a b s t r a c t

Introduction: Hypoglycemia is a common adverse effect when intravenous (IV) insulin is administered for hyperkalemia. A prolonged infusion of dextrose 10% (D10) may mitigate hypoglycemia compared to dextrose 50% (D50) bolus. Our objective was to evaluate whether D10 infusion is a safe and effective alternative to D50 bolus for hypoglycemia prevention in hyperkalemic patients receiving IV insulin.

Methods: We conducted a retrospective review of patients >= 18 years who presented to the emergency depart- ment (ED) with hyperkalemia (K⁺ N 5.5) and received IV insulin and D10 infusion or D50 bolus within 3 h. The primary endpoint was incidence of hypoglycemia, defined as blood glucose (BG) <= 70 mg/dL, in the 24 h fol- lowing IV insulin administration for hyperkalemia.

Results: A total of 134 patients were included; 72 in the D50 group and 62 in the D10 group. There was no differ- ence in incidence of hypoglycemia between the D50 and D10 groups (16 [22%] vs. 16 [26%], p = 0.77). Symptom- atic hypoglycemia, Severe hypoglycemia, and hyperglycemia rates in the D50 and D10 groups were [5 (7%) vs. 2 (3%), p = 0.45], [5 (7%) vs. 1 (2%), p = 0.22], and [34 (47%) vs. 23 (37%), p = 0.31] respectively. Low initial BG was a predictor for developing hypoglycemia.

Conclusions: In our study, D10 infusions appeared to be at least as effective as D50 bolus in preventing hypogly- cemia in hyperkalemic patients receiving IV insulin. In context of ongoing D50 injection shortages, D10 infusions should be a therapeutic strategy in this patient population.

(C) 2019

Introduction

Hyperkalemia is a potentially life-threatening metabolic abnormal- ity [1]. Insulin, which shifts K⁺ intracellularly, is administered in up to 64.4% of cases of Severe hyperkalemia to temporarily reduce serum levels [2,3]. The reported incidence of hypoglycemia when insulin is ad- ministered in this setting is variable, ranging from 8.7 to 75% [4-6]. For severe hyperkalemia (K⁺ N 6.5 mmol/L), the American Heart

Abbreviations: IV, Intravenous; D10, Dextrose 10% in water; D50, Dextrose 50% in water; BG, Blood glucose; ED, Emergency Department; K⁺, Potassium; AHA, American Heart Association; CPOE, Computerized Physician Order entry; SCr, Serum Creatinine; CrCl, creatinine clearance; REDCap, Research Electronic Data Capture; SPSS, Statistical Package for Social Sciences; IQR, Interquartile ranges.

* Corresponding author at: Emergency Medicine Clinical Pharmacotherapy Specialist, 545 1st Avenue, GBH-SC2-097, New York, NY 10016, USA.

E-mail addresses: [email protected] (I. Yang), [email protected] (S. Smalley), [email protected] (T. Ahuja),

[email protected] (C. Merchan), [email protected] (S.W. Smith), [email protected] (J. Papadopoulos).

¹ Current address: Department of Pharmacy, Hackensack University Medical Center, 30 Prospect Avenue, Hackensack, NJ 07601, USA.

Association (AHA) recommends 10 units of IV regular insulin with 50 mL of D50 [7]. However, to our knowledge, studies have not been conducted to evaluate the effects of different dextrose concentrations and infusion durations on hypoglycemia rates in management of hyperkalemic patients.

Hypoglycemia is a known adverse effect based on the pharmacolog- ical profiles of IV insulin and D50. The contrasting durations of effect be- tween IV insulin and D50 bolus may contribute to hypoglycemia in hyperkalemic patients. The effects of IV insulin begin in approximately 10-15 min and terminate at a median of about 4 h (range: 1-5 h). How- ever, blood glucose (BG) levels can return to baseline after 30 min fol- lowing a D50 bolus [8,9]. In addition, peripheral administration of D50, a hyperosmolar medication, may be associated with phlebitis and extravasation injuries [9]. Use of a lower osmolarity agent, such as D10, may decrease these adverse effects.

