Lipid Quality and Oxidative Stability in Industrial Foods for Dogs and Cats

Abstract

The global population of dogs and cats has grown, boosting the revenue of the pet industry, and has raised concerns among their owners regarding the quality of food. Lipids are essential in the diet of dogs and cats, supporting overall health when present within recommended limits. Lipid oxidation represents a challenge as it compromises the quality and safety of food. Antioxidants play an important role in minimizing these impacts and ensuring food stability. Therefore, this study analyzed the composition of fatty acids and antioxidants in dry extruded food for dogs and cats. The values of saturated fatty acids (SFA), monounsaturated fatty acids (MUFA) and polyunsaturated fatty acids (PUFA) were determined, highlighting the importance of the omega-6/omega-3 ratio in animal health. It was observed that all foods met the minimum recommended limits for fatty acids. Antioxidant analysis revealed the presence of butylhydroxyanisole (BHA), butylhydroxytoluene (BHT) and ethoxyquin within permitted limits. The peroxide index (PI) was higher in dog foods. The results reinforce the need for an adequate balance between fatty acids and strategies to control lipid oxidation, aiming to guarantee the quality of industrialized foods for dogs and cats.

Keywords:
antioxidants; fatty acids; liquid chromatography; peroxide value

Introduction

According to the Brazilian Association of the Pet Products Industry (ABINPET), the global population of dogs and cats has been increasing, resulting in a corresponding rise in industry revenue.¹ Consequently, the quality and safety of food consumed by these animals have become a major concern for their guardians, especially after incidents reported in 2022, when the government of the Brazilian state of Minas Gerais suspended sale of batches suspected of contamination following 30 cases of ethylene glycol poisoning in pets, many of which resulted in death after consumption of snacks.²

Dry foods for dogs and cats contain lipids in their formulation.³ Lipids have important biological functions in cell membranes, such as membrane transport and signal transduction.⁴ Polyunsaturated fatty acids (PUFAs), including omega-3 (n-3) and omega-6 (n-6), are necessary for the maintenance of cell membranes and brain functions. Both arachidonic acid (ARA) and linoleic acid (LA) are essential for maintaining canine health because their deficiency results in skin, coat and reproduction problems.⁵

Furthermore, PUFAs, such as eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), have previously demonstrated potential for reducing inflammation and muscle damage in dogs.⁶ Research indicates that supplementation with EPA and DHA can improve cognitive performance in puppies, without causing an increase in blood lipid oxidation.⁷,⁸

The processing and formulation of a food product can be modified according to the relationship between the oxidizable fractions and the stable part of the lipids.⁹ Lipid oxidation raises significant concerns in food safety due to the formation of oxidative degradation products.¹⁰ The nutritional consequences of lipid oxidation include partial destruction of the essential unsaturated fatty acids; partial destruction of other unsaturated lipids such as vitamins A, carotenoids and tocopherols; partial destruction of vitamin C (co-oxidation); formation of secondary products of lipid oxidation (malonaldehyde and other compounds) and Maillard compounds, capable of reacting with biomolecules (especially proteins), decreasing their absorption.

Lipids present in food generally undergo oxidation,¹¹ making antioxidants essential for maintaining the quality of the final product, ensuring food safety, nutritional values, and high levels of customer acceptance due to flavor, texture and appearance.³ Antioxidants inhibit lipid oxidation by eliminating free radicals and preventing oxidative chain reactions. Their effectiveness is altered according to concentration, chemical structure and interactions with other food components.¹² Oxidative stability and antioxidant properties influence the quality, shelf life and nutritional value of food products.¹³

According to the Brazilian Ministry of Agriculture and Livestock (MAPA), the maximum limits are 150 mg kg^-1 for butylhydroxyanisole (BHA) and butylhydroxytoluene (BHT), 100 mg kg^-1 for ethoxyquin, and no established limit for tert-butylhydroquinone (TBHQ).¹⁴ However, the Association of American Feed Control Officials (AAFCO) recommends a maximum limit of 200 mg of synthetic antioxidants (BHA, BHT and TBHQ) per kg of fat in complete feed.¹⁴ The Food and Drug Administration (FDA) prohibits the use of ethoxyquin, and the maximum limit established for BHA and BHT is 100 mg kg^-1.¹ Excessive intake of BHA and BHT has been associated with potential toxic effects in animals, including liver damage and disruption of normal metabolic processes, highlighting the importance of adhering to established safety limits.¹⁵

Considering the importance of fatty acids in the diet of dogs and cats and the oxidative degradation of lipids, which compromises food quality and generates toxic compounds that are harmful to health, this study aims to analyze the fatty acids present in commercial diets for dogs and cats, as well as the residual antioxidant levels and peroxide index (PI) to monitor oxidation. In addition, the data were evaluated using multivariate analysis using principal component analysis (PCA) to better understand the observed variations.

Experimental

Materials

Petroleum ether, ethyl ether, n-heptane, potassium hydroxide (KOH), chloroform, methanol, acetonitrile, acetic acid, sodium sulfate, potassium iodide, starch, sodium thiosulfate, and the internal standards (a fatty acid methyl esters standard mixture (FAME mix C4-C24, purity ≥ 97%) and methyl tricosanoate (23:0, purity ≥ 99%)) were obtained from Millipore-Sigma (St. Louis, USA).

Sampling

Sixteen different samples of dog and cat food, with package mass between 200 g and 1 kg, from nine commercial brands of different categories were purchased from a local store in the city of Maringá (state of Paraná, Brazil), including four samples from the premium line (AM01, AM02, CN10 and CN11), eight from the special premium line (AM05, CN06, CN07, CN08, CN09, CN14, CN15 and CN16) and four from the super-premium line (AM03, AM04, CN12 and CN13). Sample selection considered their relevance and high demand in the dog and cat food market. The samples were categorized and named according to their target audience: AM01 to AM05 correspond to cat food, while CN06 to CN16 are dog food, none of which are for specific clinical purposes. Table 1 contains general information about each sample purchased.

