Rate of Inpatient Hypoglycemia Following a 1:1 Dose Interchange Between Concentrated Insulin Glargine to Insulin Detemir

Abstract

Background

Insulin remains a mainstay of treating hyperglycemia in an acute setting. Insulin glargine 300 units/mL (Toujeo, iGlar300) has a different pharmacokinetic profile than 100 units/mL basal insulins, such as insulin detemir (iDet100) and iGlar100. While conversion from iGlar300 to iGlar100 requires a 20% dose decrease, there is currently no recommended interchange from iGlar300 to iDet100.

Objective

Compare the incidence of hypoglycemia in patients who received a 1:1 unit interchange from home iGlar300 or iGlar100 to iDet100 while admitted.

Methods

A retrospective study was conducted to evaluate adults within a multi-site network admitted between May and December 2019. Patients were included if they received at least one dose of iDet100 following interchange from home iGlar300 or iGlar100. The primary endpoint was the incidence of hypoglycemic events following a 1:1 interchange of iGlar300 vs. iGlar100 to inpatient iDet100. Secondary outcomes include overall hypoglycemic events, time to hypoglycemia, and doses given before hypoglycemia.

Results

Of 615 patients, 394 received a 1:1 unit interchange to iDet100 (52 from iGlar300 and 342 from iGlar100). Incidence of hypoglycemic events was significantly higher in those with a 1:1 interchange from iGlar300 versus iGlar100 (36.5% vs. 18.7%, p = 0.007). Significant differences were observed in overall hypoglycemic events, time to hypoglycemia, and number of doses given before hypoglycemic event.

Conclusion and Relevance

A 1:1 unit interchange from iGlar300 to iDet100 led to a higher incidence of hypoglycemic events compared to those interchanged from iGlar100. Dose reduction should be considered when transitioning from home iGlar300 to iDet100 in the inpatient setting.

Keywords

basal insulin concentrated insulin diabetes hypoglycemia

Introduction

Diabetes mellitus (DM) affects a significant proportion of the adult population in the United States, with 13% carrying a diagnosis of DM in 2018.¹ According to the Centers for Disease Control and Prevention, 16 million emergency department visits and 7.8 million hospital discharges in 2016 involved an adult patient with a confirmed diagnosis of DM, contributing to $237 billion of total direct medical costs for managing DM.¹ Recommendations from the American Diabetes Association for managing hyperglycemia in the hospital setting recommend the use of basal insulin or basal plus bolus correction insulin to achieve adequate glucose levels.² Hospitals typically select a single formulary basal insulin to manage costs and standardize dosing practices. While clinical scenario may dictate either increase or decrease from home insulin dosing, therapeutic interchanges often include a suggested unit per unit interchange from patient’s home insulin to formulary agent for stable patients; however, the variation in pharmacokinetic (PK) profiles between basal insulins makes 1:1 unit conversions controversial.^3,4

The introduction of two concentrated insulins in 2015, insulin glargine 300 units/mL (Toujeo, iGlar300) and insulin degludec 200 units/mL (Tresiba, iDeg200), marked the first Food and Drug Administration (FDA)-approved concentrated insulins since the creation of regular insulin 500 units/mL. These concentrated insulins have differing PK/pharmacodynamic (PK/PD) profiles compared to 100 units/mL insulin formulations (iGlar100 and iDet100) due to alterations in concentrations and injection depot surface areas. iGlar300 and iDeg200 have increased durations of action over 24 hours and flattened insulin peaks, better mimicking physiologic pancreatic release of basal insulin.^5,6 While iGlar300 has no structural differences to iGlar100, the higher concentration reduces the injected volume and exposed surface area of drug, which slows the release profile leading to the PK/PD differences.^5-7

iGlar300 was compared to iGlar100 in four large, multicenter, open-label trials (EDITION 1-4) evaluating safety and efficacy in various patient populations with DM (type I DM, insulin-naive type II DM, and insulin dependent type II DM). In each trial, iGlar300 was non-inferior to its U100 comparator in glycated hemoglobin (A1C) reduction and in varying capacities showed similar or reduced incidence of nocturnal and severe hypoglycemic events. The EDITION trials identified one noticeable difference between the two glargine insulins, specifically that a higher iGlar300 dose (~10%-17% unit increase) was needed to yield similar outcomes as those receiving iGlar100.^8-13 In a PD analysis of the 24-hour glucose-lowering effect at steady state, iGlar300 had a 27% reduction in potency than an equivalent administered dose of iGlar100.^14,15 As a result, a 20% dose reduction is recommended when switching from iGlar300 to iGlar100 to minimize risk of hypoglycemia.¹⁶

