89 Detection of AmpC beta-lactamases in E. coli using different screening agars cephamycins, the ACC-group remains an exception (Jacoby 2009). This beta-lactamase seems to be less common as CMY-group AmpC beta-lactamases. Nevertheless, ACCtype enzymes have been detected in several countries in Europe (Bauernfeind et al. 1999; Girlich et al. 2000; Miró et al. 2005). Outbreaks have been reported as well (Nadjar et al. 2000). The use of cefoxitin containing agars may have limitations in areas where the ACC-group is prevalent. Clearly, there are more limitations in the present study. First, this study is an analytical exploration of sensitivity and specificity based upon retrospectively selected strains. Although the panel of strains was selected on genotype, all strains were obtained from clinical or study collections in which different culture screening criteria were used. We tried to minimize selection bias based upon our screening criteria, but the distribution of our collection may not be completely comparable to a clinical setting. The use of mainly CMY-group pAmpC isolates may have led to overestimation of sensitivity in some degree. However, the CMY-group seems to be the predominant pampC gene in clinical settings, and the cefoxitin containing agars may be therefore applicable in these setting for e.g. screening of AmpC rectal carriage. We remodelled specificity for the cefotaxime 1mg/L + cefoxitin 8mg/L and ceftazidime 1mg/L + cefoxitin 8mg/L agar, as the number of ESBL isolates seems to influence agar specificity. We expect that in a low endemic setting of ESBL specificity of both cefoxitin containing agars will increase. In the Netherlands prevalence of ESBL rectal carriage has been found to range from 5% in a teaching hospital in the South of the Netherlands (25) to 8,6% in general practices in Amsterdam (Willemsen et al. 2015; E. A. Reuland et al. 2016). Numbers on rectal carriage in North-America are scarce, but a recent prevalence study on ESBL infections in a hospital in the United States found ranges between 4.7% to 13.4% (Hoffman-Roberts et al. 2016). It is likely that screenings agars in area’s similar to the Dutch setting would have high specificity. In an ESBL high endemic area specificity of the agar will be lower. Our study was performed using strains cultured in semi-selective pre-enrichment broth. This pre-enrichment step was included to mimic our laboratory algorithm for screening of multiresistant Enterobacterales in rectal samples. Former studies on the screening of third generation cephalosporin resistance have shown that a pre-enrichment strategy results in higher yield compared to direct plating (M. F.Q. Kluytmans-Van Den Bergh et al. 2015; Nathalie Jazmati, Hein, and Hamprecht 2016; N. Jazmati, Jazmati, and Hamprecht 2017). We assume that that the use of a pre-enrichment strategy in clinical setting will result in a similar detection range, though inoculum of the original sample may differ. A limit of detection is difficult to obtain, as it is not well-known which load in 5
RkJQdWJsaXNoZXIy MTk4NDMw