INVESTIGATING BODY CONDITION AND METABOLIC HORMONES IN
RELATIONSHIP TO REPRODUCTIVE CYCLICITY IN FEMALE AFRICAN
ELEPHANTS, LOXODONTA AFRICANA
by
Kari Ann Morfeld
A Dissertation
Submitted to the
Graduate Faculty
of
George Mason University
in Partial Fulfillment of
The Requirements for the Degree
of
Doctor of Philosophy
Environmental Science and Public Policy
Committee:
_______________________________________ Dr.
Geoffrey Birchard, Dissertation Director
_______________________________________ Dr. Janine
Brown, Committee Member
_______________________________________ Dr.
Cody Edwards, Committee Member
_______________________________________ Dr.
William Rosenberger, Committee Member
_______________________________________ Dr.
Albert Torzilli, Graduate Program Director
_______________________________________ Dr.
Robert Jonas, Department Chairperson
_______________________________________
Dr. Timothy Born, Associate
Dean for Student and Academic
Affairs, College of Science
_______________________________________
Dr. Peggy Agouris, Interim Dean,
College of Science
Date: ________________________________ Summer Semester 2013
George Mason University
Fairfax, VA
Investigating Body Condition and Metabolic Hormones in Relationship to Reproductive
Cyclicity in Zoo Female African Elephants (Loxodonta africana)
A Dissertation submitted in partial fulfillment of the requirements for the degree of
Doctor of Philosophy at George Mason University
By
Kari Ann Morfeld
Master of Science
University of Nebraska, 2006
Bachelor of Arts
Nebraska Wesleyan University, 2001
Director: Geoffrey Birchard, Professor
Department of Environmental Science and Public Policy
Summer Semester 2013
George Mason University
Fairfax, VA
Copyright: 2013 Kari Ann Morfeld
All Rights Reserved
ii
DEDICATION
This dissertation is dedicated to my daughter, Scarlett.
iii
ACKNOWLEDGEMENTS
I would like to thank my daughter, my parents, and my advisor-Janine Brown.
iv
TABLE OF CONTENTS
Page
List of Tables…………………………………………………………………………..…vi
List of Figures…………………………………………………………………………....vii
Abstract................................................................................................................……….viii
Chapter 1: Introduction……………………………………………………………………1
Chapter 2 : Development of a body condition scoring index for female African
elephants validated by ultrasound measuresments of subcutaneous fat…………………22
Chapter 3 : High body condition and elevated insulin and leptin levels are associated
with ovarian acyclicity in zoo African elephants……………………………...................51
Chapter 4 : Conclusions……………….…………………………………………………75
Appendix I…..……….…………………………………………………………………..77
Appendix II…..……….………………………………………………………………….79
Appendix III..……….……………………………………………………………………80
References.……………………………………………………………………………….81
v
LIST OF TABLES
Table
Page
Table 2.1. African elephant body condition scoring (BCS) index……………………….29
Table 2.2. Body condition scores (BCS) (and relative percentage) of free-ranging and
captive female African elephants in each BCS category………………………………...35
Table 2.3. Mean (SD) ultrasound measurements of subcutaneous fat thickness by body
region and body condition score (BCS) category in zoo African elephants (N = 33)…...37
Table 2.4. Spearman correlations among body condition score (BCS) and mean
ultrasound measurements of fat thickness of specific body regions……………..………39
Table 2.5. Level of intra-and inter-assessor agreement for assessment of elephant body
condition…………………………………………………………………………………43
Table 3.1. Number of cycling and non-cycling female African elephants in each body
condition score (BCS) category………………………………………………………….65
Table 3.2. Comparison of metabolic hormone concentrations [mean (SD)] between
cycling and non-cycling zoo African elephants by body condition score (BCS)
category…………………………………………………………………………………..66
Table 3.3. Logistic regression results of metabolic parameters predicting a non-cycling
status in female African elephants……………………...………………………………..67
vi
LIST OF FIGURES
Figure
Page
Fig 1.1. A general model of the reproduction-associated hormones fluctuations and
associated follicular waves during the estrous cycle in elephants………………………...9
Fig 1.2. Relationship of obesity and metabolic conditions on reproductive
abnormalities……………………………………………………………………………..14
Fig 2.1. Key areas for visually assessing body condition in female African elephants…28
Fig 2.2. Body condition scoring flow chart for female African elephants………………31
Fig 2.3. Percentage of free-ranging (n=57) and captive (n=50) female African elephants
in each body condition score category…………………………………………………...36
Fig 2.4. Scatter plot of body condition scores and overall mean ultrasound
measurements of fat thickness for all five body regions combined in female African
elephants…………………………………………………………………………………41
Fig. 3.1. Parallelism between binding inhibition curves of serially diluted African
elephant serum pools using leptin RIA…………………………………………………..57
Fig. 3.2. Recovery of leptin (1.56, 3.13, 6.25, 12.5, 25, and 50 mg/ml) added to serum
from an adult female African elephant…………………………………………………..58
Fig. 3.3. Parallelism between binding inhibition curves of serially diluted African
elephant serum pool using an anti-bovine insulin EIA…………………………………..60
Fig. 3.4. Recovery of insulin (0.05, 0.15, 0.5, 1.5, and 3.0 mg/ml) added to a pool of
serum from adult female African elephants (n=6)……………………………………….60
vii
ABSTRACT
INVESTIGATING BODY CONDITION AND METABOLIC HORMONES IN
RELATIONSHIP TO REPRODUCTIVE CYCLICITY IN ZOO FEMALE AFRICAN
ELEPHANTS, LOXODONTA AFRICANA
Kari Ann Morfeld, PhD
George Mason University, 2013
Dissertation Director: Dr. Geoffrey Birchard
African elephants (Loxodonta africana) in U.S. zoos generally appear heavier than
wild counterparts, and there are claims that obesity-related health and reproductive
problems may be contributing to population non-sustainability. A previous study found a
high body mass index (BMI) was positively correlated with ovarian inactivity in African
elephants, suggesting reproductive problems may be caused in part by metabolic
derangements associated with excessive body weight. To determine whether obesity and
related metabolic conditions are a problem in zoo-managed African elephants, body
condition, insulin, glucose, and leptin levels, and the glucose-to-insulin ratio (G:I) were
compared between breeding-aged cycling (n=23) and non-cycling (n=23) females. Body
condition scoring (BCS) is the assessment of subcutaneous fat stores based on visual or
viii
tactile evaluation of muscle tone and key skeletal elements, and provides an immediate
appraisal of the degree of fatness of an individual. An objective of this study was to
develop a visual BCS index for female African elephants and validate it using ultrasound
measures of subcutaneous fat. To develop the index, standardized photographs were
collected from zoo (n=50) and free-ranging (n=57) female African elephants to identify
key body regions and anatomical sites, which were used to visually assess body fat
deposition patterns. The visual BCS method consisted of a list of body regions and the
physical criteria used for obtaining an overall score on a 5-point scale, with 1
representing the lowest and 5 representing the highest levels of body fat. Significant
correlations were found between the visual scores and ultrasound measures of
subcutaneous fat thickness at all anatomical sites, but were highest in the backbone (r =
0.748, P < 0.01) and pelvic bone (r = 0.745, P < 0.01) regions, indicating that BCS
adequately reflects the amount of actual fat reserves. The new BCS index proved to be a
reliable and repeatable evaluation method, with a high percentage agreement (73 % to
95%) and an overall “substantial” strength of agreement determined by the weighted k
statistic (κw = 0.62 to 0.91) between and among three assessors. In comparing
photographs of wild vs. captive elephants, the median BCS in the free-ranging
individuals (BCS=3, range 1-5) was lower (P = 0.0001) than that of the zoo population
(BCS=4, range 2-5). When comparing, BCS, insulin, glucose, and leptin levels, and the
glucose-to-insulin ratio (G:I) between cycling and non-cycling females, the mean BCS of
non-cycling elephants was higher than that of cycling elephants [4.39 (SD 0.58) vs. 3.73
(SD 0.93), P = 0.009]. There were differences in the serum concentrations of leptin [4.03
ng/mL (SD 1.63) vs. 3.16 ng/mL (SD 0.88); P = 0.041] and insulin [0.65 mg/mL (SD
0.31) vs. 0.48 mg/mL (SD 0.18); P = 0.032] for non-cycling females in the BCS=5
category. Serum glucose did not differ between cycling and non-cycling elephants (P =
0.892); however, the G:I was lower in the non-cycling group [69.82 (SD 53.21) vs.
227.18 (SD 96.23); P = 0.019]. Using “non-cycling” as the outcome variable in
regression models, we examined BCS, leptin, insulin, and the G:I ratio as predictors, and
found that BCS showed the strongest predictive power (P = 0.01). The odds ratio for the
BCS coefficient is 3.15 with a 95% confidence interval of [1.36, 7.26], which suggests
that with each 1 point increase in BCS, an elephant is approximately 3 times more likely
to be non-cycling. These results demonstrate that ovarian acyclicity is associated with a
high BCS and perturbations in markers of metabolic status. Thus, monitoring BCS,
insulin, glucose, and leptin may assist in identifying elephants at risk for developing
metabolic health conditions, including fertility problems. This would allow for
management interventions to be implemented to improve body condition and metabolic
status with the goal of reinitiating ovarian activity and promoting a healthier zoo
population.
CHAPTER 1
INTRODUCTION
Overview of the study
The ethics of displaying elephants in zoos is a topic of public discussion and has
received significant media attention in recent years. Evidence that elephant welfare is not
ideal in certain zoo situations includes findings of populations that are not self-sustaining
(Faust and Marti, 2011; Wiese and Willis, 2006) high infant mortality (Saragusty et al.,
2009), stereotypic behaviors (Harris et al., 2008), a high prevalence of reproductive
issues (Brown et al., 2004a; Dow et al., 2011; Proctor et al., 2010a), and foot problems
(Lewis et al., 2010). Despite a long fascination with elephants, we still know little about
what factors impact health and reproduction, which prevents developing optimal
management strategies. At least two health issues, ovarian acyclicity and foot/joint
problems, are believed to be related to poor body condition (i.e. obesity), as a result of
combined excessive food intake and too little physical activity (Clubb & Mason, 2002;
Clubb et al., 2009). Ovarian acyclicity is a serious problem for captive, but not wild,
African elephants; over 40% of reproductive aged females in zoos exhibit abnormal
reproductive cycles and, thus are effectively infertile (Brown et al., 2004a; Freeman et al.,
1
2009; Proctor et al., 2010a). A recent study found a strong correlation between ovarian
acyclicity and a high body mass index (BMI) in zoo-maintained African elephants
(Freeman et al., 2009), suggesting reproductive problems may be caused in part by
metabolic derangements associated with excessive body fat (Brown et al., 2004a; Clubb
and Mason, 2002; Lewis et al., 2010; Proctor et al., 2010a). This seems likely given
studies in horses and humans indicate obesity can lead to metabolic changes that impair
fertility (Bray, 1997; Chang, 2007; Hartz et al., 1979; Irvine and Shaw, 2003; Norman
and Clark, 1998; Pasquali et al., 2003). Obese mares experience an extended interval
between successive ovulations (Vick et al., 2006), an issue common in elephants (Brown,
2000). Stillbirths and dystocias also are common in obese women and horses (Althabe,
2012; Vick et al., 2006), and are a major cause of calf mortality in elephants (Clubb et al.,
2009). Although circumstantial, this evidence suggests that poor reproduction in African
elephants may be related to problems associated with excessive body fat. It is rational to
extend this supposition to zoo-held elephants that too often are fed diets high in calories
and given inadequate exercise (Clubb et al., 2009; Hatt and Clauss, 2006; Lewis et al.,
2010; Mason and Veasey, 2010). Because of the well-recognized need to establish a selfsustaining population in U.S. zoos, there is increased interest in determining if obesity
and metabolic conditions are risk factors for poor fertility in African elephants held in
zoos.
This dissertation presents a series of studies designed to test the hypothesis that noncycling captive elephants display more characteristics of obesity and associated metabolic
alterations than their normal cycling counterparts. Obesity in zoo elephants has not been
2
previously assessed, although certainly warranted, due to a lack of reliable techniques for
assessment. Therefore, a visual body condition scoring (BCS) index in African elephants
was developed (Chapter 2), a validated tool to objectively assess obesity. Metabolic
parameters (BCS, insulin, glucose, and leptin) associated with cyclicity status and then
each BCS category (BCS=1, 2, 3, 4, or 5) in cycling and non-cycling elephants were
compared (Chapter 3). Finally, we examined whether obesity and altered metabolic
hormone concentrations are risk factors for poor reproduction in zoo African elephants.
An improved knowledge of factors contributing to poor reproduction will assist in the
development of management strategies that mitigate obesity and health or reproductive
risks associated metabolic conditions.
African elephants in zoos
Elephants have been maintained in animal collections for at least 3,500 years
(Croke, 1997), originally kept in royal menageries. Zoos, as we know them today,
emerged in the mid-19th century and the elephant remains one of the most popular and
sought-after exhibit animals for zoos. Zoos use elephants as ambassadors to help educate
the public about conservation. Elephants are one of the great attractions for zoo visitors
and thus an ideal flagship species for African wildlife and conservation support.
However, some suggest that confining such large, social and intelligent animals in small
spaces raises health, welfare, and ethical concerns (Clubb and Mason, 2002). Regardless
of whether elephants should or should not be kept in zoos, there is little doubt that all
animal holding facilities should provide the best husbandry and welfare for these
3
magnificant animals. The Association of Zoos and Aquariums (AZA) is committed to
animal welfare, exemplifed by misson statement as follows:
“The Association of Zoos & Aquariums (AZA) provides its members the services,
high standards and best practices needed to be leaders and innovators in animal
care, wildlife conservation and science, conservation education, the guest
experience, and community engagement.”
The AZA is continually providing zoos resources for optimal elephant welfare and care,
and provides an “Elephant Husbandry Manual”. Although the AZA and AZA-accredited
zoos are striving to provide best practices for animal care, science-based information in
several areas is needed, and a priority of the AZA to support such research.
In the wild, there are an esimated 472,000 - 690,000 African elephants remaining
(Blanc et al., 2007), and the African elephant is currently listed as „vulnerable‟ on the
International Union for the Conservation of Nature (IUCN) Red List of Threatened
Species. Southern Africa has by far the largest known number of elephants across the
four sub-regions with just over 56% of the continent‟s elephants. Eastern Africa holds
almost 27%, Central Africa 16% and West Africa 1.5%. While populations in Central
Africa have in the past been most at risk from poaching and the illegal ivory trade,
previously secure populations in Eastern and Southern Africa are now facing an
increasing threat from illegal killing. In addition, land conversion and loss of habitat
continue to be major long-term threat to the species across its range. Given the increased
threats to the population viability of African elephants, a healthy, self-sustaining captive
population is essential to long-term global conservation (Brown et al., 2004a).
4
Unfortunately, zoo populations of African elephants are not self-sustaining due to
problems of both high mortality and low reproduction, with 5 deaths to only 3 births
occurring over the past decade (Faust and Marti, 2011). If this trend continues, elephant
populations will pass through a severe bottleneck in 30 years, becoming demographically
nonviable in about 50 years (Faust and Marti, 2011). To prevent further population
declines, the Association of Zoos and Aquariums Elephant Taxon Advisory
Group/Species Survival Plan® (AZA TAG/SSP) Management Committee has endorsed
research to better understand the causes of poor health and reproduction of African
elephants (Keele and Ediger, 2011)
Conditions affecting zoo elephants. A number of health and reproductive
problems afflict zoo elephants, including cardiovascular disease, arthritis, foot problems
and ovarian cycle abnormalities (Brown, 2000; Brown et al., 2004a; Clubb et al., 2009;
Clubb and Mason, 2002; Lewis et al., 2010; Mason and Veasey, 2010). All of these are
suspected of being linked to one central problem: obesity (Clubb et al., 2009). Many zoomanaged elephants are subjectively considered to be obese (Clubb et al., 2008,2009;
Clubb and Mason, 2002; Hatt and Clauss, 2006;). Studies in horses (often used as a
model for elephants) and women provide clear evidence that obesity can lead to
biological changes related to cardiovascular disease, metabolic problems (i.e. insulin
resistance), arthritis, foot problems, and impaired fertility. A primary cause of morbidity
in zoo elephants is poor foot health; half of all captive elephants will suffer from footrelated problems in their lifetime (Clubb and Mason, 2002; Lewis et al., 2010). Arthritis
5
is also a concern in zoo-elephants (Lewis et al., 2010); both foot problems and arthritis
are suspected of being related to excess body weight and lack of exercise (Clubb et al.
2008, 2009; Lewis et al., 2010). Ovarian acyclicity is a serious reproductive problem in
captive elephants; 15% of Asian females and 42% of African females exhibit abnormal
ovarian cycles and are considered infertile (Proctor et al., 2010). As mentioned
previously, reproductive problems observed in zoo elephants are also suspected of being
associated with obesity; however, these relationships have not been properly investigated.
Reproduction in zoo elephants
Elephants display distinct differences in reproductive physiology compared to
other mammalian species. Females have the longest ovarian cycle of all mammals studied
to date: 14-16 weeks in duration, with an 8-12 week luteal phase and a 4-6 week
follicular phase (Brown, 2000; Hodges, 1998; Plotka et al., 1988). The estrous cycle is
characterized by assessing luteal steroid activity; however, instead of progesterone as in
most other mammals, the major circulating luteal progestagens are reduced pregnanes
like 5α-pregnane-3,20-dione (5α-DHP) and 5α-pregnane-3-ol-20 one (5α-P-3-OH)
(Brown, 2000; Hodges et al., 1997). A representative endocrine model for the elephant
estrous cycle is presented in Figure 1.1. Elephants also have the longest gestation of any
studied mammal; pregnancy lasts 20-22 months and can be diagnosed by elevated
progestagens beyond the normal luteal phase (about week 12).
