P481
P481
P481
Tätä julkaisua myy Denna publikation säljs av This publication is available from
VTT TIETOPALVELU VTT INFORMATIONSTJÄNST VTT INFORMATION SERVICE
PL 2000 PB 2000 P.O.Box 2000
02044 VTT 02044 VTT FIN–02044 VTT, Finland
Puh. (09) 456 4404 Tel. (09) 456 4404 Phone internat. +358 9 456 4404
Faksi (09) 456 4374 Fax (09) 456 4374 Fax +358 9 456 4374
EVALUATION OF SANITATION
PROCEDURES FOR USE IN
DAIRIES
Gun Wirtanen1, Solveig Langsrud2, Satu Salo1,
Ulla Olofson3, Harriet Alnås4, Monika Neuman3,
Jens Petter Homleid5 and Tiina Mattila-Sandholm1
1
VTT Biotechnology, Espoo, Finland
2
Matforsk, Ås, Norway
3
SIK, Göteborg, Sweden
4
Arla Foods, Göteborg, Sweden
5
TINE, Oslo, Norway
ISBN 951–38–6017–5 (soft back ed.)
ISSN 1235–0621 (soft back ed.)
ISBN 951–38–6018–3 (URL: http://www.inf.vtt.fi/pdf/)
ISSN 1455–0849 (URL: http://www.inf.vtt.fi/pdf/)
Copyright © VTT Technical Research Centre of Finland 2002
ABSTRACT
The research work for project P96049 in the second NORDFOOD programme was
carried out at the Nordic research institutes VTT Biotechnology, MATFORSK and
SIK as well as at the universities in Helsinki and Reykjavik from April 1997 to
January 2000. The companies involved in the project were the dairies Valio Ltd.
from Finland, Arla from Sweden and TINE from Norway as well as the
technochemical company Suomen Unilever Oy DiverseyLever from Finland. Dr
Gun Wirtanen, VTT Biotechnology, coordinated the project. The senior advisors at
Nordic Industrial Fund involved in the project were Maija Uusisuo and Oddur
Gunnarsson. The experiments were focused on monitoring methods in sanitation of
open and closed systems e.g. fogging, ozonation, footbath hygiene, cleaning of
cheese moulds and yoghurt pasteurizers, development of testing procedures for
measuring disinfectant efficacy, microbial resistance phenomena against
disinfectants, life-cycle assessment (LCA) and an evaluation procedure for the
functionality of the cleaning procedures. New procedures in hygiene have been
implemented in dairies, based on the results. The findings can be summarized as
follows:
• The main task of the research conducted in Sweden was to develop and
evaluate practical methods for the measure of cleaning and disinfection
efficiency. These methods are suitable for equipment surfaces used for
production of various dairy products. The analysing methods, washing out
and triphenyltetrazolium chloride (TTC) agar, worked well in testing the
cleanability of plastic cheese moulds. As part of this study a 5-step method
how to evaluate cleaning and disinfection agents was prepared at Arla
Foods.
3
indirect immunofluorescence and quartz crystal microbalance (QCM)
techniques.
• The LCA of various CIP methods, including standard CIP with lye phase,
acid phase and thermic disinfection and enzymatic cleaning followed by acid
treatment and chemical disinfection, was performed at TINE. The LCA
covered all environmental aspects including transportation and effects on the
wastewater. In an LCA all the potential environmental effects of each
emission were considered according to the worst case scenario. The LCA
provides information necessary for choosing the best CIP method, even
though widely varying assumptions must be made and limitations
4
considered. In this study enzyme-based CIP procedures showed the best
results, because enzymes are used in very small concentrations and at low
temperature.
• The aim of the sanitation study on yoghurt fermentation lines was to isolate
interfering spoilage microbes, possibly thermophilic bacteria, from a yoghurt
process and to find suitable agents for cleaning the yoghurt-soiled process
surfaces. In the pilot-scale cleaning studies, in which various cleaning agent
combinations were compared, yoghurt-milk including process isolates were
burned on stainless-steel surfaces using steam heating. The results showed
that the best cleaning effect for surfaces with the burned milk-soil was
achieved with a 2-phase cleaning procedure using chelator-based sodium
hydroxide (NaOH).
• The cheese mould hygiene studies were carried out at both pilot and process
scale. The structure of the plastic cheese moulds is complex, with long,
narrow conical channels. The ultrasound cleaning procedure was shown to
be efficient in cleaning the channels of the cheese moulds. The cleanliness of
the cheese moulds both after pilot-scale cleaning and during processing was
assessed using various methods, of which the dipslide technique proved to
be the most practical for detecting microbial contaminants. In industrial
scale, pH measurement proved to be a useful indicator for checking, as long
as the cleaning procedure was functional. The chemical oxygen demand
(COD) as well as ethylenediaminetetraacetic acid (EDTA) measurements
were useful in following up the organic load of the cleaning waters.
5
• To evaluate the influence of fluid dynamic shear (0.024–0.53 m/s) of
disinfectant solutions on the extent and mode of detachment of P.
aeruginosa biofilm, a concentric cylinder reactor (CCR) was used. The
results showed that the CCR can be used to discriminate between biocidal
and cleansing action for different disinfectant types. In testing of disinfectant
efficiency, impedance can be used for evaluating surface sterility because
very few, well-adapted, fast-growing cells as well as large amounts of
chemically treated cells change the conductance or capacitance values in the
liquid. With staining procedures, the viability and total amount of cells can
be measured on the surface. Cultivation is not a proper method for
measuring well-established biofilms on surfaces, because the cells can stick
firmly to the surface material, and thus not all cells are measured using
swabs.
6
PREFACE
The research work for project P96049 in the second NORDFOOD programme was
carried out at the Nordic research institutes VTT Biotechnology, Matforsk and SIK
as well as at the universities in Helsinki and Reykjavik from April 1997 to January
2000. The companies involved in the project were the dairies Valio Ltd. from
Finland, Arla from Sweden and TINE from Norway as well as the technochemical
company Suomen Unilever Oy DiverseyLever from Finland. The senior advisors at
Nordic Industrial Fund involved in the project were Maija Uusisuo and Oddur
Gunnarsson. Dr Gun Wirtanen, VTT Biotechnology, coordinated the project and the
project board meetings were chaired by Janna Luotola and Irma Klemetti, both from
Valio Ltd. The project group and the steering group have met 8 times during the
project. The project was carried out according to plans and the tasks to be carried
out were chosen at the steering group meetings.
Financial support for the project, which is gratefully acknowledged, was provided
by the Nordic Industrial Fund. The senior advisers for this project at the Nordic
Industrial Fund were Maija Uusisuo and Oddur Gunnarsson. The representatives at
the participating institutions were: Janna Luotola, Irma Klemetti, Matti Koivisto,
Jarmo Juurinen and Kai Hotakainen (Valio Ltd. in Helsinki, Lapinlahti and
Herajoki), Jens Petter Homleid (TINE in Oslo), Harriet Alnås and Birgitta Axelsson
(Arla FoU in Stockholm and Växjö), Urban Wiik and Kai Ahlgren (Suomen
Unilever Oy DiverseyLever in Turku and Helsinki), Solveig Langsrud, Gunhild
Sundheim and Birgitta Baardsen (Matforsk at Ås), Ulrika Husmark and Ulla
Olofson (SIK in Gothenburg), Jon Bragi Bjarnason (University of Iceland in
Reykjavik), Terhi Ali-Vehmas (University of Helsinki in Helsinki) as well as Satu
Salo, Tiina Mattila-Sandholm and Gun Wirtanen (VTT Biotechnology in Espoo).
The valuable comments of the referees Prof. Anna-Maija Sjöberg from University
ofHelsinki and Ass. Prof. Alan Friis Biocentrum-DTU are gratefully acknowledged.
Our special thanks are due to Antti Huovinen,who has painted the cover picture.
7
articles in international journals. The abstract and summary of the results achieved
in this project have also been translated into Swedish, Norwegian, Icelandic and
Finnish. The aim of the new Nordic network project DairyNET – Hygiene
control in dairy environment (P00027), in which there are partners from all
Nordic countries, is to continue the Nordic dairy hygiene research carried out in
the 2 previous NordFood programmes (1994–2000) through maintaining
contacts between industrial personnel and researchers dealing with hygiene
questions in the Nordic countries.
8
CONTENTS
ABSTRACT 3
PREFACE 7
LIST OF APPENDICES 12
9
2.7 Testing of disinfectant efficacy 50
2.7.1 Suspension tests 51
2.7.2 Surface test based on biofilms 52
2.7.3 Surface test based on biofilm constructs 55
3. MICROBIAL PHENOMENA 57
3.1 Resistance phenomenon due to use of disinfectants 57
3.1.1 Definition of resistance 57
3.1.2 Resistant strains from dairies 57
3.2 Thermophilic bacteria in yoghurt processing 58
3.3 Bacillus spores on process surfaces – Inventory study of total
contamination and B. cereus along a cream production line 59
4. ENVIRONMENTAL ASSESSMENT 61
4.1 Life-cycle analysis for assessing environmental effects of cleaning 61
4.2 Evaluation procedure for assessing the functionality of sanitation 62
10
6. CONCLUSIONS 80
6.1 Cleaning of closed systems 80
6.2 Comparison of test methods for disinfectant efficiency 80
6.3 Testing of air disinfection in industrial scale 80
6.4 Surface disinfection in industrial scale 81
6.5 Resistance phenomena due to disinfection 81
6.6 Evaluation of the cleaning of cheese moulds 81
6.7 Environmental assessment 82
REFERENCES 83
APPENDICES
11
LIST OF APPENDICES
Appendix 1. Summary of the activities in the DAIRYNI-project (P96049)
Appendix 2. Poster presentation Highlights from the NordFood2 Project
P96049 Evaluation of cleaning agents and disinfectants for use in
dairies: Methods and mechanisms (1997–2000) written by Gun
Wirtanen and presented at the conference The future for Nordic
food innovation in a European context arranged by Nordic
Industrial Fund in Stockholm 28–29 January 2002.
Appendix 3. List of publications, oral and poster presentations as well as
theses carried out in the project
Appendix 4. Sammandrag av projektet (in Swedish)
Appendix 5. Sammendrag av prosjektet (in Norwegian)
Appendix 6. Tiivistelmä ja yhteenveto projektista (in Finnish)
Appendix 7. Samantekt (Íslenska)
Appendix 8. Abstract Characterisation of Serratia marcescens surviving in
disinfecting footbaths written by Solveig Langsrud, Trond
Møretrø and Gunhild Sundheim
Appendix 9. Poster presentation Ultrasound cleaning in cheese mold hygiene
based on the Pro Gradu thesis by Antti Heino and presented at
the IAFP Annual meeting in August 2000.
Appendix 10. Abstract of the oral presentation Comparison of ultrasound based
cleaning programs for cheesery utensiles written by Wirtanen, G.,
Salo, S., Heino, A., Hattula, T. & Mattila-Sandholm, T. and
published in the proceedings of Fouling, Cleaning and
Disinfection in Food Processing (edited by Wilson, D. I., Fryer,
P. J. & Hasting, A. P. M.) held at the Jesus College in Cambridge
April 2–4, 2002, pp. 165–171.
Appendix 11. Poster presentation Effects of cleaners of biofouled stainless-steel
surfaces in yoghurt manufacturing equipment based on the
Master thesis by S. Kontulainen and presented at the IAFP
Annual meeting, August 2000.
Appendix 12. Abstract of the article Potentiation of the lethal effect of
peroxygen on Bacillus cereus spores by alkali and enzyme wash
written by Langsrud, S., Baardsen, B. & Sundheim, G. and
published in International Journal of Food Microbiology, 56
(2000) 81–86.
Appendix 13. Abstract of the article Influence of fluid dynamic forces upon the
steady-state population dynamics in microbial biofilm com-
munities written by Willcock, L., Allison, D. G., Holah, J.,
12
Wirtanen, G. & Gilbert, P. and published in Journal of Industrial
Microbiology, 25 (2000) 235–241.
Appendix 14. Poster presentation Disinfectant testing using bacteria grown in
poloxamer-hydrogel biofilm-constructs based on Bachelor theses
by Päivi Härkönen and Mervi Aalto and presented at the ASM
conference Biofilm 2000, July 2000.
Appendix 15. Colour pictures of various staining techniques presented in the
Master thesis Detection of Bacillus cereus spores on surfaces
using DEM, immunofluorescence and QCM-DTM written by
Anna Nilsson.
13
VOCABULARY AND ABBREVIATIONS
A. niger Aspergillus niger; mould used in the air disinfection
experiments
AISI American Iron and Steel Institute
amphoteric tenside agent in which the surface-active molecule contains both an
anionic and a cationic radical, either of which can be activated
under different pH conditions or in extreme environments
both simultaneously to form a zwitterion
AO acridine orange (CA index name: N,N,N',N'-tetramethyl-3,6-
acridine-diamine monohydrochloride)
ATP adenosine 5'-triphosphate (CA index name: adenosine 5'-
(tetrahydrogen triphosphate))
AU auramine O (CA index name: 4,4'-carbonimidoylbis[N,N-
dimethyl-benzenamine monohydrochloride)
AY acridine yellow (CA index name: 2,7-dimethyl-3,6-acridine-
diamine mono- hydrochloride)
B. Bacillus; Gram-positive, spore-forming rods belonging to the
Bacillus family (e.g. B. cereus, B. flavothermus, B.
licheniformis and B. subtilis)
Betane amphoteric tenside
C. albicans Candida albicans, yeast used in the air disinfection
experiments
CCR concentric cylinder reactor used in disinfectant efficacy
studies
CFU colony-forming units
cleanability ease of removal of soiling components
cleaning removal of soiling components
COD chemical oxygen demand
CIP cleaning-in-place
CTC 5-cyano-2,3-di-p-tolyltetrazolium chloride (5-cyano-2,3-
bis(p-methylphenyl)-2H-tetrazolium chloride)
D. anomala Dekkera anomala, yeast used in disinfectant efficacy
studies
DAPI 4',6-diaminidino-2-phenylindole (CA index name: 2-[4-
(aminoiminomethyl)-phenyl]-1H-indole-6-carboximidamide)
DEM direct epifluorescence microscopy
14
detergent surface-active agent containing tensides
disinfectant sanitizer; an agent reducing viable microorganisms through
destruction or removal and preventing microbial growth on
surfaces during the interproduction period in processing
DRBC agar dichloranrose bengalchloramphenicol agar
E-value capacitance or relative electrode impedance value of the
BacTrac 4100 equipment
E. coli Escherichia coli, bacterium used in disinfectant efficacy
studies
EDTA organic chelating agent; ethylenediaminetetraacetic acid (CA
index name: N,N'-1,2-ethanediylbis[N-(carboxy-methyl)-
glycine)
EHEDG European Hygienic Engineering and Design Group
ERB erythrosine B (CA index name: 3',6'-dihydroxy-2',4',5',7'-
tetraiodospiro [isobenzofuran-1(3H),9'-[9H]xanthen]-3-one
disodium salt)
ETA ethylalcohol-based disinfectants
IPA isopropylalcohol-based disinfectants
HNO3 nitric acid
KOH potassium hydroxide
L. inocua Listeria inocua, bacterium used in disinfectant efficacy
studies
L. monocytogenes Listeria monocytogenes, bacterium used in disinfectant
efficacy studies
LCA life-cycle assessment
M-value conductance or relative medium impedance value of the
BacTrac 4100 equipment
M. luteus Micrococcus luteus, bacterium used in disinfectant efficacy
studies
MIC minimum inhibitory concentration
MRD maximal recovery diluent
NaOH sodium hydroxide
NB nutrient broth
NPN 1-N-phenyl-naphthylamine
O3 ozone
OM outer membrane
ORP oxidation-reduction potential (value in mS/V)
15
Oxonia Aktiv disinfectant containing hydrogen peroxide and peracetic
acid
P. Pseudomonas; Gram-negative rods belonging to the genus
Pseudomonas
PCA plate count agar
PEI polyethyleneimine
Poloxamer F127 a diblock copolymer of polyoxyethylene and
polyoxypropylene
S. Staphylococcus Gram-positive cocci belonging to the genus
Staphylococcus
SU 560 chelating additive containing EDTA
surfactant surface-active agent
TAAS tenside-based disinfectant
TEGO disinfectant containing tetraethyleneglycol orthophtalate
tenside tensio-active substance
TGE agar tryptone glucose extract agar
TP-99 disinfectant containing alkylaminoacetate
TSA tryptone soy agar
TSB tryptone soy broth
TTC triphenyltetrazolium chloride
UHT ultrahigh-temperature treated
UV ultraviolet
QCM quartz crystal microbalance
Veterinær Ultra Des disinfectant containing quaternary ammonium compound
16
1. INTRODUCTION TO THE PROJECT
Cleanliness of surfaces, training of personnel, as well as good manufacturing and
design practices are important in combating hygiene problems in the food
industry (Holah & Timperley, 1999). Achieving a clean food plant must be
desired by the plant management, which must invest the necessary time and
money to accomplish it. Careful thought must be given to the cleaning
procedures, including the programme, cleaning agents, disinfectants and
cleaning equipment used. The key to effective cleaning of a food plant lies in
understanding the type and nature of the soiling (e.g. sugar, fat, protein and
mineral salts) and of the removal of microbial growth from surfaces. Fats are
easily removed at temperatures slightly above their melting point. Sugars and
other carbohydrates are water-soluble at elevated temperatures, but temperatures
causing caramelization should be avoided. Proteins are denatured at elevated
temperatures and may adhere strongly to surfaces at high temperatures. Cleaning
and disinfection procedures can be optimized with pilot-scale equipment for both
closed and open processes. An efficient cleaning procedure consists of a
sequence of rinse, detergent and disinfectant applications at suitable
temperatures using efficient concentrations and a final drying phase (Wirtanen &
Mattila-Sandholm, 2002).
