Systematic versus random error in laboratory surveys.

In the United States, group B streptococcus (GBS) is the leading cause of serious bacterial infections in newborns. In 1996, consensus guidelines for use of intrapartum antibiotic prophylaxis (IAP) to prevent perinatal GBS disease recommended either of two methods for identifying candidates for chemoprophylaxis: 1) late prenatal culture-based screening for GBS colonization or 2) monitoring of women intrapartum for particular risk factors associated with early-onset GBS disease. Evidence that culture-based screening was substantially more effective than the risk-based approach led to revised guidelines in 2002 recommending late prenatal GBS screening for all pregnant women. Although methods for isolation and identification of GBS from prenatal specimens remained the same as those recommended in 1996, the 2002 guidelines recommended that laboratories perform antimicrobial susceptibility testing on prenatal GBS isolates from women at high risk for penicillin anaphylaxis and clarified that laboratories should report the presence of any GBS in urine specimens from pregnant women. To assess laboratory adherence to recommendations in the 2002 guidelines, CDC's Active Bacterial Core surveillance (ABCs)/Emerging Infections Program Network surveyed clinical laboratories about prenatal culture-processing practices in 2003. This report summarizes the results of that survey, which indicated that, although adherence to GBS isolation procedures was high, opportunities exist to improve implementation of recommendations related to antimicrobial susceptibility testing and GBS bacteriuria.

Clinical Laboratory Techniques↗

PubMed2004

Blinded application of microscopy, bacteriological culture, immunoassays and PCR to detect gastrointestinal pathogens from faecal samples of patients with community-acquired diarrhoea.

A blinded trial in two different laboratories was performed to compare the detection of selected enteric pathogens in 92 unselected faecal samples collected from patients with community-acquired diarrhoea by conventional and PCR-based techniques. Conventional techniques detected a single potential etiological agent in 15% of the samples, whereas results of PCR detected evidence of at least one agent in 41% of the samples. Overall, the detection rates for the different pathogens were as follows: adenovirus serogroup F, 1%; Campylobacter spp., 7.6%; Salmonella spp., 4%; enteroaggregative Escherichia coli, 9.8%; enteropathogenic E. coli, 6.5%; enterotoxigenic Clostridium perfringens, 3%; Cryptosporidium spp., 13%; and Giardia spp., 11%. Results for the detection of Salmonella spp., Campylobacter spp. and C. perfringens were similar by both techniques, whereas Cryptosporidium and Giardia spp. were detected 22 times more often by PCR than by conventional microscopy. It was not possible to compare the results for detection of enteroaggregative E. coli and enteropathogenic E. coli since these were only investigated by PCR. The results of this small study clearly demonstrate the advantages of PCR-based methods compared to conventional techniques for the detection of gastrointestinal pathogens.

Clinical Laboratory Techniques↗