12: Diabetes

The differential diagnosis of diabetes mellitus (DM) is actually a classification of the different causes of diabetes:

1. Type 1 DM

   1. Of the persons with DM in Canada, the United States, and Europe, 5–10% have type 1.

   2. Caused by cellular-mediated autoimmune destruction of the pancreatic beta cells in genetically susceptible individuals, triggered by an undefined environmental agent

      1. Some combination of antibodies against islet cells, insulin, glutamic acid decarboxylase (GAD₆₅), or tyrosine phosphatases IA-2 and IA-2beta are found in 85–90% of patients.

      2. Strong HLA association

      3. Risk is 0.4% in patients without family history, 5–6% in siblings and children, and 30% in monozygotic twins.

      4. Patients are also prone to autoimmune thyroid disease, Addison disease, vitiligo, celiac sprue, autoimmune hepatitis, myasthenia gravis, and pernicious anemia.

   3. Occasionally occurs without a defined HLA association or autoimmunity in patients of African or Asian ancestry

   4. Insulin therapy is always necessary.

   5. Patients are at high risk for diabetic ketoacidosis (DKA).

2. Type 2 DM

3. Other, less common causes of diabetes

   1. Genetic defects of beta cell function or insulin action

   2. Exocrine pancreatic diseases (pancreatitis, trauma, infection, pancreatectomy, pancreatic carcinoma)

   3. Endocrinopathies (acromegaly, Cushing syndrome, glucagonoma, pheochromocytoma)

   4. Medications (especially corticosteroids)

   5. Infections

4. Gestational diabetes

Type 1 DM generally occurs in children, although approximately 7.5–10% of adults assumed to have type 2 DM actually have type 1, as defined by the presence of circulating antibodies. Type 2 DM is becoming more prevalent in teenagers and young adults, presumably related to the increased prevalence of obesity.

In most patients, the distinction between type 1 and type 2 DM is clear. Thus, the primary tasks of the clinician are to determine who should be tested for diabetes, who has diabetes, which complications to monitor, and how to treat the patient.

Mrs. D’s pretest probability of diabetes is high because of 2 pivotal points in her history, the polyuria and the positive family history. Excess fluid intake and diseases that cause true polyuria, defined as urinary output of > 3 L/day, should also be considered. Bladder dysfunction and urinary tract infection generally cause frequent, small volume urination. Since patients sometimes have trouble quantifying the amount of urine produced, causes of small volume urinary frequency should be kept in the differential. Table 12-1 lists the differential diagnosis.

Leading Hypothesis: Type 2 DM

Textbook Presentation

Patients with type 2 DM can have the classic symptoms of polyuria, polydipsia, and weight loss. The presentation can also be more subtle, with patients complaining that they feel tired or “just not right.” Many patients are asymptomatic; the diagnosis is made after plasma glucose testing. The complications of diabetes may already be present by the time patients seek medical attention.

Disease Highlights

1. Caused by a combination of impaired insulin secretion and insulin resistance with no evidence of autoimmunity

2. Accounts for 90–95% of cases of DM, with prevalence in the United States of about 13–14%; up to 50% of patients are unaware that they have DM.

3. The lifetime risk of diabetes developing in individuals born in 2000 is estimated to be 32.8% for males and 38.5% for females; rates are as high as 50% for African American and Hispanic women.

4. Strong genetic component

   1. In the United States, type 2 DM is 2–6 times more prevalent among African Americans, Native Americans, Pima Indians, and Latinos than among whites.

   2. 39% of patients have at least 1 parent with diabetes

   3. 60–90% concordance in monozygotic twins

   4. The lifetime risk of a first-degree relative of a patient with type 2 DM is 5–10 times higher than that of age- and weight-matched individuals without a family history.

5. The most important risk factor is obesity, which induces insulin resistance.

   1. The relative risk of diabetes developing in a woman who has a body mass index (BMI) > 35 kg/m² is 93, compared with a woman who has a BMI < 22 kg/m².

   2. The relative risk of diabetes developing in a man who has a BMI > 35 kg/m² is 42, compared with a man who has a BMI < 23 kg/m².

6. DKA develops less often in patients with type 2 DM than those with type 1; however, DKA can occur in persons with type 2 DM. Recent data show that two-thirds of patients with DKA have type 1 DM, and one-third have type 2 DM.

   Do not assume all patients with DKA have type 1 DM; DKA can develop in persons with type 2 DM.

7. Risk factors for type 2 DM include

   1. Age ≥ 45

   2. BMI ≥ 25 kg/m²

   3. A first-degree relative with diabetes

   4. Physical inactivity

   5. Being a member of a high-risk ethnic group (African American, Latino, Native American, Asian American, Pacific Islander)

   6. Having delivered a baby weighing > 9 pounds or having had gestational DM

   7. Hypertension

   8. High-density lipoprotein [HDL] cholesterol < 35 mg/dL or triglycerides > 250 mg/dL

   9. Polycystic ovary syndrome

   10. Vascular disease

   11. Prediabetes (impaired glucose tolerance [IGT], impaired fasting glucose [IFG], or mildly elevated HgbA_1c; see Evidence-Based Diagnosis section for definitions)

       1. Patients with either IGT or IFG have a 5–10% annual risk of developing diabetes; those with both have a 10–15% annual risk.

       2. Patients with an HgbA_1c in the prediabetes range have an annual risk of 5–10%.

       3. These annual risks are 10–20 times higher than in people without prediabetes.

8. Screening for diabetes

   1. American Diabetes Association (ADA) recommends screening patients beginning at age 45.

      1. Screening should be done at any age in patients with a BMI ≥ 25 and 1 of the additional risk factors listed above.

      2. Screening should be done every 3 years in patients with normal results, and every 1–2 years in patients with prediabetes.

   2. In 2008, the US Preventive Services Task Force recommended screening asymptomatic adults with sustained BP > 135/80 mm Hg. The Task Force concluded that evidence is insufficient to assess the benefits and harms of routine screening in asymptomatic patients with BP of 135/80 mm Hg or lower.

Evidence-Based Diagnosis

1. Table 12-2 lists the ADA diagnostic criteria for diabetes and prediabetes.

2. The ADA recommends that all abnormal results be confirmed with repeat testing.

3. Most physicians screen with either the fasting plasma glucose (FPG), HgbA_1c, or both; the oral glucose tolerance test is used primarily to screen for gestational diabetes.

