Translate this page into:
Use of the Glucommander® to manage young pediatric patients with diabetic ketoacidosis: A case series
*Corresponding author: Faith Christina Harris, Mercer University School of Medicine, Macon, Georgia, United States. fharris138@gmail.com
-
Received: ,
Accepted: ,
How to cite this article: Harris FC, Hutchins J, Narsinghani U. Use of the Glucommander® to manage young pediatric patients with diabetic ketoacidosis: A case series. J Pediatr Endocrinol Diabetes. doi: 10.25259/ JPED_24_2025
Abstract
Diabetic ketoacidosis (DKA) is rare in infants, and the current standard treatment is the traditional two-bag method. We aim to increase awareness of cases of DKA in three pediatric patients under 2 years, and how they were effectively treated with a computerized algorithm, the Glucommander®. Target blood glucose levels were achieved in 1, 7, and 20 h, and the time to correction of metabolic acidosis was 16, 9, and 21 h. Glucommander® is not yet approved by the United States Food and Drug Administration (US FDA) for patients under 2 years. Through this study, we wish to demonstrate the safety of Glucommander® as a treatment option in this age group.
Keywords
Diabetic ketoacidosis
Glucommander®
Pediatrics
Type 1 diabetes mellitus
INTRODUCTION
Type 1 diabetes mellitus (T1DM) is an immune-mediated disease with an indolent prodromal phase characterized by cell-mediated destruction of pancreatic β-cells that produce insulin in persons with genetic susceptibility.[1] Diabetic ketoacidosis (DKA) is a major complication of T1DM. Although there have been major advances in the understanding of the pathophysiology of T1DM and DKA, DKA still accounts for the majority of T1DM-related deaths in pediatric patients.[2] DKA is associated with ketosis, acidemia, hyperglycemia, weight loss, polyuria, and dehydration. This condition can be fatal if not recognized and treated in a timely manner.
There are a few reported cases of DKA in patients <2 years old. However, we report three unusual cases of 12-month, 14-month, and 20-month-old patients who were diagnosed with moderate/ severe DKA and were effectively treated with Glucommander® application.
The Glucommander® is an adaptable, weight-based algorithm that evaluates the aggregate blood glucose (BG) levels and uses these measurements to calculate a dosage of insulin to lower the BG to a pre-determined target range.[3] In 1982, an array of orders was suggested to predict the basal dosage of insulin needed to treat pediatric patients with insulin pumps in a study by White et al.[4] Data from the study were graphed using linear regression analysis with an intercept of 60 and a slope/multiplier of 0.02 to lower BG with infrequent complications. From these findings, insulin dose/hour = (BG – 60) × 0.02 was derived.[5] The value 0.02 represents the multiplier, or an adjustable coefficient that can be raised or lowered depending on the severity of hyperglycemia.[5] In our cases, we used Glucommander®, a proprietary algorithm that analyzes patients’ glucose levels in real time and calculates for a non-linear decrease in BG that is efficient and unique to the severity of the patient’s hyperglycemia. This algorithm uses an initial weight-based multiplier and constantly analyzes the velocity of the glycemic change and its proximity to the target BG range. The multiplier subsequently increases or decreases and recommends a specific insulin dosage based on the patient’s real-time and historic data hourly.
CASE SERIES
Patient A
A 12-month-old male (8.1 kg) presented to the emergency room (ER) with lethargy, polyuria, and polydipsia for 10 days and two episodes of non-bilious vomiting. He was afebrile, tachycardic, and ill appearing, exhibited Kussmaul breathing, dry mucous membranes, cracked lips, and prolonged capillary refill time. Laboratory findings showed a pH of 7.122, serum bicarbonate was <7 mmol/L, anion gap (AG) of 19 mmol/L, BG of 323 mg/dL, and beta-hydroxybutyrate of 9.3 mmol/L. The patient was given 20 mL/kg of normal saline (NS) bolus but received no insulin in the ER.
Patient B
An 18-month-old female (11.2 kg) presented to the outlying ER with a history of 4 days of fatigue, polydipsia, polyuria, and 3–4 episodes of non-bilious vomiting daily. She was afebrile. Laboratory findings were significant for a pH of 7.135, serum bicarbonate was 12 mmol/L, AG of 16 mmol/L, BG of 558 mg/dL, and beta-hydroxybutyrate level of 3.8 mmol/L. Patient received an NS bolus and an insulin infusion of 0.1 units/kg/h in the ER.
