Preventing the transfer of allergens from one food to another via food contact surfaces in retail food environments is an important aspect of retail food safety. Existing recommendations for wiping and cleaning food contact surfaces is mainly focused on preventing microorganisms, such as bacteria and viruses, from contaminating foods. The effectiveness of these wiping and cleaning recommendations for preventing the transfer of food allergens in retail and food service establishments remains unclear. This project investigated (i) allergen removal from surfaces by wiping with paper wipes, terry cloth, and alcohol quaternary ammonium chloride (quat) sanitizing wipes; (ii) cleaning of allergen-contaminated surfaces by using a wash–rinse–sanitize–air dry procedure; and (iii) allergen transfer from contaminated wipes to multiple surfaces. Food contact surfaces (stainless steel, textured plastic, and maple wood) were contaminated with peanut-, milk- and egg-containing foods and subjected to various wiping and cleaning procedures. For transfer experiments, dry paper wipes or wet cloths contaminated with allergenic foods were wiped on four surfaces of the same composition. Allergen-specific lateral flow devices were used to detect the presence of allergen residues on wiped or cleaned surfaces. Although dry wipes and cloths were not effective for removing allergenic foods, terry cloth presoaked in water or sanitizer solution, use of multiple quat wipes, and the wash–rinse–sanitize–air dry procedure were effective in allergen removal from surfaces. Allergens present on dry wipes were transferred to wiped surfaces. In contrast, minimal or no allergen transfer to surfaces was found when allergen-contaminated terry cloth was submerged in sanitizer solution prior to wiping surfaces. The full cleaning method (wash–rinse–sanitize–air dry) and soaking the terry cloth in sanitizer solution prior to wiping were effective at allergen removal and minimizing allergen transfer.
Wet cloths and wipes were more effective in allergen removal from surfaces than dry wipes.
Prescraping food from surfaces prior to full cleaning aided allergen removal.
Cloth storage in sanitizer solution minimized allergen transfer between surfaces.
Allergens were difficult to remove from a textured plastic surface.
The prevalence of food allergies among the U.S. population is estimated between 3 to 4%, with evidence of food allergies in children as high as 8% (1, 5, 8, 17, 18). Allergic reactions to foods are the most common cause of anaphylaxis reported in the community (5). With more than 54% of food expenditures in 2018 attributed to food purchases away from home, there is a need for evaluations of effective allergen control procedures in various food establishments to protect food-allergic consumers (20).
Recommendations for ensuring the safety and protection of food prepared in retail and food service establishments are described in the U.S. Food and Drug Administration (FDA) Food Code (23). Most state, local, tribal, and territorial regulatory agencies have adopted some edition of the FDA Food Code (hereafter “Food Code”), which is updated every 4 years by the FDA's Retail Food Protection Staff. Although many of the provisions in the Food Code were originally developed to reduce microbial risks associated with foods, the effectiveness of these practices for preventing allergen cross-contact remains unclear. The definition of major food allergens contained in the Food Allergen Labeling and Consumer Protection Act of 2004 (22) was added to the 2005 edition of the Food Code. The updated 2009 Food Code further specified that food allergy awareness must be part of the food safety training duties of the person in charge of the establishment. Additionally, the 2013 Food Code amended the cleaning and sanitizing frequency for food contact surfaces or utensils that are in contact with raw animal food that is a major food allergen, such as fish, followed by other types of raw animal foods. The 2015 supplement to the 2013 Food Code further specified that employees must be properly trained in food safety, including food allergy awareness, as it relates to assigned duties. Although recommendations are provided in Chapters 3 (Subpart 3-304) and 4 (Subparts 4-301, 4-501, 4-603, 4-703) of the 2017 edition of the Food Code (23) for manual warewashing or full cleaning and use limitations for wiping cloths, little information exists on whether they are effective at preventing allergen transfer, because these recommendations were originally developed to reduce microbial contamination risk (19, 27).
Published information on the effectiveness of cleaning and wiping procedures used in retail and food service establishments for allergen control on food contact surfaces is scarce. Previous literature reports mostly focused on peanut distribution in different environments, such as the home, school, and hospitals or investigated peanut removal from hands or surfaces by using common cleaning agents or household or hospital wipes (6, 14, 26). One of the few surveys on the occurrence of milk, egg, and gluten on food contact surfaces in school cafeterias was conducted by Ortiz et al. (13). This research team determined the presence of milk, egg, and gluten on food contact surfaces and utensils used in school cafeterias in Spain and documented the percentage of positive results by allergen and general or exclusive use of surfaces and utensils.
Several publications on cleaning and other control strategies for preventing allergen cross-contact in a food manufacturing environment highlighted dry and wet cleaning methods along with indirect (visually clean) and direct (allergen-specific tests) validation and verification procedures when developing an effective allergen control program (9, 15, 24). Additionally, the Food Code, which provides recommendations for ensuring the safety and protection of food prepared in retail and food service establishments (23), also provides some details about the cleaning of food contact surfaces, although these were originally focused to reduce microbial risks associated with foods. Although there are differences in the procedures used for allergen removal and cleaning in industrial food manufacturing operations compared with retail and food service operations, the factors influencing allergen removal are similar. Parameters that influence allergen removal include the nature of the allergenic food matrix (dry powder, wet, paste, or sticky, and high fat), allergen load applied to a surface, food contact material composition, surface characteristics (smooth, textured, or porous), and the type of wipe used in allergen removal (16). The complex set of factors that influence allergen removal, combined with the reality that staff in a retail food setting often rely on speed and efficiency with regard to wiping and cleaning surfaces, can make allergen control in food establishments difficult.
