Vitamin A stability during storage of fortified gari produced using different fortification strategies

The potential of cereal bran to stabilise RP during storage of gari

The RP retention as a function of storage time for RP-fortified gari samples without bran addition and with the addition of native wheat bran, heat-treated rice bran and toasted wheat bran is shown in Fig. 1. The initial vitamin A content of these four samples ranged from 2.5–3.8 µg retinol equivalents (REeq) per gram gari. RP degradation occurred relatively fast in the first week and then slowed down gradually. When comparing the different bran types, toasted wheat bran showed the best RP stabilisation, followed by heat-treated rice bran and then native wheat bran. This ranking corresponds to the RP stabilisation as observed in model systems in our previous studies19,20. However, for the gari samples the differences between the different samples are small and not always significant. In addition, the difference in RP retention between the control sample, without bran addition, and the samples with bran addition was relatively small. After one week of accelerated storage, no significant differences (p < 0.05) were observed between the samples. For all four samples, between 43 and 55% of RP was retained after one week. After two and four weeks of storage, a clear stabilising effect was observed for the sample with toasted wheat bran. The RP retention for this gari sample was 53 ± 5% and 34 ± 9% after two and four weeks, respectively. For the sample without bran addition, the RP retention was only 36 ± 5% after two weeks and 19.4 ± 0.3% after four weeks of accelerated storage. However, these differences were not significant (p < 0.05) due to the relatively high batch-to-batch variation. After eight weeks of storage, about 2 to 10% of RP was retained in all samples. The RP retention was 1.9 ± 1.2%, 4.9 ± 0.9%, 7.2 ± 0.2% and 10.4 ± 1.1% for the gari sample with native wheat bran, without bran, with heat-treated rice bran and with toasted wheat bran, respectively. Hereby, the RP retention of the sample with toasted wheat bran was twice as high and significantly higher than the RP retention of the control sample without bran (p < 0.05).

Fig. 1: The retention of retinyl palmitate (RP) in RP-fortified gari during accelerated storage at 60 °C and 70% relative humidity.
figure 1

Gari production was done in duplicate, and RP analysis was also done in duplicate, resulting in a 2 × 2 setup. Error bars represent standard deviations of duplicate gari production followed by accelerated storage. The initial vitamin A contents of the gari samples were 3.80 ± 0.02 µg retinol equivalents (REeq)/g (no bran), 3.2 ± 0.3 µg REeq/g (native wheat bran), 3.22 ± 0.07 µg REeq/g (toasted wheat bran) and 2.5 ± 0.2 µg REeq/g (heat-treated rice bran).

Based on these results, we can conclude that only toasted wheat bran can noticeably stabilise RP during gari storage. The stabilisation of RP by toasted wheat bran can be explained by the high antioxidant capacity of wheat bran and the absence of wheat bran endogenous enzymes due to toasting. This stabilisation mechanism was investigated in depth in our previous studies19,20,21. However, in gari, the RP-stabilising effect of cereal bran is limited, especially compared to the stabilising effect observed in our previous study where RP-enriched oil was mixed with toasted wheat bran19. The reason for this is assumed to be twofold. Firstly, the concentration of RP in the fortified gari is about a factor of 250 lower than the RP content of the model system consisting of 0.16% RP, 19.84% soy oil and 80.00% cereal bran, used in our previous study19. In another study, we showed that a high RP concentration combined with a high lipid content causes RP to act as a pro-oxidant, accelerating its own oxidation and lipid oxidation. Moreover, a negative correlation was observed between lipid oxidation and RP stability. This pro-oxidative effect is, to a certain extent, counteracted by bran antioxidants, such as polyphenols, phenolic acids and tocopherols, resulting in a pronounced RP stabilisation21,22,23. Given the low RP concentration and the low lipid content of the fortified gari, the pro-oxidative effect of RP is assumed to be absent here. This can, in part, explain the limited stabilising effect of cereal bran. It is therefore important to investigate the RP stability in the food matrix of interest fortified with the desired concentration of RP. Secondly, it is possible that the gari matrix as such stabilises RP to a certain extent, resulting in a less pronounced effect on RP stability when bran is added.

