In comparison with lower respiratory tract, urinary tract, and skin and soft tissue infections, infections of bones and joints are relatively uncommon in the elderly, accounting for no more than a few percent of the infectious diseases treated by geriatricians. However, some bone and joint infections disproportionately afflict the elderly, such as osteomyelitis contiguous to pressure ulcers, septic arthritis of joints damaged by rheumatoid arthritis, or periprosthetic hip joint septic arthritis. As is true of many infectious illnesses in the elderly, diagnosis of bone and joint infections may be complicated by subtle, masked, or atypical clinical presentations. The consequences of delay in diagnosis and treatment of these infections, moreover, are no less serious in the elderly than in younger patients, and in the frail elderly inadequate diagnosis and treatment may increase mortality and cause great morbidity. Although definitive antibiotic treatment is similar to that of bone and joint infections in younger patients, empiric therapy in most cases must be broader.
At any age, infections of the periosteum, medullary cavity, and cortex of bones are usually caused by pyogenic bacteria, and less commonly by mycobacteria or fungi; infections by other types of organisms are extremely rare. Osteomyelitis is classified primarily according to route and duration of infection, and secondarily by anatomic location and etiologic agent. Such a classification is useful in ensuring a careful evaluation of the patient, as well as determining antibiotic and surgical treatment choices. The terms acute and chronic, while clinically useful, are not sharply demarcated temporally, and are often used loosely as descriptors (1). In general, acute osteomyelitis evolves over weeks, and chronic osteomyelitis evolves over many months or years, and may be associated with dead bone (sequestra), fistulous tracts, or foreign bodies.
Three types of osteomyelitis are usually distinguished: contiguous focus infection, hematogenous osteomyelitis, and that which is secondary to vascular insufficiency. Whereas most cases of osteomyelitis in elderly patients treated by geriatricians are the result of contiguous spread from an infected wound or cellulitis, hematogenous seeding from a distant infection, postoperative complications resulting from orthopaedic surgery, or penetrating wounds may also be the underlying causes. In all circumstances,
From: Infectious Disease in the Aging Edited by: Thomas T. Yoshikawa and Dean C. Norman © Humana Press Inc., Totowa, NJ
however, healthy bone is highly resistant to infection, so trauma, ischemia, or foreign bodies are usually necessary preconditions.
Once bacteria are introduced into or near injured or ischemic bone, the phagocytic reaction that is mobilized to contain the infection releases enzymes that may lyse bone. The inflammatory response also impairs blood flow and may cause bone necrosis, which may eventually lead to the development of a devascularized sequestra of bone. Once bone infection is well established, the bacteria protect themselves from host defenses with the production of a polysaccharide-rich biofilm—especially when a foreign body is involved—making it more difficult to eradicate infection. As a result, most treatment regimens for osteomyelitis are longer than those for acute infections such as pneumonia or urinary tract infection.
Osteomyelitis in elderly patients most often develops by direct, contiguous extension of infection from infected adjacent soft tissues. Clinically, contiguous focus osteomyelitis may be indolent and its presence obscured by the skin and soft-tissue infection. The usual underlying infection in elderly patients is infected pressure ulcers (decubitus ulcers). Whereas penetrating injuries, compound comminuted fractures, and postoperative complications of surgical procedures are more typical of the type of contiguous focus osteomyelitis encountered in younger patients, they may also be responsible for contiguous focus osteomyelitis in elderly patients. When pressure ulcers or smoldering cellulitis fail to respond to wound care and antibiotic therapy, underlying contiguous focus osteomyelitis must be suspected (2).
In the usual case of contiguous focus osteomyelitis in an elderly patient, poor tissue perfusion and repeated trauma to soft tissue produce conditions that are conducive to polymicrobial and anaerobic infection. Thus, it should be assumed that contiguous focus bone infection in elderly patients is polymicrobial in etiology, although Staphy-lococcus aureus (the most prevalent cause of bone infection in younger patients) is still incriminated as a pathogen or copathogen in the majority of cases. Consequently, a mixture of staphylococci, streptococci, enteric gram-negative organisms, and anaerobic bacteria may be found in bone biopsy cultures when contiguous focus osteomyelitis is due to a diabetic foot infection or underlying pressure ulcer. Similarly, osteomyelitis from soft tissue infection of the oropharynx, paranasal sinuses, gastrointestinal tract, or female genital tract is usually due to an unpredictable mixture of anaerobic and Gram-positive and Gram-negative bacteria. When osteomyelitis results from a postoperative wound infection, the cause is most often S. aureus, although coagulase-negative staphylococci are common when osteomyelitis follows implantation of orthopedic appliances. Pseudomonas aeruginosa osteomyelitis may be the result of puncture-wound infections of the foot, especially when the foot wound is soaked in water, which may introduce pseudomonas organisms. Antibiotic-resistant organisms, such as Enterobacter aerogenes, Acinetobacter calcoaceticus, or P. aeruginosa, may be selected out when polymicrobially infected pressure ulcers are treated with narrow-spectrum antibiotics.
