Re-Evaluating Inpatient Prandial Insulin Administration Timing: From Data to Clinical Practice

In the outpatient setting, evidence links postprandial hyperglycemia with diabetes complications. ¹ Physiological studies indicate that short-term hyperglycemia may cause lasting harm, including epigenetic changes, secretion of inflammatory factors, and production of reactive oxygen species. ^{2 –4} Additionally, short-term hyperglycemia may induce glucotoxicity, which inhibits insulin secretion. ⁵ Clinically we have observed that once glucotoxicity resolves following blood glucose control, the same insulin dose could lead to hypoglycemia. In the inpatient setting, increased glycemic variability has been independently associated with longer hospital stays and higher mortality. ⁶ As continuous glucose monitoring (CGM) technology is relatively new in inpatient settings and is vital for detecting postprandial hyperglycemia, further inpatient studies utilizing CGM are required to explore the association between postprandial hyperglycemia with clinical outcomes.

Alexanian et al. conducted a post hoc analysis of the In-Fi study, revealing that delayed administration of mealtime insulin, as well as evening snacking without insulin coverage, contributes to postprandial and overnight hyperglycemia in hospitalized inpatients. ⁷ These findings align with our anecdotal clinical observations. Notably, this study identified a high prevalence (43%) of delayed mealtime insulin administration, despite instructions for nursing staff to administer it prior to meals.

Over the past decade, there has been a notable shift in the hospital from premeal to post meal insulin administration. The primary motivation for this change is that unexpected nutritional interruptions with unadjusted insulin doses resulting in hypoglycemia. ^8,9 Due to this, we have seen that some hospitals have revised their policies to intentionally administer insulin to after meal completion. While we are not aware of how many hospitals have changed their protocols to post-meal insulin administration, there has been a cultural shift among nursing staff toward post-meal insulin dosing. Furthermore, a recent report indicated that changing hospital policies from premeal to post-meal insulin administration reduced the incidence of hypoglycemia. ¹⁰ It needs to be emphasized that even with protocols calling for premeal insulin dosing, in the hospital post-meal dosing happens due to many reasons (procedures interrupting the timing of insulin and food, changes in nurse to patient ratios, etc.).

After-meal glucose levels are not routinely measured in hospitals, leaving it uncertain whether post-meal insulin administration contributes to postprandial hyperglycemia under current inpatient practices. The introduction of CGM in hospitals has brought increased attention to this issue. In our recent multicenter interventional trial (TIGHT study), which aimed to achieve improved glucose control with CGM assistance, we observed markedly high postprandial glucose levels, consistent with those reported in this study. ¹¹ Of the six study sites in the The Time in Glucose Hospital Target (TIGHT) study, three implemented post-meal insulin administration policies, while the remaining three sites followed premeal insulin administration policies—although insulin was often administered after meals, which, as noted above, is common in the inpatient setting.

In the outpatient setting, post-meal insulin administration has generally been discouraged, as data—particularly in patients with type 1 diabetes—clearly show a deterioration in glucose control. ^{12 –15} Still, post-meal administration of ultra-rapid lispro has been shown to be noninferior to premeal insulin lispro. ¹⁶ However, even in this study, post-meal ultra-rapid lispro administration resulted in higher postprandial hyperglycemia compared with the same ultra-rapid lispro administered before eating. ¹⁶ These findings from outpatient studies align with the observations made in this inpatient study. ⁷

To address unexpected reduction or interruption in oral intake during premeal insulin administration, several strategies can be considered: 1.

Partial Insulin Administration: For patients with potentially unreliable intake, a portion of mealtime insulin can be administered before meals, with the remainder given after meals are completed, based on the amount consumed. However, this approach requires nursing staff to administer insulin twice per meal, significantly increasing their workload. Additionally, it relies on nursing staff to assess the proportion of the meal completed, which lacks standardization and may result in interpersonal variations.

2.

Ultra-Rapid-Acting Insulin: Administer ultra-rapid-acting insulin after meals are completed. Future inpatient studies are needed to compare this approach with premeal administration of rapid-acting insulin and ultra-rapid-acting insulin. The ultra-rapid insulins also may be more expensive.

3.

Inhaled Insulin: Inhaled insulin, which has a rapid onset and shorter duration of action compared with injectable insulins, has demonstrated feasibility in the outpatient setting. ^17,18 However, common conditions leading to hospitalization, such as pulmonary infections and chronic obstructive pulmonary disease, are contraindications for its use. Furthermore, training inpatients to use inhaled insulin correctly poses an additional challenge, and thus, this is not a realistic option but needs to be acknowledged to be complete.

4.

Targeted Premeal Administration: Identify populations prone to postprandial hyperglycemia and administer premeal insulin accordingly. Based on our experience, patients with type 1 diabetes, insulin-requiring type 2 diabetes, or those receiving high-dose glucocorticoids are more likely to experience severe postprandial hyperglycemia if insulin is not administered before meals. However, both patients with type 1 diabetes and supraphysiological glucocorticoid use were excluded in this study and the TIGHT study, highlighting the need for further research to better identify at-risk populations. For now, we should prioritize premeal insulin administration for patients with type 1 and insulin-requiring type 2 diabetes whenever feasible—unless challenges like unpredictable oral intake, nausea/vomiting, or gastroparesis make it unsuitable.

In summary, this study corroborates our clinical observations that post-meal insulin administration in hospitalized patients contributes to postprandial hyperglycemia. These findings underscore the need to reassess the timing of insulin administration in relation to meal intake within hospital settings. Additionally, this challenge identifies key areas of potential future research.