Table 1
Segmented sampling of dry extruded food for cats (AM01-AM05 samples) and dogs (CN06-CN16 samples) by classification, indication, flavor, package mass, and fat (oil source)

Sample	Line	Indication	Flavor	Package mass / kg	Fat (oil source)
AM01	premium	adult cat	chicken	0.8	chicken oil, fish oil
AM02	premium	adult cat	meat	0.8	chicken oil, fish oil
AM03	super premium	adult cat	chicken	0.5	chicken fat, refined fish oil
AM04	super premium	kitten	chicken	0.5	chicken fat, refined fish oil
AM05	special premium	adult cat	chicken	0.5	chicken fat, refined fish oil
CN06	special premium	adult dog	chicken	1.0	refined lard, chicken fat
CN07	special premium	puppy	chicken	1.0	chicken fat, refined fish oil
CN08	special premium	mini adult dog	-	1.0	poultry by-product oil, refined fish oil, linseed oil
CN09	special premium	mini adult dog	meat	1.0	poultry oil, refined fish oil, linseed oil
CN10	premium	mini and small adult dog	meat and chicken	1.0	poultry oil
CN11	premium	medium and large adult dog	meat and chicken	1.0	chicken fat
CN12	super premium	small adult dog	chicken	1.0	poultry oil, salmon oil
CN13	super premium	small puppy	chicken	1.0	poultry oil, refined fish oil
CN14	special premium	small adult dog	meat	1.0	poultry oil
CN15	special premium	medium adult dog	meat	1.0	poultry oil
CN16	special premium	small adult dog	chicken	1.0	poultry oil, fish oil

Fatty acid composition by gas chromatography

Lipids were extracted by the Soxhlet method, as described by da Cruz et al.¹⁶ with adaptations. Briefly, 5 g of sample were weighed and subjected to extraction with a solvent mixture of petroleum ether and ethyl ether (1:1, v v^-1) under reflux at 60 °C for 6 h in a Soxhlet extractor. The obtained organic phase was concentrated by rotary evaporation, resulting in the lipid extract.

FAMEs were prepared according to the ISO 12966:20172¹⁷ protocol. An amount of ca. 100 mg of extracted lipids was mixed with 2 mL of n-heptane and vortexed for 2 min. 2 mL of 2 mol L^-1 KOH in methanol were added, and the mixture was stirred for an additional 3 min. Then, 500 μL of the internal standard methyl tricosanoate (23:0; 1 mg mL^-1 in n-heptane) were incorporated. After 24 h of phase separation, the supernatant was collected and analyzed by gas chromatograph (GC) (Thermo Scientific, model Trace GC Ultra, Waltham, USA), equipped with a flame ionization detector (FID). The separation was performed using a CP-7420 capillary column (Agilent, Saint Clara, USA). The sample injection volume was 2.0 µL, with the injector configured in split mode with a 1:40 division. The gas flow rates were set at 1.4 mL min^-1 for the carrier gas (hydrogen, H₂), 30 mL min^-1 for the makeup gas (nitrogen, N₂), and 30 and 300 mL min^-1 for the flame gases (H₂ and synthetic air, respectively). The injector and detector temperatures were set at 230 and 250 °C, respectively. The GC oven temperature was set under the following temperature gradient conditions: initial temperature at 65 °C, maintained for 4 min, followed by heating to 185 °C at a rate of 16 °C min^-1, remaining at this temperature for 12 min. Subsequently, the temperature was increased to 235 °C at 20 °C min^-1, being maintained for 9 min, totaling an analysis time of 35 min. Peak areas were integrated with ChromQuest 5.0 software. The FAME identification was performed using the internal standard FAME mix C4-C24 to compare the retention times of the mixture components. Lipid mass was determined using the internal standard 23:0 according to Visentainer and Franco.¹⁸ The fatty acid composition was expressed as mass (mg g^-1 of lipid), with all samples analyzed in triplicate.

The composition of fatty acids was determined by grouping them into specific categories based on the degree of saturation and chemical structure. The total amounts of saturated fatty acids (ΣSFA), monounsaturated fatty acids (ΣMUFA) and polyunsaturated fatty acids (ΣPUFA) were calculated by summing the individual fatty acids within each respective class (equations 1 to 3).

(1) $$\mathrm{SFA}=(14:0)+(16:0)+18:0)+(20:0)$$

(2) $$\begin{array}{rl}& \text{EMUFA}=(16:1n-7)+(16:1n-9)+(18:1n-9)+\\ & (18:1n-7)+(20:1n-9)\end{array}$$

(3) $$\begin{array}{rl}& \text{DPUFA}=(18:2n-6)+(18:3n-3)+(18:3n-6)+\\ & (20:5n-3)+(22:5n-3)+(22:6n-3)\end{array}$$

In addition, the sums of fatty acids belonging to the n-6 (Σn-6) (omega-6) and n-3 (Σn-3) (omega-3) series were determined, as well as the ratio between these two series (n-6/n-3) (equations 4 to 6). Fatty acids of the n-6 and n-3 families perform essential physiological functions in mammals. The ratio between these two classes of fatty acids is highly relevant in the nutrition of companion animals since the n-6:n-3 ratio directly influences the regulation of the inflammatory response, being a crucial aspect in the treatment and control of several diseases.¹⁹

(4) $$\sum n-6\text{series}=(18:2n-6)+(18:3n-6)$$

(5) $$\mathrm{\Sigma }n-3=(18:3n-3)+(20:5n-3)+(22:5n-3)+(22:6n-3)$$

(6) $$\frac{n-6}{n-3}\text{relationship}=\frac{\mathrm{\Sigma }n-6}{\mathrm{\Sigma }n-3}$$

where, 14:0 refers to myristic acid, 16:0 to palmitic acid, 18:0 to stearic acid, 20:0 to arachidic acid, 16:1n-7 to palmitoleic acid, 16:1n 9 to 7-hexadecanoic acid, 18:1n 9 to oleic acid, 18:1n 7 to vaccenic acid, 20:1n 9 to gondoic acid, 18:2n 6 to LA, 18:3n 3 to α-linolenic (ALA), 18:3n 6 to γ-linolenic acid, 20:5n 3 to EPA, 22:5n 3 to docosapentaenoic acid and 22:6n 3 to DHA.