When comparing the RP stability in gari to other fortified food products, both lower and higher RP retentions compared to the RP retention in gari have been reported in the literature. Kim et al.15 reported an 85% RP loss in RP-fortified cornflakes after four weeks of storage at ambient temperature. When stored at an elevated temperature of 45 °C, RP degradation occurred faster and about 85% was already lost after two weeks. When comparing the RP stability observed in the study of Kim et al.15 with the RP stability in gari, it can be stated that RP shows good RP stability in the gari matrix. After two weeks of storage at 60 °C and 70% relative humidity, about 65% RP was lost in the fortified gari. This is lower than the RP loss observed in the study of Kim et al.15, even though the storage conditions in the study of Kim et al.15 were milder. However, other studies also reported higher RP retentions. For example, Lee et al.17 reported an RP retention higher than 90% in fortified rice after six weeks of storage at 35 °C and 80% relative humidity. Despite the milder storage conditions, the RP stability in the fortified rice is presumably higher than in gari. Nevertheless, comparing storage experiments performed at different temperatures and different relative humidities is challenging, especially when the storage temperature exceeds 40 °C. The Arrhenius plot can be used to calculate the reaction rate at different temperatures, whereby a temperature increase of 10 °C corresponds to a doubling of the reaction rate24. However, it has been shown that above 40 °C, other oxidation mechanisms can occur, which impedes the use of the Arrhenius equation. Furthermore, the combined effect of temperature and humidity further complicates the extrapolation to ambient conditions25. Despite this, we can conclude that the gari matrix as such provides a moderate RP stability, making gari a good food product of choice for RP fortification. It is hypothesised that the low RP concentration and the low oil content of gari prevent fast RP and lipid oxidation, resulting in good RP stability. In addition, the gari matrix might provide physical protection against oxidation. However, there is still room to improve the RP stability. The RP stability could be further improved by optimising packaging and storage conditions. RP oxidation can be prevented by limiting the contact with oxygen using vacuum or modified atmosphere packaging. In addition, storing gari at low temperatures can be done to slow down RP degradation, although this is expected to be difficult in practice26,27. Next to storage and packaging conditions, advanced RP stabilisation techniques such as microencapsulation could be explored28,29.

Apart from the RP stability, the gari quality and sensory characteristics are important. When looking at the effect of cereal bran addition on the quality characteristics of gari, no significant differences were observed (Table 1). All investigated gari samples had a moisture content lower than 12% and an ash content lower than 2.75%, which is in accordance with the Codex Alimentarius guidelines on gari quality30. However, the TTA was slightly lower than the specified minimum of 0.6%, which suggests the fermentation time should have been increased. Moreover, all gari samples showed a swelling capacity of around a factor three, which indicates good gari quality5. When looking at the colour, the gari samples enriched with cereal bran, particularly wheat bran, had a visually observed darker colour. The colour of both native and toasted wheat bran is darker than the colour of gari, with toasted wheat bran having the most intense dark brown colour due to the Maillard reaction during toasting. However, no significant differences in colour were measured. The colour change, likely together with other sensory attributes, is assumed to be the main limitation for applying cereal bran as a stabilising agent.

Table 1 Overview of the quality characteristics of RP-fortified gari with the addition of native wheat bran, toasted wheat bran and heat-treated rice bran

Despite the limited effect of cereal bran addition on the gari quality, no pronounced stabilising effect of cereal bran on RP was observed during the storage of gari (Fig. 1). Toasted wheat bran was the only bran sample that showed a small but significant potential to stabilise RP. However, this stabilisation should be verified at ambient storage conditions. Unfortunately, wheat bran is typically not locally produced in Ghana, which likely implies a higher ingredient cost. To circumvent this, the use of other locally sourced cereals could be explored. However, given the limited effect of cereal bran addition on the RP stability, it is possible that the investment cost would not make up for the improvement in RP stability and the impact on gari sensory aspects. To investigate this, a financial analysis should be carried out. Next to the production process for RP-fortified gari used in this study, the addition of a small amount of a mixture of RP, oil and cereal bran with a high RP concentration as an additive to gari could be another valuable fortification strategy. Hereby, cereal bran is used to stabilise RP in the additive which is highly concentrated in RP. Hence, the focus, in this case, is on RP stabilisation during storage of the additive rather than RP stabilisation during gari storage. Moreover, the amount of cereal bran added to gari would be much lower in this case compared to the amount added to the gari prepared in this study. This will reduce both the effect on sensory aspects as well as the investment cost.

