Diabetic Foot Disease

Christopher E. Gross, MD

Jeannie Huh, MD

Dr. Gross or an immediate family member has received research or institutional support from Wright Medical Technology, Inc. Dr. Huh or an immediate family member serves as a board member, owner, officer, or committee member of the American Academy of Orthopaedic Surgeons and the American Orthopaedic Foot and Ankle Society.

This chapter is adapted from Thomas RL: Diabetic Foot Disease in Chou LB, ed: Orthopaedic Knowledge Update: Foot and Ankle 5. Rosemont, IL, American Academy of Orthopaedic Surgeons, 2014, pp 67-83.

ABSTRACT

As the global number of diabetes cases increases, so does its health consequences. Foot and ankle surgeons must recognize and be able to treat the myriad of often-devastating diabetic complications including infection, Charcot neuropathy, ulceration, and amputation. However, surgeons should not remain an island when it comes to patient care; rather, we must enlist the help of a team of healthcare providers to care for these complex patients.

Introduction

Diabetes mellitus first was recognized and described approximately 1,500 years ago, but the course of the disease dramatically changed with the discovery of insulin in 1922. Diabetes causes comorbidities affecting multiple systems. Foot morbidity is the most common reason for hospitalization of patients with diabetes in the United States and is responsible for substantial consumption of healthcare resources.

Diabetic foot disease is associated with the risk of ulceration, infection, foot deformity, amputation, vasculopathy, and neuroarthropathy. The success of orthopaedic care depends on the vascularity of the limb as well as patient cooperation in glycemic control and foot care. Care should be coordinated by a team that includes an orthopaedic surgeon, an endocrinologist, an infectious disease specialist, a vascular surgeon, a plastic surgeon, a physical therapist, and a pedorthotist.

Epidemiology

Diabetes is increasing at alarming rates worldwide with over half of people with diabetes unaware that they have it. Approximately 425 million of the world’s population is estimated to have diabetes, and the percentage is expected to increase by 48% to 629 million people by 2045.¹ The increasing incidence in the United States is a direct reflection of the increasing rate of obesity.² The number of individuals in the United States in whom diabetes was diagnosed is 30.3 million (9.4% of the population). In addition, an estimated 7.2 million individuals have diabetes, but the disease has not been diagnosed. More than 84.1 million people in the United States are considered to be prediabetic. Diabetes is most common among individuals older than 65 years; the aging of the population is correlated with the increasing prevalence of the disease. The direct medical cost of caring for patients with diabetes in the United States was $237 billion in 2017. This figure represents a 35% increase over the preceding 5 years and amounts to 12.5% of all expenditures for direct medical care in the United States.³

Complications associated with diabetic foot disease are the most common reason patients with diabetes are admitted to the hospital. In the United States, the annual incidence of foot ulcers in patients with diabetes is 6%, and the lifetime incidence has been reported to range between 19% and 34%. Patients with diabetes undergo almost 108,000 lower extremity amputations per year.⁴

Etiology

Diabetic neuropathy is the most important factor in the development of diabetic foot disease. As the severity of neuropathy increases, so does the risk of ulceration, amputation, and death. With increasing neuropathy, the patient’s functional ability, balance, and coordination decline. Neuropathy probably is the result of both metabolic and vascular factors. Sensation, motor control, and autonomic function are disturbed with diabetic
neuropathy, and these changes occur simultaneously. The risk of neuropathy in a patient with diabetes increases over time. Distal symmetric polyneuropathy is present in about 20% of newly diagnosed diabetics. It eventually affects 75% of patients with diabetes.⁵^,⁶