   1. FPG is widely available and inexpensive; the primary disadvantage is the need for fasting at least 8 hours.

   2. The HgbA_1c is now globally standardized and does not require fasting.

      1. Falsely low values can occur in patients with hemoglobinopathies, active hemolysis, and stage 4 or 5 chronic kidney disease.

      2. Falsely high values can be seen in iron deficiency.

      3. African American patients tend to have slightly higher HgbA_1c levels than white patients (0.2–0.3%).

4. Patients may have abnormal results on 1 or both tests; in 1 study, one-third of new diabetes cases were detected by FPG testing only, one-third by HgbA_1c testing only, and one-third by both tests.

Patients with normal HgbA_1c levels may still have diabetes by fasting glucose criteria.

Treatment of Prediabetes

1. The goals are to prevent or delay the onset of diabetes and to optimize other cardiac risk factors.

2. Large randomized trials have shown that lifestyle modification or medication can prevent or delay diabetes.

   1. Finnish patients with IGT were randomized to brief diet/exercise counseling or intensive individualized instruction.

      1. There was a 58% relative reduction in the development of diabetes in the intensive group, (NNT = 22 to prevent 1 case of DM over 1 year; NNT = 5 to prevent 1 case of DM over 5 years).

      2. The study cohort has been monitored for 10 years post intervention; the group of patients initially assigned to the intensive lifestyle intervention maintained a 43% relative risk reduction in the development of diabetes.

   2. Patients in the United States (45% African American or Hispanic) were randomized to intensive diet/exercise program, metformin, or placebo.

      1. There was a 58% relative reduction in the development of DM in the intensive diet/exercise group and a 31% relative reduction in metformin group.

      2. NNT = 7 over 3 years to prevent 1 case of diabetes for the intensive diet/exercise group, and NNT = 14 for the metformin group.

      3. The patients initially assigned to the intensive diet/exercise program maintained a 43% relative risk reduction in the development of diabetes over 10 years.

   3. Acarbose, orlistat, and pioglitazone have also been studied, but the ADA does not recommend their use in diabetes prevention.

      Lifestyle modification is the best way to prevent or delay the onset of diabetes.

3. Optimal lifestyle modification goals are 150 minutes of aerobic exercise weekly and losing 7% of the baseline body weight.

4. None of the drugs studied is FDA approved for diabetes prevention; the ADA recommends considering metformin in patients under 60 with a BMI > 35 and/or with progressive hyperglycemia.

5. The goal of hypertension therapy in patients with prediabetes is to achieve a BP < 140/90 mm Hg.

6. Lipids should be treated according to current guidelines for nondiabetic patients (see Chapter 23, Hypertension).

Mrs. D does not meet diagnostic criteria for diabetes, but she does have prediabetes. This does not cause glycosuria of a degree sufficient to cause urinary frequency. A urinary tract infection is ruled out by the normal urinalysis. She has increased her water consumption, so excess fluid intake is a likely cause of her symptoms. Bladder dysfunction should be considered if her symptoms do not resolve with reduction in fluid intake. Diabetes insipidus and primary polydipsia are rare diseases that do not need to be considered unless she has a documented urinary output of more than 3 L/day. The next diagnostic test should be reducing her fluid intake.

Clearly, Mrs. D now has type 2 DM. At this point, in addition to starting treatment, the clinician should focus on identifying and managing diabetic complications and associated cardiovascular risk factors rather than ruling out other diagnoses (Table 12-3).

Leading Hypothesis: Diabetic Complications

1. Retinopathy

Textbook Presentation

Most patients with retinopathy are asymptomatic. Other patients experience either gradual or sudden vision loss.

Disease Highlights

1. Most common cause of new cases of blindness in adults aged 20–74 years

2. Incidence and risk of progression have declined over the past 30 years, with a 77% decrease in the annual incidence of retinopathy in patients with type 1 DM.

3. Stages of diabetic retinopathy (DR)

   1. Nonproliferative (NPDR)

      1. Earlier stage of DR

      2. Earliest signs are microaneurysms and retinal hemorrhages

      3. Progressive capillary nonperfusion leads to ischemia, manifested by increasing cotton wool spots, venous beading, and intraretinal vascular abnormalities.

   2. Proliferative diabetic retinopathy (PDR)

      1. Most advanced form of DR

      2. Progressive retinal ischemia causes formation of new blood vessels on the retina or optic disk.

      3. The new vessels bleed, leading to vision loss because of vitreous hemorrhage, fibrosis, or retinal detachment.

   3. Diabetic macular edema (DME)

      1. Can develop at any stage of retinopathy

      2. Now the leading cause of vision loss in persons with diabetes

      3. Increased vascular permeability causes plasma leaks from the macular vessels, leading to swelling and formation of hard exudates at the central retina.

4. Risk factors

   1. Most consistently identified risk factors are duration of DM, elevated HgbA_1C level, hypertension, hyperlipidemia, pregnancy, nephropathy.

   2. Less consistently identified risk factors include obesity, smoking, moderate alcohol consumption, physical inactivity.

Evidence-Based Diagnosis

1. Evaluation should include dilated indirect ophthalmoscopy or fundus photography, or both, by an ophthalmologist.

2. Patients with type 1 DM should have an exam within 5 years of disease onset, followed by at least annual exams.

3. Patients with type 2 DM should have an exam at the time of diagnosis, followed by at least annual exams.

   All patients with type 2 DM need eye exams by an ophthalmologist at least annually.

Treatment

1. Glycemic control

   1. In persons with type 1 DM without retinopathy, the risk of developing DR is reduced 76% by tight control (HgbA_1C 7.2% vs 9.1% in the Diabetes Control and Complications Trial).

   2. In persons with type 1 DM with retinopathy, the risk of progression is reduced by 54% by tight control.

   3. In persons with type 2 DM, better control reduces the risk of microvascular complications (retinopathy and nephropathy) by 16–25%. (HgbA_1C 7% vs 7.9% in the United Kingdom Prospective Diabetes Study [UKPDS] [1998]; HgbA_1C 6.5% vs 7.2% in the ADVANCE trial [2008].)

   4. In persons with type 2 DM, there is a 35% reduction in the risk of microvascular complications for every percentage point decrease in HgbA_1C.

