summary writing

Research in Veterinary Science 165 (2023) 105071

Available online 30 October 2023 0034-5288/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

Assessment of the content of macronutrients and microbiological safety of dry dog foods

Jagoda Kępińska-Pacelik a,*, Wioletta Biel a, Robert Witkowicz b, Krzysztof Frączek c, Karol Bulski c

a Department of Monogastric Animal Sciences, Division of Animal Nutrition and Food, West Pomeranian University of Technology in Szczecin, Klemensa Janickiego 29, 71-270 Szczecin, Poland b Department of Agroecology and Crop Production, University of Agriculture in Krakow, Mickiewicza 21, 31-120 Krakow, Poland c Department of Microbiology and Biomonitoring, University of Agriculture in Krakow, Mickiewicza 21, 31-120 Krakow, Poland

A R T I C L E I N F O

Keywords: Yeasts Molds Quality Pet food Indicator bacteria

A B S T R A C T

In Europe, there are no legal regulations specifying maximum allowable limits for pathogenic bacteria or fungi and yeasts in dog food. For proximate composition, the European Pet Food Industry Federation (FEDIAF) nutritional guidelines provide minimum recommended levels for protein and fat only, not for other ingredients and no safe maximum levels. Therefore, the aim of this study was to evaluate the microbiological safety of 35 dry dog foods, taking into account the division into foods with grains as the main plant ingredient and those described as grain-free. We assumed that grain-included foods are more susceptible to the presence of the total number of aerobic bacteria. This parameter is an essential, hygienic criterion informing about the microbio- logical safety of the pet food. It also inform about the microbiological quality of the food ingredients used, the effectiveness of decontamination during the production process, sanitary and hygienic conditions during the acquisition, processing and marketing of pet food and its components. In the case of macronutrients, the results in some cases were multiples of the minimum recommended levels. The most important microorganisms that presence was found in the analyzed dog foods were indicator bacteria (E. coli and coliform bacteria, Clostridium perfringens), whose presence in feed constitutes microbiological contamination. They inform about sanitary and hygienic conditions during the production and storage of the pet food. The current guidelines recommend microbiological testing of food for the presence of Salmonella bacilli as the basic indicator for assessing the sanitary quality. No Salmonella spp. was detected in any of the foods. The remaining analyzed bacteria (Proteus spp., Enterococcus spp., Staphylococcus spp.) as well as molds and yeasts, were present in both grain-included and grain-free foods. The obtained results showed that in terms of microbiological safety, grain-included foods come out worse, because microorganisms were detected more frequently in this type of dry dog food.

1. Introduction

For over 15.000 years, dogs have evolved to meet human needs as “man’s best friend”, and the importance of relationship with them has also increased (Samet et al., 2022). Currently, dogs are an important member of society, they are not only companion animals, but also play

roles as guide dogs or dogs in uniformed services. There are already almost 90 million dogs in Europe, and 161 million of them all over the world and the number is constantly growing (FCI, 2022). Due to the increasing number of dogs, the pet food industry is developing dynam- ically (FEDIAF, 2022). Dog caregivers are looking for the best food that meets all their pet’s needs. Currently, there are many types of food on

Abbreviations: AGR, accessory gene regulator; ANOVA, one factorial analysis of variance; CA, crude ash; CF, crude fiber; CFU, colony forming units; CP, crude protein; DM, dry matter; EE, ether extract; FEDIAF, The European Pet Food Industry Federation; ME, metabolizable energy; NFE, nitrogen-free extracts; NRC, The National Research Council; MEA, malt extract agar; MRSA, methicillin-resistant Staphylococcus aureus; PCA, principal component analysis; RASFF, Rapid Alert System for Food and Feed; SAR, staphylococcal accessory regulator; TBX agar, tryptone bile X-glucuronide agar; TSA, trypticase-soy agar; VRBL, agar with crystal violet, red, neutral bile and lactose.

* Corresponding author. E-mail addresses: [email protected] (J. Kępińska-Pacelik), [email protected] (W. Biel), [email protected] (R. Witkowicz),

[email protected] (K. Frączek), [email protected] (K. Bulski).

Contents lists available at ScienceDirect

Research in Veterinary Science

journal homepage: www.elsevier.com/locate/rvsc

https://doi.org/10.1016/j.rvsc.2023.105071 Received 5 June 2023; Received in revised form 9 October 2023; Accepted 10 October 2023

Research in Veterinary Science 165 (2023) 105071

2

the market, but the most convenient seems to be dry food (Naughton et al., 2022), however their microbiological safety is questioned (Serhan et al., 2022). On the other hand, as reported by Runesvärd et al. (2020), raw meat-based diets are questionable because of alarming levels of bacteria (52% of analyzed samples), mainly Escherichia coli and those of the genus Salmonella spp., which are common food-borne pathogens. Dry dog foods are not safer in this respect. Nowadays there are various types of dry food available on the market, therefore it is very easy to choose a food adapted to the animal’s needs, age, size, activity, or accompanying diseases.

Contamination of pet food with pathogenic microorganisms not only creates the risk that animals eating these products may become ill or become carriers of pathogens. It also poses a health risk to pet caregivers who handle food products and interact with pets. An important fact is also the possibility of transmission of pathogens between animals. In the case of certain diseases, animals are considered to be a potential reser- voir of pathogenic organisms that can be transmitted directly or indi- rectly to pets and free-living animals in the same environment. It should be noted that the dog is on the verge of being carnivorous and omniv- orous, but generally the basis of its diet is meat and animal derivatives. Producers of “premium” type dog foods ensure that the raw materials used in them are of high quality. However it does not mean that they are free from hazards. For example, potential threats are presence of various pathogens including molds and their secondary metabolites (mycotoxins).

In addition to animal raw materials, dog foods also contain plant raw materials as carbohydrate and crude fiber sources. In general, the most common type of dog foods are those where the main plant ingredients are grains such as wheat, oat, corn or rice (Beloshapka et al., 2016). Nowadays, grain-free foods, in which the main source of carbohydrates are sweet potatoes, potatoes, and legumes, are gaining popularity. Although The European Pet Food Industry Federation (FEDIAF, 2021) does not recognize it as an essential nutrient for dogs, the presence of carbohydrates in the diet is beneficial (de Godoy et al., 2013). For example, it can help control body weight by reducing the energy density of the diet. Moreover, slowly digestible carbohydrates can provide the feeling of fullness (Gourineni et al., 2017). Carbohydrate sources (di- etary fiber fractions) can also be fermented by colon bacteria providing health benefits for the animal (Calabrò et al., 2013; Palmqvist et al., 2022; Paßlack et al., 2022).

There are many studies suggesting that cereals may be a source of undesirable microbes, such as bacteria and mold fungi (Aspergillus, Penicillium, Fusarium) that can produce mycotoxins (Savi et al., 2018; Thomas-Popo et al., 2019; Kim et al., 2021). The use of contaminated raw material may result in the presence of microbes in the final product – pet food, the consumption of which has negative health consequences (Böhm et al., 2010; Gazzotti et al., 2015).

Due to the uncertainties associated with feeding grain-included foods, one of the perceived health benefits of grain-free foods may be the elimination of chronic exposure to low-dose mycotoxins, as grains in pet food are presumed to be the main source of them (Tegzes et al., 2019). The legal regulations regarding the microbiological safety of dog food are quite scarce and most of them do not refer directly to pet food.

The aim of this study was to evaluate the microbiological safety of dry dog foods, taking into account the division into foods with grains as the main plant ingredient and those described as grain-free. We assume that grain-included foods are more susceptible to the presence of mi- croorganisms, than grain-free foods. An additional criterion was the analysis of the proximate composition of the analyzed foods, taking into account compliance with the FEDIAF (2021).

2. Material and methods

2.1. Material

The material for the study consisted of 35 complete commercial

extruded dry dog foods: 23 grain-free and 12 grain-included (Table S1). They were generally available complete foods for adult dogs of all breeds. The main plant component of grain-free food was usually po- tatoes and/or sweet potatoes, peas, while in the case of grain-included it was rice, oat, barley and corn.

The term, which is often used for foods that do not contain cereals is “grain-free”, although in fact „grains” is a much broader word than „cereals”, therefore the commonly accepted term is not entirely correct. However, for the purposes of this article, we will use the terms “grain- free” and “grain-included” (Wrigley et al., 2004). The weight of collected bags were from the range of 0.3 kg to 2.5 kg. These foods were bought at Polish pet stores, however more brands were international, than local. They were the most popular pet foods, and were selected based on the opinions of sellers in pet stores. Dog foods were bought in January 2022, all of them were produced in the fourth quarter of 2021. All bags were stored in laboratory at room temperature (approx. 18–21 ◦C) until analysis. Bags were opened on the same day.

A representative 100 g sample of each food was taken from each package for laboratory analysis (ISO, 2012). Analyzes of obtained analytical samples were performed in three repetitions.

2.2. Proximate composition

Representative samples of each product were ground in a laboratory mill type KNIFETEC 1095 (Foss Tecator, Höganäs, Sweden) to powder, whose particles had a diameter of about 0.5 mm, placed in sterile con- tainers and marked with the symbols 1–35.

The proximate composition was determined in the ground samples according to the methods of the Association of Official Analytical Chemist (AOAC, 2019). To determine dry matter (DM), samples were dried at 105 ◦C to a constant weight (method 945.15). Crude protein (CP) (method 945.18) was determined from 6.25 × total nitrogen measured by the Kjeldahl method using a Büchi Scrubber B414 digestion apparatus and a Büchi 324 distillation set (Büchi Labortechnik AG, Flawil, Switzerland). Crude fat (as an ether extract, EE) was determined by the Soxhlet method with diethyl ether used as a solvent (method 2003.06). The crude fiber (CF) was determined with an ANKOM220 Fiber Analyser (ANKOM Technology, New York, USA). Crude ash (CA) was determined by burning in a muffle furnace at 580 ◦C for 8 h (Czylok, Poland) (method 920.153). Nitrogen-free extracts (NFE) were calculated as follows: NFE = 100 − (moisture content + CP + EE + CA + CF), all values are expressed in g/kg DM, except of the content of dry matter (g/ kg of fresh matter).

2.3. Metabolizable energy calculation

The metabolizable energy (ME) value of dog foods was calculated on the basis of the determined proximate composition. According to FEDIAF (2021) and NRC (2006), for calculation of ME in prepared pet foods for dogs the 4-steps procedure have been used (Table 1).

Table 1 Calculation of the metabolizable energy (ME) in dog foods (NRC, 2006; FEDIAF, 2021).

1. Calculate gross energy (GE): GE (kcal) = (5.7 × %CP) + (9.4 × %EE) + [4.1 × (%NFE + %CF)]

2. Calculate energy digestibility (ED) (%): ED (%) = 91.2 – (1.43 × %CF in DM)

3. Calculate digestible energy (DE): DE (kcal) = (kcal GE × ED)/100

4. Calculate metabolizable energy (ME): ME (kcal) = DE – (1.04 × %CP)

ME, metabolizable energy; GE, gross energy; ED, energy digestibility; DE, digestible energy; CP, crude protein; EE, ether extract; NFE, nitrogen-free extract; CF, crude fiber; DM, dry matter.