Neuropathy begins distally and progresses proximally in a stocking glove pattern. It can herald diabetes as those with impaired glucose tolerance can also be affected.⁶ Somatic sensory neuropathy affects the large-fiber nerves and is length dependent; it affects the longest nerves and is related to the patient’s height (a tall patient is at greatest risk). Disturbance of large sensory fibers is reflected in a decreased awareness of light touch and diminished proprioception with diminished or absent reflexes. Disturbance of small sensory fibers is reflected in a loss of pain and temperature perception. Almost one-third of patients with neuropathy report pain that is worse in the evening. The pain commonly is bilateral and symmetric, and it may be characterized in terms of burning, paresthesias, allodynia, electric shock, deep shooting pain, cramping, or aching. Over time, vibratory sensation is lost, and deep tendon reflexes disappear. It is commonly associated with development of retinopathy and nephropathy. Surgical decompression can be helpful if identifiable nerve compression is present, but only limited research supports this recommendation.⁷^,⁸ A recent meta-analysis has shown the efficacy of surgical decompression in relieving neurologic symptoms and restoring some of the sensory deficits in diabetic neuropathy.⁹ Pharmacologic treatment includes gabapentin, pregabalin, tricyclic antidepressants, venlafaxine, and duloxetine. Topical medication such as capsaicin cream or gel may also benefit patients.

Autonomic neuropathy occurs in 20% to 40% of patients with diabetes. It rarely occurs in isolation as it frequently is accompanied by sensory neuropathy.⁶ Symptoms are usually mild until the late stages of diabetes. The major clinical manifestations of autonomic neuropathy include resting tachycardia, exercise intolerance, orthostatic hypotension, constipation, gastroparesis, erectile dysfunction, cutaneous sudomotor dysfunction, impaired neurovascular function, so-called brittle diabetes, and hypoglycemic autonomic failure.¹⁰ Autonomic dysfunction affects control of sweat glands, blood vessel tone, and thermoregulation. The normal hyperemic response, which is necessary to fight infection, is lost. The skin becomes dry and scaly, cracks and fissures develop, and bacteria are able to invade, causing infection.

While standing, foot pressures can become very high. In the absence of sufficient blood flow to the underlying skin and tissue, cell oxygenation is inadequate and tissue dies. The result is ulceration beneath the points of greatest pressure. Ulceration can occur during less than 1 hour of constant standing. Repetitive mild trauma, as in walking, can cause preulcerative conditions that over time progress to complete ulceration. The triad of neuropathy, foot deformity, and repetitive trauma creates a high risk of ulcer formation.

Evaluation of the Diabetic Foot and Ankle

Obtaining a thorough history is crucial in a patient with diabetes. The identification of peripheral neuropathy and vascular disease, followed by appropriate intervention, can decrease the risk of subsequent infection and ulceration.

Examination of the foot and ankle should begin with an evaluation of the patient’s gait, posture, range of motion, muscle strength, and skin coverage. Limited joint mobility can increase plantar pressures, resulting in foot ulceration. Thin, shiny, atrophic, and hairless skin is indicative of diminished vascularity. Any corns, calluses, or ulcerations should be documented by size, location, margins, and depth. Areas with intradermal hemorrhage or blistering represent preulceration. It is also important to observe any exposure of a tendon and to determine whether an ulceration can be probed to bone. Thick toenails indicate vascular or fungal disease. Any deformities of the foot and ankle should be noted, and the presence or absence of protective sensation should be documented. The accepted threshold for normal sensation is the ability to perceive a Semmes-Weinstein size 5.07 nylon monofilament wire applied perpendicular to the skin. Of note, two studies found that vibration perception threshold testing has greater sensitivity for detecting impaired sensation.¹¹^,¹² However, this requires a biothesiometer, which is an instrument that is not readily available in most offices. One should be wary of flexion contractures of the toes, as they could increase distal pressure leading to ulceration and ultimately osteomyelitis. The fit and material of the patient’s shoes also should be evaluated. The interior of the shoes should be checked for foreign objects, and the insole removed to look for blood or other fluid discharge. Examination of the sole of the shoe may reveal an abnormal wear pattern related to deformity.

Oftentimes, an examiner must distinguish between an acute episode of Charcot neuroarthropathy and foot infection. The rubor in Charcot is more pronounced in a dependent position. A diagnostic strategy is for a physician to elevate the leg for 5 minutes. If the redness dissipates, then the diagnosis is likely an acute Charcot attack.