2. Better BP control reduces the risk of progression of retinopathy.

3. Aspirin neither improves nor worsens retinopathy; the presence of DR is not a contraindication to aspirin therapy.

4. Laser photocoagulation is indicated for PDR, DME, and selected cases of severe NPDR.

5. Anti-vascular endothelial growth factor, given by intraocular injection, improves vision in patients with DME.

6. Emerging therapies include intravitreal fluocinolone and possibly fenofibrate.

2. Neuropathy

Textbook Presentation

Diabetic peripheral neuropathy (DPN) can be focal but classically presents as paresthesias or burning pain in a “glove-stocking,” symmetric distribution. Diabetic autonomic neuropathy can manifest in a variety of ways, including orthostatic dizziness, diarrhea, urinary incontinence, and gastroparesis.

Disease Highlights

1. Focal mononeuropathies

   1. Cranial (0.05% of mononeuropathies)

      1. Usually cranial nerve III or VI

      2. Usually acute and transient

      3. Caused by ischemia

   2. Thoracolumbar

   3. Limb

      1. Median nerve most common site (5.8% of mononeuropathies)

      2. Ulnar (2.1%), femoral, and peroneal also affected

2. Diabetic lumbosacral radiculoplexus neuropathy (also called diabetic amyotrophy): pain, severe asymmetric muscle weakness, and wasting of the iliopsoas and quadriceps muscles

3. Symmetric distal polyneuropathy (most common manifestation of DPN)

4. Epidemiology of symmetric distal polyneuropathy

   1. Prevalence ranges from 28% to 55%

      1. In 1 study, 55% of patients had DPN, but only about 14% were symptomatic

      2. Develops at a rate of 6.1 per 100 person years

   2. Severity is related to duration of disease, degree of glycemic control, and presence of hypertension and hyperlipidemia.

   3. DPN is an independent risk factor for foot ulceration and amputation; patients with neuropathy have a 15% lifetime risk of amputation.

5. Clinical manifestations of symmetric distal polyneuropathy

   1. History

      1. Frequently asymptomatic

      2. Burning, shooting, or lancinating pain

      3. Paresthesias, hyperesthesias

      4. Often worse at night

      5. When symptoms ascend to the knees, upper extremity symptoms start

   2. Physical exam

      1. Loss of vibration, pain, pressure, and temperature sensation

      2. Loss of ankle reflexes

      3. Distal muscle atrophy late in the course

   3. Charcot joints develop, usually in the tarsometatarsal region, in 10% of patients.

6. Differential diagnosis of symmetric distal polyneuropathy

   1. Cervical myelopathy, lumbar stenosis, tarsal tunnel syndrome, and digital neuropathies can all mimic DPN, and must be looked for on physical exam.

   2. Consider other causes of neuropathy if

      1. It develops before the onset of or early in the course of the diabetes

      2. Patient has a history of excellent glycemic control

      3. It is asymmetric

      4. Proximal or upper extremity involvement is disproportionate to distal lower extremity involvement

   3. Be sure to check for other causes of peripheral neuropathy (eg, hypothyroidism, vitamin B₁₂ deficiency, monoclonal gammopathy), even in patients with long-standing diabetes.

      Think about other causes of neuropathy in diabetic patients.

7. Diabetic autonomic neuropathy can affect any organ innervated by the autonomic nervous system.

   1. Cardiovascular autonomic neuropathy has many possible manifestations.

      1. Reduced heart rate variability; associated with increased risk of silent ischemia and cardiac death

      2. Fixed heart rate

      3. Resting sinus tachycardia

      4. Inadequate increase in heart rate/BP with exercise

      5. Postural hypotension with systolic BP drop of > 30 mm Hg, without an appropriate heart rate response

      6. Intraoperative cardiac instability

   2. Gustatory sweating

      1. Facial sweating, often accompanied by flushing, that occurs after eating

      2. Generally occurs in patients with nephropathy or peripheral neuropathy

      3. Cause unknown

   3. GI dysfunction

      1. Reduced esophageal motility

      2. Gastroparesis

         • (1) Abnormality of gastric motility leading to delayed gastric emptying

         • (2) Symptoms include nausea, vomiting, anorexia, postprandial fullness, early satiety.

         • (3) Poor correlation between demonstrated motility abnormalities and symptoms

      3. Diabetic diarrhea

         • (1) Characterized by intermittent, brown watery, voluminous stools, occasionally accompanied by tenesmus

         • (2) Can be episodic, separated by periods of normal bowel movements or constipation

         • (3) Rare in the absence of other manifestations of neuropathy, either peripheral or autonomic

      4. Constipation

         • (1) Constipation specifically resulting from autonomic neuropathy occurs in 20% of patients with type 2 DM.

         • (2) Caused by abnormality in autonomic neural control of colonic motility

      5. Anorectal dysfunction

         • (1) Results in fecal incontinence, even in the absence of diarrhea

         • (2) Patients can generally sense the presence of stool, but cannot prevent passage

   4. Genitourinary dysfunction

      1. Bladder dysfunction

         • (1) Initially motor function normal, but sensation of bladder distention impaired

         • (2) Then, detrusor muscle hypocontractility occurs, leading to urinary retention and overflow incontinence.

      2. Erectile dysfunction

         • (1) Present in 28–45% of diabetic men

         • (2) Most common organic cause of erectile dysfunction

         • (3) Risk factors include duration of DM, glycemic control, smoking, other diabetic complications.

Evidence-Based Diagnosis

1. Symmetric distal polyneuropathy

   1. Nerve conduction studies and electromyographic studies are the gold standard.

   2. A randomized trial showed that patients screened for DPN with a physical exam had lower amputation rates than those not screened

   3. The ADA recommends annual comprehensive foot exams including inspection, assessment of pedal pulses, and testing for loss of protective sensation with a 10-g monofilament plus any 1 of the following: vibration using a tuning fork, pinprick sensation, ankle reflexes.

   4. Although studies use variable physical exam techniques, the standards follow:

      1. Semmes-Weinstein monofilament examination

         • (1) Apply a 5.07/10-g monofilament to a non-callused site on the dorsum of the first toe just proximal to the nail bed.

         • (2) Repeat 4 times on both feet in an arrhythmic manner.

         • (3) Add up the total number of times the monofilament is perceived by the patient (score range = 0–8).

         • (4) Some studies test different sites with the monofilament: plantar surfaces of the first, third, and fifth toes; the first, third, and fifth plantar metatarsal heads, medial foot, lateral foot. One study showed that testing the third and fifth toes plus the first and third metatarsal heads identified 95% of the patients with abnormal results on 8-point testing.