Determination of synthetic antioxidant residual

BHA and BHT were quantified according to da Silva et al.²⁰ using a high-performance liquid chromatography (HPLC) model Alliance e2695 system coupled to with diode array detection (DAD) model 2998 (Waters, Milford, MA, USA). For extraction, 2 g of the sample were shaken with 4 mL of acetonitrile for 5 min. The solvent was removed in a rotary evaporator at ca. 40 °C, and the resulting extract was filtered through a 0.45 µm membrane prior to chromatographic analysis. The chromatographic conditions were as follows: flow rate of 1.2 mL min^-1, acetonitrile/acidified water (5% acetic acid, pH 3.1) as mobile phase in the ratio of 70:30 (v v^-1) for 3 min, 80:20 (v v^-1) for 5 min, and 70:30 (v v^-1) for 4 min, and a total run time of 12 min. According to the chromatographic analysis by HPLC-DAD, using the standards of each compound, it was found that the retention times were 5.003 min for BHA, 8.941 min for ethoxyquin and 17.502 min for BHT.

Determination of peroxide index (PI)

PI was determined according to the methodology described by Costa et al.,¹⁴ with modifications. The amount of ca. 30 g of sample was mixed with 50 mL of methanol, 25 mL of chloroform and 17 mL of water, under magnetic stirring for 2 min. Then, 25 mL of chloroform and 25 mL of 1.5% sodium sulfate solution were added, with further stirring for 2 min. The mixture was transferred to a separatory funnel for natural phase separation. The lower phase containing chloroform and lipids was filtered with filter paper and anhydrous sodium sulfate to remove traces of water. Afterwards, 20 mL of acetic acid, 0.5 mL of saturated potassium iodide solution (14 g mL-¹), 30 mL of distilled water and 1 mL of 1% starch solution were added. The samples were kept in the dark for 1 min, and then, titrated with 0.1 M sodium thiosulfate until the color disappeared. PI was calculated using equation 7:

(7) $$\mathrm{PI}\left(\mathrm{mEq}\mathrm{kg}\text{of}{\mathrm{fat}}^{-1}\right)=\left(\mathrm{M}\frac{(\mathrm{V}-\mathrm{Vb})}{\mathrm{Ma}}\right)1000$$

where, M is molarity of sodium thiosulfate, V the volume of sodium thiosulfate used in the titration (mL), Vb the volume of sodium thiosulfate used in the blank titration (mL) and Ma the mass of the sample (g).

Data analysis

The fatty acid data were organized into a joint matrix of antioxidant residue and PI values. The data were autoscaled and subjected to multivariate statistical analysis using PCA technique to identify the dimensions that best differentiate the data set, highlighting the similarities and differences present through the identification of patterns. PCA was performed using the factoextra and FactoMineR packages in the RStudio software version 2024.04.2.²¹ All data from the analyses were subjected to analysis of variance (ANOVA) and Tukey’s mean comparison test (p < 0.05).

Results and Discussion

Fatty acids composition

Figure 1 presents the results of the fatty acid composition including the sum of the unsaturation classes.

Figure 1
Quantities of saturated fatty acids (SFA), monounsaturated fatty acids (MUFA) and polyunsaturated fatty acids (PUFA) calculated for dry food for cats (AM01-AM05 samples) and dogs (CN06-CN016 samples).

The SFA concentrations ranged from 198.30 ± 2.81 mg g^-1 (CN16) to 285.07 ± 2.20 mg g^-1 (CN07). Saturated fats, commonly found in ingredients such as coconut and palm oils, are associated with increased total cholesterol and low-density lipoprotein (LDL) levels, increasing cardiovascular risk. On the other hand, unsaturated fats, found in oils such as canola and olive oil, help reduce total cholesterol and low-density lipoprotein (LDL), in addition to offering additional benefits, such as better glycemic control, weight loss, and lower risk of cancer, and are widely recommended as healthier options.²² The highest amount of MUFA was observed in sample CN07 (464.41 ± 2.21 mg g^-1), followed by sample AM03 (429.82 ± 5.16 mg g^-1), while the lowest value was recorded in CN16 (326.69 ± 5.24 mg g^-1).

Regarding PUFAs, the highest value was recorded in sample CN16 (295.92 ± 7.39 mg g^-1), followed by CN12 (288.77 ± 3.65 mg g^-1), while the lowest concentrations were found in samples AM03 (142.87 ± 1.30 mg g^-1) and CN07 (184.97 ± 0.36 mg g^-1). PUFAs are essential in the canine diet, providing an effective source of energy and facilitating the absorption of fat-soluble vitamins. In addition, the inclusion of PUFAs, particularly omega-3, contributes to the reduction of inflammatory mediators, helping to reduce inflammation. However, it is important to monitor the quantity as in high concentrations can increase susceptibility to oxidation by free radicals due to the high degree of unsaturation.⁵

The LA and ALA fatty acids are considered essential for mammals, as the body is unable to synthesize them, which must be obtained through the diet. In dogs, LA can be converted to ARA by the action of enzymes suitable for this process. In cats, this conversion is limited due to the low or non-existent activity of the ∆-6 desaturase enzyme. For this reason, the inclusion of ARA in the diet is crucial, especially at times of greater metabolic demand, such as during gestation and growth.¹⁹

ALA, in turn, is converted into EPA and, subsequently, into DHA by the same enzymes that participate in the conversion of LA to ARA. This implies that, when there is a greater amount of LA to ALA, the body tends to prioritize the production of ARA, reducing the conversion of ALA to EPA and DHA. Thus, an unbalanced ratio of n-6 to n-3 fatty acids in the diet mainly reduces the level of n-3 fatty acids in the body and can directly influence inflammatory responses, disease control, and general health.¹⁹Table 2 shows the results of the laboratory analysis for fatty acids in extruded dry diets for cats and dogs and the corresponding European Pet Food Industry Federation (FEDIAF) recommendations per 100 g of dry matter (DM).²³

Table 2
Fatty acid content in dry extruded foods for cats (AM01-AM05 samples) and dogs (CN06-CN16 samples) compared to FEDIAF recommendations²³

Fatty acid	Extruded food for cat						Recommendationa
	AM01	AM02	AM03	AM04	AM05		Growth (reproduction)	Adultb
LA (18:2n-6) / (100 g-1 DM)	23.67	22.13	17.85	21.55	14.81		0.55	0.50
ARA (20:4n-6) / (100 g-1 DM)	0.06	0.12	0.10	0.14	0.10		0.02	0.006
ALA (18:3n-3) / (100 g DM)	0.16	0.13	0.15	0.17	0.16		0.02	NR
EPA (20:5n-3) / (100 g-1 DM)	0.21	0.10	0.24	0.30	0.30		NR	NR
DHA (22:6n-3) / (100 g-1 DM)	0.42	0.35	0.46	0.58	0.84		NR	NR
EPA + DHA / (100 g-1 DM)	0.63	0.45	0.70	0.88	1.14		0.01	NR

