Modulation of airflow-controlling mechanisms in newborn infants breathing carbon dioxide


Material Information

Modulation of airflow-controlling mechanisms in newborn infants breathing carbon dioxide
Physical Description:
vii, 81 leaves : ill. ; 29 cm.
Wozniak, John Alexander, 1957-
Publication Date:


Subjects / Keywords:
Infant, Newborn -- physiology   ( mesh )
Respiration -- physiology   ( mesh )
Respiration -- Child   ( mesh )
Respiration -- Infant   ( mesh )
Carbon Dioxide -- physiology   ( mesh )
Physiology thesis Ph.D   ( mesh )
Dissertations, Academic -- Physiology -- UF   ( mesh )
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph.D.)--University of Florida, 1989.
Bibliography: leaves 72-80.
Statement of Responsibility:
by John Alexander Wozniak.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001468960
oclc - 22252916
notis - AGY0639
System ID:

This item is only available as the following downloads:

Full Text







I am very grateful to Drs. Philip Kosch, Paul Davenport, Marc

Jaeger, Wendell Stainsby, and Floyd nhampson for their critical review

of this dissertation and guidance in this research. I thank Dr. Koech

in particular for his contribution of time, facilities, resources, and

advice throughout my graduate studies. I thank Dr. Alastair Hutchison

for his assistance in recruiting infants and his comments on this


I am indebted to Ms. Karin Cross for her assistance in

performing the experiments and analyzing much of the data. Finally, a

special word of thanks to my wife for her understanding and support of

my endeavors.

These studies were performed within the Department of

Physiological Sciences, College of Veterinary Medicine, University of

Florida, Gainesville. Financial support was provided by a National

Institute of Health Grant #5 R01 HL28617-06.



ACNOWEDGMENTS......................................... ii

SYMBOLS AND ABEEV-IATINS................................ iv

ABSTRACT..................... ... ...... ............. vi


1 INTRODUCTITr........................................ 1

2 ITERATURJ REVIEW................................. 3

DURING 002 BREA7RIN IN NONAMTES................. 8

Experimental Design and Methods.................... 9
Subjects......................................... 9
Study Protocol .................................. 9
Maements..................................... 10
Data Analysis........................ ...... 10
Results...........o ...........oo o oo .... o..... o........... 12
Discussion......................................... 19

IIURING 002 BEAII IN NEATES. ................. 29

Experimental Design and Methods..................... 31
Subjects and Me!sureents........................ 31
Data Analysis................................ ... 31
Results ........................................ ..... 33
Discussion................................... ....... 42

DURATICN......................................... 49

The Inspiratory Off-Switch.......................... 58
Inspiratory-Epiraty Coupling............... ........ 62
Proposed Model of Respiratory Control...............66
Future Studies...................................... 69
Conclusions......................................... 70

REFERENCES............................................... 72

BIOGRARHICAL SKETCHO.......................................... 80


CIA Central inspiratory activity

CII Central inspiratory inhibitory

002 Carbon dioxide

CPG Central pattern generator

Di Diaphragm

EEV End-expiratory volume

Em Electranmyogram

f Frequency

Ic Intercostal

inspDI Average inspiratory activity of the diaphragm

inspIC Average inspiratory activity of the interoost

inspPCA Average inspiratory activity of the posterior
cricoarytenoid muscles

MMA Moving-time average

PCA Posterior cricoarytenoid muscle

PCAb Baseline expiratory activity of the posterior
cricoarytenoid muscles

PIIA Post-inspiratory inspiratory activity

PIIAdi Post-inspiratory inspiratory activity of the c

p)Ke Peak expiratory flow rate

pkWi Peak inspiratory flow rate

PSR Slowly adapting pulmnary stretch receptor

r Correlation coefficient

RAR Rapidly adapting receptor


TA Thyroarytenoid muscle

Te Expiratory time

Ti Inspiratory time

TPIIAdi Duration of post-inspiratory inspiratory activity of
the diaphragm

TISS Time to reach steady-state

UA Upper airway

4e Mean expiratory flow rate

fi Mean inspiratory flow rate

ain Minute ventilation

Vt Tidal volume

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




December, 1989

Chairman: Dr. Philip Kosech
Major Department: Physiology

The effects of carbon dioxide inhalation on the breathing

pattern and on mechanisms that control this breathing pattern in the

neonate are investigated in this dissertation. Diaphragm, intercostal

and posterior cricoarytenoid muscle activities were measured along

with respiratory airflow and volume changes in eight full-term newborn


Infants breathing room air exhibited breathing patterns with

varying degrees of inspiratory pump and laryngeal muscle activities.

Diaphragmatic airflow retardation (braking) during early expiration

was observed in six of eight infants. A mixture of 2-3% carbon dioxide

in room air was used as a stimulus, and a steady-state response was

measured during the fourth minute of the stimulus. The observed

breathing patterns with carbon dioxide were qualitatively similar to

those during room-air breathing. Tidal volumes were elevated with

little change in frequency of breathing, resulting in an average 38%

increase in minute ventilation.

Breathing carbon dioxide augmented inspiratory muscle activity

of the posterior cricoarytenoids (44%), the intercostals (17%) and the

diaphragm (11%). 'he relative responses of these muscles varied

greatly between individuals. No respiratory abdominal muscle activity

was recorded at any time during the experiments. Breathing carbon

dioxide did not change diaphragmatic post-inspiratory inspiratory

activity, but elevated expiratory posterior cricoarytenoid muscle

activity to a degree (40%) similar to that of the inspiratory


These experiments contribute important information on

respiratory control in the neonate, including 1) the newborn infant

adopts varied breathing strategies to defend absolute lung volume

including the braking of expiration with the diaphragm and laryngeal

muscles, 2) the newborn has the ability to augment minute ventilation

in a variety of ways when challenged with carbon dioxide, 3) improving

diaphragm-ribcage coupling through intercostal muscle activation may

provide an energy-efficient means of enhancing pump muscle

performance, 4) the defense of absolute lung volume is of critical

importance to the newborn as demonstrated by the persistance of

diaphragmatic braking of expiration throughout this chemical

challenge, and 5) the modulation of laryngeal resistance to airflow by

posterior cricoarytenoid activation plays a crucial role in the

ability of the newborn to respond to a chemical load while maintaining

absolute lung volume.



The purpose of this study is to investigate the effects of CO2

irhalaticn on the breathing pattern and mechanisms that control this

pattern in the neonate. The respiratory control system in rnewbons

differs from that of adult humans and other mammalian species. In same

ways the newborn is more similar to mammalian species other than

human, and much investigation into the respiratory control system of

different mammalian species has been performed. It is preferable to

study the human neonate itself, when possible, rather than to make

inferences from an animal model. Only in this manner can questions

about the human infant be unequivocally answered.

The effects of chemical stimulation on reflex modulation of

airflow-cntrolling mechanisms are investigated in the newborn infant.

Specifically, experiments are designed to study both transient and

steady-state responses to carbon dioxide (C02) inhalation as they

affect posterior cricoarytenoid (PCA) muscle activity, punp muscle

activity (diaphragm and inspiratory interoostals) and airflow pattern

in the healthy newborn infant while asleep. This includes the effects

of 0D2 inhalation on inspiratory PCA activity, on pump muscle

activities, and on expiratory braking mechanisms in newborn infants.


This research shows how a chemical factor shapes the

respiratory pattern in the unanesthetized sleeping infant, and it

provides useful information on neuramuscular strategies employed by

newborn infants to maintain minute ventilation (Qin) and absolute

lung volume. Indeed, many of the life-threatening breathing disorders

faced by the neonate involve changes in the control of breathing

studied in this dissertation. To better understand the genesis of

disordered breathing, one must first define the mechanisms controlling

respiration in the normal developing infant.

A review of the literature relating to hypercapnia and the

control of respiration is presented in Chapter 2. This is followed by

studies of the response of newborns to inhaled CO2. Effects of this

stimuli on inspiration and expiration are treated separately in

Chapters 3 and 4, respectively. The final chapter addresses the

integration of the inspiratory and expiratory results into a

description of the response of the newborn to CO2 inhalation.

Additionally, the relationship of these results to previously proposed

models of respiratory control, and speculations on mechanisms

underlying the observed responses, are discussed.


The control of breathing in newborns differs from that in

adult humans and animals (101,99). Maturational changes in chemical

control, mechanical reflexes and the central integrative abilities of

the respiratory neuronal network have been observed but are

inadequately understood. Behavioral states and respiratory reflexes

have been studied for many years, but the integration of these two

areas of research is recent. The majority of the speculation on

respiratory control mechanisms employed by babies comes from animal

studies. These studies often use invasive techniques, and correlations

to the control system in the infant are uncertain due to effects of

anesthesia (40,53,92,103), relative maturation, and possible species


Sleeping infants, unlike adults, exhibit a eupneic breathing

strategy to defend absolute lung volume (36,62,63,73,98). Newborn

infants have a highly compliant chest wall that aids in parturition

(42), but results in a low relaxation volume (10-15% of total lung

capacity as ccupared to 30-35% for the supine adult). Breathing from

this low lung volume could result in decreased oxygen stores (35) and

put the infant at risk of small airway closure that may result in

atelectasis of terminal lung units. The newborn normally adopts a


breathing strategy to compensate for this mechanical disadvantage by

establishing an end-expiratory lung volume (EEV) above the relaxation

volume. The newborn can defend EEV by shortening expiratory time (Te)

and/or by actively braking expiration to retard expiratory airflow. By

shortening Te, expiration is normally interrupted before relaxation

volume is reached. Methods of braking expiration include dynamic

narrowing of the upper airways (UA) and post-inspiratory inspiratory

muscle activity (PIIA) (43,66). Evidence suggests that there is an

integrative neural mechanism that modulates Te which is sensitive to

both EEV and the rate of lung deflation (59). By retarding expiratory

airflow, the rate of lung deflation is decreased and Te is prolonged.

The net effect of braking on EEV depends on the interaction between

the mechanical effect to elevate EEV and the opposing effect of Te


There is little information available on the spontaneous

occurrence of braking and its neuzamedhanical oonsequences in infants.

Radvanyi-Bouvet et al. (80) studied expiratory patterns in normal

full-term and preterm infants and found that expiration was frequently

retarded or airflow was interrupted, but they did not make direct

observations of specific braking mechanisms. It appears that even

during periods of regular breathing, infants use PIIA early in

expiration to oppose rapid emptying of the lungs, and that scue degree

of UA braking provides a variable, controlled resistance to airflow

throughout expiration.

It has been suggested that an important function of the larynx

may be to brake expiration to provide an optimal resting respiratory

frequency without an end-expiratory pause (81). In the event of an


increase in ventilatory needs, the brake could be removed to provide

lower resistance and increased ventilation. There is recent evidence

that normal adults use their larynx in this manner. The normal pattern

of vocal cord movement in the adult consists of a slight increase in

laryngeal resistance during expiration due to vocal cord addiction

(28). This expiratory narrowing in adults does not result in an EEV

above functional residual capacity as observed in infants. England and

Bartlett (27) found that this expiratory laryngeal braking is reduced

during hypercapnia and exercise in normal adults. It is reasonable to

assume that the newborn infant uses the larynx to control airflow when

challenged by chemical loads, as well.