      2. On–off vibration testing

         • (1) Apply a vibrating 128-Hz tuning fork to the bony prominence at the dorsum of the first toe just proximal to the nail bed.

         • (2) Repeat twice on each foot.

         • (3) Add up the total number of times the patient perceives the application of the vibrating tuning fork and the cessation of the vibration (score range = 0–8).

      3. Timed vibration testing

         • (1) Apply a vibrating 128-Hz tuning fork to the same location used for the on–off vibration test.

         • (2) Ask the patient to report the time at which vibration diminished beyond perception, and compare with the number of seconds perceived by the examiner when the tuning fork is applied to the examiner’s thumb.

         • (3) Record number of times patient’s perception time less than examiner’s (score range = 0–8).

      4. Superficial pain sensation

         • (1) Apply a sterile sharp to the same sites used for the monofilament.

         • (2) Repeat 4 times on each foot.

         • (3) Add up the total number of times the patient did not perceive the painful stimulus (score range = 0–8).

   5. Monofilament testing is more reproducible than timed vibration testing.

   6. Table 12-4 lists the test characteristics.

      Patients with any abnormal neurologic exam findings in the foot are likely to have DPN and are at high risk for developing ulcerations; those with normal exams may have early DPN but have a lower risk of developing ulcerations.

2. Diabetic autonomic neuropathy

   1. Cardiovascular autonomic neuropathy

      1. Cardiology consultation is necessary to evaluate heart rate variability.

      2. Postural change in systolic BP is used to diagnose orthostatic hypotension caused by diabetic autonomic neuropathy; the systolic BP is measured with the patient supine and again after 2 minutes of standing.

         • (1) A drop of < 10 mm Hg is normal.

         • (2) A drop of 10–29 mm Hg is borderline.

         • (3) A drop of > 30 mm Hg is definitely abnormal.

   2. Gustatory sweating is diagnosed by history.

   3. GI dysfunction

      1. Esophageal dysmotility: Esophagogastroduodenoscopy and manometry

      2. Gastroparesis: Diagnosed clinically or by a “gastric emptying” study, consisting of double-isotope scintigraphy of either solids or liquids.

      3. Diabetic diarrhea: Rule out other causes of chronic diarrhea.

      4. Anorectal dysfunction: Anorectal manometry and defecography can be done to document abnormalities.

   4. Genitourinary dysfunction

      1. Urinary bladder dysfunction: Ultrasound and urodynamic testing

      2. Erectile dysfunction: History

Treatment

1. Tight glycemic control

   1. Definitely prevents and improves neuropathy in persons with type 1 DM (relative risk reduction of 60%, NNT of 15 to prevent 1 case of neuropathy in tightly controlled patients)

   2. Possibly prevents and improves neuropathy in persons with type 2 DM

2. Otherwise, treatment is symptomatic.

   1. Peripheral neuropathy

      1. Tricyclic antidepressants, gabapentin, and pregabalin have all been shown to effectively reduce neuropathic pain.

      2. Tramadol and opioids are also effective.

      3. Capsaicin is possibly effective.

      4. Nonsteroidal antiinflammatory drugs generally are not effective.

         All patients with DM should receive foot care education. Those with DPN or structural foot abnormalities should be referred to a podiatrist, and screening for peripheral arterial disease with an ankle-brachial index should be considered.

   2. Autonomic neuropathy

      1. Cardiovascular

         • (1) Orthostatic hypotension is usually the most disabling symptom.

           • (a) Patients should raise head of bed, and rise slowly.

           • (b) Patients can try an elasticized garment that extends from the feet to the costal margins.

           • (c) Fludrocortisone is sometimes used but must beware of supine hypertension, and excessive salt and water retention.

         • (2) Cardioselective beta-blockers are sometimes helpful.

      2. Sweating: no specific treatment available; clonidine may be effective.

      3. Esophageal dysmotility: can try prokinetic agents such as metoclopramide.

      4. Gastroparesis

         • (1) Severe gastroparesis is very difficult to manage.

         • (2) Small meals sometimes help.

         • (3) Prokinetic agents, such as metoclopramide or erythromycin, sometimes are effective.

         • (4) Gastric electrical stimulation is being studied for refractory cases.

      5. Constipation

         • (1) Increase fiber.

         • (2) Drug choices include lactulose, polyethylene glycol, stool softeners.

         • (3) Avoid senna, cascara due to stimulant activity.

      6. Urinary bladder dysfunction

         • (1) Bethanecol

         • (2) Intermittent self-catheterization

      7. Erectile dysfunction: sildenafil and other similar agents

3. Nephropathy

Textbook Presentation

Diabetic nephropathy is asymptomatic until it is so advanced that the patient has symptoms of chronic kidney disease.

Disease Highlights

1. Occurs in 20–40% of patients with diabetes

2. The most common cause of end-stage renal disease (ESRD) in the United States

3. Definitions (based on spot collection and calculation of the albumin/creatinine ratio in mcg/mg)

   1. Normal < 30 mcg/mg

   2. Microalbuminuria = 30–299 mcg/mg

   3. Macroalbuminuria (overt nephropathy) ≥ 300 mcg/mg

4. Natural history: much better defined for type 1 than for type 2 DM

   1. Type 1 DM

      1. Renal enlargement and hyperfunction at onset of diabetes; continues for 5–15 years

      2. Microalbuminuria appears 10–15 years after onset of DM; glomerular filtration rate (GFR) and BP initially normal.

      3. Over the ensuing 10–15 years, 80% of patients progress to macroalbuminuria; GFR declines and hypertension develops.

      4. ESRD develops in 50% of patients with macroalbuminuria within 10 years and in 75% by 20 years.

   2. Type 2 DM

      1. Natural history is less well defined because onset of type 2 DM is usually not well defined, and other causes of kidney disease (such as hypertension and vascular disease) are common comorbidities.

      2. 20–40% of patients with microalbuminuria progress to macroalbuminuria.

      3. 20% have ESRD within 20 years of the onset of macroalbuminuria.

5. Risk factors for development of nephropathy

   1. Poor glycemic control

   2. Hypertension

   3. Long duration of DM

   4. Male sex

   5. Ethnic predisposition (Native American, African American, Hispanic [especially Mexican American])

6. Patients with any amount of albuminuria have an increased risk of cardiovascular events.

Evidence-Based Diagnosis

1. ADA recommends annual screening for microalbuminuria beginning at the time of diagnosis for patients with type 2 DM and starting at year 5 for patients with type 1 DM.

