Effects of sacrocaudal spinal cord lesions and transplants of fetal tissue on cutaneous reflexes of the tail

MISSING IMAGE

Material Information

Title:
Effects of sacrocaudal spinal cord lesions and transplants of fetal tissue on cutaneous reflexes of the tail
Physical Description:
xii, 180 leaves : ill. ; 29 cm.
Language:
English
Creator:
Friedman, Robert Mark, 1961-
Publication Date:

Subjects

Subjects / Keywords:
Research   ( mesh )
Fetal Tissue Transplantation   ( mesh )
Spinal Cord -- transplantation   ( mesh )
Spinal Cord Injuries -- surgery   ( mesh )
Spinal Cord Injuries -- veterinary   ( mesh )
Electric Stimulation   ( mesh )
Physical Stimulation   ( mesh )
Tail   ( mesh )
Reflex   ( mesh )
Cats   ( mesh )
Department of Neuroscience thesis Ph.D   ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Neuroscience -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1995.
Bibliography:
Bibliography: leaves 169-179.
Statement of Responsibility:
by Robert Mark Friedman.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002304088
oclc - 50179908
notis - ALQ7399
System ID:
AA00009023:00001


This item is only available as the following downloads:


Full Text









EFFECTS OF SACROCAUDAL SPINAL CORD LESIONS AND TRANSPLANTS
OF FETAL TISSUE ON CUTANEOUS REFLEXES OF THE TAIL














By

ROBERT MARK FRIEDMAN


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


1995






























to absent friends








ACKNOWLEDGEMENTS


A number of individuals made significant contributions to the successful

completion of this research. First and foremost, I would like to recognize the

patience of my mentor Dr. Charles Vierck, Jr. Even though there may be more

productive student-mentor approaches, I feel that his hands off style of minimal

cajoling has made me a scientist better prepared to achieve my future research

goals. For not always getting on my goat, I sincerely recognize the contributions

of Dr. Brian Cooper to my early graduate research endeavors. Dr. Louis Ritz

receives a special acknowledgement, for I appreciate the opportunity to study tail

behaviors under his tutelage. I especially want to thank Dr. Paul Reier who made

numerous and invaluable contributions to this research. I also appreciate the time

and support of my other committee members, past and present: Drs. Paul Brown,

Christiana Leonard, John Munson, and Alan Spector.

Thanks go to Carolyn Baum for her assistance in working with the cats

involved in this research. Special thanks go to Jean Kaufman, for her help in other

aspects of my research is greatly appreciated.

I would also like to recognize the excellent care of the felines provided by

Ms. Barbara O'Steen, Kim Foli and Dan Theele, DVM. I especially would like to

thank the technical support of Anwaral Azam. Without his help in troubleshooting

electrical and computer problems, this research would have never gotten off the

ground.








The support of past and present fellow graduate students should not go

unmentioned. Gregory Schrimscher, Jianxin Bao, Audrey Kalehua, Diana

Glendinning, David Yeomans, Doug Swanson and Laura Errante made the

struggles of graduate school bearable. I especially appreciate the support of James

Makous for engaging discussions, advice and reflections on the scientific method,

life as research scientist, and the somatosensory system.

Love and special thanks go to my family. I would like to acknowledge the

understanding and support of my parents, Carolyn and Lester Friedman, who

wanted their son to grow up to be a doctor but will have to settle for a son with a

doctorate. I especially want to recognize the love and support of my sister Renee

Friedman who took longer to receive her Ph.D. than I, but to my chagrin still

finished her's first. Alas, the unspoken love and contributions of Laura Friedman

to my life are immeasurable.

Financial support for this research was provided by the Center for

Neurobiological Sciences (NIMH grant MH15737), NIH grant NS27511 to Dr. Paul

J. Reier, the Mark F. Overstreet Fund for Spinal Cord Injury Research and the State

of Florida Impaired Drivers and Speeders Trust Fund.








TABLE OF CONTENTS


ACKNOWLEDGEMENTS ......................

LIST OF TABLES ..........................

LIST OF FIGURES ..........................

ABSTRACT .............................

CHAPTERS

1 INTRODUCTION AND BACKGROUND ...........

Consequences of Human Spinal Cord Injury ........
Treatments for Spinal Cord Injury ...... .
Behavioral Evaluation of Spinal Cord Injury .......
Animal Model Systems to Study Spinal Cord Injury .
Overview of Dissertation .................

2 CUTANEOUS REFLEXES OF THE TAIL OF CATS BEFORE
AFTER SACROCAUDAL SPINAL LESIONS ........
Introduction . .. . .
Methods . .. . .
Results . . . .
Discussion . . . .

3 EFFECTS OF FETAL SPINAL TISSUE TRANSPLANTS ON
CUTANEOUS REFLEXES OF THE CAT TAIL ........
Introduction . .. .. . .
Methods . . . .
Results . . . .
Discussion . .. . .

4 OVERALL DISCUSSION ...................

REFERENCES ............................

BIOGRAPHICAL SKETCH ......................


AND


. 4
. 6
. 7
. 11


. 13
. 13
. 15
. 25
106


115
115
117
121
154

164

.169

180








LIST OF TABLES


Table 2-1 Summary of the multiple linear regression analysis of
reflex force of normal animals and after a
chronic sacrocaudal transaction. . .


Table 2-2


Table 2-3


. 39


Summary of the multiple linear regression analysis
of peak reflex amplitude of normal animals
and after a chronic sacrocaudal transaction. .
Summary of the multiple linear regression analysis
of latency at peak reflex amplitude of normal animals
and after a chronic sacrocaudal transaction. .


Table 2-4


Table 2-5


Table 2-6


Table 2-7


Table 2-8

Table 2-9


Table 2-10


Table 2-11


Table 2-12


Summary of the multiple linear regression analysis
of rise time of normal animals and after
a chronic sacrocaudal transaction. . ... 50
Summary of the multiple linear regression
analysis of half maximal reflex duration of normal
animals and after a chronic sacrocaudal transaction 54
Summary of the multiple linear regression analysis
of quarter maximal reflex duration of normal
animals and after a chronic sacrocaudal transaction 58
Summary of the multiple linear regression analysis
of reflex latency of normal animals and after a
chronic sacrocaudal transaction . . 62
Spearman rank order correlation coefficients between
measured characteristics of the reflex waveform ... 64
Summary of the multiple linear regression analysis
of changes in reflex force over time after sacrocaudal
transaction ........................ .... 68
Summary of the multiple linear regression analysis
of changes in peak reflex amplitude over time
after sacrocaudal transaction . . 71
Summary of the multiple linear regression analysis
of changes in latency at peak reflex amplitude over time
after sacrocaudal transaction . . 73
Summary of the multiple linear regression analysis
of changes over time in the duration of decay to
0.5 maximal amplitude after sacrocaudal transaction 76








Table 2-13


Table 2-14

Table 2-15

Table 2-16


Table 2-17


Table 3-1

Table 3-2


Table 3-3


Table 3-4


Summary of the multiple linear regression analysis
of changes over time in the duration to decay to
0.25 maximal amplitude after sacrocaudal transaction .
Summary of the multiple linear regression analysis
of changes in reflex latency after sacrocaudal transaction .
Summary of the multiple linear regression analysis
of changes in reflex rise time after sacrocaudal transaction.
Comparison of 5 reflex characteristics for one animal
after a left hemisection and a group of 7 animals
pre- and post-transection . . .
Summary of the consequences of a transaction of
the sacrocaudal spinal cord on the response
characteristics of the electrocutaneous reflex .
Anatomical evaluation of transplant survivability
and integration. .......................
Summary of the multiple linear regression analysis
of reflex force of animals following transaction or
transaction plus transplantation . .
Summary of the multiple linear regression analysis
of peak reflex amplitude of animals following transaction
or transaction plus transplantation. . .
Summary of the multiple linear regression analysis
of latency at peak reflex amplitude of animals following
transaction or transaction plus transplantation .


Summary of the multiple linear regression
analysis of reflex rise time of animals following
transaction or transaction plus transplantation .


. 146


Summary of the multiple linear regression analysis
of half maximal reflex duration of animals following
transaction or transaction plus transplantation .
Summary of the multiple linear regression analysis
of quarter maximal duration of animals following
transaction or transaction plus transplantation .
Summary of the multiple linear regression analysis of
reflex latency of animals following transaction
or transaction plus transplantation. . .


149


153


157


. 77

. 81

. 83


. 105


. 108

. 122


. 134


. 138


. 142


Table 3-5


Table 3-6


Table 3-7


Table 3-8








Summary of the consequences of transaction plus
transplantation in comparison to the transection-only
lesion on the response characteristics of the
electrocutaneous reflex . . 159


Table 3-9








LIST OF FIGURES


Figure 2-1
Figure 2-2
Figure 2-3

Figure 2-4
Figure 2-5

Figure 2-6
Figure 2-7

Figure 2-8

Figure 2-9
Figure 2-10

Figure 2-11

Figure 2-12
Figure 2-13
Figure 2-14

Figure 2-15
Figure 2-16
Figure 2-17

Figure 2-18
Figure 2-19
Figure 2-20

Figure 2-21


The testing apparati. . . .. 18
Histological reconstructions of the sacrocaudal hemisections 27
Representative traces of tail reflexes to mechanical
stimulation ............. ... .. .. .. 30
Frequency of responses to mechanical stimuli . 32
Representative traces of tail reflexes to single
DC pulses of electrocutaneous stimulation. ... 35
The relationship of reflex force to stimulus intensity 38
The relationship of peak reflex amplitude to stimulus
intensity . . . 42
The relationship of latency at peak reflex amplitude to
stimulus intensity ........................ 45
The relationship of reflex rise time to stimulus intensity 49
The relationship of half maximal reflex duration to
stimulus intensity ........................ 53
The relationship of quarter maximal reflex duration
to stimulus intensity ....................... 57
The relationship of reflex latency to stimulus intensity 61
The time course of changes in reflex force . 67
The time course of changes in peak reflex amplitude
and latency at peak reflex amplitude . 70
The time course of changes in reflex duration . 75
The time course of changes in reflex latency and rise time. 80
Frequency of wind-down of the reflex with repetitive
stimulation . ... . 85
Frequency of wind-up with repetitive stimulation ... 87
Representative responses showing wind-up of the tail reflex. 90
The magnitude of changes in the reflex response
with repetitive stimulation. . . ... 92
Reflex responses of the tail to trains of
electrocutaneous stimulation . . 94







Figure 2-22

Figure 2-23
Figure 2-24

Figure 2-25

Figure 3-1
Figure 3-2
Figure 3-3

Figure 3-4
Figure 3-5

Figure 3-6


Figure 3-7


Figure 3-8


Figure 3-9


Figure 3-10


Figure 3-11


Reflex force in animals with variants of the sacrocaudal
lesion . . . 96
Reflex characteristics following dorsal hemisection at S3 99
Reflex characteristics following sacrocaudal transaction
with an ischemic episode. . .. 101
Reflex characteristics following caudal spinal transaction
at Ca3. . . . 103
Survival of transplant tissue . . 124
Coronal section through transplant tissue . 127
Anatomical integration of transplant tissue with
host spinal cord ........................ 129
Longitudinal section through a transplant . 131
Relationship of reflex force to stimulus intensity for animals
following transaction or transaction plus transplantation 133
Relationship of peak reflex amplitude to stimulus intensity
for animals following transaction or transaction plus
transplantation ........................ 137
Relationship of latency at peak reflex amplitude to stimulus
intensity for animals following transaction or transaction
plus transplantation . . 141
Relationship of reflex rise time to stimulus intensity for
animals following transaction or transaction plus
transplantation ........................ 145
Relationship of half maximal reflex duration to stimulus
intensity for animals following transaction or transaction
plus transplantation. . . ... 148
Relationship of quarter maximal reflex duration
to stimulus intensity for animals following transaction or
transaction plus transplantation. . .. 152
Relationship of reflex latency to stimulus intensity for
animals following transaction or transaction plus
transplantation ........................ 156













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

EFFECTS OF SACROCAUDAL SPINAL CORD LESIONS AND TRANSPLANTS
OF FETAL TISSUE ON CUTANEOUS REFLEXES OF THE TAIL

By

ROBERT MARK FRIEDMAN

May 1995


Chairperson: Charles J. Vierck, Jr.
Major Department: Neuroscience

Cutaneous reflexes were measured in the tail of the cat to determine whether

cutaneous hyperreflexia is present after sacrocaudal spinal lesions and whether

changes in cutaneous reflexes are similar to those observed in human patients after

spinal cord injuries. Reflexes to electrocutaneous and mechanical stimulation of

the tail were evaluated prior to and after spinal cord lesions.

