Two weeks after being diagnosed by his Primary Care Provider (PCP) with T2DM, a 34 year-old white male was referred to the Diabetes Specialty Care Team for education and evaluation. Despite a 40-pound weight loss over the past two months, the PCP diagnosed the patient with T2DM. The patient also reported having vision problems for two months prior to diagnosis of T2DM. As a result of his history of Pigmentation Disbursement Syndrome, the patient initially made an appointment with an ophthalmologist. He believed that his eye disorder was the cause of his vision changes. However, after several appointments the ophthalmologist determined that the symptoms were unrelated to his eye condition and suggested he see his PCP for further evaluation. The patient also reported symptoms of polyuria, polydipsia, fatigue and frequent headaches for several weeks in addition to blurry vision and the 40-pound weight loss. Results of laboratory tests ordered by the PCP at the initial appointment are listed in Table 1.

Figure 1

Table 1: Lab Results From PCP and Initial Appointment to Diabetes Specialty Care Team

A lipid panel and dipstick urinalysis was not available for review. Despite these lab results indicating early metabolic acidosis, the patient denied having any symptoms of diabetic ketoacidosis (DKA); a potential risk. The symptoms of DKA include lethargy, confusion, extreme fatigue, abdominal pain, vomiting, extreme thirst, or dry mucous membranes. [₂]

The fasting blood glucose during the two weeks since the patient’s initial diagnosis ranged from 227 - 455 mg/dL. The post-prandial glucose and bedtime glucose readings ranged from 240 - 595 mg/dL. At the initial appointment with the PCP, the patient had been placed on metformin (Glucophage ® ) 500 mg twice daily and pioglitazone (Actos ® ) 15 mg once daily. However, he continued to experience polydipsia, polyuria, headaches, and blurry vision.

The patient’s past medical history was significant for mild-intermittent asthma since childhood, seasonal allergic rhinitis, diverticulosis, and pigment dispersion syndrome of the eye. This chronic problem is the result of a loss of pigment from the neuroepithelial posterior surface of the iris, which is spread throughout various structures of both anterior and posterior chambers. The pigment is carried forward and deposited along the routes of aqueous flow, creating decreased outflow of the trabecular meshwork. The decreased outflow can potentially result in glaucoma. [₃₄] The patient believed that his blurry vision was related to his chronic eye problem, and initially sought care from his ophthalmologist.

Allergies included penicillin. His surgical history was significant for an appendectomy and bilateral inguinal hernia repairs as a child. Current medications included montelukast (Singular ® ) 10 mg p. o. once daily; mometasone (Nasonex ® ) nasal spray 50mcg/spray-2 sprays each nostril once daily; metformin (Glucophage ® ) 500 mg p. o. twice daily (recently started by PCP), and pioglitazone (Actos ® ) 15 mg once daily (recently started).

Family history was significant for hypertension (HTN) and elevated cholesterol. There was no family history of T1DM or T2DM diabetes in his family.

The patient was married and had one three year-old child. His wife was very supportive and accompanied him to his appointments. He denied smoking, alcohol or illicit drug use. He was employed full time and had insurance coverage.

1. Type 1 Diabetes Mellitus

2. Type 2 Diabetes Mellitus

3. Ketosis-prone Type 2 Diabetes Mellitus (KP-T2DM)

4. Latent Autoimmune Diabetes in Adults (LADA)

5. Diabetic Ketoacidosis (DKA)

6. Hyperglycemic Hyperosmolar Nonketotic State (HHNS)

7. Ketosis with metabolic acidosis

Differentiating between T1DM and T2DM is not always straightforward. The patient presented with the typical symptoms of hyperglycemia, including polyuria, polydipsia, blurry vision, and fatigue. In addition, he had experienced a 40-pound weight loss over the past two months. The in-office laboratory findings at the first appointment to the diabetes specialty care team revealed 4+ ketones in the urine, urine glucose greater than 2000+, and a random fingerstick glucose of 395 mg/dL. Review of the laboratory ordered by the PCP two weeks earlier was significant for early metabolic acidosis, as exhibited by the low C0₂ (20 mEq/L), in conjunction with extremely high urine ketones, despite a normal anion gap. The reported C0₂ in the serum refers to the total C0₂ in the blood, including that contained in bicarbonate. More than 70% of C0₂ in the blood is in the form of bicarbonate. [₅] Causes of non-anion-gap acidosis include ketosis with ketone excretion. [₆] Metabolic acidosis is defined as a decrease in bicarbonate. Metabolic acidosis may be caused by an overproduction of keto-acids with a respiratory alkalosis due to the compensatory increase in ventilation, in an effort to decrease C0₂ and bicarbonate (H₂C0₃) levels. [₅]

