Professional Documents
Culture Documents
Updates in The Management of Diabetic Ketoacidosis: Kathryn Evans Kreider, DNP, FNP-BC
Updates in The Management of Diabetic Ketoacidosis: Kathryn Evans Kreider, DNP, FNP-BC
Ketoacidosis
Kathryn Evans Kreider, DNP, FNP-BC
ABSTRACT
Diabetic ketoacidosis (DKA) is an emergency for people with diabetes characterized by
hyperglycemia, metabolic acidosis, and ketosis. DKA onset and recurrence can largely
be prevented through patient education. Nurse practitioners are well positioned to
promote patient education, self-management, and individualized patient care. This
article outlines updates in the clinical management of patients with DKA to optimize
care and reduce costs.
D
iabetic ketoacidosis (DKA) is a metabolic declined due to earlier detection and increased
derangement characterized by hyperglyce- evidence-based management. In fact, the recent
mia, metabolic acidosis, and ketosis.1 DKA CDC US Diabetes Surveillance System report indi-
occurs in patients with diabetes who have a lack of cated that the mortality for DKA has decreased
circulating insulin relative to physiologic to 0.4%.3
requirement, such as in type 2 diabetes mellitus Episodes of DKA typically require an emergency
(T2DM), or absolute depletion, such as in type 1 department visit or hospital admission for the patient
diabetes mellitus (T1DM),2 in the presence of to receive insulin, intravenous (IV) fluids, and elec-
increased counterregulatory hormones (cortisol, trolyte correction. Hospital encounters can be costly,
growth hormone, epinephrine, and glucagon). The with recent reports suggesting that a single hospital
lack of adequate insulin can occur from medication encounter for DKA treatment can cost up to
noncompliance, infection, or a precipitating $17,500,5 with an average length of stay of 3.4 days.
pathologic event, such as a myocardial infarction, that Further, the annual direct and indirect cost of DKA
causes an increased metabolic demand for insulin. treatment in the US exceeds $2.3 billion.1
The Centers for Disease Control and Prevention DKA and recurrent DKA are largely preventable
(CDC) United States Diabetes Surveillance System through better outpatient management, patient ed-
recently reported an increase in DKA episodes in the ucation, and promoting self-care behaviors.6 Nurse
US between 2009 and 2014, with an average annual practitioners (NPs) have an opportunity to educate
increase of 6.3%.3 DKA can occur at any age but patients about the risks of DKA, promote self-
primarily occurs in those aged younger than 30 years management, and offer patient-centered care. Pre-
(36% incidence) and between 30 and 50 years (27% sented here is a review the management of DKA and
incidence).2 The highest incidence of DKA is for updates in clinical care.
those aged between 11 and 15 years.4 In addition,
girls and women and the immigrant population are at DIAGNOSIS
higher risk. The diagnosis of DKA is made based on the meta-
In the past, DKA had a fatality rate of 1% to 5%. bolic triad of high blood glucose (BG) levels
Older adults and individuals who have comorbid risk (generally > 250 mg/dL), acidosis (pH < 7.2), and
factors are in the highest risk category.1 In recent the presence of urine or serum ketones (Table).5
years, however, the overall mortality of DKA has Inpatient providers commonly rely on laboratory data
to confirm DKA, whereas outpatient providers rely initiated, an increase in glucagon, or decreased
on history, presentation, BG levels, and excretion of ketone bodies.1 Other related factors
urine ketones. may be mild infection, increased activity, reduced
Common presenting symptoms include abdom- food intake, or insulin reduction or omission.10 Case
inal pain and the classic triad of hyperglycemia reports of traditional and euDKA occurring while
symptoms: polydipsia, polyphagia, and polyuria. using SGLT-2 inhibitors has been shown in patients
Physical examination findings can include any or all with T1DM and in those with T2DM.11 Many
of tachycardia, hypotension, Kussmaul respirations, patients with euDKA present with nausea and
significant dehydration, or a change in mental status.7 vomiting but are misdiagnosed due to the lack of
It is important to consider differential diagnoses of clear glucose elevation.