184.108.40.206. Hematogenous (Vertebral) Osteomyelitis
Acute hematogenous osteomyelitis primarily occurs as an infection of intravenous drug abusers and young children. The only type of hematogenous infection ordinarily encountered in older adults is the relatively indolent vertebral osteomyelitis. In vertebral osteomyelitis, bacteria reach the vertebral end plate and disk space via the spinal arteries and spread to the adjacent vertebral body. The urinary tract is often the source of the infection, and in older men bacteria are believed to reach the spine via the prostatic venous (Batson's) plexus. The source of bacteremia in cases that are unrelated to urinary tract infection is usually obvious—less-obvious primary sources of vertebral body infection include subacute bacterial endocarditis, remote soft tissue infections, or an infected intravenous line or other medical device.
Spine pain with no apparent cause or a fever of unknown origin are typical presentations for vertebral osteomyelitis; atypical presentations, such as referred pain to chest or abdomen, may result from nerve root irritation. The lumbar spine is involved most often, followed by the thoracic and cervical spine. In the case of now-rare Pott's disease, tuberculous spondylitis most often affects the thoracic vertebrae. The majority of cases of vertebral osteomyelitis have a subacute course with vague pain that gradually intensifies over several months and low-grade fever. Fever may be absent, however, and the white blood cell count is usually not elevated. There may be spasm of the paraspinal muscles and limitation of spine motion. In patients suspected of having an occult vertebral osteomyelitis, diagnosis may be reliably suggested by the simple physical examination finding of localized percussion tenderness over the involved vertebrae. Posterior extension may cause epidural and subdural abscesses or meningitis, and anterior or lateral extension may lead to paravertebral, retropharyngeal, mediastinal, subphrenic, or retroperitoneal abscesses.
Because of the protracted course typical of vertebral osteomyelitis, before it is clinically recognized the erythrocyte sedimentation rate (ESR) is elevated and plain X-rays show irregular erosions in the end plates of adjacent vertebral bodies and narrowing of the disk space—a radiographic pattern that is virtually diagnostic of bacterial infection because tumors and other diseases of the spine rarely cross the disk space. Computed tomographic (CT) or magnetic resonance imaging (MRI) scans are still prudent for confirmation of diagnosis because they may demonstrate paraspinal or epidural abscesses. When the cause of vertebral pain, radicular pain, and muscle weakness is unrecognized as coming from an epidural abscess, the clinical course may be one of rapid progression over several weeks to irreversible paralysis.
When nonvertebral hematogenous osteomyelitis is diagnosed in older adults, it is often a relapse of an infection that occurred many years or even decades after the initial episode of infection. In these cases of nonvertebral hematogenous chronic osteomyelitis, recurrent exacerbations typically punctuate long periods of quiescence. Patients are usually afebrile, and a relapse may be manifest only by increased purulence of drainage from sinus tracts between bone and skin, or an increase in the ESR. Rare late complications of chronic hematogenous osteomyelitis include pathologic fractures, squamous cell carcinoma of the sinus tract, and amyloidosis.
