Effect of high DHA and ARA fortification on lipid oxidation of infant formula powder

ABSTRACT

Our objective was to assess the impact of fortified high-dose docosahexaenoic acid (DHA) and arachidonic acid (ARA) in infant formula powders (IFs) on stability. We compared and evaluated the stability of IFs containing high and low doses of DHA and ARA over a 6-month storage period at 37°C. While most formulations showed minimal changes in unsaturated fatty acid content of during storage, significant losses of DHA and ARA were observed in some high-dose formulations. Furthermore, this study noted consistent trends in the changes of vitamin C and vitamin E across both formula groups. Notably, the high-dose group formulas exhibited a higher presence of volatile organic compounds (VOCs) during storage, along with a more pronounced increase in thiobarbituric acid-reactive substance values (TBARS). In conclusion, the inclusion of high doses of DHA and ARA in IFs may compromise their oxidative stability.

KEYWORDS:

• Infant formula
• unsaturated fatty acid
• volatile organic compounds
• docosahexaenoic acid
• antioxidant

1. Introduction

Approximately 50% of the fat in infant formula powders (IFs) is composed unsaturated fatty acids (UFA), including polyunsaturated fatty acids (PUFA) like linoleic acid, linolenic acid, docosahexaenoic acid (DHA), and arachidonic acid (ARA). Notably, the addition of a high UFA content to IFs makes them more prone to oxidation (Chávez-Servín et al., 2009). Furthermore, even when low doses of DHA and ARA are added to IFs, they still exhibit increased oxidation compared to formulas without added DHA and ARA (Maduko & Akoh, 2006).This susceptibility to oxidation is due to the presence of long-chain polyunsaturated fatty acid (LC-PUFA), which contain multiple ethylenic double bonds that readily oxidize in the presence of substances like oxygen, iron, and high temperatures (Jimenez-Alvarez et al., 2008). The oxidation of fats in IFs can result in the formation of small molecules such as aldehydes, ketones, and hydroperoxides, including hexanal, glutaraldehyde, 4-hydroxy nonaenoic acid, and malondialdehyde (Manglano et al., 2005; Stefania et al., 2015). This lipid oxidation can adversely affect the quality of fatty foods during production and storage. In the food industry, antioxidants like vitamin E, vitamin C, and tea polyphenols are commonly employed to inhibit the oxidation of fatty acids (Indrasena & Barrow, 2010; Srensen et al., 2011). Research has demonstrated that vitamin E and C are effective antioxidants for PUFA, and their combined use can more effectively extend the shelf-life stability of PUFA (Baik et al., 2004; Jang et al., 2006; Srensen et al., 2011). In IFs, vitamin E and C are essential nutrients and although they experience some degradation during storage, they effectively mitigate fat oxidation (Chávez-Servín et al., 2008a; Zou & Akoh, 2015).

Research evidence indicates that DHA and ARA play a crucial role in enhancing infant development, including cognition functions, visual acuity, and immune responses (Lien et al., 2018). Recent studies conducted by Herrmann et al. (2021) and Nieto-Ruiz et al. (2020) have demonstrated that infant formula supplemented with DHA and ARA offers several advantages. It helps reduce the incidence of infectious diseases in infants and young children, such as respiratory and gastrointestinal infections, and contributes to early behavioral development by lowering the occurrence of clinical affective problems. In addition, it has been shown that the proportion of DHA and ARA in plasma phospholipids of malnourished children is directly proportional to the degree of malnutrition, which also suggests that DHA and ARA not only have a positive effect on the growth and development of infants, but also have an impact on the nutritional status of children (Videla et al., 2022).

As a result, an increasing number of expert groups are advocating for the inclusion of long-chain polyunsaturated fatty acid, specifically ARA and DHA, in IFs. In China, the addition of DHA and ARA is optional, with specific guidelines for their maximum amounts in different IFs. For infants aged 0–6 months and those using follow-up formulas (for infants aged 6–12 months), the allowable DHA content is set between 3.6 and 9.6 mg/100 kJ, according to the National Health Commission of the People’s Republic of China in 2021a and 2021b. Furthermore, it is essential to add at least the same amount of ARA as DHA in the formula. For formulas designed for young children aged 12–24 months, the DHA content should be less than 9.6 mg/100 kJ, as specified by the National Health Commission of the People’s Republic of China in 2021c.

Previous studies have primarily focused on assessing the stability of infant formula powder containing low levels of DHA and ARA (DHA ≤0.2% and ARA ≤0.54% of total fatty acids). These investigations have examined various aspects, including the stability of fatty acids (Hageman et al., 2019; Pina-Rodriguez & Akoh, 2010; Takenaga et al., 1998), the stability of antioxidants (Chávez-Servín et al., 2008a; Manglano et al., 2004; Miquel et al., 2004), the composition of volatile organic compounds (VOCs) (Chávez-Servín et al., 2008b; Hausner et al., 2009; Van Ruth et al., 2006; Wang et al., 2019), changes in VOCs during storage (Fenaille et al., 2003), and the stability of products with different carbohydrate compositions (Masum et al., 2020).