Comparison of three strategies for the production of vitamin A-fortified gari

The local production of RP-fortified gari will be challenging as the RP used in this study is chemically synthesised on an industrial scale and not easily accessible to local producers. Therefore, two other fortification strategies were investigated. These include the addition of red palm oil to gari and the use of biofortified yellow cassava. Both red palm oil and yellow cassava are rich in provitamin A, mainly present as β-carotene1. In addition, both ingredients are locally produced in Ghana and, more broadly, in West Africa. This implies that both strategies could be feasible for local, small-scale gari producers. As the addition of cereal bran only had a limited effect on the RP stability, the use of cereal bran to stabilise vitamin A, in the form of β-carotene present in red palm oil and yellow cassava, will not be further investigated.

In Fig. 2, the vitamin A retention, expressed as a percentage of the initial RP or β-carotene content, is shown as a function of the storage time. Based on these results, it can be concluded that the red palm oil-fortified gari provided the highest vitamin A stability. After one week of accelerated storage, 94 ± 15% of vitamin A was retained for red palm oil-fortified gari, whereas the vitamin A retention was only 48 ± 9% for RP-fortified gari and 54 ± 1% for yellow cassava gari. After eight weeks of storage, the vitamin A retention for red palm oil fortified gari was 22.6 ± 0.1%, which was more than four times higher compared to the two other fortification strategies. In the literature, a limited number of studies were found that investigated the stability of β-carotene in red palm oil-fortified gari and yellow cassava gari1,31,32,33. For RP-fortified gari, on the other hand, no studies were found. In the study of Abiodun et al.31, only yellow cassava gari was investigated and β-carotene retentions ranging from 47–72% were observed after two months of storage at ambient conditions. In the study of Bechoff et al.1, both fortification strategies were compared. A β-carotene retention of around 70% was observed for red palm oil-fortified gari stored for four weeks at 40 °C. For yellow cassava gari, a substantially lower β-carotene retention of 23% was measured after four weeks of storage. Although a lower storage temperature was used in the study of Bechoff et al.1, resulting in higher β-carotene retention, the higher β-carotene stability observed in red palm oil-fortified gari compared to yellow cassava gari is in line with our results. This can presumably be attributed to the presence of natural antioxidants in red palm oil and the protective effect of the oil matrix on β-carotene34.

Fig. 2: The vitamin A retention monitored during accelerated storage of gari fortified with retinyl palmitate (RP), gari fortified with red palm oil and biofortified gari made with yellow cassava.
figure 2

The gari samples were stored for eight weeks at 60 °C and 70% relative humidity. Gari production was done in duplicate and vitamin A analysis was also done in duplicate, resulting in a 2 × 2 setup. Error bars represent standard deviations of duplicate gari production followed by accelerated storage. The initial vitamin A content of RP- fortified gari was 3.80 ± 0.02 µg retinol equivalents (REeq) per gram gari, using a conversion factor of 1.83:1. The initial vitamin A content of red palm oil-fortified gari and yellow cassava gari was 1.88 ± 0.08 µg REeq/g and 2.5 ± 0.5 µg REeq/g, respectively, using a conversion factor of 6:1 for β-carotene. When using a conversion factor of 2.4:1 for red palm oil-fortified gari and 4.5:1 for yellow cassava gari, the initial vitamin A content was 4.7 ± 0.2 µg REeq/g and 3.3 ± 0.6 µg REeq/g, respectively.

Not only the vitamin A retention but also the vitamin A content is important to evaluate the effectiveness of the investigated fortification strategies. To this end, the RP or β-carotene content of the fortified gari samples was converted to retinol equivalents (REeq). This is, however, challenging as largely different conversion factors have been reported for β-carotene35. This is partly due to the effect of the food matrix on the bioavailability of β-carotene36. Moreover, little is known about the bioavailability of RP in the gari matrix. Using different conversion factors can substantially affect the resulting conclusions. To illustrate this, the vitamin A content (µg REeq/g gari) of red palm oil-fortified gari and yellow cassava gari was calculated using two different conversion factors for β-carotene. For RP-fortified gari, one µg REeq corresponds to 1.83 µg RP37. When the generally accepted conversion factor of 6 µg β-carotene per µg REeq was used for red palm oil-fortified gari and yellow cassava gari, the RP-fortified gari sample had the highest initial vitamin A content (3.80 ± 0.02 µg REeq/g), followed by the yellow cassava gari (2.5 ± 0.5 µg REeq/g) and the red palm oil-fortified gari (1.88 ± 0.08 µg REeq/g). Assuming a portion size of 50 g, one portion of RP-fortified gari, yellow cassava gari or red palm oil-fortified gari provides respectively about 27%, 18% or 13% of the recommended daily vitamin A intake. The vitamin A content is here compared with the recommended daily intake as this value was used to calculate the amount of vitamin A added to the gari. It should be noted that a calculation based on and a subsequent comparison with the estimated average requirement would be more appropriate38. However, after one week of accelerated storage, the differences in vitamin A content between the three different samples were reduced and the vitamin A content of all gari samples ranged between 1.3 and 1.8 µg REeq/g. After eight weeks of accelerated storage, this content was further reduced to 0.1–0.4 µg REeq/g, with the red palm oil-fortified gari having the highest vitamin A content. Different conclusions can be drawn when using the β-carotene conversion factors determined by Zhu et al.11 and La Frano et al.39, specifically determined for red palm oil-fortified gari and yellow cassava gari. Hereby, one µg REeq corresponded to 2.4 µg β-carotene in red palm oil-fortified gari or 4.5 µg β-carotene in yellow cassava gari11,39. The difference in conversion factor might be explained by the presence of oil in the red palm oil-fortified gari, which can enhance the uptake of β-carotene. In addition, the possible inclusion of β-carotene in plant cell walls might reduce its bioaccessibility in yellow cassava gari40,41. Using these conversion factors, red palm oil-fortified gari had the highest initial vitamin A content (4.7 ± 0.2 µg REeq/g), followed by RP-fortified gari (3.80 ± 0.02 µg REeq/g) and yellow cassava gari (3.3 ± 0.6 µg REeq/g). This content corresponds to 34%, 27% and 24% of the recommended daily vitamin A intake per portion of 50 g, respectively. After one week of storage, the vitamin A content decreased to 4.4 ± 0.5 µg REeq/g, 1.8 ± 0.4 µg REeq/g and 1.8 ± 0.3 µg REeq/g for red palm oil-fortified gari, RP-fortified gari and yellow cassava gari, respectively. After eight weeks, 1.06 ± 0.04 µg REeq/g was retained in red palm oil-fortified gari, whereas less than 0.2 µg REeq/g was retained for the two other fortification strategies. Overall, it can be stated that the initial vitamin A content of the fortified gari samples ranged from 1.9 to 4.7 µg REeq/g, taking into account different β-carotene conversion factors. This aligns with the study of Bechoff et al.1, who measured an initial vitamin A content ranging from 2–3 µg REeq/g for red palm oil-fortified gari and yellow cassava gari. Aruna et al.14 obtained a substantially higher vitamin A content for gari fortified with moringa seed powder.

Based on the results discussed above it can be concluded that the used conversion factors largely affect the final results in terms of vitamin A content. To make a statement about the effectiveness of the three investigated fortification strategies, human intervention studies are therefore required. Nevertheless, it can be stated that the red palm oil-fortified gari provided the best vitamin A stability. In addition, local production of red palm oil in Ghana and other West African countries makes this fortification strategy a feasible approach for small-scale producers. However, apart from the vitamin A content and stability, quality and sensory aspects are crucial determinants for consumer acceptance. Therefore, the main quality characteristics were analysed and a sensory analysis was performed.

In Table 2, an overview of the quality characteristics measured for regular gari, red palm oil-fortified gari and yellow cassava gari is given. For all three gari samples, the moisture content, the ash content and TTA fell within the range specified in the Codex Alimentarius guidelines30. In addition, the pH and swelling capacity for the fortified gari samples did not differ significantly from the regular gari (p < 0.05). All gari samples had a swelling capacity factor of around three, which is desirable. The particle size for yellow cassava gari was significantly smaller than for regular, unfortified gari and red palm oil-fortified gari (p < 0.05). Most importantly, the colour of the fortified gari samples was significantly more yellow than the regular gari, as measured by the b-value. This was, however, not surprising given the intense orange colour of β-carotene. The L-value for the fortified samples was slightly lower compared to the regular gari. This indicates a less bright colour for the fortified gari, although the difference was not significant (p < 0.05).