Fatty acid	Extruded food for dog												Recommendationa
	CN06	CN07	CN08	CN09	CN10	CN11	CN12	CN13	CN14	CN15	CN16		Early growth and reproduction	Late growth	Adultc
LA (18:2n-6) / (100 g-1 DM)	25.02	22.05	22.29	21.72	21.91	22.83	30.24	24.55	27.69	27.23	34.04		1.30d	1.30	1.32
ARA (20:4n-6) / (100 g-1 DM)	0.13	0.08	0.11	0.03	0.18	0.15	0.22	0.12	0.10	0.13	0.14		0.03	0.03	NR
ALA (18:3n-3) / (100 g-1 DM)	0.17	0.11	0.14	0.14	0.17	0.13	0.14	0.15	0.11	0.14	0.17		0.08	0.08	NR
EPA (20:5n-3) / (100 g-1 DM)	0.12	0.01	0.01	0.01	0.09	0.20	0.15	0.09	0.17	0.10	0.15		NR	NR	NR
DHA (22:6n-3) / (100 g-1 DM)	0.24	0.23	0.23	0.24	0.56	0.67	0.36	0.73	0.39	0.28	0.41		NR	NR	NR
EPA + DHA / (100 g-1 DM)	0.36	0.24	0.24	0.25	0.65	0.87	0.51	0.82	0.56	0.38	0.56		0.05	0.05	NR

• DM: dry matter; FEDIAF: European Pet Food Industry Federation; NR: no recommendation; LA: linoleic acid; ALA: α-linolenic acid; ARA: arachidonic acid; EPA: eicosapentaenoic acid; DHA: docosahexaenoic acid.
• ^a
  Minimum recommended level;
• ^b
  based on the metabolizable energy requirement of 100 kcal per kg of metabolic weight (kg^0.67);
• ^c
  based on the metabolizable energy requirement of 110 kcal per kg of metabolic weight (kg^0.75);
• ^d
  maximum recommended nutritional level for LA in the growth phase of dogs is 6.50 g.

All samples for cats exceeded the recommended minimum risk limit (MRL) for LA, with the lowest value recorded in sample AM05 (14.81 g 100 g^-1 DM) and the highest in sample AM01 (23.67 g 100 g^-1 DM). Regarding ARA content, AM01 food presented the lowest value (0.06 g 100 g^-1 DM), while AM04 food presented the highest value (0.14 g 100 g^-1 DM). The MRL value for ARA is 0.014 g higher in the growth or reproduction phase compared to the adult phase, and AM04 food met this criterion.

For ALA, there is no recommendation for adult cats, but for growing or reproducing cats, the MRL value is 0.02 g 100 g^-1 DM, and AM04 food also met this criterion. Regarding EPA and DHA fatty acids, the recommendation is for the sum of the two, which should be at least 0.01 g 100 g^-1 DM. All cat foods met this specification, with AM05 presenting the highest value (1.14 g 100 g^-1 DM). In the case of dog foods, all met MRL for LA, with values of 1.30 g 100 g^-1 DM for dogs in growth or reproduction and 1.32 g 100 g^-1 DM for maintenance in adulthood. The lowest value was recorded in sample CN09 (21.72 g 100 g^-1 DM) and the highest in sample CN16 (34.04 g 100 g^-1 DM). Puppy foods, CN07 and CN13, presented amounts of ARA above the MRL value for dogs in growth (0.08 and 0.12 g 100 g^-1 DM, respectively). For adult dogs, there is no established MRL for this fatty acid.

As for ALA, the recommended minimum risk limit for growing dogs is 0.08 g 100 g^-1 DM, and both CN07 and CN13 foods met this criterion, with values of 0.11 g 100 g^-1 DM and 0.15 g 100 g^-1 DM, respectively. In addition, all foods presented the sum of EPA + DHA above the MRL value of 0.05 g 100 g^-1 DM, with the highest concentration found in the CN11 sample (0.87 g 100 g^-1 DM).

In recent years, commercial formulations for dogs have been improved with higher levels of n-3 fatty acids, reflecting a growing concern with the balance between n-6 and n-3, essential for general health and inflammation control. However, significant variations in the proportion of these acids in different products can compromise the expected benefits, highlighting the importance of maintaining an adequate balance and adopting strategies that prevent lipid oxidation, ensuring their nutritional functionality.²⁴

Determination of synthetic residues and oxidation level

One of the strategies to prevent lipid oxidation is the addition of antioxidants during production, with synthetic phenolic antioxidants, such as BHA, BHT and ethoxyquin, commonly used. This is because natural antioxidants are generally less stable during storage and more susceptible to oxidative degradation.²⁵ However, both classes of antioxidants share similar mechanisms of action, including neutralizing free radicals, chelating metals, and quenching of singlet oxygen.²⁶

Table 3 shows the antioxidant levels detected in the analyzed pet foods, with all samples containing synthetic antioxidants. According to Brazilian legislation, all the BHA and BHT levels were within the permitted limit (150 mg kg^-1). In addition, ethoxyquin was detected in only 5 samples, all within the established limit (100 mg kg^-1).

Table 3
Synthetic antioxidant concentrations and peroxide index values of dry food for cats (AM01-AM05 samples) and dogs (CN06-CN16 samples)