The microbiology of vertebral osteomyelitis is different from contiguous focus osteomyelitis in that more than 95% are caused by a single organism. The majority of all cases of hematogenous osteomyelitis are caused by S. aureus, although in vertebral osteomyelitis Escherichia coli or other enteric bacilli are often the cause. The literature is replete with cases that have unusual causes of hematogenous osteomyelitis, including fungi—such as those resulting from disseminated histoplasmosis, coccidioidomy-
cosis, blastomycosis, cryptococcosis, and Candida or aspergillus tissue infections. Atypical mycobacteria, syphilis, yaws, or even viruses such as varicella and vaccinia may also infect bone. The great variety of causes makes culture verification of etiology, either by bone biopsy (always preferred) or blood culture, imperative. Sinus or urine cultures alone are never sufficient to determine antibiotic therapy. 220.127.116.11. Osteomyelitis Secondary to Vascular Insufficiency
When osteomyelitis occurs in the setting of peripheral vascular disease or diabetic neuropathy, it nearly always involves the small bones of the feet. Poor vascular supply to the infected bone and soft tissue impairs normal healing and defenses against infection; in the case of diabetic or other types of peripheral neuropathy, there may be delayed recognition of osteomyelitis because of lack of normal sensation. Microbio-logically there is little difference in etiologies between contiguous focus osteomyelitis and that caused by vascular insufficiency: a wide range of Gram-positive cocci, Gram-negative bacilli, and anaerobic pathogens must be considered. Cultures of skin ulcers are unreliable for identification of pathogens, and cultures of bone biopsies are mandatory if precise bacteriologic data are needed (3,4).
As with other infections in the elderly, the clinical manifestations of osteomyelitis may be blunted by muted fever or reduced inflammatory response, and the diagnosis can be unrecognized or delayed by difficulties in communication with patients who frequently have neurological diseases. Moreover, because the debilitated elderly may be less mobile, the physical movements or weight bearing that elicit musculoskeletal pain in younger patients may not be present or pain may occur only when they are passively moved by caregivers. As a result, an indolent bone infection may become well-established, and the diagnosis of osteomyelitis only becomes apparent weeks or months later. The initial indication that osteomyelitis is present may thus be the occurrence of complications such as sinus tract formation or disseminated infection. The diagnosis of vertebral osteomyelitis may be the surprising conclusion of a search for the cause of a low-grade fever that had been incorrectly attributed to cellulitis, urinary tract infection, or aspiration and has failed to resolve with treatment. Although the diagnosis of osteomyelitis may be suspected clinically, the definitive diagnosis must be made on the basis of radiological studies or examination of pathological specimens.
Prompt diagnosis and institution of antibiotic therapy of osteomyelitis may prevent bone necrosis or serious septic complications. An elevated ESR or C-reactive protein (CRP) level is present in most cases of active osteomyelitis, including in patients who do not have constitutional symptoms or an elevated white blood cell count. Although these are nonspecific laboratory tests, and especially in early infections the ESR may occasionally be normal, the diagnosis of osteomyelitis should be called into doubt when the ESR and CRP are normal.
The targeted evaluation of suspected osteomyelitis usually begins with plain radiographs, although they frequently show no abnormalities during early infection because, on average, 1-2 wk are required for infection to cause necrosis of bone. Plain radio graphs may initially show nonspecific soft-tissue swelling; radiographic findings of periosteal reaction, if they occur, are not visible for at least 10 d after the onset of infection. Definitive indications of possible infection, such as lytic changes that do not appear until more than half of bone has been lost, sequestra, or sclerotic new bone formation, may require many weeks to appear on plain radiographs. A CT scan is more sensitive than plain films for the detection of sequestra, sinus tracts, and soft-tissue abscesses. Both CT and ultrasonography are useful for guiding percutaneous aspiration of subperiosteal and soft-tissue fluid collections.
Bone scans are very sensitive indicators of osteomyelitis. In nearly all cases, the technetium radionuclide scan (99mTc diphosphonate) is positive within 24 h of symptomatic infection. Consequently, a negative radioisotope bone scan virtually excludes osteomyelitis. Bone scans are, however, not at all specific for infection. Uptake of technetium merely indicates osteoblastic activity and increased vascularity, so techne-tium bone scanning cannot differentiate osteomyelitis from fracture, tumor, adjacent soft-tissue infection, or infarction. Although less convenient and more expensive than technetium scans, an abnormal gallium or indium scan (with 67Ga citrate, or 111In-labeled leukocyte or immunoglobulin isotopes) are evolving diagnostic techniques that may be helpful in distinguishing infectious from noninfectious processes found by plain radiographs and technetium scanning, and may prove especially useful in patients with diabetic ulcers (5).
