Introduction
‘Feeding value’ is a complex function of nutrient content, physical and chemical characteristics that affect digestibility, acceptability (palatability) and associative interactions with the digestive process. Consequently, processing methods that might enhance digestion might not be reflected by improved dietary net energy (NE) or growth performance. Thus, processing methods are selected that will most economically enhance digestibility and acceptability without detrimentally affecting ruminal pH or causing digestive dysfunction. Popular approaches include: (1) particle size reduction that yields dry rolled or dry ground grain with or without the addition of moisture (tempering); (2) ensiling with inherent moisture before the grain has dried in the field to yield high-moisture maize or by adding water to reconstitute maize before ensiling and/or feeding; and (3) steaming flaking. Processing criteria for specific methods vary considerably. These, in combination with differences in feeding management, nutritional balance (nutrient adequacy of dietary formulation) and dietary forage level, make summary comparisons of processing techniques difficult. The objective of the present paper is to provide a practical overview of differences among techniques in terms of digestive function and growth performance of feedlot cattle.
Components of maize grain that limit digestion
Seed coat
The seed coat or pericarp of cereal grain serves to protect the seed from moisture, insects and fungal infection that would hamper germination. Although the pericarp comprises only about 4·7 % of the weight of the maize kernel, it contains nearly half of the neutral-detergent fibre (NDF) of the kernel (pericarp averages 10·0 % NDF). For digestion of the starchy endosperm, the seed coat must be cracked to permit moisture, microbes and enzymes to enter. Even after being dry rolled, the pericarp can remain attached to vitreous starch, and thereby might limit access of the endosperm to the processes of fermentation or digestion. Any factor that introduces stress cracks in the pericarp (for example, high-temperature drying of grain, premature harvest) may increase starch exposure and the rate and extent of digestion. Steam flaking and ensiling will reduce the physical association of the pericarp with the endosperm.
The nutrient composition of various fractions of the maize kernel is shown in Fig. 1(Reference Watson, Watson and Ramstad1). Note that most of the oil and ash is located in the germ whereas the pericarp contains most of the NDF. The surface area of a sphere per unit of volume decreases by half as the diameter doubles. Thus, a larger kernel will have less pericarp as a fraction of total weight; thereby it will have less NDF than a small kernel. This relationship may explain why NE value is greatest for the maize hybrid with the largest mean kernel weight(Reference Jaeger, Luebbe and Macken2). The question is: what determines kernel weight? Grown in the same field in 2002, ten commercial Pioneer maize hybrids had mean kernel weights that ranged from 320 to 401 mg; maximum kernel size is characteristic of a hybrid but size can be reduced by water stress or other environmental stresses. Kernel weight also increases with maturity; for certain hybrids, kernel weight may increase even after kernel moisture falls below 30 % moisture, a point often used to initiate high-moisture maize harvest. Selection of maize with low NDF content or large kernel size should increase the NE value of the grain.
Fig. 1 Nutrient content of various parts of the maize kernel (from Watson(Reference Watson, Watson and Ramstad1)): starch (
); protein (
); neutral-detergent fibre (
); fat (
); ash (□); sugar (
).
Germ size
Although a decrease in the relative size of the germ will decrease the ash and NDF content of grain, the germ carries most of the oil and a large proportion of the essential amino acids in maize grain. Consequently, reducing the size of the germ may not enhance the NE value of the grain because the oil content is reduced; it also will decrease the concentration of most essential amino acids. The relative size of the germ is greater for hybrids selected for high oil content. Hybrids marketed as being ‘nutrient dense’ typically contain more oil than other hybrids. Because ruminal microbes do not ferment oil as a source of energy, microbial protein yield is reduced as oil replaces starch in grain. The value of oil in maize grain depends on its cost relative to other sources of supplemental fat and oil.
Vitreous endosperm
From 25 to 80 % of the starch is present in the horny (hard or vitreous) endosperm where starch granules are densely packed within a ruminally insoluble and very slowly degradable protein matrix. Yellow maize typically contains approximately 50 % vitreous endosperm. However, white maize may contain 80 % vitreous endosperm. The remaining starch is deposited in the soft or floury endosperm as loosely packed granules resembling basketballs held in a large mesh bag by endosperm cell walls. Vitreousness can be estimated through physical dissection of kernels, by measuring absolute density (not test weight) of the grain, or by grinding with a Stenvert mill. The horny:floury (H:F) ratio varies genetically, being greater for maize grain classified as flint (v. dent). The H:F ratio often increases with grain maturation and N fertility (that also will increase the crude protein content of maize grain). The rate and extent of digestion of isolated starch granules from the floury endosperm of maize grain are rapid and complete. Thus, primary limitations to the rate and extent of digestion of maize hybrids (in situ) are associated with the H:F ratio, or degree of vitreousness(Reference Corona, Owens and Zinn3–Reference Shaver and Majee6).
