THE GLYCEMIC AND INSULINEMIC INDEX IN
HORSES
INGRID VERVUERT and MANFRED COENEN
Introduction
In horses, the total tract apparent digestibility of starch for various types of grain
is usually very high. Arnold et al. (1981) reported values of 97.0, 96.7, and 97.0%
respectively for corn, oats, and sorghum starch. On the other hand, considerable
differences in prececal starch digestibility were observed between the different
starch sources. In general, the prececal digestibility of oat starch exceeds that of
corn starch or of barley starch (Kienzle et al., 1992; Potter et al., 1992; Meyer et
al., 1995). Prececal starch digestibility of grains is improved by different processing
techniques. The granular structure of starch might be destroyed mechanically
(e.g, rolling, crushing, or grinding) or by heat and pressure in combination with
moisture (Kienzle et al., 1992; Potter et al., 1992; Meyer et al., 1995). The effect
of thermal processing such as micronizing, steam-flaking, or popping is an
irreversible swelling and destruction of the internal crystalline structures of the
starch granules; this transformation is termed gelatinization (Holm et al., 1988;
Selmi et al., 2000). Thus, the extent of prececal digestion influences the proportion
of cereal carbohydrates absorbed as glucose in the small intestine and that are
fermented and absorbed as volatile fatty acids or lactic acids in the large intestine.
Consequently, an increase in availability of starch for enzymatic digestion in
the small intestine might alter the metabolic response as more substrate will be
absorbed. In humans, the measurement of blood glucose and insulin response is
known as a suitable tool for assessing the effects of food processing on starch
digestion (Jenkins et al., 1981; Brand et al., 1985; Granfeldt et al., 1994). In
human subjects, starchy foods have been classified over the entire range from
restrained, or low glycemic and insulin response, to rapid with respect to effects
on blood glucose and insulin responses after a meal (Jenkins et al., 1988; Granfeldt
et al., 1994). The resulting glycemic or insulinemic index utilized white bread as
the standard source and all foods were ranked accordingly (Jenkins et al., 1981).
Influence of Grain Source and Processing Techniques on the
Glycemic and Insulinemic Responses in Horses: Recent Work
Information is available regarding the influence of various grain sources on
glucose and insulin response in horses, but no such glycemic or insulinemic
index has yet been formulated for horses.
THE GLYCEMIC AND INSULINEMIC INDEX IN HORSES 56
Stull and Rodiek (1988) noticed a significant increase in plasma glucose
concentration after corn feeding as well as after a combined diet of corn and
alfalfa (50% corn and 50% alfalfa) in two-year-old Quarter Horse geldings. Insulin
concentrations closely followed the glucose curves. However, postprandial
response area for glucose did not differ between alfalfa feeding (100%), corn
feeding (100%), and combined corn and alfalfa feeding (50% corn and 50%
alfalfa).
In ponies (age 3 to 18 years), oat feeding caused higher blood glucose concentrations
in comparison to whole corn or barley. The addition of roughage to the
diet blunted the postprandial rise in blood glucose, while the increase in starch
intake (2 g starch/kg BW per meal and 4 g starch/kg BW per meal, respectively)
did not influence blood glucose response (Radicke et al., 1994). The higher glycemic
response to oat feeding was accompanied by a higher prececal starch digestibility
rate of oats. However, in the study by Radicke et al. (1994), mean peak plasma
glucose concentrations after oat feeding were lower (<5.55 mmol/L) when
compared to the mean peak plasma glucose concentrations following corn feeding
(7.85 ± 1.03 mmol/L) measured by Stull and Rodiek (1988), although starch
intake was comparable between these two studies. No differences in mean postprandial
peak plasma glucose concentrations were observed for sweet feed (45%
cracked corn, 45% whole oats, and 10% molasses), whole oats, and cracked corn
by Pagan et al. (2001) in six Thoroughbred geldings. Furthermore, area under the
postprandial glucose curve did not differ for whole oats and cracked corn, but
mean glucose concentrations were higher for whole oats (5.51 mmol/L) when
compared to cracked corn (5.4 mmol/L).
The effect of corn processing on glycemic response was investigated by Hoekstra
et al. (1999) in 6 horses (four Arabians, two Thoroughbreds; age 6 to 10 years).
This experiment was conducted to evaluate how cracking, grinding, or steam
measure of prececal starch digestibility. The glycemic response of each grain
was compared using a glycemic index in which each feed´s glucose area under
the curve (AUC) was expressed relative to cracked corn. The highest glycemic
index was noticed for the steam-flaked corn (2 g starch/kg BW in a single meal;
Figure 1). It is speculated that the high glycemic response reflects changes in
prececal starch digestibility by thermal corn processing.
