The Effect of nail bed compression and supraorbital pressure on selected physiological and motor responses in unconsciou...

MISSING IMAGE

Material Information

Title:
The Effect of nail bed compression and supraorbital pressure on selected physiological and motor responses in unconscious patients
Physical Description:
ix, 129 leaves : ; 29 cm.
Language:
English
Creator:
Aragon, Elizabeth Dale
Publication Date:

Subjects

Subjects / Keywords:
Nursing Research   ( mesh )
Pain   ( mesh )
Physiological Processes   ( mesh )
Unconsciousness   ( mesh )
Physical Stimulation   ( mesh )
Brain Injuries   ( mesh )
Nursing thesis, Ph.D   ( lcsh )
Dissertations, Academic -- Nursing -- UF   ( lcsh )
Dissertations, Academic -- College of Nursing -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1998.
Bibliography:
Bibliography: leaves 113-119.
Statement of Responsibility:
by Elizabeth Dale Aragon.
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 - 002370748
notis - ALX5433
oclc - 52006952
System ID:
AA00004706:00001

Full Text










THE EFFECT OF NAIL BED COMPRESSION AND SUPRAORBITAL
PRESSURE ON SELECTED PHYSIOLOGICAL AND MOTOR RESPONSES IN
UNCONSCIOUS PATIENTS











By

ELIZABETH DALE ARAGON


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


1998



























Copyright 1998

by

Elizabeth Dale Aragon











ACKNOWLEDGMENTS

I want to thank the members of my supervisory committee for all of their support

throughout this study. I wish to particularly express gratitude to Dr. Kathleen Smyth,

committee chairperson, for her encouragement and assistance during my doctoral education

and completion of this research. I am grateful to Dr. Lois Malasanos for her mentorship to me

in my professional career and physiological expertise. I am appreciative to Dr. Hossein

Yarandi for his contribution to the statistical analysis used in this study. I wish to thank Charles

Wood, who encouraged me to pursue a foundation in the physiological sciences. Finally, I

want to thank Dr. Donna Treloar for contributing her clinical knowledge of the care of

critically ill neurological patients. Without their support, completion of this dissertation would

be impossible.

I want to personally acknowledge Norma Shoemaker and the Society of Critical Care

Medicine for providing grant funding for this study. I also want to thank Ardith Lange and

Hewlett Packard for donating the computer and bedside monitor used for data collection.

I want to acknowledge the families who placed their loved one in this study. By doing so,

they have unselfishly contributed to the knowledge needed to care for those with severe brain

injury.

Finally, I wish to thank my family, friends, and particularly Robert, for all the support and

patience with me through the years required to complete my doctoral education.










TABLE OF CONTENTS



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

LIST O F TA BLE S.................................................................................................... iv

A B STR A C T ............................................................................................................. viii

CHAPTERS

I IN TRO D U CTION ........................ ....................... ........................................... 1

Background of the Problem.................................................................................... 1
Purpose of the Study............................................................................................... 7
Conceptual Fram ew ork............................. ............ ..................................... .................. 8
Clinical Significance of the Study.......................... ........................................ 12
R research H ypotheses.................................... .................................................... 13
Variables................... .... ........... ............................................................... 14
Definition of Terms.............................. .............................................................. 14
Assumptions............................ ................................................................ 16

II REVIEW OF LITERATURE................................................................................. 17

Cerebral Perfusion and Metabolism...................................................................... 17
Significance of Intracranial Pressure and Cerebral Perfusion Pressure.................... 20
Clinical Monitoring ofICP and CPP..................................................................... 24
Effects of Clinical Interventions and Factors on ICP and CPP.................................... 28
Physical Examination of the Unconscious Patient.................................... ........... 39
Sum m ary.......................... ....................................................................... ............. 50

Ill MATERIALS AND METHODS......................................................................... 51

Research D esign..................................................................................................... 51
Setting........................... ... ....... ................. ..... ....... ... .................... 52
Sam ple....... ... ........................... .. .... .... .............................. 53
Human Subjects................. ........................................................... 55
M easures.. .. ..................................................... ....................................... 56
Procedures.......................... .... .............................................................. 68
R isk s................................................................................................. ......... 7 0








IV R E SU L T S................................................................ ............................................... 72

Sam ple Characteristics.............................................................................................. 72
Tests of the Research Hypotheses..................................................................... 74
Measurement ofNBC and SOP Pressures............................................................ 97

V SUMMARY AND CONCLUSIONS....................................................................... 100

Discussion of Findings.... ...................................................................................... 101
Conclusions....................................... ..................................................................... 106
Recommendations for Future Research............. .............................................................. 107
Implications for Nursing Practice............................................. ............................ 111

REFEREN CE S. ....................................................................................................... 113

APPENDICES

A PILOT STUDY QUANTIFICATION OF FORCE MEASUREMENTS OF
SIMULATED NAIL BED COMPRESSION AND SUPRAORBITAL
PR E SSU R E ........................................ ...... ............................................................. 120

B DATA SOURCES AND SCALING USED FOR DATA ACQUISITION.......... 122

C DEMOGRAPHIC AND CLINICAL DATA COLLECTION TOOL...................... 123

D INFORMED CONSENT FORM.............................................. ............................ 125

BIOGRAPHICAL SKETCH ......................................................................................... 129














LIST OF TABLES


Table e

2.1 The Glasgow Coma Scale...................................................................................... 42

4.1 Concurrent Brain Injuries in Subjects with Closed Head Injury.................................... 74

4.2 Descriptive Statistics of ICP Changes with NBC.................... ......... ........... .. 76

4.3 Descriptive Statistics of ICP Changes with SOP................................................ ... 77

4.4 Descriptive Statistics of MAP Changes with NBC................................. ............. 84

4.5 Descriptive Statistics of MAP Changes with SOP................................... ............ 85

4.6 Statistics on MAP Changes From Baseline with NBC............................... ............ 85

4.7 Statistics on MAP Changes From Baseline with SOP............................................ 86

4.8 Analysis of Sequential Differences in MAP Over Time with NBC.............................. 87

4.9 Analysis of Sequential Differences in MAP Over Time with SOP............................... 87

4.10 Descriptive Statistics on CPP Changes with NBC......................................................... 89

4.11 Descriptive Statistics on CPP Changes with SOP.............................. ............ .. 89

4.12 Statistics on CPP Changes From Baseline with NBC........................................... 90

4.13 Statistics on CPP Changes From Baseline with SOP................................ ........... 90

4.14 Analysis of Sequential Differences in CPP Over Time with NBC...................................91

4.15 Analysis of Sequential Differences in CPP Over Time with SOP.................................. 92

4.16 Descriptive Statistics on Heart Rate with NBC..................................... ............ .. 93

4.17 Descriptive Statistics on Heart Rate with SOP...................................... ........... ... 93

vi






Table Page

4.18 Statistics on Heart Rate Changes From Baseline with NBC........................................ 94

4.19 Statistics on Heart Rate Changes From Baseline with SOP........................................ 95

4.20 Analysis of Sequential Differences in Heart Rate Over Time with NBC................... 96

4.21 Analysis of Sequential Differences in Heart Rate Over Time with SOP................. 96

A .1 Force M easurem ents..................................................................................................... 120

A .2 Surface Area M easurem ents......................................................................................... 120

A .3 Pressure Calculations in psi........................................................................................... 121

B.1 Channel Selection for Data Sources....................................................................... 122

B .2 Scaling Inputs..................................... ..... ................ ................................ 122














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

THE EFFECT OF NAIL BED COMPRESSION AND SUPRAORBITAL PRESSURE ON
SELECTED PHYSIOLOGICAL AND MOTOR RESPONSES IN UNCONSCIOUS
PATIENTS

By

Elizabeth Dale Aragon

May, 1998

Chairman: Kathleen A. Smyth, Ed.D., R.N., F.AA.N.
Major Department: Nursing

The purpose of this study was to determine if administering two painful stimuli used

to assess motor responses in patients with brain injury, nail bed compression (NBC) and

supraorbital pressure (SOP), had an effect on intracranial pressure (ICP), mean arterial

pressure (MAP), cerebral perfusion pressure (CPP), heart rate (HR), and motor responses,

in unconscious patients. Thirteen unconscious adult male and female subjects with brain

injury were enrolled in the study. All subjects had normal ICP and CPP, and were

hemodynamically stable. After collection of baseline values of the dependent variables,

NBC and SOP were delivered on both sides of the body while ICP, MAP, CPP and HR

were recorded from the bedside monitor. Subjects were recorded on video tape for motor

responses to NBC and SOP and later given a motor score on the Glasgow Coma Scale.








Pressures used to administer NBC and SOP were measured in psi by devices constructed

for purposes of this study.

Data were analyzed using MANOVA statistical procedures to detect differences

within subjects on all physiological values from baseline. Findings from this study were

that NBC and SOP administered on both sides of the body resulted in statistically

significant increases in ICP, MAP, CPP, and HR from baseline for a brief, yet unsustained

time period (p = < 0.05). The ICP had returned to baseline in 30 seconds, the MAP and

CPP within four minutes, and HR by the second minute after administration of NBC and

SOP. ANOVA statistical procedure was used to detect differences in motor scores when

NBC and SOP were given. There were no statistical differences between motor scores on

the GCS with NBC or SOP. The mean pressures that were measured on NBC and SOP

were 77.1130.98 and 86.18.54 psi, respectively.

These data suggest that NBC and SOP do have an effect on physiological indices of

cerebral perfusion by an increasing ICP, MAP, CPP, and HR for a brief period of time but

return to baseline quickly. Therefore, administering painful stimuli to evaluate motor

responses in unconscious patients who have normal ICP and CPP is probably safe. Also,

the data suggest that NBC and SOP produce similar motor responses and could both be

used to assess unconscious patients.














CHAPTER I
INTRODUCTION

Background of the Problem

The clinical management of unconscious patients who have incurred an acute, severe brain

injury is an important challenge. Brain injury is associated with high morbidity, mortality, and

costs. Additionally, the long term effects of severe brain injury on victims and their families can

involve extraordinary emotional and economic tolls, necessitating lifestyle changes (Coburn,

1992). The primary goals of clinical management for those with acute brain injury are to

preserve brain viability and brain function and prevent the development of secondary

complications (Walleck, 1992). Accomplishment of these goals serves to improve short and

long term outcomes of patients who have suffered an acute brain insult.

Normally, the brain requires a continuous supply of oxygen and nutrients for survival

(Jennett & Teasdale, 1981). Threats to sufficient oxygenation and nutrition to the brain

can result in ischemia, disordered metabolism, or death to that tissue within minutes.

When brain integrity is threatened, clinical changes may occur rapidly and without warning

in patients with severe neurological injury. Observation for these clinical signs affords the

opportunity for rapid clinical intervention to combat hazardous conditions for the injured

brain. Early detection of clinical signs indicating potential pathological changes in the

brain are preferred, so that treatment can be implemented in a timely fashion. Often,

appropriate treatment must be started within minutes of detection of abnormalities to

1








preserve neuronal integrity. Precise and well-timed measurement of clinical indicators of

brain function is essential. The accuracy in measurement of key clinical indicators by

clinicians is paramount for early discovery and intervention for clinical problems, reduction

of secondary complications, better outcomes, and minimized costs. Since timing and

accuracy of measurement is important in the clinical management of these patients, it is

important for clinicians to use tools that are sensitive to subtle changes in brain function

over time.

Clinical assessment of persons with severe brain injury is somewhat limited when

compared to other hemodynamic monitoring available for other critically ill patients. Although

new methods to evaluate brain metabolism and oxygenation are evolving, at present the

majority of patients with severe brain injury are evaluated by means of physical examination

and intracranial pressure monitoring. Currently estimation ofintracranial blood flow is

monitored by means of calculating cerebral perfusion pressures derived from arterial and

intracranial pressure (ICP) measurement. Accurate determination of these pressures is

important for making sound clinical decisions and modifications in treatment. Owing to the

complexity and dynamics of human physiology, rarely is one single clinical measurement used

to assess patients with severe head injury. More often, a number of clinical indicators are

considered and weighed in making clinical decisions to achieve positive outcomes, such as a

high level of functional recovery.

Intracranial pressure (ICP) is the pressure exerted within the intracranial vault by its

structures. The intracranial vault is comprised of a rigid and non-compliant skull and is

occupied by the brain, blood volume, and cerebrospinal fluid. Rises in intracranial pressure

(ICP) can encroach upon intracranial structures and create resistance to cerebral perfusion








from the arterial circulation. A major goal of care in individuals with severe brain injury is to

monitor and maintain adequate cerebral perfusion and intracranial pressure (ICP) conducive to

functional neuronal survival and preventing secondary injury (Walleck, 1992). Understanding

which factors affect ICP and cerebral perfusion is important in the clinical management of these

patients.

Cerebral perfusion pressure (CPP) is a clinical measure of blood pressure supplied to the

brain. Since ICP creates a resistive force within the intracranial vault, the mean arterial blood

pressure must sufficiently overcome that resistance to perfuse the brain. Adequate cerebral

perfusion is required to supply necessary oxygen and nutrients for brain function and survival.

Usually, arterial pressure is measured concurrently with ICP monitoring, allowing for the

opportunity to measure cerebral perfusion pressure. Clinically, CPP is estimated by subtracting

the ICP from the mean arterial pressure. Since patient outcomes are associated with changes in

ICP and CPP values (Sheilds & McGraw, 1986; Jennett & Teasdale, 1981), it is appropriate to

monitor ICP and CPP in unconscious individuals with severe head injury.

Heart rate is a clinical measurement of the number of heart beats which occur over a one

minute period. The heart rate and arterial blood pressure of vascular volume are the

controlling factors in supplying blood flow to the brain. An increase in heart rate and stroke

volume from the left ventricle of the heart augments cardiac output of blood supply to the

brain. One physiological response to painful or threatening stimuli is an increase in heart rate

and blood pressure (Clochesy, Breu, Cardin, Rudy, & Whittaker, 1993). When an individual

senses pain, an increase in heart rate and blood pressure may be evident.

Physical examination is the other common type of clinical observation used to evaluate

patients with brain injury. Measurement by physical examination is less quantifiable than








physiological parameters and its accuracy is dependent upon the clinician's knowledge and

skill. In contrast to ICP and arterial pressure monitoring, physical examination is performed

intermittently as dictated by hospital standard or at the discretion of the clinician. Thus,

information gained by means of physical assessment is potentially more variable and less

available than are physiological parameters.

The physical examination is an important tool used to evaluate a patient's condition

following head injury. Changes in the condition of patients with head injury are strongly

associated with pupillary reaction, eye movement, and motor abilities (Mamelak, Pitts, &

Damron, 1996). One advantage of physical assessment is that it is an inexpensive and

noninvasive method to evaluate patients.

Although useful clinically, intracranial and arterial pressure monitoring is invasive, costly,

and poses some risks to the patient. The potential and actual risks associated with physical

assessment of unconscious patients are unknown. The reliability of physical assessment is

dependent on the skills of the clinician conducting the examination. Therefore, physical

assessment findings are only as accurate as the expertise of measurement and interpretation.

Information regarding physical neurological evaluation of patients with acute head injury is

routinely described in textbooks and other available literature. Most often, this information

describes what should be done to evaluate patients, but less emphasis is placed on how this is

to be accomplished. Generally, most clinicians learn how to conduct physical examination of

the unconscious patient with brain injury through demonstration by clinical preceptors. The

assessment methods taught by these clinical preceptors are variable and lack standardization in

many clinical settings, which results in inconsistency in the physical examination and lack of






5

precise measurement. In order to achieve greater accuracy of physical assessment measures, it

is imperative that the most reliable and accurate methods are employed.

There are several clinical indicators of brain function that are amenable to assessment by

physical examination. The level of consciousness (LOC), movement (motor) and sensation

(sensory) ability, and cognitive abilities are significant indicators of brain function. Thus,

important clinical assessment of individuals with brain injury includes evaluation of level of

consciousness, orientation level, motor, and sensory abilities. Patients who have acute, and

severe, brain injury are sometimes rendered unconscious. Once a patient with brain injury is

unconscious, evaluation of their orientation and sensory determination is impractical.

Additionally, when one is unconscious, evaluation of movement abilities becomes difficult

because the patient is incapable of following commands.

Many brain-injured individuals demonstrate abnormal movement patterns. In unconscious

individuals, assessment of these movement (motor) responses becomes even more important,

since LOC and cognition are impaired. Along with pupillary responses, motor responses are

important indicators of outcome and survival in the unconscious patient (Mamelak et al.,

1996).

Once a patient demonstrates a reduced level of consciousness and is incapable of

following verbal commands, a more intense stimulus is typically required to elicit a motor

response. Once verbal commands and light physical stimulation have failed to arouse a patient,

a clinician ordinarily administers a painful stimulus of some type and observes for a motor

response. The types of motor responses to these painful stimuli yield important information

about brain and nervous system function. Methods employed by clinicians when delivering

these stimuli are neither standardized nor measured. These methods vary from person to






6

person, depending on personal preference, education, institutional influence, and individual

technique (Proehl, 1992).

Patients who are in a state of coma due to an acute neurological injury are typically

assessed for the level of coma through serial neurological examinations conducted by clinicians.

The Glasgow Coma Scale (GCS), developed in 1974, is a standard clinical tool used for serial

evaluation of neurological status in critically ill patients. This grading scale has a total of 15

points and evaluates three distinctive areas of neurological function: level of consciousness,

orientation, and motor response. In the unconscious individual, motor responses become more

important. The GCS includes a grading scale for motor responses to painful stimuli.

There have been studies reported in the literature regarding the effects of various stimuli

or clinical factors on intracranial and cerebral perfusion pressures. These studies followed

earlier research reports of ICP changes which occurred during routine nursing care procedures

(Lundberg, 1960). Later studies have been have conducted on the effects of various nursing

interventions and environmental stimuli on ICP and CPP, such as endotracheal suctioning,

body positions, family visitation, conversation, and hygiene maneuvers (Boortz-Marx, 1985;

Hendrickson, 1987; March, Mitchell, Grady, & Winn, 1990; Mitchell & Mauss, 1978; Rudy,

Turner, Baun, Parsons, Smith-Peard, & Page, 1985; Stone, & Brucia, 1991). Findings from

these studies have been instrumental in changing nursing practice for patients with severe head

injuries. Although the effects of various noxious stimuli are anecdotally reported, a study was

not found in the literature that specifically examined the effects of painful stimuli used in

performing motor assessment in unconscious patients on ICP and cerebral perfusion pressure.

