Factors associated with severity in invasive community-acquired Staphylococcus aureus infections in children: a prospective European multicentre study.

Staphylococcus aureus is the main pathogen responsible for bone and joint infections worldwide and is also capable of causing pneumonia and other invasive severe diseases. Panton-Valentine leukocidin (PVL) and methicillin-resistant S. aureus (MRSA) have been studied as factors related with severity in these infections. The aims of this study were to describe invasive community-acquired S. aureus (CA-SA) infections and to analyse factors related to severity of disease. Paediatric patients (aged 0-16 years) who had a CA-SA invasive infection were prospectively recruited from 13 centres in 7 European countries. Demographic, clinical and microbiological data were collected. Severe infection was defined as invasive infection leading to death or admission to intensive care due to haemodynamic instability or respiratory failure. A total of 152 children (88 boys) were included. The median age was 7.2 years (interquartile range, 1.3-11.9). Twenty-six (17%) of the 152 patients had a severe infection, including 3 deaths (2%). Prevalence of PVL-positive CA-SA infections was 18.6%, and 7.8% of the isolates were MRSA. The multivariate analysis identified pneumonia (adjusted odds ratio (aOR) 13.39 (95% confidence interval (CI) 4.11-43.56); p 0.008), leukopenia at admission (<3000/mm(3)) (aOR 18.3 (95% CI 1.3-259.9); p 0.03) and PVL-positive infections (aOR 4.69 (95% CI 1.39-15.81); p 0.01) as the only factors independently associated with severe outcome. There were no differences in MRSA prevalence between severe and nonsevere cases (aOR 4.30 (95% CI 0.68- 28.95); p 0.13). Our results show that in European children, PVL is associated with more severe infections, regardless of methicillin resistance.

Two decades of acute hematogenous osteomyelitis in children: are there any changes?

Our purpose was to compare the clinical course of acute hematogenous osteomyelitis (AHO) 20 years ago and today in the Department of Paediatric Surgery, Kaunas Medical University Hospital, Lithuania. Retrospective data analysis from patients aged 1-16 years with confirmed diagnosis of AHO was performed. The data were collected from 1982 to 2003. The incidence of AHO per year for 1,00,000 children (0-16 years) was analyzed for that period. The patients were divided into two groups-group A, treated in 1982-1983, and group B, treated in 2002-2003. The number of patients, patient age, duration of illness, complications, and length of hospital stay were compared using statistical methods for nonparametric data analysis (Mann-Whitney U test, chi-square criterion). Linear regression was used for incidence analysis. Population data were obtained from the Lithuanian Statistics Department. From 1982-2003, 758 patients were treated. The incidence of AHO increased from 1982. There was no statistically significant difference between the periods 1982-1983 and 2002-2003 in median patient age (10.36 and 10.72 years, respectively), in gender proportion (20.4% and 29.8% of the cases were girls), or in median duration of symptoms until admission (4 days and 3 days, respectively, p=0.058). Median hospital stay and duration of antibiotic therapy were longer in the period 1982-1983 (50 days and 43 days) than during 2002-2003 (29 days and 29 days). The differences in frequency of positive blood cultures (36.4% in group A and 64.9% in group B, p=0.046) and frequency of periosteal abscess (40.8% in group A and 19.3% in group B, p=0.015) were statistically significant. An increase in AHO incidence is seen when comparing contemporary data and the data from two decades ago, but nowadays the clinical course is less complicated and is marked by shorter hospital stays and shorter duration of antibiotic therapy.