Recent concepts in the management of septic arthritis
Septic arthritis is a rheumatologic emergency condition. Delayed in diagnosis and treatment can result in joint damage and deformity, the presence of osteomyelitis or even death. It is most often a consequence of systemic bacteremia. Penetrating trauma, intra-articular procedures (arthrocenthesis, intraarticular injection or arthroscopic procedures) can cause septic arthritis in minority of the cases.
The prevalence and incidence of septic arthritis depend on the criteria for the diagnosis and the type of population studied. However, the incidence of septic arthritis in tertiary care medical center is increasing. This is due to increasing age of the population (with co-morbid medical conditions), use of corticosteroid and immunosuppressive drugs, and the increase in incidence of human immunodeficiency virus (HIV) infection. Pre-existing joint condition, particularly rheumatoid arthritis, is the most common risk factor for septic arthritis.
Rheumatoid arthritis patients receiving anti-TNF therapy have been found to have increase risk of septic arthritis. Other risk factors include conditions that cause loss of skin integrity, abuse of intravenous injections, and conditions that associate with impair immunity. Gram-positive organisms (Staphylococcus aureus and Streptococcus spp.) are responsible for most cases of septic arthritis in several series. S. aureus is the most common offending organism in intravenous drug users. In contrast, gram-negative organisms are responsible for only approximately 10-20%. However, the causative organisms usually depend on the epidemiology of the organisms in that area, and the hosts’ risk factors. Neisseria gonorrhoeae, once was the most common cause of septic arthritis, is now uncommon. In Thailand, a high incidence of gram-negative septic arthritis has been reported from the university hospitals. Salmonella spp. are the most common offending organisms particularly in patients receiving corticosteroids or those with acquired immunodeficiency syndrome (AIDS). Burkholderia pseudomelli septic arthritis is also common in the north eastern part of Thailand.
The diagnosis of septic arthritis is not difficult when it presents with a classic symptoms of fever and a warm and painful swollen joint. However, in those who are receiving nonsteroidal anti-inflammatory drugs or corticosteroids, such as patients with rheumatoid arthritis or systemic lupus erythematosus, the sign of acute joint inflammation may be blunted, which delays making diagnosis and cause greater complications as a consequence. Only 60% of the cases have high grade fever, and leukocytosis is present in only 50-60% of cases. Leukocytosis might be absent in immunosuppressed patients. Acute monoarthritis of the large joint, especially the knee joint, is the most common presentation. Polyarticular septic arthritis can be seen in 10-20% of cases. Rarely, septic arthritis of the sternoclavicular joint, sacroiliac joint and symphysis pubis can be seen, particularly in intravenous drugs users.
Synovial fluid (SF) cell counts of over 50,000 cells/mm3 is usually suggestive of the possibility of septic arthritis, but other arthritic disease such as gout, pseudogout, reactive arthritis and even neuropathic joint could have high SF cell counts that mimics septic arthritis. SF cell counts of less than 50,000 cells/mm3 can also be seen in patients with septic arthritis, particularly in those who are receiving immunosuppressive drugs or those with AIDS. Concomittant septic arthritis and crystal arthritis can be seen, although it is uncommon. Gram staining of the SF is helpful in the diagnosis, but it was insensitive as it was positive in only 40-70% of cases. Although a positive SF cultures is the gold standard in the diagnosis of septic arthritis, many cases of septic arthritis do not have positive SF cultures. Cultures of SF in blood culture bottles or BECTEC media can improve positive cultures.
Radionuclide scans (Tc-99m and 111Indium) can help to differentiate septic from non-septic arthritis in deep seated joint. The combination of these 2 scans gives a very high sensitivity and specificity in the diagnosis of septic arthritis. However, these tests are generally required in those whom SF cannot be obtained. Use of molecular biology technique (such as polymerase chain reaction and specific nucleic acid probe) can improve in the identification of bacterial antigen in the joint, but these cannot confirm whether the organism is viable. SF glucose and protein are not helpful in the diagnosis of septic arthritis.
Treatment of septic arthritis requires both appropriate antibiotic and drainage. Antibiotic should be started promptly when the diagnosis of septic arthritis is suspected, until the culture data becomes available. If gram stains of SF is negative, cefazolin is the drug of choice to cover Staphylococcus spp. Streptococcus spp. and gram negative bacteria. If a patient is at risk for gram negative septic arthritis (such as elderly or immunocompromised patients), third generation cephalosporin should be used. Vancomycin should be used in patients who are at risk for methicillin-resistant S. aureus septic arthritis (such as intravenous drug users, patient on hemodialysis, diabetes mellitus and those with recent or current hospitalization). Combination of vancomycin and third generation cephalosporin should be used in critically ill patients with septic arthritis. The duration of antibiotic therapy is 4-6 weeks depending on clinical responses. For intracellular organisms such as Salmonella spp. and B. pseudomallei, the duration of antibiotic therapy should be 12 weeks.
Joint drainage can be performed by repeated arthrocentesis, arthroscopy or arthrotomy. Easily assessable joint can be drained by repeated arthrocentesis. Serial SF analysis is useful in assessing response to treatment. Those with SF cell counts of less than 20,000 cells/mm3 at the end of one week of the therapy usually have complete recovery.
Surgical drainage is indicated when the repeated arthrocentesis and antibiotic fail to control infection after 5 days. There is no good data showing the superiority of surgical drainage over arthrocetesis, although studies showed that those with arthrocentesis had better functional outcome Morbidity and mortality in patients with septic joint varies depending on the age, co-morbidity, underlying disease and immune status of the patients. The mortality is high in elderly patient (age over 65 years), diabetes mellitus, clinical sepsis at presentation, polyarticular disease, rheumatoid arthritis and HIV infection. There is no different in outcome among those with positive and negative SF cultures.
In conclusion, management of septic arthritis requires prompt diagnosis, initiation of appropriate antibiotic and drainage. Selection of antibiotic depends on prevalence and antibiotic susceptibility in each area, and risk factors and immune status of the hosts.
(Worawit Louthrenoo, M.D. Professor of Medicine, Division of Rheumatology Department of Internal Medicine, Faculty of Medicine Chiang Mai University, Chiang Mai, Thailand.)