THE SHORT- AND LONG-TERM EFFECTS
OF METHOTREXATE ON THE RAT SKELETON
DONNA L. WHEELER
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
I would like to thank Dr. Robert E. Vander Griend for suggesting the area
of chemotherapy-induced osteopenia for study, and for his guidance and
encouragement throughout this work. I would like to thank Dr. R. William Petty and
the Department of Orthopaedics for generously supplying the funds for this project.
I am also thankful for the contributions of Dr. Thomas J. Wronski to my mastery of
histomorphometry and understanding of osteoporosis. I am also indebted to Dr.
Gary J. Miller, Dr. James E. Graves, Dr. Scott K. Powers, and Dr. David Lowenthal
for their guidance and instruction, enabling me to develop as a scientist.
I would like to acknowledge the loyal support of my friend and colleague,
Ernest E. Keith. His expertise, instruction, and assistance in the care of laboratory
animals were fundamental to the completion of this project. He also provided
valuable assistance in histomorphometric processing. Special thanks are extended to
Mia Park for her assistance in animal care, tissue processing, data acquisition, data
processing, and data entry.
I am indebted to Dr. Martha Campbell-Thompson and the Department of
Gastroenterology for the use of their microscope and Vidas imaging equipment. I
would also like to thank the Department of Exercise and Sport Sciences for the use
of their dual-energy x-ray absorptiometer and to Lunar Corporation for supplying the
software needed to use this machine.
Finally, I would like to acknowledge the support of Kris Billhardt. Her love,
friendship, inspiration, patience, and editing skills were instrumental in the
completion of this research. I would also like to thank my parents, Jack and Jane
Wheeler, for their undying support and encouragement.
TABLE OF CONTENTS
ACKNOWLEDGMENTS ...... .................................. ii
LIST OF TABLES ...... ....................................... vi
LIST OF FIGURES ..... ......................................... vii
ABSTRACT ......... ....................................... xi
1 INTRODUCTION ................................... 1
Bone Remodeling ................ ........ ........ 3
Involutional Bone Loss .......................... 8
Mineral Regulating Mechanisms ................... 11
Mineral Regulating Hormones .. ............ 11
Growth Regulating Hormones ............... 15
Sex Hormones ........................... 16
Exercise ................................ 17
Types of Osteoporosis ........................... 18
Problem Statement ............................. 18
Research Objectives ....................... 19
Hypotheses ............................. 20
Delim stations ............................ 21
Lim stations ............................. 21
2 REVIEW OF THE LITERATURE ...................... 22
Clinical Research .............................. 23
Animal Research .............................. 28
Methods of Skeletal Assessment ................... 30
3 MATERIALS AND METHODOLOGY .................. 33
Anim al Care .................................. 33
Bone Histomorphometry ......................... 35
Cancellous Bone ......................... 35
Cortical Bone ............................ 38
Quantification of Bone Parameters ............ 39
Biomechanical Testing .........
Dual-Energy X-Ray Absorptiometry
Statistical Analysis .............
4 RESULTS .. .....................
Bone Histomorphometry ........
Cancellous Bone ........
Cortical Bone ..........
Dual-Energy X-Ray Absorptiometry
5 DISCUSSION .....................
Recommendations for Future Work
CANCELLOUS BONE FIXATION, DEHYDRATION
AND METHYL METHACRYLATE EMBEDDING ......
MODIFIED VON KOSSA STAIN ....................
CORTICAL BONE FIXATION, DEHYDRATION
AND EMBEDDING IN BIOPLASTIC .................
COMPUTER CODE FOR IMAGE ANALYSIS ..........
DEXA REPEATABILITY STUDY ....................
SAS PROGRAMS FOR STATISTICAL ANALYSIS .......
QUICK REFERENCE FOR ABBREVIATIONS .........
REFERENCES ........ ..............................
BIOGRAPHICAL SKETCH .............................
LIST OF TABLES
Factors associated with osteoporosis .................... ....... 2
Effects of mineral regulating hormones on
serum calcium and phosphate ............................ 13
Cancellous bone parameters ............................. 52
Femoral cortical bone parameters ................ ...... 70
Tibial cortical bone parameters ........................... 71
Femoral torsional biomechanical parameters ............... 104
Tibial torsional biomechanical parameters .................. 105
Dual-energy x-ray absorptiometry values for BMD ........... 120
Results of DEXA reliability study ........................ 160
Standard Abbreviations ............................... 164
LIST OF FIGURES
1.1 Cancellous bone remodeling .............................. 4
1.2 Cortical bone remodeling ................................ 5
1.3 Involutional bone loss ................................. 10
2.1 Mechanism of action of Methotrexate ...................... 24
3.1 Photograph of femur and tibia with ends
embedded in low melting-point metal ...................... 42
3.2 Graphical depiction of biomechanical parameters ............. .45
4.1 Rat weight changes with time ........................... 49
4.2 Tibial cancellous bone volume ........................... 53
4.3 Tibial cancellous osteoclast surface ........................ 54
4.4 Tibial cancellous longitudinal bone growth .......... ...... .55
4.5 Tibial cancellous mineralizing surface ..................... ..56
4.6 Tibial cancellous mineral apposition rate ................... 57
4.7 Tibial cancellous bone formation rate ..................... .58
4.8 Photomicrograph of baseline cancellous bone volume .......... 59
4.9 Photomicrographs of cancellous bone volume at 30 days ........ 60
4.10 Photomicrographs of cancellous bone volume at 80 days ....... 61
4.11 Photomicrographs of cancellous bone volume at 170 days ....... 62
Photomicrographs of fluorescent labels on
cancellous bone surfaces at 30 days ....................... 63
Photomicrographs of fluorescent labels on
cancellous bone surfaces at 80 days ....................... 64
Photomicrographs of fluorescent labels on
cancellous bone surfaces at 170 days ........................ ..65
Femoral total bone tissue area ..... ..................... 72
Tibial total bone tissue area ............................ 73
Femoral marrow area .................................. 74
Tibial marrow area ................................... 75
Femoral cortical bone area .............................. 76
Tibial cortical bone area ............................... 77
Femoral mean cortical bone width ........................ 78
Tibial mean cortical bone width .......................... 79
Femoral polar moment of inertia ......................... 80
Tibial polar moment of inertia ........................... 81
Femoral periosteal mineralizing surface .................... 82
Tibial periosteal mineralizing surface ................... ... 83
Femoral periosteal mineral apposition rate ................. .84
Tibial periosteal mineral apposition rate .................... 85
Femoral periosteal bone formation rate ................... .86
Tibial periosteal bone formation rate ...................... 87
Photomicrograph of the femoral cross-section
of the baseline control animal ........................... 88
Photomicrograph of the tibial cross-section
of the baseline control animal ........................... 89
of femoral cross-sections at 30 days
of tibial cross-sections at
of femoral cross-sections
of tibial cross-sections at
of femoral cross-sections
30 days ...
at 80 days .
80 days .
at 170 days
of femoral periosteal surface
of tibial periosteal surface at
of femoral periosteal surface
of tibial periosteal surface at
of femoral periosteal surface
Photomicrographs of tibial periosteal surface at
at 30 day
at 80 day
at 170 da
. ...... 95
's ...... 96
,s ...... 98
iys .... 100
Photograph of a typical fracture pattern following torsional test
Femoral breaking torque ..............................
Tibial breaking torque ................................
Femoral twist angle at failure ...........................
Tibial twist angle at failure .............................
Femoral energy absorbed at failure ......................
Tibial energy absorbed at failure ........................
Femoral torsional stiffness .............................
Tibial torsional stiffness ...............................
of tibial cross-sections at 170 days
4.54 Femoral torsional strength ............................. 115
4.55 Tibial torsional strength ............................... 116
4.56 Femoral polar moment of inertia associated with torsional fracture 117
4.57 Tibial polar moment of inertia associated with torsional fracture 118
4.58 Femoral bone mineral density .......................... 121
4.59 Tibial bone mineral density ............................. 122
4.60 Vertebral bone mineral density ......................... 123
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
The Short- and Long-term Effects of Methotrexate on the Rat Skeleton
Donna L. Wheeler
Chairman: James E. Graves, Ph.D.
Major Department: Health and Human Performance
The chemotherapy drug, Methotrexate (MTX), has been shown to decrease
bone mass and increase the incidence of bone fractures. Existing studies measure
bone parameters at one point in time following chemotherapy and do not monitor
the bone's recovery. This study's purpose was to use a rat model to determine the
long-term effects of MTX on bone volume, turnover, mineralization, density, and
Sprague-Dawley rats were randomly assigned to either control or MTX
groups. Daily MTX and saline injections were given for two five-day courses to drug
treatment and control groups, respectively. Fluorochrome compounds were injected
prior to sacrifice to monitor actively mineralizing bone surfaces. One control and
MTX group were sacrificed at 30, 80, and 170 days following treatment initiation.
Both femurs, both tibias, and 2 lumbar vertebrae were harvested for cancellous and
cortical bone histomorphometry, biomechanics (torsion), and densitometry (DEXA).
Standard cancellous and cortical histomorphometric parameters were measured from
undecalcified, methyl methacrylate embedded sections from the right proximal tibia
and tibial and femoral diaphyses. The contralateral femur and tibia were torsionally
loaded to failure and standard mechanical parameters were measured. Excised
bones were scanned using DEXA to measure bone density. All bone responses were
statistically analyzed using a 2-way ANOVA followed by Duncan's multiple
comparison procedure (significance: p=0.05).
