Septic Arthritis (SA) is considered a surgical urgency/ emergency by physicians around the world. As our understanding grows, and improved diagnostic algorithms are developed, it has become apparent that competing interests in terms of accurately diagnosing concurrent osteomyelitis may supercede rapid surgical intervention when the imaging is timely. Nevertheless, even in cases of isolated SA, many patients will require repeat surgery. We aimed to assess factors which could predict this.

A multicenter retrospective redcap database was created involving 20 pediatric centers from the CORTICES study group with the goal of better understanding pediatric musculoskeletal infection (PMSKI). All patients who met inclusion for the database were considered, surgeons for each site determined through imaging and chart review which patients met the diagnosis of isolated SA. Patients with concomitant abscesses or osteomyelitis were expressly excluded. Appropriate non parametric statistics were used to assess univariate significance. Multivariable logistic regression was used to assess clinical factors associated with an increased likelihood of more than one surgery. Receiver characteristics operating curve (ROC) analysis was used to determine optimal cutoffs to discriminate between children who required more than one surgery compared to those who required only one surgery. A probability algorithm was developed for the number of clinical factors present and the likelihood requiring more than one surgery following SA diagnosis.

“Four hundred and fifty-four patients with isolated SA were analyzed from 20 US hospitals. Patients were 5.4 +/− 4.8 years old at admission, and the cohort was 56% male. Of the 454 patients, 47 (10.4%) needed more than one surgery. Bivariate comparisons across surgery groups found significant differences in minimum platelet count (pplatelet, a patient has a 0.3% reduction in the odds of needing more than one surgery (OR=.997; p=0.04). For each additional ten units of CRP, a patient has a 0.1% increase in the odds of needing more than one surgery (OR=1.001; p<0.001). Our predictive algorithm found that children with both risk factors had a 64% chance of requiring multiple surgeries.

Higher CRP values and lower platelet values indicate more severe disease in isolated SA with a greater likelihood of repeat surgery. Higher CRP and lower platelet counts may portend multiple surgeries and caregivers of children with isolated SA should be advised as such.

Introduction

Infection following total knee arthroplasty (TKA) is a catastrophic complication. In the United States, for chronic, first time infected TKA, the gold standard remains a 2-stage reimplantation (2SR) procedure, with reported success rates approaching 90%. However, there is a lack of consensus on the treatment of subsequent reinfections.

Fractures of the femoral neck in children are rare, high-energy injuries with high complication rates. Their treatment has become more interventional but evidence of the efficacy of such measures is limited. We performed a systematic review of studies examining different types of treatment and their outcomes, including avascular necrosis (AVN), nonunion, coxa vara, premature physeal closure (PPC), and Ratliff’s clinical criteria. A total of 30 studies were included, comprising 935 patients. Operative treatment and open reduction were associated with higher rates of AVN. Delbet types I and II fractures were most likely to undergo open reduction and internal fixation. Coxa vara was reduced in the operative group, whereas nonunion and PPC were not related to surgical intervention. Nonunion and coxa vara were unaffected by the method of reduction. Capsular decompression had no effect on AVN. Although surgery allows a more anatomical union, it is uncertain whether operative treatment or the type of reduction affects the rate of AVN, nonunion or PPC, because more severe fractures were operated upon more frequently. A delay in treatment beyond 24 hours was associated with a higher incidence of AVN.

Cite this article: Bone Joint J 2013;95-B:135-41.