After a sacrocaudal spinal transaction, the overall magnitude of

electrocutaneous tail reflexes was greater than that observed normally. Major

changes in the characteristics of the response included increases in peak reflex

amplitude, longer latencies to reach peak reflex amplitude, and longer reflex

durations. A long latency reflex could be exhibited, and the magnitude of the reflex

could show wind-up in magnitude with repeated stimulation. The changes in







electrocutaneous reflexes developed progressively over a period of 3 months. The

expression of tail reflexes to mechanical cutaneous stimulation was highly variable.

However, there was a decrease in mechanical reflex threshold after sacrocaudal

transaction. These changes in cutaneous reflexes did not occur after incomplete

lesions of the sacrocaudal spinal cord.

This animal model was investigated for its potential to evaluate treatments of

spinal cord injury. Consequences for tail reflexes were determined for animals that

received a transplant of fetal spinal tissue placed into an acute sacrocaudal

transaction cavity. Eleven animals received fetal tissue. In 8 animals the transplant

successfully survived and integrated with the host spinal cord. In behavioral

studies of animals with surviving transplants, peak amplitude of the

electrocutaneous reflex was less than observed for animals with a chronic

transaction. The duration of the electrocutaneous reflex for animals with

transplants was less than observed for transected animals without transplants and

was similar to normal animals. These findings indicate that the presence of

transplant tissue can modulate the cutaneous hyperreflexia that otherwise develops

after spinal cord transaction.












CHAPTER 1

INTRODUCTION AND BACKGROUND


Improved treatment strategies for spinal cord injury represent a major goal

for the health care industry. Approximately 10,000 new cases of spinal cord injury

are reported each year in the United States. The causes of spinal cord injuries are

numerous; however, they primarily involve automobile accidents, acts of violence,

falls and sports related injuries. Typically, these individuals are men between the

ages of 19 and 26. On a cost basis, including lost wages, billions of dollars are

spent each year on medical intervention, rehabilitation and chronic health care of

individuals with spinal cord injuries (U.S. Congress, 1990).


Consequences of Human Spinal Cord Iniury

Initially after a spinal cord injury, the patient goes through a period of spinal

shock when all motor functions of spinal segments below the lesion are depressed.

After spinal shock recedes the neurological effects of spinal cord injury can be

dichotomized in terms of either negative (i.e., loss of functions) or positive (i.e.,

abnormal behaviors) signs (Young and Shahani, 1986).

Negative signs that are present below the spinal level of injury can include

a complete loss of voluntary movement (e.g., paraplegia or quadriplegia), weakness







2

or paresis of a limb, and a loss of bowel and bladder control. In addition, cutaneous

sensibility for fine touch, proprioception, pain and temperature sensations can be

disrupted or lost. Problems associated with sensory loss are significant, since a

patient with a lowered pain response to a noxious stimulus (hypoalgesia) or a

complete absence of pain (analgesia), may unknowingly experience a higher

incidence and severity of accidental injury (e.g., from a burn, cut, or pressure sore).

Positive signs after a lesion to the spinal cord can involve the development

of postural abnormalities, exaggerated flexion and tendon reflexes, muscle spasms,

and painful sensations. Both hyperreflexia and pain can interfere with the spinal

cord injured patient's ability to function. The different manifestations of

hyperreflexia that can occur after a spinal cord lesion are known collectively as the

spastic syndrome (Young and Shahani, 1986).

Alterations in muscle tone can include a fixed postural abnormality of a limb

(a dystonia or contracture), rigidity, and changes in stretch reflexes. The

exaggeration of segmental stretch reflexes is considered a crucial or defining

feature of spasticity. The alteration of segmental stretch reflexes is thought to (1)

underlie changes in threshold and amplitude of the tendon jerk (myotactic reflex),

(2) trigger a radiation of a local myotactic reflex to the entire limb, (3) trigger rapid

oscillatory movements of a limb clonuss), and (4) include modifications of tonic

stretch reflexes that cause abnormalities in posture dystoniaa) of a limb (Dimitrijevic

and Nathan, 1967). Spasticity has been characterized as a velocity sensitive

increase in resistance to passive stretch of a limb with an accompanied increase







3

in tendon reflexes and hypertonia (Landau, 1974). In human patients with cortical

lesions, quantitative behavioral studies have characterized spasticity as a shift in

threshold to elicit stretch reflexes without an increase in magnitude or gain of the

response (Lee et al., 1987). However, similar studies have not been performed to

characterize spasticity in patients with spinal cord injuries.

After spinal cord injury in humans, cutaneous hyperreflexia is evident for

forms of stimulation other than muscular stretch. Cutaneous stimulation normally

elicits flexion reflexes only when the level of stimulation is nociceptive. However,

in a spastic patient, an innocuous cutaneous stimulus on the leg can produce a

brisk, stereotyped, involuntary, flexor movement that can include triple flexion at the

hip, knee and ankle, and crossed extension (Dimitrijevic and Nathan, 1967; Roby-

Brami and Bussel, 1987). The flexion reflex is normally a limited response in one

muscle or limb, but in spastic patients the response spreads to other muscles or

other limbs (Dimitrijevic and Nathan, 1968). In addition, involuntary flexor

movements of a limb can occur in the absence of a known cutaneous stimulus.

Because flexor spasms can increase in frequency in the presence of cutaneous

stimuli, it would appear that central flexor reflex mechanisms are involved (Shahani

and Young, 1973).

Another positive sign after spinal cord injury is the possible occurrence of a

central pain syndrome, despite diminution or loss of cutaneous sensations after

damage to the long ascending spinal pathways. Ninety percent of spinal cord

injured patients report some unpleasant sensory experiences; fifty percent of these








4

patients experience some pain sensations, and twenty-five percent report disabling

pain (Davis and Martin, 1947; Botterell et al., 1953; Kaplan et al., 1962;

Nepomuceno et al., 1979). Pain sensations can involve (1) a phantom limb pain

where the limb is sensed to be in a cramped, twisted position; (2) a visceral pain

where the patient perceives a fullness of the abdomen, fecal urgency and/or

bladder disturbances; (3) a dysesthesia, where the patient feels an unpleasant

sensation that can include diffuse burning or tingling; and (4) a neuralgia, where the

patient feels a hot, sharp, knifelike pain localized to a specific dermatome (Merskey

et al., 1986). With respect to sensations elicited by cutaneous stimulation, pain

perception varies in patients following spinal cord injuries (Lindblom, 1985). A

patient can have a lowered threshold for painful sensations (allodynia), where a

painful sensation is elicited by a stimulus that normally does not provoke pain. In

addition, a patient can report that a noxious stimulus elicits an increased sensation

of pain (hyperalgesia).


Treatments for Spinal Cord Injury

Multiple approaches are employed in treatment of the acute and chronic

effects of spinal cord injuries. In an attempt to control spasticity and pain following

a spinal cord injury, pharmacological and surgical interventions are used with

varying degrees of success (Dimitrijevic and Nathan, 1967; Davidoff, 1985).

Physical therapy and specialized mechanical devices (e.g., wheel chairs, crutches,

prosthetic devices, specialized hand splints) are utilized to improve independence








5

and provide some form of mobility (Trieschmann, 1988). However, there are

currently no available treatments to compensate for the loss of cutaneous sensibility

or development of chronic pain after spinal cord injury.

A new strategy that might ameliorate some effects of spinal cord injury is the

grafting of embryonic neural tissue. An example of the success of embryonic neural

tissue transplantation is found with Parkinson's disease, where grafting of cells from

embryonic brainstem has been reported to have measurable therapeutic benefits

for some patients (Lindvall et al., 1990; Freed et al., 1992). With this approach, the

goal can be to replace lost neurons, connections or neural pathways, to rescue

neurons at risk of cell death, or to modulate neural activity in the host nervous

system. Anatomical evidence suggests that some of these mechanisms may be

associated with the benefits derived from embryonic grafts in patients with

Parkinson's disease.

With respect to the spinal cord, anatomical studies have provided evidence

for the ability of embryonic spinal tissue to integrate and form axonal projections

with the host (Jakeman and Reier, 1991; Reier et al., 1994). Behavioral studies in

cats and rats have shown that fetal spinal transplants can promote recovery in

locomotor behavior (Yakoleff et al, 1989; Kunkel-Bagden and Bregman, 1990;

Anderson et al., 1991; Goldberger, 1991; Kunkel-Bagden et al., 1991; Stokes and

Reier, 1992). In other experiments in acutely spinalized animals, long-latency

components of flexion reflexes have been shown to be enhanced by previous

transplants of fetal brain stem tissue (Moorman et al., 1990). Similarly, sexual








6

reflexes can be enhanced by transplantation of brain stem tissue (Privat et al.,

1989).

These studies are encouraging in the sense that neural grafts can modulate

or restore certain locomotor behaviors and reflexes that are diminished or lost as

a consequence of spinal cord injury. Improved recovery of locomotion as a result

of fetal cell transplantation could represent, at least in part, an attenuation of

abnormal reflexes, but little or no information is available to evaluate this possibility.

Thus, it is important to determine the effects of fetal transplantation on the

development of spasticity after spinal cord injury.


Behavioral Evaluation of Spinal Cord Injury

Evaluation of new treatment strategies for spinal cord injuries will require

quantitative evaluation in animals of both volitional and reflex behaviors in order to

fully assess the functional utility of these methodologies (Goldberger et al., 1990).

In animals and humans, evaluation of behavioral consequences of spinal cord injury

has relied primarily on subjective rating scales (e.g., Ashworth, 1964; Gale et al.,

1985). This approach has been successful in characterizing consequences of

spinal cord injuries and treatment strategies (Kerasidis et al., 1987). These rating

methods can be effective in specifying qualitative changes, but quantitative

changes in a behavioral function over time remain uncharacterized, irrespective of

issues of test sensitivity and interexaminer reliability.








7

Several different quantitative behavioral assays have been applied to

describe consequences of spinal cord injuries on volitional behaviors. Some of

these behavioral assays include the inclined plane test (Rivlin and Tator, 1977),

foot-falls during grid walking (Bresnehan et al., 1987; Kunkel-Bagden and Bergman,

1990), and quantitative footprint analysis of locomotion (Kunkel-Bagden and

Bergman, 1990). In addition, quantitative behavioral assays have been applied to

forelimb reaching and grasping after spinal cord lesions (Schrimsher and Reier,

1992, 1993). Most of these tests evaluate gross behavioral performance by noting

whether an animal achieved the goal of a task. An issue with these behavioral

measures is that an animal with a lesion may utilize an abnormal motor strategy to

achieve the goal. Only a few behavioral paradigms have used electromyographic

and kinematic analyses (Barbeau and Rossignol, 1987; Goldberger, 1988) in

attempts to describe, quantitatively, changes in volitional performance.


Animal Model Systems to Study Spinal Cord Injury

A number of different types of spinal cord lesion have been performed in

order to determine which pathways must be interrupted to produce spasticity and/or

a loss of specific volitional behaviors (Wiesendanger, 1985). However, in the

attempt to understand spasticity in animals, there have been few studies that

quantify the different components of the spastic syndrome in order to relate to the

human condition. In evaluating the consequences of specific spinal cord lesions








8

on the development of spasticity, most animal studies have focused on alterations

of stretch reflexes.

Spinal transaction. In cats with chronic spinal transaction, behavioral signs

of spasticity are readily and consistently apparent (Fulton and Sherrington, 1932;

McCouch, 1947; Kozak and Westerman, 1966; Afelt, 1970; Bailey et al., 1980), but

rarely have been quantified (Little, 1986). As determined by subjective methods of

neurological evaluations of cats with complete thoracic transaction (Bailey et al.,

1980), the limbs are hypotonic for the first few days below the level of the lesion.

