We evaluated the utility of real-time CGM for titrating intravenous (IV) insulin via a validated institutional computerized insulin infusion (CII) algorithm in the CV-ICU. We used a hybrid approach of combining CGM with periodic point-of-care blood glucose (POC-BG) tests to validate the continued accuracy of CGM. We also surveyed nurses on this care change. Dexcom G6 CGMs were applied to 61 post-surgical patients with hyperglycemia (34% with diabetes) receiving IV insulin. CGM values were validated with POC-BG every(Q) 1-2 h per the CII protocol. Once validated (i.e., within 20% of POC-BG values if BG ≥100 mg/dL or within 20 mg/dL if BG <100 mg/dl), sensor values were then used to titrate IV insulin doses per CII algorithm. POC-BG checks were then reduced to Q6h validation. Among 857 paired POC-BG and CGM values analyzed, the mean and median average relative difference between POC-BG and CGM values were 13.2% and 9.8%, respectively. 99.6% of paired CGM and POC-BG were in Zone A and B of the Clarke Error Grid (Figure). Thirty nurse respondents found CGM very or quite convenient (n=28; 93%) and favored it over POC-BG testing (n=28; 93%). This pilot study demonstrates that using CGM via a hybrid approach for CII titration protocol is feasible, has high accuracy, and higher nursing convenience.


L.Ang: None. Y.Qu: None. R.Freeman: None. N.H.Esfandiari: None. R.Busui: Board Member; American Diabetes Association, Consultant; Averitas Pharma, Inc., Lexicon Pharmaceuticals, Inc., Nevro Corp., Novo Nordisk, Roche Diagnostics, Procter & Gamble, Research Support; Novo Nordisk, Medtronic, National Institutes of Health. R.Gianchandani: None. F.Akanbi: None. L.F.Schroeder: None. Y.Lin: None. C.A.Degeorge: None. P.Arnold: None. S.Knotts: None. E.Dubois: None. N.Desbrough: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.