Evaluation of a two-hour method for screening pathogens from stool specimens.

The 2-h stool-screening test (SST) (API System S.A., Montalieu-Vercieu, France) was used to screen 231 organisms yielding suspicious colonies on stool differential agars for potential pathogens. All 54 salmonellae yielded correct screens. Of 14 shigellas, 9 keyed as possible Shigella spp.-Yersinia enterocolitica-rare Salmonella spp. (SYS), and 5 (all Shigella sonnei) keyed as possible S. sonnei-Y. enterocolitica-Arizona spp. (SYA). Three Arizona strains were identified as probable Salmonella spp., two were identified as SYA, and one was identified as SYS. Seven Y. enterocolitica strains keyed as SYS, and two keyed as SYA; one strain was screened out as a nonpathogen. The two Aeromonas hydrophila strains keyed as SYA, and the two Plesiomonas shigelloides strains keyed as SYS. All 63 Proteus, Morganella, Providencia, and Pseudomonas aeruginosa strains were screened out as nonpathogens. Among 80 coliforms, 37 were screened as SYA, 32 were screened as SYS, and 11 were screened as nonpathogens. On the basis of the SST, 69/80 coliforms would have been tested further. However, only 1/88 potential pathogens would have been missed, and all Proteus, Morganella, Providencia, and P. aeruginosa strains would have been excluded. The capability of the SST (when combined with a rapid identification method) of providing same-day identification of potential stool pathogens must be weighed against its inability to effectively screen out coliforms. The cost of this method is equivalent to that of a three-tube conventional screening method. The suitability of the SST for individual laboratories must be predicated on the incidence of stool pathogens and commensals in the specific setting as well as on factors related to work flow, technologist acceptability, and turnaround time.