Sample	Ethoxyquin / (mg kg-1)	BHA / (mg kg-1)	BHT / (mg kg-1)	Peroxide index / (mEq O2 kg-1 of fat)
AM01	ND	22.89 ± 1.55cdef	5.27 ± 0.06d	3.08 ± .0.93abcde
AM02	ND	50.55 ± 0.30a	52.21 ± 2.14a	2.86 ± 0.60abcde
AM03	11.69 ± 0.62c	26.82 ± 3.00bcde	80.95 ± 1.17b	0.24 ± 0.34a
AM04	1.05 ± 0.01a	19.26 ± 2.55cf	51.60 ± 5.2a	1.46 ± 0.28abc
AM05	ND	17.81 ± 0.88f	51.90 ± 1.78a	1.29 ± 0.22ac
CN06	ND	27.52 ± 0.61bcde	52.12 ± 0.67a	1.79 ± 0.48abcd
CN07	8.18 ± 0.1b	20.93 ± 0.59cdf	50.24 ± 1.28a	2.20 ± 0.15abcd
CN08	ND	28.83 ± 0.88bdeg	9.43 ± 0.29cd	5.76 ± 0.84bdef
CN09	0.89 ± 0.01a	32.35 ± 0.64bgh	25.94 ± 1.34e	4.55 ± 0.01abcdef
CN10	ND	8.33 ± 0.01i	40.96 ± 0.79f	3.47 ± 1.69abcdef
CN11	ND	17.44 ± 0.17f	91.36 ± 0.01g	4.22 ± 0.63abcdef
CN12	ND	24.81 ± 1.34bcdef	8.04 ± 0.09cd	8.02 ± 0.45f
CN13	ND	36.58 ± 0.96gh	12.43 ± 0.27cd	5.36 ± 1.01bcdef
CN14	ND	30.25 ± 1.35begh	26.04 ± 0.38e	3.90 ± 3.58abcdef
CN15	ND	28.97 ± 0.36bdeg	16.42 ± 0.25c	7.23 ± 0.19def
CN16	1.25 ± 0.13a	38.18 ± 3.59h	30.06 ± 0.53e	7.25 ± 0.01ef

Proper control of synthetic antioxidant use is essential given their potential health risks. BHA can cause damage to the thyroid system, metabolic disorders, growth retardation, carcinogenesis, and neurological and reproductive effects.¹⁰ It is often combined with BHT due to its low cost and thermal stability.

BHT is associated with reproductive toxicity, DNA damage, and endocrine disruption, while its transformation products are even more toxic, impacting fetal development and increasing the risk of cancer.²⁷

The reduction or elimination of ethoxyquin in formulas is motivated by the same concerns. Although ethoxyquin is not considered genotoxic or carcinogenic in its original form, its oxidation during feed processing can generate compounds, such as ethoxyquin quinone imine, which have characteristics associated with mutagenicity and interaction with DNA. These oxidation byproducts raise questions about potential health risks, especially regarding genotoxic and carcinogenic effects.²⁸ This context explains the presence of ethoxyquin in only 30% of the feeds.

The highest concentrations of ethoxyquin, BHA and BHT were detected, respectively, in samples AM03 (11.69 ± 0.62 mg g^-1), AM02 (50.55 ± 0.30 mg g^-1) and CN11 (91.36 ± 0.01 mg g^-1). None of the samples analyzed exceeded the maximum recommended limit of 150 mg g^-1 for the sum of these antioxidants.²⁹ The highest total value found for the sum of antioxidants was recorded in sample AM03, with a total of 119.46 mg g^-1, followed by sample AM02, which recorded 102.76 mg g^-1. The lowest values were found by AM01 (28.16 mg g^-1) and CN12 (32.85 mg g^-1).

Regarding PI, none of the samples exceeded the level of 10 mEq kg^-1 of fat. Although there is no legislation for peroxide limits established in dog and cat food, Brazilian National Health Surveillance Agency (ANVISA)³⁰ determines the maximum limit for fats as 10 mEq kg ¹ of fat. Samples CN12 (8.02 mEq kg^-1 of fat), CN16 (7.25 mEq kg^-1 of fat) and CN15 (7.23 mEq kg^-1 of fat) presented the highest PI values. The present results are within the range found, which varied from 2.2 to 94.10 mEq kg^-1 of fat. However, as peroxides are temporary intermediates in lipid oxidation, their levels may decrease as the process progresses. This suggests that the observed PI may be underestimated, and the product may have undergone significant lipid oxidation, despite the low values found.¹⁴

Lipid oxidation is detrimental to both the health of animals and the acceptance of food by owners. High levels of peroxides can reduce the food intake of dogs, affect the growth of puppies, cause changes in platelet function, and compromise the immune system. In addition, the undesirable odors generated by oxidation negatively impact the choice of food by owners, harming the marketing of the product.³¹

The lowest peroxide levels were in samples AM03 and AM05 with values of 0.24 and 1.29 mEq kg^-1 of fat, respectively. In general, foods intended for felines presented an average of 1.77 mEq kg^-1 of fat, significantly lower than the average for dog foods, which was 4.89 mEq kg^-1 of fat, that is, more than twice as high.

Multivariate data analysis

The PCA results based on the fatty acid composition, PI, and antioxidant residual data are presented in Figure 2. PCA is a mathematical technique used to reduce the dimensionality of multivariate data by creating linear combinations, called principal components (PCs), from the original variables. These components are defined by the eigenvalues and eigenvectors of the data covariance matrix, reflecting the directions of greatest variance. This technique provides the projection of data into a lower-dimensional space, facilitating clearer visualization and interpretation of the relationships among samples and their characteristics.³²

Figure 2
Principal component analysis. (a) distribution of in the first two principal components (PC1 and PC2) of food for cats (AMx-AMx samples) and dogs (CNx-CNx samples) rations from the datasets, and (b) contribution of variables to the first two principal components.

The distribution of the samples is shown in Figure 2a, while the contribution of the variables to the observed separation is highlighted in Figure 2b. The first principal component (PC1) explained 53.9% of the total variability of the data, and the second principal component (PC2) explained 20.2%. Together, the first two principal components accounted for 74.1% of the total variability.

PC1 was the main factor responsible for distinguishing between dog and cat foods (Figure 2a). The cat food samples (AM01-AM05) were located on the positive side of PC1, while most dog food samples were positioned on the negative side, except CN07, which presented a positive coordinate on PC1. Samples CN10 and CN11 were located near the boundary between the positive and negative PC1 quadrants but showed higher scores on PC2, indicating a greater influence of this component.

As illustrated in Figure 2b, the similarity between cat foods is related to the levels of ΣSFA, ΣMUFA, Σn-3, and residues of the antioxidants ethoxyquin and BHT. Puppy food CN07 is very similar to adult cat food AM03, sharing the same levels mentioned. In contrast, dog foods stand out for high levels of ΣPUFA, Σn-6, Σn-6/n-3 ratio, the residue of the antioxidant BHA and PI.