Floury hybrids generate more fines during grinding. The advantage of having more particles that are very fine is debatable. Increased digestibility should be beneficial, and fine particles may be flushed rapidly through the rumen with fluids, increasing postruminal starch supply. However, fine particles of the floury endosperm are very rapidly fermented, increasing the risk of acidosis. Greater in situ disappearance of floury hybrids has led to the suggestion that the extent of digestion will be greater for hybrids containing more floury and less vitreous endosperm when the grain is rolled (not extensively processed). This concept was verified by Jaeger et al. (Reference Jaeger, Luebbe and Macken2) who observed that maize hybrids with greater proportions of floury starch produced the best gain efficiency (r 0·83) when dry rolled.
The optimum H:F ratio may differ with the grain-processing technique employed. Differences in starch digestion among maize hybrids due to vitreousness is apparent with dry rolling, but not with steam flaking(Reference Corona, Owens and Zinn3). Indeed, floury hybrids may complicate the steam-flaking process, yielding more fragile flakes, producing greater fines.
Ensiling influences vitreousness. When subjected to fermentation with maize silage, the vitreousness of starch in maize kernels is reduced(Reference Johnson, Harrison and Davidson7). When high-moisture maize is prepared from a vitreous and a floury maize hybrid, ruminal and total tract starch digestibility of starch tends to be superior for the vitreous hybrid, as observed in a site of digestion study with steers(Reference Szasz, Hunt and Szasz8). Thus, extrapolation of hybrid differences from measurements on dry rolled grain to fermented grain or maize silage may prove erroneous.
Amylose content
Chemically, starch within the starch granules of maize grain is present either as amylopectin in a multi-branched structure or as amylose that is a linear structure and is less rapidly digested by enzymes. Amylose can comprise from below 2 % (waxy maize) to a high of about 70 % (high amylose) of total starch due to genetic differences in activity of the amylose extending gene and enzyme. Environmental factors such as day length also may be involved, as a circadian rhythm in amylose synthesis activity has been detected in some plants. Typical dent maize hybrids have from 24 to 30 % of starch present as amylose. The amylose:amylopectin ratio increases with maize kernel maturity but can be decreased by high environmental temperatures. Maize grain with high amylose content was found to be poorly digested in the small intestine of dogs even after the grain was extruded(Reference Gajda, Flickinger and Grieshop9). The susceptibility of flaked high-amylose maize samples to ruminal digestion also is low. Fermentation of amylose may be restricted to a limited number of bacterial strains(Reference Wang, Conway and Brown10). Starch granules have been proposed to contain consecutive rings or spheres of amylose and amylopectin, and if amylose degradation is limited, granules may resist digestion. Linkage of reducing ends of starch within the starch granule with lipid or P also may reduce the rate of digestion. Small-intestinal digestion of starch from potatoes by lactating cows was found to be nil; in contrast, the small-intestinal digestion of starch was reported to be 70 and 65 % for starch from wheat and maize in a study by van Vuuren et al. (Reference van Vuuren, Hindle and Cone11). Steam flaking removes the feed efficiency advantage of waxy hybrids (low amylose) over typical hybrids (normal amylose) noted in steer performance trials.
Resistant starch
With exposure to heat and moisture, starch granules swell and form gels, a process called gelatinisation. Swollen particles become enriched in amylopectin as amylose diffuses out of the swollen granules. Starch granules with high amylose content resist swelling. Upon storage of the gelatinised starch, amylose gels and forms retrograde starch. Ward & Galyean(Reference Ward and Galyean12) measured in vitro DM disappearance of flaked maize samples that had been held warm in order to decrease starch availability by one-third, presumably due to starch retrogradation. The rate and extent of in vitro digestion were found to be equal for flaked maize despite a reduction in amyloglucosidase reactive starch (increased starch retrogradation). This observation suggests that ruminal microbes from steers fed flaked grain must have sufficient capacity to ferment retrograde starch or at least solubilise starch that might be resistant to attack by starch-degrading enzymes.