Whereas ovulation in mammals is typically induced by a single, pre-ovulatory
luteinizing hormone (LH) surge, the elephant uniquely exhibits two LH peaks during the
6
follicular phase, referred to as the „double LH surge‟ (Brown et al., 1999; Kapustin et al.,
1996). These two surges are spaced consistently 19-21 days apart, with ovulation
occurring after the second (ovulatory) LH surge. During the first surge (anovulatory
surge), which occurs 10-20 days after the drop in progestagens, small, non-ovulating
follicles develop, however large follicles that attain ovulatory diameter have also been
observed at the first LH peak, but these large follicles never ovulate (Brown et al., 1999;
Lueders et al., 2010). During the second follicular wave, only one large dominant antral
follicle ovulates, and ovulation occurs approximately 24 hours after the ovulatory LH
surge (Brown et al., 1999). No differences in LH concentrations are found between the
two peaks and estradiol surges precede both surges (Czekala et al., 2003; Lueders et al.,
2010).
The unique double-LH surge raises a number of questions, including: 1) what is
the function of the first peak? and 2) why does ovulation occur only after the second
peak? Initial theories suggested the first LH surge prepared the endometrium for the long
reproductive cycle (Kapustin et al., 1996), or that it acted as a primer to attract roaming
bulls in advance of ovulation to ensure a mating partner was available upon ovulation
(Hermes et al., 2000). Recently, ovarian ultrasonic observations combined with hormone
measures in blood samples of Asian elephants provide a new explanation for the double
LH surge and follicle selection (Leuders et al. 2011). A major finding was evidence that
luteinized follicles (LUFs) secrete inhibins, indicating a likely role in determining
dominant follicle selection. The interaction of inhibin and follicle-stimulating hormone
(FSH) plays a major role in folliculogenesis and dominant follicle selection in many
7
species. Inhibin is secreted by the granulosa cells of follicles that develop from small
antral to preovulatory size in several species, including humans (Knight and Glister,
2001). Inhibin then suppresses FSH secretion from the pituitary gland by negativefeedback; the declining FSH results in a reduced growth of smaller follicles and a
subsequent growth advantage for the largest follicle(s). In contrast, inhibin production in
elephants originates in luteal rather than follicular tissue. Leuders et al. (2011) showed
the importance of the first LH peak, because it is associated with formation of LUFs and
secretion of inhibin derived from the luteinized granulosa cells of the LUFs. Thus, they
conclude that the development of LUFs is a precondition for inhibin secretion, which in
turn impacts the selection of a single dominant, ovulatory follicle.
8
Fig. 1.1. A general model of the reproduction-associated hormones fluctuations and
associated follicular waves during the estrous cycle in elephants. The dashed vertical line
marks the time of ovulation (Adapted from Hildebrant et al., 2011).
9
The Endocrinology Laboratory at the Smithsonian Conservation Biology Institute
serves the Elephant Taxon Advisory Group/Species Survival Plan (TAG/SSP) by
conducting routine hormonal assessments of elephants (e.g., progestagens) to determine
reproductive status. Through this service, it has been determined that 35% of zoomanaged African elephants are „flatliners‟, that is, serum progestagen concentrations
remain at baseline levels indicating complete ovarian inactivity; another 10% exhibit
irregular cycles (Brown et al., 2004; Dow et al., 2011). Of even greater concern is that
over three quarters of abnormal cycling females are of reproductive age (i.e., from 11
through 35 years of age) (Dow et al., 2011). By contrast, reproductive acyclicity is only
observed in ~15% of Asian elephants and few of these are of breeding age (i.e. 11-35
years of age ), suggesting age may be factor affecting reproductive acyclicity in Asian,
but not African elephants (Dow et al., 2011). Transrectal ultrasonography techniques
have become instrumental in monitoring reproductive tract health of zoo elephants, with
the majority of elephants having ovarian, uterine and vaginal cysts and/or tumors
(Hildebrandt et al., 2006). Older, nonreproductive African elephants are prone to
developing endometrial hyperplasia (Hildebrandt et al., 1998, 2000), whereas Asian
elephants develop uterine leiomyomas (Dow et al. 2011; Hildebrandt et al., 2006;
Hildebrandt and Goritz, 1995). Overall, however, reproductive tract pathologies are not a
significant problem affecting cyclicity in zoo elephants.
One significant endocrine finding associated with abnormal ovarian activity is
high circulating prolactin concentrations; i.e. hyperprolactinaemia (Brown et al. 2004a;
Dow et al., 2011; Meyer et al. 2004). In 2004, a comprehensive study aimed to determine
10
if specific hormonal imbalances were causative factors to ovarian acyclicity was carried
out (Brown et al. 2004a). Of the 10 hormones examined (luteinizing hormone, follicle
stimulating hormone, prolactin, thyroid stimulating hormone (TSH), estradiol, free and
total triiodo thyronine (T3), free and total thyroxin (T4) and cortisol), only prolactin
differed and was significantly elevated in one-third (11/33) of acyclic females. Given the
high rate of hyperprolactinemia in acyclic females, Dow et al. (2012) re-examined the
zoo population and found the frequency of hyperprolactinemia had increased
significantly from 37% in 2004 (Brown et al., 2004a) to 71% of acyclic African females
in 2011.
Ovarian acyclicity only appears to occur in captivity (Brown, 2000), and is not a
significant problem for wild, reproductive age African elephants (Freeman et al., 2009).
Various aspects of reproduction differ in captive versus wild elephants; for example,
females in captivity reach reproductive maturity at approximately 7 or 8 years of age,
several years earlier than those in the wild (Brown, 2000). Elephants in the wild can
conceive into their 50s (Hall-Martin, 1987; Laws et al., 1970; Moss, 2001; Whitehouse
and Hall-Martin, 2000), much later in life than observed in zoos. Conceptions in the wild
occur once every 3-5 years, equating to approximately 10 times throughout an
individual‟s lifetime. By contrast, females in captivity that have not produced a calf by
the time they are 30 years of age are not likely to conceive or successfully produce a calf
(Brown et al., 2004). These older, nulliparous females typically express reproductive tract
abnormalities like ovarian and uterine cysts and tumors, and increased rates of dystocias
and stillbirths (Brown, 2000; Hermes et al., 2004). This may be due to the fact that zoo
11
elephants cycle three to four times per year, and do so continuously since pregnancy is
rare. Thus, cycling females are exposed to the steroids associated with multiple
consecutive cycles. By contrast, free-ranging animals are generally pregnant or in
lactational anestrus at any given time, and multiple non-conceptive cycles are rare
(Hildebrandt et al., 2006). Thus, it is important to identify and breed reproductively
viable females at a young age before the chances of producing a calf significantly
decrease, and reproductive tract pathologies become a problem.
Linking energy balance and reproduction
Every level of the reproductive axis, the hypothalamus, pituitary gland, and
gonad, has the capacity to respond to metabolic cues. In humans, inadequate food intake
(starvation), excessive exercise, or increased thermoregulatory costs can all shut down
reproductive cyclicity and the secretion of gonadal steroids that are essential for the
health of many organs and tissues (Forcada & Abecia, 2006; Garcia et al., 2011; Hill et
al., 2008; Kumar, 2011). Since animals under caloric deficiency invest energy in survival
first and reproduction second, the reproductive axis has the capacity to respond to
changes in energy status. If excess leanness occurs in young women, puberty is often
delayed (Martos-Mareno et al., 2010; Roze et al., 2007). A critical amount of energy
reserve is necessary for puberty initiation, normal sexual maturation, and maintenance of
cyclicity and fertility in females of most species (Frisch et al., 1970; Hill et al., 2008;
Kennedy, 1963). Puberty is a complex process that involves changes in maturation of
reproductive organs, physical manifestations of steroid hormones and establishment of
12
reproductive capacity. The onset of puberty is marked by an increase in gonadotropinreleasing hormones (GnRH) secretion in a defined manner. When puberty develops
normally, sexual maturity is achieved and the secretion of gonadotropins becomes
pulsatile, consisting of variable but well defined secretory pulses across the estrous, or
menstrual, cycle. This pulsatile release of GnRH is obligatory to sustain normal
gonadotropin synthesis and secretion, cyclicity, and ovulation. There is a tight connection
between energy homeostasis and reproduction, which is substantiated by the interaction
between numerous neuroendocrine signals which jointly participate in the physiological
control of energy balance and reproductive maturation and function (FernandezFernandez et al., 2006; Tena-Sempere, 2003, 2007). On the opposite end of the spectrum,
obesity and excess energy reserves can also negatively affect fertility (Martos-Mareno et
al., 2010; Panidis et al., 2013). A schematic of the complex interaction of metabolic
hormones, obesity, and infertility is shown in Figure 1.2. Fertility processes involve
complex mechanisms of both ovarian and extra-ovarian origin, and by interfering with
normal gonadal and neuroendocrine function, obesity can reduce ovulatory and fertility
rates in otherwise healthy females.
The role of adipose tissue is crucial in controlling the balance of sex hormone
availability in the target non-fat tissues. Adipose tissue is able to store various lipid
soluble steroids, including androgens. Most sex hormones are preferentially concentrated
within the adipose tissue rather than in the blood. As a consequence, excessive fat tissue
stores steroids and leads to maintenance of higher plasma concentrations compared to
normal weight females (Azziz, 1989). Obesity is also considered a condition of increased
13
estrogen production, which correlates with body weight and the amount of body fat
(Kirschner et al., 1990). Excess estrogens may exert a positive feed-back regulation on
gonadotropin release, triggering a rise in ovarian androgen production (Yen, 1980). An
increase in fat tissue is associated with several abnormalities of sex steroid balance,
involving both androgens and estrogens, and the carrier protein, sex hormone-binding
globulin (SHBG). There also is a strong association between obesity and polycystic
ovarian syndrome (PCOS), the most common hyperandrogenic disorder in women (Yen,
1980; Franks, 1989).
Fig. 1.2. Relationship of obesity and metabolic conditions on reproductive abnormalities.
(Adapted from Expert Rev. of Obstet Gynecol, 2008).
14
Leptin. Among the numerous endocrine regulators involved in the integral
control of metabolism, energy stores, and reproduction, the adipose hormone, leptin, has
been universally recognized as an important signal for the proper coupling of fuel
reserves and fertility (Considine et al., 1996). Leptin has been shown to vary directly with
BMI and the percentage of body fat in several species, including women (Banks et al.,
2001; Fors et al., 1999; Garnsworthy et al., 2008; Pasquali and Casimirri, 1993). Elevated
leptin concentrations (hyperleptinemia) are associated with high planes of nutrition and
obesity (Chilliard et al., 2001; Mácajová et al., 2004), and studies suggest that leptin (in
conjunction with glucose and insulin) can be used as a predictor of body condition and
nutritional status (Brandt et al., 2007; Chilliard et al., 2001; León et al. 2004; Mácajová et
al., 2004). Leptin also plays a role in reproduction, with excesses being associated with
obesity and polycystic ovarian syndrome (Brewer and Balen, 2010; Mácajová et al.,
2004). Gentry et al. (2002) first showed that leptin was elevated in mares with a higher
BCS. Conversely, horses with low BCS also had lower concentrations of leptin. The
horses in these studies seemed to fit into 2 distinct groups based on leptin concentrations:
low (<5 ng/mL) or high (7 ng/mL or greater). There are no reports of leptin levels in
elephants, so this study will be the first.
Leptin plays a permissive role in terms of puberty onset and fertility, as threshold
leptin levels (and thus, fat reserves) are mandatory for puberty to proceed. In humans,
the mechanisms of leptin action on reproductive function have been well documented in
15
recent years, indicting the hypothalamus as a key site of action. Leptin receptors are
located within hypothalamic areas associated with appetite control, reproduction, and
growth (Barb et al., 2005). Hence, leptin signals the amount of body fat stores to neural
circuits controlling food intake, but also to a number of neuroendocrine axes, including
the reproductive axis. GnRH neurons do not express leptin receptors, but rather leptin
modulates GnRH release indirectly via interneuron impinging on GnRH-secreting cells in
the hypothalamus. Neuropeptide Y (NPY) fibers are hypothesized to act as
neuroendocrine integrators, linking perturbations in energy balance and alterations in the
activity of the reproductive axis, with the NPY fibers intimately associated with dendrites
and cell bodies of GnRH neurons in the medial preoptic area. Kisspeptin neurons are also
hypothesized key neurons acting as direct targets for leptin within the hypothalamus, and
recognized as being one of the most potent stimulators of GnRH/LH (Tena-Sempere,
2007). However, the interplay between metabolic state and reproductive function is not
well defined and it is possible that other neurons transmit leptin effects onto the GnRHsystem and hence affect ovarian cyclicity.
Insulin and glucose. Central control of reproduction requires the hypothalamus to
receive information regarding energy status of an animal, for example by sensing
hormonal signals secreted into circulation in proportion to body adipose stores. Insulin
plays a central role in the regulation of energy homeostasis, as concentrations are
proportional to adipose tissue in most mammals, and recent studies have confirmed that
insulin sensing in the brain is required for normal reproduction (Boden, 2011; Edmonds
16
et al., 2010). These actions may be mediated by direct insulin action on GnRH neurons or
by altering input from secondary insulin sensitive neurons. Impaired insulin sensitivity
(or insulin resistance) disrupts homeostasis and opens the door to a multitude of serious
and potentially fatal ailments. Insulin regulates energy homeostasis by coordinating
storage, mobilization, and utilization of free fatty acids and glucose in adipose tissue,
liver, and muscle (Ruan and Lodish, 2004). Insulin resistance is a reduced sensitivity of
target cells to insulin to induce glucose uptake into cells (Xu et al., 2003). Blood glucose
concentrations are narrowly regulated mammals, usually between 80 to 90 mg/dL in a
fasted state. Concentrations increase to 120 to 140 mg/dL in the first hour after a meal,
and usually return to resting concentrations within 2 h. Similar concentrations are evident
in the horse; blood glucose concentrations after feed deprivation are normally 60 to 90
mg/dL (Ralston, 2002). In the elephant, glucose concentrations are normally 60-116
mg/dL (Mikota, 2006). When cellular utilization of glucose is impaired, blood glucose
levels are elevated. Pancreatic insulin secretion increases in an attempt to control glucose;
however, over time, high sugar levels lead to health problems associated with diabetes
(e.g., kidney disease, stroke, heart attack, blindness, nerve damage, poor circulation and
wound healing). Many of these problems are causes of morbidity and mortality in captive
elephants (Clubb et al., 2009; Clubb and Mason, 2002). Insulin resistance in horses is
well documented and recent studies associate obesity and insulin resistance in the horse
with development of abnormal reproductive function (Vick et al., 2007). Insulin levels
and insulin resistance have not been previously studied in elephants.
17
Obesity and body condition scoring (BCS)
Once thought to play only a minor role in physiology as a storage depot for
triglycerides, adipose (i.e. fat) tissue is now believed to have very important role as an
endocrine organ and a regulator of metabolism. However, these relationships have not
been investigated in elephants. Therefore, this study will assess body condition (amount
of stored body fat) of elephants to quantify adiposity. A major constraint in determining
this metric in elephants is the lack of a technical approach to accurately assess body fat
and deposition patterns. Standard methods used to measure body fat in humans are
impractical for most zoo animals, as they rely on potentially stressful, costly and invasive
manipulations (e.g., heavy water dilution, hydrostatic weighing). Skinfold calipers have
been widely used in humans, but with mixed results, and these have not yielded good
results in elephants (Wemmer et al., 2006).
Body condition scoring is an indirect means of measuring body condition and has
become an integral part of assessing body fat in veterinary practices. It is widely used in
veterinary medicine as an effective, inexpensive method for quantifying patient
condition. Body condition scoring is a subjective assessment of subcutaneous body fat
stores, based on visual or tactile evaluation of muscle tone and key skeletal elements
(Burkholder, 2000; Otto et al., 1991). Scores are commonly based on an ordinal 5- or 9point scale (Burkholder, 2000; Schröder and Staufenbiel, 2006). Low scores represent
animals with less body fat, and higher scores represent animals with more body fat. Body
condition scoring systems are routinely used in the management and care of many
species, including horses, cattle, sheep, mice and dogs (Henneke et al., 1983). This
18
technique has been useful for assessing the degree of obesity in nondomestic carnivores
(Cook et al., 2001), but relatively few reports exist in the wildlife literature (Gerhart et
al., 1996; Hynes et al., 2004). A visual body condition score, based on direct observation
or assessment of standardized photographic views, is a quick and inexpensive way to
screen for body condition, but needs to be properly validated for every species. For this
study, we developed a BCS index for elephants to assess adiposity and validated the BCS
index with ultrasound measures of subcutaneous fat thickness.
Ultrasound is a widely used method for estimating body composition in domestic
animals, and the techniques used in cattle, pigs, and horses have been validated
extensively (Domecq et al., 1995; Mitchell, 2007). An ultrasound technique for
measuring backfat thickness has been developed in dairy cattle and has been correlated to
total body fat content (Schröder and Staufenbiel, 2006). Ultrasound has been used to
successfully estimate body fat in moose (Alces alces) (Stephenson et al., 1998), elk
(Cervus elaphus) (Cook et al., 2001), mule deer (Odocoileus hemionus) (Bishop et al.,
2009), and woodland caribou (Rangifer tarandus caribou) (Gustine et al., 2010).
Ultrasound imaging has also been used to measure blubber thickness in pinnipeds (Gales
and Burton, 1987; Hall and McConnell, 2007; Mellish et al., 2004), using biopsy
measurements for validation (Mellish et al., 2004). This study will use ultrasound to
measure the thickness of the subcutaneous fat layer in elephants to determine if this
measurement is a good predictor of body condition to validate the visual body condition
scoring index.
