ECDC reveals Salmonella typing assessment findings
It presents the results of the sixth round of the External Quality Assessment (EQA) scheme for typing of Salmonella enterica spp. enterica.
The EQA (full report here) covers Pulsed Field Gel Electrophoresis (PFGE) and Multiple Locus Variable number of tandem repeat Analysis (MLVA).
It was for laboratories in the European Food-and Waterborne Diseases and Zoonoses Network (FWD-Net).
Molecular typing methods used for EU-wide surveillance are PFGE for all serovars and MLVA for Typhimurium.
PFGE and MVLA typing techniques
For PFGE, it focused on the production of raw PFGE gels of high quality, normalisation of PFGE images and interpretation of final results.
ECDC said PFGE is still the gold standard for high discriminatory typing of Salmonella and the only generic molecular method for typing of all serovars.
For MVLA, it covered laboratory procedure and correct data analysis (calibration of raw data into MLVA profiles according to the nomenclature).
A total of 29 laboratories signed up for the EQA-6 in October 2014–March 2015 however three did not submit any results. 25 laboratories produced PFGE results and 14 MLVA results, 13 completed PFGE and MLVA.
Comparing EQA-5 and EQA-6 there was a small increase in participating laboratories in the PFGE part (from 22 to 25), but the number of gels not useful in an inter-laboratory comparison increased from four to seven.
Mixed PFGE results
Two of seven participants were ‘new’ in the PFGE part and three of these also achieved results which were not useful for inter-laboratory comparison in EQA-5.
“Unfortunately, there were a relatively high number of participants (28%), compared with last year (14%), that scored ‘Poor’ (1) in the parameter ‘Bands’. The low scores…were primarily due to fuzzy bands on the gels,” said ECDC.
“This problem can be caused by mistakes in almost every step in the protocol, and so the participants are encouraged to go through the protocol minutely and evaluate the individual steps.
Salmonellosis is the second most commonly reported zoonotic disease in EU with 82,694 cases by the 27 EU Member States in 2013.
Salmonella is the most frequently detected causative agent in foodborne outbreaks reported to ECDC and the European Food Safety Authority (EFSA) (22.5% of outbreaks in 2013 or 1,168 outbreaks).
“It is also important to have equipment that is running properly as well as making sure that the running temperature is as described in protocol.
“In the PFGE part, only 28% (TIFF file) and 18% (BN) of the laboratories produced results that need to be improved for inter-laboratory exchange of data; in the MLVA part only three laboratories produced results that need to be improved for inter-laboratory exchange.
“However, for the majority of the identified technical issues an acceptable quality is within reach by optimisation of procedures within laboratories, trouble- shooting assistance and training.”
Comparability of profiles between laboratories primarily relies on the use of correct running conditions, good quality image acquisition, and distinct bands.
Data analysis and MLVA testing
Subsequent normalisation and interpretation of the profiles was performed using software (BioNumerics).
The PFGE part had a high participation rate and many laboratories produced fairly good gels.
However, only 68% were capable of performing the data analysis part, which indicates there is still a need for capacity building in gel analysis and interpretation by the use of BioNumerics.
Of the 14 laboratories, 79% MLVA typed all strains correctly and 93% reported correct MLVA profiles for at least nine strains.
There is a slight increase of typing all strains correctly from 71% in EQA-5 to 79% in EQA-6.
“Two of the five errors were related to missing the presence of a locus or vice versa, i.e. a false positive allele number for an absent locus,” said ECDC.
“One laboratory accounted for the remaining three errors, however the laboratory did not have any errors in the EQA-5 so maybe they changed procedures.”
EQA schemes are targeted to national reference level laboratories expected to already be performing molecular surveillance at national level. It was arranged by Statens Serum Institut (SSI) in Denmark.
Eleven laboratories (79%) reported correct MLVA profiles for all 10 strains and thirteen (93%) found the correct profile for at least nine of the ten strains.
Results indicate the majority of laboratories were able to perform the critical calibration of raw data and use the agreed nomenclature.
Gels were graded according to the TIFF Quality Grading Guidelines where seven parameters are used.
Surveillance data, including some basic typing parameters for the isolated pathogen, are reported by the Member States to the European Surveillance System (TESSy).
