at the momentof consumption.B. cereuscan causefoodpoisoning if present at
highlevels, and can produce quality issuesin milksuchas sweetcurdling.For
thesereasonsmost European countrieshaveset limitvalues of 10^4 or 10^5 cfu/
mL forB. cereusat the timeof consumption.Pasteurised milknormally contains
about100 cfu/LofB. cereus. At a storagetemperatureof 7 ÎC thesewill multiply
in 8 to 10 daysto 10^4 or 10^5 cfu/mL. Pasteurised milkmaycontain 0 cfu/L of
GNP,but oftencontains a few cfu/LGNP.At 7 ÎC thesewill multiply in 8 to 10
daysto 10^6 or 10^7 cfu/mL. It is thus clearthat aftera storage time of about9 days
boththe levelofB. cereusand the level of GNPmaybe at its maximumand
spoilage can occurby eitherof thesebacteria groups.
Theexposureassessmentstep is the part of the risk assessmentthat relates to
the specifics of the production line.It identifiesthe sources of (re)contamination
by the two groupsof bacteria and evaluatesthe possible levelof eachgroup
presentin the finishedproduct.A detailedon-siteinspectionof the production
line indicatedeight possiblesources of (re)contamination:(1) raw milk,(2)
poorly cleanable heat exchangers/gaskets in the pasteuriser, (3) constant
pressurevalve,(4) poorly cleanabletemperaturesensor on T-piece,(5) direct
air contactin buffertanks,(6) waterusedat fillingmachine,(7) air nearfilling
machine, and (8) packagingmaterial(gable-tops). The contributionof eachof
thesesourcesof contaminationwasinvestigated by microbiological testing.
Threekindsof sampleswere collected: (i) milksamplesof 1 L eachfrom
varioussamplepointsalongthe production line duringproduction, (ii) swab
samples, taken after a production run and cleaning-in-place (CIP), after
dismantlingsomepartsof the production line,and (iii) environmental samples,
suchas air samples and watersamples. Eachsample was analysedfor GNPsand
forB. cereus.
Basedon the results(given in Table20.1),the effectof contaminationfrom
the eight identified sources can be evaluated. These results indicate that
recontaminationof milkwithB. cereustakesplacein the pasteuriser:upon
passingthroughthe pasteuriser the levelofB. cereusincreasesfrom100 to
200 cfu/L. The mostprobablecauseof this recontamination is a foulinglayer
whichleaksB. cereus. The occurrenceof difficult to cleanlayerscontaining
milksoil andB. cereussporeson the plateheatexchangersand gaskets of
pasteurisers is wellknown. The results in Table20.1 also show thatthe
pasteurisationstep works welland effectively killsoff the GNPs,thus reducing
the risk associatedwith GNPsin raw milkto a minimum.
Whenduringthe inspectionthe constant pressurevalvewas dismantled,it
was seenthat the diaphragm was not perfectlyin place and that it contained
crazes.The diaphragm was swabbedand this swab showed50 000 cfu of GNPs.
If thesewere to be released overone batchof milk(200000 L), it would result in
a contaminationof 0.25 cfu/L.It is generally believedthat swabsestimate only
approximately10%of the microflora sampled,i.e. the real number of GNPs in
the diaphragmwouldbe about500 000 cfu. It is, however, veryunlikelythat all
thesebacteriawouldend up in one batchof produced milk.For the processline
considered in this example assessment, it can be estimated that the totalnumber
Riskassessmentin hygiene management 319