Differentiating a case of transient synovitis from septic arthritis of the hip in children can be difficult because the presentations of these two entities are similar. Fever, a limp, unwillingness to bear weight, limited motion and joint effusion can occur with both. Kocher and associates conducted a retrospective study of presenting signs in children with septic arthritis and transient synovitis of the hip.

The medical records of 282 children with signs of hip inflammation were analyzed. Thirty-eight of the 282 children were considered to have had true septic arthritis as identified by a positive culture of the joint fluid or at least 50,000 white blood cells per mm³ (50 × 10⁹ per L) in the joint fluid, accompanied by positive findings on blood culture. The diagnosis of presumed septic arthritis was given to 44 of the 282 children. These patients had a white blood cell count in the joint fluid of at least 50,000 per mm³ with negative findings on joint fluid and blood cultures. The diagnosis of transient synovitis was given to 86 of 282 children who had a white blood cell count in joint fluid of less than 50,000 per mm³ and negative findings on culture, resolution of symptoms without antimicrobial therapy and no further problems. The 114 children excluded from further analysis were those in atypical groups, such as those with renal failure, an immunocompromised state, neonatal sepsis, postoperative infection of the hip and other disorders.

The presenting signs in the 82 patients with septic arthritis differed significantly from those in the 86 children with transient synovitis. Compared with children with transient synovitis, those with septic arthritis had a higher erythrocyte sedimentation rate (51.6 ± 23.5 versus 21.3 ± 12.5 mm per hour) and a higher serum white blood cell count (15,000 ± 5,700 versus 9,000 ± 3,100 per mm³ [15.0 ± 5.7 versus 9.9 ± 3.1 × 10⁹ per L]). As many as 78 (95 percent) of the patients with septic arthritis could not bear weight on the hip, compared with 30 (35 percent) of the patients with transient synovitis. A history of fever was present in 67 (82 percent) of the children with septic arthritis but in only seven (8 percent) of those with transient synovitis. There was a strong relationship between an increasing number of these predictors and the proportion of patients with septic arthritis.

The authors conclude that the risk of septic arthritis of the hip can be accurately predicted by the presence of three or four specific factors (see the accompanying table). The authors state that children presenting with such signs (three or four predictors), which indicate a high probability of septic arthritis of the hip, may be good candidates for aspiration in the operating room, given the likelihood that arthrotomy and drainage will subsequently be needed. Patients with a low probability of septic arthritis can be safely observed without aspiration. Validation of this clinical prediction algorithm awaits prospective studies.