It is important for the examiner to determine the patient’s palpable tibialis posterior pulse, dorsalis pedis pulse, and capillary filling time. Any evidence of ischemia or a nonhealing wound warrants a vascular evaluation. Arterial Doppler ultrasonography is effective in assessing the adequacy of circulation. This test is reproducible and is not operator dependent. The results are reported in
terms of toe pressures and the ratio of ankle pressure to arm Doppler arterial pressure. An acceptable level for healing is toe pressures higher than 40 mm Hg or an ankle-brachial index higher than 45 mm Hg. Absolute toe pressures are a better predictor of healing than the ankle-brachial index.¹³ A triphasic waveform is present within normal vessels. When the vessel is calcified, the waveform is monophasic, and the reading can be falsely elevated. Transcutaneous oxygen tension values higher than 30 mm Hg indicate an acceptable potential for healing; this technique does not produce false readings with calcified vessels. If screening reveals limb ischemia, arteriography can identify the site of occlusion. This test is expensive, however, and has possible complications including allergic dye reaction, pseudoaneurysm, and acute renal failure in patients with compromised renal function or dehydration.

Vascular disease is 30 times more common in patients with diabetes than in other individuals.¹⁴ Heart disease continues to be the leading cause of death in patients with diabetes; approximately 74% of patients with diabetes have concurrent hypertension.¹ The risk of cerebrovascular accident is as much as four times greater in patients with diabetes than in the general population.¹ The typical lower extremity atherosclerotic findings in patients with diabetes include bilateral, diffuse, circumferential, with plaque formation in the medial layer of blood vessels. Atherosclerosis in the nondiabetic population usually is patchy, with plaque formation occurring in the intimal layer of blood vessels. Patients with diabetes are affected by atherosclerosis at a younger age, and the disease progression is more rapid. The iliac and femoral vessels often are affected. Involvement typically occurs at or just distal to the popliteal trifurcation involving the anterior tibialis, posterior tibialis, and peroneal arteries. Compromised blood flow to either lower extremity, combined with neuropathy, dramatically increases the risk for subsequent foot ulceration.

When evaluating vascular supply to the foot, no screening tool is completely accurate. When vascular compromise is of concern, a vascular consultation should be obtained to determine whether revascularization (transluminal angioplasty and vascular bypass) is an option.

Imaging

The imaging of the diabetic foot begins with plain weight-bearing radiographs of the foot and ankle, which are primarily used to evaluate major structural changes, joint alignment, soft-tissue gas, vascular calcifications, foreign bodies, and osteomyelitis. Focal demineralization, a reflection of underlying marrow changes, is the earliest radiographic change related to neuroarthropathy or osteomyelitis. In diabetic neuroarthropathy, localized osteopenia increases the risk of fracture with continued weight bearing. Excessive osteoclastic activity has been identified in Charcot-reactive bone, with cytokine mediators inciting bone resorption.¹⁵ The classic triad of osteomyelitis, which includes periosteal reaction, osteolysis, and bone destruction, may not be evident during the early stages of the disease (within the first 20 days).¹⁶

Charcot Neuroarthropathy and Infectious Radiographic Characteristics

CT scan is preferable to plain radiography for identifying osteomyelitic cortical erosions, but it has limited value in diagnosing early osteomyelitis. CT also is unable to distinguish between the changes of chronic infection and neuroarthropathy. CT is most useful for assessment of deformity and bone stock during surgical planning.