Patient C
A 20-month-old female (9.3 kg) was transferred from an outside hospital after several hours of lethargy and one episode of non-bilious vomiting. She had a recent history of loss of balance, difficulty in walking, polyuria, and polydipsia. BG was 945 mg/dL. Basic metabolic panel (BMP) showed a serum bicarbonate of 4 mmol/L and AG of 35.5 mmol/L. Beta-hydroxybutyrate was not measured. She was given an NS bolus and started on an insulin infusion of 0.05 units/kg/h. On transfer to our hospital, vital signs were stable, except for sinus tachycardia. Laboratory findings were significant for venous pH of 7.13, serum bicarbonate of 5 mmol/L, base deficit of 22 mmol/L, and BG of 521 mg/dL.
Treatment
All three patients were diagnosed with DKA secondary to new onset T1DM, transferred to the pediatric intensive care unit (PICU), and placed on insulin infusion through the Glucommander® protocol. Vital signs and neurological assessments were done hourly. BMP was obtained every 4–6 h. Per consultation with pediatric endocrinology, the target BG for patient A and B was set to 150–200 mg/dL, and for patient C, it was set to 100–140 mg/dL. Using weight-based multipliers, the initial insulin infusion rates were calculated. As prompted by the Glucommander®, the patients’ BG was checked hourly and entered into the application by the bedside nurse. Based on each patients’ average glucose velocity change, the generative artificial intelligence (AI) component of the Glucommander® adjusted the multiplier up or down to continue to recommend appropriate insulin dosages that would lower BG in a safe and predictable manner [Figure 1]. Treatment consent was obtained, and all three patients were treated similarly per PICU protocol with isotonic fluids and added potassium supplementation (phosphate and acetate). An estimated fluid deficit of 5–7% was replaced over 24– 48 h. 5% dextrose was added to 0.45–0.9% sodium chloride intravenous fluids when BG decreased to <250 mg/dL.
- Patient A, B, and C, blood glucose levels while on the Glucommander® protocol on admission to the pediatric intensive care unit. Stars correspond to each patient’s time to target blood glucose and circles correspond to time to correction of metabolic acidosis.
Outcome and follow-up
Discontinuation criteria for Glucommander® involved DKA correction, which was defined as serum bicarbonate ≥15 mmol/L, closure of AG, stable glycemic control, and resolution of clinical symptoms of DKA.
For each patient, target BG range, time to target BG, time to correction of metabolic acidosis (MA), total time on the Glucommander®, total IV insulin, and corresponding values are shown in Table 1. There were no episodes of hypoglycemia, electrolyte imbalances, or other complications. Once DKA was corrected, all three patients were taken off the Glucommander®. Patients were then started on individualized subcutaneous maintenance insulin and an appropriate low-carbohydrate diet. After teaching and education, patients were discharged home and recommended follow-up with pediatric endocrinology.
| Patient | TBG range (mg/dL) | Time to TBG (h) and related BG level (mg/dL) | Time to correction of MA (h) | Serum bicarbonate level at time to correction of MA (mmol/L) | AG at time to correction of MA (mmol/L) | Total time on Glucommander (h) | Total IV insulin used on Glucommander (Units) |
|---|---|---|---|---|---|---|---|
| Patient A | 150–200 | 1 (193) | 16 | 15 | 12 | 17.5 | 4.30 |
| Patient B | 150–200 | 7 (198) | 9 | 16 | 11 | 9 | 7.65 |
| Patient C | 100–140 | 20 (138) | 21 | 16 | 5 | 21 | 7.37 |
TBG: Target blood glucose, BG: Blood glucose, AG: Anion gap, MA: Metabolic acidosis
DISCUSSION
Goals of treatment for DKA are to correct dehydration, acidosis, reverse ketosis, gradually restore hyperosmolality and BG to near normal levels, monitor for acute complications, and identify and treat any precipitating event. Insulin therapy is essential to restore normal cellular metabolism, suppress lipolysis and ketogenesis, and to normalize BG concentrations.[6] Technology influences healthcare more than any other approach, and in the future, will continue to develop and evolve in dramatic ways. Part of Glucommander®’s efficiency in treating hyperglycemic crises involves the generative AI component and embedded calculator that eliminates complex computations and minimizes human error with insulin administration. Evidence suggests that computer-based algorithms lead to decreased adverse effects and faster reduction of blood ketones and hyperglycemia into the euglycemic range as opposed to a written scale-based algorithm like the twobag method.[7] Glucommander® tool is readily available in many hospitals but only United States Food and Drug Administration (US FDA) approved for use in patients over the age of 2 years. This case series illustrates that it is safe and reliable for use in children under the age of 2.