2. The recommended screening is a spot urinary albumin/creatinine ratio.

   1. There is diurnal variation, so first-void or early-morning specimens are best; otherwise, try to obtain confirmatory specimen at same time of day as initial specimen.

   2. Short-term hyperglycemia, exercise, urinary tract infection, marked hypertension, heart failure, and acute febrile illness can cause transient elevations in albumin excretion.

   3. Because of variability, 2–3 specimens in a 3- to 6-month period should be abnormal before diagnosing new or progressive diabetic nephropathy.

   4. For morning specimens, sensitivity ranges from 70% to 100% and specificity ranges from 91% to 98%.

   5. For random specimens, sensitivity ranges from 56% to 97% and specificity ranges from 81% to 92%.

3. It is not clear whether it is necessary to measure the albumin/creatinine ratio annually in patients being treated with an angiotensin-converting enzyme (ACE) inhibitor or angiotensin receptor blocker (ARB).

4. All patients should have a serum creatinine checked at least annually.

Treatment

1. Tight glycemic control reduces nephropathy.

   1. Type 1 DM: incidence of microalbuminuria is reduced by up to 43% and that of macroalbuminuria by 56%

   2. Type 2 DM

      1. Better control reduces the risk of microvascular complications (retinopathy and nephropathy) by 16–25%.

      2. NNT = 36 over 10 years in the UKPDS; NNT = 66 over 5 years in the ADVANCE trial

      3. The microvascular complication rate was 58% for patients with an HgbA_1C ≥ 10% and 6.1% for patients with an HgbA_1C < 6.0% (UKPDS).

      4. Microvascular complication rate decreases by 37% for every 1% reduction in HgbA_1C.

2. BP control and choice of agents

   1. BP should be < 140/80 mm Hg; aiming for a systolic BP < 130 mm Hg is appropriate if it can be achieved without adverse medication effects.

   2. Either ACE inhibitors or ARBs should be used.

      1. ACE inhibitors have been shown to reduce

         • (1) Progression to overt nephropathy in persons with type 1 and type 2 DM who have hypertension and microalbuminuria

         • (2) Progression to microalbuminuria in persons with type 2 DM who have hypertension and normoalbuminuria

         • (3) Cardiovascular events in patients with type 2 DM

      2. ARBs have been shown to reduce progression to overt nephropathy in persons with type 2 DM who have hypertension and albuminuria.

3. There are conflicting data regarding the efficacy of dietary protein restriction.

4. Diabetic Foot Ulcers

Textbook Presentation

A patient with peripheral neuropathy is unaware of minor trauma and the beginning of plantar ulceration. By the time the ulcer is discovered incidentally, it is often advanced, often with associated osteomyelitis.

Disease Highlights

1. Lifetime risk of developing an ulcer is about 15%.

2. Nearly all patients with ulcers have neuropathy, and over 50% have peripheral arterial disease, a strong predictor of nonhealing ulcers.

3. Tend to occur at pressure points, so plantar surface and sites of calluses are common locations

   1. Venous ulcers generally occur above the medial or lateral malleolus.

   2. Arterial ulcers generally occur on the toes, metatarsal heads, or shins.

4. Risk factors

   1. Previous amputation or foot ulcer

   2. Peripheral neuropathy

   3. Foot deformity

   4. Peripheral arterial disease

   5. Visual impairment

   6. Diabetic nephropathy (especially patients on dialysis)

   7. Poor glycemic control

   8. Smoking

5. Pathophysiology

   1. Repetitive mechanical stress occurs as a result of altered biomechanics, foot deformities, ill-fitting shoes.

   2. Peripheral neuropathy causes loss of protective sensation, so the patient is unaware of the incipient ulceration.

   3. Ischemia, resulting from macrovascular peripheral arterial disease (commonly in the tibioperoneal vessels) or microvascular dysfunction from autonomic neuropathy, inhibits healing and promotes progression.

6. Classification of diabetic foot infections

   1. Mild

      1. Local swelling or induration, erythema, local tenderness or pain, local warmth, purulent discharge

      2. Involves skin and subcutaneous tissue, with no involvement of deeper tissues and no systemic signs

      3. Erythema extends ≤ 2 cm from the ulcer

   2. Moderate: signs of local infection, plus

      1. Erythema extends more than 2 cm from the ulcer, or deep structure involved (abscess, osteomyelitis, septic arthritis, fasciitis)

      2. No systemic inflammatory response signs (SIRS)

   3. Severe: local infection plus at least 2 SIRS criteria (temperature > 38°C or < 36°C; pulse > 90 bpm; RR > 20 breaths/min or PaCO₂ < 32 mm Hg; WBC > 12,000 or < 4000 cells/mcL or ≥ 10% band forms)

7. Microbiology (see Table 12-5)

8. Osteomyelitis develops in up to 20% of patients with mild foot infections and 50–60% of those with moderate to severe infections.

Osteomyelitis can still develop in patients with mild foot infections.

Evidence-Based Diagnosis

1. Patients with neuropathy should have a foot exam at every visit.

   You cannot examine the feet of your diabetic patients too often, and you cannot examine them with their shoes on!

2. All patients with ulcers should have an ankle-brachial index to look for peripheral arterial disease.

3. Culturing ulcers

   1. Do not culture clinically uninfected ulcers.

   2. Do not obtain a specimen by swabbing the wound or wound drainage.

   3. Cleanse and debride the wound before obtaining a specimen.

      1. If purulent secretions are present, aspirate using a sterile needle or syringe.

      2. Obtain a tissue specimen by scraping the base of a debrided ulcer with a sterile scalpel or dermal curette.

4. Diagnosing complications

   1. Cellulitis: clinical diagnosis (see Chapter 17, Edema)

   2. Osteomyelitis (Table 12-6)

      1. Open bone biopsy with culture is the gold standard.

      2. Needle bone biopsy subject to sampling error (sensitivity, 87%; specificity, 93%; LR+, 12.4; LR–, 0.14)

      3. Being able to see bone or to probe the ulcer down to bone substantially increases the probability the patient has osteomyelitis.

      4. C-reactive protein, erythrocyte sedimentation rate, CBC, blood cultures not sufficiently sensitive or specific to diagnose osteomyelitis.

      5. MRI is the imaging procedure with the best test characteristics; bone scan and WBC scans are less specific but are sometimes done in patients who cannot undergo MRI.