J. Kępińska-Pacelik et al.

Research in Veterinary Science 165 (2023) 105071

3

2.4. Microbiological analyzes

Samples for microbiological analyzes were collected from 35 com- mercial dry dog foods: 23 grain-free and 12 grain-included foods. Food samples were collected for analysis from the bottom, middle and top of the package. They were combined and thoroughly mixed to obtain the laboratory sample. Then, the material for analysis was collected from the laboratory sample, which constituted the analytical sample. Dog food samples for microbiological analyzes were prepared in accordance with ISO 6887–1 (2017b), ISO 7218 (2008). The analyzes included the determination of the total number of mesophilic aerobic bacteria (ISO, 4833–1, 2013), general coliform bacteria (confirmation of lactose fermentation capacity) (ISO, 4832, 2007), Enterobacteriaceae spp. (confirmation test for glucose fermentation and the presence of oxidase) (ISO, 21528–2, 2017a), Staphylococcus spp. (confirmation coagulase test) (ISO, 6888–1, 2021), Salmonella spp. (ISO, 6579, 2020), Proteus spp. (ISO, 21528, 2017a), Enterococcus spp. (ISO, 6579, 2020), Escher- ichia coli (confirmation test for the detection of oxidase) (ISO, 9308, 2014), Clostridium perfringens (confirmation on LS medium) (ISO, 7937, 2005), and yeasts and molds (microscopic examination confirmation) (ISO, 21527–1, 2009). The microbiological analyzes of the dog foods were carried out by the serial dilution method (Pepper and Gerba, 2005). The procedure was the same. Two 10 g portions were prepared from the dog food samples, and then transferred to flasks containing 90 cm3 of NaCl (concentration: 0.85%). The samples were shaken for 30 min. From the obtained suspension, 4 successive 10-fold dilutions were made, which were then inoculated (in 3 replications) in a volume of 1 cm3 onto previously prepared microbiological media. For mesophilic bacteria – trypticase-soy agar (TSA) (incubation conditions: 1 day (24 h) at 37 ◦C ± 1 ◦C, + 3 days at 22 ◦C ± 1 ◦C + 3 days at 4 ◦C ± 1 ◦C, general coliform bacteria – agar with crystal violet, red, neutral bile and lactose (VRBL), Enterobacteriaceae spp. – agar with crystal violet, red, neutral bile and lactose (VRBL), Staphylococcus spp. – Baird-Parker agar, Proteus – MacConkey agar (1 day at 37 ◦C ± 1 ◦C), Salmonella spp. – SS agar, Clostridium perfringens – Wilson-Blair agar, Enterococcus spp. – Slanetz Bartley agar (2 days in 37 ◦C ± 1 ◦C), Escherichia coli – tryptone bile X- glucuronide agar (TBX agar) (1 day at 44 ◦C ± 1 ◦C), and yeasts and molds – malt extract agar (MEA) (4 days at 30 ◦C ± 1 ◦C + 4 days at 22 ◦C

± 1 ◦C). In the research, microbiological media from Biomaxima (Poland) were used. Microbiological media were prepared in accordance with the preparation instructions provided by their manufacturers or in the references. Tests confirming the belonging of bacteria to each of the isolated genera were also performed in accordance with the methods contained in Polish ISO normative acts adapted from EU law. After in- cubation, the number of bacterial and fungal microorganisms (C) per 1 g of dog food was calculated according to the formula (Eq. (1)):

C = N

10− DA ×

V1

V2 (1)

N – average number of colonies grown on the microbiological me- dium [CFU];

D – dilution factor; A – sample weight [g]; V1 – volume of the extraction solution [cm3]; V2 – volume of the plated sample [cm3]. The number of CFU (colony forming units) of microorganisms was

determined by the microbial counting method, converting the results of the determination into kg of dry mass of dog food (CFU/kg DM of dog food).

2.5. Statistical analyzes

One factorial analysis of variance (ANOVA) and principal component analysis (PCA) were carried out using the STATISTICA 13.3 software (TIBCO, 2021). The significance of differences between the means was

assessed using the Newman-Keuls test at P = 0.05. In order to determine the intensity of the relationship between GF/GI and the presence or absence of the identified microorganisms in foods, analyzes of contin- gency tables were performed. These were 2 × 2 contingency tables ((GF/ GI) × (microorganism no/microorganism yes)). Pearson’s χ2 test, Pearson’s χ2 test with Yates’ correction due to the occurrence of ex- pected numbers smaller than 5, and Kendall’s τb test for square arrays were used to assess the above-mentioned relationships. The analysis of contingency tables was performed using STATISTICA 13.3 software. Contrast analysis was used to divide the object variability into two groups (GF and GI).

In order to compare the macronutrients content of the dog foods, their composition (CP, EE, CF, CA, NFE, ME) was determined. The percentage of a given nutrient or metabolic energy in the profile is expressed by an arithmetic mean converted into units on a 9-point scale. For profile comparison, Cohen’s profile similarity coefficient rc was used (Cohen, 1969; Kazimierska et al., 2021). This coefficient value was measured in the range − 1.00 to 1.00, and its interpretation depends on the value: x ≥ + 0.75 (high similarity); + 0.75 > x > +0.30 (moderate similarity); + 0.30 ≥ x ≥ − 0.30 (no similarity); − 0.30 > x > − 0.75 (moderate dissimilarity); x ≤ − 0.75 (high dissimilarity). The closer were the values of rc to boundary values (1/− 1), the stronger was the eval- uated similarity/dissimilarity. Inter-profile analysis was conducted using MS Office 2017.

3. Results

The average protein content differed in a statistical significance for grain-free and grain-included foods (317.8 and 308.5 g/kg DM, respectively) (Table 2). The protein content of the individual foods ranged from 182.8 g/kg DM in food 28 to 482.2 g/kg DM in food 19. In terms of the range of variability, 14 homogeneous groups were sepa- rated. Among the analyzed contrasts (GF versus GI) determined for in- dividual macronutrients, only the content of CF was the same in both types of food (GF and GI).

The average level of fat differed in a statistical significance in grain- included (142.5 g/kg DM) and grain-free foods (149.1 g/kg DM). The fat content ranged from 34.4 g/kg DM (food 28) to 236.9 g/kg DM (food 23). These both values were obtained for grain-free food. Furthermore, food 28 with a fat level of 34.4 g/kg DM was the only food that did not meet the minimum recommended level (MRL). The variability of EE content in the tested set of foods was divided into as many as 24 ho- mogeneous groups (Table 2). The most numerous, because they had three components, were groups m, n and o.

Crude fiber in foods ranged from 25.6 g/kg DM in food 7 to 149.8 g/ kg DM in food 34. The mean level of crude fiber in the grain-free foods (71.2 g/kg DM) did not differ significantly than that in the grain- included (68.0 g/kg DM). The lowest crude ash level of 45.9 g/kg DM was identified in grain-included food (food 22), while the highest was 125.4 g/kg DM in grain-free food (food 33). Statistically significant differences in the results were observed. The grain-free foods contained more crude ash than the grain-included foods (88.9 g/kg DM and 76.3 g/ kg DM, respectively).

The average NFE content of the grain-free diets was significantly lower than that of the grain-included foods (373.1 g/kg DM and 404.7 g/ kg DM, respectively). The lowest value was obtained for grain-free food 19 (190.8 g/kg DM), and the highest for grain-included food 4 (574.3 g/ kg DM).

The significantly lowest metabolic energy level of 13.8 MJ/kg DM was identified in grain-free (foods 13 and 28), while the highest 18.7 MJ/kg of DM in grain-included food (food 22). The values of the metabolic energy did not differ in a statistical significance between the group of grain-free and grain-included foods (15.72 MJ/kg DM and 15.86 MJ/kg DM, respectively).

In order to generalize the differentiation of proximate composition of the analyzed dog foods, a comparison of multivariate proximate

J. Kępińska-Pacelik et al.

Research in Veterinary Science 165 (2023) 105071

4

composition profiles was carried out (Table 3). Out of 595 presented Cohen’s coefficients, as many as 217 (36.47%) had values of ±0.30 (no similarity – white), which means that the remaining 378 coefficients described negative or positive similarities of the analyzed proximate composition profiles (138 (23.19%) and 240 (40.34%) respectively). Only five pairs of the analyzed dog foods were characterized by negative similarity coefficients (negative similarity (dissimilarity) – red color). These were pairs 5 and 28, 11 and 22, 18 and 34, 23 and 28 as well as 25 and 29.

When assessing the composition of the food, profiles in the context of their similarities, the foods with the highest amount of Cohen’s co- efficients in green were foods 21, 31 and 17, and the numbers of these coefficients were 20, 19 and 18, respectively. A certain universalism of their composition is also confirmed by their central location in the graph of factor coordinates for cases (dog foods) (Fig. 1B). PCA analysis shown that the first two components account for 79.59% of the total variance (Fig. 1A). Based on this analysis, seven food groups were distinguished, as illustrated by the ellipses in the plot of factor coordinates (Fig. 1B). The three one-piece set were made of foods 10, 22 and 28. Food 28 was low in fat and food 22 was very high in fat. Another two-element group was made up of foods 13 and 34 with a high Cohen similarity index (0.774), whose common feature was a high crude fiber content, as indicated by the PCA analysis. The next group included five foods (12,

14, 24, 26, 33), the common feature of which was moderate fiber and protein content. The recipe common to these foods was potatoes, assuming, of course, that the hydrolyzed starch in food 26 came from potatoes. The next group included six foods with a very high fat content (1, 5, 19, 20, 23, 25). In the case of four of them (1, 20, 23, 25), the main source of fat seemed to be raw materials obtained from fish, and in the other two cases (5, 19), animal fat supplemented in food 5 with sun- flower oil. The last group consisted of foods located in the third and fourth quadrants of the coordinate system. Quarter 4 food was charac- terized by a high NFE content, while the third quarter food has a low metabolizable energy value (ME) (Fig. 1A, B). Therefore, this group consisted of foods with the most varied recipe compositions, both with and without a grain ingredient. As expected, foods from the most distant ellipses in the coordinate system shown an average or high dissimilarity of the proximate composition profiles expressed by Cohen’s coefficients, and this mainly resulted from an absolutely different recipe composition.

Analysis of 35 dry complete foods for adult dogs for the presence of potentially pathogenic bacteria and fungi showed the growth of at least microbiological indicator in each product. The largest group of micro- organisms detected in dog foods were mesophilic bacteria (Table 4). Their amounts have been found in all tested in our research dog foods. The amounts of bacteria in grain-free foods were in the range of 6.0 ×

Table 2 Proximate composition (g/kg DM) and metabolizable energy (MJ/kg DM) of analyzed dog foods1.

