RESUMEN
We aimed to assess the accuracy of BD Phoenix for determining carbapenem susceptibility because we observed a decline in carbapenem susceptibility rate from the biannual cumulative data, after we transitioned to the BD Phoenix form Vitek 2 system. Between October 2021 and May 2022, we collected 82 non-duplicated Enterobacterales showing non-susceptible to at least one of the three carbapenems by BD Phoenix. We performed the broth microdilution (BMD) and disk diffusion (DD) according to the CLSI guideline. Compared to BMD, the categorical agreements for ertapenem (ERT), imipenem (IPM) and meropenem (MEPM) was 58.8%, 56.8% and 91.5% for BD Phoenix and it was 85.4%, 89.0%, and 97.6%, respectively, for DD (p value; 0.0001 for ERT and IPM, p value; 0.17 for MEPM). The major errors/minor errors for ERT, IPM, and MEPM were 14.0%/31.7%, 2.94%/40.7%, and 2.56%/6.10%, respectively for BD Phoenix, compared to 0%/14.6%, 0%/9.8%, and 0%/2.5%, for DD. While errors in the BD Phoenix showed tendency towards resistance, those in DD displayed no tendency towards either resistance or susceptibility. With DD, 21 out of the 27 isolates showing susceptible/intermediate/susceptible pattern (ERT/IPM/MEPM) and 13 out of the 16 isolates showing intermediate/susceptible/susceptible pattern (ERT/IPM/MEPM), were correctly categorized by DD. However, for 22 isolates showing resistant/susceptible/susceptible pattern (ERT/IPM/MEPM), only 13 isolates were correctly categorized by DD. In conclusion, to mitigate the risk of overcalling carbapenem non-susceptibility with BD Phoenix, it will be helpful to perform a complementary test using DD and to provide comments on the DD results to clinicians.
RESUMEN
We evaluated the performance of the BioFire® FilmArray® Pneumonia panel (PN-panel) in detecting bacterial pathogens by comparing it to cultures and to the usefulness of the leukocyte esterase (LE) urine strip test. Between January and June 2022, a total of 67 sputum specimens were obtained from community-acquired pneumonia patients. The PN-panel and LE test were performed simultaneously with conventional cultures. The pathogen detection rates of the PN-panel and culture were 40/67 (59.7%) and 25/67 (37.3%), respectively. The concordance rate between the PN-panel and culture was high (76.9%) when the bacterial burden was high (107 copies/mL), but it was low (8.6%) when it was 104-6 copies/mL, irrespective of the sputum quality. According to the LE positivity, the overall culture positive rate and PN-panel positive rate were significantly higher among the LE-positive specimens (23/45, 31/45) than among the LE-negative specimens (2/21, 8/21). Moreover, the difference in concordance rate between the PN-panel test and culture was significant according to the LE positivity, but not the Gram stain grading. In conclusion, the PN-panel showed high concordance when the bacterial burden was high (107 copies/mL) and ancillary use of LE test will be helpful in interpreting the PN-panel results, especially when the copy number of bacterial pathogens is low.