Cancellous bone volume was significantly lower in the MTX treated group at
80 and 170 days. Cancellous mineralizing surface and longitudinal bone growth were
significantly depressed at all time points yet mineral apposition rate was only
depressed at the 170 day point. Cancellous osteoclast surface was increased at all
time points for the MTX treated animals. Cortical cross-sectional area and
periosteal mineral apposition rate were significantly lower for both the femur and
tibia in the MTX groups at all time points. However, periosteal mineralizing surface
was significantly depressed in the MTX group only for the femur. MTX had minimal
effects on the biomechanical parameters and bone density measurements.
Methotrexate had long-term negative effects on both cortical and cancellous
bone. The drug decreased bone volume, decreased bone formation, and decreased
The disease osteoporosis results from a decrease in bone mass and bone
strength leading to an increase in the incidence of bone fractures. Osteoporosis is
an enormous public health problem which is responsible for over a million bone
fractures in the U.S. each year (61). The most common sites of osteoporotic
fractures are the vertebrae, hip, and distal forearm. One-third of all women over 65
will have vertebral fractures and, by the eighth decade of life, one of every three
women and one of every six men will have had a hip fracture (61). The direct and
indirect costs of osteoporosis are estimated to be over $6.1 billion annually in the
Peak bone mass is achieved by the third decade of life and then slowly
declines from this point. The magnitude of peak bone mass as well as the rate at
which bone is lost contribute to the development of osteoporosis. Many factors have
been shown to be associated with osteoporosis with the most important being age,
gender, race, and hormonal status. Osteoporosis is most commonly seen in
postmenopausal white women, Asian women, and older men (60). Other factors
such as genetics, lifestyle, nutritional factors, medical disorders, or drugs have been
shown to significantly affect bone health. A summary of osteoporosis risk factors are
presented in Table 1.1.
Table 1.1 Factors Associated with Osteoporosis
White or Asian ethnicity
Positive family history
Small body frame (<127 lbs)
Excessive exercise (producing amenorrhea)
Early natural menopause
Nutritional Factors Milk intolerance
Low calcium intake
Excessive alcohol intake
Consistently high protein intake
Medical Disorders Anorexia nervosa
Type I diabetes
Abnormal gastrointestinal function
Abnormal hepatobiliary function
Occult osteogenesis imperfecta
Drugs Thyroid replacement therapy
Chronic lithium therapy
GnRH agonist or antagonist therapy
The following sections will discuss bone remodeling, involutional bone loss,
physiological mineral regulating mechanisms, types of osteoporosis, and will conclude
with the development of the problem being studied.
Throughout the body, bone is continually being remodeled, where bone is
resorbed and replaced by new bone. In order to maintain bone mass it is essential
that this remodeling process remain balanced. The amount of bone resorbed should
be replaced by an equal or greater amount of new bone or there will be a net loss
of bone mass. If this imbalance remains uncorrected this condition can lead to
osteoporosis. This section will describe the normal bone remodeling sequence and
discuss problems which might affect the normal cycle.
Figure 1.1 and Figure 1.2 describe the normal sequence of cancellous and
cortical bone remodeling, respectively. Remodeling occurs in a programmed
sequence at discrete bone foci called bone remodeling units (BMU) (27). The steps
of bone remodeling can be described as ACTIVATION, RESORPTION, and
FORMATION and are similar for cancellous and cortical bone.
At the beginning of each cancellous remodeling cycle (Figure 1.1), activation
occurs at previously inactive bone surfaces. These surfaces are covered with bone
lining cells, presumably of osteoblastic lineage. These lining cells respond to bone-
resorbing hormones and expose the bone surface to osteoclast progenitor cells.
Osteoclast progenitors then replicate and differentiate into mature osteoclasts. The
Cancellous bone remodeling
Oc=osteoclast; Opc=osteoprogenitor cells;
Ob=osteoblast; BLC=bone lining cells
(a) Resorption is completed before formation
(b) Resorption followed closely by formation
(Figure adapted from W.S.S.Jee (37))
a II ||
E mewU 4
S ZII II
osteoclasts secrete lysosomal enzymes that degrade the collagen matrix and hydrogen
ions which dissolve the bone mineral. After the osteoclasts resorb the bone there is
a reversal phase in which macrophages may appear to either complete resorption or
initiate the next stage of remodeling, the formation phase. During formation,
osteoblasts replace the bone that was resorbed by the osteoclasts. For successful
bone formation the osteoblasts must be able to adequately replicate, differentiate,
and produce normal bone matrix. Since osteoblasts lay down bone on existing bone
spicules, it is important for there to be a template of unresorbed bone in the
resorption cavity on which the osteoblasts may build the new bone matrix.
Cortical bone remodeling is similar, in principle, to cancellous remodeling.
The osteoclasts first resorb a canal or tunnel which is then refilled centripetally by
osteoblastic apposition, forming a new osteon or secondary Haversian system (Figure
The completion of the osteoclastic resorption phase in humans requires
approximately 21 days for cancellous bone and 24 days for cortical bone. The
osteoblasts then create a new structural bone unit in approximately 91 days and 124
days for cancellous and cortical bone, respectively. Therefore, the completion of a
cancellous BMU requires 112 days, and a cortical BMU requires 147 days (39).
In normal young adults, the resorption and formation phases are tightly
coupled and bone mass is maintained. However, there is a natural bone loss with
age which implies an uncoupling or imbalance in the phases of bone remodeling, with
an increase in resorption over formation. The slow age-dependent phase of bone
loss results mainly from impaired bone formation, where the osteoclasts create
resorption cavities of normal depth but the osteoblasts fail to refill them completely
(21,56). In other words, age-related bone loss occurs due to reduction in the rate of
bone formation with no change in the rate of bone resorption (39). Over time, the
uncoupled remodeling cycles can cause significant thinning of spicules of cancellous
bone until the osteoclastic resorption cavities penetrate the thickness of bone. These
resorptive perforations create discontinuities in the cancellous scaffold, resulting in
loss of mechanical integrity and strength of the bone.
Increased bone turnover (activation of new BMUs) can also result in net bone
loss. Postmenopausal accelerated bone loss is associated with a high rate of bone
turnover (80), where more osteoclasts are present and these cells create deeper
resorption cavities (21,56). Although there is more bone being played down due to
the increased bone turnover, there is even greater bone resorption (38).
Drug induced osteoporosis may affect the normal remodeling cycle in a variety
of ways depending on the cellular actions of the drug. The drug can either increase
or decrease the number of osteoblasts or osteoclasts or affect the quality and quantity
of bone being played down by osteoblasts. Cytotoxic drugs may also affect bone
metabolism indirectly by inducing gonadal damage, thus decreasing the levels of
circulating sex hormones. The mechanism of drug-induced bone loss is typically due
to diminished osteoblastic number, activity, or quality of bone.
The focus of this study is on the chemotherapy drug, Methotrexate, which
interferes with the replication of DNA within the cells. Faulty DNA replication
would decrease the number of osteoblasts and osteoclasts and would, therefore,
decrease the overall bone formation and resorption. Based on Methotrexate's
mechanism of action, this drug would decrease the surface of bone lined by
osteoblasts and osteoclasts and reduce the overall bone turnover.
Involutional Bone Loss
Both men and women experience age-related bone loss. Bone mass continues
to increase until approximately the age of 30, then after a transient period of
stability, age-related bone loss begins. Women lose approximately 35% of their
cortical bone mass and 50% of cancellous bone mass over their lifetime, whereas
men lose only 23% and 33% of their cortical and cancellous bone, respectively (50).
Bone loss occurs in a biphasic pattern for both cortical and cancellous bone in
women; with a sustained slow phase that occurs in both sexes and a transient
accelerated phase that occurs in women after menopause.
For cortical bone, bone is lost at approximately 0.3 to 0.5 %/year and
increases with aging until it levels out late in life (50). For women, an accelerated
postmenopausal phase of cortical bone loss is superimposed on the slow phase to
increase bone loss to a rate of 2 to 3 %/year immediately after menopause. This
accelerated bone loss decreases exponentially to baseline slow phase levels after
about 8 to 10 years (50).
Cancellous bone loss starts to occur earlier in life than cortical bone loss.
Controversy exists as to the rate of cancellous bone loss. Some researchers claim a
linear decrease of 0.6 %/year (61), while others predict a curvilinear decrease of
2.4 %/year in women (11) and a linear 1.2 %/year for men (51). Because cancellous
bone has a greater surface area than cortical bone, it is more metabolically active
and, therefore, more responsive to changes in external loading (exercise), mineral,
hormonal, and metabolic status. The effect of hormonal status is reflected in the
earlier onset of bone loss for cancellous bone and the more rapid rate of cancellous
bone loss following menopause compared to cortical bone. This accelerated
cancellous bone loss, however, is shorter in duration than the accelerated
postmenopausal phase of cortical bone loss (61). In addition to involutional (age-
related) bone loss, drugs and diseases can exacerbate the bone loss by either
increasing the rate of bone loss, diminishing peak bone mass or both. In general, the
development of osteoporosis is related to two factors: (1) level of peak bone mass,
and (2) the rate of loss of bone after achieving peak bone mass.
Figure 1.3 graphically represents natural bone loss in humans. As discussed
above, peak bone mass occurs at the approximate age of 30 and slowly declines at
a constant rate due to age-associated changes. In women, an accelerated bone loss
occurs at menopause where the contributions of age and hormonal status combine.
