Content uploaded by Jessica Suagee-Bedore
Author content
All content in this area was uploaded by Jessica Suagee-Bedore on May 09, 2019
Content may be subject to copyright.
gastrointestinal disease. It was hypothesized that diets with
decreasing DPS would result in decreased fecal particle size (FPS)
and fecal pH. Eight geldings (528 ±54 kg BW) ranging from 3 to
15 yr were used in a Latin Square design. Four dietary treatments
fed at 2% BW were 100% hay (HA), 40% hay and 55% whole oats
(HO), 40% hay, 27.5% whole oats, and 27.5% pellets (HOP), and
40% hay and 55% pellets (HP). Differences in DPS are clarified by
the % feed particles <1.18 mm in the HA, HO, HOP, and HP diets
(1, 5, 33, and 60%; respectively). Pellets were formulated to have
similar nutrient makeup to whole oats. To meet nutrient re-
quirements, HO, HP, and HOP diets had a balancer pellet added
at 5%, that was unneeded for HA diet. Treatments were divided
into 4 14 d periods; each proceeded by a 14 d hay only washout
period. The DPS profiles of feed were obtained using the Penn
State Particle Separator and feed company data. These values
were used to calculate DPS distribution of the 4 diets (Table 1).
Fresh fecal samples were collected from the stall on d 10 during
treatment periods and pH was analyzed. Remaining feces were
frozen at 20C until FPS was measured by a modified Nasco
Digestion Analyzer sieving method (Table 1). A mixed ANOVA
was used to investigate differences due to fixed effects of diet
and period. Differences were defined at P<0.05. There was an
overall effect of diet on fecal pH and fecal pH was different be-
tween each diet group (HA 6.7 ±0.3, HO 6.3 ±0.1, HOP 6.0 ±0.1,
and HP 5.8 ±0.1). The range of pH may be due to different
availability of dietary NSC across diets. The horses fed whole oats
had the largest % of 4.8 mm (largest diameter sieve) fecal par-
ticles, a characteristic that may be due to the indigestible oat
hulls. This study reports feasible methods to quantify FPS and
DPS and results indicate that certain characteristics of whole oats
may increase hindgut pH.
Key Words: pH, Oats, particle size
125
Equine microbiome project:
Understanding differences in the horse
gut microbiome related to diet
J.T. Berg*, B. Chambers, H. Siegel, A. Biddle
University of Delaware, Newark, DE, USA
In the hindgut of a horse, changes in the bacteria levels through
dietary disruption, seasonal changes, stress, or age can lead to
colic and laminitis, causing lameness or even death in severe
cases. In this study, we profiled bacterial communities from fecal
samples collected and submitted by horse-owners participating
in the Equine Microbiome Project (EMP) and correlated differ-
ences in community structure with feed type, specifically horses
eating exclusively hay, pasture, a hay-concentrate mix, or a
combination of pasture, hay, and concentrate as reported in EMP
metadata surveys. Feed categories were based on reported diet,
and no surveys reported feeding a hay and pasture diet, thus the
absence of a hay/pasture group. To participate in the EMP, horse
owners submitted a fresh fecal sample collected in a provided kit
according to standard instructions, and were required to com-
plete an informational survey including horse diet, past medical
history, and other metadata. Genomic DNA from fecal samples
from 184 horses were analyzed using the 16S rRNA gene. Se-
quences were clustered against the Greengenes database, and
b
diversity was calculated using weighted UniFrac metric in QIIME.
Significant differences in bacterial community structure of
pasture fed horses were found using PERMDISP (999 permuta-
tions, P-value <0.05). Spearman rank correlation (999 permuta-
tions, P-value <0.05) identified Christensenellaceae,Oscillospira,
and Prevotella taxa to be more highly abundant in only pasture
whereas RFN20, Streptococcus, and Lactobacillus taxa were
differentially represented in hay/concentrate. These results point
to functional differences in these communities that could lead to
understanding how diet affects normal microbiome structure and
hypotheses regarding functional differences leading to equine
digestive disorders such as colic, laminitis, and equine metabolic
syndrome.