In a recent study by our own research group, mechanical or thermal processing
of oats, barley, or corn did not clearly influence glycemic or insulinemic response
in horses (Bothe 2001; Vervuert et al., 2002; Coenen et al., 2002). In a crossover
design, six Standardbred horses (age 4 to 15 years, mean body weight 450 ± 37
kg) were fed in random order: untreated, finely ground, steam-flaked, and popped
grain. All diets were adjusted to a starch content of 630 g starch per day from
oats, barley, or corn (1.2-1.5 g starch/kg BW in a single meal). Grain feeding
resulted in a significant increase in plasma glucose and insulin concentrations,
but glucose and insulin peaks as well as AUC were not clearly influenced by
Insulin concentrations followed the glucose curves, but the correlation of mean
plasma glucose and plasma insulin concentrations was not as close as expected
and best described by the following equation: y = 12.6 x – 49.24, where y = mean
plasma insulin and x = mean plasma glucose (r2 = 0.47, p<0.001).
The glycemic index where each feed´s glucose area under the curve was
expressed relative to untreated oats varied between 90.44 ± 13.55% (steam-flaked
corn) and 112.02 ± 18.71% (popped corn, P< 0.05.). The insulin index where
each feed´s insulin area under the curve was expressed relative to untreated
oats ranged between 78.06 ± 33.75% (untreated barley) and 114.76 ± 34.06%
(popped corn, n.s.). The glycemic and insulinemic indexes of oat, barley, and
In our study one striking feature was the high variation in plasma glucose and
insulin response between the horses (e.g., peak plasma insulin concentrations for
oat treatment; Figure 3). The reasons for the great individual variation are not
fully understood. Ralston (1992) noticed a similar individual reaction in horses
after feeding pelleted concentrates with a high or low level of soluble carbohydrates.
In accordance, a high variation in glycemic response was monitored by
Venner and Ohnesorge (2001) after an oral glucose load in healthy horses. A
great variation in behavioral response to grain feeding is monitored by several
horse owners and might be related to the variable glycemic and insulinemic
responses to starch feeding.
In contrast to the investigation by Hoekstra et al. (1999), the effects
of mechanical or thermal grain processing seemed to be of minor importance
for the metabolic reaction and were not reflected in a raised glycemic
or insulinemic response. However, untreated barley or corn was known
to have a low prececal starch digestibility, and thermal treatment enhanced
starch digestibility in the small intestine about threefold (Kienzle et al., 1992;
Potter et al., 1992; Meyer et al., 1995). On the other hand, the improvement in
prececal starch digestibility by grain processing and the increase in substrate
availability might have been masked by other factors. Some of these factors
include the chemical nature of the grain (amylose-amylopectin ratio), interactions
with proteins and fat, the presence of dietary fiber, the rate of gastric emptying,
and amylase availability in the small intestine (Rooney and Pflugfelder, 1986;
Granfeldt et al., 1994).
Several studies have been conducted in humans to compare the amylose-amylopectin
content of starch and its effect on glycemic and insulinemic response.
The glucose and insulin response to high-amylose starch is significantly lower
when compared to starch with moderate levels of amylose (Byrners et al., 1995;
Kabir et al., 1998). More research is necessary to compare the digestibility of and
Clinical Application of the Glycemic and Insulinemic Index in Horses
The nutritional importance of postprandial glucose and insulin response with
regard to different sources of cereals and processing techniques is gaining greater
awareness. On one hand, a high prececal starch digestibility is important to
minimize starch flow into the large intestine, which might lead to considerable
alterations in the microbial fermentation. On the other hand, exaggerated plasma
glucose and insulin responses after carbohydrate intake have been associated
with noninsulin-dependent diabetes and cardiovascular diseases in humans. Foods
that elicit low postprandial glycemic responses are considered beneficial in
subjects with metabolic diseases as well as in healthy human subjects. In horses,
glucose and insulin control may be impaired in a number of life stages and/or
conditions such as diabetes, obesity, gestation, pituitary dysfunction, laminitis,
and aging. In horses with impaired glucose metabolism, plasma glucose concentration
remains higher for longer periods of time, and horses were less sensitive
to insulin than control individuals.