metabolic acidosis that may include lactic acidosis or
hyperchloremic acidosis. Differential diagnoses for DIAGNOSTIC WORKUP
ketosis include starvation ketosis (dietary history, Providers in outpatient settings where laboratory re-
weight trends) or alcoholic ketoacidosis (alcohol sults are not readily available should rely on BG and
consumption history), hyperemesis, isopropyl ketone values for the initial diagnosis. BG values will
alcohol, or ketotic hypoglycemia.1 typically be > 250 mg/dL, although up to 10% of
Euglycemic DKA (euDKA), which occurs when patients in DKA may present with euDKA.9 For this
the patient presents with acidosis and ketosis but has a reason, some experts argue that the cutoff BG value
glucose 200 mg/dL, has become an emerging for diagnosing DKA should be decreased to
concern. Causes of euDKA can include recent insulin 200 mg/dL.12
administration, decreased caloric intake, substantial Ketone measurement is an important diagnostic
alcohol consumption, chronic liver disease, or rarely, component and severity classification of DKA for
glycogen storage issues.8 In addition, there have patients in the outpatient setting. Urine ketones will
been increasing reports of euDKA caused by a new test positive, although may be as minimal as “trace”
class of drugs for diabetes, sodium glucose ketones. Urine dipsticks measure acetoacetate (AcAc)
cotransporter 2 (SGLT-2) inhibitors. In May 2015, and acetone but not b-hydroxybutyrate (b-OHB).
the US Food and Drug Administration added a Given this, the measurement of AcAc in the urine
warning about the risk of DKA with use of these tends to underestimate the severity of DKA, because
drugs. One study suggested that the risk of DKA for the ratio of AcAc to b-OHB can be 1:10 during
patients using SGLT-2 inhibitors was twice as high as ketoacidosis.13 Other limitations with urine ketone
those prescribed a dipeptidyl peptidase IV inhibitor, testing include lag time in change in urine ketones,
after controlling for other risk factors, although the difficulty obtaining urine from dehydrated patients,
risk of hospitalization was low.9 The exact cause of and subjective measurements by the patient. Many
this relationship is unknown, but several theories experts agree that measurement of blood or capillary
include reduced insulin doses when SGLT-2 is ketones is preferred due to these limitations.9
2 The Journal for Nurse Practitioners - JNP Volume -, Issue -, -/- 2018
Some outpatient glucometers, measure b-OHB in
addition to BG levels. Several of these types of moni- Box 1. Common Precipitants of Diabetic
toring systems are available on the market. These Ketoacidosis9
meters measuring capillary ketones have been shown to
Common Precipitants of Diabetic Ketoacidosis9
have high sensitivity, specificity, positive predictive
The Five “I’s”
value, and negative predictive value in identifying
Infection
DKA compared with urine ketone testing.14 Patients
Infarction (myocardial, heart attack)
with diabetes who have access to b-OHB meters may
Infant (pregnancy)
have reduced emergency department visits, time to
Indiscretion (such as street drugs)
recovery from DKA, and the potential for reduced
Insulin lack (noncompliance, dose reduction,
costs.15 Currently, these systems are not widely used
inability to afford, pump malfunction)
and are expensive but may be beneficial to consider for
patients who are at risk for DKA. Other:
Intravenous fluids
1,000e2,000 mL 0.9% NaCl over 1e2 h.
Continue 0.9% NaCl or switch to 0.45% NaCl at 250e500 mL/h depending on serum sodium.
When BG level ¼ 200-250 mg/dL, change to 5% dextrose in 0.45% NaCl.
Insulin
Regular human insulin intravenous bolus of 0.1 U/kg (optional), followed by continuous insulin infusion at
0.1 U/kg/h.
When BG 250 mg/dL, reduce insulin rate to 0.05 U/kg/h; thereafter, adjust rate to maintain glucose level
at w200 mg/dL.