In the food industry, equipment design and choice of surface materials (Fig. 1)
are crucial to combating biofilm formation (Holah & Timperley, 1999; Wirtanen
17
& Mattila-Sandholm, 2002). The most useful material component in processing
equipment is steel, which can be treated, e.g. with mechanical grinding,
brushing, lapping and electrolytic or mechanical polishing. It was reported that
although the grain boundaries of AISI (American Iron and Steel Institute) 316L
acid-resistant stainless steel constitute 3–20% of the total surface area, over 90%
of the adherent bacteria were found attached to the grain boundaries (Bryers &
Weightman, 1995).
Dead ends, corners, cracks, crevices, gaskets, valves and joints are vulnerable
points for biofilm accumulation (Pirbazari et al., 1990; European Hygienic
Engineering and Design Group (EHEDG), 1993a, b; Chisti & Moo-Young,
1994). Poorly designed sampling valves can destroy an entire process or give
rise to incorrect information due to biofilm effects at measuring points. As a
result of the construction, valves are vulnerable to microbial growth and thus
constitute a hygienic risk (Chisti & Moo-Young, 1994; EHEDG, 1994). Hoses,
tubes, filters etc. containing polyvinylchloride increase the risk of contamination
(Price & Ahearn, 1988). Problems with accumulation of particulates and cells
will occur whenever cleaning is inappropriate for any reason (Mettler &
Carpentier, 1998). Inadequate cleaning and sanitation of surfaces coated with
biofilms represents a source of contamination within the process (Wirtanen et
al., 2000). In practice, a biofilm left on improperly cleaned surfaces is a barrier
between microbes and the disinfectants, antibiotics or biocides used against them
18
(Kinniment & Wimpenny, 1990; Nichols, 1991; Brown & Gilbert, 1993;
Wirtanen et al., 2000).
Legislation on food hygiene and the hygienic design of food machinery, together
with public awareness of product quality and manufacturers' desires to improve
product safety, makes reliable cleanability testing an important issue. In this type
of testing it must be possible to assess the relative cleanability of various
equipment components to facilitate the design, testing and maintenance of
hygienic food-processing equipment. Assessment must be carried out using
standardized test procedures on a sound scientific basis (Cnossen & Wirtanen,
2002). The aim of EHEDG, which is an independent consortium of
representatives from research institutes, the food industry, equipment manufac-
turers and government organizations, is to develop hygienic equipment on a
scientifically and technologi-cally sound basis. The effects of cleaning
procedures used in the food industry can be evaluated using LCA. All environ-
mental aspects including the process and energy consumped in producing the
cleaning chemicals, transportation, properties of chemicals before and after
cleaning, water amount used, organic and inorganic loads in wastewater and the
recipient. The present report deals with detection and elimination of microbes on
surfaces in dairy environments. The project plan is given in Fig. 2 and a
summary of the activities in Table 1 (Appendix 1). The contact addresses of the
members in the research group and the industrial partners are given in Table 1.
19
The purposes of the Nordic Industrial Fund project P96049 ”Evaluation of
cleaning agents and disinfectants for use in dairies: methods and mechanisms”
are to develop environmentally less harmful cleaning agents and disinfectants
than those currently available on the market and reliable methods with which the
mechanisms of these agents can be studied on microbes in suspensions as well as
on surfaces in the laboratory, at pilot and process scales (Appendix 2). The
20
Table 1. Updated contact addresses of members in P96049 ”Evaluation of
cleaning agents and disinfectants for use in dairies: Methods and mechanisms”.
21
new agents should be effective in both releasing soil from processing surfaces
and killing microbes. The specific topics of the project were:
A list of publications, oral and poster presentations as well as theses carried out
in the project is given in Appendix 3. The summary of the project was translated
into Swedish (Appendix 4), Norwegian (Appendix 5), Finnish (Appendix 6) and
Icelandic (Appendix 7) languages.
22
2. EVALUATING THE EFFICACY OF SANITATION
PROCEDURES
2.1 LITERATURE REVIEW IN DAIRY HYGIENE
The tendency of microbes to adhere to and colonize inert food contact surfaces is
a matter of concern in the food-processing industry due to the significant health
consequences that can arise from the relatively low numbers of microbes
remaining on such surfaces after cleaning (Mattila-Sandholm & Wirtanen, 1992;
Carpentier & Cerf, 1993). Different types of layers can form on surfaces in dairy
manufacturing plants. Biofilms and biofouling are 2 terms used to describe surface
accumulation of organisms. Biofilm is a generic term for positive and negative
implications of microbial adhesion. A biofilm is an aggregation of microbial
cells and their associated extracellular polymeric substances, actively attached
to, growing and multiplying on a surface (Flint et al., 1997). Biofouling contains
both biofilm and organic soil. The term biofouling describes instances in which
biologically active films are considered deleterious (Zottola & Sasahara, 1994),
while fouling is used for thin milk component layers formed inside processing
equipment (Visser, 1997). Fouling is the major problem encountered in dairies,
making cleaning efficacy more difficult and thus resulting in additional costs (de
Jong, 1997; Visser & Jeurnink, 1997). Fouling in heat exhangers reduces heating
efficacy. Contamination problems caused by biofouling can be solved with
regular cleaning (Holah & Gibson, 1999). Biofilms on dairy processing lines are
characterized by rapid development (<12 h) and the predominance of single
species of bacteria, e.g. Streptococcus thermophilus or Bacillus spp. (Flint et al.,
1997). The base of the biofilm formed on gaskets removed from dairy pipelines
consisted of nonviable Gram-negative cells, while the outer surface of the
biofilm consisted of healthy Gram-positive cocci (Austin & Bergeron, 1995).
Sasahara and Zottola (1994) observed that the Gram-negative bacterium
Pseudomonas fragi can act as a primary colonizing microbe that may entrap
Listeria monocytogenes. It has been observed that one heat-resistant microbe,
Streptococcus thermophilus, can adhere to the pasteurized milk section of a
pasteurizer, inoculating the milk at a rate of 106 cells/ml (Carpentier & Cerf,
1993).
23
The soil to be removed consists mainly of milk residues, which include fat,
proteins, lactose and milk stone (Kessler, 1981). Milk stone begins to form when
milk is heated above 60 °C. The deposits adhere tightly to the surfaces, and after
runs of more than 8 h a change in colour from whitish to brownish can also be
observed (Bylund, 1995).
Surface irregularities such as roughness, crevices and pits have been shown to
increase bacterial adherence by both increasing bacterial cell attachment and
decreasing removal of attached cells by cleaning (Characklis, 1981). Regions of
the gaskets that make the seal with the pipe were shown to be more heavily
colonized than the inner diameter of the gasket (Austin & Bergeron, 1995).
Milk used for milk products is normally pasteurized (15–20 s, 72–75 ºC). Some
cheese products are made of milk that is just heated (≥ 15 s, 63–65 ºC).
Pasteurizing eliminates most of the pathogens and spoilage organisms from milk
(Chapman & Sharpe, 1990). Post-contamination may occur, if proper levels of
hygiene are not maintained in the dairy. Pathogenic bacteria found in milk and
milk products include Escherichia coli, L. monocytogenes, Yersinia entero-
colitica, Staphylococcus aureus; Salmonella sp. and Bacillus cereus (Ahmed et
al., 1983b; Johnson et al., 1990; Zottola & Smith, 1991; Zuniga-Estrada et al.,
1995; Benkerroum et al., 2002). Some examples of pathogens found in dairy
products are listed in Table 2 (Hasting, 1995).
24
Bacillus cereus in dairy products
B. cereus occurs widely in dairy products. Ahmed et al. (1983b) found that 9% of
raw milk, 35% of pasteurized milk, 14% of cheese and 48% of ice-cream samples
were contaminated with B. cereus, while Wong et al. (1988) isolated B. cereus
organisms from 29% of milk powder, 17% of fermented milk, 52% of ice-cream,
35% of soft-ice and 2% of pasteurized milk samples. Studies performed in Scotland
showed that 75% of pasteurized milk samples contained spores of B. cereus (Davies
& Wilkinson, 1973). However, food-poisoning outbreaks caused by dairy products
contaminated with B. cereus have been rare (Wong et al., 1988).
Raw milk often contains E. coli, which is killed in the pasteurization process, and
products can be contaminated due to poor production hygiene. E. coli has been
detected in fresh soft cheeses, Camembert and Brie cheeses. However, food-
poisoning outbreaks caused by dairy products contaminated with E. coli have been
rare (Heeschen & Hahn, 1996). E. coli O157:H7 has been isolated from raw milk
and bulk tank milk samples in the range of 0 to 10% (Hancock et al., 1994; Keene
et al., 1997; Mechie et al., 1997; Murinda et al., 2002). The results obtained by
Murinda et al. (2002) however show that incidences of E. coli O157:H7 are more
often associated with undercooked ground beef than with the consumption of raw
milk.
L. monocytogenes has been found in cheese, milk, ice cream and butter (Zuniga-
Estrada et al., 1995; Benkerroum et al., 2002). The main contamination routes are
surfaces of process equipment on which biofilm containing Listeria can form
(Farber & Peterkin, 1991; Farber et al., 1992; Miller et al., 1997). Listeria is known
to survive high salt concentrations. Its survival during pasteurization treatment
varies (Donnelly & Briggs, 1986; Mackey & Bratchell, 1989; Lovett et al., 1990;
Bradshaw et al.,1991; Farber & Peterkin, 1991; Farber et al., 1992; Donnelly,
1994).
25
Staphylococcus aureus in dairy products
Parts of equipment and other utensils such as cheese moulds and transportation
frames for milk cartons can be washed in closed washing tunnels. The utensils are
transported to the washer with conveyors, and nozzels in the tunnel shower the
equipment under light pressure (< 2 bar) using large volumes of detergent
solution (Daufin et al., 1987). Ultrasound cleaning can also be used to clean
utensils (Kivelä, 1996), while open surfaces can be cleaned using foam cleaning
(Hansen, 1986).
26
2.2 HYGIENE OF FOOTBATHS
Some bacteria have natural properties that enable them to circumvent the action
of a disinfectant, while others have acquired resistance mechanisms from other
bacteria (McDonnell & Russell, 1999). Such bacteria may be selected in
disinfecting footbaths. In addition, bacteria can become adapted to tolerate
higher concentrations of a disinfectant, thus making sensitive strains into
resistant ones. Bacterial growth in aqueous solutions of disinfectants has been
reported (Lowbury, 1951; Heinzel & Bellinger, 1982) and footbaths could thus
serve as a contamination source.
27
A questionnaire was distributed to 30 Norwegian dairy plants using footbaths.
They were asked the type and number of footbaths, disinfectant used (type,
concentration) and routines for use (refilling, change of disinfectant). The
number of footbaths in each dairy varied, as did the area of the footbaths (0.17–
1.8 m2) and the depth (2.5–12 cm). The mean amount of disinfectant used varied
5–15 l/m2. The most commonly used disinfectant in the footbaths was
hypochlorite; 20 out of 30 dairies used it (Table 3). TEGO and TP-99 were used
by 8 and 6 dairies, respectively. One dairy alternated between the amphoteric
tenside Betane and the quaternary ammonium compound Veterinær Ultra Des
and one used the peracetic acid-based Oxonia Aktiv.
The samples often contained spore-forming bacteria. They were not analysed
further, since bacterial spores were expected to survive in the concentrations of
disinfectants used. Vegetative bacteria were isolated from 10 out of 14 footbaths
with hypochlorite. Bacteria isolated from used chlorine compounds were
identified as Acinetobacter-like, Staphylococcus-like or unknown. The bacteria
were found both in the disinfectant (6 samples) and on the surfaces (9 samples).
Ten out of 11 footbaths containing TEGO demonstrated bacteria, both surviving
in the disinfectant and on the surfaces. Footbaths with TEGO 103G appeared to
select for Pseudomonas, Cedecea, Serratia or Proteus resistant to both TEGO
28
103G and other surface-active disinfectants e.g. benzalkonium chloride
(C6H5CH2(CH3)2NRCl). We therefore question the effectiveness of TEGO 103G
as a footbath disinfectant. Alternation with TP-99 in one dairy did not improve
the efficacy. No bacteria were isolated from 3 footbaths with 3% TP-99;
however, in footbaths with lower concentrations (1–3%), 8 out of 10 were
positive for bacteria in suspension and on surfaces. In summary, bacteria were
isolated from about 75% of the footbaths tested and none of the disinfectants
totally prevented bacterial survival. Isolation of viable bacteria from the
disinfectant used indicated that the disinfectant was neutralized by soil or that
bacteria had developed resistance to the disinfectant applied. It is difficult to
compare the efficacy of different disinfectants due to differences in routines and
the level of contamination. Depressions and scars from wear and tear of the
footbaths may influence survival of bacteria as well. The concentration used, and
the frequencies of emptying and refilling also influenced the survival of
microbes in the footbaths (Appendix 8).
Swab
In the Swab after
Number of In the before
Disinfectant neutralized rinsing of
footbaths disinfectant rinsing of
disinfectant the footbath
the footbath
Chlorine
14 3 7 10 10
compounds1
TEGO2 11 9 10 9 9
TP-993 13 5 8 8 8
Betane4 1 0 0 0 0
Oxonia Aktiv5 4 0 1 0 1
Veterinær
Ultra Des6 3 0 1 0 0
1
Disinfectants containing hypochlorite
2
Based on amphoteric tensides in the TEGO group
3
Based on alkylaminoacetate
4
Based on amphoteric tensides
5
Based on hydrogen peroxide and peracetic acid
6
Based on quaternary ammonium compound.
Reports on the isolation of bacteria from disinfectants and equipment used for
disinfection have long been published (Lowbury, 1951; Heinzel & Bellinger,
1982). Therefore, it is not surprising that we isolated bacteria from footbaths and
29
that the disinfectants sampled often contained heavy microbial loads. The use of
footbaths may kill most bacteria, but some will survive and increase the risk of
bacteria spreading from footbath to floors in the critical zones and to the
environment with aerosols in locations where the footbath is emptied, rinsed and
refilled. The use of footbaths also results in additional wetness on floors.
Bacteria on the footwear will be protected due to higher survival rates of
attached bacteria (Langsrud & Møretrø, 2001). It is important to remember that
the majority of the strains isolated were easily killed by the disinfectant after
laboratory cultivation. This has also been noted earlier when cultivating bacteria
isolated from disinfectants and disinfection equipment (Lowbury, 1951; Carson
et al., 1972; Heinzel & Bellinger, 1982). Therefore, bacteria considered sensitive
to disinfectant, e.g. L. monocytogenes, could survive and spread from footbaths.
The main objective of this study was to obtain information on the use of
footbaths in the Norwegian dairy industry and to propose methods for hygienic
control. We have documented the need for more effective hygienic control by
determining the occurrence and location of bacteria in disinfecting footbaths.
Although bacteria on the footwear could be killed by the disinfectant, a change
in footwear when entering a critical zone is recommended as a more effective
hygienic measure and should always be carried out when entering a high-risk
area. For those dairies that prefer to continue the use of footbaths as a single
measure or in combination with change in footwear, we recommend that the
concentration of disinfectant be higher than that in general use and that both the
concentration and the frequencies of refilling be documented. A bacteriological
control should also be included in the routines; this could easily be done by
swabbing about 10 cm2 of the footbath after emptying. If a consistent biofilm is
developed, the purchase of a new footbath using higher concentrations is needed.
30
on a Petri dish with plate count agar (PCA) and another with blood agar. The
Petri dishes were incubated at 20 ºC, 5–7 d nd 3–5 d, respectively.
The concentration of chlorine was tested on the same occasion. The chlorine
concentration varied from zero to 1000 ppm and there were no correlations
between the concentration and numbers or types of microbes. The test showed
that the footbaths contained many microbes of different genera, e.g. moulds,
yeasts, Enterobacteriaceae, Pseudomonas, Bacillus and Micrococcus.
The conclusion reached is that the footbaths did not function in the way they
were intended, probably because biofilms may have formed on the walls of the
footbath and that neither the disinfectant nor the cleaning routines used may
have been effective enough. The question of whether the footbath was more
harmful than useful was raised. Suggested improvements included better control
of the chlorine dosage and, if necessary, use of another type of disinfectant as
well as cleaning of the footbaths at regular intervals.
31
eliminate (Morton et al., 1998). Therefore, the efficacy of the fogging
disinfection system must also be evaluated in the plant after installment.
The 5 cheese plants included in this survey used fogging disinfection systems
from Henkel Ecolab AS, Norway, or mobile equipment from Arcon AS, Norway
(Clean Tech aps, Denmark) installed. The effect of fogging disinfection was
measured using contact agar plates. A total of 10–19 control points were
sampled, before and after disinfection using contact plates with PCA. The plates
were incubated at 20 °C for 7 d before counting. Examples of control points
included walls, ceilings, conveyor belts, electric switches, packing machines,
ventilation ducts (outside) and tanks (outside). The controls included samples
from different heights and undersides of objects. Samples were taken at different
distances from the nozzle. The microbial controls were taken in the evening
before disinfection and after disinfection in the morning just before work began.
An overview of the microbial counts is given in Tables 5 and 6. Plant 1 was not
audited and the efficacy of washing is not known. TP-99 was used for
disinfection. Microbes appeared to be reduced in 6 out of 9 control points in
Plant 1. The control points without reduction were places that were difficult to
reach, e.g. behind shelves, under a washing machine and on the wall behind the
washing machine. Yeast and spore-forming Gram-positive rods (possibly
Bacillus) were found in the microflora isolated after disinfection.
