Case Study: Type 1 Diabetes Mellitus & Diabetic Ketoacidosis

Diabetes mellitus is one of the most common chronic diseases of childhood and youth (PHAC, 2011). Canada has one of the highest rates of Type 1 Diabetes Mellitus (T1DM) in the world (Metzger, 2010). It is estimated to effect 21.7 in 100,000 Canadian children aged 0 to 14 years of age (Metzger, 2010). T1DM is becoming increasingly common. The prevalence has been increasing worldwide at an alarming rate of 2 to 9% annually (Leung, Perlman, Rumantir, & Freedman, 2015). T1DM is a disease that results from destruction of insulin-producing cells in the pancreas, leading to insulin deficiency and hyperglycemia (Gregory, Moore, & Simmons, 2013). If left untreated, T1DM progresses to acidosis and death.

Children who present with T1DM have a broad range of clinical features, ranging from extremely subtle symptoms to life-threatening diabetic ketoacidosis (DKA) (Merger, Leslie, & Boehm, 2013). Studies have shown that 25% of children age 5 to 15 years with new onset T1DM present with DKA (McCance & Heuther, 2014). DKA is the leading cause of morbidity and mortality in children with T1DM.  Due to the increasing prevalence of T1DM and its severity if left untreated, health care providers in virtually all medical specialties need to be familiar with diagnosing and managing this disease (Palmer & Clegg, 2015). This paper will explore the case of 6-year-old Amy (Appendix A), who presented to the emergency department with clinical symptoms of T1DM and DKA. The differential diagnosis, pathophysiology, diagnostic approach, management and health promotion strategies for T1DM with DKA will be reviewed.

Differential Diagnosis

It is important to consider the differential diagnosis when approaching a patient presenting with illness. This process involves, “distinguishing between two or more diseases with similar symptoms by systematically comparing their signs and symptoms” (Baid, 2006). It is key to examine Amy’s clinical presentation, past medical and family history closely to determine the most probable, most treatable, and most dangerous potential causes of her illness. Amy’s most pertinent presenting symptoms include hyperglycemia, enuresis, fatigue, nausea, and vomiting. The top three differential diagnoses for this presentation include T1DM with DKA, urinary tract infection, and Hyperosmolar Hyperglycemic State (Appendix B).

Type 1 Diabetes with Diabetic Ketoacidosis (DKA)

Amy displayed classic symptoms of new onset T1DM with DKA, including enuresis, or bed wetting, which is likely due to an increase in urine output, called polyuria. Polyuria coupled with increased drinking, or polydipsia, is present in 98% of patients presenting with T1DM (Westerberg, 2013). Vomiting and abdominal pain are present in 46% and 32% of patients respectively (Westerberg, 2013). Amy is hyperglycemic with a fasting plasma glucose (FPG) of 30 mmol/L. FPG over 7 mmol/L meets diagnostic criteria for T1DM (American Diabetes Association (ADA), 2017). Her reported 10-day history of illness is consistent with a diagnosis of T1DM, as patients and families usually report a one to two-week duration of symptoms before seeking medical attention (Gregory et al., 2013). Amy does not have a family history of T1DM, however only 13% of those with T1DM have a parent or sibling with the disease (Hanas, 2015).  Amy’s symptoms strongly suggest that her diagnosis is likely T1DM with DKA.

Urinary Tract Infection (UTI)

Amy’s symptoms of enuresis and abdominal pain make a UTI the second most likely diagnosis. A UTI occurs when a pathogen colonizes anywhere in the usually sterile urinary tract (Zorc, Kiddoo, & Shaw, 2005). UTIs are common in children and account for 5 to 14% of emergency department visits annually (Shaikh, Morone, Bost, & Farrell, 2008). Additionally, 8% of children presenting with urinary symptoms will be diagnosed with a UTI. While UTIs are thought to be relatively benign, if left they can have significant long term complications including renal scarring and hypertension (Zorc et al., 2005). Therefore, it is important to consider a UTI on Amy’s differential. However, Amy is lacking hallmark signs of UTI including fever and dysuria, or painful urination. For this reason, UTI is the second most likely diagnosis for Amy.

Hyperosmolar Hyperglycemic State (HHS)

While HHS is a rare condition, accounting for <1% of hospital admissions, it carries a mortality rate as high as 50-60% and must be considered in patients presenting with hyperglycemia and dehydration (Pasquel & Umpierrez, 2014). While Amy has these two symptoms, HHS most commonly occurs in elderly patients with type 2 diabetes and it is rare for children with new onset T1DM to present with it (Cho et al., 2017). HHS has a more insidious onset. The patient experiences osmotic diuresis from a prolonged period of hyperglycemia, causing extreme fluid and electrolyte depletion (Cho et al., 2017). It can be difficult to distinguish HHS from DKA on clinical exam alone and laboratory testing is needed to differentiate the two disorders. If Amy did have HHS her results would show hyperglycemia and hyperosmolarity with an absence of severe ketosis (Cho et al., 2017). Due to the rarity of this disorder this would be the third diagnosis on the differential for Amy.

Pathophysiology of T1DM

Etiology

T1DM is the most common form of diabetes in the paediatric population (Newhook et al., 2004). Two subtypes of T1DM have been identified, autoimmune (1a) and nonimmune (1b) (ADA, 2009). T1DM type 1a is much more common and will be discussed in detail. T1DM type 1b occurs secondary to other disease processes, such a chronic pancreatitis, or is idiopathic (ADA, 2009). In healthy individuals, the body regulates how much glucose enters the cell using two hormones that are produced in the pancreatic islet of Langerhans. -Cells produce glucagon to increase blood glucose, and β-cells produce insulin to decrease blood glucose by allowing it to enter the cell. T1DM type 1a occurs following selective autoimmune destruction of β-cells, resulting in insulin deficiency and hyperglycemia (Guthrie & Guthrie, 2004).