Discrepancies in insulin potency also exist between non-concentrated insulins; however, a 1:1 unit conversion when switching between iGlar100 and iDet100 does not appear to have a significant impact on rates of hypoglycemia.^4,17 Transitioning concentrated insulin to a non-concentrated basal insulin may further expose known variations in insulin unit potency and PK/PD differences and potentially contribute to undesired outcomes.^18,19 Clinical opinion suggests that dose reductions may be appropriate when transitioning from outpatient concentrated insulins to formulary agents.^20-22 Despite the recommendations for dose reduction when converting between glargine formulations, there are no specific data to guide conversion from iGlar300 to iDet100. This study aimed to evaluate incidence of hypoglycemic rates in patients receiving a 1:1 unit conversion from iGlar300 or iGlar100 to iDet100 on admission.

Materials and Methods

This retrospective cohort study included nine hospitals within the Ascension Seton network in Central Texas and was approved by the local institutional review board. Adult patients (≥ 18 years) admitted to a network hospital between May 2019 and December 2019 with a documented home medication of iGlar300 or iGlar100 were eligible for inclusion. Patients were identified if a therapeutic interchange occurred in the electronic medical record from home insulin glargine to iDet100 while in the inpatient setting. Those with multiple admissions during the specified time period had each encounter considered for inclusion separately. Patients were excluded if they did not receive at least one dose of iDet100 in the inpatient setting following conversion, had no confirmed outpatient prescription or dose for iGlar300 or iGlar100 on external fill history and admission medication history, or had documented prescriptions for other basal insulins (iDeg200 and iDet100) in the outpatient setting at the time of admission. Any patient who received a continuous infusion of regular insulin before and/or during long-acting insulin administration was also excluded. If inclusion criteria were met, the following baseline demographics were collected: age, sex, weight, A1C, diagnosis of end-stage renal disease (ESRD, defined by documented diagnosis or receipt of chronic dialysis), documented home insulin dose, and initial inpatient basal insulin dose. Serum creatinine measurements were collected for each patient and evaluated for incidence of acute kidney injury (AKI) at any point during inpatient stay according to the Acute Kidney Injury Network (AKIN) criteria.²³

The primary outcome was incidence of hypoglycemic events among patients receiving a 1:1 unit conversion from iGlar300 versus iGlar100 to iDet100 in the inpatient setting. A hypoglycemic event was defined as a blood glucose of < 70 mg/dL following the unit per unit interchange. Any hypoglycemic events that occurred prior to first inpatient dose or after a dose adjustment of iDet100 were not classified as a hypoglycemic event for either the primary or the secondary outcome analysis. Secondary outcomes evaluating all patients regardless of interchange ratio included time to first hypoglycemic event following interchange, number of doses of iDet100 received prior to first hypoglycemic event, number of hypoglycemic events per patient, and total hypoglycemic events that occurred during admission. A comparison of hypoglycemic events in each cohort (iGlar300 and iGlar100) based on the presence or absence of a 1:1 unit conversion was performed. In those patients on a twice daily regimen of iDet100, the time to hypoglycemia was based on the time of first dose given starting a 24-hour insulin administration period. When determining number of doses received prior to an event, each dose of a twice daily regimen was considered as a half dose.

After initial data collection was completed and statistics were evaluated, additional patient characteristics were collected only in patients with a confirmed hypoglycemic event regardless of interchange ratio. These characteristics include enteral (nil per os [nothing by mouth]; NPO) status, time of last short-acting insulin (SAI) dose prior to hypoglycemic event, total SAI before hypoglycemic event, type of insulin sliding scale ordered by prescriber (low, medium, or high dose), and receipt of intravenous (IV) regular insulin for hyperkalemia. NPO status was defined as having a confirmed NPO order in the patient chart and both of the following criteria were met: iDet100 was administered and a timed meal would have been missed. These characteristics were then used to determine whether the hypoglycemic event could have been caused by a confounding variable, independent of the multivariate logistic regression. Patients with a dose of SAI given within 6 hours before the event and/or NPO status were considered to have potential confounders. Subsequently, a tertiary analysis was performed to reevaluate the primary and secondary outcomes with these confounding patients removed.