The three primary objectives of this study were to investigate (i) the effectiveness of wiping on the removal of peanut, egg, and milk allergen from stainless steel (SS), textured polyethylene plastic, and maple hardwood surfaces; (ii) the impact of a manual wash–rinse–sanitize–air dry full cleaning method on allergen removal from allergen-contaminated surfaces; and (iii) the extent of allergen transfer to surfaces when using allergen-contaminated wipes or cloths. The materials and methods in this study were chosen with the main intent to mimic and study dry, wet, or sticky and paste food compositions of certain major food allergens that may be commonly found on food contact surfaces in various retail and food service establishments.
MATERIALS AND METHODS
Food-grade SS (304 alloy, 2B finish, Online Metal Supply, Houston, MO), textured polyethylene plastic cutting boards (15.24 by 25.4 cm; Food Service Warehouse, Greenwood, CO) and hard maple wood cutting boards (Carlisle-HLA800, 40.64 by 60.96 by 3.81 cm; Food Service Warehouse) were used for the study. The SS, plastic, and wood were cut to form coupons (∼12 by 12 cm) prior to use. Coupons and surfaces were cleaned prior to each set of experiments by using the following procedure. All items were first rinsed individually under running warm tap water (∼45°C), followed by applying a 2% solution of Micro-90 alkaline detergent (International Products Corporation, Burlington, NJ). Disposable paper towels (Scott C-Fold, Kimberly-Clark, Roswell, GA) were used to scrub the coupon surface, and warm tap water was used to remove the detergent solution. The cleaning procedure was repeated twice, and a final rinse step with deionized water was used before the coupons or items were placed on a dish rack to air dry.
Dry or powdered, wet, and sticky or paste forms of foods containing milk, egg, and peanut allergens were purchased at local grocery stores or online. The foods included Carnation nonfat dry milk powder (NFDMP; Nestlé, Solon, OH), Philadelphia cream cheese (Kraft, Northfield, IL), fluid whole milk (Dean Foods, Dallas, TX), whole egg crystals (Hoosier Hill Farm, Fort Wayne, IN), Hellmann's mayonnaise (Unilever, Englewood Cliffs, NJ), Jif Peanut Powder (The J.M. Smucker Company, Orrville, OH), and Skippy Creamy Peanut Butter (Hormel Foods Corporation, Austin, MN). The protein content (percentage) of each allergenic food was measured with the Kjeldahl test by a contract laboratory (Merieux NutriSciences, Crete, IL). Protein concentrations of nonfat dry milk, cream cheese, fluid whole milk, whole egg crystals, mayonnaise, peanut powder, and peanut butter were 35.3, 5.0, 3.2, 42.2, 1.0, 45.6, and 21.6% (on an as-is basis), respectively. The various protein concentrations are important to note because the different allergenic foods contained different amounts of protein, the analyte detected in the lateral flow device (LFD) assays.
WypAll X60 dry paper wipes (31 by 40 cm; Kimberly-Clark, Roswell, GA), dry terry dish cloths (86% cotton and 14% polyester blend; 30 by 30 cm; Central Restaurant Products, Indianapolis, IN), and sanitizing wipes saturated with 5.48% isopropyl alcohol and 175 ppm of quaternary ammonium chloride (quat; 20 by 26 cm; Table Turners Sani-Professional no-rinse hard, nonporous surface sanitizing wipes, PDI, Inc., Orangeburg, NY) were used in the wiping and transfer studies. Wet terry dish cloths soaked in warm tap water (∼43°C) or in a 50 ppm of total chlorine bleach sanitizer solution (∼43°C) for 5 min were also used in the experiments. Wet terry cloth was gently squeezed to remove excess water or sanitizer solution prior to use. Total chlorine levels in the tap water and sanitizer were measured by using the Hach thiosulfate drop test (product CN-21P; Hach, Loveland, CO) and test strips (product 2745050). The concentration of total chlorine used for sanitizing solution in this study (50 ppm total) is within the concentration range (25 to 100 ppm of total chlorine) specified in the Food Code (Subpart 4-501.114) (23).
Coupon surfaces were swabbed after conducting the wiping, cleaning, and transfer experiments by using the instructions provided with Neogen allergen LFD kits. The presence of milk, egg, and peanut from swabbed surfaces was determined with allergen-specific Reveal 3-D (Neogen, Lansing, MI) LFD tests for total milk (product 8479), egg (product 902082Q), and peanut (product 901041L).
A set of experiments evaluated the effects of sanitizer residue (chlorine or quat) on LFD results. Tap water or chlorine sanitizer solutions (0 or 1 mL; 50 ppm or 100 ppm of total chlorine) were applied to clean, allergen-free surfaces. The surfaces were then swabbed and tested for responses with the LFD tests. Similarly, clean SS, plastic, and wood surfaces were also wiped with the quat sanitizing wipe for 5 s and then tested with a premoistened swab to determine if residual quat affected the LFD responses with the milk, egg, and peanut LFD test kits.