From a drug shortage perspective, D50 injections are commonly un- available from wholesalers, forcing clinicians to rely on alternative or contingency hypoglycemic management strategies such as D10 infu- sions. Beginning April 2017, our health system was unable to obtain suf- ficient supply of D50 injections from our wholesaler supplier. Subsequently, we implemented an alternative use alert in our

https://doi.org/10.1016/j.ajem.2019.09.003

0735-6757/(C) 2019

computerized physician order entry (CPOE; Epic, Epic Systems Corpora- tion, Verona, Wisconsin) system, recommending D10 infusion over D50. Therefore, we sought to evaluate whether D10 infusion is a safe and ef- fective alternative to D50 bolus for prevention of hypoglycemia in hyperkalemic patients treated with IV insulin.

Methods

Study design and setting

This was an institutional review board approved, retrospective re- view conducted at three urban emergency departments (ED) within a single health-system. Due to the retrospective nature of this review, in- formed consent was not required by the institutional review board. Chart reviews were conducted by a pharmacy resident and overseen by three clinical pharmacy specialists. The pharmacy resident was trained with our data collection tool prior to commencement of data ab- straction; periodic monitoring for accuracy of data collection was com-

Primary and Secondary Outcomes

The primary endpoint was the incidence of hypoglycemia, defined as BG <= 70 mg/dL in the 24 h following IV insulin administration for hyperkalemia. Secondary endpoints included the incidences of severe hypoglycemia, symptomatic hypoglycemia, hyperglycemia, use of res- cue agents for the management of hypoglycemia, and extravasation. Se- vere hypoglycemia was defined as a BG <= 40 mg/dL. Symptomatic hypoglycemia was defined as documented symptoms in the EHR attrib- uted to hypoglycemia and treated with a carbohydrate or rescue agent, regardless of BG level. Rescue agents were defined as glucose gel, glu- cose tablets, carbohydrates in the form of snacks or a meal, intramuscu- lar glucagon, D10 infusion or D50 bolus with the intent to treat hypoglycemia. Hyperglycemia was defined as a BG >= 180 mg/dL.

Table 1

Baseline characteristics and past medical history.

pleted by a clinical pharmacist.	D50 Bolus	D10 Infusion	p-Value
	(n = 72)	(n = 62)

Selection of patients

electronic health records (EHRs) were used to identify patients who presented to the ED and received IV insulin with concomitant dextrose for the management of hyperkalemia, defined as a K⁺ N 5.5. At our insti- tution, there is no standardized protocol for the management of hyperkalemia. IV insulin is typically administered with 25-50 g of dex- trose, as either a D50 bolus or D10 intermittent infusion. Inclusion criteria consisted of patients >=18 years of age who received a D50 bolus or a D10 infusion lasting a minimum of 30 min within 3 h of a noted K⁺ N 5.5 mmol/L between August 20, 2016 and August 20, 2018. Our D50 injection shortage began in April 2017. Patients who re- ceived >=25 g of dextrose for hypoglycemia prevention received all dex- trose as one single dose without spaced dosing. Although providers were routed to select D10 in our CPOE during our D50 shortage, this alert could be overridden leading to more patients receiving D50 com- pared to D10 in our study period. Therefore, D50 patients were selected in a 1:3 ratio utilizing a random number generator to keep a balanced sample size. Exclusion criteria included receipt of a D50 bolus and D10 infusion simultaneously or a lack of at least 1 follow-up BG. Data collec- tion included baseline characteristics, administered doses of insulin and dextrose, medications with dextrose-containing diluents, and labora- tory values including serum creatinine (SCr), K⁺ levels and BG levels. Creatinine clearance (CrCl) was calculated using the Cockcroft-Gault equation [10]. Data was managed with REDCap (version 8.10.18; Re- search Electronic Data Capture, Nashville, Tennessee) [11].