However, there is a notable gap in the literature regarding a comprehensive comparison and analysis of the effects of varying amounts of DHA and ARA within the same formula. Specifically, there is a lack of comparative concerning their degree of oxidation and oxidation products, as well as an assessment of changes during storage. In this study, we sought to address this gap by evaluating the oxidative stability of IFs with differing DHA and ARA content. We conducted analyses of fatty acid profiles, antioxidant stability, degree of oxidation (measured using thiobarbituric acid-reactive substance values, TBARS), and the presence of volatile oxidation products during storage at 37°C cover a period of 6 months. Our aim was to provide a comprehensive understanding of the impact of high-dose of DHA and ARA on the oxidative stability of IFs.

2. Materials and methods

2.1. Sample preparation

Three types of IFs (infant formula, follow-up infant formula, young children formula) based on bovine milk, namely IF1, IF2 and IF3, were produced by wet mixing-spray-drying process (Bright Dairy & Food Co., Ltd). The first type (IF1) is infant formula (for infants aged 0–6 months), which contains 25.6 g of fatty acids per 100 g (including 4.18 g of linoleic acid and 0.540 g of α-Linolenic acid). The second type (IF2) is follow-up infant formula (older infant formula, for infants aged 6–12 months), which contains 21.4 g of fatty acids per 100 g (including 3.41 g of linoleic acid and 0.442 g of α-Linolenic acid). The third type (IF3) is young children formula (for young children aged 12 to 36 months), which contains 20.6 g of fatty acids per 100 g (including linoleic acid 3.26 g, α-Linolenic acid 0.420 g). Then, IF1, IF2 and IF3 fortified different doses of DHA and ARA through the dry mixing process to form a low-dose group (IF1-L, IF2-L and IF3-L; additional fortified DHA 105 mg; ARA, 130 mg per 100 g of milk powder per formula) and a high-dose group (IF1-H, IF2-H and IF3-H; additional fortified DHA 168 mg; ARA, 180 mg per 100 g of milk powder per formula).The formulas were packed in airtight containers flushed with an N₂-modified atmosphere, and the residual oxygen content of the product was controlled to be less than 3%.

2.2. Storage

To evaluate the oxidation reaction during IF storage, the products were kept at 37°C (75% relative humidity) for 6 months. Samples were taken at 0, 2, 4, and 6 months to analyze oxidation indicators. After storage, analytical determinations were conducted. Each time, new independent packaging was opened for analysis.

2.3. Determination of unsaturated fatty acids

The fatty acid (FA) composition of the IFs was determined by fast gas chromatography (GC) after derivatization to FA methyl esters (FAMEs) according to GB 5009.168–2016 (National Health Commission of the People’s Republic of China, 2016).

2.4. Determination of vitamin E and C

Vitamin E (α-tocopherol equivalent) and Vitamin C content in the formulas were measured according to a previously reported HPLC method reported by Chávez-Servín (Chávez-Servín et al., 2008a).

2.5. Determination of TBARS

TBARS were determined according to the method described by Manglano (Manglano et al., 2005), with slight modifications. 12.00 g of IF was accurately weighed and dissolved in water, and the solution was made up to 100 mL using water in a volumetric flask. To 17.6 mL reconstituted IF into test tubes and heated in a water bath at 30°C for 5 min, then 1 mL (1 g/mL) trichloroacetic acid (TCA) in water and 2 mL 95% ethanol was added, the tubes were shaken vigorously for 20 s and left to stand for 5 min. After 5 min, the contents were filtered through quantitative filter paper. Then, 1.0 mL of TBA solution (1.4 g 2-thiobarbituric acid in 95% ethanol to 100 mL) was added to 4.0 mL of the clear filtrate, followed by oscillation for 10 s, and placed in a 60°C water bath for 60 min, followed by cooling at room temperature, and finally, the optical density at 532 nm was determined with a spectrophotometer.

2.6. Determination of VOCs

In a sample bottle with a capacity of 20 mL were placed 1.5 g of IFs (accurate to 0.01 g); 2.0 mL of 0.9% NaCl solution was added separately and homogenized. Then, the headspace bottle was placed in a water bath (60°C) and the solid-phase microextraction (SPME) needle was inserted, which was Carboxen®/Polydimethylsiloxane (CAR/PDMS) fiber, adsorbed for 30 min under water bath conditions. This was followed by desorption at 250°C in the gas phase sampler for 5 min.

VOCs were identified using gas chromatography-mass spectrometry (GC-MS, 5977b−7890b, Agilent Technologies, U.S.A.). An HP-5 MS 30 m × 0.25 mm ID, 0.25 μm film thickness, capillary column was used. The GC condition was as follows: splitless injection mode; initial column temperature of 40°C, maintained for 1 min, increased at a rate of 4°C/min to 160°C, then increased at a rate of 10°C/min to 250°C, and maintained for 3 min. High-purity helium (99.999%) was used as the carrier gas at a flow rate of 1.2 mL/min, and pressure of 60 kPa. The MS condition was as follows: electron energy 70 eV, ion source temperature 230°C, quadrupole temperature 150°C, detector temperature 250°C, GC/MS interface temperature 280°C, and filament emission current 200 μA. The detector voltage was 1.2 kV, and the quality scanning range was m/z 50–450.

After the GC-MS analysis, the chromatographic data of the volatiles were determined using the NIST library (version 17.0) and the Wiley MS library. The spectrum of volatile substances was qualitatively analyzed using the NIST 2008 database, and only results with positive and negative matching degrees greater than 750 (maximum value of 1000) were reported. The area normalization method was used to calculate the relative content of volatiles in the IFs, where the content of measured volatiles was equal to the peak area/total peak area of volatiles.