MRI scan is the best imaging procedure to diagnose osteomyelitis in a patient with a diabetic foot ulcer.

A normal CBC, C-reactive protein, or erythrocyte sedimentation rate does not rule out osteomyelitis.

Treatment

1. Preventive foot care

   1. Improve glycemic control to reduce risk of neuropathy.

   2. Reduce vascular risk factors (smoking cessation, BP control, lipid management, glycemic control).

   3. Examine the feet of high-risk patients at every visit.

   4. Examine the feet of low-risk patients at least annually.

   5. Ensure patients wear well-fitted shoes.

   6. Educate patients regarding need for daily visual inspection of feet.

   7. Refer to podiatrist for débridement of calluses, assessment of bony deformities.

2. Treatment of ulcers

   1. Treat any infection (see Table 12-5).

   2. Determine need for revascularization, and revascularize as early as possible in patients with treatable peripheral vascular disease.

   3. Heal the ulcer.

      1. Off loading: use orthotics or fiberglass casts to remove pressure from the wound while allowing the patient to remain active.

      2. Débride ulcers (surgically or with débriding agents such as hydrogels).

      3. Control edema.

      4. Growth factors are being studied.

   4. Institute preventive measures once the ulcer has healed.

      A multidisciplinary approach, including internal medicine, vascular surgery, and podiatry is necessary for the optimal treatment of diabetic foot ulcers.

The evaluation for diabetic complications is complete. Mrs. D has no evidence of retinopathy, neuropathy, or diabetic foot disease. She does have microalbuminuria. However, before formulating a treatment plan for Mrs. D, it is necessary to assess for the presence or absence of other cardiovascular risk factors and cardiovascular disease:

1. Dyslipidemia

2. Hypertension

3. Obesity

4. Smoking

5. Coronary artery disease (CAD)

6. Cerebrovascular disease

7. Peripheral vascular disease

Table 12-7 outlines a summary of testing that must be performed on all patients with diabetes.

Treatment of Type 2 DM

The treatment of type 2 DM involves not only the treatment of the hyperglycemia but the management of associated complications and cardiovascular risk factors as well. According to survey data, only 51% of participants reach HgbA_1C goals, 51% reach BP goals, and 57% reach cholesterol goals; only 14.3% reach all 3 goals.

It is common for patients to require 6–7 medications to meet accepted treatment goals.

Treatment of Hyperglycemia

1. Treatment goals for patients with type 2 DM

   1. The ADA recommends treating to a HgbA_1C < 7.0% in most patients.

      1. HgbA_1C levels < 7% have been clearly shown to reduce microvascular events in patients with type 2 DM (see data above).

      2. Intensive control has not been consistently shown to reduce macrovascular events; intensive control may be harmful in older diabetics with cardiovascular disease, and may be beneficial in younger persons in whom diabetes was recently diagnosed.

         • (1) UKPDS (1998)

           • (a) About 4200 persons with newly diagnosed type 2 DM without cardiovascular disease, mean age 53, randomized to conventional therapy (HgbA_1C 7.9%) vs intensive therapy with sulfonylureas with or without insulin (HgbA_1C 7.0%) and monitored for 10 years

           • (b) Relative risk of myocardial infarction (MI) in the intensive group = 0.84 (95% CI, 0.71–1.00)

           • (c) A separate study arm randomized obese patients to receive metformin or conventional therapy; there was a significant reduction in MI with metformin (relative risk = 0.61, 0.41–0.89)

         • (2) UKPDS 10-year follow up (2008)

           • (a) About 3200 of the original 4200 patients were monitored for an additional 10 years after the intervention trial ended, with patients returning to their personal physicians for diabetes care.

           • (b) End points included death from any cause, diabetes-related death, MI, stroke, peripheral arterial disease, microvascular disease, any diabetes-related end point.

           • (c) The mean HgbA_1C was about 8% in all groups (conventional, sulfonylurea/insulin, metformin) 1 year after the intervention trial ended.

           • (d) Patients originally assigned to sulfonylurea/insulin had significant reductions in all end points measured (relative risk for MI 0.85, 95% CI 0.74–0.97; relative risk for death from any cause = 0.87, 0.79–0.96).

           • (e) Patients originally assigned to metformin had significant reductions in any diabetes-related end point, death from any cause (relative risk =0.73, 0.59–0.89), diabetes-related death, and MI (relative risk = 0.67, 0.51–0.89).

           • (f) These results suggest a “legacy” effect from initial intensive therapy in type 2 DM, similar to that seen in long-term follow-up of type 1 diabetics.

         • (3) ADVANCE trial (2008)

           • (a) About 11,000 type 2 diabetics with cardiovascular disease or multiple risk factors, mean age 66, randomized to intensive control with a sulfonylurea-based regimen (HgbA_1c 6.5%) vs conventional control (HgbA_1c 7.2%) and monitored for 5 years

           • (b) No difference in macrovascular events (relative risk = 0.94, 0.84–1.06)

         • (4) ACCORD trial (2008)

           • (a) About 10,000 type 2 diabetics with cardiovascular disease or multiple risk factors, mean age 62, randomized to intensive control (HgbA_1c 6.4%) vs conventional control (7.5%) and monitored for 3.5 years

           • (b) No difference in primary end point (nonfatal MI, nonfatal stroke, or cardiovascular death)

           • (c) Increase in any cause death with intensive treatment (relative risk = 1.22 (1.01–1.46), NNH = 100)

           • (d) Increase in cardiovascular death with intensive treatment (relative risk = 1.35 (1.04–1.76), NNH = 125)

           • (e) Most patients in the intensive group received rosiglitazone, which is associated with increased risk of MI.

   2. Goals should be modified for frail elderly, in whom avoidance of hypoglycemia and optimization of functional status may be more important than tight glycemic control.

   3. Goals should be individualized based on the overall health and age of the patient (Figure 12-1).

2. Monitoring

   1. HgbA_1c levels every 6 months in stable patients meeting goals; every 3 months in patients not meeting goals or undergoing changes in therapy

      1. Table 12-8 shows correlation between plasma glucose and HgbA_1C.

      2. 50% of HgbA_1c is determined by glycemia during the month before the measurement, 25% from the 30–60 days before, and 25% from 60–90 days before.