Fig. 1. Flow chart of patient enrollment

Demographics

Age (yr), median (IQR) 71 (61, 78) 67 (57, 76) 0.243

Male 46 (64) 40 (65) 0.94

Weight (kg), median (IQR) 74 (61, 90) 71 (62, 84) 0.906

Height (cm), median (IQR) 168 (160, 173) 167 (163, 174) 0.641

Hospital admission 67 (93) 51 (83) 0.055

Past Medical History Renal Function

SCr at time of initial K⁺ (mg/dL) 3.3 (1.9, 8.0) 3.1 (1.7, 6.3) 0.18

CrCl (mL/min) 13 (8, 29) 21 (8, 41) 0.085

CrCl <= 30 mL/min 56 (78) 42 (68) 0.191

ESRD 19 (26) 15 (24) 0.771

Hemodialysis 14 (19) 14 (23) 0.656

AKI* 48 (67) 38 (61) 0.589

Diabetes 37 (51) 23 (37) 0.097

Type I 1 (1) 1 (2) 0.915

Type II 34 (47) 19 (31) 0.050

Not stated 2 (3) 3 (5) 0.662

Heart failure 25 (35) 18 (29) 0.482

HFrEF 11 (15) 8 (13) 0.694

HFpEF 14 (19) 8 (13) 0.308

NPO at time of hyperkalemia 17 (24) 8 (13) 0.113

Prior To Admission Medications Causing Hypoglycemia Diabetes Medications

Medication route

Oral	13 (18)	11 (18)	0.276
IV	13 (18)	7 (11)	0.962
Combination Oral and IV	6 (8)	2 (3)	0.285

Medication classes

Insulin	19 (26)	13 (21)	0.544
Sulfonylurea	6 (8)	2 (3)	0.285
Metformin	9 (13)	3 (5)	0.121
Thiazolidinedione	1 (1)	1 (2)	1
DPP4 Inhibitors	11 (15)	4 (7)	0.106
GLP-1 Agonists	1 (1)	0 (0)	1
Meglinitides	2(3)	1 (2)	1
SGLT-2 Inhibitors	0 (0)	0 (0)
Alpha-glucosidase inhibitors	0 (0)	0 (0)
Pramlinitide	0 (0)	0 (0)

Other medications which can cause hypoglycemia

Pentamidine	1 (1)	0 (0)	1
Sulfamethoxazole-Trimethoprim	1 (1)	4 (6)	0.182

All data presented as (n, %) unless otherwise noted.

AKI defined as: increase in SCr by >=0.3 mg/dL or >=1.5 times baseline.

IQR: interquartile range; CrCl: creatinine clearance; ESRD: end stage renal disease; AKI: acute kidney injury; HFrEF: heart failure with reduced ejection fraction; HFpEF: heart fail- ure with preserved ejection fraction; NPO: nothing by mouth; DPP4: dipeptidyl peptidase- 4; GLP-1: glucose-dependent insulinotropic peptide; SGLT-2: sodium-glucose cotransporter-2.

Adverse events included those adverse events documented in the EHR, such as extravasation, attributed to dextrose administration.

Statistical analysis

Statistical analysis was performed using SPSS (software for Win- dows, version 25.0; Statistical Package for Social Sciences, Chicago, Illi- nois) comparing patients who received D50 with those who received D10. Categorical variables were described as frequencies and propor- tions and compared using Chi-square test or Fisher’s exact test. Contin- uous variables were described as medians with interquartile ranges (IQR) and analyzed using Mann-Whitney U test. Timing of hypoglyce- mic episodes was compared using a paired-samples Sign test. A two- sided alpha of b 0.05 was considered statistically significant. A univari- ate analysis was conducted to determine predictors of hypoglycemia. Baseline characteristics including renal function, insulin doses and dex- trose doses were included in the univariate analysis. A power analysis was not performed due to the lack of previous similar studies.

Results

Characteristics of study participants

A total of 140 patients presenting to the ED with a K⁺ N 5.5 mmol/L who received insulin and dextrose were screened for inclusion. Two were excluded for lack of at least 1 follow-up BG; 4 were excluded for initial simultaneous administration of D50 and D10. Therefore, 134 pa- tients were included for statistical analysis (n = 72, D50; n = 62, D10; Fig. 1). There were no differences in baseline characteristics (Table 1). Most patients in the D50 and D10 groups had renal dysfunction, with an elevated SCr (3.3 mg/dL [IQR 1.9, 8.0] vs. 3.1 mg/dL [IQR 1.7, 6.3], p = 0.18), respectively. Prior to insulin administration, patients receiv- ing D10 had lower median BG versus those receiving D50 (112 mg/dL [IQR 89, 133] vs. 123 mg/dL [IQR 100, 177], p = 0.023). The median ini- tial dextrose dose was 25 g in both groups (25 g [IQR 25, 50] vs. 25 g [IQR 25, 25], p = 0.022), with a median D10 infusion rate of 250 mL/hr (IQR 250, 625) (Table 2).

