Lipid and carbohydrate sources for elemental, enteral diets for neonates


Material Information

Lipid and carbohydrate sources for elemental, enteral diets for neonates
Physical Description:
ix, 147 leaves : ill. ; 29 cm.
Rice, Lori P., 1957-
Publication Date:


Subjects / Keywords:
Dissertations, Academic -- Animal Science -- UF
Animal Science thesis Ph. D
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1989.
Includes bibliographical references (leaves 138-145).
Statement of Responsibility:
by Lori P. Rice.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001561201
oclc - 22655257
notis - AHH4882
sobekcm - AA00004785_00001
System ID:

Full Text







This dissertation is dedicated to my father,

Bernard Rudin,

who passed away in November, 1987. His humor,
wisdom, support, and encouragement, unselfishly
given to all who knew him, inspired great deeds.
He believed that I could accomplish anything I
set out to do. All things were possible.


Sincere gratitude and appreciation are extended to the

chairman of my supervisory committee, Dr. E. A. Ott. He

always had patience, tolerance, and good spirits, even

during difficult times (such as trying to catheterize an

uncooperative foal at 3 a.m.). His guidance and knowledge

were essential to the completion of this project.

A special thanks is given to Dr. Peggy Borum, who

introduced me to the challenging field of neonatal

nutrition. Her dedication and high research standards were

an inspiration to do the very best job possible.

Support, encouragement, and advice were liberally given to

me by the other members on my advisory committee: Dr. E.

Johnson, Dr. D. Beede, and Dr. S. Lieb.

I am extremely grateful for the assistance of Fred

Buhl, who solved all my computer problems, Dr. Wilcox, who

very patiently helped with the statistical analyses, Mel

Tooker, and the crew at the Horse Research Center (C. Roden,

farm manager, Rose, Cindy, Jim, and Sharon). Maria

Fernandez and Sandy Duyck performed miracles restraining

fiesty foals and provided vital feedback. The piglet

projects benefitted from the experienced help of Janet


Baltzell, Mike Sullivan, Sergio Quintana, and Robin Adkins.

My friends, Ty and Scott McClung, have been especially

helpful and supportive during my entire program.

Unwavering support, love, and encouragement were given

to me by my family. My husband, Brett, allowed himself to

be dragged 1000 miles to sweltering Florida and put up with

a student for a wife for longer than he'd like to remember.

My mother, Ellie Rudin, my grandmother, Ida Rudin, my sister

and her husband, Gail and Chuck Willis, were always there

when I needed them.


ACKNOWLEDGMENTS...................................... iii

ABSTRACT............ ............. ......... ........... viii


I REVIEW OF THE LITERATURE...................... 1

Introduction.................................... 1
Formulating Enteral Diets for Neonatal
Foals................. .................... 2
Foals as Research Animals................. 3
Piglets as Models in Foal Research........ 5
Digestive Problems of Neonates................ 7
Infectious Causes of Diarrhea ........... 7
Non-Infectious Causes of Diarrhea......... 8
Lipid Metabolism in Neonates.................. 9
Fat in the Neonatal Diet................. 9
Fatty Acid Digestion and Absorption...... 11
Metabolism of Medium Chain and Long
Chain Fats by Neonates................. 13
Disaccharidase Activity in Neonatal Small
Intestine........... ............... .......... 15
Introduction............................. 15
Brush Border Disaccharidases.............. 19
Developmental Patterns of Enzyme Activity
in Neonates... ........................... 20
Distribution of Disaccharidases.......... 25
Sucrase.................................... 26
Lactase.................................... 27
Maltase................ .......... .... .... 28
Determining Disaccharidase Activity in
Experimental Animals................... 30
Responses of the Neonate to Stress............ 31
Failure to Acquire Passive Immunity....... 31
Cold Stress.................... ............ 33
Compromised Neonates...................... 36
Hematology and Blood Chemistry in Neonates.... 37
Hematology................................ 37
Chemistry.... ............... ......... 39
Milk Composition and Intake.................... 40


Milk Production in Mares................. 41
Special Nutrient Requirements of
Neonates................................ 43


Introduction................................... 45
Materials and Methods........................... 48
Animals. ................................. 48
Surgery.......... ..................... 50
Diets. .... .....* .. .. ... ............ 52
Development of Feeding Regimen............ 56
Daily Care............... ................. 57
Tissue Collection........................ 58
Analyses.................................. 59
Additional Information................... 60
Statistical Analysis...................... 61
Results........... ........... ............... 61
General Observations...................... 61
Gross Observations at Necropsy........... 64
Weight Gain.............................. 65
Small Intestine Length................... 66
Blood Glucose............................. 67
Small Intestine Weight.................... 67
Disaccharidases in the Small Intestine... 67
Complete Blood Counts.................... 73
Urinalysis............................... 75
Discussion. ......... ....... ................... 75
Disaccharidases in Neonatal Pigs......... 76
Growth Rate in Neonatal Pigs.............. 79
Development of the Small Intestine in
Neonatal Pigs........................... 79
Blood Glucose............................. 80
Hematology. ... ............... .......... 81
Summary of Piglet Trial................... 81


Introduction................................ .. 83
Materials and Methods.......................... 87
Preliminary Study........................ 87
Animals and Design....................... 88
Analyses......................... .. .. 89
Statistical Analyses..................... 89
Results........ .. ... ... ............ ... ....... 93
General Observations........................ 93


Fasting Blood Glucose.................... 94
Glucose Absorption Time Curves.......... 94
Treatment Curves Pooled Over All Days.... 95
Comparison of Day-Treatment Curves....... 100
Discussion...................................... 108
Fasting Blood Glucose in Neonatal Foals.. 108
Treatment Curves Pooled Over All Days.... 110
Day 1 Treatment Curves................... 112
Day 3 Treatment Curves................... 112
Day 5 Treatment Curves.................... 114

IV CONCLUSIONS............................... ......118


A PIGLET AND FOAL DATA........................... 122


LITERATURE CITED...................................... 137

BIOGRAPHICAL SKETCH.................................. 145


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Lori P. Rice

December 1989

Chairman: Edgar A. Ott
Major Department: Animal Science

Colostrum-deprived piglets were fed either elemental,

enteral diets, containing hydrolyzed whey proteins and

malto-dextrins (Group E), or a conventional milk-based diet

(Group C). Two litters were euthanized at birth (BO) to

provide baseline values. Normal values were obtained from

two litters of sow-fed (SF) piglets.

Group E piglets were surgically fitted with gastric,

bladder, and umbilical artery catheters, and fed by bottle

or continuous infusion pump. These piglets experienced

diarrhea. Diet and flow rate modifications, made for each

litter, reduced but did not eliminate the diarrhea.

To investigate diaccharidase activity in the small

intestine (SI) of these piglets, tissue homogenates were

incubated with maltose, lactose or sucrose. Maltase

activity was low in the SI of BO piglets suggesting that the

malto-dextrin in the elemental diet was probably not well


digested by Group E piglets. Undigested maltose reaching

the large intestine probably caused the diarrhea.

Maltase activity was detected in the SI of Group C and

SF piglets even though they had never ingested maltose.

Similarly, lactose activity was detected in the SI of Group

E piglets that had never ingested lactose. Sucrase activity

was very low in all of the piglets.

To investigate disaccharidase activity in neonatal

foals, an oral tolerance test was conducted on 13 foals.

Maltose, lactose, sucrose, and glucose solutions were

administered orally on d 1, 3, and 5 postpartum. Changes in

plasma glucose from venous blood were measured over time.

Large peaks in plasma glucose occurred shortly after dosing

with lactose and glucose on all days. This suggests that

glucose from both substrates was absorbed. No rise in

plasma glucose was detected after oral sucrose

administration, even on d 5. Maltose was not well digested

on d 1. However, some digestion may have occurred on d 3

and 5, as evidenced by small increases in plasma glucose

after dosing with the maltose solution.

The colostrum-deprived piglet can be used successfully

as an animal model for nutritional investigations. The

results of the piglet and foal trials suggest that maltose

and sucrose would not be suitable substrates for these

neonates in the first few days of life.



Advances in medicine have decreased greatly the

mortality rate of weak or sick neonatal humans and animals.

New techniques in neonatology, especially in the field of

respiratory therapy, have enabled veterinarians to save many

diseased or premature foals that otherwise would not have

survived. The value, whether monetary or sentimental, of

many horses today has resulted in a greater number of high

risk foals receiving intensive care. As a result of the

great strides made in saving these foals, a growing number

of veterinary hospitals are developing or expanding their

facilities for the care of neonatal foals.

Once conditions that immediately threaten the life of

the newborn are stabilized, the primary concern becomes

administration of nutrients in a form that can be readily

utilized. In many cases, the foal can not or should not

consume mare's milk. Some foals must be separated from

their dams while they are in an intensive care setting.

Sick or weak foals that can not stand to nurse must be hand-

fed an appropriate enteral or parenteral diet.


Many of the feeding programs used on neonatal foals were

originally developed for human infants in intensive care

nurseries. Unfortunately, due to the difficulties in

working with these neonates, many of the formulas and

supplements being fed have not been well tested on very

young or stressed foals (Koterba and Drummond, 1985).

Formulating Enteral Diets for Neonatal Foals

Before an appropriate diet for stressed foals can be

formulated, basic research is needed to provide an

understanding of their metabolism. Many questions about the

efficiency of utilization of different nutrients remain


Ensuring adequate caloric intake is of primary

importance. In the past, some foals were inadvertently

starved to death while being hand-fed because their energy

requirements were underestimated (Koterba and Drummond,

1985). There is disagreement in the literature about the

amount and type of enteral diet that will supply adequate

nutrients during the critical neonatal period (Naylor and

Bell, 1985).

Stressed or sick foals often have poorly functioning

digestive systems and may not absorb nutrients or antibodies

well. Some of these foals may even need parenteral feeding

to increase the supply of nutrients available to the cells.

Young foals are especially susceptible to infection when

they do not get an adequate supply of colostral antibodies

from their mothers, or when they are premature or dysmature

at birth (Rossdale, 1987). It has been shown in horses and

in other species that the nutritional needs of premature and

sick neonates can be different from that of term, healthy

newborns. While their need for nutrients remains the same

or even higher than that for healthy foals, their appetites

may be poor (Koterba and Drummond, 1985). Therefore, the

conclusions drawn from studies done on normal foals would

not necessarily be applicable to the weak or sick patients

treated in veterinary clinics.

Foals as Research Animals

Many problems present themselves when trying to design

experiments involving a relatively large number of research

foals. The expense of buying and feeding broodmares is

great, and there is no assurance that each mare will produce

a foal every year. Horses have a long gestation and tend to

foal during the same time of year. This makes it difficult

to spread foaling out over a long enough period of time to

facilitate handling many foals.

The labor required for round-the-clock care of even a

single foal is extensive. Even very young foals are

relatively large and require more elaborate facilities than

do small laboratory animals. Also, large radioactive

carcasses resulting from studies involving labeled compounds


would be difficult to dispose of. These concerns tend to

limit the number of foals that can be successfully studied

each year. These problems can be overcome, in part, by

using an appropriate animal model for preliminary


Using animals as experimental models for human studies

is common practice. Ethical considerations have led to

extensive use of animal models for research concerning human

infants. Although the rat is frequently used as a model for

adults, rat pups are not similar to infants or foals.

Newborn rats have intestines that are fragile and difficult

to work with, and the rats' stage of development and rate of

growth at birth are very different from newborn humans,

pigs, horses, and ruminants (Widdowson, 1984). Calves,

while similar to foals, would not offer many advantages as a

model for foals.

Piglets have been used successfully as models for

premature infants. It has been suggested that they are

closer to human newborns than any other animal (Glauser,

1966). One of the objectives of this research is to

determine the appropriateness of using piglets as models

for the neonatal foal.


Piglets as Models in Foal Research

One of the major limitations of any animal model is, of

course, that the model is not going to behave exactly like

the real subject. Therefore the model chosen should always

come as close as possible to resembling the animal of

interest. There are no references in the literature to

previous work comparing piglets and foals, but information

on each shows that these neonates share many common traits.

The pig and the foal both tend to nurse frequently during

the day for short periods of time (Pegorier et al., 1983).

They have similar digestive functions at this age, both

being adapted to a milk diet. Also, both species depend

heavily on the passive immunity provided by the mother's

colostrum, which must be ingested shortly after birth.

It has been shown by many workers that foals and

piglets who do not acquire colostral antibodies are at

severe risk from many diseases including pneumonia,

septacemia, arthritis, and infectious diarrhea (Rossdale and

Leadon, 1975; Naylor and Bell, 1985; Rossdale, 1985; Koterba

et al., 1984; Mouwen, 1971; Bergeland and Henry, 1982). The

colostrum from both sows and mares contains predominantly

IgG (Pegorier et al., 1983; Naylor, 1979); however, IgA is

present in greatest quantities in non-colostral mare's milk.

