1

Pre-Screening for Hypoglycemic Events in Intensive Care Units

Sarah Bachmeier, SN, Maury Burke, SN, and Julia Vetter, SN

Department of Nursing, University of Mary

NUR 435: Research and Evidence Based Practice

Professor Sarah Berreth, MSN, RN

Pre-Screening for Hypoglycemic Events in Intensive Care Units

Hypoglycemia is no stranger in the healthcare world. When a patient has hypoglycemia,

eight ounces of orange juice is given as well as carbohydrates to bring up their blood glucose to

the normal range of 70-100 mg/dL. If the patients’ blood sugar reaches this range, typically

nurses go about their day. However, further investigation to find out why this event occurred and

the risk for it to happen again, is not typically investigated. Hypoglycemic events are common in

inpatient healthcare settings and can happen on any unit, from the newborn nursery to the

intensive care unit. These events directly affect patients’ health status and are of utmost

importance for nurses to know and understand. The article review, Inpatient Hypoglycemia: The

Challenge Remains, states “A recent study reported the prevalence of threshold hypoglycemia

and SH [severe hypoglycemia] to be 10.1% and 1.9% respectively” (Cruz, 2020, p. 560). As

nurses, we need to investigate why hypoglycemia is occurring in our patients. A risk assessment

for patients admitted to the ICU would benefit not only the patients, but the nurses as well. It

could directly affect how patient care is being delivered and improve patient outcomes. There are

obstacles that will need to be overcome, such as understanding the challenge of detecting

hypoglycemia (Fitzgerald et al. 2021). Throughout this paper we will discuss the history and

evolution, the clinical significance of hypoglycemia, our clinical question, evidence-based

literature relating to hypoglycemia, what was learned, recommendations that will impact the

clinical world, and how to create change.

History and Evolution of Topic

Hypoglycemic events are a healthcare concern that has always been present. According

to the article review Inpatient Hypoglycemia: The Challenge Remains, hypoglycemia is one of

the most common complications in an inpatient setting (Cruz, 2020). It can cause effects, such as
 3

“increased morbidity and mortality, increased length of stay, increased readmission rate, and

increased cost” (Cruz, 2020, p. 560). Hypoglycemia in intensive care unit (ICU) inpatient

settings has a high prevalence, “Studies using tight control (BG 80-100 mg/dL) in critically ill

patients have shown an incidence as high as 45%” (Cruz, 2020, p. 560). 

The standard of preventing hypoglycemia has not changed significantly through the

years. In practice, trying to reach the target goal blood glucose of 140-180 mg/dL is, and has

been, the recommendation. Other strategies are “Standardized reporting, order sets, and

hypoglycemia protocols are highly effective to reduce inpatient hypoglycemia” (Cruz, 2020, p.

561). However, one thing that has been changing is the equipment and technology that we have

now that may be able to aid in screening patients for hypoglycemic risk. 

An important piece in learning about the evolution of our topic was talking to current

ICU nurses to see what policies and guidelines are standard practice. In interviewing an ICU

nurse, E. Morgan (personal communication, October 27, 2022) mentioned that upon admission

all nurses are instructed and required to get a blood glucose level. From there, they follow their

facility and department specific sliding scale which she stated has not changed since she started

six years ago. At this hospital the ICU nurses have the autonomy to give six units less or six units

more insulin than required. This requires them to use clinical judgment and could be a risk to the

patient if they give too little or too much. Morgan stated that as a new nurse when a patient was

admitted she only looked at the blood glucose. As she gained experience, she would look at

factors that could indicate a possible hypoglycemic event but stated there is no universal risk

assessment upon admission. This shows a gap in the standard of care because although there are

factors the nurse may look at due to having experience in the field, there is no standard of care to

assess risk factors other than a blood glucose reading. 

Clinical Significance

Hypoglycemia affects all patient populations outcomes whether mildly, with patients

only needing juice to bring up their blood sugar, or drastically, becoming unresponsive. No

matter the area of nursing, hypoglycemia needs to be understood as it is clinically significant to

care. We chose this topic because each of us have dealt with hypoglycemia as a certified nursing

assistant and in our nursing internships. Additionally, we all have seen the impact hypoglycemia

can have on patient recovery within the hospital. One patient that comes to mind fell after

standing up and experiencing low blood sugar. The patient then had to get an x-ray of their hip

because they had hip surgery. This fall caused the patient’s rehab to last longer when the fall

could have been prevented altogether. Our interest in the topic of hypoglycemic events and the

detrimental impact they can have on ICU patients led us to wonder if there is something that

could be implemented into the inpatient setting to reduce these events. After reading research

studies and talking to current staff in the intensive care unit, we created our PICOT research

question: Does pre-screening in adult ICU patients at risk for poor glycemic control compared to

no screening have an effect on the number of hypoglycemic events during ICU hospital stay?