The washing procedure in Plant 2 was very exhaustive and the room was
relatively humid and cool after full washing. The room was visibly filled with
disinfectant fog during disinfection with peracetic acid-based Oxonia Aktiv. The
plant alternated with alkylaminoacetate-based TP-99. The plant used an
automatically operated unit and the interval between disinfection and rinsing was
for unknown reasons omitted. The contact time was reduced from the planned 20
min to the fog-producing period. The fogging disinfection appeared very
efficient, with reduction at all control points. However, some bacteria survived
on a water hose, wallboard joints and on the underside of a table. The microflora
after disinfection consisted of slowly growing pink colonies identified as
Methylobacterium (identification based on fatty-acid content and 16S rDNA
analysis) as well as slimy colonies of a Gram-positive bacterium identified as
Rhodococcus sp. by 16S rDNA analysis.
32
The washing process in Plant 3 was exhaustive except for a conveyor belt which
was not loosened, washed or dried on the underside. The room was humid and
warm after the washing process and visibly filled with disinfectant fog during
disinfection. However, the fog could not be expected to penetrate on the under
side of the transport band. Less fog was also seen near the ceiling in Plant 3
compared with Plant 2. The agents used were a alkylaminoacetate-based
disinfectant, TP-99, alternated with a disinfectant based on hydrogen peroxide
and peracetic acid, Oxonia Aktiv. The disinfecting efficacy was also high in
Plant 3, having only one control point with less than 50% reduction in viable
counts. The microflora consisted of small red and yellow colonies and also
different moulds after disinfection. The microflora under the conveyor belt
mainly consisted of Gram-negative bacteria.
Only equipment and tables used during the day of the audit were washed in Plant
4. All surfaces were visibly clean and the room was relatively dry with normal
temperature after washing. The fogging system did not function as planned in
Dairy 4, because the droplet size was too large to make efficient fog. The fog
produced was mainly concentrated around, and precipitated close by, the
nozzles. The disinfectant used was Oxonia Aktiv alternated with TP-99. The
contact agar plates taken at the 15 control points had visible growth before
disinfection, with reduction on 10. The microflora consisted of yeast, different
moulds and bacteria (yellow, pink and colourless colonies). The pink and yellow
colonies appeared after approximately 1 week.
The washing process in Plant 5 was mainly performed using an automatic CIP
system. The exterior of the equipment, walls and packing machines were not
thoroughly washed but all surfaces were visibly clean. The room was relatively
dry and with normal temperature. The quality of the fog produced appeared
normal, but it was concentrated around the 2 nozzle holders. The majority of
production equipment and packing lines were located in the fog, but equipment
and tanks were also located elsewhere in the production hall. The disinfectant
used was the amphoteric tenside-based Betane (Arcon AS) and the fogging
disinfection was mainly used to control the quality of the air at the premises.
Only 10 control points were sampled before and after disinfection at Plant 5, all
of them only in direct contact with the product. None of the 7 control points with
growth before disinfection indicated full reduction in viable counts, although one
point indicated about 50% reduction. The microflora mainly consisted of slow
growing small red Methylobacterium colonies, Gram-positive spore-forming
33
bacteria (possibly Bacillus sp.), Gram-positive catalase-negative rods (possibly
lactobacilli) and slow-growing yellow colonies. In summary, the efficacy of the
washing and the extent to which the fog filled the room varied greatly among the
dairies, and this was reflected in the microbial counts. About 70% of a total of 75
agar plates had microbial growth (bacteria, yeast or moulds) before disinfection
(Tables 5 and 6). Only 2 agar plates had colonies after, but not prior to,
disinfection.
Table 5. Total number of samples in each plant and numbers of contact agar
plates with and without growth before disinfection. The plates were incubated at
20 °C for 2.5 d and 7 d.
Growth after,
Plant Samples No growth1 Growth before2
not before3
1 15 5 9 1
2 16 4 12 0
3 19 5 14 0
4 15 0 15 0
5 10 2 7 1
1
Number of samples without growth before or after disinfection
2
Number of samples with growth before disinfection
3
Number of samples with growth after disinfection, but not before.
34
The disinfectants are documented using standardized suspension methods, in
which killing of more than 99.999% of laboratory-grown microbes are needed.
However, this study has shown that bacteria can survive after fogging with
disinfectants. Few isolates were highly resistant to the in-use concentration of
oxidizing disinfectants. The slow-growing bacterium isolated in several plants
was resistant to tenside-based disinfectants if grown under nutrient limited-
conditions. Thus, survival was probably caused by lack of contact between
microbes and disinfectant, e.g. insufficient washing process, inefficient fog
droplets, too low concentration of disinfectant in the fog or inadequate amount
of fog in the room and bacterial resistance to disinfectants.
35
The results suggest that the technical performance of the fogging disinfection
should be monitored frequently to ensure optimal disinfection. The highest
efficacy of the fogging disinfection on environmental surfaces was seen in plants
with an exhaustive washing process preceding the disinfection. Our results
demonstrate that visual and microbial control can be an effective tool enabling
improved hygiene. Critical control points should be identified both for technical
performance of the fogging system and disinfection efficacy. Based on these
results we recommend:
• regular control of the nozzle and complete filling of the room with fog,
• regular control to ensure that the disinfection programme is functioning as
planned,
• regular control of the amount of disinfectant consumed,
• random sampling of the disinfectant concentration in the fog and after
rinsing,
• monitoring disinfection efficacy,
• auditing of the washing programme preceding disinfection and evaluation of
the washing process,
• random sampling using 15–20 different control points as well as air
sampling. The fogging system could be optimized until the necessary level
of disinfection is attained,
• regular control using a few control points to show that the hygienic
performance of the system is maintained.
36
cereus and S. aureus) in the same trial emphasizes the reliability of the results
achieved. The placement of the stainless-steel coupons appears to affect the
result. A slightly higher reduction was found on the coupons on top of the shelf.
The results show wide differences between the 2 agents used.
8
7
Survival on surfaces
6
5
(log cfu)
4
3
2
1
0
Ref 2h 3h 4h 6h
Treatment time
37
Preliminary experiments, in which B. subtilis, Micrococcus luteus, E. coli and
yeast were grown on agar plates and the agar plates were treated directly with
ozone, showed that the microbes were killed. To verify these results, microbes
growing on stainless-steel coupons were removed from the hard surface through
shaking in a neutralizer for 30 min and various dilutions were pipetted onto agar
plates. These agar plates were finally kept in a cheese storage room during air
disinfection. A reduction of all 4 microbes was noticed. The reduction in C.
albicans was approximately 3 log-units and for S. aureus approximately 4 log-
units. However, all the surviving microbes were found near the edge of the
coupon and the edge could have prevented the ozone from reaching the agar
plate. The number of B. cereus spores and A. niger spores were in this case
reduced by 1 log-unit.
38
8
7
Viable count (log cfu/cm)
6
2
0
Reference upper shelf on the top of below the mid below the
the mid shelf shelf lowest shelf
8 A. niger spores
B. cereus spores
7 S. aureus
Viable count (log cfu/cm)
6
2
0
Reference upper shelf on the top of below the mid below the
the mid shelf shelf lowest shelf
A. niger spores
B. cereus spores
S. aureus
39
2.4 HYGIENE IN AN OPEN DAIRY PROCESS –
CHEESE PRODUCTION
Cheeses have been made for many centuries to preserve the milk. Cheese is a
concentrate of milk comprising casein, particles soluble in water and milk fat. As
a by-product whey is formed, since the casein and fat are concentrated into
cheese. Some examples of cheese types are ripened hard or semihard cheese,
blue cheese, unripened cheese as well as whey cheese. The use of rennets,
cooking of the cheese matrix and forming under pressure in moulds are common
phases in the production of most cheese types (Fox, 1987).
There are many crucial phases in the production of semihard and hard cheeses.
According to Frandsen (1992) the control procedure during production includes
analysis of the milk content, determination of the mammalian cell number,
bacterial count and freezing point as well as possible residues of antibiotics. The
cheese-milk is heat-treated either by pasteurization (70–72 ºC/15–20 s) or by
heating (65 ºC/15 s), homogenized and preacidified using a starter culture typical
for the cheese type produced. The heat treatment is performed to ensure that
neither spoilage microbes nor pathogens such as L. monocytogenes are
transferred from the raw milk into the cheese (Bertrand, 1987; Chapman &
Sharpe, 1990; Frandsen, 1992). During acidification rennet is added to the
cheese-milk, the casein in the milk is precipitated and the cheese structure is
formed. The jelly stucture or curd is cut into small pieces with special cutting
blades charasteristic for each cheese. During the syneresis phase whey is worked
out of the granular curd and the whey and granules are separated from each other.
The cheese moulds are used to remove the last part of the whey from the cheese
curd under pressure. In this phase the characteristic shape and surface of the
cheese is formed. Its main functions are to obtain a certain consistency, promote
ripening and adjust the amount of whey secreted (Bertrand, 1987; Chapman &
Shape, 1990). In earlier days cheese cloth and wooden moulds were used to form
the shape and skin of the cheese. At present these moulds are mostly made of
plastic. They are convenient because they are strong, light, cleanable, decrease
the process noise and are suitable for automation (Tamine, 1993). Moulds typical
for each type of cheese are used to obtain the characteristics of the particular
cheese type (Anon., 1980).
40
2.4.2 Methods in evaluation of cheese mould cleanliness
The design of cheese moulds is unfortunately not very hygienic. The moulds are
perforated by small holes for drainage, and made of a plastic material that cannot
withstand temperatures higher than 70 °C and their surface finish will in time be
quite rough. Therefore, it is important to have cleaning and disinfecting
procedures that are effective enough to ensure safe cheese production. The aim
here was to select and improve methods for the evaluation of washing and
disinfection effect of cheese moulds. These experiments have been performed on
cheese-moulds on 5 different occasions. Two methods, of the original 8 given in
Chapter 5.6 were selected. These 2 methods, washing out and TTC-method,
measure the remaining numbers of living microbes. We are continuing to search
for an effective method for measuring the remaining protein fraction. Four
different cleaning agents were tested.
The results using Check Pro protein kits are greatly dependent on the pH,
leading to possible false-negative reactions. Therefore, to determine whether
Check Pro is a useful method, the protein measurements should be repeated with
a neutral pH in the rinsing water. Macroscopic ultraviolet (UV) illumination was
also difficult in combination with the plastic material in the cheese moulds. UV
illumination was evaluated in 2 trials and then rejected. It may be a potentially
useful method after additional training and improvements in application. It was
difficult to use the direct epifluorescence microscopy (DEM) technique on
cheese moulds. Some bacterial cells were seen but no quantification was
possible.
Swabbing showed the same effects as washing out, but with much lower
recovery of bacteria. We chose washing out as the better of these 2 methods. The
Bioscreen measurements showed the same trends as the other culture techniques,
but with quite large standard deviations between trials as well as between
replicates within the same trial. After 4 trials the Bioscreen was rejected in
favour of the 2 other culturing-based techniques: washing out and TTC (Table
7). The adenosinetriphosphate (ATP) measurements are only relevant for lactic
acid bacteria because the spores do not contain ATP. The values obtained after
cleaning, however, were very low and it is doubtful if ATP measurements are
sensitive enough in this application.
41
Table 7. Results obtained in full-scale cheese mould studies.
Washing out
TTC moulding
Strain Sample (cfu/mould)
Mean St. Dev. Mean St. Dev.
Reference 5.0 0.0 7.5 0.0
SU436 2.5 0.0 4.6 0.1
Bacillus
Horolit CIP 2.1 0.2 4.6 0.1
P3 VR 2.4 0.4 4.5 0.0
Reference 4.0 0.0 6.6 0.0
Lactic acid SU436 1.6 0.6 2.5 0.1
bacteria Horolit CIP 1.2 0.6 2.4 0.3
P3 VR 1.4 0.6 2.5 0.2
The washing out technique gave much higher numbers of bacterial cells than
traditional swabbing. It appears to be a simple and promising method. Moulding
with TTC is also promising and quite simple to perform. Visual inspection is
even more sensitive than when using a camera. The camera, however, is simple
and fast to use and gives a quantitative value of the result.
We conclude that the methods ATP, swabbing, washing out, and moulding with
TTC, all indicated that the traditional agent (Horolit CIP) was more efficient in
cleaning than the others tested. When using an effective cleaning agent and lactic
acid bacterial contaminants, no growth could be detected after washing. The
Bacillus strain used did show some growth, which is logical since the spores
survive and adhere better than lactic acid bacterial cells. Comparing cleaning and
killing for different cleaning agents using the washing out method gave results
that were consistent for both organisms and during all repetitions. The results
from the TTC method correlate with the washing out method. Similar results are
also achieved when both indicator organisms were used.
42
2.4.3 Ultrasound cleaning procedure in cheese production
Narrow holes in the cheese mould walls serve for removal of the whey (Kivelä,
1996). These holes penetrate the walls and through them the whey is removed
during the pressing phase. When the whey flows through the narrow cavities and
perforated holes, the mould surface becomes soiled and later the whey can no
longer penetrate through these holes, thus lowering cheese quality. Furthermore,
the cheese material left on the surface also functions as a good substrate for
microbial growth (Chapman & Sharpe, 1990; Kivelä, 1996). Clogging of the
cavities in the moulds interferes with removal of the whey and the cheese
structure remains soft (Koivisto, 1999).
Yoghurt manufacturing is a long process beginning with fixing the fat content of
milk to 2.0–3.5%. Milk is fortified by adding nonfat milk solids or concentrated
43
by evaporation to obtain the final texture desired (Matalon & Sandine, 1986).
The basic mix is then homogenized and heat-treated before fermentation
(Matalon & Sandine, 1986; Savello and Dargan, 1995). After heat treatment the
milk is cooled to 40–45 °C, inoculated with 2% yoghurt bulk culture and
incubated until sufficient acidity is attained (Matalon & Sandine, 1986).
Consistency of yoghurt is dependent on the starter culture used, storage
temperature and addition of stabilizing agents (Sinha et al., 1989; Savello &
Dargan, 1995). Yoghurt processing is a very demanding task, especially when
probiotic bacteria such as Lactobacillus acidophilus and Bifidobacterium
bifidum are used, because the growth rate of these bacteria may differ
considerably and affect the fermentation process. Long incubation periods at
temperatures around 37 °C are not attainable industrially if milk has not been
sterilized in advance. After the heat treatment the facilities in the processing
plant must be aseptic to avoid the risk of microbial growth in the final product
(Driessen & Loones, 1992).
44
were cleaned with 3 different detergents, 0.7% NaOH, 0.7% NaOH containing
0.2% SU 560 (chelator) and 1.5% potassium hydroxide (KOH), all with or
without nitric acid (HNO3) treatment (as single-phase cleaning and a 2-phase
cleaning).
The study showed that bacterial numbers of harmful thermophilic bacteria will
increase strongly when processing times are long. In cleaning efficiency tests the
results showed that there are differences between detergents. In these trials the
best cleaning result was achieved using 2-phase cleaning with 0.7% NaOH
containing 0.2% SU 560 and 1.0% HNO3. The 2-phase cleaning procedure using
an alkaline mixture containing a chelator as well as HNO3 was the most efficient
combination for cleaning burned milk from stainless steel. In general, the results
showed that the acidic treatment enhanced the cleaning result. The harmful
thermophilic bacteria did not survive the cleaning treatments, but the remaining
soil was also a risk factor offering an attachment site for new contaminants. The
detachment of burned yoghurt-milk is therefore a very important but difficult
task to achieve without strong mechanical forces. Some results are presented in
the poster presentation Effects of cleaners of biofouled stainless-steel surfaces in
yoghurt manufacturing equipment in Appendix 11.
The efficiency of cleaning and disinfection was evaluated using 3 different CIP
procedures, which included a standard alkaline/acid wash, an enzyme-based
treatment as well as ozonated water (see Chapter 5.8). Two strains of B. cereus
were used in the 2 soiling procedures. The evaluation was performed measuring
the number of germinating spores, i.e the disinfection effect, and the amount of
remaining milk-soil, i.e. the cleaning effect.
45
Figure 5. Pilot equipment in simu- Figure 6. Pilot equipment in soiling
lation of soiling surfaces with burned surfaces with heated yoghurt-milk.
yoghurt-milk.
46
The sprayed spores displayed higher total cell numbers and smaller standard
deviations, compared with the naturally adhered spores. The logarithmic
reduction (4–6 units) of colony-forming units(CFU) on the steel surface after the
different CIP procedures measured with the swabbing clearly showed that the
standard wash with alkaline and acid was the most effective washing procedure.
With the enzyme-based agent no significant reduction in cell numbers was
observed. The ozonated water displayed reductions between 1–3 log-units. All of
these measurements showed lower reductions for the naturally adhered spores in
milk compared with the spores sprayed without milk. The TTC method worked
very well for low contamination levels, but all reference surfaces were over-
grown and a quantitative value could not be obtained. This method also showed
that the standard washing procedure was the most effective one. Using the
enzyme-based agent the surfaces were still overgrown after the cleaning, while
with ozonated water the reduction was estimated to be 1–3 log-units.
47
2.5.4 Potentiation of the disinfectant effect with alkali and
enzyme wash
Blakistone et al. (1999) determined the lethal effect of Oxonia Aktiv (2%,
40 °C) on a number of spore-forming bacteria and showed that B. cereus spores
were the most resistant. The resistance of Bacillus spores to disinfection varies
with the strain, sporulation, harvesting and washing procedure, and storage
conditions and recovery conditions (Waites & Bayliss, 1980). B. cereus ATCC
9139 spores were used in this study, because preliminary experiments revealed
this strain to be the most resistant to Oxonia Aktiv compared with 8 B. cereus
strains isolated from the Norwegian dairy industry (not shown). The higher
resistance shown by of B. cereus ATCC strains than by dairy isolates to chlorine
disinfectants was demonstrated by Te Giffel et al. (1995).