The breathing strategy employed by the newborn infant to

increase ventilation during CD2 inhalation is quite variable and

different from that used by the adult. The adult shortens both

inspiratory time (Ti) and Te and increases tidal volume (Vt) during

C02 inhalation (39). The response of infants breathing 2 appears to

depend on sleep state and gestational age. Several groups of

investigators have studied the effect of C02 inhalation on ventilator

parameters in full-term infants, and have arrived at different

conclusions. All investigators agree that there is a marked increase

in minute ventilation with CO2 inhalation, but there is disagreement

as to the contribution of frequency and Vt to this increase. Some

investigators found an increase in both Vt and frequency

(10,21,52,64,69), others noted an increase in Vt alone

(2,22,25,45,94,97), while still others found the response to be quite

variable and dependent on sleep state (8,9,84). In addition to these

differences, there is debate as to the influence of sleep state on the


magnitude of the ventilatory response to 002 Sme investigators have

found a reduced ventilatory response to 002 in REM sleep (2,15,46,52)

while others found similarity in the 002 response in REM and NRWM

sleep (22,32,45). Overall, most investigators have observed an

increase in Vt with no timing change due to 02 a ring

nanREM sleep.

Studies in premature infants have found that the response to

002 is less than in the full-term (64,70,85), but there is
disagreement as to how this change in ventilation is mediated and what

effect sleep state has on the aD2 response. Scme investigators have

found that both frequency and Vt contribute to the ventilator

response (21,22,24,26,85). Others have found only an increase in Vt

(10,37,64,86,87). Others, still, have found the response to be

variable and dependent on sleep state, breathing pattern, or time

course of the stimulus (8,9,55,67). Two things that all investigators

appear to agree on is that there is a graded increase in ventilation

with increasing inspired 002 concentration, and that premature infants

respond less than full-term infants.

Investigations to elucidate the medcanisms responsible for the

increase in ventilation with 002 inhalation have led to the recording

of the activity of respiratory pump muscles and airflow-controlling

muscles during C02 inhalation in babies. There is an increase in pump

muscle activity during 002 inhalation (17,46,84), and an increase in

activity of two upper airway dilating muscles (17,88), the alae nasi

and genioglossus muscles. Since the larynx has been implicated as a

primary site for airflow control, the response to 002 of the principal


laryngeal abductors, the CA muscles, should be examined along with

their interaction with pump muscle activation during 002 inhalation.

The mechanisms responsible for changes in the ventilator

pattern of the neonate due to D02 breathing are not fully understood.

Experiments performed on animal models allow more invasive techniques

to be used to examine the respiratory control system, and to this end,

models have been proposed and tested. It is difficult to relate

decerabrate or vagotacized animals to sleeping infants, yet these

animal studies form a framework on which a general model describing

the mechanisms controlling neonatal respiration can be built. The

following studies examine the mechanisms responsible for the

ventilatory pattern of the sleeping neonate and how these mechanisms

are influenced by OD2 breathing.


The ventilator response of the term infant to inspired 002

during quiet sleep consists of increased tidal Vt with no change in

frequency (2,22,25,45), resulting in increased flow rates. These

increased flows can be achieved by increasing the activity or

effectiveness of the respiratory pump, or by decreasing resistance to

airflow. In premature (16,17) and term infants (61), as well as adults

(76,77,96), the UA muscles are activated in synchrony with respiration

and are critical in modulating airflow. The intrinsic laryngeal

muscles produce changes in UA resistance by altering laryngeal caliber

in animals (4,6,23) and adult humans (12,13). The paired PCA muscles

abduct the vocal cords and exhibit primarily inspiratory burst

activity in both adult humans (12,13) and newborn infants (16,61). A

lowered resistance to airflow during the inspiratory phase of

spontaneous respiration has been well documented in several species,

including humans (28).

During CD002 inhalation, electraygraphic studies in premature

infants have documented increased diaphragmatic (17,46,84),
genioglossus (88), alae nasi (17) and laryngeal (16) muscle

activities. The percent increase in minute ventilation during OD2

breathing is less for the premature infant as compared to the term

infant, although both populations exhibit an increased Vt with 002


breathing during quiet sleep (10,70). Electrc uyographic inflation is

lacking in the term infant; thus, experiments were designed to

determine the roles of the laryngeal and pump muscles in the

ventilator response of the term neonate to 002 inhalation.

Experimental Design and Methods


Subjects consisted of 8 healthy term infants with gestational

age 39.9 + 0.4 wk (mean + SD), postnatal age 55 + 23 hr and

birthweight 3.4 + 0.3 kg. Delivery was by cesarean section for 4

subjects and vaginal for the other 4. Four subjects were male and 4

were female. Informed parental consent was obtained, and a

pediatrician and nurse were present throughout the studies.

Study Protocol

Studies were performed 1 hour after a feeding when the infant

was sleeping supine. The head was turned to one side and in a neutral

position with little flexion or extension of the neck. Sleep state was

assessed by standard behavioral criteria (79) and confirmed in four

subjects by measuring the electrooculogram with gold cup surface

electrodes. Subjects were stimulated with a mix of 2.4 + 0.4% 002 in

raon air by fixing a bag filled with the gas to the inspiratory port

of a two-way valve attached to the facemask. The total deadspace of

the system was less than 3 ml.

Studies consisted of at least 2 minutes of room-air breathing

followed by at least 4 minutes of CD2 breathing, with another 2-minute

control period after the CD2 run. Breathing runs were used for

analysis if 90% of the ccnputed steady-state ventilatory response was

reached (see Data Analysis). The level of 002 used for each infant was


determined by the response of that individual. The first run consisted

of approximately 3% 002. If the infant was aroused by this mixture, a

2% 002 mixture was added to the 3% mixture in the bag to decrease the

inspired 002 level. Only runs where the subject remained in nonREM

sleep throughout the sampling period were included for analysis.


Diaphragm, intercostal and abdominal muscle electrc~yograms

(XsB) were measured with bipolar skin surface electrodes. Diaphragm

leads were placed over the sixth and seventh intercostal spaces in the

right anterior axillary line. Intercostal leads were placed

parasternally over the second or third rib space. Electrodes were

placed on the abdomen just to the right of the navel. The PCA muscle

EM was measured with bipolar silver electrodes built into a 5-French

infant feeding tube (61). This tube was placed in the esophagus

immediately posterior to the larynx where the PCA muscles are located.

A fast-response 002 analyzer (Beckman IB-2) with a nasal

catheter was used to continuously measure CO2 concentrations. Airflow

and airway pressure were measured using a previously described low-

deadspace facemask-pneumotach system (61) and differential pressure

transdcxers (Validyne MP45). Volume was obtained by digitally

integrating the flow signal.

Data Analysis

Data were digitized and analyzed using a laboratory computer

(DEC LSI-11/23). IMscle ENs were digitized at a sampling frequency of

1 kHz and all other signals were sampled at 200 Hz. A 100 ms moving

time average (MMA) was computed for each EM after the electrocardio-

gram artifacts in the diaphragm and intercostal EMs were subtracted


using the technique of Bloch (7). Values were computed continuously

and averaged over 5-second intervals to determine the steady-state

response time. The 15-second averages of Vt and Imin for the time of

CD2 breathing were fitted to the equation y=a-b*exp(-c*t) using a

least-squares nonlinear assymptatical regression (Gauss-Newton

algorithm) (67). In this equation, a is the steady-state level, a-b is

the value at time t=0, and c is the rate constant. From this equation,

90% of the computed steady-state response was calculated.

Onset of WE activity and neural inspiratory time were

determined from the MIA for each muscle by computer-assisted detection

of increasing and decreasing slope as previously described (61). The

mean inspiratory activity of the diaphragm (inspDI), intercostal

(inspIC) and PCA (inspPCA) muscles were computed as the area under the

corresponding MIA above electrical zero during neural inspiratory

time, divided by that time.

Data were combined into 60-second epochs which consisted of 2

control epochs, 4 or 5 epochs of C02 breathing and up to 2 control

epochs following CO2 inhalation. To determine statistical differences

between epochs, a repeated-measures ANOVA with Tukey's multiple range

test was used. Simplifying the ANOVA procedure necessitated taking an

equal number of breaths from each epoch. The number of breaths used

differed between individuals, but was kept constant within a subject

(ranging from 28-46 breaths/epoch). The number of breaths/epoch for a

subject equaled the lowest number of breaths in any epoch for that

individual. Within an epoch, breaths were chosen from the middle so

that an equal number of breaths were eliminated from the beginning and


end of a period. All correlations were determined using a least-

squares method. Group correlations were performed on the percent

changes in muscle activities and ventilatory parameters (n=8

subjects). To analyze the relationship between muscle activities and

changes in flow rates within individuals, simple and multiple

least-squares linear regressions were performed sequentially on the

15-second period data for 1 minute before and 4 minutes during 002

breathing (n=20 time periods), for each subject. Mean inspiratory (vi)

and peak inspiratory (pkvi) flow rates, as well as Vt and rinin were

used as dependent variables, and EM3 data were used as independent



All 8 subjects reached 90% of the computed steady-state

ventilatory response by the fourth C002 epoch. Times to reach 90% of

steady-state were 98 + 87 s (mean + SD) for V'in and 110 + 100 s

for Vt. Thus, all subsequent comparisons of 00C2 to control values will

refer to the fourth 002 epoch. The 002 stimulus was sufficient to

raise end-tidal CO2 significantly from a control level of 38.3 + 1.4

to 40.2 + 1.3 torr (mean + SE). Average values for mechanical

variables during roanm-air and 002 breathing are shown in Table 3-1 in

absolute units. Table 3-2 lists electrayographic variables in

standardized units (percent change) and compares them to standardized

mechanical variables. There was no change found in frequency of

breathing or respiratory cycle timing, but there was a 42% increase in

Vt during CD2 inhalation (Tables 3-1 and 3-2). The increased Vt

resulted in a 38% increase in Vmin. Changes in tnin were highly

correlated (r=0.94, P<0.001) with changes in Vt. The augmentation of

Mechanical Variables During Roam-Air and CO2 Breathing

TTSS (sec)


230 + 22

5.0 + 0.6

8.5 + 0.9

12.3 + 1.4

48 + 4

0.58 + 0.03

0.71 + 0.07

*318 + 30

*6.9 + 0.7

*11.5 + 1.0

*17.5 + 1.6

47 + 3

0.60 + 0.03

0.72 + 0.06








Values are means + SE for the minute preceding and the
fourth minute of 00 breathing. Significant difference from room-air
breathing (P<0.05) Inicated with "*".













+ 31

+ 35

+ 43

+ 23

Inspiratory Electrcyographic and Ventilator3

Subject inspDI inspIC inspPCA

x + SE











































y Variables












Values are expressed as percent change fram rowm-air
breathing. Significant changes (P<0.05) are indicated with "*".

l I m m

Vt with no change in timing resulted in elevated vi and pk'i (Table

3-1). Changes in tuin were also highly correlated with increases in

flow rates (r=0.96 for pkfi, r=0.97 for vi, P<0.0001).

Mean inspiratory pump muscle activities were elevated, with

inspDI and inspIC increasing an average of 11 + 3% and 17 + 5%,

respectively. Both parameters had a similar rate of rise to reach

steady-state (TTSS) as 4tin and Vt (Table 3-2). There was also a large

increase in inspPCA (44+12%) that exhibited a somewhat longer time

to reach steady-state (Table 3-2). Activity in phase with respiration

was not recorded at any time from the abdominal leads.

