Professor Medeiros presented a joint collaboration between Spanish and US researchers investigating the incidence and relevance of macrolide-resistant S. pneumoniae (MRSP) bacteremia. Microbiological data from 1,074 patients with S. pneumoniae bacteraemia was analyzed with 93 patients identified as having MRSP. The antibiotic histories of these patients were then reviewed.

Seventeen of the 93 MRSP patients developed their bacteremia while taking a macrolide. Fifteen had pneumonia and 2 had no identifiable source of infection. One patient also had meningitis. Patients failed therapy with newer macrolides and azalides, as well as with erythromycin.

Twelve isolates had the MLS phenotype which was associated with the erm gene. Five US isolates had the M phenotype which was documented to be due to the mef gene. Nearly all of the MLS phenotype strains had minimum inhibitory concentrations to erythromycin of 128 micrograms per milliliter or greater. In contrast, those with the M phenotype had lower MICs to erythromycin. The 17 clinical failures presented in this study indicate that infection with MRSP is associated with reduced outcome, for both the MLS and M phenotypes.

The clinical relevance of resistance among uropathogens was investigated in a trial of empiric therapy of acute pyelonephritis comparing 7 days of fluoroquinolone therapy with 14 days of trimethoprim sulfamethoxazole. Compared to 90% for the trimethoprim sulfamethoxazole group, almost 100% of patients treated with the fluoroquinolone had a bacteriological cure a statistically better outcome. The fluoroquinolone group maintained a statistically better outcome at 22 to 48 days as well.

This also correlated with improved clinical cure for the fluoroquinolone group both at 4 to 11 days after treatment, and at an even greater statistical advantage at 22 to 48 days post-therapy.

The difference in outcome was due to resistance to trimethoprim sulfamethoxazole. The bacteriological cure rate was 50% in those with infection caused by resistant Escherichia coli and the clinical cure rate was only 35%. In patients with trimethoprim sulfamethoxazole-susceptible E. coli, the bacteriological and clinical cure rates were 96% and 92%, respectively. This data demonstrates that in vitro resistance of E. coli to a specific antimicrobial correlates with reduced efficacy in treatment with that agent.