   2. Home glucose monitoring

      1. Patients taking multiple doses of daily insulin should test blood levels several times a day (fasting, before lunch, before dinner, and before bed); those taking bedtime long-acting insulin should test in the morning.

      2. Optimal frequency for patients taking oral agents is unclear; data regarding effects on control are mixed but show only slight improvement at best.

3. Lifestyle modification

   1. Weight loss (goal of at least 10% of body weight), diet modification, and exercise (goal of at least 150 minutes/week) are the foundations of all treatment for patients with type 2 DM.

   2. Best instituted in conjunction with a certified diabetes educator or dietician

4. Oral hypoglycemics

   1. Sulfonylureas

      1. Examples: glyburide, glipizide, glimepiride

      2. Increase insulin secretion.

      3. Average decrease in HgbA_1c about 1–1.5%

      4. Side effects include weight gain (2–5 kg) and hypoglycemia, especially in the elderly, patients with reduced renal function, and those with erratic eating habits.

      5. Shown to reduce microvascular outcomes; no change in cardiovascular events.

      6. Can be used as monotherapy or in combination with insulin or other oral agents (except non-sulfonylurea secretagogues)

      7. May become less effective with time, as beta cell function decreases

   2. Biguanides

      1. Example: metformin

      2. Reduce hepatic glucose production.

      3. Average decrease in HgbA_1c about 1–2%

      4. Associated with weight loss (or at least no weight gain); hypoglycemia rare

      5. Most common side effects are GI (abdominal pain, nausea, diarrhea).

      6. Associated with an increased risk of vitamin B₁₂ deficiency

      7. Because of risk of lactic acidosis, relatively contraindicated in patients with creatinine ≥ 1.5 mg/dL, decompensated heart failure, significant hepatic dysfunction, metabolic acidosis, and alcoholism.

         Metformin should be withheld in patients with acute illness and those undergoing surgery or procedures using radiocontrast.

      8. Has been shown to decrease microvascular and macrovascular outcomes, and total mortality in obese patients with type 2 DM (UKPDS, 1998)

      9. Can be used as monotherapy or in combination with all other oral agents and insulin

   3. Alpha-glucosidase inhibitors

      1. Example: acarbose

      2. Delay and decrease intestinal carbohydrate absorption, decreasing postprandial glucose swings

      3. About 50% less effective than sulfonylureas and metformin in reducing HgbA_1C (average reduction 0.5–0.9%)

      4. Side effects include flatulence, abdominal discomfort, and diarrhea.

      5. No studies of effects on macrovascular or microvascular outcomes

      6. Can be used as monotherapy, but this is rarely done because of relatively poor efficacy; can also be used in combination with sulfonylureas

   4. Thiazolidinediones

      1. Example: pioglitazone

      2. Increase insulin-stimulated glucose uptake by skeletal muscle cells and decrease hepatic glucose production

      3. Average decrease in HgbA_1c about 0.5–1.4%

      4. Tend to increase HDL and decrease triglycerides

      5. Can take weeks or months to obtain maximum effect

      6. Side effects include weight gain (as great as or more so than that seen with sulfonylureas) and edema.

      7. Increased risk of heart failure (relative risk ∼ 3)

      8. Current data do not show that pioglitazone increases the risk of cardiovascular events (as was shown with rosiglitazone).

         Do not use thiazolidinediones in patients with heart failure or edema.

         1. Can be used as monotherapy or in combination with sulfonylureas, metformin, and insulin.

   5. Non-sulfonylurea secretagogues (meglitinides)

      1. Examples: repaglinide, nateglinide

      2. Because of short half-life, cause brief, episodic increases in insulin secretion

      3. Primarily reduce postprandial glucose, with less risk of hypoglycemia than with sulfonylureas

      4. Average decrease in HgbA_1c 0.5–1%

      5. No long-term studies of effects on macrovascular or microvascular outcomes

      6. Must be dosed with every meal

      7. Should be used cautiously in patients with liver or kidney dysfunction

      8. Can be used as monotherapy or in combination with metformin

   6. Dipeptidyl peptidase 4 (DPP4) inhibitors

      1. Incretins (glucose-dependent insulinotropic polypeptide [GIP] and glucagonlike peptide 1 [GLP-1]) are intestinal peptides that augment insulin secretion in the presence of glucose or nutrients in the gut; they are inactivated by the enzyme DPP4.

      2. Sitagliptin and vildagliptin are selective DPP4 inhibitors.

      3. Decrease HgbA_1c by ∼0.75%

      4. No GI side effects; average weight gain < 1 kg

      5. Less effective than metformin in a direct comparison

      6. No data on macrovascular or microvascular outcomes

5. GLP-1 receptor analogues

   1. Example: exenatide

   2. Given subcutaneously twice daily

   3. Decrease HgbA_1c by 0.5–1.5%

   4. Most common side effects are nausea and vomiting.

   5. Average weight loss of ∼1.5 kg when compared with placebo and of ∼4.75 kg when compared with insulin

   6. No data on macrovascular or microvascular outcomes

6. Insulin

   1. Types of insulin (Table 12-9)

   2. Reduces HgbA_1c by 1–2.5%

   3. Adverse effects of insulin

      1. Hypoglycemia, especially with short-acting forms

      2. Weight gain of 2–4 kg

7. Choosing a medication to treat type 2 DM (Figure 12-2)

   1. Most studies compare an agent to placebo, so direct comparison data are limited.

   2. Sulfonylureas, metformin, and insulin have the best long-term outcome data.

   3. 75% of patients require more than 1 drug by 9 years; there is no evidence that any specific combination is better than another.

   4. Metformin is the preferred initial therapy in most patients with type 2 DM.

   5. Consider starting metformin plus a second agent in patients who present with an HgbA_1c > 9%.

   6. Patients who present with DKA, hyperosmolar hyperglycemic state (HHS), or an HgbA_1c > 10% should be given insulin as the initial therapy, although some patients may eventually be able to stop insulin and reach goal with oral agents.

   7. Using insulin to manage type 2 DM

      1. Beta cell function declines over time in type 2 DM, so many patients will eventually need insulin.

      2. The first step is to add long-acting basal insulin to oral agents, titrating the insulin dose to the fasting blood sugar.

         • (1) There are fewer nocturnal hypoglycemic episodes with bedtime glargine than with bedtime NPH.

         • (2) There is less weight gain with metformin and insulin than with sulfonylureas or thiazolidinediones and insulin.