The strategy adopted by an individual to increase nmin varied

considerably. Tracings of 3 different strategies are displayed in

Figure 3-1; predominantly elevated PCA activity (Panel A),

predominantly increased pump muscle activity (Panel B) and increased

pump muscle combined with elevated PCA activity (Panel C). As shown in

Table 3-2, 4 subjects (nos. 1,6,7 and 8) exhibited increased inspPCA

with insignificant changes in inspDI. Two of these 4 (nos. 1 and 6)

also showed increased inspIC. One subject (no. 2) only increased

inspDI and inspIC, and 3 subjects (nos. 3,4 and 5) increased all

inspiratory muscle activities.

Increases in inspDI were significant in 4 subjects (Table

3-2). Thus, group analysis found a weak correlation between inspDI and

vi (r=0.43, ns) (Figure 3-2). Within individuals, inspDI was

correlated (P<0.05) with vi in 4 subjects (r=0.75 + 0.06, mean+rSD),

with pkvi in 5 subjects (r=0.76 + 0.14), and with Vt in 4 subjects

(r=0.68 + 0.10). An example of an individual regression is

illustrated in Figure 3-3.

FIGURE 3-1: Breathing Strategies with CD2 Breathing.
Representative examples of 3 strategies to increase
ventilation during D02 breathing (right tracings) as compared to rocm
air (RA) breathing (left tracings). Strategies include increased PCA
activity (Panel A, subject 1), increased pump muscle activity (Panel
B, subject 2) and increased activity of all muscles (Panel C, subject
3). Shown in each panel are the volume spirogram, PCA EM, intercostal
(Ic) EM3 and diaphragm (Di) E13. Values to the right, starting at the
tcp, are % changes from room-air breathing for Vt, inspPCA activity,
inspIc activity and inspDi activity.


Volume CMNV""

OW w- -, Mhl A- -A&~Li _, ILl LI ILIJ l,16 idflk.

-v$4 44 Lot,29%

0 N 4%









14 %



Time -
I sec

M 67%

,r,_ 66%

000*00OV0 26%X
. k& Ih .i A, I"
-r--", '", ,177,,. p



..,.,i ..1-'116"166 ,,16 j JIU. i I. -., ,w A
r-, -- /

w 6*~40 6 6 i






r=0.43, NS

2 -



% Change

Mean Insp.

FIGURE 3-2: Airflow and Diaphragm Activity Changes.
Values are expressed as percent change from rocm-air
breathing. Points are mean + SE for each of 8 subjects. A regression
line is plotted ( _j with a correlation coefficient (r). Points
right of the vertical line (-) are significantly elevated diaphragm
activities (P<0.05). All values for fi are significantly elevated.







I i


Increases in inspIC were significant in 6 subjects (Table

3-2). Group analysis found a good correlation between inspIC and vi

(r=.77, P<0.05) (Figure 3-4). Within individuals, inspIC was

correlated (P<0.05) with vi in 4 subjects (r=0.58 + 0.08), with pkJi

in 5 subjects (r=0.61 + 0.10), and with Vt in 5 subjects (r=0.61 +


When inspPCA was analyzed, 7 subjects exhibited a rise in

activity (Table 3-2), and the increased activities were correlated

with changes in vi (r=.66, P<0.05) (Figure 3-5). Individual

regressions found inspPCA correlated (P<0.05) with vi in 6 subjects

(r=0.65 + 0.14), with pk&i in 7 subjects (r=0.69 + 0.18), and

with Vt in 7 subjects (r=0.66 + 0.14).

Multiple regressions were performed within individuals in an

attempt to better define changes in inspiratory flow rates. When

inspDi, inspIC and inspPCA were combined as independent variables in

these regressions, good correlations with vi (range=0.78-0.95,

mean=0.85 + 0.06 (SD)) and with pk)i (range=0.76-0.95, mean=0.89 +

0.07) were found in all 8 subjects.


The healthy, term infant in nonREM sleep augmented %nin during

CO2 breathing by increasing Vt with no change in frequency or timing

as previously reported (2,22,25,45). In this study, it was found that

the breathing strategy employed by an individual to meet this

challenge varied considerably. Large differences in the relative

contributions of the various respiratory muscle activities to the

hypercapneic response of the present subjects were observed; yet

despite these differences, the combinations of muscle activities

3 14, 5
17 18


5- 1

0_ I I I I





.Mean Insp.

FIGURE 3-3: Example of an Individual Regression.

Data is from Subject 2, showing the relationship between
inspDI and vi. Points are mean values for successive 15-second
intervals starting 1 minute before CO breathing and ending after 4
minutes of C00 breathing (20 points t)tal). The correlation
coefficient for this regression was 0.83 (P<0.0001).








80 r .77 P.05
r=0.77, P<0.0500

60 F


/ -.
:i-- 6----






% Change

Mean Insp.

FIGURE 3-4: Airflow and Intercostal Muscle Activity Changes.
Values are expressed as percent change from room-air
breathing. Points are mean + SE for each of 8 subjects. A regression
line is plotted (_ _) with a correlation coefficient (r). Points
right of the vertical line (-) are significantly elevated
intercostal muscle activities (P<0.05). All values for 1i are
significantly elevated.




30 I














z L

0 25 75 100
0 25 50 75 100

% Change

Mean Insp

FIGURE 3-5: Airflow and PCA Muscle Activity Changes.
Values are expressed as percent change from roca-air
breathing. Points are mean + SE for each of 8 subjects. A regression
line is plotted ( ) with a correlation coefficient (r). Points
right of the vertical line (-) are significantly elevated PCA muscle
activities (P<0.05). All values for fi are significantly elevated.





resulted in similar ventilatory responses of increased Vt with no rate

or timing change for all subjects.

A change in ventilatory pattern of elevated Vt without timing

changes necessitates increased 4i. Inspiratory airflow can be

increased through augmented pump muscle activity and/or decreased UA

resistance. An increase in activity of the diaphragm, the primary pump

muscle of inspiration, was observed along with increased inspiratory

parasternal intercostal muscle activity. Overall, changes in inspIC

correlated better with 4i than did inspDI, and within individuals,

inspIC correlated with Vt in 5 subjects as compared to 4 for

inspDI. This is consistent with studies in adult humans where it was

observed that the increased Vt during 002 breathing was closely linked

to recruitment of the rib cage (78) as measured with magnetometers.

These data also support the conclusions reached in a study on term

infants where it was found that the slope of the 002 response curve

during rebreathing was determined primarily by rib cage recruitment

(52) as measured by respiratory inductive plethysmography. In addition

to direct inspiratory action, the increase in inspIC observed may

reflect a stiffening of the chestwall to improve mechanical coupling

between the diaphragm and rib cage. This would improve the inherent

instability of the neonatal chestwall and make the diaphragm a more

effective pump (50,66,73). All 4 subjects that increased inspDI with

002 breathing also increased inspIC (Table 3-2), but 2 subjects
increased inspIC without a significant increase in inspDI (subjects 1

and 6). Thus, increased inspiratory airflow can be achieved with

little change in diaphragmatic activation by improving coupling of the

diaphragm and rib cage. Additionally, this suggests that augmented

intercostal muscle activation is necessary for increased inspm to be
effective in elevating Vt.

Increased inspPCA is well correlated with inspiratory flow

(Figure 3-5). This suggests active laryngeal dilation during
CD2 breathing in most subjects. This is consistent with observed

increases in the activity of UA muscles with hypercapnia in animals

(5,6,29,49,91) and adult humans (12,13,76,77,96), as well as infants.

Specifically, investigators have observed increased genioglossus
activity in term infants (88) and elevated alas nasi (17) and

laryngeal muscle (PCA) activities (16) in premature infants with

hypercapnia. In the present study an increase in measured PCA activity

of 44% was found. This is more than twice the increase in pump muscle

activity (11 and 17%), and is consistent with the animal and adult

human data. Carlo et al. (16) observed that peak diaphragmtic

activity increased 50% and peak laryngeal EM increased 142% in
premature infants breathing 4% 002 in ron air. Ihe relatively larger
response of the PCA muscles to 002 breathing, in addition to a more

consistent correlation with inspiratory flow, suggests that control of

laryngeal caliber plays a vital role in elevating inspiratory airflow

rates in the newborn. T ~meet an increased ventilator demand, tapping

the intrinsic flow reserves afforded by baseline laryngeal caliber

provides an efficient means to bring about increased inspiratory
airflow, thus sparing the inspiratory pump muscles.

A striking feature of this study was the variation found in
breathing strategies employed by individuals in response to CD2

challenge, perhaps reflecting an imature respiratory control system.

Subject no. 1 had the largest increase in Vt and greatest ventilator


response (Table 3-2). Little change in inspDI was observed, while this

infant had the largest increases in inspIC and inspPCA (Figure 3-1,

Panel A). Thus, a combination of increased chestwall stability and

active laryngeal dilation, without a significant increase in inspDI,

appeared to be more than adequate to significantly increase Vt and

'Vin in this individual. This is illustrated in Figures 3-2, 3-4, and

3-5, where what appears to be an inadequate diaphragmatic response of

subject no. 1 (Figure 3-2) is offset by large increases in inspIC

(Figure 3-4) and inspPCA (Figure 3-5).

Infant no. 2 had the third largest increase in Vt and the

second greatest increase in 4min. Little change was observed in PCA

activity in this individual (Figures 3-1, Panel B, and 3-5), which may

reflect a lack of active laryngeal dilation in response to C002

Alternatively, although one cannot determine this from these results,

one would not expect a large increase if laryngeal caliber was near

maximum prior to the hypercapneic stimulus. In Figure 3-2 it can be

seen that, despite a small PCA response (Figure 3-5), subject no. 2

elevated fi 60%. This was accomplished by increasing inspDI and insplC

19% and 29% (Figures 3-2 and 3-4), respectively. This suggests an

additive effect of diapragmatic and intercostal activity in

increasing Vt and augmenting tuin.

Subject no.3 exemplifies a strategy of shared contributions

from all three muscle groups (Figures 3-1, Panel C, 3-2, 3-4, and 3-5)

to meet the 002 challenge, resulting in the second largest increase in

Vt and the third greatest increase in Vmin. Data from the remaining 5

subjects also show variability in individual breathing strategies in

response to hypercapnia.


The breathing strategy employed by a subject to respond to the

002 challenge may be predetermined by that subject's baseline rocm-air
breathing strategy. This baseline strategy also varies between

individuals. Particularly, the laryngeal caliber during roan-air

breathing directly effects the degree to which further dilation is

possible. It is conceivable that in healthy infants, and more so in

infants with respiratory disease, the response of the inspiratory pump

muscles is predetermined by their corresponding baseline activities.

Substantial activation during roam-air breathing would limit the

potential response of that muscle to the chemical load.

One would assume that the healthy infant would have flow

reserves afforded by the baseline activity of both UA and inspiratory

punp muscles. Submaximal baseline PCA activity associated with

submaximal laryngeal caliber would allow increased activation to

utilize this flow reserve during times of increased ventilatory

demands. Likewise, the punp muscles could also be augmented in this


The baseline and stimulated activities of UA and pump muscles

may be influenced by mechanical factors. In subjects with good

mechanical coupling of the ribcage and diaphragm, one might expect

proportional responses of all the muscles to an applied chemical load.

In subjects with poor mechanical coupling, two possible baseline

strategies may be employed that in turn may effect that subject's

potential OD2 response. A baseline breathing strategy of augmented

punp muscle activity to compensate for existing chestwall instability

may limit further increases in pump muscle activities in response to

an added chemical load. Thus, this individual may exhibit a

substantial increase in PCA activity to decrease UA resistance and

increase airflow during the CD2 challenge. Alternatively, an infant

with a baseline breathing strategy of increased PCA activity to allow

greater flow rates in the face of existing chestwall instability may

have no alternative but to increase pump muscle activation when faced

with a chemical load. These two scenarios are extreme cases, and of

course, a variety of baseline breathing strategies composed of varying

contributions of the pump and UA muscles are observed. Thus, a variety

of response strategies are also found.