Good management and sanitary conditions can help save

hypoglobulinemic neonates, but the risk of infection is

still high.


It has been suggested that mare's milk is closer to

that of primates than any other animal, being high in

lactose and low in fat and protein (Weaver, 1986). If

mare's milk is similar to human milk, and piglets are used

to simulate the response of infants, then perhaps the theory

that foals can also be compared to piglets has some


The use of the pig as a model has both advantages and

limitations. Being able to do research on piglets that are

littermates and on pigs from different litters, but by the

same parents, allows experiments to be performed using large

numbers of animals with minimal genetic variation. Also,

many animals can be studied in a relatively short period of

time, as the gestation of the sow (approximately 114 d) is

much less than that of the mare (approximately 335 d). Sows

can be bred to farrow all year, while mares have a foaling

season of several months.


Digestive Problems of Neonates

After birth, mammalian neonates undergo a period of

rapid acclimation to extrauterine life. They must adapt to

using enterally rather than parenterally supplied nutrients.

The neonate is now a free-living organism and it must be

able to ingest and absorb nutrients from its environment.

In mammals, of course, this usually means learning to nurse

from its mother. Serious problems occur when the newborn

has difficulty suckling or has an incompetent digestive

tract. Immature or damaged intestinal tissues will not

function normally, therefore, it is necessary to consider

the nutritional needs of neonates with digestive problems

separately from those of healthy neonates.

Diarrhea is a potentially serious problem for neonates

and can occur for several reasons. The causes are usually

classified as infectious or non-infectious.

Infectious Causes of Diarrhea

Diarrhea in piglets and foals has been attributed to an

assortment of viruses and bacteria. Rotavirus, in

particular, seems to be present in over 30% of the cases,

while coronavirus and adenovirus are also common infectious

agents (Palmer, 1985; Bergeland and Henry, 1982). These

authors suggest that rotavirus acts by causing damage to the

epithelium of the small intestine and leaving the neonate

vulnerable to a secondary bacterial infection. The bacteria


usually cultured in cases of piglet diarrhea include

Salmonella, and E. coli, while the latter is not considered

a problem in foals. They also suggested that in foals,

bacterial diarrhea may be secondary to septacemia and

enteritis. Piglets may also be plagued by coccidiosis or

infestations of strongyles.

Because of the ever present threat of invasion by

infectious organisms, the area where newborns are housed

must be kept extremely clean. It has been suggested that

foals usually become infected through contact with adult,

asymptomatic shedders on the same farm. Treatment for

diarrhea in piglets and foals usually includes

administration of antibiotics to cause a change in gut

flora, supplemental vitamins to offset decreased gut

absorption and bacterial synthesis, and, in piglets,

antispasmodics (Mouwen, 1971; Palmer, 1985).

Non-Infectious Causes of Diarrhea

Stress of all kinds, including exposure to low ambient

temperatures, or malfunctioning of the small intestine can

predispose newborn animals to diarrhea (Bergeland and Henry,


Many foals suffer from diarrhea on d 6 to 14

postpartum. Because this is also the time when the mare

returns to estrus after giving birth, it has been called

"foal heat" diarrhea. However, there was no difference in


the amount or severity of diarrhea in hand-fed or mare-fed

foals observed by Palmer (1985). Therefore, the theory that

hormonal changes in the mare are reflected in the milk is

unfounded. It is important, though, to monitor the foal

closely at this time, because the onset of infectious

diarrhea can mimic foal heat diarrhea.

Piglets may excrete thin, watery feces called "white

scours" when they are very young. They usually seem normal

in attitude, but may be less-active. The feces, upon

analysis, usually contains greater than 50% fat, but it is

not known if this is the cause or effect of the condition.

One experimenter suggests that it is the presence of

abnormally structured gut villi that is responsible for the

diarrhea (Mouwen, 1971).

Lipid Metabolism in Neonates

Fat in the Neonatal Diet

The addition of fat to human and animal enteral

formulas has proven to be an effective means of increasing

caloric density in an easily digestible form without also

increasing osmolarity (Koterba and Drummond, 1985). By

increasing the caloric density, the volume of formula to be

fed can be kept low. Large fluid loads can be detrimental

to sick or premature newborn patients. Fats are also added

to enteral and parenteral solutions to provide essential


fatty acids and to provide a carrier for fat soluble


Interest in studying the type of fat most readily

utilized by neonates has increased because the composition

of fat in fetal pig tissues and in sow's milk and colostrum

can be influenced by the source of dietary lipids fed to the

sow. Therefore, it may be possible to manipulate the diet

of piglets to affect the composition of body lipids. This is

important because there is evidence that piglets metabolize

fats differently according to chain length and degree of

saturation (Miller et al., 1971; Lloyd and Crampton, 1957).

This also appears to be true in human infants (Faber et al.,

1988; Bach and Babayan, 1982; Lammi-Keefe and Jensen, 1984).

The fat content of milk and the extent of body lipid

stores at birth vary with species. There is usually an

increase in body fat of young animals after nursing.

Neonates who are energy-deprived must depend on fat reserves

for metabolizable energy and for thermal regulation

(Pegorier et al., 1983).

It has been suggested that within 2 d of birth neonatal

piglets can utilize fat as efficiently as glucose and the

ability to use fat increases with age (Miller et al., 1971;

Wolfe et al., 1978). Piglets have low fat stores at birth

(approximately 1% of their body weight) that rapidly

increase to about 15% of their body weight by day 14.

Wolfe et al. (1977) fed isocaloric diets to colostrum-


deprived, newborn piglets. The diets differed only in the

proportion of calories provided by carbohydrate and

butterfat. The fatty acid pattern of the backfat of the

piglets resembled the fatty acid composition of the diet,

especially when high fat diets were fed. Lipogenic enzyme

activity decreased with higher fat diets, which the authors

suggest is due to the inhibiting effect of increased lipid

intake rather than to decreased carbohydrate intake. Wolfe

and coworkers in 1978 determined that on low fat diets,

newborn piglets will resort to de novo fat synthesis and

attain a level of fat deposition similar to piglets fed a

high fat diet.

Fatty Acid Digestion and Absorption

Researchers have investigated the digestion and

absorption of lipids in neonates of many species, but more

information on how different types of fats are utilized by

premature and compromised newborns is needed. Data on lipid

metabolism in foals are limited.

Of particular interest is the effect of chain length on

neonatal absorptive and metabolic processes. In the past,

infant formulas contained exclusively long chain

triglycerides (LCTs), but recent work suggests that

compromised neonates may benefit from the replacement of

some of the LCT with medium chain length fats. It has been

suggested that medium chain triglycerides (MCT), usually


saturated fats consisting of 6 to 12 carbons, are readily

hydrolyzed and absorbed by infants, rat pups, and piglets

(Newport et al.; 1979, Putet et al., 1987; Jandacek et al.,


Medium chain fatty acids (MCFA) from the diet are

transported directly to the liver via the portal vein

whereas long chain fatty acids (LCFA) travel through the

lymph and then through the general circulation before

reaching the liver (Newport et al., 1979). MCT being more

polar and therefore more water soluble than LCT, are more

easily attacked by pancreatic lipase and thus, are a

desirable alternative to LCT in cases of pancreatic

insufficiency or in other situations when LCT would be

poorly absorbed (Jandacek et al., 1987).

Specifically designed triglycerides have been used to

benefit from both MCT and LCT in a single lipid source. A

triglyceride having medium chain fatty acids at the one and

three positions and a long chain fatty acid at the two

position may be the solution. This takes advantage of the

pancreatic lipase specificity for cleavage at the one and

three sites, leaving a 2-monoglyceride. Even in cases where

LCT absorption is poor, the 2-monoglycerides seem to be

easily transported across the intestinal mucosa (Newport et

al., 1979), thereby avoiding essential fatty acid (EFA)

deficiency. Human milk triglycerides have a 16:0 at the 2


position, which is thought to enhance the absorption of

dietary fats (Lammi-Keefe and Jensen, 1984).

Metabolism of Medium Chain and Long Chain Fats
by Neonates

The different metabolic pathways available to MCFA and

LCFA affect the energy sources available to peripheral

tissues. Interest in substituting MCT for LCT as part or

all of the lipid content of enteral formulas stems from the

products of metabolism of MCT. Several groups of

investigators have begun feeding MCT to neonates in an

effort to determine its benefits and drawbacks.

There is evidence to suggest that infants, especially

preterm infants, do better on MCT-added formulas than on

cow's milk alone, even though MCT are not found in large

amounts in mature human milk (Putet et al., 1987). Lammi-

Keefe and Jensen (1984) found considerably more 10, 12, and

14 carbon fatty acids in the milk of mothers that had very

premature or premature babies when compared to the milk of

mothers with full-term babies. They speculate that this may

be of some special benefit to premature infants. Some

researchers advocate feeding MCT to infants that have

malabsorption of lipids (Bach and Babayan, 1982).

MCT and MCFA are not normally found in the blood even

when they are added to the diet. They are thought to be

rapidly metabolized by the liver and do not reach the


peripheral tissues intact. However, 15-20% of the fatty

acids found in the umbilical cord blood of infants have

eight carbons or less. MCT tend to increase carbohydrate

tolerance, insulin secretion, and decrease glucose output

from the liver when given orally (Bach and Babayan, 1982).

A study on neonatal puppies (Cotter et al., 1987)

suggests that insulin causes LCFA to be removed from the

blood for storage, while MCFA are removed for immediate use

as an energy substrate. The authors have shown that low

doses of MCT given intravenously (i.v.) are utilized more

rapidly than LCT. The calories are available faster because

the lipoprotein lipase (LPL) in the capillary walls quickly

releases the MCFA, which then bind to albumin and are

absorbed by peripheral tissues. In the liver, MCFA can

enter the mitochondria rapidly, in contrast to LCFA which

must be activated and be transported with the help of a

carnitine carrier. MCFA can be oxidized so quickly that the

Kreb's cycle is overwhelmed. When this happens, the 2-carbon

units are funneled into ketone bodies which then diffuse

into the blood. The authors suggest that MCT is a more

flexible energy source than LCT. They also observed an

increase in plasma clearance rate of long chain fats when

medium chain fatty acids were present.

In a similar study, piglets fed either a 25:75 or 50:50

mixture of dietary MCT and LCT had higher blood glucose

concentrations and increased insulin secretion than pigs fed


a diet in which the fat source was 100% LCT (Wolfe et al.,


Newport et al. (1979) fed 2 day old neonatal pigs

various ratios of MCT and LCT until they were euthanized at

day 28. The pigs fed a 50:50 ration had similar growth

rates to pigs fed 100% LCT as a dietary fat source.

However, 90% MCT feeding resulted in a significantly slower

growth rate. The pigs fed 90% or 50% MCT had higher liver

weights but lower carcass weights than the 100% LCT group.

The addition of MCT resulted in lower plasma lipid and free

fatty acids (FFA), but higher body water, total body lipids,

and cholesterol. They found no increase in nitrogen

retention due to MCT feeding as had been suggested by


MCT feeding is not without possible disadvantages.

Urinary output of dicarboxylic acids (DCA) may be seen.

These end products of omega oxidation of medium chain fatty

acids, such as seberic, sebacic, and adipic, are often

detected in the urine during fasting, during ketosis, or in

children with defects in fatty acid oxidation or carnitine

metabolism. Henderson and Dear (1986) fed preterm infants

averaging 29 wk gestational age either parenterally (no

medium chain fats included) or enterally. The enteral

formulas contained either a high amount of MCT (1.8g/100ml)

or a low amount (.44 g/100 ml). Some infants received

breast milk containing .15 g MCT/100 ml. The infants fed

the high MCT formula excreted larger amounts of DCA than

infants fed the other feeding regimens, although no other

symptoms were evident. It is unclear if this would be a

problem for healthy infants. The authors stated that MCT

feeding would be unadvisable in infants with #-oxidation

defects and that more investigation is needed to determine

the effects, if any, of increased urinary excretion of DCA.

They did not comment on the possibility that DCA excretion

resulted from inefficient use of MCT by the infants.

Ketones from MCT provide a quick source of energy for

neonatal extrahepatic tissues. As their concentration in

the blood increases, utilization by extrahepatic tissues

increases. Therefore, they are a good substrate for energy

production in times of high energy needs, such as growth or

undernutrition (Bach and Babayan, 1982). Ketogenesis is

controlled by a feedback system: when the concentration of

ketones in the blood increases, negative feedback to the

liver decreases production (Robinson and Williamson, 1980).

