A Survey of Ph and Water Activity Levels in Processed Salamis and Sausages in Metro Toronto

1007

Research Note

A Survey of pH and Water Activity Levels in Processed Salamis and Sausages in Metro Toronto Downloaded from http://meridian.allenpress.com/jfp/article-pdf/59/9/1007/1666213/0362-028x-59_9_1007.pdf by guest on 30 September 2021 MARILYN B. LEE1* and STEVE STYLIADIS2

[School of Environmental Health, Ryerson Polytechnic University, 350 Victoria St., Toronto, Ontario M5B 2K3; and 2Food SafetylWater Quality Unit, Disease Control Service, Public Health Branch, Ontario Ministry of Health, 5700 Yonge St., 8th Floor, North York, Ontario M2M 4K5, Canada

(MS# 95-225: Received 14 September 1995/Accepted 6 February 1996)

ABSTRACT then there is a potential for food borne disease, since pathogenic organisms such as salmonellae, staphylococci, or Eighty-four processed (various combinations of cold-cured, pathogenic strains of Escherichia coli may survive and even cold-smoked, fermented, hot-smoked, cooked, dried) salami and grow if processed meat is left unrefrigerated. Numerous sausage samples from Metro Toronto were analyzed for pH and outbreaks have been documented as originating from con- water activity (a ). Results from 62 non-heat-treated meat products w sumption of contaminated dry fermented products (2, 3, 5-7, were compared with previously published criteria of pH and a for w 10, 17), including a recent (1994) outbreak of 20 laboratory- safety of shelf-stable products that require no refrigeration or thermal processing. Comparisons revealed that 31 to 82% of confirmed cases of diarrhea caused by E. coli 0157:H7 samples analyzed met the previously published criteria, which acquired from consumption of a dry-cured salami (2). Meats varied widely. Additionally, it was found that non-heat-treated that have been cooked or hot smoked to sufficient tempera- products made on-site at small deli-type establishments were more tures to inactivate pathogens may also present a risk if not likely to be left constantly unrefrigerated (chi-square, P = 0.05) properly refrigerated at <5°C. Such products may have been than those made off-site at larger commercial plants. Those exposed to postcooking contaminants which may survive products made on-site had a mean pH of 5.8 and a mean aw of 0.90, and even multiply. The main purpose of this study was to while products made off-site had lower mean values of pH (5.1) analyze non-heat-treated meat products for pH and/or aw and and aw (0.87), These differences were both statistically significant to compare these results with previously published criteria (Student's t test), for pH (P = .000005) and a (P = .03). It is w for safety of shelf-stable products. suggested that larger commercial establishments may recognize the value of lowering the pH and aw and have equipment for such measurements. MATERIALS AND METHODS