For baseline characteristics, means and standard deviations (SDs) were used to describe continuous variables with normal distribution, and median and interquartile range (IQR) were used for non-normally distributed or ordinal data. The Shapiro–Wilk test was used to determine normality of the data. Outcomes with ordinal data, including the primary endpoint, were evaluated using a chi-squared test. Parametric and non-parametric data were analyzed using student’s t-test or Mann–Whitney test, respectively. A multivariate logistic regression was used to assess for the following potential confounders: age, gender, weight, A1C, presence of AKI, and diagnosis of ESRD. An alpha of 0.05 was considered statistically significant. Statistical analyses were conducted using SAS version JMP Pro 15 (SAS Institute, Cary, NC).

Results

Study Patients

A total of 663 patients were screened, with 615 patients meeting inclusion criteria (80 and 535 patients in the iGlar300 and the iGlar100 groups, respectively) for analysis (Figure 1). A 1:1 unit conversion from home insulin glargine to inpatient iDet100 occurred in 52 patients (65%) in the iGlar300 arm and 342 patients (64%) in the iGlar100 arm.

Baseline characteristics (Table 1) were similar between groups, with a few exceptions. Notable differences include a higher percentage of male patients in the iGlar300 arm and a lower body weight in the iGlar100 arm. Doses of insulin glargine both at home and following interchange were higher in the iGlar300 arm.

Table 1.

Baseline Characteristics.

Baseline characteristic	iGlar300 (n = 80)	iGlar100 (n = 535)	P value
Age, yr (median, IQR)	64.5 (56.3-72)	65 (57-72)	0.872
Male, n (%)	53 (66.3%)	228 (42.6%)	< 0.001
Weight, kg (median, IQR)	98.4 (80.8-111.5)	89.4 (73.2-107.6)	0.009
TDD of home glargine, units (median, IQR)	50 (28.5-72)	30 (20-50)	0.001
TDD of home glargine by weight, units/kg (median, IQR)	0.5 (0.3-0.7)	0.4 (0.2-0.6)	0.007
TDD of inpatient detemir, units (median, IQR)	36.5 (20.3-60)	26 (16-40)	0.002
TDD of inpatient detemir by weight, units/kg (median, IQR)	0.4 (0.3-0.5)	0.3 (0.2-0.5)	0.012
Percentage change in TDD following interchange (mean, SD)^a	–35.3% (+13.6)	-29.1% (+5)	0.67
A1C, % (median, IQR)	7.4 (6.5-8.9)	7.8 (7-9.5)	0.139
AKI, n (%)^b	16 (20%)	132 (24.7%)	0.353
ESRD, n (%)^c	5 (6.3%)	59 (11%)	0.166

Abbreviations: A1C, glycated hemoglobin; AKI, acute kidney injury; ESRD, end-stage renal disease; iGlar100, insulin glargine 100 units/mL; iGlar300, insulin glargine 300 units/mL; IQR, interquartile range; kg, kilogram; SD, standard deviation; TDD, total daily dose; yr, year.

Represents patients who did not undergo a 1:1 unit interchange.

AKI is defined per the Acute Kidney Injury Network (AKIN) criteria as absolute increase in serum creatinine of more than or equal to 0.3 mg/dL and/or a percentage increase in serum creatinine of more than or equal to 50% from baseline.

ESRD is defined as a documented diagnosis in the patient’s chart or receipt of chronic dialysis.

Primary Analysis

The incidence of hypoglycemic events following a 1:1 unit interchange was significantly higher in the iGlar300 group compared to the iGlar100 group (36.5% vs. 18.7%; OR = 2.5, 95% CI: 1.34-4.68; P = 0.006) (Table 2). A multivariate logistic regression was performed to determine whether any other variables may have influenced the primary outcome. None of the measured variables previously described demonstrated a statistically significant association with incidence of hypoglycemic events (Table 3). When using a stepwise approach, no statistically significant associations were found.

Table 2.

Results.