Subcutaneous rapid-acting insulin might be an alternative to intravenous insulin in patients with
mild-to-moderate DKA.
Potassium
Serum potassium (K+) level > 5.0 mEq/L (no supplement required; monitor every 2 hours)
K+ level 3.3e5 mEq/L (add 20-40 mEq potassium chloride to replacement fluid)
K+ level <3.3 mEq/L (hold insulin; give 20-30 mEq/h until > 3.3)
Goal K+ level 4-5 mEq/L
Oral potassium may be considered depending on patient status
Bicarbonate
Not routinely recommended. If pH < 6.9 consider 50 mmol/L in 500 mL of 0.45% normal saline over 1 hour
until pH increases to 7.0
Transition to subcutaneous insulin
Continue insulin infusion until resolution of ketoacidosis. To prevent recurrence of ketoacidosis or rebound
hyperglycemia, continue intravenous insulin for 2e4 h after subcutaneous insulin (basal) is given.
For patients treated with insulin before admission, restart previous insulin regimen and adjust doses as
needed. For patients with newly diagnosed diabetes mellitus, start total daily insulin dose at 0.6 U/kg/d.
Consider multidose insulin given as basal and prandial regimen.
4 The Journal for Nurse Practitioners - JNP Volume -, Issue -, -/- 2018
that rapid-acting subcutaneous (SQ) insulin, such as causing an influx of sodium into the cell and extra-
lispro or aspart, given every 1 to 2 hours, can be cellular depletion of sodium. The opposite occurs
effective in the management of mild DKA.1 SQ with potassium: Acidosis causes extracellular shifts of
insulin can also be administered in lower levels of potassium, creating high serum potassium levels but
care, which can assist in the cost reduction of the relative cellular depletion. In otherwise healthy pa-
hospitalization. In fact, using insulin analogs may tients, hyponatremia and hyperkalemia are both
reduce hospitalization costs by up to 30%.1 generally resolved with fluid replacement. Patients in
An earlier study evaluating SQ vs IV regular in- DKA may occasionally present with hypokalemia,
sulin used 0.3 U/kg SQ rapid-acting insulin as a particularly if they are taking diuretics or were
loading dose, followed by 0.1 U/kg/h until the BG vomiting. In the case of hypokalemia upon presen-
was < 250 mg/dL. The dose was then reduced to tation, fluids and potassium replacement should be
0.05 U/kg/h until the DKA resolved. Compared given before starting insulin because the insulin
with IV insulin, there was no difference in length of administration may cause a further drop in potassium
stay or total amount of insulin needed.19 Other more levels.1 Patients with severe hypokalemia require
recent studies have supported these findings, cardiac monitoring.
including no difference in rates of hypoglycemia Resolution of DKA occurs when BG is < 200
when using SQ insulin compared with IV mg/dL and 2 of the following have occurred: a serum
insulin.20,21 Recommended doses of rapid-acting SQ bicarbonate level 15 mEq/L, a venous pH > 7.3,
insulin analogs are between 0.075 and 0.1 U/kg/h. and a calculated AG 12 mEq/L.16 Box 2 contains a
This is similar to the recommended dose of IV insulin summary of recommendations regarding the
that is initiated in the hospital setting. Alternately, 0.1 treatment of DKA in adult patients. Comprehensive
to 0.2 U/kg can be administered every 2 to algorithms detailing the comprehensive management
3 hours.22 of DKA have been published elsewhere.1,2
For patients in DKA, the goal BG at the time of
treatment is 150 to 200 mg/dL.16 TREATING THE UNDERLYING PRECIPITANT
Successful identification of precipitating factors is
METABOLIC ACIDOSIS paramount to effective DKA management. Common
In the case of DKA, metabolic acidosis occurs when triggers include medication noncompliance, infec-
hyperglycemia precipitates an increase in free fatty tion, myocardial infarction, pancreatitis, alcohol
acid breakdown, which produce ketone bodies when abuse, cerebrovascular accident, trauma, hyperthy-
metabolized. These ketone bodies cause a decrease in roidism, or new diagnosis of T1DM.7 Any stress or
the alkali reserve causing ketoacidosis. Metabolic inflammatory response in the body could trigger the
acidosis can be determined by calculating the AG, the production of ketones, resulting in DKA. Patients
difference between serum cations and anions.1 A should be thoroughly evaluated for underlying causes
normal AG level is < 9 mEq/L. Patients who present based on history, presentation, physical examination,
in DKA typically have an AG > 10 mEq/L.16 and laboratory findings. Patients with new-onset
Patients who present with symptomatic metabolic diabetes represent w20% of those presenting with
acidosis (Kussmaul respirations, confusion, lethargy) DKA.5 Newly diagnosed patients will require a
should be managed as high-acuity patients. comprehensive care plan, including
The calculation of AG is (serum cations [sodium] diabetes education.