Key words: pH, a , salamis, dry sausages, processed meats, w Eighty-four processed meat products, representing 30 brands fermented meats, dried meats from 16 meat delicatessens, were purchased in metropolitan Toronto, Ontario, from August 1994 to January 1995. Premises selected were small « 150 m2) and processed on-site at least one The recent immigration to Ontario and the change in ready-to-eat deli meat product. A minimum of 200 g of each meat eating habits of Ontarians have put pressure on the meat product was purchased. Other processed meat products prepared by industry to develop innovative products in order to satisfy larger, commercial manufacturers (called "off-site") were pur- the taste buds of the consumers of this large multinational chased when available. Evaluation in the laboratory was completed diverse community. Ready-to-eat dried meat products, which within 2 h of purchase. may or may not be fermented, present a unique concern The aw values were measured with an Aqualab CX-2 (Deca- since they are prepared from raw meat which has not been gon Devices, Pullman, WA) using 2.5 to 4.0 g samples. A minimum cooked. Control of foodborne pathogens is achieved during of four readings were taken from each product. The accuracy of the processing by adding curing agents to ground meat and/or aw measurement device was checked before and after use against a saturated KCl solution, aw 0.843 at 24°C and a saturated NaCI by dehydration to reduce water activity (aw)' In the case of properly fermented products, production of lactic acid solution, aw 0.752 at 24°C. All measurements for KCl were ::'::0.003 and for NaCI ::'::0.002at 21.5 to 24.0°C and were within published reduces the pH, resulting in the control of pathogens. water activity limits for saturated solutions in this temperature If a suitable reduction in pH and/or a is not attained, w range (8). A Sargent-Welch Model 4090 pH meter with a glass electrode probe was used to assess the acidity of each fermented meat * Author for correspondence. Tel: 416-979-5154; Fax: 416-979-5377. sample. Samples were weighed to 25 g, placed with 50 ml of 1008 LEE AND STYLIADIS deionized water, and blended for 30 s in a Waring blender set on TABLE 1. Survey results of cold-cured and cold-smoked salamis, liquify. The pH meter was calibrated before use against pH sausages, or ham standards of 7.0 and 5.0. The probe was washed with detergent and Manu- pH a water and thoroughly rinsed with tap water between each reading. w factured Refrig- mean mean Two measurements of pH were recorded for each product. Branda Type of productb on-siteC eratedd (n = 2) (n = 4) Statistical analyses were performed using Epi-Info version 5.0 from the Epidemiology Program Office, Centers for Disease 1 farmer salami no no 5.4 0.89 Control and Prevention, Atlanta, GA. 1 pepperseed salami no no 4.6 0.84 1 pepperseed salami no no 4.9 0.78 RESULTS 1 summer salami no no 5.5 0.82 2 Portuguese salami yes no 5.6 0.95 3 Genoa style beef salami no no 5.1 0.88 Tables 1 to 3 show pH and aw values according to the 4 pick salami yes 6.1 0.82 type of processing. For cold-cured and cold-smoked salamis, no 4 spicy paprika salami yes no 5.4 0.84 sausages, or ham (Table 1) pH readings ranged 4.3 to 6.4, Downloaded from http://meridian.allenpress.com/jfp/article-pdf/59/9/1007/1666213/0362-028x-59_9_1007.pdf by guest on 30 September 2021 5 chabi sausage yes no 5.7 0.94 while aw values ranged from 0.65 to 0.97. Table 2 shows 6 domace smoked sausage yes no 5.0 0.86 dry-cured Italian style salamis (not smoked) with pH 7 Italian salami no no 5.8 0.86 readings ranging from 4.8 to 5.6 and aw from 0.78 to 0.92. 8 deli sticks yes no 5.7 0.87 All Italian style salamis were made off-site. Table 3 shows 8 landjaegar salami yes no 5.8 0.87 cured and hot-smoked salamis or sausages with pH readings 8 smoked ham yes yes 6.0 0.93 ranging between 4.8 to 7.0 and aw from 0.80 to 0.98. In 9 Hungarian salami no no 6.3 0.94 contrast to the Italian salamis, all of these products were 9 Hungarian salami no yes 6.4 0.95 made on-site. Three cured and hot-smoked salamis and 10 Portuguese salami yes no 4.8 0.88 sausages were also evaluated. Their pHs ranged from 6.1 to 10 Portuguese salami yes no 5.5 0.95 11 Portuguese salami yes yes 5.6 0.97 6.4 while aw values ranged from 0.95 to 0.98. All 3 products 12 beef salami no yes 4.6 0.93 were made on-site and were not being refrigerated. 12 German style salami no yes 6.3 0.93 Of the 84 products analyzed (Tables 1 to 3, plus the 3 12 Hungarian salami no no 6.2 0.95 cooked and/or hot smoked sausages and salamis) 60 were 13 pick salami no yes 6.3 0.85 being continuously stored at room temperature (about 22 to 14 German salami no no 4.6 0.89 25°C), usually hanging from a hook against the wall, while 14 gypsy salami no yes 4.7 0.90 the other 24 were being stored in a refrigerated «5°C) 14 Jugoslavian salami no no 4.9 0.90 display case. The non-heat-treated products, made on-site at 14 Jugoslavian salami no no 4.9 0.65 small deli-type establishments, were more likely to be left 14 kaiser salami no no 4.3 0.93 constantly unrefrigerated (chi-square, P = 0.05) compared 14 Mailander salami no no 4.9 0.69 to those made off-site at larger commercial plants. 14 old forest salami no no 4.6 0.79 14 paparika salami no yes 4.9 0.89 Those products made on-site had a mean value of pH of 14 szegedi no no 5.1 0.83 5.8 and a of 0.90, while those products made off-site had a w 14 szegedi no no 5.5 0.79 lower mean value of pH of 5.1 and aw of 0.87. These 14 Yugoslavian salami no yes 5.1 0.86 differences were both statistically significant (student's t 14 Yugoslavian salami no yes 5.0 0.88 test), for pH (P = .000005) and aw (P = .03). 15 Genoa salami no yes 4.9 0.87 16 dry salami no no 4.5 0.90 DISCUSSION 17 csabai yes no 4.9 0.89 18 original Hungarian salami no no 5.2 0.83 Determining the safety of a product at endpoint rather 19 camping salami no yes 5.0 0.91 19 csabai yes 5.3 0.87 than by analysis of the process has limitations, since earlier no 20 German style salami no yes 4.7 0.93 abuse of the product may go undetected. For example, if an 20 German style salami no yes 4.6 0.95 initially high inoculum of Staphylococcus aureus contami- 20 Hungarian salami no yes 4.8 0.95 nates raw meat, growth and release of enterotoxin may 20 Hungarian salami no yes 4.7 0.96 follow. After fermentation and drying, the product may have 20 kaiser salami no yes 4.7 0.94 an acceptable aw and pH and low numbers of S. aureus, but 20 Mexican style salami no yes 4.8 0.90 the preformed enterotoxin remains in the product. Con- 20 mustardseed salami no yes 4.9 0.90 versely, a product for sale with high aw and pH suggests the 20 pepper salami no no 4.4 0.77 possibility of an initial faulty process. If such products are a Every brand was assigned a number. Occasionally the same type temperature abused, pathogenic microorganisms, if present, of product of the same brand was purchased at another establish- may grow. ment: hence the replicates. There is much debate over the precise pH and/or a w b Names and spellings reflect the grocery labels. levels required for a shelf-stable meat product which has not C Some products were manufactured on-site (yes) where purchased undergone heat processing. Many published criteria suggest while other products were manufactured off-site (no) where differing threshold levels of pH and/or aw for safety. For purchased. example, Ledward (12) suggests an ~ :S 0.85 or a pH < 5 as d Refrigeration refers to continuous storage at <5°C. A SURVEYOF PROCESSEDMEATPRODUCTS 1009