One of the main goals in the clinical management of patients with severe brain injury is to

prevent secondary brain injury by maintaining a low ICP and adequate cerebral perfusion








(Walleck, 1992). It is clinically relevant to determine which activities are safe to perform on

patients opposed to those which may result in intracranial hypertension and reduced perfusion

pressures. Although useful clinically, the physiological effects of performing painful stimuli to

determine motor abilities on ICP and CPP in patients with severe brain injury are unknown.

In summary, ICP and CPP monitoring, along with serial physical assessment, are

clinically used for early detection of acute neurological events that may require prompt

and specific intervention for individuals who have sustained a brain injury. Accuracy of

these assessment techniques for the unconscious patient is crucial, since findings from

them are often used to make important clinical decisions. Identification of those clinical

factors that may influence ICP and CPP is important. Standardization of methods used to

assess motor responses in comatose individuals is essential if accurate interpretation and

effective interventions are to be implemented to promote maintenance of neuronal

integrity and survival.

Purpose of the Study

Selected physiological measurements were recorded to determine responses of

unconscious patients with brain injury to painful stimuli. The purposes of this study were to

determine the effects of delivering two different painful stimuli on ICP, CPP, mean arterial

pressure (MAP), heart rate (HR), and motor responses. The two types of painful stimuli that

were employed for this study were nail bed compression (NBC) and supraorbital pressure

(SOP) on both sides of the body. Another purpose of this study was to quantify the average

forces used when administering SOP and NBC as painful stimuli.








Conceptual Framework

The conceptual foundation for this study was a combination of several models. The

primary conceptual framework used as an underpinning structure for responses to stimuli was

Roy's Adaptation Model (1984). As a supplemental substructure, Monro's (1783) classic

physiological proposition regarding intracranial pressure dynamics is introduced to illustrate

principles inherent to cerebral dynamics. Finally, an integrated model designed by the

investigator is presented in order to show how the concepts of Roy and Monro illustrate the

"balance of cerebral perfusion."

The nursing conceptual framework chosen for this study was Roy's Adaptation Model

(1984). In this model, humans are continuously being subjected to internal and external stimuli

to which they are constantly adapting through various processes. The person either positively

or negatively adapts to stimuli using these processes. When these processes result in a positive

outcome, this is seen as adaptation. When coping processes result in a negative outcome, the

result is viewed as ineffective adaptation.

The process by which effective or ineffective adaptation takes place starts with the

adaptation level of the individual. The adaptation level at the time of the stimulus encounter

affects the capacity to adapt to it. This initial adaptation level is a result of feedback

mechanisms from adaptation to previous stimuli. Once stimulated, the regulator or cognator

adapting processes are activated. The regulator process is primarily neurochemical and

physiological in nature, while the cognator process involves reasoning and thought. The effects

of initiating the regulator and cognator processes are demonstrated in various responses.

Regulator effectors are seen in physiological responses, while cognitive effectors are concerned

with self concept and role identification. The final outcome of encountering stimuli will either








support the person positively (adaptive response) or negatively (ineffective response). An

encounter with a stimulus results in the person achieving a particular adaptation level. In Roy's

model, the physiological mode of adaptation promotes responses to increase physical integrity.

There are five basic needs for human survival, including activity and rest, nutrition, elimination,

oxygenation, and protection. Five regulator processes for these basic needs are the senses,

fluids and electrolytes, neurological and endocrine functions. The regulator process is primarily

driven by biological functions to maintain homeostasis within the individual (Meleis, 1991).

Roy's Adaptation Model is applicable as a theoretical framework for this study since the

purposes of this investigation involved the effects of applying painful stimuli to unconscious

individuals with head injury. The painful stimuli serve as input to the person who has potential

for initiating the regulator, physiological processes to cope with the stimulus. This results in

effector responses to which the subject will have either an adaptive or ineffective physiological

response (output). Roy's Adaptation Model integrated the variables studied in this research:

stimuli (input) and physiological responses of ICP, MAP, CPP, heart rate, and motor behavior

effectorss) with evaluation of effective or ineffective adaptation (output).

In Roy's model, the clinician is capable of selecting stimuli which influence an individual's

adaptive state and response (Nicoll, 1986). The nurse is a potential change agent in the process

of adaptation. The effectiveness of a nurse's intervention is dependent on the ability to

promote positive, adaptive coping mechanisms and reduce negative, ineffective responses. It is

known that certain stimuli to those with severe brain injury can cause increases in ICP and

threaten neuronal integrity and function (Mitchell & Mauss, 1978; Rising, 1993). It is also

known that other activities and treatments augment physiological function and adaptive

capacity in those individuals. Individualizing care to those with severe brain injury by






10

observing for signs of effective or ineffective adaptation is a meaningful clinical application of

Roy's Adaptation Model.

Mitchell (1986) proposed "decreased intracranial adaptive capacity" as nursing diagnosis

to describe disproportionate increases in ICP in response to a variety of noxious and

nonnoxious stimuli in patients with increased ICP. Rauch, Mitchell, and Tyler (1990)

conducted a study validating two risk factors predisposing one to a decreased adaptive ICP

response to stimuli in children. Applying Roy's Adaptation Model, this illustrates an ineffective

adaptation outcome to stimuli. In Roy's model, human capacity to adapt to stimuli is partially

related to their adaptive capacity at the onset of stimulation. Persons with acute brain injury

may have decreased adaptive capacity to respond to stimuli due to common pathophysiological

sequelae.

In 1783, Monro (Adams & Victor, 1989; Hickey, 1997; Jennett & Teasdale, 1981)

theorized that there was a relationship between three aggregate components within the

intracranial vault. The three components within the intracranial vault include the brain, blood,

and cerebrospinal fluid (CSF). The intracranial vault is a rigid, mostly enclosed structure and

has a limited and finite capacity for volume. As one or more volume components increase,

there must be a concomitant decrease in another volume to sustain equal pressure. For small

increases in a given volume, one or more components may shift to accommodate the expanding

mass. For example, if brain volume is increased by edema, CSF may be shunted caudally

toward the spinal cord to compensate, thereby reducing CSF volume and maintaining low

intracranial pressure. This process represents intracranial compliance. There is a point at

which further compensation cannot be made, and a sharp rise in ICP occurs (Hickey, 1997).

At this point, intracranial compliance is lost. Increases in ICP create resistance to arterial blood








flow to the brain, thereby threatening adequate cerebral perfusion. Clinical application of this

theoretical model is evident in treatment aimed at maintaining low brain volumes and

preserving cerebral perfusion. Performing CSF drainage is a common clinical practice to

reduce CSF volume to accommodate rises in other volumes.

Incorporation of Monro's theoretical model to this study is based upon factors which

influence ICP and their effects on cerebral perfusion pressure. Intracranial pressure is

determined by changes in volumes within the intracranial vault. No change in ICP may

indicate compliance or no effect on intracranial volume. In this study, ICP and CPP were

measured before, during, and after the application of painful stimuli. Changes in their values

may indicate a alteration in cerebral volume and cerebral compliance. An increase in ICP or

decrease in CPP in the person with severe brain injury is a clinical manifestation of ineffective

adaptation. Conversely, no significant changes in ICP and CPP may indicate effective

adaptation.

This investigator introduces the concept of" the balance of cerebral perfusion" to combine

Roy's and Monro's models. Physiological responses to stimuli are not related to a single

mechanism. Cerebral perfusion is the net result ofmultifactoral influences working

simultaneously and continuously (Jennett & Teasdale, 1981). The resultant ICP and CPP of a

patient with brain injury is mostly due to the balance between driving and resisting forces of the

MAP and the intracranial pressure. In addition, it includes factors which affect both of those

forces, for example hemodynamic stability and body position. Conceptually, the author views

the "balance of cerebral perfusion" as a seesaw model with "driving forces of MAP" and its

influential factors on one end and "resisting forces of ICP" on the other end. Painful stimuli

regularly upset this balance since regulators are set into motion to restore cerebral perfusion to








adaptive levels. In this study, painful stimuli are inputs which affect the balance of cerebral

perfusion. Successful and effective adaptation occurs when there is balance and adequate

cerebral perfusion. Ineffective adaptation results in imbalance of cerebral perfusion and

threatens neuronal integrity.

Clinical Significance of the Study

Painful stimuli are administered to unconscious, brain-injured patients to determine their

motor responses, yet the physiological effects of these stimuli are unknown. Findings have

been published on the effect of various stimuli and conditions on ICP and CPP responses;

however, no research which specifically examined motor responses to painful stimuli in

unconscious patients was found. There are suggested methods for application of painful

stimuli to assess motor function in the literature, yet it is not clear which method yields the best

results of neurological function. Additionally, there are differences in motor responses the

patient may exhibit, depending upon the site assessed.

It is of clinical value to determine factors which affect neurological function in patients

with brain injury. Measurement of ICP and CPP are commonly used to evaluate the patient's

response to treatment. Previous studies on environmental factors that affect ICP and CPP have

resulted in establishment of nursing care standards. Patients have varying responses to the

external environment; therefore, certain stimuli that may result in increases in ICP and are

avoided or controlled in the brain-injured patient. One means of controlling stimuli is to

maintain a quiet, darkened environment and reduce noxious stimuli to prevent increases in ICP.

Administering painful stimuli by the clinician to determine neurological status is employed

frequently as part of routine assessment of patients with brain injury. It is important to know

how this assessment affects changes in ICP and cerebral perfusion pressure. It is the only






13

clinical maneuver performed to deliberately cause pain in the unconscious individual. Usually,

emphasis is placed on providing care that avoids rises in ICP. Giving painful stimuli to assess

motor function may cause an increase in ICP, thus may be in conflict with maintaining an

environment with low levels of stimuli.

Consistency in methods used to determine motor responses in unconscious individuals is

important. There are many clinical variations in how a clinician delivers painful stimuli for

assessment of motor responses. Motor response in unconscious patients is one factor used to

determine neurological function in brain-injured patients, along with ICP, CPP measurements,

and pupillary reaction. Physical assessment of responses is routinely observed and documented

over time with the course of the patient's treatment. Results from this assessment are weighed

with diagnostic findings to make clinical decisions about care. Since physical assessment is

important in ascertaining neurological function and treatment response, it is important to

standardize neurological assessment techniques that foster consistency in measurement.

Prompt management of threats to cerebral perfusion is of utmost importance in the care of

patients with acute brain injury. Determining factors which jeopardize or augment cerebral

perfusion is useful in providing a climate conducive to healing and reduction of secondary

injury to the brain. Development of accurate evaluation tools for prompt recognition of threats

to cerebral perfusion may be helpful to the application of prompt and effective treatment in the

patient with brain injury.

Research Hypotheses

The following research hypotheses were tested in this study. Nail bed compression and

supraorbital pressure will result in

(1) a significant increase in intracranial pressure when compared to baseline values.








(2) a significant increase in mean arterial pressure when compared to baseline

values.

(3) a significant increase in cerebral perfusion pressure when compared to baseline

values.

(4) a significant increase in heart rate when compared to baseline values.

(5) a significant difference in motor responses

In this study, NBC and SOP were measured during stimulus delivery and recorded. The

mean pressures used for NBC and SOP were determined.

Variables

The independent variables in this study were nail bed compression (NBC) and supraorbital

pressure (SOP) given on both sides of the body in four separate trials for each patient for up to

a period often seconds or until a motor response was seen. The dependent variables were

intracranial pressure, mean arterial pressure, cerebral perfusion pressure, heart rate, and motor

response score.

Definition of Terms

The following terms are defined for this research:

Painful stimuli, in this study, consisted of two randomly assigned physical maneuvers:

a) finger nail bed compression (NBC) on either side of the body consisting of vertical

compression of one of the subject's finger nail beds with a device that consisted of a hard,

plastic, flat surface, and b) supraorbital pressure (SOP) consisting of upward pressure against

the supraorbital notch on either side of the upper orbital bony rim with the investigator's

thumb. Each stimulus was performed until a motor response was demonstrated by the patient,






15

or until 10 seconds had passed. The stimuli were performed only by the principal investigator

with two hand-held devices used to measure the pressure exerted with each stimulus.

Intracranial pressure (ICP) is a measure of the strain within the intracranial vault relative to

atmospheric pressure (Clochesy et al., 1993; Hickey, 1997). This pressure can be measured via

several means: intraventricular catheters, parenchymal and intraventricular fiberoptic

catheters, subarachnoid bolt, and epidural sensors. Clinically, ICP monitoring is used to

determine effectiveness of therapeutic interventions and clinical management of patients with

severe brain injury. Normal ICP is between 0 and 15mmHg. ICP produces a resisting force to

arterial driving pressures. In this study, ICP was measured by the use of a ventriculostomy

catheter attached to the bedside Hewlett Packard monitor.

Mean arterial pressure (MAP) refers to the average arterial perfusion pressure over a

given cardiac cycle. Mean arterial pressure is derived from direct arterial pressure

measurement. Mean arterial pressure is a calculated value using the formula: systolic +

2(diastolic) pressure divided by three. Normal MAP is between 80 and 95mmHg (Clochesy et

al., 1993). In this study, arterial pressure was monitored by means of the bedside Hewlett

Packard monitoring system.

Cerebral perfusion pressure (CPP) is a calculated value derived by subtracting ICP from

MAP. Cerebral perfusion pressure reflects the net arterial pressure to the brain when driving

forces (MAP) overcome resisting forces (ICP) and is a clinical guide for adequate cerebral

blood flow. In this study, CPP was calculated from the measured ICP and arterial monitoring

readings. Satisfactory CPP is greater than 60-70mmHg. Cerebral perfusion pressures were

calculated from the ICP and arterial blood pressure measurements taken from the bedside

Hewlett Packard monitoring equipment.






16

Heart rate (HR) is the number of heart beats counted over the duration of one minute. In

this study, heart rate was determined by measuring the timed intervals from the monitored

electrocardiographic monitor readings. An increase in HR is often seen as a response to painful

stimuli and is the result of activation of the sympathetic nervous system. In this study, heart

rate measurement was one indicator of physiological responses to painful stimuli.

Acute and severe brain injury as a result of head trauma, stroke, cerebral aneurysm

rupture, or surgical procedure. "Acute" refers to within seven days of admission. Severe

injury refers to patients who have acute brain injury with a Glasgow Coma Score of eight or

less and are rendered unconscious. Unconscious patients are those who cannot be aroused by

strong verbal commands or non-noxious physical stimuli.

Motor responses are the movement responses that a patient displays when subjected to a

painful stimulus. In this study, motor responses included any movement that an unconscious

patient demonstrated when NBC or SOP was administered. In this study, motor responses

were recorded by video taping the subject's movement during application of the stimuli.

Assumptions

(1) Measured ICP and CPP accurately reflect the degree of pressure within the

intracranial compartment and cerebral perfusion, respectively.

(2) Physiological and motor responses are a result of the painful stimuli given.

(3) The patient is an open system, capable of sensing painful stimuli and responding with

effective or ineffective adaptation.

(4) An increase in ICP is a result of an increase in one or more of the three volume

components within the intracranial vault.













CHAPTER II
REVIEW OF THE LITERATURE

The purpose of this chapter is to review pertinent literature related to the clinical

assessment of unconscious patients with severe brain injury. In order to provide a foundational

background as rationale for evaluating the dependent variables in this study, an overview of the

significance of cerebral perfusion and clinical measurement of ICP and CPP is presented. The

next section reviews selected studies conducted on the effect of nursing activities and clinical

conditions on cerebral hemodynamics. The final section provides an background on physical

assessment of the unconscious patient and clinical tools used to evaluate patients with brain

injury.

Cerebral Perfusion and Metabolism

There are several areas of importance in the clinical management of patients with severe

brain injury. Maintenance of cerebral perfusion and oxygenation is crucial when promoting

neuronal survival. Monitoring key indicators of cerebral perfusion is important to evaluate

patient response to treatment and to make clinical decisions. The following includes a

background for this study with regards to cerebral perfusion measurement in the patient with

severe brain injury.

Characteristics of Cerebral Metabolism and Perfusion

The brain is a complex organ with a multitude of important physiological and

psychological functions. Under normal conditions, the brain has a high and stable metabolic

rate and requires a continuous supply of oxygen and glucose to support its function (Jennett &

17






18

Teasdale, 1981; Guyton, 1996). The brain is highly sensitive to reductions in both oxygen and

glucose (Guyton, 1996; Jennett & Teasdale, 1981). Sudden disruptions in energy provisions

result in neuronal failure and structural damage within minutes. A reduction of oxygen can

occur with insufficient cerebral perfusion (ischemia) or oxygen content (hypoxemia). A

physiological adaptation to hypoxemia is an increase blood flow to the brain to maintain

oxygen delivery. The brain can survive hypoxia for several minutes and still recover. The

period of time the brain can be hypoxic and still partially recover is extended when blood flow

to the brain is maintained. Reductions in blood flow rate to brain tissue below 20-25 ml per

minute causes a decreased ATP production and loss of consciousness. Further reductions

below 10 ml per minute result in electrical and metabolic failure with ultimate cell lysis and

death (Jennett & Teasdale, 1981).

A constant supply of both oxygen and glucose is necessary for brain survival. Threats to

oxygenation and perfusion of the brain can result in ischemia or death to that tissue within

minutes. Once brain integrity is threatened, clinical changes may occur rapidly and without

warning in patients with severe brain injury.

The Mechanisms of Cerebral Perfusion

Cerebral perfusion is accomplished by blood flow overcoming the resistance within the

intracranial vault. Blood flow to the brain is provided by pressure within the arterial circuit

from the aorta into the extracranial arteries: the carotid and vertebrobasilar systems. The

resistance within the skull is produced by the intracranial contents and a resultant intracranial

pressure. Arterial pressure must significantly overcome ICP to support adequate cerebral






19

perfusion. An increase in ICP must be followed by an increase in arterial pressure to maintain

stable cerebral perfusion pressure.