Evidenced Based Literature

Study One

Head-to-Head Comparison of Two Continuous Glucose Monitoring Systems on a Cardio-

Surgical ICU

The purpose and aim of the quantitative study by Punke et al. (2019) was to evaluate the

accuracy and precision of two continuous glucose monitoring (CGM) sensors in twenty patients

who are on a cardio-surgical ICU unit in a head to head comparison. In order to be considered

for the study, the patients had to be eighteen years old or older, have no premedical history of
 5

steroid therapy, and scheduled for elective cardiac surgery between April 16, 2014 and August 8,

2014. Two sensors, a subcutaneous sensor Sentrino, Medtronic and an intravasal sensor

Glucoclear, Edwards were the continuous glucose monitoring sensors that were utilized in this

study. The CGM data that was collected was recorded for up to forty-eight hours and those

values were then compared with “blood-gas-analysis (BGA) values, analyzed with Bland-

Altman-plots and color-coded surveillance errors-grids'' (Punke et al., 2019, p. 895). The

subcutaneous sensor was placed in the upper thigh with fat tissue and the intravasal was placed

in a peripheral vein and the sensors had “novel drug interference with a wide array of

pharmaceuticals used in the critical care unit” (Punke et al., 2019, p. 896). 

The researchers found that “glucose variability per patient differed between the sensors

and was 18.1 + - 7.3 mg/dL, 24.5 + -10.3 mg/dL and 19.7 +- 7.7 mg/dL for intravasal sensor,

subcutaneous sensor and BGA respectively” (Punke et al., 2019, p. 897).  During the study,

Punke et al. (2019) found that there were no severe hypoglycemic events (below 40 mg/dL),

though there were moderate hypoglycemic events of 70 mg/dL and below. In the subcutaneous

sensor measurements, there were eight moderate hypoglycemic events, two moderate events

occurred from the intravasal sensor, and one moderate event with the BGA measurements

(Punke et al., 2019). The researchers noted that severe hyperglycemic events occurred (above

180 mg/dL) as well; thirteen with subcutaneous sensors, four with the intravasal sensors, and

nine with the BGA values. Despite these events, the researchers had positive outcomes with

reaching a target range as 83.1% of values were in the 100-180 mg/dL target range (Punke et al.,

2019). Missing values from the subcutaneous and intravasal sensors did occur; 12.3% and 7.5%

respectively. It was also found in this study that comparison of CGM sensors with the clinical

gold standard of POC testing, the CGM sensors had an acceptable accuracy “with a slight
 6

overestimation of the mean BG level in case of the subcutaneous sensor, or rather

underestimation in case of the intravasal sensor” (p. 898). Subcutaneous sensors showed an

ascending trend on the linear regression with an underestimation of blood glucose levels that

were low. Simultaneously, there was an overestimation of high blood glucose values compared

to the intravasal sensor that did not show any trend (Punke et al., 2019). 

The authors concluded that to reduce dangerous hypo- and hyperglycemic events, the

ability to continuously monitor blood glucose levels would allow for earlier intervention.

However, continuous monitoring is not standard in practice and point-of-care testing is

considered the gold standard in clinical settings. The study also concluded that “both sensors

showed acceptable accuracy. In the case of the subcutaneous sensor precision was slightly above

the clinically acceptable limits'' (Punke et al., 2019, p. 900). The CGMs were unique devices that

help recommend intervention and interpretation of hypoglycemia and hyperglycemia while the

gold standard POC testing should be considered for extreme values (Punke et al., 2019). 

Study Two

The Paradox of the Glycemic Gap: Does Relative Hypoglycemia Exist in Critically Ill

Patients? 