There is a paucity of published cases of patients with DKA under 24 months of age. DKA in this younger pediatric population is associated with a higher mortality and is difficult to treat, due to misdiagnosis as respiratory or gastrointestinal illness, prolonging the duration of DKA and worsening the acidemia and dehydration.[8] It is also difficult to correct hyperglycemia due to unpredictable sensitivity to insulin in this age group. Effectual treatment for DKA is crucial in these young patients due to increased risk of cerebral edema, a dreaded complication of DKA, which has a mortality rate of 20–30%.[9]
Glucommander® has been shown to decrease time and quantity of insulin utilized, fluid changes, and other interventions needed to reach target BG.[6] It has also been shown to decrease time spent in the PICU and hospital compared to patients managed with manual insulin protocols.[3]
Glucommander® features safety guardrails and a cloud-based platform and is suitable for any setting where patients require IV insulin. In addition, it offers personalized insulin dosing recommendations that ensure precise control of BG levels while minimizing risk of hypoglycemia; the decrease in BG typically does not exceed 50–100 mg/dL/h.
There are some limitations to our study. All patients were treated with boluses of isotonic fluids, and two of the three patients received insulin in the ER before their admission to the PICU for treatment through Glucommander® protocol. As a result, their BG levels at initial presentation to the PICU were lower than the values in the ER. Patients B and C were transferred from outlying ERs; therefore, the total time spent before admission was difficult to quantify. In addition, since very young patients are often more sensitive to insulin, we relied on recommendations from endocrinologists for specific target BG ranges that were tailored for our patients. Finally, all three patients received long-acting basal insulin as a plan to transition to a subcutaneous insulin regimen while being on the Glucommander® which could have led to shorter time to glycemic control.
CONCLUSION
Although not approved in children under the age of 2 years, Glucommander® provides a predictable, safe, and efficient method of treating DKA and hyperglycemia in this particularly vulnerable population. While cases of DKA in patients younger than 2 years of age are rare, they are often more difficult to treat and associated with more complications. Glucommander® offers a proven tool for precision dosing suitable for any setting where patients require IV insulin, it is easy to order, features automatic adjustments, and is intuitively less error-prone.
Ethical approval:
Institutional Review Board approval is not required.
Declaration of patient consent:
The authors certify that they have obtained all appropriate patient consent.
Conflicts of interest:
There are no conflicts of interest.
Use of artificial intelligence (AI)-assisted technology for manuscript preparation:
The authors confirm that there was no use of artificial intelligence (AI)-assisted technology for assisting in the writing or editing of the manuscript and no images were manipulated using AI.
Financial support and sponsorship: Nil.
References
- Environmental triggers and determinants of type 1 diabetes. Diabetes. 2005;54(Suppl 2):S125-36.
- [CrossRef] [PubMed] [Google Scholar]
- Risk of death following admission to a UK hospital with diabetic ketoacidosis. Diabetologia. 2016;59:2082-87.
- [CrossRef] [PubMed] [Google Scholar]
- Evaluating the safety and efficacy of Glucommander, a computer-based insulin infusion method, in management of diabetic ketoacidosis in children, and comparing its clinical performance with manually titrated insulin infusion. J Pediatr Endocrinol Metab. 2009;22:119-25.
- [CrossRef] [PubMed] [Google Scholar]
- Practical closed-loop insulin delivery. A system for the maintenance of overnight euglycemia and the calculation of basal insulin requirements in insulin-dependent diabetics. Ann Intern Med. 1982;97:210-3.
- [CrossRef] [PubMed] [Google Scholar]
- Glucommander: A computer-directed intravenous insulin system shown to be safe, simple, and effective in 120,618 h of operation. Diabetes Care. 2005;28:2418-23.
- [CrossRef] [PubMed] [Google Scholar]
- ISPAD clinical practice consensus guidelines 2022: Diabetic ketoacidosis and hyperglycemic hyperosmolar state. Pediatr Diabetes. 2022;23:835-56.
- [CrossRef] [PubMed] [Google Scholar]
- Computer-based versus paper-based insulin infusion algorithms in diabetic ketoacidosis. Curr Diabetes Rev. 2020;16:628-34.
- [CrossRef] [PubMed] [Google Scholar]
- Diabetic ketoacidosis in the pediatric ICU. Pediatr Clin North Am. 2008;55:577-87. x
- [CrossRef] [PubMed] [Google Scholar]
- Brain injury in children with diabetic ketoacidosis: Review of the literature and a proposed pathophysiologic pathway for the development of cerebral edema. Pediatr Diabetes. 2021;22:148-60.
- [CrossRef] [PubMed] [Google Scholar]