Primary outcome

No difference was observed in the incidence of hypoglycemia be- tween the D50 and D10 groups (16 [22%] vs. 16 [26%], p = 0.769; Fig. 2). The median incidence of hypoglycemic events per patient was numerically higher in the D50 group (2 [IQR 1, 3] vs. 1 [IQR 1, 2]; p =

Fig. 2. Incidence of hypoglycemia

0.168). The median time to the first hypoglycemic event occurred ear- lier in the D50 group (3.8 h [IQR 1.3, 16] vs. 6.8 [IQR 1.6, 8.9]; p =

0.873) (Table 3, Fig. 3).

Secondary outcomes

Symptomatic hypoglycemia, severe hypoglycemia, and hypergly- cemia rates between D50 and D10 were: 5 (7%) vs. 2 (3%), 5 (7%) vs. 1 (2%), and 34 (47%) vs. 23 (37%), respectively. Incidence and timing of rescue agent administration were similar in the D50 and D10 groups ([15% vs. 16% patients, p = 1], [3.8 vs. 4.0 h, p = 0.515])

(Table 3).

Univariate analysis for predictors of hypoglycemia

In the univariate analysis, baseline BG <= 100 mg/dL was identified as a risk factor for developing hypoglycemia (OR 4.90, 95% CI 2.09-11.49). More patients who experienced hypoglycemia were >= 60 years old com- pared to those who did not (21 [64%] vs. 12 [12%]; p = 0.195). We did not see any differences in rates of hypoglycemia when examining base- line characteristics including weight, gender, diabetes status, renal func- tion, initial dose of insulin or dextrose administered (Table 4).

Table 2

IV insulin and IV dextrose doses and administrations.

	D50 Bolus (n = 72)	D10 Infusion (n = 62)	p-Value
Potassium (mmol/L) Initial K+	6.4 (6.2, 7.1)	6.4 (6.1, 6.9)	0.621
Repeat K+ after insulin treatment	5.5 (5.0, 6.4)	5.6 (5.0, 6.1)	0.994
IV Insulin Regular Dose
For initial hyperkalemia treatment (units)	8 (5, 10)	9 (5, 10)	0.995
For initial hyperkalemia treatment (units/kg)	0.11 (0.07, 0.13)	0.10 (0.07, 0.14)	0.995
Patients receiving repeat insulin for hyperkalemia treatment (n, %)	15 (21)	16 (26)	0.542
For repeat hyperkalemia treatment (units)	10 (7.5, 10)	9 (5, 10)	0.281
For repeat hyperkalemia treatment (units/kg)	0.11 (0.08, 0.12)	0.11 (0.06, 0.14)	0.74
Time to repeat insulin administration (hours)	7.8 (5.1, 9.3)	6.5 (4.9, 13.2)	0.83
Total insulin administered for hyperkalemia treatment in 24 h (units)	10 (5, 10)	10 (5, 10)	0.574
Total insulin administered for hyperkalemia treatment in 24 h (units/kg)	0.12 (0.07, 0.16)	0.12 (0.08, 0.18)	0.643
Dextrose Administration
Dextrose administered during initial hyperkalemia treatment (g)	25 (25, 25)	25 (25, 50)	0.022
Total dextrose administered during hyperkalemia treatment in 24 h (g)	25 (25, 50)	50 (25, 50)	0.075
Median D10 infusion rate (mL/hr)		250 (250, 625)

All data presented as median (IQR) unless otherwise specified. IQR: interquartile range; kg: kilogram.

Discussion

In our study, we found the incidence of hypoglycemia for the man- agement of hyperkalemia to be 24% at 24 h following IV insulin adminis- tration. The reported incidence of hypoglycemia with IV insulin for the management of hyperkalemia in the ED predominantly ranges from 8.7% to 17% at 4 to 8 h post insulin administration [4-6]. This may explain the high rates of hypoglycemia in our study as we may have captured de- layed hypoglycemia events from IV insulin, which other studies did not. Our incidence of hypoglycemia within 8 h of insulin administration was 13%, which is similar to the rates reported in literature [4,5,12,13].

We aimed to identify patient characteristics for the risk of hypogly- cemia. Similar to a recent study, we found low pre-treatment BG to be a predictor of hypoglycemia [13]. We also found age >=60 years to be a predictor of hypoglycemia. Although other studies have found higher doses of insulin, no history of diabetes mellitus, Female gender, abnor- mal renal function and low body weight to be predictive of hypoglyce- mia, we did not observe these findings in our limited sample size [4,5, 12,14].