2.7. Statistical analysis

Statistical analysis was conducted using IBM SPSS Statistic 22, and one-way analysis of variance (ANOVA) was used. Statistical significance are represented by the term p < .05. All assays were performed in duplicate or triplicate, and data are presented as mean ± standard deviation (SD).

3. Results and discussion

3.1. UFA profiles

Tables 1–3 display the composition of UFAs in IFs at 0, 2, 4, and 6 months of storage at 37°C. The results indicate that DHA and ARA remained stable in both infant and follow-up IFs. After 6 months, the IF1-L, IF1-H, IF2-L and IF2-H showed no significant changes in DHA and ARA levels compared to the initial values (0 months). However, in IF3-H, both DHA and ARA exhibited a significant decrease after 2 months of storage. Oleic acid (C18:1-9c), linoleic acid (C18:2-9c,12c), and linolenic acid (C18:3-9c,12c,15c), which are among the UFAs with the highest content in the formula, did not significantly change during the storage of IF1-L, IF2-L, IF1-H, IF2-H, and IF3-H. Nevertheless, in IF3-L, linoleic acid and linolenic acid decreased significantly (from 4.05 ± 0.042 to 3.90 ± 0.049 and 0.37 ± 0.006 to 0.35 ± 0.004, respectively) after 2 months of storage, and oleic acid also showed a significant decrease (from 7.37 ± 0.071 to 6.95 ± 0.078) after 4 months of storage. Furthermore, monounsaturated fatty acids (MUFA), PUFA, and UFA in IF3-L significantly decreased after 6 months of storage compared to the initial values (0 months).

Table 1. Unsaturated fatty acid profiles of infant formula during storage (g/100 g).

fatty acid	IF1-L-0	IF1-L-2	IF1-L-4	IF1-L-6	IF1-H-0	IF1-H-2	IF1-H-4	IF1-H-6
C14:1-9c	0.05 ± 0.000^a	0.05 ± 0.001^a	0.05 ± 0.003^a	0.05 ± 0.000^a	0.05 ± 0.003^a	0.04 ± 0.000^a	0.04 ± 0.000^a	0.05 ± 0.001^a
C16:1-9c	0.13 ± 0.003^a	0.13 ± 0.005^a	0.13 ± 0.010^a	0.14 ± 0.004^a	0.15 ± 0.005^a	0.14 ± 0.001^a	0.13 ± 0.006^a	0.14 ± 0.004^a
C18:1-9c	9.58 ± 0.035^a	9.30 ± 0.219^a	9.43 ± 0.028^a	9.68 ± 0.042^a	8.25 ± 0.240^a	8.72 ± 0.007^a	8.47 ± 0.042^a	8.76 ± 0.255^a
C20:1-11c	0.04 ± 0.000^a	0.04 ± 0.001^a	0.04 ± 0.001^a	0.04 ± 0.001^a	0.04 ± 0.002^a	0.04 ± 0.000^a	0.04 ± 0.001^a	0.04 ± 0.002^a
C22:1-13c	0.01 ± 0.001^a	0.01 ± 0.001^a	0.01 ± 0.000^a	0.01 ± 0.002^a	ND	ND	ND	ND
C18:2-9c,12c	5.55 ± 0.028^a	5.39 ± 0.120^a	5.46 ± 0.007^a	5.62 ± 0.021^a	4.73 ± 0.148^a	4.98 ± 0.028^a	4.84 ± 0.014^a	5.02 ± 0.134^a
C18:3-9c.12c,15c	0.51 ± 0.001^a	0.49 ± 0.011^a	0.50 ± 0.006^a	0.51 ± 0.003^a	0.43 ± 0.013^a	0.45 ± 0.004^a	0.44 ± 0.002^a	0.45 ± 0.011^a
C20:3-8c,11c.14c	0.03 ± 0.000^a	0.02 ± 0.000^a	0.03 ± 0.002^a	0.03 ± 0.001^a	0.03 ± 0.000^a	0.02 ± 0.000^a	0.03 ± 0.001^a	0.03 ± 0.001^a
C20:4-5c,8c.11c,14c	0.215 ± 0.008^a	0.186 ± 0.006^a	0.206 ± 0.011^a	0.211 ± 0.013^a	0.253 ± 0.011^a	0.281 ± 0.003^a	0.242 ± 0.013^a	0.271 ± 0.012^a
C20:5-5c.8c,11c,14c,17c	0.02 ± 0.001^a	0.02 ± 0.002^a	0.03 ± 0.001^a	0.03 ± 0.004^a	0.04 ± 0.003^a	0.04 ± 0.001^a	0.03 ± 0.004^a	0.04 ± 0.001^a
C22:6-4c,7c,10c,13c,16c,19c	0.118 ± 0.008^a	0.115 ± 0.006^a	0.133 ± 0.001^a	0.120 ± 0.007^a	0.183 ± 0.010^a	0.200 ± 0.010^a	0.177 ± 0.006^a	0.193 ± 0.001^a
C18:1t	0.12 ± 0.003^a	0.12 ± 0.003^a	0.11 ± 0.000^a	0.12 ± 0.002^a	0.12 ± 0.008^a	0.13 ± 0.006^a	0.12 ± 0.000^a	0.13 ± 0.002^a
C18:2-9t,12t	0.09 ± 0.001^a	0.09 ± 0.002^a	0.10 ± 0.006^a	0.09 ± 0.004^a	0.11 ± 0.004^a	0.11 ± 0.004^a	0.09 ± 0.004b	0.09 ± 0.003b
MUFA	6.44 ± 0.028^a	6.23 ± 0.134^a	6.35 ± 0.035^a	6.51 ± 0.028^a	5.65 ± 0.184^a	5.98 ± 0.021^a	5.76 ± 0.042^a	6.00 ± 0.134^a
PUFA	9.81 ± 0.028^a	9.53 ± 0.219^a	9.66 ± 0.021^a	9.92 ± 0.042^a	8.49 ± 0.247^a	8.95 ± 0.007^a	8.69 ± 0.049^a	8.99 ± 0.262^a
UFA	16.25 ± 0.071^a	15.75 ± 0.354^a	16.00 ± 0.000^a	16.45 ± 0.071^a	14.10 ± 0.424^a	14.90 ± 0.000^a	14.45 ± 0.071^a	15.00 ± 0.424^a
TFA	0.205 ± 0.002^a	0.207 ± 0.004^a	0.211 ± 0.008^a	0.213 ± 0.006^a	0.230 ± 0.012^a	0.238 ± 0.010^a	0.210 ± 0.004^a	0.223 ± 0.006^a

