In this article entitled, “An electronic health record-integrated computerized intravenous insulin infusion protocol: clinical outcomes and in silico adjustment,” Park et al. [6] aimed to investigate the clinical outcome of a computerized intravenous insulin infusion (CII) protocol integrated to the electronic health record (EHR) system and to adjust the CII protocol in silico using the EHR-based predictors of the outcome. They showed that EHR-integrated CII protocol successfully maintained target glucose level while minimizing the risk of hypoglycemia. After EHR-based adjustment, the CII protocol reduced the delayed responses in silico patients without increasing hypoglycemia incidence. However, there are still some issues not covered in this study.

First, the reasons for selection of the Yale insulin infusion protocol should be mentioned in this study. There are a number of published insulin infusion protocols. However, the single most effective and safe insulin infusion protocol for controlling blood sugar in hospitalized patients has not been identified [7]. The Yale insulin infusion protocol was developed for the management of critically ill hyperglycemic patients in the medical intensive care unit, not specifically adjusted for those patients with diabetic emergencies, such as diabetic ketoacidosis or hyperglycemic hyperosmolar states or noncritical illness [14]. However, the reason why this protocol was applied to a very heterogeneous inpatient care settings, including perioperative glycemic control, continuous enteral feeding, diabetic ketoacidosis, and fasting for other reasons, was not discussed in the present study. Second, there is a discrepancy in target glucose levels in the study. The initial target glucose levels of the CII protocol is between 140 and 180 mg/dL. However, the authors used a different target glucose levels between 70 and 180 mg/dL in silico simulation. It is reasonable to apply the same target glucose levels throughout the study to reach a consistent and definite conclusion. The readers might be confused with inconsistent target glucose levels, which requires explanations for changes of the target value. Finally, discontinuation of the intravenous insulin infusion often leads to recurrence of hyperglycemia. Therefore, a description of post-CII management to minimize this repulsive effects, such as subcutaneous insulin protocols, would improve the level of perfection of the study [14578].

Nevertheless, the present study is very provocative and interesting. Further head-to-head comparisons of the different CII protocols in the large-scale randomized trials including in silico simulation will be required for development of the best-tailored CII protocol for generalized use.