MRI has a high sensitivity for the detection of early soft-tissue and bone marrow edema associated with both Charcot neuroarthropathy and osteomyelitis; however, in isolation, MRI is unable to differentiate between these two pathologic processes.¹⁷^,¹⁸ The basis of diagnosis of osteomyelitis by MRI includes abnormal bone marrow signals typified by confluent hypointensity on T1-weighted images, which reflect infiltration of the infectious process.¹⁹ Gadolinium contrast injection can improve the ability to detect the abscess or necrosis associated with osteomyelitis. MRI can be used to differentiate a sterile joint effusion from septic arthritis.¹⁹^,²⁰ Secondary findings such as direct spread from an ulcer or the presence of a sinus tract can contribute to the diagnosis. Because of its multiplanar capability and high spatial and contrast resolution, MRI is considered the best modality when the diagnosis of infection has already been confirmed for defining the extent of soft-tissue and bone marrow involvement for surgical planning.¹⁸^,¹⁹^,²⁰^,²¹^,²² The role of MRI in determining disease resolution and guiding management is also important and remains under investigation.²³

In isolation, triple-phase bone scanning using technetium Tc-99m (⁹⁹Tc) phosphonates has 94% sensitivity and 95% specificity for the diagnosis of osteomyelitis, in the absence of another abnormality.¹⁸^,²⁴ However, specificity is reduced to 33% in the presence of neuroarthropathy, trauma, recent surgery, or malignancy. ⁹⁹Tc bone scanning is advantageous because its high sensitivity has a high negative predictive value; therefore, a negative bone scan essentially rules out infection. Indium-111 white blood cell (WBC) scintigraphy is even more accurate than ⁹⁹Tc bone scanning for diagnosing osteomyelitis, and a negative result strongly supports the absence of infection. If trying to differentiate between Charcot neuroarthropathy and osteomyelitis, the best diagnostic test is a
combination of nuclear medicine scans (ie, ⁹⁹Tc bone scan and indium-111-labeled WBC scan).¹⁷^,²³^,²⁵ Positron emission tomography (PET)-CT with 18-fluorodeoxyglucose, which indicates an increase in glucose metabolism, also has high accuracy and specificity for the differentiation of osteomyelitis from neuroarthropathy and is superior to leukocyte-labeled studies for the diagnosis of chronic osteomyelitis.²⁶^,²⁷

Ultrasonography has a limited role in diabetic imaging, although it can help identify a foreign body or with identifying and aspirating an abscess. If there is a need to differentiate between infected and uninfected fluid surrounding a tendon sheath, ultrasonography can reveal the abnormal internal echogenicity characteristic of an infected fluid collection.¹⁸

Laboratory Studies and Vital Signs

If cellulitis or a deeper infection is suspected, appropriate laboratory testing should be done. An elevated WBC count indicates infection, but the WBC count can be normal even if an infection is present, because of an impaired immune response.²⁸ Leukocytosis higher than 11 × 10⁹/L is associated with a 2.6-fold increase in the risk of amputation, and fever higher than 100.5°F (38°C) is associated with a 1.3-fold increase in the risk of amputation.²⁹ Interestingly, more than 50% of patients with acute osteomyelitis of the foot had a normal WBC count, and 82% had a normal oral body temperature.³⁰ An evaluation of 400 patients with moderate or severe diabetic foot infection found a mean WBC count of 8.24 × 10⁹/L; those who were unsuccessfully treated had a mean WBC count of 9.98 × 10⁹/L, and those who favorably responded to treatment had a mean WBC count of 7.93 × 10⁹/L.³¹

A total lymphocyte count lower than 1.5 × 10⁹/L is correlated with immunocompetence in wound healing. Albumin levels above 30 g/dL support a nutritional status satisfactory for healing; healing can occur but is less predictable with lower levels.³² Patients requiring transtibial amputation have substantially lower serum albumin levels than patients who underwent successful limb salvage.³³

An elevated erythrocyte sedimentation rate often is correlated with inflammation or infection; usually, a level less than 40 is more common with cellulitis or a local soft-tissue infection. A level higher than 67 (mm/hr) suggests underlying osteomyelitis. The C-reactive protein (CRP) level is highly sensitive but not specific for inflammation. When the CRP >14 mg/L, the sensitivity and specificity are 0.85 and 0.83.²⁸ CRP level is a better parameter than WBC count or neutrophil count for diagnosis and monitoring of treatment in deep diabetic foot infections.³⁴ The erythrocyte sedimentation rate can remain elevated 3 months after initiation of infection treatment, although it still can be used to monitor efficacy of treatment.²⁸ Worsening glycemic control may be one of the earliest signs of diabetic foot infection.