Another study also investigated the possibility of false-negative LFD responses when allergens were in the presence of sanitizers. This series of experiments used the liquid sampling procedure described in the allergen-specific test kits and did not involve swabs or coupons. The protocol used for milk allergen involved mixing 0.1 to 5 mL whole liquid milk with 5 mL of 100 ppm of total chlorine sanitizer solution for 30 s. One milliliter of the mixture was then added to the extraction buffer provided and then tested for the presence of milk with the procedure described in the milk LFD test kit. In a similar manner, 0.1 g of peanut butter was mixed for 30 s with 0.5 to 5 mL of 100 ppm of total chlorine sanitizer for the peanut allergen interference tests, but 0.25 mL of the mixtures were added to the extraction buffer, followed by testing for peanut by LFD. Egg allergen sanitizer interference studies examined the addition of 0.1 to 0.5 g of mayonnaise to 0.5 to 5 mL of 100 ppm of total chlorine sanitizer for 30 s, and 0.25 mL of the mixture was added to the extraction buffer. Similarly, whole egg crystals (0.05 g) added to 5 mL of sanitizer solution was also evaluated with a 30-s contact time with the egg LFD liquid sampling procedure. Overall, various ratios of the allergenic food (each containing different amounts of protein) to 100 ppm of total chlorine sanitizer solution were explored and ranged from a 1:1 to 1:100 ratio of allergen to chlorine sanitizer solution to simulate conditions near the maximum use limit for sanitizer solution. The 30-s mixing time was selected on the basis of the time frame used in the full cleaning study.
Each allergenic food was applied individually to the SS, plastic, and wood coupons to cover a surface area (10 by 10 cm) and spread as evenly as possible with a disposable spatula. The amounts of foods used to contaminate the coupons were as follows: peanut powder (0.05 g); peanut butter (0.1 g); NFDMP (0.05 g); cream cheese (0.1 to 4.0 g); fluid whole milk (1 mL); whole egg crystals (0.05 g); and mayonnaise (0.5 to 2.0 g).
Immediately after foods were applied to the coupons, each surface was then manually wiped for 5 s with a single dry paper wipe, dry terry cloth, or wet terry cloth (soaked in water or 50 ppm of total chlorine sanitizing solution prepared with bleach). Experiments in this study used sanitizing solutions near the mid-concentration level of 50 ppm of total chlorine instead of the upper limit of 100 ppm of total chlorine. After wiping, the presence of the residual allergen was determined by visually inspecting the coupon under room lighting by the same individual (typical of a food establishment) and by swabbing the surface with a premoistened swab by using the procedure described in each allergen-specific LFD test kit. For experiments evaluating sanitizing quat wipes, multiple wipes per surface were used if the surface tested positive for allergens after one wipe was used. Wiping experiments for each experimental condition (food contact surface, type of wipe, allergenic food type, and amount) were completed in triplicate.
A wiping time of 5 s was selected because experiments with 0.1 g of peanut butter or 0.05 g of whole egg crystals on the SS, plastic, and wood surfaces were visually clean on most surfaces after using the dry paper wipe. Wiping for 1 s did not yield a visually clean surface, but a 5- and 10-s wipe time removed most of the food soil from the coupons on the basis of visual inspection. The only exception was a very faint, light yellow stain noted after wiping peanut butter on the textured plastic surface in all triplicate trials.
Full manual cleaning by using the wash–rinse–sanitize–air dry method
Three contaminated coupons for each allergenic food and coupon type (SS, plastic, and wood) were prepared for the full cleaning study. The amounts of food applied to each coupon were peanut powder (0.5 g), peanut butter (1 g), whole egg crystals (1 g), mayonnaise (4 g), cream cheese (4 g), fluid whole milk (5 mL), and NFDMP (0.1 g). The manual ware-washing method with a three-bay sink as outlined in the Food Code was simulated in the laboratory by using three pails. The first pail was designated as a wash pail and contained 10 L of warm tap water (∼43°C) mixed with 5 mL of detergent (Dawn Ultra, Procter and Gamble, Cincinnati, OH). The second pail acted as the rinse pail with 10 L of warm tap water (∼43°C). The third pail contained 50 ppm of total chlorine sanitizer solution, prepared by mixing 6.6 mL of bleach with 10 L of warm tap water (∼43°C). The Hach thiosulfate drop test was used to measure the total chlorine level, as described in the test kit. The full cleaning procedure involved submerging one SS coupon in the wash pail and manually wiping the surface under water in the wash pail with a clean terry cloth for 30 s. The coupon was then immersed in the rinse pail for 30 s, followed by submerging it in the sanitizer pail for 30 s. The final step was to air dry the coupons on a drying rack for a minimum of 30 min. The full cleaning procedure was repeated until all three SS, plastic, and wood coupons, having the same allergen load per surface, were washed consecutively by using the same wash, rinse, and sanitizer pails. After air-drying coupons for a minimum of 30 min, each surface was sampled with one premoistened swab and analyzed for allergen residue with the appropriate LFD test. All full cleaning experiments were conducted without scraping the surfaces with a plastic spatula (prescrape step) prior to washing the coupons. An exception was made for coupons contaminated with peanut butter, which were evaluated with and without a prescrape step. The full cleaning experiment was repeated three times.
Allergen transfer experiments
For the dry wipe transfer study, allergenic food was applied to the center of a dry paper wipe (WypAll X60). The amount of dry foods used to soil the dry wipe were as follows: whole egg crystals (0.01 to 0.05 g); peanut powder (0.01 to 0.05 g); and NFDMP (0.05 g). Sticky, paste, and wet foods were also evaluated in the study and included mayonnaise (0.5 to 2.0 g), peanut butter (0.1 g), fluid whole milk (1 mL), and cream cheese (0.5 g). The contaminated wipe was then used to wipe four consecutive coupon surfaces of the same composition for 5 s of contact time between the wipe and each surface. The wiped surfaces (1 to 4) were then sampled with a premoistened swab and analyzed for presence of allergen by using the appropriate LFD test.