Regulation of Cerebral Blood Flow

There are a number of physical circumstances which result in adaptation of the cerebral

vasculature to preserve blood flow. Physiological factors, including oxygen tension, pH,

carbon dioxide levels, neurogenic influences, and metabolic demand, can affect the size of

cerebral blood vessels and blood flow to various degrees (Kontos, 1981; Lassen &

Christensen, 1976).

Another important factor associated with cerebral blood flow is related to the local

distribution of blood flow to the brain by characteristics inherent to cerebral blood vessels.

Cerebral blood vessels have local, myogenic properties with the capacity to constrict or dilate

over various ranges of blood pressures to maintain steady and constant perfusion. This

phenomenon is known as cerebral autoregulation (Guyton, 1996; Jennett & Teasdale, 1981;

Kontos, 1981; Lassen, 1964; Reivich, 1968). The mechanism of autoregulation involves active

vascular responses of vasoconstriction to perfusion pressures rise and vasodilation to perfusion

pressures decrease. This system of autoregulation functions within mean arterial pressure

(MAP) of 60 to 160mmHg in the normal brain. Values outside those pressure ranges result in

passive blood flow with pressure (Guyton, 1996; Jennett & Teasdale, 1981, p.54). Patients

with brain injury often have dysfunction of autoregulation, particularly in localized areas

(Jennett & Teasdale, 1981, p.133). This results in maldistribution of blood flow to brain tissue,

particularly to regions in which it would be most beneficial, such as injured brain tissue.






20

Therefore, one goal in the clinical management of the patient with brain injury is to maintain a

steady and adequate arterial blood pressure within the limits ofautoregulation.

Other factors that may affect cerebral blood flow include influences from external stimuli.

Lassen, et al. (1978) conducted studies to examine cerebral blood flow under different mental

activities. In this study, xenon 133, a radioactive isotope that emits gamma rays, was injected

into the carotid artery and blood flow was measured by scintillation detectors that recorded

washout of the isotope over two minutes time. One finding in the study was there were

changes in blood flow in the cerebral cortex when the subject was subjected to electrical

stimulation. When the stimulation was delivered to the thumb at a moderately painful level,

there was a 20 percent increase in mean hemispheric blood flow and oxygen uptake,

particularly in the frontal lobes. One conclusion of this finding was that moderate pain

appeared to make the brain more conscious. Also, general brain activation increases

electroencephalographic activity and blood flow and this was related to a heightened level of

awareness. Risberg and Ingvar (1967) reported a study, also using radioactive isotopes, in

which the premotor region of the frontal lobe had increased blood flow during intellectual

functions. Aaslid (1987) measured cerebral blood flow velocity during visually evoked stimuli

and found rapid responses in increased velocity during stimulation. In effect, blood flow may

be increased to meet metabolic needs during increases in cerebral activation.

Significance ofIntracranial Pressure and Cerebral Perfusion Pressure

Patients who have sustained a severe brain injury are at serious risk for decreases in

cerebral perfusion and ischemia. The rationale for monitoring clinical indicators of cerebral

perfusion and oxygenation is substantiated by their relevance to neuronal survival. This








monitoring allows for swift and optimal management of patients with brain injury. Thus, the

clinical management of the unconscious patient with brain injury involves continuous

measurement, evaluation, and treatment designed to preserve tissue perfusion and oxygenation.

The Mechanism of Intracranial Pressure

Intracranial pressure (ICP) is the pressure exerted within the intracranial vault by its

structures. The intracranial vault is comprised of a rigid and noncompliant skull and is

occupied by the brain, blood volume, and cerebrospinal fluid. This vault, or potential space,

is likened to "a wobbling lump of fat enclosed inside a rigid box, buoyed up by the

cerebrospinal fluid" (Jennett & Teasdale, 1981, p.59). The normal components of the

intracranial vault are brain tissue, blood vessel volume, and cerebrospinal fluid (CSF)

(Kontos, 1981; Lassen & Christensen, 1976). The brain occupies about 80% of the skull,

followed by 10% CSF and 5-10% blood vessels (cerebral arterial and venous volume).

The brain is comprised of about 75% water, making it neither rigid nor compressible.

The physical properties of the intracranial contents actually contribute to the

phenomenon of increased intracranial pressure. The pressure exerted within the

intracranial vault is owed primarily to the spacial relationships of these volumes. If there is

an increase in one or more volumes, there must be a concomitant decrease in another to

maintain a low intracranial pressure. The most frequent accommodation occurs by

reduction in CSF volume, as it is expressed from the skull into the spinal dural sac, or

drained therapeutically through ventriculostomy catheter (Butterworth & DeWitt, 1989;

Jennett & Teasdale, 1981). Other potential volume reductions may occur as a result of

clinical intervention, such as removal of extracellular fluid from the brain with mannitol or








reductions in cerebral blood volume through therapeutic hyperventilation.

Hyperventilation reduces arterial carbon dioxide and produces vasoconstriction.

Compensation, or effective adaptation, occurs when the ICP remains low as volume shifts

occur. Rises in ICP occur when there is an increase in one volume and an insufficient

reduction of another volume within the vault.

Initially, as one particular volume increases within the intracranial vault, only small

incremental changes in the intracranial pressure occur. This phenomenon represents a

compensatory state as the skull contents shift and low pressure is preserved. As volume

increases without sufficient decreases in other volumes, there is a sharp and exponential

rise in ICP as decompensation occurs. At this point, even very small increases in volume

result in steep rises in intracranial pressure (Butterworth & DeWitt, 1989). The

physiological consequences of these events may include impaired cerebral function,

paralyzed cerebral vasomotor responses, and the development of irreversible brain damage

(Jennett & Teasdale, 1981).

Adjustment of the intracranial contents without significant rises in ICP is expressed as

compliance or adaptive capacity (Kerr & Brucia, 1993; Rauch, Miller, & Tyler, 1990).

Expansion of intracranial volume compartments can result in sharp rises in ICP with loss

of intracranial compliance. Subsequent compression on intracranial structures and

vasculature may result, impeding cerebral perfusion and resulting in neuronal ischemia and

death. Monitoring ICP may yield important clinical information about the degree of

intracranial compliance. Rapid detection of ICP increases can aid in early treatment to

prevent further neuronal deterioration.








Intracranial pressure is influenced by factors affecting the components within the

intracranial vault (Hickey, 1997). Both brain edema and obstruction to CSF outflow can

result in an increase in intracranial pressure. Additionally, other masses, such as

hematomas and tumors, contribute to intracranial volume and pressure. Cerebral blood

volume is partially controlled by autoregulation and cerebrovascular reactivity to various

stimuli, such as pH, vasoactive metabolites, carbon dioxide and oxygen tension. Cerebral

blood volume is partially dependent upon venous outflow through the jugular veins.

Venous outflow is passive and is contingent upon resistance to this outflow into the

central venous system. Venous outflow can be influenced by physical factors, such as

posture, intrathoracic and abdominal pressures, and venous kinks. Peripheral venous tone

can affect cerebral venous pressure through passive pressure resistance to right atrial

inflow, as seen in congestive heart failure and elevated central venous pressure.

Sudden changes in ICP are usually the result of sudden changes in one of the volume

components of the intracranial vault. Cerebrospinal fluid and brain volume most often increase

gradually and allow for compensation in other volumes. Blood volume can acutely increase as

a result of venous outflow obstruction or an increase in the vascular bed size with dilation.

Other sources of volume that acutely affect ICP are blood volume masses from vascular

bleeding and acute cerebral edema. Hematomas can abruptly cause an increase in

intracranial volume, thus producing precipitous rises in ICP and neuronal compromise.

Measurement of the ICP to detect the degree and rate of change aids in determining

cerebral adaptive capacity.








Clinical Monitoring of ICP and CPP

Lundberg published classical studies on ICP measurement and the effect of clinical

interventions on those measurements (Lundberg, 1960). Monitoring ICP is one form of

clinical measurement used in intensive care units for selected patients with brain injury to

evaluate their response to treatment.

There are several methods available to measure ICP in the clinical setting. The types

of ICP monitoring devices are either known for their anatomical location within the

cranium or the method of measurement (Clochesy et al., 1993; Germon, 1994; Williams &

Hanley, 1998). Some devices use fluid- or hydrostatic-coupled systems with external

transducers to measure ICP. The types of devices used with these systems include

intraventricular and subdural catheters, and subarachnoid bolts. Other type systems use

miniature internal transducers placed within the lateral ventricle in the brain, or brain

parenchyma, and fiberoptic systems within the brain parenchyma or lateral ventricle

(Germon, 1994; Hickey, 1997; McQuillan, 1991; Williams & Hanley, 1998). The "gold

standard" for measuring ICP is the intraventricular catheter (Germon, 1994). The

intraventricular catheter (IVC) is the most invasive form of monitoring, but its use is

associated with low risk when inserted and maintained appropriately. The IVC provides

excellent waveform of ICP and allows for therapeutic drainage of CSF to reduce

intracranial pressure and for direct access to obtain CSF for laboratory analysis.

Normal ICP ranges between 0 and 15 mm Hg (Clochesy et al., 1993; Hickey, 1997;

McQuillan, 1991; Williams & Hanley, 1998), although transient increases from 15-20 mm

Hg are clinically insignificant and could be the result of many factors. Pressure rises






25

lasting for more than a few minutes are considered clinically important and require prompt

intervention (Richmond, 1993).

When using a fluid-coupled system with an external transducer, the ICP measurement

is in reference to atmospheric pressure. The zero point for referencing the transducer is

inferred at the anatomical level of the foramen ofMonro (Clochesy et al., 1993; Rosner et

al., 1995). This anatomical reference can be made at the external auditory meatus or

between the lateral canthus of the eye and top of the ear. This leveling point should be

adjusted for the position of the patient's head during monitoring. Miniature internal

transducers and fiberoptic devices are calibrated before insertion only (McQuillan, 1991,

Williams & Hanley, 1998)

The measurement of ICP is considered unrefined data and never should be interpreted

in isolation of systemic hemodynamics (Richmond, 1993). It is clinically important to

determine signs of cerebral blood flow since the brain consumes high amounts of oxygen

and has limited energy stores. When measuring ICP alone, the practitioner only observes the

pressure resulting from the resistive forces within the intracranial vault. Arterial blood pressure

measurement gives an indication of the driving forces of blood flow directed into the cerebral

circulation. In order to estimate the degree of actual perfusion to the brain, one must consider

the net result from these driving and resisting forces. Cerebral perfusion pressure measurement

provides a means to evaluate the pressure produced after taking both ICP and arterial pressure

into consideration. Clinically, cerebral perfusion is measured by subtracting the ICP (resisting

forces) from mean arterial pressure (driving forces) (Andrus, 1991; Germon, 1994; Rosner &

Daughton, 1990; Williams & Hanley, 1998).






26

Normal CPP ranges from 60 to 100mmHg (Germon, 1994), although some investigators

advocate maintaining CPP above 70mmHg (Rosner & Daughton, 1990; Rosner, Rosner, &

Johnson, 1995) to produce better outcomes. Since CPP reflects overall brain perfusion, it has

become an important parameter for monitoring patients with brain injury. As CPP is dependent

upon both ICP and MAP, treatment is focused upon maintaining appropriate arterial pressure

and reducing intracranial pressure. Arterial blood pressure (ABP) drives cerebral blood

flow and must overcome the ICP in order to perfuse the brain. When the ICP is elevated,

more arterial pressure is needed to preserve adequate cerebral perfusion pressure.

The usual reference point for positioning the air-fluid interface of the transducer when

measuring ABP is located at the phlebostatic axis at the mid-chest level or at the location

of the arterial catheter (Clochesy et al., 1993). When measuring CPP, it is important to

capture the measure of pressure entering the intracranial vault. If the ABP transducer is

leveled at the phlebostatic axis, it is measuring pressure from the reference point of heart.

When measuring CPP, this reference point may not be adequate. Since the purpose of CPP

measurement is to determine the blood pressure to the brain, a better reference point for

the transducer when measuring CPP is from the reference point for cerebral circulation.

While monitoring ABP to determine CPP, the reference point for the external transducer

air-fluid interface should be located at the external auditory meatus or lateral canthus of

the eye and the top of the ear (Nates, Niggenmeyer, Anderson, & Tuxen, 1997; Rosner &

Daughton, 1990). This location is also anatomically near the basal cerebral circulation,

where cerebral blood flow is distributed.








Intracranial Hemodynamics and Outcomes

Much of the foundational knowledge basis supporting the clinical management of

individuals with brain injury stems from studies on intracranial pressure and cerebral perfusion

(Jennett & Teasdale, 1981). Following brain injury, ICP is often elevated and impairs cerebral

perfusion. The net result of this combination is ischemic brain damage. Jennett and Teasdale

(1981) conducted a study in which they found that levels of ICP were important in the

outcomes of patients with head injury.

Cerebral ischemia is the most significant event in determining outcome from brain injury

(Rosner, Rosner, & Johnson, 1995). In postmortem analysis of 151 persons with brain injury,

ischemic brain injury was found in 91%, and 83% showed histological evidence of an increased

ICP. Seventy-nine percent of those patients had both lesions, and only 5% had no evidence of

either kind of secondary insult (Jennett & Teasdale, 1981, p. 31). Most studies in the last 10

years have demonstrated that clinical detection and treatment of elevated ICP results in

improved patient outcomes in those with brain injury (Williams & Hanley, 1998). Repeated

ICP elevations can produce secondary and cumulative brain insults, even when the pressure

returns to normal in between. Since ischemia and increased ICP are so prevalent a finding in

fatal brain injury, it is clinically relevant to monitor cerebral perfusion and intracranial pressures.

There are no pragmatic means to monitor cerebral blood flow in the clinical setting.

Sole monitoring of the ICP for the basis of treatment for patients with brain injury has been

found insufficient (Rosner et al.,1995). Treatment for ICP is highly limited, and

approximately 50% of patients who die from brain injury have uncontrollable intracranial

pressure (Rosner et al., 1995).






28

Clinical studies on patients with brain injury have demonstrated that active management of

CPP as the primary therapeutic end point resulted in improved outcomes (Rosner, 1990;

Rosner et al., 1995). These studies indicated that treatment of CPP is superior to traditional

management of ICP. The current goal in managing the care of those with severe brain

injury is to sustain cerebral perfusion and prevent further neuronal injury. This is partially

accomplished by maintaining an adequate driving force of perfusion from arterial

circulation and reducing resistant forces from intracranial contents under pressure. The

ultimate clinical goals are to maintain a low, controlled ICP (minimize resistance), and an

adequate MAP (driving force) to perfuse the brain, and manipulate factors affecting both

(Hickey, 1997; Sullivan, 1990).

Effects of Clinical Interventions and Factors on ICP and CPP

Initial studies conducted on ICP measurement were focused upon correlation of

pathological findings and outcomes. Later, more emphasis was placed upon research

depicting the influence of care activities on intracranial pressure. As the technical capacity

to measure ICP and CPP improved, the ability to more precisely evaluate factors that affect

intracranial dynamics has emerged.

There are a number of clinical studies that have been conducted regarding the effects of

various clinical conditions on ICP and cerebral perfusion pressures. Some studies assess the

effects of general nursing care on ICP, while others evaluate the effect of procedures, body

position, clinical interventions. Knowing the effects that clinical factors have on ICP and CPP

allows for understanding of which interventions are safe or those which pose potential risks to

the patient with brain injury.






29

Lundberg (1960) conducted a hallmark study on the effects of nursing procedures on

ventricular fluid pressure (VFP) and found associated rises with several aspects of nursing

care. He anecdotally reported an association of bodily activity and nursing interventions

and the onset of transient and sustained VFP in those patients with ICP between 15 and 20

mm Hg. Other clinical factors associated with rises in ICP included painful procedures,

manipulation of the patient in bed, emotional upsets, and straining during elimination of

urine and stool. Shallit and Umansky (1977) noted fluctuations in ICP in 21 patients with

brain edema during routine bedside procedures, such as positioning, endotracheal

suctioning, and head rotation. Tsementzis, Harris, and Loizou (1982) reported findings

from a study on the effect of routine nursing care procedures on ICP in patients with

severe head injury. In that study, there was an increase in ICP with endotracheal

suctioning, insertion of an nasogastric tube, and intramuscular injection in 14 out of 33

subjects. The increases in ICP were transient and reversible. In that study, the MAP also

increased with these procedures. Other "nursing care stimuli" were evaluated for an effect

on ICP in 19 out of 33 subjects in that study, but were not found to significantly

contribute to increases in ICP. These activities included pupillary assessment with a bright

light, loud noises, blood pressure measurement, taking a rectal temperature, and skin

hygiene.

Mitchell and Mauss (1978) conducted the first published descriptive nursing study to

identify the relationship of patient-nurse activities on ventricular fluid drainage (VFD).

For the purposes of that study, VFD was an indirect indication of an increasing intracranial

pressure. In that study, the investigators observed for the effect of a wide variety of








nursing activities on VFD in nine patients. In eight out of the nine patients, there were

significant differences (p = < 0.001) in the predicted VFD and actual drainage amount

when certain nursing activities were performed. Nursing activities discovered to affect

VFD were turning the patient in bed, giving the patient the bedpan, flexion of extremities,

and painful procedures, for examples. Other interesting findings included changes in VFD

during conversations about the patient's condition. Four of the nine patients demonstrated

increases in VFD after five or more minutes. Activities generated by the patient also

increased VFD, such as restless and spontaneous movement, coughing, snoring, and

chewing. This study was pivotal in the development of subsequent studies to determine

the effect of patient care activities on intracranial pressure.

Since the Mitchell and Mauss study (1978), a number of nursing studies have been

conducted to evaluate the effect of various nursing interventions on ICP and CPP.

