Resistance – a big problem in little children

Keith Klugman

The Rollins School of Public Health, Emory University, Atlanta, USA

Monitoring the prevalence of antibiotic resistance provides information from which empirical antibiotic therapy strategies should be reassessed.

Latest data from the Alexander Project, a worldwide surveillance programme of susceptibility in community-acquired lower respiratory tract infection (LRTI), show that the prevalence and degree of resistance in key pathogens – Streptococcus pneumoniae, Haemophilus influenzae and Moraxella catarrhalis – vary across Europe and the USA. There are continued high rates of penicillin-non-susceptible S. pneumoniae (MIC ³0.12 mg/l) in some regions, particularly centres in Spain (47%) and France (62%). Of further concern is the recent emergence of strains exhibiting penicillin MICs ³8 mg/l. In contrast, the prevalence of penicillin-non-susceptible S. pneumoniae strains does not exceed 10% in centres in Germany and the Czech Republic. There are also an increasing number of reports of macrolide-resistant S. pneumoniae (erythromycin MIC ³1 mg/l) in some European countries, in many cases the rates of macrolide resistance are higher than those of penicillin. Data from the Alexander Project show high levels of beta-lactamase-producing H. influenzae strains. In M. catarrhalis, beta-lactamase-producing strains dominate with the prevalence exceeding 90% in many European countries. There is evidence that in isolates from children the prevalence of resistance is even higher than that seen in adults.

Analysis of MIC90 data for a range of antibiotics shows that some agents, such as amoxicillin, have maintained good activity against S. pneumoniae, including penicillin-non-susceptible strains, compared with some cephalosporins and with macrolides. The combination of amoxicillin plus clavulanate is highly active against H. influenzae and M. catarrhalis, including beta-lactamase-producing strains. In contrast, the higher MIC90 of macrolides against H. influenzae indicates their relatively low intrinsic activity, suggesting they are less suitable empirical therapy choices for RTIs.

These findings emphasize the need for continued epidemiological surveillance of antibiotic resistance, both globally and locally and within defined patient populations such as children.