Background:
CGM is commonly used in the management of Type 1 diabetes. Experience in T2DM is limited.

Aim:
To determine the utility of CGM in T2DM.

Method:
6-day data from the Medtronic iPRO-Carelink CGM was audited from 37 consecutive subjects with poorly-controlled T2DM attending Blacktown Hospital between September 2015-April 2016. Subjects simultaneously self-monitored blood-glucose(SMBG) 4 times/day. Clinical and biochemical information was obtained from hospital records.

Results:
57% of subjects were male, mean age 64.7yrs. 80% had T2DM>10yrs. All but one were using insulin, average total daily dose 89U/day±59U. Insulin treatment: 69% basal-bolus, 22% pre-mixed. Additional oral agents: sulphonylureas 6%, metformin 61%, SGLT2-i 17%, DPPIV-i 6%.

There was no difference between mean(±SEM) glucose on CGM vs SMBG (10.1±0.42 vs 10.2±0.46 mmol/L, p=0.6). However mean lowest CGM-glucose level was significantly lower than SMBG(4.3±0.25 vs 5.1±0.29 mmol/L, p<0.01) and mean highest CGM-glucose was higher than SMBG(18.8±0.67 vs 17.3±0.73 mmol/L, p<0.01). In total, CGM detected 62 episodes of glucose<4.0mmol/L compared to 22 episodes from SMBG. Of the 40 excess episodes detected by CGM, 34 occurred in 4 patients. At least 1 episode of hypoglycaemia was detected in 40% of subjects using CGM compared to 31% using SMBG. No severe hypoglycaemia(glucose<2.5mmol/L) was detected by SMBG, but CGM revealed 5 episodes, all occurring in 3 subjects. 438 episodes of hyperglycaemia(glucose>10.0mmol/L) were recorded on CGM vs 407 by SMBG.

To date, 27 subjects have post-CGM management information recorded. CGM prompted a decrease in insulin dose in 9 subjects and an increase in 12. 1 further subject had redistribution of daily insulin dose and another switched from pre-mixed to basal-bolus. Changes to diet and activity were recommended in 11 subjects, based on CGM.

Discussion:
CGM detects unrecognised glycaemic excursions and has clinical utility in influencing both pharmacological and lifestyle management in T2DM.