Values are the mean of two measurements. IF1, infant formula (for infants aged 0–6 months). For each row considered, values of the same formula with the same letter are not statistically different at the 5% level (ANOVA test). UFA, unsaturated fatty acids; MUFA, monounsaturated fatty acids; PUFA, polyunsaturated fatty acids; TFA, total trans fatty acids.

Table 2. Unsaturated fatty acid profiles of follow-up infant formula during storage (g/100 g).

fatty acid	IF2-L-0	IF2-L-2	IF2-L-4	IF2-L-6	IF2-H-0	IF2-H-2	IF2-H-4	IF2-H-6
C14:1-9c	0.03 ± 0.000^a	0.04 ± 0.001^a	0.03 ± 0.000^a	0.03 ± 0.002^a	0.03 ± 0.001^a	0.03 ± 0.002^a	0.03 ± 0.000^a	0.03 ± 0.000^a
C16:1-9c	0.10 ± 0.002b	0.11 ± 0.002^a	0.10 ± 0.000b	0.10 ± 0.002c	0.12 ± 0.007^a	0.12 ± 0.005^a	0.12 ± 0.007^a	0.12 ± 0.003^a
C18:1-9c	7.61 ± 0.191^a	7.69 ± 0.014^a	7.72 ± 0.106^a	7.38 ± 0.269^a	7.38 ± 0.042^a	7.35 ± 0.212^a	7.18 ± 0.205^a	7.33 ± 0.028^a
C20:1-11c	0.04 ± 0.000^a	0.04 ± 0.001^a	0.04 ± 0.000	0.04 ± 0.002^a	0.04 ± 0.002^a	0.04 ± 0.001^a	0.04 ± 0.001^a	0.04 ± 0.000^a
C22:1-13c	0.01 ± 0.000^a	0.01 ± 0.002^a	0.01 ± 0.000^a	NDb	0.01 ± 0.000^a	NDb	NDb	NDb
C18:2-9c, 12c	4.47 ± 0.120^a	4.52 ± 0.014^a	4.53 ± 0.071^a	4.33 ± 0.170^a	4.29 ± 0.035^a	4.25 ± 0.127^a	4.17 ± 0.120^a	4.25 ± 0.021^a
C18:3-9c.12c,15c	0.40 ± 0.014^a	0.41 ± 0.001^a	0.41 ± 0.001^a	0.39 ± 0.016^a	0.39 ± 0.002^a	038 ± 0.012^a	0.37 ± 0.011^a	0.38 ± 0.002^a
C20:3-8c,11c.14c	0.02 ± 0.002^a	0.02 ± 0.000^a	0.02 ± 0.000^a	0.02 ± 0.002^a	0.03 ± 0.002^a	0.03 ± 0.000^a	0.03 ± 0.002^a	0.03 ± 0.001^a
C20:4-5c,8c.11c,14c	0.202 ± 0.011^a	0.196 ± 0.003^a	0.180 ± 0.003^a	0.184 ± 0.003^a	0.264 ± 0.018^a	0.250 ± 0.001^a	0.245 ± 0.013^a	0.238 ± 0.013^a
C20:5-5c.8c,11c,14c,17c	0.02 ± 0.000^a	0.03 ± 0.001^b	0.02 ± 0.001^a	0.02 ± 0.004^a	0.04 ± 0.000^a	0.04 ± 0.003^a	0.04 ± 0.002^a	0.04 ± 0.001^a
C22:6-4c,7c,10c,13c,16c,19c	0.121 ± 0.003^a	0.116 ± 0.008^a	0.112 ± 0.005^a	0.114 ± 0.001^a	0.201 ± 0.001^a	0.190 ± 0.012^a	0.187 ± 0.007^a	0.174 ± 0.007^a
C18:1t	0.09 ± 0.002b	0.09 ± 0.001^b	0.09 ± 0.000^b	0.11 ± 0.001^a	0.09 ± 0.005^a	0.09 ± 0.002^a	0.10 ± 0.004^a	0.09 ± 0.003^a
C18:2-9t,12t	0.08 ± 0.011^a	0.08 ± 0.010^a	0.07 ± 0.004^a	0.07 ± 0.001^a	0.08 ± 0.001^a	0.08 ± 0.008^a	0.08 ± 0.000^a	0.08 ± 0.007^a
MUFA	5.24 ± 0.148^a	5.32 ± 0.028^a	5.28 ± 0.064^a	5.07 ± 0.184^a	5.20 ± 0.057^a	5.13 ± 0.127^a	5.03 ± 0.156^a	5.10 ± 0.042^a
PUFA	7.79 ± 0.191^a	7.89 ± 0.014^a	7.90 ± 0.106^a	7.55 ± 0.276^a	7.59 ± 0.049^a	7.54 ± 0.219^a	7.37 ± 0.198^a	7.51 ± 0.028^a
UFA	13.05 ± 0.354^a	13.20 ± 0.000^a	13.20 ± 0.141^a	12.60 ± 0.424^a	12.80 ± 0.141^a	12.65 ± 0.354^a	12.45 ± 0.354^a	12.65 ± 0.071^a
TFA	0.169 ± 0.009^a	0.171 ± 0.009^a	0.160 ± 0.004^a	0.186 ± 0.003^a	0.171 ± 0.005^a	0.169 ± 0.006^a	0.185 ± 0.004^a	0.170 ± 0.009^a