The results of this study in healthy infants may provide

useful information for a better understanding of the mechanisms

employed by sick infants attempting to compensate for abnormalities

associated with respiratory disease. Very high ventilator demands are

often associated with respiratory disease. The UA may be critical in
modulating airflow in cases involving low lung ocmpliance where the

inspiratory pump muscles are considerably loaded.

A similar correlation can be drawn with premature infants

where shape instability of the ribcage is prominent and varies with

maturity. Premature infants may depend on UA mechanisms to adjust

airflow resistance when confronted with ventilatory demands in an

effort to spare punp muscles that are operating at a mechanical

disadvantage. Additionally, these infants may be highly dependent on

increases in inspIC to meet respiratory challenges while minimizing

further ribcage distortion.

Overall, the importance of combined increases of inspIC and

inspPCA with chemical loading cannot be overemphasized. Minimizing

paradoxical rib cage movement may be crucial in sparing the primary


inspiratory punp muscle additional work. The present results support

the idea that intercostal muscle recruitment to stabalize the ribcage

plays an important role in the response to 002 in the neonate. In

fact, the small change in inspDI measured may actually be

overestimating diaphragmatic activation due to contamination of the

signal by lower inspiratory interostal muscle activity. Augmented

inspPCA, suggesting active laryngeal dilation, provides a cost-

effective way to elevate inspiratory flow rates, and additionally

spares the primary pump muscle.

In summary, the response of the newborn to C02 inhalation

depends on the coordinated activation of the diaphragm, intercostal

muscles, and laryngeal muscles to bring about increased Vt with little

change in respiratory cycle timing. The relative contributions of

these muscles to a canon ventilatory response varies greatly,

although PCA and intercostal muscle activity consistently exhibited a

higher correlation than the diaphragm with changes in Vt and

inspiratory flow rates. In conclusion, the newborn infant is able to

use a variety of breathing strategies to augment Vt and Imin during

times of increased ventilatory demands.


Resistance to airflow is higher during expiration than during

inspiration, coinciding with laryngeal narrowing as described in both

animals (4,6) and adult humans (13,28). Augmenting expiratory airflow

resistance may play an important role in defending absolute lung

volume in the neborn infant who is at a mechanical disadvantage at

birth due to a highly ocupliant chest wall (42). The term nenate

adopts a breathing strategy to establish EEV above an intrinsically

low relaxation volume by retarding (braking) expiratory airflow


Methods of braking expiratory airflow include PIIA, which

opposes passive lung recoil during early expiration, and dynamic UK

narrowing, which increases resistance to expiratory airflow.

Electrmyographic evidence of PIIA, specifically PIIA of the diaphragm

(PIAdi), has been obtained from studies of both premature (17,66,95)

and term infants (63). The braking of expiratory airflow using UA

narrowing in addition to PIIA is known to occur in newborn animals

(29,47,48,51). An airflow pattern of UA braking is also seen in

newborn humans and activation of the thyroarytenoid (TA) muscles,

laryngeal adductors, has been implicated as the mechanism responsible

for the reduced airflow (36,71,80). In fact, there is no direct

evidence of adductor activity in infants, but recordings of

PCA muscle activity in both premature and term infants using an

esophageal elctrode have correlated reduced abductor activity with a

retardation of expiratory airflow (16,61). This is similar to the

pattern seen in the newborn qpposum (33) and most adult animals (6,48)

and humans (12).

Radvanyi-Bouvet et al. (80) studied expiratory patterns in

normal full-term and preterm infants and found that expiration was

frequently retarded or airflow was interrupted, but they did not make

direct observations of specific braking mechanisms. Expiratory

patterns in full-term infants were recently categorized and correlated

with diaphragm and PCA electromyograms providing clear evidence of

PIIAdi and illustrating the important role of intrinsic laryngeal

muscles in providing controlled braking of expiration in the sleeping

neonate (61). These patterns included 1) expiratory braking with

PIIAdi, 2) expiratory braking using the UA, 3) expiratory braking

using both PIIAdi and the UA, and 4) little expiratory braking with no

PIIAdi. It appears that during periods of regular breathing, infants

use PIIA early in expiration to oppose rapid emptying of the lungs,

and that some degree of UA braking provides a variable, controlled

resistance to airflow throughout expiration.

In the event of increased ventilatory demands, any significant

braking could be reduced to provide a lower resistance leading to

increased flow rates and Omin. There is recent evidence that normal

adults use their larynx in this manner. England and Bartlett (27)

found that expiratory laryngeal braking is reduced during hyperpnea

brought about by either C0D breathing or exercise in normal adults. It


is reasonable to assume that the newborn infant also uses the larynx

to control expiratory airflow when challenged by chemical loads.

It has been shown in the preceedig chapter that the term

neonate augments inspiratory airflow when challenged with a chemical

load through the coordinated activation of pump and laryngeal muscles.

Expiratory airflows are elevated as well, and the purpose of this

investigation is to determine the extent to which changes in PIIAdi

and PCA activity influence the expiratory pattern of the term neonate

during 002 inhalation.

Experimental Design and Methods

Subjects and Measurements

Subjects consisted of 8 healthy term infants with demographics

described in the proceeding chapter. The study protocol and

measurements were also the same.

Data Analysis

Analytical techniques were similar to those in the proceeding

paper. Values for PIIAdi were computed as the area under the diaphragm

MMA above electrical zero from the onset of expiratory airflow until

the MMh activity returned to within 1 SD of mean baseline activity.

The time frame the onset of expiratory airflow until the MRA returned

to baseline is referred to as TPIIAdi, and this variable was used as

another index of diaphragmatic braking (74). The mean and SD of PCA

baseline (PCAb) activity was cupuated using 50 ms of MEA data around

the point of minimum PCA activity during expiration.

Statistics were performed as in the proceeding paper with mean

expiratory (Te), and peak expiratory (pk)e) flow rates, as well as Vt

and rinin used as dependent variables for regressions. Electrinyograhic

data were used as independent variables. Additional regressions over

time were performed within individuals in an effort to further

describe the relationship between changes in muscle activities and

changes in expiratory flow rates. A single expiration was divided into

10 equal time segments and average values for airflow, PCA activity

and diaphragm activity were ocoputed for each segment. his was done

for each breath and mean values for these segment variables were

cauputed for each 15-second period. Simple and multiple linear

regressions were performed on each segment over time with flow as the

dependent and PCA and diaphragm activities as the independent

variables. Twenty sequential 15-second time periods were used for the

regression beginning 1 minute before C2 breathing and ending after 4

minutes of C02 inhalation.

Ensemble flow/volume loops were created for each individual

for the minute preceding 02 breathing and the fourth minute (steady-

state) of C02 inhalation. The loops were computed by dividing

individual inspired and expired volumes into 50 equal segments. These

volumes and the corresponding flow rates were averaged for each

segment, so that the composite loop consisted of 100 pairs of average

points (50 inspired and 50 expired). Points were plotted with volume

on the ordinate and flow on the abscissa to produce the composite


All 8 subjects reached 90% of the camputed steady-state

ventilatory response by the fourth 02 epoch. Times to reach 90% of

steady-state are presented in Table 4-1. All subsequent comparisons of

C02 to control values will refer to the fourth 002 epoch. he


individual and average data for changes in mechanical variables during

CO2 breathing are shown in Table 4-1. An average 42% increase in Vt

with no timing change resulted in an average 38% increase in trin that

was highly correlated (r=.94, P<0.001) with the change in Vt. lhe

augmentation of Vt with no change in timing resulted in significantly

elevated ve and pkIe (Table 4-1). The flow rate changes were also

highly correlated with increases in hidn (r=0.95 for pkWe, r=0.97 for

ve, P<0.0005) since no change in Te occurred.

A variety of breathing patterns were apparent during roam-air

breathing as described in the proceeding paper, but no matter what

pattern an infant used breathing roam air, the strategy remained

basically unchanged with OD2 breathing. This is apparent fram an

analysis of the overall shape of the composite flow-volume tracings

from the individual subjects (Figure 4-1). Panels A-H of Figure 4-1

correspond to subjects 1-8, respectively. Six subjects adopted a

pattern during room-air breathing (dotted curve) that included

noticable braking using PIIAdi resulting in retardation of early

expiratory airflow (Figure 4-1, panels B,C,D,E,F and H). This is seen

as a steeper slope during early expiration (upper right quadrant of

the flow-volume loop) indicating lower flow rates during the initial

part of expiration. These subjects exhibited PIIAdi well into

expiration (TPIIAdi=0.34+0.09 sec) which was maintained or

prolonged during 00 breathing (TPIIAdi=0.38+0.06 sec). The other 2

subjects (Panels A and G) exhibited little diaphragmatic braking of

expiration during both roam-air and CD2 breathing (TPIIAdi=0.13+0.04

and 0.13+0.05 sec, respectively).

Expiratory Electrnyographic and Ventilatory Variables

Subject mtin Vt Te fe pkSe PCAb PIIAdi TPIIAdi

1 74* 91* 10 76* 87* 112* -8 -12

2 56* 49* -2 52* 51* 8 11 -6

3 53* 67* 5 59* 76* 88* 32 15

4 50* 49* -1 55* 58* 33* 66* 24

5 27* 23* -7 32* 36* 34* -6 -9

6 18* 31* 17* 4 7 32* 73* 39*

7 18* 21* 5 15* 15* 0 7 9

8 15* 11 -6 19* 15 15* 21 31

x + SE 38+8* 42+10* 3+3 39+9* 43+11* 40+15* 24+12 11+7

TISS 98+31 110+35 98+21 92+25 154+17

Values are expressed as percent change frcan rom-air
breathing. Significant changes from roan-air breathing (P<0.05)
indicated with "*".

FIGURE 4-1: composite Flow-Volume Loops.
Panels A-H are data from subjects 1-8, respectively. Ioops are
composites for rocu-air and Xb -- breathing. The number of
breaths averaged for each subject is displayed (n), as well as the
percent change in 4min, 4e, and PCAb. Significant changes are
indicated with "*" (P<0.05).


Mean Ve=76"




n=27 Mean Ve=32
9mln=27" PCRb= 34"


F n=28

(N I

n=27 Mean Ve=59"
mIn=53" PCAb= 88"

n=37 Mean /e=55"
m I n=50" PCRb= 33"

-50 Insp exp 50
FLOW (mi/sec)

Mean Ve= 4
PCAb= 32"

n=46 Mean Ve=15"
U mIn=18" PCAb= 0

ca I

(s : :


n=29 Mean 'e=19"
lmIn=15" PCRb= 15"

-50 Insp exp 50
FLOW (ml/sec)







w m
-j L
0 -


To more thoroughly analyze the relationships between
expiratory muscle activities and airflow, expiration was divided into

10 equal time segments and time-series regressions were performed as
described in the methods. A summary of correlation coefficients is
presented in Table 4-2. Data for diaphragm activity versus airflow are
displayed for the first 2 segments, or 20% of mechanical expiration.

Correlations between PCA activity and airflow are shown for the first

8 segments, or 80% of expiration. The first 2 segments will be
collectively referred to as "early expiration", and the next 6

segments (20-80% of expiration) will be called "mid-expiratian".