48
Potentiation of spores by peroxygen in alkali wash
The sporicidal effect of 1% Oxonia Aktiv was generally poor at 20 oC and 30 oC,
even after an exposure time of 30 min. The sporicidal effect increased with
higher temperature and exposure time; a log-reduction of more than 2 log-units
was obtained after 30 min at 40 oC. A concentration of 0.2 % Oxonia Aktiv had
little effect (< 1 log-unit) on the spores even at 40 o C and 30 min. Exposure to
1% NaOH (10–30 min, 60 °C) did not reduce the viability of B. cereus
significantly (< 0.2 log-unit reduction). However, pretreatment of spores with
1% NaOH at 60 oC made the spores susceptible to Oxonia Aktiv, even when this
was applied at a relatively low concentration (0.2%). The lethal effects of 0.2%
and 1% Oxonia Aktiv were similar, indicating that a subpopulation of the spores
was potentiated by the pretreatment, whereas the reminder of the population was
unaffected even after 30 min of alkali treatment. Nevertheless, it appears that if
contact between spores and warm alkali can be established, significant reduction
in spore numbers can be expected even when using a relatively low
concentration of Oxonia Aktiv.
The effect of Oxonia Aktiv on alkali-treated spores (1%, 60 °C, 20 min) was
approximately 3 log-unit reduction (Appendix 12), which was equal to the
exposure of intact spores to 1% HNO3 (65 oC, 10 min). Alkali-treated spores
were killed, showing > 5-log-units reduction by warm HNO3 (65 oC, 10 min).
49
was generally higher when using 1% alkali at 60 °C for 20 min. A 3 log-unit
reduction was obtained using the recommended in-use concentration of Parades,
if the spores were pre-exposed to 1% NaOH. No lethal effect was observed if the
pre-exposure was carried out using 60 °C water for 20 min (Appendix 12).
It is known that monocomponent enzymes can be used for biofilm removal. The
heterogenicity of the biofilm matrix limits the potential of these enzymes for use
in effective cleaning. The proteinase samples, e.g. chemotrypsin were shown to
be effective in reducing and inactivating pure-culture biofilms, but when milk
residues were present no effect of the proteinases could be observed. The
different enzymatic cleaning procedures tested were also shown to be ineffective
in inhibiting growth and metabolic activities of bacterial strains isolated from
dairies. Based on the varying results obtained for removal and inactivation of
microbes on surfaces by enzyme preparates, one possibility could therefore be to
combine various types of enzymes to attain efficient cleaning. The use of
enzymes is also limited due to the lack of techniques for quantitative evaluation
of the enzymatic effects and the accessibility of the different enzymatic
activities. The results showed that the resazurin-based fluorometric assay tested
during that part of the project performed at the Faculty of Veterinary Medicine at
the University of Helsinki can be used for estimating the enzymatic activities on
process surfaces. This method can be recommended especially when a rapid,
high-throughput capacity system is needed (Mikkola, 1999; Augustin, 2000).
50
Whitby reported chlorhexidine (C22H30Cl2N10) mixtures contaminated with
Pseudomonas sp. (Marrie & Costerton, 1981). Increased amounts of free chlorine
(2.0 mg/l) did not kill E. coli grown in biofilm. The capsular Klebsiella pneumoniae
has been shown to have a 150-fold resistance to chlorine when growing on glass
surfaces compared with suspensions. Microbial contamination has also been found
in solutions of aldehydes, quaternary compounds and amphotensides (Heinzel,
1988).
51
specified time and temperature, e.g. at 20 °C for 5 min, is determined by plate
spreading. The main advantages of these methods are that all types of
disinfectants can be tested and the effects of temperature, exposure time and
interfering substances may be included. Bactericidal tests are easy to perform,
but more time-consuming and less reproducible than the MIC test (Nicoletti et
al., 1993). For this reason bactericidal tests are mostly used when investigating a
few strains and a small number of antibacterial agents.
52
hydrogel, which protects the cells (Wirtanen et al., 1998). Denaturation can be
very rapid in suspensions (Härkönen et al., 1999); however, alcohols are not
efficient against spores, which means that they do not destroy enzymes taking
part in spore formation (Larson & Morton, 1991).
All methods used showed that the tenside-based disinfectant treatment left a
large number of viable cells on the surface. The area covered by biofilm
remained as it was before this disinfectant treatment. The tenside-based
disinfectant was not effective against biofilm bacteria both in these experiments
and in the hydrogel-construct test (Wirtanen et al., 1998; Härkönen et al., 1999).
These findings are in agreement with those from a disinfectant efficiency study
using dried bacterial cells on stainless-steel surfaces, in which a cationic tenside-
based agent also proved to be ineffective in killing bacteria (Grönholm et al.,
1999) and in results achieved using P. fragi biofilm (Wirtanen, 1995).
Cultivation and microscopy of CTC-DAPI stained biofilm on stainless-steel
surfaces confirmed that the tenside-based agent was not effective on cells in
biofilms. Impedance measurement, however, showed that the disinfectant was
effective in solution. This effect could also be seen from the permeabilization
results and from suspension tests (Wirtanen et al., 1998), which suggests that this
tenside-based agent affects naked cells but not cells embedded in biofilms or
attached to surfaces.
53
hydrogel test (Wirtanen et al., 1998). The NPN uptake assay showed that the
oxidizing peroxide-based disinfectant permeabilized the cell envelopes of the
Pseudomonas strains used. This study proved that the peroxide-based
disinfectant was the most effective disinfectant against Pseudomonas biofilms
when the microbiological activity was measured using conventional cultivation
and DEM with CTC-DAPI after a 30-min treatment. The impedance
measurement showed, however, that some viable cells were left on the surface.
Some unharmed cells can grow very rapidly, resulting in very short detection
times (Wirtanen et al., 1997).
54
discriminate between biocidal and cleansing action for different disinfectant types
(Appendix 13).
The results show that the Gram-positive bacteria tested in poloxamer hydrogels
underwent killing that varied in extent from ~ 0.1 to ~ 2-log-unit reductions. The
least susceptible organisms were M. luteus E-215 and L. monocytogenes. The
most effective agent against these 2 bacteria was a peroxide-based disinfectant
HPPA-1. This treatment was the poorest against L. innocua. It was also the most
effective of the formulations against the Salmonella strains tested, showing a
reduction of approximately 1 log-unit. The isopropyl alcohol-based IPA-L was
effective against most of the tested bacteria, except E. coli and
L. monocytogenes. The tenside-based disinfectant showed poor efficacy with all
the microbes tested. In the suspension tests using 4 the above-mentioned
disinfectants against the 11 bacteria, a 5-log-unit reduction in viable count
(5 min) was achieved in all instances, in many cases with no recoverable viable
cells. The susceptibility of the poloxamer gel constructs to HPPA, was further
evaluated over a range of concentrations, representing the extremes and mid-
point of the recommended use levels, using P. aeruginosa, E. coli, L. innocua, B.
subtilis and S. epidermidis. In most instances the degree of effectiveness
increased with increasing exposure to the agent. The results in the second study,
in which various commercial formulations of the same type were evaluated
against biofilm constructs inoculated with P. fragi, Enterobacter sp.,
L. monocytogenes, B. subtilis and Dekkera anomala, confirmed the earlier
results, showing that there is a pattern of susceptibility varying as a function both
of the organism and the disinfectant type. The results gainedfrom testing 13
commercial disinfectants showed agreement with the general observation that
Gram-negative bacteria are more resistant to disinfectant treatments than Gram-
positive bacteria (Vaara, 1992; McKane & Kandel, 1996). In all the gels with
Gram-negative bacteria, significant levels of surviving bacteria were detected.
The most effective formulations in these tests were the oxidizing hydrogen
peroxide based disinfectants. However the activity of this type of agent, unlike
the other formulations, was much lower against L. innocua and
L. monocytogenes. The killing activity generally increased with greater exposure
to the agent. They also performed best against Gram-negatives, e.g.
Enterobacter spp., Salmonella spp. and P. fragi as well as against the Gram-
55
positives L. monocytogenes and B. subtilis (vegetative cells). The isopropyl
(IPAs) and ethyl (ETA-B) alcohol-based disinfectants proved to be more
effective against vegetative cells of the Gram-positive B. subtilis than against the
other microbes tested (Appendix 14). The tenside-based disinfectant TAAS was
also the least effective in the hydrogel tests against all the microbes chosen (P.
fragi, Enterobacter spp., L. monocytogenes, B. subtilis, and D. anomala), giving
a log kill of < 0.3. These results agree with earlier studies using bacterial cells
dried on stainless steel surfaces as the inocula (Grönholm et al., 1999) and with
biofilm studies (Wirtanen et al., 1998, 2001).
56
3. MICROBIAL PHENOMENA
3.1 RESISTANCE PHENOMENON DUE TO USE OF DISINFECTANTS
For users of disinfectants in the food industry and in other applications, it would
be most relevant to define resistance as ‘survival in practical use’. However,
survival after disinfection may be explained by factors not related to the
properties of the microbes themselves, but to external factors, such as soil
neutralizing the disinfectant. Survival may also be explained by other failures in
the cleaning procedures, e.g. leaving too much water after cleaning which will
dilute the disinfectant to sub-lethal levels or using temperatures too low or
exposure times too short during disinfection.
57
use concentrations of disinfectants based on chlorine, amphotheric tenside,
alkylamine, quaternary ammonium compounds or peroxygen in a bactericidal
suspension test. Possible cross-resistance was also studied. Spore-forming strains
were not tested because they are intrinsically resistant to most disinfectants. In
general, bacteria from the footbaths or from the disinfectants used did not
survive exposure to the recommended in-use concentrations of chlorine,
alkylamine or peroxygen. Resistance to oxidative disinfectants has mainly been
associated with biofilm growth (Bolton et al., 1988; Clark et al., 1994; Mead &
Adams, 1986). In essence, this indicated that survival in footbaths containing
hypochlorite and Oxonia Aktiv was not mainly caused by development of high
resistance, but by biofilm formation.
The causes for delays in yoghurt fermentation were studied by isolating harmful
thermophilic bacteria from the yoghurt process. Bacterial numbers were determined
from hot- and cold-mixing equipment used in yoghurt manufacturing. Milk samples
were taken from the evaporator funnel, sampling valve in the pasteurizing apparatus
58
and the fermentation tanks. Milk samples were taken 0, 2, 4, 5, 6, 7, 8, 9 and 10 h
after the beginning of processing. Milk samples were cultivated on milk plate count
agar and the plates were incubated at 55 °C for 48 h. The results showed that the
numbers of thermophilic bacteria were lower when using the cold-mixing than the
warm-mixing method. The numbers of thermopholic bacteria increased
significantly after an 8-h production.
59
levels of contamination. Some sealings and gaskets, especially sealings in the
pumps located before the filling stage, were heavily contaminated with Gram-
positive rods and cocci. Also Gram-negative bacteria were found at some sites.
There were 7 sites at which B. cereus was found both in the spore and in the
vegetative form.
60
4. ENVIRONMENTAL ASSESSMENT
4.1 LIFE-CYCLE ANALYSIS FOR ASSESSING ENVIRONMENTAL
EFFECTS OF CLEANING
In this LCA 3 different transport cases (Oslo, Stavanger and Alta) and 3 different
effluent treatment methods (no treatment, internal biological effluent treatment
plant at the dairy and municipal effluent treatment plant and in addition a method
with filtration of CIP solutions) were used. The functional unit is defined as:
Satisfactorily cleaning, based on experience, of an average Norwegian dairy
with 30 cleaning operations a day (excepting pasteurizers) through one year.
The required amount of energy, water and cleaning agents was based on earlier
measurements and practical experiences. The chemical oxygen demand (COD),
phosphorus (P) and nitrogen (N) emissions were calculated from the content of
these elements in the cleaning agents. The production data of the cleaning agents
was site-specific data from the producers.
Based on the assumptions and limitations of this LCA the method of enzyme-
based cleaning has the lowest environmental impact. It has the best result for the
evaluation methods EPS and Ecoindicator, and also for the impact categories
global warming potential and ozone formation. The enzyme method and the
single-phase alkaline method had almost the same and lower use of energy than
the other methods, and also lower acidification potential. The enzyme-based
cleaning agent is used in very small concentrations and at low temperatures,
which are the main reasons for these results. The single-phase alkaline method
showed the best result for eutrophication.
61
Transport did not greatly influence the results. Filtration of CIP solutions
directly impacts the produced and transported amounts of cleaning agents, which
leads to better results for all parameters considered. The other effluent treatment
method, cleaning of wastewater, influences only the potential for eutrophication
and the Ecoscarcity evaluation method. Cleaning of the dairy and production of
cleaning agents and their delivery, have the greatest influences on the results.
Several sources of error occur in this analysis. The origin of the data is not the
same for all the CIP methods, and many assumptions have been used. Emissions
of various chemicals occur that cannot be evaluated or characterized in an LCA
e.g. phosphonates and different tensides, which are part of some of the cleaning
agents (the disinfectants of the enzyme method and the single-phase alkaline
method) used in this analysis.
The first estimate of suitability for use is made based on fundamental theoretical
aspects:
• The product content from the specification sheet
• Environmental influence should be assessed
• Theoretical assessment of functional properties based on data from the
supplier’s tests performed on a small scale. The criteria for checking include
the influence on different types of goods, ability to remove and transfer soil,
killing effect on microbes as well as technical properties (foaming
properties, stability etc.)
• A cost estimate.
b. Supplier audit
62
• Delivery certainty, product safety, responsibility of the producer, quality
aspects as well as company economy should be assessed.
d. Practical evaluation
e. Final decision
63
Theoretical phase
Working
Detergent
Environment/pollution
Function
Declaration of functional
1. Data collection properties
-Effect on materials
-Range of effectiveness
-Detergent properties
Yes No
Experience
2. Experience - Dairies
- Supplier
- Experts
- Documentation
Yes No
Final approval
64
5. DETECTION METHODS USET IN THE STUDIES
5.1 ISOLATION OF RESISTANT MICROBES
The MIC method has been used extensively to determine resistance to antibiotics
and is also used for disinfectants. In this method, one or several bacterial strains
are inoculated in a range of concentrations of disinfectant in nutrient broth (NB).
The lowest concentration allowing bacterial growth after a specified time, e.g. 24
h, is termed the MIC-value of the disinfectant for the microbe. The main
advantages of the MIC method are that it is easy to perform and many strains or
antibacterial agents can be tested in the same experiment. The applicability of
the MIC method is limited because many commonly used disinfectants cannot be
tested because their pH is either too high or low for growth or precipitation of
the disinfectant in the NB (Nicoletti et al., 1993; Sundheim & Langsrud, 1995).
The relevance of the method has also been questioned since the aim of
disinfection is not primarily to prevent growth, but to kill microbes. Therefore,
the MIC method is most commonly used in screening of strains for resistance,
comparing the efficacy of antibacterial agents or studying synergy effects.
65
more realistic views of the levels of disinfectant that will kill the microbe in
practical surface disinfection in factories with low hygienic level. In practice,
surfaces are cleaned before disinfection and the exposure to cleaning agents may
reduce the resistance of the microbes to disinfectant (Holah et al., 1998;
Langsrud et al., 2000).
Potential resistant isolates were identified. It is well known that spores and
mycobacteria survive high levels of disinfectants and identification could be
sufficient to explain survival in many cases. For certain types of bacteria
intrinsic resistance to some disinfectants and biofilm formation have been
reported, e.g. pseudomonads, and this could give some indications of why they
survive practical disinfection levels.
If it was not evident from the literature that the bacteria have high intrinsic
resistance to the disinfectant applied, we tested the level of resistance. The
strains were analysed in a simple suspension test in distilled water using the
recommended in-use concentration of the actual disinfectant. In some cases we
tested type strains or reference strains from other sources of the same species. If
type strains and reference strains were susceptible in the suspension test, this
indicated that the species generally was not expected to survive disinfection. If
they survived the test, the recommended in-use concentration of the disinfectant
may have been too low to kill the actual species. If the isolate survived the
suspension test it was termed resistant. If it was killed, we did not know if it was
susceptible or had low-level resistance.
If the isolate survived the suspension test it was necessary to determine if it had
survived or grew in the disinfectant solution for a longer time, especially if it had
been isolated from a disinfectant solution. If the isolate was killed by the
suspension test, it was on some occasions tested to determine if other growth
66
conditions or disinfection procedures more similar to those in practical use made
the isolate resistant. To find measures to eliminate a resistant isolate, potentiation
of disinfectant by a cleaning agent and/or cross-resistance to other disinfectants
was on some occasions evaluated.