For a more detailed analysis of these relationships, it is possible to observe the distribution of the samples in the different quadrants. Samples CN12, CN13, CN14 and CN16 are positioned in the negative quadrant of PC1 and in the positive quadrant of PC2 (Figure 2a), which indicates the predominance of the levels of ΣPUFA, Σn-6, the residual of the antioxidant BHA, and PI (Figure 2b). These diets present the highest levels of Σn-6 and ΣPUFA, in the following decreasing order: CN13, CN14, CN12 and CN16. On the other hand, the samples with the highest concentration of ΣPUFA (CN12 and CN16) also presented the highest PIs (8.02 and 7.25 mEq kg^-1 of fat, respectively).

Samples AM02, AM03 and CN07 are located in the positive quadrant for both PC1 and PC2 (Figure 2a), reflecting a greater influence of the ΣMUFA, Σn-3 and ethoxyquin contents (Figure 2b). Among these samples, AM03 stood out for presenting a high concentration of ΣMUFA (429.82 mg g^-1), a greater amount of ethoxyquin (11.69 mg g^-1), and the highest content of total antioxidants (119.46 mg g^-1), in addition to exhibiting the lowest peroxide index (0.24 mEq kg^-1 of fat). These results indicate that the amount of antioxidants added was effective in preventing lipid oxidation in this product. Compared to BHT and BHA, ethoxyquin is more resistant to processing conditions, such as heat, pressure and humidity, which may influence its performance in protecting against oxidation.¹⁴

A similar trend was observed for sample CN07, which presented a ΣMUFA of 464.41 mg g^-1 and ethoxyquin of 8.18 mg g^-1. However, a lower number of total antioxidants was added (79.35 mg kg-¹), which resulted in a higher PI value (2.2 mEq kg^-1 of fat) compared to AM03.

Samples CN08 and CN09, in turn, are in the negative quadrant for both PC1 and PC2 (Figure 2a), being mainly associated with the Σn-6/n-3 ratio (Figure 2b). Higher Σn-6/n-3 ratios indicate high amounts of omega-6 fatty acids and low amounts of omega-3 fatty acids. However, as previously demonstrated, these samples met the FEDIAF recommendations regarding n-6 and n-3 fatty acids.

Samples AM04, AM05, CN06, CN10, and CN11, located in the positive quadrant for PC1 and negative for PC2 (Figure 2a) are predominantly influenced by the ΣSFA and residual BHT contents (Figure 2b). On average, BHT was used more (37.81 mg kg^-1) than BHA (26.97 mg kg^-1) in the products.

Sample AM01 is positioned on the dividing line between the positive and negative quadrants of PC1. However, it has a positive coordinate in PC2, being influenced mainly by the levels of Σn-3 and by the residues of BHA and ethoxyquin. Despite being the feed with the lowest number of total antioxidants (28.16 mg kg^-1), PI was not so high (3.08 mEq kg^-1 of fat).

Sample CN15, in turn, is positioned on the dividing line between the positive and negative quadrants of PC2, with a negative coordinate in PC1. It is influenced by the levels of ΣPUFA, Σn-6, Σn-6/n-3 ratio, residual of the antioxidant BHA and PI. PI of this sample was 7.23 mEq kg^-1 of fat, thus being one of those with the highest levels of peroxide, along with samples CN12, CN13, CN14 and CN16.

However, when comparing the levels of BHA, BHT and ethoxyquin in AM03 with those of the other samples, it becomes evident that, in addition to containing the combined total of the three antioxidants, this sample also shows a higher BHT index than the others (Figure 3).

Figure 3
Results of comparative peroxide index with butylhydroxyanisole (BHA), butylhydroxytoluene (BHT) and ethoxyquin indexes in 16 samples of dry food for cats (AM01-AM05 samples) and dogs (CN06-CN16 samples).

The relationship between the omega-6 and omega-3 levels of sample AM03 (8.85 ± 0.97) obtained the lowest value, below 10:1, while the other samples presented values greater than 10:1, being in increasing order the samples CN07 (15.02 ± 0.45), AM04 (24.14 ± 1.83) and AM01 (23.79 ± 5.52) (Table 4). According to Hand,³³ the relationship between the omega-6 and omega-3 levels when less than 10:1, demonstrates immunomodulatory effects, when the absolute levels of EPA and DHA are also adequate.

Table 4
Results of fatty acid analysis in 16 samples of dry food from different segments for cats (AM01-AM05 samples) and dogs (CN06-CN16 samples)

Sample	SFA / (mg g-1 of lipid)	MUFA / (mg g-1 of lipid)	PUFA / (mg g-1 of lipid)	∑n-6 / (mg g-1 of lipid)	∑n-3 / (mg g-1 of lipid)	n-6/n-3 / (mg g-1 of lipid)
AM01	237.77 ± 1.12a	395.36 ± 3.67ab	216.49 ± 3.59bcdf	207.76 ± 3.04ad	8.73 ± 0.55bd	23.79 ± 5.52g
AM02	263.40 ± 1.53a	397.13 ± 2.15ab	207.24 ± 1.66abcd	195.24 ± 0.98a	11.99 ± 0.67ab	16.27 ± 1.45 a
AM03	267.14 ± 5.30b	429.82 ± 5.16c	142.87 ± 1.30a	128.37 ± 0.64b	14.49 ± 0.66cd	8.85 ± 0.97 b
AM04	245.34 ± 2.52abc	380.66 ± 2.90ab	197.28 ± 0.83abc	189.44 ± 0.54a	7.84 ± 0.29c	24.14 ± 1.83b
AM05	236.02 ± 0.37abc	354.55 ± 7.84ac	192.77 ± 3.85e	188.07 ± 3.12c	4.70 ± 0.73ce	39.99 ± 4.25c
CN06	245.52 ± 12.90ac	390.28 ± 7.43bde	208.90 ± 11.59cdf	202.39 ± 10.92d	6.51 ± 0.67a	31.05 ± 16.28d
CN07	285.07 ± 2.20abc	464.41 ± 2.21bdef	184.97 ± 0.36ab	173.43 ± 0.11a	11.54 ± 0.25a	15.02 ± 0.45e
CN08	249.26 ± 6.50abc	373.36 ± 7.49bd	200.60 ± 0.80ab	196.84 ± 0.60ab	3.76 ± 0.20a	52.29 ± 2.92d
CN09	268.67 ± 1.26bc	363.23 ± 1.85ab	199.92 ± 0.80abc	196.16 ± 0.60a	3.76 ± 0.20a	52.11 ± 2.92d
CN10	258.33 ± 4.91a	380.23 ± 3.98bd	196.83 ± 0.54ab	192.18 ± 0.23e	4.64 ± 0.30f	41.33 ± 0.78f
CN11	283.41 ± 10.27a	391.14 ± 13.90bdef	202.96 ± 6.36ab	198.14 ± 6.22e	4.82 ± 0.14f	41.08 ± 42.13f
CN12	238.57 ± 6.92a	371.43 ± 8.48bde	288.77 ± 3.65g	280.41 ± 1.69e	8.36 ± 1.95e	33.53 ± 0.86f
CN13	243.18 ± 6.70a	363.20 ± 12.25bdef	222.85 ± 7.87bcdf	213.07 ± 6.94e	9.77 ± 0.93fg	21.79 ± 7.40f
CN14	237.31 ± 7.49a	335.03 ± 8.84ef	243.97 ± 8.23f	237.07 ± 7.65e	6.89 ± 0.58f	34.36 ± 13.18f
CN15	234.73 ± 2.13a	339.23 ± 5.57def	238.68 ± 4.64df	233.10 ± 4.53e	5.57 ± 0.11eg	41.83 ± 40.66f
CN16	198.30 ± 2.81d	326.69 ± 5.24f	295.92 ± 7.39g	289.32 ± 6.29e	7.60 ± 1.09e	37.90 ± 5.73f