The patellar tendon reflex is slightly exaggerated, and the cutaneous flexor reflex

is mildly exaggerated, with the possible presence of cross extension. By the end

of the first week, an increase in tone during passive stretch of a limb is apparent,

and a clasp-knife reflex can be elicited. Tendon reflexes are exaggerated with the

possible presence of clonus. Cutaneous flexor reflexes are exaggerated, and

flexor-extensor spasms might be present. Reflex stepping and standing begin to

appear. By 14 days, muscle tone is only mildly increased, yet all spinal reflexes are

exaggerated. Several early phases of reflex exaggeration have been distinguished

by quantitative methods (Little, 1986), but after 30 days muscle tone is markedly

increased, and spinal reflexes are easily elicited and markedly exaggerated.

A similar time-course has been reported for spinal monkeys. However,

depending on the quality of post surgical care, the lower limbs can become

contracted, rigid, and atrophied (Sahs and Fulton, 1940; Liu et al., 1966). In

paraplegic humans the sequence of hyporeflexia followed by hyperreflexia is similar








9

to that observed in the cat and monkey, except that the time-course of these events

is expanded.

Hemisection. The clinical picture of animals with a hemisection is less clear.

Some studies in rats and cats have reported signs of spasticity (Teasdall et al.,

1958, lesions at T4-L1; Murray and Goldberger, 1974, lesions at T13; Carter et al.,

1991, lesions at L5); whereas, others have not (Hultborn and Malmsten 1983,

lesions at T12 and L2; Carter et al., 1991, lesions at T12). When spasticity is

present after incomplete spinal lesions, the time-course of events is similar to that

of an animal with a transaction except that weight bearing occurs in two weeks and

locomotion occurs by three weeks. In addition, hyperreflexive responses can

resolve at later post-injury time periods.

Staggqqered lateral hemisections. The consequences of staggered lateral

hemisection on reflexes and locomotion also have been investigated (Kato et al.,

1985). Initially there is a depression of segmental reflexes at 1 to 2 days. From 3

to 7 days, segmental reflexes are exaggerated, and unusual reflexes are observed

with the presence of airstepping and alternating limb movements. However, from

7 to 53 days, when the animal begins to stand and walk, abnormal reflexes subside

and segmental reflexes become less exaggerated. In primates, staggered bilateral

hemisection results in a clinical picture similar to that observed after complete

spinal transaction (Turner, 1891; Lassek and Anderson, 1961). This difference

among species after staggered lateral hemisections may reflect differences in the








10

organization of the descending systems that control the spinal cord (Holstege and

Kuypers, 1987; Chung and Coggeshall, 1988).

Contusion injury. In animals, a weight drop technique (Allen, 1911) has been

used to model the rapid and traumatic contusive or compressive sort of insult that

causes damage in most patients with spinal cord injury. After a compression or

contusion injury, a hemorrhagic necrosis begins at the center of the insult and is

followed by neural degeneration and the development, over a period of weeks, of

a central cavitation. Modifications of the original technique have improved upon the

reproducibility of the injury (Wrathall et al., 1985; Anderson et al., 1988; Beattie et

al., 1988, Stokes et al., 1992). The behavioral consequences of contusion depend

on the degree of the injury (Gale et al., 1985). With respect to spasticity, both

cutaneous hyperreflexia (Gale et al., 1985) and alterations of physiologically

recorded monosynaptic reflexes (Thompson et al., 1992), develop over a period of

weeks after a contusion injury of the rat spinal cord.

Lesion factors in spasticity. The factors that are critically important to the

development of signs of spasticity in an animal include size, spinal level, and time

after the lesion. Together, these factors determine which pathways are spared, the

extent of deafferentation of the spinal cord, and the excitability of neurons caudal

to the lesion.

Variability in the chronic manifestation of hyperreflexia is apparent in humans

(Mailis and Ashby, 1990). For example, spasticity can be differentiated on the

bases of whether the lesion is spinal or supraspinal (Burke et al., 1972;









11

Wiesendanger, 1985). Following cortical lesions, extensor reflexes are released

over flexors, whereas flexor reflexes become exaggerated after spinal cord lesions.

In addition, not all hemiplegias from cortical or capsular lesions involve a sequential

development from early hypotonia to chronic hypertonia. And, recovery of volitional

control usually results in a decrease in hypertonia in these patients.


Overview of Dissertation

Preliminary to this dissertation research, Rhoton et al. (1988) adapted

traditional methods of human neurological examination to investigate the

consequences of caudal spinal lesions on the tails of cats. Soon after a caudal

spinal transaction, the tip of the tail was flaccid. Then, over a period of weeks, the

tail developed heightened cutaneous flexion reflexes and increased tone

(resistance to stretch of tail muscles). Rigidity of the tail appeared, as did clonus-

like responses to cutaneous stimulation. Subsequently, quantitative measurements

confirmed the impression from neurological testing that tone was greater for tail

muscles of lesioned cats than for normal cats (Ritz et al., 1992).

The dissertation research developed quantitative behavioral methods for

testing cutaneous reflexes. These procedures were evaluated and utilized to study

functional consequences of spinal cord injury and potentially therapeutic

interventions. The key aspect of this research was utilization of sacrocaudal spinal

lesions to evaluate effects of spinal cord lesions on cutaneous reflexes of the tail.

The advantage of a sacrocaudal spinal lesion is that effects of spinal transaction








12

can be studied in the absence of bowel and bladder dysfunctions or disruptions in

gait or hindlimb reflexes. The lesions are made caudal to the lumbar enlargement

and are one to two segments caudal to the parasympathetic region (S1 and S2) of

the spinal cord that controls the bowel and bladder (Thor et al., 1989).

Chapter 2 describes studies designed to acquire precise measurements of

cutaneous reflexes in order to determine the time-course and magnitude of any

changes in cutaneous reflexes of the tails of cats after sacrocaudal spinal lesions.

Cutaneous reflexes of the tail were evaluated in normals and animals with spinal

cord lesions. This research determined that lesions of the sacrocaudal cord (S3-

Cal) provide an appropriate model for investigating effects of transaction of the

spinal cord on segmental reflexes of the limbs.

In the studies described in Chapter 3, this animal model was utilized to

evaluate a potential treatment for some of the debilitating effects of spinal cord

injury. Fetal spinal tissue was transplanted into an acute spinal cord transaction

injury. The effects of transaction and transplantation on cutaneous reflexes were

compared with effects of transaction without introduction of a transplant.

In concluding, Chapter 4 discusses possible mechanisms for the cutaneous

hyperreflexia that is typically observed after a spinal cord injury. Also, mechanisms

for influences of fetal spinal transplantation on the consequences of spinal cord

injury are discussed.













CHAPTER 2

CUTANEOUS REFLEXES OF THE TAIL OF CATS BEFORE
AND AFTER SACROCAUDAL SPINAL LESIONS


Introduction

The consequences of a spinal cord transaction, following a period of

areflexia (spinal shock), include a gradual development of heightened reflexes

caudal to the site of the lesion (Riddock, 1917; Dimitrijevic and Nathan, 1967). Both

phasic and tonic stretch reflexes are affected and are clinically defined as

hyperreflexic by the appearance of exaggerated tendon reflexes and a heightened

resistance to passive stretch of a muscle, respectively (Landau, 1974; Lance,

1980). Additionally, a limb can exhibit hyperreflexive cutaneous reflexes (Riddoch,

1917; Kuhn and Macht, 1949; Rushworth, 1966; Dimitrijevic and Nathan, 1968;

Shahani and Young, 1971, 1973; Meinck et al., 1985; Roby-Brami and Bussel,

1987). Clinically, cutaneous hyperreflexia is observed when an innocuous stimulus

to the skin of the foot or leg induces an exaggerated triple flexion of the hip, knee

and ankle. The Babinski reflex (dorsiflexion of the great toe with a fanning of the

other toes) is a cutaneous flexion reflex of clinical importance that can be exhibited

after a lesion to the spinal cord (Kugelberg et al., 1960; Roby-Brami et al., 1989).








14

Hyperreflexive reflex responses to cutaneous stimulation or muscle stretch are two

components of the spastic syndrome (Young and Shahani, 1986).

In order to study neural mechanisms underlying the development of

spasticity after spinal cord injury, studies in animals have induced hyperreflexia with

complete (Fulton and Sherrington, 1932; Afelt, 1970; Burke et al., 1972; Nelson and

Mendell, 1979; Bailey et al., 1980; Munson et al., 1986; Ritz et al., 1992) and

incomplete lesions (Turner, 1891; Wagley, 1945; Teasdall et al., 1958; Lassek and

Anderson, 1961; Fujimori et al., 1966; Murray and Goldberger, 1974; Kato et al.,

1985; Little, 1986; Carter et al., 1991; Thompson et al., 1992). These studies

primarily have characterized alterations in stretch reflexes. Only a few animal

studies have examined chronic changes in cutaneous (or polysynaptic) reflexes

after spinal cord lesions (Ranson and Hinsey, 1930; McCouch, 1947; Kozak and

Westerman, 1966; Liu et al., 1966; Hultborn and Malmsten, 1983).

Previously, the consequences of a lesion in the cat sacrocaudal spinal cord

which only affects the tail were investigated to develop a minimally disruptive model

of spinal cord injury. Rhoton et al. (1988) observed that a hemisection or complete

transaction in the cat sacrocaudal spinal cord produced clinical signs of hypertonia,

cutaneous hyperreflexia, and clonus-like movements. These classical signs of

spasticity occurred in the absence of bowel and bladder dysfunctions or disruptions

in gait and hindlimb reflexes. Subsequent quantitative measurement (Ritz et al.,

1992) confirmed the changes in muscle tone. After a complete transaction of the

spinal cord, the tail becomes actively ventroflexed in a midline position, and there








15

is a loss of sensation and volitional control. These functional alterations are akin

to paraplegia with spasticity that occurs in human patients with spinal cord injuries

(Young and Shahani, 1986).

To further investigate alterations in tail reflexes after lesions in the cat

sacrocaudal spinal cord, quantitative measurements of cutaneous reflexes of the

tail to electrocutaneous and mechanical stimuli were performed in normal animals

and in animals with lesions of the sacrocaudal spinal cord. Specifically, this study

addressed whether cutaneous reflexes are hyperactive after spinal transaction, as

observed caudal to spinal cord injuries of human patients.


Methods

Subjects. A total of 11 adult cats (10 female, 1 male) served as subjects.

For 8 animals, data were collected before and after lesions of the sacrocaudal

spinal cord. Three animals were part of a prior study (Ritz et al., 1992) and had

previously received either a complete sacrocaudal transaction (n=2) or a left

hemisection (n=1). When the animals were not performing in the behavioral

experiments, they were housed at the University of Florida Animal Resources

Facilities.

Surgery. All surgeries were performed under sterile conditions and gas

anesthesia (Halothane or Isoflourane). Some cats received 50 mg/kg of ketamine

as a preanesthetic. The animal was intubated and a venous catheter was inserted

for infusion of fluids. Dorsal laminectomies were performed at the L6 and L7








16

vertebrae. Cats received a complete transaction or a subpial transaction cavity (~2-

3 mm long) at the S3/Cal level. The lesions were made with precision forceps and

suction. Four to 6 dural sutures secured the spinal cord after the lesion, and when

a subpial transaction was performed, 4 pial sutures were added. Visual inspection

at the time of surgery confirmed the completeness of the lesion.

Postoperative care included antibiotic treatments (Amoxicillin, 50 mgs) for

7 days. Canned food was added to the diet. If necessary, the bladders were

manually expressed, but in most cases normal bladder function returned by the

second postoperative day.

At the termination of the behavioral experiments, the cats were anesthetized

with a lethal dose of sodium pentobarbital (100mg/kg) and transcardially perfused

with isotonic saline solution (0.9% with sodium nitroprusside; 0.5g/L) and heparin

sulfate (1.5cc/L), followed by a solution of 4.0% paraformaldehyde in 0.1 M

phosphate buffer (pH 7.6). Upon extraction of the spinal cord, the completeness

of the lesion was determined, and the level of the lesion was identified.

Testing apparatus. The testing apparatus provided comfortable restraint of

the body and tail of each cat (Figure 2-1). Either a cat sack or a restraint harness

was used to situate the body of the cat, and the tail was bracketed at the base. To

provide quantitative measures of cutaneous reflexes, the tip of the tail was tethered

to a copper-beryllium leaf (2.5 cm by 8 cm by 0.5 mm) with an attached strain

gauge. The strain gauge was incorporated into a Wheatstone bridge circuit, and

the signal was amplified 1000 times by a direct current (DC) amplifier. The tether






























Figure 2-1. The testing apparati. The cats were positioned with a cat sack (top) or
a restraint harness (bottom). The tail was bracketed at the base. Cutaneous
reflexes of the tail were measured by tethering the tip of the tail to a copper-
beryllium leaf containing a strain gauge.
























