Ahmed, Salman M.D.; Ayoub, Elia M. M.D.
“The patient with arthritis who does not respond [to aspirin] within one to three days does not have acute rheumatic fever.”
Denny FW. In: T. Duckett Jones and Rheumatic Fever 1
Chronic arthritis in children usually results from rheumatologic disease, most often juvenile rheumatoid arthritis (JRA). Other causes include other collagen vascular disease, malignancy, metabolic disease, septic arthritis and occasionally acute rheumatic fever (ARF). Arthritis is the most common major manifestation of ARF, occurring in ~70% of patients with this disease. The arthritis is acute, migratory, and exquisitely responsive to salicylates. It is self-limited in duration, rarely lasting more than a few days.
In 1959 Crea and Mortimer described 21 children who developed bacterial culture-sterile arthritis during the course of scarlet fever and who failed to meet the Jones criteria for ARF. 2 In an effort to distinguish these patients from children with ARF, they coined the term “scarlatinal arthritis.” Almost two decades later, Goldsmith and Long 3 reported 12 children who presented with arthritis after group A streptococcal infection but failed to satisfy the Jones criteria for ARF. The arthritis was described as prolonged, protracted and poorly responsive to salicylates and other nonsteroidal antiinflammatory drugs (NSAIDs). Because of the similarity to the reactive arthritis associated with enteric pathogens, Goldsmith and Long designated this form of arthritis “poststreptococcal reactive arthritis” (PSRA).
During the past 20 years, additional patients with PSRA have been described, expanding understanding of this disease. 4–10 PSRA has an acute onset and typically develops 3–14 days after streptococcal pharyngitis. Fever and a scarlatiniform rash are often present during the acute phase of pharyngitis but are usually absent by the time arthritis develops. The arthritis is nonmigratory and can affect both large and small joints, most often the knees, ankles, wrists and proximal interphalangeal joints. Involvement of the axial skeleton occurs in ~20% of patients. 10 Morning stiffness is present in some cases. Unlike the arthritis of ARF, PSRA responds poorly to salicylates and NSAIDs, lasting from 1 week to 8 months, with a mean duration of ~2 months. 10 The course of arthritis is unaffected by antimicrobial therapy or prophylaxis. In ~5–6% of patients with PSRA, carditis has been reported to have developed at a later time. 10 Mitral valve disease is most common and is often diagnosed only by echocardiography. Table 1 compares the clinical characteristics of PSRA, ARF and JRA.
| Table 1. Clinical characteristics of PSRA, ARF and JRA** Adapted from Ayoub EM and Majeed HA. 11NA, not applicable. |
The pathogenesis of PSRA is poorly understood. Antibodies that develop against the group A streptococcus and cross-react with joint synovial tissue or cartilage are believed to be fundamental, similar to the situation with ARF. Recent evidence indicates that in Caucasians PSRA is associated with the class II HLA antigen DRB1*01. 10 In contrast ARF is associated with HLA-DR*16, 10 whereas reactive arthritis due to enteric pathogens (Salmonella, Shigella, Yersinia, Campylobacter) is associated with HLA-B27, a class I antigen. 11 This suggests that PSRA is more closely related to ARF than to other forms of reactive arthritis.
In patients with PSRA, the peripheral leukocyte count is generally within normal limits, and the erythrocyte sedimentation rate is elevated. In addition rheumatoid factor is negative, and the frequency of antinuclear antibodies and HLA-B27 is similar to that in the general population. 12
To confirm the diagnosis of PSRA, evidence of antecedent group A streptococcal infection is necessary. Some patients may present with history of recent pharyngitis and a positive throat culture or rapid antigen test for group A streptococci. If this is lacking, a throat culture or rapid antigen test should be obtained, especially in patients who present early in their course of arthritis. The most reliable method to establish antecedent group A streptococcal infection is to measure anti-streptolysin O and anti-DNase B titers, 13 which minimizes the confounding contribution of group A streptococcal carriage detected by throat swab. Because of the short interval between the inciting infection and the development of arthritis, patients who are seen early in their course may not manifest an antibody response initially, but they should do so on a repeat serum sample obtained 3–4 weeks later.
Because of the potential for “silent carditis,” some (but not all) experts recommend an echocardiogram at the time of initial presentation and at follow-up 1 year later. 9
Patients with PSRA can be treated with NSAIDs such as ibuprofen or naproxen to ameliorate the features of arthritis. NSAIDs should be continued for the duration of arthritis and then tapered as tolerated.
Based on recommendations by the American Heart Association, 14 patients with PSRA should receive antimicrobial prophylaxis to prevent recurrent streptococcal infection. Penicillin is preferred as first line therapy, and erythromycin is appropriate for penicillin-allergic patients. The current recommendation is to continue prophylaxis for 1 year and to follow for evidence of carditis. If no carditis is observed during the year, prophylaxis can be discontinued. Patients who develop carditis should be classified as having ARF and should continue to receive antimicrobial prophylaxis. Because the timing of carditis is poorly defined, an alternative proposal is to maintain prophylaxis until the age of 21 years and for a minimum of 5 years, analogous to the prophylaxis regimen recommended by the American Heart Association for patients with ARF who have arthritis without carditis. 10, 14 This alternative recommendation awaits approval by the American Heart Association.
1. Denny FW. Circulation 1987; 76: 963–70. Ovid Full Text Bibliographic Links [Context Link]
2. Crea MA, Mortimer EA Jr . Pediatrics 1959; 23: 879–84. Bibliographic Links [Context Link]
3. Goldsmith DP, Long SS. Arthritis Rheum S18:1982;25:S18. [Context Link]
4. De Cunto CL, et al. Pediatr Infect Dis J 1988; 7: 683–6. [Context Link]
5. Gibbas DL, Broussard DA. Arthritis Rheum 1986; 29: S92. Bibliographic Links [Context Link]
6. Birdi N, et al. J Pediatr Orthop 1995; 15: 661–5. [Context Link]
7. Moon RY, et al. J Rheumatol 1995; 22: 529–32. [Context Link]
8. Gutierrez-Urena S, et al. J Rheumatol 1995; 22: 1710–13. Bibliographic Links [Context Link]
9. Schaffer FM, et al. Pediatrics 1994; 93: 837–9. [Context Link]
10. Ahmed S, et al. Arthritis Rheum 1998; 41: 1096–102. [Context Link]
11. Ayoub EM, Majeed HA. Curr Opin Rheumatol 2000; 12: 306–10. Ovid Full Text Bibliographic Links [Context Link]
12. Ayoub EM, Ahmed S. Curr Probl Pediatr 1997; 27: 90–101. [Context Link]
13. Ayoub EM, Harden E. In: Rose NR, et al., eds. Manual of clinical laboratory immunology. Washington, DC: American Society for Microbiology, 1997: 450–7. [Context Link]
14. Dajani A, Taubert K, Ferrieri P, et al. Pediatrics 1995; 96: 758–64. [Context Link]