In this quantitative retrospective observational study conducted by Guo et al. (2021) the

purpose was to find “the association of different glycemic gaps in mortality in critically ill

patients” (p. 4654). Stress-induced hyperglycemia for diabetics and non-diabetics can be

considered an indicator of disease severity. Individualized care for each patient's blood glucose

levels was of high importance because hypoglycemia’s frequency and severity in ICU patients

are indicators of mortality. This study included nine-hundred and thirty-five adult patients who

were admitted to medical and surgical ICUs at a tertiary medical center, between December 2015
 7

and July 2017. These patients were divided into three groups that consisted of both diabetics and

non-diabetics. Data was gathered by screening patient medical records for patient characteristics

and comparing in-hospital mortality (Guo et al., 2021). 

  The researcher’s hypothesis for this study was that “low glycemic gaps at the time of ICU

admission, even without absolute hypoglycemia, would be associated with higher rates of

mortality during and after hospitalization” (Guo et al., 2021, p. 4655). Guo et al. (2021) found an

association between relative hypoglycemia and increased mortality due to lower glycemic gaps,

reflecting lower plasma glucose levels at the time of ICU admission. This association was

significant. These may be associated with greater disease severity. Sepsis and high Acute

Physiology and Chronic Health Evaluation II (APACHE II) scores were prevalent among

patients with low glycemic gaps (Guo et.al., 2021). Other findings were that excessive use of

glucose-lowering agents during the initial treatment of high glucose could be associated with low

glycemic gaps upon ICU admission. This study showed a U-shaped relationship between

glycemic gap and mortality and showed that hypoglycemia can be harmful to patients who are

critically ill. These findings “validated the usefulness of adapting glycemic gap at ICU admission

for risk stratification of critically ill patients” (p. 4659). Hypoglycemia in the critical care

settings tends to be higher among older adult patients with more comorbidities as well as those

who receive “intensive antidiabetic treatment (especially insulin)” (p. 4659). The researchers’

found that hypoglycemia and in-hospital mortality remain significant with “crude OR: 1.82, 95%

CI: 1.04-3.20, p = 0.036” (p. 4659). There were limitations to the study, including the institution

did not have a standardized protocol for glycemic control throughout hospitalization and lacked

information about hypoglycemic events after the study. The researchers also addressed that the

HbA1c levels could be altered if a recent blood transfusion was given, but this was unavailable. 
 8

Guo et al., (2021) concluded that further research is needed. More research would

confirm the “possible impact of relative hypoglycemia on adverse outcomes and its potential

clinical applications” (Guo et al., 2021, p. 4660). There needs to be a larger-scale study to reach

a final conclusion. When a patient is admitted to the ICU with low glycemic gaps, as well as

hypoglycemia, this was associated with greater mortality in the hospital, confirming the

researcher’s hypothesis for the study (Guo et al., 2021). 

Study Three

Hypoglycemia in a Surgical Intensive Care Unit

Switzer et al. (2021) conducted a quantitative, retrospective, observational study on

hypoglycemia within a surgical intensive care unit (SICU). This study focused on patients

admitted to the Los Angeles County + University of Southern California Medical Center

(LAC+USC) SICU from June 1, 2019, to July 31, 2020. Inclusion criteria was one or more

episodes of hypoglycemia consisting of a blood sugar less than 60 mg/dL (Switzer et al., 2021). 

Switzer et al. (2021) identified the objectives of this study were to use and collect data on

patient events of hypoglycemia to identify the rate and risk factors in the LAC+USC SICU.

Switzer et al. hoped to determine opportunities to improve care and SICU patient outcomes. The

purpose of this study originated from the associated neurologic and vascular complications

noticed in patients and increased rates of mortality and hospital stay. Patient records were studied

to collect data due to the retrospective nature of the study based on the dependent variable,

hypoglycemic events, to identify correlations and commonalities between patients. 

In the Switzer et al. (2021) study, 32 out of 2133 SICU patients experienced

hypoglycemia , a 1.5% incidence rate, “which is on par with existing literature” (Switzer et al.,

2021, p. 1582). Five of these patients were excluded as a result of a false reading which resulted
 9

in 41 hypoglycemic events amongst 27 patients. The authors found 26% of patients had more

than one episode of hypoglycemia. Of the hypoglycemic episodes, 59% were determined to have

been iatrogenic and preventable with nursing interventions. The researchers established that 39%

of incidents could have been avoided by adding dextrose to IV fluids and 22% with “more

judicious correction of hyperglycemia” (Switzer et al., 2021, p. 1582). 