Site of
Electrical
Stimulation


Tether


Strain
Gauge


Stimulation


Tether


Strain
Gauge








19

tension was held at 1.5 Newtons (N), which held the tail straight in an axial position.

Muscle contractions that moved any portion of the tail away from the axial position

were detectable as an increase in force as measured by the strain gauge. The

measured force assessed the strain produced by a tail reflex. Since the tail was

tethered in parallel, the setup was incapable of measuring the torque of the tail

reflex, which depends on the length of the tail distal to the pivot point (acting as a

moment arm), the angle of the movement, and the reflex force exerted. Attempts

to directly measure torque would encounter an inherent problem in that the pivot

point of tail reflexes were not fixed, but instead depended on stimulus intensity,

since recruited motoneurons course along the length of the tail. The response

forces were collected at a 1 KHz sampling rate and stored digitally in a file, using

computer hardware and software (RC-Electronics).

The timing of all stimulation and event triggering were performed by the

Stoelting Protege System's Stimulus Laboratory Controller with Laboratory

Automation System software.

Training paradigm. Cats were adapted to the testing room and then to the

testing apparatus. For restraint in cat sacks, adaptation to the testing apparatus

included rewards of dry food (BONKERS Cat-Treats or POUNCE Treats for Cats)

as well as physical affection. Training cats restrained with the harness entailed

rewards of liquid food. Subsequently, the proximal tail was braced, and then the

distal tail was tethered. Next, the cats were accustomed to passively receive

mechanical and electrocutaneous stimulation.








20

Testing regimen. When possible, the animals were adapted and tested prior

to receiving a lesion (n=8). Following a sacrocaudal lesion, to evaluate possible

changes in the reflex over time, these animals were tested repeatedly over a period

of at least 7 months and up to a year. Five post-lesion time periods were

specifically examined: 7-14 days, 21-35 days, 70-90 days, 140-165 days, and

greater than 210 days. Due to the methods of restraint and concern for the welfare

of the animal, testing was only rarely attempted during the first week after surgery.

Three other animals were tested only at 3 years post-lesion.

Mechanical stimulation and reflex testing. All mechanical stimulation was

presented on the dorsal surface of the tail at dermatomal level Ca3/Ca4,

approximately 5 cm from the tip of the tail. Mechanical stimulation was presented

with a solenoid and consisted of a 50 ms cutaneous tap elicited through a blunt

metallic probe (1 mm in dia.) or an edge (a Steel T Pin, Create-A-Craft). The probe

was either in continuous contact with the skin or first made contact with the skin

during the tap. In all testing sessions, each stimulus condition was repeated 8

times with an interstimulus interval of 8 s.

Cutaneous response measures and data analysis. Different characteristics

of the response to the mechanical stimuli were tabulated. Individual trials were

evaluated with respect to (1) occurrence of a response, (2) reflex latency, (3) peak

reflex amplitude, and (4) complexity of the response. To tabulate reflex latency,

reflexes shorter than 30 ms were scored in one bin, responses greater than 30 ms

but less than 250 ms were placed in another bin, and reflexes occurring later than








21

250 ms after stimulus onset but less than 750 ms later were scored in a third bin.

Initial responses that occurred at latencies greater than 750 ms were considered as

responses unrelated to the mechanical stimulus. The duration of the reflex was

tabulated into 4 bins: (1) less than 500 ms, (2) greater than 500 ms but less than

1 s, (3) greater than 1 s but less than 2 s, and (4) greater than 2 s. The amplitude

of the reflex was also tabulated into 4 bins: (1) less than 0.15 N, (2) greater than

0.15 N but less than 0.25 N, (3) greater than 0.25 N but less than 0.5 N, and (4)

greater than 0.5 N. To characterize the complexity of the reflex response 3 bins

were created. Simple responses, reflexes with less than 3 peak envelopes were

placed in one bin. Responses with greater than 3 peak envelopes but less than

7 peak envelopes were placed in a second bin. And responses with greater than

6 peak envelopes were placed in a third bin.

For each animal 3 to 4 blocks of trials were grouped together and scored to

acquire frequency values for the reflex measures. The population of frequency

values from pre-lesion and post-lesion animals with chronic transaction lesions

were statistically compared with the Mann-Whitney Rank Sum Test.

Electrocutaneous stimulation. Innocuous electrocutaneous stimulation

consisted of a 5 ms pulse of constant current (DC) delivered via two wells (5 mm

in dia.) of electrode paste (EC 2 Electrode Cream; Grass Instrument Co.) which

were separated by 1 cm. The wells were punched out of a 1/8" thick polycushion

pad (Smith Nephew Rolyan Inc.) and were secured on the tail with elastic tape

(Elastikon, Johnson and Johnson Medical, Inc.). The stimulation site was the








22

dorsal surface of the tail at dermatomal level Ca3, approximately 8 to 10 cm from

the tip of the tail. Electrocutaneous stimulation was presented by Grass Instrument

Company equipment: a stimulator (Model S8), a stimulus isolation unit (Model SIU

5B), and a constant current unit (Model CCU 1A). The amplitude of the

electrocutaneous stimulation was manually set. A current monitor (built in house)

was utilized to measure the amount of current delivered by each stimulus.

Electrocutaneous reflex testing. Tail reflexes were elicited by a single DC

pulse of electrocutaneous stimulation. To determine intensity-response

relationships, stimuli were presented in blocks of 6 to 8 trials at each of 6 different

stimulus intensities in each testing session. Intensities of 2, 3, 4, 6, 8, and 10.5

mAmps/mm2 (in mAmps, these values were 0.4, 0.6, 0.8, 1.2, 1.6, and 2.1) were

presented in an ascending order of blocks. The intertrial intervals were 7 s.

Occasionally, reflex responses were recorded at various intensities of a 10 pulse,

50 Hz train of electrocutaneous stimulation.

Electrocutaneous response measures and data analysis. The stability of

reflex responses to repeated stimulation was evaluated by comparing the force of

responses on the first trial in a block of trials at one intensity with responses on all

subsequent trials presented at that intensity. For this purpose the force of the reflex

was determined by summing the amplitudes of responses over the first 500 ms after

stimulus onset. For a block of trials to be characterized as showing either wind-up

or wind-down to repeated stimulation, two criteria had to be satisfied. First, the

average change in reflex force had to be greater than 20 percent of the reflex force








23

estimate of the first response. And second, either the average change in reflex

force had to be greater than one standard deviation away from the reflex force

estimate of the first trial, or a consistently increasing or decreasing trend in reflex

force must be present.

For all other further characterizations of the reflex, averaged waveforms were

constructed for each stimulus intensity and period of testing. Individual trials were

not included in the average if there was an unstable baseline due to a volitional

body movement. Starting 10 ms prior to stimulus onset and ending 110 ms later,

data were entered into a waveform data file at the 1 KHz sampling rate. For the

next 400 ms of the response, the file received data at 5 ms intervals. From 400

msec until the end of the data collection window (1.5 s, or 3.5 s), the file received

data at 20 ms intervals.

Different characteristics of the averaged reflex waveforms were evaluated.

Reflex latency was measured by extrapolating back to the X-intercept of a line

made through the 10 and 90 percent peak amplitude points on the ascending limb

of the reflex. Peak reflex amplitude and the latency at peak reflex amplitude were

measured directly. The rise time of the ascending limb of the reflex was estimated

by calculating the slope of a line through the 25 and 75 percent peak amplitude

points on the ascending limb of the reflex. Two measures of reflex duration were

the latency of the reflex amplitude to decay to 50 percent and to 25 percent of the

peak reflex amplitude. An estimate of the overall force of the reflex was determined

by summing the reflex amplitudes from the time of stimulus offset until the decay of








24

the reflex response reached 25 percent of the peak reflex amplitude. The different

characteristics of the averaged waveform were calculated by a computer program

written in the C programming language.

Statistical analysis was performed with SigmaStat computer software (Jandel

Scientific Software). A mixed repeated-measures design was utilized for

comparison of a given characteristic of the tail reflex for normal animals and

animals with chronic sacrocaudal transaction. The presence or absence of the

sacrocaudal lesion was treated as the categorical between-subjects factor.

Stimulus intensity was treated as a continuous within-subjects factor. Each

characteristic of the reflex was treated as a continuous dependent variable.

Multiple linear regression analysis was used to evaluate the results (Pedhazur,

1982). To perform the multiple linear regression analysis, the independent and

dependent variables were transformed, when appropriate, to pass tests of normality

(Kolmogorov-Smirnov test) and homoscedasticity (constant variance) and to

linearize the relationship between stimulus intensity and a characteristic of the

reflex. Criterion scaling was used to code for the subjects.

Multiple linear regression analysis was used to evaluate changes in the

reflex over time after sacrocaudal lesions. In the repeated-measures design, the

pre-lesion and 5 post-lesion time periods were treated as a categorical within-

subjects factor. Effect coding was utilized to vector code the categorical variable

for the multiple linear regression analysis. Stimulus intensity was treated as a

continuous within-subjects factor. Each characteristic of the reflex was treated as








25

a continuous dependent variable. Transformations of the independent and

dependent variables were the same as those used for comparisons of pre-lesion

and late post-lesion reflex characteristics.


Results

For 7 of the animals tested, spinal transactions were successfully placed at

sacrocaudal levels between the caudal half of S2 and the caudal half Cal. These

7 animals comprised a group for analysis of chronic changes in cutaneous reflexes

after sacrocaudal transaction. For 5 of these animals, pre-lesion and acute and

chronic post-lesion data were collected to evaluate the time-course of the effects

of the lesions. The reflex responses from 4 additional animals were treated

separately. For one animal, the sacrocaudal transaction was located at rostral Cal,

but evidence of an ischemic episode was present in caudal segments, where cresyl

violet staining revealed an abundance of macrophages and few neurons. In

another animal, the sacrocaudal transaction was mistakenly placed at the Ca3

segment. In another animal, an incomplete sacrocaudal lesion at S3 cut through

the dorsal aspect of the spinal cord, disrupting the dorsal columns, Lissauer's tract,

and the dorsal aspect of the dorsolateral white matter, bilaterally (Figure 2-2,

bottom). A near perfect left hemisection at S3 was histologically confirmed for the

fourth animal (Figure 2-2, top).






























Figure 2-2. Histological reconstructions of the sacrocaudal hemisections. The top
drawing shows the lesion from the animal with the left hemisection. The bottom
drawing shows the lesion from the animal with the bilateral dorsal hemisection.








27










Responses to Mechanical Stimuli in Normal and Chronic Transection Animals

Examples of tail responses to mechanical stimuli are shown in Figure 2-3.

When tail reflexes could be elicited, various types of responses were expressed

which depended on both the type of mechanical stimulation and the specific animal

tested. Due to the variability observed in the reflexes produced by mechanical

stimuli, different characteristics of the reflex response were tabulated. And,

because of the inability of a single type of mechanical stimulus to consistently elicit

reflexes in all animals, 4 different types of mechanical stimulation were used.

Reflex threshold to the mechanical stimuli in normal animals (n=7) and

animals with a chronic transaction (n=7) was evaluated by determining the reflex

response frequency (Figure 2-4). The trend observed for a decrease in mechanical

reflex threshold for 3 of the 4 mechanical stimuli, as measured by an observed

increase in response frequency, was significant for stimulation with the blunt probe

that remained in contact with the skin (Mann-Whitney Rank Sum test; T = 24.0, P

= 0.008). Over 80 percent of the reflexes elicited were made up of less than 3

envelope peaks, and no differences were observed in the complexity of the reflex

between pre-lesion and chronic post-lesion animals. In post-lesion animals

increases were observed when compared with pre-lesion animals in the frequency

of reflexes with latencies between 30 and 250 ms which paralleled a decline in the

frequency of longer latency responses in post-lesion animals (e.g., edge stimulus,

on skin; P = 0.026). Long latency responses to the blunt probe (off skin) declined

in post-lesion animals (median frequency 0.00) from the frequency observed (0.26)




























Figure 2-3. Representative traces of tail reflexes to mechanical stimulation. From
a normal (top) and an animal with a chronic sacrocaudal transaction (bottom), the
traces show various reflex responses of the tail to mechanical stimulation. For
clarity, the reflex traces are offset. The bottom trace (top diagram) shows the
stimulus artifact when the mechanical probe displaces a passive tail.






