The immunomodulatory effects are more notable when the omega-6/omega-3 ratio is less than 5:1, acting as modulators of intestinal inflammation and other systems, aiding in the treatment of gastrointestinal disorders. However, although low ratios are recommended, care should be taken when working with ratios lower than 2:1, as Wander et al.³⁴ reported that such proportions led to a reduction in the cellular immune response, in the serum concentrations of plasma α-tocopherol, and an increase in lipoperoxidation products, considered deleterious in dogs by these authors.³³ In addition, supplementation with high concentrations of PUFAs can increase susceptibility to free radical oxidation.⁵

Studies³⁵ have also shown that dogs fed with a n-6/n-3 ratio of 5:1 exhibited increased activation of isolated T and B cells, along with breed-specific differences in fatty acid metabolism, that may explain why not all atopic dogs respond to fatty acid therapy. Additionally, it is important to note that higher dietary levels of PUFAs should be accompanied by increased vitamin E intake, as this vitamin acts both as an antioxidant and as an enzymatic cofactor in the metabolic pathways involved in eicosanoids synthesis.³³

Conclusions

All food samples met the FEDIAF standards for omega-3 and omega-6 fatty acid contents recommended for growing, reproducing, or adult cats and dogs. All food samples analyzed contained synthetic antioxidants at levels above current legislation. Ethoxyquin was identified in some samples, but was not mentioned on the labels, which only declared the use of BHT and BHA as antioxidants. All food presented peroxide values at acceptable levels for fats, according to ANVISA (below 10 mEq kg^-1 of fat). The obtained results in this study reinforce the need for an adequate balance between fatty acids and strategies to control lipid oxidation, aiming to guarantee the quality of industrialized food intended for cats and dogs.

Acknowledgments

The authors thank Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) and Fundação Araucária for financial support.

Data Availability Statement

Data will be made available on request.

