In vitro susceptibility of Clostridium difficile to rifaximin and rifampin in 359 consecutive isolates at a university hospital in Houston, Texas.

AIM: This was an in vitro study to analyse the susceptibility of Clostridium difficile isolates to rifampin and rifaximin. METHODS: Stool samples from patients who had nosocomial diarrhoea and C difficile toxin B at a university hospital between August 2006 and December 2007 were cultured for C difficile. Susceptibility of C difficile isolates to rifaximin and rifampin was determined by agar dilution and E strips, respectively. C difficile isolates were analysed via PCR for genes encoding toxins A and B, for binary toxin (BT), and for partial deletions of the tcdC gene (tcdC-del). RESULTS: Rifaximin exhibited high-level activity against 359 C difficile isolates, with MIC(50) <0.01 microg/ml and MIC(90) 0.25 microg/ml; rifampin had MIC(50) <0.002 microg/ml and MIC(90) 4 microg/ml. Among isolates analysed, 55 (15%) were positive for BT and tcdC-del. 28 (8% of 359) isolates were resistant to rifampin (> or = 32 microg/ml), of which 6 (2% of 359) were resistant to rifaximin and rifampin with MIC values > or = 32 microg/ml. 2 of the 28 isolates resistant to rifampin were A(+)/B(+)/BT(+)/tcdC-del(+), 5 were A(+)/B(+)/BT(-)/tcdC-del(+), 4 were A(+)/B(+)/BT(+)/tcdC-del(-), 13 were A(+)/B(+)/BT(-)/tcdC-del(-), and 4 had no detectable toxin genes. Of the 11 isolates resistant to rifaximin alone, 1 was A(+)/B(+)/BT(-)/tcdC-del(+), 2 were A(+)/B(+)/BT(+)/tcdC-del(-), 6 were A(+)/B(+)/BT(-)/tcdC-del(-), and 2 had no detectable toxin genes. CONCLUSIONS: The study demonstrates that rifaximin has high-level activity against C difficile in vitro. Determination of resistance to rifampin by E strip did not predict rifaximin resistance.

Evaluation of a commercial rRNA amplification assay for direct detection of Mycobacterium tuberculosis in processed sputum.

A commercial assay (Amplified Mycobacterium tuberculosis Direct Test, Gen Probe) which combines transcription-mediated amplification of target rRNA with amplicon detection by a chemiluminescent DNA probe for the rapid detection of Mycobacterium tuberculosis in sputum was evaluated. The test was applied to consecutively collected, NALC/NaOH processed sputum sediments from two laboratories (H and L), each serving a different population of patients with pulmonary tuberculosis. Results were compared to those of fluorochrome staining and culture. A total of 760 specimens obtained from 246 patients were used for the study. The test was positive in 141 of 144 (98%) specimens that were fluorochrome-positive and culture-positive for Mycobacterium tuberculosis. Fifteen of 31 specimens that were fluorochrome-negative, culture-positive were also assay-positive. A total of 312 specimens (100 patients) from laboratory H (prevalence = 10%) and 448 specimens (146 patients) from laboratory L (prevalence = 34%) were analyzed. Compared to culture, test sensitivity, specificity, positive predictive and negative predictive values were 65%, 99%, 94% and 97%, respectively, for laboratory H and 93%, 99%, 99% and 97%, respectively, for laboratory L. If the results were analyzed on the basis of at least one concordant result between the amplification assay and culture in three sputum samples per patient, then the sensitivity, specificity, positive and negative predictive values for identifying infected patients was 70%, 99%, 87% and 97%, respectively, for laboratory H, and 100%, 98%, 96% and 100%, respectively, for laboratory L.