A positive correlation between diaphragm activity and early
expiratory airflow was found in 3 subjects (Table 4-2, nos. 2,4 and

8). These 3 subjects all had diaphragmatic braking and all exhibited

a positive correlation between PCA activity and airflow during early

expiration. A negative correlation was Observed between diaphragm
activity and early expiratory airflow in 3 subjects (nos. 1,3, and 6).

Two of these 3 subjects had diaphragmatic braking (nos. 3 and 6) and
no correlation between PCA activity and airflow during early

expiration. Both subjects with little diaphragmatic braking (nos. 1

and 7) had a positive correlation between PCA activity and airflow

during early expiration.
During all 6 mid-expiratory segments, 3 subjects (nos. 4,6 and

8) exhibited a positive correlation between PCA activity and airflow.
One subject (no. 5) had positive correlations in 5 of 6 segments, and

1 subject (no. 1) had positive correlations in 3 of 6 segments. All 5

subjects with positive correlations during mid-expiration had
significantly increased PCAb (Table 4-1). The remaining subject with

ABLE 4-2
Correlation Coefficients Between Expiratory Flow Rates and
Muscle Activities

Subject 1 2 3 4 5 6 7 8


0-10% -0.07 0.55* -0.22 0.73* 0.04 -0.63* 0.25 0.42*

10-20% -0.31 0.33 -0.39* 0.68* -0.22 -0.44* -0.27 0.47*


0-10% 0.66* 0.61* 0.21 0.83* 0.02 0.06 0.45* 0.46*

10-20% 0.64* 0.56* 0.06 0.69* -0.04 0.36 0.38* 0.53*

20-30% 0.58* 0.43* -0.05 0.62* 0.30 0.60* 0.25 0.62*

30-40% 0.52* 0.26 -0.01 0.62* 0.47* 0.74* 0.02 0.73*

40-50% 0.38* -0.10 0.07 0.70* 0.58* 0.84* -0.16 0.85*

50-60% 0.33 -0.15 0.10 0.73* 0.79* 0.89* -0.12 0.85*

60-70% 0.34 -0.20 0.19 0.74* 0.84* 0.89* -0.35 0.81*

70-80% 0.34 -0.23 0.31 0.72* 0.82* 0.84* -0.28 0.75*

Values are correlation coefficients between diphragm and
airflow for segments comprising the first 20% of expiration, and
between PC and airflow for segments comprising the first 80% of
expiration for all 8 subjects. Regressions were performed on average
values of these variables for 15-second time intervals beginning 1
minute before 00D2 inrhalticn and ending 4 minutes after the initiation
of the challenge (n=20 time intervals). Significant relationships
(P<0.05) are marked with "*".


increased PCAb (no. 3) showed no significant correlation between PCA

activity and airflow during mid-expiration. However, the combination

of diaphragm and PCA activity in a multiple regression resulted in

uich higher regression coefficients for the first 5 mid-expiratory

segments (20-70% of expiration, r=0.79, 0.71, 0.65, 0.68 and 0.54,

respectively) in this individual.

The changes in muscle activities during expiration with 002

breathing are sunmarized in Table 4-1. No significant change in PIIAdi

with CO2 inhalation was found, although there tended to be an increase

(24 + 12%, significant in 2 subjects). Individual regressions found

PIIAdi correlated with 4e and pk)e in only one subject (r=-0.58 and

-0.55, respectively). A slight overall increase in TPIIAdi was

measured (11 + 7%), but this was only significant in one subject.

Values for TPIIAdi were correlated with fe in 3 subjects (r=-0.49 +

0.02) and with pkge in 1 subject (r=-0.49).

An elevation of mean PCAb of 40 + 15% was measured, with

significant increases in six subjects. The changes in PCAb correlated

with increases in ve (r=.70, P<0.05) (Figure 4-2). Individual

regressions found PCAb positively correlated with ve in 4 subjects

(r=0.79 + 0.06) and with pkJe in 5 subjects (r=0.77 + 0.11). An

example of an individual regression is shown in Figure 4-3.
Diaphragmatic braking parameters (PIIAdi and TPIIAdi) were combined

with PCAb in a multiple regression in an attempt to better define

changes in expiratory airflow. A significant multiple regression

coeffient resulted in 5 subjects for both fe and pk~e (r=0.82 + 0.11

and 0.84 + 0.09, respectively) when all three parameters (PCAb,

PIIAdi, TPIIAdi) were used as independent variables. Potential changes


90 -
r=0.70, P<0.0500 /

80 /
r4J7-- 1-

*> /

C 60 -
50 /

o 40 -
U /

20 /

10 ----- / -------- r ---------------------i--------------

-50 0 50 100
% Change PCR baselI ne

FIGURE 4-2: Expiratory Airflow versus PCA Muscle Activity.
Relationship between 4e and PCAb. Values are expressed as %
change from roc-air breathing. Points are mean + SE for each of 8
subjects. A regression line is plotted (_ _) and the correlation
coefficient (r) given. Points right of the vertical line (-) are
significantly elevated PCAb (P<0.05). Points above the horizontal line
(-) are significantly elevated ve (P<0.05).

/ 121714

50. 60. 70. 80. 90. 100. 110. 120. 130.

PCR baselIne

FIGURE 4-3: Example of an Individual Expiratory Regression.

Data are from Subject 8, showing the relationship between
Ve and PCAb. Points are mean values for successive 15-second intervals
starting 1 minute before 00 breathing and ending after 4 minutes of
CD- breathing (20 points total). The correlation coefficient for this
regressin was 0.76 (P<0.0001).

3. 5






2. 5



1 .




in lung recoil pressure were quantified by adding Vt to this

regression. The resulting correlations with Ve and pk)e were

significant in all 8 subjects (r=0.96 + 0.02 for both).


The healthy term infant in rncREM sleep augmented min during

C2 breathing by increasing Vt with no change in frequency or timing

as previously reported (2,22,25,45). An elevated Vt without timing

changes requires increased inspiratory and expiratory flow rates.

Three mechanisms by which expiratory airflow can be increased during

hypercapnia are: 1) increased passive lung recoil pressure, 2)

decreased expiratory braking, and 3) expiratory pump muscle

activation. No respiratory activity from abdominal muscles was

observed, and no change in expiratory activity of the intercostal

muscles was found in any subject. Thus, one can conclude that the term

infant does not use expiratory muscles when challenged with this level

of 002 he present study supports the idea that changes in lung

recoil pressure combined with changes in UA braking are largely

responsible for the observed increase in expiratory flow rates.

Individual breathing strategies must be examined in

conjunction with passive mechanical changes influencing expiratory

airflow. The increased vt during CO2 inhalation should increase

passive lung recoil pressure unless EEV were lowered. No measure of

EEV was made in the present study, and the flow-volume loops were

normalized to EEV to better visualize changes in Vt and flow rates.

There is no evidence to suggest that EEV is lowered, but a decay of

expiratory flow at end-expiration which would suggest a lowering of

EEV is not observed. It is, therefore, reasonable to assume lung


recoil pressure is elevated due to increased Vt. High correlations

were observed between increased Vt and both pk)e and fe that were

significant in all subjects (r=0.88 + 0.09 (SD) and 0.86 + 0.08,

respectively). In a passive system, this would suggest that lung

recoil pressure is a major contributing factor to changes in

expiratory airflow. In the infant, with no served expiratory muscle

activity, increased lung recoil pressure is the driving force behind

increased expiratory airflow. This passive mechanical influence on

airflow acts in concert with changes in active expiratory braking

mechanisms to shape the expiratory airflow pattern.

Changes in expiratory braking mehanisms with 002 bathing

were observed. Expiratory braking is of particular importance to the

neonate in maintaining an EEV above their low relaxation volume

(63,71,72). One expiratory braking mechanism, PIIAdi, did not decrease

during hypercapnia. In fact, it increased somewhat. This is contrary

to findings in other animals (41,49,74), but consistent with the

findings of Carlo et al. (16) in pretermn infants and suggests that the

elevation of EEV continues to be of vital importance in the newborn

even during hypercapnia. The persistence of PIIAdi serves to retard

early expiration and may partially counter the effect of increased

passive recoil forces to augment pk)e. The data provide evidence that

another mechanism, decreased laryngeal braking, is also involved in

augmenting expiratory airflow.

The larynx serves as a variable resistor to control expiratory

airflow in the spontaneously breathing infant (61) and adult human

(12,28). laryngeal braking of expiration can be accomplished through

active laryngeal adduction (48), loss of active abduction (6,12) or a


combination of these factors (29). Conversely, decreased expiratory

laryngeal resistance can result from decreased adductor activity

and/or increased abductor tone (PCAb). Increases in PCAb coincident

with decreased expiratory laryngeal resistance have been reported in

both unanesthetized and anesthetized cats (4,6) and adult humans (13).

Activities of other UA muscles have been shown to be related to

changes in UA resistance in animals (29,48) and adult humans (96), but

the larynx, acting as a flow valve, appears to be the primary UA

region involved in the regulation of UA resistance (75). Unlike the

PCA muscles, TA activity does not appear to play a role altering

expiratory airflow in most adult animals (6,48) and humans (12).

However, in newborn puppies (29) and lambs (48,51,54), TA activity is

present during expiration, especially during quiet sleep.

It has been suggested that newborns use laryngeal adductors to

brake expiratory airflow (36,61,71,80), however, it has not yet been

possible to record adductor activity in infants. The present study

indicates that elevated PCA activity correlates well (r=0.70) with,

but does not fully explain, changes in ie during hypercapnia. However,

a combination of changes in Vt (lung recoil pressure), PCAb and

diaphragm braking consistently correlate highly with fe (r>0.94 in

each subject). This high correlation leaves little to add to the

description of expiratory flow rates, suggesting that either laryngeal

adductor activity is of little importance in the response of the

newborn to 002, or that the adductors act conversely to the abductors

in a highly synchronized fashion. The reciprocal activation of these

muscles was evident in previous studies in newborn puppies (29) and

indicates a high degree of intercoorellation between the activities of


the laryngeal adductors and abductors. Therefore, if it was possible,

the addition of adductor activity to the present description of

expiratory flow rates would likely be insignificant whether or not the

muscles were actively contributing to the response. However, one

cannot exclude the possibility of laryngeal adductors playing a role

in shaping expiratory airflow and the reciprocal activation of these

muscles may be necessary to bring about any appreciable change in

laryngeal caliber.

In the proceeding chapter it was found that a variety of

breathing strategies may be invoked to augment inspiratory airflow.

The same can also be said for expiratory airflow. Three basic

relationships were observed between muscle activities and expiratory

airflow. These are: 1) positive correlations of both PIIAdi and PCA

with early expiratory airflow, 2) negative correlations between PIIAdi

and early expiratory airflow with no relationship to PCA activity, and

3) good correlations between PCA and expiratory airflow during early

and/or mid-expiration along with little PIIAdi.

A positive correlation between PIIAdi and early expiratory

airflow indicates that augmented PIIAdi coincided with increased

airflow. This could only be possible if another mechanism was involved

to counter the effect of increased diaphragmatic braking. Three of the

subjects with substantial PIIAdi while breathing both room air and CO2

(nos. 2,4 and 8) had positive correlations between diaphragm activity

and airflow during early expiration. These 3 subjects also had

significant correlations between PCA activity and airflow during the

same time period. This suggests that increased diapragm activity is

coupled with increased PCA activation in these subjects, and that any


retardation of pk e that may have been afforded by PIIAdi is offset by

increased PCA activity.