Complications and Treatment

Diabetic Foot Ulcer

Foot ulcerations are the most frequently recognized lower extremity complication of diabetes, accounting for significant morbidity, mortality, and healthcare expenditures. According to 2015 prevalence data from the International Diabetes Federation (IDF),³⁵ foot ulcers develop annually in 9.1 million to 26.1 million people with diabetes worldwide. Between 19% and 34% of patients with diabetes are likely to be affected with a foot ulcer in their lifetimes.¹ Diabetic foot ulcers are the leading cause of lower extremity amputation in these patients,³⁴ and more than half of all diabetic foot ulcers become infected.³⁶ The risk of death at 5 years is over twice as high in diabetic patients with foot ulcers compared with nonulcerated diabetic patients.³⁷ In conjunction with its heavy morbidity and mortality burden, the cost of care for those with foot ulcers is five times higher than those without ulcers.³⁵ Management of diabetic foot ulcers accounts for approximately one-third the total cost of diabetic care, which in the United States was estimated to be $348 billion in direct healthcare expenditures in 2017.³⁵ The causes of diabetic foot ulceration are multifactorial and include autonomic, sensory, and motor neuropathy; foot deformity; decreased vascularity; loss of skin integrity; poor glucose regulation; impaired leukocyte function; and repetitive trauma.

Autonomic neuropathy leads to decreased sweating with resultant dry, cracked skin, vulnerable to open wounds. Sensory neuropathy leads to loss of protective sensation. Motor neuropathy can result in clawing and hammering of the toes secondary to loss of intrinsic muscle balance. Fixed toe deformities often result in pressure over the proximal interphalangeal joints and beneath the metatarsal heads, which are common sites for ulcer development. Progressive Achilles tendon contracture further increases forefoot pressures. In a comparison of 32 patients with diabetes and 32 healthy control subjects, diabetics had a thicker Achilles tendon and stiffer plantar soft tissue than the control subjects.³⁸ As much as a fivefold increase in the mechanical stiffness of the plantar soft tissues has been reported in patients with diabetic foot disease.³⁹ This results in increased tissue hydrostatic pressure in the loaded foot, effectively reducing the blood supply to the areas of highest pressure.³⁸

Risk factors for development of a diabetic ulcer include loss of sensation, diminished pulses, foot deformity,
abnormal reflexes, advanced age, history of a foot ulcer, and an abnormal neuropathy disability score.

The most common reasons a diabetic ulcer does not heal include decreased vascular supply, deep infection, and failure to mechanically unload the affected area. Wound size, wound duration, and wound grade are directly associated with the likelihood of wound healing by the 20th week of care.⁴⁰ The Brodsky classification system, which is based on ulcer depth and ischemia, helps determine the need for hospitalization and possible surgery⁴¹ (Figure 1). The Scottish Foot Ulcer Risk Score, which classifies an ulcer as mild, moderate, or high risk based on palpable pulses, sensory neuropathy, foot deformity, and a history of ulcer or amputation, can be used to predict ulcer development and ulcer healing.⁴² A prospective analysis of 1,000 patients was used to develop a diabetic ulcer severity score in which the examiner assigns one point for each of four parameters: inability to palpate a pedal pulse, ability to probe to bone, presence of an ulcer anywhere other than a toe, and multiple ulcerations.⁴³ A relatively high score was associated with an increased risk of amputation, and an increase of one point in a patient’s score predicted a 35% decrease in the likelihood of healing. A higher score was correlated with a larger initial wound area, a longer wound history, and a greater likelihood that surgery or hospitalization will be required.