Primary outcome	iGlar300 (n = 52)	iGlar100 (n = 342)	P value
Hypoglycemic events following a 1:1 interchange n (%)	19 (36.5%)	64 (18.7%)	0.006
Secondary outcomes	iGlar300 (n = 80)	iGlar100 (n = 535)	P value
Overall hypoglycemic events, n (%)	24 (30%)	103 (19.3%)	0.033
Time to hypoglycemic events, hr (median, IQR)	10 (8-14)	18 (10-33)	0.001
Number of doses before hypoglycemic event (mean + SD)	1.1 (+ 0.3)	1.7 (+ 1.6)	0.043
Number of hypoglycemic events per patient (mean + SD)	0.76 (+1)	0.49 (+1.6)	0.427

Abbreviations: iGlar100, insulin glargine 100 units/mL; iGlar300, insulin glargine 300 units/mL; IQR, interquartile range; SD, standard deviation.

Table 3.

Multivariable Logistic Regression.

Variable	LogWorth	P value
AKI^a	0.918	0.121
A1C%	0.747	0.179
ESRD^b	0.626	0.236
Gender	0.453	0.352
Age	0.397	0.401
Weight	0.083	0.827

Abbreviations: A1C, glycated hemoglobin; AKI, acute kidney injury; ESRD, end-stage renal disease.

ESRD is defined as a documented diagnosis in the patient’s chart or receipt of chronic dialysis.

Secondary Analysis

A secondary analysis was completed to investigate the relationship of hypoglycemic events and dose changes regardless of interchange ratio (Table 2). The median time to the first hypoglycemic event was significantly shorter in the iGlar300 group versus the iGlar100 group (10 vs. 18 hours, P = 0.001). The mean number of doses given prior to the first hypoglycemic event was statistically fewer in the iGlar300 cohort when compared to the iGlar100 cohort (1.1 vs. 1.7 doses, P = 0.043).

Total incidence of hypoglycemic events regardless of interchange ratio was also significantly higher in the iGlar300 group compared to the iGlar100 group (30.0% vs. 19.3%; OR = 1.8, 95% CI: 1.06-3.04; P = 0.033) (Table 2). Patients in the iGlar100 group who did and did not have a dose reduction had similar rates of hypoglycemia (19.2% vs. 18.1%, respectively); however, in the iGlar300 group, patients with dose reductions had numerically lower rates of hypoglycemia than those without reductions, though this finding did not reach statistical significance (18.5% vs. 36.5%, P = 0.09).

Tertiary Analysis

A tertiary analysis was completed to evaluate specific characteristics and potential confounders only in the population of patients who had a hypoglycemic event (Table 4). Of the potential confounders evaluated, only total SAI before hypoglycemic event was found to be significantly different with those in the iGlar100 arm receiving more units than the iGlar300 group (3.5 vs. 6, P = 0.016).

Table 4.

Hypoglycemia Subgroup Characteristics.

Hypoglycemia subgroup characteristics	iGlar300 (n = 24)	iGlar100 (n = 103)	P value
Number of patients with potential confounder (SAI usage < 6 hours or NPO^a), n (%)	5 (20.8%)	30 (29.1%)	0.402
NPO status, n (%)	1 (4.2%)	11 (10.7%)	0.461
Time from last SAI dose, hours (median, IQR)	8 (5.9-11)	9.5 (6.9-13.9)	0.154
Total SAI before HG event, units (median, IQR)	3.5 (0-6)	6 (1-19)	0.016
High-dose sliding scale insulin ordered, n (%)	4 (16.7%)	9 (9.09%)	0.437
Regular insulin IV push usage for hyperkalemia, n (%)	1 (4.2%)	5 (4.85%)	1.000

Abbreviations: HG, hypoglycemic; iGlar100, insulin glargine 100 units/mL; iGlar300, insulin glargine 300 units/mL; IQR, interquartile range; IV, intravenous; NPO, nil per os (nothing by mouth); SAI, short-acting insulin; SD, standard deviation.

NPO was defined as having a confirmed NPO status in the patient chart where basal insulin had been given and the patient would have skipped a timed meal.

The number of patients with potential confounders was not significantly different between groups. When those patients with a hypoglycemic event and a potential confounder were removed and statistical tests were reperformed on the primary and secondary outcomes, no changes in status of statistical significance were noted in any of the outcomes (Table 5).