These results by Ralston (1996) were supported in a field study with 218
Thoroughbred weanlings where a high glucose and insulin response to a concentrate
meal was associated with an increased incidence of OCD (Pagan et al., 2001). At
foals with a high risk of developing OCD by their glycemic and insulinemic
response (Ralston, 2001). Based on these results it would be beneficial to feed
young growing horses feedstuffs that are known to elicit a moderate or low glycemic
and insulinemic response.
Metabolic dysfunctions and the glycemic and insulinemic index
Ponies that were fat or had previously suffered laminitis were found to be more
intolerant to oral glucose loads than healthy ponies or Standardbred horses (Jeffcott
et al., 1986). These ponies exhibited a far greater response in plasma glucose and
insulin levels after glucose loading (1 g glucose/kg BW). Furthermore, aged
horses often exhibit a relative glucose intolerance characterized by hyperglycemia
and hyperinsulinemia following a glucose challenge (Ralston et al., 1988).The
glucose intolerance in old horses is caused by a high incidence of pituitary adenomas.
The pituitary adenomas cause excess corticosteroid secretion and impaired
glucose metabolism.
The dietary management of glucose intolerance in horses is not well defined.
However, there are two different ways of influencing dietary glycemic load:
reducing carbohydrate intake or using feedstuffs with a low glycemic index.
However, in humans there is good evidence that a high-carbohydrate intake with
a low glycemic index improved pancreatic -cell function in subjects with
impaired glucose tolerance in comparison to a low-carbohydrate and high-
monounsaturated fat diet (Wolever et al., 2002). In consequence, diets with a low
glycemic index might be preferable in dietary prevention and management of
glucose intolerance in horses.
Glycemic and insulinemic response prior to exercise and training
Dietary management prior to exercise may affect performance by altering energy
metabolism during exercise in horses. In several investigations, corn feeding one,
two, three, or four hours prior to exercise resulted in a marked drop in plasma
glucose and insulin concentration below pre-feeding levels during exercise (Rodiek
et al., 1991; Lawrence et al., 1995; Stull and Rodiek 1995; Pagan and Harris,1999).
In general, horses that began exercise with high blood glucose and insulin levels
(e.g., after corn feeding) showed a transient hypoglycemia during exercise, but
the size of the meal (1, 2, or 3 kg) did not affect the response, although higher
pre-exercise glucose levels were observed when the horses received 3 kg of corn
glucose and insulin concentrations (e.g., after alfalfa feeding) maintained steady
glucose and insulin levels throughout exercise.
A drop in blood glucose concentration may indicate a lack of glucose availability
for the muscle or brain and might have a deleterious effect on performance.
In addition, FFA concentrations during exercise as a major substrate for
energy metabolism were very low when horses received a pre-exercise meal of
corn (Lawrence et al., 1993; Pagan and Harris 1999). In the performance horse, it
might be useful to develop feeding strategies that include feedstuffs with a high
carbohydrate content and low glycemic index.
Conclusion
In humans, starchy foods have been classified over the entire range from
restrained to rapid with respect to effects on blood glucose and insulin responses
after a meal. The resulting glycemic or insulinemic index utilized white bread as
the standard source, and all foods were ranked accordingly. In horses, the classification
of starchy foods with respect to their effects on blood glucose and insulin
responses after a meal might prove useful in developing appropriate feeding
strategies for horses with impaired glucose control or for performance horses.
The glycemic or insulinemic index for horses should use untreated oats as the
standard source. Nutritional factors like amylose-amylopectin ratio, interactions
with proteins, or the presence of dietary fiber that might influence the glycemic
and insulinemic index should be developed, and the great individual differences
in response to starch feeding need further investigation. Furthermore, research is
necessary to investigate the relationship between prececal starch digestibility and
the glycemic and insulinemic index as an indirect measurement of prececal starch
digestibility in horses.
References
Arnold, F. F., G. D. Potter, J. L. Schelling, and W. L. Jenkins. 1981. Carbohydrate
digestion in the small and large intestine of the equine. In: 7th Equine Nutr.
Physiol. Sym., p. 19.
Bothe, C., 2001. Effekte unterschiedlicher Stärketräger und deren Bearbeitung auf die
postprandiale Glucose- und Insulinreaktion beim Pferd.
Hochschule, Diss.
Brand, J. C., P. L. Nickolson, A. W. Thorburn, and A. S. and Truswell. 1985. Food
processing and the glycemic index. Am. J. Clin. Nutr. 42:1192.
Byrners, S. E., J. C. Brand, and G. S. Denyer. 1995. Amylopectin starch promotes the
development of insulin resistance in rats. J. Nutr. 125:1430.
Coenen, M.,
glycemic and insulin responses in horses. J. Appl. Physiol. Nutr., in preparation.
Granfeldt, Y., H. Liljeberg, A. Drews, R. Newman, and
insulin responses to barley products. Influence of food structure and amyloseamylopectin
ratio. Am. J. Clin. Nutr. 59:1075.
Hoekstra, K. E., K. Newman, M. A. Kennedy, and J. D. Pagan. 1999. Effect of corn
processing on glycemic responses in horses. In: 16th Equine Nutr. Physiol.
Sym., p. 144.
Holm, J., I. Lundquist,
starch gelatinization, digestion rate of starch in vitro, and metabolic response
in rats. Am. J. Clin. Nutr. 47:1010.
Jeffcott, l. B., J. R. Field, J. G. McLean, and K. O´Dea. 1986. Glucose tolerance and
insulin sensitivity in ponies and Standardbred horses. Equine Vet. J. 18:97.
Jenkins, D. J. A., T. M. S. Wolever, R. H. Taylor, H. Barker, H. Fielden, J. M. Baldwin,
A. C. Bowling, H. C. Newmann, A. L. Jenkins, and D. V. Goff. 1981. Glycemic
index of foods: A physiological basis for carbohydrate exchange. Am. J.
Clin. Nutr. 34:362.
Jenkins, D. J., T. M. Wolever, and A. L. Jenkins. 1988. Starchy foods and glycemic
index. Diabetes Care 11:149-159.
Kabir, M., S. W. Rizkalla, M. Champ, J. Luo, J. Boillot, F. Bruzzo, and G. Slama. 1998.
Dietary amylose-amylopectin starch content affects glucose and lipid metabolism
in adipocytes of normal and diabetic rats. J. Nutr. 128:35.
Kienzle, E., S. Radicke, S. Wilke, E. Landes, and H. Meyer. 1992. Praeileale
Stärkeverdauung in Abhängigkeit von Stärkeart und -zubereitung. 1. Europ.
Conf. Horse Nutr., Pferdeheilkunde, 103.
Lawrence, L. M., L. V. Soderholm, A. Roberts, J. Williams, and H. Hintz. 1993.
Feeding status affects glucose metabolism in exercising horses. J. Nutr.
123:2152.
.Lawrence, L. M., J. Williams L. V. Soderholm, A. Roberts, and H. Hintz. 1995. Effect
of feeding state on the response of horses to repeated bouts of intense exercise.
Equine Vet. J. 27:27.
Meyer, H., S. Radicke, E. Kienzle, S. Wilke, D. Kleffken, and M. Illenseer. 1995.
Investigations on preileal digestion of starch grain, potato and manioc in horses. J. Vet. Med. A 42:371.
Pagan, J. D., R. J. Geor, S. E. Caddel, P. B. Pryor, and K. E. Hoekstra. 2001. The
relationship between glycemic response and the incidence of OCD in Thoroughbred
weanlings: A field study. In: Proc. 47th AAEP Conv., pp. 322-325.
Pagan, J. D. and P. A. Harris. 1999: The effects of timing and amount of forage and
grain on exercise response in Thoroughbred horses. Equine Vet. J. Suppl.
30:451.
Potter, G. D., F. F. Arnold,
Digestion of starch in the small or large intestine of the equine. 1. Europ. Conf.
Horse Nutr., Pferdeheilkunde, 107.
Radicke, S., H. Meyer, and E. Kienzle. 1994. Über den Einfluß von Futterart und
Fütterungszeitpunkt auf den Blutglucosespiegel bei Pferden. Pferdeheilkunde
10:187.
Ralston, S. L. 1992. Effect of soluble carbohydrate content of pelleted diets on postprandial
glucose and insulin profiles in horses. 1. Europ. Conf. Horse Nutr.,
Pferdeheilkunde, 112.
Ralston, S. L. 1996. Hyperglycemia/hyperinsulinemia after feeding a meal of grain to
young horses with osteochondritis dissecans (OCD) lesions. 2. Europ. Conf.
Horse Nutr., Pferdeheilkunde, 320.
Ralston, S. L. 2001. Glucose intolerance and developmental orthopedic disease in
foals–A connection? In: J.D. Pagan and R.J. Geor [Eds.] Advances in Equine
Nutrition II. p. 397-401.
Ralston, S. L., C. F. Nockels, and E. L. Squires. 1988. Differences in diagnostic test
results and hematologic data between aged and young horses. Am. J. Vet. Res.
49:1387.
[Back]