While the etiology of T1DM is not fully understood, it is widely accepted that it results from a interaction between genetic and environmental factors that promote disease (Xie, Chang, & Zhou, 2014). Although most patients do not have a family history of T1DM, genetic susceptibility plays an important role (Xie et al., 2014). Siblings of those with the disease have a 6% risk of developing the disease, which is 15 times higher than the general population (Xie et al., 2014). Genomic studies have confirmed that T1DM is a multifactorial polygenic disease and identified over 40 loci associated with it (Nosikov & Seregin, 2008, Zhang & Eisenbarth, 2011). The most important determinant of disease is an abnormality in chromosome 6 in the region that controls Major Histocompatibility Complex (MHC) and Human Leucocyte Antigens (HLA) class II alleles, specifically HLA-DP, DQ, and DR (Guthrie & Guthrie, 2004, Zhang & Eisenbarth, 2011). HLA class II genes encode molecules that are active in antigen presentation to CD4+ T cells, and variants in these genes determine 40-60% of T1DM genetic susceptibility (Zhang & Eisenbarth, 2011).

Though genetics play a role, it is not conclusive in identifying which patients will develop T1DM. Studies have shown that only 23 to 50% of monozygotic twins of patients with T1DM will develop the disease (Abel & Krokowski, 2001). Environmental factors such as viral infections, early exposure to cow’s milk, gluten and vitamin D deficiency have been implicated in the development of T1DM (Xie et al., 2014). These environmental antigens have similar amino acid sequences to β-cells, and can induce molecular mimicry. This triggers an activation of T cells prompting a cross-reactive autoimmune response (Xie et al., 2014).

Autoimmune β-cell Destruction

The selective destruction of pancreatic β-cells occurs via two autoimmune mechanisms. The first mechanism involves lymphocyte and macrophages infiltrating the islets causing insulinitis, or inflammation, and death of islet β-cells (McCance & Heuther, 2014). Autoantigens are expressed on the surface of the islet cells and enter systemic circulation and the lymphatic system (McCance & Heuther, 2014). Antigen presenting cells ingest these circulating autoantigens resulting in activation of CD4+ T helper lymphocytes (Th1 cells) (Calderon & Unanue, 2012). When Th1 cells are activated they secrete interleukin-2 (IL-2), which in turn causes β-cell autoantigen specific T cytotoxic lymphocytes (CD8+ cells) to multiply and secrete toxic perforin and granzyme inducing β-cells apoptosis, or cell death (Roep & Peakman, 2011). It has been discovered that individuals with deficiencies in perforin or granzyme are protected against T1DM, and manipulating these enzymes is being explored as the target for future therapies (Thomas, Trapani, & Kay, 2010). Finally, Th1 lymphocytes release interferon stimulating the secretion of inflammatory cytokines, including tumor necrosis factor (TNF ) and interleukin-1 (IL-1), activating macrophages and further contributing to the destruction of β-cells (Haskins & Cooke, 2011).

The second mechanism involved in the destruction of pancreatic of β-cells involves the production of autoantibodies against insulin, islet cells, glutamic acid decarboxylase 65 (GAD65), and other cytoplasmic proteins (McCance & Heuther, 2014). T helper 2 lymphocytes (Th2) are activated and release interleukin 4 (IL-4) stimulating the activation of B lymphocytes, which produce islet cell autoantibodies and anti-glutamic acid decarboxylase (anti-GAD) antibodies (need reference). GAD is an enzyme that is involved in coordinating insulin release in the β-cells (need reference).

In addition to these two mechanisms of destruction, patients with T1DM appear to have abnormalities in their regulatory T cells (Suarez-Pinzon & Rabinovitch, 2001). Regulatory T cell counterbalance the activity of autoreactive T cells to maintain immune homeostasis (Cabrera, Rigby, & Mirmira, 2012). When an imbalance in this relationship occurs, there is a breakdown in self-tolerance allowing the immune system to incorrectly attack healthy cells (Cabrera et al., 2012).

Physiology in Healthy Individuals

In individuals who do not have T1DM, insulin predominates glucagon activity and is produced by the pancreatic β-cells in response to ingested glucose and amino acids (Aronoff, Berkowitz, Shreiner & Want, 2004). Insulin then enters circulation and acts on insulin sensitive cells, muscle and adipose tissue throughout the body. At the cellular level, insulin binds to insulin receptors on cell surfaces stimulating a sequence of enzymatic reactions, resulting in the production of a glucose transporters (GLUT4) (Glaser, 2005). GLUT4 adheres to the cell surface allowing large molecule nutrients like protein and glucose to enter the cell providing it with energy. Excess of glucose promotes gluconeogenesis, storage in a polymerized form known as glycogen in the liver (Aronoff et al., 2004). During fasting the opposite, glycogenolysis, occurs to make glucose available in circulation (Aronoff et al., 2004). Other glucoregulatory hormones in the body including amylin, glucose-dependent insulinotropic peptide (GIP), glucagon-like peptide-1, epinephrine, cortisol and growth hormone work together to maintain serum glucose within a tight range (Aronoff et al., 2004).

Pathophysiology of DKA

Polyphagia and Weight loss

Amy has likely experienced destruction of over 80-90% of her β-cells and is now showing clinical symptoms (McCance & Heuther, 2014). At this stage, insulin deficiency ensues and GLUT4 transporters fail to translocate to the cell membrane to allow glucose to enter the cell (Glaser, 2005). Cells are depleted of glucose needed for energy and the body enters starvation mode. For energy it taps into alternative energy stores and decreases tissue use of glucose (McCance & Heuther, 2014). Pancreatic -cells secrete glucagon inducing glycogenolysis. Counter regulatory hormones including cortisol, catecholamines, and growth hormone are released (Westerberg, 2013). The lack of usable energy stimulates polyphagia, as seen in Amy’s case, as the body attempts to increase glucose (Ozougwu et al., 2013). These efforts to produce more glucose and decrease its consumption further contributes to hyperglycemia.