Table 5.

Primary and Secondary Outcomes With Potential Confounders Removed.

Primary outcome with potential confounders removed	iGlar300(n = 49)	iGlar100(n = 329)
Hypoglycemic events following a 1:1 interchange	16 (32.7%)	51 (15.5%)	0.006
Secondary outcomes with potential confounders removed	iGlar300(n = 75)	iGlar100(n = 505)
Overall hypoglycemic events, n (%)	19 (25.3%)	73 (14.5%)	0.022
Time to hypoglycemic events, hr (median, IQR)	11 (8-19)	20 (10-32.3)	0.002
Number of doses before hypoglycemic event (mean + SD)	1.1 (+ 0.2)	1.7 (+ 1.2)	0.027

Abbreviations: iGlar100, insulin glargine 100 units/mL; iGlar300, insulin glargine 300 units/mL; IQR, interquartile range; SD, standard deviation.

Discussion

With the approval of additional concentrated insulins, including iGlar300, there is a need to determine best practices when transitioning patients from these concentrated basal insulins to hospital formulary options while admitted. Unlike the concentrated regular insulin 500 units/mL, which is usually reserved for highly insulin-resistant patients and not interchanged to formulary 100 units/mL basal insulin products at admission, iGlar300 has the potential to be prescribed to a greater number of patients in the community, making this a particularly relevant issue for inpatient practitioners. While PD analysis and results from the EDITION trials highlight the need for a 20% dose reduction when converting iGlar300 to iGlar100, it is not well elucidated how to transition patients on iGlar300 to iDet100.^10-13 This is the first study, to the authors’ knowledge, to evaluate conversion of iGlar300 to iDet100 in the inpatient setting while comparing to a control group of patients interchanged from iGlar100 during the same timeframe.

When converting from iGlar300 to iDet100 at a 1:1 unit dose conversion, patients in our study had almost twice the risk of hypoglycemia compared to a 1:1 unit conversion from iGlar100. This means that for every five patients treated with a 1:1 dose conversion, one more patient will experience a hypoglycemic event. This trend was similar when we evaluated all patients regardless of the ratio of dose interchange on admission, with 30% of iGlar300 patients having a hypoglycemic event following the interchange compared to only 19.3% of iGlar100 patients. This finding was likely influenced by the large proportion of patients converted on a 1:1 unit basis, which was around 65% in both arms. While patients in the iGlar300 arm had higher weight and total daily insulin doses by weight, it is not suspected to have had an effect on hypoglycemia events. Rather, the lower median A1C in the iGlar300 arm would potentially suggest greater risk for hypoglycemia; however, the difference in median A1C between groups was small, not statistically significant, and likely not clinically significant to our findings. It is noteworthy that hypoglycemia rates were similar in the iGlar100 arm regardless of dose reduction, while hypoglycemia was numerically lower in patients who had a reduction from home dose of iGlar300 when compared to 1:1 interchange. The study was not powered to detect differences within the iGlar300 arm, but this finding indicates that an empiric dose reduction may be effective at reducing hypoglycemia risk.

The hypoglycemic event rate seen in the iGlar100 arm in our study (19.3%) was lower than the event rates seen in a similar analysis conducted by Capson, et al evaluating the 1:1 conversion from iGlar100 to iDet100 or maintained on iGlar100 in an inpatient setting (46% and 49%).¹⁷ However, our study did not intend on capturing the totality of hypoglycemic events during an inpatient admission and excluded any hypoglycemic events occurring before insulin administration, after a dose change inpatient, or any that occurred after the first hypoglycemic event. While this likely resulted in a comparatively lower overall hypoglycemic event rate recorded, both studies align in support that a 1:1 conversion between iGlar100 to iDet100 in the inpatient setting is likely appropriate. In a recent analysis evaluating iGlar300 vs. iGlar100 in the inpatient setting, hypoglycemia rates occurred at a lower incidence than our study (8.7% and 9.5%, respectively) which highlights the wide range of hypoglycemia rates that may be observed when evaluating insulin use in the inpatient setting.²¹