The ideal strategy to decrease hypoglycemic events in hyperkalemic patients receiving IV insulin has evaded clinicians to date. One study ex- ploring dextrose dosing reported lower rates of hypoglycemia 1 h after IV insulin administration comparing 50 g versus 25 g, but no difference at 4 h [15]. However, limited data exists on alternative methods of dex- trose administration in the hyperkalemia setting. Other authors have

Table 3

Blood glucose, incidence of hypoglycemia and safety.

	D50 Bolus (n = 72)	D10 Infusion (n = 62)	p-Value
BG prior to insulin administration
BG (mg/dL)	123 (100, 177)	112 (89, 133)	0.023
Baseline hypoglycemia* (n, %)	2 (3)	2 (3)	1
Baseline BG <= 100 mg/dL	18 (25)	26 (42)	0.041
Time from BG to insulin	0.8 (0.4, 1.6)	1.3 (1.6, 2.0)	0.528

administration (hours)

First BG post insulin administration

BG (mg/dL) 120 (104, 190) 135 (96, 202) 0.918

retrospectively evaluated patients receiving continuous infusions of D10 500 mL over 4 h for the management of hyperkalemia and reported hypoglycemia in 6.1%, suggesting D10 to be a safe and effective modality [12]. However, all patients received 10 units of rapid acting insulin and no comparator arm was present. Other existing literature utilizing D10 infusions for hypoglycemia management is not specific to hyperkalemia and derived from the prehospital setting [16].

To our knowledge, our study is the first to compare a D10 infusion to a D50 bolus strategy for hypoglycemia prevention in patients receiving IV insulin for hyperkalemia.

We aimed to compare these approaches due to the frequent drug shortages of D50 injections. The Joint Commission recommends identi- fication of an appropriate Alternative medication as a potential strategy to combat drug shortages [17]. Starting April 2017, due to continued shortages of D50, our institution began utilizing D10 infusions as an al- ternative agent for preventing hypoglycemia in the treatment of hyperkalemia with IV insulin. In addition to being a potential viable al- ternative to D50 injections, there are several theoretical advantages to utilizing D10 infusions. Besides prolonging the anti-hypoglycemic ef- fects of dextrose, D10 infusion may be associated with decreased stim- ulation of additional insulin release compared to D50 bolus. When D50 is used specifically for the treatment of sulfonylurea-induced hypo- glycemia, adverse insulin release may be induced as excess glucose sup- presses gluconeogenesis and glycogenolysis [18]. While this mechanism may be beneficial for the management of hyperkalemia, it could poten- tially result in rebound hypoglycemia. Furthermore, there is a lower risk of thrombophlebitis and extravasation with D10 as the osmolarity of D10 is 506 mOsm/L, as compared to D50 which has an osmolarity of 2525 mOsm/L. Reported cases of D50 extravasation have been severe, with some patients requiring treatment with hyaluronidase and emer- gent fasciotomy [19,20].

Limitations

There are several limitations applicable to our study. First, this was a retrospective chart review at a single-center on the basis of an unantic- ipated drug shortage. A power analysis was not performed due to the

First BG <= 100 mg/dL, (n, %)	17 (24)	18 (29)	0.476	limited information available from previous studies. We had a small
Time to first BG check post insulin	1.3 (0.7, 3.2)	1.6 (1.0, 2.4)	0.899	sample size and this may have limited the ability to find a difference
(hours) Primary Outcome Incidence of hypoglycemia (n, %)	16 (22)	16 (26)	0.769	in hypoglycemic rate. No institutional protocol for management of hyperkalemia or hypoglycemia in this context existed at the time of our study; therefore, prescriber insulin dosing and hypoglycemia treat-
Time of first hypoglycemic event	3.8 (1.3, 16)	6.8 (1.6, 8.9)	0.873	ment was not constrained and BG checks were not performed in a stan-

dardized fashion. Patients who received D10 during initial hyperkalemia management may have received D50 for hypoglycemia treatment (and vice-versa) within 24 h. Lastly, report of symptomatic hypoglycemia was based on chart review and subject to Accurate documentation.

