Richard E. Goldstein,     New York, New York

Michael R. Lappin,     Fort Collins, Colorado

Differential Diagnosis

Dogs with septic joints generally are systemically ill and have swollen, hot, red, and painful joints. Synovial fluid cytologic examination reveals increased numbers of nondegenerative neutrophils, degenerative neutrophils, and macrophages; bacteria, fungi, or protozoans may or may not be observed. In dogs, bacterial arthritis generally involves only one joint, and often there is a portal of entry such as a laceration or bite wound.

In most dogs with nonseptic, suppurative polyarthritis the clinical signs include intermittent fever, general malaise, anorexia, joint pain, and stiffness. However, some affected dogs can have normal findings on joint palpation. It is important to remember that the infectious causes of nonerosive, nonseptic, suppurative polyarthritis often are clinically indistinguishable from primary immune causes.

A number of vector-borne or non–vector-borne infectious agents causing polyarthritis are relatively common in the United States and include Anaplasma phagocytophilum (dogs or cats), Bartonella spp. (dogs), Borrelia burgdorferi (dogs), Ehrlichia canis (dogs or cats), Mycoplasma spp. (dogs or cats), and Rickettsia rickettsii (dogs). Since these organisms generally are not seen in the joints, the joint fluid usually is classified cytologically as nonseptic. Secondary immune-mediated, suppurative, nonerosive polyarthritis commonly occurs due to infectious diseases that do not directly involve the joints but cause inflammation from type III hypersensitivity. Examples are brucellosis, ehrlichiosis, bacterial endocarditis, chronic otitis externa, pyometra, actinomycosis, coccidioidomycosis, and leishmaniasis. Nonerosive, nonseptic, suppurative, polyarthritis also is believed to occur as a manifestation of vaccine reactions, although this is not well documented in dogs. It occurs most frequently with modified live calicivirus vaccination of kittens.

There may be other historical data and clinical signs associated with infectious causes of nonerosive, nonseptic, suppurative polyarthritis that vary with the infectious agent. For example, R. rickettsii replicates in endothelial cells, which leads to severe inflammation in multiple tissues, and so most clinically affected dogs are extremely ill. R. rickettsii infection (Rocky Mountain spotted fever) has only an acute stage; this finding, combined with the fact that tick vectors are not active in the winter, make this a seasonal syndrome (spring through fall). Most cases of R. rickettsii infection are reported in the midwestern and southeastern United States. In contrast, Ehrlichia ewingii is not seasonal, clinical signs other than polyarthritis are minimal, and infection occurs mostly in dogs living in states harboring the Lone Star tick, Amblyomma americanum. Another classic finding is that polyarthritis associated with A. phagocytophilum or B. burgdorferi infection occurs almost exclusively in dogs that have visited areas endemic for the deer tick, Ixodes spp. Finally, A. phagocytophilum, E. ewingii, and R. rickettsia primarily are associated only with acute disease, which leaves Bartonella spp., B. burgdorferi, E. canis, and Mycoplasma spp. as the most likely infectious agent differentials if polyarthritis has been present long term.

Diagnostic Plan

When polyarthritis is suspected, the initial diagnostic plan generally includes radiography of the affected joints to determine whether osteoarthritis or erosive changes are present and a complete blood cell count, serum biochemical testing, and urinalysis to evaluate for possible causes or sites of inflammation. Arthrocentesis usually is performed to determine whether the joint fluid is septic or nonseptic and whether the predominant cells are mononuclear cells, degenerative neutrophils, or nondegenerative neutrophils.

If bacteria are seen, synovial fluid should be cultured for aerobes, anaerobes, and Mycoplasma spp. In all dogs in which the predominant cells are neutrophils, urine also should be cultured regardless of whether the neutrophils are degenerative or nondegenerative, even if bacteria are not seen. Transport media and specific culture media are required for successful culture of Mycoplasma spp. Thoracic and abdominal radiography or ultrasonography should be considered for dogs with nonerosive, nonseptic, suppurative polyarthritis to evaluate for occult infectious or neoplastic antigen sources. Urine and blood cultures should be considered. Protein : creatinine ratio should be determined if proteinuria without hemorrhage or inflammation is detected on initial urinalysis. Echocardiography can be performed to evaluate heart valves for changes associated with bacterial endocarditis. In some dogs with valvular endocarditis murmurs may not be auscultated.

The decision to perform tests for primary immune diseases depends on the individual case. It has been shown that positive results on tests for antinuclear antibodies and rheumatoid factors do not correlate with the presence of primary immune-mediated diseases (Smith et al, 2004). However, because most dogs with SLE are antinuclear antibody positive, a negative test result may rule out this syndrome as a cause of the polyarthritis.

The signalment, risk of vector exposure, vector disease control history, travel history, and clinical findings should be used to rank the differential list. The clinician then can select tests for individual infectious disease agents to complete the diagnostic workup. Several laboratories now offer panels of infectious agent serologic tests or polymerase chain reaction (PCR) assays that amplify the DNA of infectious disease agents. However, the results of these assays can be positive in healthy dogs as well as in dogs with polyarthritis, so positive test results do not always correlate with illness. See the later sections on each individual agent for a discussion of currently available diagnostic tests.

Treatment

Animals with septic polyarthritis should be treated with antibiotics that kill aerobes, anaerobes, and gram-positive and gram-negative organisms until culture results return. The combination of a quinolone with a penicillin or a cephalosporin is a logical first choice. Quinolones often are effective against gram-negative bacteria as well as Bartonella spp., L-form bacteria, Mycoplasma, and R. rickettsii. Penicillin derivatives are effective against gram-positive bacteria, anaerobic bacteria, and B. burgdorferi.

Animals with nonseptic, suppurative polyarthritis often are administered doxycycline (5 to 10 mg/kg q12-24h PO) while the workup is completed. Doxycycline generally is effective for management of infections with A. phagocytophilum, B. burgdorferi, Ehrlichia spp., Mycoplasma spp., and R. rickettsii. However, Bartonella spp. infections of dogs rarely respond to doxycycline alone, and so this agent should stay on the differential list for dogs with polyarthritis in which a doxycycline trial fails (see the section on Bartonella). An additional benefit of doxycycline is that it has antiinflammatory properties and may lessen discomfort in some affected dogs nonspecifically. If doxycycline is unavailable or not tolerated, alternative options do exist such as minocycline (at the same dose as doxycycline), penicillins, or cephalosporins for B. burgdorferi; minocycline or imidocarb for E. canis; and so forth. Because the inflammation usually is neutrophilic, low-dose prednisone administered at 1 mg/kg q12-24h PO initially may give dramatic improvement in clinical signs of disease. However, excessive doses can activate some infectious agents such as B. burgdorferi. If the syndrome turns out to be a primary immune-mediated disorder, increased doses of prednisone with or without other immunosuppressive drugs may be required.