Following this accelerated phase, bone loss resumes a slow decline until late in life,
while bone mass appears to stabilize at an osteopenic level. If a disease requires
drug therapy such as glucocorticoids, anticonvulsants, or chemotherapy (4,7,24,30,61),
the age-related bone loss is confounded by drug induced bone loss. As can be seen
in Figure 1.3, the bone fracture threshold is also an important consideration in the
5 35 45 55 65 75 85
-Males -- Females
Involutional bone loss
(curves based on 1.2%/year)
bone density spectrum. This threshold represents the level of bone density at which
a person's statistical risk for osteoporotic fractures increases. As depicted in Figure
1.3, the lower the peak bone mass and the greater the rate of bone loss the sooner
a person will reach this fracture threshold.
Mineral Regulating Mechanisms
Endocrine status, endogenous mineral balance, and mechanical loading
regulate both acquisition of peak bone mass and rates of bone loss and gain. The
following sections will discuss the effects of systemic mineral-regulating hormones,
growth-regulating hormones, sex hormones, and exercise on bone.
Bone is the reservoir for 99% of the body's calcium. Therefore, hormones
which regulate calcium homeostasis are extremely important for the regulation of
bone mass. In addition to the hormones responsible for calcium metabolism,
adequate dietary intake is essential for healthy bones. Only 30% of ingested calcium
is absorbed into the system; therefore, the consumption of at least 800 mg of calcium
a day to prevent the bone reserve of calcium from being depleted is very important.
With age, less calcium is absorbed, and the daily intake of calcium should be
increased in order to maintain calcium homeostatis and healthy bones. Calcium is
the most common nutritional supplement given to postmenopausal and osteoporotic
patients. Studies continue to show, however, that calcium is not effective by itself in
preventing postmenopausal bone loss or reversing osteoporosis. Calcium intake may
be more important during the period of growth and development to allow
maximization of peak bone mass (37).
The goal of mineral-regulating hormones is to maintain physiologic serum
calcium, magnesium, and phosphate levels. The three main mineral-regulating
hormones are parathyroid hormone (PTH), calcitonin (CT) and 1,25dihyroxyvitamin
D (calcitriol or 1,25D). These hormones act on three target tissues: bone, intestine,
and kidney. Table 1.2 supplies a summary of the actions of these mineral-regulating
hormones in the regulation of serum calcium (Ca) and phosphate (Pi) at the different
target tissues. Magnesium is not included in this table; its regulation by these
hormones is either nonexistent or inadequately defined. The role of calcitonin in
mineral regulation is modest compared to the effects of PTH and 1,25D.
Vitamin D is either obtained through dietary foodstuffs or absorbed from the
sun. In both cases this form of "raw"vitamin D is not active, and a series of enzymes
in the liver and kidney must manufacture the active form, 1,25D. PTH is secreted
by the parathyroid gland in response to low levels of serum calcium and 1,25D.
Calcitonin is secreted by the thyroid gland in response to hypercalcemia. The actions
and interaction of these hormones maintain mineral homeostasis.
Ca and Pi enter the blood through the intestine, leave it through the kidney,
and are stored in the bone. In order to maintain homeostasis, the net absorption of
Ca and Pi in the intestine must be equal to the net excretion of these ions by the
kidney. The absorption of Ca and Pi depends on the dietary intake and the
Effects of Mineral-Regulating Hormones on Serum Calcium and Phosphate
Bone Gut Kidney NET
Ca Pi Ca Pi Ca Pi Ca Pi
PTH + 4' 4 ,
CT 4, 4, *, ,- 4 ,
1,25D t + +
KEY: PTH = parathyroid hormone
CT = calcitonin
1,25D = 1,25-dihydroxyvitamin D
Ca = calcium
Pi = phosphate
t = increase in serum mineral level when hormone acts on target tissue
= decrease in serum mineral level when hormone acts on target tissue
= no effect in serum mineral levels when hormone acts on target tissue
(Figure adapted from D.D. Bikle (6))
efficiency of absorption. Glomerular filtration of these ions in the kidney is relatively
constant, so the kidney reabsorbs Ca and Pi relative to the body's needs. Bone
provides the major buffer for maintaining constant levels of Ca and Pi in the blood.
This is achieved by balancing bone formation (which deposits these ions in bone)
with bone resorption (which releases these ions into the bloodstream). PTH, 1,25D,
and calcitonin act on the intestine, kidney, and skeleton and maintain mineral
balance. 1,25D has the positive effect of increasing serum Ca and Pi levels by
increasing absorption in the intestines and increasing reabsorption in the kidney.
However, 1,25D in combination with PTH mobilizes Ca and Pi from bone. The net
effect of 1,25D is to increase both serum Ca and Pi.
PTH regulates serum Ca and Pi by stimulating the release of these ions from
the skeletal reservior and by stimulating Ca reabsorption and inhibiting Pi
reabsorption in the kidney. PTH also affects 1,25D levels by stimulating its renal
synthesis which, in turn, leads to increased intestinal absorption of Ca and Pi. The
net effect of PTH is to increase serum Ca and decrease serum Pi. PTH levels have
been shown to increase with age, which may partly explain age-related bone loss (45).
Chronically high PTH level (hyperparathyroidism) increases the number and activity
of osteoclasts (6). However, animal studies have shown that intermittent (pulsatile)
doses of PTH have the opposite effect, stimulating bone formation and resulting in
significant increases in cancellous and cortical bone (85,86). Human studies have
also confirmed that pulsatile PTH administration increases bone mineral density in
postmenopausal women and proves to be a promising prevention and treatment for
Calcitonin, although of less importance to mineral homeostasis than PTH and
1,25D, is a potent inhibitor of bone resorption by decreasing the activity and number
of osteoclasts. The secretion of CT is regulated by serum calcium levels. High
serum Ca stimulates CT secretion, which decreases the release of Ca to the blood
by bone resorption. Calcitonin is currently used for the treatment of osteoporosis.
Treatment of high turnover osteoporosis (postmenopausal) with calcitonin has been
shown to increase bone mass but its effects are less consistent and sustained than
estrogen therapy (60).
1,25D, PTH, and calcitonin act on the skeleton, kidney, and intestines to
regulate blood levels of Ca and Pi. Regulation entails control of how much comes
into the body from the diet, how much leaves the body through the kidney, and how
much is stored and released from the bone. The different hormones, ions, and target
tissues involved communicate and interact to ensure the precise regulation of these
Several systemic hormones associated with growth are important in the
development of peak bone mass and strength. These hormones include growth
hormone, glucocorticoids, thyroid hormone, and insulin. Growth hormone is a very
important determinant of skeletal mass and acts through changes in the production
of insulin-like growth factor 1 (IGF-1), which is a potent stimulator of skeletal growth
(10,60). Glucocorticoids have complex effects on bone metabolism. Excess
glucocorticoids produce bone loss primarily through suppression of osteoblast
function. They also inhibit intestinal calcium absorption leading to secondary
hyperparathyroidism and increased bone resorption. Glucocorticoids also promote
bone loss by inhibiting renal calcium reabsorption which may contribute to increased
PTH secretion (34). Thyroid hormones increase bone turnover but tend to stimulate
resorption more than formation, resulting in net bone loss. Insulin has been shown
to stimulate osteoblastic collagen synthesis at physiological concentrations and may
also affect calcium transport and vitamin D metabolism (60).
It has been well established that estrogen deficiency leads to the development
of osteoporosis. Accelerated bone loss has been noted in postmenopausal women
or after oophorectomy (43,54,79-82,84,85), as well as in ammenorrheic women
(19,57). Estrogen replacement therapy has been helpful in preserving bone mass
under estrogen deficient conditions in humans (18,36) and animals (29,80). A
similar link to osteoporosis has been noted in androgen deficient men (22). Upon
withdrawal of estrogen or androgen, bone turnover is increased, where bone
resorption is greater than bone formation, resulting in a net loss of bone mass.
Another important factor promoting bone health is the presence of
mechanical stimulation which provides impetus for modeling, remodeling, and
mineralization (12,26,63-65,75). Weightbearing exercise, therefore, becomes an
important modality to preserve or increase bone mass. Weight-bearing exercise has
been shown to augment bone mass in athletes (41,46,49,78)and maintain or improve
bone mass in aging individuals and postmenopausal women (1,14,32,72). In fact, men
and women subjected to high intensity physical training have been shown to increase
their bone mineral content (BMC) and bone mineral density (BMD) between 5 and
20% when compared to age-matched sedentary controls (46,49). Some studies have
shown that exercise tends to be site specific, augmenting bone mass only in the
specific bones loaded during the exercise routine (41,78). Other researchers (32)
have found an increase in BMC in both appendicular and axial skeleton with a
combined exercise program which incorporated both weight-bearing aerobic exercise
and strength training.
As described in the previous paragraphs, bone homeostasis is affected by many
factors including natural aging, nutritional state, hormonal status, and activity level.
When the equilibrium is upset by alterations in body function due to disease, surgery,
or drug therapy, drastic changes in bone mass can occur.
Types of Osteoporosis
Osteoporosis can be classified into three types: (1) Type I or Postmenopausal,
(2) Type II or Senile, and (3) Type III or Therapy Induced (5). Type I and type II
osteoporosis result from either lack of estrogen at menopause (women) or from
natural aging (men and women), respectively. Type I osteoporosis affects mainly
cancellous bone with the most common fracture sites being the vertebrae (crush
fractures) and the distal radius. Type II osteoporosis affects both cancellous and
cortical bone with the most common fracture sites being the vertebrae (multiple
wedge fractures) and hip. Type III osteoporosis can result from chronic
administration of pharmacological agents which affect circulating levels of
parathyroid hormone, vitamin D metabolites, osteocalcin, calcium absorption, and
renal conservation of calcium. Type III osteoporosis can also occur from short- or
long-term treatment of cytotoxic (chemotherapy) drugs such as Adriamycin or
Drugs are given to cure disease or improve the function of organ systems. In
the case of cancer, chemotherapy drugs are used to kill actively growing cells. While
cancer cells are usually actively growing and, therefore, affected by the
chemotherapy, the drugs do not discriminate between neoplastic and normal rapidly
growing cells. Chemotherapy, therefore, can have deleterious effects on non-targeted
tissue such as epithelial tissue (i.e. hair follicles, lining of the gut), hematopoietic
marrow cells, and bone cells, resulting in hair loss, gastrointestinal distress,
immunosuppression, and osteoporosis.