TABLE 2. Survey results of dry cured Italian style salamis TABLE 3. Survey results of cured and hot-smoked salamis (not smoked) and sausages

pH aw Manu- pH aw Typeof Manufactured Refrig- mean mean factured Refrig- mean mean Branda productb on-siteC eratedd (n = 2) (n = 4) Brand Typeofproducta on-siteb eratedC (n=2) (n=4)

21 Italian salami no no 5.2 0.86 2 Portuguese sausage, dry yes no 6.2 0.91 21 salametti no no 5.5 0.90 2 Portuguese sausage, fresh yes no 6.3 0.95 21 sopprossata no no 5.4 0.86 5 dry drzewnia yes no 6.1 0.81 22 Calabrese salami no yes 5.4 0.87 5 dry kabanossy yes no 6.2 0.95 22 suppressata no yes 5.6 0.85 5 krakowska yes no 6.2 0.87 23 Calabrese salami no no 5.2 0.88 5 dry zyriecka yes no 4.9 0.83 23 Genoa salami no yes 4.9 0.92 6 sremski hot smoked sausage yes no 5.9 0.96 23 Genoa salami no no 4.9 0.88 8 kabanosse yes no 5.9 0.87 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/59/9/1007/1666213/0362-028x-59_9_1007.pdf by guest on 30 September 2021 24 Italian salami no yes 5.2 0.78 28 dry mysliwsa yes no 5.6 0.96 25 Genoa salami no yes 5.2 0.78 28 goralska yes no 6.3 0.87 26 Genoa salami no no 4.8 0.88 28 kabanosy yes no 6.0 0.81 27 casalingo salami no no 5.0 0.83 28 kabanosy dry yes no 6.2 0.98 27 Genoa salami no no 4.8 0.86 28 krakowska yes no 6.1 0.94 28 krawkowska dry yes no 6.0 0.81 a Every brand was assigned a number. Occasionally the same type 28 wiejskha yes no 6.8 0.91 of product of the same brand was purchased at more than one 29 kabanosy yes no 6.4 0.96 establishment, hence the replicates. 29 dry krakowska yes no 7.0 0.94 b Names and spellings reflect the grocery labels. 29 pepperoni yes no 4.9 0.80 C Some products were manufactured on-site (yes) where purchased 29 Polish sausage yes no 4.8 0.80 while other products were manufactured off-site (no) where purchased. a Names and spellings reflect the grocery label. d Refrigeration refers to continuous storage at <5°C. b Some products were manufactured on-site (yes) where purchased while other products were manufactured off-site (no) where purchased. the criteria while Canadian Federal Guidelines (1) suggest C Refrigeration refers to continual storage at <5°C. an aw ::; 0.90 and a pH < 5.4. One wants to balance safety with unnecessarily rejecting products because the criteria are growth, Noterman and Heuvelman's (15) criteria may be too too restrictive. restrictive. However, Holley et al. (11) in sampling Italian Smith et al. (16) showed that trichinae larvae are not dry sausages at two plants in Canada found toxigenic viable in the preparation of dry cured pork products at aw < staphylococci « 1,OOO/g of meat) in both Genoa and

