Kocher Criteria differentiates pediatric septic arthritis and transient synovitis of the hip [Classics Series]

This study summary is an excerpt from the book 2 Minute Medicine’s The Classics in Medicine: Summaries of the Landmark Trials

1. Children with septic arthritis of the hip could be accurately differentiated from transient synovitis of the hip on the basis of the presence or absence of four major clinical predictors: fever, lack of weight-bearing status, erythrocyte sedimentation rate (ESR) > 40 mm/hr, and leukocytosis greater than 12 000 cells/cm3.

2. Several other variables increased the odds of septic arthritis, including radiographic evidence of joint effusion, fever with chills, female gender, and recent antibiotic use.

Original Date of Publication: December 1999

Study Rundown: When evaluating a child with acute hip pain, only a few diagnoses can be made primarily by radiographic appearance, including Legg-Calve-Perthes disease, slipped capital femoral epiphysis, and fracture. Differentiating between septic arthritis and transient synovitis of the hip remains a diagnostic dilemma. The appropriate management of septic arthritis requires timely initiation of intravenous antibiotics and surgical drainage of the infected joint, thus making rapid and accurate diagnosis vital in avoiding both the sequelae of a failure to treat and of the morbidity associated with treatment of those without a septic joint.

In the reviewed article, Kocher et al. set out to simplify the diagnostic process in differentiating between a septic hip and transient synovitis by developing and validating a set of clinical criteria by which an individual’s risk for septic arthritis may be stratified. In a retrospective review of all cases of an acutely irritable hip in children between 1979 and 1996 at a major pediatric tertiary-care hospital, several variables were found to significantly differ between those with true septic arthritis and those with transient synovitis of the hip. The four variables that were most predictive of true septic arthritis were a history of fever, an inability to bear weight on the affected hip, an ESR greater than 40 mm/hr, and a leukocytosis of greater than 12 000 cells/cm³. Patients with all four criteria were 99.6% likely to have septic arthritis, while those with zero of four criteria had septic arthritis effectively ruled out with less than 0.2% probability of an infected joint. Patients with true septic arthritis were also significantly more likely to be female, present with chills, have recent antibiotic use, and demonstrate radiographic evidence of a joint effusion at the affected site, although these features were not included in the final clinical decision algorithm due to reduced specificity. Following publication, Caird et al. prospectively evaluated these criteria in a separate pediatric population, finding that the algorithm remained effective, but inclusion of a C-reactive protein > 2.0 mg/dL was additionally predictive of septic arthritis as a fifth criterion and was independently more valuable than an ESR. However, with further prospective use of the Kocher criteria in alternative pediatric populations, such as that published in Luhmann et al. and rebutted by Kocher et al. in 2004, the clinical prediction rules demonstrated reduced but still reasonable diagnostic performance, possibly due to suboptimal modeling of the selected variables within a new population. This algorithm was the first and remains a lasting and effective evidence-based guideline for the clinical differentiation of septic arthritis from transient synovitis of the hip across multiple pediatric populations.

Click to read the study in The Journal of Bone & Joint Surgery

In-Depth [retrospective cohort]: A total of 282 children presenting with acute irritability of the hip between 1979 and 1996 at a major Northeastern pediatric hospital were included in this retrospective review. Patients were excluded if they presented with complicating factors, including immunocompromised status, renal failure, neonatal sepsis, postoperative hip infection, juvenile rheumatoid arthritis, subsequent development of Legg-Calve-Perthes disease or associated proximal femoral osteomyelitis. Patients were stratified into one of three categories of disease: True septic arthritis (38 patients), proven by positive bacterial growth from synovial fluid aspirate with associated fluid white blood cell (WBC) count > 50 000 cells/cm³; presumed septic arthritis (44 patients) on the basis of a synovial fluid aspirate with >  50 000 WBCs/cm³ but no bacterial growth on culture; or transient synovitis of the hip (86 patients), with < 50 000 WBCs/cm³ on synovial aspiration, no bacterial growth on culture, and spontaneous resolution of symptoms without antibiotics or surgery. Multiple clinical variables were tracked to determine which differed most significantly between groups by univariate analysis and multiple logistic regression analysis.

Overall, children with true septic arthritis were significantly more likely to be female, and present with a history positive for fever > 38.5 degrees Celsius, chills, recent antibiotic use, radiographic evidence of a joint effusion, non-weight bearing status on the affected side, elevated ESR, reduced hematocrit, and leukocytosis (p < 0.05 for all). On further analysis, the four variables that were most predictive of septic arthritis formed the final Kocher Criteria and included fever > 38.5C, non-weight bearing status, ESR > 40 mm/hr, and leukocytosis > 12 000 WBC/cm³. Meeting zero of the four criteria effectively ruled out septic arthritis (probability < 0.2%), while meeting all four criteria effectively ruled it in (probability of 99.6%). With three of four criteria met, probability of disease was 93.1%, which reduced to 40.0% for two criteria and 3% for one criteria.

Kocher MS, Zurakowski D, Kasser JR. Differentiating between septic arthritis and transient synovitis of the hip in children: An evidence-based clinical prediction algorithm. The Journal of Bone and Joint Surgery. 1999 Dec 1;81-A(12):1662–70.

Additional Review:

Kocher MS, Mandiga R, Murphy JM, Goldmann D, Harper M, Sundel R, et al. A clinical practice guideline for treatment of septic arthritis in children: Efficacy in improving process of care and effect on outcome of septic arthritis of the hip. The Journal of Bone and Joint Surgery. 2003 June 1;85-A(6):994–99.

Caird MS, Flynn JM, Leung YL, Millman JE, D’Italia JG, Dormans JP. Factors distinguishing septic arthritis from transient synovitis of the hip in children: a prospective study. The Journal of Bone and Joint Surgery. 2006 June 1;88-A(6):1251–57.

Luhmann SJ, Jones A, Schootman M, Gordon JE, Schoenecker PL, Luhmann JD. Differentiation between septic arthritis and transient synovitis of the hip in children with clinical prediction algorithms. The Journal of Bone and Joint Surgery. 2004 May 1;86-A(5):956–62.

©2022 2 Minute Medicine, Inc. All rights reserved. No works may be reproduced without expressed written consent from 2 Minute Medicine, Inc. Inquire about licensing here. No article should be construed as medical advice and is not intended as such by the authors or by 2 Minute Medicine, Inc