67
5 min at 25 °C in the hydrogel test. After 5 min, samples (1 ml) were transferred
to a neutralizer solution (9 ml) which was kept under refrigeration (10 ± 1 °C)
for the hydrogel test and an ambient temperature (22 ± 1 °C) for the suspension
test. The aliquots were left for 5–15 min before further dilution and viable
counting on TSA plates (incubation 2–3 d at 30 °C) and yeast and malt extract
agar plates (incubation 4–5 d at 25 °C). The suspension test was performed
twice. The poloxamer-grown cultures were transferred directly to prewarmed
solutions of disinfectant, together with the stainless-steel discs. These tests were
carried out in triplicate. After 5 min the discs were removed and the gels
transferred to solutions of neutralizer (10 °C) for 5 min. This was sufficient time
for the gels to liquify and disperse. Serial dilutions were made and viable counts
estimated as before. The results were expressed as survival relative to
appropriate controls following exposure to sterile water (hydrogel test) or saline
(suspension test). The neutralizer solution contained lecithin (0.6% w/v), Tween
80 (6% w/v), sodium thiosulphate (0.8% w/v), L-histidine hydrochloride
(0.5% w/v), and bovine serum albumen (0.72% w/v) in Sorenson’s phosphate
buffer (1.25 mmol). The neutralizer solutions were sterilized by filtration
through a 0.45-µm filter. The efficiency of the neutralizer had previously been
tested using suspension cultures of M. luteus VTT-E-91474.
68
instead of the disinfectant in the control test. The following determinations were
performed:
• Conventional cultivation: The bacteria were scraped from the test surfaces
(2.5 x 7 cm2) with a cotton-tipped swab (see sample treatment in previous
chapter), which was transferred into a test tube containing 5 ml maximal
recovery diluent (MRD; LabM, UK). The test tube containing the swab was
stirred thoroughly for 1 min to release the cells into the MRD solution. The
samples were diluted as logarithmic series in MRD and cultivated at 30 °C on
PCA for 3 d.
• Impedance: The change in impedance (resistance) of the growth medium
due to microbial growth was automatically measured using a BacTrac 4100
instrument (Sy-Lab, Purckersdorf, Austria). The instrument consists of
autoclavable glass measuring vessels containing 2 pairs of electrodes
connected to a microprocessor. The metabolism caused by microbial growth
changes the concentration of ions in the growth medium and in the layers
surrounding the electrodes. Changes in conductance (the M-value; relative
medium impedance) are brought about by bacterial metabolism, whereby
weakly charged substrates in the growth medium are transformed into highly
charged end products. Capacitance (the E-value; relative electrode
impedance) can be altered by factors such as changes in the pH of the
medium. The E-value should also be used if the growth medium contains a
high concentration of salts, which may make the M-value inaccurate. The
test surfaces (12 x 55 mm2) were placed in the measuring vessels, each of
which contained 9 ml growth medium. The measuring vessels were
incubated in the BacTrac incubator block (30 °C, 48 h), and detection time
for the samples was measured when the sample reached the E-value set.
• Epifluorescence image analysis of acridine orange-stained samples: For the
steel surfaces covered with biofilm stained with AO, the stain was allowed to
act for 2 min at room temperature, after which the surfaces were rinsed with
sterile water, air-dried and stored at 4 °C. The stained surfaces were examined
under an Olympus AH-2 epifluorescence microscope (Olympus, Japan) with a
suitable filter combination, using a 100x oil-immersion objective. Area
measurements were carried out with a microcomputer system. The images (50
fields/sample) obtained in the microscope were analysed as grey scale
interpretations. The areas covered with biofilm were converted into
percentages of the total area analysed per coupon (Wirtanen, 1995).
69
• Epifluorescence image analysis of samples stained in a metabolic indicator
system: CTC and DAPI staining was performed on biofilms attached to
stainless steel surfaces. For direct staining of the coupons, 2 ml of 5 mmol
CTC (Polysciences, Inc., USA) were pipetted onto the surface and incubated
without shaking at 30 °C for 2 h. After incubation, the surfaces were rinsed
with sterile distilled water, fixed in formaldehyde (CHCHO) solution, rinsed
with sterile distilled water and air-dried at room temperature for about 20
min. A total of 2 ml of 1 µg/ml DAPI (Sigma, USA) was added to the
surface. The stain was incubated for 20 min at room temperature, after which
it was poured off. The coupons were rinsed with sterile distilled water and
air-dried. The stained biofilm samples were stored at 4 oC. The samples were
analysed using the BH-2 epifluorescence microscope (Olympus) witha total
magnification of 1000x. The total number of cells and the number of living
cells were counted from 15 microscopic fields (= 0.032 mm2) for each
sample. The results are given as actual number of cells per area; the
detection limit in this experiment was 3000 cfu/cm2. Image analysis was
carried out with the Image-Pro PlusTM program (version 3.0 for WindowsTM;
USA) using an Optronics OPDEI-470T Cooled Color charge-coupled device
(CCD) camera (Optronics Engineering, USA) and a system consisting of a
Targa 64+ ADC card image processor and a High Resolution SVPVM1353
13" RGB colour monitor image display (Sony, USA).
The bacterial suspensions were treated in in-use concentrations except for the
alcohol-based agent, of which the concentration was 0.5%, in 5 mmol N-[2-
hydroxyethyl]piperazine-N’-[4-butanesuöphonic acid] (HEPES), pH 7.2 (Sigma).
The effects of the disinfectants were assayed using the hydrophobic probe
uptake method, in which a hydrophobic fluorescent NPN probe (Merck-
Schuchardt, Germany) is added to the bacterial suspension and the resulting
fluorescence is measured at 420 nm (excitation wavelenght 350 nm; Shimadzu
RF-5000, Japan). An increase in fluorescence is associated with disruption of the
OM of the bacterial cell, since NPN only fluoresces in a lipid environment. In an
aqueous milieu the fluorescence is zero. Normally the OM prevents the entry of
NPN into the cell’s lipid layers; thus low fluorescence indicates the presence of
an intact OM (Helander et al., 1997; Helander & Mattila-Sandholm, 2000).
70
5.5 CONTROL METHOD FOR TESTING FOOTBATH HYGIENE
The sampling kit contained sterile test tubes, test tubes with NB, sterile swabs
and pipettes. The managers were asked to take 4 samples (a–d) from 3 or 4
footbaths, preferentially from the most contaminated footbath, but with cleaner
ones also included. We sampled about 50% of the footbaths in 11 dairies. The 4
samples were:
A total of 10 µl of samples a–d were plated on PCA after arrival of the samples
in the laboratory. The plates were incubated for 3–7 d at 20 °C. Colonies of
differing morphology were further cultivated and stored in Lurcia-Berthani broth
containing 15% glycerol at –80 °C.
Contamination of cheese mould surfaces was carried out using autoclaved cheese
slurry prepared from 50+ cheese. The bottom parts of the cheese forms were
contaminated with the cheese slurry in combination either with lactic-acid
bacteria from a starter culture for cheese or with B. cereus spores. The pieces of
the moulds were immersed in the solution for 15 min and rinsed by dipping 5
times in clean water. After 1–2 h of drying the bottom parts were assembled and
washed.
The cleaning equipment used in the pilot scale was a Jeros 5120 washing
machine. Washing at 70 °C lasted 6 min with a water rinse for 1.5 min. The
cleaning equipment used in the full-scale operation was a Tuchenhagen tunnel.
The cleaning programme included prerinsing with water, cleaning and finally
rinsing with water. The programme time was 3 min and the washing temperature
71
was 70 °C. After washing the level of microbial contamination was measured
with different methods:
• In washing out (measures cleaning and killing effect) one piece of the mould
was placed in a sterile bag containing sodium chloride (0.85%, 100 ml) and
Tween (0.1%). The bag with the mould piece was shaken in a shaking
device for 30 min. From this suspension the number of surviving bacteria
was measured using the traditional pour plate technique.
• The TTC method measures the cleaning and killing effect. The cheese
moulds with B. cereus spores or lactic-acid bacteria were moulded with TSA
containing the indicator TTC. The cheese moulds covered with agar were
incubated at 30 °C for 2 d. After this incubation the degree of red
colouration was estimated manually (scale with 5 degrees) and measured
with a colour camera. Ten measurements were taken of each mould piece.
• The ATP method was used to measure the cleaning effect. The special swabs
were wetted in swabbing solution; the entire test piece (12 x 12 cm2) was
swabbed. The swab was immersed and shaken for 10 s in a releasing agent
and the relative luminescence was measured in a luminometer.
• UV illumination can be used to measure the cleaning effect. A UV lamp was
used for macroscopic visualization of contaminants. The cheese mould was
illuminated with the lamp in a dark room.
• In the DEM method, which measures the cleaning and killing effect, the
cheese moulds were incubated for 6 h in NB before staining with AO and
analysis in a fluorescent microscope.
• Measurements of protein residues to measure the cleaning effect were
performed using the SwabNCheck monitoring kit (LabDesign). The cheese
mould was swabbed according to the test protocol. The colour change was
visually inspected after 10 min and was also measured in a spectro-
photometer at 562 nm. According to the technical information the sensitivity
varied between 15 and 1000 µg protein.
• Swabbing, which measures the cleaning and killing effect, was carried out
by shaking the swabs with which the surfaces were sampled in sodium
chloride (0.85%) and the number of bacteria was measured through a
conventional pour plate technique.
• Bioscreen measurements can be used to measure the cleaning and killing
effect. The Bioscreen measures bacterial growth spectrophotometrically.
This method was modified between the different trials. The cheese-mould
was immersed either in various types of NBs or sodium chloride. From this
72
suspension assays of 0.1 ml were transferred to the Bioscreen for
measurements lasting 24–45 h. The integrated area under the growth curves
was calculated.
Before the disinfection treatments, i.e. ozone treatment and fogging, the
ventilation was shut off and the experiment performed at room temperature. The
ozone concentration was approximately 1 ppm and the relative humidity 43% in
the first treatment, in the second treatment the parameters were 2 ppm and 53%,
respectively. The fogging treatment was performed with a Disinfector 2000
(Clean Tech aps) device using both hydrogen peroxide and tenside based
disinfectants. Three replicates of each microbe at each level were used. After the
disinfection treatments the coupons were placed in a tube with neutralizer and
shaken for 30 min and the number of microbes determined using culture
technique. Tryptone glucose extract (TGE) agar was used for all microbes except
A. niger, for which dichloranrose bengalchloramphenicol (DRBC) agar was
used. The TGE agar plates were incubated for 2 d at 30 °C and the DRBC agar
plates for 7 d at 20 °C. After the incubation, the number of colonies was
enumerated.
73
5.8 EVALUATION OF CLEANING AND DISINFECTION EFFICACY OF
VARIOUS CIP PROCEDURES USING SPORE-SOILED SURFACES
Stainless-steel surfaces were contaminated with B. cereus spores and milk. The
test surfaces of stainless steel (ss2343, 2B; 45 x 45 mm2) were soiled in 3
different ways:
• surfaces contaminated with B. cereus spores of the dairy strain 341. The
spores were suspended in buffered saline solution (5 x 108 spores/ml) and
sprayed with a paint-brush in 6 thin layers. After spraying the surfaces were
dried overnight at room temperature in a sterile bench whose surfaces were
gently rinsed 4 times in cold water just before the CIP cleaning.
• surfaces with B. cereus spores (NVH1) in milk. The test surfaces were
mounted vertically in the spore-containing skim milk suspension, using a
magnetic stirrer in the soiling vessel. Spores were allowed to adhere for 2 d
at low temperature (4 °C). The surfaces with adhered spores were gently
rinsed 4 times in cold water before the CIP cleaning.
• surfaces with milk-soil were soiled using 1 ml of 3% fat milk which was
poured and spread, covering the entire surface, and then air-dried at 50 °C
overnight. These surfaces were gently rinsed 4 times in cold water before the
CIP cleaning.
After drying, the surfaces were cleaned and disinfected using 3 different CIP
procedures in the vertical section of the pilot-plant CIP rig (120 l) at a flow of
1.5 m/s. A standard alkaline/acid procedure was compared with a formulated,
enzyme-based agent, and to ozonated water (Ozotech, Norway). Ozone was
produced and bubbled directly into the test-rig tank. Ozonated water was built up
during 1–2 h. The conductivity was measured and the concentration of ozone
was calculated from the oxidation-reduction potential (ORP) value. The
concentrations of alkaline and acid substances (conductivity) were measured at-
line during cleaning. The pH value was also determined by titration. It was not
possible to determine the concentration of the enzyme-based agent either with
conductivity or titration, and therefore an adequate amount of the agent was
directly added to the buffer tank. The concentration of ozone was calculated
using the ORP values. After cleaning and disinfection the number of surviving
spores was enumerated and the amount of soil remaining estimated; all surfaces
were rinsed with a neutralizing solution before the analyses were performed. The
CIP procedures tested were:
74
• Standard rinse (prerinse with cold water for 5 min, 70 °C 0.9% NaOH for 10
min, 65 °C 1% nicric acid for 5 min and a final rinse with cold water for 5
min)
• Enzyme-based cleaning (prerinse with cold water for 5 min, treatment at 50
°C using 0.09% enzyme-based cleaner, rinse with cold water for 5 min,
treatment with peroxygen-based 0.2% disinfectant and a final rinse with cold
water for 5 min)
• Ozonated water (prerinse with cold water for 5 min, 0.5–1.0 ppm ozone
rinse for 5–15 min and a final rinse with cold water for 5 min).
The number of surviving spores, i.e. the disinfection effect, was measured using
swabbing of the surfaces and moulding with TTC agar. The amount of remaining
milk-soil, i.e. the cleaning effect, was estimated in 5 different ways using a
protein-measuring kit (SwabNCheck), ATP measurement, measurement of the
contact angle of water, a UV microscopy method and staining the surfaces with
fluorescent AO and microscopy.
75
In the fluorochrome experiments the surfaces tested (stainless steel, glass, teflon,
and silicone) were washed with ethanol and dried. The B. cereus spores (108
spores/ml) were prepared in three different mixtures: in saline and in milk-NaCl
(1:10 and 1:50) solutions. The mixtures were spread onto surfaces by spraying
twice. The surfaces were dried, whereafter they were stained with fluorochromes
at different concentrations and the excess stains removed from the surfaces by
rinsing with distilled water. The surfaces were then investigated using
fluorescence microscopy. A total of 40 images (equivalent to 0.001 mm2/surface)
within each sample were calculated manually; the analysis was also performed
automatically using an image analysis system (0.0081 mm2/surface). The results
were expressed in amount of spores/mm2.
Glass surfaces (1 cm2) were inoculated with spore suspension and dried for use
in immunofluorescence studies. These surfaces were then placed in a humidity
chamber. Phosphate buffered saline containing 1% bovine serum albumine was
added and after a 15-min incubation at room temperature the excess buffer was
removed. The primary antibody solution was added to the surfaces followed by
incubation at 37 ºC for 60 min, washing with buffer and addition of the
secondary antibody solution. These surfaces were incubated at 37 ºC for 30 min
and washed with buffer before the surfaces were analysed with fluorescence
microscopy. In all, 10 images (equivalent to a total of 0.625 mm2) within the
measurement frame were calculated manually, and the image analysis system
calculated a total area of 0.529 mm2. The experiments were performed with 2
different B. cereus strains, with and without addition of milk in the spore
suspension (1:20 and 1:50, respectively). The results were expressed in amount
of spores/ mm2.
In the QCM-DTM the cell suspension was injected into the QCM chamber,
which contained saline solution. The response was registered at the same
moment as the cells were in contact with the sensor surface (gold). Both
vegetative cells of B. cereus (3 x 108 cells/ml) and B. cereus spores (2 x 108
spores/ml) were examined. The measurements were performed twice without
cleaning of the chamber between the 2 measurements.
76
5.9.2 Cleaning of a production line
The swab assays were split into 2 parts. One part was directly analysed and the
other part heat-treated (80 °C, 10 min) before analysing. The heat treatment was
performed for the enumeration of B. cereus spores. The swab assays were
inoculated on TGE and blood agar. All samples were incubated at room
temperature for 14 d. Colonies with a Bacillus appearance were transferred for
confirmation on B. cereus agar and Mossel agar.
Parts, e.g. sealings and gaskets, that were dismounted were moulded with Mossel
agar containing polymyxin which is quite selective for B. cereus. The gaskets
were split into 2 parts, one of which was moulded directly and the other heat-
treated at 80 °C for 10 min before analysing. The colour indicator TTC was
added to the agar just before moulding occurrred. These samples were incubated
at room temperature for 14 d. Some colonies were transferred to TGE agar and
then to B. cereus agar and Mossel agar for confirmation.
77
SucAAPFpNA was used for alkalase from Novo and chymotrypsin from cod.
The measurements were performed at 25 °C using equilibrated buffer
(Ásgeirsson et al., 1992; Bjarnason & Ásgeirsson, 1993; Bjarnason et al., 1997).
In the determinations the increase in absorption rate at 410 nm was recorded, and
the activity corresponds to that section of the curve having the greatest slope. In
some cases the initial rate can be low and the optimum is achieved in the second
or third minute of the reaction, which is especially true for crude enzyme
samples: one unit of enzyme activity is the amount of enzyme that hydrolyses 1
mmol of substrate per minute The comparative experiments were performed
using equivalent activity concentrations of the enzymes in the comparative
cleaning and disinfection experiments. Furthermore, a pH of approximately 8.0
was used in all experiments at refrigerated temperatures for 2 h and also
extended duration of 24 h (Bjarnason & Ásgeirsson, 1993; Bjarnason et al.,
1997).
78
polysaccharides were measured using ruthenium red stain. The changes in
turbidity using Bioscreen equipment (Labsystem) of the samples were measured
to determine the bacterial growth, which can be seen as an increase in turbidity.
This measurement was used to measure chemical residues in the samples after
washing and rinsing. The fluorogenic method was chosen for use in further
studies (Mattila-Sandholm et al., 1991).
79
6 CONCLUSIONS
6.1 CLEANING OF CLOSED SYSTEMS
The tests with the use of ozone for CIP cleaning and disinfection gave poor
results in microbial reduction. The main reason for this is probably the failure of
reaching an effective ozone concentration in the CIP system.