One would suspect that those subjects that relied on PIIAdi to

brake early expiration (nos. 2,3,4,5,6 and 8) might have good negative

correlations between diaphragm activity and early expiratory airflow,

such that reduced PIIAdi would coincide with increased expiratory

airflow. In fact, only 2 of 6 subjects with substantial PIIAdi

exhibited such a relationship. Those 2 subjects (nos. 3 and 6) showed

little correlation between PCA activity and airflow in early

expiration. Without an early increase in PCA activity, braking of

expiratory airflow by the diaphragm goes unopposed.

Subject 6 exhibited the largest increase in diaphragmatic

braking with 002 breathing and this resulted in no change in pkJe

(Table 4-1). This suggests that increased PIIAdi fully ccopensated for

any increased recoil pressure due to elevated Vt. An elevation of PCAb

was also observed in this subject (Table 4-1), and this is reflected

in good correlations between PCA activity and flow rates during

mid-expiration (Table 4-2). This expiratory PCA activity may reflect

an effort to shorten Te in the face of augmented diapragmatic braking

(39% increase in TPIIAdi). In fact, Te was prolonged 17% in this

subject, but increased tmin was still achieved through elevated Vt.

The other subject that exhibited a negative diaphragm-airflow

correlation in early expiration (no. 3), used a ccnbination of

increased pump muscle and PCA activity to augment inspiratory airflow

(Table 3-2). This subject also appears to use a combination of

diaphragm and PCA muscles to modulate expiratory airflow. The negative

correlations between diaphragm or PCA activity and airflow during

early expiration and mid-expiration were unimpressive (Table 4-2),
however, a multiple regression using both muscle activities to

describe airflow resulted in significant contributions of both muscles
to correlation coefficients that averaged 0.71 for the first 6
expiratory time segments. This suggests that activities of the
diaphragm and PCA muscles alone cannot explain much of the variability
in expiratory airflow in this subject, but the coordinated action of
these muscles together can explain 50% (r2=0.50) of the observed

variability in airflow.

Both subjects (nos. 1 and 7) that established a relatively
uritaked ventilator pattern during both roam-air and 002 breathing,
exhibited significant correlations between PCA activity and airflow
during early and, for subject 1, mid-expiration. In addition, both of
these subjects showed no increase in diaphragm activity during
inspiration (Table 3-2) suggesting that laryngeal modulation of

airflow throughout the respiratory cycle was the dominant factor in
the hypercapnic response of these individuals.

In spite of the variations in breathing strategy between
individuals, the expiratory flow-volume patterns of individual subjects

varied little between roam-air and 002 breathing (Figure 4-1). The
predominant pattern exhibited a retardation of early expiratory

airflow related to PIIAdi, but no matter what strategy was adopted to
maintain EEV, no appreciable change in strategy occurred within the
time course of these experiments. This suggests that preserving an
expiratory pattern to defend EEV is critical and is maintained in
spite of changes associated with this level of 002 inhalation.


In suomary, the response of the newborn to 002 inhalation

depends on the coordinated activation of the diaphragm and laryngeal

muscles to bring about increased expiratory airflow with little change

in timing. The relative contributions of these muscles to individual

subject's ventilatory responses varied greatly, yet resulted in a

common outocme of increased Vt and mtain with little change in

respiratory cycle timing. The most frequently observed breathing

strategy involved maintained PIIAdi and increased PCAb, with PCA

activity consistently exhibiting higher correlations than the

diaphragm with changes in expiratory flow rates. The persistent PIIAdi

may serve to offset changes in lung recoil pressure and appears to be

more tightly coupled to the maintenance of EEV, while other airflow-

controlling muscles of the UA appear active in modulating resistance

to airflow to help bring about the Vt changes necessary to increase

%Ain during CD2 breathing.


The newborn infant responds to low levels of CD2 inhalation by

elevating 4min throuu i increased Vt with little change in Ti, Te or

frequency of breathing. This is achieved by augmenting PCA muscle

activity throughout the respiratory cycle and by increasing

inspiratory pump muscle activity. Augmentation of inspPCA and insplC

exceeded that of inspDI and correlated well with inspiratory flow

rates. Changes in expiratory braking consisted of minimal modulation

of PIIAdi and increased PCAb reflecting a decrease in laryngeal

braking. Thus far, inspiration and expiration have been treated

separately, yet to fully appreciate the responses of the individuals a

discussion of the total response is necessary.

The use of EMGs as an estimate of neurcauscular response has

potential problems. Poor electrode placement, deterioration of

electrode contact and, especially with the laryngeal electrode,

movement of the electrode can all contribute to erroneous

measurements. Good electrode placement was confirmed by the

measurement of phasic inspiratory activity of the muscles in question.

Studies did not last more than 2 hours, and in this time period no

deterioration of the E3 was observed. The prevention of faulty

measurements due to movement was achieved by analyzing only those

experiments where the subject did not move or swallow throughout the

recording period.


Another ccmplicating factor in using EMs to describe

mechanical changes is that the relationship between muscle activation

and mechanical dcanges is most likely alinear. The amount of force

generated for a given increase in EM3 measured would depend on several

factors, including the resting length of the muscle itself.

Additionally, resistance to airflow is related to the inverse of the

radius of the tube to the fourth power. Even if the relationship

between PCA muscle activation and laryngeal caliber is linear, airflow

resistance would not be linearly related to the PCA M changes. Thus,

any correlations between the measured changes in DB activity and

changes in mechanical variables like airflow and Vt are first

approximations, at best.

There appears to be an order to the recruitment pattern of the

various muscles with the exception of only one subject (no. 2). All

subjects that increased inspDI also increased inspIC. Likewise,

excluding subject 2, those subjects that increased inspIC elevated

inspPCA. Those subjects that augmented PCAb also increased inspPCA,

and the two subjects with increased PIIAdi exhibited elevated PCAb.

This suggests that the general response of the newborn is to increase

primarily PCA muscle activity during both inspiration and expiration.

Secondarily, pump muscle augmentation can contribute to the

hypercapneic response in the form of intercostal muscle recruitment to

stiffen the chestwall and allow the diaphragm to pump more

effectively. This increased mechanical efficiency of the diaphragm may

account for the smaller augmentation of inspDI observed. Lastly, the

diaphragm itself can be recruited to elevate inspiratory flow rates.

The relationship between increases in inspPCA and PCb is
significant, as shown in Figure 5-1. The correlation coefficient of

0.94 means that 88% (r2=0.88) of the variability in PCpb is reflected

in changes in inspPCA. This implies that all of the subjects modulated

inspPCA and PCAb to the same relative degree. Bartlett (4) found that

with hyperc(dc 002 breathing, both inspiratory and expiratory
laryngeal resistance fell in conjunction with increased PCA activity
in anesthetized cats. When changes in inspiratory and expiratory
resistance from Bartlett's (4) individual animals are plotted against

each other (Figure 5-2) a linear relationship is found with r=0.7

(P<0.05). This relationship is markedly improved when, what appears to
be an outlier (no.5), is removed. The correlation coefficient then
becomes 0.93 (P<0.001), almost identical to the correlation in PCA
activities found in the present study (Figure 5-1).
The relationship between Ti and TPIIAdi is also of importance.
A study on anesthetized dogs (74) breathing 100% oxygen found that
those animals with little PIIAdi (TPIIAdi<0.24 sec.) had significantly

shorter inspiratory times than those animals with TPIIAdi>0.32 seconds

(Ti=0.51+0.07 (meantSE) and 0.92+0.13, respectively, P<0.01).
When the present data are plotted in this manner, a similar

relationship is seen (Figure 5-3). The correlations, slopes and
intercepts are similar between rocm-air and C02 breathirq (Table 5-1),
hence, all data were combined in a regression resulting in a
correlation between Ti and TPIIAdi (r=0.67, P<0.005). In the study on
dogs (74) this relationship was maintained during hyperoaic o2
rebreathing with both Ti and TPIIAdi shortening in those animals that
had substantial diaphragmatic braking initially. Those animals with

r=0.94, P<0.000S /

Li T
-.- ---i

*- /

7S -

01 /X--3--


/ /


0l-------^------4 -----------------------------------------
0 1I
-50 0 50 100
% Change PCR baselIne

FIGURE 5-1: Inspiratory versus Expiratory PCA Muscle Activity.
Values are expressed as % change from roam-air breathing.
Points are mean + SE for each of 8 subjects. A regression line is
plotted (_ _- and the correlation coefficient (r) given. Points
right of the vertical line (-) are significantly elevated PCAb
(P elevated inspPCA (P<0.05).







/ n=


% of Control

Exp. Laryngeal

Res I s.

FIGURE 5-2: Laryngeal Resistance in Cats.
Data fram Bartlett (4) from anesthetized cats breathing CO2.
Plotted as percent of control expiratory resistance versus percent of
control inspiratory laryngeal resistance. A regression line (_ )
is plotted for both 9 and 8 (no. 5 excluded) subjects.

100 r





80 B

7 /








r=0.67, (P<0.005)


4 /
/ 8 4C


iC /




FIGURE 5-3: Inspiratory Time versus Diapragmatic Braking.
Subject numbers are displayed for values during roam-air and
OD, (C) breathing. A regression line ( ) is plotted
for all data (n=16).





Ti vs. TPIIAdi, rom air


sessions Using TPIIAdi

n Ooefficient

8 0.67

8 0.67

16 0.67

e vs. TPIIdi, roam air




















little PIIAdi during 02 breathing showed no change in Ti or TPIIAdi

with M2 rebreathing. Vagotcmy abolished the shortening of TPIIAdi and

Ti during rebreathing suggesting a vagal modulation of PIIAdi and Ti.

The action of the diaphragm to brake expiration suggests that

a similar relationship between TPIIAdi and Te may exist, where Te is

prolonged by increased TPIIAdi. Oliven et al. (74) found that there

was no relation between Te and TPIIAdi in anesthetized dogs, but their

subjects were using expiratory muscles to actively expire which may

have profound effects on Te. In the present study, such a relationship

exists between Te and TPIIAdi (Figure 5-4). Correlations, slopes and

intercepts were similar between roam-air and 002 breathing (Table

5-1), and again all data were combined in a regression resulting in a

high correlation (r=0.95, P<0.0001). The correlation coefficient was

much higher than for Ti and TPIIAdi suggesting a tighter linkage

between TPIIA and the control of expiratory duration. This further

suggests that the newborn actively controls Te using diaphragmatic

braking. However, when individual changes in Te and TPIIAdi are

considered, subjects 1,4 and 8 exhibit Te/TPIIAdi relationships with

negative slopes (Figure 5-4), contrary to the overall response. The

tight coupling between diaphragmatic braking and Te, and, thus EEV,

observed in Figure 5-4 suggests that the overall strategy of breathing

and the pattern of respiratory cycle timing that an infant adopts is

related to the level of diaphragmatic braking. However, the individual
responses to C02 breathing depend primarily on modulation of

expiratory airflow resistance through another mechanism; active

laryngeal dilation.

r=0.95, (P<0.0001)


8 /
2 8C
4 /







FIGURE 5-4: Expiratozy Time versus Diaphragmatic Braking.
Subject nrubers are displayed for values during rocm-air and
00 (C) breathing. A regression line ( ) is plotted
for all data (n=16).


0.9 -




0.5 1

0. .