Values are the mean of two measurements. IF2, follow-up infant formula (for infants aged 6–12 months). For each row considered, values of the same formula with the same letter are not statistically different at the 5% level (ANOVA test). UFA, unsaturated fatty acids; MUFA, follow-upmonounsaturated fatty acids; PUFA, polyunsaturated fatty acids; TFA, total trans fatty acids.

Table 3. Unsaturated fatty acid profiles of young children formula during storage (g/100 g).

fatty acid	IF3-L-0	IF3-L-2	IF3-L-4	IF3-L-6	IF3-H-0	IF3-H-2	IF3-H-4	IF3-H-6
C14:1-9^c	0.05 ± 0.000^a	0.05 ± 0.001^a	0.04 ± 0.001^a	0.04 ± 0.000^a	0.04 ± 0.000^a	0.04 ± 0.000^a	0.05 ± 0.004^a	0.04 ± 0.002^a
C16:1-9^c	0.12 ± 0.001^a	0.12 ± 0.004^a	0.12 ± 0.004^a	0.12 ± 0.003^a	0.13 ± 0.001^a	0.12 ± 0.000^b	0.12 ± 0.002^b	0.13 ± 0.004^a
C18:1-9^c	7.37 ± 0.071^a	7.27 ± 0.106^a	6.95 ± 0.078^b	6.99 ± 0.071^b	6.86 ± 0.106^a	6.54 ± 0.078^a	6.51 ± 0.120^a	6.74 ± 0.184^a
C20:1-11^c	0.04 ± 0.000^a	0.04 ± 0.001^a	0.03 ± 0.000^b	0.03 ± 0.000^b	0.03 ± 0.000^a	0.03 ± 0.001^a	0.03 ± 0.000^a	0.03 ± 0.000^a
C22:1-13^c	ND	ND	ND	ND	ND	ND	ND	ND
C18:2-9c,12^c	4.05 ± 0.042^a	3.90 ± 0.049^b	3.81 ± 0.042^b	3.85 ± 0.035^b	3.72 ± 0.057^a	3.55 ± 0.035^a	3.54 ± 0.064^a	3.67 ± 0.106^a
C18:3-9c.12c,15^c	0.37 ± 0.006^a	0.35 ± 0.004^b	0.35 ± 0.004^b	0.35 ± 0.004^b	0.33 ± 0.003^a	0.32 ± 0.003^a	0.32 ± 0.006^a	0.33 ± 0.009^a
C20:3-8c,11c.14^c	0.02 ± 0.000^a	0.02 ± 0.000^a	0.02 ± 0.001^a	0.02 ± 0.001^a	0.03 ± 0.000^a	0.03 ± 0.000^b	0.03 ± 0.001^b	0.03 ± 0.000
C20:4-5c,8c.11c,14^c	0.197 ± 0.007^a	0.199 ± 0.012^a	0.188 ± 0.007^a	0.199 ± 0.011^a	0.264 ± 0.000^a	0.243 ± 0.001^c	0.247 ± 0.001^c	0.259 ± 0.004^b
C20:5-5c.8c,11c,14c,17^c	0.03 ± 0.001^a	0.02 ± 0.003^a	0.03 ± 0.000^a	0.03 ± 0.000^a	0.04 ± 0.001^a	0.03 ± 0.000^b	0.04 ± 0.002^b	0.04 ± 0.001^a
C22:6-4c,7c,10c,13c,16c,19^c	0.130 ± 0.008^a	0.128 ± 0.006^a	0.127 ± 0.001^a	0.122 ± 0.006^a	0.194 ± 0.000^a	0.175 ± 0.002^b	0.176 ± 0.007^b	0.183 ± 0.006^b
C18:1t	0.14 ± 0.003^a	0.13 ± 0.002^a	0.13 ± 0.001^a	0.13 ± 0.006^a	0.13 ± 0.006^a	0.12 ± 0.001^a	0.13 ± 0.004^a	0.13 ± 0.003^a
C18:2-9t,12t	0.08 ± 0.000^a	0.08 ± 0.004^a	0.08 ± 0.006^a	0.08 ± 0.002^a	0.07 ± 0.007^a	0.07 ± 0.008^a	0.08 ± 0.011^a	0.08 ± 0.002^a
MUFA	4.81 ± 0.064^a	4.62 ± 0.049^b	4.53 ± 0.049^b	4.57 ± 0.042^b	4.58 ± 0.057^a	4.35 ± 0.049^a	4.35 ± 0.064^a	4.52 ± 0.106^a
PUFA	7.58 ± 0.071^a	7.47 ± 0.113^a	7.15 ± 0.078^b	7.19 ± 0.071^b	7.06 ± 0.113^a	6.73 ± 0.071^a	6.71 ± 0.113^a	6.95 ± 0.191^a
UFA	12.40 ± 0.141^a	12.10 ± 0.141^a	11.70 ± 0.141^c	11.75 ± 0.071^bc	11.65 ± 0.212^a	11.10 ± 0.141^a	11.05 ± 0.212^a	11.50 ± 0.283^a
TFA	0.213 ± 0.003^a	0.213 ± 0.006^a	0.212 ± 0.007^a	0.209 ± 0.004^a	0.200 ± 0.003^a	0.187 ± 0.007^a	0.207 ± 0.015^a	0.216 ± 0.005^a