Polyuria is a result of osmotic diuresis secondary to sustained hyperglycemia. Blurry vision and increased thirst are also a result of this hyperosmolar state. Weight loss is due to the breakdown of fat and muscle stores for the production of energy. [₂] According to Gardner and Shoback, even a small amount of insulin can prevent the development of ketosis by inhibiting lipolysis in the adipose stores. Therefore, development of ketonuria is an indication of absolute insulin deficiency. [₂] Ketonuria is very common in T1DM, but is rarely seen in T2DM, with the exception of KP-T2DM, where this finding can be seen. [₇] The presence of urinary ketones alone should cause the practitioner to consider T1DM as a differential diagnosis, and insulin initiation is warranted.

Acute life-threatening complications of uncontrolled diabetes are hyperglycemia with ketoacidosis or nonketotic hyperosmolar syndrome. [₈₉] Of the two, only diabetic ketoacidosis (DKA) presents with elevated ketone bodies in the blood and urine ketones. In DKA, the lack of insulin stimulates lipolysis in the adipose tissue, and ketogenesis in the liver, resulting in increased free fatty acids and a rise in hydrogen ions leading to metabolic acidosis. [₁₀] The hyperglycemic, hyperosmolar, nonketotic state (HHNS) is characterized by severe hyperglycemia, hyperosmolality, and dehydration without significant ketosis. A partial or relative insulin deficiency leads to decreased glucose utilization by the muscles, and increased hepatic glucose output. The result is a hyperglycemia that leads to glycosuria, osmotic diuresis, and dehydration. The presence of even a small amount of circulating insulin can prevent ketosis by inhibiting lipolysis. The inhibition of lipolysis and subsequent lack of available free fatty acids restricts the rate at which ketones are formed, thereby preventing severe ketosis. If a patient is unable to maintain adequate fluid intake, severe dehydration can occur, leading to renal insufficiency, decreased glucose excretion, and a marked rise in serum glucose and osmolality. Serum osmolality above 320-330 mosm/kg results in water being drawn out of cerebral neurons, resulting in mental obtundation and coma. [₂]

1. T1DM

2. Ketosis with metabolic acidosis.

Laboratory Tests

• Complete metabolic profile

• Lipid profile

• Urine microalbumin

• Islet cell autoantibodies (ICAs) included: 1) insulin autoantibodies (IAA), 2) protein

• tyrosine phosphatase (IA-2A), and 3) glutamic acid decarboxylase – 65 (GAD-65).

• C-peptide level

Diabetes mellitus is a chronic disorder of the metabolism due either to a deficiency of insulin secretion, a decreased effectiveness of insulin, or both. [₂] The etiology is believed to be the result of an interaction between the environment and genetics of susceptible individuals. The fasting plasma glucose is the preferred test for diagnosing diabetes in children and non-pregnant adults. [₈] Although the 75 gm oral glucose tolerance test is slightly more accurate, the inconvenience and time required make this diagnostic test less desirable. The provisional diagnosis of diabetes requires one of the following criteria: 1) fasting plasma glucose greater than 126 mg/dL after an eight hour fast; OR 2) symptoms of hyperglycemia (polyuria, polydipsia, and unexplained weight loss) and a random plasma glucose of greater than 200mg/dL; OR 3) a 75 gm 2-hour oral glucose tolerance test (OGTT) of greater than 200 mg/Dl (Table 2). The diagnosis must be confirmed on a subsequent day in the absence of unequivocal hyperglycemia. Pre-diabetes is the term now used for patients with impaired fasting glucose (IFG) or impaired glucose tolerance (IGT). [₈]

Figure 2

Table 2: Diagnostic Criteria for Impaired Fasting Glucose, Impaired Glucose Tolerance and Diabetes Mellitus []