Gentz et al. (1970) have shown moderate increases in

plasma ketones during starvation in newborn and 16 day old

pigs during times when fat utilization is limited. They

suggest that this may not be detrimental because 10 kg pigs

are not prone to ketosis during starvation. Pegorier et al.

(1983) question the extent of ketogenesis and #-oxidation in

newborn pigs since they do not show hyperketonemia as do

rabbits, rats, guinea pigs, and humans fed high fat diets.


In summary, the MCT system has a fast turnover rate but

is easily overloaded. The presence of medium chain fats

stimulate the absorption of plasma FFA. The 2-carbon units

resulting from #-oxidation of MCFA are ketogenic, thereby

providing the flexibility of an alternative energy source.

The LCT system handles large quantities well but has a slow

turnover rate. The resulting increase in plasma FFA may

lead to acidosis. Some LCT are needed in the diet to

prevent essential fatty acid deficiency. MCFA may improve

the utilization and absorption of LCFA when there is

impaired lipid digestion.

Disaccharidase Activity in Neonatal Small Intestine


Milk is a complex mixture of nutrients, hormones, and

other substances designed to provide optimal nutrition for a

normal, healthy newborn. Consequently, it is a difficult

task to formulate an artificial diet that will allow the

hand-raised neonate to grow and develop normally. The

challenge is even greater when the neonate is sick, weak or


The developmental pattern of intestinal disaccharidases

has been used to assess the nutritional status of neonates.

It has been shown that improper nutrition results in a

malnourished intestinal tract that does not show the same

pattern of enzyme activity changes as a healthy neonate


nursing its mother (Rossi et al., 1986). These authors

experimentally malnourished rats adding extra pups to a

normal-sized litter. The pattern of disaccharidase activity

in the malnourished rat pups resembled that of lesser

developed pups. The maltase and sucrase activities were

lower, and the lactase activity higher than in the control


Sick neonates often lose the ability to digest lactose

(Tzipori et al., 1984) and in infants, prematurity may

result in low total intestinal lactase activity (Mayne et

al., 1986). Therefore, it is important to assess the value

of alternative carbohydrates when developing artificial

formulas Dairy farmers are also interested in the use of

economical energy sources for use in calf milk replacers

(Dollar and Porter, 1957). The efficient utilization of any

energy source will depend on the activity of appropriate

intestinal enzymes for digestion and absorption. Digestion

of complex carbohydrates, such as corn starch, are cleaved

by amylases into smaller units, into maltose, and finally

into single glucose molecules that can be absorbed (Code,

1968). Therefore, starch can not be utilized unless there

is adequate amylase and maltase present. Roberts (1974)

detected only low levels of both heat-sensitive pancreatic

a-amylase and heat-stable glucoamylase in neonatal foals.

One way to determine the digestibility of various sugars for


use in neonatal formulas is to estimate the amount of

disaccharidase activity present in the intestine.

Brush Border Disaccharidases

Disaccharidases are synthesized in membrane-bound

polysomes within enterocytes of the small intestine. There

is evidence to show that they are present in an inactive

form in crypt cells, becoming activated when the cells have

migrated to the top of the villus. Active lactase enzyme

has only been found in the microvillus membrane on "mature"

enterocytes at the tip of the villi in neonatal rats

(Mackenzie, 1985). The enzymes are activated when proteases

cleave a polypeptide chain from the large pro-disaccharidase

molecule. Extracellular pancreatic proteases are

responsible for activating maltase and the

sucrase-isomaltase complex, while lactase is activated by

intracellular proteases. The disaccharidases are attached

to the brush border of the epithelial cells lining the

intestine. In the rat, the rate of turnover for these large

proteins is short, about 11.5 h. The enterocytes themselves

take only 4 to 5 d to migrate from the crypt to the tip of

the villus, where they are shed (Dahlqvist and Semenza,

1985). The rapid rate of cell division and protein

synthesis makes these cells and their attached enzymes very

susceptible to adverse conditions and poor nutrition.


Disaccharidase enzymes, primarily sucrase, maltase, and

lactase, tend to follow the same pattern of development in

most mammals, but the rate of development, pre- and post-

partum, differs by species (Paige and Bayless, 1981). The

amount of each disaccharidase present in the intestine will

determine the carbohydrates that can be successfully

incorporated into milk replacer formulas. Milk contains

many substances that can affect the development of the

intestines. Therefore, artificial formulas must be able to

support the normal pattern of maturation seen in neonates

nursing their mother. Determination of enzyme activity will

allow assessment of the development and general health of

the intestine. In this way, the nutritional value of

artificial diets can be compared to that of milk and


Developmental Patterns of Enzyme Activity in Neonates

The relative progression of the increase and decrease

in disaccharidase enzymes is similar for many animals, but

the timetables are very different for each species. For

most mammals, lactase is high at birth, increasing during

the nursing period, and decreasing near weaning. Lactase,

which develops in human infants later in gestation than

sucrase and maltase, reaches a peak at 36 to 40 wk gestation

(Roggero et al., 1986). Therefore, very premature infants

will have a limited ability to digest this sugar. Birth and


feeding each contribute to a postnatal rise in lactase in

both term and preterm infants (Weaver et al., 1986).

Studies on neonates of several species indicate

that prematurity or damage to the intestines can greatly

reduce lactase activity. Mayne and coworkers (1986)

developed a technique for assessing disaccharidase activity

in living human infants. They demonstrated a significant

correlation between the enzyme activities measured in

jejunal fluid with enzyme activities in adjacent mucosal

tissue. Using this method, they studied eight premature,

but appropriate for gestational age infants 17 times during

the first 3 wk after birth. The infants were fed

pasteurized breast milk via nasogastric tube. The lactase

activity in these infants at 1 wk of age was lower than the

activity present at 2 and 3 wk of age (P < .001). The 1 wk

old infants had enzyme activities in the normal range for

adults. This work agreed with other published reports

indicating that lactase activity increases greatly near the

end of gestation in humans. This results in a lowered

capacity of preterm infants to digest a lactose-

based diet.

Damage to the intestinal tissue may also result in

reduced lactase activity. Biopsies were obtained proximally

and distally to a complete or incomplete congenital blockage

of the small intestines of 12 newborn infants during surgery

(Serrano and Zetterstrom, 1987). The activities of all


disaccharidases in tissue samples proximal to the

obstruction site were reduced when compared with previously

published reference values. Lactase activity was more

markedly reduced than the activities of the other

disaccharidases. There was no difference in lactase

activities in tissues from the four infants with complete

blockages compared the eight infants with incomplete


However, when tissue samples from sites distal to the

obstruction were compared, lactase activity was lower (P <

.03) in infants with complete blockage when compared with

infants with incomplete blockage. The authors suggest that

distension and damage to the small intestine proximal to the

blockages resulted in impaired development of brush border

enzymes, particularly lactase. Minimal passage of ingesta

through an incomplete blockage permitted greater development

of lactase activity than detected in tissue distal to a

complete blockage.

To investigate the effect of infectious pathogens on

disaccharidase activities in neonates, Tzipori et al. (1984)

inoculated seven 4 day old foals with various combinations

of Streptococcus durans, Staphylococcus epidermis, and

Escherichia coli isolated from feces of foals with naturally

occurring diarrhea. Three foals received saline

inoculations. Inoculations were administered via

nasogastric tube and were followed with 200 ml evaporated


cow's milk. The foals were necropsied 1 to 3 d later. The

seven foals inculated with pathogens had experienced

diarrhea. S. durans bacteria were found to be adhered to

the intestinal mucosa. This probably contributed to the

significantly lowered lactase activity measured in these

foals compared with the lactase activity of the control

foals. These authors then conducted a similar study on

newborn piglets, inoculating them with the same bacterial

strains isolated from foals. These piglets also experienced

diarrhea and reduced lactase activity compared with saline

inoculated controls.

Maltase and sucrase do not appear in any appreciable

amount until at least a few days after birth (Code, 1968;

Veum and Mateo, 1986; Manners and Stevens, 1972; Ahrene et

al., 1969). This is the case for rabbits, calves, pigs,

rats, dogs, and cats (Code, 1968; Paige and Bayless, 1981).

Guinea pigs, however, are very mature at birth, nurse

infrequently, and can digest maltose and sucrose immediately

(James et al., 1987). Human infants develop considerable

maltase and sucrase activity before birth (Paige and

Bayless, 1981).

It has been reported that newborn piglets can utilize

lactose and glucose equally well at birth (Dollar et al.,

1957; Ahrene et al., 1969), while sucrase and maltase

activities increase during the first few days of life (Veum

et al., 1986). Dahlqvist (1961a) reported high lactase


activity in the pig at birth and a rapid increase in maltase

by day 2, but he found no appreciable sucrase at this age.

In contrast, Bellis (1957) reported that by day 3, piglets

could digest sucrose, lactose, and glucose equally well.

James et al. (1987) found low levels of maltase and sucrase

in sow-fed piglets on d 1 to 4 of life, with levels

increasing from day 3 through day 10. Sucrase activity was

twice that of maltase. When they gave Epidermal Growth

Factor (EGF), which may be present in colostrum, to

hand-raised piglets, they found an increase in sucrase and

maltase in the middle and distal portions of the small

intestine. EGF had no effect on lactase.

Dairy calf producers have tried to formulate milk

replacers using economical energy sources other than

lactose. Dollar and Porter (1957) fed newborn calves skim

milk using starch as an energy source with and without

amylase. Calves under 3 wk of age did poorly on this diet.

The authors suggest that the calves were unable to digest

the dextrins and maltose resulting from amylase digestion.

They report only very low activities of maltase and amylase

in the newborn calf and no sucrase activity. They state

that calves under 4 wk old have only been shown to digest

lactose and glucose. They did not see an appreciable rise

in maltose digestion until the calves were 9 wk old.

Distribution of Disaccharidases

The distribution of disaccharidase activity in the

small intestine has been investigated in several species.

In calves, lactase activity is highest in the proximal small

intestine. With age, lactase activity is reduce, mostly in

the distal third (Huber et al., 1961). Manners and Stevens

(1972) studied enzyme distribution in the newborn pig.

While they found great variation from pig to pig, they were

able to draw some conclusions. Enzyme activities were

measured at various sites along the small intestine, with

the proximal and distal ends represented as the 0% and 100%

sites, respectively. Lactase activity was highest at the

20% site along the small intestine, decreasing to a low at

the 95% site. From birth to 1 wk old, much of the decrease

seen in lactase activity occurred in the proximal 50%, with

higher activities found in the more distal section. Sucrase

activity was only barely detectable until the pigs were 1 wk

old, with the highest values in the proximal two-thirds of

the small intestine.

In foals, lactase activity was reported to be highest

in the duodenum and upper jejunum, decreasing distally

(Roberts et al., 1974). Maltase activity was low in the

equine fetus, increasing slowly during the first month after

birth mostly in the proximal and middle segments of the

small intestine, then increasing markedly until the end of

the first year. Sucrase was very low at all sites along the


intestine in the newborn foal, increasing in the proximal

sections during the first year, with the pattern of

distribution parallel to that of maltase.

Postmortem studies of the entire length of small

intestine of preterm and near-term human infants provided

information on the development of disaccharidases (Raul,

1986). The relative distribution of sucrase and lactase was

similar for infants in both age groups throughout the

intestine. The greatest amount of lactase activity was

found in the jejunum and duodenum, with little found in the

distal small intestine. Term infants had five times greater

lactase activity than the preterms, with the largest

differences found in the proximal intestine. The greatest

amount of sucrase activity was found in the jejunum. The

term infants had more enzyme activity in all segments of the

intestine. Glucoamylase was present in the tissues of

preterm infants, increasing distally, as it does in adults.

The authors suggest that this enzyme, along with salivary

amylase (which, in preterm infants, is still functional in

the small intestine) enhances digestion of glucose polymers.


Veum and Mateo (1986) fed pigs a sucrose-added formula

but could not stimulate sucrase activity by day 7. No

appreciable amounts of the enzyme were detected until day

14. Manners and Stevens (1972) found that pigs fed an


artificial diet had higher sucrase activity than sow-fed

controls. This may suggest that diet influences the rate of

development of the intestine. Huber et al. (1961) reported

that even when calves were fed a sucrose-added diet, there

was no notable digestion of sucrose until day 44, in

contrast to their study with pigs, where sucrase digestion

was evident by day 10.

Very low sucrase activity was detected in the small

intestine of equine fetuses and newborn foals. The activity

increased slowly during the first months after birth,

reaching adult activities by 7 mo of age (Roberts, 1974).


