Osteoarticular Infections in Children: Accurately Distinguishing between MSSA and Kingella kingae.

Introduction

Osteoarticular infections (OAIs) constitute serious paediatric conditions that may cause severe complications. Identifying the causative organism is one of the mainstays of the care process, since its detection will confirm the diagnosis, enable adjustments to antibiotic therapy and thus optimize outcomes. Two bacteria account for the majority of OAIs before 16 years of age: Staphylococcus aureus is known for affecting the older child, whereas Kingella kingae affects infants and children younger than 4 years old. We aimed to better define clinical characteristic and biological criteria for prompt diagnosis and discrimination between these two OAI. Materials and methods: We retrospectively studied 335 children, gathering 100 K. kingae and 116 S. aureus bacteriologically proven OAIs. Age, gender, temperature at admission, involved bone or joint, and laboratory data including bacterial cultures were collected for analysis. Comparisons between patients with OAI due to K. kingae and those with OAI due to S. aureus were performed using the Mann−Whitney and Kruskal−Wallis tests. Six cut-off discrimination criteria (age, admission’s T°, WBC, CRP, ESR and platelet count) were defined, and their respective ability to differentiate between OAI patients due to K. kingae versus those with S. aureus was assessed by nonparametric receiver operating characteristic (ROC) curves.

Results

Univariate analysis demonstrated significant differences between the two populations for age of patients, temperature at admission, CRP, ESR, WBC, and platelet count. AUC assessed by ROC curves demonstrated an exquisite ability to discriminate between the two populations for age of the patients; whereas AUC for CRP (0.79), temperature at admission (0.76), and platelet count (0.76) indicated a fair accuracy to discriminate between the two populations. Accuracy to discriminate between the two subgroups of patients was considered as poor for WBC (AUC = 0.62), and failed for ESR (AUC = 0.58). On the basis of our results, the best model to predict K. kingae OAI included of the following cut-offs for each parameter: age < 43 months, temperature at admission < 37.9 °C, CRP < 32.5 mg/L, and platelet count > 361,500/mm3.

Conclusions

OAI caused by K. kingae affects primarily infants and toddlers aged less than 4 years, whereas most of the children with OAI due to MSSA were aged 4 years and more. Considering our experience on the ground, only three variables were very suggestive of an OAI caused by K. kingae, i.e., age of less than 4 years, platelet count > 400,000, and a CRP level below 32.5 mg/L, whereas WBC and ESR were relatively of limited use in clinical practice.