The explanation for the increased hypoglycemic events seen in the iGlar300 arm of our study may be multifactorial but can likely be attributed to the differences in the PK/PD profile of the concentrated insulin. Given the peakless and longer duration of action of iGlar300 compared to iGlar100 (30 vs. 24 hours), insulin stacking may have occurred leading to hypoglycemic events. If a patient was on iGlar300 prior to admission and was given a dose of inpatient insulin 24 hours after the last outpatient dose, then the patient’s overall exogenous insulin administered in a 24-hour period is presumably greater than if the patient was on iGlar100 at home. This theory is somewhat supported by the timing of the hypoglycemic event after the first inpatient insulin dose was given, occurring within 10 versus 18 hours in patients in the iGlar300 and iGlar100 arms, respectively. In addition, in the iGlar300 arm, a hypoglycemic event occurred after an average of 1.1 doses of iDet100 compared to an average of 1.7 doses of iDet100 in the iGlar100 arm.

While there is known potency variation between iGlar300 and iGlar100, there is limited existing data on the clinical impact of any potency variation between iGlar300 and iDet100. However, previous studies have established that there are differences in insulin potencies between the iGlar100 and iDet100 formulations.¹⁸ Human insulin has a defined potency of 1 unit per 6 nmol with a relative binding affinity of 1.0 for the insulin receptor.¹⁵ iGlar100 is formulated to have similar potency with 1 unit of insulin equaling 6 nmol of iGlar100 and a relative binding affinity of 0.67. Comparatively, insulin detemir has a formulation with 1 unit insulin equaling 24 nmol of iDet100 and a relative binding affinity of 0.25. This indicates a weaker potency on a per mole basis with detemir compared to glargine. While insulin potencies are standardized to an international unit scale, these standardization processes are done in healthy individuals or in vitro assays, which do not reflect the condition of hospitalized patients.¹⁸ Patient characteristics in an acute, inpatient setting may cause an increased sensitivity to the amount of insulin administered further exposing any PK/PD differences between the insulins.^5,7,24,25

Previous literature evaluating conversion between basal insulins have primarily occurred in the outpatient setting with patients who may have less risk factors for hypoglycemia than those admitted in an inpatient setting and where there is more time to implement titrations and adjustments to achieve goal glucose control.²⁶ Patients admitted to the hospital are at higher risk of hypoglycemic events due to change in enteral status, impaired renal function, use of strict carbohydrate-controlled diets, hypoalbuminemia, severe illness, such as sepsis, and change in traditional glycemic therapy, including the use of sliding scale insulin and discontinuation of oral antihyperglycemic agents.²⁷ Even when removing potential confounders, such as NPO status or SAI usage, the hypoglycemic event rate in our study was still significant between the two groups, occurring in 32.7% of the iGlar300 patients with a 1:1 dose interchange compared to 15.5% of those in the iGlar100 arm. New data are needed to further evaluate modifiable hypoglycemic risks and identify strategies to reduce inpatient hypoglycemia.

Our study is not without limitations. Given the retrospective nature of this study, certain factors that may have impacted glycemic control could not reliably be considered like other concomitant medications (ie, corticosteroids), reduced oral intake, time since last dose of any home oral antidiabetic agents or dietary changes compared to home. While corticosteroid use would have presumably contributed to less hypoglycemia, the study attempted to control for any factors that may have played a role in precipitating hypoglycemic events. To reduce confounding, any events that occurred prior to first basal insulin administration were excluded. In addition, the median time to first hypoglycemic event occurred within the first 24 hours of admission. Any doses of SAI administered within 6 hours preceding a hypoglycemic event were considered but did not appear to significantly contribute to the event rate. There was a statistical difference between the number of units of SAI administered prior to the hypoglycemic event, though this was higher in the iGlar100 arm (3.5 vs. 6 units; P = 0.016) and did not impact hypoglycemia rate. Our study did not evaluate any potential impact on patient outcomes (eg, length of stay, mortality) as our study was not adequately powered for such comparisons, and the extent of confounding variables would be significant. In addition, majority of patients included were admitted to a medical-surgical floor where low mortality rates were anticipated. Variation in prescriber preference and familiarity as it pertains to insulin dosing could also not be controlled, which could influence dose changes on a patient-by-patient basis. For instance, the ratio of short acting to long-acting insulin was noted to be higher in the iGlar100 group than the iGlar300, which intuitively could have increased the rates of hypoglycemia. Overall, 20.8% and 29.1% of iGlar 300 and iGlar100 patients, respectively, had a potential confounder, but none of the variables evaluated statistically altered our findings.