Item2 DM g/kg CP EE CF CA NFE ME

1 GF 954.4op 366.7o 227.0α 81.3defghi 94.5n 230.5b 16.8klmn

2 GI 929.8ghijkl 297.9efgh 129.6k 95.1ghij 62.7b 414.8hi 15.0bcdef

3 GI 922.9efgh 252.6c 143.9p 61.0abcdefg 66.5c 476.0klmn 16.2ghijklm

4 GI 914.8cde 233.3b 88.3b 29.4ab 74.7fg 574.3o 15.9fghijkl

5 GF 957.3p 397.1p 208.1x 84.4efghij 71.2de 239.3b 16.7jklmn

6 GF 934.4ijkl 246.5bc 139.8mn 67.5bcdefgh 114.0q 432.2hij 15.3bcdefgh

7 GF 926.2fghi 282.0de 137.2l 25.6a 89.3kl 466.0jkl 16.8lmn

8 GF 944.1klmn 321.2ijk 100.2ef 90.9fghij 76.8gh 410.9hi 14.5abc

9 GF 888.4b 235.4b 179.2s 54.1abcdef 61.6b 469.7kl 17.1mn

10 GI 914.9cde 343.0lmn 183.1t 55.1abcdef 70.1de 348.8g 17.2mn

11 GI 920.4def 275.4d 115.1i 90.0fghij 115.2q 404.4h 14.2ab

12 GF 940.3klm 326.3jk 140.8no 111.1ij 80.3i 341.5fg 14.5abcd

13 GF 950.3nop 306.9ghi 143.0p 118.1j 122.1r 309.9cdef 13.8a

14 GF 947.9mno 343.3lmn 187.7u 95.2ghij 92.9mn 281.0c 15.7defghijk

15 GI 920.8defg 304.1gh 100.8f 89.4fghij 72.9def 432.9hij 14.6abcd

16 GF 936.1jkl 285.6def 102.9h 56.2abcdef 70.1de 485.2lmn 15.6cdefghij

17 GI 943.7klmn 271.8d 99.4e 44.8abcde 77.6h 506.4mn 15.7efghijkl

18 GI 951.3nop 292.5efg 138.9m 29.3ab 67.8cde 471.6klmn 17.1mn

19 GF 941.4klm 482.2r 172.6r 72.3cdefgh 82.1ij 190.8a 16.4ijklm

20 GF 938.3kl 422.2q 197.3w 48.9abcde 106.8p 224.8b 17.2mn

21 GF 930.2hijkl 254.9c 138.9m 70.9cdefgh 91.9lmn 443.4ijk 15.5cdrfghi

22 GI 910.9c 417.8q 210.6y 37.3abc 45.9a 288.4cd 18.7o

23 GF 926.2fghi 355.7mno 236.9β 104.5hij 83.6j 219.4b 16.3hijklm

24 GF 943.2klmn 363.5o 124.4j 62.2abcdefg 122.1r 327.9efg 15.1bcdefg

25 GI 950.7nop 405.2p 222.6z 52.6abcdef 98.9o 220.7b 17.6n

26 GF 847.8a 314.3hij 193.0v 110.6ij 63.6b 318.5defg 15.6defghij

27 GF 974.4q 326.2jk 92.7c 35.4abc 90.2klm 455.6jkl 15.7efghijkl

28 GF 956.8p 182.8a 34.4a 54.7abcdef 97.3o 630.9p 13.8a

29 GI 939.0klm 274.9d 94.4d 82.2defghij 72.2def 476.3klmn 14.6abcde

30 GF 938.1kl 300.5fgh 152.0q 43.7abcd 87.8k 416.1hi 16.6jklmn

31 GF 924.5fgh 287.5def 101.8g 55.3abcdef 93.8n 461.6jkl 15.2bcdefgh

32 GF 941.5klm 242.9bc 136.9l 49.7abcde 63.8b 506.7mn 16.4ijklm

33 GF 953.7op 356.7mno 141.1o 74.1cdefgh 125.4s 302.7cde 15.0bcdefg

34 GI 925.9fghi 333.5kl 183.9t 149.8k 90.6klm 242.3b 13.9a

35 GF 928.8fghij 308.4ghi 140.6no 70.7cdefgh 63.4b 416.9hi 15.9fghijkl

Contrast Mean of GF 935.8b 317.8b 149.1b 71.2a 88.9a 373.1a 15.72a

Mean of GI 928.8a 308.5a 142.5a 68.0a 76.3b 404.7b 15.86b

MRL – 180.0 55.0 – – – –

DM, dry matter; CP, crude protein; EE, ether extract; CF, crude fiber; CA, crude ash; NFE, nitrogen free extract; ME, metabolizable energy; GF, grain-free dog food; GI, grain-included dog food; MRL, Minimum Recommended Level for adult dogs based on MER of 110 kcal kg0.75 (FEDIAF, 2021);

1 Means with at least one same letter in the superscript (a, b, c…) not differ statistically at P = 0.05 (and in columns, separately for foods and contrast); 2 the main ingredients of analyzed dog foods and division into cereal and cereal-free foods are shown in Table S1.

J. Kępińska-Pacelik et al.

Research in Veterinary Science 165 (2023) 105071

5

104–1.9 × 107 CFU/kg DM, and in grain-included foods in the range of 7.0 × 104–4.6 × 106 CFU/kg DM. The post-hoc analysis allowed to distinguish ten homogeneous groups in which both GF and GI foods were located, as evidenced by, among others, group of the most contaminated foods composed of foods 14 (GF) and 34 (GI). The assessment of the microbiological purity of dog foods included groups of microorganisms acting as the so-called hygiene indicators. These in- dicators allow to determine the intensity of microbial contamination and to determine the origin of contamination and the approximate time in

which contamination took place. The presence of Salmonella spp. is the basic indicator of the sanitary and hygienic condition of the pet food. However, in our study, no Salmonella spp. was detected in any of the foods. Other microbiological indicators of dog food quality are co- liforms. The coliform bacteria include genera: Escherichia, Citrobacter, Klebsiella and Enterobacter. Purple red colonies of at least 0.5 mm in diameter grown on VRBL medium were considered typical coliform colonies requiring no further confirmation. However, in the case of atypical colonies, their presence was confirmed in accordance with the

Table 3 Cohen’s profile similarity coefficient.

No 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

2 -0.348 -

3 -0.292 0.506 -

4 -0.585 0.277 0.826 -

5 0.865 -0.041 -0.175 -0.564 -

6 -0.038 -0.090 0.189 0.407 -0.452 -

7 -0.040 -0.192 0.660 0.791 -0.168 0.456 -

8 -0.481 0.799 0.179 0.283 -0.310 0.206 -0.187 -

9 0.021 0.245 0.906 0.606 0.097 0.040 0.653 -0.218 -

10 0.598 -0.082 0.413 0.101 0.721 -0.265 0.505 -0.439 0.660 -

11 -0.164 0.138 -0.128 0.136 -0.471 0.840 -0.001 0.564 -0.395 -0.615 -

12 0.058 0.686 -0.168 -0.317 0.149 0.095 -0.593 0.756 -0.375 -0.357 0.517 -

13 0.168 0.075 -0.465 -0.349 -0.111 0.592 -0.389 0.436 -0.613 -0.574 0.875 0.703 -

14 0.853 -0.015 -0.372 -0.636 0.712 0.147 -0.306 -0.023 -0.228 0.234 0.247 0.547 0.596 -

15 -0.570 0.867 0.357 0.419 -0.376 0.199 -0.098 0.978 -0.040 -0.392 0.491 0.678 0.306 -0.146 -

16 -0.596 0.565 0.827 0.931 -0.433 0.252 0.617 0.546 0.539 0.104 0.128 -0.035 -0.298 -0.538 0.665 -

17 -0.560 0.346 0.796 0.986 -0.520 0.428 0.769 0.403 0.538 0.096 0.210 -0.202 -0.265 -0.558 0.516 0.962 -

18 -0.101 0.060 0.817 0.784 -0.003 0.063 0.889 -0.153 0.832 0.679 -0.365 -0.568 -0.699 -0.416 -0.021 0.708 0.754 -

19 0.734 -0.216 -0.456 -0.562 0.849 -0.393 -0.160 -0.170 -0.286 0.518 -0.290 0.129 0.023 0.623 -0.312 -0.423 -0.472 -0.106

20 0.852 -0.600 -0.341 -0.365 0.692 0.038 0.247 -0.507 -0.085 0.601 -0.140 -0.230 0.049 0.606 -0.623 -0.436 -0.326 0.098

21 -0.217 0.311 0.610 0.682 -0.456 0.874 0.560 0.399 0.398 -0.106 0.680 0.162 0.344 -0.020 0.471 0.612 0.701 0.342

22 0.531 -0.135 0.242 -0.052 0.752 -0.551 0.314 -0.498 0.514 0.944 -0.812 -0.410 -0.703 0.127 -0.464 -0.021 -0.066 0.569

23 0.921 -0.055 -0.282 -0.703 0.886 -0.180 -0.330 -0.295 -0.009 0.468 -0.168 0.337 0.241 0.903 -0.370 -0.620 -0.673 -0.288

24 0.273 -0.361 -0.393 -0.049 -0.048 0.661 0.207 0.166 -0.493 -0.174 0.706 0.161 0.669 0.419 0.015 -0.105 0.047 -0.191

25 0.916 -0.551 -0.235 -0.389 0.790 -0.063 0.235 -0.613 0.085 0.725 -0.296 -0.267 -0.075 0.619 -0.695 -0.462 -0.373 0.157

26 0.460 0.589 0.110 -0.419 0.656 -0.316 -0.457 0.184 0.199 0.302 -0.172 0.640 0.133 0.621 0.192 -0.185 -0.381 -0.211

27 -0.340 0.018 0.448 0.804 -0.380 0.474 0.795 0.312 0.218 0.127 0.295 -0.263 -0.120 -0.380 0.331 0.760 0.859 0.652

28 -0.741 0.363 0.479 0.789 -0.838 0.654 0.401 0.613 0.118 -0.480 0.642 0.101 0.215 -0.496 0.682 0.757 0.809 0.244

29 -0.694 0.808 0.555 0.649 -0.548 0.312 0.123 0.882 0.160 -0.361 0.469 0.466 0.165 -0.345 0.954 0.813 0.711 0.180

30 0.209 -0.193 0.618 0.648 0.072 0.439 0.965 -0.238 0.670 0.659 -0.034 -0.501 -0.324 -0.062 -0.170 0.501 0.642 0.857

31 -0.453 0.212 0.473 0.807 -0.591 0.744 0.645 0.531 0.158 -0.180 0.641 0.024 0.215 -0.292 0.560 0.770 0.864 0.426

32 -0.346 0.399 0.986 0.879 -0.234 0.172 0.735 0.110 0.898 0.418 -0.178 -0.308 -0.552 -0.484 0.289 0.846 0.840 0.880

33 0.394 -0.355 -0.487 -0.226 0.042 0.652 0.049 0.125 -0.543 -0.211 0.732 0.274 0.785 0.580 -0.036 -0.276 -0.136 -0.354

34 0.347 0.367 -0.441 -0.653 0.317 0.056 -0.742 0.403 -0.485 -0.333 0.436 0.896 0.774 0.753 0.296 -0.454 -0.572 -0.778

35 -0.145 0.727 0.840 0.575 0.183 -0.170 0.390 0.381 0.729 0.539 -0.286 0.122 -0.466 -0.162 0.502 0.759 0.610 0.692

No 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34

20 0.805 -

21 -0.545 -0.228 -

22 0.600 0.558 -0.412 -

23 0.654 0.602 -0.266 0.430 -

24 0.315 0.536 0.324 -0.315 0.051 -

25 0.775 0.969 -0.263 0.682 0.706 0.337 -

26 0.294 -0.007 -0.103 0.273 0.747 -0.374 0.144 -

27 -0.104 0.077 0.542 -0.010 -0.589 0.460 -0.062 -0.594 -

28 -0.719 -0.586 0.788 -0.640 -0.763 0.215 -0.693 -0.436 0.660 -

29 -0.528 -0.707 0.623 -0.470 -0.543 -0.054 -0.762 0.035 0.458 0.833 -

30 0.032 0.432 0.527 0.445 -0.075 0.260 0.440 -0.272 0.703 0.229 0.008 -

31 -0.392 -0.182 0.821 -0.392 -0.601 0.498 -0.321 -0.477 0.886 0.894 0.690 0.542 -

32 -0.465 -0.317 0.568 0.267 -0.385 -0.378 -0.226 -0.037 0.527 0.495 0.510 0.667 0.502 -

33 0.335 0.548 0.287 -0.355 0.224 0.972 0.374 -0.217 0.253 0.100 -0.136 0.137 0.350 -0.500 -

34 0.243 -0.013 -0.031 -0.362 0.604 0.186 -0.018 0.680 -0.571 -0.215 0.064 -0.600 -0.289 -0.582 0.372 -

35 -0.036 -0.225 0.298 0.454 -0.053 -0.426 -0.128 0.408 0.340 0.214 0.554 0.414 0.269 0.798 -0.512 -0.203

x ≥ +0.75 (high similarity); +0.75 > x > +0.30 (moderate similarity); +0.30 ≥ x ≥ -0.30 (no similarity); -0.30 > x > -0.75 (moderate dissimilarity); x ≤ -

0.75 (high dissimilarity).