A set of experiments evaluated the extent of transfer from terry cloth to wiped surfaces when the cloths were stored in sanitizer solution before use. The Food Code, Subparagraph 3-304.14 (B)(1), recommends that cloths in use for wiping counters and other equipment surfaces are held between uses in a chemical sanitizer solution. A sanitizer solution (50 ppm of total chlorine) was prepared by adding 2.5 mL of bleach to 3.78 L of warm tap water (∼40 to 45°C), and residual chlorine level was measured. A clean terry cloth was soaked in sanitizer solution for 5 min and then gently squeezed to remove excess sanitizer solution. The center of the wet cloth was loaded with individual allergenic foods (0.05 g of whole egg crystals, 0.05 g of peanut powder, 0.05 g of NFDMP, 2.0 g of mayonnaise, 0.1 g of peanut butter, 1 mL of fluid whole milk, and 2.0 g of cream cheese), and the allergen-contaminated cloth was then wiped on the surface of one coupon type for 5 s. The same cloth was submerged in sanitizer solution for 15 s and then wiped on a second coupon of the same composition as the first. The same procedure was followed to wipe the remaining two other coupons. All four surfaces were sampled by using a premoistened swab (one swab per surface) and analyzed for the presence of peanut, milk, or egg residue with an LFD test. Transfer experiments were repeated in triplicate.
RESULTS AND DISCUSSION
Food service and retail food establishments often handle a wide variety of food allergens in various forms that routinely contact SS, as well as plastic or hardwood food contact surfaces, such as cutting boards, bowls, cookware, and utensils during food preparation. Allergenic food matrices selected for this study were chosen on the basis of an attempt to evaluate various forms of milk (NFDMP, whole liquid milk, and cream cheese), egg (whole egg crystals and mayonnaise), and peanut (peanut powder and peanut butter) allergens in a dry, wet, or sticky and paste composition, that may be commonly found in kitchens of food establishments in preparation of sandwiches or bakery items. Additionally, these foods were chosen because milk, eggs and peanuts are identified as “major food allergens” in the Food Allergen Labeling and Consumer Protection Act of 2004 and in the Food Code (22, 23). The coupons or surfaces selected for use were chosen to reflect different finishes (smooth, textured, and porous) and materials of composition (SS, polyethylene plastic, and hard maple wood) of food contact surfaces used in food establishments. Similarly, the dry paper wipes, terry cloth, and disposable quat wipes chosen for the study reflect items described in Chapter 4 of the Food Code and are commonly used in food establishments for wiping surfaces with or without use of a bleach-based sanitizing solution (23). The wiping and allergen transfer studies were designed to provide information on the effectiveness of some practices that may be used outside of the Food Code recommendations. The full cleaning method, as described in Chapter 4 of the Food Code, used the manual three-compartment warewashing method incorporating a detergent containing wash (compartment 1), clean water rinse (compartment 2), chlorine-based sanitizing step (compartment 3) and was followed by air drying the surfaces (23). To simulate a practical use application of this cleaning method, three SS, three plastic, and three wood surfaces each having high allergen loads on the individual surfaces were manually cleaned and evaluated for allergen residue by using allergen-specific LFDs.
Use of LFDs to detect allergen residues
Allergen-specific LFD tests used in this study provided a rapid, qualitative assessment regarding the presence of allergen residue rather than quantitative results. Positive control experiments were conducted to ensure that the lowest amount of each allergenic food used in the experiments could be detected on the coupons prior to any wiping or cleaning. For all allergenic foods (0.01 g of peanut powder, 0.1 g of peanut butter, 0.05 g of NFDMP, 0.1 g of cream cheese, 1.0 mL of fluid whole milk, 0.05 g of whole egg crystals, and 0.1 g of mayonnaise), positive LFD responses (3 of 3) were recorded. The limit of detection (LOD) for the peanut, milk, and egg LFD tests were not determined for each of the allergenic foods evaluated in this study.
Negative control experiments were used to confirm that the presence of chlorine sanitizer did not result in positive LFD results or interfere with the immunochemical tests. For example, testing 100 ppm of total chlorine sanitizer directly, by mixing with the extraction buffer included in the milk, peanut, and egg LFD kits, tested negative and showed no interference with the LFD test response (Supplemental Table S1). “High-positive” LFD results reflect an overloaded sample having a high allergen concentration. Additional experiments were also conducted to determine if the ability to detect allergenic food was influenced by residual sanitizer solution. Varying ratios of whole liquid milk, peanut butter, mayonnaise, or whole egg crystals and 100 ppm of total chlorine sanitizer solution were mixed for 30 s and analyzed with the appropriate LFD, after dilution with extraction buffer included with each LFD kit. The results of the LFD tests are shown in Table S1. All triplicate responses were positive or high positive (as described in Table S1 and the test kit insert on reading LFD results) for the presence of the allergens that indicated that 100 ppm of total chlorine sanitizer solution did not interfere with the LFD tests under the tested conditions. Additionally, sanitizer residue (chlorine or quat) swabbed from clean surfaces tested negative with the peanut, milk, and egg LFD tests.
It is common practice within retail and food establishments to routinely wipe surfaces with disposable wipes or reusable cloths. The current (2017) edition of the Food Code (23) provides recommendations and use limitations of wiping cloths from a microbial control perspective. An important distinction for this study is to note that “wiping” for allergen removal is not equivalent to “cleaning” as described in the Food Code. Both Tebbutt (19) and Welker et al. (27) examined cleaning and wiping from a microbial control perspective and concluded that wiping surfaces having a food soil is different from cleaning a surface.