Values are the mean of two measurements. IF3, young children formula (for young children aged 12 to 36 months). For each row considered, values of the same formula with the same letter are not statistically different at the 5% level (ANOVA test). UFA, unsaturated fatty acids; MUFA, monounsaturated fatty acids; PUFA, polyunsaturated fatty acids; TFA, total trans fatty acids.

Notably, PUFAs are easily attacked by free radicals that react with their double bonds, thereby yielding several products such as short-chain aldehydes. The susceptibility of fatty acids to oxidation largely depends on their degree of unsaturation (Jia et al., 2019). This is consistent with the conclusion that only DHA and ARA in IF3-H significantly decreased during the 6-month storage period (Table 3). In addition, during the 6-month storage process, there was no significant decrease in DHA and ARA in IF3-L, whereas there was a significant decrease in linoleic acid, linolenic acid, and oleic acid, which may be related to the lower contents of DHA and ARA in the formula. Chávez-Servín et al. (2009) found that during the 18 months storage period of IFs, only linoleic acid showed a significant decrease in all fatty acids, while other fatty acids, including DHA and ARA (0.02–0.04 g/100 g, 0.03–0.08 g/100 g, respectively), did not show significant changes. According to Tables 1–3 and section 2.1. Sample preparation, the percentage of DHA and ARA in total fatty acids was IF3 > IF2 > IF1(and IF1-H > IF1-L, IF2-H > IF2-L, IF3-H > IF3-L), which may further indicate that in the case of similar content (mass percentage), the higher the percentage of DHA and ARA in total fatty acids of IFs, the lower their stability.

3.2. Vitamin E and vitamin C content

The vitamin E content (α-tocopherol equivalents) of the analyzed formulas stored at 37°C for 6 months is presented in Figure 1a. Initially, IF1-L and IF1-H contained 15.53 mg α-TE/100 g and 15.50 mg α-TE/100 g of vitamin E, respectively. No significant changes were observed relative to the initial values (0 months) at the end of the study period. In IF2-L and IF2-H, vitamin E exhibited a significant decrease after 2 and 4 months of storage, respectively. Conversely, in the 6-month storage of IF3-L and IF3-H formulas, vitamin E in IF3-L significantly decreased in the second month. After 6 months, the vitamin E loss rates for IF1-L, IF1-H, IF2-L, IF2-H, IF3-L, and IF3-H were −0.6%, 1.5%, 10.1%, 6.3%, 8.1%, and 4.7%, respectively. These trends align with findings reported by Chávez-Servín et al. (2008a), where IFs stored at 25°C and 40°C for 6 months with an initial vitamin E content of 27.3 mg α-TE/100 g, experienced losses of 4.8% and 6.5%, respectively. For IFs with an initial vitamin E content of 12.1 mg α-TE/100 g, the losses were 8.2% and 12.3%, respectively. This research suggests that lower initial vitamin E content in IFs correlates with higher vitamin E loss during storage.

Figure 1. Vitamin E and vitamin C content of the analyzed samples.

Figure 1b displays the vitamin C content of the IFs. After six months of storage, there was a significant decrease in vitamin C in all formulas, with most of this decrease occurring in the fourth month of storage, except for IF1-H,which experienced it in the sixth month. After 6 months, the vitamin C loss rates for IF1-L, IF1-H, IF2-L, IF2-H, IF3-L, and IF3-H were 3.9%, 5.7%, 6.0%, 6.0%, 7.7%, and 6.1%, respectively. This results align with those reported by Chávez-Servín (Chávez-Servín et al., 2008). In their study, when IFs were stored at 40°C for 6 months, vitamin C decreased by approximately 12.5%. However, the findings slightly differ from Manglano (Manglano et al., 2004), whose study observed a sharp 20% decrease in vitamin C in IFs stored at 37°C for 1 month, followed by a gradual decrease as the storage period extended to 7 months.