e anions [chloride þ bicarbonate]).1
PREVENTION
ELECTROLYTE MONITORING DKA commonly reoccurs in the same cohort of
Hyponatremia and hyperkalemia are the primary patients, and poor adherence to treatment is the
electrolyte abnormalities that occur in DKA. Hypo- number one precipitant of DKA in the US.5
natremia is caused when the acidosis precipitates a Discussing medication regimens, feasibility, cost, and
shift of potassium out of the intracellular space, patient satisfaction is imperative to encouraging
call their health care provider at the first detection of ketoacidosis hospitalizations and in-hospital mortality—United States,
2000-2014. MMWR Morb Mortal Wkly Rep. 2018;67:362-365.
ketones. Mild DKA caught early can occasionally be 4. Fritsch M, Rosenbauer J, Schober E, Neu A, Placzek K, Holl RW; German
Competence Network Diabetes Mellitus and the DPV Initiative. Predictors of
treated by patients at home (in consultation with a diabetic ketoacidosis in children and adolescents with type 1 diabetes.
6 The Journal for Nurse Practitioners - JNP Volume -, Issue -, -/- 2018
6. Mills LM, Stamper JE. Adult diabetic ketoacidosis: diagnosis, management, 22. Vincent M, Nobecourt E. Treatment of diabetic ketoacidosis with
and the importance of prevention. J Diabetes Nurs. 2014;18:8-12. subcutaneous insulin lispro: review of the current evidence from clinical
7. Perelli G, Saraceni C, Daniels MN, Ahmad A. Diabetic ketoacidosis: a review studies. Diabetes Metab. 2013;39:299-305.
and update. Curr Emerg Hosp Med Rep. 2013;1:10-17. https://doi.org/10.1007/ 23. Auchterlonie A, Okosieme O. Preventing diabetic ketoacidosis: do patients
s40138-012-0001-3. adhere to sick day rules? Clin Med (Lond). 2013;13:120.
8. Modi A, Agrawal A, Morgan F. Euglycemic diabetic ketoacidosis: a review. 24. Elliott J, Jacques RM, Kruger J, et al. Substantial reductions in the number of
Curr Diabetes Rev. 2017;13:315-321. diabetic ketoacidosis and severe hypoglycaemia episodes requiring
9. Fralick M, Schneeweiss S, Patorno E. Risk of diabetic ketoacidosis after emergency treatment lead to reduced costs after structured education in
initiation of an SGLT2 inhibitor. N Engl J Med. 2017;376:2300-2302. adults with type 1 diabetes. Diabet Med. 2014;31:847-853.
10. Peters AL, Buschur EO, Buse JB, Cohan P, Diner JC, Hirsch IB. Euglycemic 25. AADE Practice Paper. Continuous subcutaneous insulin infusion (CSII) with and
diabetic ketoacidosis: a potential complication of treatment with without sensor integration. American Association of Diabetes Educators, 2018.
sodiumeglucose cotransporter 2 inhibition. Diabetes Care. https://www.diabeteseducator.org/practice/practice-documents/practice
2015;38:1687-1693. -papers. Accessed May 17, 2018.