19
Policy and conservation implications
Because of the urgency to determine the cause(s) of ovarian acyclicity and other
health problems in elephants, the Elephant TAG/SSP has strongly endorsed this research,
stating that “Validating these [body condition and metabolic biomarkers] techniques as
quickly as possible is critical to the understanding of factors associated with the high
incidence of acyclicity with African elephants in North America”. There is a legitimate
concern within the zoo elephant community that high rates of reproductive problems may
lead to the eventual extinction of the captive population. A goal of this study is to
produce scientific data that will aid decision-making with regard to best practices in
elephant management. A summary of our research findings and recommendations will be
provided to the Elephant TAG Steering Committee, which makes recommendations to
the Association of Zoos and Aquarium (AZA) Board of Directors about revising elephant
standards and guidelines. We expect that the resulting outcomes will impact the
accreditation standards for all AZA institutions, especially given that this is an AZAsanctioned project. AZA zoos are accredited every 5 years and must conform to updated
and revised standards. In this way, the accreditation process drives facility and
management changes in AZA institutions.
Understanding the relationship between metabolic status and reproduction in
elephants is vital so that science-based strategies can be developed to improve fertility
and improve overall health. This study will produce published reference intervals for
several biomarkers that coincide with metabolic status and infertility, data that currently
do not exist for elephants. Our results will provide the basis for designing future studies
20
of metabolic syndrome conditions in elephants. For example, studies can be designed to
more objectively assess the biological effects of altering diets or increasing exercise on
metabolic biomarkers in elephants, with the aim of re-initiating ovarian activity to
promote a sustainable zoo population. In addition, there are numerous other health
problems of elephants suspected of being related to obesity and metabolic disease that
could be studied using these tools: foot problems, arthritis, and cardiovascular disease.
Exhibiting elephants that are healthy and reproductively fit is inspiring to the public, and
essential to the mission of zoos to promote connections between animals and zoo visitors.
This project will result in an understanding of obesity and metabolic health in elephants
to combat a potentially serious problem to promote their continued survival.
21
CHAPTER 2
Development of a body condition scoring index for female African elephants validated by
ultrasound measurements of subcutaneous fat
ABSTRACT
African elephants (Loxodonta africana) in U.S. zoos generally appear heavier than
wild counterparts, and there are claims that obesity-related health and reproductive
problems may be contributing to population non-sustainability. However, a major
constraint in screening for obesity is the lack of a practical method to accurately assess
body fat in elephants. Body condition scoring (BCS) is the assessment of subcutaneous
fat stores based on visual or tactile evaluation of muscle tone and key skeletal elements,
and provides an immediate appraisal of the degree of obesity of an individual. The
objective of this study is to develop a visual BCS index for female African elephants and
validate it using ultrasound measures of subcutaneous fat. To develop the index,
standardized photographs were collected from zoo (n=50) and free-ranging (n=57)
female African elephants to identify key body regions and anatomical sites, which were
used to visually assess body fat deposition patterns. The visual BCS method consists of a
list of body regions and the physical criteria used for obtaining an overall score on a 5point scale, with 1 representing the lowest and 5 representing the highest levels of body
22
fat. Significant correlations were found between the visual scores and ultrasound
measures of subcutaneous fat thickness at all anatomical sites, but were highest in the
backbone (r = 0.748, P < 0.01) and pelvic bone (r = 0.745, P < 0.01) regions, indicating
that BCS adequately reflects the amount of actual fat reserves. The new BCS index
proved to be a reliable and repeatable evaluation method, with a high percentage
agreement (73 % to 95%) and an overall “substantial” strength of agreement determined
by the weighted k statistic (κw = 0.62 to 0.91) between and among three assessors. In
comparing photographs of wild vs. captive elephants, the median BCS in the free-ranging
individuals (BCS=3, range 1-5) was lower (P < 0.001) than that of the zoo population
(BCS=4, range 2-5). Results indicate that this new BCS index could serve as a valuable
tool for assessing the physical state of individual elephants, and examining how body
condition impacts the health and reproductive potential of zoo and free-ranging elephants.
INTRODUCTION
Efforts to maintain a self-sustaining population of African elephants (Loxodonta
africana) in zoos have met with limited success (Wiese and Willis, 2006), in part because
of high mortality and low reproduction; over the past decade, there have been only three
births to five deaths annually in the U.S. (Faust and Marti, 2011). If this trend continues,
the captive population will pass through a bottleneck in 30 years, and could become
demographically nonviable in about 50 years (Faust and Marti, 2011). Contributing to
this problem is a dramatic increase in the rate of ovarian cycle problems, which in
African elephants has increased from 25% to 46% in just 7 years (Dow et al., 2011). To
23
prevent further population declines, the Association of Zoos and Aquariums Elephant
Taxon Advisory Group/Species Survival Plan® (AZA TAG/SSP) Management
Committee has endorsed research to better understand causes of poor health and
reproduction of African elephants (Keele and Ediger, 2011). One condition that is
suspected of having a significant effect on both of these problems is excessive body
weight (Clubb et al., 2008; Clubb et al., 2009).
Studies in horses (often used as a model for elephants) and women show that
being overweight can lead to biological changes related to metabolic conditions (i.e.,
insulin resistance), arthritis, foot problems and impaired fertility, and that many of these
can be alleviated by implementing weight reduction programs (e.g., reduced caloric
intake and/or exercise) (Clark et al., 1995; Hill et al., 2008; Lievense et al, 2002; Miller et
al., 2008). One problem in zoo elephants is poor foot health and arthritis (Clubb and
Mason, 2002; Clubb et al., 2008, 2009; Lewis et al., 2010), with a strong inverse
relationship existing between foot pathology and levels of exercise (Lewis et al., 2010).
Ovarian acyclicity also is a serious reproductive problem in captive African elephants,
and a previous study found a relationship between a high body mass index and ovarian
acyclicity in this species (Freeman et al., 2009). In women, the incidence of infertility is
significantly higher in obese compared to normal weight females (Pasquali et al., 2003).
Obese women also have an increased risk of high fetal birth weights, miscarriage and
maternal death (Bray, 1997), all of which are causes of infant and maternal mortality in
elephants (Clubb et al., 2009). Circumstantial, evidence suggests there may be a link
between obesity-related changes and poor health and reproduction in elephants.
24
A major constraint in screening for obesity in African elephants is the lack of a
practical method to accurately assess body condition. Standard methods for measuring
body fat in humans are impractical for elephants, as they rely on potentially stressful,
costly and invasive manipulations (e.g., heavy water dilution, hydrostatic weighing)
(Deurenberg and Yap, 1999; Ellis et al., 2000). Skinfold calipers are widely used in
humans, but with conflicting results, and these have not yielded good results in elephants
(Wemmer et al., 2006). An alternate method used in veterinary medicine is body
condition scoring (BCS), a subjective assessment of subcutaneous body fat stores based
on visual or tactile evaluation of muscle tone and key skeletal elements (Burkholder,
2000; Otto et al., 1991). BCS methods rely on a numeric scoring system, with higher
scores representing animals with more body fat. A variety of BCS systems have been
developed for farm, exotic, and research animal management (Burkholder, 2000;
Henneke et al, 1983; Schröder and Staufenbiel, 2006), including an 11-point scale
developed for Asian elephants (Wemmer et al., 2006).
Validating BCS methods often relies on ultrasound measures of actual fat thickness,
an approach used in a number of domestic and non-domestic species, including cattle
(Schröder and Staufenbiel, 2006), moose (Alces alces) (Stephenson et al., 1998), elk
(Cervus elaphus) (Cook et al., 2001), mule deer (Odocoileus hemionus) (Bishop et al.,
2009), woodland caribou (Rangifer tarandus caribou) (Gustine et al., 2010), and
pinnepeds (Gales and Burton, 1987; Hall and McConnell, 2007; Mellish et al., 2004). The
AZA Standards for Elephant Management and Care (AZA, 2012) recommends the BCS
system of Wemmer et al. (2006) be used for routine assessment of body condition in
25
Asian elephants, although it has yet to be validated ultrasonographically. Furthermore, no
recommendations are provided for assessing body condition in African elephants. Given
the health and reproductive problems facing African elephants in zoos (Brown et al.,
2004a; Clubb et al., 2009; Clubb et al., 2008; Dow et al., 2011; Freeman et al. 2009;
Lewis et al., 2010; Proctor et al., 2010), especially the high rate of ovarian acyclicity, a
reliable and validated method to assess body condition for this species is needed.
The objective of this study was to develop a simple, reliable method to score body
condition in African elephants using either direct observations or the assessment of
standardized photographic views, and validate it by ultrasound measures of subcutaneous
fat. Ultimately, the goal is to have a validated BCS index to determine if excessive body
weight is related to morbidity, mortality or infertility in zoo African elephants.
MATERIALS AND METHODS
Body condition scoring index
This study was approved by Institution Animal Care and Use Committees of the
Smithsonian Conservation Biology Institute (SCBI) and all participating zoos. A total of
107 elephants were utilized in the development of the BCS index, about half (n=50; age
range, 10 - 45 years) of which were housed in AZA-accredited zoos. Each participating
institution submitted a series of photographs taken by zoo staff using a “Photographing
Guide” (Appendix 1) that provided detailed instructions for obtaining a set of three
standardized photographs of each elephant from three angles (rear view, side view, and
26
rear-angle view). To compare the body condition of captive vs. free-ranging African
elephant females, similar sets of photographs from a photographic dataset at South
Africa‟s Kruger National Park (KNP), Tanda Tula Camp were examined (n = 57; age
range, 10 – 45 years). Photographs of free-ranging elephants were age-matched and
chosen randomly from the database to provide a representative sample without bias for
comparison to the captive population. Using photographs from captive and free-ranging
elephants, key body areas were identified to serve as the anatomical regions for assessing
body fat deposition patterns (Fig. 2.1). From these, the BCS Index was created (Table
2.1), as was a BCS Flow Chart (Fig. 2.2) to systematically guide an assessor through the
scoring process.
27
VR
LD
PB
R
Figure 2.1. Key areas for visually assessing body condition in female African elephants.
R = ribs, PB = pelvic bone, VR= vertebral ridge of backbone, LD= lumbar depression
alongside backbone.
28
Table 2.1. African elephant body condition scoring (BCS) index.
BCS
Description and example photographs
Ribs: Clearly visible
Pelvic Bone: Protrudes, deep depression in front and depression or flattened
area behind pelvic bone
Backbone: Prominent from tail head to shoulders, deep depression alongside backbone in
lumbar region
1
Ribs: Not visible and appear to be covered by a very thin fat layer
Pelvic Bone: Clearly visible, gradual sunken area in front and flattened area behind pelvic bone
Backbone: Clearly visible from tail head to mid-back, depression alongside
backbone in lumbar region
2
Ribs: Not visible
Pelvic Bone: Visible as a ridge, entire pelvic bone may not be visible, slight sunken or flattened
area in front and/or behind pelvic bone
Backbone: Visible from tail head to mid-back, sloping alongside backbone in lumbar region
3
29
Ribs: Not visible
Pelvic Bone: Not visible
Backbone: Visible as ridge from tail head to mid-back, no obvious depression and fat beginning
to accumulate alongside backbone in lumbar region
4
Ribs: Not visible
Pelvic Bone: Not visible
Backbone: Difficult to differentiate, may be visible from tail head to pelvic bone region and
appears to be covered with a thin fat layer; area alongside backbone in lumbar region filled in
5
30
Fig.2.2. Body condition scoring flow chart for female African elephants.
31
Ultrasound measures of subcutaneous fat
A subset of 33 captive adult female African elephants at 12 AZA-accredited zoos
was used for ultrasound measures of subcutaneous fat thickness to validate the visual
BCS index. Anatomical regions for assessing body fat reserve depositions were: 1)
directly along the vertebral ridge; 2) lumbar depression alongside the vertebral ridge; 3)
directly on the iliac crest (pelvic bone); 4) in front of the iliac crest (pelvic bone); and 5)
behind the iliac crest (pelvic bone) (Fig. 2.1). From a preliminary study, the „ribs‟ were
found to be too difficult to locate consistently, especially in elephants with more fat
reserves, and so were not included in the ultrasound validations.
An ultrasonographic unit (A-scan mode) (Preg-Alert Pro, Renco Corporation, MN,
USA) was used to measure subcutaneous fat thickness. This equipment was specifically
calibrated for use in elephants to measure fat depth up to 20 cm. Ultrasound
measurements were obtained by one individual (KM) throughout the study. To perform
the exam, ultrasound gel was applied to the desired location and five measurements were
obtained at each body region. The mean thickness of the subcutaneous fat layer was
calculated for each location and an overall mean of the five body regions for each
elephant was calculated. Ultrasound measurements were obtained for either the right or
left side for regions involving the pelvic bone and lumbar depression, whereas vertebral
ridge measurements were obtained at the body midline. Ultrasound assessments (n = 5
each) were conducted in the following order: 1) vertebral ridge – the length of the lumbar
region of the vertebral ridge, spanning a length of approximately 10 inches; 2) lumbar
depression - 2-3 cm to either side of the vertebral ridge, spanning a length of
32
approximately 10 inches; 3) pelvic bone - probe oriented directly on the iliac crest and
moved along the entire length; 4 and 5) in front of and behind the pelvic bone,
respectively - probe was placed 2 to 3 cm on either side (front or behind) of the iliac crest
and moved around these regions spanning a length of approximately 5 inches.
Measurements were obtained by holding the transducer for 2-3 seconds in each location
until the subcutaneous fat layer measurement (in mm) was displayed.
Reliability study
To determine inter-assessor and intra-assessor reliability of the BCS method, three
assessors scored sets of photographs (side view, rear-angle view, and rear view) in a pilot
study of 40 captive African elephants. Raters included a graduate student that developed
the BCS index (Assessor A), an elephant specialist who has worked with elephants for
>30 years and contributed to the development of the Asian elephant BCS index
(Wemmer et al., 2006) (Assessor B), and an animal behaviorist with no prior experience
in scoring body condition of any species (Assessor C). Each assessor independently
scored the photographic sets twice within a 1-month period. Assessors remained blinded
to the previous scores and were not permitted to discuss study results.
Statistical methods
The five ultrasound measurements were averaged to provide a mean ultrasound
measurement of fat thickness for each body region. An overall mean ultrasound score for
all five body regions for each elephant was then calculated. The Cochran-Mantel-
33
Haenszel statistic was used to test for linear trend by testing the null hypothesis that BCS
was randomly distributed between the free-ranging and captive groups. The relationships
between BCS and ultrasound measurements were investigated using regression analyses.
Spearman correlation coefficients were used to determine relationships among BCS and
ultrasound measurements of the five body regions. Correlations were considered to be
different from zero at P < 0.01.
For the reliability study, the overall percentage (%) agreement between paired intraand inter-assessor assessments was calculated as (100 × m)/n, where n = total number of
samples examined and m = number of cases of exact agreement. A weighted kappa (κw)
statistic was also used to analyze intra- and inter-assessor variability (Cohen, 1968).
Standards proposed by Lanidis and Koch (1977) were used to interpret resulting kappa
values, where perfect agreement equates to a kappa of 1 and chance agreement equates to
0. The following standards for interpreting kappa values for strength of agreement were
used: kappa values ≤ 0 = poor, 0.10 to 0.20 = slight, 0.21 to 0.40 = fair, 0.41 to .60 =
moderate, 0.61 to 0.80 = substantial and 0.81 to 1.00 = almost perfect agreement.
RESULTS
Body condition scoring of captive and free-ranging elephants
There is strong evidence for a very strong correlation between BCS and group
(captive vs. free-ranging) (Table 2.2; P = 0.0001). There is an increasing trend in BCS for
captive elephants and a decreasing trend BCS for free-ranging elephants across levels of
34
BCS. The median BCS for elephants in the captive population (n = 50) was a 4 (range 25, Table 2.2), and higher than that for free-ranging elephants (n = 57) (BCS=3, range 1-5,
Table 2.2). Graphical representation of the percentage of elephants in each BCS category
for captive and free-ranging elephants is shown in Figure 2.3. The most prevalent (mode)
BCS observed in the free-ranging study population was a 2 (39%, n=22), lower than the
BCS=5 (40%, n=20) observed for the captive population. None of the captive elephants
had a BCS=1, whereas 9% (n=5) of the free-ranging elephants scored in that category.
Table 2.2. Body condition scores (BCS) (and relative percentage) of free-ranging and
captive female African elephants in each BCS category.
P value
Free-ranging elephants
Captive elephants
BCS
Number (%)
Number (%)
1
5 (9)
0 (0)
2
22 (39)
2 (4)
3
18 (33)
12 (24)
4
10 (18)
16 (32)
5
2 (4)
20 (40)
57
50
3 (1-5)
4 (2-5)
3.74 (0.96)
4.39 (0.58)
Total
Median (range)
Mean (SD)
*Cochran-Mantel-Haenszel test for linear trend
35
0.0001*
Fig 2.3. Percentage of free-ranging (n=57) and captive (n=50) female African elephants
in each body condition score category.
Ultrasound validation
Table 2.3 shows the ultrasound measurements of mean fat thickness for each body
region and the mean of all five regions for elephants of different BCSs. Regression
analyses indicated all models were significant in explaining the variation in BCS.
Ultrasound measurement of fat thickness of all five body regions and for the means of the
five regions were significantly associated with BCS (Table 2.4, P < 0.01). As the BCS
increased, the fat thickness also increased (P < 0.01), indicating that BCS adequately
reflected the amount of fat reserves.
36
Table 2.3. Mean (SD) ultrasound measurements of subcutaneous fat thickness by body region and body condition score (BCS)
category in zoo African elephants (N = 33).