0.0 ms


800 ms/div


.0 s


800 ms/div


0.0 ms











'-0
0) L 0)
o -D
a oa -

cE
0. 0)0
.c ca



.0 0
0) C





o (--




". cs 0








S .--c .
0E c:
0o.0
c( l .














6-
o ) 0 0.
Se)0 'U-

C
C CH

m 0-0 >










cg .
0 U a)
* ) U )




.0 )-


a) C a) C)
" 0 C












iI E -c E















I I


Z:=

o
0


CD
'a
w


i-


(ueipaAVL) Af!l!qeqoJd


C
0
.0
0








C
C)





.0(
0














.0

a-


I -








33

in pre-lesion animals (Mann-Whitney Rank Sum test; T = 54.5, P = 0.0086). This

finding confirmed that the responses with latencies longer than 250 ms were most

likely volitional tail movements in response to the mechanical stimuli. And although,

in comparison to pre-lesion animals, there were trends in post-lesion animals for an

increase in reflex duration and peak reflex amplitude in response to the mechanical

stimuli, none were found to be statistically significant (Mann-Whitney Rank Sum

test).


Responses to Electrocutaneous Stimuli in Normal and Chronic Transection Animals

Representative traces of tail reflexes to single electrocutaneous stimuli are

shown in Figure 2-5. Reflexes occurred at short latencies with rapid rise times that

dissipated with a gradual decay. For normal animals and animals with a chronic

sacrocaudal transaction, the tail responded similarly, allowing direct statistical

comparisons between pre- and post-lesion measurements of different features of

the reflex. For the following comparisons, averages from 4 pre-lesion sessions

were compared with averages from 2 or 3 sessions at the end of the testing period

for each animal.

Stimulation at 2 mA/mm2 was near threshold for elicitation of the reflex, and

responses were variably absent. Therefore, this intensity was not utilized for the

analyses of different characteristics of the reflex across stimulus intensities.

Comparing response probabilities for stimulation at 2 mA/mm2 during the pre-lesion

and chronic post-lesion periods, significant differences in reflex threshold were not





























Figure 2-5. Representative traces of tail reflexes to single DC pulses of
electrocutaneous stimulation. From a pre-lesion recording (top) and after a chronic
sacrocaudal transaction (bottom), the top 3 traces show reflex responses of the tail
to electrocutaneous stimulation at different stimulus intensities. For clarity, the
reflex traces are offset. The bottom trace shows the timing of the electrocutaneous
stimulus.




















z
(0
0
U1)
Q
L..
0
LL







0.


800 ms/div


0 ms


800 ms/div


z

0



0
1L


4.0 s


0.0 ms


4.0 s


. . .

8 mA/MM2! ^




6t mA/mm2 /



3 mA/mm2 A *

o . . .

iStimulus


8 mA/mm-2




6mA/ mm . --.


3 mA/mm2 ..
| . .. .


Stimulus

. . .








36

indicated (Mann-Whitney Rank Sum Test (T=60.0, P=0.694). For pre-lesion testing

at 2 mA/mm2, the median response probability was 0.6 (n=8), and the post-lesion

response probability was 0.75 (n=7).

In an overall evaluation of the electrocutaneous reflex, the force of the reflex

was observed to be larger for animals with a chronic transaction than for normal

arimals. The force of the reflex increased for both groups with higher levels of

stimulus intensity (Figure 2-6). Multiple regression analysis (Table 2-1) was

performed after the inverse exponential relationship was linearized by applying an

inverse transformation to the independent variable and a natural log transformation

to the dependent variable. The multiple linear regression analysis, which contained

stimulus intensity as a repeated continuous variable and the presence or absence

of the sacrocaudal lesion as the between-subject variable accounted for over 90

percent of the variance and was significant (r2 = 0.906; F = 30.59, P < 0.01). The

possible interaction between stimulus intensity and the presence of a lesion was not

significant (F= 3.04, P > 0.05). The main findings of the analysis were that: (1)

reflex force significantly increased with increases in stimulus intensity (F = 70.79,

P < 0.01); and (2) over this stimulus intensity range, there was a post-lesion

increase in reflex force (F = 7.34, P < 0.05).

An increase in peak reflex amplitude in animals with a chronic transaction in

comparison to normal animals was found to be contributing factor for the

differences observed in reflex force between the two groups. Peak reflex amplitude

was observed to increase for both groups with higher levels of stimulus intensity,










c 0). e -n
0E E' o 0
.c c EOD


St c c a. "
0) a 8) 0 0
S- -C c o


U, 0 ) C

. C4 ( a) 3





a o



c a) .. X 0 U) CU
C 0'-
to a) U. o E3

a) o. a








- a) ) C ^


E'o0-E)mo
a) 0 a -ao




S--- as) E
0 .a 4- C
SxO.*- -. m 0






_0 0 C 0 a)
.C ) 0c o
| .> .0











S _o C a)


"3 -- 0 a..0

Sa --
AE (n 0
o x a ) -.












(IoS













0


01


E

E


(0
C
C)
C% 4




CO,

CM


o 0 0 0 0 0 0 0 0 0 0 0 0
o0 0 0 0 0 0 0 0 0 0 0 0
(N o- -0 M N

(N) 8ojoj xeIJ9>j


T-




0)
r-




E
E




C
c


0 0 0 0C 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0
v, rC C- o o oo ,- D Co O CT M C '


(N) eoJOJ xeiJe8





















Table 2-1. Summary of the multiple linear regression analysis of reflex force of
normal animals and after a chronic sacrocaudal transaction.

Reflex Force
2 Equation Std.
Source DF SS F P R2 Equation Std.
Coefficient Error

Constant 1.33 0.234
Treatment 1 39.4 7.34 <0.05 0.279
Subject 13 69.8 20.44 <0.01 0.773 0.94 0.041
Intensity 1 18.6 70.79 <0.01 0.904 -6.05 0.638
Treatment X 1 0.8 3.04 >0.05 0.906
Intensity
Residual 51 13.4
Total 67 127.8 30.59 <0.01 0.906
Transformations: Reciprocal of Intensity; natural log of the dependent variable.
Normality Test: Failed; Kurtosis=13.3, Skewness=2.34.
Homoscedasticity Test: Passed.








40

in what appeared to be a saturating nonlinear function (Figure 2-7). Multiple

regression analysis (Table 2-2) was performed after the relationship was linearized

by taking the reciprocal of the independent variable. The multiple linear regression

analysis, which contained stimulus intensity as a repeated continuous variable, the

presence or absence of the sacrocaudal lesion as the between-subject treatment

variable, and an interaction term between stimulus intensity and the possession of

a sacrocaudal lesion, accounted for over 90 percent of the variance and was

significant (r2 = 0.924; F = 39.06, P < 0.01). The main findings of the analysis were

that: (1) peak reflex amplitude significantly increased with increases in stimulus

intensity (F = 82.7, P < 0.01); (2) over this stimulus intensity range, there was a

post-lesion increase in the peak reflex amplitude (F = 9.13, P < 0.01); and (3) for

animals with a lesion, peak amplitude increased at a greater rate with increases in

stimulus intensity (F = 13.13, p < 0.01). The point of intersection of the ordinal

interaction was estimated be at 1.8 mA/mm2. Between normal animals and animals

with a chronic transaction, peak reflex amplitudes were significantly different at

intensities greater than 2.4 mA/mm2 (Johnson-Neyman Technique).

Along with the differences observed between the two groups in peak reflex

amplitude, the latency at peak reflex amplitude was observed to remain relatively

constant in normal animals, but for animals with a lesion, latency at peak reflex

amplitude increased with higher levels of stimulus intensity (Figure 2-8). Multiple

regression analysis (Table 2-3) was performed after the relationship was linearized

by applying an inverse transformation to the independent variable and log











mh,) ->C C







3 0) = 0-- -
() C ( Q)













75 c c -"
e --0
> Q. 0)2






(-| 0












._ "U)
C0 C (D-






0-- ) c c
E. 00 S cu




















cuece


o co
c e
a) C)U

t-^l





S-0 a


a) i 3 2





" +CO


0o ( UO ) ICT O C T- 0
o o o d d 0 o 6 o 6

(N) epnl!ldwV lee9d










Jx
\ \







\ ^x\
N\ V^
X 4< \X


Oc co I1- o LO I Co Co o

(N) epnl!ldwv W)ted


0

a)
01
E
E



c
a)
10


CO)


42






-co E
-I)





C





















Table 2-2. Summary of the multiple linear regression analysis of peak reflex
amplitude of normal animals and after a chronic sacrocaudal transaction.

Peak Amplitude
Source DF S F R2 Equation Std.
Coefficient Error

Constant 0.186 0.025
Treatment 1 1.230 9.13 <0.01 0.323
Subject 13 1.751 23.90 <0.01 0.781 1.015 0.039
Intensity 1 0.466 82.70 <0.01 0.904 -0.965 0.094
Treatment X 1 0.074 13.13 <0.01 0.924 0.382 0.094
Intensity
Residual 52 0.293
Total 68 3.521 39.06 <0.01 0.924
Transformations: Reciprocal of Intensity.
Normality Test: Failed; Kurtosis=0.105, Skewness=0.836.
Homoscedasticity Test: Passed.










) 0) ) 0 )


a) .c o =-
.C .

E0 E

x E >xm


-" _. -% v
) a- ay o
_ 0 -- CV
a) v


co c


c 75 c
Sc b- C C
.u --- mo








---o 0 0
.: 2 co-



S..-c o JOU
. 0 ,-n Ci CU












" o ." L "a .C
-c c -a


SOm c0 O_ 0.





a) C o IC -
0 0 C
cu a) U) a)





U C) C L "
4- 0) a CU 0 C



U ) c x C
Mco a)







0 O- _C (D

































; I I I I I



(oesw) epnl!IdwV lead le Aougie-I


I I I I ; I I I i
0 0 0 0 0 0 0 0 0 0 C
0 eo s pn 0lV o co co i

(oesw) apnllidwV N89d Ie Aouele1


r- 45

0


1

co E
E

>I

C








CN
0










c0


- co







CN
3
T-
- I


(NJ.





















Table 2-3. Summary of the multiple linear regression analysis of latency at peak
reflex amplitude of normal animals and after a chronic sacrocaudal transaction.

Latency at peak reflex amplitude

Source DF SS F P R2 Equation Std.
Coefficient Error

Constant 0.609 0.312
Treatment 1 3.39 13.99 <0.01 0.323
Subject 13 3.15 5.30 <0.01 0.781 0.898 0.068
Intensity 1 0.39 8.54 <0.01 0.904 -0.963 0.273
TreatmentX 1 0.77 16.85 <0.01 0.924 1.258 0.272
Intensity
Residual 51 2.33
Total 67 10.30 10.53 <0.01 0.924
Transformations: Reciprocal of Intensity; natural log of the dependent variable.
Normality Test: Passed.
Homoscedasticity Test: Passed.








47

transformation to the dependent variable. The multiple linear regression model,

which contained stimulus intensity as a repeated continuous variable, the presence

or absence of the sacrocaudal lesion as the between-subject treatment variable,

and an interaction term between stimulus intensity and the presence of a

sacrocaudal lesion, accounted for over 90 percent of the variance and was

significant (r2 = 0.92; F = 10.53, P < 0.01). The main findings of the analysis were

that: (1) latency at peak reflex amplitude significantly increased with increases in

stimulus intensity for animals with a lesion (F = 8.54, P < 0.01); (2) over this

intensity range, there was a post-lesion increase in latency at peak reflex amplitude,

compared to pre-lesion values (F = 13.99, P < 0.01); and (3) for animals with a

lesion, latency at peak reflex amplitude increased at a greater rate with increases

in stimulus intensity (F = 16.85, P < 0.01). The point of intersection of the ordinal

interaction was estimated to be at 2.65 mA/mm2. Between normal animals and

animals with a chronic transaction, the latency at peak reflex amplitude values were

significantly different at intensities greater than 3.1 mA/mm2 (Johnson-Neyman

Technique).