Mitchell, Ozuna, and Lipe (1980) studied the effect of eight nursing activities on the ICP

in 18 patients. These activities included turning the body in four positions, passive range

of motion of the extremities, and rotation of the head. Findings from that study indicated

that turning resulted in an increase in ICP for at least five minutes in 100% of subjects

after one of the four turns, and in 88% of the subjects after a half the turns. Large

increases in ICP were observed in the five subjects for whom head rotation was done, but

there was only a minimal increase in ICP during passive range of motion. Boortz-Marx

(1985) published a study on the effects of 365 occurrences on ICP in four unconscious

patients with severe head injuries. In that study, occurrences were categorized as either

health care activities, patient-initiated activities, and environmental stimuli. The healthcare








activities that were associated with an increase in ICP were endotracheal suctioning,

turning, and head flexion. Decreases in ICP were found with elevating the head of the bed

and repositioning the patient. Patient-initiated activities that resulted in an increase in ICP

were extremity flexion, neck rotation, and coughing. Environmental stimuli, such as

conversation to and around the patient, loud voices, banging the side rails, monitor and

ventilator alarms, and touching activities, did not result in an increase in ICP. Parsons,

Smith-Peard, and Page (1985) studied the effect of hygiene interventions on the

cerebrovascular status in 19 patients with severe head injury. The majority of the subjects

in that study were unconscious. The types of hygiene interventions investigated were oral

care, body hygiene, and urinary bladder catheter care. All three hygiene interventions

were associated with an increase ICP, MAP, CPP, and HR (p < 0.05). The ICP, MAP,

CPP and HR returned to within one minute. Snyder (1983) conducted a descriptive

observational study on the relationship nursing activities on ICP in nine patients. In that

study, respiratory care activities, repositioning, and neurological assessment were

associated with an increase in intracranial pressure. Conversation towards the patient

resulted in an increase in ICP in 33 observations, and when conversation was about the

patient in 21 observations. There were more increases in ICP with concurrent activities.

The ICP was back to baseline by two to nine minutes in those patients after care activities.

Longer increases in ICP were associated with invasive procedures, restless behavior and

posturing by the patient. A study reported by Rising (1993) showed similar findings to

other studies. Endotracheal suctioning and turning both resulted in an increase in ICP,

and bathing was not associated with an increase in intracranial pressure. Some of the






32

studies on the effect of activities on ICP examined the association of cumulative activities

and the time activities are spaced apart. In the Mitchell study (1980) an unexpected

finding was a cumulative increase in ICP when activities were spaced 15 minutes apart, yet

there was no cumulative increase in ICP when procedures were spaced one hour apart.

Bruya (1981) studied the effect of planned rest between nursing activities on intracranial

pressure. With allowing for ten minutes of rest between a variety of patient care activities,

there was no significant difference in ICP. Muwaswes (1984) examined the temporal

pattern of the recovery of ICP to baseline after passive range of motion and turning in

twelve patient with brain injury. In her study, activities were spaced at least 15 minutes

apart and the effect of those activities on ICP and ABP were recorded. The ICP was

found to elevate with nursing activities. There were no significant increases in arterial

blood pressure. Findings from this study demonstrated that patients' recovery time from

activities that increase ICP varied and was associated with the initial degree of increase in

ICP with the activity.

Findings from these studies indicated that many activities performed by the nurse can

result in an increase in ICP in patients with severe brain injury. Performance of these

activities are important for maintenance of the patient care and prevention of

complications. There are other clinical factors that involve nursing care of patients with

brain injury which may affect ICP. These factors include any procedure or condition

which can limit venous outflow from the brain, such as head position and intrathoracic

pressure. Any clinical factor that may result in an increase in ICP is important to the

overall care and monitoring by the nurse. Studies have developed in concentrated areas






33

of patient care activities to more specifically evaluate their effect on intracranial pressure

following these observational and descriptive studies.

The effect of body position on ICP has been reported further in the literature.

Parsons and Wilson (1984) studied the effects of six passive body position changes on

cerebrovascular status of patients with severe closed head injury. All body position

changes resulted in increases in HR, MAP, ICP, and CPP, except for elevation of the head

of the bed. The physiological changes seen were transient and returned to baseline within

one minute after interventions. March, Mitchell, Grady, and Winn (1990), studied the

effect of backrest position on ICP, ABP, CPP, and transcranial doppler flow velocity in

four subjects with brain injury. In that study, subjects were placed in four different head

positions and evaluated for changes in the physiological measurements. Findings from

that study were that responses from subjects were not consistent across individuals over

time. Subjects demonstrated highly individual responses in ICP, CPP, HR and ABP when

placed in different positions. In a study reported by Ropper, O'Rourke, and Kennedy

(1982), 19 patients who had brain injury were evaluated for changes in ICP and

intracranial compliance with elevation of the head of the bed to 60 degrees and a flat head

of bed position. Raising the head of the bed only resulted in a decrease in ICP in one half

of the patients. In contrast, McQuillan (1987) studied the effects of placing the head of

the bed in a Trendelenburg position for postural drainage on the ICP, MAP, CPP and HR

in 20 patients. Findings from that study was that ICP was higher when the patient was

receiving postural drainage therapy (p < 0.05), but returned to baseline more rapidly than

when the position of the head of the bed was flat. A clinical implication gained fromthese






34

studies and others (Simmons, 1997) was that the optimal position for the head should be

established for each individual, depending on the pressures observed when changing the

position of the head.

Another area of study on the effect of clinical factors on ICP is endotracheal

suctioning. Original studies, described above, indicated that endotracheal suctioning

(ETS) resulted in an increase in ICP in many of the subjects. A study was conducted to

determine the method of ETS that resulted in the least compromise to the cerebrovascular

status in adult patients with head injuries (Rudy, Turner, Baun, Stone, & Brucia, 1991).

In that study, patients were assigned to groups in which two protocols were evaluated for

the effect on ICP, MAP, CPP, HR, and oxygen saturation (SaO2). The different

treatments included preoxygenation with 100% and 135% tidal volume, and two versus

three ETS passes. There was a statistically significant increase in ICP, MAP, HR, and

CPP with both hyperoxygenation protocols from baseline (p < 0.001). No significant

differences were found in the Sa02 with either of the two protocols. No significant

differences were found between the two- and three-ETS suction passes, however, all

groups demonstrated cumulative increases in ICP, MAP, and CPP with each consecutive

suction pass. Parsons & Shogan (1984) reported a study on 20 patients in which the

effects of manual hyperventilation with a ventilation bag with 100% oxygen for 20 to 30

seconds and ETS for no more than ten seconds on ICP, MAP, CPP and heart rate.

Findings from that study were that all physiological values increased during the procedures

done in the study. An important finding was that the CPP remained above 70mmHg

during the procedures, indicating that ETS is safe to perform in patients with ICP between








0 and 20mmHg. Metcalfand Mitchell (1987) studied the effects ofpresuctioning and

postsuctioning hyperinflation treatment on ICP, CPP, and MAP in six patients with brain

injury. There was no significant difference in ICP and CPP between the two methods

used. The MAP increased significantly with ETS, thereby increasing the CPP. This

increase in MAP counteracted the effects that ETS might have on cerebral perfusion

pressure. Based on the findings from these and other studies, Kerr, Rudy, Brucia, and

Stone (1993) made recommendations for ETS in patients with head injury.

Recommendations included preoxygenation prior to ETS, limiting suction duration to ten

seconds or less, hyperventilating with caution, maintaining neutral head position during

suctioning, maintaining suction pressure under 120 mm Hg, and using suction catheters

with small diameter ratios.

Other studies have been reported that examine the effect of clinical interventions or

procedures on cerebral hemodynamics. In one study, the effect of using intermittent

pneumatic leg compression on ICP, MAP, CPP, CVP and HR in 24 patients with brain

injury was examined (Davidson, Willms, & Hoffman, 1993). There were no significant

changes in MAP, CVP, ICP, or CPP in those subjects. Stevens and Johnston (1994)

studied the effect of phases of a routine heel stick procedure on ICP, SaO2, and HR in

124 premature infants. Analysis of the results from that study demonstrated a statistically

significant multivariate main effect of phase on the physiological measures when

considered together. A significant increase in HR and ICP occurred during phases of heel

stick and squeezing of the foot (p < 0.0001). Davies, Deakin, and Wilson (1996)

conducted a study to determine whether the use of a rigid collar used to protect the






36

cervical spine in patients with concurrent head injury resulted in an increase in intracranial

pressure. They reported a significant increase in the ICP after application of the collar

(p = < 0.001), followed by a decrease to baseline after the collar was removed. In that

study, the MAP did not change significantly with application of the collar.

Two, recent laboratory animal studies illustrate interventions currently practiced in

the clinical setting that may have an effect on ICP in the human with brain injury. One

study was conducted on pigs on the influence of diagnostic laparoscopy, where air is

administered into the peritoneum for direct visualization for injuries, on intracranial and

cerebral perfusion pressure (Josephs, Este-MacDonald, Birkett, & Hirsch, 1994). This

administration of air was thought to contribute to intraabdominal pressure and

intrathoracic pressure. The findings from that study demonstrated that there was an

increase in ICP from baseline (p = 0.0001) when air was introduced into the abdomen.

When ICP was artificially raised to a higher baseline level using an epidural balloon, there

was a higher increase in the ICP (p = 0.0001). Hiriri, Firlick, Shepard, Cohen, Barie,

Emery, & Ghajar (1993) conducted a study on pigs in which were volume resuscitated for

hypovolemia and evaluated for increases in central venous pressure (CVP) and intracranial

pressure. There was a statistically significant increase in ICP from 20mmHg in pigs not

aggressively volume resuscitated, and 24mmHg in those animals who had undergone

volume resuscitation (p = < 0.05).

In the earlier studies on nursing activities and their effect on ICP described above, there

were reports that certain environmental and conversational factors resulted in an increased

intracranial pressure. These findings were not consistent between different studies. Further








research was published on the effect of family visits and familiar conversation on cerebral

hemodynamics. Johnson and Nikas (1989) investigated the effects of conversation on

intracranial pressure. In that study, eight patients were read conversations of different content

types. One type of conversation was an emotionally referenced conversation designed to

simulate an actual nursing report that would be given at the change of shifts. The second type

of conversation was social in nature, and not related to clinical topics. There were no

significant differences between the ICP when compared with the two groups. An unexpected

finding was that the ICP recording during the socially referenced content decreased. Also

found was that patients who had an increase or decrease in ICP had a GCS of six or greater.

Treloar, Jermier-Nalli, Guin, and Gary (1991) conducted a study on the effect of verbal

stimulation on intracranial pressure. In this study, a message taped of a familiar family voice

and an unfamiliar voice of the researcher was played while ICP was being recorded. Results

from this study concluded that there was no statistically significant differences in ICP when

exposed to these different voice types. Schinner, Chisholm, Grap, Hallinan, and LaVoice-

Hawkins (1995) studied the effects of auditory stimuli on ICP and CPP in 15 unconscious

patients with brain injury. Three types of auditory stimuli, earplugs, a music tape, and a tape

recording of ICU environmental noise, were played to subjects while recording ICP and

cerebral perfusion pressure. The results showed no statistically significant change in ICP or

CPP during the data collection period. Hendrickson (1987) studied the effect of family visits

on ICP in 24 patients. In those subjects, a decrease in ICP was seen during family presence in

all but six patients. In those six patients, an increase in ICP occurred, but were statistically

insignificant. A study was, also, conducted by Prins (1989) on the effect of family visits on






38

ICP. Findings determined there was no increase in ICP associated with family visits. Although

statistically insignificant, there was a decrease in ICP values with family visitation. The

implication of these studies is that family visitation does not adversely affect the ICP in patients

with brain injury and, in certain circumstances, results in a decrease in ICP.

Studies on the effects of tactile stimuli on ICP have been conducted. Walleck (1983,

1990) studied the effect of rubbing the side of the face and hands with a circular motion for

two minutes on ICP. She noted a statistically significant reduction of ICP in 26 out of 30

subjects. Pollack and Goldstein (1981) studied the effects of gentle tactile and auditory

stimulation and found consistent reduction in ICP to below 10mmHg in seven patients with

Reye's syndrome.

Findings from this review of the literature illustrated that patients have variable,

individual responses to diverse types of stimuli. Clinical application of the results from

these studies is limited due to small sample sizes. Also, in most of these studies, subjects

had normal ICP and cerebral perfusion pressure. Information about patients with

increased ICP are not as well studied. More research is needed to verify and explore those

findings. Painful stimuli used to determine the degree of responsiveness of unconscious

patients is a common clinical practice by nurses and other critical care personnel.

Although an important indicator of state of arousal and awareness, little is known about

specific physiological effects of applying painful stimuli to assess unconscious patients

with severe brain injury.








Physical Examination of the Unconscious Patient

In addition to monitoring for ICP and CPP, the physical examination adds other important

information regarding the status of patients with severe brain injury (Hickey, 1997; Kelly,

1998). The following section reviews pertinent literature regarding the assessment of

unconscious patients.

Consciousness

The phenomenon of consciousness is a state of awareness of the self and the

environment. The physiological basis for consciousness is complex and involves states of

arousal and cognitive awareness (Plum & Posner, 1986). Conscious behavior is

dependent upon functional cerebral hemispheres and deeper structures of the upper

brainstem, hypothalamus, and thalamus. The conscious state is influenced by the reticular

activating system (RAS). The RAS is an intricate network of neuronal projections

beginning in the brainstem and distributing widely over the cortical and midbrain

structures (Kelly, 1998; Plum & Posner, 1986; Stewart-Amidei, 1991). The RAS is

stimulated by major and specialized sensory pathways. Sensory inputs from spinothalamic

collateral tracts, responsible for pain and temperature sensory transmission, are numerous

and are probably associated with the arousing capacity of noxious stimuli. Global and

diffuse brain injury causes dysfunction of the reticular activating system, resulting in

altered consciousness.

Coma and Unconsciousness

In contrast to consciousness, coma is the total absence of awareness of self and the

environment, and failure to respond when stimulated externally (Plum & Posner, 1986).









An impaired level of consciousness (LOC) may indicate diffuse, multifocal involvement of

both cerebral hemispheres and brain stem (Kelly, 1998). In those with acute and severe

brain injury, the duration and degree of unconsciousness is associated with the amount of

brain tissue injury involved, rather than a specific focal location. Loss of consciousness

implies that there is widespread cerebral hemispheric or brainstem dysfunction.

A reduced level of consciousness or coma is associated with severe brain dysfunction.

(Plum & Posner, 1986). Unconsciousness is the equivalent of advanced brain failure and

its presence should prompt clinicians for urgent attention. There are many attributable

causes of coma, including severe brain injury, metabolic and diffuse cerebral disorders,

ischemic states, and psychogenic origin. Additionally, there are variations in the degree of

reduced consciousness, such as stupor, obtundation, and coma. Coma is described as a

state in which the person is unarousable and unresponsive to external noxious stimuli or

internal need.

Clinical Evaluation of the Unconscious Patient

The level of consciousness, or degree of sustained coma, is strongly correlated with

patient outcomes (Jennett & Teasdale, 1974). Severe brain injury resulting in coma yields

a mortality rate of approximately fifty percent. Thus, determination of the LOC is

considered an important aspect to appraise in patients with acute brain injury.

Clinical tools have been developed to assess the degree of consciousness and other

important aspects of neurological assessment. These instruments are used by clinicians to

assess neurological function following an acute brain injury. Scores derived from these

tools allow for systematic evaluation of neurological condition and severity of illness









designation. Since brain injury is accompanied to a life-threatening event, the instruments

used to assess neurologic function should be easy to perform in a short period of time,

noninvasive, not harmful, and have the capability to be used and interpreted by different

types of clinicians (Stanczak, White, Gouview, Moehle, Daniel, Novack, & Long, 1984).

In 1974, Jennett and Teasdale (1974) conducted a study on patients with brain injury.

They developed a clinical assessment tool, known as the Glasgow Coma Scale (GCS), to

standardize and quantify the level of consciousness and coma as part of that study. This

clinical tool was instrumental in predicting outcomes in patients with brain injury. Since

then, the GCS has been adopted for use by clinicians to assess patients with severe brain

injury (Jones, 1979; Mason, 1989; Teasdale, 1975; Williams & Hanley, 1998). There are

three areas of assessment included in the GCS. Figure 2.1 outlines the important elements

of the GCS.

The Glasgow Coma Scale is the most widely used scoring system used by clinicians to

evaluate patients with brain injury. The GCS is not always practical when evaluating

neurological status of patients with severe brain injury in the intensive care unit setting

(Hilton, 1991, Kelly, 1998; Knaus, 1994; Sullivan, 1990). Many patients with severe brain

injury are supported on mechanical ventilation, eliminating the ability to grade the level of

verbal ability. Also, it is not uncommon for patients with head injury to have concurrent

orbital edema and injury, rendering their eyelids closed. Assessment of eye opening in

those individuals is precluded by this situation. The accuracy of clinical measures are

dependent upon the precision, reliability, reproducibility, and validity of the tools used,









as well as correct utilization by the clinician (Pollit & Hungler, 1989, Brockopp & Hastings-

Tolsma, 1995, Gift & Soeken, 1988).

Table 2.1

The Glasgow Coma Scale

Categories of Assessment Description of Responses
Eye Opening 4 = Spontaneously open Level of Consciousness
3 = Open to speech
2 = Open to painful stimuli Not valuable if orbital
1= Closed; no eye edema or injury
opening
Best Verbal Response 5 = Oriented Orientation
4 = Confused
3 = Inappropriate words Not valuable if
2 = Incomprehensible sounds endotracheally intubated
1 = No verbal response
Best Motor Response 6 = Obeys commands Motor Responses
5 = Localizes painful stimulus
4 = Flexion withdrawal to Record the best arm
painful stimulus response
3 = Abnormal, spastic flexion;
Flexion posturing to
painful stimulus
2 = Abnormal extension
posturing to painful
stimulus
1 = No motor response
Total GCS Range 3 to 15

Note: Adapted from the Glasgow Coma Scale developed by Jennett and Teasdale (1981).

The GCS has been studied for its reliability in measurement. Initial interrater

reliability studies on the use of the GCS between nurses, junior physicians, and

neurosurgeons revealed a high degree of concordance between examiners (Teasdale,

Knill-Jones, & Van Der Sande, 1978). More recently, the GCS has been reexamined.