Rowe et al. (Reference Rowe, Choct and Pethick13) summarised specific responses to various grain-processing techniques on seed components that may limit the site and extent of digestion (Table 1). As will become clear further on in the present paper, processing maize to increase endosperm exposure (usually by particle size reduction) will have very little net impact on feeding value. Enhancement of feeding value will be most predictable with maize-processing techniques that disrupt the endosperm matrix.
Table 1 Impact of various processing techniques on maize and its digestion
+++, Major impact; ++, medium impact; +, minor impact; − , no impact.
Effects of processing on site and extent of starch digestion by cattle
Results from published trials between 1987 and 2007 that tabulated the site and/or extent of starch digestion for feedlot cattle were assembled (Table 2)(Reference Owens and Zinn14). Data from trials where mixtures of grains or where grains processed by multiple methods were fed simultaneously were excluded from analysis. Site of digestion measurements with 135 different diets were identified; each measurement was the average of several cattle (mean of 4·5) fed each diet. To evaluate the site and extent of starch digestion, factors of primary concern were: (1) percentage of dietary starch apparently digested in the rumen; (2) percentage of starch flowing out of the rumen digested in the intestines; (3) total tract starch digestion; and (4) site of starch digestion (fraction of total tract starch digested that disappeared in the rumen).
Table 2 Effect of processing method on site of starch digestion from maize grain (derived in part from Owens & Zinn(Reference Owens and Zinn14))
Total tract starch digestion was similar for dry processed maize. However, in studies directly comparing dry rolled v. whole shelled maize(Reference Murphy, Fluharty and Loerch15–Reference Gorocica-Buenfil and Loerch17), dry rolling increased (2 to 12 %) total tract starch digestion. The postruminal digestion of intact maize kernels is nil. Therefore, intact kernels that escape the rumen are passed as such in the faeces(Reference Corona, Rodriguez and Ware16). Expected enhancements in starch digestion due to dry rolling maize may be greater in light-weight calves than in yearlings(Reference Gorocica-Buenfil and Loerch17), and when comparisons are made at lower levels of intake ( < 1·5 % of body weight; Murphy et al. (Reference Murphy, Fluharty and Loerch15)).
In our experience there continues to be considerable controversy regarding the benefit of finely grinding v. dry rolling maize. This concept, although appealing, is not supported by digestion trials, due perhaps, to extensive particle size reduction during the normal course of feeding. Taking into consideration that cattle spend approximately 2 h/d eating, with seventy-four jaw movements per min during eating bouts (including the 10–15 s laps of no jaw movement between boluses; Kononoff et al. (Reference Kononoff, Heinrichs and Lehman18)), a steer consuming 10 kg of DM will masticate the feed at a rate of 5328 chews/kg. This of course does not include any additional chewing that might occur during rumination.
Average total tract starch digestion was found to be similar for high-moisture and steam-flaked maize. However, the proportion of starch digested ruminally was greater (8 %) for high-moisture maize (Table 2).
Effects of processing on growth performance and net energy value of maize
Results from published trials between 1987 and 2007 that directly compared the various processing methods with dry processed (coarsely rolled or cracked) maize were evaluated. In all cases, a NE maintain (NEm) value of 9·12 MJ (2·18 Mcal)/kg was assigned to the control, dry process treatment to which all of the process treatments below were compared using the replacement technique. Likewise, changes in average daily gain (ADG) and DM intake (DMI) are with reference to the control (dry process treatment). The comparative effects of whole dry shelled maize v. dry rolled maize are summarised in Table 3(Reference Corona, Rodriguez and Ware16, Reference Gorocica-Buenfil and Loerch17, 19–Reference Searle, Casella and McCulloch22). Feeding whole shelled maize tends to decrease ADG by 2·5 % and increase DMI by 3·2 %. The NEm and NE gain (NEg) values for whole shelled maize are 8·83 and 6·02 MJ (2·11 and 1·44 Mcal)/kg, respectively. The only slight reduction in NE (3·2 %) due to not processing maize is consistent with expected values based on starch digestion studies (Table 2). The National Research Council (NRC)(19) does not differentiate the NE value of maize for whole v. dry processes. Likewise, the comparative effects of finely rolling or grinding dry maize v. coarse processing (Table 4)(Reference Corona, Rodriguez and Ware16, 19, Reference Scott, Milton and Erickson20, Reference Loe, Bauer and Lardy23) on growth performance and maize NE are minimal ( < 1 % change).