While our findings suggest the need for implementing a dose reduction upon conversion of iGlar300 to iDet100, the degree of percent reduction is not immediately clear. Based on the post-hoc analyses in the EDITION trials evaluating the iGlar300 to iGlar100 interchange, implementing a similar 20% minimum dose reduction when converting iGlar300 to iDet100 may be appropriate.^10-13,16 Operationalizing a minimum dose reduction in an electronic medical record and incorporating this practice change in the health care system may be challenging. Potential options could be an electronic alert and/or extensive staff education; however, these solutions can often be limited by alert fatigue and employee turnover, respectively. Following this study, our institution implemented a combination of an electronic alert with an automatic minimum 20% dose reduction at order entry where pharmacists serve as the final check to intervene when necessary. This workflow is effective but currently only takes effect when long-acting insulin is continued off the admission medication reconciliation versus independently ordered by the provider as a single order from the hospital database. As with any inpatient continuation of medications from a home medication list, the quality of the medication history is paramount. An erroneous documentation in home insulin unit dose could impact hypoglycemia rates upon transition to any inpatient insulin product. Institutions should strive to develop their own best practices around insulin conversion that can be seamlessly incorporated into workflow amid existing constraints.

Conclusion and Relevance

A 1:1 unit interchange from home iGlar300 to iDet100 led to a higher incidence of hypoglycemic events compared to those interchanged from iGlar100. Hypoglycemic events occurred in a shorter time from admission and following fewer insulin doses in the iGlar300 arm. These results were consistent after adjusting for potential confounders encountered in the acute setting. A dose reduction should be strongly considered when transitioning from home iGlar300 to iDet100 upon hospital admission. Additional studies are needed to further elucidate a specific percentage reduction in insulin dosage when interchanging from iGlar300 to iDet100 in the inpatient setting.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

S. D. Wulfe

J. Addison

References

Centers for Disease Control and Prevention. National Diabetes Statistics Report, 2020. Atlanta, GA: Centers for Disease Control and Prevention, U.S. Dept of Health and Human Services; 2020.

American Diabetes Association Professional Practice Committee. 16. Diabetes care in the hospital: standards of medical care in diabetes—2022. Diabetes Care. 2022;45(suppl 1):S244-S253.

Levemir [package insert]. Plainsboro, NJ: Novo Nordisk; 2005. Revised July 2022.

Mehta

Goldenberg

Katselnik

Kuritzky

. Practical guidance on the initiation, titration, and switching of basal insulins: a narrative review for primary care. Ann Med. 2021;53(1):998-1009. doi:10.1080/07853890.2021.1925148.

Cheng

AYY

Patel

Reid

Wyne

. Differentiating basal insulin preparations: understanding how they work explains why they are different. Adv Ther. 2019;36(5):1018-1030. doi:10.1007/s12325-019-00925-6.

Lindauer

Becker

. Insulin depot absorption modeling and pharmacokinetic simulation with insulin glargine 300 U/mL. Int J Clin Pharmacol Ther. 2019;57(1):1-10. doi:10.5414/CP203269.

Porcellati

Lucidi

Candeloro

, et al. Pharmacokinetics, pharmacodynamics, and modulation of hepatic glucose production with insulin glargine U300 and glargine U100 at steady state with individualized clinical doses in type 1 diabetes. Diabetes Care. 2019;42(1):85-92. doi:10.2337/dc18-0706.

Lamos

Younk

Davis

. Concentrated insulins: the new basal insulins. Ther Clin Risk Manag. 2016;12:389-400. doi:10.2147/TCRM.S99855.

Davis

Fleming

Malinowski

Brown

Fleming

. Ultra-long-acting insulins: a review of efficacy, safety, and implications for practice. J Am Assoc Nurse Pract. 2018;30(7):373-380. doi:10.1097/JXX.0000000000000076.

10.

Riddle

Bolli

Ziemen

, et al. New insulin glargine 300 Units/mL versus glargine 100 Units/mL in people with type 2 diabetes using basal and mealtime insulin: glucose control and hypoglycemia in a 6-month randomized controlled trial (EDITION 1). Diabetes Care. 2014;37(10):2755-2762. doi:10.2337/dc14-0991.