References

• ¹ Associação Brasileira da Indústria de Produtos para Animais de Estimação (ABINPET); Manual Pet Food Brazil, 10^th ed.; ABINPET: São Paulo, Brasil, 2019. [Link] accessed in October 2025
  » Link
• ² Costa, M.; Vigilância Rastreia Uso de Produto que Matou Cães na Indústria Alimentícia; Estado de Minas, 2022. [Link] accessed in October 2025
  » Link
• ³ European Pet Food Industry Federation (FEDIAF); Guide to Good Practice for the Manufacture of Safe Pet Foods; FEDIAF: Brussels, 2018. [Link] accessed in October 2025
  » Link
• ⁴ Xue, Z.; Zhang, C.; Liu, J.; Li, Q.; Yao, Y.; Yang, Y.; Ran, C.; Zhang, Z.; Zhou, Z.; LWT 2023, 184, 115004. [Crossref]
  » Crossref
• ⁵ Baritugo, K. A.; Bakhsh, A.; Kim, B.; Park, S.; J. Funct. Foods 2023, 109, 105744. [Crossref]
  » Crossref
• ⁶ Burri, L.; Wyse, C.; Gray, S. R.; Harris, W. S.; Lazzerini, K.; Res. Vet. Sci. 2018, 121, 18. [Crossref]
  » Crossref
• ⁷ Rodrigues, R. B. A.; Zafalon, R. V. A.; Rentas, M. F.; Risolia, L. W.; Macedo, H. T.; Perini, M. P.; da Silva, A. M. G.; Marchi, P. H.; Balieiro, J. C. C.; Mendes, W. S.; Vendramini, T. H. A.; Brunetto, M. A.; Animals 2023, 13, 2938. [Crossref]
  » Crossref
• ⁸ Jacuńska, W.; Biel, W.; Witkowicz, R.; Maciejewska Markiewicz, D.; Piątkowska, E.; Appl. Sci. 2023, 13, 11791. [Crossref]
  » Crossref
• ⁹ Silva, F. A. M.; Borges, M. F. M.; Ferreira, M. A.; Quim. Nova 1999, 22, 94. [Crossref]
  » Crossref
• ¹⁰ Zhang, X.; Diao, M.; Zhang, Y.; J. Sci. Food Agric. 2023, 103, 6150. [Crossref]
  » Crossref
• ¹¹ Asido, E.; Zeigerman, H.; Verman, M.; Argov-Argaman, N.; Kanner, J.; Tirosh, O.; Curr. Res. Food Sci. 2024, 8, 100652. [Crossref]
  » Crossref
• ¹² Baschieri, A.; Pizzol, R.; Guo, Y.; Amorati, R.; Valgimigli, L.; J. Agric. Food Chem. 2019, 67, 6902. [Crossref]
  » Crossref
• ¹³ Villeneuve, P.; Bourlieu-Lacanal, C.; Durand, E.; Lecomte, J.; McClements, D. J.; Decker, E. A.; Crit. Ver. Food Sci. Nutr. 2023, 63, 4687. [Crossref]
  » Crossref
• ¹⁴ Costa, J. L. G.; Pedreira, R. S.; Gomes, A. C. P.; Restan, A. Z.; Vasconcellos, R. S.; Loureiro, B. A.; Anim. Feed Sci. Technol. 2022, 294, 115499. [Crossref]
  » Crossref
• ¹⁵ Ren, J.; Li, Z.; Li, X.; Yang, L.; Bu, Z.; Wu, Y.; Li, Y.; Zhang, S.; Meng, X.; Foods 2025, 14, 1095. [Crossref]
  » Crossref
• ¹⁶ da Cruz, V. H. M.; Priori, R. L. G.; Santos, P. D. S.; Ferreira, C. S. F.; Piccioli, A. F. B.; da Silveira, R.; Visentainer, J. V.; Santos, O. O.; Chem. Pap. 2021, 75, 515. [Crossref]
  » Crossref
• ¹⁷ ISO 12966:2/2017: Animal and Vegetable Fats and Oils - Gas Chromatography of Fatty Acid Methyl Esters - Part 2: Preparation of Methyl Esters of Fatty Acids; ISO: Geneva, 2017. [Link] accessed in October 2025
  » Link
• ¹⁸ Visentainer, J. V.; Franco, M. R. B.; Ácidos Graxos em Óleos e Gorduras: Identificação e Quantificação, 1^st ed.; Varela: São Paulo, 2006.
• ¹⁹ Burron, S.; Richards, T.; Krebs, G.; Trevizan, L.; Rankovic, A.; Hartwig, S.; Pearson, W.; Ma, D. W. L.; Shoveller, A. K.; J. Anim. Sci. 2024, 102, skae143. [Crossref]
  » Crossref
• ²⁰ da Silva, I. C.; dos Santos, P. D. S.; dos Santos Júnior, O. O.; Rocha, M.; Janeiro, V.; Volpato, J. A.; Lazzari, A.; Vasconcellos, R. S.; Anim. Feed Sci. Technol. 2024, 315, 115997. [Crossref]
  » Crossref
• ²¹ RStudio, version 2024.04.2+764; RStudio, Boston, MA, USA, 2024.
• ²² Voon, P. T.; Ng, C. M.; Ng, Y. T.; Wong, Y. J.; Yap, S. Y.; Leong, S. L.; Yong, X. S.; Lee, S. W. H.; Adv. Nutr. 2024, 15, 100276. [Crossref]
  » Crossref
• ²³ European Pet Food Industry Federation (FEDIAF); Nutritional Guidelines for Complete and Complementary Pet Food for Cats and Dogs; FEDIAF: Brussel, Belgium, 2024. [Link] accessed in October 2025
  » Link
• ²⁴ Bosch, G.; Beerda, B.; Hendriks, W. H.; van der Poel, A. F. B.; Verstegen, M. W. A.; Nutr. Res. Rev. 2007, 20, 180. [Crossref]
  » Crossref
• ²⁵ Pacheco, G. F. E.; Bortolin, R. C.; Chaves, P. R.; Moreira, J. C. F.; Kessler, A. M.; Trevizan, L.; J. Anim. Sci. 2018, 96, 4590. [Crossref]
  » Crossref
• ²⁶ Barden, L.; Decker, E. A.; Crit. Rev. Food Sci. Nutr. 2016, 56, 2467. [Crossref]
  » Crossref
• ²⁷ Wang, W.; Xiong, P.; Zhang, H.; Zhu, Q.; Liao, C.; Jiang, G.; Environ. Res. 2021, 201, 111531. [Crossref]
  » Crossref
• ²⁸ Aquilina, G.; Bampidin, V.; Bastos, M. L.; Bories, G.; Chesson, A.; Cocconcelli, P. S.; Flachowsky, G.; Gropp, J.; Kolar, B.; Kouba, M.; Puente, S. L.; López-Alonso, M.; Mantovani, A.; Mayo, B.; Ramos, F.; Rychen, G.; Saarela, M.; Villa, R. E.; Wallace, R. J.; Wester, P.; EFSA J. 2015, 13, 4271. [Crossref]
  » Crossref
• ²⁹ Ministério da Agricultura e Pecuária (MAPA); Instrução Normativa No. 51 de 03 de agosto de 2020, Critérios e Procedimentos para Fabricação, Fracionamento, Importação e Comercialização dos Produtos Dispensados de Registro para Uso na Alimentação Animal; Diário Oficial da União (DOU), Brasília, 05/08/2020, seção 1, p. 26. [Link] accessed in October 2025
  » Link
• ³⁰ Agência Nacional de Vigilância Sanitária (ANVISA); Instrução Normativa No. 87 de 15 de março de 2021, Lista de Espécies Vegetais Autorizadas, as Designações, a Composição de Ácidos Graxos e os Valores Máximos de Acidez e de Indice de Peróxidos para Óleos e Gorduras Vegetais; Diário Oficial da União (DOU), Brasília, 2021. [Link] accessed in October 2025
  » Link
• ³¹ Chanadang, S.; Koppel, K.; Aldrich, G.; Animals 2016, 6, 44. [Crossref]
  » Crossref
• ³² Granato, D.; Santos, J. S.; Escher, G. B.; Ferreira, B. L.; Maggio, R. M.; Trends Food Sci. Technol. 2018, 72, 83. [Crossref]
  » Crossref
• ³³ Hand, M. S.; Small Animal Clinical Nutrition, 7^th ed.; Mark Morris Institute: Topeka, USA, 2010. [Link] accessed in October 2025
  » Link
• ³⁴ Wander, R. C.; Hall, J. A.; Gradin, J. L.; Du, S.-H.; Jewell, D. E.; J. Nutr. 1997, 127, 1198. [Crossref]
  » Crossref
• ³⁵ Kearns, R. J.; Hayek, M. G.; Turek, J. J.; Meydani, M.; Burr, J. R.; Greene, R. J.; Marshall, C. A.; Adams, S. M.; Borgert, R. C.; Reinhart, G. A.; Vet. Immunol. Immunopathol. 1999, 69, 165. [Crossref]
  » Crossref

Edited by

• Editor handled this article:
  Andrea R. Chaves (Executive)

Publication Dates

• Publication in this collection
  28 Nov 2025
• Date of issue
  2025

History

• Received
  26 Aug 2025
• Published
  20 Oct 2025