Table 3 Comparative effects of whole dry maize v. conventional dry processing (i.e. coarsely rolled, cracked) on average daily weight gain (ADG), DM intake (DMI) and net energy (NE) value of maize (estimated using the replacement technique and given that NE maintain (NEm) of dry rolled maize=9·12 MJ (2·18 Mcal)/kg; National Research Council(19))*
* In this Table, and Tables 4 to 6, the comparative effects of different maize processing v. conventional dry processing (i.e. coarsely rolled, cracked) on ADG and DMI were evaluated using a meta-analysis approach that combines means from individual studies, while taking into account differences in sample sizes. The mean difference or ‘effect size’ method described by Whitehead & Whitehead(Reference Whitehead and Whitehead21) was used. The effect size was calculated as the difference between the means of two studies for ADG and DMI variables. The standardized effect size was estimated by dividing this difference by its pooled sd. Reported se, and the number of animals in the treatment groups, were used to calculate the pooled sd of each mean difference (Searle et al. (Reference Searle, Casella and McCulloch22)). A fixed-effects statistical model was used and this model included the following factors: study, process, and sex groups. Comparisons between studies were done within sex group. General linear model procedures from SAS were used in the analysis.
Table 4 Comparative effects of finely rolling or grinding dry maize v. conventional dry processing (i.e. coarsely rolled, cracked) on average daily weight gain (ADG), DM intake (DMI) and net energy (NE) value of maize (estimated using the replacement technique and given that NE maintain (NEm) of dry rolled maize=9·12 MJ (2·18 Mcal)/kg; National Research Council(19))
The growth performance response of feedlot cattle to feeding of high-moisture maize is highly variable. Previous studies indicate that high-moisture maize increases diet acceptability over that of dry processed maize, increasing ADG and DMI by 2·9 and 3·2 %, respectively(Reference Mader, Dahlquist and Britton24–Reference Utley and McCormick26). However, the NE value of whole high-moisture maize is similar to or slightly less than that of dry processed maize, averaging 9·04 and 6·44 MJ (2·16 and 1·54 Mcal)/kg, respectively(Reference Mader, Dahlquist and Britton24–Reference Utley and McCormick26). Grinding or rolling high-moisture maize before ensiling increases its NE value; however, these processes also increased the likelihood for depressed ADG compared with ensiling the maize whole. The comparative effects of processed high-moisture v. conventional dry processed maize on the ADG, DMI and NE value of maize are shown in Table 5(19, Reference Scott, Milton and Erickson20, Reference Huck, Kreikemeier and Kul27–Reference Stock, Sindt and Cleale30). Substituting processed high-moisture maize for dry processed maize does not affect ADG, but reduces (P < 0·05) DMI by 6·6 %. Consequently, the NEm and NEg values for high-moisture maize are greater than that of dry processed maize, averaging 9·67 and 6·74 MJ (2·31 and 1·61 Mcal)/kg, respectively. These values are consistent with those suggested for high-density high-moisture maize (25·4 kg (56 lb)/bushel) by the NRC(19). The improvements in NE value of ensiled high-moisture maize are also consistent with the expected enhancement in starch digestion (Table 1).
Table 5 Comparative effects of high-moisture v. conventional dry processing (i.e. coarsely rolled, cracked) on average daily weight gain (ADG), DMI and net energy (NE) value of maize (estimated using the replacement technique and given that NE maintain (NEm) of dry rolled maize=9·12 MJ (2·18 Mcal)/kg; National Research Council(19))
Consistent with previous reviews, substituting steam-flaked maize for dry processed maize increases (P < 0·05) ADG (6·3 %) and decreases DMI (5 %). The comparative NEm and NEg values for steam-flaked maize are 10·29 and 7·32 MJ (2·46 and 1·75 Mcal)/kg, respectively (Table 6) (Reference Corona, Rodriguez and Ware16, 19, Reference Scott, Milton and Erickson20, Reference Huck, Kreikemeier and Kul27, Reference Zinn31–Reference Brown, Krehbiel and Duff34). The NE values are greater (5·5 %) than current tabular values(19), being more consistent with earlier standards(35).