Switzer et al. (2021) concluded that the hypoglycemic incidence was most prevalent in

certain patient groups. The groups identified to be at greatest risk for hypoglycemia were those

with multisystem organ failure, nothing per os patients (NPO), and those receiving insulin

infusions or long-acting subcutaneous insulin. The researchers recommended “increased

vigilance with strict glycemic control and frequent blood glucose monitoring” in populations at

greater risk for hypoglycemia (Switzer et al., 2021, p. 1582). The researchers noted their next

step of study would be to implement hourly glucose checks for at-risk patients. 

Study Four

Individualised versus Conventional Glucose Control in Critically Ill Patients: the

CONTROLING Study-a Randomized Clinical Trial

The quantitative study conducted by Bohé et al. (2021) was a double-blind randomized

control trial. The authors established that intensive care unit (ICU) patients were randomly

assigned to either individualized glucose control by the computer application Contrôle

Personnalisé de la Glycémie (CPG) which targeted the pre-admission glucose calculated by the

A1c, or conventional glucose control which used sliding scale insulin to try and keep the blood

glucose under 180 mg/dL. According to Bohé et al. (2021), “CPG is an algorithm based on

multiple insulin infusion rate sliding scales and rules to move within a scale and from one scale

to another. CPG gave the nurse (without physician intervention) all the instructions for the pre-
 10

scheduled time of glycaemia sampling, administration of the intravenous insulin, and correction

of hypoglycaemia” (p. 1272). This study included 2075 participants admitted from May 19th,

2015, to July 6th, 2016, from 12 ICUs within France (Bohé et al., 2021).

The aim of this research performed by Bohé et al. (2021) was to determine if the

glycemic targets for patients in the ICU could be individualized and how it would impact

mortality and morbidity. The CPG informed dynamic sliding-scale for participants in the

intervention group was individualized and compared next to the conventional insulin therapy

control group. The researchers then looked to see if after 90 days there was a comparative

decrease in morbidity and mortality.

Bohé et al. (2021) were transparent in disclosing that the data and safety monitoring

board recommended stopping the trial due to the low likelihood of benefit and potential for harm.

At the time that the trial was stopped, 2075 of 4200 patients had been included (Bohé et al.,

2021). After adjustment had been made for variables such as sex, admission type, diabetes

mellitus status etc. researchers did not find a significant difference in mortality at 90 days, p=0.1.

However, there was a statistically significant difference between the intervention and control

groups when looking at hypoglycemic events. 31.2% of the intervention group had some form of

a hypoglycemic event and 15.8% of the control experienced some form of hypoglycemia, a

significance of p=0.0001 (Bohé et al., 2021). 

         Upon conclusion, Bohé et al. (2021) found that there was no significant difference in

mortality when targeting ICU patients’ pre-admission blood sugar while using a “dynamic

sliding-scale insulin protocol” (Bohé et al., 2021, p. 1281). The individualization of blood

glucose control did however result in an increased incidence of hypoglycemia in the ICU. Bohé

et al. disclosed that to maintain anonymity and blinding of the two groups the ICU nurses did not
 11

have access to the patient’s history or the changes in blood glucose or insulin therapy. The

researchers believe this may have limited the opportunities for the nurses to override insulin

therapy recommendations generated by CPG, potentially allowing for better control of blood

sugar. The authors of the study also determined that “the individualized glycemic target was

maintained constant during the entire ICU stay, which might not match the changes in insulin

resistance that can occur during the course of critical illness” (Bohé et al., 2021, p. 1281). While

they did not conclude that individualization surrounding blood sugar control in totality is a

problem, the independent variable proposed by Bohé et al. was not beneficial in the effect it had

on patient outcomes.

Study Five

Reducing Hypoglycemia in Critical Care Patients Using a Nurse-Driven Root Cause Analysis

Process

Shea et al. (2019) conducted a quantitative, retrospective investigation that examined the

impact of completing a root-cause analysis process after hypoglycemic events in critical care

patients. It was conducted at Stamford Hospital which is a 305-bed facility, and about 30 miles

Northeast of New York City. The intensive care unit is a 16-bed unit described as, “Care in the

ICU is highly protocol based and nurse driven. These factors set the stage for development of an

RCA [root cause analysis] of hypoglycemic events” (Shea et al., 2019, p. 30). 