Information is currently lacking on the effectiveness of wiping methods on the removal of peanut, milk, and egg allergens from common food contact surfaces used in food establishments. This wiping study investigated removal of allergens in dry, wet, paste, and sticky forms and used five different types of wipes: a dry wipe (WypAll X60), a dry terry cloth, a wet terry cloth soaked in tap water, a wet terry cloth soaked in 50 ppm of total chlorine sanitizer solution, and a sanitizing disposable quat alcohol wipe. In general, the dry wipe and dry terry cloth were not effective in completely removing the different forms of peanut-, milk-, or egg-containing foods from most of the surfaces under the conditions tested as shown in Tables 1 to 7. Use of the dry wipe or cloth on the dry forms of the allergenic foods (i.e., peanut powder, nonfat dry milk, and egg crystals) was generally not adequate in removing allergens, because positive LFD results were detected on many of the surfaces in the triplicate trials, although the surfaces appeared visually clean (Tables 1, 3, and 6). For instance, as shown in Table 1, when the dry wipe was used to wipe peanut powder (0.5 g) from the SS, plastic, and wood, peanut residue was detected by LFD on all surfaces in triplicate trials. The dry terry cloth was used in the same manner, and peanut residue was detected on the SS, wood, and plastic surfaces in all three trials, except for one replicate trial for the plastic surface that showed complete removal of peanut powder. Similar to the results observed with the dry and powdered form of allergens, use of the dry wipe and dry terry cloth was not effective at removing allergenic food pastes (i.e., peanut butter, cream cheese, and mayonnaise) from the SS, plastic, and wood coupons (Tables 2, 4, and 7), although in some cases, the surfaces appeared visually clean.
The effectiveness of the wet terry cloth soaked in either tap water or 50 ppm of total chlorine sanitizer solution at allergen removal depended on the amount and form of the food allergen (dry, wet, paste, or sticky) and the composition of the coupon. For example, as shown in Tables 1, 4, and 7, the use of a wet terry cloth (soaked in tap water or sanitizer solution) to remove 0.05 g of peanut powder, 0.5 g of cream cheese, or 0.5 g of mayonnaise from coupon surfaces resulted in no detectable peanut, milk, or egg residues, respectively, on SS, plastic, and wood surfaces in triplicate trials. However, when higher amounts of cream cheese (Table 4) and mayonnaise (Table 7) were loaded on the wood or plastic surfaces, the wet terry cloth was not always effective at allergen removal. The coupons appeared to be visually clean, unless noted otherwise in the tables.
In general, disposable quat wipes were effective for allergen removal from the various surfaces, especially when multiple wipes were used (Tables 1 to 7). In most cases, two, three, or four wipes were needed to effectively remove allergens from surfaces and test negative (0 of 3) with the LFDs. The textured plastic surface was more difficult to wipe clean than the SS or wood surfaces when contaminated with sticky or paste forms of the allergenic foods, and additional wipes were often required to completely remove the allergen to levels below the LFD detection limit. As shown in Table 2, three wipes were required to remove 0.1 g of peanut butter from the SS and wood surfaces, but the textured plastic required four wipes to test negative for peanut by using the LFD tests. An early study by Tebbutt (19) and Welker et al. (27) also found that it was challenging to remove microbial contaminants from polypropylene plastic and wood surfaces. All quat-wiped surfaces were visually clean after using one wipe to remove 0.1 g of peanut butter, with the exception of a slightly oily sheen on the SS surface. Overall, these results are similar those reported by Watson et al. (26) who demonstrated the effectiveness of using one or more sanitizer wipes to remove peanut butter from a variety of different surfaces (a nonporous plastic table, a plastic toy, and plastic ball).
Although SS and plastic surfaces are commonly found in food establishments, the use of hardwood surfaces has been a subject of debate, mainly due to microbiological safety concerns. Research on the cleanability of different food contact surfaces showed that it is was more difficult to recover bacteria inoculated onto the surfaces of hardwood (maple, beech, oak, or walnut) coupons than from plastic (polyethylene or polyacrylic) surfaces (2, 3, 7). The researchers attributed their findings to the porosity of hardwood coupons. Additionally, Gehrig et al. (7) found through scanning electron microscopy that surfaces of polyethylene cutting boards after heavy use, had rough “cavernous” surfaces that could retain and later release bacteria.
In contrast, a study by Lucke and Skowyrska (11) found no significant differences between the hardwood and polyethylene cutting boards, with respect to cleanability from a microbial control perspective. A recent review by Aviat and Gerhards (4) suggests that in addition to the porosity of hardwood surfaces, reduced recovery of bacteria inoculated onto hardwood food contact surfaces can be attributed to the presence of antimicrobial compounds in wood. On the basis of recent research, wood surfaces may pose a lesser relative risk from a microbiological point of view, and it appears from this study that the same may also be true for allergen transfer.
The success of cleaning procedures on removal of allergenic foods from food contact surfaces depends on several factors, including the types of surfaces and cleaning methods available, especially because both factors are interrelated (11, 16). The effectiveness of wipes for allergen removal may also be impacted by the absorbency of the wipe, the solvent used for wet wipes, the state of the allergen matrix (wet, sticky or paste, or dry), and the amount of food or allergen loaded on the surface. For parameters evaluated in this study, use of a wet wipe, cloth, or quat wipe to remove a dry allergen from a surface appeared to be more effective than use of a dry wipe. The food contact surface condition (smooth versus textured) appeared to play a role in determining the degree of effectiveness when wiping allergens from surfaces, similar to the results of studies that evaluated removal of microbial contaminants from food contact surfaces (19, 27).