Vitamin C and vitamin E are added to IFs, both to improve vitamin content and to prevent lipid oxidation during manufacture and storage, thereby helping to extend product shelf-life (Chávez-Servín et al., 2008a). Previous studies have shown that the presence of vitamin E can reduce the loss of vitamin C (Manglano et al., 2004). In this study, significant reductions in vitamin E in IF2-H and IF3-L occurred as early as the second month, while significant reductions in vitamin C in IFs (IF1-L, IF2-L, IF2-H, IF3-L and IF3-H) occurred as early as the fourth month. In addition, up to six months of storage, all formulas showed a significant decrease in vitamin C compared to the initial value, and the significant decrease in vitamin C in IF1-H (with high vitamin E content) only occurred in the sixth month. This indirectly indicates that when vitamin E and vitamin C are present simultaneously in IFs, vitamin E may preferentially play an antioxidant role, which can reduce the loss of vitamin C.

3.3 TBARS

Table 4 displays the TBARS content of reconstituted IFs at different storage durations. The highest TBARS values for all six formulas were observed in the fourth or sixth month of storage, while the lowest values corresponded to the initial measurements. ANOVA analysis indicated that in the sixth month of storage, TBARS values for all formulas were significantly higher than their initial levels., Some specific formulas, such as IF1-H, IF2-L, and IF2-H, exhibited a notable increase in TBARS levels in the second month of storage. A paired T-test revealed significantly differences in TBARS content between IF1-H and IF1-L at the fourth and sixth months, a pattern also observed IF2-H and IF2-L. Conversely, paired T-tests for IF3-L and IF3-H showed no significant differences between the two groups at each storage time point, though TBARS levels in IF3-H were consistently higher than those in IF3-L.Several studies have used TBARS to assess the stability of milk powder and IFs (Mccluskey et al., 1997; Semeniuc, 2009). The determination of TBARS does not require calibration curves, making it an effective means of evaluating product oxidation status (Semeniuc, 2009). As shown in Table 4, during the storage of IFs, TBARS levels increased over time for almost all formulas (except for IF1-L and IF2-L), aligning with the trends observed in a study by Mccluskey on whole milk powder stored at 15°C and 30°C for 12 months (Mccluskey et al., 1997). This trend is also consistent with malondialdehyde (MDA) values, determined through the TBA test, for IFs stored at 20°C for 14 days after opening, as studied by Stefania et al. 2015.

Table 4. TBARS contents of IFs stored at different storage time (λ = 532 nm absorbance).

		Storage time
items	groups	0 mon	2 mon	4 mon	6 mon
TBARS	IF1-L	0.234 ± 0.045^bc	0.390 ± 0.006^ab	0.488 ± 0.011^a	0.473 ± 0.057^a
	IF1-H	0.253 ± 0.025^c	0.475 ± 0.010^b	0.558 ± 0.006^ab*	0.605 ± 0.057^a*
	IF2-L	0.219 ± 0.002^b	0.316 ± 0.056^a	0.294 ± 0.010^ab	0.325 ± 0.010^a
	IF2-H	0.237 ± 0.008^d	0.384 ± 0.016^c	0.542 ± 0.004^b*	0.657 ± 0.021^a*
	IF3-L	0.209 ± 0.007^b	0.229 ± 0.003^b	0.240 ± 0.016^b	0.331 ± 0.007^a
	IF3-H	0.221 ± 0.022^c	0.280 ± 0.051^bc	0.346 ± 0.003^b	0.474 ± 0.015^a

Values are the mean of two measurements. For each row considered, values with the same letter are not statistically different at the 5% level (ANOVA test). * There is a significant difference compared to the corresponding formula (p <  .05).

3.4 VOCs in IFs

As shown in Figure 2 and Supplement 1, a complex mixture of volatiles including 27 aldehydes, 13 ketones, 14 alkanes, 14 alcohols, 17 olefins, 3 carboxylic acids, 7 esters, and 14 other compounds, was observed in the IFs after 6 months of storage. Among the 109 volatiles identified (Supplement 1), hexanal was the most abundant, followed by heptanal and nonanal. Research shows that hexanal, heptanal and pentanal are the main decomposition product of n-6 polyunsaturated fatty acid (n-6 PUFA) (Frankel, 1993; Jimenez-Alvarez et al., 2008). As shown in Tables 1–3, the highest content of UFA in the six formulas (IF1-L, IF1-H, IF2-L, IF2-H, IF3-L and IF3-H) was linoleic acid (n-6 PUFA), and ARA (n-6 PUFA) was also added to the formulas, which may be the reason why the content of hexanal in the volatile products of IFs was the highest. In addition, the main decomposition product of n-3 PUFA oxidation is propanal (Romeu-Nadal et al., 2004). However, in this study, propanal was not detected in any formulation during the 6-month storage period. This finding is consistent with those of previous studies (Hausner et al., 2009; Romeu-Nadal et al., 2004; Van Ruth et al., 2006; Zou & Akoh, 2015), which also did not detect propanal in the headspace of IFs products.