Shea et al. (2019) wrote, “The purpose of this article is to describe a successful nurse-

driven initiative to reduce episodes of hypoglycemia in critically ill patients using an RCA

format” (p. 30). The authors mentioned the ICU glycemic protocols, which is important to

comprehend their process. They wrote, “Concerns over episodes of hypoglycemia in this setting

were ever present despite safety measures in the protocol” (p. 33). The goal was to eliminate
 12

hypoglycemia if possible. The researchers implemented a program requiring nurses to fill out a

root-cause analysis form after the occurrence of a blood sugar of 60 mg/dL or less as “a more

intentional process of data collection and analysis was identified as valuable for improving out-

comes” (p. 33). This form was piloted by a group of nurses after which the form was shared with

nursing staff and put into action. They reviewed RCA reports monthly to identify causes of

hypoglycemic events and provide education to nurses on prevention measures to think upstream,

instead of being reactionary (Shea et al., 2019).

The study ended when the data showed a positive impact on glycemic outcomes,

demonstrating that performing a root cause analysis after a hypoglycemic event in the ICU is a

beneficial process. According to Shea et al. (2019):

The number of patients without diabetes who had at least 1 occurrence of blood glucose

level less than 40 mg/dL, less than 60 mg/dL, and less than 70 mg/dL decreased by

61.3%, 38.5%, and 21.6%, respectively, with corresponding decreases in the percentage

of blood glucose values in these strata. Similarly, the number of patients with diabetes

who had at least 1 occurrence of blood glucose level less than 40 mg/dL, less than 60

mg/dL, and less than 70 mg/dL decreased by 45.8%, 45.0%, and 38.1%, respectively. (p.

35)

These results showed the significance and success of their research. The authors state, “We

compared the pre-intervention and implementation cohorts to a “continuation” cohort, the 500

patients admitted between March 1, 2016, and August 31, 2016, to determine the sustained effect

of the intervention” (Shea et al., 2019, p. 33). It proved that the intervention was sustainable,

showing positive results, even after the time of the study. 

         Shea et al. (2019) concluded that implementing a root cause analysis in the ICU at

Stamford Hospital to reduce hypoglycemic events was a sustainable change. The authors stated,

“Among patients with diabetes, the substantial reduction in hypoglycemia events was not

accompanied by an increase in mean blood glucose level” (Shea et al., 2019, p. 37). This is

important because when focusing on avoiding hypoglycemia one does not want to cause

hyperglycemia as an alternative. Shea el al. (2019) concluded by saying:

 By finding a process to drill down to a new level of detail regarding the variety of factors

that can lead to hypoglycemia, nurses were able to examine their own practice and make

relatively minor adjustments that produced positive results. (p. 38)

Study Six

  Incorporating Real-World Evidence into the Development of Patient Blood Glucose

Prediction Algorithms for the ICU

Fitzgerald et al. (2021) conducted a quantitative, cohort study that used an algorithm to

predict blood glucose. They performed this in the intensive care unit at Beth Israel Deaconess

Medical Center in Boston, Massachusetts. The aim was to present a data-driven method to

predict glycemic control in ICU patients. The researchers performed this by using electronic

medical records of 18,961 admissions from the MIMIC-III data set, including 318,574 blood

glucose measurements. They stated, “we train and validate a gradient boosted tree machine

learning (ML) algorithm to forecast patient blood glucose and a 95% prediction interval at 2-

hour intervals” (p. 1642). In this process, they used data related to recent patient medical history,

glycemic control, insulin dosing, and nutrition. According to Fitzgerald et al. (2021):

Glycemic control, an important component of critical care, may benefit from all of these

ML tasks from the development of algorithms which provide individual patient glucose
 14

predictions to the detection of early risk of hypo- or extreme hyperglycemia, through

individualized control algorithms. (p. 1643)

According to Fitzgerald et al. (2021), the final sample size was 14,742 patients. They

found that their process had a high degree of accuracy and “perform[ed] best in the most

frequently observed glycemic range (100-200 mg/dL), followed by the hyperglycemic (> 200

mg/dL), and then hypoglycemic ranges (< 70 mg/dL)” (p. 1645). The mean absolute percentage

errors were 12.4%, 22.5%, and 96.3% respectively. It is stated that 97% of their predictions were

clinically acceptable. However, they write “difficulty of forecasting hypoglycemia clearly

warrants further research and analysis” (p. 1649). The authors state that though their research

was beneficial, there were limitations because the model should have detected both

hyperglycemia and hypoglycemia to have clinical significance (Fitzgerald et al., 2021).