Humans are an unusual species. For some races, the

ability to digest lactose remains long after weaning, but

the percent of the population with adult lactase activity is

small (Dahlqvist and Semenza, 1985). In humans, rabbits,

and rats prolonged feeding or absence of feeding of lactose

does not affect the postnatal changes in lactase activity

(Paige and Bayless, 1981). In calves, up to 44 d of age, no

increase in lactase activity occurs as a result of adding

lactose to the diet (Huber et al., 1961).

In human infants, lactose hydrolysis by the

f-galactosidase enzyme is affected by other mono- and

disaccharides that may be concurrently ingested (Paige and

Bayless, 1981).


Lactase activity in the neonatal piglet has been shown

to be high at birth, reaching a maximum at day 15, then

decreasing throughout the nursing period with low activities

still detectable in the adult (Paige and Bayless, 1981; Veum

and Mateo, 1986; Manners and Stevens, 1971; Dollar et al.,

1957; Dahlqvist, 1961b).

Lactase activity was detected in the mucosa of equine

fetuses as early as 105 d of gestation, increasing after the

ninth month, reaching a maximum at birth and remaining high

for 4 mo (Roberts et al., 1974). The activity declined,

especially in the duodenum, until 1 yr of age.


Several researchers have investigated maltose as an

energy source for neonates. Cunningham and Brisson (1957)

fed 1 day old piglets a purified casein-based diet with

either maltose or glucose as an energy source. In piglets 3

to 7 d old, greater than 95% of the maltose in the diet, and

greater than 99% of the glucose disappeared from the small

intestine. The glucose diet resulted in greater

digestibility for protein and carbohydrate than the maltose

diet (P < .01). When the entire small intestine of neonatal

piglets of various ages was ligated and injected with a 5%

maltose solution, .66 pmol of maltose was hydrolyzed and

absorbed per h per kg body weight. This increased to 1.05


Jmol by day 6 and 7. The authors conclude that maltose

digestion during the first wk may only be borderline in

supporting the animal's energy needs, but after day 7

maltose is used just as efficiently as glucose. Huber et

al. (1961) could not increase the maltase activity of calves

by feeding diets of whole milk plus lactose, or sucrose and

starch for 44 d. They note that this is in contrast to

other workers who have reported an increase in maltase

activity beginning in the third or fourth wk after birth.

Using intestines obtained from slaughterhouses, Roberts

et al. (1974) detected low maltase activity in equine

fetuses during the first 3 mo of gestation. Activities

remained low until birth. Enzyme activity doubled during

the first month, with the greatest increase occurring in the

proximal small intestine. Adult activities were achieved by

7 mo of age. No data were given specifically for the early

neonatal period.


Determining Disaccharidase Activity in Experimental Animals

Intestinal disaccharidase activity can be determined by

homogenizing samples of tissue from segments along the small

intestine and incubating the homogenate with disaccharide

substrates. The glucose produced is a measure of enzyme

activity. Many researchers have used this method, as

modified from Dahlqvist (1964). The major disadvantage of

this method is that the experimental animal must be

euthanized. Economic constraints usually limit the number

of foals that can be euthanized, making it impossible to

conduct large scale equine trials of this type. Therefore,

it would be helpful to have an alternate method, at least

during preliminary trials. Studies on calves have shown

oral tolerance tests to correlate well with the results

obtained using tissue homogenates when comparing relative

enzyme activities (Dollar and Porter, 1957).

Roberts (1975) used an oral disaccharide tolerance test

to determine the digestibility of various sugars in adult

horses and ponies. Although foals were not included in the

study, Roberts suggested that an oral lactose tolerance test

would be a useful tool for determining digestive competence

in the sick or premature foal. Testing the digestibility of

other sugars (for possible inclusion in artificial formulas)

in this way would allow animals to be studied without



Responses of the Neonate to Stress

Stress factors include a wide variety of situations

that can cause problems for neonates. Poor maternal

nutrition and dystocia are examples of stresses can occur

before or during birth, but this project will be concerned

with stresses that occur during the neonatal period.

Failure to Acquire Passive Immunity

In the horse and pig there is little transport of

immunoglobulins across the placenta. So, these species must

depend on colostral transfer of antibodies to protect them

from a wide variety of infectious organisms until their own

immune systems are competent. The risk is high that these

neonates will be severely stressed by infections if they are

deprived of colostrum or are unable to absorb antibodies

during the first day of life. This becomes even more

serious if the neonate is also exposed to other stresses in

its environment or is weak, sick, or immature at birth.

Many researchers have emphasized the importance of

colostrum ingestion for newborn foals and have suggested

reasons why foals may be hypogammaglobulinemic even after

they have nursed (Naylor and Bell, 1985; Naylor, 1979;

Rossdale, 1985). Their findings show that immunoglobulins

in the colostrum decline rapidly, becoming 15% of its

concentration at parturition in 4 to 8 h.


Mares may produce poor quality colostrum for several

reasons. Older broodmares may leak colostrum before

parturition and so lose the high antibody content necessary

to provide the foal with adequate protection. Some foals

are born prematurely, before the mammary gland has completed

concentrating the immunoglobulins. Foals must ingest

colostrum within 24 h of birth. After this time, the

digestive tract "closes" and will no longer absorb large

molecules. Closure is not delayed by food-withholding.

Foals may be deprived of colostrum if they are orphaned

or because of a problem with the mare or with the foal

itself that prevents it from nursing. Foals that need to be

hospitalized may be separated from their mothers. Frozen

colostrum may either be unavailable or unwise to feed,

depending on the condition of the foal. Colostrum-deprived

foals have been shown to remain almost completely

agammaglobulinemic for 2 wk, then begin rapid synthesis of

antibodies until they appear the same as colostrum-fed foals

by 5 wk of age (Naylor, 1979). These workers feel that over

50% of colostrum-deprived foals will develop an infection

resulting in septacemia, diarrhea, pneumonia,

omphalophlebitis, or arthritis within 1 wk of birth. They

suggest prophylactic treatment of 1.2 liters of plasma taken

from an adult horse, given intravenously over 2 h, to supply

an adequate amount of IgG.


Piglets are also likely to suffer similar diseases if

not provided with passive immunity from colostrum (Svendsen

and Bille, 1981; Bergeland and Henry, 1982) or kept in

isolated, aseptic conditions (Pond, 1978). In addition to

providing passive immunity to neonates, colostrum also

contains epidermal growth factors that promote mucosal

development in the gastrointestinal tract (Svendsen and

Bille, 1981). Unfortunately, it has been shown that

colostrum-deprived neonates fed either milk from other

species or soy-based formulas sustain damage to the

intestine, as the immune system responds to the foreign

proteins ingested (Weaver, 1986). The use of elemental

diets would prevent this adverse reaction as they contain

hydrolyzed rather than whole proteins.

Cold Stress

Exposure to a low ambient temperature shortly after

birth is a common problem for many neonates including

piglets, foals, and human infants. Management practices

used on many breeding farms encourage mares to begin cycling

and ovulating earlier in the year than they would in nature,

so foals are born more and more frequently in late

winter/early spring when temperatures are low. Therefore,

it is important to study the ability of neonates to regulate

their body temperature and the effects of cold stress on

their health and nutritional requirements.


Normal body temperature for foals is 38.05 C 1, with

the lower boundary of the thermal neutral zone being 140 C

during the first 48 h postpartum. After that, the foal can

tolerate an ambient temperature as low as 8 C, provided he

has nursed and the environment is free of draught (Rossdale,

1985). Premature foals often have difficultly in

maintaining their body temperature, which may drop to 36.60C

or even lower (Naylor, 1979). Foals lack significant

deposits of brown adipose tissue and must rely on shivering

for thermogenesis (Rossdale, 1985).

The normal body temperature of piglets is 390C at

birth. The thermal neutral zone is very small (34-350C)

(Stanton et al., 1973) and they have greater difficulty

maintaining their body temperature under cold stress than do

foals and calves (McCance and Widdowson, 1959).

Piglets are easily cold stressed for several reasons:

they are poorly insulated at birth, having almost no brown

adipose tissue and only 1% of their body weight (about 10 g)

is fat; they have a small body size with a large surface

area; their metabolic rate is low at birth; and they are

born with no protective hair or fur (McCance and Widdowson,

1959; Gentz et al., 1970; Curtis and Rogler, 1970; Stanton

et al., 1973). However, their ability to thermal regulate

does increase with age and is fairly well developed by 1 wk

of age (Gentz et al., 1970; Curtis and Rogler, 1970; Stanton

et al., 1973).


Piglets will experience a 2 degree decrease in body

temperature, to 370C, shortly after birth. Adverse effects

may be seen if the temperature reaches 35 C (Svendsen and

Bille, 1981).

An ambient temperature of 50C is close to the cold

limit for newborn pigs. At this temperature, there should

be maximum utilization of their thermal regulatory

capabilities. Several investigators have exposed newborn

pigs to this temperature for short periods of time. Results

suggest that starving piglets undergoing cold stress

experience hypoglycemia and will not survive longer than 30

h unless fed (McCance and Widdowson, 1959). These workers

report that at birth, pigs are able to increase oxygen

consumption during cold stress whether they are fed or not,

whereas the newborn rat will not show an increase until it

is 5 d old. Rabbit young improve their ability to respond

during the first few days of life.

Carbohydrate is the main energy source of newborn

piglets, even though sow's milk contains 30-40% fat on a dry

matter basis (Allee et al., 1971) because fat utilization is

limited during the first few days postpartum, regardless of

the ambient temperature. However, cold exposure results in

rapid depletion of energy stores and death from hypothermia

unless the piglets are fed. It is estimated that the

carbohydrate and fat stores endogenous to the piglet would


yield about 72 h worth of energy with ambient temperatures

in the thermal neutral zone (Gentz et al., 1970).

Compromised Neonates

Being born premature or dysmature will compromise the

neonate's ability to adapt to extrauterine life. Dystocia

and neonatal maladjustment syndrome (where there is reduced

ability to suck, swallow, or move about) also predispose

newborns to septacemia and in the case of foals, septic

arthritis (Koterba et al., 1984). Premature foals usually

have abnormal blood gases and an impaired acid/base status

(Rossdale et al., 1987). They often are weak and take a

longer-than-normal time to stand and nurse. They commonly

show hypoflexion of the metacarpal and metatarsal joints,

and a bright red tongue, but they may maintain normal

respiration and heart rate for the first 24 h before their

condition deteriorates (Rossdale, 1987).

Intestinal maturation will not be complete in preterm

neonates. Enterocyte function, villi development, and

enzyme activity are often different from that found in term

newborns, resulting in decreased digestibility of some

nutrients (Raul et al., 1986; Mayne et al., 1986; Mackenzie,



Hematoloqy and Blood Chemistry of Neonates


It has been suggested that evaluating the packed cell

volume (PCV), hemoglobin (Hb), red blood cell count (RBC),

white blood cell count (WBC), and differential leukocyte

count can be helpful in assessing the condition of neonates

and estimating the degree of maturity and viability of

neonates (Rossdale, 1985).

Becht and Semrad (1985) found that in foals, PCV, Hb,

and RBC counts peak at birth and then begin to decrease

within 12 h postpartum. They also noticed that mean

corpuscular volume (MCV) decreased slightly in the fetus

just before birth, then remained steady for the first 2 wk

of extrauterine life. They suggest that some of these

indices may be useful for establishing the maturity of the

foal at birth.

Several workers have reported lower leukocyte counts (4

x 109 /liter) for premature foals as compared to full term

foals (6 x 109/liter). Erythrocyte counts were also lower (6

x 1012 vs 12 x 1012/liter) as was the PCV (33 vs 45.6%).

Globulin (<10 gm/liter) and gammaglobulin (4 gm/liter) were

also lower than for normal foals (Jeffcott et al., 1982;

Kitchen and Rossdale, 1975; Rossdale, 1983; Becht and

Semrad, 1985).

Foals have an intact granulocytic system at birth. The

increase in WBC reported during the first day of life is

thought to be due to an increase in mature polymorphonuclear

neutrophils (PMN) and lymphocytes. The increase in PMN

continues during the first 3 days postpartum. During the

first wk, monocytes begin to appear. Basophils are not

normally seen in the neonatal period.

Premature foals tend to have lower WBC (4 x 109 /liter),

RBC (6 x 1012/liter), PCV (33%), globulin (<10gm/liter),

gammaglobulin (4gm/liter) and a narrower neutrophil to lymph

ratio (1.2:1) than normal foals. A severe leukopenia, left

shift, and the appearance of toxic neutrophils may indicate

sepsis (Becht and Semrad, 1985).

Jeffcott et al. (1982) and Rossdale (1983) found that

premature foals consistently show a lower than normal

neutrophil to lymphocyte ratio (N/L). Normal foals have a

ratio > 2:1 while premature foals show a reversed N/L ratio

of > 1:1. A dose of short acting exogenous ACTH will elict

a neutrophilic response in mature foals but not in immature

foals that lack normal adrenal function.