The Inspiratory Off-Switch

the reflex mechanisms responsible for the observed increases

in inspiratory muscle activation and expiratory laryngeal dilation

with CD2 inhalation are unknown. It is speculated that the neonatal

respiratory control system is similar to that described in

experimental animals. Ihe results of this study will be related to the

inspiratory off-switch model. Several studies have attempted to model

the central pattern generator (CPG) for rhythmic respiration. Studies

to localize various respiratory neuronal pools have utilized several

techniques in experimental animals including transactions, electrical

stimulation and intracellular recording of respiratory related

neuronal activity. Further studies have been performed to describe the

interactions between neuronal groups as well as the reflex responses

of these cells to a variety of stimuli. Numerous models have been

proposed to integrate these studies into a functional description of

the CPG and its reflex responses. Most descriptions of the CPG divide
the respiratory cycle into four phases (30): 1) inspiratory phase with

augmented inspiratory activity, 2) inspiratory off-switch phase, 3)

expiratory phase 1 with PIIA, and 4) expiratory phase 2 with increased

inspiratory inhibition and augmented expiratory motor activity with

increased drive. Ihe inspiratory off-switch controls the amplitude and

duration of a central inspiratory activity (CIA) integrator while this

integrator is kept inhibited during expiration by mechanisms

controlling PIIA and expiratory muscle recruitment thresholds. No

expiratory punp muscle activity was observed in these infants, so the

final phase including expiratory motor activity will not be addressed.

Chemical Drive
--- -------------------



I +



To 5ptnol
P5R Motoneurons

FIGUR 5-5: Model of Inspiratory Off-Switch.
A hypothetical model adapted fran Bradley et al. (11)
depicting one version of the off-switch mechanism of the CpG.
Inhibitory (-) and excitatory (+) cnnections are noted.


A general model describing the functional organization of the

inspiratory off-switch mechanism of the CPG is shown in Figure 5-5.

The amplitude and duration of inspiration are determined by the

integration of negative influences frum the off-switch and stimulation
from chemical and other drives. The output from the off-switch is

augmented by a combination of feedback from the CIA and lung volume

related pulmonary stretch receptor (PSR) activity and is inhibited by

chemical and other drives. The relationship between amplitude and

timing of inspiration, or Vt-Ti has been extensively studied in

numerous species. The Hering-Breuer inspiratory-inhibitory inflation

reflex (14) describes a PSR-mediated termination of inspiration. The

volume threshold for the reflex termination of inspiration has been

found to fall rapidly during the course of inspiration in most animals

(19,34,44) and infants (60). The Vt-Ti relationship depicting a

declining volume threshold for termination of inspiration is shown in

Figure 5-6.

Stimulation of respiration through CO2 breathing causes

increased rate of rise of phrenic nerve activity in the lightly

anesthetized cat (19). This increased rate of rise would result in a

shortening of Ti, as seen with increased temperature in cats (30),

unless the off-switch threshold is raised. The latter appears to be

the case in cats (19) with an elevation of the off-switch threshold

allowing a larger Vt. In the present study, the increase in Vt was

accompanied by no change in Ti. This suggests that in the term infant,

the inhibitory effects of C02 on the off-switch (raising the

off-switch threshold) balance the increased excitatory feedback from

the CIA and PSRs that would act to shorten Ti. There are two ways to


35 -



25 \

E \

> 20- \ R1


1.4 0.5 0.6 0.7 0..8
Ti (sec)

FIG RE 5-6: Declining Inspiratory Off-Switch Threshold.

Plot of Vt versus Ti adapted from Kosch et al. (60). Data is
from term infants during control breathing (C) and three levels of
single-breath inspiratory resistive loading (R1,R2,R3).


interpret these results; either there is a range of the Vt-Ti

relationship where Vt continues to increase without a shortening of

Ti, or the increased Vt is due to a shift of the entire Vt-Ti curve to

the right as shown in Figure 5-7. Additionally, these data provide

evidence of a variable amplitude effect of CD2 on the neuronal pools

mediating the activity of the diaphragm, intercostal and PCA muscles.

Inspiratory-Expiratory Coupling

The expiratory phase of respiration is generally thought of as

a passive mechanical event during eupneic breathing, with a decay of

volume being brought about by elastic recoil of the lungs. In fact,

the control of expiratory duration is far from passive. There are

neuronal pools that discharge in a declining manner in early

expiration in close correspondence with PIIA (82,83). This PIIA

counters elastic recoil and slows emptying of the lung and is coupled

to a prolongation of Te as described earlier (Figure 5-4). This PIIA

is an efficient means of braking expiration, for it is achieved by the

negative work of activating the diaphragmatic muscle fibers while they

are stretching. This stretching contraction retards expiratory airflow

at a low energy cost to the infant. The prolongation of Te through the

retardation of volume decay is hypothesized to be mediated by the

feedback of PSR activity on a neuronal pool that integrates factors

controlling central inspiratory-inhibition (CII) (56,57,104). Factors

that serve to slow the emptying of the lungs, including PIIA and

increased UA resistance, reflexly prolong expiration. These influences

on CII are integrated into a more comprehensive model of respiratory

control in Figure 5-8.

<- WITH C02

\ \



I I i I




Ti [sec)

FIGURE 5-7: Effect of 002 an Vt-Ti Curve.

Hypothetical effect of CD breathing on the declining
inspiratory off-switch threshold in infants. Roon-air breathing curve
adapted from Kosch et al. (60). A "?" indicates unknown data.

35 r

30 1

25 1

15 F-






Chemical Drive

-- -- -----^ c^

I 1SwI ch



To 5pinal
PSR Motoneurons

FIGURE 5-8: Model of Central Inspiratory Inhibition.
Model adapted from Knox (57) depicting the interaction of a
central inspiratory inhibitory neuronal pool controlling Te with the
off-switch model from Figure 5-5.


The CII neuronal pool has been added to this model and the

output from these neurons has a inhibitory effect on CIA. CII

neurons are shown to be excited by PSR feedback as well as by

off-switch output in this model. An additional input to the CYG,

rapidly adapting receptor (RAR) feedback, is introduced in this model.

These receptors are generally thought to respond to high flow rates

and low lung volumes (30,57). They are shown as having an inspiratory

excitatory effect on the CIA and a corresponding inhibitory input to

the CII neurons.

The manner in which PIIA neurons and neurons controlling

laryngeal aperature fit into this model is largely unknown. Sane

information does exist on the vagal modulation of PIIA and laryngeal

caliber or muscle activity, as well as the effects of increased

respiratory drive on these potential expiratory braking mechanisms.

This information, canbined with results from the present studies,

allow one to speculate on the relationship of the neurons controlling

these functions with the CEG.

A reduction in vagal afferent input results in decreased PIIA

in the lamb (51), suggesting that breathing at higher lung volumes

augments PIIA. Studies on anesthetized (74) and awake (93) dogs

breathing various C02 gas mixtures found a decrease in PIIAdi along

with decreasing Te. A study on premature infants reported no change in

PIIAdi with 002 breathing (16), with no mention of changes in

respiratory cycle timing. The lack of PIIA modulation with D002 in the

premature infants is contrary to the findings in animals, however

there is also no change in Te in the studies on premature infants as

measured from the tracings of raw data. These data, in combination

with the present studies, suggest that the relationship between Te andm

PIIAdi in the infant is unchanged by a)2 breathing.

The effect of lung inflation on laryngeal caliber or PCA

activity is complex and variable. Investigators have reported

augmentation of PCA activity or laryngeal widening with lung inflation

(6,23,68) while others have observed or inferred reflex addiction of

the vocal cords with this stimulus (31,38). Reflex increases in

laryngeal abduction or abductor muscle activity with a2 breathing are

well defined in the literature. Increases in PCA activity or laryngeal

caliber have been observed with CD2 breathing during both inspiration

and expiration in cats (4,6), adult humans (27), and premature infants

(18). This information, ocubined with the present data, indicates a

clear augmentation of laryngeal abductor activity with o02 beating.

Modulation of laryngeal caliber by PSR feedback mechanisms cannot be

explicitly described from the current body of knowledge.

Proposed Model of Respiratory Control in the Newborn

A model is proposed in Figure 5-9 that incorporates PIIA and

PCA controllers, as well as the motoneuron pools of the diaphragm,

interoostal and PCA muscles. This model shows diaphragm and

intercostal muscles acting synergistically to drive the respiratory

pump. Activation of the PCA muscles is shown to decrease laryngeal

resistance to airflow which ocabines with inspiratory drive to

increase lung volume. Changes in lung volume, in turn, modulate PSR

feedback to the CFG.

The PIIA neurons are most likely no different from the

phrenic motaoneurns. The observation of PIIA probably reflects a

lowering of the inhibitory influences on the CIA during early

Chemical Drive

FIGURE 5-9: Proposed Model of Respiratory Control.

Model incorporating PIIA and laryngeal control mechanisms into
the model froma Figure 5-8.


expiration. Increased lung volume may have an indirect excitatory

effect through RAR activation and the CII pool. In infants, the

increased Vt associated with 002 inhalation and subsequent stimulation

of PITA through RAR feedback may be offset by PSR feedback on the CII

pool. This results in the observed insignificant changes in PIIAdi

with 002 inhalation.

This model proposes that the PCA-controlling neurons are

closely tied to the CPG through common neural inputs. The PCA muscle

is activated prior to diapragmatic activation and this timing

difference has been shown to be due to earlier activation of laryngeal

notoneurons rather than differences in conduction velocities or

contraction times (20,89,90). This activation sequence may be the

result of fewer interneuronal connections in the laryngeal motoneuron

activation pathway or a lower threshold for activation of the

laryngeal notcneurons. This model also proposes direct stimulation of

laryngeal premotor neurons by increased chemoreceptor feedback. This

is consistent with the observation of similar changes in inspPCA and

PCAb within individual infants during 002 inhalation, and is supported

by similar changes in inspiratory and expiratory resistance in

hyperoxic cats breathing 002 (4). The laryngeal premotor neurns in

Figure 5-9 are driven directly by the CIA and CII pools. This means

that 002 drive stimulation of the CIA pool augments the direct

stimulation of the PCA neurons resulting in a greater sensitivity to

002 of the laryngeal muscles as ccapared to the pump muscles.

The observed differences in intercostal and diaphragmatic

motor output remain to be addressed. The close association of

diaphragmatic and intercostal muscle activity are reflected in a

caonmon source in the CIA integrator of the CFG. The observed

difference in the response to 002 inhalation may be the result of

separate neuronal pools within the CIA integrator that have slighlty

different inputs from chdimreceptors. Other reasons for the observed

responses include different spinal reflexes for the various muscles

and changes in pH with hypercapnia that may differentially affect the

resting potential of premotor neurons or other interneurons.

Future Studies

These studies emphasize the importance of the control of

absolute lung volume to the newborn. Reflex modulation of volume and

respiratory cycle timing in the newborn is complex and cannot be

thoroughly addressed in a study of this nature. The breathing strategy

of the newborn to defend lung volume necessitates future studies where

absolute lung volume is measured rather than inferred as in this

study. The measurement of lung volume greatly complicates an already
complex experimental protocol, but sequential studies on the same

infant may help to answer this question. The increase Vt observed with

D02 inhalation evokes reflex responses of itself. Future studies on
reflexes induced by mechanical loading or augmentation would serve to

clarify the response to OD2 breathing. The ability to respond to added

mechanical loads could be quite important in the newborn where partial

obstruction to airflow can occur from substances in the upper airway,
with periodic loss of genioglossus muscle tone, or with spontaneous

neck flexion.