Fitzgerald et al. (2021) concluded that, “we demonstrate that EMRS can be used to

develop blood glucose prediction models that achieve a high degree of accuracy” (p. 1649). They

stated that their algorithms may be suitable for decision making in ICU patients. However, a

challenge remained in detecting hypoglycemic events which warrants further research. This

should be done using real world data to continue to find ways to improve the standard of care. 

Synthesis

In examining our research articles, we have identified four primary themes. The first

theme identified was the seriousness of hypoglycemia in hospitalized patients and the affect on

patient outcomes (Bohé et al., 2021; Fitzgerald et al., 2021; Guo et al., 2021; Punke et al., 2019;

Shea et al., 2015; Switzer et al., 2021).The second theme amongst the primary sources was that

hypoglycemic events are hard to predict and prevent (Bohé et al., 2021; Fitzgerald et al., 2021;

Guo et al., 2021; Punke et al., 2019; Shea et al., 2015; Switzer et., 2021). The need for further
 15

research is the third theme we identified. Four out of the six primary research articles pointed to

the fact that further research is needed regarding hypoglycemia and glucose monitoring (Bohé et

al., 2021; Fitzgerald et al., 2021; Guo et al., 2021; Switzer et., 2021). This points to a gap in the

standard of care regarding hypoglycemia. The final theme was that assessment and identifying

risk factors for hypoglycemic events was shown to be beneficial in decreasing its prevalence, as

supported by two of our research articles (Shea et al., 2015; Switzer et al., 2021). 

Recommendations for Clinical Impact

After analyzing and synthesizing our research articles, we came up with a

recommendation for clinical change. There are a few components to our recommendation. This

change includes implementing a hypoglycemic risk assessment, see Appendix A, that would be a

requirement for a nurse to fill out upon every intensive care unit admit. This would allow for

early identification of patients at risk for poor glycemic control and prevent hypoglycemic

episodes. The risk assessment would include an admission sugar, diabetic status, diet status, tube

feeds, infusions including dextrose fluids and medications, and health diagnoses. By having a

risk assessment sheet to fill out upon admission, risk factors for poor glycemic control would be

identified early and nurses would not solely rely on their ICU experience to avoid a

hypoglycemic event. This added level of awareness will positively impact nurse clinical

judgment when making decisions for the patients’ health and safety. 

Creating Change

Kotter’s Eight Stages of Change model was determined to best format our proposed

change (Paul, 2022). The first stage of Kotter’s model is to establish urgency surrounding the

proposed change. Urgency is created by educating ICU staff on the prevalence of hypoglycemic

events, as well as the negative impacts hypoglycemia can have on patient outcomes. 
 16

Kotter’s second stage, putting a team together, would involve identifying ICU staff that

are interested in hypoglycemia prevention. This team would help to spearhead the

implementation of the risk assessment on the unit and encourage utilization by other staff. The

third stage, developing a vision and strategies for change, would involve obtaining feedback

from the team members as stakeholders in patient care within the ICU and altering the

assessment as appropriate. The team would then discuss how to best introduce the risk

assessment to ICU staff for effective implementation. Communicating the change would be done

by holding staff and unit meetings as well as sending emails about the meetings being held with

an introduction of the risk assessment. The team members would be important in this stage to

inspire others and lead by example in the early adoption.

An obstacle to implementing the admission risk assessment on the unit, the fifth stage,

would be staff resistance to the change. To address this obstacle, we would ask staff members

about the reasons they are not adopting the assessment and discuss the benefits of the change

adoption for the patient and staff.

The sixth stage, setting short-term goals, helps generate benchmarks to reach during the

change implementation. One of these goals is staff attendance at an educational meeting about

the risk assessment. Another goal is for 50% of ICU nurses to consistently utilize the risk

assessment upon admission. Consolidating gains and keeping momentum, the seventh stage of

Kotter’s model, could be done by having daily check-ins with staff in the shift change huddle.

Feedback would be sought to determine how the risk assessment is being utilized, potential

changes, and the benefits seen so far. The final stage is making the change stick (Paul, 2022). To

ensure the pre-screening checklist continues to be utilized within the ICU, data regarding the
 17

benefit and effect it has had on hypoglycemic event occurrence would be shared and staff would

be affirmed for assessment use. 