Table 6 Comparative effects of steam flaking v. conventional dry processing (i.e. coarsely rolled, cracked) on average daily weight gain (ADG), DM intake (DMI) and net energy (NE) value of maize (estimated using the replacement technique and given that NE maintenance (NEm) of dry rolled maize=9·12 MJ (2·18 Mcal)/kg; National Research Council(19))
Optimising processing conditions
Due to maize's high starch content, assessment of changes in starch enzymic reactivity, and/or digestibility have been used to optimise processing techniques. However, when grain is the primary or sole source of starch in the diet, concentration of starch in the faeces (faecal starch (FS), % of DM) of feedlot steers can serve as an indicator of total tract starch digestion, and, hence, the feeding value of maize. Zinn et al. (Reference Zinn, Barreras and Corona36), using a regression analysis, evaluated data from thirty-two metabolism trials involving 637 individual starch digestibility measurements. Percentage FS explained 96 % of the variation in total tract starch digestion (TSD (%) = 99·9 − 0·413 FS − 0·0131 FS2 (Zinn et al. (Reference Zinn, Barreras and Corona36); Fig. 2). Zinn et al. (Reference Zinn, Owens and Ware37) observed that starch digestion was closely associated (r 2 0·88) with the NE value of maize (Fig. 3): maize NEm (MJ/kg) = − 3·14+0·134 TSD (maize NEm (Mcal) = − 0·75+0·032 TSD). Thus, where maize is the principle source of dietary starch: maize NEm (MJ/kg) = 10·21 − 0·0552 FS − 0·00 175 FS2 and NEg (MJ/kg) = 3·67 NEm − 1·72 (maize NEm (Mcal) = 2·44 − 0·0132 FS − 0·000418 FS2), and NEg (Mcal) = 0·877 NEm − 0·41). This relationship also applies to sorghum and wheat.
Fig. 2 Relationship between faecal starch (FS) percentage and total tract starch digestion (starch digestion (%) = 99·9 (se 0·0540) − 0·413 (se 0·0306) FS − 0·0131 (se 0·00 192) FS2; R 2 0·96) (from Zinn et al. (Reference Zinn, Barreras and Corona36)).
Fig. 3 Relationship between change in total tract starch digestion (CSD, %) and change in grain (maize (♦), wheat (▲), sorghum (■)) net energy maintain (NEm): change in grain NEm (MJ/kg) = 0·1372 CSD (%) (change in grain NEm (Mcal/kg) = 0·0328 CSD (%)) (r 2 0·97; standard error of the estimate (Sy.x) = 0·022; P < 0·001) (from Zinn et al. (Reference Zinn, Owens and Ware37)).
Example of assessment of maize processing method
For practical assessments of the adequacy of maize processing under feedlot conditions, consider the following example:
The problem
A feedlot is steam flaking maize but is not certain what flake density (as measured directly beneath the rolls) is optimal under their production system.
The evaluation
Set the ram pressure to produce flaked maize at a density of 0·36 kg/l (12·7 kg (28 lbs)/bushel). After feeding the flaked maize for 5 d, collect twelve faecal grab samples (about 100 g/sample) from three pens (four samples/pen) receiving the finishing diet. Faecal samples may be collected in the early daylight hours when cattle are just getting up. Dry the faecal material as soon as possible following collection, and determine percentage FS in the dried sample.
The result
FS analysis for the composite sample averaged 4·5 %. Thus, starch digestion = 97·8 % and maize NEm = 9·92 MJ (2·37 Mcal).
The solution
Increase ram pressure to produce a flake density of 0·34 kg/l, feed maize for 5 d and repeat FS evaluation. A practical target for FS might be < 3·0 %. This will result in an expected NEm value for flaked maize of 10·04 MJ (2·40 Mcal)/kg.
Conclusions
The NE values of whole and dry rolled maize are consistent with current standards (NRC(19)). The NE value of whole high-moisture maize is only slightly (2·3 %) greater than that of dry processed maize, and 5·5 % less than that suggested by the NRC(19). Softening the kernel before rolling by tempering or steaming enhances the growth performance response and the NE value of maize as a result of increased total tract starch digestion. The feeding value of maize may be affected by the intensity of processing only in steam-flaked processing. The comparative NEm and NEg values for steam-flaked maize at optimal processing (density = 0·34 kg/l) are 10·04 and 7·07 MJ (2·40 and 1·69 Mcal)/kg, respectively. These NE values are greater (3 %) than current tabular values(19), being more consistent with earlier standards(35). When maize is the primary or sole source of starch in the diet, the concentration of starch in faeces (FS, % of DM) of feedlot steers can serve as an indicator of total tract starch digestion, and, hence, the feeding value of maize.
Acknowledgements
The present review received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.
The authors contribution to the manuscript were as follows: R. Z. and F. O., design and draft of the work; A. B. statistical analysis; L. C., literature review and A. P., statistical analysis and style revision.
There are no conflicts of interest in the subject of the present review.