Subcutaneous fat thickness (mm)
BCS1
Lumbar depression
Pelvic bone
Pelvic bone-front
2 (N=2)
9.90 (1.41)
10.90 (0.14)
11.50 (2.40)
11.20 (1.31)
11.10 (1.56)
10.92 (1.41)
3 (N=8)
12.25 (1.41)
12.33 (2.04)
12.08 (1.35)
12.15 (1.21)
12.10 (1.74)
12.18 (0.99)
4 (N=15)
12.83 (1.55)
13.06 (1.56)
13.55 (2.07)
13.80 (2.18)
13.69 (2.75)
13.39 (1.13)
5 (N=8)
21.75 (7.46)
20.80 (6.99)
20.03 (5.28)
22.20 (6.63)
19.08 (6.34)
20.77 (4.33)
P-value
0.0003*
0.0003*
0.007*
0.0023*
0.0007*
0.0008*
1
Pelvic bone-back
Mean of all body
regions
Vertebral ridge
BSC (1=thinnest to 5=most body fat); no zoo elephant had a BCS=1.
Resulting P-value from regression analyses with each body region and the means of all body regions analyzed separately
37
Spearman correlation coefficients for comparison of BSC with the five body
regions and mean of the five regions evaluated by ultrasound are shown in Table 2.4.
There were positive correlations (P < 0.01) between the BCS and mean ultrasound
measures of fat thickness at each body region, and for the overall mean of the five
regions. Of the five specific body regions, the vertebral ridge had the strongest
correlation to BCS (r = 0.748, P < 0.0001), and the mean of the five regions showed an
even stronger correlation (r = 0.816, P < 0.0001). A scatter plot of BCS and the mean
ultrasound measurement of fat thickness for the five body regions is presented in Fig. 2.4,
which shows that the mean ultrasound measures of fat thickness increased as BCS
increased.
38
Table 2.4. Spearman correlations among body condition score (BCS) and mean
ultrasound measurements of fat thickness of specific body regions.1
BCS
BCS
VR
LD
PB
PB-F
VR
LD
PB
PB-F
PB-B
Mean
1.000 0.748* 0.707* 0.665* 0.745* 0.658* 0.816*
1.000
0.615* 0.549* 0.657* 0.623* 0.791*
1.000
0.555* 0.717* 0.490* 0.794*
1.000
0.725* 0.562* 0.769*
1.000
PB-B
0.615* 0.897*
1.000
Mean
0.809*
1.000
1
VR=vertebral ridge, LD= depression alongside lumbar region of backbone, PB=directly
on pelvic bone, PB-F=in front of pelvic bone, PB-B=behind pelvic bone
Mean=mean of all body regions
*Significant (P < 0.01)
39
40
Fig. 2.4. Scatter plot of body condition scores and overall mean ultrasound
measurements of fat thickness for all five body regions combined in female African
elephants.
Reliability study
Intra-assessor reliability: The level of percentage agreement ranged from 88-95%
between repeat BCS assessments of the photograph set of 40 captive elephants for the
three assessors (Table 2.5). Weighted kappa values ranged from 0.8-0.9, and the
interpretation of κw values for repeat assessment for all assessors was “excellent”,
according to the criteria of Landis and Koch (1997).
Inter-assessor reliability: The level of percentage agreement for assigning a BCS to
the set of 40 elephants among assessors ranged from 73%-93%, with the greatest
agreement between assessors A and B (Table 2.5). Weighted kappa values for
41
assessments between assessors A and B were interpreted as “almost perfect” agreement,
whereas all other inter-assessor agreements were interpreted as “substantial” agreement
when applying the methods of Landis and Koch.
42
Table 2.5. Level of intra-and inter-assessor agreement for assessment of elephant body condition.
Intra-assessor agreement
Inter-assessor agreement
Assessor
A
B
C
A and B
A and C
B and C
Percentage (%)
agreement
95
90
88
93
73
73
0.81 (0.74-0.88)
0.82 (0.78-0.86)
0.89 (0.82-0.96)
0.67 (0.58-0.76)
0.62 (0.52-0.72)
κw (95% CI)
0.93 (0.88-0.98)
κw = weighted kappa; 95% CI = 95% confidence interval
43
DISCUSSION
A new visual BCS index was developed for assessing body fat and condition in
female African elephants. The scoring method consists of a list of body regions and the
physical criteria used for obtaining an overall score on a 5-point scale, with 1 being the
least and 5 being the most body fat. The index includes example photographs of
elephants representing each BCS, and a „Body Condition Scoring Flow Chart‟ to
systematically guide an assessor through the scoring process when using direct
observations or sets of standardized photos. The BSC was validated by ultrasound
measures of actual body fat, and found to be a simple, reliable assessment tool.
A BCS system (1-6 scale) was previously developed for African elephant bulls
(Poole, 1989); however, that publication did not include photographs representing each
body condition category and only brief anatomical descriptions were provided. By
contrast, the Wemmer index for Asian elephants (Wemmer et al., 2006) included
numerous photographs of elephants at each BCS, but was based on a rather extensive
scoring system - six regions of the body scored using two or three criteria per region, and
then the six scores totaled to obtain an overall score ranging from 0-11 (where 0 is the
thinnest). Our goal was to simplify the Wemmer technique by reducing the number of
scoring criteria, and to validate the method with direct measures of fat thickness using
ultrasound, something that was not done by Wemmer. First, we excluded several body
regions (e.g., head and shoulder regions) because no consistent variations in fat thickness
were observed in these areas, especially for elephants with a higher BCS. Second, the
Wemmer BCS index requires observing the entire elephant from all angles and adding
44
the scores from specific body regions, which can be difficult under field conditions. By
contrast, our new BCS index does not require observing the head or shoulder regions, and
an overall BCS can be assigned cumulatively by assessing only the backbone, pelvic
bone, and rib areas.
The new BCS index was validated with ultrasound measures of subcutaneous fat
thickness, and significant positive correlations were observed at all five body regions. As
the BCS increased, subcutaneous fat thickness increased, similar to that described for
other species (Bishop et al. 2009; Domecq et al., 1995; Gales and Burton, 1987; Hall and
McConnell, 2007; Kearns et al., 2002; Westervelt et al., 1976; Wilkinson and McEwan,
1991). These findings suggest that appropriate body regions were identified for
developing the new BCS index. The backbone region (vertebral ridge) provided the
highest correlation to the BCS (r = 0.748, P < 0.001, Table 2.4), although all correlations
were statistically significant. Thus, any of the designated body regions could be used as a
single validated indicator of body condition. In humans, certain anatomical regions such
as the abdominal region, give a better prediction of overall body fat than others (Fanelli
and Kuczmarski, 1984; Volz and Ostrove, 1984), and the same has been found for cattle
(Schröder and Staufenbiel, 2006), horses (Carter et al., 2009) and dogs (Wilkinson and
McEwan, 1991). Recently, ultrasound was used to validate a new visual 9-pt BCS index
for Asian elephants (Treiber et al., 2012). That study assessed ultrasonic fat thickness in
the rump region only (n = 12), whereas we evaluated ultrasound fat thickness at multiple
body regions. Nevertheless, results were comparable in both studies in showing a
significant relationship between measured fat thickness and BCS. The median ultrasonic
45
measure of subcutaneous rump fat in the Asian elephants was 1.07 cm (range, 0.41-3.11
cm) and the median BCS was 6.25. These findings are in accordance with our
measurements of fat thickness at the most comparable region assessed (directly behind
the pelvic bone), which had a median fat thickness of 1.31 cm (range, 1.0-3.28 cm). Our
ultrasound fat thickness measures observed in elephants with high BCS (~2.5 cm) were
also comparable to the highest fat thickness reported in other herbivore species, including
cattle (~3.5 cm) (Schroder and Staufenbiel, 2006), horses (~3.1 cm) (Westervelt et al.,
1976), elk (~3.7 cm) (Cook et al., 2010; Gustine et al., 2006;), deer (~3.5 cm) (Cook et
al., 2010), and moose (~7 cm) (Cook et al., 2010; Stephenson et al., 1998).
The new BCS index proved to be a reliable method for assessing female African
elephant body condition based on high intra- and inter-assessor agreement across three
assessors. Determining the accuracy of a new diagnostic test can be difficult when there
is no accepted reference or “gold” standard, and none of the previous elephant BCS
studies tested assessor reliability. Two types of reliability exist: agreement between
assessments made by two or more assessors (inter-assessor reliability); and agreement
between assessments made by the same assessor on two or more occasions (intra-assessor
reliability). Agreement can be measured using several statistics: percent agreement to
provide an overall agreement rate; and the kappa statistic, which is a measure of
agreement that indicates the proportion of agreement expected by chance. A weighted
kappa was used to reflect the degree of disagreement so that a greater emphasis was
placed on large differences between or among assessments compared to small
differences, which is commonly adopted in ordinal scale reliability assessments (Cohen,
46
1968). The resulting weighted kappa value was then interpreted in terms of the strength
of agreement among assessments. Landis and Koch proposed the following as standards
to determine the strength of agreement in reliability studies: kappa values ≤ 0 = poor, 0.10.20 = slight, 0.21 - 0.40 = fair, 0.41 - 0.60 = moderate, 0.61 - 0.80 = substantial and 0
.81 – 1 = almost perfect agreement. Similar formulations exist to interpret weighted
kappa results for reliability studies, but with slightly different descriptors. For example,
an alternate scale was developed by Fleiss (1981), where kappa value < 0.41=poor,
<0.75-0.41=good-fair, and >0.74-1.0=excellent agreement. We chose the formulation of
Landis and Koch because it is often cited in studies assessing reliability in body condition
scoring systems (Clingerman and Summers, 2012; Phythian et al., 2012). Regardless of
the scale used, kappa values above 0.40 are considered to be clinically useful (Sim and
Wright, 2005), and all assessments in our study had values above this clinically relevant
level. The current study demonstrated high percentage agreement (73% to 95%) and an
overall “substantial” strength of agreement determined by the weighted k statistic (κw =
0.62 to 0.91) between and among assessors using the new BCS index.
Body condition scoring has become an integral part of assessing body fat in
veterinary practices because it is an effective and inexpensive way to quantify patient
condition. In other species, BCS indexes are used to categorize an individual in terms of
body fat (i.e. thin, normal, overweight, obese), and within the same species various BCS
methods may be employed with the middle score commonly representing the “ideal” or
“normal” body condition. For example, two numeric scales are typically used and
accepted in veterinary practices for assessing body condition in dogs (5-pt and 9-pt
47
scales) (Gebin et al., 2012; Laflamme, 1997;). When using a 5-point scale, the
“ideal/normal” BCS = 3, BCS = 1-2 equates to “underweight/thin” and
“overweight/obese” includes BCS = 4-5. When using a 9-point scale, the “ideal/normal”
BCS = 4-5, whereas “underweight/thin” is represented by BCS = 1-3 and
“overweight/obese” include BCS = 6-9. Finally, in cattle, both 9-point (Lassen et al.,
2003;Wagner et al., 1988) and 5-point (Edmonson et al., 1989; Wildman et al., 1982)
scales are used, and the middle scores represent the “ideal” distribution of body fat.
Based on these criteria, a BCS = 4 or 5 in elephants using our 5-point scale would equate
to “overweight/obese” categories, whereas a BCS = 3 would indicate an “ideal/normal”
body condition. If this classification holds, 40% (n=20) of the elephants in our zoo study
were obese, differing from the free-ranging elephants where only 4% (n=2) of the study
population had a BCS=5. Furthermore, 32% (n=16) of zoo elephants could be classified
as overweight (BCS = 4), whereas only 18% (n=10) of free-ranging elephants scored in
this category. Last, only 12 zoo elephants (24%) were in “ideal” condition (BCS = 3) and
two (4%) were “thin” (BCS = 2), compared to 72% of wild African elephants that had a
BCS of 2 or 3. We used the free-ranging population in South Africa‟s Kruger National
Park (KNP), Tanda Tula Camp as a reference or benchmark because the elephants are
protected, and in general the habitat quality is good (Ganswindt, 2008; de Knegt et al.,
2011). In addition, Save the Elephants at Tanda Tula has an existing and extensive
photographic database on hundreds of elephants. Photographs were chosen randomly and
across seasons with standard views to match those of the captive population study. While
additional scoring of wild African elephants certainly is needed to define optimal or
48
suboptimal body condition status under varied conditions, the data are beginning to
suggest that captive elephants are on the higher end of the scale, and thus may not
represent the “ideal/normal” condition. Further investigations also are warranted to define
elephant obesity in terms of associated health complications, similar to what has been
done in humans, horses, and other species (Frank et al., 2006, 2010; Lievense et al., 2002;
Lund et al., 2005, 2006; Nguyen et al., 2008; Vick et al., 2006). This new BCS index
should enable better monitoring of body condition in individual and populations of
elephants, allowing for more directed medical and management decisions based on
scientific data and validated tools. For example, managers can monitor the effects of
changes in diet formulations or exercise programs on BCS to optimize zoo husbandry
practices and promote a more healthy and sustainable population.
Ultimately, our goal was to develop an easy and reliable means of assessing BCS
because of the need to investigate possible relationships between obesity and health and
reproductive problems in zoo-managed elephants (Clubb, 2002; Clubb et al., 2009;
Mason and Veasey, 2010). The new BCS method includes examples photographs of
elephants representing each BCS category and a flow chart to systematically guide an
assessor through the body condition scoring process. The relationships between BCS and
ultrasound measurements determined that scores adequately reflect the amount of
subcutaneous fat of elephants, and that most zoo elephants are in the higher condition
categories. Given the need to increase reproduction in the captive population (Faust and
Marti, 2011), the Endocrinology Laboratory at SCBI has served the Elephant TAG/SSP
since 1996 by identifying reproductively viable females through routine serum
49
progestagen assessments. The observation that 30% of zoo-managed African elephants
maintain consistent baseline levels of progestagens, indicative of complete ovarian
inactivity, and another 16% exhibit irregular cycles is a serious concern, especially
because over three quarters of abnormal cycling females are of reproductive age (Dow et
al., 2011). The link between a high BMI and ovarian acyclicity (Freeman et al., 2009),
suggests that reproductive problems may be caused in part by metabolic derangements
associated with excessive body fat. This seems possible given studies in horses and
humans indicate obesity can lead to metabolic changes that are correlated with impaired
fertility (Bray, 1997; Chang, 2007; Hartz et al., 1979; Irvine and Shaw, 2003; Pasquali et
al., 2003; Norman and Clark, 1998). Our new elephant BCS index is a quick and reliable
tool that has been validated to assess body condition in female African elephants for
routine health monitoring. By instituting obesity screening of zoo elephants, facilities can
monitor and potentially identify at-risk elephants in time to mitigate health and
reproductive problems.
50
CHAPTER 3
High body condition and elevated insulin and leptin levels are associated with ovarian
acyclicity in zoo African elephants
ABSTRACT
A previous study found a high body mass index (BMI) was positively correlated
with ovarian inactivity in African elephants, suggesting reproductive problems may be
caused in part by metabolic changes associated with excessive body weight. This seems
possible given studies in horses and humans that show obesity can lead to metabolic
changes that impair fertility. This study measured body condition, concentrations of
insulin, glucose, and leptin of breeding-aged cycling (n=23) and non-cycling (n=23)
females. To assess body condition, a new body condition score index (BCS; 5-point scale
with 1 = thinnest and 5 = fattest), validated with ultrasound measures of fat depth, was
used. The mean BCS of non-cycling elephants was higher than that of cycling elephants
[4.39 (SD 0.58) vs. 3.73 (SD 0.93), P = 0.009].There were differences in the serum
concentrations of leptin [4.03 ng/mL (SD 1.63) vs. 3.16 ng/mL (SD 0.88); P = 0.041] and
insulin [0.65 mg/mL (SD 0.31) vs. 0.48 mg/mL (SD 0.18); P = 0.032] for non-cycling
females in the BCS=5 category. Serum glucose did not differ between cycling and noncycling elephants (P = 0.892); however, the G:I was lower in the non-cycling group
51
[69.82 (SD 53.21) vs. 227.18 (SD 96.23); P = 0.019]. Using “non-cycling” as the
outcome variable in regression models, we examined BCS, leptin, insulin, and the G:I
ratio as predictors, and found that BCS showed the strongest predictive power (P = 0.01).
The odds ratio for the BCS coefficient is 3.15 with a 95% confidence interval of [1.36,
7.26], indicating that with each 1 point increase in BCS an elephant is approximately 3
times more likely to be non-cycling. These results suggest that ovarian acyclicity is
associated with a high BCS and perturbations in markers of metabolic status. Thus,
monitoring insulin, glucose, and leptin may assist in identifying elephants at risk for
developing metabolic health conditions, including fertility problems. This would allow
for management interventions to be implemented to improve body condition and
metabolic status with the goal of reinitiating ovarian activity and promoting a healthier
zoo population.
INTRODUCTION
Poor reproductive success due to ovarian cycle disruptions in zoo African elephants
(Loxodonta africana) is a problem that has been recognized for over a decade (Brown,
2000). In a recent survey, 46% of zoo-managed female African elephants of reproductive
age exhibited abnormal ovarian activity, ranging from irregular cycles to a complete lack
of cyclicity (Dow et al., 2011). Studies of cycle abnormalities have yet to find a definitive
cause, although it appears they are not related to elevated cortisol (Proctor et al., 2010),
altered androgen production (Mouttham et al., 2011), thyroid problems (Brown et al.,
2004b), or pituitary gonadotropin function (Brown et al., 2004b).
52
There is growing recognition and concern that obesity and metabolic conditions are
negatively impacting the health of many species, including humans, companion and
domestic animals. A similar health concern may exist for zoo-held species, including
elephants, that often are fed diets high in calories and given inadequate exercise (Ange et
al., 2001; Clubb et al., 2009; Hatt and Clauss, 2006; Lewis et al., 2010). A recent study
found a high body mass index (BMI) is positively correlated with ovarian inactivity in
African elephants (Freeman et al., 2009), suggesting reproductive problems may be
caused in part by metabolic derangements associated with excessive body fat. This seems
plausible given studies in horses and humans that show obesity can lead to metabolic
changes that impair fertility (Bray, 1997; Chang, 2007; Clark et al., 1998; Hartz et al.,
1979; Irvine and Shaw, 2003; Norman and Clark, 1998; Pasquali et al., 2003). For
example, obese mares experience an extended interval between successive ovulations
(Vick et al., 2006), not unlike the irregular cycles observed in elephants (Brown, 2000).