In piglets, RBC (from 6.18 x 1012/liter to 4.4 x

1012/liter), PVC (from 40 to30%), and Hb (12.5 to 10.0g/dl)

are decrease during the first wk after birth. However, the

number of WBC tends to increase from about 6.2 to 17 x

103/ sl. The proportion of PMN to lymphocytes percentages

shifts from about 38:60 to about 53:42. Granulocytes are

rarely seen during the first wk (Schmidt and Tumbleson,



Various clinical chemistry analyses are useful for

evaluating the general health and nutritional status of

newborn animals. Both foals and piglets are susceptible to

hypoglycemia shortly after birth, especially if they are

weak, sick or stressed.

Normal blood glucose in the neonatal piglet is 60 to 80

mg/dl. However, it may be as high as 100 mg/dl immediately

after birth. The decrease may be due to limited glycogen

stores in the liver (Pond, 1978). Blood glucose values as

low as 48 mg/dl have been reported for newborn pigs,

increasing to 114 mg/dl by day 7.

Foals are susceptible to hypoglycemia shortly after

birth, especially if they are weak, sick or stressed. Foals

that have suckled during the first 2 h after birth were

reported to have the following blood glucose values: 95.5 +

17.44 mg/100 ml at birth, 83 + 3.28 mg/100 ml 30 min

postpartum, and 131 + 12.4 by 14 h (Kitchen and Rossdale,

1975). Normal, healthy foals that have suckled have blood

glucose levels that are higher than adult levels for the

first 24 h and remain in the high normal adult range for the

first 30 d. Hypoglycemia may result in sick foals not

receiving sufficient nutrient intake, or from poor digestive

and absorptive function even with adequate intake (Becht and

Semrad, 1985).


Blood lipid components such as free fatty acids,

trigycerides, ketones, blood urea nitrogen (BUN), and

circulating liver enzymes are important parameters to

consider when evaluating the utilization of dietary fat and

nitrogen status in the neonate.

Milk Composition and Intake

Some researchers, using milk replacers currently

available for foals, have expressed dissatisfaction with the

foals' growth rate and general appearance. They report that

foals fed milk replacers according to directions usually

remain small (Naylor and Bell, 1985). Since no other farm

animal species produces milk similar to that of the mare,

finding a suitable formula is more complicated than just

substituting cow's milk, for example, for mare's milk.

However, goat's milk has been used successfully to raise

orphan foals. It is higher in fat and protein and lower in

carbohydrate than mare's milk, but foals drink it readily

and thrive on it, although some may experience diarrhea

(Koterba and Drummond, 1985). Interestingly, goat's milk is

similar to sow's milk, so perhaps this would further suggest

that piglets would be good models for foals.

Even though goat's milk and colostrum contain slightly

more protein and much more fat than that of the mare, foals

have been successfully raised at the veterinary teaching

hospital at the University of Florida on goat's milk

(Koterba and Drummond, 1985). Goat's milk is high in fat

and is similar to sow's milk (Glauser, 1966). Also, many

commercially available milk replacers for foals are fed at

twice the recommended concentration, which significantly

increases the fat intake (Naylor and Bell, 1985).

The compositions of colostrum and early lactation milk

from mares and sows are compared in Table 1.1. Ullrey et

al. (1966) and Pagan and Hintz (1985) reported data on

equine milk constituents, while Widdowson (1984) and

Widdowson (1985) reported on both sows and mares. While

similar in protein, sow's milk has higher total solids,

higher fat, and lower lactose than mare's milk.

Milk Production in Mares

Compared to other species, mare's milk is unusually

high in water. To compensate for this, foals must consume

large quantities of milk and have a rapid body water

turnover rate (the halflife of body water for foals under 1

wk old is 2.5 d). From birth to 11 d, foals will usually

drink about 16.2 kg of milk per day, ingesting about 422 g

protein and about 9830 kcals (Palmer, 1985).

When mares were fed iso-caloric diets that differed

only in the proportions of fiber and fat, milk compositions

did not differ (Pagan and Hintz, 1986). Mares fed diets

that contained 1.25 times the National Research Council

Table 1.1. Comparison of cslostrum and milk produced by
sows and mares .

Mi 1 'Ic Ci-~rt~4- i i-11~vb4-

M2 ~

.. Ile nn-.. .-+- w ar *So"*w

% Dry Matter
% of Dry Matter:

Colostruim Consti tient

(gm/10Oml milk)








aWiddowson, 1984
bUllrey et al., 1966

Coosru CositetWW

<"*^ .

<" _- -


(NRC) recommendations for energy requirements for lactation

produced greater volumes of more dilute milk than did mares

fed adequate calories. The growth rates of the foals were

not different for the two groups. The authors concluded

that feeding excess energy does not improve foal

performance, and results in obese mares.

Special Nutrient Requirements of Neonates

It is is important to determine the special nutrient

needs of neonates when developing a suitable enteral

formula. Premature neonates, for example will have

different nutrient requirements from healthy, full term

newborns (Koterba and Drummond, 1985). These authors

suggest that besides the major nutrients, requirements for

folate, Vitamin E, cysteine, calcium, and phosphorous

requirements for premature foals need to be evaluated. They

have found that stressed foals require about 120 kcal/kg/d

and may be expected to consume 20-28% of their body weight

(BW) in milk per day. They also recommend feeding an

elemental type of diet, because stressed newborns may not

tolerate milk-based diets which can cause bloat, colic, and

diarrhea. Elemental diets may prove to be a good

alternative to feeding cow's milk or soy-based formulas that

have been shown to have damaging effects on the intestinal

villi of infants (Weaver, 1986).


By examining prepartum mammary gland secretions, we may

be able to determine which nutrients are of special benefit

to premature neonates. In humans, the mammary gland

produces considerable amounts of medium chain fatty acids

(C8-C14) after parturition (Bitman et al., 1986). Linoleic

(18:2) also increases when compared to prepartum mammary

gland secretions. The secretion of several long chain

polyunsaturated fatty acids (PUSFA) decreases. Prepartum

secretions contained large amounts of C16 and C14 when

compared to normal milk. After parturition, the milk

contained increasing amounts of short and medium chain

length fatty acids (4:0, 6:0, 8:0, 10:0) as well as 18:0 and

18:1. Similar patterns of increased lipid synthesis and a

shift towards medium-chain fats after parturition were also

observed in cows. These authors suggest that the increase

in MCT in the milk may be of some advantage to preterm

neonates whose capacity for lipid digestion may not be

completely matured.



As recently as 20 yr ago, premature or sick human

infants were not expected to survive the neonatal period and

often were allowed to starve to death (Koterba et al.,

1985). Today, however, the advent of improved respiratory

therapy, refined surgical techniques, neonatal intensive

care facilities, and specially trained personnel gives the

very low birth weight or compromised infant a dramatically

improved survival rate. Once these infants could be kept

alive for longer periods of time, the question of nutrition

became important.

The field of equine neonatology, which has expanded

considerably over the past few years, has benefitted from

the advances made in human neonatology. Today, premature

and sick foals are treated successfully using much of the

same equipment and techniques developed for infants. Many

of these foals have gone on to achieve success on the race

track and in the show ring. Veterinarians are now faced

with the same questions faced by neonatologists treating


infants. Now that life threatening infections and

respiratory problems can be overcome (Webb et al., 1984;

Baker and Ames, 1987) what type of nutrition is appropriate

for these patients?

Ethical considerations make some types of research

impossible on human infants. Animal models have proven

useful in situations where invasive techniques or use of

radioisotopes is necessary. It was suggested that neonates

of different species may be more similar to each other than

they are to the adults of their species (Koterba et al.,

1985). Therefore, it is possible to use information gained

from work in one species for use in another. For example,

Rossdale (1987) used the human neonatology terms premature,

dysmature, and small for gestational age, to develop

guidelines for assessing readiness for birth in newborn


Although the elemental, enteral diets used in this

trial was designed originally for use in humans,

scientifically controlled trials on neonates were needed.

It was thought that this formula might be suitable for

premature or sick foals which were unable to drink mare's

milk. These compromised neonates may have poorly

functioning or damaged intestinal tracts.

Nutrients in enteral diets are delivered to the

digestive tract usually by mouth or by catheters inserted in

the stomach or small intestine. In contrast, parenteral


formulas are delivered intravenously. In an elemental diet,

individual nutrients are added in easily digestible forms,

in contrast to conventional, milk-based formulas. The base

formula tested in this study contained hydrolyzed whey

proteins rather than whole milk proteins. The formula was

lactose-free because, as stated previously, lactase may be

deficient when there is damage to the intestines. Milk-

based formulas contain long chain lipids almost exclusively,

but this formula included some medium chain fats. This was

because the digestion, absorption, and metabolism of long

chain fats may be inefficient in compromised neonates.

Piglets were used as a model for the preliminary

research in this trial. The information gained from these

trials would then be used as the basis for studies on foals

and infants. Colostrum-deprived piglets were used for

several reasons. If the piglets were allowed to nurse the

sow, it would not be possible to ensure that each piglet

received the same amount of colostrum. By hand-feeding the

piglets from birth, the amount and type of each nutrient

consumed can be controlled. Many sick neonates do not

receive colostrum shortly after birth and so are deprived of

its growth-promoting factors. The colostrum-deprived piglet

model permits the formula to be tested on stressed animals.

The restrictions and limitations of research projects with

foals and infants encourages the use of an appropriate

animal model for all preliminary work.


This study was part of a multi-project program to

develop an enteral diet for premature or sick foals.

Designing such a diet is a complex task requiring a series

of separate trials to determine the ideal formulation.

The goals of this specific study were:

1. to learn techniques and methods used in the investigation

of neonatal nutrition using a piglet model;

2. to determine the effect of an elemental, enteral diet on

the growth, intestinal development and general health of

colostrum-deprived neonatal piglets and, secondarily, to

learn to formulate diets with different lipid sources;

3. to evaluate and improve surgical and daily care

procedures for colostrum-deprived piglets hand-raised for 7

d; and,

4. to use the information obtained from the piglet study

towards possible development of a diet for neonatal foals.

Materials and Methods


This study was conducted using newborn Yorkshire x

Hampshire x Duroc crossbred piglets. The sows were bred and

cared for by the staff of the Reproductive Physiology Unit

under the direction of Dr. Fuller Bazer of the Animal

Science Department. Labor was induced 2 d prior to the

expected farrowing date to ensure that the birth would be

attended. The sow was injected with Prostaglandin F2a at


730 h on d 111 of gestation. Oxytocin was administered at

900 h on d 112. Farrowing usually occurred within several

hours after the oxytocin injection. All piglets were

removed from the sow immediately after birth, before

nursing. The umbilical cord was tied and cut. Identifying

markings, sex, weight, and general condition were recorded.

Only piglets greater than 750 g were admitted to the trial.

The protocol that subsequently was followed depended on the

treatment assignment of the litter.

Piglets from a total of 10 different litters were used

for the trial. Only one litter was on trial at any one time

due to time, labor, and facility constraints.

Baseline values were obtained from 18 piglets from two

litters removed to the animal facilities at the Food Science

and Human Nutrition Building and euthanized within a few

hours after birth. These non-fed piglets are referred to as

birth piglets (BO).

Normal values for sow-fed animals (SF) were obtained

from 22 piglets from two litters were returned to the sow

and allowed to nurse for 7 d, after which they were

euthanized. The sow and piglets were housed in an

environmentally controlled room. The sow was confined to a

farrowing stall of pipe construction. The piglets could

enter the stall easily under the bottom pipe or retreat out

of the stall to a corner of the room to avoid the sow.

Rubber floor mats were used to reduce conductive heat loss

and improve footing. Clean hay was provided for bedding in

a corner of the room where the piglets often slept. A heat

lamp provided additional warmth. The SF were monitored and

weighed daily.

Thirty-eight experimental piglets from six litters were

transported to the Piglet Neonatal Intensive Care Unit

(PNICU), an environmentally controlled room (37*C, 70%

humidity) similar to hospital nurseries designed for human

neonates. These colostrum-deprived piglets were hand-raised

in individual plexiglass boxes (measuring 14" x 14" x 17")

in the PNICU for 7 d.


All experimental piglets except those bottle-fed the

milk-based diet underwent surgery. Umbilical artery and

bladder catheters were inserted to permit collection of

blood and urine, respectively, during the trial. The

samples were used to evaluate the health of the piglets.

Gastric catheters were inserted to permit feeding by

constant infusion pump.

Development of anesthesia procedures. Piglets were

anesthetized for surgery with intramuscular injection (75

mg/kg BW) of ketamine HC1. Lidocaine was given

subcutaneously at the incision site. Once an intravenous

(i.v.) catheter was placed in the umbilical artery, an

Acepromazine drip was started at 60 drops/min. Full


recovery from this anesthesia was slow, delaying the

initiation of enteral feeding.

