Diabetic foot infections are common, costly, and potentially catastrophic for patients. The average patient with a diabetic foot infection undergoes three surgical procedures, including toe amputation in 19% and leg amputation in 14%.1 The annual cost of diabetic foot osteomyelitis in the United States is $2.8 billion.2
Diabetics are uniquely predisposed to foot infections. Because of neuropathy, minor repetitive injury leads to large foot ulcers, especially over the metatarsal heads. Foot deformities result in soft tissue injury from poorly fitting shoes. Because of autonomic neuropathy, the diabetic foot sweats less, leading to dry, cracked skin for bacteria to invade. Resistance to infection is lower because of neutrophil dysfunction in hyperglycemia. Vascular insufficiency impairs both wound healing and the immune response.3
The major principles of therapy are as follows:
1. Use an empiric antibiotic regimen with broad coverage against gram-positive, gram-negative, and anaerobic organisms. Diabetic foot infections are usually polymicrobial. Except for mild infections, in which therapy directed at gram-positive organisms may suffice, initial therapy should cover streptococci, methicillin-sensitive Staphylococcus aureus, E. coli, Proteus, and anaerobes.
Many studies of antibiotic therapy in diabetic foot infections have been performed, without demonstrating a clear superiority for any one regimen. The 2004 guidelines of the Infectious Diseases Society of America list 13 acceptable regimens for moderate diabetic foot infections.4 Useful single drug regimens include ampicillin-sulbactam, piperacillin-tazobactam, levofloxacin, and cefoxitin. For the penicillin-allergic patient, clindamycin and ciprofloxacin is a useful combination. Empiric vancomycin should be reserved for patients with a history of MRSA, treatment failure, or severe infection. (I would also reserve carbapenems, such as imipenem-cilastatin, for more severe infections, to prevent antibiotic resistance against a class of drugs representing our last defense against highly resistant gram-negative organisms.)
Once the patient is clinically improved and results of adequate cultures are available, consideration can be given to narrowing the course of therapy. Most diabetic foot infections can be treated with some combination of intravenous and oral therapy.
2. Have a high clinical suspicion for osteomyelitis. Osteomyelitis is extremely common in diabetic foot infections due to plantar ulceration and poor soft tissue coverage. Cure rates are reduced in osteomyelitis because dead bone acts as a nidus for persistent infection. The most useful diagnostic maneuver for osteomyelitis is deep probing of the wound with a sterile swab at the bedside. If a gritty sensation is felt, osteomyelitis is likely, and further diagnostic testing is probably unnecessary.
If the physical examination is equivocal, plain radiographs should be obtained to look for bony erosions. If these are negative, MRI is the next most useful diagnostic step. Bone scans are of limited value due to their low specificity. (Charcot arthropathy is a common cause of false-positive bone scans in this setting.)
Because of the high prevalence of osteomyelitis in diabetic foot infections, my own practice is to err on the side of longer, rather than shorter, antibiotic therapy.
3. Surgical debridement is required for many diabetic foot infections. Develop a collaborative relationship with a vascular or orthopedic surgeon with interest and expertise in the management of diabetic foot infections. Necrotic and gangrenous material should be removed. Ideally, dead bone should be debrided, both therapeutically and to help establish a bacteriologic diagnosis. (When removal of dead bone would result in loss of function or might create a non-healing wound, the option of long-term antibiotic suppression could be explored with an infectious disease specialist.)
4. When osteomyelitis is present, bone cultures help to define the optimal antibiotic therapy. Recent studies have confirmed older data regarding the poor correlation between surface cultures and bone cultures. The latter are preferred, when feasible.
5. Consider revascularization. Most diabetic foot infections arise in the setting of vascular insufficiency. At a minimum, patients with diabetic foot infections should have ankle-brachial indices performed for screening. Because diabetics may have falsely elevated ankle pressures due to calcified and non-compressible arteries, additional diagnostic studies may be useful, such as segmental pressures and Doppler pulse volume recordings.
Measurement of the transcutaneous oxygen concentration (TcP02) has been recommended, particularly in assessing which patients may benefit from hyperbaric oxygen. However, the TcP02 is not widely available, and the benefit of hyperbaric oxygen in this setting remains controversial.
6. Patients should be educated in meticulous foot care to prevent recurrences and reinfections. A diabetic foot infection may indicate that the patient lacks the knowledge, resources, or motivation for proper foot care. It also suggests that something is seriously awry with the patient’s diabetic regimen, compliance, or both. Hospital admissions for diabetic foot infections provide an opportunity to revise the patient’s diabetic medications; to educate the patient regarding wound care, skin care, and daily foot self-examination; to provide additional resources such as visiting nurses; and to refer patients for podiatric care, including tailored shoes, orthotics, and, if necessary, casting to off-load ulcers. TH
Dr. Ross is an instructor in medicine at Harvard Medical School.
- Hill SL, Holtzman GI, Buse R. The effects of peripheral vascular disease with osteomyelitis in the diabetic foot. Am J Surg. 1999 Apr;177(4):282-286.
- Gordois A, Scuffham P, Shearer A, et al. The health care costs of diabetic peripheral neuropathy in the United States. Diabetes Care. 2003;26(6):1790-1795.
- Lazzarini L, Mader JT, Calhoun JH. Diabetic foot infection. In: Calhoun JH, Mader JT, eds. Musculoskeletal Infections. New York, NY. Marcel Dekker. 2003.
- Lipsky BA, Berendt AR, Deery HG, et al. Diagnosis and treatment of diabetic foot infections. Clin Infect Dis. 2004 Oct;39(7):885-910.