0.92 and pH < 5.3. In the present study, only 50% of dried salametti sausages in all stages of production. The lowest aw and/or fermented pork products met these criteria (Table 4). and pH of products for retail that they found was 0.837 and However, if pork is frozen at - 20°C for 1 to 2 weeks 4.75, respectively. according to thickness prior to processing or during process- Other criteria may be too permissive. Canadian Federal ing, or if 3.3% salt is added and subsequent curing is carried Guidelines suggest that shelf-stable products have an aw ::; out in accordance with Schedule "A" of the Canada Meat Inspection Act (4), trichinae larvae are inactivated and there TABLE 4. Percentages of cured and cold-smoked or dry is little risk of trichinosis regardless of final a or pH. w and/or fermented meat (non-heat-treated) products attaining Messier et al. (14) studied the survival of Salmonella in various published criteria for a shelf-stable product not Genoa salami and determined that salmonellae did not requiring refrigeration survive when the aw was <0.96 and pH was <4.84. Glass et al. (9) found that Escherichia coli 0157:H7 inoculated into a % incurrentstudy Author(reference) Criteria thatmetcriteria sausage batter and fermented to pH 4.8 did survive subsequent drying and storage. However, criteria for the safety of Smith et al.a (16) aw < 0.92 and 30/60 = 50% processed meats are generally based on the more restrictive pH < 5.3 growth and enterotoxin production requirements of S. au- Notermans & aw ::; 0.85 or pH < 4.6 19/62 = 31% reus compared to other bacteria or survival of trichinae Heuvelman (15) larvae. Canadian Federal aw ::; 0.90 and 31/62 = 50% Notermans and Heuvelman (15) observed that S. aureus Guidelines (l) pH < 5.4 Ledward (12) aw ::; 0.85 or pH < 5 37/62 = 60% growth in brain heart infusion broth did not occur at an aw ::; a ::; 0.95 and 51/62 = 82% 0.85 or at a pH < 4.6. Using these criteria to assess Leistner (13) w pH::; 5.2 or cold-processed meats in this study would result in an a ::; 0.91 or acceptance rate of only 31%. Since curing agents, such as w pH::; 5.0 sodium nitrates and/or nitrites, smoking, and competitive bacteria have a synergistic effect in inhibiting microbial a Criteria only apply to pork products. 1010 LEE AND STYLIADIS