The rise time of the reflex was measured in order to investigate a possible

contribution to observed increases in peak amplitude at higher stimulus intensities

and in animals with a sacrocaudal lesion. Reflex rise time was observed to

increase for both groups with higher levels of stimulus intensity in what appeared

to be a saturating nonlinear function (Figure 2-9). Multiple regression analysis

(Table 2-4) was performed after the relationship was linearized by taking the










(D a) a) = --.C

* E .m c-

a) 0



C I-
CU 0 .-









- > -a
a) .o 0) -a M







--ey
LOCO)





SL. o c Q.
a) LoU








-C 0 C


S0) C--
m C) >' C"







". .2. 2O.Q
" mc







. C -0 .C C(

ro E o o

'- Oc a) C.
.0- r 8 0-



*.- 9 Q c) o *C
-0 0 c -








S.c n m c C -c









o .g e '- -
4- n fn- C





C"- (D C-


a- 0 ()0 C.) .D
* 0.C i--
































C..4 Ci C


(oes/seuAp) ewil esjI


I. I I I I I I
c(O r o 0 o (0 '" No 0 co( (D Ir 0c o
CM. C' r C4 v -- '- 0 0 o 0 0


(oes/seuAp) ewil esmIu


0
r-C
- 0C

-o E
E

E


')


4
/

7


4-


0




E-





Lo C
E



(0
a) J


l I I r I i I I 1 1


J





















Table 2-4. Summary of the multiple linear regression analysis of reflex rise time
of normal animals and after a chronic sacrocaudal transaction.

Rise Time

Source DF SS F P R2 Equation Std.
Coefficient Error

Constant 0.082 0.012
Treatment 1 0.1410 2.52 >0.05 0.140
Subject 13 0.7280 55.57 <0.01 0.861 0.991 0.030
Intensity 1 0.0833 82.66 <0.01 0.943 -0.402 0.040
Treatment X 1 0.0048 4.76 <0.05 0.949 0.096 0.040
Intensity
Residual 52 0.0524
Total 68 1.0095 59.36 <0.01 0.949
Transformations: Reciprocal of Intensity, log of the dependent variable plus one.
Normality Test: Passed.
Homoscedasticity Test: Passed.








51

inverse of the independent variable and applying a log transformation to the

dependent variable. The multiple linear regression analysis, which contained

stimulus intensity as a repeated continuous variable, and an interaction term

between stimulus intensity and the possession of a sacrocaudal lesion, accounted

for over 90 percent of the variance and was significant (r2 = 0.95; F = 59.36, P <

0.01). The main findings of the model were that: (1) reflex rise time significantly

increases with increases in stimulus intensity (F = 82.7, P < 0.01); and (2) for the

between-subject variable, significant differences were not obtained for the presence

or absence of a lesion (F= 2.52, P > 0.05); and (3) for normal animals reflex rise

time increased at a greater rate with increases in stimulus intensity than in animals

with a lesion (F = 4.76, P < 0.05). The point of intersection of the ordinal interaction

was estimated be at -3.4 mA/mm2. Between normal animals and animals with a

chronic transaction, the rise time values were significantly different at intensities

greater than -0.4 mA/mm2 (Johnson-Neyman Technique).

Two different measures of reflex duration were obtained in order to

determine the contribution of reflex duration to the reflex force estimate. It took

longer for animals with a lesion to reach half-maximal reflex duration when

compared to normal animals. Half maximal reflex duration was observed to

increase for both groups with higher levels of stimulus intensity (Figure 2-10). The

nonlinear relationships between stimulus intensity and half maximal reflex duration

were fit with hyperbola functions. Multiple regression analysis (Table 2-5) was

performed after the hyperbola relationship was linearized by applying an inverse










^ -a 4-; U) A


c 0 w o 0
E 0- 0



C .. V c




E M ( "
c _Ea










"c r c .- Co
oE ,-0 4



* a ()- o )=
-C O C "
X- 0 -




S.c (U .c >
(D ^ ) a) -
* (o .c m ..)
c E 0





* '.. r h. o E



.( .. o -
TC "- ,2



1-- > o
Ec a).. (D g


r- a (
c -c










0-c O E 0

"0 C ..
(o q (o > 0 E
. 4- 0 C .. U


0) U) U)

































I l I I i I I I I I

00000000000000000
N-o o U l) c CM J o 0) co N- co m ttv co c'J T-


(oesw) xew 9-0 uoilejna


00000000000000
00000000000000
N 0 CO a O 0 0 CON O 0 0 0
-esw) xew- r u- -n


000
000
CO')NJ T-


SIf




\ /
\ I
\ I



I~~ ~~ ,l. l ..l


0

0)

C4
co E
E
-E-



(D
U)
c-





















Table 2-5. Summary of the multiple linear regression analysis of half maximal
reflex duration of normal animals and after a chronic sacrocaudal transaction.

Half Maximal Duration
Source DF SS F P Equation Std.
Source DF SS F P Coefficient Error

Constant -1.04E-3 3.41 E-4
Treatment 1 1.01 E-4 8.21 <0.05 0.298
Subject 13 1.60E-5 12.94 <0.05 0.769 0.933 5.01E-2
Intensity 1 2.97E-5 31.23 <0.05 0.857 7.65E-3 1.21E-3
Treatment 1 1.40E-6 1.47 >0.05 0.861
X Intensity
Residual 51 4.85E-5
Total 67 2.92E-4 19.20 <0.01 0.861
Transformations: Reciprocal of Intensity minus 1; reciprocal of the dependent variable.
Normality Test: Failed; Kurtosis=7.45, Skewness=2.16.
Homoscedasticity Test: Passed.








55

transformation to both the independent variable and the dependent variable. The

multiple linear regression analysis, which contained stimulus intensity as a repeated

continuous variable and the presence or absence of the sacrocaudal lesion as the

between-subject variable, accounted for over 80 percent of the variance and was

significant (r2 = 0.86; F = 19.2, P < 0.01). The interaction between stimulus

intensity and the presence of a lesion was not significant (F = 1.47, P > 0.05). The

main findings of the analysis were that: (1) half maximal reflex duration significantly

increased with increases in stimulus intensity (F = 31.23, P < 0.05); and (2) over

this stimulus intensity range, there was a post-lesion increase in half maximal reflex

duration (F = 8.21, P < 0.05).

One quarter maximal reflex duration values were observed to be greater in

animals with lesions in comparison to normal animals (Figure 2-11). Multiple

regression analysis (Table 2-6) was performed after the relationship was linearized

by taking the reciprocal of the independent variable and performing a natural log

transformation of the dependent variable. The multiple linear regression analysis,

which contained stimulus intensity as a repeated continuous variable, the presence

or absence of the sacrocaudal lesion as the between-subject treatment variable,

and an interaction term between stimulus intensity and the possession of a

sacrocaudal lesion accounted for 90 percent of the variance and was significant (r2

= 0.899; F = 27.92, P < 0.01). The main findings of the analysis were that: (1)

quarter maximal reflex duration significantly increased with increases in stimulus

intensity (F = 3.54, P < 0.01); (2) over this stimulus intensity range, there was a










U) D -o 7 "o V
0) C C 0.
w C 0 ( "
C=_ 0- -z c


0 o 0o


' ( C a.


a) o. 2 .v o


o.-
(0 0






SC- o (:
0 -- O) "--



cu c c u.9
E-- :3 03o


-u
CD. ..
X -50 0







,- 0 C I
o U) CU

o.-. u .



C) 0 0E
)C ( a 0--





0c) c

.c ;-_ -- C 0
L- 0) -o


E m ao E
C C 0-C =
a) c a )




0_ 0- C

0) M % I ..0 C
c c- o-cc m
- c -0)
.cn -- --


C :Z *- C .-


c- a 0 o


U n CLt
EU > mO c










I


CN



-E

0c
U)
-L
cD
10


I I I I I I I t I I
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 A 0 0 0 0 0 L 0 0 U) 0 0L
0 CN 0 1- UL (N 0 f'- LA (N 0 r- L (N
C) CO C) (M ( w (I (N '- '- *- -

(oasw) xew cz'O uoiieJna


0






0)
CN
co E


E-


(IS
-LA^


0 0 0 0 0 0 0 0 0 0 0 0 0
0 Ln 0 nO 0 A 0 LA 0 Lo 0 Ln 0 Ln
Lo ( ,,,I O N I L .. 0 N- LA ( I 0 N- I L ( ,

(oesw) xew 970 uoriejnc




















Table 2-6. Summary of the multiple linear regression analysis of quarter
maximal reflex duration of normal animals and after a chronic sacrocaudal
transaction.

Quarter Maximal Duration
2 Equation Std.
Source DF SS F P R2 Equation Std.
Coefficient Error

Constant 0.595 0.280
Treatment 1 11.60 5.55 <0.05 0.242
Subject 13 27.19 21.70 <0.01 0.810 0.965 0.043
Intensity 1 3.54 36.72 <0.01 0.884 -1.601 0.234
Treatment X 1 0.07 7.68 <0.01 0.899 0.723 0.233
Intensity
Residual 50 4.82
Total 66 43.07 27.92 <0.01 0.899
Transformations: Reciprocal of Intensity minus one; natural log of the dependent
variable.
Normality Test: Passed.
Homoscedasticity Test: Passed.








59

post-lesion increase in quarter maximal reflex duration (F = 5.55, P < 0.05); and (3)

for animals with a lesion, quarter maximal reflex duration increased at a greater rate

with increases in stimulus intensity (F = 7.68, P < 0.01). The point of intersection

of the ordinal interaction was estimated be at 2 mA/mm2. Comparing normal

animals and animals with a chronic transaction, the quarter maximal durations were

significantly different at intensities greater than 2.3 mA/mm2 (Johnson-Neyman

Technique).

Reflex latency was observed to be shorter in normal animals in comparison

to animals with a transaction of the sacrocaudal spinal cord. The latency of reflex

onset was observed to decrease for both groups with higher levels of stimulus

intensity in what appeared to be a saturating nonlinear function (Figure 2-12). The

nonlinear relationships between stimulus intensity and reflex latency were fit with

inverse exponential functions. Multiple regression analysis (Table 2-7) was

performed after the inverse exponential relationship was linearized by applying a

natural log transformation to the dependent variable. The multiple linear regression

analysis, which contained stimulus intensity as a repeated continuous variable and

the presence or absence of the sacrocaudal lesion as the between-subject variable

accounted for over 80 percent of the variance and was significant (r2 = 0.834; F =

16.24, P < 0.01). The possible interaction between stimulus intensity and the

presence of a lesion was not significant (F= 0.53, P > 0.05). The main findings of

the analysis were that: (1) reflex latency significantly decreased with increases in










A-me



C 0 0 >
- 4. CC- C
w : o .-"





.- e- 2_--


E .ca =Z-wC
(.*I E
=) .Eo






c ~CO
_" ) C' v
r- U) *& h A



. (0 E ^
e.c 0 _









a. b
'o P-' o v
C:meU







E v,
D 0 0).0












L m "


Z 0.-






-a .o- .c
0.0 c 0,- (D

C) a .b
re oa




































S0 L)O 0 0


(oeswu) Aoualel


o U) o

(oesw) Aouelel


/



* /"


61


0
0-



co E
E




-I










(0
















- *E
E
co












IC
- 0


co E





















Table 2-7. Summary of the multiple linear regression analysis of reflex latency
of normal animals and after a chronic sacrocaudal transaction.

Reflex Latency
2 Equation Std.
Source DF SS F P R2 Equation Std.
Coefficient Error

Constant 4.403 1.158
Treatment 1 202.5 12.66 <0.01 0.343
Subject 13 208.0 8.46 <0.05 0.697 0.985 0.060
Intensity 1 80.3 42.43 <0.01 0.834 -5.664 0.772
TreatmentX 1 1.0 0.53 >0.05 0.834 0.382 0.094
Intensity
Residual 51 98.4
Total 67 589.2 16.24 <0.01 0.834
Transformations: Log of Intensity.
Normality Test: Passed.
Homoscedasticity Test: Passed.








63

stimulus intensity (F = 42.43, P < 0.01); and (2) over this stimulus intensity range,

there was a increase in post-lesion reflex latency (F = 12.66, P < 0.01).

How the different measured characteristics of the reflex waveform were

related was examined with Spearman rank order correlation coefficients as shown

in Table 2-8. In normal prelesion animals, measures of force, half maximal

duration, quarter maximal duration and peak amplitude were significantly

intercorrelated. Reflex rise time was significantly correlated with reflex force, peak

amplitude and latency at peak reflex amplitude, but not with the measures of

duration. Latency at peak reflex amplitude was significantly correlated with the

measures of duration, but not with peak amplitude or reflex force. Reflex latency

was not significantly correlated with any of the other measured characteristics. In

contrast, all of the reflex characteristics were significantly intercorrelated in chronic

post-lesion animals. The correlations among the reflex characteristics were

significantly greater in the chronic post-lesion animals (Mann-Whitney Rank Sum

Test; T = 306.0, P =< 0.0001).