To more closely simulate what would be done in retail and food service operations, visual inspection of wiped surfaces was conducted by the same individual who performed the wiping experiments. Although surfaces that were visually clean did not always correspond to negative LFD test results, visual inspection provided a first step for evaluating the effectiveness of wiping treatments. For example, as shown in Table 5, wiping 1 mL of liquid milk with one quat wipe resulted in positive LFD responses on all surfaces, although no visible residue was apparent on the plastic or wood, and only a very slight sheen was apparent from an angled view on the SS surface. Use of two quat wipes resulted in all surfaces appearing visually clean, but the textured plastic surface contaminated with 1 mL of whole liquid milk still resulted in 3 of 3 very faint positive LFD results, and three quat wipes were required to correspond to negative LFD results. Similarly, 0.5 g of mayonnaise was easily wiped from each surface with one quat wipe, and all surfaces were visually clean and had negative LFD results (0 of 3; Table 7). Increasing the amount of mayonnaise to 2 g and use of a quat wipe resulted in faint positive LFD responses on all surfaces, which indicated that the amount of egg residue was near the LOD of the egg-specific LFD kit. Although all plastic and wood surfaces were visually clean, a slightly oily smear was initially visible only on the SS coupons, which then appeared visually clean after the mayonnaise residue dried. Two quat wipes were required to remove 2 g of mayonnaise from each surface to obtain a visually clean and negative LFD (0 of 3) response on all SS, plastic, and wood surfaces, as noted in Table 7.
Limitations that exist with visual assessment of cleaning effectiveness include the type and adequacy of the lighting, the color and textural differences between the food contact surface and the allergen residue, and the visual acuity of the examiner. In this study, the use of white plastic coupons hindered visualization of light-colored foods, such as milk, cream cheese, mayonnaise, and NFDMP. In these circumstances, visual inspection may not provide adequate assessment of the presence of food residues. Also, we found instances in which the surfaces appeared visually clean but still tested positive for allergen residue on the basis of the LFD test results. The significance of these results is not clear because the allergen-specific LFD tests used in this study provide qualitative rather than quantitative results. Thus, it is difficult to determine the amount of hazardous allergenic residue. It was observed that most allergen LFD results on some visually clean surfaces were faintly positive, suggesting that the amount of allergen present was close to the LOD of the LFD test and thus likely to be quite low. However, more research is needed to understand the significance of these positive residue results.
Full cleaning study
A full cleaning method, also referred to as the “wash–rinse–sanitize–air dry” procedure simulated the process of using a three-bay sink and air-drying surfaces on a dish rack after cleaning. The entire wash–rinse–sanitize–air dry procedure was repeated for a total of three independent trials. In this experiment, the amounts of food allergen added to each coupon was substantially greater than those used in the wiping study. As shown in Table 8, results demonstrated that the full cleaning method was consistently effective in allergen removal in triplicate trials (0 of 3, 0 of 3, 0 of 3 positive LFD test results for each type of surface and all surfaces were visually clean) for all types of coupons and for all allergenic foods, with the exception of peanut butter. The textured plastic coupons retained peanut residue as detected by the peanut-specific LFD in all three trials (3 of 3, 3 of 3, and 3 of 3), but two faint positive residues and negative responses were found for wood surfaces in the triplicate trials (f+ 1 of 3, 0 of 3, f+ 1 of 3). Note that during washing, peanut butter from the contaminated coupons (1 g of peanut butter per coupon) was transferred into the wash water (10 L). Because nine coupons were consecutively washed, the wash water contained up to 900 ppm of peanut butter at the conclusion of each trial. Also, because wood coupons were washed last in this study, the faint positive LFD results in two of the independent trials may be attributed to peanut butter present in the wash water that may have redeposited on the wood surfaces. The wood surfaces appeared visually clean except for a slightly oily and wet stain, yet the wood surfaces tested negative or registered faint positive LFD results for peanut residue.
All the SS surfaces appeared visually clean and tested negative for peanut in the LFD tests, which is most likely attributed to the smooth SS surface finish and because the SS surfaces were washed first in all trials. The white, polyethylene plastic coupons on the other hand, tended to retain peanut butter within the grooves of the textured surface and displayed a faint yellow color stain in five of the nine plastic coupons. Thus, approximately 44% of the textured plastic surfaces appeared visually clean, but all of the LFDs were positive for peanut residue. Implementing a prescrape step to remove the bulk of the peanut butter residue prior to washing improved the effectiveness of the cleaning procedure for the textured plastic coupons, with faint positive (f+ 2 of 3, f+ 3 of 3, f+ 3 of 3) LFD responses recorded in the three trials.
Relatively few studies report the effectiveness of a full manual cleaning procedure on allergen removal. The presence of milk, egg, and gluten on utensils, cookware, and other food contact surfaces present in school cafeterias and kitchens in Spain was examined by Ortiz et al. (13). In that study, where the food contact surfaces were either washed with an automatic dishwasher or manually washed, milk residue was not found on the surfaces with LFD tests, but 15% of egg and 45% of gluten LFD results were positive. Cleaning conditions (i.e., time and temperature of the cleaning procedures, detergent concentrations, and use of three basins for manual washing) were not described. In addition, it was also unclear whether the positive results were due to recontamination of the surfaces by use of allergens in daily operation and management of the cafeteria. Miller et al. (12) found food contact surfaces and food prepared in a commercial kitchen could become contaminated with gluten if controls were not in place to prevent dispersal of gluten-containing ingredients.