FIGURE 1 Schematic drawings showing a classification system for diabetic foot lesions based on ulcer depth (A) and ischemia (B). (Adapted with permission from Brodsky JW: The diabetic foot, in Coughlin MJ, Mann RA, Saltzman CL, eds: Surgery of the Foot and Ankle, ed 8. Philadelphia, PA, Mosby, 2007, vol 2, pp 1281-1368.)

Ulcer Management Strategies

The principles of diabetic foot ulcer treatment are local wound care with sharp débridement and wound off-loading. First, however, treatment of any associated active infection and vascular insufficiency, if present, is necessary. The first step in ulcer management is to remove all necrotic tissue and surrounding callus. This step converts a chronic wound into an acute wound and encourages healing with granulation tissue formation and reepithelialization. Débridement also contributes to infection control, because devitalized tissue may provide a nidus for bacterial proliferation. Elliptical wounds usually heal better than circular wounds. The wound should be débrided weekly or more frequently.

The next step is to off-load plantar shear stress and plantar pressure from the ulcer by distributing both over a larger surface area, thus reducing mechanical stress and promoting an environment conducive to healing. Although off-loading can be accomplished by avoiding weight
bearing on the involved extremity, patient compliance is problematic. Total-contact casting (TCC) has historically been considered the benchmark method of pressure off-loading for diabetic foot ulcers based on shorter healing times and better healing rates compared with a half shoe or removable cast walker.⁴⁴ Despite its effectiveness in ulcer healing, the application of a TCC requires technical skill, and the reported complication rates are as high as 17%.¹³ Alternatively, there has been growing evidence that other nonremovable knee-high device options, in the form of commercially available walking devices rendered irremovable by overwrapping with plaster (i-RWDs), can achieve similar results as TCC.⁴⁵^,⁴⁶^,⁴⁷ More recently, some studies have also reported positive outcomes with removable walker devices (RWDs) compared with TCC and i-RWD.⁴⁷ The removable option is attractive in patients who are unable to tolerate a nonremovable device or in situations where they are contraindicated, such as significant peripheral artery disease and infection.

Surgery may be used as a method of off-loading. Lengthening of the gastrocnemius or Achilles tendon in the setting of an equinus contracture can decrease plantar pressure and allow forefoot ulcers to heal.⁴⁸^,⁴⁹ However, overcorrection can lead to an increase in peak pressure under the heel, a calcaneal gait, and a subsequent heel ulcer that eventually may require partial calcanectomy or transtibial amputation.⁴⁸ Plantar fascia release has been shown to result in healing of forefoot ulcers.⁵⁰

For ulcers that fail to demonstrate improvement by >50% wound area reduction after a minimum of 4 weeks of standard wound therapy, the IDF recommends adjunctive wound therapy options.³⁵ Multiple adjuvant therapies are available and being studied for the care of diabetic foot ulcers. These include nonsurgical débridement agents, dressings and topical products, oxygen therapies, negative-pressure wound therapy (NPWT), acellular bioproducts, human growth factors, skin grafts and bioengineered skin, energy-based therapies, and systemic therapies. Although systematic reviews of various wound dressings have found no evidence that any specific type is better than others, for foot ulcers with heavy exudate, the IDF recommends a dressing that absorbs moisture, while dry wounds need topical treatments that add moisture. Hyperbaric oxygen therapy (HBOT) is among the adjuvant therapies for diabetic foot ulcers that has been available and studied for multiple decades. A 2015 Cochrane review showed a significant increase in healing rate with HBOT at 6 weeks, but no difference at 1 year follow-up, and no statistically significant difference in major amputation rate.⁵¹ In 2016, a double-blinded randomized controlled trial concluded that 12 weeks of HBOT does not reduce indication for amputations in patients with Wagner grade 2 to 4 diabetic foot ulcers.⁵² NPWT is another well-studied adjuvant therapy. A systematic review of 11 randomized controlled trials comparing NPWT with standard dressing changes showed that NPWT had a higher rate of healing, shorter healing time, and fewer amputations, with no difference in incidence of treatment-related adverse effects.⁵³

Only gold members can continue reading. Log In or Register to continue