Prevalence, clinical expression, invasiveness and outcome of Staphylococcus aureus containing Panton-Valentine leukocidin in children treated in a university hospital of Lithuania

Abstract

Background: There is a high prevalence of Staphylococcus aureus virulence factor Panton-Valentine leukocidin (PVL) in North-East parts of Europe. The aim was to evaluate data regarding the PVL occurrences in Lithuania, determine the relationship with Methicillin resistant Staphylococcus aureus (MRSA), association with demographic and clinical conditions, invasiveness and severity of the disease in children treated in hospital Kauno klinikos (KK).

Methods: We performed a prospective case-cohort single-center study on paediatric patients hospitalized from 2012 to 2015 to KK. We compared characteristics in PVL positive [SA-PVL(+)] and PVL negative [SA-PVL(–)] groups among non-invasive and invasive infections. Logistic regression was performed to detect PVL predicting factors and Cox regression was presented to define factors associated with admission to intensive care unit (ICU).

Results: PVL was detected in 51.5%, MRSA in 7.0% and MRSA-PVL(+) in 4.8% of cases. In general, PVL was associated with older age comparing with SA-PVL(–) (median 8.5 vs. 4.0 years, p < .001). Skin and soft tissue infections were presented in 87.9% of all SA-PVL(+) cases. Invasive infections (44.7% vs. 12.1%, p < .001) and co-morbidities (20.5% vs. 2.9%, p < .001) were associated with SA-PVL(–) infections compared to SA-PVL(+), but ICU admission number was higher in invasive SA-PVL(+) cases comparing to invasive SA-PVL(–) cases (41.2% vs. 10.2%, p = .007).

Conclusions: There was a high prevalence of pvl gene in patients treated in KK. SA-PVL(+) infections were associated with SSTI and were not common in invasive infections, but the invasive infections caused by SA-PVL(+) were related to severe disease progression and admission to ICU.

Keywords:

• PVL toxin
• pvl gene
• S. aureus
• children
• prevalence
• invasiveness

Acknowledgements

We wish to thank all the team working in the microbiology laboratory in LSMU KK and Hospital Universitario 12 de Octubre in Madrid for their assistance. Special thanks to Irena Norviliene for professional molecular testing, Violeta Simatoniene for assistance in statistics and Shahin Saleh for language editing.

Disclosure statement

The authors report no conflict of interest. The authors alone are responsible for the content and writing of the paper.