The Time-course of Changes in Reflex Characteristics

The time-course of changes in the electrocutaneous tail reflex after a

sacrocaudal spinal transaction was evaluated for 5 animals that were tested

preoperatively and at 5 different post-lesion time periods. Post-lesion, alterations

in the electrocutaneous reflex progressively developed over a period of months.











Table 2-8. Spearman rank order correlation coefficients between
measured characteristics of the reflex waveform.


Latency at Duration Duration Rise Reflex Reflex
Peak
Am pu 0.25 max 0.5 max Time Latency Force
Amplitude

Normal

Peak 0.013* 0.451 0.297 0.840 -0.255 0.840
Amplitude 0.937 0.008 0.082 <0.001 0.133 <0.001

Latency at Peak 0.650 0.610 -0.448 -0.004 0.208
Amplitude <0.001 <0.001 0.006 0.980 0.082

Duration 0.710 0.038 -0.311 0.740
0.25 max <0.001 0.829 0.073 <0.001

Duration -0.112 -0.211 0.620
0.5 max 0.520 0.223 <0.001

Rise -0.238 0.518
Time 0.162 0.002

Reflex -0.326
Latency 0.056

Postlesion

Peak Amplitude 0.750 0.870 0.810 0.740 -0.470 0.940
<0.001 <0.001 <0.001 <0.001 0.006 <0.001

Latency at Peak 0.730 0.750 0.540 -0.720 0.730
Amplitude <0.001 <0.001 0.002 <0.001 <0.001

Duration 0.950 0.640 -0.650 0.970
0.25 max <0.001 <0.001 <0.001 <0.001

Duration 0.570 -0.630 0.930
0.5 max <0.001 <0.001 <0.001

Rise -0.394 0.660
Time 0.023 <0.001

Reflex -0.542
Latency 0.001

*For each cell, the top value is the correlation coefficient while
the bottom number is the corresponding P value.








65

The force estimate of the reflex was observed to increase over time after the

lesion for individual animals (Figure 2-13). For the group of subjects, a multiple

linear regression analysis (Table 2-9) accounted for over 70 percent of the variance

and was significant (r2 = 0.747; F = 30.8, P < 0.0001). Interactions between

stimulus intensity and reflex force at different times after the lesion were not

significant (P > 0.05). The main findings of the multiple regression analysis were:

(1) reflex force significantly increased with increases in stimulus intensity (t = -9.18,

P < 0.0001), and (2) the grouping of reflex force values into different post-lesion

time periods accounted for a significant proportion of the variance (Table 2-9). A

post hoc Scheffe analysis found that pre-lesion reflex force values were not

significantly different from post-lesion values at 7-14 and 21-35 days (P > 0.05), but

the reflex force at those 3 time periods was significantly less than at 70-90 days,

140-165 days, and >210 days (P < 0.05).

Peak reflex amplitude was observed to increase over time after the lesion for

individual animals (Figure 2-14). For the group of subjects, the multiple linear

regression analysis (Table 2-10) accounted for over 70 percent of the variance and

was significant (r2 = 0.743; F = 30.6, P < 0.0001). Interactions between stimulus

intensity and peak amplitude at different times after the lesion were generally not

significant (P > 0.05; except for the period 70-90 days, P = 0.0217). The main

findings of the multiple regression analysis were: (1) peak reflex amplitude

significantly increased with increases in stimulus intensity (t = -10.142, P < 0.0001),

and (2) the grouping of peak reflex amplitudes into different post-lesion time periods



























Figure 2-13. The time course of changes in reflex force. Reflex force, averaged
across intensities for each of 5 animals (symbols), is shown at 6 different time
periods. The columns represent the average reflex force determined across
animals. A post hoc Scheffe analysis found that pre-lesion values were not
significantly different from the 7-14 and 21-35 day values (P > 0.05), but reflex force
at those 3 time periods was significantly less than the mean force at 70-90 days,
140-165 days, and >210 days (P < 0.05).























1200
*

1000
*
*
Z 800



lu 600 -
x


O 400 -



200- 0




Prelesion 7-14 21-35 70-90 140-165 >210

Time period (days)










Table 2-9. Summary of the multiple linear regression analysis of changes in
reflex force over time after sacrocaudal transaction.

Reflex Force
Source DF SS F P R2
Regression 15 167.6
Residual 122 56.7
Total 137 224.3 30.8 <0.0001 0.747


Equation Standard
Variable t PCoefficient Error


Constant 7.236
Subject 12.825
Intensity -9.181
7-14 days -3.857
21-35 days -2.444
70-90 days 3.648
140-165 days 4.233
210 days plus 4.811
Intensity X 3.374
7-14 days
Intensity X 1.126
21-35 days
Intensity X -1.727
70-90 days
Intensity X -1.211
140-165 days
Intensity X -0.319
> 210 days
Transformations: Reciprocal of
Normality Test: Passed.
Homoscedasticity Test: Failed.


<0.0001
<0.0001
<0.0001
0.0002
0.0159
0.0004
<0.0001
<0.0001

0.7088

0.2622

0.0867

0.2281

0.7501


2.120
0.703
-6.069
-0.487
-0.314
0.498
0.535
0.608


0.546

1.647

-2.695

-1.747

-0.460


0.2930
0.0548
0.6610
0.1263
0.1285
0.1366
0.1263
0.1263


1.4586

1.4620

1.5610

1.4421

1.4421


Intensity; natural log of the dependent variable.


























Figure 2-14. The time course of changes in peak reflex amplitude and latency at
peak reflex amplitude. Peak amplitudes (top) and latency at peak amplitude
(bottom) for each of 5 animals (symbols) are shown at 6 different time periods,
averaged across intensities. The columns represent the average peak amplitude
and latency at peak reflex amplitude determined across animals. A post hoc
Scheffe analysis found that pre-lesion peak amplitudes were significantly less than
post-lesion amplitudes at 21-35, 70-90, 140-165 and >210 days (P < 0.05). Post
hoc Scheffe analysis found that pre-lesion latencies at peak amplitude were
significantly less than post-lesion latencies at 70-90, 140-165 and >210 days (P <
0.05).























E
mQ.
<
cc
0)
0.

















0
V)
E
()

E


cc
<
0





-j


0.8


0.7


0.6


0.5


0.4


0.3


0.2


0.1


0






160


140


120


100


80


60


40


20


0


140-165 >210


Prelesion 7-14 21-35 70-90 140-165

Time Period (days)


>210


A


A A



----
0 0
A


Prelesion


7-14 21-35 70-90

Time Period (days)










Table 2-10. Summary of the multiple linear regression analysis of changes in
peak reflex amplitude over time after sacrocaudal transaction.

Peak Amplitude
Source DF SS F P R2
Regression 15 3.75
Residual 124 1.29
Total 139 5.04 30.6 <0.0001 0.743


Equation Standard
Variable t PCoefficient Error


Constant
Subject
Intensity
7-14 days
21-35 days
70-90 days
140-165 days
210 days plus
Intensity X
7-14 days
Intensity X
21-35 days
Intensity X
70-90 days
Intensity X
140-165 days
Intensity X
> 210 days


6.220
13.413
-10.142
-1.946
0.093
1.599
2.962
4.133

0.982

-0.416

-2.323

0.012

-0.253


<0.0001
<0.0001
<0.0001
0.0538
0.9258
0.1123
0.0037
<0.0001

0.3281

0.6780

0.0217

0.9897

0.8004


0.2030
0.9837
-0.9987
-0.0368
0.0018
0.0328
0.0560
0.0781


0.2144

-0.0909

-0.5430

0.0028

-0.0547


0.0326
0.0733
0.0985
0.0189
0.0189
0.0205
0.0189
0.0189

0.2184

0.2184

0.2337

0.2159

0.2159


Transformations: Reciprocal of Intensity.
Normality Test: Passed.
Homoscedasticity Test: Failed.








72

accounted for a significant amount of the variance (Table 2-10). Scheffe analysis

found that pre-lesion values were significantly less than post-lesion peak amplitude

levels at 21-35 days, 70-90 days, 140-165 days, and >210 days (P < 0.05).

Latency at peak reflex amplitude was observed to increase over time after

the lesion for the individual animals (Figure 2-14). The multiple linear regression

analysis (Table 2-11) for latency at peak reflex amplitude accounted for over 50

percent of the variance and was significant (r2 = 0.537; F = 12.1, P < 0.0001).

Interactions between stimulus intensity and time after the lesion were generally not

significant (P > 0.05; except for the period of >210 days, P = 0.0067). The main

findings of the multiple linear regression analysis were: (1) latency at peak reflex

amplitude significantly increased with increases in stimulus intensity (t = -3.736, P

= 0.0003), and (2) the grouping of latency at peak reflex amplitude values into post-

lesion time periods accounted for a significant amount of the variance (Table 2-11).

Post hoc analysis found that pre-lesion values were significantly less than the mean

peak reflex latencies at 70-90 days, 140-165 days, and >210 days (Scheffe, P <

0.05).

Changes over time for several measures of reflex duration are shown in

Figure 2-15. Multiple linear regression analyses were performed on the measures

of half maximal reflex duration (Table 2-12, R2 = 0.539, F = 12.3, P < 0.0001) and

quarter maximal reflex duration (Table 2-13, R2 = 0.599, F = 15.7, P < 0.0001). For

half maximal reflex duration, interactions between stimulus intensity and duration

for different times after the lesion were generally not significant (P > 0.05; except










Table 2-11. Summary of the multiple linear regression analysis of changes in
latency at peak reflex amplitude over time after sacrocaudal transaction.

Latency at peak reflex amplitude
Source DF SS F P R2
Regression 15 6.17
Residual 122 5.31
Total 137 11.48 12.1 <0.0001 0.537


Variable t Equation Standard
Coefficient Error
Constant 1.408 0.1615 0.9063 0.6435
Subject 5.668 <0.0001 0.8200 0.1447
Intensity -3.736 0.0003 -0.7634 0.2043
7-14 days -1.813 0.0723 -0.0701 0.0386
21-35 days 0.269 0.7887 0.0104 0.0386
70-90 days 1.913 0.0580 0.0795 0.0416
140-165 days 3.005 0.0032 0.1161 0.0386
210 days plus 4.299 <0.0001 0.1689 0.0393
Intensity X1.666 0.0981 0.7452 0.4472
7-14 days
Intensity X
25day X0.867 0.3875 0.3879 0.4472
21-35 days
Intensity X
7ntenidtys -1.304 0.1947 -0.6240 0.4786
70-90 days
Intensity X
-1.380 0.1700 -0.6103 0.4422
140-165 days
Intensity X
Itotays -2.757 0.0067 -1.2758 0.4628
> 210 days
Transformations: Reciprocal of Intensity; natural log of the dependent variable.
Normality Test: Passed.
Homoscedasticity Test: Failed.


























Figure 2-15. The time course of changes in reflex duration. Half maximal (top) and
quarter maximal (bottom) durations for each of 5 animals (symbols) are shown at
6 time periods, averaged across intensities. The columns represent the average
half maximal and quarter maximal durations determined across animals. A post hoc
Scheffe analysis found that pre-lesion half maximal durations were significantly less
than the values at 70-90, 140-165 and >210 days (P < 0.05). Post hoc Scheffe
analysis found that pre-lesion quarter maximal durations were not significantly
different from the 7-14 and 21-35 day values (P > 0.05), but quarter maximal
durations at these 3 time periods were significantly less than at 70-90 days, 140-
165 days, and >210 days (P < 0.05).










1400

1200

1000

800

600

400

200

0




3500


3000

2500


2000


1500


1000

500

0


Prelesion 7-14 21-35 70-90

Time Period (days)


140-165 >210


Prelesion 7-14 21-35 70-90 140-165 >210
Time Period (days)


*






0 0
0




rf'li lnfliWD_


*
*
*








0

0
0
-1.Q.












Table 2-12. Summary of the multiple linear regression analysis of changes over
time in the duration of decay to 0.5 maximal amplitude after sacrocaudal
transaction.