When the disinfectant used contained hydrogen peroxide and peracetic acid the
fogging was efficient while fogging with agents containing tensides was not
efficient. There is a need to test new agents in a controlled way. These tests
80
should be based on surface tests. In performing surface test it is very important
to know the behaviour of the microbes in different environments and situations.
The disinfection efficiency of ozone on surfaces was not good. If the capacity of
the equipment is sufficient and the surfaces are clean, fogging disinfection can
be used to eliminate bacteria from surfaces. It is recommended that the
disinfection efficacy in practical applications should be monitored frequently
using several sampling sites.
Spores and some resistant vegetative bacteria can survive disinfection due to
build-up of resistance against disinfectants. Some bacteria are intrinsically
resistant whereas other adapt to the disinfectant used. It was also noticed that
spores or bacteria forming biofilms may not be eliminated by ordinary cleaning
and disinfection procedures.
The cheese moulds were washed with different agents and washing-methods and
then analysed with different methods. None of the tested methods (protein-kit
Check Pro, ATP, swabbing, washing out, UV-illumination, DEM and TTC agar)
worked well for detection of soil. The TTC agar and washing out was most
reliable for detection of microbial growth. These methods generate compatible
and repeatable results. In the pilot-scale trials the contact agar method gave the
most reliable results. There is a need for improvement of methods with which the
level of organic residues can be detected so that the pure cleaning effect can be
evaluated. In the full-scale trial the microbiological tests showed that the rinsing
water was contaminated with a high number of bacteria. The methods can
therefore also be used to evaluate parameters in the supply systems e.g. as the
efficacy of the cleaning agents.
81
6.7 ENVIRONMENTAL ASSESSMENT
The main goal of cleaning and disinfection in the food industry is to remove dirt
and fouling and to destroy any remaining microbes on the cleaned surfaces.
Other aspects of cleaning are economy, environmental impact, safety and
corrosion. Life cycle analysis (4.1) is a useful method for looking at the whole
range of environmental impact of different CIP-methods, starting with raw
materials and ending with effluent handling. The weakness of LCA in this
context is the uncertainties in valuation and characterization of environmental
impact from certain chemicals such as phosphonates and tensides.
82
REFERENCES
Ahmed, A. A.-H., Mustafa, M. K. & Marth, E. 1983a. Growth and enterotoxin
production by Staphyloccus aureus in whey from the manufacture of Domiati
cheese. J. Food. Prot., Vol. 46, pp. 235–237.
83
Bertrand, F. 1987. The main steps in manufacture. In: Eck, A. (ed.).
Cheesemaking: Science and Technology. Paris: Lavoisier Publishing. Pp. 413–
421.
Blakistone, B., Chuyate, R., Kautter, D., Charbonneau, J. & Suit, K. 1999.
Efficacy of Oxonia active against selected spore-formers. J. Food Prot., Vol. 63,
pp. 262–267.
84
Boulange-Peterman, L. 1996. Processes of bioadhesion on stainless steel
surfaces and cleanability: a review with special reference to the food industry.
Biofouling Vol. 10, pp. 275–300.
Bredholt, S., Maukonen, J., Kujanpää, K. Alanko, T., Olofson, U., Husmark U.,
Sjöberg, A.-M. & Wirtanen, G. 1999. Microbial methods for assessment of
cleaning and disinfection of food-processing surfaces cleaned in a low-pressure
system. Eur. Food Res. Technol., Vol. 209, pp. 145–152.
Burfoot, D., Hall, K., Brown, K. & Xu, Y. 1999. Fogging for the disinfection of
food processing factories and equipment. Food Sci. Technol., Vol. 10, pp. 203–
210.
Bylund, G. 1995. Dairy processing handbook. Lund: LP Grafiska AB. Pp. 403–
413.
Carpentier, B. & Cerf, O. 1993. Biofilms and their consequences, with particular
reference to hygiene in the food industry. J. Appl. Bact., Vol. 75, pp. 499–511.
85
Chapman, H. R. & Sharpe, M. E. 1990. Microbiology of cheese. In: Robinson,
R.K. (ed.). Dairy microbiology: The microbiology of milk products. Great
Yarmouth: Elsevier Science Publishers Ltd. Pp. 203–291.
Clark, R., Rice, E., Pierce, B., Johnson, C. & Fox, K. 1994. Effect of aggregation
on Vibrio cholerae inactivation. J. Environ. Eng., Vol. 120, pp. 875–887.
Cloete, T. E., Smith, F. & Steyn, P. L. 1989. The use of planktonic bacterial
populations in open and closed recirculating water cooling systems for the
evaluation of biocides. Int. Biodeterior., Vol. 25, pp. 115–122.
Daufin, G., Ecalard, J. P. & Kerherve L. 1987. Cleaning and disifection. In: Eck,
A. (ed.). Cheesemaking: Science and technology. Paris: Lavoisier Publishing.
Pp. 444–449.
Davies, F. L. & Wilkinson, G. 1973. Bacillus cereus in milk and dairy products.
In: Hobbs, B. C. & Christian, J. B. H. (eds.). The microbial safety of food.
London: Academic Press. Pp. 57–67.
de Jong, P. 1997. Impact and control of fouling in milk processing. Trends Food
Sci. Technol., Vol. 8, pp. 401–405.
86
Donnelly, C. W. & Briggs, E. H. 1986. Psychrotrofic growth and thermal
inactivation of Listeria monocytogenes as a function of milk composition. J.
Food Prot., Vol. 49, pp. 994–998.
Farber, J. M., Daley, E., Coates, F., Emmons, D. B. & McKellar, R. 1992.
Factors influencing survival of Listeria monocytogenes in milk in a high-
temperature short-time pasteurizer. J. Food Prot., Vol. 55, pp. 946–951.
87
Flemming, H.-C. 1991. Biofouling in water treatment. In: Biofouling and
biocorrosion in industrial water systems. Flemming, H.-C. & Geesey, G. G.
(eds.). Berlin–Heidelberg: Springer–Verlag. Pp. 47–80.
Frandsen, T. 1992. Milk for cheese production. Scand. Dairy Inf., Vol. 1, pp.
12–13.
Gilbert, P., Jones, M. V., Allison, D. G., Heys, S., Maira, T. & Wood, P. 1998.
The use of poloxamer hydrogels for the assessment of biofilm susceptibility
towards biocide treatments. J. Appl. Microbiol., Vol. 85, pp. 985–991.
Grönholm, L., Wirtanen, G., Ahlgren, K., Nordström, K. & Sjöberg, A.-M.
1999. Screening of antimicrobial activities of disinfectants and cleaning agents
against foodborne spoilage microbes. Z. Lebensm. Unters. Forsch. A, Vol. 208,
pp. 289–298.
Hancock, D. D., Besser, T. E., Kinsel, M. L., Tarr, P. I., Rice, D. H. & Paras,
M. G. 1994. The prevalence of Eschericia coli in dairy and beef cattle in
Washington State. Epidemiol. Infect., Vol. 113, pp. 199–207.
Hansen, R. 1986. Surface cleaning of dairy equipment. North Europ. Dairy J.,
10, p. 285.
Hasting, T. 1995. Good enough to eat. The Chemical Engineer, 9 Nov. 1995, pp.
19–20.
88
Heeschen, W. & Hahn, G. 1996. Escherichia coli: alter Freund – neuer Feind?
Deutsche Milchwirtschaft, Vol. 10, pp. 448–454.
Holah, J., Lavaud, A., Peters, W. & Dye, K. 1998. Future techniques in
disinfectant testing. Int. Biodeterior. Biodegrad., Vol. 41, pp. 273–279.
89
Hood, S. K. & Zottola, E.A. 1995. Biofilms in food processing. Food Control,
Vol. 6, pp. 9–18.
Härkönen, P., Salo, S., Mattila-Sandholm, T., Wirtanen, G., Allison, D. G. &
Gilbert, P. 1999. Development of a simple in vitro test system for the
disinfection of bacterial biofilms. Water Sci. Technol., Vol. 39, pp. 219–225.
Keene, W. E., Hedberg, K., Herriott, D. E., Hancock, D. D., McCay, R. W.,
Barrett, T. J. & Flemming, D. W. 1997. A prolonged outbreak of Escherichia
coli O157:H7 infections caused by commercially distributed raw milk. J. Infect.
Dis., Vol. 176, pp. 815–818.
Kivelä, T. 1996. Easier cheese mould cleaning by ultrasonics. Scand. Dairy Inf.,
Vol., 1, pp. 34–35.
90
Langsrud, S. & Møretrø, T. 2001. Characterisation of bacteria surviving in
disinfecting footbaths. SFAM summer Conference 2001, 9–12 July, Swansea
(UK). Poster abstract. 1 p.
Lovett, J., Wesley, I. V., Vandermaaten, M. J., Bradshaw, J. G., Francis, D. W.,
Crawford, R. G., Donnelly, C. W. & Messer, J. W. 1990. High-temperature
short-time pasteurization inactivates Listeria monocytogenes. J. Food Prot., Vol.
53, pp. 734–738.
91
Mattila-Sandholm, T. & Wirtanen, G. 1992. Biofilm formation in the industry, a
review. Food Rev. Int., Vol. 8, pp. 573–603.
Mettler, E. & Carpentier, B. 1998. Varations over time of microbial load and
physicochemical properties of floor materials after cleaning in food industry
premises. J. Food Prot., Vol. 61, pp. 57–65.
92
Mucchetti, G. 1995. Biological fouling and biofilm formation on membranes.
Int. Dairy Fed. Special Issue No. 9504, pp. 118–124.
Murinda, S. E., Nguyen, L. T., Ivey, S. J., Gillespie, B. E., Almeida, R. A.,
Draughon, F. A. & Oliver, S. P. 2002. Prevalence and molecular characterization
of Escherichia coli O157:H7 in bulk tank milk and fecal samples from cull
cows: a 12-month survey of dairy farms in east Tennessee. J. Food Prot., Vol.
65, pp. 752–759.
Nicoletti, G., Boghossian, F., Gurevitch, F., Borland, R. & Morgenroth, P. 1993.
The antimicrobial activity in vitro of chlorhexidine, a mixture of iso-
thiazolinones ('Kathon' CG) and cetyl trimethyl ammonium bromide (CTAB).
J. Hosp. Infect., Vol. 23, pp. 87–111.
93
Storgårds, E., Simola, H., Sjöberg, A.-M. & Wirtanen, G. 1999. Hygiene of
gasket materials used in food processing equipment. Part 2: Aged materials.
Trans. Inst. Chem. Eng. Part C, Vol. 77, pp. 146–155.
Te Giffel, M. C., Beumer, R. R., van Dam, W. F., Slaghuis, B. A. & Rombouts,
F. M. 1995. Sporicidal effect of disinfectants on Bacillus cereus isolated from
milk processing environment. Int. Biodeter. Biodegrad., Vol. 36, pp. 421–430.
Vaara, M. 1992. Agents that increase the permeability of the outer membrane.
Microbiol. Rev., Vol. 56, pp. 395–411.
Visser, H. 1997. Fouling and cleaning of heat equipment. IDF Bull. Doc., Vol.
328, p. 4.
Visser, H. & Jeurnink, Th. J. M. 1997. General aspects of fouling and cleaning.
IDF Bull. Doc., Vol. 328, pp. 5–6.
Waites, W. M. & Bayliss, C. E., 1980. The preparation of bacterial spores for
evaluation of sporicidal activity of chemicals. In: Gould, G. W. & Corry, J. E. L.
(eds.). Microbial growth and survival in extremes of environment. London:
Academic Press. Pp. 159–172.
94
Wirtanen, G. 1995. Biofilm formation and its elimination from food processing
equipment. Espoo: VTT Publications 251. 106 p. + appendices 48 p. ISBN 951-
38-4789-6.
Wirtanen, G., Salo, S., Allison, D., Mattila-Sandholm, T. & Gilbert, P. 1998.
Performance-evaluation of disinfectant formulations using poloxamer-hydrogel
biofilm-constructs. J. Appl. Microbiol., Vol. 85, pp. 965–971.
Wirtanen, G., Salo, S., Maukonen, J., Bredholt, S. & Mattila-Sandholm, T. 1997.
NordFood Sanitation in dairies. Espoo: VTT Publications 309. 47 p. +
appendices 22 p. ISBN 951-38-5055-2.
Wong, H.-C., Chang, M.-H. & Fan, J.-Y. 1988. Incidence and characterization of
Bacillus cereus isolates contaminating dairy products. Appl. Environ.
Microbiol., Vol. 54, pp. 699–702.
Wood, P., Jones, M. Bhakoo, M. & Gilbert, P. 1996. A novel strategy for control
of microbial biofilms through generation of biocide at the biofilm-surface
interface. Appl. Environ. Microbiol., Vol. 62, pp. 2598–2602.
95
Zuniga-Estrada, A., Lopez-Merino, A. & Mota de la Garza, L. 1995. Behavior of
Listeria monocytogenes in milk fermented with a yoghurt starter culture. Rev.
Latinoam. Microbiol., Vol. 37, pp. 257–265.
96
Summary of activities in the project.
APPENDIX 1
APPENDIX 2
1997
♦ Gilbert, P., Wirtanen, G. & Allison, D. (1997). Standard laboratory test methods
and their relevance in the evaluation of disinfectants. Autumn Meeting of the
Society for Applied Microbiology (October 22, 1997). SFAM, London. 2 p.
♦ Wirtanen, G., Salo, S., Maukonen, J., Bredholt, S. & Mattila-Sandholm, T.
(1997). NordFood Sanitation in dairies. Espoo: VTT Publications 309. 47 p. +
appendices 22 p. ISBN 951-38-5055-2.
1998
♦ Wirtanen, G., Salo, S., Allison, D., Mattila-Sandholm, T. & Gilbert, P. (1998).
Performance-evaluation of disinfectant formulations using poloxamer-hydrogel
biofilm-constructs. Journal of Applied Microbiology, 85, pp. 965–971.
♦ Mikkola, J., Ahlgren, J. & Ali-Vehmas, T. (1998). A fluorometric screening
method for analysing bacterial activities in biofilms – Efficacy of enzymatic
cleaning procedures. Tiedevinkki, Faculty of Veterinary Medicine, University
of Helsinki (Poster presented Dec. 12, 1998).
1999
♦ Bredholt, S., Maukonen, J., Kujanpää, K., Alanko, T., Olofson, U., Husmark, U.,
Sjöberg, A.-M. & Wirtanen, G. (1999). Microbial methods for assessment of
cleaning and disinfection of food-processing surfaces cleaned in a low-pressure
system. European Food Research and Technology, 209, pp. 145–152.
♦ Härkönen, P., Salo, S., Mattila-Sandholm, T., Wirtanen, G., Allison, D.G. &
Gilbert, P. (1999). Development of a simple in vitro test system for the
disinfection of bacterial biofilms. Water Science and Technology, 39, pp. 219–
225.
♦ Høgaas Eide, M. & Homleid, J. P. (1999). Livsløpsanalyse (LCA) av CIP-
metoder. Meieriposten nr. 2/99, pp 32–35.
♦ Sundheim, G. & Eide, O. K. (1999). Bakteriologiske undersøkelser av fotbad.
Meieriposten, 7, pp. 190–192.
♦ Sundheim, G. & Homleid, J. P. (1999). Forhold som påvirker effekten av
tåkedesinfeksjon. Meieriposten, 9, pp. 248–255.
♦ Wirtanen, G., Salo, S. & Mattila-Sandholm, T. (1999). Microbial methods in
evaluation of cleaning procedures. In Wirtanen, G., Salo, S. & Mikkola, A.
(eds.). 30th R3-Nordic Contamination Control Symposium. VTT Symposium
193. Libella Painopalvelu Oy, Espoo. Pp. 207–218.
2000
♦ Langsrud, S., Baardsen, B. & Sundheim, G. (2000). Potentiation of the lethal
effect of peroxygen on Bacillus cereus spores by alkali and enzyme wash.
International Journal of Food Microbiology, 56, pp. 81–86.
♦ Maukonen, J., Mattila-Sandholm, T. & Wirtanen, G. (2000). Metabolic indicators
for assessing bacterial viability in hygiene sampling using cells in suspensions
and swabbed biofilm. Lebensmittel-Wissenschaft und -Technologie, 33, pp. 225–
234.
♦ Willcock, L., Allison, D. G., Holah, J., Wirtanen, G. & Gilbert, P. (2000).
Influence of fluid dynamic forces upon the steady-state population dynamics in
microbial biofilm communities. Journal of Industrial Microbiology, 25, pp.
235–241.
♦ Wirtanen, G., Salo, S., Aalto, M. & Gilbert, P. (2000). Disinfectant testing using
microbes grown in biofilm-constructs. In: Proceedings of 15th ICCCS
International Symposium and 31st R3-Nordic Symposium on Contamination
Control. Pp. 319–324.
♦ Wirtanen, G., Heino, A. & Salo, S. (2000). Ultrasound cleaning in cheese mold
hygiene. Proceedings of the 87th Annual Meeting of International Association
for Food Protection. Atlanta, Georgia. Poster P9. Pp. 40.
♦ Wirtanen, G., Kontulainen, S. & Salo, S. (2000). Effects of cleaners of
biofouled stainless-steel surfaces in yoghurt manufacturing equipment.
Proceedings of the 87th Annual Meeting of International Association for Food
Protection. Atlanta, Georgia. Poster P6. Pp. 39.
♦ Wirtanen, G., Saarela, M. & Mattila-Sandholm, T. (2000). Biofilms – Impact on
hygiene in food industries. In Biofilms II: Process Analysis and Applications,
Bryers, J. (ed.). New York: John Wiley-Liss Inc. Pp. 327–372. ISBN 0-471-
29656-2.