11.

Yki-Järvinen

Bergenstal

Ziemen

, et al. New insulin glargine 300 Units/mL versus glargine 100 Units/mL in people with type 2 diabetes using oral agents and basal insulin: glucose control and hypoglycemia in a 6-month randomized controlled trial (EDITION 2). Diabetes Care. 2014;37(12):3235-3243. doi:10.2337/dc14-0990.

12.

Bolli

Riddle

Bergenstal

, et al. New insulin glargine 300 U/ml compared with glargine 100 U/ml in insulin-naïve people with type 2 diabetes on oral glucose-lowering drugs: a randomized controlled trial (EDITION 3). Diabetes Obes Metab. 2015;17(4):386-394. doi:10.1111/dom.12438.

13.

Home

Bergenstal

Bolli

, et al. New insulin glargine 300 Units/mL versus glargine 100 Units/mL in people with type 1 diabetes: a randomized, phase 3a, open-label clinical trial (EDITION 4). Diabetes Care. 2015;38(12):2217-2225. doi:10.2337/dc15-0249.

14.

Toujeo (Insulin Glargine) [Package Insert]. Bridgewater, NJ: Sanofi; 2015. Revised February 2015.

15.

Heise

Nørskov

Nosek

Kaplan

Famulla

Haahr

. Insulin degludec: lower day-to-day and within-day variability in pharmacodynamic response compared with insulin glargine 300 U/mL in type 1 diabetes. Diabetes Obes Metab. 2017;19(7):1032-1039. doi:10.1111/dom.12938.

16.

Lantus (Insulin Glargine) [Package Insert]. Bridgewater, NJ: Sanofi; 2000. Revised May 2019.

17.

Capson

Cade

Avanesyan

. Detemir vs glargine: comparison of inpatient glycemic control. J Am Osteopath Assoc. 2019;119(2):89-95. doi:10.7556/jaoa.2019.014.

18.

Takayama

Yamauchi

Aizawa

. Quantification of insulin. Diabet Med. 2014;31(3):375-376. doi:10.1111/dme.12337.

19.

Knopp

Holder-Pearson

Chase

. Insulin units and conversion factors: a story of truth, boots, and faster half-truths. J Diabetes Sci Technol. 2019;13(3):597-600. doi:10.1177/1932296818805074.

20.

D’Souza

Kruger

. Considerations for insulin-treated type 2 diabetes patients during hospitalization: a narrative review of what we need to know in the age of second-generation basal insulin analogs. Diabetes Ther. 2020;11(12):2775-2790. doi:10.1007/s13300-020-00920-z.

21.

Pasquel

Lansang

Khowaja

, et al. A randomized controlled trial comparing glargine U300 and glargine U100 for the inpatient management of medicine and surgery patients with type 2 diabetes: glargine U300 hospital trial. Diabetes Care. 2020;43(6):1242-1248. doi:10.2337/dc19-1940.

22.

Hirsch

Draznin

. Transition of patients to and from insulin degludec: a clinical challenge. J Clin Endocrinol Metab. 2020;105(6):dgz280. doi:10.1210/clinem/dgz280.

23.

Mehta

Kellum

Shah

, et al. Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Crit Care. 2007;11(2):R31. doi:10.1186/cc5713.

24.

Gosmanov

. A practical and evidence-based approach to management of inpatient diabetes in non-critically ill patients and special clinical populations. J Clin Transl Endocrinol. 2016;5:1-6. doi:10.1016/j.jcte.2016.05.002.

25.

Davidson

Hebblewhite

Steed

Bode

. Analysis of guidelines for basal-bolus insulin dosing: basal insulin, correction factor, and carbohydrate-to-insulin ratio. Endocr Pract. 2008;14(9):1095-1101. doi:10.4158/EP.14.9.1095.

26.

Wallace

McFarland

. Comparing dosing of basal insulin analogues detemir and glargine: is it really unit-per-unit and dose-per-dose? Ann Pharmacother. 2014;48(3):361-368. doi:10.1177/1060028013518420.

27.

Pratiwi

Mokoagow

Made Kshanti

Soewondo

. The risk factors of inpatient hypoglycemia: a systematic review. Heliyon. 2020;6(5):e03913. doi:10.1016/j.heliyon.2020.e03913.