0.90 and a pH < 5.4 before release (1). Ledward (12) has 2. Alexander, E. R., J. Boase, M. Davis, L. Kirchner, C. Osaki, and T. suggested a aw ::; 0.85 as being satisfactory or a pH < 5.0. In Tanino. ]995. Escherichia coli 0157:H7 outbreak linked to commer- cially distributed dry-cured salami-Washington and California, this study 60% of dried and/or fermented products would be 1994. Morbid. Mortal. Weekly Rep. 44:157-160. considered safe according Ledward's criteria. Leistner (13) 3. Aoust, J. Y., and A. Evans. 1978. Suspect case of human salmonellosis has suggested criteria for shelf-stable meat products be an from a dry cured meat product-Ontario. Can. Dis. Weekly Rep. aw ::; 0.95, low enough to inhibit the growth of Clostridium 4:27-28. 4. Canada Meat Inspection Act. 1990. R.S., C.M-7 amended 1976-77. botulinum, salmonellae, and spoilage organisms, coupled C.28 Meat inspection regulation. Schedule A. Queens Printer for with achieving a pH ::; 5.2. Leistner (13) has also suggested Canada, Ottawa, Ontario. alternative criteria of aw ::; 0.91 or pH::; 5.0. Neither the aw 5. Center for Disease Control. 1975. Staphylococcal food poisoning or pH alone would inhibit S. aureus growth. Using Leist- associated with Italian dry salami. Morbid. Mortal. Weekly Rep. 24:374,379. ner's criteria 82% of meats sampled would be considered 6. Center for Disease Control. 1979. Staphylococcal food poisoning safe in this study. associated with Genoa and hard salami-United States. Morbid. Mortal. Weekly Rep. 29: 179-180. The criteria against which non-heat-treated, processed Downloaded from http://meridian.allenpress.com/jfp/article-pdf/59/9/1007/1666213/0362-028x-59_9_1007.pdf by guest on 30 September 2021 meats are evaluated for safety need to reflect the ability of 7. Cowden, J. M., M. O'Mahony, C. L. R. Bartlett, B. Rana, B. Smyth, D. Lynch, H. Tillett, L. Ward, D. Roberts, R. J. Gilbert, A. C. Staphylococcus aureus to grow. Values of awand pH, preserva- Baird-Parker, and D. C. Kilsby. 1989. A national outbreak of tives, smoking, and various by-products produced by the lactoba- Salmonella typhimurium DT 124 caused by contaminated salami cilli and other fermenting microorganisms have a synergistic sticks. Epidemiol. Infect. 103:219-225. effect in inhibiting microbial growth. Further studies exam- 8. Decagon Devices. 1994. Aqualab Operator's Manual. Decagon De- vices, Pullman, Washington. ining the survival of S. aureus inoculated into various types 9. Glass, K. A., J. M. Loeffelholz, J. P. Ford, and M. P. Doyle. 1992. Fate of salamis may better clarify criteria for safety. of Escherichia coli 0157:H7 as affected by pH or sodium chloride and Of the 62 non-heat-treated products (Tables 1 and 2) in fermented, dry sausage. Appl. Envir. Microbial. 58:2513-6. those made on-site were more likely to be left constantly 10. Harbottle, J. E. 1980. An unusual outbreak of staphylococcal food poisoning associated with fermented salami-United States. CDC unrefrigerated (chi-square, P = 0.05) posing a risk to the Vet. Notes April: 1-2 .. consumer if inadequately processed or contaminated postpro- 11. Holley, R. A.,A. M. Lammerding, and F. Tittiger. 1988. Microbiologi- cess. Those products made on-site also had higher pH and aw cal safety of traditional and starter-mediated processes for the values than those made off-site at larger commercial estab- manufacture of Italian dry sausage. Int. J. Food Microbial. 7:49-62. 12. Ledward, D. A. 1985. Novel intermediate moisture meat products, p. lishments, suggesting that these establishments may recog- 447-450. In D. Simatos and J. L. Multon (ed.), Properties of water in nize the value oflowering the pH and awand have equipment foods. Martinus Nijhoff Publishers, Dordrecht, Netherlands. for such measurements. 13. Leistner, L. 1978. The microbiology of ready-to-serve foods. Fleischwirtschaft 58:2008-2011. 14. Messier, S., H. J. Smith, and F. Tittiger. 1989. Survival of Salmonella ACKNOWLEDGMENTS typhimurium and Staphylococcus aureus in Genoa salami of varying salt concentrations. Can. J. Vet. Res. 53:84-86. The authors would like to thank the Office of Research Services and 15. Notermans, S., and C. J. Heuvelman. 1983. Combined effect of water the Faculty of Community Services, Research and Scholarship Committee activity, pH and sub-optimal temperature on growth and enterotoxin at Ryerson Polytechnic University for financial support for this project. production of Staphylococcus aureus. J. Food Sci. 48:1832-1836. 16. Smith, H. J., S. Messier, and F. Tittiger. 1989. Destruction of Trichinella spiralis during the preparation of "dry cured" pork REFERENCES products, proscuitto, proscuittini and Genoa salami. Can J. Vet. Res. 53:80-83. I. Agriculture Canada. 1983. Meat hygiene manual, curing 4.9.3. Food 17. Taplin, J. 1982. Salmonella newport outbreak-Victoria. Commun. Production and Inspection Branch, Ottawa, Ontario. Dis. Intell. 1:3-6.