; 55% of BGLs were within target (4–10 mmol/l) with a significant trend to worsening glycaemia in the later postoperative period (Figure 1). There were no episodes of DKA or HONK. A minority of patients (12%) received care from the Endocrinology team.

Conclusions: The perioperative management of inpatient diabetes did not always follow established best practice. This audit identifies perioperative care as a major issue; given the numbers of hospital inpatients with diabetes a comprehensive inpatient diabetes service accompanied by guidelines may promote consistency and improve outcomes in the perioperative management of diabetes. 

163-BGL.jpg

1 2 3 4 

  1. Fourlanos S et al. International Congress on Obesity, 2010.
  2. National Diabetes Inpatient Audit Study 2010, National Health Service, United Kingdom
  3. Golden SH et al. Diabetes Care 1999; 22:1408-14
  4. 4. Moghissi E et al. Diabetes Care 2009,32:1119-1131