PHE reveals results of pilot E. coli proficiency testing scheme
The Food and Environmental Proficiency Testing Unit (FEPTU) of PHE said to help deliver accurate and reliable results and allow laboratories to evaluate the method they use, they should be part of a suitable scheme, especially due to the importance of the pathogen and changes in legislation.
40 labs requested to take part and 32 returned results from 13 countries including Austria, Cyprus, Hong Kong, Italy, Slovenia, Spain and Ukraine.
The scheme provides external quality assessment samples for laboratories that examine foods products for Shiga toxin-producing E. coli in accordance with European legislation Regulation (EC) 2073/2005 Microbiological Criteria for Foodstuffs associated with Regulation (EC) 852/2004 amendments.
PHE is applying for it to be accredited with the UK Accreditation Service (UKAS) to ISO/IEC 17043:2010.
The aim was to produce a proficiency testing scheme for STEC organisms in food samples that supports Commission Regulation (EU) No 209/2013 amending Regulation (EC) No 2073/2005 as regards microbiological criteria for sprouts and the sampling rules for poultry carcases and fresh poultry meat.
There has been a significant increase in the EU for STEC from 2008–2012, from 3,000 to 6,000 reported cases. PHE said this is probably due to implementation of rapid techniques and increasing awareness of non-O157 STEC organisms in addition to strains of STEC O157 in testing laboratories.
Results of pilot scheme
Samples StxP1A&B contained E. coli O26:H11, which was a stx 1 & 2 and eae positive strain.
A total of 84% (27/32) detected Shiga toxin in sample StxP1A and 71% (22/32) found it in sample StxP1B.
Samples StxP2A&B contained E. coli O157:H7, which was a stx 2 and eae positive strain.
Results showed 66% failed to detect Shiga toxin in sample StxP2A and 69% failed to detect it in sample StxP2B.
PHE said analysis of questionnaire responses did not show a common methodology linked to this failure rate.
“The reason for this failure rate may be due to a lower level of genomic material in the samples (StxP2A&B) when compared to samples StxP1A&B. This is based on comparing individual laboratory CT values for stx 2 and eae targets which are consistently higher in StxP2A&B (Graph 1a-c).
“This comparison makes the assumption that participants used the same procedure and analysis for both sets of samples.”
Half the participants received samples that contained a blue dye and half contained no dye, to determine if the dye had an effect on the PCR process and the subsequent results reported.
There was no significant difference in results between Lenticule discs that contained dye and those that did not for StxP1A&B, and StxP2B.
Those for StxP2A showed a higher number of participants correctly reporting the detection of Shiga toxin (56%) for discs that contained dye, compared to discs that did not (31%).
PHE said this indicates the dye used in the samples did not affect the PCR process or results.
Objectives were to produce Lenticule discs of non-viable STEC at sufficient levels to be detectable in molecular assays used in accordance with ISO/TS 13136:2012, provide samples that challenge the detection limit of the molecular methods, determine the variability of the methods and determine if laboratories are testing for STEC in accordance with ISO/TS 13136:2012.
Variation in methods
PHE said there was a large variation in the methods used to examine the samples.
“Participants should know the limits of detection for the method they are using. This would include knowing the impact such as the volumes used from enrichment broth and DNA extraction, reagent ratios, cycle runs, would have on results obtained.”
There was a large variation of primers used by participants but the majority used a commercial kit 66% including Bax Systems/USDA, biomerieux/VIDAS, Bio-Rad iQ-Check and Biotecon.
For DNA extraction the majority (72%) of participants used a commercial kit but there was a large variation of extraction methods (none were identical), but many (>30%) used a variation of the Qiagen method/kit.
The volume of the 1ml reconstituted sample added to elution buffer varied from 20μl to 1ml, 31% used a centrifugation step and the volume of elution buffer varied from 100-600μl.
Kits included Applied Biosystems, Bax, Bio-Rad, Genelute, Instagene, Maxwell, Qiagen and Vidas.
Four participants used a pre-incubation step: one for four minutes at 37 °C, two for two minutes at 50 °C and one for three hours at 50 °C.
Five reported a cooling time and temperature: three used 30 seconds at 40 °C, one used two minutes at 40 °C, and one indefinitely at 4 °C.