In general, manual warewashing appeared to be effective for allergen removal when practiced according to the procedures outlined in the Food Code. Using a prescrape step (Subpart 4-603.12 Precleaning) to remove the bulk of allergenic food residues and decreasing food load in the wash water improved overall effectiveness of the full cleaning procedure (23). Although not studied here, changing the wash water frequently to maintain clean solutions is another factor that can improve cleaning effectiveness. Other factors that may impact cleaning effectiveness include the amount and type of allergenic food on the surface, time and temperature of the wash solution, type and concentration of detergent in the wash sink, composition and finish of food contact surface material, and the mechanical and manual force used during the washing step. Other strategies to clean and minimize cross-contact include washing the prescraped allergen surface more than once, increasing the submersion time in wash water, or simply maintaining dedicated cutting boards or surfaces when possible, especially if using textured plastic materials with peanut butter. The U.S. Department of Agriculture cutting boards and food safety fact sheet (21) also suggests using a dedicated cutting board for raw meat, poultry, and seafood and maintaining a separate food contact surface for fresh produce to prevent microbial cross-contamination, despite the ability to effectively clean cutting boards from a microbial control perspective. This concept can also be extended to sticky allergenic foods, such as peanut butter and other similar foods, which can be problematic for effective manual warewashing on select materials.
A limitation of the full cleaning study design involved the use of a single order to wash the coupons (SS, plastic, and wood). Future experiments should randomize the order of cleaning the different surfaces to allow for exposure to wash water having varying levels of food soils. Another limitation of this washing study was the absence of food soils that were dried, cooked, or heated on the surfaces. Cooked food soils tend to require more manual force and cleaning effort in removing denatured proteins, such as heated milk, which can adhere to equipment and surfaces (16, 25).
Allergen transfer study
The focus of this series of experiments was to determine the extent of allergen transfer to surfaces from a contaminated wipe or cloth. Unlike previous studies in which coupon surfaces were directly contaminated with allergenic foods, the allergenic foods were placed on dry wipes or sanitizer-soaked terry cloth for transfer experiments. In the experiments that used dry wipes, one allergen-contaminated dry wipe was used to wipe four consecutive coupon surfaces of the same material composition, followed by testing all four surfaces for the presence of allergens with allergen-specific LFD tests.
Most dry or powdered allergens transferred from the dry wipe to all four wiped surfaces as shown in Table 9. Whole egg crystals (0.01 g) on the dry wipe showed a mixed degree of egg transfer to surface 4, while a higher allergen load of 0.05 g of whole egg crystals on the dry wipe, consistently transferred egg to all surfaces with (3 of 3) positive LFD results. Peanut powder (0.01 g) resulted in no detectable transfer (0 of 3) on wood coupon 2 and SS coupon 3, respectively. However, peanut residue was present on all textured plastic surfaces in all three trials. The NFDMP (0.05 g) also transferred from the dry wipe to all SS, plastic, and wood coupon 4, with positive LFD responses in all three trials.
Wet, paste, and sticky forms of allergens also transferred from the dry wipe to many of the subsequently wiped surfaces, as shown in Table 10. Only mayonnaise (0.5 g) resulted in minimal egg allergen transfer to subsequent surfaces, with no egg detected on all SS, plastic, and wood surface 3 (0 of 3). Increasing the food load to 2 g of mayonnaise on the dry wipe led to extended allergen transfer to some surface 4 plastic and wood coupons, but egg LFD responses were only faintly positive. In general, allergen absorption by the dry wipe and the porous wood surface may provide one explanation for the many faint positive LFD results detected on wood, compared with the positive LFD results registered on the smooth SS or textured plastic surfaces. Additionally, the lower protein content in the mayonnaise compared with the egg powder may have been responsible for the mixed and faint positive results for allergen transfer on surface 4. One disparity of note in Table 10 is with the 0.1 g of peanut butter transfer experiment between SS surface 3 in which 0 of 3 LFD results were observed and SS surface 4, with 1 of 3 positive LFD responses. A possible explanation is that peanut butter present on the wipe did not make contact with SS coupon 3 but was able to transfer to SS surface 4 during the wiping step. Experiments with whole fluid milk and cream cheese showed milk transfer to all SS, and plastic surface 4 from the dry wipe, with only faint positives noted on the wood surface.
Prior studies have shown that reusable wiping cloths harbored bacteria when they were not stored in sanitizing solutions (10, 19). The Food Code guidelines on use limitations for wipe cloths, as discussed in Subparagraph 3-304.14 (B)(1), were followed to determine the extent of allergen transfer from a wet terry wipe cloth that is contaminated with allergen (23). The objective was to simulate current recommendations for use and storage of a cloth, by submerging the allergen-contaminated wipe cloth in sanitizer solution before wiping each surface. Storage of the cloth in sanitizer solution prior to wiping each surface resulted in no dry allergen transfer to some surface 2 and no transfer to surface 3 (Table 11) for the dry forms of peanut and egg allergens investigated in this study. The NFDMP, on the other hand, showed no transfer to surface 2 when the cloth was stored in sanitizer solution prior to wiping surfaces. The detection of allergen residue on surface 1 was expected because the allergen was added directly to the wet sanitizer-soaked cloth and transferred immediately to surface 1, with the intentional objective to show allergen transfer from wet allergen contaminated terry cloth to the initial surface. Note that the peanut powder and NFDMP both had minimal transfer of allergen from the cloth to wood surface 1, which may be attributed to the porous nature of the wood surface.