3/2
♦ Wirtanen, G., Salo, S., Aalto, M. & Gilbert, P. (2000). Disinfectant testing
using microbes grown in poloxamer-hydrogel biofilm-constructs. Proceedings
of the ASM Conference on Biofilm 2000 (B2K). Big Sky, Montana. Poster
113. Pp. 55.
♦ Wirtanen, G., Storgårds, E., Saarela, M., Salo, S. & Mattila-Sandholm, T.
(2000). Detection of Biofilms in the Food and Beverage Industry. In Industrial
Biofouling, Walker, J., Surman, S. & Jass, J. (eds.). John Wiley & Sons, Ltd.,
Chichester. Pp. 175–203. ISBN-0-471-98866-9.
2001
♦ Bore, E. & Langsrud, S. (2001). Identification of microorganisms isolated from
dairy industy after cleaning and disinfection. SFAM Summer Conference, 9–12
July, Swansea (UK). Poster abstract. 1 p.
♦ Langsrud, S. & Møretrø, T. (2001). Characterisation of bacteria surviving in
disinfecting footbaths. SFAM Summer Conference 2001, 9–12 July, Swansea
(UK). Poster abstract. 1 p.
♦ Wirtanen, G., Salo, S., Helander, I. M. & Mattila-Sandholm, T. (2001).
Microbiological methods for testing disinfectant efficiency on Pseudomonas
biofilm. Colloids and Surfaces B: Biointerfaces, 20, pp. 37–50.
♦ Wirtanen, G., Aalto, M., Härkönen, P., Gilbert, P. & Mattila-Sandholm, T.
(2001). Efficacy testing of commercial disinfectants against foodborne
pathogenic and spoilage microbes in biofilm-constructs. European Food
Research and Technology, 213, 409–414.
2002
♦ Wirtanen, G. (2002). Highlights from the NordFood2 Project P96049
"Evaluation of cleaning agents and disinfectants for use in dairies: Methods and
mechanisms" (1997–2000). Symposium on "The future for Nordic food
innovation in a European context" in Stockholm 28–29 January 2002. Oslo:
Nordic Industrial Fund. Poster abstract. 2 p.
♦ Wirtanen, G., Salo, S., Heino, A., Hattula, T. & Mattila-Sandholm, T. (2002).
Comparison of ultrasound based cleaning programs for cheesery utensils. In:
Fouling, Cleaning and Disinfection in Food Processing. Wilson, D. I., Fryer, P. J.
& Hastings, A. P. M. (eds.). Cambridge: City Services Design and Print. Pp.
165–171. ISBN 0 9542483 0 9.
3/3
♦ Wirtanen, G., Salo, S. & Storgårds, E. 2002. Microbial assessment
methods.used.in cleaning efficacy evaluation. In: Matuszek, T. (ed.). Food,
packaging, equipment and building surfaces in their contribution to food
products contamination and process safety. Gdansk: Gdansk University of
Technology Publishing Office. Pp. 49–57. ISBN 83-88579-40-1.
♦ Sundheim, G. & Langsrud, S. (2002). Characterization of Cedecea like strains
surviving in disinfecting footbath. Manuscript in preparation.
1997
♦ Kallio, J. (1997). Gram-negatiivisten bakteerien solunläpäisevyyttä edistävien
aineiden vaikutukset pesu- ja desinfiointiaineiden tehokkuuteen (The effect of
permeability-increasing agents and disinfectants on Gram-negative bacteria).
Bachelor thesis, Helsinki: Pikku-Sitomo. 75 p. + 5 appendices.
1999
♦ Härkönen, P. (1999). In vitro -testin kehittäminen desinfiointiaineiden tehon
testaamiseen biofilmibakteereilla (Development of an in vitro test system for
disinfection of biofilm bacteria). Bachelor thesis, Helsinki: Pikku-Sitomo. 60 p.
+ 1 appendix.
♦ Mikkola, J. (1999). Uusia seulontamenetelmiä puhdistusaineiden tehon
mittaamiseksi meijeripesuja varten. – Fluorometrinen, turbidometrinen ja
kolorimetrinen menetelmä (New screening methods for the analysis of the
efficiency of enzymatic cleaning agents used in dairies). Bachelor thesis,
Helsinki: EVTEK. 58 p. + 9 appendices.
♦ Nilsson, A. (1999). Detektion av Bacillus cereus sporer på yta med hjälp av
DEM, immunofluorescens och QCM-D™ (Detection of Bacillus cereus spores
on surfaces using DEM, immunofluorescence and QCM-DTM). Master thesis
1999:L4, Kalmar. 33 p. + 5 appendices.
3/4
2000
♦ Kontulainen, S. (2000). Termofiilisten haittabakteerien vaikutus jogurtin
kypsymisen hidastumiseen ja valmistuslaitteistojen pintojen peseytyvyyteen
(Effect of harmful thermophilic bacteria in yoghurt fermentation and cleaning
of surfaces in manufacturing equipment). Pro Gradu thesis (University of
Helsinki), EKT-series 1175, Pikku-Sitomo, Helsinki. 89 p.
♦ Heino, A. (2000). Juustomuottien puhtauden tutkiminen pilot- ja prosessi-
mittakaavassa (Cleanliness of cheese moulds in pilot and process scale). Pro
Gradu thesis (University of Helsinki), EKT-series 1182, Pikku-Sitomo,
Helsinki. 117 p.
♦ Aalto, M. (2000). Desinfiointiaineiden teho biofilmimikrobeihin (Disinfectant
efficacy against microbes in biofilm-constructs). Bachelor thesis, Helsinki,
Pikku-Sitomo. 66 p.
♦ Augustin, M. (2000). Rapid and sensitive fluorometric method for evaluation
of the enzymatic cleaning agents against biofilm bacteria. Bachelor thesis,
Pikku-Sitomo, Helsinki. 38 p. + 7 appendices.
3/5
3/6
APPENDIX 4
PROJEKTSAMMANDRAG (SVENSKA)
Forskningsarbetet i projektet P96049 inom ramprogrammet NORDFOOD2
utfördes vid de nordiska forskningsinstituten VTT Bioteknik, MATFORSK och
SIK samt vid Helsingfors och Reykjavik universitet fr.o.m. april 1997 t.o.m.
januari 2000. De deltagande företagen var mejerierna Valio från Finland, Arla från
Sverige och TINE från Norge samt teknokemiföretaget Suomen Unilever Oy
DiversyLever från Finland. Dr. Gun Wirtanen, VTT Bioteknik, koordinerade
projektet. Maija Uusisuo och Oddur Gunnarsson skötte om projektet från Nordisk
industri fonds sida. Experimenten, vilka utfördes i projektet, fokuserde på
mätningsmetoder inom rengöring av öppna och slutna system, t.ex. dim-
desinfektion, ozonering, fotbadshygien, rengöring av ostformar och yoghurtlinjer
samt utveckling av metoder för testning av desinfektionsmedels effekt,
mikrobiologiska resistensfenomen, livscykelanalyser (LCA) och rengörings-
procedurens funktionsduglighet. Nya hygienlösningar baserade på projektresultaten
har införts i mejerierna. Projektresultaten kan summeras enligt följande:
• Det huvudsakliga målet i det svenska projektet var att utveckla och utvärdera
praktiska metoder för mätning av hur effektiv rengörings- och desinfektion-
processen är. Kravet var att metoderna skulle kunna användas på ytor i olika
typers mejeriapparater. Analysmetoder baserade på utsköljning och ingjutning
med trifenyltetrazoliumklorid (TTC) agar fungerade bra för testning av
rengöringen av plast ostformar. En 5-stegs metod för evaluering av rengörings-
och desinficeringsmedel har utvecklats på Arla Foods.
4/2
lämpliga rengöringsmedel för ytor nedsmutsade med yoghurt. Olika
kombinationer av rengöringsmedel testades i pilotskala användande mikrober
isolerade ur yoghurtprosesslinjen samt vidbränd yoghurtmjölk på rostfria
stålytor. Bästa rengöringseffekt på ytor med vidbränd yoghurtmjölk erhölls
med en tvåstegs-rengöring med lut innehållande kelat.
4/3
mätning av väletablerad biofilm, eftersom dessa celler ofta är hårt fästade vid
ytorna och därför inte kan lösgöras genom svabbning.
4/4
SAMMANFATTNING
RENGÖRING AV SLUTNA SYSTEM
Även om ozonet i sig är ett effektivt medel vid desinficering finns det fortfarande
problem med att uppnå tillräckligt höga koncentrationer i CIP-systemet. CIP-
systemet i detta försöket var placerat för långt från ozon-generatorn. Ozonet bröts
ner när det var i kontakt med luft och därmed minskade dess effekt. För att man
skall kunna använda ozon bör man först få fram riktiga, tekniska lösningarna.
4/5
Ett desinficeringsmedel innehållande väteperoxid och perättiksyra var mera
effektivt än det tensid-baserade medlet vid dimning. Nya medel bör testas under
kontrollerade förhållanden med ytprover. Då ytprov utförs är det mycket viktigt att
känna till hur mikroberna uppför sig i olika förhållanden.
Resultaten av de utförda proven visade att ozonet inte var effektivt vid den utförda
ytdesinficeringen. I det fall att utrustningens kapacitet är tillräcklig och ytorna är
fria från organisk smuts kan dimdesinficering användas för avdödning av bakterier
på ytor. Det anbefalles att effektiviteten av desinficeringen övervakas i praktiska
applikationer och att man då använder sig av flera provställen per mätning.
4/6
att utvärdera kvaliteten i tillförselsystem ss. kvaliteten på ånga, vatten och
kemikalier.
UTVÄRDERING AV MILJÖEFFEKTEN
4/7
4/8
APPENDIX 5
PROSJEKTSAMMENDRAG (NORSK)
Forskningsarbeidet i prosjekt P96049 i det andre NORDFOOD programmet ble
utført ved de nordiske forskningsinstituttene VTT Biotecnology, MATFORSK og
SIK samt ved universitetene i Helsinki og Reykjavik fra april 1997 til Januar 2000.
Meieriene som var involvert i prosjektet var Valio Ltd fra Finland, Arla fra Sverige
og TINE fra Norge. I tillegg deltok det teknokjemiske firmaet Suomen Unilever Oy
DiversyLever fra Finland. Dr. Gun Wirtanen, VTT Biotechnology koordinerte
prosjektet. Eksperimentene fokuserte på måle metoder i renhold av åpne og
lukkede systemer, for eksempel tåkelegging, ozonering, fotbad hygiene, vask av
osteformer og yoghurt linjer, utvikling av metoder for å teste effekten av
desinfeksjonsmidler, mikrobielle resistens fenomener, livs-syklus-analyser (LCA)
og evalueringsprosedyrer for funksjonaliteten av vaskeprosedyrer. Nye hygiene
løsninger basert på resultatene i prosjektet har blitt implementert i meieriene.
Funnene kan oppsummeres som følgende:
• Mer miljøvennlige vaske prosedyrer ved bruk av ozonert vann eller enzym
baserte vaskemidler ble utprøvd i et CIP-system. Ved disse metodene kan man
spare energikostnader i tillegg til mindre forurensing av miljø. Lovende
resultater med 90.0–99.9% reduksjon i sporetall ble oppnådd ved en
ozonkonsentrasjon på 0.1–0.3 ppm.
• Tåkelegging med desinfeksjonsmiddel og ozonering av luft er potensielle
metoder for desinfeksjon av luft og overflater. Det var imidlertid problemer
med eliminere bakterier på visse områder i rommet, slik som utsiden av rør,
gummislanger og tak. På disse områdene sitter mikrobene muligens fast i
underlaget og har er naturlig resistens eller utvikler høyere resistens mot
desinfeksjonsmidler. Resultatene vist at ozonering hadde liten effekt på
mikrober på overflater, spesielt når disse er tørket inn. Ytelsen til tåkeleggings-
utstyret må måles jevnlig for å sikre optimal desinfeksjon
5/2
• Det ble utført uttesting av vask av osteformer i pilot- og full skala forsøk.
Strukturen til osteformer i plast er kompleks, med lange, trange, koniske
kanaler. Ultralyd viste seg å være en god metode for å vaske osteformer.
Renheten til osteformene etter vask i pilotskala og under prosessering ble målt
ved ulike metoder og dipslide teknikken var den mest praktiske metoden for å
finne mikrobiell forurensing. I industriell skala var måling av pH beste metode
for å bestemme om vaskeprosedyren fungerte. Kjemisk oksygen behov (COD)
metoden og EDTA målinger var nyttige for å bestemme organisk belastning av
vaskevann.
• I forsøk for å utvikle nye miljøvennlige vaskemidler basert på enzymer ble det
renset proteinase prøver (for eksempel cryotin fra torsk, trypsin fra antarktisk
krill og chymotrypsin fra torsk) på Universitetet på Island. En mikrotiter plate
metode med fluorometriske, kolorimetriske og turbidometriske målinger av
effektiviteten av de enzymatiske vaskemidlene på Bacillus biofilmer ble
utviklet. En metode basert på fluorogeniske redoks indikatorer (for eksempel
resazurin) ble brukt for å evaluere effekt av enzymvask av biofilmer med
melkesyrebakterier, Escherichia coli, og Pseudomonas aeruginosa. Proteinase
prøvene fjernet biofilmer i fravær av melk.
5/3
viste at sensitiviteten for permeabiliserende stoffer varierer meget mellom
arter. Resultatene viste også at sitronsyre var en effektiv permeabilisator mens
natrium sitrat virket dårligere.
KONKLUSJONER
RENGJØRING AV LUKKEDE SYSTEMER
Selv om ozon er betraktet som et effektivt middel for desinfeksjon er det fortsatt
problemer med å få til konsentrasjoner som er høye nok til CIP-systemet. CIP-
systemet i dette forsøket var plassert for langt unna ozon-generatoren. Ozonen ble
dekomponert når den kom i kontakt med luft og dermed ble effektiviteten av
ozonet redusert. For at man skal kunne bruke ozon må de tekniske løsningene på
plass først.
5/4
SAMMENLIKNING AV TESTMETODER FOR EFFEKTIVITET MTP
DESINFEKSJON
5/5
RESISTENSFENOMENER GRUNNET DESINFEKSJON
EVALUERING AV OSTFORMSVASK
Ostformer ble vasket med ulike midler og metoder og deretter analysert vha ulike
metoder. Ingen av testmetodene (protein-kit Check Pro, ATP, svabring, skylling,
UV-bestråling, DEM og TTC-agar) fungerte bra for å detektere smuss. TTC-agaren
og skylling var de metodene som virket best for deteksjon av mikrobiell vekst.
Disse metodene ga sammenliknbare og reproduserbare resultater. I pilot-forsøkene
ga kontakt-agaren de sikreste resultatene. Det er behov for forbedring av metoder
der nivået av organiske rester kan detekteres, slik at vaske-effekt kan evalueres. I
full-skala forsøket viste de mikrobielle testene at rensevannet var kontaminert med
et høyt antall bakterier. Metoden kan altså også brukes for å vurdere kvaliteten på
vann.
MILJØASPEKT
5/6
APPENDIX 6
6/2
valintaan tarvittavat tiedot, vaikkakin arviossa oli tehty useita oletuksia ja
rajoituksia. Tämän tutkimuksen mukaan entsyymipohjainen CIP osoittautui
parhaaksi menetelmäksi, koska siinä käytetyt konsentraatiot olivat pieniä ja
lämpötilat alhaisia.
6/3
proteaasia sisältävät aineet olivat tehokkaita maitolikaa sisältämättömien
biofilmien poistoon.
6/4
teerit. Vetyperoksidipohjaiset desinfiointiaineet tehosivat useimpiin tutkittuihin
mikrobikantoihin.
JOHTOPÄÄTÖKSET
SULJETTUJEN JÄRJESTELMIEN PUHDISTUS
6/5
olevat mikrobit olivat helposti tuhottavissa. Tätä vaihetta hyväksikäyttäen voitai-
siin otsonointi saada tehokkaammaksi. Tutkimusten mukaan on myös tärkeää
käyttää suurta konsentraatiota ja tarpeeksi pitkää vaikutusaikaa.
PINTADESINFIOINTI TEOLLISUUSMITTAKAAVASSA
6/6
näytteestä, viljely huuhtelemalla otetusta näytteestä, visuaalinen tarkastelu UV-
valolla, värjätyn pinnan tutkiminen epifluoresenssimikroskoopilla (DEM) sekä
kontaktiviljely TTC-väriainetta sisältävällä agarilla. Mikään käytetyistä mene-
telmistä ei pystynyt määrittämään pinnalla ollutta likaa hyvin. TTC-agar- ja
huuhtelumenetelmä olivat luotettavimmat menetelmät mikrobikasvun määrittä-
miseen. Näillä menetelmillä saatiin keskenään yhteneviä ja toistettavia tuloksia.
Pilottimittakaavan kokeissa saatiin luotettavimmat tulokset kontaktiagar-
menetelmällä. Jotta varsinaista puhdistuksen tehokkuutta pystyttäisiin arvioimaan,
on tarpeen kehittää menetelmiä orgaanisten jäämätasojen määritykseen.
Teollisuusmittakaavan kokeissa mikrobiologiset testit osoittivat, että huuhteluvesi
oli saastunut ja sisälsi runsaasti bakteereita. Täten todettiin, että tutkittuja
menetelmiä voidaan käyttää myös arvioitaessa ylläpitojärjestelmiin kuuluvia
osioita, kuten pesuaineiden tehokkuuksia.