Osteoporosis is a silent disease, since the health of the skeleton is not
outwardly apparent and the symptoms are minimal until catastrophic fracture.
Chemotherapy may attend to the more obvious disease, cancer, while creating other
diseases in the process. The focus of this work is on the secondary disease,
osteoporosis, resulting from administration of the chemotherapy drug, Methotrexate.
This project will explore the pathophysiology and severity of Methotrexate-induced
osteoporosis using an animal model. The recovery of bone after withdrawal of drug
treatment will also be investigated. Histomorphometry, biomechanics, and dual-
energy x-ray absorptiometry will be used to asess bone quantity and quality.
MTX has been shown to have immediate adverse effects on bone turnover
and fracture healing using a rat model (23,24,35,74). However, administration of
more than one course of MTX and long-term bone recovery have not been
investigated. Therefore, the objectives of this study are outlined below:
(1) To study the short- and long-term effects of MTX on the skeleton by using 3
month old male Sprague-Dawley rats.
(2) To quantify the effects of MTX on bone characteristics at different times after
withdrawal of MTX treatment using histomorphometric, biomechanical, and
(3) To verify the pathophysiology of MTX-associated osteoporosis proposed by
(4) To propose changes in the existing chemotherapy protocols to prevent or treat
The Sprague Dawley rat will be used to test the following hypotheses:
(1) Two courses of Methotrexate will induce osteopenia and depress bone
formation 30 days following treatment as reported by others after one course
of Methotrexate (24).
(2) Osteoblast and osteoclast function will return to normal (age-matched control
levels) after the drug is cleared from the tissues at approximately 170 days.
(3) Although cellular function will be restored after 170 days, the osteopenia
resulting from Methotrexate influence will persist. Methotrexate will,
therefore, cause sustained loss of cancellous and cortical bone, bone mineral
density, and mechanical strength.
The following demarcations are recognized in this study:
(1) Only male Sprague-Dawley rats were used in this study.
(2) Two 5-day courses of Methotrexate chemotherapy were given to the rats.
(3) The quantity and quality of the skeleton was examined at three points in time
following completion of chemotherapy: 30, 80, and 170 days.
The following weaknesses are acknowledged in this study:
(1) Animals were studied instead of humans.
(2) Longitudinal measures of bone quality and quantity were not made due to the
invasive nature of some of these measurements.
(3) Complete clearance of methotrexate from bone was not achieved at 170 days,
therefore, recovery was not fully recognized.
REVIEW OF THE LITERATURE
Cancer is a far-reaching disease with often tragic outcomes. Cancer cells can
arise in any body tissue, at any age. Cancer cells can invade local tissues by direct
extension or they can spread throughout the body by way of lymphatic or vascular
channels. The size and scope of cancer disease is overwhelming. Approximately 56
million Americans will be diagnosed with cancer -- this is approximately 1 in 4.
These statistics make it extremely likely that each of us will face the disease at some
point in our life, either personally or through the care of a loved one. Although two
thirds of those diagnosed with cancer will die of their disease, the percentages of
survivors are increasing due to improved efficacy of adjuvant therapies such as
radiation therapy and chemotherapy.
Chemotherapy agents, although often effective in destroying and controlling
neoplastic cell growth, have many adverse side effects on normal tissue and body
function. Many drugs cause nausea, vomiting, alopecia, cardiotoxicity, anorexia,
myelosuppression, renal damage, liver toxicity, and osteopenia. The primary interest
of this investigation concerns the chemotherapy drug Methotrexate and the adverse
effects of this drug on the skeleton.
Methotrexate (MTX) is a common antineoplastic agent used to treat acute
lymphoblastic leukemia, choriocarcinoma in women, breast carcinomas, testicular
carcinomas, head and neck carcinomas, osteosarcomas, chondrosarcoma,
fibrosarcoma, liposarcoma, lymphosarcoma, Hodgkin's disease, lung cancer
(squamous and small cell types), and, at lower doses, severe psoriasis and rheumatoid
arthritis. MTX is classified as an antimetabolite and is a folic acid analog. It
competitively inhibits the enzyme dihydrofolate reductase. This enzyme, involved in
protein synthesis, catalyzes a reaction to convert nucleic acids to DNA (see Figure
2.1). Therefore, MTX inhibits DNA synthesis by depleting the cell of the DNA
building blocks. MTX is carried by the blood and about 90% is filtered out of the
bloodstream within 48 hours. However, a large percentage of the drug remains in
the tissues where it was captured and its effects are very long-lasting and often
considered irreversible, especially with chronic MTX therapy (35).
MTX has been shown to have an adverse effect on the skeleton in both
humans and animals. The following sections will detail existing research in this area,
both clinical and animal. Methods to quantify bone changes will be reviewed.
Nesbit et al. (52) studied children with acute lymphocytic leukemia who were
treated with high doses of MTX. The most apparent toxicity with these children
were associated with pulmonary inflammation (pneumonia) as well as inflammation
of bladder, vaginal, and pleural epithelium. Ulceration of the gastrointestinal tract
was also noted. Chronic administration of MTX resulted in liver cirrhosis and
osseous changes. MTX associated hepatotoxicity (16) and osteoporosis (59) has also
THYMIDYLIC ACID SYNTHESIS
Figure 2.1 Mechanism of action of Methotrexate
NH l N
C-n- CONH CH
been reported by others. Approximately 20% of Nesbit's patients receiving chronic
MTX treatment experienced bone pain and/or fractures associated with osteopenia;
the researchers attribute these osseous changes to MTX induced abnormalities in
calcium metabolism hypercalcemiaa).
Atkinson et al. (4) studied children with acute lymphoblastic leukemia in
attempts to elucidate the physiologic mechanism leading to osteoporosis. It was
uncertain if the cancer itself was affecting the bone mineral or if the chemotherapy
used to treat the disease was affecting the gastrointestinal and renal handling of
nutrients, causing alterations in mineral homeostasis and leading to abnormal
turnover of bone mineral and osteoporosis. The chemotherapy protocols in this study
used a combination of the following drugs over a 24 month period: prednisone,
vincristine, L-asparaginase, methotrexate, 6-mercaptopurine, and doxorubicin. This
study provided evidence that the chemotherapy protocol mentioned above caused an
imbalance in the mineral homeostasis including hypomagnesium, hypocalcemia, and
hypoparathyroidism. These abnormal calcium and magnesium levels, caused by
chemotherapy, indirectly altered bone turnover.
Ragab et al. (59) reported on 11 children with acute lymphoblastic leukemia
being treated with MTX therapy for more than 6 months who developed severe bone
pain and/or fractures in their lower extremities. These children were diagnosed
radiographically with osteopenia. Four of the patients were withdrawn from MTX
therapy. Serial radiographs indicated improvements in bone density after 6 drug-free
months. In contrast to Nesbit's study, these children did not experience MTX-
associated hypercalcemia. Ragab postulated that MTX alters either protein
metabolism and/or bone cell activity to induce osteopenia and speculated that these
changes are reversible.
Similar results were noted by Stanisavljevic and Babcock (73) in their review
of 37 children treated for leukemia with MTX. A high incidence of bone fractures
and bone pain were noted in these children. Fracture healing was delayed and non-
unions were common. Those children taken off of MTX went on to achieve normal
fracture repair. Stanisavljevic and Babcock (73) surmised MTX inhibits osteogenesis;
however, this effect appears to be reversible. MTX binds to dihydrofolate reductase
months after a single MTX dose, yet it has been postulated that the free unbound
intra-cellular MTX inhibits DNA synthesis (13). If this is true, the cells affected by
MTX therapy would be capable of resuming normal function after MTX withdrawal.
Gnudi et al. (31) studied the bone mineral content of 59 osteosarcoma
patients treated with different doses of MTX using single photon absorptiometry.
They analyzed the radius at the mid-shaft (primarily cortical bone) and the distal
metaphysis (rich in cancellous bone). These researchers found a dose-dependent
reduction in bone mineral content at the cancellous-rich distal radius with higher
doses of MTX having a more detrimental effect. They concluded that high doses of
MTX or low doses over long periods of time may severely compromise bone mass
and strength. Long-term follow-up bone mineral content measurements after
withdrawal of MTX treatment were not reported for these patients.
Clinical research (4,52,73) has verified the detrimental effects of
chemotherapy, specifically MTX, on bone through documentation of fracture
incidence. More quantitative evidence of MTX's effect on bone has been presented
in a recent study using single photon absorptiometry showing a reduction in bone
mineral content with MTX treatment (31). The mechanism of action of MTX on
bone and the permanancy on the osteoporotic effects, however, are unknown. Some
researchers (4,52) speculate the high incidence of fractures in MTX patients is due
to the drug's affect on mineral homeostasis through either alterations in protein
metabolism or cellular activity. Other possible explanations for the high fracture
incidence associated with cancer and MTX treatment include (1) the neoplastic
disease itself weakening the bone (not related to MTX treatment); (2) inflammation
reaction due to MTX-associated cell necrosis affecting the bone matrix; (3)
malnutrition associated with abnormal absorption of nutrients in the intestines caused
by MTX; (4) decreased physical activity due to illness; or (5) a direct relationship to
the cellular alterations in osteoblasts and osteoclasts caused by MTX treatment.