11. Bonora BM, Avogaro A, Fadini GP. Sodium-glucose co-transporter-2 26. Goldenberg RM, Berard LD, Cheng AY, et al. SGLT2 inhibitoreassociated
inhibitors and diabetic ketoacidosis: an updated review of the literature. diabetic ketoacidosis: clinical review and recommendations for prevention
Diabetes Obes Metab. 2018;20:25-33. and diagnosis. Clin Ther. 2016;38:2654-2664.
12. Dhatariya KK, Umpierrez GE. Guidelines for management of diabetic 27. MacArthur C. Ketoacidosis in diabetes: recognition and avoidance. Pract
ketoacidosis: time to revise? Lancet Diabetes Endocrinol. 2017;5:321-323. Nurs. 2015;26:393-399.
13. Misra S, Oliver NS. Utility of ketone measurement in the prevention, 28. Pearce MJ, Pereira K, Davis E. The psychological impact of diabetes: a
diagnosis and management of diabetic ketoacidosis. Diabet Med. practical guide for the nurse practitioner. J Am Assoc Nurse Pract.
2015;32:14-23. 2013;25:578-583.
14. Brooke J, Stiell M, Ojo O. Evaluation of the accuracy of capillary 29. Schmitt A, Reimer A, Kulzer B, Haak T, Ehrmann D, Hermanns N. How to
hydroxybutyrate measurement compared with other measurements in the assess diabetes distress: comparison of the Problem Areas in Diabetes Scale
diagnosis of diabetic ketoacidosis: a systematic review. Int J Environ Res (PAID) and the Diabetes Distress Scale (DDS). Diabet Med. 2016;33:835-843.
Public Health. 2016;13:837. 30. van Dijk SE, Adriaanse MC, van der Zwaan L. Measurement properties of
15. Klocker AA, Phelan H, Twigg SM, Craig ME. Blood b-hydroxybutyrate vs. depression questionnaires in patients with diabetes: a systematic review.
urine acetoacetate testing for the prevention and management of Qual Life Res. 2018;27:1-16.
ketoacidosis in Type 1 diabetes: a systematic review. Diabet Med.
2013;30:818-824.
16. Kitabchi AE, Umpierrez GE, Miles JM, Fisher JN. Hyperglycemic crisis in adult
patients with diabetes. Diabetes Care. 2009;32:1335-1343. Kathryn Evans Kreider, DNP, FNP-BC, is from the Division
17. Evans KJ, Thompson J, Spratt SE, Lien LF, Vorderstrasse A. The
implementation and evaluation of an evidence-based protocol to treat of Endocrinology, Metabolism & Nutrition, Duke University
diabetic ketoacidosis: a quality improvement study. Adv Emerg Nurs J.
2014;36:189-198.
Medical Center, and is an assistant professor at Duke University
18. Ullal J, McFarland R, Bachand M, Aloi J. Use of a computer-based insulin School of Nursing, Durham, NC. She is available at kathryn.
infusion algorithm to treat diabetic ketoacidosis in the emergency
department. Diabetes Technol Ther. 2016;18:100-103. evans@duke.edu In compliance with national ethical guidelines,
19. Umpierrez GE, Latif K, Stoever J, et al. Efficacy of subcutaneous insulin lispro
versus continuous intravenous regular insulin for the treatment of patients the author reports no relationships with business or industry that
with diabetic ketoacidosis. Am J Med. 2004;117:291-296. would pose a conflict of interest.
20. Barski L, Kezerle L, Zeller L, Zektser M, Jotkowitz A. New approaches to the
use of insulin in patients with diabetic ketoacidosis. Eur J Intern Med.
2013;24:213-216.
21. Cohn BG, Keim SM, Watkins JW, Camargo CA. Does management of diabetic 1555-4155/18/$ see front matter
ketoacidosis with subcutaneous rapid-acting insulin reduce the need for © 2018 Elsevier Inc. All rights reserved.
intensive care unit admission? J Emerg Med. 2015;49:530-538. https://doi.org/10.1016/j.nurpra.2018.06.013