(hours)
Hypoglycemia events per patient	2 (1, 3)	1 (1, 2)	0.168
Patients with more than one	10 (14)	7 (11)	0.652
hypoglycemia event (n, %)
Secondary Outcomes Symptomatic hypoglycemia (n, %)	5 (7)	2 (3)	0.45
Severe hypoglycemia (n, %)	5 (7)	1 (2)	0.216
Hyperglycemia (n, %)	34 (47)	23 (37)	0.307
Other adverse events, i.e.,	0 (0)	0 (0)
extravasation (n, %)
Rescue items given (n, %)	11 (15)	10 (16)	1
	D50 infusion	D10 bolus
Glucose tablets (n, %)	(n = 11)	(n = 10)	1
Snack or meal (n, %)	1 (9)	0 (0)	0.249
Glucagon (n, %)	2 (18)	5 (50)	0.463
D50 bolus (n, %)	0 (0)	1 (10)	0.339
g, median	7 (64)	3 (3)
D10 infusion (n, %)	50 (50, 50)	25 (25, 25)	1
g, median	2 (18)	1 (10)
Time to initial rescue (hours)	50 (50, 271)	50	0.515
Number of interventions for each	3.8 (2.1, 13.8)	4.0 (2.1, 10.7)	0.606
hypoglycemia encounter	1 (1, 1)	1 (1, 1.3)

All data presented as median (IQR) unless otherwise noted. IQR: interquartile range; BG: blood glucose; g: grams.

^* Hypoglycemia defined as BG <= 70 mg/dL. Fig. 3. Timing of hypoglycemic episodes

Table 4

Univariate predictors of hypoglycemia.

	Hypoglycemia (n = 33)	No hypoglycemia (n = 101)	p-Value
Baseline Characteristics Age >= 60 years	21 (64)	12 (12)	0.195
Female	11 (33)	37 (37)	0.731
Weight <= 60 kg	23 (70)	81 (80)	0.209
Hemodialysis	7 (21)	21 (21)	0.959
Diabetes	14 (42)	46 (46)	0.754
Type I	1 (3)	1 (1)	0.401
Type II	11 (33)	42 (42)	0.400
Heart failure	9 (27)	34 (34)	0.495
HFrEF	6 (18)	13 (13)	0.448
HFpEF	2 (6)	20 (20)	0.064
NPO at time of hyperkalemia	6 (18)	19 (19)	0.936

BG <= 100 mg/dL 17 (52) 18 (18) 0.001 OR

4.90, 95% CI

2.09-11.49)

Declaration of Competing Interest

The authors declared that there is no conflict of interest.

References

1. Pierce DA, Russell G, Pirkle Jr JL. Incidence of hypoglycemia in patients with Low eGFR treated with insulin and dextrose for hyperkalemia. Ann Pharmacother 2015;49(12):1322-6.
2. Singer AJ, Thode Jr HC, Peacock WF. A retrospective study of emergency department potassium disturbances: severity, treatment, and outcomes. Clin Exp Emerg Med 2017;4(2):73-9.
3. Li T, Vijayan A. Insulin for the treatment of hyperkalemia: a double-edged sword? Clin Kidney J 2014;7(3):239-41.
4. Schafers S, Naunheim R, Vijayan A, Tobin G. Incidence of hypoglycemia following insulin-based acute stabilization of hyperkalemia treatment. J Hosp Med 2012;7 (3):239-42.
5. Apel J, Reutrakul S, Baldwin D. Hypoglycemia in the treatment of hyperkalemia with insulin in patients with end-stage renal disease. Clin Kidney J 2014;7(3):248-50.
6. Estep P, Efird LE. Evaluation of hypoglycemia incidence and risk factors in patients treated with IV insulin aspart for hyperkalemia. Endocrinol Diab Res 2015;1:1.
7. Vanden Hoek TL, Morrison LJ, Shuster M, et al. Part 12: cardiac arrest in special sit- uations: 2010 American Heart Association Guidelines for Cardiopulmonary Resusci-