Most clinical researchers speculate the osteoporotic effects are transitory
(59,73); however, they base this theory soley on fracture incidence and observations
of fracture repair. No longitudinal quantitative clinical studies have been conducted
to investigate the bone's response over time to MTX treatment. It is uncertain if
bone recovers after chronic chemotherapy. If bone density remains depressed after
withdrawal of chemotherapy, this would decrease the peak bone mass the patient
would accrue in their lifetime, making them at risk for osteoporotic fracture
throughout their life. In order to further investigate the pathophysiology and the
time course of MTX-associated osteoporosis, animal models have been used to
facilitate quantification of bone characteristics through histomorphometry and
destructive biomechanical tests.
An abstract by Tross et al. (74) was the first to report the effects of the
chemotherapy drugs, Methotrexate (MTX) and Adriamycin (ADR), on bone turnover
and strength using a rat model. One 5-day course of daily chemotherapy injections
were given intraperitoneally, the animals were given fluorochrome label on day 7 and
13, and sacrificed on day 14 for histomorphometric and biomechanical assessment
of bone morphometry and strength. No changes in the torsional biomechanical
strength, stiffness, and energy absorbed at failure were noted for the drug treated
rats. However, significant decreases were noted in cancellous bone volume, osteoid
surface, and osteoblast surface in the drug treated animals compared to controls.
MTX and ADR treatment was not reported to affect osteoclast surface or mean
Freidlaender et al. (24) expanded Tross's previous work into manuscript form.
As outlined in the previous paragraph, the researchers administered one 5 day course
of MTX and ADR to rats and, using quantitative histomorphometric techniques,
measured the bone's response 2 weeks after initiation of the protocol. These
researchers found significant reductions in cancellous bone volume and bone
formation rate during this short-term study. This confirms the osteotoxic effects of
MTX and ADR immediately following acute administration, but does not determine
the ability of bone to recover from the drug insult.
Freidlaender's group (23) continued their research into the skeletal effects of
MTX by evaluating fracture healing. In this study Sprague Dawley rats received a
transverse fracture of the femur using a bone saw. The fracture was fixed internally
with an intramedullary K-wire. Rats were divided into 3 groups: (1) control rats
which received no treatment following fracture fixation; (2) MTX treated rats which
received one 5-day course of MTX injections; and (3) Radiation treated rats which
were irradiated with 250 rad fractions for 10 days following surgery. Groups of rats
in each of the 3 treatment groups were sacrificed at 1,2, 4, 8, and 12 weeks following
surgical fixation. Following sacrifice, bone harvest, and embedding procedures,
longitudinal sections were made through the callus. Callus formation was graded
based on the amount of repair present. Radiation and MTX treatments retarded
callus formation at all time intervals when compared to the control animals. MTX
treated animals also failed to regain femoral torsional strength following fracture
even at extended time periods (24 weeks). Similar results were also reported by
Hajj et al. (35) using a similar model where both bending strength and histological
grade of callus formation were significantly compromised in animals receiving weekly
injections of MTX compared to control animals. Burchardt et al. (8) also reported
suppressed bone/callus formation and reduced junction strength in chemotherapy
treated dogs in repair of segmental cortical non-unions.
Existing animal studies have elucidated the pathophysiology of MTX-
associated osteoporosis. MTX directly affects the cellular activity of osteoblasts as
observed by a significant decrease in bone volume, osteoblast surface, osteoid
surface, and bone formation rate (24,74). However, no study, animal or human, has
investigated the recovery of bone following completion of chemotherapy treatment.
All existing studies measure bone parameters at only one point in time after
treatment. It is unknown if the bone mass remains depressed indefinitely, slowly
recovers to normal levels, or recovers to osteopenic levels below the normal range.
Methods of Skeletal Assessment
Quantification of skeletal changes due to aging, menopause, exercise, or drug
treatment depends on the type of study (human or animal), resources available, and
desired accuracy of measurement. In vivo studies involving humans require
noninvasive means to assess bone changes unless skeletal biopsies are indicated.
Non-invasive methods commonly used to measure bone density include: (1) single-
photon absorptiometry (SPA), (2) dual-photon absorptiometry (DPA), (3) dual-
energy x-ray absorptiometry (DEXA), and (4) quantitative computed tomography
(QCT). DEXA, with it's high accuracy, quick scan times, and low radiation dose is
the most popular method to assess bone density in humans (15,28,40,55,68). DEXA
has also proven to be accurate for either in vivo or ex vivo animal studies measuring
either whole skeleton or appendicular density (2,3,33,48,67). These radiographic
methods measure the amount of mineral in the skeleton but do not offer the
capability of evaluating bone architecture or structure, assessing actively mineralizing
bone surfaces, or measuring bone quality or strength. Invasive measurements supply
more information but require painful biopsies for human studies or animal sacrifice.
Methods commonly used to assess bone quantity and quality following animal
sacrifice are histomorphometry and mechanical testing. Histomorphometric
techniques enable quantification of cancellous bone volume, cortical bone area,
cortical thickness and bone surface lined with osteoblasts or osteoclasts (42). When
fluorescent compounds are used to label the actively mineralizing bone,
histomorphometry can also quantify mineralizing surface, mineral apposition rate,
bone formation rate, and longitudinal bone growth (25).
Whole bones, cortical sections, or cancellous bone blocks can be mechanically
tested to determine the failure load, ultimate strength and stiffness of the specimen.
Weight bearing activities apply a complex loading environment to the skeleton
consisting of bending, torsion, and compressive loads. Mechanical tests of long bones
are typically loaded either in torsion or bending. Compressive tests are also used for
testing cancellous blocks or vertebral segments.
Bone mass, assessed either through DEXA or histomorphometry, provides
important information concerning bone health. However, the quality or strength of
the bone is also very important. The amount of cancellous or cortical bone mass can
be depressed or the amount of bone mineral compromised and not affect the
structural strength of the bone due to compensatory changes in bone geometry.
Similarly, adverse changes in bone geometry can decrease the structural strength of
bone without changes in bone mass or mineral content.
MATERIALS AND METHODOLOGY
The following sections detail the experimental protocol concerning animal
care, bone histomorphometry, biomechanical testing, dual-energy x-ray
absorptiometry (DEXA), and statistical analysis.
The protocol for this experiment was approved by the University Animal Use
Committee to assure humane treatment of animals and prevent undo suffering.
Sixty-nine male Sprague-Dawley rats (120 days old) with an average body weight of
415 g were randomly assigned to 4 control groups and three drug treatment groups.
Six rats were selected for the baseline control group and nine rats were chosen for
three control groups to be sacrificed at 30, 80, and 170 days. Similarly, three drug
treatment groups were also randomly chosen with 12, 10, and 14 rats in the 30, 80,
and 170 day groups, respectively. More animals were used in the 170 day group in
anticipation of death from MTX toxicity. Following a 2 week acclimation period,
treatment was initiated and baseline control rats were euthanized. Methotrexate
(MTX) was administered intraperitoneally (i.p.) in a 0.5 ml bolus injection using a
dose of 0.75 mg/kg/day. This dose is comparable to approximately one-third the
daily dose used for humans (69,70). Two courses of MTX were administered to the
treated rats in a cycle involving 5 consecutive days of drug injections, followed by 9
drug-free days, followed by another 5 days of injections (5-ON/9-OFF/5-ON). Two
courses of chemotherapy were used in this protocol to provide consistency with
typical clinical multiple-course sequelae used with human patients. Control rats
received a 0.5 ml i.p. bolus injection of saline using the same injection cycle (5-
ON/9-OFF/5-ON) as the MTX-treated rats.
All rats were injected with fluorochrome compounds on 2 separate occasions,
15 mg/kg of Demeclocycline (Lederle Laboratories, Pearl River, NY) on the 14th
day prior to sacrifice and 15 mg/kg of Calcein (Sigma Chemical Co., St. Louis, MO)
on the 7th day prior to sacrifice in order to monitor actively mineralizing bone
(dynamic histomorphometric parameters).
The rats were housed individually, with an ambient temperature of 24o-26 C
and a 12hr/12hr light/dark cycle. The activity of the rats was not monitored but was
limited to the confines of relatively small cages. The rats were fed Purina Rat
Laboratory Chow (St. Louis, MO) with 1.0% calcium and 0.9% phosphorus and
water ad libitum.
The rats were euthanized by i.p. injections of sodium pentobarbital
(100mg/kg). Bilateral femurs and tibiae as well as the 2nd and 3rd lumbar vertebrae
were harvested. The left tibia and femur and the vertebral segments were stripped
of all soft tissue. The tibia was sawed into 3 segments to allow infiltration of fixative:
the proximal third, middle diaphysis including the tibiofibular junction, and the small
distal segment which was discarded. The cortical bone of the anterior aspect of the
proximal tibia was shaved to expose cancellous bone tissue to the fixative solution.
The proximal and distal metaphyses of the femur were removed by cross sectional
saw cuts and discarded leaving the femoral diaphysis for analysis. All bone segments
were immediately fixed in 10% formalin-alcohol for future histomorphometric
analysis (see Appendix A). The right limb (tibia and femur) was disarticulated at the
hip, wrapped in saline soaked gauze and frozen at -700 C for future biomechanical
analysis. DEXA evaluation was performed on all excised bones prior to histological
and biomechanical testing.