J. Kępińska-Pacelik et al.

Research in Veterinary Science 165 (2023) 105071

6

recommendation of the ISO standard, 4832, 2007. In addition to coli- form bacteria, the function of indicator organisms is also performed by bacteria of the genus Enterococcus spp., Proteus spp., and Clostridium perfringens. In the dog food samples, their presence was found at the following levels: in the case of coliform bacteria – from 5.0 × 103 CFU/ kg DM to 2.9 × 105 CFU/kg DM, for Proteus spp. From 1.0 × 104 CFU/kg DM to 7.0 × 104 CFU/kg DM, for Escherichia coli from 1.0 × 104 CFU/kg DM to 2.4 × 105 CFU/kg DM, for Clostridium perfringens from 2.0 × 104

CFU/kg DM to 4.7 × 105 CFU/kg DM. Enterococcus spp. were identified in only one grain-included food (food 15) at the level of 5.0 × 104 CFU/ kg DM. The presence of Clostridium perfringens was identified in 17.39% of the grain-free and 16.66% of the grain-included foods. Coliform bacteria were present in 52.17% of the analyzed grain-free foods and 66.66% of the grain-included foods. Bacteria of the genus Proteus spp. were identified in 34.78% of grain-free foods and 58.33% of grain- included foods. The bacteria of the genus Staphylococcus spp. were the most numerous group of bacteria, 91.30% in grain-free foods and 91.66% in grain-included foods. Based on our research, the presence of both coagulase-positive staphylococci (which makes them an indicator of dangerous contamination) and coagulase-negative staphylococci were found in the tested samples. Due to their nature, staphylococci are often included in the so-called opportunistic or relatively commensal microbiota. They can change their character from saprophytic to

parasitic, provided that there is an additional “external” factor, unre- lated to the properties of the microorganism (the so-called infection- promoting factor). The total number of yeasts and molds in the dog food samples was in the range from 2.5 × 104 CFU/kg DM to 5.0 × 104 CFU/ kg DM (Table 4). The presence of fungi was identified in four dog foods only. The analysis of the results showed that the types of fungi such as Aspergillus, Penicillium and Fusarium were the dominant organisms in the tested foods.

The analysis of the contingency tables prepared for individual microbiological contaminants basically showed no relationship between the degree of microbiological contamination and the absence or pres- ence of the grain component in the food. Relatively close to doc- umenting such a relationship was the content of TYMC. This is confirmed by the value of Pearson’s χ2 coefficient close to the signifi- cance level (α = 0.068). Unfortunately, the presence of expected numbers less than five in the contingency table requires Yates’s correction to be taken into account. The level of significance after its introduction drops to 0.207, and Kendall’s τb coefficient is 0.308, which shows a slight correlation between the presence or absence of a grain component and the presence in TYMC food. It should be emphasized that in this research the relationship between GI/GF and TYMC preva- lence has not been studied. It seems, therefore, that a thesis (statistical tendency) can be formulated stating that TYMC is more frequent in foods containing a grain component. The GF/GI relationship and the presence of Proteus spp. were at a slightly lower level than the above-mentioned one. In this case, the expected numbers were not less than five, which allows to recall the significance level of the analyzed compound as being 0.181. This allows us to formulate a thesis (statistical tendency) that, as in the case of TYMC, the addition of a grain component increases the probability of Proteus spp. (τb of Kendall = 0.226) appearing in the food. The remaining microbiological contamination will not be discussed due to the very low values of the significance levels of the analyzed relationships.

4. Discussion

4.1. Proximate composition

The FEDIAF (2021) provides minimum recommended levels only for protein (180.0 g/kg DM) and fat (55.0 g/kg DM). All dog foods, both grain-included and grain-free, met the MRL for protein. Our results confirm the results obtained by Kahraman and Inal (2021) who found a significantly higher level of protein in grain-free foods (370.0 g/kg DM) compared to grain-included foods (290.8 g/kg DM). This dependency is confirmed by Stercova et al. (2022), whose analyzes showed that grain- included foods had 237.5–270.2 g/kg DM protein, while the grain-free foods – 324.9 g/kg DM, which means that grain-free foods might be a better source of protein compared to grain-included foods. However, Corsato Alvarenga and Aldrich (2020), reported that grain-included foods were richer in protein compared to grain-free foods (282.4 g/kg DM and 266.2 g/kg DM, respectively). FEDIAF does not specify the maximum allowable protein level. However, as shown by Gebreselassie and Jewell (2019), high protein consumption can lead to increased fecal levels of indole and branched-chain fatty acids and reduce levels of the short-chain fatty acids. Moreover, the consumption of high protein food increased the proportions of the genera Clostridium and Streptococcus in the intestine, that are implicated to have negative health associations. This was accompanied by a reduction in the beneficial bacteria belonging to the genera Prevotella and Bifidobacterium (Pieper et al., 2012; Gebreselassie and Jewell, 2019).

Another essential ingredient is fat, which determines the palatability of the pet food (Inal et al., 2020). Fat is also the most energy dense of the macronutrients, providing more than double the calories per gram provided by either protein or carbohydrate (Tolbert et al., 2022). FEDIAF does not specify the maximum allowable level of fat. Deficiency of fat and fatty acids in the diet leads to body weight loss, weakening of

Fig. 1. Biplot based on first two principal component axes for proximate composition and metabolic energy of dog foods (A) and distribution of 35 dry dog foods based on the first two components obtained from principal compo- nent analysis (B). CP, crude protein; EE, ether extract; CF, crude fiber; CA, crude ash; NFE, ni- trogen free extract; ME, metabolizable energy.

J. Kępińska-Pacelik et al.

Research in Veterinary Science 165 (2023) 105071

7

the quality of the hair and skin (Lenox, 2016). Excessive fat consumption leads to the development of obesity (Muñoz-Prieto et al., 2018). In the studies by Kahraman and Inal (2021), the level of fat in grain-free foods averaged 140.8 g/kg DM and was higher than in grain-included foods (112.5 g/kg DM), however in the studies of Stercova et al. (2022) the grain-free diets also had more fat than these grain-included (149.9 g/kg

DM, 137.1 g/kg DM, respectively). The same situation was observed in our study – grain-free foods were richer in crude fat in comparison with grain-included foods.

Usually, the main source of carbohydrates in grain-included dog foods are corn, wheat, rice, oats, barley, while in grain-free – potatoes, sweet potatoes, peas, lentils, and tapioca (Table S1). Importantly, results

Table 4 Analysis of the content of microorganisms in dry dog food (CFU/kg DM).

Item TAMBC* TYMC E. coli Salmonella spp. Clostridium perfringens

Coliform bacteria

Proteus spp.

Enterococcus spp.

Staphylococcus spp.

1 (GF) abc2.1 × 105

ND 1.0 × 104 ND ND ND ND ND ND

2 (GI) fg8.1 × 105 ND ND ND ND ND ND ND 2.0 × 105

3 (GI) bcde4.3 × 105

ND ND ND ND 4.5 × 104 3.5 × 104 ND 1.3 × 105

4 (GI) abc2.2 × 105

ND ND ND 3.0 × 104 9.0 × 104 6.0 × 104 ND 8.0 × 104

5 (GF) abc1.7 × 105

ND ND ND ND 4.0 × 104 1.5 × 104 ND 8.5 × 104

6 (GF) k3.6 × 106 ND 6.5 × 104 ND ND 2.0 × 104 ND ND 8.0 × 104

7 (GF) cde4.4 × 105

ND 1.5 × 102 ND ND 2.0 × 104 7.0 × 104 ND 1.4 × 105

8 (GF) n1.9 × 107 ND ND ND ND 3.0 × 104 ND ND 1.7 × 105

9 (GF) i2.4 × 106 ND 2.0 × 104 ND ND ND ND ND 2.3 × 105

10 (GI) l4.6 × 106 ND ND ND ND 3.0 × 104 1.0 × 104 ND 2.0 × 104

11 (GI) efg6.8 × 105

ND ND ND ND 2.0 × 104 ND ND 1.0 × 104

12 (GF) m5.1 × 106 ND 4.0 × 104 ND ND 2.0 × 104 ND ND 2.4 × 105

13 (GF) fg7.6 × 105 ND ND ND ND 2.0 × 104 1.5 × 104 ND 1.8 × 105

14 (GF) j2.7 × 106 ND 1.0 × 104 ND ND 3.0 × 104 3.0 × 104 ND 1.6 × 105

15 (GI) abcd3.4 × 105

4.5 × 104 1.0 × 104 ND ND 2.9 × 105 ND 5.0 × 104 3.0 × 104

16 (GF) defg6.2 × 105

ND ND ND ND 2.0 × 104 ND ND 2.5 × 105

17 (GI) a7.0 × 104 2.5 × 104 ND ND ND ND ND ND 8.5 × 104

18 (GI) h1.8 × 106 3.0 × 104 1.5 × 104 ND ND 5.0 × 104 2.5 × 104 ND ND 19 (GF) a7.5 × 104 ND 2.4 × 105 ND ND 1.0 × 104 1.0 × 104 ND ND 20 (GF) abc2.3 ×

105 ND ND ND ND ND ND ND 2.6 × 105

21 (GF) abc1.2 × 105

ND 2.5 × 104 ND 4.7 × 105 ND 4.0 × 104 ND 4.0 × 104

22 (GI) abc1.5 × 105

ND ND ND 1.3 × 105 1.0 × 104 1.0 × 104 ND 4.0 × 104

23 (GF) abc2.5 × 105

ND ND ND ND 5.0 × 103 ND ND 3.5 × 104

24 (GF) abcd3.7 × 105

ND ND ND 1.0 × 105 ND ND ND 1.3 × 105

25 (GI) cdefg4.7 × 105

ND 1.5 × 104 ND ND ND ND ND 7.0 × 104

26 (GF) cde4.5 × 105

ND 1.0 × 104 ND ND ND ND ND 4.5 × 104

27 (GF) a6.0 × 104 ND ND ND 2.9 × 105 3.0 × 104 ND ND 6.0 × 104

28 (GF) i2.3 × 106 ND ND ND ND 7.0 × 104 4.0 × 104 ND 1.9 × 105

29 (GI) fg7.4 × 105 ND ND ND ND 4.0 × 104 5.0 × 104 ND 1.0 × 105

30 (GF) abc2.9 × 105

ND ND ND ND ND ND ND 7.0 × 104

31 (GF) abc2.4 × 105

ND ND ND 2.0 × 104 ND ND ND 4.0 × 104

32 (GF) abc1.9 × 105

ND ND ND ND ND ND ND 1.3 × 105

33 (GF) fg7.3 × 105 ND 3.0 × 104 ND ND ND 6.0 × 104 ND 4.0 × 104

34 (GI) j2.7 × 106 ND 1.0 × 104 ND ND ND 7.0 × 104 ND 3.0 × 104

35 (GF) ab8.5 × 104 5.0 × 104 ND ND ND ND ND ND 2.1 × 105

χ2 Pearson’ – α = 0.068 α = 0.561 – α = 0.957 α = 0.411 α = 0.181 – α = 0.971 χ2 Yates’ – α = 0.207 α = 0.827 – α = 0.676 α = 0.643 α = 0.329 – α = 0.549 τb Kendall’ – 0.308 0.098 – 0.009 0.139 0.226 – 0.006 Rejection

limit – 1.0 × 101

CFU/kg absence in 10− 2 kg

absence in 2.5 × 10− 2 kg

– – – – –

Reference – GMP (2005) CR (EU) No 142/2011

GMP (2005) – – – – –

%GF 100 4.34 43.48 0.00 17.39 52.17 34.78 0.00 91.30 %GI 100 25.00 33.33 0.00 16.66 66.66 58.33 8.33 91.66

CFU, colony forming unit; DM, dry matter; ND, not detected; %GF, percentage of contaminated grain-free dog foods; %GI, percentage of contaminated grain-included dog foods; TAMBC, total mesophilic bacterial count; TYMC, total yeast and mold count; * means with at least one some letter in the superscript (a, b, c…) not differ statistically at P = 0.05 (Newman-Keuls test)

J. Kępińska-Pacelik et al.

Research in Veterinary Science 165 (2023) 105071

8

showed that the sources of carbohydrates and use of balanced diets affected only some biochemical parameters and did not alter the levels of taurine in healthy adult dogs which was previously questionable (Gizzarelli et al., 2021). Potato starch is reportedly more digestible and palatable than corn starch, partly as a result of its lower density (Murray et al., 1999; Thompson, 2008). Therefore, potato is now considered as an especially appropriate carbohydrate source for puppies and dogs with food allergies (Domingues et al., 2019). Studies have determined diets utilizing potato allow for greater nutrient digestion than those con- taining grains (Chiofalo et al., 2019). As reported by Chiofalo et al. (2019), the grain-free dry foods offer higher apparent nutrient di- gestibility of protein (+ 10%) and fat (+ 7%) and more stable large intestinal fermentation of carbohydrate compared to the commercial grain-included dry diet. Some caregivers believe that dogs – as a result of their carnivorous ancestry – do not require carbohydrates which are provided by grain components. Consuming carbohydrates is beneficial in many ways. What is more, once the carbohydrates are absorbed, they will be used by the dog’s body to meet its glucose needs (Laflamme et al., 2014). As expected, the average carbohydrate content of the grain-free foods in our study was significantly lower than that of the grain- included foods, which confirm the research of other authors. The same relation was observed by Kahraman and Inal (2021), in which the grain- free foods contained 311.5 g/kg DM carbohydrates, while grain- included foods – 407.4 g/kg DM. Stercova et al. (2022) in their studies also obtained an average lower level of carbohydrates in grain-free foods compared to grain-included foods (418.3 and 527.0 g/kg DM, respectively).