Duration at Half Maximal
Source DF SS F P R2
Regression 15 0.000320
Residual 123 0.000274
Total 138 0.000594 12.3 <0.0001 0.539



Variable t P Equation Standard
Coefficient Error
Constant 3.741 0.0003 0.0015 0.0004
Subject 5.906 <0.0001 0.4847 0.0821
Intensity 6.549 <0.0001 0.0057 0.0009
7-14 days 2.136 0.0346 0.0006 0.0003
21-35 days 0.523 0.6019 0.0001 0.0003
70-90 days -1.833 0.0692 -0.0005 0.0003
140-165 days -2.954 0.0037 -0.0080 0.0003
210 days plus -3.999 0.0001 -0.0011 0.0003
Inten7-14 days -0.943 0.3475 -0.0018 0.0019

Intensity X
21-35 days -1.102 0.2724 -0.0021 0.0019
Intensity X
70-90 days 2.276 0.0245 0.0047 0.0021
Intensity X
140-165 days 1.002 0.3180 0.0019 0.0019
Intensity X 0.0019
> 210 days -0.160 0.8733 -0.0003 0.0019
Transformations: Reciprocal of Intensity-1; reciprocal of the dependent variable.
Normality Test: Passed.
Homoscedasticity Test: Failed.












Table 2-13. Summary of the multiple linear regression analysis of changes over
time in the duration to decay to 0.25 maximal amplitude after sacrocaudal
transaction.

Duration at Quarter Maximal
Source DF SS F P R2
Regression 15 55.0
Residual 123 36.8
Total 138 91.8 15.7 <0.0001 0.599



Variable t P Equation Standard
Coefficient Error
Constant 4.069 <0.0001 2.100 0.5160
Subject 8.743 <0.0001 0.698 0.0798
Intensity -5.033 <0.0001 -1.603 0.3185
7-14 days -3.359 0.0010 -0.340 0.1013
21-35 days -3.303 0.0012 -0.334 0.1013
70-90 days 4.191 <0.0001 0.457 0.1091
140-165 days 2.967 0.0036 0.300 0.1012
210 days plus 3.536 0.0006 0.358 0.1012
7-14 Intensitdays 2.093 0.0384 1.470 0.7023

Intensity X
21-35 days 1.788 0.0762 1.256 0.7023
Intensity X
70-90 days -1.917 0.0575 -1.446 0.7543
Intensity X
140-165 days -2.384 0.0186 -1.661 0.6966
Intensity X
> 210 days -0.754 0.4520 -0.526 0.6966
Transformations: Reciprocal of Intensity-1; natural log of the dependent variable.
Normality Test: Passed.
Homoscedasticity Test: Failed.










for the time period of 70-90 days, P = 0.0245). For quarter maximal reflex duration,

some of the interactions between stimulus intensity and duration at different times

after the lesion approached significance (Table 2-13). The main findings of the

multiple regression analyses were that: (1) reflex duration significantly increased

with increases in stimulus intensity (Table 2-12, t = 6.549, P < 0.0001; Table 2-13,

t=-5033, P < 0.0001), and (2) the grouping of reflex durations at most of the post-

lesion time periods accounted for a significant amount of the variance (Tables 2-12,

2-13). For the increasing trend over time in half maximal reflex duration, a post hoc

Scheffe analysis found that pre-lesion values were significantly less than post-

lesion values at 70-90 days, 140-165 days, and >210 days (P < 0.05). For quarter

maximal duration, a post hoc Scheffe analysis found that pre-lesion values were not

significantly different from post-lesion values at 7-14 and 21-35 days (P > 0.05).

The quarter maximal durations at these 3 time periods were significantly less than

the mean quarter maximal reflex durations at 70-90 days, 140-165 days, and >210

days (P < 0.05).

The changes over time in reflex latency and rise time are shown in Figure 2-

16. For reflex latency, the multiple linear regression analysis accounted for over

60 percent of the variance and was significant (Table 2-14; r2 = 0.634; F = 18.2, P

< 0.0001). Interactions between stimulus intensity and latency at different times

after the lesion were not significant. The main findings of the multiple regression

analysis were: (1) reflex latency significantly decreased with increases in stimulus

intensity (t = -8.504, P < 0.0001), and (2) the grouping of reflex latency values into



























Figure 2-16. The time course of changes in reflex latency and rise time. Reflex
latency (top) and rise time (bottom) for 5 animals (symbols) are shown at 6 time
periods, averaged across intensities. The columns represent the average reflex
latencies and reflex rise times determined across animals. A post hoc Scheffe
analysis found that pre-lesion latencies were significantly less than at all post-lesion
time periods (P < 0.05). Pre-lesion rise times were not significantly different from
post-lesion values at any time period after the lesion.









80
25


A A o
20 A A








5-
Q--









0





2.0)

1.8 A

1.6 -
-0 0
...- 1 -

S 1.2-
5*
Prelesion 7-14 21-35 70-90 140-165 >210

Time Period (days)








21.0

10.8 -

10.6

0) 1.4

0 1.2 -
1.0













Time Period (days)










Table 2-14. Summary of the multiple linear regression analysis of changes in
reflex latency after sacrocaudal transaction.

Reflex Latency
Source DF SS F P R2
Regression 15 817
Residual 123 472
Total 138 1289 18.2 <0.0001 0.634


Variable t P Equation Standard
Coefficient Error


Constant 3.104
Subject 8.560
Intensity -8.504
7-14 days -0.104
21-35 days 0.005
70-90 days 1.660
140-165 days 3.393
210 days plus 2.527
Intensity X
7-14 days
Intensity X
21-35 days
Intensity X
70-90 days
Intensity X
140-165 days
Intensity X
> 210 days
Transformations: Log of Intensity.
Normality Test: Failed; Passed.
Homoscedasticity Test: Passed.


0.0024
<0.0001
<0.0001
0.9170
0.9958
0.0994
0.0009
0.0128

0.8544

0.0900

0.1700

0.1627

0.9329


6.3303
0.9445
-7.1937
-0.0379
0.0019
0.6473
1.2295
0.9156


2.040
0.110
0.846
0.363
0.363
0.390
0.362
0.362


0.3441

3.1984

-2.7382

-0.8901

0.1545


1.871

1.872

1.984

1.831

1.831








82

post-lesion time periods accounted for a significant amount of the variance (Table

2-14). A post hoc Scheffe analysis of latency found that pre-lesion values were

significantly less than that observed for all post-lesion time periods (P < 0.05).

Reflex rise time at the different time periods is shown in Figure 2-16. The

multiple linear regression analysis accounted for over 70% of the variance and was

significant (Table 2-15, R2 = 0.74, P < 0.0001). Except for a significant effect of

intensity (P = < 0.0001) and an interaction between intensity and rise time at >210

days (P = 0.0387), no other significant differences were obtained for changes in rise

time.


Stability of the Electrocutaneous Reflex with Repetitive Stimulation

When electrocutaneous stimulation was presented in blocks of trials at a

single intensity, there were instances of wind-down (a decrease) or wind-up (an

increase) of reflex force estimates, when comparing the response to the first

stimulus with subsequent responses. The frequency of wind-down varied in normal

animals, ranging from 10 percent to 36 percent, with a median of 13.9 percent

(Figure 2-17). Post-lesion, wind-down increased for some animals and decreased

for others. For the population of 5 animals there was a slight decrease in the

occurrence of wind-down over the 5 different time periods after the lesion.

In normal animals wind-up occurred less often than wind-down, varying from

0 percent to 14 percent, with a median of 8 percent across subjects (Figure 2-18).

Post-lesion, the frequency of wind-up increased for most animals. Also, as early










Table 2-15. Summary of the multiple linear regression analysis of changes in
reflex rise time after sacrocaudal transaction.

Rise Time
Source DF SS F P R2
Regression 15 1.021
Residual 123 0.361
Total 138 1.382 29.7 <0.0001 0.739



Variable t P Equation Standard
Coefficient Error


Constant
Subject
Intensity
7-14 days
21-35 days
70-90 days
140-165 days
210 days plus
Intensity X
7-14 days
Intensity X
21-35 days
Intensity X
70-90 days
Intensity X
140-165 days
Intensity X
> 210 days


4.632
15.896
-8.828
-0.377
-1.648
0.762
0.446
1.897

-0.662

-1.549

-1.397

1.840

2.089


<0.0001
<0.0001
<0.0001
0.7068
0.1019
0.4473
0.6561
0.0601

0.5091

0.1240

0.1648

0.0681

0.0387


0.0949
0.9712
-0.4633
-0.0038
-0.0165
0.0083
0.0045
0.0190


-0.0767

-0.1795

-0.1733

0.2109

0.2394


0.0205
0.0611
0.0525
0.0100
0.0100
0.0109
0.0100
0.0100


0.1159

0.1159

0.1240

0.1146

0.1146


Transformations: Reciprocal of Intensity; log of the dependent variable+1.
Normality Test: Passed.
Homoscedasticity Test: Passed.




























Figure 2-17. Frequency of wind-down of the reflex with repetitive stimulation. For
the 6 time periods, the graphs show the median probability of occurrence of a
diminished reflex response with repetitive stimulation at one stimulus intensity for
each animal (top) and for the group of 5 subjects (bottom). A decreased frequency
of wind-down was observed following sacrocaudal lesions. The lines represent 75
percent and 25 percent confidence intervals.










0.70


0.60


0.50 Pre-lesion
- DO7to 14
O
a. 0.40 B21 to 35
cO 70 to 90
0.30 El 140 to 165
0M>210
C

0.20


0.10


0.00
Subject 1 Subject 2 Subject 3 Subject 4 Subject 5

Time Period (days)



0.40


0.35


0.30


O 0.25
0

c 0.20
0
0.15


0.10-


0.05


0.00 -
Pre- 7 to 21 to 70 to 140 to >210
lesion 14 35 90 165

Time Period (days)




























Figure 2-18. Frequency of wind-up with repetitive stimulation. For the 6 time
periods, the graphs show the median probability of occurrence of wind-up of the
reflex response with repetitive stimulation at a constant intensity for each subject
(top) and for the group (bottom). The increase in frequency of wind-up developed
progressively over time after sacrocaudal lesions. The lines represent 75 percent
and 25 percent confidence intervals.









87
0.70


0.60


0.50 -
"I 0.50 U| Pre-lesion
CO07 to 14
2 0.40 B21 to 35
C-
. ||M 70 to 90
-6 0.30 B 140 to 165
S>210

0.20


0.10 .


0.00
Subject 1 Subject 2 Subject 3 Subject 4 Subject 5
Time Period (days)


0.50

0.45 -

0.40

>. 0.35

c0 0.30

1. 0.25 -

0.20

0.15



0.05

0.00 ------
Pre- 7 to 21 to 70 to 140 to >210
lesion 14 35 90 165
Time Period (days)








88

as 70 to 90 days post-lesion, a new form of wind-up was observed, involving

increases in reflex force over a number of stimulus presentations (Figure 2-19).

This contrasted with the normal pattern, involving stable reflex force values after the

first response. The percent change in the magnitude of reflex force during wind-up

were similar pre- and post-lesion but varied across the 5 post-lesion periods (Figure

2-19). The percent change in the magnitude of the reflex force during wind-down

was generally reduced post-lesion (Figure 2-20).

Post-lesion responses of the tail to trains of low intensity electrocutaneous

stimulation (50 Hz for 100 msec) sometimes included a late response at latencies

on the order of 400 to 500 ms (Figure 2-21). The late onset reflex was observed

in half (3/6) of the animals tested, with an average frequency of 26 percent. When

the late reflex occurred, it was present on the first few trials of a block and then

dissipated. At high stimulus intensities, the responses to trains of stimulation

occurred at latencies similar to those observed for a single pulse (Figure 2-21).


Reflex Characteristics in Animals with Variants of the Standard Sacrocaudal Lesion

Changes in properties of the electrocutaneous reflex were examined for

lesions other than the intended sacrocaudal transaction. A decrease in reflex force

was observed for animals that received either a dorsal hemisection at S3, or a S3

transaction with evidence for an ischemic episode, or a transaction at rostral Ca3

(Figure 2-22). This effect was opposite to the post-lesion increase in reflex force

observed for other animals with sacrocaudal transactions.




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E72TJCD7D_MW8ZKZ INGEST_TIME 2012-02-20T22:06:27Z PACKAGE AA00009023_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES