True metabolizable energy (TME)

Oil true metabolizable energy (TME) was analyzed according to Sibbald¹¹ (Table 1). Experimental animals were twenty-four Bovans White line roosters of 33 weeks of age with an average individual weight per bird of 2.06 ± 0.06 kg. Animals were randomly distributed in three treatments, eight per treatment, with each rooster representing a replicate. Administration of pure oil causes poultry to regurgitate¹², and its liquid state prevents quantification of dry matter (DM)¹³. Due to these physical characteristics, the oils were mixed with ground sorghum at a 90:10 proportion. Sorghum DM was therefore quantified simultaneously with the treatments using six roosters.

Metabolic and endogenous energy were measured. The roosters were allowed to rest for five days and then fasted for 24 hours. Total manure (endogenous and metabolic material) was collected from each animal to ensure that the endogenous portion used in the calculations came from the same animal¹⁴. Ingredient and excreta gross energy (GE) were measured in two replicates using a isoperibolic calorimetric pump (Parr 1266, model Moline, Illinois, USA).

Production variables and egg quality

A total of 240 Bovans White hens, 30 wk old, were used in this assay. Animals were distributed into six treatments, five replicates per treatment, and eight animals per replicate. Hens were placed two per cage (30 x 45 cm), with linear feeders and automatic drinking troughs in a conventional hut. Photoperiod was 16 h daylight^-1, provided by artificial lighting. The experimental period was 16 wk.

Diets were isoenergetic and based on a sorghum-soybean paste (Table 2). They met the laying hen nutritional requirements of the NRC¹⁵ and Cuca et al¹⁶. The diets were kept isoenergetic by varying proportions of sorghum, soybean paste and sand (sterilized in autoclave). Crude soy oil (CSO), soybean soapstock T (SST) and soybean soapstock Y (SSY) were evaluated at two inclusion levels (2 and 4 %), resulting in six treatments: 2%CSO; 4%CSO; 2%SST; 4%SST; 2%SSY; and 4%SSY. During the growth period hens had been vaccinated against newcastle, smallpox, gumboro, bronchitis, encephalomyelitis and infectious coryza. Water and food were freely available.

Data were collected weekly on five production variables: food intake (FI, g/bird/d); laying percentage (LP, %); egg weight (EW, g/d); feed conversion (FC); and egg mass (EM, g). Egg quality was measured using twenty eggs (four per replicate) from each treatment at the beginning of the period and at wk 4, 8 and 12. Four parameters were used to characterize egg quality: albumin height (AH); Haugh units (HU); yolk color (YC) using an Egg Multi Tester (QCM System, Technical Services and Supplies, Dunnington, United Kingdom) which measures yolk color based on the DSM range; and eggshell thickness (ET), taken with a micrometric screw.

Fatty acids analysis

Oil fatty acid profile (Table 3) was analyzed using the AOAC total lipids technique¹⁸. Egg fatty acids composition was measured using the same eggs used to measure egg quality. These were manually mixed with a blender to create a pooled sample. Lipid extraction was done using the AOAC total lipids technique¹⁹ (923.07), with a gas chromatographer (model 3380 CX, Varian) equipped with a DB23 column (30 m x 0.25 mm id), a CP8400 Autosampler and a flame ionization detector (FID)(USA).

Cost per kilogram of eggs

The cost of each diet was calculated by multiplying the price of each ingredient by the quantity of each in each feed formula. The cost of one kilo of eggs per feed was calculated based on the FI of each treatment and multiplied by the feed cost. Ingredient prices (/kilo) were sorghum, $3.58; soy paste, $7.96; CSO, $16.00; SST, $12.00; SSY, $10.00; DL-methionine, $70.00; threonine, $30.00; CaCO₃, $1.50; dicalcium phosphate, $16.00; vitamins, $75.00; minerals, $20.00; salt, $3.50; and pigment, $30.00.

Statistical analyses

Data were analyzed with a completely random design employing a 3x2 factorial arrangement in five replicates: oils (CSO, SST and SSY), and inclusion levels (2 and 4%). Using the SAS statistics package²⁰, the MIXED procedure was applied and differences between the treatment means compared with a Tukey test (P<0.05).

Egg quality

Inclusion of both SST and SSY increased HU values (P<0.05), but no differences were observed between different inclusion levels (Table 5). This contrasts with a study in which substitution of CSO (2.6%) with sunflower soapstock (25, 50, 75 and 100%) tended to lower HU values as inclusion level increased²¹. However, another study reported that use of soybean soapstock in hen diets had no effect on HU values²⁶. Neither oil type (CSO, SST, SST) nor level (2 and 4%) affected AH or ST (P>0.05); this agrees with previous studies^21,26.

Yolk color (YC) was modified by oil type (P<0.05) but not by oil inclusion level. Addition of SST improved yolk color, whereas no changes were observed with the CSO and SSY treatments (Table 5). How an added oil affects YC depends on the xanthophyll content of the seeds from which it was extracted, and the process used to produce the soapstock since bleaching of soybean soapstocks can eliminate xanthophylls⁶. The present results coincide with a study in which YC improved in response to replacement of CSO with sunflower soapstock, a phenomenon attributed to oil tocopherol content²¹. Soy soapstock is also reported to be an important natural pigment in broilers³¹. However, another study found CSO and sunflower soapstock to have no effect on skin pigmentation in chickens³².

Egg fatty acid composition

Fatty acid composition was affected by oil type (P<0.05). Inclusion of SSY increased concentrations of C14:0 and C16:0 (P<0.05) in the egg (Table 6). In contrast, addition of SST lowered C14:0 by 14% and C16:0 by 2%, and CSO lowered C14:0 by 25% and C16:0 by 3%. This is to be expected because these fatty acids were deposited in the egg according to their levels in each oil (Table 3). Neither soybean soapstock modified egg C18:0 levels. Oil diet inclusion levels had no effect on C14:0 or C18:0 levels, but C16:0 (P<0.05) did increase at the 4% level. These results contrast with a previous report in which egg SFA (C14:0, C16:0 and C18:0) composition did not vary between treatments containing different levels of soybean soapstocks²⁶. It is yet unclear why some fatty acids are more readily deposited in the egg. Some fatty acids are better metabolized than others, and high SFA content decreases when oils with lower SFA content are added to diets³³.

Inclusion of SSY increased concentrations of the monounsaturated fatty acids (MUFA) C16:1 and C18:1 (P<0.05). Addition of SST decreased C16:1 by 18% and C18:1 by 6%, while CSO reduced C16:1 by 22% and C18:1 by 12%. Concentrations of C16:1 responded to oil inclusion level since levels were higher at the 2% level (P<0.05); C18:1 concentration was unaffected by inclusion level. These results differ somewhat from those of a study in which no changes were observed in C16:1 and C18:1 concentrations in eggs when CSO was substituted by soybean soapstock at 25, 50, 75 and 100%²⁶.

Content of the polyunsaturated fatty acid (PUFA) C18:3 ω3 was higher (P<0.05) with addition of CSO in the diet and decreased with inclusion of SST (23%) and SSY (61%). This is to be expected since yolk PUFA composition, and especially C18:3 ω3, is influenced by feed oil profile^34,35,36. Levels of C18:3 ω3 increased at the 4% oil inclusion level (P<0.05). This is consistent with a reported increase in C18:3 ω3 when diet oil content was raised from 1.5 to 3%²³.

Eicosapentaenoic acid (EPA) levels did not change (P>0.05) in response to addition of different oils or inclusion level. However, docosahexaenoic acid (DHA) and docosapentaenoic (DPA) acid levels tended to increase in the egg (P<0.05) when CSO and SST were added to the diet, whereas they decreased with addition of SSY. This was probably due to the high C18:3 ω3 content in the CSO and SST (Table 3), which desaturase and elongase enzymes transform into EPA and subsequently DHA and DPA^37,38. Soapstock inclusion level had no effect on DHA levels (P>0.05), but DPA levels did increase at the 4% level (P<0.05).

Levels of the PUFA C18:2 ω6 in the CSO treatment were 25 % higher than with SST and 40 % higher than with SSY (P<0.05); this was probably due to the respective contents of this acid in each oil. The content of C18:2 ω6 was not affected by oil inclusion level (P>0.05). Addition of CSO and SST reduced (P<0.05) C20:4 ω6 content in the egg, but SSY increased it. This may be because the SSY contained 0.23% C18:3 ω3 while the CSO had 7.52 % and the SST 6.59 % (Table 3). High C18:3 ω3 concentrations are known to limit synthesis of C20:4 ω6 since both acids use the Δ-desaturase enzyme³⁹ due to competition between n-3 and n-6 for the same enzymes for biosynthesis^34,40.

Total egg SFA and MUFA contents were unaffected by oil type and inclusion level (P>0.05). This was not true for the PUFA n-3, which decreased 16 % with SST and 47 % with SSY, and n-6, which decreased 27 % with SST and 38 % with SSY (P<0.05). Higher oil inclusion level increased (P<0.05) both n-3 and n-6 contents (Table 6).

Both n-6 and n-3 fatty acids are important in human nutrition, and maintaining a 4:1 n-6/n-3 ratio is vital to overall human health^41,42. During gestation n-3 fatty acids function as structural components in the brain and retina, and contribute to normal growth and development in the infant⁴³. High levels of n-6 promote cardiovascular diseases, and an adequate n-6/n-3 balance can diminish and prevent obesity⁴⁴. Addition of oils rich in n-3 (e.g. flax seed) to hen diets can raise n-3 levels in the egg and help to improve the n-6/n-3 ratio³³. Compared to eggs from the CSO treatment, those from the soybean soapstock treatments had a lower n-6/n-3 (P<0.05); these eggs had a lower n-3 content as well as a lower n-6 content. Diet oil inclusion level did not influence the n-6/n-3 ratio, which agrees with previous findings of no effect on this ratio in response to addition of CSO (11.90) and soybean soapstock (13.75)²⁶.

Cost per kilogram of eggs

Compared to the cost per one kilogram of eggs in the CSO treatment, the cost in the SST dropped 2.68% and that in the SSY by 2.03% (P<0.05) (Table 7). At the 4% inclusion level the cost per one kilogram increased by 1.8 % over the 2% level (P<0.05).