Crude fiber is the part of the dietary fiber that increases the volume of fecal masses, accelerates peristalsis, and reduces digestibility (Kienzle et al., 2001). But on the other hand, it gives a feeling of fullness, what can be important for dogs with excess body weight (Heuberger and Wakshlag, 2011; Weber et al., 2007). As showed by Alegría-Morán et al. (2019), high levels of crude fiber content have negative impact on dogs’ food preferences, and dogs are more likely to intake foods with a lower crude fiber content. Legumes such as peas contain greater concentration of soluble fibers compared to cereal grain (de Oliveira et al., 2012). Similarly as in our study, a higher average level of crude fiber in grain- free foods compared to grain-included foods was obtained by Stercova et al. (2022), (23.0 g/kg DM and 18.2 g/kg DM, respectively). Kahraman and Inal (2021) observed an opposite relation. In these studies, the average level of crude fiber in grain-free foods was 41.2 g/kg DM, while in grain-included foods – 57.2 g/kg DM.

In our study, grain-free foods contained significantly more crude ash than grain-included. Similar results were obtained by Kahraman and Inal (2021), (83.2 g/kg DM, 77.0 g/kg DM, respectively). In studies by Stercova et al. (2022) the grain-free foods also had more crude ash compared to the grain-included foods (86.0 and 66.7 g/kg DM, respec- tively). Higher quality meat will have a lower crude ash level because it contains less bone than bone meal, the crude ash content may also vary depending on the type of animal component used, chicken and fish have naturally less crude ash, while red meat as beef has more. A high crude ash content may also suggest that bone/animal meal is the main component of the food, which is poorly digestible. The excess of bone elements increases the crude ash content. As reported by Johnson et al. (1998), bone meals contain 24–34% crude ash, poultry by-product meals – 7–16%, lamb meals – 15–24%. Crude ash is inorganic matter that is left over from the combustion of organic matter. In its form, minerals are expressed. Importantly, pet food manufacturers have a legal obligation to list its content on the product label (FEDIAF, 2019).

In our study, grain-free foods had a insignificantly lower metabolic energy level than grain-included foods (15.72 MJ/kg DM and 15.86 MJ/ kg DM, respectively). Kahraman and Inal (2021) observed an inverse relation (grain-included 15.1 g/kg DM, grain-free 15.8 g/kg DM), as did Stercova et al. (2022) – grain-free foods had an energy content of 17.3 g/ kg DM, and grain-included foods had the content of 16.7 g/kg DM. Precise calculation of the ME allows the caregiver to individually dose

the pet food. The doses provided on the food packaging do not take into account the individual specific energy demand, individual housing conditions, and other factors (FEDIAF, 2021). Providing too much metabolic energy and too much food in a short time can lead to obesity. Metabolic energy is not an ingredient – it is provided in the form of protein, fat and carbohydrates. It has been shown that dogs when choosing food are guided not only by palatability but also by energy content (Hall et al., 2018).

4.2. Microbiological safety

Safety of pet foods is understood as the lack of physical, chemical and microbiological threats. Food safety in the biological context is mainly determined by the absence of pathogenic microorganisms. Assessment of the microbiological condition of companion animal foods is an important element of nutritional safety for the animals themselves and their caregivers. Dry food can be a source of infection, especially in the elderly, who often have reduced immunity (White et al., 2018). An important aspect is also the possibility of transmission of pathogens between animals. Direct contact is one of the main methods of disease transmission between animals. It occurs when a susceptible animal comes in a direct contact with an infected animal, its body fluids or tissues. Depending on the microorganism, it may be transferred directly by blood, saliva, urine, or feces (Fong, 2017).

The most popular type of dog food are kibbles – dry food which is easily available, easy to dose and store, and can be also stored for a long period of time. Kibbles are most often obtained by extrusion. Extrusion has been used to produce pet foods since 1954. Since then, plant-based ingredients have been included in pet food formulas to provide the necessary starch for proper processing and kibble formation. In addition to starch, these plant-based feedstuffs also contain different types and concentrations of fiber. Often their size is adapted to the size of the breed for which the type of pet food is intended (Donadelli et al., 2021).

Cereals and cereal products have been used for years in the case of dry pet food as a source of energy, but most of all they are a structure- forming raw material that determines the viscosity, brittleness, mois- ture, and strength of the kibble. It is not possible to produce dry food without such ingredients, and here comes the problem of choosing the type of carbohydrate sources in dry food. For some time, grain-free in- gredients (such as legumes seeds) have been used. It is presumed that grain-included foods pose a greater risk of the presence of certain bac- teria and fungi due to the fact that they may be present in the cereal raw material, but also the raw material may become contaminated during storage and transport (Donadelli et al., 2021).

According to the FEDIAF Guide to Good Practice for the Manufacture of Safe Pet Food (2018), the list of biological hazards that may occur in pet food includes the following types and species of bacteria: Aeromonas, Campylobacter, Clostridium perfringens, Clostridium botulinum, Enterobac- teriaceae, Listeria monocytogenes, Escherichia coli, Salmonella, Staphylo- coccus aureus. In recent 20 years in RASFF (2022) 190 cases of detections of microorganisms in pet food and feed materials were reported. As reported in the literature on the subject, monitoring these biological threats is important as there is a risk of spreading resistant bacterial genes to other animals and also to people in the household and the environment. Antibiotic resistance of pathogenic microorganisms has been a topic discussed in the scientific literature for many years. In the developing world, multi-resistant bacteria to various antibiotics is becoming a serious problem in the treatment of diseases (Bacanlı and Başaran, 2019). Worryingly, research shows that dogs are one of the animal species from which drug-resistant pathogens are isolated (Stolle et al., 2013; de Jong et al., 2018; Loncaric et al., 2020). According to Ma et al. (2021) antimicrobial agents are now widely used in food- producing animals, which can be considered as a reservoir of antibiotic-resistant bacteria and can be transmitted to humans and an- imals. Homologous relationships between drug-resistant bacteria in humans and animals have been identified in the most common food-

J. Kępińska-Pacelik et al.

Research in Veterinary Science 165 (2023) 105071

9

borne pathogens, such as E. coli and Salmonella, different types of enterococci, and methicillin-resistant Staphylococcus aureus (MRSA), (Ma et al., 2021).

Currently, there are no strict regulations regarding maximum limits for individual biological contaminants in pet food. According to EU law, “the feed business is primarily responsible for feed safety” (Regulation (EC) No 183/2005), but many publications have revealed problems with the quality of pet food and its impact on human and animal health (Witaszak et al., 2020; Bottari et al., 2020). In the EU, the safety of raw materials and food products is controlled by relevant authorities, including the Rapid Alert System for Food and Feed (RASFF).

In the assessment of the microbiological purity of dog foods, in- dicators were used, which are also used in the assessment of food safety. Determination of the presence of indicator microorganisms in food in- dicates the possible presence of enteric pathogens. These microorgan- isms are part of the physiological microbiota of the digestive tract of humans and animals. The presence and number of indicator microor- ganisms informs about the microbiological quality of raw materials, additives and final products, as well as the hygienic conditions of the technological process (Nemser et al., 2014; Kępińska-Pacelik and Biel, 2021). The most important biological threat in pet foods are bacteria of the genus Salmonella, which are the main bacterial zoonotic agent (Behravesh et al., 2010; Girio et al., 2012).

Dogs could be silent carriers and spread Salmonella species to the environment. Salmonella species can persist in food bowls despite cleaning and disinfection, and the risk is as high as 71% (Bottari et al., 2020). In our study the presence of these bacteria was not detected in any of the analyzed dog foods. However, it does not mean that dry foods are safe in this regard. In study by Selmi et al. (2011) nine of ten samples (90%) of dehydrated pet food were positive. Hellgren et al. (2019) research suggests that the presence of these bacteria is more likely in a dog’s diet based on raw meat and offal that have not received any treatment to reduce the probability of food contamination with patho- genic bacteria.

The next place among the microbiological hazards occurring in dog foods is occupied by microorganisms that have developed the ability to produce spores resistant to adverse environmental conditions (bacteria of the genus Clostridium). Clostridium perfringens is the most important pathogen among spore anaerobes. It is a constant component of the microbiota of the digestive tract of humans and warm-blooded animals, with the number of Clostridium perfringens cells excreted with feces is much lower than the number of fecal streptococci or Escherichia coli (Ryan et al., 2010; Milanov et al., 2018; Kępińska-Pacelik and Biel, 2021). A study indicated that up to 28% cases of diarrhea in dogs may occur due to Clostridium perfringens poisoning (Weese, 2001). Poisoning with these bacteria can even lead to fatal non-traumatic gas gangrene (Sprohnle-Barrera et al., 2022). In the studies of Viegas et al. (2020) it was shown that dogs fed RMBD were more likely to be positive for C. perfringens than those receiving commercial dry food. In the study by Kazimierska et al. (2021) bacteria of the genus Clostridium were not detected in any of the dry dog foods. However, in our research we found that 17.39% of the grain-free and 16.66% of the grain-included foods contained C. perfringens. Although these numbers are not large, it should nevertheless be worrying.

Escherichia coli is the best known and most widespread representative of the Enterobacteriaceae family, and at the same time the most commonly used microorganism to assess the degree of contamination of the sanitary and hygienic environment (water, soil) or food, pet food and feed. Most representatives belong to the commensal microbiota of humans and animals, but there are also pathogenic groups (Nemser et al., 2014). The research conducted in 2020 identified the presence of Escherichia coli at the level of 12.5% in pet foods and feeds (Ge et al., 2020). In our research, the levels of E.coli found exceeded the safe level regulated by the EU regulation 142/2011. E. coli levels found varied from 1.0 × 104 CFU/kg DM to 2.4 × 105 CFU/kg DM, while the legal limit is 1.0 × 10− 2 (Regulation 142/, 2011).

In our study, grain-free foods were found to be contaminated with E. coli to a greater percentage than grain-included foods (43.48% and 33.33% respectively). We assume that grain-included foods tend to be low-price, which is due to less labor involved in their production, which may be related to their microbiological safety.