The right tibia and femur segments were transferred from 10% formalin-
alcohol to 70% ethyl alcohol (ETOH) two days after harvest. The proximal tibia was
chosen over the femur for cancellous histomorphometry evaluation due to the ability
to standardize the sampling area relative to the growth plate. The tibial diaphysis
proximal to the tibiofibular junction and the mid-diaphysis of the femur were used
for analysis of cortical bone.
The proximal tibia was prepared for undecalcified cancellous bone
histomorphometry through dehydration in graded solutions of ETOH (70%, 95% and
100% for at least two days at each concentration) and xylene (for one day). The
sample was then infiltrated with a series of solutions containing methyl methacrylate,
dibutyl phthalate, and benzoyl peroxide. The recipes and detailed procedures for
methyl methacrylate embedding are included in Appendix A. The methyl
methacrylate solution was polymerized and the anterior aspect of the embedded
proximal tibia specimen was ground flat using a dental grinding wheel (Buffalo
Dental Manufacturing Co. Inc., Syosset, NY) to approximately one-third the depth
of the metaphysis. Thin longitudinal sections, 4 and 8 micrometers thick, were then
cut using an AO Autocut/Jung 1150 microtome (Cambridge Instruments, West
Germany). The thinner sections were placed on slides and stained according to the
Von Kossa method with a tetrachrome counterstain (Polysciences, Inc., Warrington,
PA). The recipe and procedure for the modified Von Kossa stain is included in
Appendix B. The 8 lim sections were left unstained and illuminated under ultraviolet
light to analyze the fluorescent labels for cancellous dynamic histomorphometry.
Two bone sections of the proximal tibia from each animal were analyzed for
both static and dynamic bone parameters. The standardized sample site (3mm
square window) was taken approximately 1 mm distal to the growth
plate/metaphyseal junction. The following static histomorphometric measurements
(25) were made on the 4 Am sections:
(1) Cancellous Bone Volume Cn.BV/TV (%) Cn.BV/TV is the percentage of
cancellous bone tissue (bone, marrow, and unmineralized osteoid) composed
of mineralized bone matrix.
(2) Osteoclast Surface Cn.Oc.S./BS (%) Cn.Oc.S./BS is the percentage of
cancellous bone surface with osteoclasts (bone resorbing cells) present and is
considered an index of bone resorption.
The following dynamic bone measurements (25) were made on the unstained, 8 pm-
thick sections under ultraviolet illumination to enable observation of the
(3) Longitudinal Bone Growth Cn.LBG (tm/day) Cn.LGB is the mean
distance between final fluorochrome label (Calcein) and the growth
plate/metaphyseal junction divided by the time between last label injection
(4) Mineralizing Surface Cn.MS (%) Cn.MS is the percentage of cancellous
bone surface with double fluorochrome labels and is an index of bone
(5) Mineral Apposition Rate Cn.MAR (pm/day) Cn.MAR is the mean
distance between the two fluorochrome markers divided by the time interval
between administration of the labels and is an index of osteoblast activity.
(6) Bone Formation Rate Cn.BFR/BS (m3/nm2/day) Cn.BFR/BS is
calculated by multiplying mineralizing surface by mineral apposition rate and
is an index of bone turnover.
Cortical bone segments of the tibia and femur were dehydrated in serial
solutions of 70%, 95%, and 100% ETOH and acetone. The bone segments were
then embedded in bioplastic (TAP Plastics, Inc., Dublin, CA) as described in
Appendix D. The bone blocks were then sectioned to 100 J/m using a Bueler Isomet
low-speed bone saw (Lake Bluff, IL) with a diamond chip blade (#801-137 LECO
Corp., St. Joseph, MI). The sections were mounted on slides for analysis. The
following static bone measurements were made:
(1) Total Bone Tissue Area Ct.T.Ar (mm2) Ct.T.Ar is the area within
periosteal perimeter which includes cortical bone and marrow.
(2) Marrow Area Ct.Ma.Ar (mm2) Ct.Ma.Ar is the area within the
endocortical perimeter which contains marrow.
(3) Cortical Bone Area Ct.Ar (mm2) Ct.Ar is the area of only cortical bone
(marrow area subtracted from the total bone tissue area).
(4) Mean Cortical Width Ct.Wi(mm) Ct.Wi is the average width or thickness
of the cortical bone sampled at 12 positions around the circumference.
(5) Polar Moment of Inertia J (mm4) J is the geometric property which is
calculated by modeling the cross-section of the bone as a hollow ellipse and
is represented by the following equation:
J = [r.(ab3 +a3b-(a-t)(b-t)3-(a-t)3(b-t))]/4
where a is the minor axis of the ellipse, b is the major axis of the ellipse, and
t is the mean cortical thickness (Ct.Wi) of the cross-section.
Under ultraviolet illumination the following dynamic bone parameters were
(6) Periosteal Mineralizing Surface Ps.Ms (%) Ps.MS is the percentage of the
cortical periosteal surface with double fluorochrome labels and is an index of
(7) Periosteal Mineral Apposition Rate Ps.MAR (im/day) Ps.MAR is the
average distance between the two fluorochrome markers divided by the time
interval between administration of the labels and is an index of osteoblast
activity on the periosteal surface.
(8) Periosteal Bone Formation Rate Ps.BFR (Jm3l/Im2/day) Ps.BFR/BS is
calculated by multiplying periosteal mineralizing surface by the periosteal
mineral apposition rate.
Quantification of Bone Parameters
Two methods were used to quantify the cancellous and cortical bone
parameters listed in the previous section. The static bone parameters were measured
using a Vidas imaging system (Kontron Electronics, West Germany) and customized
software programs. The dynamic bone parameters were measured using a bone
histomorphometry package by Bioquant (R & M Biometrics Corp., Nashville, TN).
Previous work has shown the Vidas imaging system to be more consistent and
reliable than the Bioquant hand digitizing system for static histomorphometric
measurements (76). The Vidas system was not used for dynamic histomorphometric
measurements due to the absence of proper fluorescent filters to enable viewing of
the demeclomycin and calcein labels.
The Vidas imaging system consisted of a Zeiss Axiophot (Zeiss, West
Germany) microscope, Hamamatsu C2400 high resolution black and white CCTV
video camera (Hamamatsu City, Japan), and 386 IBM-based computer with expanded
memory capabilities. The cancellous and cortical bone sections were imaged with the
video camera and the pixel information of these images were stored on the 386
microcomputer. The density contrast between the dark stained bone spicules and the
lighter blue stained marrow allowed semi-automatic differentiation between the bone
and marrow for calculation of cancellous bone volume. The cortical cross-sections,
however, were not stained and the periosteal and endocortical perimeters were
digitized to calculate bone areas. Cortical thickness and polar moment of inertia
were also calculated by a special program written to identify the geometry of the
cross section and sample, at 300 increments, the cortical width to calculate mean
cortical width and major and minor diameters for the elliptical model. The computer
codes written to make all cancellous and cortical calculations are presented in
Dynamic parameters were measured with the Bioquant Bone Morphometry
System. Surfaces of interest on the bone were viewed with a Nikon Labophot
microscope (Nikon, Inc., Garden City, NY) and digitized with a Hipad digitizing
tablet (R & M Biometrics Corp., Nashville, TN). A camera lucida enabled the
digitizing cursor to be visible in the microscopic field of view. An interface between
the digitizing tablet and an Apple HIe microcomputer (Cupertino, CA) allowed the
transfer of bone surface lengths to the Bioquant software where calculations were
performed to obtain the desired histomorphometric parameters.
The right tibia and femur, complete with surrounding soft tissue, were stored
at -70 C following harvest. Bones were thawed slowly to 25oC, cleaned of soft
tissue, and saturated in saline solution and kept chilled until testing. DEXA scans
were made of these bones immediately after removing the soft tissue and were
biomechanically tested the next morning.
The distal and proximal metaphyses were embedded in low temperature metal
potting material in a specially designed mold. When the metal was poured into the
mold, it produced a metal block surrounding the bone ends which fit into the torsion
tester's grip. The distance between the proximal and distal ends were standardized
to provide tibia and femur gauge lengths of 23.6 mm and 20.75 mm, respectively.
See Figure 3.1 for a photograph of a femur and tibia with bone ends embedded in
low melting point metal.
The right tibia and femur from each animal were tested to failure in external
torsion about the long axis of the bone on a rapid loading torsional testing machine
(A.H. Burstein, Cleveland, OH) at a loading rate of 12 rad/sec (9). Torsional tests
were selected for mechanical assessment because these tests apply uniform torque
Figure 3.1 Photograph of femur and tibia with ends
embedded in low melting point metal
along the bone length, allowing failure to occur at the weakest part of the test
Raw data, including torque applied (N m) and angular displacement (degrees)
at failure, were collected at 1000 Hz per channel and subsequently stored on a 386-
IBM microcomputer through an analog to digital interface with the torsional testing
machine. The cross-sectional geometry of the bone was analyzed in order to
calculate the polar moment of inertia and the torsional strength. The cross-sections,
therefore, were taken immediately adjacent to the torsional fracture. Cross-sections,
approximately 1000 pm in thickness, were prepared from undecalcified bone using
a Unimat-SL circular saw (model #DB200, American Edelstaal, Inc., NY, NY). Two
cross-sections were cut from each bone and glued with cyanoacrylate to glass slides.
These sections were then ground flat to approximately 300 Am using a Dremel Moto-
tool (model 285, Emerson Electric Co., Racine, WI). These cross sections were then
analyzed with a Vidas imaging system, as mentioned in the previous
histomorphometry section, to obtain the following parameters:
(1) Total Bone Tissue Area (mm2) = A
(2) Marrow Area (mm2) = A.