Rowley and Fielding (1991) found that there were consistent errors in determining GCS

among inexperienced users of the tool.

Ellis and Cavanagh (1992) found that registered general nurses had a higher rate of

agreement than students in judging GCS following review of a videotape. Janosik and

Fought (1992) conducted a study for inter-rater reliability in scoring the GCS when

performed by registered and licensed practical nurses and physicians after they observed a

videotape. The highest percent of correct agreement was obtained by registered nurses,

followed by physicians (21 74%). Segatore and Way (1992) evaluated the GCS for

psychometric properties and found that it has pragmatic utility, but suffers serious

limitations for clinical monitoring and prediction. In a study conducted by Ingersoll and

Leyden (1987), an educational program was presented to nurses on the GCS with

subsequent evaluation of their performance using the GCS. There was considerable

agreement between expert evaluators and staff nurses who both did and did not attend the

educational session. However, one interesting finding from this study was that as the

complexity of the examination increased, the discrepancy between examiners increased.

The investigators indicated that when examiners are faced with assessment findings that

are difficult to place on the GCS, they may have used other indicators for the level of

consciousness. Marion and Carlier (1994) conducted a descriptive study to examine the

practice of determining the GCS in the early evaluation of patients with brain trauma who

had received drugs which impair mobility. Discrepancies were found among 17 major

neurological centers in the United States in regard to the timing and grading of the GCS

and the personnel who performed the initial and serial Glasgow Coma Scores.









There are other clinical neurological assessment tools available to evaluate the

unconscious patient with brain injury. Stalmark, Stalhammar, & Holmgren (1988)

introduced the Reaction Level Scale (RLS85) as a reliable coma scale to assess for overall

reaction level in patients with acute brain injury. The RLS85 is an eight level single line

scale that evaluates the patient on the degree of consciousness and motor responses to

pain. This scale was developed to eliminate the problems associated with the use of the

GCS, such as with patients who are intubated or have swollen eyelids. The RLS85

(RLS82) was tested for reliability measurement. A Kappa value of 0.69 was observed in a

multicenter study. The tool was also assessed for its validity (Starmark, Stalhammar,

Holmgren, & Rosander, 1988). A Spearman's rank correlation between the GCS sum

score and the RLS85 was -0.94 and the ability for the clinician to assign patients to unique

categories was significantly better for the RLS85 compared to the GCS sum score.

The Comprehensive Level of Consciousness Scale (CLOCS) was developed as an

alternative to the GCS as a clinical measure of consciousness (Stanczak, et al., 1984).

This tool was developed to address the deficiencies found when using the GCS. The

CLOCS is an eight-item scale designed to more comprehensively evaluate the behaviors of

the unconscious patient. The investigators evaluated the use of the CLOCS for

psychometric properties of reliability and validity and compared this tool to the GCS. This

analysis revealed the CLOCS was a more reliable and sensitive instrument than the GCS

for assessing patients with severe impairment of neurological function. With some

adjustment to the scale, the investigators recommended the use of the CLOCS as an

alternative to the GCS to evaluate unconscious patients.









Born (1988) evaluated the use of the Glasgow-Liege Scale (GLS) for its prognostic

value and incorporation of motor responses and brain stem reflexes in patients after severe

head injury. The GLS is identical to the GCS with the addition of a brain stem reflex

score. The purpose of the study was to determine which factors, motor responses and

brain stem reflexes, were predictive for patient outcomes at six months. Born found that

brain stem reflex responses in the first 24 hours had the best prognostic ability for short

periods of time. The motor scores were more prognostic over a longer period of time.

The longer the patient demonstrated a motor response deficit, the more likely they were to

have poorer outcomes. Both motor and brain stem responses were predictive of outcome.

Way and Segatore (1994) developed and evaluated the Neurological Assessment

Instrument (NAI) for evaluation of the patient with severe brain injury. This instrument

was developed as an alternative to the GCS that was a valid and reliable measure of the

level of consciousness (LOC), neurological function, and predictive of outcome. The NAI

consists often items and includes assessment areas of LOC, pupillary and ocular signs,

communication, and orientation, and motor responsitivity. The interrater reliability of the

NAI was high between observers and the NAI scores were strongly correlated to the

Glasgow Coma Score. In addition, the construct validity of the NAI was high. This

preliminary evaluation of the NAI demonstrated it to be a valid, reliable, and feasible

measure of the LOC in patients with brain injury. Crosby and Parsons (1989) assessed a

neurological assessment tool known as the clinical neurologic assessment tool (CNA).

This tool consists of 21 items, and evaluates response to verbal and tactile stimulation,

ability to follow commands, muscle tone, body position, movement, chewing and yawning









in patients with head trauma. The CNA was found to be a highly reliable and valid

instrument. Additional findings in this study were that the voluntary body movement was

the best indicator of the level of consciousness. The investigators recommended this tool

as an alternative to the GCS to evaluate the LOC in the patient with brain injury.

To summarize, although the initial use of the GCS was beneficial in developing

scoring systems to evaluate degree of consciousness in patients with brain injury, it has

been subjected to further study and scrutiny in terms of its overall effectiveness and use in

the current treatment environment (Marion, 1994). Other tools have been developed to

address the deficiencies found in the practical use of the GCS in evaluating unconscious

patients. Studies have demonstrated that these tools are reliable and valid instruments for

neurological assessment, however, the GCS continues to be the most widely used tool to

evaluate unconscious patients.

Stimuli Used to Assess Motor Responses

There are a number of different types of painful stimuli used to assess motor response

in unconscious patients (Proehl, 1992; Starmark & Heath, 1988). A preferred method to

deliver painful stimuli would be one in which produces the least amount of tissue injury

and yields the best information regarding level of consciousness. Jennett & Teasdale

(1981) recommended using pressure against the nail bed using a pencil and supraorbital

pressure as stimuli to assess motor response. Stalmark, Stalhammer, and Holmgren

(1988) recommend nail bed pressure and retromandibular pressure using the index finger

behind the jawbone under in the direction of the chin. They recommended patience in

administration of the painful stimulus until the patient demonstrated a motor response,









since some patients exhibit a delay in motor response to painful stimuli. In one study, the

motor scores elicited by SOP were more liable to disagreement than with fingertip

stimulation (Teasdale, Knill-Jones, & Van der Sande, 1978). The SOP was found to be a

less effective stimulus when compared to responses elicited by nail bed compression. In

this study, the NBC never failed to yield some kind of response in their subjects. In the

Born study on the GLS (1988), specific painful stimuli were used to evaluate motor

responses. They either applied digital pressure of increasing intensity to the medial part of

the supraorbital ridge, followed by facial nerve compressions and progressive pinching of

the thorax. The motor score was derived from the "best" motor response seen from these

stimuli. Born (1988) determined that the most important parameters to determine

neurological status were motor responses and brain stem reflexes.

In the study on the CLOCS neurological evaluation tool (Stanzcak et al., 1984), the

stimuli used to assess the unconscious patient were standardized. The types of noxious

stimuli used were the sternal rub vigorously with the thumb, followed by pressing on the

nail bed of fingers on both hands, and squeezing the webbed tissue between the thumb and

index finger on both hands. The motor response scale was an eight point scale and

included facial grimace and degrees of withdrawal. One goal of the study was to

standardize administration of the stimuli by degree of intensity. The investigators had

well-defined instructions and developed hierarchy of stimuli in successive order.

The motor responses seen with administration of painful stimuli vary. A "normal"

response to painful stimuli consists of localizing the site of pain, or attempting to remove

the stimulus. The next best level of response would be to demonstrate flexion withdrawal









from the painful stimulus. Spastic, or classical flexion and extension posturing are

abnormal responses to painful stimuli. No response to painful stimuli is the worst clinical

finding. It is common for patients to have mixed and varied motor responses to painful

stimuli. The "best motor response" was suggested to use when scoring patients on any

type coma scale as it correlated better with patient outcomes (Teasdale & Jennett, 1974).

There have been few studies conducted in which the type of stimulus used and its

effectiveness in eliciting motor responses were addressed. Starmark and Heath (1988)

evaluated six different types of painful stimuli in patients used to evaluate motor responses

of patients with drug overdose. The six stimuli included earlobe pressure, sternal rub,

trapezial grip, SOP, NBC, and retromandibular pressure. The patients were classified in

terms of whether the response was clear, unclear, and no response. They studied the best

response of the patients to the stimuli. They analyzed data and found that there was a

significant effect of order found for nail bed pressure. Patients had decreased responses

with repeated stimulation. The most effective stimulation in producing a motor response

were sternal rub and retromandibular pressure. The earlobe and SOP were the mildest

techniques. In this study, 26 out of 146 subjects had either bruising from sternal rub,

subungal hematoma, and skin excoriation from the delivery of painful stimuli. They

concluded that different painful stimuli complemented each other. If only one technique

for a scale is used, the patients may be underscored on different scales. The stimulation

hierarchy suggested by the investigators was sternal rubbing, retromandibular pressure,

trapezius grip, nail bed pressure, SOP, and earlobe pressure. They also suggested that if

no response was seen with one stimulus, one should continue with others. Sternal rubbing






49

and nail bed pressure caused hematomas. The stimuli which yielded best motor responses

were sternal rub and retromandibular pressure.

Guin (1997) conducted a study to evaluate the clarity of motor responses when three

different types of painful stimuli were given in 21 subjects with altered levels of

consciousness. The scale for determining clarity was the same as in the Starmark and

Heath study (1988). The stimuli tested in this study were nail bed pressure, sternal rub,

and trapezius squeeze. There were no significant differences between the three types of

stimuli on motor responses on the GCS, nor clarity in the response. However, when

multiple pairwise comparisons were analyzed in the three stimuli types, the rank sum for

sternal rub was significantly greater than trapezius squeeze, and the nail bed rank sum fell

between those two extremes.

Guin used a strain gauge device to measure nail bed pressure, identical to the one

used in this study. When measuring nail bed pressure in her study, Guin found that the

pressures used to administer nail bed stimuli ranged from 10 to 89 psi with a mean of 65 +

19.06 and median of 72 psi in all subjects. Nail bed pressure in patients who were

hemiparetic measured a mean of 70 16.51 psi (median 73 psi), and in non-hemiparetic

patients, 59.6 21.01 psi (median 68 psi). Patients in that study had a GCS from 4 tolO,

with a mean of seven. The time over which the stimuli were delivered was 0.6 to 9.6

seconds.






50

Summary

In summary, it is known that certain stimuli affect physiological measures ofICP and

CPP. Painful stimuli are intentionally administered for routine assessment of unconscious

patients with severe neurological injury. The physiological effects of applying these

painful stimuli to assess motor responses are unknown. There is considerate variability in

the manner that clinicians apply painful stimuli and there is inadequate clinical

standardization for administering painful stimuli. Finally, information about the amount of

force required to elicit motor responses in unconscious individuals is limited.














CHAPTER III
MATERIALS AND METHODS

In this chapter, a description of the research design, study setting, sample selection,

and instrumentation for measurement of the independent research variables is discussed.

Protocols used in this study are included. The risks to human subjects with regard to

participation in the study are presented.

Research Design

This study employed a quasi-experimental repeated measures design. Each patient served

as his/her own control. Patients were selected from a sample of unconscious individuals who

met inclusion criteria and had appropriate consent. Independent variables included in the study

were: finger nail bed compression (NBC) and supraorbital notch pressure (SOP) given on both

sides of the body, unless there was a contraindication to perform the maneuver at that site. A

randomized order for the delivery of each of the four stimuli (NBC and SOP on either side of

the body) was accomplished by the following procedure: each of the four stimuli to be

performed were written on separate pieces of paper and placed in a box. Each potential

sequence of stimulus delivery was determined by drawing the four stimulus types from the box

one at a time and recorded in the order drawn. This predetermined, randomized order was

assigned to subjects as they were enrolled in the study. Dependent, physiological variables

measured or calculated in this study included mean arterial pressure (MAP), intracranial

pressure (ICP), cerebral perfusion pressure (CPP), and heart rate (HR). These values were

51








52

recorded before, during and after administration of the stimulus for a total period equaling five

minutes. Motor responses to the total painful stimuli were recorded on videotape and graded

according to the motor scale portion of the Glasgow Coma Scale. Demographic information

was collected on each subject. (Appendix A) Other important clinical information about the

patient was recorded, including admission and most recent Glasgow Coma Scores, results of

diagnostic computerized tomography of the head, type of brain injury, admission diagnoses,

baseline physiological measurements ofICP, CPP, MAP, and HR, admission and current

pupillary size and reaction, and documentation of their most recent motor response on the

Glasgow Coma Score.

Setting

This study was conducted in an intensive care unit (ICU) at a Level I trauma center in

Orlando, Florida. This setting is typically utilized for patients with severe brain injury.

Equipment for monitoring ICP, MAP, CPP, and HR is readily available. This ICU employs

nursing and ancillary staff appropriately educated and trained in the management of patients

with severe brain injury. The staff are prepared to perform comprehensive neurological

assessment, including the evaluation of motor responses in unconscious patients. The data

were collected in this study in the patient's room. In order to control for environmental

stimuli, such as light and noise, the room was darkened, the doors closed, and other noise

reduced as much as possible at least 15 minutes prior to administration of the painful stimulus.

Employees and family were asked not to enter the room during the procedure and any patient

care activities were completed at least 15 minutes before the experiment.










Sample

A sample of 10 males and three women who had sustained brain injury were recruited for

the study from a group of patients admitted to the ICU. Unconscious patients with diagnoses

of acute neurological disorders, including traumatic brain injury and intracranial

hemorrhage were recruited. The patients were required to be unconscious with a Glasgow

Coma Score between three and eight and have a functioning ICP monitoring device in place for

serial measurement. Since the subjects were unable to consent for participation in the study,

the appropriate designated family member making healthcare decisions was asked to sign a

written consent form.

Inclusion Criteria

Thirteen men and women, age 18 to 85 years, were selected for the study. Subjects

were unconscious as indicated by neurological assessment and were unable to respond to

strong verbal commands or non-painful stimuli, such as gentle shaking. All subjects were

required to need neurological assessment in which a painful stimulus was used to elicit

motor responses as part of their routine neurological evaluation. Selected subjects had a

GCS of <8. Each subject was required to have a functioning ICP monitoring device in

place either directly connected to the bedside Hewlett Packard monitor or connected by a

slave cable attachment for readings. Functional direct arterial pressure and

electrocardiographic monitoring devices on the bedside monitor were required for

measurement of CPP and physiological response variables. Subjects were required to be

hemodynamically stable with systolic blood pressure readings of at least 100mmHg and

CPP readings of>60 mmHg. ICP readings were required to be stable at 20mmHg or less.










Oxygenation requirements for inclusion were arterial blood gas readings ofPaCO2

<40mmHg and PaO2>80mmHg.

Exclusion Criteria

Subjects who were excluded from this study included anyone under the age of 18 or

greater than 85 years of age. Subjects with conditions which would preclude their ability

for a motor response, such as high cervical spinal cord injury, or bilateral upper extremity

trauma resulting in paralysis were excluded.

Patients with conditions that could impair the response to stimuli or contribute to

hemodynamic responses were excluded. Those patients who were undergoing active

treatment to stabilize blood pressure or treat acute rises in ICP, significant ventilator

changes for oxygenation problems on the day of study were excluded. Patients whose

ICP were > 20mmHg or CPP < 60mmHg were excluded. Patients who had received

pharmacological suppression of motor or sensory responses, such as heavy sedation,

neuromuscular blockade, and barbiturate-induced coma were excluded. Patients were

excluded who had clinical evidence of seizures within two hours of the experiment.

Finally, any patient who could follow commands or be aroused in any way other than by

delivery of a painful stimulus was excluded.

Exceptions

The purpose of the study was to examine physiological and motor responses to

stimulus of both sides of the patient with NBC and SOP. If a patient met inclusion for

some of the data collection, and one or more sites were unacceptable to stimulate, they

were enrolled with collection of data based upon the acceptable sites. This allowed for










data collection in several of the subjects, who might have been otherwise excluded. For

example, if a patient had unilateral orbital injury, only the stimulus to the unaffected side

would be administered.

Human Subjects

Permission and Protection of Human Subjects

Permission to conduct the study was obtained from the University of Florida Health

Center Institutional Review Board in Gainesville, Florida and the Institutional Review Board in

the hospital where the study was undertaken in Orlando, Florida. The informed consent, which

included a complete description of the purposes, procedures, and risks and benefits of the study

was obtained from the appropriate person responsible for making treatment decisions for

unconscious subjects (Appendix D). A full verbal description of the study procedures was

provided by the investigator and questions answered before signing the consent form. This

person was informed that participation in the study was strictly voluntary and should he/she

choose not to enroll the unconscious patient in the study, this decision would not affect the

care rendered in the ICU. In addition, withdrawal from the study could occur at any time. A

written consent form for video-taping of the motor responses was obtained as required by

policy of the study facility.

Each subject was assigned a code number for identification to assure anonymity. The

subjects' names were not used in any reports of this research. The person consenting for the

patient was informed that the information obtained would be kept in a locked place only

accessible to the investigator. The videotape would be only viewed by the investigator for data

analysis only.










Measures

In this section, the instruments used to measure the dependent and independent

variables are presented. The independent variables measured in this study include ICP,

MAP, CPP, HR, and motor responses. Additionally, the dependent variables of NBC and

SOP were measured.

Physiological Measurements

All of the subjects in the study had physiological measurements being continuously

monitored from an individual, bedside Component Monitoring System (Hewlett Packard

CMS Ml 176A). The physiological measurements of arterial pressure and

electrocardiographic (EKG) tracings were consistently measured by the Hewlett Packard

(HP) CMS. Calibration and balancing of the transducers used to measure ICP and MAP

was accomplished according to standard HP product specifications within five minutes of

data collection.

Intracranial Pressure Measurement

In this study, the ICP was measured with a ventriculostomy catheter that was

connected to a fluid-filled pressure transducer and module system. The module was

attached directly to the HP CMS for simultaneous display. The air-fluid interface at the

transducer stopcock was placed between the lateral canthus of the eyelid and the top of

the ear.