Stillbirths and dystocias also are common in obese women and horses (Althabe, 2012;
Vick et al., 2006), and are causes of calf mortality in zoo elephants (Clubb et al., 2009;
Clubb and Mason, 2002; Dale, 2009; Mason and Veasey, 2010). Such evidence suggests
elephants may be experiencing reproductive problems associated with excessive body
weight, including ovarian acyclicity. Thus, investigations that compare obesity-related
metabolic hormones in cycling and non-cycling elephants may help identify factors that
impact reproductive potential.
In the equine field, various methods have been developed for assessing obesity,
including visual body condition scoring (BCS), calculating BMI, and use of ultrasound to
53
measure subcutaneous fat thickness to predict total body fat content (Donaldson et al.,
2004; Henneke et al., 1983; Kearns et al., 2002). In particular, BCS has become an
integral part of assessing body fat in veterinary practices as an indirect, effective, and
inexpensive method for quantifying patient condition. Other methodologies to assess
altered metabolic states associated with obesity include measuring levels of glucose,
insulin, and leptin, and calculating a glucose-to-insulin ratio (G:I). Impaired insulin
sensitivity (insulin resistance) leads to a multitude of ailments related to high blood sugar
(glucose) and diabetes (e.g., kidney disease, stroke, heart attack, blindness, nerve
damage, poor circulation and wound healing), some of which are reported causes of
morbidity and mortality in zoo elephants (Clubb and Mason, 2002; Clubb et al., 2009;
Mason and Veasey, 2010). The relationship between insulin resistance and infertility has
been well documented in women and horses (Hartz et al., 1979; Johnson, 2002; Pasquali
et al., 2003; Vick et al., 2006), but to date has not been examined in elephants.
Leptin, a protein produced by adipose tissue, is important in the regulation of food
intake and energy balance (Considine et al., 1996), and also plays a role in reproduction
with excess levels being associated with obesity and ovarian dysfunction (Brewer and
Balen, 2004; Mácajová et al., 2004). Leptin has been shown to vary directly with the
percentage of body fat in several species (Banks et al., 2001; Fors et al., 1999;
Garnsworthy et al., 2008; Pasquali and Casimirri, 1993), and in women, elevated serum
leptin is a sign of energy imbalance, poor diet, insulin resistance, or changes in other
metabolic risk factors (Leyva et al., 1998; Martins et al., 2012). Taken together, it is not
unreasonable to question whether that obesity and altered metabolic states may also be, at
54
least in part, responsible for the reproductive problems observed in a high percentage of
female African elephants.
The objective of this study is to determine whether excessive body fat and altered
metabolic hormone levels are associated with ovarian acyclicity in zoo African elephants.
The hypothesis is that non-cycling African elephant females display more characteristics
of metabolic aberrations than normal-cycling zoo counterparts. The study goals were to:
1) compare measures of body and metabolic condition (BCS, leptin, glucose, insulin, G:I
ratio) between cycling and non-cycling elephants; 2) compare metabolic hormone
concentrations across BCS categories in cycling and non-cycling elephants; and 3)
determine if obesity and altered metabolic hormone concentrations are risk factors for
abnormal ovarian function. An improved knowledge of factors contributing to poor
reproduction might enable the development of management strategies to mitigate obesity
and health or reproductive problems associated metabolic conditions.
MATERIALS AND METHODS
Animals and sample collection
Reproductive-aged female African elephants housed in Association of Zoo and
Aquariums (AZA) accredited-zoos that exhibited normal ovarian cycles served as the
control, reference population (n = 23 elephants at 12 zoos; mean age, 31.6 years; age
range, 22-44 years). Non-cycling, breeding-aged female elephants (n = 23 elephants at 10
zoos; mean age, 33.7 years; age range, 24-44 years) served as the experimental
55
population. This study was approved by Institution Animal Care and Use Committees of
the Smithsonian Conservation Biology Institute (SCBI) and all participating zoos.
To assess body condition, photographs were used to assign a BCS to each
elephant using a recently validated BCS index (Chapter 2). Participating zoos were
provided a photographic guide containing detailed instructions on how to obtain three
standardized photos for each elephant. To assess metabolic hormone status, five serum
samples were collected weekly for 5 weeks from each elephant (cycling and non-cycling)
within approximately 1 month of obtaining photographs. Blood was collected from an ear
vein by collaborating veterinarians at the facilities in which the animals resided. All
elephants were well-conditioned to the blood sampling procedure, which was part of the
normal management routine. Blood was maintained at ~4˚C and centrifuged within a few
hours of collection. Serum samples were stored at -20˚C or colder until analysis.
Serum analyses
Leptin. Serum leptin concentration was measured using a multi-species doubleantibody radioimmunoassay (RIA) (Xl-85K, Linco Research Inc., St. Louis, MO, USA)
that relies on an 125I-human leptin tracer and a guinea pig anti-human leptin antiserum.
Serum samples (100 μl), analyzed in duplicate, were incubated with 100 μl antiserum at
4°C overnight; then 100 μl of 125I-leptin was added to each tube and incubated at 4°C for
18-24 hr. The following day, 1.0 ml of cold precipitating reagent (goat anti-guinea pig
IgG serum, 3% polyethylene glycol and 0.05% Triton X-100 in 0.05 M phosphatebuffered saline) was added to all tubes except those for measurements of total counts.
56
Tubes were vortexed, incubated at 4°C for 20 min, centrifuged at 4°C, and the resulting
supernatant decanted without disturbing the precipitate. Bound radioactivity was
measured for 1 minute in a gamma counter (IsoData 20, New York, New York, USA).
The leptin RIA was validated for use in African elephant serum by demonstrating: 1)
parallelism between the leptin standard curve and serial serum samples (100 μl; 1:2, 1:4,
and 1:8) diluted with kit assay buffer (Fig. 3.1); and 2) significant recovery (94%;
y=1.089x – 0.531) of leptin standard (50 μl; 1.56, 3.13, 6.25, 12.5, 25, and 50 ng/mL)
added to 50 μl of low leptin concentration serum (Fig 3.2). The inter- and intra-assay
coefficients of variations were <10%.
Figure 3.1: Parallelism between binding inhibition curves of serially diluted African
elephant serum pools using leptin RIA.
57
Figure 3.2. Recovery of leptin (1.56, 3.13, 6.25, 12.5, 25, and 50 ng/ml) added to serum
from an adult female African elephant.
Insulin. Serum insulin concentration was measured using a solid-phase, two-site
bovine insulin enzyme immunoassay (EIA) (10-1113-01, Mercodia Inc., Uppsala,
Sweden). Bovine insulin standards, controls, and serum samples (25 μl) were incubated
in duplicate in a 96-well anti-bovine insulin antibody-coated plate after immediate
addition of 100 μl of enzyme conjugate solution containing peroxidase-conjugated antibovine insulin antibodies. Plates were incubated for 120 min on a shaker at room
temperature. To remove unbound enzyme-labeled antibody, plates were washed six times
by hand with wash buffer. Bound conjugate was detected by reaction with 200 μl of
58
3,3′5,5′-tetramethylbenzidine (TMB) added to each well. After a 15-minute incubation at
room temperature, 50 μl 0.5 M H2SO4 was added to stop the enzymatic reaction and the
optical density was determined spectrophotometrically (450 nm filter) with a Opsys MR
Microplate Reader (Dynex Technologies, Chantilly, VA, USA). Final insulin
concentrations were calculated with an immunoassay data management program,
Revelation QuickLink (Dynex Technologies, Chantilly, VA, USA). The insulin EIA was
validated for use in African elephant serum by demonstrating: 1) parallelism between the
insulin standard curve and serial 25 μl serum samples (1:2, 1:4, and 1:8) diluted with 25
μl zero calibrator (Fig. 3.3); and 2) significant recovery (88%; y = 0.976x + 0.019) of
insulin standard (12.5 μl; 0.05, 0.15, 0.50, 1.5, and 3.0 mg/mL) added to 12.5 μl of low
insulin concentration serum (Fig. 3.4). The inter- and intra-assay coefficients of
variations were <10%.
59
Figure 3.3. Parallelism between binding inhibition curves of a serially diluted African
elephant serum pool using an anti-bovine insulin EIA.
Figure 3.4. Recovery of insulin (0.05, 0.15, 0.5, 1.5, and 3.0 mg/ml) added to a pool of
serum from adult female African elephants (n=6).
60
Glucose. Serum glucose was determined using an automated glucose analyzer
(One Touch Ultra, LifeScan, Inc., Milpitas, CA, USA) and a glucose-to-insulin ratio (G:I)
was calculated.
Determination of reproductive cyclicity status
Serum progestagens were analyzed using a solid-phase progesterone I125 RIA
(Seimens Medical Solutions Diagnostics, Los Angeles, CA) validated for elephant serum
(Brown and Lehnhardt, 1995). At least 2 years of current weekly progestagen data were
used to categorize the ovarian cycle status of each elephant (“cycling” or “non-cycling”),
with normal cycling represented by a 14- to 16-week cycle, consisting of an 8-12 week
luteal phase and a 4-6 week folicular phase (Brown et al., 2004b). Non-cycling females
exhibited consistent baseline concentrations (<0.1 ng/ml) for 2 years prior to the study.
Body condition scoring
Zoos submitted a set of three standardized photographs to score the body
condition of each elephant using a BCS index recently validated for elephants (Chapter
2). The visual scoring method consisted of a list of key body regions (ribs, backbone, and
pelvic bone) and the physical criteria used for obtaining an overall score on a 5-point
scale, with 1 being the thinnest and 5 being those with the most body fat.
61
Data analysis
Statistical analyses were performed using SAS software (SAS Institute, Cary,
NC). Power analysis was conducted (80% power, GPOWER) to determine that the study
sample size (n = 23 non-cycling and n = 23 cycling) was sufficent for 95% confidence in
the utility of the models (F-test). The Kolmogorov Smirnov test was used to test for
normality of the hormone data and the Levene Median test was used to measure for equal
variances. Data are expressed as mean and standard deviation (SD). Five serum samples
from each elephant were analyzed for glucose, insulin, and leptin levels and a mean value
calculated for each elephant. Metabolic parameters (BCS, leptin, insulin, glucose, G:I)
were compared between cycling and non-cycling elephants using an unpaired t-test. The
Cochran-Mantel-Haenszel statistic was used to test for linear trend by testing the null
hypothesis that BCS was randomly distributed between the cycling and non-cycling
groups. Relationships between BCS and hormone concentrations were analyzed using
Spearman correlation coefficients. Correlations were considered to be different from zero
at P < 0.01. To predict cycling status from the study variables, “non-cycling” was used as
the outcome variable in regression models, and BCS, leptin, insulin, and G:I were
examined as predictors.
RESULTS
The numbers of cycling and non-cycling elephants in each BCS category are
presented in Table 3.1. None of the females had a BCS=1, and relatively few had a
BCS=2. There is evidence for a strong correlation between BCS and group (cycling or
62
non-cycling) suggesting BCS of non-cycling elephants was higher than that of cycling
elephants (Table 3.1, P = 0.0228). The most frequent (mode) BCS observed in cycling
elephants was a 3 (35%), whereas the most frequent BCS observed in the non-cycling
group was a 4 (52%). The category representing the most body fat, BCS=5, was observed
in 44% of non-cycling compared to 26% of cycling elephants. The median scores for
both cycling and non-cycling elephants was a 4. Two cycling elephants had BCS=2,
whereas none of the non-cycling elephants scored in this category.
When comparing mean serum hormone concentrations including all BCS
categories, there were differences in the concentrations of serum leptin [4.03 ng/ml (SD
1.63) vs. 3.16 ng/ml (SD 0.88); P = 0.041] and insulin [0.65 mg/mL (SD 0.31) vs. 0.48
mg/ml (SD 0.18); P = 0.032] between cycling and non-cycling elephants, respectively.
Serum glucose levels did not differ between cycling and non-cycling elephants [95.0
mg/dl (SD 19.89) vs. 94.3 mg/dl (SD 15.3); P = 0.892); however, the G:I was lower in
the non-cycling group [169.82 (SD 53.21) vs. 227.18 (SD 96.23); P = 0.019].
Comparative data are shown in Appendix III. Comparisons of metabolic hormones by
cyclicity and BCS categories are shown in Table 3.2, with significant differences
observed in all hormones except glucose between cycling and non-cycling elephants with
BCS=5. Because so few elephants were represented in the BCS =2 or 3 categories (and
no elephants with BCS=1), we were unable to calculate statistical comparisons for these
categories.
There were positive correlations between BCS and mean serum leptin (r2 = 0.39, P
= 0.005) and insulin (r2 = 0.30; P = 0.06) concentrations. Glucose nor the G:I ratio were
63
correlated with BCS (r2 = 0.18, P = 0.34; r2 = -0.19, P = 0.09; respectively). Insulin
concentration also was correlated with the G:I (r2 = -0.87; P < 0.001) and glucose alone
(r2 = 0.74, P < 0.001), and glucose was related to the G:I (r2 = -0.38; P < 0.05).
Spearman correlation coefficients for comparison of BCS and the metabolic biomarkers
in cycling and non-cycling elephants are presented in Appendix III.
64
Table 3.1. Numbers of cycling and non-cycling female African elephants in each body
condition score (BCS) category.
BCS1
Cycling elephants
Non-cycling elephants
1
0
0
2
2
0
3
8
1
4
7
12
5
6
10
Total
23
23
Median (range)
4 (2-5)
4 (1-5)
Mean (SD)
3.73 (0.93)
4.39 (0.58)
Mode
3
4
1
*
Body condition score assessed on a 5-point scale
Cochran-Mantel-Haenszel test for linear trend
65
P-value
0.0228*
Table 3.2. Comparison of metabolic hormone concentrations [mean (SD)] between
cycling and non-cycling zoo African elephants by body condition score (BCS) category.
Hormone
Leptin (ng/ml)
BCS1
Cycling
Non-Cycling
P-value
2
2.90 (0.71)
NA
NA
3
2.89 (0.93)
3.03 (NA)
NA
4
3.34 (0.98)
3.57 (0.71)
0.62
5
3.58 (0.64)
4.75 (2.12)
0.05*
2
0.59 (0.42)
NA
NA
3
0.45 (0.19)
0.23 (NA)
NA
4
0.45 (0.10
0.56 (0.23)
0.19
5
0.49 (0.18)
0.78 (0.33)
0.04*
2
105 (17)
NA
NA
3
93 (18)
70 (NA)
NA
4
87 (25)
92 (15)
0.56
5
104 (17)
100 (13)
0.59
2
219 (125)
NA
NA
3
206 (65)
299 (NA)
NA
4
199 (54)
182 (42)
0.37
5
227 (64)
143 (43)
0.02*
Insulin (mg/ml)
Glucose (mg/ml)
2
G:I
NA= not applicable, sample size too small for statistical analysis/comparisons between
groups
1
Body condition score assessed on a 5-pt scale
2
Glucose-to-insulin ratio
* Significant (P < 0.05)
66
A logistic regression was performed to determine which model variables
contribute most to the occurrence of non-cycling ovarian status (Table 3.3). The odds
ratio for the BCS coefficient is 3.15 with a 95% confidence interval of [1.36, 7.26] and
predicts that with each 1 point increase in BCS, an elephant is approximately 3 times
more likely to be non-cycling.
Table 3.3. Logistic regression results of metabolic parameters predicting a non-cycling
status in female African elephants.
Predictors
SE
P
Odds Ratio (95% CI)
BCS
0.43
0.01
3.15 (1.36 - 7.26)
Leptin
0.35
0.04
2.05 (1.03 - 4.08)
Glucose
0.017
0.89
0.99 (0.97 - 1.03)
Insulin
1.45
0.04
19.50 (1.08 – 350.84)
G:I
0.006
0.04
0.99 (0.98 – 0.99)
SE, standard error of beta coefficient; P, significance level; BCS, body condition score;
G:I, glucose-to-insulin ratio
CI, confidence interval
67
DISCUSSION
Non sustainability of zoo managed African elephants is due in part to high
mortality and low reproduction, with five deaths to only three births occurring annually
over the past decade (Faust and Marti, 2011). In particular, reproduction is compromised
in zoo African elephants because of ovarian cycle problems, which appear to be on the
rise, having increased from 22% in 2002 (Brown et al., 2004a) to 46% in 2008 (Dow et
al., 2011). If these trends continue, the U.S. zoo population is predicted to become
reproductively nonviable in about 50 years (Faust and Marti, 2011). Not surprisingly,
understanding causes of morbidity, mortality and poor reproductive performance to
prevent further population declines are high priorities of the AZA/Elephant Species
Survival Plan Management Committee (Keele and Ediger, 2011). This study examined
one suspected cause of ovarian inactivity – obesity-related metabolic conditions, and is
the first to report measures of body condition in the context of reproduction and serum
values for insulin and leptin in African elephants. Similar to Freeman et al. (2009), who
found a significant relationship between high BMI and ovarian acyclicity in this species,
higher BCS scores were associated with ovarian dysfunction. Furthermore, there were
significant differences in metabolic hormone concentrations between cycling and noncycling zoo African elephants, with non-cycling females having higher concentrations of
insulin and leptin, and a lower G:I. The highest BCS also was associated with elevated
insulin and leptin concentrations compared to the other BCS categories, for both cycling
and non-cycling elephants. Finally, logistic regression analysis predicted that an elephant
is approximately 3 times more likely to be non-cycling with each 1 point increase in
68
BCS. Together, these results are highly suggestive of high body condition and metabolic
perturbations being significant risk factors for ovarian cycle problems in zoo-held
African elephants.