YMPÄRISTÖVAIKUTUSTEN ARVIOINTI
6/7
6/8
APPENDIX 7
SAMANTEKT (ÍSLENSKA)
Rannsóknarvinnan i verkefninu P96049 í öðrum hluta NORDFOOD verkefnisins
var unnin af háskólum, opinberum stofnunum og fyrirtækjum á Norðurlöndunum.
Aðilarnir sem um ræðir eru VTT Biotechnology, Valio Ltd, Suomen Unilever Oy
DiverseyLever og Háskólinn í Helsinki (Finnlandi), Matforsk og TINE (Noregi),
SIK og ARLA (Sviþjóð) og Háskóli Íslands (Íslandi). Verkefnið stóð yfir frá
apríl 1997 til janúar 2000 og var dr. Gun Wirtanen frá VTT Biotechnology
yfirumsjónarmaður verkefnisins. Yfirstjórn þessa verkefnis að hálfu norræna
Iðnaðarsjóðsins voru Maija Uusisuo og Oddur Gunnarsson. Tilraunir sem voru
framkvæmdar í verkefninu beindust einna helst að því að athuga þær aðferðir sem
notaðar eru við þrif og sótthreinsun á opnum og lokuðum kerfum. Má þar t.d. nefna
þokuúðun, ósoneringu, notkun sótthreinsimotta, hreinsun á gerilsneyðingartækjum,
þróun á aðferðum til að meta virkni sótthreinsiefna, þolni örvera gegn
sótthreinsiefnum, líftímaákvörðun (LCA) og þróun á aðferðum sem meta árangur
þrifa og sótthreinsunar í mjólkursamlögum. Út frá þeim niðurstöðum fengist hafa í
verkefninu hefur verið breytt um aðferðir hvað varðar hreinsunarferli í
mjólkursamlögum. Helstu niðurstöður verkefnisins eru raktar hér á eftir:
• Þokuúðun með mismunandi efnum sem innihalda osón er aðferð sem er oft
notuð til þess að sótthreinsa framleiðslu- og geymslurými. Það er hins vegar
erfitt að ganga úr skugga um að efnið nái til allra svæða eins og t.d. á yfirborð
pípulagna, á gúmmípakkningar og á loft. Á þessum stöðum eru oft bakteríur
sem eru búnar að koma sér fyrir (biofilmur) og eru orðnar þolnar gegn þeim
efnum sem eru notuð. Niðurstöðurnar sýndu að osón er ekki eins virkt gegn
örverum á þurrum stöðum. Einnig þarf að hafa stöðugt eftirlit með virkni
þokuúðunarinnar.
7/2
kom í ljós að CIP-hreinsun með ensímum gaf bestu niðurstöðuna vegna þess að
ensímin eru notuð í mjög litlu magni og við lágt hitastig.
• Við Háskóla Íslands fóru fram tilraunir til þess að þróa ný umhverfisvæn
þvottaefni til að hreinsa próteinmengaða fleti (sýni). Sem dæmi um þau efni
sem athuguð voru má t.d. nefna cryotín og chymotrysín úr þorski og trypsín úr
uppsjávarfiskum. Þróuð var örtítrunaraðferð sem er byggð á flúoriserandi
mælingum, ljósmælingum og þéttnimælingum til þess að athuga virkni þessara
ensíma á Bacillus biofilmur. Aðferð til þess að meta sérhæfni og næmni ensíma
á biofilmur annarra baktería (mjólkursýrubaktería, Eschericia coli, Pseudomonas
aeruginosa) byggðist á flúoriserandi afoxunar indikatorum, eins og t.d.
resazurin. Þeir próteasar sem voru prófaðir reyndust mjög vel til þess að
hreinsa biofilmur sem innihéldu ekki mjólkurleifar.
7/3
• Til þess að meta áhrif eiginleika sótthreinsilausna á P. aeruginosa biofilmur
var s.k ”concentric cylinder reactor (CCR) notaður. Niðurstöðurnar sýndu að
CCR má nota til þess að greina á milli bakteríudrepandi þvottaáhrifa
mismunandi sótthreinsiefna. Til þess að athuga virkni sóthreinsiefnanna má
nota leiðnimælingar til þess að meta hvort yfirborð sé dauðhreinsað vegna þess
að mjög fáar, vel aðlagaðar, hraðvaxandi frumur sem og mikið magn af
frumum sem hafa verið hreinsaðar, breyta leiðninni í vökvanum. Með
litunaraðferðum má fá fram lifandi frumur og heildarfrumufjöldann á
yfirborðsflötum. Ræktun örvera er ekki nægjanlega góð aðferð til þess að mæla
biofilmur sem hafa náð sér vel á strik á yfirborðsflötum vegna þess að
frumurnar geta verið mjög fastar og losna ekki þó að strokusýni sé tekið.
• Mælieiningar í prófi sem byggt er á s.k. ”hydrogeli” var rannsakað með því að
nota mismunandi sótthreinsiefni. Niðurstöðurnar sýndu að 5 klst. ræktunartími
100 l ræktar gaf bestu niðurstöðuna. Virkni sótthreinsiefnanna á biofilmur má
"optimera" með þessari aðferð. Niðurstöðurnar sýndu að Gram neikvæðar
bakteríur hafa meira þol gegn sótthreinsimeðferð en Gram jákvæðar bakteríur.
Sótthreinsiefni sem innihéldu vetnisperoxíð reyndust vera virk gegn flestum
örverum sem voru athugaðar í þessari rannsókn.
Ljóst er að halda verður áfram að þróa aðferðir til þess að finna og greina
örverufræðilega mengun á tækjum sem notuð eru í matvælavinnslu, í hráefninu
sjálfu, í loftinu í vinnslunni, á þeim pakkningum sem notaðar eru og í
lokaafurðunum. Samvinna á þessu sviði mun halda áfram á Norðurlöndunum í
verkefninu ”DairyNET ´Hygiene control in dairy environment” sem er styrkt af
Norræna Iðnaðarsjóðnum (P00027).
7/4
NIÐURSTÖÐUR
ÞVOTTUR/SÓTTHREINSUN Í LOKUÐUM KERFUM (CIP)
Þær tilraunir sem gerðar voru með notkun á ósoni til þvotta og sótthreinsunar í
lokuðum kerfum reyndust ekki sem skyldi hvað varðar fækkun á örverum. Ástæðuna
má trúlega rekja til þess að ekki hafi tekist að dreifa efninu í nægilegum styrkleika í
lokuðu kerfi.
Þrátt fyrir að notkun ósons sé talin afar árangursrík aðferð til sótthreinsunnar þá er enn
nokkuð vandamál að dreifa ozoni í nægilega háum styrkleika í lokuðum kerfum.
Lokaða kerfið, sem notað var í þessari rannsókn, var of langt frá ósontækinu og þ.a.l.
Spilltist ósonið þegar það komst í snertingu við loft og virkni þess minnkaði.
Nauðsynlegt er því að leysa tæknileg atriði í þessu sambandi áður en hægt verður að
nota óson með fullnægjandi árangri.
Það er ekki hægt að nota örverudrepandi lausnarpróf (suspension test) til að bera
saman áhrif mismunandi sótthreinsiefna eða við samanburð á næmum tegundum
baktería. Þær aðferðir sem annars vegar byggjast á því að bera saman áhrif
sótthreinsiefna á óvarðar (exposed) bakteríur sem eru ræktaðar sem biofilmur og hins
vegar bakteríur verndaðar innan í biofilmu gáfu sambærilegar niðurstöður og veittu
jafnframt upplýsingar um örverudrepandi áhrif sótthreinsiefna auk upplýsinga um
víxlverkun sótthreinsiefnanna og grunnmassa biofilmunnar.
7/5
styrk og nægilega langan virknitíma. Þegar notuð voru sótthreinsiefni sem innihéldu
vetnisperoxíð og peredikssýru þá reyndist þokuúðun skila nægilega góðum árangri, en
úðun með sótthreinsiefnum sem innihéldu tensíð skilaði ófullnægjandi árangri.
Nauðsynlegt er að prófa ný efni við stýrðar aðstæður. Slíkar rannsóknir þurfa að
byggja á yfirborðsmælingum. Við framkvæmd yfirborðsmælinga er afar nauðsynlegt
að þekkja vel viðbrögð og hegðun örvera við mismunandi aðstæður og í ólíku
umhverfi.
YFIRBORÐSSÓTTHREINSUN Í IÐNAÐARSKALA
Árangur af sótthreinsun yfirborðs með ósoni skilar ekki góðum árangri. Ef geta
tækis er nægjanleg og yfirborðin hrein er hægt að nota sótthreinsun með þokuúðun
til að eyða bakteríum af yfirborðum. Ráðlagt er að fylgjast vel með árangri
sótthreinsunnarinnar við raunaðstæður og að taka sýni á mörgum mismunandi
stöðum.
Gró og nokkrar ónæmar bakteríur geta lifað af sótthreinsun þar sem þær hafa byggt
upp þol gegn sótthreinsiefnum. Sumar bakteríur hafa eðlislægt þol en aðrar hafa
aðlagað sig að þeim sótthreinsiefnum sem notuð eru. Eftir því var tekið að
hefðbundnar þvotta- og sótthreinsiaðferðir dugðu oft ekki til að eyða gróum og
bakteríum sem höfðu myndað biofilmu.
Þrif á ostaformum voru framkvæmd þar sem notast var við ýmsar tegundir
þvottaefna og þvottaaðferða og árangurinn síðan metinn með mismunandi
aðferðum. Engin þeirra aðferða sem notuð var (prótein-próf (Check Pro), ATP-
ljósmæling, penslun, úthreinsun, UV-lýsing, DEM og TTC agar) skilaði góðum
árangri við mat á óhreinindum. TTC-aðferð og s.k. úrheinsun reyndust þó skila
áreiðanlegustu niðurstöðunum við greiningu á örverum. Þessar aðferðir skila
niðurstöðum sem eru bæði samanburðarhæfar og hægt er að endurtaka þær aftur og
aftur. Í tilraunum þar sem líkt er eftir raunverulegum aðstæðum (pilot-scale)
7/6
reyndist snertiskálaaðferðin gefa áreiðanlegustu niðurstöðurnar. Nauðsynlegt er að
bæta aðferðir við greiningu leifa af lífrænum uppruna þannig að hægt sé að
sannreyna áhrif þrifa til hins ýtrasta. Í tilraunum í fullum skala sýndu öruveru-
mælingar að mikið fannst af bakteríum í því vatni sem notað var til skolunnar.
Aðferðirnar má þ.a.l. einnig nota til að meta breytur í dreifikerfum eins og t.d. áhrif
hreinsiefna.
Í kafla 4.2 er lýst hagnýtri nálgun við mat á virkni þrifa. Eftirfarandi aðgerðir eru
ráðlagðar, en þær eru mat á skriflegum upplýsingum, endurskoðun birgja, mat á
virkum eiginleikum, hagnýtt mat og lokaskoðun.
7/7
7/8
APPENDIX 8
Methods and results: Bacterial strains were isolated from disinfecting footbaths
containing TEGO 103G (amphoteric disinfectant) or TP-99 (alkylaminoacetate-
based disinfectant) in five out of six dairy factories. Fourteen strains identified as
Cedecea spp. by their fatty acid composition were further characterised. The
reactions in the Rapid ID 32 E API analysis and 16S-rDNA-sequensing showed
that all strains were Serratia marcescens. In contrary to Ser. marcescens ATCC
13880 the isolates from disinfecting footbaths were not killed (<5 log reduction) by
the recommended in-use concentration of TEGO 103G, TEGO 51 or benzalkonium
chloride. Survival and multiplication in tap water with in-use concentration of
TEGO103G was demonstrated for one of the strains. All strains were killed by the
in-use concentrations of commercial disinfectants based on peracetic acid,
hypochlorite, quaternary ammonium compounds and alkyl aminoacetate (TP-99).
There were no indications of cross-resistance between disinfectants and antibiotics.
Poster presentation Ultrasound cleaning in cheese mold hygiene based on the Pro
Gradu thesis by Antti Heino and presented at the IAFP Annual meeting in August
2000.
APPENDIX 10
The cleaning of the plastic cheese moulds is a challenging task. The structure of the
parts used in these utensils is often complex, with long, narrow conical channels,
which are hard to clean with conventional cleaning procedures. Furthermore the
cleaning procedure must be performed quickly, efficiently, economically and
environmentally friendly without harming the surface material. If harmful microbes
remain in the channels after cleaning they may be transferred into the product
causing quality and shelflife problems. An ultrasonic washing system based on
cavitation has been applied in automatized cheese mould cleaning systems in a
Finnish cheesery. The aim of this study was to compare various parameters in the
ultrasound cleaning procedure using different detection methods to optimise the
washing procedure. Parameters in the test series were type and concentration of
cleaning agents and ultrasound frequency. The results showed that it is important to
know the principles of the measuring method used to be able to interpret the results
correctly. The organic load determinations for the cleaning agent experiments were
performed using the EDTA (ethylenediamine tetraacetic acid) and COD (chemical
oxygen demand) methods. The organic load in the cleaning liquid affected the
efficiency of the various cleaning agents used. Experiments with cleaning agent
concentrations of 0.5%, 1.0% and 1.5% showed that the concentration should be
maintained at least at a level of 1.0%. An increase in the ultrasound intensity from
460 W to 740 W enhanced cleaning especially in experiments using artificially
aged cheese moulds.
APPENDIX 11
Figures A-B. Spores of Bacillus cereus 229 stained for 2 min with 0.1% erytrosine B (left) and
0.1% acridine yellow (right). The magnification used 1000x.
Figures C-D. Spores of Bacillus cereus 229 stained for 5 min with 0.1% auramine O (left) and
for 15 min with 0.42% trans-4-(p-N,N-dimethylaminostyryl)-N-butoxylcarbonylmethylpyri-
dinium bromide (right). The magnification used 1000x.
Figures E. Spores of Bacillus cereus 229 stained for 2 min with 0.1% acridine orange. The
magnification used 1000x.
Figures F-G. Spores of Bacillus cereus 229 in milk soil (diluted 1:10) stained for 2 min with
0.1% acridine yellow (left) and for 5 min with 0.1% auramine O (right). The magnification
used 1000x.
Figures H-I. Spores of Bacillus cereus 229 in milk soil (diluted 1:10) stained for 15 min with
0.42% trans-4-(p-N,N-dimethylaminostyryl)-N-butoxylcarbonylmethylpyridinium bromide (left)
and for 2 min with 0.1% acridine orange (right). The magnification used 1000x.
15/2
Published by Series title, number and
report code of publication
Vuorimiehentie 5, P.O.Box 2000, FIN–02044 VTT, Finland
Phone internat. +358 9 4561 VTT Publications 481
Fax +358 9 456 4374 VTT–PUBS–481
Author(s)
Wirtanen, Gun, Langsrud, Solveig, Salo, Satu, Olofson, Ulla, Alnås, Harriet, Neuman, Monika,
Homleid, Jens Petter & Mattila-Sandholm, Tiina
Title
Evaluation of sanitation procedures for use in dairies
Abstract
The research work for project P96049 in the second NORDFOOD programme was
mainly carried out at VTT Biotechnology, Matforsk and SIK together with
representatives from the Nordic dairies Valio Ltd., Arla and TINE as well as the
technochemical company Suomen Unilever Oy DiverseyLever. The senior advisors at
Nordic Industrial Fund involved in the project were Maija Uusisuo and Oddur
Gunnarsson. The experiments carried out in the project focused on monitoring methods in
sanitation of open and closed systems e.g. fogging, ozonation, footbath hygiene, cleaning
of cheese moulds and yoghurt pasteurizers, development of testing procedures for
measuring disinfectant efficacy, microbial resistance phenomena against disinfectants,
life-cycle assessment and an evaluation procedure for the functionality of the cleaning
procedures. New procedures in hygiene have been implemented in dairies based on the
results. Development of detection and identification methods for assessing microbial
contaminants on or in process equipment, raw material, process air, packaging material
and final products is continued in the Nordic dairy hygiene network project DairyNET –
Hygiene control in dairy environment, which has partners from all Nordic countries and
which is partly funded by the Nordic Industrial Fund (P00027). The contacts between
industrial personnel and researchers dealing with hygiene questions in the Nordic
countries, which have been built up in the 2 previous NordFood programmes (1994–
2000), are thus continued.
Keywords
NORDFOOD, dairies, sanitation, hygiene, microbes, detection, isolation, disinfection, cleaning,
life-cycle analysis, environmental assessment
Activity unit
VTT Biotechnology, Tietotie 2, P.O. Box 1500, FIN–02044 VTT, Finland
ISBN Project number
951–38–6017–5 (soft back ed.) B1SU00200
951–38–6018–3 (URL:http://www.inf.vtt.fi/pdf/ )
Date Language Pages Price
November 2002 English 96 p. + app. 43 p. C
Name of project Commissioned by
DairyNET – Hygiene control in dairy Nordic Industrial Fund
environment (P00027)
Series title and ISSN Sold by
VTT Publications VTT Information Service
1235–0621 (soft back ed.) P.O.Box 2000, FIN–02044 VTT, Finland
1455–0849 (URL: http://www.inf.vtt.fi/pdf/) Phone internat. +358 9 456 4404
Fax +358 9 456 4374
ESPOO 2002 VTT PUBLICATIONS 481
Tätä julkaisua myy Denna publikation säljs av This publication is available from
VTT TIETOPALVELU VTT INFORMATIONSTJÄNST VTT INFORMATION SERVICE
PL 2000 PB 2000 P.O.Box 2000
02044 VTT 02044 VTT FIN–02044 VTT, Finland
Puh. (09) 456 4404 Tel. (09) 456 4404 Phone internat. +358 9 456 4404
Faksi (09) 456 4374 Fax (09) 456 4374 Fax +358 9 456 4374