(3) Cortical Bone Area (mm2) = AC
(4) Mean Cortical Thickness (mm) = t
(5) Polar Moment of Inertia (mm4) = J
The following biomechanical parameters were either measured or calculated
using custom designed algorithms created for digital processing software (DADisp,
DSP Development Corporation, Cambridge, MA):
(1) Torque at failure (N m) = Tu
(2) Twist angle at failure (degrees) = en
(3) Energy absorbed at failure (N degrees) = Eu
(4) Stiffness (N/degrees) = S
(5) Torsional strength at failure (N/m2) = 7,
To and eu were measured directly during testing. Tu was plotted against eu to
calculate Eu and S which represent the area under the torque-angle curve and the
slope of the elastic region of the torque-angle curve, respectively. The biomechanical
parameters are represented in graphical form in Figure 3.2. Due to the lamellar
nature of cortical bone in the rat, there is minimal plastic deformation, therefore, a
linear regression from the minimum torque and angle to the maximum torque and
angle was used to determine stiffness. r, was then calculated using the hollow
ellipse model (20,62), where
r7 = Tu-b/J
and the polar moment of inertia,
J = [Tr.(ab3 +a3b-(a-t)(b-t)-(a-t)(b-t))]/4
Twist Angle (e)
Figure 3.2 Graphical depiction of biomechanical parameters
where T. is the torque at failure, a is the minor axis of the ellipse, b is the major axis
of the ellipse, and t is the mean thickness of the cross-section calculated from 12
positions around the circumference of the ellipse.
Dual Energy X-Ray Absorptiometrv
Ex-vivo measurements of bone mineral density were made for the excised
femurs, tibiae, and vertebrae using a dual energy x-ray absorptiometer (DEXA)
(Lunar Radiation, Madison, WI). The DEXA, with special high resolution software
supplied by the company, is capable of accurately scanning small bone samples (3,48).
The DEXA machine was calibrated prior to each use with a standard calibration
block consisting of known density materials. Repeatability tests were run using
excised rat tibiae and femurs to determine the coefficient of variation for this
procedure. The results from these repeatability tests are presented in Appendix E.
The cortical bone segments of the left tibial and femoral diaphyses, which
were stored in 70% ETOH for histomorphometric analyses at the time of scan, were
positioned on a 2 cm thick piece of acrylic with the anterior aspects of the bone
facing the scan sensor. The acrylic material was required in the scan area to provide
a contrast density, similar to tissue density, to the excised bones. The bones were
scanned side-by-side in the scanning area. The vertebral segments were scanned in
a manner similar to the left tibia and femur.
The intact right tibia and femur (which were kept moist and chilled in saline
baths and ice in order to prevent material degradation prior to mechanical testing)
were scanned in a saline filled petri dish on top of the acrylic contrast medium. The
bones were positioned and scanned in the same manner as the left tibia and femur.
The scans were analyzed by software provided by Lunar. The density scans
produced a pixel map of density gradients. The edges of the bone were detected by
an automatic algorithm. The edges were often misplaced by this algorithm and were
corrected by manual intervention. The bone mineral density (g/cm2) was calculated
for each excised bone.
A two-way analysis of variance (ANOVA) was used to determine if drug
treatment (MTX or CTL) or time following treatment (30, 80, and 170 days) or the
interaction between drug treatment and time had significant effects on the response
variables which include the histomorphometric, DEXA, and biomechanical
measurements detailed in the previous sections. If significant interactions were
noted, a one-way ANOVA was run to determine the effects of drug treatment at
each time interval and time for each treatment group, separately. If differences
among groups were noted with the ANOVA, a Duncan's multiple comparison
procedure was used to determine which groups were different from one another. A
significance level was set for all tests at the p = 0.05 level. PC-SAS version 6.03
(SAS, Inc., Cary, NC) was used for all the statistical analyses. A copy of the code
used for these analyses is presented in Appendix F.
The weights of the animals were monitored every week throughout the study.
A plot of the weight changes with time are presented in Figure 4.1. Although the
methotrexate treated rats had lower body weights throughout the study, this
difference was not significant (p> 0.05). There was a significant increase in weight
for both the MTX and CTL rats with time (p<0.05). Animals experienced no
noticeable side effects from the MTX treatment other than the slight decrease in
The following sections present the results of the histomorphometric analysis
cancellouss and cortical bone), biomechanical testing, and dual-energy x-ray
absorptiometry on the bones of control and methotrexate treated rats. Abbreviations
of bone parameters are used when presenting the results and in the graphical
presentation. In order to make these abbreviations more readable, a glossary of
abbreviations is provided in Appendix G for quick reference.
Figure 4.1 Rat weight changes with time
0 50 100 150
The means and standard deviations for the cancellous bone parameters are
presented in Table 4.1. A graph of tibial cancellous bone volume (Cn.BV/TV) with
time in shown in Figure 4.2. The statistical results of the 2-way and 1-way ANOVAs
are summarized on the right-hand side of the graph. The results for all parameters
are presented in a similar manner in subsequent graphs. The 2-way ANOVA
indicated a significant interaction between treatment and time (p=0.003), therefore,
separate 1-way ANOVAs were run for treatment by time and time by treatment. As
indicated in the graph, CTL rats had significantly higher CN.BV/TV at 80 and 170
days. Peak Cn.BV/TV for MTX rats occurred at 30 days with significant decreases
after this point, whereas, the CTL rats' peak Cn.BV/TV occurred at 80 days.
Figure 4.3 presents a graph of cancellous osteoclast surface (Cn.Oc.S), an
indicator of bone resorption. The MTX treated rats had significantly higher Cn.Oc.S
at 30, 80, and 170 days. Cn.Oc.S declined with age in both MTX and CTL rats.
Cancellous longitudinal bone growth (Cn.LBG) is presented in Figure 4.4. As
indicated in the statistical summary, there is not a significant interaction between
treatment and time. Therefore, the general effects of treatment (CTL or MTX) and
time (30, 80, and 170) are evaluated. In this case both treatment and time do have
a significant effect on Cn.LBG. The CTL rats had generally higher rates of
longitudinal growth than MTX and there was a general decline in longitudinal bone
growth with age in both groups.
The dynamic parameter of cancellous mineralizing surface (Cn.MS), indicative
of active bone forming surface, is shown in Figure 4.5. CTL rats had significantly
higher Cn.MS at all measurement intervals. The Cn.MS was highest at 30 days and
decreased to lower levels at 80 and 170 days for both CTL and MTX rats.
Osteoblastic activity was quantified by cancellous mineral apposition rate
(Cn.MAR) in Figure 4.6. The osteoblastic activity between CTL and MTX rats was
significantly different only at 170 days, with CTL rat's Cn.MAR exceeding MTX rats.
Again, there was an age-associated decline in Cn.MAR with highest values at 30 days
and decreasing to the lowest levels at 170 days for both CTL and MTX rats.
Cancellous bone formation rate (Cn.BFR), an index of bone turnover, is
presented in Figure 4.7. CTL rats had higher bone turnover at all time intervals (30,
80, and 170 days). Bone turnover was also age-dependent with highest levels
occurring at 30 days followed by decrease and plateau at 80 days for both CTL and
Comparative photomicrographs of cancellous bone volume are presented in
Figures 4.8, 4.9, 4.10, and 4.11 for the 0, 30, 80, and 170 day time intervals,
respectively. Each figure presents the control animal's bone in part "a"and the MTX
treated animal in part "b" for ease of comparison. Photomicrographs capturing the
fluorescent labels on the cancellous surface are shown in Figures 4.12,4.13,and 4.14
for the 30, 80, and 170 time intervals, respectively. Again, these figures are
composite photographs showing both the control and drug treated animal.
C ..-.4 00 *
'r,* 't C .4
c o oo o00 0
"a+1 -H +1 +I
+1 4- -H
Nt (ONI ~
'U- 0% r- ~ 0 0
^ I 9r l mC In
c) C 0%0
+1 -H +1 +1 +
^ tn 0o0~oo
06 00 0 6O6 N
,< o o '
+ +1 +1 +1 +1 -
"-i ~ r e 'U-* t t- N 'U
c~~~ c f r -^ 'U ~o
C 00 -H +1 +1 +1
00 C0 *
C4 00 t en
rpm -t -rn
c~9-H +i -H -H H +
i E C. 0
\0 t 00 .C
Sl S 0 5U H
6 0 0 6
M en ON 00 c. tn (U
u^ ^ 0; 0 d 0 n ; 1
e 0% C ; -4 o f- *
S+1 +1 +1 +i +i +1 +
C Cl< O 0 +1 v+ -
m Cc 0r~ ~lr
S Ch 0\
U I II +I +I I
0o CD 'D
SS u u 2 S s
6 -- 0
5s "- \ 11 '"
I- 0 m 00 v 00
HO UUQU 2
Orc o' ~ ^ S
^ou u u u S
2 o O x
0 C m 000
0 0 =
m ur L
Q-".aCTO U g
.0 6 ,, 0
0 en 00 c00 0 u0 -
0 o 0l
u II 00
c d d
trt x time: p=0.003
CTL>MTX @80,170 (*)
Figure 4.2 Tibial cancellous bone volume
trt x time: p=0.05
0 50 100 150
Figure 4.3 Tibial cancellous osteoclast surface
- 25 trt x time: p=0.643
* trt: p=0.0001
E 20- CTL>MTX (*)
0 so loo 15o CTL
Time (days) MTX
Figure 4.4 Tibial cancellous longitudinal bone growth
trt x time: p=0.0003
Figure 4.5 Tibial cancellous mineralizing surface
50 100 150
trt x time: p=0.045
Figure 4.6 Tibial cancellous mineral apposition rate
trt x time: p=0.0001
Figure 4.7 Tibial cancellous bone formation rate
~ !r. r U
2* U -L i
'' a~i. ii
~~i~r! -.' r ; : it 7 "
..? ... u~12
a. P .. -
^ ... a ^ t
A1~ 'f'^'J^l^ v SL
;k *"* jr < *0 ^ 1
I, .c r 4
i.Lw .'d-" -; "--
'i.'g ~. 0b.
