Implementation of fully closed-loop insulin delivery for inpatients with diabetes: Real-world outcomes

Aims: Fully closed-loop insulin delivery has been shown in clinical trials to be safe and improve glucose control compared with standard insulin therapy in the inpatient setting. We investigated the feasibility of implementing the approved CamAPS HX fully closedloop system in a hospital setting. Methods: This implementation project was conducted in a large teaching hospital in Cambridge, UK. Healthcare professional training was multimodal including face-to-face workshops, online learning modules and supported by standard operating procedures. Set-up and maintenance of closed-loop devices were undertaken by the inpatient diabetes team. Selection of suitable patients was multidisciplinary and prioritised those with more challenging diabetes management. Demographic and clinical data were collected from electronic health records and diabetes data management platforms. Results: In the 12 months since the closed-loop system was implemented, 32 inpatients (mean +/- SD age 61 +/- 16 years, 8 females, 24 males) used closed-loop insulin delivery during their admission, across medical and surgical wards in the hospital with a total of 555 days of closed-loop glucose control (median [IQR]: 14 [6, 22] days per inpatient). The time spent in target glucose range 3.9-10.0 mmol/L was 53.3 +/- 18.3%. Mean glucose was 10.7 +/- 1.9 mmol/L with 46.0 +/- 18.2% of time spent with glucose >10.0 mmol/L. Time spent with sensor glucose below 3.9 mmol/L was low (median [IQR]: 0.38 [0.00, 0.85]). There were no episodes of severe hypoglycaemia or diabetic ketoacidosis during closed-loop use. Conclusions: We have demonstrated that the fully closed-loop system can be safely and effectively implemented by a diabetes outreach team in complex medical and surgical inpatients with challenging glycaemic control.

Automated summary

The study investigated the feasibility of implementing a fully closed-loop insulin delivery system in a hospital setting and found that it can be safely and effectively implemented in complex medical and surgical inpatients with challenging glycaemic control.