Key Words: diet, microbiome, multivariate analysis
126
Morphometric comparisons of draft,
pony, and horse breeds
J.K. Suagee-Bedore*, D.R. Linden
The Ohio State University, Wooster, OH, USA
Estimating body fatness is an important measurement tool;
however, some measurements are subjective and even repeated
trainings fail to increase accuracy. Objective measurements,
including equations based off morphometric measurements, may
provide a better estimate. Ponies (n ¼30), horses (n ¼32), and
draft horses (n ¼27) were evaluated for neck length (NL), neck
circumference (NC), heart girth (HG), belly girth (BG), wither
height (WH), body length (BL), and body condition score (BCS).
Calculations included the ratio of NC to NL (NCNL), BG to HG
(BGHG), BG to WH (BGWH), and BG to BL (BGBL) and body con-
dition index (BCI). Breed types differed by BCS, with draft horses
(6.0 ±0.2) averaging higher (P<0.001) scores than horses (5.2 ±
0.2) or ponies (5.2 ±0.2). The ability of the BCI equation to
accurately predict BCS scores across breed type was evaluated by
classifying animals as having acceptable agreement (absolute
difference 1; AGR) or having unacceptable agreement (absolute
difference >1; NAG), and were analyzed using Chi Square.
Agreement between the BCI and BCS was noted for 66% of horses,
Table 1
Number of horses with agreement between BCS and BCI scores using
different equations
Type Original BCI Adjusted BCI
AGR NAG P-value AGR NAG P-value
Horse 22 10 0.034 28 4 0.001
Pony 16 14 0.715 23 7 0.004
Draft 9 18 0.083 22 5 0.001
Table 1
Distribution of particle size in feed and feces expressed on % DM basis
Feed PS (mm) Diet P-value
HA HO HOP HP
>19.0 82 33 33 33
8.1e19.0 9 4 4 4
4.1e8.0 5 46 24 2
1.18e4.0 4 12 7 1
<1.18 1 5 33 60
Fecal PS (mm)
4.8 22±3
a
30±4
b
27±2
b
10±4
c
<0.001
2.4 10±3
ab
9±2
b
7±2
a
8±1
ab
<0.05
1.6 14±3
a
11±2
ab
10±1
b
13±3
a
<0.05
Cheesecloth
1
23±2
a
18±3
b
20±2
b
31±4
c
<0.001
Remainder
2
32±6
a
32±5
a
38±5
a
39±7
a
<0.05
1
Cheesecloth was folded 4 times.
2
Calculated % of DM that was not caught on any of the layers.
Abstracts / Journal of Equine Veterinary Science 52 (2017) 76e9594
53% of ponies, and 33% of drafts horses (Table 1). To improve the
rate of acceptable agreement, morphometric comparisons were
made within breed type. For horses, the HGWH ratio tended (P¼
0.085) to be greater for NAG horses (1.22 ±0.04) than AGR horses
(1.13 ±0.03), and in horses where the HGWH ratio was >1.16,
altering the BCI equation to [(HG
0.5
+BG+NC
1.2
)/WH
1.06
)]
2.2
, resul-
ted in 88% agreement between BCI and BCS (Table 1). For ponies,
the NCNL was greater (P¼0.016) in NAG (1.37 ±0.04) than AGR
(1.23 ±0.04) ponies, and in ponies were the NCNL was >1.30,
altering the BCI equation to [(HG
0.5
+BG+NC
1.155
)/WH
1.05
)]
2.2
resulted in a 77% agreement between BCI and BCS (Table 1). In
drafts, the BGHG ratio was greater (P¼0.039) in NAG (1.07 ±0.01)
than AGR (1.03 ±0.01), and in drafts with a BGHG >1.051, altering
the BCI equation to [(HG
0.5
+BG
0.95
+NC
1.2
)/WH
1.05
)]
2.2
resulted in
81% agreement between BCI and BCS (Table 1). Equations to
predict BCS have the potential to improve nutritional manage-
ment across all equine types, especially for equine managers with
minimal training. Further evaluation is warranted.
Key Words: body condition, body fat, morphometrics
Abstracts / Journal of Equine Veterinary Science 52 (2017) 76e95 95