   Renal function CrCl <= 30 mL/min	27 (82)	71 (71)	0.195	tation and Emergency Cardiovascular Care. Circulation 2010;122(18 Suppl 3): S829-61. [8] Hipszer B, Joseph J, Kam M. Pharmacokinetics of Intravenous insulin delivery in
   AKI	22 (67)	64 (64)	0.731	humans with type 1 diabetes. Diab Technol Ther 2005;7(1):83-93.
   ESRD	10 (30)	24 (24)	0.453	[9] Reynolds PM, MacLaren R, Mueller SW, Fish DN, Kiser TH. Management of extrava-
   Insulin Administration Initial insulin dose >= 0.1	20 (61)	53 (53)	0.415	sation injuries: a focused evaluation of noncytotoxic medications. Pharmacotherapy 2014;34(6):617-32. [10] Botev R, Mallie JP, Couchoud C, Schuck O, Fauvel JP, Wetzel JF, et al. Estimating glo-
   units/kg				merular filtration rate: Cockcroft-Gault and Modification of Diet in Renal Disease
   Total insulin dose >= 0.1	21 (64)	67 (67)	0.777	formulas compared to renal inulin clearance. Clin J Am Soc Nephrol 2009;4(5):
   units/kg				899-906.
   Patients receiving repeat	6 (18)	25 (25)	0.437	[11] Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic
   insulin doses				data capture (REDCap)-a metadata-driven methodology and workflow process for
   				providing translational research informatics support. J Biomed Inform 2009;42(2):
   Dextrose Administration				377-81.

   Initial dextrose dose <= 25 g 25 (76) 68 (67) 0.362

   All data presented as (n, %).

   AKI defined as: increase in SCr by >=0.3 mg/dL or >=1.5 times baseline.

   OR: odds ratio; CI: confidence interval; CrCl: creatinine clearance; AKI: acute kidney in- jury; BG: blood glucose; ESRD: end stage renal disease; HFpEF: heart failure with pre- served ejection fraction; HFrEF: heart failure with reduced ejection fraction; kg: kilograms; mg: milligrams; NPO: nothing by mouth.

   Conclusion

   Hypoglycemia in the setting of insulin administration for hyperkalemia management remains problematic. We found D10 infu- sion to be an acceptable alternative to bolus D50 injection for hypogly- cemia prevention. Although the optimum total glucose dose remains to be determined, D10 infusion may lead to fewer incidents of dysglycemia. D10 administration warrants consideration, especially in light any reported D50 injection shortages. The ideal D10 infusion rate and duration to minimize hypoglycemia should continue to be ex- plored. Patients with low initial BG are at higher risk of hypoglycemia and require more aggressive glucose treatment and monitoring.

   Funding

   This research did not receive any specific grant from funding agen- cies in the public, commercial or not-for-profit sectors

   Coca A, Valencia AL, Bustamante J, Mendiluce A, Floege J. Hypoglycemia following in- travenous insulin plus glucose for hyperkalemia in patients with impaired renal function. PLoS ONE 2017;12(2):e0172961.

8. Jacob BC, Peasah SK, Chan HL, Niculas D, Shogbon Nwaesei A. Hypoglycemia associ- ated with insulin use during treatment of hyperkalemia among emergency depart- ment patients. Hosp Pharm 2019;54(3):197-202.
9. Moussavi K, Fitter S, Gabrielson SW, Koyfman A, Long B. Management of hyperkalemia with insulin and glucose: pearls for the emergency clinician. J Emerg Med 2019. https://doi.org/10.1016/j.jemermed.2019.03.043.
10. Farina N, Anderson C. Impact of dextrose dose on hypoglycemia development fol- lowing treatment of hyperkalemia. Ther Adv Drug Saf 2018;9(6):323-9.
11. Moore C, Woollard M. Dextrose 10% or 50% in the treatment of hypoglycaemia out of hospital? A randomised controlled trial. Emerg Med J 2005;22(7):512-5.
12. The Joint Commission. medication management (MM) (critical access hospitals/crit- ical access hospitals). medication selection and procurement-managing medication shortages. What are The Joint Commission requirements for managing medication shortages? 2017. https://www.jointcommission.org/standards_information/ jcfaqdetails.aspx?StandardsFAQId=1530&StandardsFAQChapterId= 27&ProgramId=0&ChapterId=0&IsFeatured=False&IsNew=False&Keyword= [accessed May 16, 2019].
13. McLaughlin SA, Crandall CS, McKinney PE. Octreotide: an antidote for sulfonylurea- induced hypoglycemia. Ann Emerg Med 2000;36(2):133-8.
14. Lawson SL, Brady W, Mahmoud A. Identification of highly concentrated dextrose so- lution (50% dextrose) extravasation and treatment-a clinical report. Am J Emerg Med 2013;31(5) 886.e3-5.
15. Chinn M, Colella MR. Prehospital Dextrose extravasation causing forearm compart- ment syndrome: a case report. Prehosp Emerg Care 2017;21(1).