Nowadays, many of pet foods are produced by extrusion. The main advantages of extrusion are undoubtedly the process parameters (high temperature and pressure – usually 150–180 ◦C and 5.5 MPa), which are conducive to greater retention of amino acids and vitamins, lower lipid oxidation and high protein and starch digestibility. Moreover, the extrusion process sterilizes the product (Singh et al., 2007). It has been shown that the extrusion process effectively reduced pathogenic mi- croorganisms, therefore it seems all the more justified to analyze whether dry commercial dog food is safe in this respect (Leiva et al., 2019).

The presence of bacteria of the genus Proteus and Enterecoccus in the pet food samples may be treated as an indicator of their reduced hy- gienic condition (Hołda et al., 2017). It is difficult to find information about contamination of Proteus spp. in dry dog foods in the latest available literature on the subject. Our research has shown that they were present in 34.78% of grain-free foods and in 58.33% of grain- included foods. The genus Enterococcus is often used as an indicator of antibiotic resistance in food chain studies. Multidrug resistance is re- ported among foods of animal and plant origin, which are common in- gredients in pet food (Bacci et al., 2019; Davies et al., 2019). Ge et al. (2020) found that 463 of 1025 (45.2%) pet foods and feeds were contaminated with these bacteria. In the studies by Finisterra et al. (2021) Enterococcus spp. were found in 7 out of 8 (87.5%) commercial dry dog foods tested. In our research, these bacteria were found in the only one grain-included food, contrary to the results cited before.

Bacteria of the genus Staphylococcus found favorable living condi- tions, which resulted in their presence in 32 dog foods. Their presence in the tested samples may indicate the extraordinary ability of staphylo- cocci to colonize the products in which they survive, the adaptability resulting from the flexibility of biochemical features, which is condi- tioned by the systems of genetic regulation of metabolism, such as the sigma system of normal metabolism, as well as the accessory gene regulator (AGR) and staphylococcal accessory regulator (SAR). Among them, the most numerous species were coagulase-negative staphylo- cocci, which are naturally part of the permanent or transient microbiota of humans and other mammals, which makes them an indicator of a specific and dangerous biological contamination occurring in pet food (Kępińska-Pacelik and Biel, 2021). Moreover, there is no legal act setting the maximum allowable level of these bacteria in dog food. In our research, the presence of these bacteria was found in 91.30% of grain- free and 91.66% of grain-included foods. For example in the study by Filipović et al. (2010), no bacteria of this species have been detected in extruded dog foods. On the other hand, in other study in five (14%) dog foods the presence of Staphylococcus was detected (Kazimierska et al., 2021). Certain strains of the genus Staphylococcus spp. are important pathogens causing a broad range of diseases including community ac- quired and nosocomial skin and soft tissue infections, and life threat- ening conditions such as bloodstream infections (Sahin-Tóth et al., 2021). It is an opportunistic pathogen frequently isolated from healthy canines and, more importantly, associated with numerous infections in animals. Dogs are the most common animal species infected with e.g. S. pseudintermedius, with 84.7% of all S. pseudintermedius isolates origi- nating from canine diseases including skin, ear and urinary tract in- fections (Ruscher et al., 2009; Lynch and Helbig, 2021).

According to Andrade and Nascimento (2005), one of the risk factors for animal health is contamination of food by fungi. The analysis of the tested dog food samples in this study showed the presence of both molds and yeasts. Certain fungi that coexist with cereals in the field (often referred to as “field fungi”) can form mycotoxins immediately before or just after harvesting, especially species of the genera Alternaria, Fusa- rium, Aspergillus and Penicillium (Hocking, 2003). Our research showed

J. Kępińska-Pacelik et al.

Research in Veterinary Science 165 (2023) 105071

10

the presence of fungi in some dog food samples. The conducted research showed that the types of fungi such as Aspergillus, Penicillium and Fusa- rium were the dominant organisms in the tested dog foods. Fungi of the genus Penicillium were the most numerous. It was found that grain- included foods were more susceptible to their presence than grain- free. The cited literature confirm that fungi can be a microbiological contamination of dog food. The composition of pet food may contain several raw materials of plant origin contaminated with various species of fungi at the same time. This is not a new problem, as already in the research conducted by Abarca et al. (1994), which included checking the microbiological condition of dry dog food, it was found that the molds present in the food were mainly of the genus Aspergillus, Penicillum and Fusarium. Similar results were obtained by Bueno et al. (2001), who found fungi belonging to the genus Aspergillus, Rhizopus and Mucor in pet food samples. Current literature data indicate that up to 38% of pet food may be contaminated with mold fungi (Witaszak et al., 2020). Our research has shown that grain-included foods are five times more contaminated with fungi compared to grain-free foods (25.00% and 4.34% respectively). In our research, the levels of fungi found exceeded the safe levels regulated by Good Manufacturing Practice (GMP, 2005).

Previous study conducted by our team (Kazimierska et al., 2021) showed that all analyzed grain-free and grain-included dry dog foods met the MRL established by FEDIAF (2021). In the previous study, 36 commercial dry extruded complete grain-free and grain-included foods for adult dogs were analyzed. However, in present study, not all of the foods met the nutritional guidelines (FEDIAF, 2021) (Table 2). When it comes to microbiological safety, in the previous study in none of the analyzed foods Enterobacteriaceae spp. were found, including coliforms, Escherichia coli and Salmonella spp. Bacteria of the genus Listeria and Clostridium as well as yeasts were also not detected. In our present study, these microorganisms were present (except of Salmonella spp.), which may indicate that the microbiological situation of extruded dog foods has worsened on the Polish market during 2 years. This may be due to economic factors. Due to the global crisis related to the SARS-CoV-2 pandemic, producers may use cheaper, less hygienic food processing techniques. They may use cheaper raw materials, possibly from other sources, with reduced microbiological quality. Further research should be carried out in this regard, as this situation may worsen further.

5. Conclusion

In our research, we focused on evaluating the microbiological safety of 35 dry dog foods, taking into account the division into foods with grains as the main ingredient and those described as grain-free. We assumed that grain-included foods are more susceptible to the presence of bacteria (Proteus spp., Enterococcus spp., Staphylococcus spp.) and fungi than grain-free foods. An additional criterion was the analysis of the proximate composition of the analyzed foods, taking into account compliance with the FEDIAF nutritional guidelines (2021). Our results confirm the research of other authors – unfortunately, dry dog foods, are contaminated with microorganisms. The presence of microorganisms in the analyzed dog foods indicates the need for systematic control of the quality of raw materials and the food production process in order to minimize the risk to animal and human health. All foods met the MRL for protein recommended by FEDIAF (2021), however the values in some foods were multiples of the minimum levels recommended by FEDIAF (2021). However, current nutritional guidelines do not set maximum levels for proximate components in dog food. Any improperly balanced diet carries a risk of negative health effects. The demonstration of a certain differentiation of the microbiota between GI and GF proves the legitimacy of the comparison of foods, because the similarity of the profiles may indirectly characterize the microbiological purity of the foods.

This study, while confirming the need for in-depth further analysis, has limitations that must be taken into account. The microbiological safety of dog food on the Polish market was assessed only using classical

methods, without further confirmatory tests such as the identification of serotypes.

In the next stage of the risk study, it would be worth carrying out in vivo tests to determine to what extent microbiologically contaminated food affects the health of the animal. It would be useful to analyze the hygiene practices of the caregivers.

Funding sources

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability statements

The data used to support the findings of this study are included within the article.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.rvsc.2023.105071.

References

Abarca, M.L., Bragulat, M.R., Castellá, G., Cabañes, F.J., 1994. Ochratoxin A production by strains of Aspergillus niger var. niger. Appl. Environ. Microbiol. 60, 2650–2652. https://doi.org/10.1128/aem.60.7.2650-2652.1994.

Alegría-Morán, R.A., Guzmán-Pino, S.A., Egaña, J.I., Muñoz, C., Figueroa, J., 2019. Food preferences in dogs: effect of dietary composition and intrinsic variables on diet selection. Animals. 9, 219–231. https://doi.org/10.3390/ani9050219.

Andrade, R.M., Nascimento, J.S., 2005. Presença de fungos filamentosos em ração para cães comercializadas na cidade de Pelotas. RS. Arqu. Instit. Biol. São Paulo, São Paulo 72, 10–12 (in Portuguese).

AOAC (Ed.), 2019. Official Methods of Analysis of the AOAC, 21th ed. AOAC, Gaithersburg, USA.

Bacanlı, M., Başaran, N., 2019. Importance of antibiotic residues in animal food. Food Chem. Toxicol. 125, 462–466. https://doi.org/10.1016/j.fct.2019.01.033.

Bacci, C., Vismarra, A., Dander, S., Barilli, E., Superchi, P., 2019. Occurrence and antimicrobial profile of bacterial pathogens in former foodstuff meat products used for pet diets. J. Food Prot. 82, 316–324. https://doi.org/10.4315/0362-028X.JFP- 18-352.

Behravesh, C.B., Ferraro, A., Deasy, M., Dato, W., Moll, C., Sandt, C., Rea, N.K., Rickert, R., Marriott, C., Warren, K., Urdaneta, V., Salehi, E., Villamil, E., Ayers, T., Hoekstra, R.M., Austin, J.L., Ostroff, S., Williams, I.T., 2010. Human Salmonella infections linked to contaminated dry dog and cat food, 2006–2008. Pediatrics. 126, 477–483.

Beloshapka, A.N., Buff, P.R., Fahey, G.C., Swanson, K.S., 2016. Compositional analysis of whole grains, processed grains, grain co-products, and other carbohydrate sources with applicability to pet animal nutrition. Foods. 5, 23. https://doi.org/10.3390/ foods5020023.

Böhm, J., Koinig, L., Razzazi-Fazeli, E., Blajet-Kosicka, A., Twaruzek, M., Grajewski, J., Lang, C., 2010. Survey and risk assessment of the mycotoxins deoxynivalenol, zearalenone, fumonisins, ochratoxin A, and aflatoxins in commercial dry dog food. Mycotoxin Res. 26, 147–153. https://doi.org/10.1007/s12550-010-0049-4.

Bottari, B., Bancalari, E., Barera, A., Ghidini, S., Gatti, M., 2020. Evaluating the presence of human pathogens in commercially frozen, biologically appropriate raw pet food sold in Italy. Vet. Rec. 187, 1–6. https://doi.org/10.1136/vr.105893.

Bueno, D., Silva, J.O., Oliver, G., 2001. Mycoflora in commercial pet foods. J. Food Prot. 64, 741–743. https://doi.org/10.4315/0362-028X-64.5.741.

Calabrò, S., Carciofi, A.C., Musco, N., Tudisco, R., Gomes, M.O.S., Cutrignelli, M.I., 2013. Fermentation characteristics of several carbohydrate sources for dog diets using the in vitro gas production technique. Ital. J. Anim. Sci. 12 https://doi.org/10.4081/ ijas.2013.e4.

Chiofalo, B., De Vita, G., Lo Presti, V., Cucinotta, S., Gaglio, G., Leone, F., Di Rosa, A.R., 2019. Anim. Nutr. 5, 297–306. https://doi.org/10.1016/2Fj.aninu.2019.05.001.

Cohen, J., 1969. RC: A profile similarity coefficient invariant over variable reflection. Psychol. Bull. 71, 281–284. https://doi.org/10.1037/h0026865.

Commission Regulation (EU) No 142/2011, 2011. Commission Regulation (EU) No 142/ 2011 of 25 February 2011 Implementing Regulation (EC) No 1069/2009 of the European Parliament and of the Council Laying Down Health Rules as Regards Animal By-Products and Derived Products Not Intended for Human Consumption

J. Kępińska-Pacelik et al.

Research in Veterinary Science 165 (2023) 105071

11

and Implementing Council Directive 97/78/EC as Regards Certain Samples and Items Exempt From Veterinary Checks at the Border under that Directive.

