Abstract
Timely accurate diagnosis of Shiga toxin-producing Escherichia coli (STEC) infections is important. We evaluated a laboratory-developed real-time PCR (LD-PCR) assay targeting stx1, stx2, and rfbEO157 with 2,386 qualifying stool samples submitted to the microbiology laboratory of a tertiary care pediatric center between July 2011 and December 2013. Broth cultures of PCR-positive samples were tested for Shiga toxins by enzyme immunoassay (EIA) (ImmunoCard STAT! enterohemorrhagic E. coli [EHEC]; Meridian Bioscience) and cultured in attempts to recover both O157 and non-O157 STEC. E. coli O157 and non-O157 STEC were detected in 35 and 18 cases, respectively. Hemolytic uremic syndrome (HUS) occurred in 12 patients (10 infected with STEC O157, one infected with STEC O125ac, and one with PCR evidence of STEC but no resulting isolate). Among the 59 PCR-positive STEC specimens from 53 patients, only 29 (54.7%) of the associated specimens were toxin positive by EIA. LD-PCR differentiated STEC O157 from non-O157 using rfbEO157, and LD-PCR results prompted successful recovery of E. coli O157 (n = 25) and non-O157 STEC (n = 8) isolates, although the primary cultures and toxin assays were frequently negative. A rapid "mega"-multiplex PCR (FilmArray gastrointestinal panel; BioFire Diagnostics) was used retrospectively, and results correlated with LD-PCR findings in 25 (89%) of the 28 sorbitol-MacConkey agar culture-negative STEC cases. These findings demonstrate that PCR is more sensitive than EIA and/or culture and distinguishes between O157 and non-O157 STEC in clinical samples and that E. coli O157:H7 remains the predominant cause of HUS in our institution. PCR is highly recommended for rapid diagnosis of pediatric STEC infections.