At present, however, the definitive radiographic study for delineating the presence and extent of infection is the MRI scan, which is as sensitive as radioisotope bone scans for the diagnosis of osteomyelitis but is far more specific for infection. It yields better anatomic resolution of epidural abscesses and other soft-tissue processes, and it is preferable to CT scan for delineation of anatomical changes. An MRI scan gives detailed information about the activity and the anatomic extent of infection but cannot always distinguish osteomyelitis from healing fractures and tumors. The MRI scan is particularly useful in distinguishing cellulitis from osteomyelitis in the diabetic foot.
"Probing to bone" is a relatively specific procedure for verifying the presence of osteomyelitis, and may be very useful in the evaluation of nursing facility patients with pressure ulcers. If bone is palpable during examination of the base of an ulcer with a blunt surgical probe, osteomyelitis is highly likely, and further diagnostic testing may be avoidable.
Blood cultures are positive in more than 25% of cases of vertebral osteomyelitis. Cultures of sinus tract drainage or the base of an ulcer correlate poorly with the organisms infecting the bone (4,6). The gold standard is aerobic and anaerobic culture of a biopsy specimen obtained under direct vision during surgery. Multiple aerobic and anaerobic cultures by percutaneous needle biopsy under ultrasound or radiographic guidance or biopsy at the time of debridement are mandatory before antibiotic therapy for chronic osteomyelitis is initiated. Cultures positive for multiple organisms and organisms of low virulence, such as coagulase-negative staphylococci, should not be assumed to be contaminants, especially if foreign bodies are present. Special culture media for less common pathogens, such as mycobacteria or fungi, may be required in some cases, and ultimately histopathologic examination of biopsy specimens may be the only way to make a diagnosis.
Antibiotic treatment should not be initiated until appropriate cultures have been obtained. The antibiotics selected should be bactericidal and in most cases, at least initially, should be given intravenously. When possible, empiric therapy may be initiated by Gram staining of a specimen from the bone or abscess. If Gram stains are unavailable, after cultures are obtained, prompt broad-coverage empiric therapy against the most likely pathogens is prudent. Empiric therapy should always include high doses of an antistaphylococcal antibiotic, such as nafcillin, a cephalosporin, or vancomycin; if Gram-negative organisms are likely to be involved, as is usually the case for vertebral osteomyelitis, a third-generation cephalosporin such as ceftazidime or a newer fluoroquinolone such as levofloxacin, may be used. If, methicillin-resistant staphylococci and P. aeruginosa are possible infecting organisms, a combination of vancomy-cin and an antipseudomonal antibiotic, such as a parenteral fluoroquinolone or an antipseudomonal third-generation cephalosporin, may be needed (7,8). Success in animal models of staphylococcal osteomyelitis using rifampin have led some to use the drug in combination with usual antistaphylococcal therapy (9). Typical initial antibiotic regimens for elderly patients based on the presumed infecting organism are listed in Table 1.
Doses must be modified for the reduced lean-body weight and reduced renal function of elderly patients, especially those residing in long-term care facilities. Amino-glycosides should be avoided in the elderly because of their unpredictable ototoxicity and nephrotoxicity, both of which are more frequent and, when they occur, more debilitating in the elderly. If they must be used, serum levels of aminoglycosides should be monitored closely and renal function checked frequently to minimize toxicity.
Duration of intravenous and oral therapy for osteomyelitis is 4-6 wk. When well-absorbed antibiotics such as fluoroquinolones are employed, the switch to oral therapy may be made after defervescence to reduce the complications and expense of immobilization and parenteral therapy. When more than a week of parenteral antibiotic therapy is required, use of agents that require infrequent dosing, such as ceftriaxone, vancomycin, or once-daily newer fluoroquinolones, is preferred (10,11).
When oral therapy is prescribed, the bioavailability and gastrointestinal tolerance of antibiotics must be considered carefully. For example, most oral penicillins or cephalosporins are incompletely absorbed, absorption may be reduced by food (or, paradoxically, by fasting) and the doses required for the treatment of osteomyelitis are several times higher than the doses of these drugs prescribed for other common infections, which may not be tolerated by older adults. Except for the fluoroquinolones, such as ciprofloxacin (500-750 mg twice daily) or levofloxacin (500 mg once daily) there are limited clinical data to support the use of oral antibiotics by adults for the treatment of chronic osteomyelitis. Caution should be exercised in the use of fluoroquinolones as the sole agents for treatment of infection caused by S. aureus or P. aeruginosa because resistance may develop during therapy. Clindamycin (300-450 mg every 6 h) or metronidazole (500 mg every 6-8 h) are also highly bioavailable and may be useful for oral regimens.