In an effort to decrease recovery time, the last two

litters on the trial were given the inhalant isoflurane by

mask as the sole anesthetic. Schieber et al. (1986) found

that although isoflurane caused a reduction in blood

pressure, peripheral resistance was reduced, preventing a

decrease in cardiac output. These workers suggest that

isoflurane, well tolerated by newborn piglets, has a clear

advantage over the use of halothane. The piglets in the

present study tolerated isoflurane well, recovering quickly

(some were quite active within minutes of removal of the


Catheter placement. Each piglet was fitted with an

umbilical artery catheter for administration of i.v. fluids

(.45% saline and 5% dextrose) during and after surgery, and

for collection of blood on d 3 and d 5 of the trial. The

piglets received the i.v. fluids until they were fully

recovered from the anesthetic and enteral feeding was

initiated, at least 12 h post-surgically.

Piglets also were fitted with bladder catheters for

urine collection. The ureters were left open to permit

normal urination.

The introduction of gastric catheters and the use of

continuous infusion pumps allowed enteral feeding to be


All catheters were externalized through a single

abdominal incision.

Post-operative care. After surgery, the piglets were

returned to their plexiglass boxes in the PNICU and observed

closely. A chart was kept beside each box for recording

data from individual piglets. Respiration, temperature,

color, activity level, and urine output, were recorded every

30 minutes until the piglets had recovered fully.


Hand-raised piglets were fed one of three enteral

formulations. A conventional, cow's milk-based formula was

bottle-fed to a total of 12 piglets from two litters. These

piglets, designated Group C, did not undergo surgery. In

addition, four piglets were fed this formula via gastric

catheter. Previous trials had shown that the growth rate of

piglets fed this formula was similar to the rate of sow-fed


The two other diets used in this study were based on

the liquid elemental formula Peptamen (Carnation). All

piglets fed diets containing the elemetal formula were

designated Group E. Hydrolyzed whey proteins provide amino

acids and malto-dextrin is the carbohydrate source. The

formula was fed in two different, isocaloric forms based on

the predominant fat source. These were designated as MCT

(medium chain triglyceride) or LCT (long chain triglyceride)


the predominant fat source. These were designated as MCT

(medium chain triglyceride) or LCT (long chain triglyceride)

formulas. Because Peptamen is much higher in carbohydrate

than sow's milk, it was diluted to one-half concentration

before use. However, this necessitated the addition of whey

(27.4 g/liter for the MCT formula and 35 g/liter for the LCT

formula), calcium citrate (4.38 g/liter), and calcium

phosphate (5.77 g/liter) to obtain a composition similar to

sow's milk. The wrong type of whey mistakenly was added to

the formulas fed to one of the litters, invalidating the

data obtained from these pigs. The diets resembled sow's

milk in the amount of fat (60 g), protein (75 g), and

carbohydrate (63.5 g) provided per liter. Carnitine was

added at 486 pl/liter of formula. The MCT formula was

obtained by adding 15 g medium chain triglyceride oil (MCT

Oil, Mead Johnson and Co., Evansville) and 6 g sunflower oil

(SFO) to 1 liter of Peptamen. All of the lipids in SFO are

long chain fats. The final MCT:LCT ratio was 70:30. The

LCT formula required the addition of .8 g MCT oil and 51.8 g

SFO to 1 liter of low-fat (LF) Peptamen. This resulted in a

90:10 LCT:MCT ratio. The composition of sow's milk and the

dietary treatments used in this trial are presented in

Tables 2.1-2.2.













p -4




0 -4

OH 0

0 0


uo -


















0 0
tD %D

Ln Ln
rN N-

0 0

C0o 0 0

4 E E-1
h 04Nb


H l 0

N o

0 0
* .



N m


9 n
o a%

Ni co
CT 0
r-H LA

co N
*s** H

SIo a
\o -^

C Hr
qw Ln
In m

1 o'
tp 0 %

Nc 0
N- W

c H

6 l
a i

04 '-

. .

ra ,1

EP E-4

0 a

4 E
0 0
44 4-

O 0

0 0
0 4

0 0

(0 (

Table 2.2 Composition of sow's milk (Pond and Houpt, 1978)
and Peptamen (Carnation) on a per liter basis.

Nutrient Sow's Milk Peptamen

Protein, g

Fat, g

Carbohydrate, g

Folic Acid, g

Thiamin, g

Riboflavin, g

Niacin, mg

B6, 9

B12, g
Biotin, g
Acid, mg
Vitamin A, IU
Vitamin D, IU
Vitamin E, IU
Vitamin C, mg
Calcium, mg
Phosphorus, mg

Potassium, mg
Sodium, mg
Magnesium, mg
Chloride, mg
Iron, mg
Zinc, mg
Manganese, mg
Copper, mg
Iodine, mg
Energy, kcal




1370 -
4.3 9.0

500 8500
1000 -





6 x 10-6




Development of Feeding Regimen

The elemental diets were fed by bottle to the first

litter of piglets that received these formulas. Subsequent

piglets fed these diets were fed via gastric catheters. The

formulas were fed to the bottle-fed piglets at the rate of

15 ml/h on d 1 and 35 ml every 2 h on d 2. The rate was to

be increased by 5 ml/feeding on each successive day,

remaining at 50 ml/2h on d 6 and 7. Because the piglets in

this litter suffered from severe, watery diarrhea, the

formula was changed to Peptamen alone, at two-thirds

strength on d 5. Subsequent litters of Group E piglets, fed

via gastric infusion, received 10 ml h-1 kg BW-1 on d

1. On each successive day, the amount fed to healthy

piglets was increased by 3 h-1 kg BW-1. The flow rate

was not increased for piglets that were not thriving. The

last two litters were fed the MCT and LCT diets at half-

strength. This decreased the production of watery feces.

With ketamine anesthesia, long post-surgery recovery

times delayed oral feeding. These piglets were often slow

to show normal sucking and swallowing behavior after being

anesthetized. Placement of a gastric catheter during

surgery allowed commencement of enteral feeding much sooner

than was possible with bottle-feeding. Enteral nutrition

was initiated on an individual basis when the piglet

appeared to be awake and kicking motions of the hindlegs

were observed. Usually, these animals remained on i.v.


fluids post-surgically for about 8 to 12 h before receiving

enteral nutrition.

To determine the effect of PNICU conditions on piglets,

the Group C controls were bottle-fed a milk-based formula

designed to be similar to the composition of sow milk. To

determine the effect of surgery on PNICU piglets, four

animals (surgery controls) were fed the milk-based diet via

gastric catheters using a continuous infusion pump. The use

of infusion pumps reduced the variation in post-absorptive

state that might occur when the animals are euthanized at

different times.

In summary, a total of 13 Group E piglets were fed the

elemental diets. Four piglets were bottle-fed, while the

rest received enteral nutrition via gastric catheter. Two

piglets were fed the base formula, Peptamen, with no

additions, via gastric catheter, to determine if the

additions of calcium and phosphorus were contributing to the

diarrhea. Of the 16 piglets fed the milk-based control

formula, 12 were bottle-fed and four underwent surgery for

gastric catheter placement and were fed via constant

infusion pump.

Daily Care

The piglets were monitored continuously throughout the

trial. Urine was collected asceptically from the bladder

catheters every 30 min. At morning and evening "rounds" the

incision site was checked and swabbed with an iodine

solution, the venous catheter was flushed, bandages changed,

and temperature, respiration, and girth measurements were

recorded. When piglets were fed via gastric catheter, new

bags of formula were started after rounds. Piglets were

weighed during morning rounds. On d 3 and d 5 blood was

collected from the umbilical catheter.

Every effort was made to keep stress on the piglets to

a minimum. Each piglet was provided with a rubber baby-

bottle nipple taped to the inside of the box, a plastic

ball, and a soft cloth.

Tissue Collection

At sacrifice, the animals were anesthetized with Na

pentobarbital and then exsanguinated by cardiac puncture.

Organs were removed as quickly as possible, weighed, and

frozen for later analyses. The intestines were measured,

weighed, cleaned of contents, and then re-weighed. The

small intestine of each piglet was divided into three

segments of equal length, designated proximal (top)

intestine (TI), middle intestine (MI) and distal (bottom)

intestine (BI).


Leukocyte (WBC) counts were performed using a Unopette

dilution chamber and a hemocytometer. Microhematocrit (Hct)

determinations performed shortly after blood was collected.

Hemoglobin (Hb) was determined using the modified

Drabkins method with a Beckman DU-7 spectrophotometer at a

wavelength of 540 nm.

The plasma glucose was assayed using Trinder reagent

(Sigma Chemical Co., St. Louis), based on the glucose

oxidase reaction. The OD was read with the Beckman DU-7

spectrophotometer at 505 nm.

Non-collagen protein in intestinal homogenates was

determined using the micro-Biuret method (Itzhaki and Gill,

1964) after precipitation with .2 N NaOH. The OD was

measured with a Gilford spectrophotometer at 310 nm.

The activities of the intestinal disaccharidases

maltase, lactase, and sucrase were determined using a

slightly modified technique by Dahlqvist (1964).

Preparation of the tissue included homogenization in a

glass-on-glass hand homogenizer with .05 M sodium phosphate

buffer in a 1:5 weight/volume ratio. Each tissue homogenate

was incubated with each substrate to determine the

micromoles of glucose produced by the disaccharidase enzyme

per microgram tissue, during a 1 h incubation. The OD due to

the glucose produced was read at 505 nm with the Beckman DU-

7 spectrophotometer.


Urine was collected from Group E piglets every 30 min

throughout the trial. All of the urine collected during a 6

h period was pooled, unless it was unusual in appearance.

In this case, the abnormal urine was placed in a separate

vial. The urine was kept on ice or refrigerated until


The volume of urine collected as well as a description

of its general appearance were recorded for each piglet for

each 6 h period. Fresh urine was analyzed using Ames

Multistix SG reagent strips (Miles, Inc., Ames). These

strips provided a quick assessment of specific gravity, pH,

protein, glucose, ketones, bilirubin, blood, and urobiligen.

The remaining urine was frozen, pending further analysis.

The following equation was used to calculate the % BW

gained by the 7-d old piglets:

% BW wt gain = ((d 7 BW birth BW) birth BW) 100.

The percent of body weight of each section of the small

intestine (TI, MI, and BI) was calculated using the

following equation: SI section wt(% of d 7 BW) =

(SI section wt d 7 BW) 100. These results will be

referred to as %TIWT, %MIWT, and %BIWT.

Additional Information

Additional information and details of surgery, daily

care, necropsy procedures, and tissue analyses are described

by Baltzell (1988). The PhD dissertation involved the the


colostrum-deprived neonatal piglet model that was further

developed in this current work.

Statisical Analysis

Treatment means for SF and BO data only were obtained

by method of least squares ANOVA (Snedecor and Cochran,

1969) using the GLM procedures of the SAS Statistical

Software (Freund and Littell, 1981). The model included

treatment, litter(treatment), sex, sex*treatment, and

sex*litter(trt). Data from Group C and Group E treatments

were not analyzed because of the low numbers of animals in

each group.


Numerical results are expressed as least squares means

SE, unless stated to be observed means.

General Observations

Sow-fed piglets. All of the SF piglets were active,

healthy and growing well at the end of the 7 d trial period.

They did not seem unduly stressed by the daily weighing


PNICU piglets. Thirty-eight piglets were started on

trial. All of the Group C piglets were healthy and active

through by d 7. Although the elemental diets were

formulated to be similar to sow milk, the Group E piglets

did not thrive. These piglets experienced diarrhea that may

were made for each litter. These changes reduced but did

not eliminate the diarrhea. The survival rate of Group E

piglets was low, as only nine of 20 survived for 7 days.

This was probably due, in part to diet and to dislodged or

faulty catheters. Piglets that had dislodged gastric or

bladder catheters were humanely euthanized. Because of the

low survival rate of the Group E piglets, only descriptive

results will be presented.

Occasionally, piglets were born with umbilical hernias.

This was not generally not a problem, except for one piglet,

who diet after a portion of the small intestine

strangulated. Blockages or strictures of the small

intestine were observed in four animals.

Behavior of PNICU piglets. The hours of care required

by the piglets allowed for close observation of their

behavior. Each piglet had a particular activity pattern

that was known to all the caretakers. Therefore, early

detection of problems was possible.

The non-surgery piglets, bottle-fed the milk based

diet, quickly became accustomed to the routine in the PNICU.

They were noisy and wanted attention from the caretakers.

Even during the first day, they began to recognize the pre-

feeding routine. They had strong suckling reflexes. Play

was part of their daily activity, just as it was for the

sow-raised piglets. Although they were kept in individual

boxes, the plexiglass allowed them to see each other and the


boxes, the plexiglass allowed them to see each other and the

caretakers. Objects such as crumpled paper towels, plastic

balls, and pacifiers were used as toys.

The healthy surgery piglets also exhibited play and

attention-seeking behavior and did not seem hampered by the

i.v. or enteral infusion lines attached to them. Soft

cloths were used as bandages to cover the incision and keep

lines in place. These were ignored by the animals. Swivels

in the infusion lines kept the twisting and tangling to a

minimum. These lines were changed twice daily, when fresh

formula bags were set up.

Activity level (sleeping, lying down, alert, moving,

etc.) of each piglet was recorded on the individual charts

every half hour and changes from the previous record noted.

Nasal oxygen was available, if needed. An attempt was made

to cool (water or alcohol swabs) or warm (heating pad or

heat lamp) animals whose body temperatures were not in the

normal range.

Healthy animals generally produced an adequate amount

of relatively clear urine. Animals that had decreased urine

output or bloody urine generally had enlarged kidneys on

necropsy and often had fluid accumulation in the abdomen and


In an effort to improve the survival and growth rate of

the piglets fed the elemental diets, modifications in the

formulas and daily intakes were made for each successive


litter. After the first litter, the calcium and phosphorus

additions were discontinued. The amount of whey added to

the base formulas was decreased. The formula strength and

flow rate during the first few days after surgery were

decreased. It was thought that these changes might decrease

the diarrhea, improve tolerance of the formula, and smooth

the transition to enteral nutrition after surgery.

Gross Observations at Necropsy

Sow-raised and BO piglets. There were no abnormalities

noted in the organs of any of the SF or the BO piglets. The

intestinal tissue of the BO piglets was very thin compared

with that of the SF and could be easily torn when separating

the mesentery from the small intestine.

The stomachs of the BO piglets essentially were empty,

while the stomachs of SF piglets contained milk curd. The

proximal intestinal contents of the SF piglets were very

liquid and yellowish. The contents became thicker in the

more distal areas. The cecal contents were pasty and darker

yellow. Large intestine contents were very thick and became

darker in the more distal regions. The distal colon

contained formed fecal pellets, dark in color.

Post-surgical, formula-fed piglets. In all cases, the

positions of the catheters were verified on necropsy. No

abnormalities, other than occasional adhesions, were noted


with the catheters. The lining of the stomachs and bladders

did not show irritation from the catheters.

Some piglets exhibited respiratory distress and had

reddish, congested lungs at necropsy. Some of these animals

had displayed swollen shoulders, hindlegs, and abdomens with

bluish blotches seen beneath the skin. While cause of these

blotches is unknown, it may have been due to caretakers

holding the piglets too tightly during rounds.

Many animals with bladder catheters had enlarged

kidneys. In some cases, the ureters also were enlarged.

The livers of all of the piglets appeared normal, but

several piglets had gall bladders that contained a dark,

thick material. This was not seen in the SF or BO groups.

Non-surgical, formula-fed piglets. The Group C

piglets, fed the conventional, milk-based diet, were healthy

and active at the end of the 7 d. The organs of these

piglets appeared normal at necropsy.

Some piglets bottle-fed the elemental formulas had

slightly reddish, congested lungs.

Weight Gain

The means reported in this section are observed means +


Sow-fed litters. The sow-fed piglets (SF) increased

their birth weight by 65.75%. One litter (n=13) gained an

average of 74.49% 4.24, with a minimum of 50.76% and a


maximum of 115.18%. The other litter (n=9) gained an

average of 53.13% 5.90. The least amount gained was

27.58%, and the most 79.16%.

Formula-fed piglets. The birth weight of Group C

piglets (n=12) increased by an average of 78.56% during the

week long trial. The smallest gain was 54.40%, while the

largest was 96.41%. The surviving surgery controls, fed the

milk-based diet via gastric catheters (n=2), gained only


One piglet fed one of the MCT-based diets gained 5.63%

of its birth weight and one piglet increased its weight by

21.91%. The other piglets experienced losses of 6.89%,

11.09%, and 8.5%. The MCT-fed piglets were all very thin by

the end of 7 d. None of the four surviving piglets fed

one of LCT-based lost weight, but they gained very little.

The average gain was 11.99%. The two bottle-fed piglets

gained 6.65% and 1.85% of their birth weight. The other

two, fed by gastric infusion, gained 35.29% and 4.18%. All

but two Group E piglets weighed greater than 1 kg at birth,

and these exceptions were over 900 g.

Small Intestine Length

The total length of the small intestine (SI) was

measured at necropsy. The length of the SI in the BO

piglets was used as a reference baseline value.


The least squares means (LS means) for one litter of BO

and one litter of SF piglets were 284.39 cm 11.27 and

371.73 cm 9.34, respectively.

Blood Glucose

Blood glucose may be expected to vary depending on an

animal's nutritional status, health, and physical condition.

The use of infusion pumps may have reduced variations due to

differences in times after feeding at death. This was

difficult to control with bottle feedings due to time and

labor constraints. The results in this section are reported

as observed means unless stated otherwise.

The LS means of blood glucose concentration of two

litters of SF piglets and one litter of BO piglets were

13.37 mM 1.81 and 6.55 mM 2.03, respectively.

Small Intestine Weight

The %TIWT, %MIWT, %BIWT observed for BO piglets, and

sow-fed piglets are presented in Table 2.3. The BO mean was

obtained from data from one litter and the SF mean was

obtained from two litters.

Disaccharidases in the Small Intestine

Results of the disaccharidase analyses are calculated

based on the amount of glucose produced by 1 g of tissue as

a result of substrate hydrolysis occurring during a 1 h


incubation period. The means for the enzyme activities for

SF and BO piglets are presented in Tables 2.4-2.6. The

results for the SF and BO piglets are from one litter of

each group. Therefore, the standard errors in the tables

also include litter effects.

Group C piglets (n=12), fed a milk-based formula, and

the Group E piglets had very low intestinal sucrase

activities. In tissue samples from several Group E piglets

no sucrase activity was detected.

Maltase activity was detected in Group E piglet

intestines, as well as in tissue obtained from Group C

piglets even though the Group C piglets had not ingested

maltose. Similarly, lactase activity was detected in the

Group E piglets that had never ingested lactose.

Table 2.3. Least squares means for the proximal, middle, and
distal sections of the SI, expressed as a
proportion of body weight.


Sow-fed (SF) .83 .03 .84 .03 .87 .03

Birth (BO) .58 .04 .63 .03 .62 .04


() 4-T



c 0


H 0

(0 (.


4-1 .4.


En *4

a) -
0 04 -1
01 0 H

OI -) 4

0) a 0

En (
0 *












O <
4 1



o co

M to>


i. c.


co rz4

* c



0 t.

*14 r-i

O 1
c I

e 0



0 0

)C 0

0 H

U 0


0 *)
a 1

W-4 r


^ <(- r
(0 0 *
O ^-

0) (U *














4* \


0 0
' -
w o



o FrA
( U




cI c

x 0
0 U)
z 0
6 a


-4 I

40 4-
0 0


C 0

41 1-I a)

>4 (n 0

U 0m


n -
fl *-
3 ^




> -4
41 0







-4 44

o m

Complete Blood Counts

The complete blood counts (CBC), consisting of Hb, Hct,

and white blood cell counts, were performed to determine if

the blood components measured fell within published normal

ranges for piglets of this age. The CBCs provided a method

of monitoring the general health of the piglet during and

after 7 d on the trial. However, it was difficult to

collect blood from the umbilical artery catheter on d 3 and

5 of the trial as the catheters often became blocked and

only a very small amount of blood, if any, could be

withdrawn. Not enough data was collected on these days to

provide meaningful results. Data from SF and BO piglets is

presented in Table 2.7.

Table 2.7. Results of CBCs from blood collected at
necropsy via cardiac puncture from sow-fed
piglets and piglets killed at birth.

Treatment n a Hb Hct WBC
(g/dl) (%) (cells/ml3)

SF 22 5.6-10.5 18.7-35.3 7000-18,700

BO 9 3.3-20.5 10.3-42.2 4050-10,100

sow-fed 31 3.6-5.2 24.1-34.9 8900-12,700

sow-fed 70 12-12.7 39.6-43.5 6270-17,600

number of samples obtained

bSchmidt and Tumbleson, 1986

cd 28 data


After surgery to introduce a bladder catheter, the

piglets routinely had some amount of blood in the urine.

Piglets with low urine outputs tended to have urine than was

visibly bloody, while piglets with substantial urine output

tended to have clear, yellow urine that was found to contain

blood only by chemstrip analysis.

The urine routinely contained varying amounts of

protein. This also may be expected after surgery. The pH

of the urine varied among piglets, but an individual piglet

tended to have urine in a small pH range throughout the

trial. The apparent health of the piglets did not seem

related to the changes in urine pH or specific gravity.


The purpose of the work using piglets was to learn to

develop techniques and procedures to be used in a trial

designed to evaluate an elemental, enteral diet for sick or

premature neonates. To achieve this, two elemental formulas

and one milk-based formula were fed to the neonatal

colostrum-deprived piglets for 7 d. The elemental formulas

differed in the amount of fat provided by LCT and MCT

lipids. The elemental diets contained malto-dextrin as the

carbohydrate source, rather than lactose because compromised

neonates may have reduced lactase activity.


The animals in the first litter were bottle-fed the

elemental diets. None of them appeared to thrive during the

7 d study period, probably due to diarrhea and respiratory

problems. All subsequent Group E piglets were fed via

gastric catheters.

The diets were modified for each litter in an attempt

to develop a formulation suitable for the piglets. Changes

in the diet and flow rate reduced but did not eliminate the

diarrhea. Some workers had suggested that maltase may be

low in neonatal piglets. Therefore, it was likely that

undigested maltose reaching the large intestine may have

caused the diarrhea. This led to the investigation of the

carbohydrate source, maltose.

Disaccharidases in Neonatal Piglets

Human infants have adequate maltase at birth to digest

maltose even when they are born prematurely (Paige and

Bayless, 1981). However, there are conflicting reports in

the literature as to when maltase becomes active in the

newborn piglet. Very low maltase activity activities have

been detected in the small intestine of piglets as early as

2 d of age (Dahqvist, 1961a), and the activity appears to

increase from d 3 to 10 (James et al., 1987). The maltose

fed to the Group E piglets was within the amount reported to

be digested by normal, sow-fed piglets, but it is possible


that these colostrum-deprived piglets lacked sufficient

total tract maltase activity to meet their energy needs on a

maltose-containing diet.

Investigation of the tissue disaccharidase activities

in the small intestine of the experimental piglets was

undertaken to determine the ability of newborn and 7-d old

piglets to digest sucrose, maltose, and lactose.

Lactase activity was measured for reference as this

enzyme should be very active during the first week after

birth. Sucrase also would be a useful reference because

only low activities have been reported in 1-wk old piglets

(Veum and Mateo, 1986).

The large variation in disaccharidase activities

between piglets in this trial was also reported by othe

workers (Manners and Stevens, 1972). It is difficult to

interpret the data when the large variation is coupled with

low numbers of animals per treatment.

The lactase activity along the SI was similar for the

BO and SF piglets. This is to be expected, because mammals

must be prepared, at birth, to digest lactose. The Group E

piglets, although not thriving, did have lactase activity,

even though they never consumed lactose. The intestine

remained prepared for the ingestion of this sugar throughout

the week.

The BO piglets had low maltase activities, leading to

the conclusion that piglets may not be able to digest


maltose efficiently enough to meet their energy needs when

this sugar is the sole carbohydrate source. The sow-fed

piglets appeared to have a numerically higher maltase

activity than did the piglets killed at birth even though

the SF piglets had never ingested maltose. This suggested

that age is a factor in enzyme development. Birth and

feeding, especially of colostrum, have been shown to

stimulate enzyme and intestinal development (Widdowson,


Sucrase activities were very low, compared with the

other disaccharidases measured. Dahlqvist (1961a) found no

appreciable sucrase in 2-d old piglets, but Bellis (1957)

reported equal digestion of lactose and sucrose by d 3.

Manners and Stevens (1972) were barely able to detect

sucrase activity until the piglets were 1 wk old. James et

al. (1987) found that EGF increased sucrose in neonatal

piglets. EGF is secreted in colostrum. In the present

study, only the SF ingested colostrum and this may explain

the generally higher enzyme activities measured in this


Growth Rate in Neonatal Piglets

The Group C controls grew at a rate comparable with,

that of the sow-fed piglets. While the hand-raised piglets

did not have to expend energy to maintain body temperature

or compete for food, they did not have the benefit of

colostrum or sow's milk. Even though the growth rate of

Group E piglets was poor, the growth rate of the Group C

piglets indicated that colostrum-deprived piglets can be

raised successfully raised for 7 d in an intensive care

facility. The unit used in this study was kept extremely

clean but not sterile.