Conclusion

Hypoglycemic events have been and will continue to be detrimental to ICU patient

outcomes. This problem is clinically significant because it affects all patient populations. The

primary research articles we examined showed four primary themes, the seriousness of

hypoglycemia, the difficulty of predicting and preventing these events, the need for further

research, and the benefit of assessing and identifying risk factors for these events. Implementing

a hypoglycemic risk assessment will help nurses provide better patient centered care and

improve patient outcomes. Although this topic is complex, this one change is a step toward

managing hypoglycemia more effectively in the intensive care unit setting.

Appendix A 

Intensive Care Unit Hypoglycemic Risk Assessment 

Patient name: 

Patient DOB: 

Date:

Admit time:

Primary diagnosis: 

Patient admission blood sugar:

What are the patient’s chronic diagnoses?

Does the patient have cirrhosis, polytrauma, multisystem organ failure, and/or skin/soft tissue

infections? (Switzer et al., 2021) YES (+1) NO (+0)

Is the patient diabetic? YES (+1) NO (+0)

Is the patient eating by mouth? YES (+0) NO (+1)

Is the patient eating by tube feeds? YES (+1) NO (+0)

Is the patient receiving insulin? YES (+1) NO (+0)

Has the patient received a dextrose infusion? YES (+0) NO (+1)

SCORE: 0-low risk; 3 moderate risk; 6 high risk

NURSE SIGNATURE
 19

References

Bohé, J., Abidi, H., Brunot, V., Klich, A., Klouche, K., Sedillot, N., Tchenio, X., Quenot, J.P.,

Roudaut, J.B., Mottard, N., Thiollière, F., Dellamonica, J., Wallet, F., Souweine, B.,

Lautrette, A., Preiser, J.C., Timsit, J.F., Vacheron, C.H., Hssain, A.A., & Maucort-

Boulch, D. (2021). Individualised versus conventional glucose control in critically-ill

patients: the CONTROLING study-a randomized clinical trial. Intensive Care Med, 47,

1271-1283. https://doi.org/10.1007/s00134-021-06526-8

Cruz, P. (2020). Inpatient hypoglycemia: the challenge remains. Journal of Diabetes Science and

Technology, 14(3), 560-566. Doi: 10.1177/1932296820918540

Fitzgerald, O., Perez-Concha, O., Gallego, B., Saxena, M. K., Rudd, L., Metke-Jimenez, A., &

Jorm, L. (2021). Incorporating real-world evidence into the development of patient blood

glucose prediction algorithms for the ICU. Journal of the American Medical Informatics

Association, 28(8), 1642–1650. https://doi.org/10.1093/jamia/ocab060 

Guo, J. Y., Chou, R. H., Kuo, C. S., Chao, T. F., Wu, C. H., Tsai, Y. L., Lu, Y. W., Kuo, M. R.,

Huang, P. H., & Lin, S. J. (2021). The paradox of the glycemic gap: does relative

hypoglycemia exist in critically ill patients. Clinical Nutrition, 40(7), 4654-4661.

https://doi.org/10.1016/j.clnu.2021.06.004 

Paul, J.C. (2022). Transitioning evidence to practice. In N.A. Schmidt & J.M. Brown (Eds.),

Evidence-based practice for nurses: Appraisal and application of research (5th ed., pp.

469-493). Jones & Bartlett Learning. 

Punke, M. A., Decker, C., Petzoldt, M., Reuter, D. A., Wodack, K. H., Reichenspurner, H.,

Kubik, M., & Kluge, S. (2019). Head-to-head comparison of two continuous glucose
 20

monitoring systems on a cardio-surgical ICU. Journal of Clinical Monitoring and

Computing, 33, 895-901. https://doi.org/10.1007/s10877-018-0221-5

Shea, K. E., Gerard, S. O., & Krinsley, J. S. (2019). Reducing hypoglycemia in critical care

patients using a nurse-driven root cause analysis process. Critical Care Nurse, 39(4), 29–

38. https://doi.org/10.4037/ccn2019876 

Switzer, E., Schellenberg, M., Lewis, M., Owattanapanich, N., Lam, L., & Inaba, K. (2021).

Hypoglycemia in a surgical intensive care unit. The American Surgeon, 87(10), 1580-

1583. https://doi.org/10.1177/00031348211024972