The most frequently observed BCS in cycling elephants was a 3, whereas in noncycling elephants it was a 4. However, there were cycling elephants in both 4 and 5 BCS
categories, so higher body condition does necessarily mean a female will be acyclic. This
is similar to humans, in which most obese women are not infertile (Grodstein et al., 1994;
Rich-Edwards et al., 1994). However, obesity can exert effects upon the hypothalamicpituitary-ovarian (HPO) axis and hence disturb menstrual cyclicity and ovulation (Hartz
et al., 1979; Lake et al., 1997) and obese women are three times more likely to suffer
from infertility than women with a normal BMI (Rich-Edwards et al., 1994). Although
obese women may have normal menstrual cycles, they often experience other
reproductive problems, such as reduced fecundity (Fedorcsak et al., 2004) and increased
miscarriage rates (Gesink Law et al., 2007) compared to non-obese women. These are
common reproductive problems observed in zoo elephants (Clubb et al., 2009; Clubb and
Mason, 2002; Mason and Veasey, 2010). Therefore, although cyclicity per se may not
always be affected because of excess weight, other reproductive functions could be
altered. Ovarian cyclicity was the only reproductive parameter analyzed in our study, so
additional studies certainly are warranted to determine how high body condition affects
other aspects of reproductive health. Furthermore, in women, the impact of excess weight
on reproductive function is influenced by age at obesity onset, time taken to become
obese, and diet (Pasquali et al., 2003), and these may well be the case for elephants.
69
In women, losing excess weight can improve fertility and lead to conception
(Bray, 1997; Clark et al., 1995, 1998; Norman et al., 2004; Pasquali et al., 2003).
However, feed restriction does not necessarily alter estrous cycle activity in other species
(horses, Vick et al., 2006; rats, Marin-Bivens and Olster, 1999). These latter studies
suggest that excess body weight may not cause reproductive dysfunction per se, but
rather may only be a contributing factor (Marin-Bevins and Olster, 1999; Wade et al,
1996; Vick et al, 2006). Therefore, interpretation of the finding of BCS associated with
ovarian acyclicity in elephants should be done with caution. Whether acyclicity in
elephants is due to direct or indirect effects of BCS, it is rationale to control the weight of
zoo elephants and determine if high BCS is a result of excessive caloric intake (Hatt and
Clauss, 2006) or metabolic conditions that impact regulation of adipose tissue, especially
given our findings of altered metabolic status in elephants with BCS=5.
Of particular interest was the finding that non-cycling elephants with a BCS=5
exhibited significant alterations in metabolic hormones compared to cycling elephants in
that category. Specifically, insulin concentrations were higher and G:I was lower in noncycling compared to cycling elephants at a BCS=5. The G:I was included to account for
the non-fasting state of our study animals, and is a common proxy for counteracting the
effects of changes in glucose and/or insulin due to feeding status (Ralston, 2002). In
humans, reproductive dysfunction due to abnormal insulin and glucose levels results from
perturbations at different levels of the gonadotropic axis, including the hypothalamus,
pituitary, and ovary (Codner and Cassorla, 2009). The ovary is a target organ for insulin,
acting via the insulin receptor and via the insulin-like growth factor 1 (IGF1 receptor),
70
and insulin receptors found in granulosa, theca, and ovarian tissue in humans (Poretsky et
la., 1999). The potential impact of disturbed insulin secretion (from low levels to
hyperinsulemia) on ovarian development and function are numerous. Because insulin is
an important regulator of the gonadotropic axis, insulin deficiency or insulin receptor
problems may lead to infertility from low gonadotropin levels due to decreased
gonadotropin-releasing hormone (GnRH) secretion, as shown in animal models with
insulin receptor depletion (Bruning et al., 2000). Under normal conditions, insulin plays a
role in ovarian function by promoting follicular development and ovarian steroidogenesis
via insulin receptors in granulosa cells (Sirotkin, 2011). In addition, insulin enhances the
recruitment and growth of pre-ovulatory follicles (Poretsky et al., 1999) and promotes
follicle maturation and ovarian growth (Hsueh et al., 1994; Portesky et al., 1999).
Obesity induces a state of hyperinsulinemia and insulin resistance in humans (Matthaei et
al., 2000), where there is an increase in insulin secretion. The altered insulin metabolism
leads to hyperandrogenemia and perturbations in the IGF system, thereby increasing the
likelihood of menstrual and ovulatory disturbance in obese women (Butzow et al. 2000).
Excess insulin may also lead to overstimulation of the insulin receptors in the ovary,
increasing androgen secretion causing premature follicle atresia and anovulation, and
fostering the development polycystic ovarian syndrome (PCOS) (Poretsky et al., 1999;
Willis et al., 1996). This condition is not believed to be a significant problem in captive
African elephants (only ~15% of females); however, there is one case study of ovarian
cysts being associated with acyclicity (Brown et al., 1999), so more work could be done
in this area. In horses, often used as a model for elephants, insulin resistance (i.e.,
71
elevated insulin) and obesity compromises reproduction through a direct action on the
ovaries, rather than at the hypothalamic-pituitary axis ( Johnson et al., 2012; Vick et al.,
2007). This assumption is based on the finding that no differences in secretion of
luteinizing hormone (LH) or follicle-stimulating hormone (FSH) were observed in mares
with insulin resistance (Sessions et al., 2004; Vick et al, 2006). Acyclic African elephants
have been shown to have normal, baseline concentrations of LH and FSH, which respond
normally to gonadotropin-releasing hormone challenge (Brown, 2000; Brown et al.,
2004a), suggesting pituitary function is not compromised in elephants either.
Nevertheless, our findings of normal glucose but elevated serum insulin levels provide
evidence of disrupted insulin regulation in non-cycling African elephant females with a
high BCS, which could indicate that these females may either have or be at risk for
developing insulin resistance.
Leptin concentrations also were higher in non-cycling females. In women, this
hormone plays a major role in energy balance and reproduction, with excess levels being
associated with ovarian cycle problems and amenorrhea (Brewer and Balen, 2010; Hill et
al., 2008; Mácajová et al., 2004; Tena-Sempere, 2007). Leptin has a regulatory role in
reproductive function, it has stimulatory effects on the HPO axis at normal serum
concentrations but can have inhibitory effects upon folliculogeneiss and its control when
levels are elevated, such as seen during obesity (Tamer Erel and Senturk, 2009). In rats,
leptin at high concentrations interferes with normal follicular development, the
development of a dominant follicle, oocyte maturation, and hence ovulation (Duggal et
al., 2000). A 10% reduction in body fat was found to lead to 53% reduction in serum
72
leptin concentration (Considine et al., 1996). Perhaps weight loss in elephants with high
BCS and elevated leptin could improve reproductive potential in part by reducing
circulating leptin concentration and the detrimental effect upon the HPO axis it exerts at
higher concentrations. Because leptin is a hormone produced by adipose cells,
concentrations vary directly with the percentage of body fat in several species, including
humans (Banks et al., 2001; Fors et al., 1999; Garnsworthy et al., 2008; Pasquali and
Casimirri, 1993). We found a similar correlation with leptin and BCS in female African
elephants. Hence, assessing serum leptin levels and/or BCS could serve as useful
screening tools to identify overweight elephants in zoos. Although mechanisms of
metabolic hormone perturbations in elephants are unknown, based on insulin and leptin
data, we theorize that derangements in metabolic activity, rather than the amount of body
fat alone, may be altering ovarian activity in the fattest elephants.
In conclusion, the potential negative effects of poor metabolic status and excess
body weight on reproductive and health function in zoo elephants have been largely
overlooked, but could be significant. This is the first study to assess body condition and
metabolic hormones as they related to reproductive function in African elephants. We
found that BCS, leptin and insulin concentrations, and G:I all were significantly altered in
non-cycling compared to normal cycling females, suggesting that factors related to
obesity may be compromising reproductive function. Studies in humans and horses
indicate that similar changes in body weight and metabolic markers can severely
compromise fertility, and that even small reductions in body fat and weight loss may
effectively restore ovarian activity (Clark et al., 19985, 1998; Hill et al., 2008; Miller et
73
al., 2008). In fact, overweight and obese women are advised to participate in lifestyle
intervention programs to lose weight before entering fertility programs (National Institute
for Clinical Excellence, 2004). Given that the zoo African elephant population is not selfsustaining, a logical next step would be to promote weight loss in non-cycling females by
changes in diet and/or exercise with the goal of improving metabolic health and
resumption of ovarian activity. We also need to assess body condition, insulin, glucose,
and G:I in the context of other causes of morbidity and mortality in zoo elephants (e.g.,
kidney disease, foot and joint problems) (Clubb and Mason, 2002; Clubb et al., 2009;
Lewis et al., 2010; Mason and Veasey, 2010), which commonly are associated with
insulin-glucose dysfunction in other species (Baily et al., 2008; Dimitrolous et al., 2013;
Geor and Frank, 2009; Roberts et al., 2013). Because of potential health effects, facilities
housing elephants are encouraged to monitor body condition and metabolic status
regularly and, if necessary, to implement mitigating management strategies for at risk
individuals.
74
CHAPTER 4
CONCLUSIONS
The results of this study provide the first evidence associating obesity and
metabolic hormones with reproductive acyclicity in the elephant. Methodologies for
obesity and metabolic hormone assessments were developed and validated in this
dissertation for use in elephants, including a body condition scoring (BCS) index, and
leptin and insulin immunoassays. Our new elephant BCS Index is a quick and reliable
tool that has been validated to assess body condition in female African elephants for
routine health monitoring. In addition, the BCS method includes example photographs of
elephants representing each BCS category and a flow chart to systematically guide the
assessor through the body condition scoring process. The relationships between BCS and
ultrasound measurements determined that BCS adequately reflects the amount of
subcutaneous fat of elephants, and that most zoo elephants are in the higher scoring
categories. We propose that this BCS index could be a valuable tool for examining the
relationship between body condition and various factors affecting elephant health and
reproduction in zoo and free-ranging elephant populations.
This study can form the basis of further investigations to determine pathogenic
mechanisms and other health significances for elephants, in terms of disease associations
75
contributing to high morbidity and mortality in zoo elephants. As in commonly reported
in human medical literature, the incidence of metabolic alterations in elephants might be
linked to management practices that included prolonged periods of physical inactivity
and the provision of rations that are nutritionally excessive with regard to the animal‟s
energy requirements. Curiously, in many cases, provision of a healthier ration for
elephants would represent a considerable cost saving for zoos who often expend
considerable financial resources on rations that, in terms of protein and energy, may
exceed the animals‟ nutritional requirements. By instituting an obesity monitoring system
such as this, facilities can monitor and identify at-risk elephants in time for mitigating
actions to be undertaken. Finally, studies are highly warranted to determine if weight loss
and/or altered diets improve metabolic conditions in elephants with the ultimate goal of
reinitiating ovarian activity. Our ultimate goal is to provide practical, cost-effective, and
scientifically validated tools and recommendations for those caring for elephants to
achieve optimal health and welfare to promote sustainable elephant populations.
76
APPENDIX I
PHOTOGRAPHING GUIDE FOR ELEPHANT BODY CONDITION SCORING
Kari A. Morfeld and Janine L. Brown, Smithsonian Conservation Biology Institute
Introduction: The purpose of this guide is to obtain standardized photos for body
condition scoring of elephants.
Equipment: Any digital camera is suitable, and natural lighting is recommended (a flash
may reflect poorly). A step stool or ladder is needed to obtain photos from a higher
vantage point. A color printer is recommended for printing the guide.
Location: Subjects can be photographed either indoors or outdoors (outdoors
recommended), but ensure there is sufficient natural light that does not require a flash and
that does not cast shadows across the elephant body. Eliminate as much background
clutter (bars, other elephants, people, etc.) as possible. The floor should be a flat surface.
Elephant position: The elephant should be standing in a relaxed position as straight
forward as possible with equal weight distribution on all four feet, eliminating excessive
bend in any knee. The trunk should be down and directed forward, and the head directed
forward. The elephant should be clean and dry, minimizing dirt and water.
Photographs: A series of 3 digital photographs are required for each elephant. Ensure
the elephant encompasses ~ ¾ of the entire photograph area. This can be done while
photographing, or using appropriate software to edit the photograph. Include the entire
elephant in each photo. There should be spacing between the elephant and the edge of the
photograph. Descriptions of photograph criteria are below, including position of
photographer and targeted body regions to include in each photo. The area of interest to
include in each photo is shown as a boxed region.
Photograph submissions: An email will be sent from Kari Morfeld (Smithsonian) to the
institution contact person with an invitation to upload the photos via a secure website
using “Accellion Secure File Transfer”. The email will include simple instructions on
uploading the photos using this system.
77
Instructions for Taking Body Condition Photos
Photo 1: SIDE
Position: Stand at eye-level from either side
of the elephant.
Targeted Regions: Entire length of elephant,
backbone ridge.
Photo 2: REAR
Position: Stand at eye-level directly behind
the elephant
Targeted Regions: Both hips, top ridge of
backbone, tailhead.
Photo 3: REAR-ANGLE
Position: Stand eye-level at a 45° angle from
rear (either side), while the elephant remains
facing forward (the head should not be turned
toward you, but should be directed forward).
Targeted Regions: Entire backbone, entire
rear, hip.
78
APPENDIX II
Comparison of body condition and metabolic hormones [mean (SD)] between cycling
and non-cycling African elephants.
Cycling
n = 23
Non-cycling
n = 23
P-value
BCS1
3.73 (0.58)
4.39 (0.93)
0.009*
Insulin (ng/ml)
0.48 (0.18)
0.65 (0.31)
0.032*
Glucose (mg/dl)
95.0 (19.89)
94.3 (15.3)
0.892
G:I2
211.38 (62.91)
169.82 (53.21)
0.024*
Leptin (ng/ml)
3.16 ± 0.88
4.03 ± 1.63
0.041*
Parameter
1
Body condition score assessed on a 5-pt scale
Glucose-to-insulin ratio
*P < 0.05
2
79
APPENDIX III
Spearman correlations among body condition score (BCS) and metabolic hormones in
female African elephants (n=46).
BCS
BCS
Leptin
Insulin
Glucose
G:I
1.00
0.386a
0.294b
0.183
-0.188
1.00
-0.021
-0.065
-0.083
1.00
0.738a
-0.870a
1.00
-0.376a
Leptin
Insulin
Glucose
G:I
a
b
1.00
Significant (P < 0.05)
Significant (P < 0.10)
80
REFERENCES
Althabe, F. 2012. Modified risk factors for stillbirth in high-income countries include
overweight and obesity, maternal age and smoking. Evidence Based Medicine,
April 17(2): 65-6.
Ange, K., S.D. Crissey, C. Doyle, K. Lance, and H. Hintz. 2001.A survey of African
(Loxodonta afriana) and Asian (Elephas maximus) elephant diets and measured
body dimensions compared to their estimated nutrient requirements. Proceedings
of the Nutrition Advisory Group Conference, 4: 5-14.
AZA Standards for elephant management and care: Association of Zoos and Aquariums,
2011: 16.
Bailey S.R., J.L. Habershon-Butcher, K.J. Ranson, J. Elliott, and N.J. Menzies-Gow.
2008. Hypertension and insulin resistance in a mixed-breed population of ponies
predisposed to laminitis. American Journal of Veterinary Research, 69(1): 122–
129
Banks, W. A., J.E. Phillips-Conroy, C.J. Jolly, and J.E. Morley. 2001. Serum leptin levels
in wild and captive populations of baboons (papio): implications for the ancestral
role of leptin. The Journal of Clinical Endocrinology and Metabolism, 86(9):
4315–4320.
Barb, C.R, G.J. Hausman, and K. Czaja.2005. Leptin: a metabolic signal affecting central
regulation of reproduction in the pig. Domestic Animal Endocrinology, Jul 29(1):
186-92.
Bishop, C. J., B.E. Watkins, L.L. Wolfe, D.J. Freddy, and C.C. White. 2009. Evaluating
mule deer body condition using serum thyroid hormone concentrations. Journal of
Wildlife Management, 73(3): 462–467.
Blanc, J. 2008. Loxodonta africana. In: IUCN 2013. IUCN Red List of Threatened
Species. Version 2013.1.
Boden G. 2011. Obesity, insulin resistance and free fatty acids. Current Opinion in
Endocrinology, Diabetes, and Obesity, 18:139–143.
Brandt, M. M., D. H. Keisler, D.L. Meyer, T.B. Schmidt, T. B., and E.P. Berg. 2007.
81
Serum hormone concentrations relative to carcass composition of a random
allotment of commercial-fed beef cattle. Journal of Animal Science, 85(1): 267275.
Bray, G. A. 1997. Obesity and reproduction. Human Reproduction (Oxford, England), 12
Suppl 1: 26–32.
Brown J.L., D.J. Schmitt, A. Bellem, L.H. Graham, and J. Lehnhardt. 1999. Hormone
secretion in the Asian elephant (Elephas maximus): characterization of ovulatory
and anovulatory luteinizing hormone surges. Biology of Reproduction, 61: 1294–
1299.
Brewer, C. J., and A. H. Balen. 2010. The adverse effects of obesity on conception and
implantation. Reproduction (Cambridge, England), 140(3): 347–364.
Brown, J. L. 2000. Reproductive endocrine monitoring of elephants: an essential tool for
assisting captive management. Zoo Biology, 19(5): 347–367.
Brown, J. L., D. Olson, M. Keele, M., and E.W. Freeman. 2004a. Survey of the
reproductive cyclicity status of Asian and African elephants in North America.
Zoo Biology, 23(4): 309–321.
Brown, J. L., S. Walker, and T. Moeller. 2004b. Comparative endocrinology of cycling
and noncycling Asian (Elephas maximus) and African (Loxodonta africana)
elephants. General and Comparative Endocrinology, 136: 360–370.
Bruning, J.C., D. Gautam, D.J. Burks, J. Gillette, M. Schubert, P.C. Orban, R. Klein, W.