Proteins are catabolized into amino acids (Westerberg, 2013). Lipids in adipose tissue are mobilized into circulation and largely taken up by the liver, where they are converted through lipolysis into free fatty acids (FFAs) (Westerberg, 2013). FFAs then undergo -oxidation into acetyl coenzyme A, which is normally completely oxidized into carbon dioxide (CO2) and water creating ATP, or energy, for cells in the Krebs cycle (King, 2015). This process is limited by insulin in healthy individuals. In DKA, the liver cannot keep up with the excessive, unrestricted amount of acetyl coenzyme A and diverts the compound to be broken down into ketone bodies (acetoacetate, acetone and β-hydroyburyrate) in a process called ketogenesis (Westerberg, 2013). Ketone bodies are converted into energy through ketolysis and fuel intracellular metabolic functions (Laffel, 1999). This breakdown of lipids and proteins, and lack of glucose leads to weight loss as seen in Amy.

Kussmaul’s Breathing and Fruity Breath Odor

Ketones are strong acids and quickly accumulate increasing hydrogen ion (H+) concentration in the blood resulting in an anion gap metabolic acidosis (King, 2015). The anion gap is a calculated value of negative cations, chloride (Cl) and HCO3, and positive anions, sodium (Na2+) and potassium (K+) and reflects the electroneutrality of the serum (Gabow, 1985). Acidosis has many effects on the body. Central and peripheral chemoreceptors that control respiration are stimulated producing distinctive Kussmaul’s breathing, or alveolar hyperventilation (Kaufmann, Smolle, Flec & Lueger, 1994). This explains Amy’s deep and rapid breathing pattern. The increase in respiration is a compensatory mechanism and results in the patient expiring CO2 through the lungs in attempt to increase pH. Acetone is volatile substance that has a fruity scent, and can often be detected on the breath of patients with DKA. Additionally, acidosis disrupts K+ balance. H+ ions move from the extracellular fluid (ECF) to the intracellular fluid (ICF), in turn K+ shifts out of the cell to maintain cation balance resulting in a normal or elevated K (McCance & Heuther, 2014). However, K+ in the ECF is renally excreted

Polyuria, Polydipsia and Tachycardia

While the cells are starving for glucose, it is accumulating in the serum. Glucose is an osmotically active substance, and its elevation increases plasma tonicity, generating a driving force that shifts water into the extracellular space diluting serum Na (Glaser, 2005, Palmer & Clegg, 2015). Additionally, when glucose concentration in the blood rises above 11.1mmol/L, as Amy’s has, it surpasses the kidneys’ threshold for glucose reabsorption and disrupts the normal concentration gradient (Glaser, 2005, Liamis, Liberopoulos, Barkas, & Elisaf, 2014). Instead of glucose being filtered in the glomerulus and reabsorbed in the proximal tubules an osmotic diuresis occurs (McCance & Heuther, 2014). Instead, glucose along with Na, K, water, ketones and other electrolytes are exerted in the urine in large amounts. This explains Amy’s symptoms of polyuria and enuresis which leads to hypovolemia or dehydration. Due to the volume deficit, patients like Amy, baroreceptors are stimulated and increase heart rate in attempt to maintain adequate cardiac output.

Na is the main cation, or negative ion, in bodily fluids and determines osmolality. The subfornical organ (SFO) area of the hypothalamus lacks the normal blood brain barrier and osmolreceptors are stimulated when osmolality increases or volume is decreased in the ECF (Matsuda et al., 2017). Granular cells in the kidney detect the drop in volume and stimulate the renin-angiotensin-aldosterone system to increase angiotensin II, driving the patient to crave salt and water (Matsuda et al., 2017). The patient experiences a decrease in saliva production and the urge to drink. Usually this works as a feedback loop that is turned off when water is absorbed in the gastrointestinal tract. Due to the massive fluid losses, patients like Amy develop polydipsia in an effort to restore fluid homeostasis.

Abdominal Pain & Vomiting

Diagnosis

Screening for T1DM

The progression towards T1DM can be detected before clinical symptoms appear. Children at high risk of developing the disease can be identified on the basis of HLA genotype and successive development of autoantibodies (Meehan, Fout, Ashcraft, Sachatz, & Haller, 2015). Islet cell antibodies (ICA), anti-tyrosine phosphate antibodies (IA-2), glutamic acid decarboxylase antibodies (GADA), insulin autoantibodies (IAA), and antibodies against zinc transporter (ZnT8) are detectable long before patients present (Abel & Krokowski, 2001, Zhang & Eisenbarth, 2011). Insulin autoantibodies are present in approximately 70% of pediatric patients at the time of diagnosis with T1DM (Williams, 2014). While screening for these markers is only currently performed in research settings, early screening may be part of future clinical practice, particularly for individuals with first-degree relatives with T1DM (American Diabetes, 2017). The use of preventative therapy, such as immune modulation, may be utilized and education can be provided to recognize symptoms and prevent the progression to life-threatening DKA (Zhang & Eisenbarth, 2011).

However, a study by Meehan et al, (2015) suggest that screening for T1DM may not be an economically viable intervention to reduce DKA presentations. Public awareness campaigns and education for primary care providers have been shown in some studies to reduce the cost of T1DM to the health care system. In an Italian study by Vanelli et al., (1999) an awareness program was implemented at a cost of $23,470 over 8 years. The lifetime cost to the system for a patient presenting with DKA was $196,465 compared to $53,365 for patients presenting without DKA (Vanelli et al., 1999). In this study, the frequency of new diagnosis T1DM with DKA fell from 78% to 12.5% (Vanelli et al., 1999). While these results have not been universal in follow up studies, education and awareness can be a powerful tool in preventing DKA in new diagnosis.

Diagnostic Testing for T1DM with DKA

Amy’s disease process has progressed to the point where she is displaying symptoms of DKA. She needs rapid diagnosis and timely treatment to prevent complications. A finger stick capillary glucose was already performed on admission and showed significant hyperglycemia. Since Amy has not had anything orally for 8 hours, this is considered a fasting plasma glucose (FPG) level. To confirm the diagnosis of T1DM with DKA, blood should be collected and tested for electrolytes, blood gas, hemoglobin A1c (HbA1c), complete blood count with differential (CBCD), ketones, urea and creatinine.