Development of the Small Intestine in Neonatal Piglets

It was suggested that the ingestion of colostrum and

milk affect the growth and development of the small

intestine in newborn mammals (Widdowson, 1985). An ideal,

enteral formula should supply adequate nutrients and growth

factors, especially to the small intestinal tissue. The

piglets killed at birth had very small, fragile intestines

that could be distinguished easily from the intestines of

older, fed piglets. By 1 wk of age, the small intestines of

SF piglets were longer and heavier than the BO intestines.

Because these weights are on a wet tissue basis, it is not

known whether the heavier weights are due to fluid or dry


Blood glucose

Although blood glucose is regulated by insulin and

glucagon, it can be affected by nutritional state, general

health, and time after feeding. Piglets are born with low

energy stores that must be replenished shortly after birth.


Blood glucose was low in BO piglets. These piglets often

were not sacrificed for several hours after birth, so

partial depletion of their energy stores probably occurred.

Normal blood glucose for neonatal piglets has been

reported by Tumbleson and Schmidt (1986). The values had

large standard errors, which is expected as glucose varies

with time after feeding. They reported mean values for

birth and 1-wk old piglets of 2.67 1.06 and 6.67 1.11

mM/L, respectively. The birth value is lower than the 6.63

1.54 mM measured in piglets killed shortly after birth in

this study. All of the Group E piglets had blood glucose

concentrations similar to the reported values for 1-wk old

piglets. However, the 1-wk old SF and Group C piglets had

concentrations averaging about 13 mM. Method of blood

collection, stress, nutrition, breed effects, and time of

sampling have been known to affect results (Tumbleson and

Schmidt, 1986).


Results of the CBCs were compared with values reported

by Schmidt (1986). Hemoglobin in neonatal piglets was

reported to be 12.5 .2 g/dl at birth, and 10.0 1.8 g/dl

by d 7. The Hct values at birth are 40.3 .7% and 29.5

.4% by d 7. The piglets in the present study had

considerable variation in the Hb and Hct values, but most

were within or close to the reported ranges. However, the


PNICU surgery controls (n=2) had low Hb and Hct values.

These piglets had been very weak and may have had problems

unrelated to the treatment or surgery.

At birth, piglets would be expected to have about 6000-

1700 WBC/pl, and 7000-10,000 WBC/Al by d 28 (Schmidt, 1986).

Generally, the piglets that survived until d 7 had CBC

values that were close to the normals for their age. Many

of the Group E piglets had severe diarrhea but did not have

extremely high packed cell volumes, suggesting that the

formula flow rate was sufficient to prevent dehydration.

Summary of Piglet Trial

As a result of this study, some techniques and

procedures for performing trials on colostrum-deprived

piglets were refined. Although each successive study will

probably result in further improvements, the colostrum-

deprived piglet model currently can be used successfully to

investigate the nutritional needs of neonates.

The poor growth rates of the piglets fed the MCT and

LCT diets may have been due to the carbohydrate source. No

information regarding the benefits of the addition of MCT to

neonatal formulas could be obtained from this trial.

However, using an appropriate base formula should allow this

information to be obtained from future studies using the

piglet model.


deprived piglet model currently can be used successfully to

investigate the nutritional needs of neonates.

The poor growth rates of the piglets fed the MCT and

LCT diets may have been due to the carbohydrate source. No

information regarding the benefits of the addition of MCT to

neonatal formulas could be obtained from this trial.

However, using an appropriate base formula should allow this

information to be obtained from future studies using the

piglet model.

It is clear that only digestible carbohydrates should

be included in neonatal enteral formulas. However,

information on the digestibilities of various sugars may not

be available for each species. Development of an enteral

diet for foals based on piglet data was a goal of this

research. However, it was not known whether newborn foals

had sufficient maltase activity to adequately digest the

malto-dextrins in the base elemental diet. Therefore, an

oral disaccharide tolerance test was performed on neonatal




Neonates that are sick, stressed, or born prematurely

may have poorly functioning digestive systems. The pattern

of small intestine disaccharidase development may not be the

same for these newborns as for healthy neonates (Rossi et

al., 1986). Lactase activity can be reduced greatly as a

result of prematurity or disease (Rossi et al., 1986;

Tzipori et al., 1984). Consequently, lactose may not be

suitable for use in enteral diets designed to support non-

healthy neonates. The use of other carbohydrate sources

needs to be considered carefully. Including monosaccharides

in enteral diets would increase the osmolarity of the

formula higher than neonatal intestines could handle well,

possibly causing diarrhea. Complex sugars must be cleaved

into smaller and smaller units, and finally from

disaccharides into single sugar molecules before absorption

can take place. It is not known if neonatal foals can

digest and absorb disaccharide sugars other than lactose.

Therefore, it is important to determine which enzymes are

likely to be active in the intestine of equine neonates.

Enzyme activity varies with age and species and can vary

among similar individuals (Paige and Bayless, 1981). There

is very little information in the literature about the

disaccharidase activity in the small intestine of foals,

although it was studied in neonatal humans, pigs, calves,

and rats (Rossi et al., 1986; Dollar and Porter, 1957; Code,

1968; Paige and Bayless, 1981).

Dollar and Porter (1957) concluded from their study of

newborn calves that enzyme activities determined from

mucosal cell extracts were correlated closely to the results

of an oral disaccharide tolerance test. Using a similar

procedure, Roberts et al. (1974) studied lactase, maltase,

and sucrase activities in horses of various ages by

homoginizing samples of intestinal mucosa. Small intestine

tissue was obtained from the horses slaughtered or

euthanized. However, the data does not show the pattern of

disaccharidase development during the critical first week

postpartum. Lactase was detected as early as 105 d

gestation. Peak activity was found in foals at birth and it

remained high for 4 mo. Sucrase was barely detected in

fetuses from 3 to 9 mo gestation, remaining low at birth.

Adult activities were attained by 7 mo of age. Similarly,

maltase was very low until the ninth month of gestation.

Activities at birth were 12 to 15% of the enzyme activities

of adult equines. The authors reported a gradual increase,

until adult activities were reached at 7 mo of age.


In a subsequent study, Roberts (1975) determined the

ability of horses to digest and absorb disaccharides using

an oral disaccharide tolerance test. These results agreed

with those of his earlier study. However, this study did

not include data on very early neonatal life.

A non-invasive method of estimating disaccharidase

activity, such as an oral tolerance test, would allow foals

to be studied at more than one point in time and would not

require the experimental animals to be euthanized.

The present study was done using healthy foals, nursing

their mothers. Although the results will not necessarily be

applicable to sick or premature foals, it may eliminate some

disaccharides as possible ingredients for enteral diets; if

healthy foals are unable to digest a particular

disaccharide, then it is unlikely that a stressed foal would

have that ability. The results of this study will provide

information on the digestive ability of newborn, healthy

foals non-invasively, and help determine possible sources of

carbohydrates for use in enteral formulas. Because the

animals do not need to be euthanized to obtain samples, each

foal can be used on different days to determine the changes

in enzyme activity during the first week of life. There can

be considerable variation in activities among animals

(Manners and Stevens, 1972).

Selection of a carbohydrate source for inclusion in

elemental, enteral formulas will depend on the activity of

the necessary enzymes and the desired osmolarity of the

formula. Disaccharidases are important in the digestion of

all sugars larger than monosaccharides.

It was presumed that healthy foals would have adequate

lactase activity while relying on milk for nourishment. It

was also presumed that glucose would be easily absorbed by

these neonates. Therefore, the changes in plasma glucose

occurring after administration of glucose and lactose would

be used for comparison with test sugars. Maltose and

sucrose were chosen as test sugars. Maltose was included

because it was the carbohydrate source in the base formula

fed to the piglets. Sucrose was tested because it was

thought that this sugar would be poorly digested and would

provide a reference for a non-absorbed substrate.

The goals of this trial were:

1. to investigate the ability of newborn, healthy

foals to digest maltose, lactose, and sucrose during the

first 5 d postpartum;

2. to investigate the change in blood glucose in

healthy foals after a short fast and the effects of age on

the ability of the foals to maintain blood gluocse

concentration after a short fast; and,

3. to determine the changes in blood glucose in

healthy foals in response to an oral glucose challenge.


Materials and Methods

Preliminary Study

A preliminary study to determine the effect of nursing

on blood glucose was performed on a normal, healthy foal at

8 h postpartum. Care of the mare and foal during and after

birth followed standard procedures at the University of

Florida Horse Research Center.

At 8 h of age, the foal was fitted with a Teflon

jugular vein catheter (Quik-Cath, Travenol Laboratories,

Inc.), in an aseptic procedure, to allow periodic blood

sampling. To minimize stress to the foal, a small amount of

local anesthetic was injected at the catheter site. The

plastic wings of the catheter were fixed to the skin with a

superglue adhesive. This was easily removed with acetone at

the conclusion of each trial period. The catheter was

filled with sterile, heparinized saline between samplings.

The foal was then allowed to return to the mare and resume

normal activity. After the foal nursed the mare, the time

was noted and the foal was muzzled to prevent nursing.

Blood was drawn from the catheter at this time (time = 0).

Blood was then drawn every 30 min thereafter for 2 h. This

procedure was repeated on the same foal on d 5. This study

suggested that blood glucose reached a baseline

concentration after a 1 h fast. It was concluded that a 2 h

fast prior to administration of the treatment, and a 4 h

fasting collection period would be sufficient to monitor


changes in blood glucose after a test meal in the subsequent


Animals and Design

Eleven Quarter Horse and two Thoroughbred foals were

used for the main trial. The foals remained with their dams

throughout the trial. Foals were weighed shortly after

birth and on d 5. After each foal was born, it was randomly

assigned, by sex, to one of four treatment groups until the

block was filled. This procedure was repeated for

subsequent foals, assigning foals of each sex to fill the

treatment blocks evenly.

On d 1, at 6 h postpartum, each foal was fitted with a

jugular vein catheter, as described above. Foals were

muzzled to prevent nursing. The foals were then fasted for

2 h to allow blood glucose concentrations to reach a

baseline level. Each animal was then given the appropriate

treatment consisting of oral administration of either

maltose, lactose, or sucrose at 1 g (5.56 mmol) per kg BW in

a 20% solution (w/v), or a glucose solution at .5g (2.78

mmol) per kg BW. These dosages were similar to dosages used

by Roberts (1975).

The solution was offered to the foal in a bottle fitted

with a suitable nipple. If the foal did not voluntarily

consume the solution, it was administered via dosing syringe


or via nasogastric tube. Blood was collected from the

catheter immediately before the solution was given

(time = 0), every 15 min for the first hour, and every 30

min thereafter for the second, third, and fourth h of the

trial period. Each foal received the same treatment when

the procedure was repeated on d 3 and d 5 postpartum.


The blood was collected in tubes containing

ethylenediaminetetraacetic acid (EDTA), plasma removed after

centrifugation at 3500 g for 20 min and frozen. Plasma

glucose was determined using the Trinder peroxidase method

(Sigma Chemical Co.). Optical density was measured at 505

nm with a Beckman DU-20 spectrophotometer.

Statistical Analyses

Statistical analyses were performed by method of least

squares ANOVA with time as a class variable (Snedecor and

Cochran, 1969,) using the GLM procedures of SAS (Freund and

Littell, 1981). The complete mathematical model is given in

Table 3.1. The appropriate error term is shown for each

independent variable.


Table 3.1. Complete mathematical model for ANOVA of changes
in plasma glucose.

Error term

Treatment 3

Sex 1

Treatment*Sex 3

Foal(Treatment*Sex) 5

Day 2

Treatment*Day 6

Day*Sex 2

Treatment*Sex*Day 6

Day*Foal(Treatment*Sex) 10

Time 10

Time*Treatment 30

Time*Foal(Treatment*Sex) 50

Day*Time 20

Sex*Time 10

Day*Sex*Time 20

Treatment*Sex*Time 30

Day*Treatment*Time 60

Treatment*Day*Sex*Time 60

Remainder 101





















Equations for the time curves were obtained using

reduced models with time as a continuous variable. The

reduced pooled model is presented in Table 3.2. The models

for comparison of day curves, treatment curves and

day*treatment curves were similar to the pooled model except

that daytimee, treatment*timen, and day*treatment*timen,

respectively, were used in place of time. F tests for

homogeneity of the time curves were calculated based on the

techniques described by Snedecor (1956). Rejection of the

null hypothesis would indicate that the curves differed,

i.e. were not parallel.