Intravenous Antibiotic Regimens for the Initial Treatment of Bone and Joint Infections in Elderly Patients, Based on Presumed or Known Pathogens"
• Penicillin-sensitive staphylococci and streptococci:
Penicillin G (3-4 million units every 4-6 h)
• Penicillin-resistant, methicillin-sensitive staphylococci
Nafcillin or oxacillin (2 g every 4-6 h), or cefazolin (1-2 g every 8-12 h)
• Methicillin-resistant staphylococci
• Gram-negative bacilli (based on in vitro susceptibility)
Ampicillin (2 g every 4-6 h), cefazolin, cefuroxime (1.5 g every 8-12 h), or a fluoroquinolone such as levofloxacin (500 mg every 24 h)
• Resistant Gram-negative bacilli, such as P. aeruginosa:
Two antibiotics known to be active against the organism, such as piperacillin (3 g every 6-8 h), or ceftazidime (1 to 2 g every 8-12 h), plus gentamicin or tobramycin (1.7 mg/kg every 8 h, or 5-7 mg/kg every 24 h) or a fluoroquinolone, such as ciprofloxacin (750 mg every 12 h) or levofloxacin (500 mg every 24 h), or aztreonam (1-2 g every 8-12 h) have been recommended by some clinicians aUsually continued for 4-6 wk, although shorter courses may be appropriate for some infections. If response to therapy is rapid, a shorter course of parenteral therapy followed by high doses of oral antibiotics known to be active against the infecting bacteria for the remainder of the course of therapy may be effective.
The ESR may be useful to monitor treatment of chronic osteomyelitis, and in particular, vertebral osteomyelitis. Failure of the ESR to drop to about half of pretreatment levels after a 4-6 wk course of treatment with an appropriate antibiotic is an indication for longer treatment.
Surgery is seldom necessary for noncontiguous-focus osteomyelitis, even in cases of many months' duration, except in instances of spinal instability, new or progressive neurologic deficits, large soft-tissue abscesses that cannot be drained percutaneously, or a failure of medical treatment. Immobilization, which is always hazardous for elderly patients, should be prescribed only for relief of symptoms—when back pain has declined to the point at which ambulation is possible, bed rest should be terminated. Spontaneous fusion of involved vertebrae occurs in the majority of cases after successful treatment.
Even when diagnosed early, however, cure of contiguous-focus osteomyelitis usually requires at least surgical debridement in addition to 4-6 wk of antibiotic therapy (12). Risks and benefits of aggressive therapy for chronic osteomyelitis in the elderly, however, should be considered carefully. For nursing home patients, for example, intermittent courses of suppressive therapy with oral antibiotics are an attractive option when compared with enduring multiple surgical procedures, prolonged courses of parenteral antimicrobial therapy, and in some cases the risk of loss of an extremity. If surgical intervention is chosen, biopsy and culture performed preoperatively permits the administration of targeted antibiotic therapy before surgery to reduce active infec tion and inflammation. If the specific etiology is not known, antibiotics should be given intraoperatively—after debridement with culture obtained for definitive therapy. Management of "dead space" created by debridement surgery is mandatory—the goal of dead-space management is to replace dead bone and scar tissue with durable vascularized tissue, as scar tissue that fills the defect may later become avascular. Complete wound closure should be attempted.
The effectiveness of treatment of osteomyelitis of the small bones of the feet related to vascular insufficiency is obviously is limited by the poor blood supply and impaired ability to heal. When appropriate, revascularization should be considered. In the case of decreased perfusion due to diabetic small-vessel disease, the only likely options are suppressive antibiotic therapy or amputation. If infected bone is entirely surgically removed, routine preoperative surgical prophylaxis is sufficient. Although employed at some medical centers, the routine use of hyperbaric oxygen to enhance the killing of microorganisms by phagocytes remains controversial.
Prevention of osteomyelitis depends primarily on controlling the predisposing con-ditions—bacteremia from acute infection elsewhere, pressure ulcers, and the complications of diabetic neuropathy and vascular disease. Prevention of the latter conditions rests mostly with good skin care, rather than the use of antibiotic therapy once ulcers have developed. The role of antibiotics in the treatment on pressure or trophic ulcers has, moreover, always been controversial, and is, at best, only marginally beneficial unless acute cellulitis is present.
Was this article helpful?