Arterial Blood Pressure

In this study, arterial pressure was measured by means of direct arterial pressure from

an intravascular catheter, positioned in a peripheral artery and connected to a fluid-filled








57

transducer and module system. Arterial pressure was monitored by the HP CMS bedside

monitor In this study, mean arterial pressure was calculated using the following formula:

MAP= (ASP) + [(ADP) 21

3

where ASP = arterial systolic pressure; and ADP = arterial diastolic pressure (Rudy et al.,

1993). The reference point where arterial pressure was monitored was alongside the

transducer used for measuring the ICP: at the level of the lateral canthus of the eye and

the top of the ear.

Cerebral Perfusion Pressure

In this study, the values of MAP and ICP were collected from the HP CMS and were

exported into a computer for continuous sampling. After the experiment, the investigator

calculated the CPP from those values by using the following formula: CPP = MAP ICP.

These CPP values were incorporated into the physiological data base with the other

physiological variables (ICP, MAP, and HR) and used for subsequent analysis.

Heart Rate

The subjects' heart rates were measured by collecting continuous EKG recordings

from the HP CMS monitor. The EKG signals were acquired from the HP CMS through a

channel with an analog to digital board and were downloaded to the study computer. The

Labtech Notebook software program supplied data control and filing of the EKG signal

amplitude. These data were later converted into heart rate per minute, discussed later in

this section.








58

The Labtech Notebook replay mode was responsible for reduction and conversion of

heart rate data computer program to identify spikes indicating ventricular depolarization

(the QRS). Heart rate was calculated by measuring the intervals between QRS spikes with

subsequent interpretation of that cycle in beats per minute.

Measure of Motor Response

Motor responses of the patient were recorded during the delivery of NBC or SOP by

means of a 8mm Sony video camera recorder. The patients were recorded on video so

that the investigator could concentrate on delivering of the stimulus and allow for

subsequent detailed review and grading of the response at a later time. Videotaping motor

responses was a method used in a study conducted by Jennett and Teasdale (1978) when

testing the interrater reliability of grading motor responses by practitioners on the

Glasgow Coma Scale. The videotapes were reviewed later by the investigator, who then

graded the motor response using the motor section of the Glasgow Coma Scale. The

motor response portion of the GCS was presented in Chapter II (Table 2.1).

Delivery of the Painful Stimuli

One of the purposes of this study was to evaluate the force used when administering

painful stimuli to the subject. Since there was neither a standardized or measurable

method to deliver painful stimuli, tools were developed for this study to measure the force

applied while administering NBC and supraorbital pressure. A pilot study to quantify the

range of force used when administering these stimuli was conducted.

With the assistance of a researcher from the Department of Anesthesiology at the

University of Florida Health Center, a pilot study to simulate and measure the stimuli was








59

accomplished. First, it was important to determine what types of measurements could be

taken on the stimuli. There were no known available means to measure these stimuli. The

next best alternative to direct measurement of the clinical stimuli was to conduct a

representative simulation of them in a controlled, laboratory setting by scientific means.

Before any clinical measurement of the stimulus could occur, it was important to quantify

the range of the simulated force so that measurement devices could be created to measure

these forces in the study. Measurement of the simulated stimuli by only the investigator

was done to determine the reproducibility of the investigator in applying force. Statistical

evaluation for the means and standard deviations of those findings was done to determine

significant variability in performance by the same investigator.

Next, it was important to determine the method to measure the stimulus. The

investigator determined that the measurement would be in the form of pressure since both

NBC and SOP are applications of pressure to an anatomical site. The formula for pressure

calculation (Pressure = Force/Surface area) was used to quantify the simulated stimuli

performed by the investigator.

Measurement of Force

Weight was chosen to be used as the unit to measure force. A calibrated, precise scale

(Olympic Smart Scale, model #56320, Olympic Medical, 1988) used for neonatal research

was used to measure the force in weight. The investigator stood in front of the scale in

the same position and applied pressure in the marked center of the scale for all

measurements. The investigator pressed on the surface of the scale, attempting to simulate

the pressure given clinically. Both NBC and SOP techniques were imitated. The scale








60

took an average of weight applied over a 6 second interval. The investigator did not have

visual access to the readout of the force (weight) until the mean value was digitally frozen

and displayed. The digital mean values were recorded between each pressure simulation.

During this laboratory trial, the investigator reproduced the clinical procedures of

NBC and SOP as closely as possible. Finger NBC is customarily accomplished by

pressing a pencil, or like object, on top of one of the nail beds. A pencil with standard six

flat sides was used for the trial. The pencil was taped to the finger to avoid movement of

the pencil during weight measurements. Supraorbital pressure is ordinarily delivered by

pressing the thumb upwards and at a right angle against the supraorbital notch where a

branch of cranial nerve V exits. Vertical, downward compressions with the investigator's

thumb on the scale was used to imitate this maneuver.

Measurement of each force was taken for 32 trials and the highest and lowest values

dropped. Thirty trials of each simulation were evaluated using simple statistics for mean,

standard error, standard deviation, and confidence interval (95%) on both type stimuli

(Appendix A, Table A. 1). The findings were that the pressure exerted by the investigator

was consistent within a narrow range.

Measurement of Surface Area

The surface areas (SA) over which SOP and NBC are administered were different and

required separate analysis. The SA of SOP delivered to the scale was much like a thumb

print, so the investigator placed her right thumb in ink and made 30 separate compressions

on plain paper on a hard surface in the same manner used during the force measurement

trials. The ovoid surface of the thumb allowed force to be distributed over a large surface










area. The SA of the pencil for NBC simulation was different from supraorbital surface

area. The SA over which NBC force is delivered is a narrow, flat region. Nearly the same

force was delivered over a much smaller area, greatly increasing pressure calculations (P =

F/SA). The width of the pencil in contact with the nail bed during NBC varies, depending

on the size of the nail bed and shape of the pencil. Therefore, 15 men and 15 women

volunteered for dominant thumb nail measurements. A mean nail bed size for men (16.8

mm) and women (14 mm) was calculated. Control for pencil size was accomplished by

using the same #2, six-sided pencil with flat surfaces for compression. Segments of

pencils representing the mean nail bed width size measured for both men and women were

cut and used to estimate SA for nail bed compression. The SA for NBC was determined

by taping the pencil sections to the finger and dipping them in ink. Thirty impressions for

each pencil size were made using downward vertical compressions (30 men, 30 women

sizes) on plain paper on a hard surface the same way as the force measurements were

taken.

SA of the regions of SOP and NBC for both men and women were measured by a

computerized system (PC microcomputer, IBM, Armonk, NY) with an electronic

planimeter (Hewlett Packard Graphics Tablet). Thirty samples representing SOP and

NBC were measured for SA and simple statistics for mean and standard deviations were

recorded in millimeters squared.

Pressure Calculations of the Stimuli Simulation

The numerical values taken from the force and SA measurements underwent the

following adjustments and calculations. First, the mean weight in ounces for both SOP










and NBC was converted to pounds. Next, the measured SA in squared millimeters were

converted to square inches (Appendix A). These conversions allowed for calculation of

force in pounds per square inch (psi). Final results of the pressures achieved by the

principal investigator are found in Appendix A.

Measurement of NBC and SOP

The information gained from the pilot study on pressure measurements were used to

develop two instruments to quantify NBC and SOP in psi. Services by the University of

Florida Health Center's bioengineering department were contracted to develop

instruments to measure NBC and supraorbital pressure. These instruments were custom-

made devices capable of measuring either NBC or supraorbital pressure. For patient

safety and reducing the size in implementation, a hydraulic system using two different

custom probes, each feeding into a Motorola pressure transducer (MXP 7100G; 0 to 100

psi) was assembled. Both probes shared a common power supply and amplifiers were

designed and built. The probe used to measure NBC was a small hand-held device

encased in hard plastic with two separate sides. One side was hard and flat, simulating a

pencil edge that the investigator placed against the nail bed during compression. The

other side contained elastic tubing filled with silicone that was connected to the Motorola

transducer for pressure measurement in psi. This side was pressed by the investigator's

thumb when delivering NBC with the other side of the probe.

Supraorbital pressure measurements were made using a rubber nipple filled with

water that was connected with pressure tubing to the same transducer as the NBC probe.








63

The investigator placed the nipple directly on the thumb within a rubber glove for stability

and control when applying pressure against the supraorbital rim.

Pressures delivered for both stimuli were quantified in pounds per square inch (psi)

using a Motorola (MXP 7100G) transducer system. The same transducer was used for

both NBC and SOP measurements with analog recordings exported to the personal

computer for data acquisition. Selection of NBC or SOP was accomplished by the use of

an up and down switch. Values of NBC and SOP in psi were used for analysis of the

pressures used for both NBC and SOP. The principal investigator was the only one to

deliver the stimuli in the study. In the study, both NBC and SOP were delivered with

increasing intensity until a clear motor response was seen, or until 10 seconds had lapsed.

The interval over which NBC or SOP was given was also recorded.

Data Acquisition and Synchronization

In order to capture incremental changes in the physiological variables (ICP, CPP,

MAP, and HR) and pressure measurements of NBC and SOP over time, a system for

continuous data acquisition was required. Additionally, it was important to collect

simultaneous measurements of the physiological values and stimuli. An engineer was

consulted from The University of Florida Health Center's bioengineering department to

assist with the development of computerized programming and synchronization of data

acquisition for this study. The consultant also developed instruments to measure the

forces used in psi when administering NBC and SOP and coordinate that data collection

with physiological measurements.








64

Timing and synchronization of continuous physiological measurements was crucial in

this study to detect changes in them over time. Since this study was an original inquiry

regarding potential changes in physiological measurements during administration of painful

stimuli, the time over which these changes might occur was unknown. Findings from

previous studies on the influence of clinical activities on ICP and CPP indicated that

changes typically occurred over a few minutes, and rarely over ten minutes before

returning to baseline. Thus, in this study, the goal of the investigator was to use a data

acquisition system capable of collecting data in small, continuous time increments to

detect particular events as they occurred.

All physiological variables, ICP, MAP, and HR, were measured and displayed by the

HP CMS (M1176A) monitor. The HP CMS receives analog signals from the pressure and

EKG modules and averages them over a three-second interval. This averaged value is

displayed in a digital format on the bedside monitor. For usual monitoring purposes, this

is adequate. All pressure and EKG impulses are rhythmical and vary in amplitude over

time. Ordinarily, clinical measures of HR, MAP, and ICP are recorded as averaged signals

over some specified time period. A sampling rate of analog data every three seconds is

sufficient for clinical monitoring. However, since the hypothesized changes in ICP, MAP

and HR in response to painful stimuli were unknown, a more rapid sampling of the

waveforms with higher resolution of data was required. The HP CMS data retrieval

system could only display data in increments of every five minutes, so this was an

unacceptable method to capture small, incremental changes in the physiological variables.

The digital display showed averaged values over a three second interval, so it would be










difficult to capture events exactly as they occurred. Thus, the use of either of these two

methods from the HP CMS would not suffice in this study. It was also essential that all

physiological data be collected simultaneously during the experiment. The data

acquisition was accomplished using an eight channel, 12 bit, analog to digital (A to D)

board (Advantech PCL-71 Is) connected to the HP CMS for direct retrieval of analog

voltage signals.

In addition to physiological data collection over brief time intervals, it was important

in this study to detect the onset, duration, and amplitude of the painful stimuli as it

occurred. Thus, it was essential to coordinate the timing of data acquisition of NBC and

SOP pressures with the physiological measurements. To accomplish this, two of the eight

channels on the A to D board were dedicated for NBC and SOP measurements. This

allowed for synchronization of the stimulus measurements with the physiological signals.

Heart Rate, MAP, and ICP measurements were exported from the HP CMS

(M1 176A) monitor via an RS232 port accessing the eight channel A to D card.

Measurements of NBC and SOP forces were made using the probes designed for this

study. Nail bed compression and SOP were measured in psi from the Motorola transducer

(MXP 7100G) and were also exported to the study computer with the physiological

variables. Using the analog output from the patient monitor yielded timely data retrieval

at the cost of relative displayed values. These analog outputs were actual voltage signals

as they occurred during the sampling time. Intervals between cardiac events and

pulsations demonstrate a waveform pattern and are and not computed values. (i.e., EKG

voltage trace, not Heart Rate). Clinically meaningful values were detected by the










measurement of average pressure or the rate of heart beats calculated for a minute. The

computer operating system used was Windows 3.1 on MS-Dos 6.22 with programs either

commercial or developed in Visual Basic 4.0.

Precision of data acquisition was based on three factors: resolution, repeatability, and

timeliness. Resolution was a function of the analog to digital conversion and scaling. The

12 bit card that was used resolved 2047 points across the applied positive voltage and the

same for negative values. This translates to 2.44 millivolts per bit of binary count at five

volts full scale. Repeatability of data acquisition was achieved by using a board that was

self-calibrating when the system is initialized. Preliminary testing of the program

demonstrated the data recorded was clean, valid, and repeatable as to calibration.

Timeliness of synchronized data acquisition was a function of the delay of recording

multiple signals as they were processed and synchronized through the system. By using a

sampling rate of fifty hertz (cycles per second), the maximum spread of signals was less

than two hundredths of one second. This resolution of signal voltage is highly acceptable

for detection of changes in the physiological variables. The timing and recording of data

acquisition was provided by the computer using the following data sources and

summarized in Appendix B.

Data control and filing was handled by the Labtech Notebook runtime configuration

"DA50.LTC" according to specifications designed by the engineer. Data were simultaneously

collected through the eight channels described in Appendix B, Table B. 1. These data were

displayed as waveforms during the experiment and saved to an automatically assigned a file name

(C:\DATASETS\Da50nnnn.CSV, where nnnn = the sequence number). Data acquisition was








67

achieved at a sampling rate of 50 times per second for five minutes. This yielded 15000

data points and a file size of 661k for each stimulus period. Data was saved as voltage

values separated by commas. Saving values using this method allowed for the data file to

fit into an EXCEL spreadsheet format with a 16,000 row limit.

The data files for each stimulus yielded 15,000 entries and were too large for data

analysis in a practical fashion. In order to simplify the data analysis, files were compressed

to a manageable size and converted from voltage signals into units of pressure and heart

rate per minute. Data review and analysis was performed by the Labtech configuration

"REPLAYS0.LTC" for data conversion and reduction. Collected data files were copied

to the Windows NoteBook subdirectory as "WFSOHZ.PRN" (Wave Form at 50 Hertz).

The REPLAY50 function reads this file at a rate of 25 samples per second, and displays

average Heart Rate, ICP, and MAP at half-second rates. The pressure voltage used during

NBC or SOP was resolved to pounds per square inch according to which probe was used

and was recorded simultaneously with ICP, MAP, and heart rate. After data reduction,

the resulting file (AD50nnnn.CSV) contained measurements in half-second time intervals

of average ICP, MAP, and HR, and pressure applied with the stimulus. The condensed

file size of 600 data points of these five variables was about 21k. The numerical values

obtained in the condensed file were used for data analysis of values in a more practical

format. These compressed files were then converted into an EXCEL file for data

organization and graphic display. Comparative graphs for selected periods of time

(stimulus application) showed good correlation of the shape and impact of the stimulus

and validated the data reduction without causing statistical shift or data loss.










Heart rate was derived from measuring the interval between ventricular

depolarizations (the QRS) over ten beats and calculating the minute rate. The data

acquisition system to detect raw EKG signals was sensitive to motion artifacts and skin

resistance. In some subjects, the EKG voltage spikes were too low for recognition by the

standard voltage detection. The converse was also true in that some patients

demonstrated very high peak levels of EKG amplitude, which resulted in abnormally high

readings of heart rate. In order to capture all levels of EKG signals, there were two

additional REPLAY50 modes designed with higher and lower triggerpoints to detect QRS

spikes for those who demonstrated variances in EKG voltage. This yielded a higher

resolution of heart rate data acquisition.

Procedures

Description of experiment

Once potential subjects were identified for the study, informed consent from the

appropriate legal representative for the patient was obtained. Prior to initiation of the

experiment, the patient was reevaluated to ensure that they still met inclusion criteria. The

patient's demographic and biographical data was collected on the patient information sheet

(Appendix C).

Once the patient was enrolled into the study, the room was prepared for the

experiment. To control for potential environmental influences on ICP during the study

period, the room was darkened, extraneous noise eliminated as much as possible, and the

door closed. The investigator consulted with all caregivers of the patient to request that no

patient care procedures be done 15 minutes prior to data collection. The caregivers were










also asked to avoid any unnecessary interruptions or entry into the room while data

collection occurred. The room remained as dark and quiet as possible with minimum

verbalization during the procedure by the investigator or assistants. The patient was

prepared for the experiment by unrestraining their limbs, if restrained and their body

exposed for maximum viewing of motor response. The head of the patient was placed at a

30 degree angle unless otherwise contraindicated. The video recorder was positioned

prior to the data collection period so as to have the patient's whole body visible in the

display screen.

The pressure modules from the bedside HP CMS were transferred onto an identical

portable Hewlett Packard CMS used for the study. It was configured with an RS 232 port

to export ICP, ABP, and HR to the study computer for data retrieval and storage.

Waveforms of the ICP, ABP, and HR were evaluated for clarity and presence of artifact.

Any distortions in waveforms were corrected prior to the start of data collection. The

transducers for ICP and ABP were balanced, calibrated and positioned at the lateral

canthus of the eye and top of the ear. The personal computer used to collect and store

data from the HP CMS was turned on and readied for data collection.

The delivery of all stimulus was performed by the investigator using devices

specifically constructed to measure forces used for purposes of this study summarized in

Chapter II (Appendix A). The investigator determined the order of the four stimuli by the

random selection sequence determined prior to the study. Once the program for data

collection had been initiated, the investigator waited for approximately 45 seconds. This

time period was used to collect baseline data. The video recorder was turned on during










this time period. After baseline data collection, the investigator began delivering the

selected stimulus in an accelerating fashion, observing for motor response. Once a motor

response was seen clearly, the investigator stopped administering NBC or SOP. In the

case that the patient did not demonstrate a motor response, stimulus delivery was stopped

at ten seconds. The video recording of the subject was discontinued after the stimulus

administration was completed. The physiological values were continuously measured for

rest of the data collection period, equaling a total of five minutes. The timing for data

collection was programmed and controlled by the Labtech Notebook program. After the

program discontinued, the data collection process was repeated, and the other three

stimuli were administered, unless otherwise contraindicated. The total time used for data

collection was 20 minutes. After collecting the data, subjects were repositioned as they

were before data collection. The pressure modules returned to the bedside monitor and

recalibrated, balanced and leveled.