Figure 2. Cluster analysis of the formulas after 6 months of storage data (VOCs) as obtained by the GC-MS analysis. (a) IF1 (infant formula), (b) IF2 (follow-up infant formula) and (c) IF3 (young children formula). Each sample has been analyzed in triplicates.

* There is a significant difference compared to the 0 mon value (p < .05)

Previous studies have shown that the enhancement of PUFA in milk increases the types of VOCs in milk and increases the types of VOCs in milk storage, thus, making the VOCs of milk more complex (Venkateshwarlu et al., 2004). From Supplement 1 and Figure 2, it can be seen that the higher the fat content, the more types of VOCs can be detected during formula storage, and high-dose fortified DHA and ARA will cause the formula to produce more VOCs during storage. During the 6-month storage process, 46, 78, 26, 45, 24, and 29 VOCs were detected in IF1-L, IF1-H, IF2-L, IF2-H, IF3-L, and IF3-H, respectively. Van Ruth et al. (2006) also found a similar finding that formulas rich in fat can detect more VOCs than formulas with low fat content. In addition, in this study, some VOCs were observed in samples stored only in IF1-H for the fourth month, and in IF2-H and IF3-H for the second month. Such as “2,3-Octanedione”, “Cyclohexanecarboxaldehyde”, “ Carbamic acid, 2-(dimethylamino)ethyl ester”, “Cycloheptanol, 2-methylene”, “1,4-Hexadiene, 4-methyl-” and “4-trifluoroacetoxyhexadecane” only appear in the fourth month of IF1-H storage; “2,4-Nonadienal, (E, E)-”, “2,4-Decadienal, (E,E)-”, “Octadecane, 1-chloro-”, “2-Hexenal, (E)-”, “1-Tetradecanol, 14-chloro-”, “Decanal”, “4-Heptenal, (Z)-” and “2-Pentanamine, 2,4,4-trimethyl-” only appeared in the second month of IF2-H storage; “1-Penten-3-one”, “p-Xylene”, “2,4-Nonadienal, (E,E)-”, “3-Octen-2-one”, “Pentanal, 2,4-dimethyl-” and “2-Hexenal, (E)-” only appeared in the second month of IF3-H storage. These compounds, which only appear in the high DHA and ARA formulas, indicate that an increase in DHA and ARA dosages can also affect the types of VOCs in the formula.

Partial Least Squares-Discriminant Analysis (PLS-DA) showed that samples low in DHA and ARA were separated from those high in DHA and ARA (Figure 3a,b). Figure 3(a,b) depict the kinetics of the selected IFs over the entire storage period. The evolution of product deterioration as a function of time was clearly observed for all four samples. A similar result has been reported in the literature. Fenaille et al. (2003) utilized PCA to distinguish various IFs based on changes in VOCs during storage. In their study, VOCs were treated using multivariate analysis, achieving an intuitive differentiation of the differences in IFs at different storage periods. However, this study did not clearly distinguish the changes in IF3 formula throughout the storage period using PLS-DA (Figure 3c), thereby further indicating that IF3 has a lower degree of oxidation. This is consistent with the TBARS results measured using IF3-L and IF3-H (Table 4), which indicates that IF3-L and IF3-H have similar oxidation levels.

Figure 3. PLS-DA of the formulas after 6 months of storage data (VOCs) as obtained by the GC-MS analysis. (a) IF1 (infant formula), (b) IF2 (follow-up infant formula) and (c) IF3 (young children formula). Each sample has been analyzed in triplicates.

4. Conclusions and outlook

The addition of high doses of DHA and ARA did not impact the stability of vitamin C in IFs. However, it did lead to a reduction in the stability of vitamin E in follow-up infant formula. Furthermore, DHA and ARA exhibited a high degree of stability in IFs, with only IF3-H showing a notable decrease in DHA and ARA levels during the 6-month storage period. Nevertheless, there were variations in the profiles of VOCs and TBARS between the IFs containing high doses of DHA and ARA compared to those containing low doses of these components. Overall, these finding from our study affirm that elevated levels of DHA and ARA can diminish the oxidative stability of IFs, resulting in an increase in both VOCs and TBARS.

As the marine environment has deteriorated dramatically in recent years, the safety of the sources and the levels of high-risk contaminants of DHA and ARA, which are widely used as food fortifiers in infant and children’s foods, have become one of the hotspots of concern. In Europe, there is strict regulation by EU legislation specifically applicable to the field of veterinary health and food safety (Bondoc, 2016a, 2016b, 2016c, 2016d). In China, there are also corresponding regulations, such as the requirements for the limitation of lead, chromium, mercury and other heavy metal elements in raw materials, etc. However, the limitation of the current regulation lies in the low sensitivity of the detection methods, especially in the detection of the end products. Therefore, in the future, the active development of testing methods with higher sensitivity and better stability, as well as the increase of the whole chain of products from raw materials to the end of the regulatory system will be an important development direction of food quality and safety risk assessment.

Acknowledgments

This project was supported by Shanghai State-owned Assets Supervision and Administration Commission Enterprise Innovation Development and Capacity Enhancement Program (No. 2022013). We thank Brightdairy (Shanghai, China) for providing the samples of infant formula powders.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplemental material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/19476337.2023.2300812.

Additional information

Funding

The work was supported by the Shanghai State-owned Assets Supervision and Administration Commission Enterprise Innovation Development and Capacity Enhancement Program [No. 2022013].