Krone, D. Muller-Wieland, and C.R. Kahn. 2000. Role of brain insulin receptor in
control of body weight and reproduction. Science, 289: 2122–2125.
Burkholder, W. J. 2000. Use of body condition scores in clinical assessment of the
provision of optimal nutrition. Journal of the American Veterinary Medical
Association, 217(5): 650–654.
Butzow, T.L., M. Lehtovirta, R. Siegberg, O. Hovatta, R. Koistinen, M. seppalo, and D.
Apter. 2000. The decrease in luteinizing hormone secretion in response to
weight reduction is inversely related to the severity of insulin resistance in
overweight women. Journal of Clinical Endocrinololgy and Metabolism.,
Sep;85(9): 3271-5.
82
Carter, R. A., R.J. Geor, W. Burton Staniar, T.A. Cubitt, and P.A. Harris. 2009. Apparent
adiposity assessed by standardised scoring systems and morphometric
measurements in horses and ponies. Veterinary Journal, 179(2): 204–210.
Chang, R. J. 2007. The reproductive phenotype in polycystic ovary syndrome. Nature
Clinical Practice Endocrinology & Metabolism, 3(10): 688–695.
Chen, Y., F. Ono, T. Yoshida, and Y. Yoshikawa. 2002. Relationship between body
weight and hematological and serum biological parameters in female cynomolgus
monkeys (Macca fascicularis). Journal of Experimental Animal Science, 51:
125-31.
Chilliard, Y., M. Bonnet, C. Delavaud, Y. Faulconnier, C. Leroux, J. Djiane, and F.
Bocquier. 2001. Leptin in ruminants. Gene expression in adipose tissue and
mammary gland, and regulation of plasma concentration. Domestic Animal
Endocrinology, 21(4): 271-295.
Clark, A.M., W. Ledger, C. Galletly, L. Tomlinson, F. Blaney, X. Wang, and R. J.
Norman. 1995. Weight loss results in significant improvement in pregnancy and
ovulation rates in anovulatory obese women. Human Reproduction, 10: 2705–
2712.
Clark A.M., B. Thornley, L. Tomlinson, C. Galletley, and R. J. Norman.1998.Weight
loss in obese infertile women results improvement in reproductive outcome for all
forms of fertility treatment. Human Reproduction, Jun13(6): 1502-5.
Clingerman, K., and L. Summers. 2012. Validation of a body condition scoring system in
Rhesus Macaques (Macaca mulatta): Inter- and interater variability. Journal of
American Association of Laboratory Animal Science, 51(1): 31-36.
Clubb, R., and G. Mason. 2002. A review of the welfare of zoo elephants in Europe.
Report commissioned by the RSPCA. Oxford: University of Oxford, Animal
Behaviour Research Group.
Clubb, R., M. Rowcliffe, P. Lee, K. U. Mar, C. Moss, and G.J. Mason. 2009. Fecundity
and population viability in female zoo elephants: problems and possible solutions.
Animal Welfare, 18(3): 237–247.
Clubb, R., M. Rowcliffe, P. Lee, K.U. Mar, C. Moss, and G.J. Mason. 2008.
Compromised survivorship in zoo elephants. Science, 322(5908): 1649.
83
Codner, E., and F. Cassorla. 2009. Puberty and ovarian function in girls with type 1
diabetes mellitus. Hormone Research, 71(1):12-21.
Cohen, J. 1968. Weighted kappa: a nominal scale agreement with provision for scaled
disagreement or partial credit. Psychological Bulletin, 70(4): 213-20.
Considine, R. V., M.K. Sinha, M.L. Heiman, A. Kriauciunas, T.W. Stephens, M.R. Nyce,
and J.P. Ohannesian. 1996. Serum immunoreactive-leptin concentrations in
normal-weight and obese humans. The New England Journal of Medicine, 334(5):
292–295.
Cook, R. C., J.G. Cook, D.L. Murray, P. Zager, B.K. Johnson, and M.W. Gratson, 2001.
Development of Predictive Models of Nutritional Condition for Rocky Mountain
Elk. The Journal of Wildlife Management, 65(4): 973–987.
Croke, 1997. The Modern Ark: The story of zoos, past, present, and future. Scribner,
New York. p 21-26.
Czekala N., E. MacDonald, K. Steinman, S. Walker, N. W. Garrigues II, D. Olson, and J.
L. Brown. 2003. Estrogen and LH dynamics during the follicular phase of the
estrous cycle in the Asian elephant. Zoo Biology, 22: 443–454
Dale, R. 2009. Birth statistics for African (Loxodonta africana) and Asian (Elephas
maximus) elephants in human care: history and implications for elephant welfare.
Zoo Biology, 29(2): 87-103
de Knegt, H.J, van Langevelde, F., Skidmore, A.K., Delsink, A. Slotow, R., Henley, S.,
Bucini, G. de Boer, W.F., Coughenour, M.B., Grant, C.C., Ignas, M.A., nig, H.,
Henley, M., Knox, N.M., Kohl, E.M., Mwakiwa, E., Page, B.R., Peels, M.,
Pretorius, Y., van Wieren, S.E., and H.T. Prins. 2011. The spatial scaling of
habitat selection by African elephants. Journal of Animal Ecology, 80: 270–281.
Deurgenber, P. and M. Yap. 1999. The assessment of obesity: methods for measuring
body fat and global prevalence of obesity. Clinical Endocrinology Metabolism,
13: 1-11.
Diamanti-Kandarkis, E., and A. Dunaif. 2012. Insulin resistance and the polycystic ovary
syndrome revisited: an update on mechanisms and implications. Endocrine
Review, 33(6): 981-1030.
Dimitroulas, T., K.M. Douglas, V.F. Panoulas, T. Tomas, J.P. Smith, G.J. Treharne, P.
Nightingale, J. Hodson, and G.D. Kitas. 2013. Derangement of hemostasis in
84
rheumatoid arthritis: association with demographic, inflammatory and metabolic
factors. Clinical Rheumatology, 15May2013.
Domecq, J. J., A. L. Skidmore, J. W. Lloyd and J.B. Kaneene. 1995. Validation of Body
Condition Scores with Ultrasound Measurements of Subcutaneous Fat of Dairy
Cows. Journal of Dairy Science, 78(10): 2308–2313.
Donaldson, M. T., D. McFarlane, A.J.R. Jorgensen, and J. Beech. 2004. Correlation
between plasma alpha-melanocyte-stimulating hormone concentration and body
mass index in healthy horses. American Journal of Veterinary Research, 65(11):
1469–1473.
Dow, T.L., I. Holaskova, and J.L. Brown. 2011. Results of the third reproductive assessment
survey of North American Asian (Elephas maximus) and African (Loxodonta
africana) female elephants. Zoo Biology, 30: 699-711.
Dow, T.L., and J.L. Brown. 2012. Evidence that hyperprolactinaemia is associated with
ovarian acyclicity in female zoo African elephants. Reproduction, Fertility, and
Development, 24(8): 1019-1027.
Duggal P.S, K. H. Van Der Hoek, C.R. Milner, N.K. Ryan, D.T. Armstrong, and D.A.
Magoffin. 2000. The in vivo and in vitro effects of exogenous leptin on ovulation
in the rat. Endocrinology, 141(6):1971-6.
Edmonds, J.W., J.K. Prassain, D. Dorand, Y.Yang, H.D. Hoang, J. Vibbert, H.M.
Kubagawa, and M.A. Mille. 2010. Insulin/FOXO signaling regulates ovarian
prostaglandins critical for reproduction. Developing Cell, 19(6): 858-71
Edmonson, A.J., I.J. Lean, L.D. Weaver, T. Farver, and G. Webster. 1989. A body condition
scoring chart for holstein dairy cows. Journal of Dairy Science, 72(1): 68-78.
Ellis, K.J., A. Scheimann, and P.D.K. Lee. 2000. Body composition measurements: clinical
and technical considerations. Endocrinology, 10: 57S-62S.
Ely, J.J., T. Zavaskis, and M.L. Lammey. 2013. Hypertension increases with aging and
obesity in chimpanzees (Pan troglodytes). Zoo Biology, 32(1): 79-87.
Fanelli, M., and R. Kuczmarski.1984. Ultrasound as an approach to assessing body
composition. The American Journal of Clinical Nutrition, 39(5): 703 –709.
Faust, L., and K. Marti. 2011. Technical report on ZooRisk modeling of the North
American African elephant SSP population. Lincoln Park Zoo, Chicago.
85
Fedorcsak, P., P.O. Dale, R. Storeng, G. Ertzeik, S. Bjercke, N. Oldereid, A.K. Omland,
T. Abyhold, and T. Tanbo. 2004. Impact of overweight and underweight on
assisted reproduction treatment. Human Reproduction, 19: 2523–2528.
Fernandez-Fernandez, R., M. Tena-Sempere, V.M. Navarro, M.L. Barreiro, J.M.
Castellano, E. Aguilar, and L. Pinilla. 2005. Effects of ghrelin upon gonadotropinreleasing hormone and gonadotropin secretion in adult female rats: in vivo and in
vitro studies. Neuroendocrinology, 82(5-6): 245-55.
Fleiss JL. 1981. Statistical Methods for Rates and Proportions. 2nd ed. New
York, NY: John Wiley & Sons Inc.
Forcada, F., and J.A. Abecia. 2006. The effect of nutrition on the seasonality of
reproduction in ewes. Reproduction, Nutrition, and Development, 46(4): 355-65.
Fors, H., H. Matsuoka, I. Bosaeus, S. Rosberg, K.A. Wikland,and R. Bjarnason. 1999.
Serum leptin levels correlate with growth hormone secretion and body fat in
children. The Journal of Clinical Endocrinology and Metabolism, 84(10): 3586–
3590.
Frank, N., S. B. Elliott, L.E. Brandt, and D.H. Keisler. 2006. Physical characteristics,
blood hormone concentrations, and plasma lipid concentrations in obese horses
with insulin resistance. Journal of the American Veterinary Medical Association,
228: 1383–1390.
Frank, N. R.J. Geor, and A.J. Bailey. 2010. Equine Metabolic Syndrome. Journal of
Veterinary Internal Medicine, 24: 467-475.
Franks, S. 1989. Polycystic ovary syndrome: a changing perspective. Clinical
Endocrinology (Oxf), 31(1): 87-120.
Freeman, E. W., G. Guagnano, D. Olson, M. Keele, and J.L. Brown. 2009. Social factors
influence ovarian acyclicity in captive African elephants (Loxodonta africana).
Zoo Biology, 28(1): 1–15.
Freeman, E. W., I. Whyte, and J.L. Brown. 2009. Reproductive evaluation of elephants
culled in Kruger National Park, South Africa between 1975 and 1995. African
Journal of Ecology, 47(2): 192–201.
Frisch R.E., and R. Revelle. 1970. Height and weight at menarche and a hypothesis of
critical body weights and adolescent events. Science, 169(3943): 397–399.
86
Gales, N., and H. Burton. 1987. Ultrasonic Measurement of Blubber Thickness of the
Southern Elephant Seal, Mirounga-Leonina (Linn). Australian Journal of
Zoology, 35(3): 207–217.
Garcia, C., M.A. Huffman, K. Shimizu, and J.R. Speakman. 2011. Energetic
consequences of seasonal breeding in female Japanese macaques (Macaca
fuscata). American Journal of Physical Anthropology, 146(2): 161-70.
Garnsworthy, P. C., A. Lock, G.E. Mann, K.D. Sinclair, and R. Webb. 2008. Nutrition,
metabolism, and fertility in dairy cows: 2. Dietary fatty acids and ovarian
function. Journal of Dairy Science, 91(10): 3824–3833.
Gebin, L., P. Lee, N. Mori, I. Yamamoto, K. Kawasumi, H. Tanabe, and T. Arai. 2012.
Supplementing five-point body condition score with body fat percentage increases
the sensitivity for assessing overweight status of small to medium sized dogs.
American Veterinary Medicine Research, 3: 71-78.
Gesink Law, D.C., R.F. Maclehose, and M.P. Longnecker. 2007. Obesity and time to
pregnancy. Human Reproduction, 22: 414-420.
Gentry L.R., D.L. Thompson Jr, G.T. Gentry Jr, K.A. Davis, R.A. Godke, and J.A.
Cartmill. 2002. The relationship between body condition, leptin, and reproductive
and hormonal characteristics of mares during the seasonal anovulatory period.
Journal of Animal Science, 80(10): 2695-703.
Geor, R., and N. Frank. 2009. Metabolic syndrome–From human organ disease to
laminar failure in equids. Veterinary Immunology and Immunopathology, 129(34): 151–154.
German, A.J., S.L. Holden, and G.L. Moxham. 2006. A simple, reliable tool for owners
to assess the body condition of their cat or dog. Journal of Nutrition, 136: 2031S2033S.
Grodstein, F., M.B. Goldman, and D.W. Cramer. 1994. Body mass index and ovulatory
infertility. Epidemiology, 5(2): 247-50.
Gustine, D. D., Parker, K. L., and D.C. Heard. 2010. Using ultrasound measurements of
rump fat to assess nutritional condition of woodland caribou in northern British
Columbia, Canada. Rangifer, 27(4): 249.
Hall-Martin, A. J. 1987. Role of musth in the reproductive strategy of the African
elephant (Loxodonta africana). South African Journal of Science, 83: 616-620.
87
Hamilton, W. D. 1964. The genetic evolution of social behavior. I. Journal of Theoretical
Biology, 7: 1-16.
Hall, A. J., and B.J. McConnell. 2007. Measuring changes in juvenile gray seal body
composition. Marine Mammal Science, 23(3): 650–665.
Harris, M., C. Sherwin, and S. Harris. 2008. The welfare, housing and husbandry of
elephants in UK zoos: final report. University of Bristol, DEFRA Science and
Research Project WC05007, Bristol, UK.
Hartz, A. J., P.N. Barboriak, A. Wong, K.P. Katayama, and A. A. Rimm. 1979. The
association of obesity with infertility and related menstural abnormalities in
women. International Journal of Obesity, 3(1): 57–73.
Hatt, J.M. and M. Clauss. 2006. Feeding Asian and African elephants (Elephas maximus
and Loxodonta africana) in captivity. International Zoology Yearbook 40: 88-95.
Henneke, D. R., G.D. Potter, J.L. Kreider, and B. F. Yeates. 1983. Relationship between
condition score, physical measurements and body fat percentage in mares. Equine
Veterinary Journal, 15(4): 371–372.
Hermes R., D. Olson D, F. Goeritz, J.L. Brown, D.L. Schmitt, D. Hagan, J.S. Peterson,
G. Fritsch, and T.B. Hildebrandt. 2000. Ultrasonography of the sexual cycle in
female African elephants (Loxodonta africana). Zoo Biology, 19:v369–382.
Hermes, R., T.B. Hildebrandt, and F. Goritz. 2004. Reproductive problems directly
attributable to long-term captivity-asymmetric reproductive aging. Animal
Reproduction Science, 83: 49-69.
Hildebrandt, T.B., and F. Goritz. 1995. Ultrasonography as a tool in propagation of zoo
animals and wildlife. Proceedings of the Annual Meeting of American
Association of Zoo Veterinarians, 356-364.
Hildebrandt, T. B., F. Goritz, N. Pratt, D.L. Schmitt, S. Quandt, J. Raath, and R.R.
Hofmann. 1998. Reproductive assessment by ultrasonography in elephants
(Loxodonta africana and Elephas maximus). Journal of Zoo and Wildlife
Medicine, 29: 114-128.
Hildebrandt, T. B., F. Goritz, F., N. Pratt, J.L. Brown, R.J. Montali, D.L.Schmitt,
G. Fritsch, and R. Hermes. 2000. Ultrasonography of the urogenital tract in
elephants (Loxodonta africana and Elephas maximus): An important tool for
assessing female reproductive function. Zoo Biology, 19: 321-332.
88
Hildebrandt, T. B. 2006. Reproductive and diagnostic ultrasonography. In Biology,
medicine, and surgery of elephants (ed. M. Fowler & S. Mikota), pp. 357–376, 1st
edn. Ames, IA: Blackwell Publishing.
Hill, J.W., J.K. Elmquist, and C.F. Elias. 2008. Hypothalamic pathways linking energy
balance and reproduction. American Journal of Physiology, Endocrinology, and
Metabolism, 294(5):E827-32.
Hodges, 1998. Endocrinology of the ovarian cycle and pregnancy in the Asian (Elephas
maximus) and African (Loxodonta africana) elephant. Animal Reproductive
Science, 53: 3-18.
Hsueh, A.J., H. Billig, and A. Tsafriri. 1994. Ovarian follicle atresia: a hormonally
controlled apoptotic process. Endocrine Reviews, 15(6):707-24.
Irvine, L. M., and R. W. Shaw. 2003. The effects of patient obesity in gynaecological
practice. Current Obstetrics and Gynaecology, 13(3): 179–184.
Johnson, P. J. 2002. The equine metabolic syndrome peripheral Cushing‟s syndrome. The
Veterinary Clinics of North America. Equine Practice, 18(2): 271–293.
Johnson, P.J., C.E. Wiedmeyer, A. LaCarrubba, V.K. Ganjam, and N.T. Messer. 2012.
Diabetes, insulin resistance, and metabolic syndrome in horses. Joural of
Diabetes, Science, and Technology, May 1;6(3): 534-40.
Kearns, C. F., K.H. McKeever, H. John-Alder, T. Abe, and W.F. Brechue. 2002.
Relationship between body composition, blood volume and maximal oxygen
uptake. Equine Veterinary Journal, Supplement, (34): 485–490.
Keele, M., and N.D. Ediger. 2011. AZA elephant master plan. AZA publication number
1999 10285.zooe. Portland, OR: Oregon Zoo.
Kennedy, G.C., J. Mitra, and G.C. Kennedy. 1963. Body weight and food intake as
initiating factors for puberty in the rat. Journal of Physiology, 166:408–418.