Serum electrolytes including sodium, potassium, glucose, chloride, calcium, magnesium, phosphate and a venous blood gas (VBG) are obtained to detect electrolyte and acid-base disturbances (Gregory et al., 2013). Sodium is commonly low due to excessive diuresis. Potassium may be low or normal but should be monitored closely throughout treatment. Arterial blood gas (ABG) monitoring has traditionally been the gold standard for diagnosis and managing patients with suspected DKA. However, ABG sampling can be difficult to obtain, be painful for the patient, and carries risks of complications such as arterial injury, thrombosis, embolism, infection and local hematoma (Kelly, 2006). Additionally, there is a small but significant risk of needle stick injury for the provider obtaining the sample, with the potential consequence of transmitting blood borne illnesses (Kelly, 2006). Instead, venous blood gases (VBG) can be used to monitor acid-base balance in hemodynamically stable patients with DKA who do not present with respiratory failure (Kelly, 2006). Clinicians should be mindful that peripheral venous pH is approximately 0.02 to 0.04 pH units below arterial pH, bicarbonate (HCO3) is around 1 to 2 mEq/L higher, and carbon dioxide (pCO2) is around 3 to 8 mmHg higher than arterial (Brandenburd & Dire, 1998).

HbA1c should be drawn and provides an indirect measure of average blood glucose levels over the preceding 2 to 3 months (ADA, 2017). HbA1c is advantageous over FPG in diagnosis as it does not require fasting, has enhanced pre-analytical stability, and fewer fluctuations during illness and stress (ADA, 2017). However, clinicians should be aware that HbA1c is a slightly more expensive test and has lower sensitivity in some individuals due to age, race, anemias and hemaglobinopathies (ADA, 2017). A HbA1c >6.5% identifies one-third fewer cases of new diagnosis T1BM than a FPG level of >7 mmol/L, and FPG is the gold standard for diagnosing acute presentations (ADA, 2017).

Amy also needs serum ketones drawn. While ketones can be measured in the urine, urinalysis for ketonemia has a sensitivity of 98%, a specificity of 35%, and a positive predictive value of 15% (Westerberg, 2013). This is because urinalysis can only measure acetone and acetoacetate, not -hydroxybutyrate which is the main ketone present in DKA (Westerberg, 2013). Testing for serum -hydroxybutyrate has a sensitivity of 98%, a specificity of 79% and a positive predictive value of 34% (Westerberg, 2013). Since we are poking Amy for blood work anyway this would be an important test to include and will help to rule out the third differential diagnosis of HHS.

A urinalysis should still be performed to assess for the presence of glucose and ketones. Additionally, a urinalysis positive for leukocytes, or white blood cells, and nitrites would raise suspicion of a UTI, the second differential diagnosis for Amy. While urine culture is the gold standard for diagnosing UTIs, urinalysis for either leukocytes or nitrites has a sensitivity of 88% and specificity of 93% (Zorc et al., 2005). Should Amy have these markers a urine culture would be indicated, and due to her age, could be easily collected with a mid-stream sample.

Additionally, infection often precipitates new presentations of T1DM. It is prudent to draw a complete blood count with differential to assess for a potential underlying infection. In patients with DKA, leukocytosis may be present in the absence of infection due to stress and dehydration (Kearney & Dang, 2007). An elevation in bands more accurately predicts infection (Westerberg, 2013). An elevated hematocrit is suggestive of dehydration (Metzger, 2010). Since Amy does not display clinical signs of infection, blood and urine cultures would not be indicated (Westerberg, 2013). Amy’s respiratory status is stable. If she had presented in respiratory distress a chest radiograph would need to be performed to rule out pneumonia (Westerberg, 2013). Renal function can be impacted by dehydration and decreased renal perfusion (Westerberg, 2013). Therefore, a urea and creatinine should be collected. The results of Amy’s laboratory testing, along with the cost of each test, can be found in Appendix C.

Interpretation of Amy’s Diagnostic Test Results

The Canadian Diabetes Association’s 2013 clinical guidelines state that a FPG greater than or equal to 7.0mmol/L, or a random blood glucose greater than or equal to 11.1mmol/L in the presence of hallmark symptoms (polyuria, polydipsia, polyphagia) is sufficient to confirm a diagnosis of diabetes. Additionally, the criteria for DKA includes hyperglycemia with a glucose greater than or equal to 11.1mmol/L, acidosis with venous pH < 7.30 or HCO3 < 15mmol/L, and the presence of ketones in the blood, urine or both (Metzger, 2010). DKA can be further classified into mild (pH 7.20-7.29), moderate (pH 7.10-7.29) and severe (pH <7.10) (Metzger, 2010).

Amy’s laboratory results are consistent with a diagnosis of T1DM with mild DKA. Her blood glucose remains elevated with ketonemia. Amy’s blood gas shows a metabolic acidosis with a low pH, pCO2 and HCO3. Her hematocrit, urea and creatinine are all marginally elevated. These values along with her clinical exam signify dehydration. Amy’s sodium and potassium are on the low end of normal. Her management should involve careful monitoring and replacement of electrolytes. Amy’s urinalysis results do not show evidence of a UTI. The laboratory tests collected also enable the clinician to calculate Amy’s anion gap and serum osmolality, a measure of particles in a fluid compartment (Westerberg, 2013).

Management

Amy is unwell and needs rehydration and insulin therapy. A senior medical staff member should be notified of her condition (Wherrett, Huot, Mitchell, & Pacaud, 2013). She should be admitted to an inpatient ward for treatment and close monitoring. If Amy’s presentation was more severe she would require admission to the intensive care unit. DKA is a complex condition that is associated with significant fluid and biochemical imbalances. Following an established paediatric clinical practice guideline or protocol in the management of DKA is imperative to ensure structured and thoughtful management (Metzger, 2010). For the purposes of this case, British Columbia Children’s Hospital’s DKA Protocol was used to make decisions about Amy’s care.

Ideally, two large bore peripheral intravenous (PIV) lines should be inserted in the emergency department when Amy’s initial blood work is drawn. One PIV can be used for intravenous fluid (IVF) and insulin administration. The other PIV can be used to draw frequent blood work to reduce the need for repeated venipuncture. If unable to secure the second venous access, capillary samples can be used for monitoring. Children experience less stress during venipuncture when seated upright supported by a parent, opposed to being supine and restrained (Sparks, Setlik, & Luhman, 2007). This strategy should be utilized for Amy to increase comfort.

Amy is dehydrated but not in shock as she is maintaining adequate cardiac output to support her blood pressure. Therefore, a fluid bolus should not be administered due to the increased risk of cerebral edema (CE). Normal saline 0.9% with 40mmol/L of potassium chloride should be started immediately and run at 95ml/hr (need to rationalize this amount). The fluid management goal is to correct dehydration over 48 hours. Potassium is added to fluids despite a normal serum potassium as hydration dilutes potassium concentration in the blood and improved renal perfusion increases excretion (Westerberg, 2013). Additionally, correction of acidosis and insulin therapy increases cellular uptake of potassium (Westerberg, 2013).

Amy has presented with a history of abdominal pain, nausea, and vomiting. These symptoms should subside once her treatment has started to take effect and her underlying condition resolves. However, it is important to manage Amy’s symptoms and keep her comfortable. Ondansetron, a selective serotonin reuptake inhibitor (SSRI), is a non-sedating antiemetic that can be used safely in children with DKA (Leung et al., 2015). SSRIs work by blocking presynaptic serotonin receptors on sensory vagal fibers in the gut wall, effectively managing acute nausea and vomiting (Dipiro, 2008). It is available in both an oral wafer and IV formulations. Side effects of ondansetron are limited. Ondansetron 2.4mg IV or PO Q8H PRN can be ordered and administered as needed for comfort.

Evidence suggests insulin administration within the first hour of treatment increases the risk of CE (Wherrett et al., 2013). After one to two hours of fluid replacement, a short-acting insulin infusion should be commenced at 0.1units/kilogram/hour IV without a bolus, as this has also been shown to increase risk of CE and does not result in quicker resolution of acidosis (Wherrett et al., 2013). Insulin is an anti-catabolic and anabolic hormone that, as previously stated, plays an important role in carbohydrate, fat and protein metabolism (Dipiro, 2008). It can only be administered IV or subcutaneously and degraded in the muscle, liver and kidneys (Dipiro, 2008). There are numerous injectable preparations that have variable pharmacologic profiles. Insulin is a high-risk medication, and great care should be taken when ordering and administering it.

Once Amy’s blood glucose reaches 14-17mmol/L, or if her glucose falls more than 5mmol/L in one hour, her fluids should be changed to 5% dextrose with saline and potassium chloride to avoid hypoglycemia (Wherrett et al., 2013). Dextrose should further be increased if her glucose falls below 8mmol/L, instead of discontinuing the insulin infusion as it is needed to resolve acidosis. If Amy’s sodium is climbing (>140) she can be switched to normal saline 0.45%, and her potassium can be adjusted based on her most recent serum level to maintain within normal parameters. Amy has slight hypophosphatemia. Correcting phosphate is not recommended as this is associated with hypocalcemia and hypomagnesemia (Kearney & Dang, 2007). Should Amy’s urine output drop, her fluid and electrolyte management would need to be adjusted accordingly to avoid imbalances.

Amy should be given nothing by mouth until her acidosis has normalized, but can be allowed ice chips for comfort as this decreases the thirst response. She should be monitored with continuous cardiorespiratory monitoring including oxygen saturation, and have her neurological status and vital signs assessed every hour. Amy should be positioned with the head of her bed elevated due to the risk of cerebral edema to encourage optimal venous drainage. Her intake and output should be monitored hourly.

A 12-lead electrocardiogram (ECG) should be performed and Amy’s ECG should be monitored for T wave changes. Peaked T waves indicate hyperkalemia, and flattened or inverted T waves indicate hypokalemia. Her blood glucose should be monitored hourly using a bedside glucometer, and venous blood gas monitored every two hours to assess electrolytes. The administration of sodium bicarbonate is not recommended as it does not improve outcomes and can cause hypokalemia (Adeva-Andany et al., 2014).

Once Amy is awake and metabolically stable (pH >7.30, HCO3 > 18) with glucose stable between 8-12mmol/L, her IV fluids can be transitioned to oral. Subcutaneous (SC) insulin should be started just before a meal when she is feeling well enough to eat normally. This can be a difficult transition and can result in a period of rebound hyperglycemia. The half-life of short-acting IV insulin is 10 minutes. For this reason, SC insulin should be administered 30 minutes to 1 hour before stopping the IV insulin infusion and Amy’s BG should be monitored more frequently during the transition (Kearney & Dang, 2007).

Complications of DKA

Children with diabetes, especially those presenting with DKA, need vigilant care to avoid complications such as cerebral edema (CE), hypoglycemia, renal failure, electrolytes imbalances and shock (Westerberg, 2013). CE is a dreaded complication that is unique to the paediatric population, occurring in 0.7 to 4% of patients with DKA and carrying a mortality rate of 21 to 24% (Westerberg, 2013, Wherrett et al., 2013). The mechanism is CE in DKA is poorly understood. Children have relatively immature auto-regulatory systems (Glaser et al., 2004). They are also less likely to recognize and communicate symptoms and are more likely to have a more severe presentation. Symptoms of CE include headache, irritability, decreased level of consciousness, recurrent vomiting and slowing heart rate (Wherrett et al., 2013).

Developmental and Social Considerations

Early school-aged children like Amy should be encouraged to gain confidence in accomplishing tasks but often lack the fine motor skills, impulse control and cognitive development needed to actively participate in some aspects of their diabetes care (Silverstein et al., 2004). Amy may be able to partake in her care by learning how to test her blood glucose, help in keeping records, and potentially counting carbohydrates (Silverstein et al., 2004). Parents and providers should avoid labelling blood glucose measurements as “good” or “bad” as children of this age group may be sensitive to this kind of language (Halvorson, Yauda, Carpenter & Kaiserman, 2005). Hypoglycemia is a concern in this age group due to inconsistent dietary intake, and can have adverse effects on brain function and development (Silverstein et al., 2004).

Follow Up and Referral

T1DM is a chronic condition and the average lifespan of paediatric patients diagnosed with T1DM is 20 years shorter than the general population (Nosikov & Seregin, 2008). Ideally, all newly diagnosed patients should be referred to a multidisciplinary diabetes team, which consists of a paediatric endocrinologist, dietician, nurse educator, social worker and mental health professional (Silverstein et al., 2004). Amy’s parents, Anne and Dave, should receive comprehensive diabetes management training and should receive regular follow up.

When Amy is stable she should be screened for comorbid conditions that are associated with T1DM, including Autoimmune Thyroid Disease (AITD), Addison’s, and Celiac Disease (CD). AITD occurs in 15 to 30% of children with T1DM, and early detection and treatment prevents poor growth related to hypothyroidism (Wherrett et al., 2013). CD effects 4 to 9% of children with T1DM, with 60 to 70% of cases being asymptomatic (Wherrett et al., 2013). Children with celiac disease require a gluten free diet, which can complicate the management of their diabetes. Additionally, Amy should be screened regularly for hypertension and, when appropriate, for nephropathy, retinopathy, neuropathy, dyslipidemia and psychosocial or psychological disorders (Wherrett et al., 2013).

Health Promotion and Protection

Influenza vaccination should be encouraged to prevent illness that could complicate management of Amy’s T1DM (Wherrett et al., 2013). All children with diabetes should wear a medical alert bracelet to indicate that they have diabetes (Silverstein et al., 2004). Fifty-seven percent of Canadians with diabetes reported that they are unable to adhere to their prescribed treatment plan because of the high out-of-pocket costs associated with medications, supplies and equipment. The average annual cost of these supplies in Alberta is $1,733.61, and $2,046.59 in Ontario.

Appendices

Appendix A

Case Study

Amy, a six-year-old Caucasian female, presents to the Edmonton emergency department with her parents, Anne and Dave, on New Year’s Day. Anne says that Amy is usually a healthy child and has lots of energy. For the past two weeks, Amy has been tired and sleeping in late. Her parents say Amy looks like she’s lost weight despite constant drinks and snacks. Her weight two months ago was measured as 23 kg. Anne and Dave say that she has recently started wetting the bed. Amy has been complaining of abdominal pain for three days, and today she started vomiting and she is refusing to eat or drink anything. Anne and Dave are concerned that she might have a parasite as they visited Mexico on vacation 6 weeks ago.

Past Medical History: healthy, no known allergies

Family History: Amy is an only child. She lives with parents who are both healthy, non-smokers in their mid 30s. Maternal grandmother died of heart disease in her 60s, maternal grandfather is in his 70s and has arthritis. Paternal grandmother is in her 70s and has interstitial cystitis, paternal grandfather is in his 70s and has celiac disease.

Social History: attends local public school, has lots of friends

Immunizations: childhood vaccinations are up to date, has not received the influenza vaccine this year, received additional recommended travel immunizations before vacation to Mexico

Exam: VS T: 36.7 PO, P: 130 bpm, RR: 28, BP: 90/50, oxygen saturation 99% in room air. Reports 4/10 abdominal pain. Weight 20kg (50^th percentile). Height 120 cm (85^th percentile). BMI 15.3 (50^th percentile). She is alert but tired looking, pale, thin, and non-toxic appearing

Head, Eyes, Ears, Nose, Throat: Glascow Coma Score 15/15, eyes and ears are clear, no adenopathy, cracked lips and dry mucous membranes with fruity breath odor

Cardiovascular: Regular rate and rhythm, slight tachycardia, S1/S2

Respiratory: Lung fields clear, effortless tachypnea

Abdominal: Abdomen flat, bowel sounds present x 4 quadrants, generalized tenderness on light palpation, negative Rovsing sign, no palpable mass, spleen or liver edge

Genitourinary/Gastrointestinal: Normal genitalia and anus. Tanner stage 1.

Integumentary: Moderate pulses, capillary refill 4 seconds peripherally, 2 seconds centrally

No rash, edema, or petechiae

Laboratory: finger stick capillary glucose 30 mmol/L (fasting)

Appendix B

Differential Diagnosis	Supporting Data	Non-Supporting Data
Type One Diabetes with Diabetic Ketoacidosis   Incidence: Prevalence:	Polyuria with polydipsia – present in 98% of patients (Westerberg, 2013)   Weight loss – present in 81% of patients (Westerberg, 2013) Fatigue – present in 62% of patients (Westerberg, 2013) Vomiting – present in 46% of patients (Westerberg, 2013) Polyphagia – present in 33% of patients (Westerberg, 2013) Abdominal pain – present in 32% of patients (Westerberg, 2013) Two-week history (mean duration of symptoms before presentation is 10 days) Hyperglycemia Age (diagnosis of diabetes peaks at age 4-6 years) Weight 75th percentile – patients with T1DM are usually normal or underweight (McCance & Heuther, 2014) Dry mucous membranes Winter presentation (Moltchanova et al., 2004) Caucasian – 1.5-2 times more common in white than non-white children (McCance & Heuther, 2014) Geography – more common in Edmonton, Alberta than other areas of Canada (Newhook et al., 2004) Delayed cap refill Tachycardia (120 bpm) Tachypnea (28 bpm)	Recent travel to Mexico   No dyspnea – present in 57% of patients (Westerberg, 2013) Kussmaul breathing Fruity breath
Urinary Tract Infection   Incidence: Prevalence:	Polyuria/enuresis   Polydipsia Abdominal pain and vomiting Dry mucous membranes Delayed cap refill Fatigue Female gender – UTIs are more common in girls after 1 year of age (Zorc et al., 2005) Caucasian background – white children have a 2-4 fold increase compared to black children (Shaw, Gorelick, McGowan, Yakscoe, & Schwartz, 1998) Tachycardia (120 bpm) Tachypnea (28 bpm) – could be related to pain and discomfort	No report of dysuria   No report of malodorous urine Afebrile (Temp 36.7) Age – prevalence is highest in uncircumcised infant males and females <4 years of age (Shaikh, Morone, Bost, & Farrell, 2008) Vomiting Polyphagia with weight loss Two-week history No reported history of UTI No reported history of bowel dysfunction – bladder and bowel dysfunction is present in 40% of toilet trained children presenting with their first UTI Weight loss
Hyperosmolar Hyperglycemic   State (HHS) Incidence: Prevalence:	Abdominal pain and vomiting   Dry mucous membranes Polyuria/enuresis Polydipsia Polyphagia Weight loss Delayed cap refill Fatigue	Age – More common in overweight adolescents and the elderly   Usually hyperglycemia is more pronounced with FPG >33.3mmol/L Fruity breath – sign of ketonemia which is absent in HHS Normal blood pressure – those presenting with HHS are often hypotensive due to excessive fluid loss

Appendix C

Relevant Lab Values

Finger stick capillary glucose on admission 30 mmol/L

Test	Result	Normal Value	Cost
Venous Blood Gas
pH	7.26	7.35-7.45	$6.20
pCO2	30	35-45
HCO3	17	22-26
Electrolytes
Sodium	136	135-145
Potassium	3.4	3.5-5
Glucose	30.2
Chloride	97
Phosphate			$5
Magnesium
Calcium
HbA1c	10%
Renal function			$7
Creatinine	36
Urea	9
			$5
CBC-D			$5
Hemoglobin
Hematocrit			$7
Leukocytes
Bands
Urinalysis
Specific gravity
Leukocytes		Negative	$10
Protein
Nitrites	Negative	Negative
pH
Blood	Negative	Negative
Glucose	Positive	Negative
Ketones	Positive	Negative
Calculated values
Anion gap   Na-(Cl + HCO3)		7 to 13 mEq/L
Serum osmolality   2(Na + K) + 2(glucose/18) + (urea/2.)		285 to 295 mOsm/kg

References

Abel, M., & Krokowski, M. (2001). Pathophysiology of Immune-Mediated (Type 1) Diabetes Mellitus. BioDrugs, 15(5), 291–301.

Adeva-Andany, M. M., Fernández-Fernández, C., Mouriño-Bayolo, D., Castro-Quintela, E., & Domínguez-Montero, A. (2014). Sodium Bicarbonate Therapy in Patients with Metabolic Acidosis. The Scientific World Journal, (1), 1–13. http://doi.org/10.1155/2014/627673

American Diabetes Association (ADA). (2009). Diagnosis and classification of diabetes mellitus. Diabetes Care, 32(1), S62-S67.

Aronoff, S. L., Berkowitz, K., Shreiner, B., & Want, L. (2004). Glucose Metabolism and Regulation: Beyond Insulin and Glucagon. Diabetes Spectrum, 17(3), 183–190. http://doi.org/10.2337/diaspect.17.3.183

Baid, H. (2006). Differential diagnosis in advanced nursing practice. British Journal of Nursing, 15(18), 1007–1011. http://doi.org/10.12968/bjon.2006.15.18.22027

Brandenburg, M., & Dire, D. (1998). Comparison of arterial and venous blood gas values in the initial emergency department evaluation of patients with diabetic ketoacidosis. Annals of Emergency Medicine, 41(4), 459-465.

Calderon, B., & Unanue, E. R. (2012). Antigen presentation events in autoimmune diabetes. Current Opinion in Immunology, 24(1), 119–128. http://doi.org/10.1016/j.coi.2011.11.005

Cabrera, S., Rigby, M., & Mirmira, R. (2012). Targeting regulatory T cells in the treatment of type 1 diabetes mellitus. Current Molecular Medicine, 12(10), 1261-1272.

Cho, Y. M., Park, B. S., & Kang, M. J. (2017). A case report of hyperosmolar hyperglycemic state in a 7-year-old child: An unusual presentation of first appearance of type 1 diabetes mellitus. Medicine, 96(25), e7369. http://doi.org/10.1097/MD.0000000000007369

Devillé, W. L. J. M., Yzermans, J. C., van Duijn, N. P., Bezemer, P. D., van der Windt, D. A. W. M., & Bouter, L. M. (2004). The urine dipstick test useful to rule out infections. A meta-analysis of the accuracy. BMC Urology, 4, 4. http://doi.org/10.1186/1471-2490-4-4

Dipiro, J., Talbert, R., Yee, G., Matzke, G., Wells, B., & Posey, M. (2008). Pharmacotherapy: A Pathophysiologic Approach (7^th ed). McGraw-Hill.

Gabow, P. (1985). Disorders associated with an altered anion gap. Kidney International, 27, 472-487.

Glaser, N. S., Wootton-Gorges, S. L., Marcin, J. P., Buonocore, M. H., DiCarlo, J., Neely, E. K., Barnes, P. B., Bottomly, J., & Kuppermann, N. (2004). Mechanism of cerebral edema in children with diabetic ketoacidosis. The Journal of Pediatrics, 145(2), 164–171. http://doi.org/10.1016/j.jpeds.2004.03.045

Gregory, J. M., Moore, D. J., & Simmons, J. H. (2013). Type 1 diabetes mellitus. Pediatrics in Review, 34(5), 203–215. http://doi.org/10.1542/pir.34-5-203

Guthrie, R., & Guthrie, D. (2004). Pathophysiology of Diabetes Mellitus. Critical Care Nursing Quarterly, 27(2), 113–125.

Haskins, K., & Cooke, A., (2011). CD4 T cells and their antigens in the pathogenesis of autoimmune diabetes. Current Opinion in Immunology, 23(6), 739–745. http://doi.org/10.1016/j.coi.2011.08.004

Kaufmann, P., Smolle, K. H., Fleck, S., Leuger, A. (1994). Ketoacidotic diabetic metabolic dysregulation: pathophysiology, clinical aspects, diagnosis and therapy. Win Klin Wochenschr, 106(5), 119-127.

Kelly, A. (2006). The case for venous rather than arterial blood gases in diabetic ketoacidosis. Emergency Med Australas, 18(1), 64-67.

Kearney, T., & Dang, C. (2007). Diabetic and endocrine emergencies. Postgrad Med J, 83(976), 79-86.

Laffel, L. (1999). Ketone Bodies: a Review of Physiology, Pathophysiology and Application of Monitoring to Diabetes. Diabetes Metac Res Rev, 15, 412–426.

Liamis, G., Liberopoulos, E., Barkas, F., Elisaf, M. (2014). Diabetes mellitus and electrolyte disorders. World J Clin Cases, 2(10), 488-496.

McCance, K.L., & Huether, S.E. (2014). Pathophysiology: The Biological Basis for Disease in Adults and Children (7th ed). Mosby/ Elsevier.

Meehan, C., Fout, B., Ashcraft, J., Schatz, D., & Haller, M. (2015). Screening for T1D risk to reduce DKA is not economically viable. Pediatric Diabetes (16), 565–572.

Merger, S. R., Leslie, R. D., & Boehm, B. O. (2013). The broad clinical phenotype of Type 1 diabetes at presentation. Diabetic Medicine, 30(2), 170–178. http://doi.org/10.1111/dme.12048

Metzger, D. (2010). Diabetic ketoacidosis in children and adolescents: An update and revised treatment protocol. BC Medical Journal, 52(1), 24–31.

Moltchanova, E. V., Schreier, N. (2009). Seasonal variation of diagnosis of Type 1 diabetes mellitus in children worldwide. Wiley Online Library. http://doi.org/10.1111/j.1464-5491.2009.02743.x

Newhook, L. A., Curtis, J., & Hagerty, D., (2004). High incidence of childhood type 1 diabetes in the Avalon Peninsula, Newfoundland, Canada. American Diabetes Association

Nosikov, V. V., & Seregin, Y. A. (2008). Molecular genetics of type 1 diabetes mellitus: Achievements and future trends. Molecular Biology, 42(5), 773–783. http://doi.org/10.1134/S0026893308050142

Ozougwu, J. C., Obimba, K. C., Belonwu, C. D., & Unakalamba, C. B. (2013). The pathogenesis and pathophysiology of type 1 and type 2 diabetes mellitus. Journal of Physiology & Pathophysiology, 4(4), 45-57. http://doi.org/10.5897/jpap2013.0001

Palmer, B. F., & Clegg, D. J. (2015). Electrolyte and Acid–Base Disturbances in Patients with Diabetes Mellitus. The New England Journal of Medicine, 373(6), 548–559. http://doi.org/10.1056/NEJMra1503102

Pasquel, F. J., & Umpierrez, G. E. (2014). Hyperosmolar Hyperglycemic State: A Historic Review of the Clinical Presentation, Diagnosis, and Treatment. Diabetes Care, 37(11), 3124–3131. http://doi.org/10.2337/dc14-0984

Roep, B. O., & Peakman, M., (2011). Diabetogenic T lymphocytes in human type 1 diabetes. Elsevier. http://doi.org/10.1016/j.coi.2011.10.001

Shaikh, N., Morone, N. E., Bost, J. E., & Farrell, M. H. (2008). Prevalence of Urinary Tract Infection in Childhood. The Pediatric Infectious Disease Journal, 27(4), 302–308. http://doi.org/10.1097/INF.0b013e31815e4122

Shaw, K. N., Gorelick, M., McGowan, K. L., Yakscoe, N. M., & Schwartz, J. S. (1998). Prevalence of urinary tract infection in febrile young children in the emergency department. Pediatrics, 102(2), e16.

Silverstein, J., Klingensmith, G., Copeland, K., Plotnick, L., Kaufman, F., Lafeel, L., Deeb, L., Grey, M., Anderson, B., Holzmeister, L. A., & Clark, N. (2004). Care of Children and Adolescents with Type 1 Diabetes. Diabetes Care, 28(1), 186–212.

Sparks, L. A., Setlik, J., Luhman, J. (2007). Parental holding and positioning to decrease IV distress in young children: a randomized controlled trial. Journal of Pediatric Nursing, 22(6), 440-447.

Suarez-Pinzon, W. L., & Rabinovitch, A. (2001). Approaches to type 1 diabetes prevention by intervention in cytokine immunoregulatory circuits. International Journal of Experimental Diabetes Research, 2(1), 3–17.

Thomas, H. E., Trapani, J. A., & Kay, T. W. H. (2010). The role of perforin and granzymes in diabetes. Cell Death and Differentiation, 17(4), 577–585. http://doi.org/10.1038/cdd.2009.165

Vanelli, M., Chiari, G., Ghizzoni, L., Costi, G., Giacalone, T., & Chiarelli, F. (1999) Effectiveness of a prevention program for diabetic ketoacidosis in children. An 8-year study in schools and private practices. Diabetes Care, (22), 7–9.

Westerberg, D. P. (2013). Diabetic ketoacidosis: evaluation and treatment. American Family Physician, 87(5), 337–346.

Wherrett, D., Huot, C., Mitchell, B., & Pacaud, D. (2013). Type 1 Diabetes in Children and Adolescents. Canadian Journal of Diabetes, 37(S1), S153–S162. http://doi.org/10.1016/j.jcjd.2013.01.042

Williams, A., (2014). Insulin autoantibodies. Diapedia. Accessed January 6, 2018 from https://www.diapedia.org/21042821233/rev/11

Xie, Z., Chang, C., & Zhou, Z. (2014). Molecular Mechanisms in Autoimmune Type 1 Diabetes: a Critical Review. Clinical Reviews in Allergy & Immunology, 47(2), 174–192. http://doi.org/10.1007/s12016-014-8422-2

Zhang, L., & Eisenbarth, G. (2011). Prediction and prevention of Type 1 diabetes mellitus. Journal of Diabetes, 3(1), 48–57. http://doi.org/10.1111/j.1753-0407.2010.00102.x

Zorc, J. J., Kiddoo, D. A., Shaw, K. N. (2005). Diagnosis and Management of Pediatric Urinary Tract Infections. Clinical Microbiology Reviews, 18(2), 417-422.