Corsato Alvarenga, I., Aldrich, C.G., 2020. Starch characterization of commercial extruded dry pet foods. Transl. Anim. Sci. 4, txaa018. https://doi.org/10.1093/tas/ txaa018.

Davies, R.H., Lawes, J.R., Wales, A.D., 2019. Raw diets for dogs and cats: a review, with particular reference to microbiological hazards. J. Small Anim. Pract. 60, 329–339. https://doi.org/10.1111/jsap.13000.

de Godoy, M.R.C., Kerr, K.R., Fahey Jr., G.C., 2013. Alternative dietary fiber sources in companion animal nutrition. Nutrients. 5, 3099–3117. https://doi.org/10.3390/ nu5083099.

de Jong, A., Muggeo, A., El Garch, F., Moyaert, H., de Champs, C., Guillard, T., 2018. Characterization of quinolone resistance mechanisms in Enterobacteriaceae isolated from companion animals in Europe (ComPath II study). Vet. Microbiol. 216, 159–167.

de Oliveira, L.D., Takakura, F.S., Kienzle, E., Brunetto, M.A., Teshima, E., Pereira, G.T., Vasconcellos, R.S., Carciofi, A.C., 2012. Fibre analysis and fibre digestibility in pet foods – a comparison of total dietary fibre, neutral and acid detergent fibre and crude fibre. J. Anim. Physiol. Anim. Nutr. 96, 895–906. https://doi.org/10.1111/j.1439- 0396.2011.01203.x.

Domingues, L., Murakami, F., Zattoni, D., Kaella, G., de Oliveira, S., Félix, A., 2019. Effect of potato on kibble characteristics and diet digestibility and palatability to adult dogs and puppies. Ital. J. Anim. Sci. 18, 292–300. https://doi.org/10.1080/ 1828051X.2018.1512385.

Donadelli, R.A., Dogan, H., Aldrich, G., 2021. The effects of fiber source on extrusion parameters and kibble structure of dry dog foods. Anim. Feed Sci. Technol. 274, 114884 https://doi.org/10.1093/tas/txaa185.

FCI, 2022. Fédération Cynologique Internationale. available online. https://www.fci.be/ en/statistics/FCI.aspx (assessed 29 June 2022).

FEDIAF, 2018. Guide to Good Practice for the Manufacture of Safe Pet Food. Bruxelles, Belgium.

FEDIAF, 2019. Code of Good Labelling Practice for Pet Food. Bruxelles, Belgium, The European Pet Food Industry.

FEDIAF, 2021. Nutritional Guidelines from Complete and Complementary Pet Food for Cats and Dogs. Bruxelles, Belgium.

FEDIAF, 2022. Facts & Figures. The European Pet Food Industry; Bruxelles, Belgium. Filipović, S., Radović, V., Okanović, Đ., Ristić, M., Kormanjoš, Š., Dosković, V., 2010.

Production of dog food from protein meal obtained from processed poultry slaughter by products. Acta Agric. Serb. 29, 25–30.

Finisterra, L., Duarte, B., Peixe, L., Novais, C., Freitas, A.R., 2021. Industrial dog food is a vehicle of multidrug-resistant enterococci carrying virulence genes often linked to human infections. Int. J. Food Microbiol. 109284 https://doi.org/10.1016/j. ijfoodmicro.2021.1.

Fong, I.W., 2017. Animals and mechanisms of disease transmission. Emerg. Zoon. 8, 15–38. https://doi.org/10.1007/978-3-319-50890-0_2.

Gazzotti, T., Biagi, G., Pagliuca, G., Pinna, C., Scardilli, M., Grandi, M., Zaghini, G., 2015. Occurrence of mycotoxins in extruded commercial dog food. Anim. Feed Sci. Technol. 202, 81–89. https://doi.org/10.1016/j.anifeedsci.2015.02.004.

Ge, B., Domesle, K.J., Gaines, S.A., Lam, C., Bodeis Jones, S.M., Yang, Q., Ayers, S.L., McDermott, P.F., 2020. Prevalence and antimicrobial susceptibility of indicator organisms Escherichia coli and Enterococcus spp. isolated from U.S. Animal Food, 2005–2011. Microorganisms. 8, 1048. https://doi.org/10.3390/ microorganisms8071048.

Gebreselassie, E.E., Jewell, D.E., 2019. Long-term consumption of high protein disrupts dog gut microbiome and metabolites. FASEB J. 33, lb248. https://doi.org/10.1096/ fasebj.2019.33.1_supplement.lb248.

Girio, T.M.S., Nader Filho, A., Rossi Junior, O.D., Amaral, L.A., Girio, R.J.S., 2012. Microbiological quality of dog feed sold in sealed packages and in bulk. Ars. Vet. 28, 36–40.

Gizzarelli, M., Calabrò, S., Vastolo, A., Molinaro, G., Balestrino, I., Cutrignelli, M.I., 2021. Clinical findings in healthy dogs fed with diets characterized by different carbohydrates sources. Front. Vet. Sci. 8, 667318 https://doi.org/10.3389/ fvets.2021.667318.

GMP, 2005. Regulations on Product Standards in the Animal Feed Sector. Good Manufacturing Practices.

Gourineni, V., Stewart, M.L., Skorge, R., Sekula, B.C., 2017. Slowly digestible carbohydrate for balanced energy: in vitro and in vivo evidence. Nutrients 9, 1230. https://doi.org/10.3390/nu9111230.

Hall, J.A., Vondran, J.C., Vanchina, M.A., Jewell, D.E., 2018. When fed foods with similar palatability, healthy adult dogs and cats choose different macronutrient compositions. J. Exp. Biol. 221, 173450 https://doi.org/10.1242/jeb.173450.

Hellgren, J., Hästö, L.S., Wikström, C., Fernström, L.L., Hansson, I., 2019. Occurrence of Salmonella, Campylobacter, Clostridium and Enterobacteriaceae in raw meat-based diets for dogs. Vet. Rec. 184, 442. https://doi.org/10.1136/vr.105199.

Heuberger, R., Wakshlag, J., 2011. The relationship of feeding patterns and obesity in dogs. J. Anim. Physiol. Anim. Nutr. 95, 98–105.

Hocking, A.D., 2003. Microbiological facts and fictions in grain storage. In: Wright, E.J., Webb, M.C., Highley, E. (Eds.), Stored Grain in Australia. Proceedings of the Australian Postharvest Technical Conference, Canberra, 25–27 June 2003. CSIRO Stored Grain Research Laboratory, Canberra.

Hołda, K., Wiczuk, W., Hać-Szymańczuk, E., Głogowski, R., 2017. Comprehensive microbiological evaluation of dry foods for growing dogs marketed in Poland. Anim. Sci. 56, 81–89. https://doi.org/10.22630/AAS.2017.56.1.10.

Inal, F., Alataş, M.S., Kahraman, O., Inal, S., Uludag, M., 2020. Determination of fat preferences of adult dogs. Turk. J. Vet. Anim. Sci. 44, 481–486. https://doi.org/ 10.3906/vet-1906-56.

ISO, 2005. Method 7937, Microbiology of Food and Animal Feeding Stuffs — Horizontal Method for the Enumeration of Clostridium perfringens — Colony-Count Technique. International Organization for Standardization, Switzerland.

ISO, 2007. Method 4832, Microbiology of Food and Animal Feeding Stuffs — Horizontal Method for the Enumeration of Coliforms — Colony Count Technique. International Organization for Standardization, Switzerland.

ISO, 2008. Method 7218, Microbiology of Food and Animal Feeding Stuffs — General Requirements and Guidance for Microbiological Examinations — Amendment 1 International Organization for Standardization Switzerland.

ISO, 2009. Method 21527–1, Microbiology of Food and Animal Feeding Stuffs — Horizontal Method for the Enumeration of Yeasts and Moulds — Part 1: Colony Count Technique in Products with Water Activity Greater than 0,95. International Organization for Standardization, Switzerland.

ISO, 2012. ISO Method 6498:2012, Animal Feeding Stuffs – Guidelines for Sample Preparation. International Organization for Standardization, Switzerland.

ISO, 2013. Method 4833–1, Microbiology of the Food Chain — Horizontal Method for the Enumeration of Microorganisms — Part 1: Colony Count at 30 ◦C by the Pour Plate Technique. International Organization for Standardization, Switzerland.

ISO, 2014. Method 9308–1, Water Quality — Enumeration of Escherichia Coli and Coliform bacteria — Part 1: Membrane Filtration Method for Waters with Low Bacterial Background Flora. International Organization for Standardization, Switzerland.

ISO, 2017a. Method 21528–2, Microbiology of the Food Chain — Horizontal Method for the Detection and Enumeration of Enterobacteriaceae — Part 2: Colony-Count Technique. International Organization for Standardization, Switzerland.

ISO, 2017b. Method 6887–1, Microbiology of the Food Chain — Preparation of Test Samples, Initial Suspension and Decimal Dilutions for Microbiological Examination — Part 1: General Rules for the Preparation of the Initial Suspension and Decimal Dilutions. International Organization for Standardization, Switzerland.

ISO, 2020. Method 6579–1:2017/Amd 1:2020, Microbiology of the Food Chain — Horizontal Method for the Detection, Enumeration and Serotyping of Salmonella — Part 1: Detection of Salmonella spp. — Amendment 1: Broader Range of Incubation Temperatures, Amendment to the Status of Annex D, and Correction of the Composition of MSRV and SC. International Organization for Standardization, Switzerland.

ISO, 2021. Method 6888–1, Microbiology of the Food Chain — Horizontal Method for the Enumeration of Coagulase-Positive Staphylococci (Staphylococcus aureus and Other Species) — Part 1: Method Using Baird-Parker Agar Medium. International Organization for Standardization, Switzerland.

Johnson, M.L., Parsons, C.M., Fahey Jr., G.C., Merchen, N.R., Aldrich, C.G., 1998. Effects of species raw material source, ash content, and processing temperature on amino acid digestibility of animal by-product meals by customized roosters and ideally cannulated dogs. Anim. Sci. J. 76, 1112–1122. https://doi.org/10.2527/ 1998.7641112x.

Kahraman, O., Inal, F., 2021. Comparison of digestibility parameters of commercial dry dog foods with different contents. Arq. Bras. Med. Vet. Zootec. 73, 469–476. https:// doi.org/10.1590/1678-4162-12167.

Kazimierska, K., Biel, W., Witkowicz, R., Karakulska, J., Stachurska, X., 2021. Evaluation of nutritional value and microbiological safety in commercial dog food. Vet. Res. Commun. 45, 111–128. https://doi.org/10.1007/s11259-021-09791-6.

Kępińska-Pacelik, J., Biel, W., 2021. Microbiological hazards in dry dog chews and feeds. Animals. 11, 631. https://doi.org/10.3390/ani11030631.

Kienzle, E., Dobenecker, B., Eber, S., 2001. Effect of cellulose on the digestibility of high starch versus high fat diets in dogs. J. Anim. Physiol. Anim. Nutr. 85, 174–185. https://doi.org/10.1046/j.1439-0396.2001.00315.x.

Kim, Y.J., Park, J.Y., Suh, S.H., Kim, M.G., Kwak, H.S., Kim, S.H., Heo, E.J., 2021. Determination of statistical sampling criteria for microbial contamination of cereal and Saeng-sik products. LWT. 151, 112129 https://doi.org/10.1016/j. lwt.2021.112129.

Laflamme, D., Izquierdo, O., Eirmann, L., Binder, S., 2014. Myths and misperceptions about ingredients used in commercial pet foods. Vet. Clin. North Am. Small Anim. Pract. 44, 689–698. https://doi.org/10.1016/j.cvsm.2014.03.002.

