Comment on Hirsch et al. Results From a Randomized Trial of Intensive Glucose Management Using CGM Versus Usual Care in Hospitalized Adults With Type 2 Diabetes: The TIGHT Study. Diabetes Care 2025;48:118–124 Free

We commend Hirsch et al. (1) on their well-designed and clinically pertinent Time in Glucose Hospital Target (TIGHT) study, comparing intensive glucose management using adjunctive continuous glucose monitoring (CGM) with standard care using intermittent point-of-care blood glucose testing. While the overall results did not demonstrate significant differences in mean glucose between groups, they nonetheless highlight important learnings for the potential future application of CGM in hospital-based diabetes care.

While outpatient CGM studies have demonstrated glycemic benefit in a broad range of populations, they have the advantages of long study duration, clinical stability, and predictable daily routine. Glycemic benefit is achieved via medication adjustments by health care providers as well as behavior change by people living with diabetes. In contrast, the challenges of acute hospital-based CGM studies include the limited duration of study participation (days rather than months) as well as marked glycemic variability in response to rapid, unpredictable changes in clinical status. Previous studies of CGM use in the hospital have demonstrated hypoglycemia reduction in patients at high risk of hypoglycemia (2) as well as hyperglycemia reduction in patients with recurrent hyperglycemia (3). Thus, the primary benefit of CGM use in the hospital may be in reducing glycemic extremes in select high-risk patients rather than being applied to broad patient cohorts with the intent of achieving tight glycemia.

In the TIGHT study, insulin dosages were, on average, no different between groups. Indeed, they were only clearly higher in the intensive group on days 8 and 9, when fewer than one-third of the original participants remained in the study. This suggests that local insulin intensification protocols may have been insufficiently responsive to derive value from real-time CGM data, or clinician concerns for iatrogenic hypoglycemia precluded intensification. Real-time alerts (2), protocolized insulin dosing algorithms, and use of CGM trend arrows may allow the full potential of CGM use in the hospital to be realized. Beyond glycemia, in-hospital CGM use likely improves both patient experience, through reducing requirements for finger-stick testing, and nursing workflows.

Finally, despite the TIGHT study’s overall negative result, the intensive group experienced lower mean glucose during CGM use compared with prerandomization (170 vs. 186 mg/dL), while no change occurred in the standard group (175 vs. 175 mg/dL). Although this magnitude of glucose reduction may appear small, over an entire inpatient stay such glucose reductions are sufficient to reduce rates of hospital-acquired infection (4,5).

Overall, the TIGHT study’s intensive glucose target of 90–130 mg/dL appears challenging with the current state of subcutaneously administered multidose insulin regimens for the complex and dynamic in-hospital terrain. Hence, adding CGM to existing standard multidose insulin management approaches appears to be insufficient to improve glycemia. Further studies are needed to identify which inpatient populations are likely to benefit most from CGM use and how that can best be achieved. Such studies should target patients at the highest risk of adverse glycemia and use all aspects of CGM data (including glucose alerts and trend arrows) for clinical decision-making to mitigate the risk of glycemic extremes.