Risks

Application of SOP and NBC is commonly used in the clinical setting to evaluate

motor responses as part of routine assessment in the intensive care unit. The methods

used to administer NBC or SOP posed no more risk to the subject than the usual standard

of care. The risk of applying painful stimuli for purposes of neurological evaluation of

motor responses is unknown. One purpose of the study was to determine whether

delivery of painful stimuli resulted in increased and sustained rises in ICP or reduction in

cerebral perfusion pressure.








71

The monitoring devices for MAP and ICP were already in place and posed no more

risk to the patient than routine monitoring of those values. Heart rate monitoring is also a

routine measure and is non-invasive, thus, posing no more risks than customary EKG

monitoring. Data collection was accomplished through analog to digital recordings from

the HP CMS monitor to a personal computer, thus posed no threat to the subject. There

were no known financial risks associated with this study.














CHAPTER IV
RESULTS

In this chapter, characteristics of the sample recruited into the study are summarized.

Additionally, this chapter presents the results of statistical analyses of the physiological

variables related to each of the five research hypotheses. The results of the pressure

measurements of nail bed compression (NBC) and supraorbital pressure (SOP) obtained in this

study are also presented. The hypotheses were tested a 0.05 level of significance.

Sample Characteristics

Subjects included in the study were adult unconscious patients in the intensive care unit

diagnosed with acute brain injury. Subjects had electrocardiographic, ICP and arterial pressure

monitoring in progress. The subjects were hemodynamically stable and not undergoing active

resuscitation. They were not receiving drugs that might have altered their ability to

physiologically respond to painful stimuli. Subjects had acceptable values of cerebral perfusion

indices and intracranial pressure. Finally, acceptable subjects were routinely evaluated for level

of consciousness and motor responses by administering a painful stimulus.

Patients excluded from the study had potential safety threats to neurological stability.

Customary caution in the care of those patients to avoid any precipitous changes in cerebral

hemodynamics precluded inclusion of those individuals. Those patients with traumatic

disruption of sensoroneural connection, such as complete cervical spinal cord injury, were

excluded.






73

This study involved administering nail bed compression and supraorbital pressure on both

sides of the body. If the patient had trauma to one of the orbits, that site for painful stimulus

evaluation was excluded. In that case, administering painful stimuli to the other acceptable

sites were performed. The same held true for those patients who had an injury to one of the

nerve tracts in the arm.

In this sample of patients, 10 of the 13 subjects had all four stimuli administered. In one

patient, SOL was not administered due to orbital injury on the left side. Another patient had an

increase in ICP after the first 3 stimuli, so the NBL stimulus was deferred. One patient had

SOL administered twice due to dampening of the arterial line pressure during the first

administration. Data collected from ICP and HR were included in the first attempt. During

data acquisition of three patients, there were technical difficulties with the sensitivity recording

on HR, thus in those subjects there was no heart rate data available. The other physiological

values were recorded and analyzed.

Demographic Characteristics of the Sample

There were a total of 13 subjects included in the study. There were ten males and three

females enrolled. The ethnicity of the subjects were as follows: ten Caucasian, one Asian, and

two Hispanic. The ages of subjects ranged from 18 to 77 years ( mean = 32, SD = 15.76).

Only one subject was 77 years of age. Most of the subjects were considerably younger. The

remainder of the sample (n = 12) were ages between 18 to 43 years (mean = 28, SD = 8.47).

Admission Diagnoses of the Sample

The subjects in this sample presented with a variety ofpathophysiological disorders of the

brain. Twelve subjects presented with closed head injuries, sustained by motor vehicle crashes,

being struck by a motor vehicle as a pedestrian, or falls. These twelve subjects had several








concurrent injuries to the brain, summarized in Table 4.1. One subject was admitted to the

hospital with a diagnosis ofintracranial hemorrhage due to hypertensive crisis.


Table 4.1

Concurrent Brain Injuries in Subjects with CHI

Subject Concurrent Pathological Findings

1 SAH, IVH, cerebral edema
2 IVH, EDH, DAI, skull fracture
3 SAH, DAI, skull fracture
4 Edema, skull fracture
5 SDH, IVH, edema
6 SDH, EDH, edema, skull fracture, contusion
7 EDH, IVI, skull fracture
9 SDH, skull fracture
10 SAH
11 SDH, EDH, SAH, edema, skull fracture, contusion
12 SDH, IVH, SAH, edema, skull fracture, contusion
13 SDH, SAH, skull fracture, edema


Note: SAH = Subarachnoid Hemorrhage, SDH = Subdural Hematoma, IVH = Intraventricular
Hemorrhage, DAI = Diffuse Axonal Injury, EDH = Epidural Hematoma

Prestudy Motor Responses

Prior to the study period, the most recent Glasgow Coma Score (GCS) documented on the

patient's chart was recorded. The GCS of the subjects ranged from four to eight (mean = 5.85,

SD= 1.41).

Tests of the Research Hypotheses

The dependent variables in this study included changes in intracranial pressure (ICP),

mean arterial pressure (MAP), cerebral perfusion pressure (CPP), heart rate (HR) and motor

responses. All subjects had these variables measured continuously throughout a five minute

period using the bedside monitor connected to a personal computer for data acquisition.








For the first 45 to 55 seconds prior to administration of a randomly selected stimulus of

NBC or SOP, continuous data were collected to serve as baseline values for comparing

each of the timed interval measurements. Baseline values were made by averaging all

measurements taken during the pre-stimulus period. There were five time periods evaluated

for changes in the dependent variables at the onset of the stimulus: peak value over the first 30

seconds, averaged values over the first 30 seconds, averaged values over the first minute,

averaged values between the first and second minute, and averaged values over the fourth and

fifth minute of entire data collection. Peak values were determined by depicting the highest

value that was measured or calculated over the first 30 seconds after the onset of delivering the

stimulus. All measurements taken over that same 30 second period were averaged and

compared to baseline.

In answering the research questions, multivariate repeated analysis of variance

(MANOVA) was the statistical method used to detect within-subject changes over time.

Averaged baseline values were used to detect differences in physiological measures of

ICP, MAP, CPP, and heart rate over the following time periods: peak value over 30

seconds beginning at the time of the stimulus onset, averaged values over the first 30

seconds from the stimulus onset, averaged values over the first minute from the stimulus

onset, averaged values over the first to second minute after the stimulus onset, then

averaged values over the last four to five minutes of the experiment. The results of the

statistical analyses are reviewed according to each of the research hypotheses.

In this study, NBC and SOP were measured during stimulus delivery and recorded. The

mean pressures used for NBC and SOP were determined to answer the research question

about the pressures used when delivering NBC and SOP in this study. Quantification of the








stimulus was a separate measure important in this study and is summarized with descriptive

statistics.

Research Hypothesis 1

The first hypothesis stated that there would be a significant increase in ICP when

compared to baseline values. This hypothesis was supported by the results from repeated

measures MANOVA to detect changes in ICP in response to NBC and SOP performed on

the right side, NBC and SOP performed on the left side, and combined responses from all

NBC (NBT) and supraorbital pressure (SOT). Tables 4.2 and 4.3 summarize the values

obtained over the time periods described above for NBC and SOP. All values are

measured in mm Hg, and include the mean and standard deviation (top values) and range

of values (bottom values).


Table 4.2

Descriptive Statistics of ICP Changes With NBC

Stimulus Baseline Peak 30 Average Average Average Average
Type seconds 30 seconds 1 minute 1-2 minute 4-5 minute
M SD M SD M SD M SD M SD M SD
NBR 14.46 4 20.25 5.16 14.28 3.34 14.06 4.13 13.78 4.35 15.07 4.5
(n=13) (9.35-20.52) (12.2-7.08) (9.01-22.69) (8.73-22.6) (9.51-22.56) (9.8-24.83)
NBL 15.49 4.26 19.72 5.42 13.9 4.18 13.45 3.56 13.11 3.51 14.79 3.88
(n=12) (9.6-22.79) (12.04-31.56) (8.72-24.32) (8.69-21.84) (8.51-20.2) (9.44-22.23)
NBT 14.95 4.08 20 5.18 14.1 3.98 13.77 3.8 13.46 3.9 14.94 4.13
(n=25) (9.35-22.79) (12.04-31.56) (8.72-24.32) (8.69-22.6) (8.51-22.56) (9.44-24.83)

Note: NBT = Combined total responses from NBC.







Table 4.3

Descriptive Statistics of ICP Changes With SOP


Stimulus Baseline Peak 30 Average Average Average Average
Type seconds 30 seconds 1 minute 1-2 minute 4-5 minute
M SD M SD M SD M SD M SD M SD
SOR 15.13 4.28 22.11 4.66 15.2 3.46 14.69 3.32 13.66 3.44 15.34 3.54
(n=14) (7.8-21.81) 14.36-30.24) (9.91-21.73) (9.2-22.18) (9.63-20.63) (10.26-22.1)
SOL 15.1 3.53 22.11 5.22 15.03 3.25 14.27 3.18 13.69 3.94 15.5 3.81
(n=13) 10.82-22.51) 13.24-29.16) (10.51-22.1) (10.36-21.1) (6.89-20.5) (9.49-22.79)
SOT 15.11 3.86 22.11 4.84 15.12 3.3 14.48 3.2 13.67 3.61 15.42 3.6
(n=27) (7.8-22.51) 13.24-30.24) (9.91-22.1) (9.2-22.18) (6.89-20.63) (9.49-22.79)

Note: SOT = Combined total responses from SOP.

Using multivariate repeated measures analysis of variance (MANOVA), the values

summarized in Tables 4.2 and 4.3 were statistically tested for significant changes in ICP

from baseline with each of the time periods. This analysis was used to test the hypothesis

that NBC or SOP would result in an increase in ICP from baseline values.

With the NBR stimulus, there was a significant change in ICP within subjects

(F = 10.905, p=0.0021). When the ICP was tested for differences between baseline and

specified time periods, the following was found: Between baseline and peak 30 second

values, there was a significant difference (F = 49.41, p = 0.0001). When contrasting

baseline to averaged pressures over the first 30 seconds, there was no significant

difference in ICP (F = 0.62, p = 0.4459). When baseline ICP was compared to the

averaged pressures over one minute, there was no significant difference (F = 3.4,

p = 0.0901). The same findings were discovered at the first to second minute (F = 2.2,

p = 0.164), and the fourth to fifth minute (F = 2.15, p = 0.1679).

With the NBL stimulus, there was a significant change in ICP within subjects (F =

10.83, p = 0.0034). When ICP was evaluated for differences between baseline and

specified time periods, the following was found: between baseline and peak ICP values








over the first 30 seconds, there was a significant difference (F = 25.71, p = 0.0004).

When contrasting baseline ICP to averaged ICP over the first 30 seconds, there was no

significant difference (F = 4.78, p = 0.0513). When baseline ICP was compared to the

averaged ICP over one minute, there was a significant difference (F = 9.03, p = 0.012).

When comparing between the averaged ICP over the first minute and the first to second

minute, there was a significant difference from baseline ICP (F = 6.69, p = 0.0253).

However, when contrasting baseline ICP and values over the fourth to fifth minute, there

was no significant difference (F = 1.21, p = 0.295).

The data obtained on NBR and NBL were subjected to MANOVA testing to

determine the rate of decline of ICP to baseline with each successive time period. This

analysis was used to determine whether there were sustained elevations in ICP after the

initial peak increases.

When comparing successive difference in ICP with NBR over the time periods, the

following was found: Between peak 30 second values and average 30 second values,

there was a significant difference in ICP (F = 51.92, p = 0.0001). When averaged 30

second ICP values were compared to averaged one minute values, there was no significant

difference found (F = 0.8, p = 0.389). Between one and two minute values of ICP, there

was no significant difference (F = 0.75, p = 0.403). However, between the one to two

minute values of ICP and the fourth to fifth minute values, there was a significant

difference in ICP values (F = 10.38, p = 0.0073). This difference represented a decrease

in the ICP from baseline.








In comparing successive difference in ICP with NBL with each time period, the

following was found: Between peak 30 second values and averaged 30 second values,

there was a significant difference in ICP (F = 77.79, p = 0.0001). When averaged 30

second ICP values were compared to averaged one minute values, there was no significant

difference found (F = 1.54, p = 0.241). Between averaged one minute and one to two

minute values of ICP, there was no significant difference found (F = 0.57, p = 0.4667).

Between the one to two minute values of ICP and the fourth to fifth minute values, there

was a significant difference in ICP values (F = 6.89, p = 0.024).

Analysis of this data revealed that the NBR stimulus did cause a significant peak rise

in ICP over the first 30 second time period, but rapidly declined to baseline and remained

there, except at the final fourth to fifth minute, when there was a small increase in ICP,

which was significant. This small rise in ICP was not significantly different from baseline

(F = 2.15, p = 0.168). Data analysis on the ICP values with the NBL stimulus did cause a

significant peak rise in ICP over the first 30 second time period, but rapidly declined to

below baseline ICP and remained there. There was a decrease in the final fourth to fifth

minute ICP, which was significantly different. The small decrease in ICP did not,

however, was not significantly different from baseline ICP.

When collapsing all nail bed pressure data (NBR and NBL), there was a significant

difference within subjects in ICP values (F = 22.9534, p = 0.0001). When all supraorbital

values of ICP were evaluated for difference between baseline and the specified time

periods, the following was found: between baseline and peak 30 second values, there was

a significant difference from baseline to peak ICP in 30 seconds (F = 71.62, p = 0.0001).

When contrasting baseline ICP values to averaged ICP over the first 30 seconds, there






80

was a significant difference (F = 4.87, p = 0.0371). This difference represented a decrease

in ICP values from baseline. When baseline ICP values were compared to the averaged

ICP over one minute, there was also a significant difference detected (F = 9.91,

p = 0.0044). This difference also represented a decrease in ICP from baseline values.

When comparing baseline ICP values to the first to second minute, there was a significant

difference from baseline ICP (F = 8.26, p = 0.0084). This, too, reflected a decrease in ICP

when compared to baseline ICP. There was no significant difference in baseline ICP and

the fourth to fifth minute ICP values (F = 0.00, p = 0.9718).

When comparing successive difference in ICP with all NBC over the time periods, the

following was found: between peak 30 second values and averaged 30 second values,

there was a significant difference (F = 126.71, p = 0.0001). When averaged 30 second

ICP values were compared to averaged one minute values, there was no significant

difference found (F = 2.41, p = 0.1337). Between one minute ICP values and from one

to two minute ICP values, there was no significant difference (F = 1.33, p = 0.2598).

Between the one to two minute values of ICP and the fourth to fifth minute values, there

was a significant difference in ICP (F = 16.4, p = 0.0005). This significant difference

represented a decline in ICP, but was not significantly different from baseline values

(F = 0.00, p = 0.9718). Analysis of this data revealed that nail bed compression did cause

a significant peak rise in ICP over the first 30 second time period, but rapidly declined to

below and at baseline values and remained there.

The analysis of the data obtained on ICP by the supraorbital stimuli was conducted in

the same fashion as the NBC data. The results of that analysis follows.

With the SOR stimulus, there was a significant change in ICP within subjects






81

(F = 25.438, p = 0.0001). When SOR was evaluated for a difference between baseline and

specified time periods, the following was found: between baseline and peak 30 second

ICP values, there was a significant difference (F = 56.96, p = 0.0001). However, when

comparing baseline to averaged ICP over the first 30 seconds, there was no significant

difference (F = 0.03, p = 0.876). When baseline ICP was compared to the averaged ICP

over one minute, there was no significant difference (F = 0.72, p = 0.41). The same

findings were discovered when comparing baseline ICP to the averaged first to second

minute values (F = 3.45, p = 0.086), and the averaged fourth to fifth minute values

(F = 0.2, p = 0.6629).

When testing for successive difference in ICP with SOR over the time periods, the

following was found: Between peak 30 second ICP values and averaged 30 second

values, there was a significant difference detected (F = 87.01, p = 0.0001). When

averaged 30 second ICP values were compared to averaged one minute values, there was

no significant difference found (F = 2.29, p = 0.154). Between averaged one minute ICP

values and one to two minute values of ICP, there was no significant difference found

(F = 3.86, p = 0.0713). However, between the one to two minute values of ICP and the

fourth to fifth minute values, there was a significant difference in ICP (F = 11.51, p =

0.0048). This significant difference represented small rise in ICP, however, this value was

not significantly different from baseline ICP (F = 0.2, p = 0.663). Analysis of this data

revealed that SOR stimulus did cause a significant peak rise in ICP over a 30 second time

period, but rapidly declined to baseline and remained there.

With the SOL stimulus, there was a significant change in ICP within subjects

(F = 17.3008, p = 0.0004). When SOL was evaluated for difference between baseline and








specified time periods, the following was found: between baseline and peak 30 second

values, there was a significant difference in ICP (F = 62.79, p = 0.0001). However, when

comparing baseline to averaged ICP over the first 30 seconds, there was no significant

difference (F = 0.03, p = 0.8677). When baseline ICP was compared to the averaged ICP

over one minute, there was no significant difference (F = 4.48, p = 0.0559), representing a

decline in ICP below baseline. When comparing baseline ICP values to the averaged first

to second minute values, there was no significant difference from baseline ICP (F = 3.39,

p = 0.0903). There was no significant difference in baseline ICP and the fourth to fifth

minute ICP values (F = 0.85, p = 0.3759).