A wet terry cloth contaminated with wet, paste, or sticky allergens (Table 12) that was submerged in sanitizer solution before wiping surfaces transferred allergens to a lesser extent than the dry paper wipes (Table 10). Minimal fluid milk transfer was noted on SS and plastic surface 1, and no detectable milk transfer on surface 2 was observed for all surfaces (Table 12). Interestingly, fluid milk (1 mL) was not detected by LFD on wood surface 1 in all three trials, which may be due to absorption of the milk by the wood surface and/or the wet terry cloth. Cream cheese (2 g) was not detectable on SS or wood surface 3 but was detected in 1 of 3 trials on textured plastic surface 3. The wipe cloths contaminated with 2 g of mayonnaise showed no detectable transfer of egg allergen to surface 3 for SS, plastic, and wood when the cloth was submerged in the sanitizer pail between wiping surfaces. Peanut butter (0.1 g) resulted in the greatest extent of allergen transfer from the wipe cloth to surface 3 SS, plastic, and wood in triplicate tests. However, surface 4 (plastic and wood) resulted in no peanut transfer (0 of 3), while the SS surface 4 had one very faint positive (1 of 3) peanut LFD response.
Overall, the results of the allergen transfer study indicate that the current Food Code (23) recommendations for use limitations requiring wipe cloth storage in sanitizer pails between use minimizes allergen transfer from the wipe cloths to surfaces. When soiled wipe cloths are stored in the sanitizer pail, the food present on cloths is likely transferred to the sanitizer solution and increases the food load to the solution. This results in a depletion of active sanitizer (chlorine) in the sanitizer solution and a need to replace the solution when concentrations are below the specific temperature or sanitizer guidelines as stated in the Food Code (23). The practice of preparing fresh sanitizer solution helps prevent the buildup of food soils and allergens in the sanitizer solution, which potentially could contaminate food contact surfaces and also ensures that sanitizer levels are at appropriate levels to address microbial concerns. Although most of the coupons were visually clean when examined after wiping, allergens were detected with LFD tests on some of the surfaces. The inability to visually detect food residue on surfaces during the transfer study may be due the very low amounts of allergenic foods on the surfaces and the color and texture of the coupons that prevented visual detection of residue.
Some limitations of this study include the absence of blinded tests for determination of visually clean surfaces, a lack of uniformity of how the allergenic foods were applied to the surfaces, an inability to quantify allergens remaining on the surface, and focusing on a single allergen matrix instead of food allergen mixtures, among others. In addition, the wiping, cleaning, and allergen transfer study was performed on freshly applied food soils. The results would likely have been different if foods were dried onto surfaces prior to wiping because dried food soils can be difficult to remove (16). The manual cleaning process is also subjective and typically conducted to a specific end point, which is often the visually clean standard. Although efforts to conduct the experiments in the same manner were made, subtle differences in the amount of pressure used in wiping and cleaning, absorbency of the wipe, and varying saturation levels of the cloth may impact the effectiveness of allergen removal and transfer. Additionally, the surfaces used in this study were similar in color (white polyethylene plastic and natural maple hardwood) to some of the allergens (NFDMP, whole liquid milk, cream cheese, mayonnaise, peanut butter, and peanut powder) used, which occasionally made visual inspection for allergen residue challenging at times. Future experiments may explore different combinations of allergen food soils, other allergen-specific LFD tests, quantitative tests, various colored surfaces and topologies, as well as a range of different detergent concentrations, including varying time and temperature parameters for cleaning and wiping.
Overall, the nature and amount of allergen on a surface, as well as the type and state of wipe cloth, food contact surface texture and material composition, influenced the effectiveness of wiping and washing treatments on allergen removal and the extent of allergen transfer on surfaces. In summary, the wiping study suggested that wet terry cloth (soaked in tap water or sanitizer solution) and alcohol quat wipes were generally more effective in allergen removal than dry wipes. Additionally, allergenic foods in this study appeared to be more difficult to remove from the textured plastic surface than the SS or wood surfaces. In general, the full cleaning method (wash–rinse-sanitize–air dry) for manual warewashing with detergent and sanitizer was effective at removing most allergenic food residues and tended to be more effective at removing higher allergen loads from surfaces than using wipes or cloths alone. A prescrape step prior to washing improved the removal of peanut butter on surfaces. Due to the nature of peanut butter and its adherence to textured plastic, multiple washings or use of dedicated cutting surfaces are recommended. Contaminated dry paper wipes tended to transfer allergens to subsequently wiped surfaces under the conditions of this study. However, storage of cloths in sanitizer solution between wiping surfaces, as prescribed in the Food Code (23), minimized allergen transfer. Many of the surfaces tested in this study had only faint positive responses for the allergen, suggesting that the amount of allergen residue may be near the LOD of the LFD. Although more research is needed to understand the potential health hazard of residues detected by LFDs in this study, using a visibly clean end point in combination with other food safety measures appears to be prudent approaches for allergen removal.
Further research is needed to quantify the amount of allergen present on surfaces when faint positive results are registered. Additional research is also needed to evaluate the amount of transfer from surfaces with low amounts of allergenic residue to other food items.
The authors appreciate the conversations and insights provided by Dr. Hal King and his suggestion on expanding the scope of the study to include alcohol quat sanitizer wipes (Table Turners) on the basis of current retail food establishment practices.
Supplemental material associated with this article can be found online at: https://doi.org/10.4315/JFP-20-025.s1