Table 2 Overview of the quality characteristics of regular gari, gari fortified with red palm oil and biofortified yellow cassava gari

Next to the analysis of the quality characteristics, a sensory analysis was performed. The average scores attributed during this sensory analysis are shown in Table 3 and Fig. 3. As shown in Table 3, the taste and smell of both fortified gari samples were rated significantly lower than the regular gari. In addition, the colour of the yellow cassava gari was given a significantly lower score than the regular gari. The colour of red palm oil-fortified gari was also scored lower than the regular gari, yet not significantly lower (p < 0.05). This is likely due to the more yellow colour of the fortified samples. For the mouthfeel and sourness, no significantly different scores were given for the three gari types. When looking at the overall appearance of the fortified gari samples (Fig. 3), both samples were scored lower than the regular gari for both preparation methods. Although when prepared as if the gari would be used for gari with beans or shito, the attributed scores were not significantly different from the regular gari. For the preparation of gari soakings, on the other hand, the fortified gari samples were rated significantly lower. Consequently, it is advisable to use the fortified gari for the preparation of gari with beans or shito rather than for soakings.

Table 3 Sensory evaluation of regular gari, gari with red palm oil and yellow cassava gari
Fig. 3 : Overall appearance of regular gari, gari with red palm oil and yellow cassava gari.
figure 3

Overall appearance of three gari samples prepared with the addition of a small amount of water as if it would be for gari with beans/shito (A), and with the addition of a large amount of water as if it would be for soakings (B). The three gari samples included regular gari (control), gari with red palm oil and gari made from yellow cassava. An untrained panel of 77 participants scored the samples based on a hedonic scale, whereby 1 is the lowest and 9 is the highest score. Samples indicated with a different letter are significantly different (p < 0.05).

The results from the sensory analysis corresponded well with the main conclusion from the focus group discussion. During this discussion, the overall appearance of the regular gari was rated the highest and attributed an average score of 8.3. The fortified samples were rated slightly lower. The red palm oil-fortified gari and yellow cassava gari were given an average score of 7.8 and 7.3, respectively. Despite the slightly lower scores, the participants indicated that they would be more willing to buy the fortified gari because of its associated health benefits. When comparing the obtained results regarding the gari quality and sensory aspects with existing literature, similar observations are described in the literature13,42. As stated by Adinsi et al.13, the main gari characteristics affected by fortification include colour and odour, whereas other characteristics, such as swelling capacity, texture and sourness, are not affected. Despite the negative impact of vitamin A fortification on the quality and sensory attributes, consumers with previous knowledge of fortification tend to associate the yellow colour of fortified gari with health benefits such as a positive influence on eyesight42. Consequently, as stated by Bechoff et al.42, campaigns to inform consumers about the health benefits of gari fortification are crucial and might overcome the limited negative effect on sensory attributes.

In conclusion, the goal of this study was to explore different strategies to fortify gari, a cassava-based West African food product, with vitamin A whereby obtaining a high vitamin A stability during storage was the main focus. In the first part of this study, the potential of cereal bran to stabilise RP during gari storage was investigated. Unfortunately, cereal bran addition only resulted in a limited improvement of the RP stability. This effect was significant but small for toasted wheat bran. For native wheat bran and heat-treated rice bran, RP stabilisation was not observed. Therefore, it is advised to use cereal bran to stabilise RP in a highly concentrated food additive consisting of RP, oil and cereal bran rather than during storage of the fortified food product itself. In the second part, three different strategies for the fortification of gari were tested. These strategies included the addition of RP, the addition of red palm oil and the use of yellow cassava. Amongst these three approaches, the incorporation of red palm oil is the most promising fortification strategy. This approach resulted in the highest vitamin A stability during accelerated storage. Moreover, the effect of red palm oil addition on the gari quality and sensory aspects was limited. In addition, as red palm oil is locally produced in many African countries, this fortification strategy is easily applicable for small-scale local producers. However, more research is needed to accurately estimate the vitamin A degradation in red palm oil-fortified gari during storage at ambient conditions.