When comparing successive difference in ICP with SOL over the time periods, the

following was found: between peak 30 second ICP values and averaged 30 second values,

there was a significant difference (F = 90.17, p = 0.0001). When averaged 30 second ICP

values were compared to averaged one minute values, there was a significant difference

found (F = 6.04, p = 0.0302). Between averaged one minute ICP values and from one to

two minute values of ICP, there was no significant difference found (F = 0.87,

p = 0.3692). Between the one to two minute values of ICP and the fourth to fifth minute

values, there was a significant difference in ICP values (F = 8.61, p = 0.0125). This

significant difference represented a small rise in ICP, but it was not significantly different

from baseline values of ICP (F = 0.85, p = 0.3579). Analysis of this data revealed that

SOL did result in a significant peak rise in ICP over a 30 second time period, but rapidly

declined to baseline and remained there after the first minute.

When collapsing all supraorbital data (SOR and SOL), there was a significant

difference within subjects in ICP values (F = 38.738, p = 0.0001). When all supraorbital








values of ICP were evaluated for the difference between baseline and specified time

periods, the following was found: between baseline and peak 30 second values, there was

a significant difference from baseline to peak ICP in 30 seconds (F = 123.65, p = 0.0001).

However, when baseline values were compared to averaged pressures over the first 30

seconds, there was no significant difference in ICP (F = 0.00, p = 0.989). When baseline

ICP values were compared to the average ICP over one minute, there was no significant

difference (F = 3.73, p = 0.0645). When baseline ICP values were compared to the

averaged first to second minute values, there was a significant difference from baseline

ICP (F = 7.09, p = 0.0131). This difference represented a decrease in ICP values from

baseline. There was no significant difference in baseline ICP and the fourth to fifth minute

ICP values (F = 0.89, p = 0.3552).

When comparing successive difference in ICP with all SOP over the time periods, the

following was found: between peak 30 second values and averaged 30 second values,

there was a significant difference found (F = 183.7, p = 0.0001). When averaged 30

second ICP values were compared to averaged one minute values, there was a significant

difference found (F = 7.71, p = 0.01). Between averaged one minute ICP and from one

to two minute ICP values, there was a mildly significant difference (F = 4.15, p = 0.0518).

This also represented a further decline in ICP. Between the averaged one to two minute

values of ICP and the fourth to fifth minute values, there was a significant difference in

ICP (F = 20.5, p = 0.0001). This significant difference represented a small rise in ICP,

but it was not significantly different from baseline values of ICP (F = 0.89, p = 0.3552).

Analysis of this data revealed that supraorbital pressure did cause a significant peak rise in








ICP over the first 30 second time period, but rapidly declined to baseline and remained

there.

The conclusion reached after analysis of the data is that the null hypothesis of no

difference in ICP with NBC and SOP various stimuli was rejected, and the alternative

hypothesis that there was a statistically significant difference in ICP with NBC and SOP

was supported.

Research Hypothesis 2

The second hypothesis stated that there would be an increase in MAP when compared

to baseline values. This hypothesis was supported by analysis of the data. Multivariate

repeated measures analysis of variance was used to detect changes in MAP in response to

NBC and SOP performed on the right side, NBC and SOP performed on the left side, and

combined responses from all NBC (NBT) and supraorbital pressure (SOT). Tables 4.4

and 4.5 summarize the data obtained on MAP over the time periods described previously

for NBC and SOP. All values in the table are measured in mm Hg, and include the mean

and standard deviation (top values) and range of values (bottom values).


Table 4.4

Descriptive Statistics of MAP Changes with NBC

Stimulus Baseline Peak 30 Average Average Average Average
Type seconds 30 seconds 1 minute 1-2 minute 4-5 minute
M SD M SD M SD M SD M SD M SD
NBR 90.93 11.67 112.94 14.57 97.75 12.79 96.58 11.87 92.68 11.04 91.23 10.82
(n=13) (68.4-113.07) (89.64-138) (77.97-120.34) (77.48-117.83) (74.75-113.31) (70.6-113.25)
NBL 90.79 9.8 116.33 13.18 99.88 11.85 97.64 12.11 98.84 12.54 91.09 10.8
(n=12) (71.99-111.56) (95.52-141.04) (81.08-120.06) (80.59-119.72) (74.75-117.65) (69.25-111.1)
NBT 90.86 10.59 114.57 13.74 98.77 12.14 97.09 11.75 93.24 11.55 91.16 10.58
(n=25) (68.4-113.07) (89.64-141.04) (77.97-120.34) (77.48-119.72) (74.75-117.65) (69.25-13.25)

Note: NBT = Combined totals responses from NBC.









Table 4.5

Descriptive Statistics of MAP Changes with SOP


Stimulus Baseline Peak 30 Average Average Average Average
Type seconds 30 seconds 1 minute 1-2 minute 4-5 minute
M SD M SD M SD M SD M SD M SD
SOR 89.89 12.47 118.9 14.67 99.3 12.32 97.43 11.72 91.1610.95 89.56 11.55
(n=14) (68.4-12.82) (91.64-35.56) (77.97-19.25) 77.48 -116.42) (74.67-114.14) (69-111.8)
SOL 0.27 11.1 124.27 15.7 102.7 14.33 99.1 12.98 94.62 13.95 91.74 13.47
(n=13) (68.5-113.4) (102.52-160) (81.1-130.14) (78.43-119.42) (72.77-122.7) 68.59-118.27)
SOT 90.07 11.6 121.49 15.12 100.97 13.18 98.24 12.13 92.83 12.36 0.61 12.31
(n=27) (68.4-113.4) (91.64-160) (77.97-130.14) (77.48-119.42) (72.77-122.7) (68.59-118.27)

Note: SOT = Combined total responses from SOP.

Using multivariate repeated measures analysis of variance (MANOVA), the data

summarized in Tables 4.4 and 4.5 were statistically tested for significant changes in MAP

from baseline that occurred within subjects with each stimulus type. First, baseline values

of MAP were compared to the MAP in each of the time periods to detect for differences.

This analysis determined whether the MAP values in each time period differed from

baseline values. The MANOVA results on MAP are summarized in Table 4.6 and 4.7.


Table 4.6

Statistics on MAP Changes from Baseline with NBC


Stimulus Change in Baseline Baseline Baseline Baseline Baseline
Type MAP within and Peak and 30 and 1 and 1-2 and 4-5
subjects second minute minute minute
over time average average average average
NBR F=16.13 F=66.68, F=17.12, F=13.05, F=8.26, F=0.14
(n=13) p=0.0005* p=0.0001* p=0.0014* p=0.0036* p=0.014* p=0.7116

NBL F=27.8475, F=107.15, F=35.74, F=18.63, F=3.85, F=0.34,
(n=12) p=0.0002* p=0.0001* p=0.0001* p=0.0012* p=0.0755 p=0.5724

NBT F=51.5321, F=167.33, F=49.31, F=32.17, F=8.72, F=0.41,
(n=25) p=0.0001* p=0.0001* p=0.0001* p=0.0001* p-0.0069* p=0.5294

Note: indicates statistical significance at the 0.05 level. NBT = Combined total
responses from NBC.









Table 4.7

Statistics on MAP Changes from Baseline with SOP


Stimulus Change in Baseline Baseline Baseline Baseline Baseline
Type MAP and Peak and 30 and 1 and 1-2 and 4-5
within second minute minute minute
subjects average average average average
over time
SOR F=18.894, F=91.48, F=30.86, F=23.18, F=0.6, F=0.27,
(n=14) p=0.0002* p=0.0001* p=0.0001* p=0.0003* p=0.4532 p=0.612

SOL F=18.8979, F=80.2, F=26.81, F=25.67, F=10.63, F=2.18,
(n=13) p=0.0003* p=0.0001* p=0.0002* p=0.0003* p-0.0068* p=0.1656

SOT F=40.3882, F=169.38, F=56.07, F=50.2, F=6.39, F=0.83,
(n=27) p=0.0001* p=0.0001* p=0.0001* p=0.0001* p=0.0179* p=0.3696


Note: indicates statistical significance at the 0.05 level. SOT = Combined total
responses from SOT

The data were also analyzed to determine the rate of change over time by sequential

evaluation of each successive time period in the experiment. This analysis evaluated the

changes of ICP values over time in order to detect the rate of return of MAP to baseline.

Tables 4.8 and 4.9 summarize the analysis of the difference between sequential values of

mean arterial pressure with both stimuli.









Table 4.8

Analysis of Sequential Differences in MAP Over Time with NBC


Peak and 30
second
average


30 second
average
and 1 minute


1 minute and
1-2 minute
average


1-2 minute
average
and 4-5 minutes


........ ............ ..................... ...................... ............ ............................................. .......... ...
NBR F = 66.68, F = 114.78, F = 1.51, F = 8.49, F = 3.17,
(n=13) p =0.0001* p =0.0001* p = 0.2421 p = 0.013* p = 0.1002
......... ..... .............. =. ^ ..... . ............ .iT. ............ .... .. ..f........................
NBL F= 107.15, F = 184.31, F= 18.14, F = 41.40, F = 5.12,
(n=12) p =0.0001* p = 0.0001* p = 0.0013* p = 0.0001* p = 0.0449*
......... ... ......... .... ..^...^ ^ ........ .. ... ... ........... .... .................. ....................... ....................
NBT F = 167.33, F = 289.35, F = 9.16, F = 27.37, F = 8.35,
(n=25) p = 0.0001* p = 0.0001* p = 0.0058* p = 0.0001* p = 0.008*


Note: indicates statistical significance at the 0.05 level. NBT = Combined total
responses from NBC


Table 4.9

Analysis of Sequential Differences in MAP Over Time with SOP

Stimulus Baseline Peak and 30 30 second 1 minute and 1-2 minute
Type and Peak second average 1-2 minute average
average and 1 minute average and 4-5 minutes
....................................... .......... .... ... aveKage ........... d lm inute and 4-5 m s.... .... d
SOR F = 91.48, F = 115.68, F = 6.10, F = 14.65, F = 1.80,
(n=14) p = 0.0001* p =0.0001* p = 0.0282* p = 0.0021* p= 0.2029
............i................ .......... ............ ......... ......... f i ............. ............ .................... ..................
SOL F = 80.20, F = 117.36, F = 17.10, F = 8.42, F = 18.58,
(n=13) p = 0.0001* p = 0.0001* p = 0.0014* p = 0.0133* p = 0.001*

SOT F = 169.38, F = 237.00, F = 21.14, F = 23.28, F = 10.19,
n=27 p = 0.0001* p = 0.0001* p = 0.0001* p = 0.0001* p = 0.0037*


Note: indicates statistical significance at the 0.05 level. NS = not statistically
significant. SOT = Total supraorbital stimulus data added together

After reviewing the data in Tables 4.8 and 4.9 on the sequential effects of MAP over

time, both NBC and SOP gradually decreased over time until baseline was reached. As

with MAP data compared to baseline in Tables 4.6 and 4.7, the MAP did return to


Stimulus
Type


Baseline
and Peak








baseline by the end of the fourth to fifth minute. The conclusion reached was that MAP

did rise statistically from baseline when NBC or SOP was administered, but declined with

each time period until MAP was at baseline at the last minute of the experiment.

The null hypothesis was rejected that there would be no increase in MAP with NBC

or SOP, and the alternative hypothesis was supported that there was an increase in MAP

from baseline with both NBC and SOP stimuli.

Research Hypothesis 3

The third hypothesis stated that there would be an increase in CPP when compared to

baseline values. This hypothesis was supported by the data analysis. Multivariate

repeated measures analysis of variance was used to detect changes in CPP in response to

NBC and SOP performed on the right side, NBC and SOP performed on the left side, and

combined responses from all NBC (NBT) and supraorbital pressure (SOT). Tables 4.10

and 4.11 summarize the values obtained over the timing intervals. All values are measured

in mm Hg, and include the mean and standard deviation (top values) and range of values

(bottom values). The CPP was calculated by subtracting the ICP from MAP. It is

understood that the values of CPP are directly influenced by the MAP and ICP, since it is

a mathematical derivation of both physiological parameters.









Table 4.10

Descriptive Statistics of CPP Changes With NBC


Note: NBT = Combined total responses from NBC.


Table 4.11

Descriptive Statistics of CPP Changes With SOP


Stimulus Baseline Peak 30 Average Average Average Average
Type seconds 30 seconds 1 minute 1-2 minute 4-5 minute
M SD M SD M SD M SD M SD M SD
SOR 75.15 14.9 100.19 16.37 84.39 13.79 82.27 12.9 77.86 12.27 74.56 13.53
(n=14) (51.27-101.8) (73.92-121.64) (60.8-107.41) (62.13-104.92) (61.42-102.66) (50.94-99.99)
SOL 75.03 12.78 106.41 16.79 87.67 14.87 85.17 13.93 80.93 15.00 76.24 15.00
(n=13) (52.67-102.59) (89.2-145.32) (64.45-112.5) (63.07-108.67) (58.28-106.29) (51.17-105)
SOT 74.85 13.39 103.02 16.29 85.81 13.92 83.52 13.02 79.15 13.32 75.19 13.8
(n=27) (51.27-102.59) (73.92-145.32) (60.8-112.5) (62.13-108.67) (58.28-106.29) (50.94-105)

Note: Note: SOT = Combined total responses from SOP

Using multivariate repeated measures analysis of variance (MANOVA), the above

values were statistically tested for significant changes that occurred within subjects with

each stimulus type. Baseline values of CPP were compared with CPP in each of the time

period values. This analysis determined whether the CPP values were different from


baseline. Tables 4.12 and 4.13 summarize the MANOVA results on CPP.


Stimulus Baseline Peak 30 Average Average Average Average
Type seconds 30 seconds 1 minute 1-2 minute 4-5 minute
M SD M SD M SD M SD M SD M SD
NBR 76.47 13.95 95.57 16.15 83.47 15.04 81.7 14.42 78.9 12.93 76.15 13.06
(n=13) (51.27-102.34) (73.92-122.08) (60.8-110.54) (62.13-107.47) (61.66-102.8) (54.71-102.63)
NBL 75.30 12 99.19 14.03 85.98 12.65 84.2 12.92 80.73 13.54 76.33 13.19
(n=12) (55.84-100.01) (77.2-123.84) (65.44-109.48) (67.03-107.05) (63.37-102.41) (53.36-100.33)
NBT 75.91 12.79 97.31 14.97 84.68 13.72 82.9 13.5 79.78 12.98 76.24 12.85
(n=25) (51.27-102.34) (73.92-123.84) (60.8-110.54) (62.13-107.47) (61.66-102.8) (53.36-102.63)








Table 4.12

Statistics on CPP Changes from Baseline With NBC


Stimulus Change in Baseline Baseline Baseline Baseline Baseline
Type CPP and Peak and 30 and One and 1-2 and 4-5
within 30 seconds second minute minute minute
subjects average average average average
over time
NBR F = 17.1688 F = 54.28, F = 18.46, F = 14.22, F = 6.87, F = 0.10
(n=13) p = 0.0004* p =0.0001* p =0.001* p = 0.0027* p = 0.0223* p = 0.7628

NBL F = 15.2927, F = 89.75, F = 43.96, F = 26.02, F = 9.24, F = 1.17,
(n=12) p =0.0012* p =0.0001* p =0.0001* p =0.0003* p =0.0112* p= 0.3025

NBT F = 35.2843, F = 135.27, F = 54.83, F = 37.34, F= 14.72, F = 0.22,
(n=25) p = 0.0001* p =0.0001* p =0.0001* p = 0.0001* p = 0.0008* p = 0.647


Note: indicates statistical significance at the 0.05 level. NBT = Combined total
responses from NBC.


Table 4.13

Statistics on CPP Changes from Baseline With SOP

Stimulus Change in Baseline Baseline Baseline Baseline Baseline and
Type CPP within and Peak and 30 and One and 1-2 4-5 minute
subjects 30 seconds second minute minute average
over time average average average
SOR F =13.1256, F = 54.8, F = 22.74, F = 18.3, F=1.81, F = 0.32,
(n=14) p =0.0011* p =0.0001* p =0.0005* p =0.0011* p=0.2036 p= 0.5842

SOL F = 13.31 F = 63.39 F = 34.66 F = 39.04 F = 15.90 F = 1.47
(n=13) p = 0.001* p = 0.0001* p = 0.0001* p = 0.0001* p =0.0018* p= 0.2487

SOT F = 30.2026, F = 124.58, F = 61.13, F = 58.4, F = 12.49, F = 0.23,
(n=27) p = 0.0001 p = 0.0001* p= 0.0001 p = 0.0001* p = 0.0016* p= 0.635


Note: indicates statistical significance at the 0.05 level. SOT = Combined total
responses from SOP








In evaluating the data for CPP presented in Tables 4.12 and 4.13, there was a

significant difference in CPP from baseline within subjects over time. Although the CPP

did peak early and continued to decline, their rate of decline is less dramatic over time.

However, at the fourth to fifth minute time interval, all CPP values were back to baseline.

This indicated that the CPP had more of a sustained effect after introduction ofNBC or

SOP, but the values were back to baseline before five minutes.

The data were also analyzed to determine the rate of change over time by sequential

evaluation of each successive time period in the experiment. This analysis evaluated the

rate of change of CPP values over time in order to determine the rate of return of CPP to

baseline. Tables 4.14 and 4.15 summarize the MANOVA analysis of difference between

sequential values of CPP.


Table 4.14

Analysis of Sequential Differences in CPP Over Time With NBC

Stimulus Baseline Peak and 30 30 second 1 minute 1-2 minute
Type and Peak second average and average
average and 1 minute 1-2 minute and 4-5
average minutes
NBR F = 54.28, F = 104.15, F = 11.53, F = 10.54, F = 6.32,
(n=14) p = 0.0001* p = 0.0001* p = 0.0053* p = 0.007* p = 0.0272*

NBL F = 89.75, F = 97.98, F = 9.21, F = 34.15, F = 12.98,
(n=12) p = 0.0001* p = 0.0001* p = 0.0113* p = 0.0001* p = 0.0042*

NBT F = 135.27, F = 207.18, F= 21.52, F = 35.34, F = 18.8,
(n=25) p = 0.0001 p = 0.0001* p = 0.0001* p = 0.0001* p = 0.0002*


Note: indicates statistical significance at the 0.05 level. NBT = Combined total
responses from NBC