An important clinical sign differentiating T1DM from T2DM is insulin deficiency, which most often presents with ketonuria and/or ketoacidosis. This patient was suspicious for T1DM due to his weight loss, metabolic acidosis, and the ketones present in the urine. Although most patients with T1DM are either children or adolescents, the ADA [₈] states that age should not be a determining factor in T1DM. Immune mediated diabetes can occur at any age, even in the 8^th and 9^th decades of life. [₈] KP-T2DM and LADA were differential diagnoses for this patient. Diagnostic testing for islet cell antibodies is a definitive way to determine if a patient has autoimmune T1DM versus KP-T2DM. [₁₁₁₂] A pressing concern was that this patient was in metabolic acidosis and could potentially develop DKA. Electrolytes were drawn to evaluate the patient’s acid/base status and determine if there was any worsening in his metabolic acidosis. A lipid profile was ordered since diabetic patients are at higher risk for lipid abnormalities and should be screened. Urine microalbumin was also ordered to evaluate for diabetic nephropathy. [₉]

According to the ADA, the four clinical categories of diabetes are 1) Type 1 diabetes mellitus (T1DM); 2) Type 2 diabetes mellitus (T2DM); 3) Other specific types of diabetes, such as single genetic defects in beta-cell function or action, diseases of the exocrine pancreas, or chemical or drug induced diabetes; and 4) Gestational diabetes. T1DM accounts for only 5-10% of the cases of diabetes worldwide. T1DM is further divided into two categories: Type 1-A (Immune-mediated), and Type 1-B (Idiopathic, non-autoimmune). [₈] The majority of cases of T1DM are Type 1-A, which is caused by destruction of the beta cells in the islets of Langerhans through either a humoral or cellular autoimmune response. Type 1-A, or immune mediated T1DM is responsible for 90-95% of all cases of the disease. [₁₃] Positive ICAs are a strong indicator of immune mediated T1DM. Antibodies to several of the autoantigens, such as glutamic acid decarboxylase-65 (GAD65), protein tyrosine phosphatase IA-2 (IA-2), and insulin autoantibody (IAA), are highly predictive of T1DM. [₁₄]

Some researchers have suggested additional categories of diabetes such as Latent Autoimmune Diabetes of Adults (LADA) and a nonautoimmune subcategory of Type1-B known as Fulminant Type-1. [_11121516] LADA has a more insidious onset, and can usually be managed with oral hypoglycemics initially. Patients with LADA also have a phenotypic presentation more closely resembling T2DM, including obesity and central adiposity. Patients with LADA typically do not require insulin at the time of initial diagnosis. The term LADA was created to describe a subset of adult patients that are initially non-insulin requiring, but have the autoimmune markers of T1DM and go on to require insulin within a few years. [₁₁] The Immunology of Diabetes Society has defined LADA as follows: 1) age over 30; 2) positive for at least one of the four antibodies commonly found in T1DM (ICA, IAA, IA-2, and GAD-65); and 3) not requiring treatment with insulin within the first 6 months after diagnosis. [₁₆₁₇]

Ketosis-prone T2DM is characterized by an acute presentation of ketoacidosis, a strong family history of T2DM, negative autoantibodies, and an intermediate insulin response to meals. In addition, these patients do not require long-term insulin administration to maintain glycemic control. Once euglycemia is achieved with insulin at the initial event, patients can be switched to oral hypoglycemics. [₁₂] Other researchers have labeled this subtype “Flatbush diabetes”, Idiopathic type-1 diabetes, and type 1-b diabetes. These patients are usually obese, African-American or Hispanic, with negative genetic and autoantibody markers. [₁₈] Although these patients may be managed with oral hypoglycemics for some time, they progress to insulin requirement more rapidly than patients with T2DM. [₁₂]

The patient was asked to monitor blood glucose four times a day for the next two weeks: fasting each morning, at bedtime each night, and before and two-hours after one meal each day. He was instructed to alternate which mealtime blood sugar to check each day. In addition, he was asked to test his glucose at 3:00 or 4:00 AM at least two times in the next two weeks. He was instructed to keep a three-day food diary and glucose measurements and bring these to the next appointment. The timing and frequency of self-measurement of blood glucose (SMBG) depends upon the type of diabetes, the needs and goals of the patient and the level of glucose control desired. [₉] One of the main reasons for self-glucose measurement is to assist in medication adjustment, diet changes, activity level, or all 3 factors. [₁₉] T1DM patients ideally should monitor their blood sugar before meals, after meals and at bedtime. Since this schedule can mean checking glucose levels 7-8 times a day, a realistic alternative is to have the patient check plasma glucose before every meal, two hours after one (different) meal each day, and at bedtime every day for a limited time to establish where the elevations occur most. This invaluable information allows the practitioner to adjust the patient’s medications and insulin appropriately. Diabetes is a disease that is primarily controlled by the patient. Diet, exercise, and blood glucose monitoring as well as appropriate medication management are important in the control of diabetes. Table 3 lists the American Diabetes Association (ADA) recommendations for blood glucose targets. The American Association of Clinical Endocrinologists (AACE, 2002) has stricter guidelines for blood glucose targets [₂₀] (see table 3 below).

Figure 3

Table 3: ADA and AACE Plasma Glucose Goals for Patients with Diabetes Mellitus

For diabetic patients taking two-to-three insulin injections per day or more, the blood sugar should be monitored three to four times per day, usually before giving an injection and at bedtime. [₉] For patients with continued elevation of fasting glucose, or patients who may have low blood glucose reactions during the night, a 3 AM check may also be needed to differentiate between the “dawn phenomenon”, the “Somogyi effect”, and waning bedtime or evening insulin levels. [₂₉₁₉]

Correct differentiation between these phenomenon is important for proper insulin adjustment. Pre-breakfast hyperglycemia is sometimes due to the Somogyi effect in which nocturnal hypoglycemia causes release of counter-regulatory hormones that produce hyperglycemia by 7:00 AM. However, the more common cause of pre-breakfast hyperglycemia in patients with T1DM and the majority of patients with T2DM is the waning of evening or bedtime insulin and/or the “dawn phenomenon”. The “dawn phenomenon” is due to the normal physiologic reduction in insulin sensitivity to insulin between 5 AM and 8 AM as a result of spikes of growth hormone that were released at the onset of sleep. [₂] An extremely low blood glucose level at 3 AM indicates that pre-breakfast hyperglycemia is a result of the “Somogyi effect”, while a normal or slightly elevated glucose level is indicative of the “dawn phenomenon”. An elevated 3 AM blood glucose level indicates waning evening or bedtime insulin levels. Table 4 illustrates the differences and recommended insulin adjustments.

Figure 4

Table 4: Patterns of Overnight Glucose Levels Due to Various Causes []

1. Initiate Insulin

2. Discontinue Metformin and Pioglitazone:

3. Begin Aspirin (ASA) 81 mg once daily

The patient was given one-on-one counseling and education regarding the pathophysiology of diabetes and the signs, symptoms and actions to take for hypoglycemia, as measured by a blood sugar level of 70mg/dl or less. A prescription for glucagon 1mg kit was given, and his wife was instructed to give him 0.5 to 1mg subcutaneously or intramuscularly if he should become unconscious. The patient was also counseled to use either the glucose tablets or the glucose gel available at the pharmacy or to drink some orange juice or eat a small piece of candy if his blood glucose fell below 70 mg/dl if he became symptomatic; i.e., shaky, diaphoretic, or confused. Gaining an understanding of the earliest symptoms of not only the general symptoms of hypoglycemia such as hunger, perspiration, nervousness, and confusion, but also their individual initial symptoms may help prevent more severe hypoglycemia episodes from occurring, and allows these events to be easily managed with a snack or another glucose source. Educating patients on the differences between major and minor hypoglycemia and how to manage each one is an important part of diabetes education. [₂₈] The patient and his wife were referred for Diabetes Self-Management Education (DSME) with a Certified Diabetes Educator (CDE) for meal planning, carbohydrate counting, and insulin administration education. DSME can help patients obtain tighter glycemic control and achieve more positive outcomes. [₂₉]

The patient was scheduled to follow-up with the Diabetes Specialty Team in 2 days and asked to bring his logs and food diary. Intense education and frequent follow-up would be needed to ensure that he had an adequate understanding of proper insulin administration, carbohydrate counting, dietary changes, his medication regimen, and to ensure that his ketosis and metabolic acidosis were resolving. Adjustments in his medication would continue until his blood glucose levels were controlled and his symptoms alleviated. [₈]

This author would like to thank the following individuals for their support and assistance in the editing of this manuscript: Pamela Willson, PhD, RN, FNP-BC, Terry Dickey, MSN, RN, FNP/GNP-BC, and Amber Jones, ME. In addition, this author would like to extend many heartfelt thanks to Joanne Hickey, PhD, RN, ACNP-BC, FCCM, FAAN for her continuous support and motivation, and to Susan Ruppert, PhD, RN, ANP-BC, NP-C, FCCM, FAANP for her continued encouragement during the writing of this manuscript.