      3. If the HgbA_1c target is not achieved, options include adding a short-acting insulin, such as lispro, with meals, or switching to twice daily biphasic insulin.

      4. Sulfonylureas should be stopped when short-acting insulins are used because of increased hypoglycemia.

Treatment of Hypertension

The treatment goal is to achieve a BP < 140/80 mm Hg. See Nephropathy section and Chapter 23, Hypertension for details.

Treatment of Hypercholesterolemia

1. American Diabetes Association guidelines

   1. Statin therapy should be used, regardless of baseline lipid levels, in diabetic patients with cardiovascular disease, and those without cardiovascular disease who are over age 40 and have at least 1 other cardiovascular risk factor (family history, hypertension, smoking, dyslipidemia, or albuminuria).

   2. Low-risk patients (those under age 40 without overt cardiovascular disease) do not require statin therapy if their LDL is < 100 mg/dL, although it should be considered in patients with multiple risk factors.

   3. The LDL goal is < 100 mg/dL for patients without overt cardiovascular disease.

   4. The LDL goal is < 70 mg/dL for patients with overt cardiovascular disease.

   5. Reduction of LDL by 40% is an alternative therapeutic goal in patients who cannot achieve targets on maximal tolerated doses of statins.

   6. Combination lipid-lowering therapy has not been shown to further lower cardiovascular event rates when compared to statin therapy alone and is not recommended by the American Diabetes Association guidelines.

      1. The ACCORD Lipid trial randomized 5518 patients with type 2 DM to simvastatin + fenofibrate vs. simvastatin + placebo; there were no differences in cardiovascular outcomes between the 2 groups.

      2. The subgroup with the lowest HDL (≤ 34 mg/dL) and highest triglycerides (≥ 204 mg/dL) benefited from combination therapy (relative risk = 0.72, NNT = 20); addition of fenofibrate could be considered in such patients.

2. See Chapter 23, Hypertension (Table 23-7) for new ACC/AHA guidelines for treating cholesterol in patients with diabetes.

Antiplatelet Therapy

1. Low-dose aspirin (75–162 mg/day) is indicated for secondary prevention in all patients with cardiovascular disease.

2. Low-dose aspirin is indicated for primary prevention in patients with diabetes and increased cardiovascular risk (Framingham 10-year risk > 10%; see Chapter 2, Screening and Health Maintenance for information on calculating risk).

   1. Patients with low cardiovascular risk (10-year risk < 5%) should not take aspirin for primary prevention; the risk of bleeding is higher than the potential benefit.

   2. Aspirin should be considered in patients with intermediate risk (10-year risk of 5–10%).

3. Clopidogrel, 75 mg/day, should be used in patients with aspirin allergies.

The differential diagnosis at this point is very broad and difficult to organize. It is helpful to recognize that Mr. G appears to be suffering from the syndrome of delirium and to use the framework for delirium to organize your thinking (see Chapter 11, Delirium and Dementia). It is also reasonable to consider Mr. G’s underlying chronic medical problems as important clinical clues and initially focus on the serious complications of these conditions; in other words, initially focus on diseases for which he has a high pretest probability:

1. Diabetes: DKA, HHS, infection with or without sepsis.

2. CAD: recurrent MI, possibly with HF or cardiogenic shock

3. Cerebrovascular disease: recurrent stroke

4. Chronic hepatitis B: hepatic encephalopathy

Table 12-10 lists the differential diagnosis.

Leading Hypothesis: Hyperosmolar Hyperglycemic State

Textbook Presentation

Patients who have HHS are usually older type 2 diabetics with the gradual onset of polydipsia, polyuria, and lethargy. They become extremely dehydrated, with reduction in urinary output, and have very high serum glucose levels, accompanied by alterations in mental status.

Disease Highlights

1. Epidemiology

   1. Risk factors include older age, nursing home residence, inability to recognize thirst, and lack of access to fluids.

   2. Mortality rate is 5–20%, compared to 1–5% in patients with DKA.

2. Pathogenesis

   1. Reduced effective insulin concentrations and a concomitant increase in counterregulatory hormones lead to increased hepatic and renal glucose production and impaired glucose utilization in peripheral tissues.

   2. Glycosuria leads to an osmotic diuresis with loss of free water in excess of electrolytes, leading to hyperosmolality.

   3. As volume depletion occurs, urinary output drops, and hyperglycemia worsens.

   4. Insulin levels are higher than in DKA and are adequate to prevent lipolysis and ketogenesis.

3. Precipitating factors

   1. The 3 most common precipitants are infection, lack of compliance with insulin, and first presentation of diabetes.

   2. Other precipitants include postoperative state, cerebrovascular accident, MI, pancreatitis, alcohol abuse, trauma, thyrotoxicosis, and medications (eg, corticosteroids, antipsychotic drugs, total parenteral nutrition).

4. Clinical manifestations

   1. History

      1. Symptoms and signs usually evolve over several days or even weeks.

      2. Common findings include polyuria followed by decreased urinary output, polydipsia, fatigue, and weight loss.

      3. Abdominal pain generally does not occur in HHS, as it does in DKA.

      4. Neurologic manifestations

         • (1) Lethargy and disorientation common

         • (2) Focal neurologic findings, including seizures, can occur with hyperglycemia and resolve with normalization of serum glucose.

         • (3) Changes in mental status correlate with the degree of hyperosmolarity.

           • (a) 20–25% present with coma.

           • (b) Coma is present in half of patients with effective serum osmolality of ≥ 320 mOsm/L.

           • (c) Must search for another cause of altered mental status if osmolality < 320mOsm/L

   2. Physical exam

      1. Hypothermia often seen resulting from peripheral vasodilation

      2. Signs of dehydration often seen (see Chapter 28, Acute Kidney Injury)

      3. Tachycardia and hypotension suggest severe dehydration or underlying sepsis.

Evidence-Based Diagnosis

1. Typical total body water deficit is 20–25% (about 9 L).

2. See Table 12-11 for laboratory findings in HHS compared with DKA.

Treatment

1. Patients with HHS generally need more fluid and less insulin than those with DKA.

2. Figure 12-3 outlines the treatment approach.

Mr. G fulfills the diagnostic criteria for HHS. It is not necessary to consider other diagnoses, but it is essential to determine the precipitant for this event. Considering Mr. G’s complicated history, he is at risk for many of the precipitants of HHS, especially infection, MI, and cerebrovascular accident.

Always look for the precipitant when patients present with either HHS or DKA.