Kirschner, M.A., E. Samojlik, M. Drejka, E. Szmal, G. Schneider, and N. Ertel. 1990.
Androgen-estrogen metabolism in women with upper body versus lower body
obesity. Journal of Clinical Endocrinology and Metabolism, 70(2):473-9.
Klop, B., J.W. F. Elte, and M.C. Cabezas. 2013. Dyslipidemia in obesity: mechanisms
and potential targets. Nutrients, 5(4):1218-1240.
Knight, P.G., and C. Glister. 2001. Potential local regulatory functions of inhibins,
activins and follistatin in the ovary. Reproduction, 121:503–512
89
Kumar, S., and G. Kaur. 2013. Intermittent fasting dietary restriction regimen negatively
influences reproduction in young rats: a study of hypothalamo-hypophysialgonadal axis. PLoS One., 8(1):e52416.
Kapustin, N., J.K. Crister, D. Olson, and P.V. Malven. 1996. Nonluteal estrouc cycles of
3-week duration are initiated by anovulatory luteinizing hormone peaks in African
elephants. Biology of Reproduction, 55:1147-1154.
Lake, J.K., C. Power, and T.J. Cole. 1997. Women's reproductive health: the role of
body mass index in early and adult life. International Journal of Obesity and
Related Metabolic Disorders, 21: 432–438.
Laflamme D. 1997. Development and validation of a body condition score system for
dogs. Canine Practice, 22:10–5.
Landis, J.R. and G.G. Koch. 1977. The measurement of observer agreement for
categorical data. Biometrics, 33(1): 159-74.
Lassen, J., M. Hansen, M.K. Sorensen, G.P. Aamand, L.G. Christensen and P. Madsen.
2003. Genetic analysis of body condition score in first-parity Danish Holsetein
cows. Journal of Dairy Science, (86): 4123–4128.
Laws, R. M., I.S.C. Parker, and R.C. B. Johnston. 1970. Elephants and habitats in North
Bunyoro, Uganda. East African Wildlife Journal, 8: 163-180.
Laws, R. M. 1970a. Biology of African elephants. Science Progress, 58: 251-262.
Laws, R. M. 1970b. Elephants as agents of habitat and landscape change in East Africa.
Oikos, 21: 1-15.
León, H. V., J. Hernández-Cerón, D.H. Keislert, and C.G. Gutierrez. 2004. Plasma
concentrations of leptin, insulin-like growth factor-I, and insulin in relation to
changes in body condition score in heifers. Journal of Animal Science, 82(2):
445-451.
Lievense, A.M., S.M.A. Bierma-Zeinstra, A.P. Verhagen, M.E. van Baar, J.A.N. Verhaar,
and B.W. Koes. 2002. Influence of obesity on the development of osteoarthritis of
the hip: a systemic review. Rheumatology, 41: 1155-62.
Lewis, K. D., Shepherdson, D. J., Owens, T. M., and M. Keele. 2010. A survey of
elephant husbandry and foot health in North American zoos. Zoo Biology, 29(2):
221–236.
90
Leyva F., I.F. Godsland, and M. Ghatei. 1998. Hyperleptinemia as a component of a
metabolic syndrome of cardiovascular risk. Arteriosclerosis Thrombosis and
Vascular Biology,18: 928-933.
Lueders, I., Taya, K., Watanabe, G., Yamamaoto, Y., Tamamoto, T., Kaewmanee, S.,
Niemuller, C., Gray, C., Jurgen Streich, W., and T. Hildebrandt. 2011. Role of
the double luteinizing hormone peak, luteinizing follicles, and the secretion of
inhibin for dominant follicle selection in Asian elephants (Elephas maximus).
Biology of Reproduction, 85 (4): 714-720.
Lueders I, C. Niemuller, C. Gray, P. Rich, and T.B. Hildebrandt TB. 2010. Luteogenesis
during the estrous cycle in Asian elephants (Elephas maximus). Reproduction,
140:777–786.
Lund, E.M., P.J. Armstrong, and C.A. Kirk. 2005. Prevalence and risk factors for obesity
in adult cats form private veterinary practices. International Journal of Applied
Research in Veterinary Medicine, 3; 88-96.
Lund, E.M., P.J. Armstrong, C.A Kirk, and J.S. Klausner. 2006. Prevalence and risk
factors for obesity in adult dogs from private veterinary practices. International
Journal of Applied Research in Veterinary Medicine, 4: 177-186.
Meyer, J., Ball, R., and Brown, J. L. 2004. Why are so many elephants in captivity not
cycling? Evidence of an association between ovarian inactivity and
hyperprolactinaemia. Journal of Elephant Managers Association, 15: 9–15.
Plotka, E.D., U.S. Seal, F.R. Zarembka, L.G. Simmons, A. Teare, L.G. Phillips, K.C.
Hinshaw, and D. G. Wood, 1998. Ovarian function in the elephant: Luteinizing
hormone and progesterone cycles in African and Asian elephants. Biology of
Reproduction, 28:309-314.
Mácajová, M., D. Lamosová, and M. Zeman. 2004. Role of leptin in farm animals: a
review. Journal of Veterinary Medicine. A, Physiology, Pathology, Clinical
Medicine, 51(4): 157–166.
Marin-Bevins, C.L., and D.H Olster. 1999. Food restriction neither improves nor
exacerbates reproduction in obese female Zucker rats. Physiology Behavior,
66:893–897.
Matthaei, S., M. Stumvoll, M. Kellerer, and H.U. Haring. 2000. Pathophysiology and
pharmacological treatment of insulin resistance. Endocrine Review, 21(6):585618.
91
Martins, M.C., L. Lima Falerio, and A. Fonseca. 2012. Relationship between leptin and
body mass and metabolic syndrome in an adult population. Revista Portuguesa de
Cardiologia, 31(11): 711-9.
Martos-Moreno, G.A., J.A. Chowen, and J. Argente. 2010. Metabolic signals in human
puberty: effects of over and undernutrition. Molecular and Cellular
Endocrinology, Aug 5; 324(1-2): 70-81.
Mason, G. J., and J.S. Veasey. 2010. What do population-level welfare indices suggest
about the well-being of zoo elephants? Zoo Biology, 29: 256–273.
Mellish, J. A., Tuomi, P. A., and M. Horning. 2004. Assessment of ultrasound imaging as
a noninvasive measure of blubber thickness in pinnipeds. Journal of Zoo and
Wildlife Medicine, 35(1): 116–118.
Mikota, S. 2006. Hemolymphatic System. In The Biology, Medicine, and Surgery of
Elephants (pp. 325-345). Blackwell Publishing.
Miller, P., D. Forstein, and S.Styles. 2008. Effect of short-term diet and exercise on
hormone levels and menses in obese, infertile women. Journal of Reproductive
Medicine, 53:315-319.
Moss, A. M., T. H. Clutton-Brock, and S. L. Monfort. 2001. Longitudinal gonadal steroid
excretion in free-living male and female meerkats (Suricata suricatta). General
and Comparative Endocrinology, 122: 158-171.
Mouttham, L. L., M. Buhr, E.W. Freeman, T.M. Widowski, L.H. Graham, and J.L.
Brown. 2011. Interrelationship of serum testosterone, dominance and ovarian
cyclicity status in female African elephants. Animal Reproduction Science,
126(1-2): 115-121.
National Institute for Clinical Excellence. 2004. Fertility Update. National Institute for
Clinical Excellence, Holborn, London.
Nguyen N.T., C.P. Magno, K.T. Lane, M.W. Hinojosa, and J.S. Lane. 2008. Association
of hypertension, diabetes, dyslipidemia, and metabolic syndrome with obesity:
findings from the national health and nutrition examination survey, 1999 to 2004.
Journal of the American College of Surgeons, 207(6):928–934.
Norman, R. J., and A.M. Clark. 1998. Obesity and reproductive disorders: a review.
Reproduction Fertility and Development, 10(1), 55–64.
Olson, D. 2006. African Elephant North American Regional Studbook. Indianapolis, IN:
Indianapolis Zoo.
92
Otto, K. L., J.D. Ferguson, D.J. Fox, and C.J. Sniffen. 1991. Relationship between body
condition score and composition of ninth to eleventh rib tissue in Holstein dairy
cows. Journal of Dairy Science, 74(3): 852–859.
Panidis, D., K. Tziomalos, E. Papadakis, C. Vosnakis, P. Chatzis, and I. Katsikis. 2013.
Lifestyle intervention and anti-obesity therapies in the polycystic ovary
syndrome: impact on metabolism and fertility. Endocrine, April 27
Pasquali, R, and F. Casimirri. 1993. The impact of obesity on hyperandrogenism and
polycystic ovary syndrome in premenopausal women. Clinical Endocrinology,
39(1), 1–16.
Pasquali, R., C. Pelusi, C., S. Genghini, M. Cacciari, and A. Gambineri. 2003. Obesity
and reproductive disorders in women. Human Reproduction Update, 9(4), 359 –
372.
Phythian, C.J., P.J. Cripps, E. Michalopoulou, P.H., Jones, D. Grove-White, M.J.
Clarkson, A.C. Winter, L.A. Stubbings, and J.S. Duncan. 2012. Reliability of
indicators of sheep welfare assessed by a group observation method. Veterinary
Journal, 193(1): 257-63.
Poole, J. 1989. Announcing intent: the aggressive state of musth in African elephants.
Animal. Behavior, 37: 140-152.
Poretsky, L., N.A. Cataldo, Z. Rosenwaks, and L.C. Giudice. 1999. The insulin-related
ovarian regulatory system in health and disease. Endocrine Reviews, 20: 535–
582.
Proctor, C. M., E. W. Freeman, and J.L. Brown. 2010a. Results of a second survey to
assess the reproductive status of female Asian and African elephants in North
America. Zoo Biology, 29, 127–139.
Proctor, C.M., E. W. Freeman, and J.L. Brown. 2010b. Influence of dominance status on
adrenal activity and ovarian cyclicity status in captive African elephants. Zoo
Biology, 29, 168-178.
Ralston, S. L. 2002. Insulin and glucose regulation. The Veterinary Clinics of North
America. Equine Practice, 18(2): 295-304.
Rich-Edwards, J.W., M.B. Goldman, W.C. Willett, D.J. Hunter, M.J. Stampfer, G.A.
Colditz, and J.E. Manson. 1994. Adolescent body mass index and infertility
caused by ovulatory disorder. American Journal of Obstetrics and Gynecology
171: 171–177.
93
Roberts C.K., A.L. Hevener, and R.J. Barnard. 2013. Metabolic syndrome and insulin
resistance: underlying causes and modification by exercise training.Comparative
Physiology, 3(1): 1-58.
Ruan, H., and H. F. Lodish. 2003. Insulin resistance in adipose tissue: Direct and indirect
effects of tumor necrosis factor-alpha. Cytokine Growth Factor Reviews, 14: 447–
455.
Roze, C., C. Doyen, M.F. Le Heuzey, P. Armoogum, M.C. Mouren, and J. Leger. 2007.
Predictors of late menarche and adult height in children with anorexia nervosa.
Clin Endocrinol (Oxf)., 67(3): 462-7.
Saragusty, J., R. Hermes, F. Göritz, D.L. Schmitt, and T.B. Hildebrandt. 2009. Skewed
birth sex ratio and premature mortality in elephants. Animal Reproduction
Science, 115(1-4): 247–254.
Schröder, U. J., and R. Staufenbiel, 2006. Invited Review: Methods to Determine Body
Fat Reserves in the Dairy Cow with Special Regard to Ultrasonographic
Measurement of Backfat Thickness. Journal of Dairy Science, 89(1): 1–14.
Sessions, D. R., S. E. Reedy, M. M. Vick, B. A. Murphy, and B. P. Fitzgerald. 2004.
Development of a model for inducing transient insulin resistance in the mare:
Preliminary implications regarding the estrous cycle. Journal of Animal Science,
82: 2321–2328
Sim, J. and C.C. Wright. 2005. The kappa statistic in reliability studies: use,
interpretation, and sample size requirements. Physical Therapy 85(3): 257-68.
Sirotkin, A.V. 2011. Growth factors controlling ovarian functions. Journal of Cellular
Physiology, 226(9): 2222-5.
Solomon, C.G. and J.E. Manson. 1997. Obesity and mortality: a review of the
epidemiologic data. American Journal of Clinical Nutrition, 66: S1044-50.
Stephenson, T. R., K.J. Hundertmark, C.C. Schwartz, and V.V. Ballenberghe. 1998.
Predicting body fat and body mass in moose with ultrasonography. Canadian
Journal of Zoology, 76(4): 717–722.
Tamer Erel, C., and L.M. Senturk. 2009. The impact of body mass index on assisted
reproduction. Current Opinions in Obstetrics and Gynecology, 21(3): 228-35.
Tena-Sempere, M., and I. Huhtaniemi. 2003. Gonadotropins and gonadotropin receptors,
Reproductive Medicine, Molecular, Cellular and Genetic Fundamentals, New
York, NY: Parthenon Publishing, pp. 225-244.
94
Tena-Sempere, M. 2007. Roles of ghrelin and leptin in the control of reproductive
function. Neuroendocrinology, 86(3):229-41.
Treiber, K.H., D.S. Kronfeld, and T.M. Hess. 2006. Evaluation of genetic and metabolic
predispositions and nutritional risk factors for pasture associated laminitis in
ponies. Journal of the American Veterinary Medical Association, 228: 15381545.
Treiber, K., N. A. Reppert, and A. Ward. 2012. Transuctaneous rump ultrasound of Asian
elephants (Elephas maximus): body fat, body condition and body weight. In: The
7th Crissey Zoological Nutrition Symposium, Raleigh, NC. p 1-6.
Vick, M. M., D.R. Sessions, B.A. Murphy, E.L. Kennedy, S.E. Reedy, and B.P.
Fitzgerald. 2006. Obesity is associated with altered metabolic and reproductive
activity in the mare: effects of metformin on insulin sensitivity and reproductive
cyclicity. Reproduction, Fertility, and Development, 18(6): 609–617.
Vick, M. M., A. A. Adams, B.A. Murphy, D.R. Sessions, D.W. Horohov, R.F. Cook, and
B.J. Shelton. 2007. Relationships among inflammatory cytokines, obesity, and
insulin sensitivity in the horse. Journal of Animal Science, 85(5): 1144-1155.
Volz, P. A., and S.M. Ostrove. 1984. Evaluation of a portable ultrasonoscope in assessing
the body composition of college-age women. Medicine and Science in Sports and
Exercise, 16(1): 97–102.
Wade G.N., J.E. Schneider, and H.Y. Li. 1996. Control of fertility by metabolic cues.
American Journal of Physiology, Endocrinology and Metabolism, 270:E1–E19.
Wagner, J.J., K.S. Lusby, J.W. Oltjetn, J. Rakestraw, R.P. Wettemann, and L.E. Walters.
1988. Carcass composition in mature Hereford cows: Estimation and effect on
daily metabolizable energy requirement during winter. Journal of Animal
Science, 66: 603-612.
Wemmer, C., V. Krishnamurthy, S. Shrestha, L.A. Hayek, M. Thant, and K.A. Nanjappa.
2006. Assessment of body condition in Asian elephants (Elephas maximus). Zoo
Biology, 25(3): 187–200.
Westervelt, R. G., J.R. Stouffer, J.F. Hintz, and H.F. Schryver. 1976. Estimating Fatness
in Horses and Ponies. Journal of Animal Science, 43(4): 781–785.
Whitehouse, A. M. and A.J. Hall-Martin. 2000. Elephants in Addo Elephant National
Park, South Africa: Reconstruction of the population's history. Oryx, 34: 46-55.
95
Wielebnowski, N. 1998. Contributions of behavioral studies to captive management and
breeding of rare and endangered mammals. In: Behavioral Ecology and
Conservation Biology (Ed. by Caro, T.), pp. 130-162. New York: Oxford
University Press.
Wiese, R. J., and K. Willis. 2006. Population management of zoo elephants. International
Zoo Yearbook, 40(1): 80–87.
Wildman, E. E., G.M. Jones, P.E. Wagner, R.L Boman, H.F. Troutt, and T.N. Lesch.
1982. A dairy cow body condition scoring system and its relationship to selected
production characteristics. Journal of Dairy Science, 65: 495-501.
Wilkinson, M. J., and N.A. McEwan. 1991. Use of ultrasound in the measurement of
subcutaneous fat and prediction of total body fat in dogs. The Journal of
Nutrition, 121(11 Suppl): S47–50.
Willis, D., H. Mason, C. Gilling-Smith, and S. Frank. 1996. Modulation by insulin of
follicle-stimulating hormone and luteinizing hormone actions in human granulosa
cells of normal and polycystic ovaries. Journal of Clinical Endocrinology and
Metabolism, 81(1): 302-9.
Xu, H., G. T. Barnes, G. Yang, G. Tan, D. Yang, C. J. Chou, J. Sole, A. Nichols, J. S.
Ross, L. A. Tartaglia, and H. Chen. 2003. Chronic inflammation in fat plays a
crucial role in the development of obesity-related insulin resistance. Journal of
Clinical Investigation, 112: 1821–1830.
Yen, S.S. 1980. The polycystic ovary syndrome. Clinical Endocrinology, (Oxf), 12(2):
177-207.
96
BIOGRAPHY
Kari Ann Morfeld is from Creston, Nebraska and attended St. Edward High School in St.
Edward, Nebraska. She earned her Bachelor of Arts in Biology from Nebraska Wesleyan
University in Lincoln, Nebraska in 2001 and her Master‟s of Science in Animal Sciences,
specializing in Reproductive Physiology, from the University of Nebraska in Lincoln,
Nebraska. She is the director of a new Wildlife Endocrinology Laboratory focused on
elephant health she recently established at the Lincoln Children‟s Zoo in Lincoln,
Nebraska in collaboration with the Smithsonian‟s National Zoological Park and the
Smithsonian Conservation Biology Institute.
97