Leiva, A., Molina, A., Redondo-Solano, M., Artavia, G., Rojas-Bogantes, L., Granados- Chinchilla, F., 2019. Pet food quality assurance and safety and quality assurance survey within the Costa Rican pet food industry. Animals. 9, 980–1005. https://doi. org/10.3390/ani9110980.

Lenox, C., 2016. Role of dietary fatty acids in dogs and cats. TVP. 8, 83–90. Loncaric, I., Misic, D., Szostak, M.P., Künzel, F., Schäfer-Somi, S., Spergser, J., 2020.

Broad-spectrum cephalosporin-resistant and/or fluoroquinolone-resistant enterobacterales associated with canine and feline urogenital infections. Antibiotics. 9 (7), 387. https://doi.org/10.3390/antibiotics9070387.

Lynch, S.A., Helbig, K.J., 2021. The complex diseases of Staphylococcus pseudintermedius in canines: where to next? Vet. Sci. 8, 11. https://doi.org/10.3390/vetsci8010011.

Ma, F., Xu, S., Tang, Z., Li, Z., Zhang, L., 2021. Use of antimicrobials in food animals and impact of transmission of antimicrobial resistance on humans. Biosaf. Health. 3 (1), 32–38. https://doi.org/10.1016/j.bsheal.2020.09.004.

Milanov, D.S., Petrović, T.R., Todorović, D.S., Aleksić, N.R., Čabarkapa, I.S., 2018. Toxin genotypes of Clostridium perfringens in animal feed and their role in the ethiology of enterotoxemia in domestic animals. Food Feed Res. 45, 67–76. https://doi.org/ 10.5937/FFR1801067M.

Muñoz-Prieto, A., Nielsen, L.R., Dąbrowski, R., Bjørnvad, C.R., Söder, J., Lamy, E., Monkeviciene, I., Ljubić, B.B., Vasiu, I., Savic, S., Busato, F., Yilmaz, Z., Bravo- Cantero, A.F., Öhlund, M., Lucena, S., Zelvyte, R., Aladrović, J., Lopez-Jornet, P., Caldin, M., Lavrador, C., Karveliene, B., Mrljak, V., Mazeikiene, J., Tvarijonaviciute, A., 2018. European dog owner perceptions of obesity and factors

J. Kępińska-Pacelik et al.

Research in Veterinary Science 165 (2023) 105071

12

associated with human and canine obesity. Sci. Rep. 8, 13353. https://doi.org/ 10.1038/s41598-018-31532-0.

Murray, S.M., Fahey, G.C., Merchen, N.R., Sunvold, G.D., Reinhart, G.A., 1999. Evaluation of selected high-starch flours as ingredients in canine diets. J. Anim. Sci. 77, 2180–2186. https://doi.org/10.2527/1999.7782180x.

Naughton, V., Grzelak, T., Mulhern, M.S., Moffett, C., Naughton, P., 2022. Caring practices of pet cat and dog owners in Northern Ireland vs potential implications for animals’ health and welfare: caring practices of pet cat and dog owners. Anim. Welf. 30, 131–144. https://doi.org/10.7120/09627286.30.2.131.

Nemser, S.M., Doran, T., Grabenstein, M., McConnell, T., McGrath, T., Pamboukian, R., Smith, A.C., Achen, M., Danzeisen, G., Kim, S., Liu, Y., Robeson, S., Rosario, G., McWilliams, K., Reimschuessel, W.R., 2014. Investigation of Listeria, Salmonella, and toxigenic Escherichia coli in various pet foods. Foodborne Pathog. Dis. 11, 706–709. https://doi.org/10.1089/fpd.2014.1748.

NRC, 2006. Absorption and Bioavailability of Dietary iron in Dogs and Cats. Nutrient Requirements of Dogs and Cats. The National Academic Press, Washington.

Palmqvist, H., Ringmark, S., Höglund, K., Pelve, E., Lundh, T., Dicksved, J., 2022. Effects of rye inclusion in dog food on fecal microbiota and short-chain fatty acids. Res. Sq. https://doi.org/10.21203/rs.3.rs-1411694/v1.

Paßlack, N., Galliou, F., Manios, T., Lasaridi, K., Zentek, J., 2022. In vitro digestion and microbial fermentation of dried food residues, a potential “new” component for pet food, and different non-digestible carbohydrate sources. PLoS One 17, e0262536. https://doi.org/10.1371/journal.pone.0262536.

Pepper, I.L., Gerba, C.P., 2005. Environmental Microbiology: A Laboratory Manual, second ed. Elsevier Academic Press, Amsterdam.

Pieper, R., Kroger, S., Richter, J.F., Wang, J., Martin, L., Bindelle, J., Htoo, J.K., von Smolinski, D., Vahjen, W., Zentek, J., Van Kessel, A.G., 2012. Fermentable fiber ameliorates fermentable protein-induced changes in microbial ecology, but not the mucosal response, in the colon of piglets. J. Nutr. 142, 661–667.

RASFF, 2022. RASFF Portal. Available online. https://webgate.ec.europa.eu/rasffwind ow/portal/?event=SearchForm&cleanSearch=1.

Regulation (EC) No 183/2005 of the European Parliament and of the Council of 12 January 2005 Laying Down Requirements for Feed Hygiene, 2005.

Runesvärd, E., Wikström, C., Fernström, L.L., Hansson, I., 2020. Presence of pathogenic bacteria in faeces from dogs fed raw meat-based diets or dry kibble. Vet. Rec. 187, 1–6. https://doi.org/10.1136/vr.105644.

Ruscher, C., Lübke-Becker, A., Wleklinski, C.G., Soba, A., Wieler, L.H., Walther, B., 2009. Prevalence of methicillin-resistant Staphylococcus pseudintermedius isolated from clinical samples of companion animals and equidaes. Vet. Microbiol. 136, 197–201. https://doi.org/10.1016/j.vetmic.2008.10.023.

Ryan, K.J., Ray, C.G., Nafees, A., Drew, W.L., Plorde, J.J., 2010. Sherris Medical Microbiology, fifth ed. McGraw Hill, New York.

Sahin-Tóth, J., Kovács, E., Tóthpál, A., Juhász, J., Forró, B., 2021. Whole genome sequencing of coagulase positive staphylococci from a dog-and-owner screening survey. PLoS One 16, e0245351. https://doi.org/10.1371/journal.pone.0245351.

Samet, L.E., Vaterlaws-Whiteside, H., Harvey, N.D., Upjohn, M.M., Casey, R.A., 2022. Exploring and developing the questions used to measure the human–dog bond: new and existing themes. Animals. 12, 805. https://doi.org/10.3390/ani12070805.

Savi, G.D., Piacentini, K.C., Rocha, L.O., Carnielli-Queiroz, L., Furtado, B.G., Scussel, R., Zanoni, E.T., Machado-de-Ávila, R.A., Corrêa, B., Angioletto, E., 2018. Incidence of toxigenic fungi and zearalenone in rice grains from Brazil. Int. J. Food Microbiol. 270, 5–13. https://doi.org/10.1016/j.ijfoodmicro.2018.02.004.

Selmi, M., Stefanelli, S., Bilei, S., Tolli, R., Bertolotti, L., Marconi, P., Giurlani, S., De Lucia, P.G., Ruggeri, G., Pagani, A., 2011. Contaminated commercial dehydrated

food as source of multiple Salmonella serotypes outbreak in a municipal kennel in Tuscany. Vet. Ital. 47, 183–190.

Serhan, M., Hadid, M., Dimassi, H., Deghel, M., Hassan, H.F., 2022. Microbiological safety of commercial canned and dry pet food products in Lebanon. Front. Vet. Sci. 9, 995184 https://doi.org/10.3389/fvets.2022.995184.

Singh, S., Wakeling, L., Gamlath, S., 2007. Retention of essential amino acids during extrusion of protein and reducing sugars. J. Agric. Food Chem. 55, 8779–8786.

Sprohnle-Barrera, C.H., Gibson, J.S., Price, R., Graham, R.M., Jennison, A.V., Ricca, M. R., Allavena, R.E., 2022. Fatal non-traumatic gas gangrene caused by Clostridium perfringens type A in a Siberian Husky dog. J. Vet. Diagn. Investig. 34, 518–522. https://doi.org/10.1177/10406387221079066.

Stercova, E., Strakova, E., Tsponova, J., Grmelova, M., Janacova, K., Muchova, K., 2022. Nutritional evaluation of commercial dry dog foods available on the Czech market. J. Anim. Physiol. Anim. Nutr. 106, 614–621. https://doi.org/10.1111/jpn.13573.

Stolle, I., Prenger-Berninghoff, E., Stamm, I., Scheufen, S., Hassdenteufel, E., Guenther, S., Bethe, A., Pfeifer, Y., Ewers, C., 2013. Emergence of OXA-48 carbapenemase-producing Escherichia coli and Klebsiella pneumoniae in dogs. J. Antimicrob. Chemother. 68, 2802–2808.

Tegzes, J.H., Oakley, B.B., Brennan, G., 2019. Comparison of mycotoxin concentrations in grain versus grain-free dry and wet commercial dog foods. Toxicol. Commun. 3, 61–66. https://doi.org/10.1080/24734306.2019.1648636.

Thomas-Popo, E., Mendonça, A., Misra, N.N., Little, A., Wan, Z., Moutiq, R., Coleman, S., Keener, K., 2019. Inactivation of Shiga-toxin-producing Escherichia coli, Salmonella enterica and natural microflora on tempered wheat grains by atmospheric cold plasma. Food Control 104, 231–239. https://doi.org/10.1016/j. foodcont.2019.04.025.

Thompson, A., 2008. Ingredients: where pet food starts. Top. Companion Anim. Med. 23, 127–132. https://doi.org/10.1053/j.tcam.2008.04.004.

TIBCO, 2021. Available online: https://www.scribd.com/document/321061529 /STATISTICA-Electronic-Manual (accessed 23 October 2021).

Tolbert, M.K., Murphy, M., Gaylord, L., Witzel-Rollins, A., 2022. Dietary management of chronic enteropathy in dogs. J. Small Anim. Pract. 63, 425–434. https://doi.org/ 10.1111/jsap.13471.

Viegas, F.M., Ramos, C.P., Xavier, R.G.C., Lopes, E.O., Junior, C.A.O., Bagno, R.M., 2020. Fecal shedding of Salmonella spp., Clostridium perfringens, and Clostridioides difficile in dogs fed raw meat-based diets in Brazil and their owners’ motivation. PLoS One 15, e0231275. https://doi.org/10.1371/journal.pone.0231275.

Weber, M., Bissot, T., Servet, E., Sergheraert, R., Biourge, V., German, A.J., 2007. A high- protein, high-fiber diet designed for weight loss improves satiety in dogs. J. Vet. Intern. Med. 1, 1203–1208. https://doi.org/10.1892/07-016.1.

Weese, H.E., 2001. The roles of Clostridium difficile and enterotoxigenic. Clostridium perfringens in diarrhea in dogs. J. Vet. Intern. Med. 15, 374–378. https://doi.org/ 10.1111/j.1939-1676.2001.tb02332.x.

White, B.L., Johnson, B.L., Morgan, R., Shields, R.G., 2018. Changes in the food safety landscape of pet foods in the United States. In: Ricke, S.C., Atungulu, G.G., Rainwater, C.E., Park, S.H. (Eds.), Food and Feed Safety Systems and Analysis. Academic Press, Cambridge, pp. 3–23.

Witaszak, N., Waśkiewicz, A., Bocianowski, J., Stępień, Ł., 2020. Contamination of pet food with mycobiota and fusarium mycotoxins-focus on dogs and cats. Toxins. 12, 130–144. https://doi.org/10.3390/toxins12020130.

Wrigley, C.W., Corke, Ch., Walker, Ch.E., 2004. Encyclopedia of Grain Science, second ed. Elsevier Academic Press, Amsterdam.

J. Kępińska-Pacelik et al.