.1** *4 ~L~.
The means and standard deviations for the cortical bone parameters are
presented in Table 4.2 and 4.3 for the femur and tibia, respectively. Figures 4.15 and
4.16 present the cortical total tissue area (Ct.T.Ar) as a function of time for the
femur and tibia, respectively. For both bones, CTL had significantly higher Ct.T.Ar
than MTX at 170 days but equivalent bone areas at prior measuring times. The
Ct.T.Ar steadily increased with time in the femur for both CTL and MTX animals.
The Ct.T.Ar of the CTL rats' tibia also followed a similar steady increase, however,
the MTX rats increased only until the 80 day time point and then experienced a
drastic decrease in Ct.T.Ar to baseline levels.
The marrow area (Ct.Ma.Ar) of the femur and tibia are presented in Figures
4.17 and 4.18, respectively. For both bones the MTX animals had larger marrow
areas than the CTL animals when taken over all time intervals. There was a general
trend for increased marrow areas with time in the femur. However, in the tibia, the
marrow area indicated a significant increase over baseline levels only at 170 days.
The cortical bone area (Ct.Ar) is the portion of the total bone tissue area
comprised of cortical bone. Figure 4.19 and 4.20 show the Ct.Ar for the femur and
tibia, respectively. The CTL animals had significantly greater femoral and tibial
cross-sectional cortical areas than the MTX rats at 30, 80, and 170 days. The femurs
of the CTL rats showed increases in Ct.Ar with time, whereas the MTX rats showed
no significant changes in Ct.Ar from baseline levels. The Ct.Ar of the tibia increased
with time for the CTL rats but only increased up to 80 days for the MTX rats at
which point a significant decrease in bone area was noted to below baseline levels.
The mean cortical width (Ct.Wi) of the femoral and tibial cross-section are
shown in Figures 4.21 and 4.22,respectively. Femoral Ct.Wi was significantly greater
for the CTL rats compared to the MTX rats at all time intervals. Femoral Ct.Wi
increased with age for the CTL rats yet didn't change for the MTX rats. Tibial
Ct.Wi was also significantly greater for the CTL compared to MTX rats at 30, 80,
and 170 days. Peak Ct.Wi occurred in the tibia at 80 days for both MTX and CTL
rats. The tibial Ct.Wi of CTL rats then stabilized, whereas, the Ct.Wi of the MTX
rats decreased to below baseline levels.
The polar moment of inertia (J) is presented in Figures 4.23 and 4.24 for the
femur and tibia, respectively. The CTL rats had significantly higher J-values than the
MTX rats at 170 days for both the femur and tibia. The J-value for CTL rats
experienced an age-related increase in the femur and tibia. The MTX rats had peak
femoral and tibial J-values at 80 days.
The dynamic cortical bone parameter, periosteal mineralizing surface (Ps.MS),
is presented in Figures 4.25 and 4.26for the femur and tibia, respectively. There was
a significant difference in femoral Ps.MS between CTL and MTX rats with CTL
exhibiting higher Ps.MS values at 30, 80, and 170 days. There was an age-related
decline in femoral Ps.MS for both CTL and MTX rats with the MTX rats exhibiting
a greater rate of Ps.MS loss with time. The tibia demonstrated a significant
decrease in Ps.MS with time with the CTL and MTX rats exhibiting similar rates of
Periosteal mineral apposition rate (Ps.MAR), an index of osteoblast activity,
is presented in Figures 4.27 and 4.28 for the femur and tibia, respectively. Although
there was not a significant interaction between treatment and time for the femur, the
CTL had significantly greater Ps.MAR than MTX treated animals and a decrease in
Ps.MAR with time. The statistical results from the tibia indicated that CTL animals
had higher Ps.MAR than MTX rats at all time intervals and a significant decrease
in Ps.MAR with time for both CTL and MTX animals.
Graphs for femoral and tibial periosteal bone formation (Ps.BFR) rate are
presented in Figures 4.29 and 4.30, respectively. Femoral Ps.BFR was significantly
lower for the MTX rats at all time intervals than the CTL animals and exhibited a
steady decline with time for both CTL and MTX animals. Tibial Ps.BFR was only
significantly lower for the MTX at the 30 day time interval; after that point both
CTL and MTX exhibited similar periosteal bone formation. Tibial Ps.BFR was
higher at 30 days for both MTX and CTL animals than at 80 and 170 days with a
rapid age-related decrease between 30 and 80 days.
Representative photomicrographs of cortical bone area for the femur and tibia
are presented in Figures 4.31 and 4.32,respectively, for the baseline control animals.
Photographs of the femoral and tibial cortical bone areas for the 30 day time interval
are presented in Figures 4.33 and 4.34, 80 day time interval in Figures 4.35 and 4.36,
and the 170 day time interval in Figures 4.37 and 4.38. Each figure presents the
control animal's bone in part (a) and the MTX treated animal in part (b) for ease
of comparison. Photomicrographs capturing the fluorescent labels on the periosteal
cortical surface of the femur and tibia are shown in Figures 4.39 and 4.40 for the 30
day time interval, Figures 4.41 and 4.42 for the 80 day time interval, and Figures 4.43
and 4.44 for the 170 day time interval. Again, these figures are composite
photographs showing both the control and drug treated animal.
CN e4' C 000 e Lej t-
Str- 00 N* -o
E qc ; 1o6 v o o z t
r- rI 03 0. 6 ;
+ -( +- +1 +1 +1 +1 M +1
qe 4 0q *c 6 o
03 ocn Clkr- o ot
tn m 0 mr- on 00 W
E o oo*go ro -o on
(O + 0+l l +1 + +1 6 +I
nc C4 tn gt r o00
-: m6 ci r-M r-* "0* C
P +l + +1 cn +1 m +1+,t +l en +l (n +l
q' o5 ci c, q r
01 +1 +1 5+ 0 +1 o0 + 0 H 0 s
h 00 C'! c) C 0t 0c
+1 -- +-1+1 +- +- -c ++I'
J It 00 0 0 't
T H- +1 +1 H +1
w 0 O0 E
S00 J 00
u uU U
u O3o oU
S< ^C 0 0
0 r o
4+ +1 +1 ++
C C M h 00 f 4 0N
c; r0i CD (=; *0 0 0 6C
-H +1 +1 +1 +1
0 n 0 0 4
O 0 4o t0 : m Mn o r
S N o6 m ooN 0 0
r +1 + +1 +1 + +1
n C14 1-
c-0- -- 00 -- %0 -
-H +1 t- +1od0 +1 o+1 +1
W1 o ul e o s n -, cn 0
--+ -- +1 -H + +1
r- (1 0 00 c) *'t t- 0 W) n 00 _a m
o0 0 0 0 0 CU
+ +1 -H +1aO -H t +
+1 4A _H H 1 -H H j H 1 0
U U _
---+ ------- --------- ^
uEd d d0 o~+~
.5 ^ H H ^ E- &i E
==== = = === = ===== ==== ======== ====== ^
n 0 00
I 0 U
0 0 U U CO
< ca oo en 00
o *o .- oo &o
a o o O O4 ) a.)
SO 3 iS
3 o oa. a
0s c 0 a -
trt x time: p=0.002
0 50 100 150
Figure 4.15 Femoral total bone tissue area
0 50 100 150
trt x time: p=0.0005
Figure 4.16 Tibial total bone tissue area
50 100 150
trt x time: p=0.283
Figure 4.17 Femoral marrow area
o trt x time: p=0.349
E MTX>CTL (*)
0 C p=0.021
0 50 100 150 CTL
Time (days) MTX
Figure 4.18 Tibial marrow area
trt x time: p=0.0001
Figure 4.19 Femoral cortical bone area
trt x time: p=0.0001
0 50 100 150
Figure 4.20 Tibial cortical bone area
trt x time: p=0.0001
Figure 4.21 Femoral mean cortical bone width
0 50 100 150
trt x time: p=0.0001
Figure 4.22 Tibial mean cortical bone width
0 50 100 150
trt x time: p= 0.0003
Figure 4.23 Femoral polar moment of inertia
trt x time: p=0.019
0 50 100 150
Figure 4.24 Tibial polar moment of inertia
trt x time: p=0.0035
*s @30,80,170 (*)
"V CTL: 30>80,170
60- MTX: 30>80>170
0 50 100 150 -- CTL
Time (days) + MTX
Figure 4.25 Femoral periosteal mineralizing surface
70- trt x time: p=0.704
60 -trt: p=0.337
0 50 100 150 -"- CTL
Time (days) + MTX
Figure 4.26 Tibial periosteal mineralizing surface
i 1 -
0 50 100 150
trt x time: p=0.07
Figure 4.27 Femoral periosteal mineral apposition rate
r------- __ *
trt x time: p=0.015
0 50 100 150
Figure 4.28 Tibial periosteal mineral apposition rate
trt x time: p=0.039
Figure 4.29 Femoral bone formation rate
0 50 100 150
trt x time: p=0.0024
Figure 4.30 Tibial periosteal bone formation rate
:'~~~-~~ 8~-~b~ses~ 0