Future studies to elucidate the mechanisms behind the aO2

response of newborns should include varying the level of 002 used
within an infant to determine if there is a differential response to


a graded drive of the various pump and airflow-controlling muscles.

Rebreathing studies in premature infants found different response

thresholds for the genioglossus and alae nasi muscles as ccupared to

the diaphragm (18). The present study has shown differences in the

response of select muscles to a single level of stimulation, and the

response to multiple levels would be useful information.

Studies that address the interaction between chemical and

mechanical reflexes could also be performed. Maintained resistive

loads could be applied to mimic partial obstruction of the airway.

Some studies of this nature with ENM recordings have been done in

adult humans (65,77) and both adult (1) and newborn (100) animals, but

no information is available from infants. Additionally, single-breath

resistive loading could be performed with and without 002 breathing to

gain more information on mechnical-chemical load interactions. Similar

studies have been done on adult humans (102) and animals (44,58), but,

once again, not in infants. Experiments of this type could provide

valuable information on the control of breathing in the newborn human.


The newborn infant adopts a strategy of breathing to defend

absolute lung volume, and this strategy can vary greatly between

individuals. Inspiration is controlled through the coordinated

activation of pump and UA muscles. The use of PIIA and active

modulation of laryngeal caliber play important roles in controlling

expiratory airflow and Te. The inhalation of 002 augments inspiratory

activity of the PCA, intercostal and diaphragm muscles to varying

degrees in different subjects, although the common result is an

elevated Vt with no change in Ti. Expiratory time also remains


unchanged through the ccbined effects of increased lung recoil

pressure, PIIA and laryngeal dilation. The pattern of activation of

these muscles depends somewhat on the baseline breathing strategy of

the individual. The C02 stimulus appears to have a greater effect on

the PCA muscles than on the pump muscles, and there is a tendency to

stiffen the chestwall through intercostal muscle activation in

preference to augmenting diaphragm activation.

In summary, one can conclude from these studies: 1) the

newborn infant adopts varied breathing strategies to defend absolute

lung volume including the braking of expiration with the diaphragm and

laryngeal muscles, 2) the newborn has the ability to augment minute

ventilation in a variety of ways when challenged with carbon dioxide,

3) improving diaphragm-ribcage coupling through intercostal muscle

activation may provide an energy efficient means of enhancing pump

muscle performance, 4) the defense of absolute lung volume is of

critical importance to the newborn as demonstrated by the persistence

of diaphragmatic braking of expiration throughout this chemical

challenge, and 5) the modulation of laryngeal resistance to airflow by

posterior cricoarytenoid activation plays a crucial role in the

ability of the newborn to respond to a chemical load while maintaining

absolute lung volume.


1. Altose, M.D., N.N. Stanley, N.S. Cherniack, and A.P. Fishman.
Effects of mechanical loading and hypercapnia on inspiratory
muscle W. J. Appl. Physiol., 38: 467-473, 1975.

2. Anderson, J.V., R.J. Martin, E.F. Abboud, I.Z. Dyme, and E.N.
Bruce. Transient ventilator response to COD as a function of
sleep state in full-tenr infants. J. Appl. &hysiol. 54:
1482-1488, 1983.

3. Arvidsson, A., A. Grassino, and L. Lindstri=. Autaoatic selection
of uncontaminated electrncyogram as applied to respiratory muscle
fatigue. J. Appl. Physiol. 56: 568-575, 1984.

4. Bartlett, D., Jr. Effects of hypercapnia and hypoxia on laryngeal
resistance to airflow. Respir. Physiol. 37: 293-302, 1979.

5. Bartlett, D., Jr., S.L. Enuth, and K.V. 1muth. Effects of
pulmonary stretch receptor blockade on laryngeal responses to
hypercapnia and hypoxia. Respir. Physiol. 45: 66-77, 1981.

6. Bartlett, D., Jr., J.E. Remmers, and H. Gautier. Laryngeal
regulation of respiratory airflow. Respir. Physiol. 18: 194-204,

7. Bloch, R. Subtraction of the electrocardiographic signal from
respiratory electronyogram. J. Appl. Physiol. 55: 619-623, 1983.

8. Bodegard, G. Control of respiration in newborn babies. III.
Developmental changes of respiratory depth and rate responses to
002. Acta Paediatr. Scand. 64: 684-692, 1975.

9. Bodegard, G. Control of respiration in newborn babies. IV. Rib
cage stability and respiratory regulation. Acta. Paediatr. Scand.
65: 257-266, 1976.

10. Boychuk, R.B., H. Rigatto,and M.M.K. Seshia. The effect of lung
inflation on the control of respiratory frequency in the neonate.
J. Physiol. 270: 653-660, 1977.

11. Bradley, G.W., C. van Euler, I. Martilla, and B. Roos. A model of
the central and reflex inhibition of inspiration in the cat.
Biol. Cybern. 19: 105-116, 1975.

12. Brancatisano, T.P., D.S. Dodd, and L.A. Engel. Respiratory
activity of posterior cricoarytenoid muscle and vocal cords in
humans. J. Apple. Physiol. 57: 1143-1149, 1984.

13. Brancatisano, T.P., D.S. Dodd, and L.A. Engel. Responses of the
posterior cricoarytenoid and alae nasi muscles to increased
chemical drive in man. Respir. Physiol. 64: 177-189, 1986.

14. Breuer, J. Die selbststeuerung der atnung durch den nervous vagus.
Sitzungsber. Akad. Wiss. Wien Abt. 2 58: 909-937, 1868.

15. Bryan, H.M., R. Hagan, G. Gulston, and A.C. Bryan. Oa2 response
and sleep state in infants. Clin. Res. 24: 689A, 1976.

16. Carlo, W.A., P.C. Kosch, E.N. Bruce, K.P. Strohl, and R.J.
Martin. Control of laryngeal muscle activity in preterm infants.
Pediat. Res. 22: 87-91, 1987.

17. Carlo, W.A., R.J. Martin, E.F. Abboud, E.N. Bruce, and K.P.
Strohl. Effect of sleep state and hypercapnia on alae nasi and
diaphragm EWs in preterm infants. J. Appl. Physiol. 54:
1590-1596, 1983.

18. Carlo, W.A., R.J. Martin, and J.M. Difiore. Differences in C02
threshold of respiratory muscles in preterm infants. J. Appl.
Physiol. 65: 2434-2439, 1988.

19. Clark, F.J., and C. van Euler. On the regulation of depth and
rate of breathing. J. Physiol. London 222: 267-295, 1972.

20. Cohen, M.I. Phrenic and recurrent laryngeal discharge patterns
and the Hering-Breuer reflex. Am. J. Physiol. 228: 1489-1496,

21. Cross, K.W., J.M.D. Hooper, and T.E. Oppe. The effect of
inhalation of carbon dioxide on the respiration of the full-term
and premature infant. J. Physiol. 122: 264-273, 1953.

22. Davi, M., K. Sankaran, M. MacCallum, D. Cates, and H. Rigatto.
Effect of sleep state on chest distortion and on the ventilator
response to CO2 in neonates. Pediat. Res. 13: 982-986, 1979.

23. Dixon, M., M. Szereda-Przestaszewska, J.G. Widdeocube, and J.C.M.
Wise. Studies on laryngeal calibre during stimulation of
peripheral and central chemoreceptors, pneumthrax and increased
respiratory loads. J. Physiol. 39: 347-363, 1974.

24. Duara, S., S. Abbasi, T.H. Shaffer, and W.W. Foax. Preterm
infants: ventilation and P100 changes with CO2 and inspiratory
resistive loading. J. Appl. Physiol. 58: 1982-1987, 1985.

25. Durand, M., F.N. Inahy, M. NaoCallum, D.B. Cates, H. Rigatto, and
V. Ohernick. Effect of feeding on the chemical control of
breathing in the newborn infant. Pediat. Res. 15: 1509-1512,

26. Durand, M., E. MoCann, and J.P. Brady. Effect of continous
positive airway pressure on the ventilatory response to 0D in
preterm infants. Pediatrics 71: 634-638, 1983.

27. England, S.J., and D. Bartlett, Jr. Changes in the respiratory
movements of the human vocal cords during hyperpnea. J. Appl.
Rhysiol. 52: 780-785, 1982.

28. England, S.J., D. Bartlett, Jr., and J.A. Daubenspeck. Influence
of human vocal cord movements on airflow and resistance during
eupnea. J. Appl. Rhysiol. 52: 773-779, 1982.

29. England, S.J., K. Geraldine, and H.A.F. Stogryn. laryngeal muscle
and diaphragm activities in conscious dog pups. Respir. Physiol.
60: 95-108, 1985.

30. Euler, C. van. Brain stem mechanism for generation and control
of breathing pattern. In: Handbook of physiology. Section 3: the
respiratory system, edited by A.P. Fishman. Bethesda, MD:
American physiological Society, 1986, p. 1-54.

31. Eyzaguirre, C., and J.R. Taylor. Respiratory discharge of same
vagal mutonmeurns. J. Neurophysiol. 26: 61-78, 1963.

32. Fagenholz, S.A., K. O'Connell, and D.C. Shannon. Ch ep
function and sleep state in apnea. Pediatrics 58: 31-36, 1976.

33. Farber, J.P. laryngeal effects and respiration in the suckling
opposum. Respir. Physiol. 35: 189-201, 1978.

34. Feldman, J.L., and H. Gautier. Interaction of pulmonary afferents
and pneunotaxic center in control of respiratory pattern in cats.
J. Neurqphysiol. 39: 31-44, 1976.

35. Findley, L.J., A.L. Ries, G.M. Tisi, and P.D. Wagner. Hypzxamia
during apnea in normal subjects: Mechanisms and lupact of lung
volume. J. Appl. Physiol. 55: 1777-1783, 1983.

36. Fisher, J.T., J.P. Mortola, J.B. Smith, G.S. Fox, and S. Weeks.
Development of the control of breathing. Am. Rev. Respir. Dis.
125: 650-657, 1982.

37. Frantz, I.D., S.M. Adler, B.T. Thach, and H.W. Taeusch. Natura-
tional effects on respiratory responses to carbon dioxide in
premature infants. J. Appl. Physiol. 41: 41-45, 1976.

38. Fukuda, H., C.T. Sasaki, and J.A. Kirchner. Vagal afferent
influences on the phasic activity of the posterior cricoarytenoid
muscle. Acta Oto-laryngol. 75: 112-118, 1973.

39. Gardner, W.N. The relation between tidal volume and inspiratory
and expiratory times during steady-state carbon dioxide
inhalation in man. J. Physiol. 272: 591-611, 1977.

40. Gautier, H. Pattern of breathing during hypoxia or hypercapnia of
the awake or anesthetized cat. Respir. Physiol. 27: 193-206,

41. Gautier, H., J.E. Remmers, and D. Bartlett, Jr. Control of the
duration of expiration. Respir. Thysiol. 18: 205-221, 1973.

42. Gerhardt, T., and E. Bancalari. Chestwall compliance in full-term
and premature infants. Acta Paediatr. Scand. 69: 359-364, 1980.

43. Griffiths, G.B., A. Nowaraj, and J.P. Mortola. End-expiratory
level and breathing pattern in the newborn. J. Appl. Physiol. 55:
243-249, 1983.

44. Grunstein, M.M., M. Younes, and J. Milic-Emili. Control of tidal
volume and respiratory frequency in anesthetized cats. J. Appl.
Physiol. 35: 463-476, 1973.

45. Haddad, G.G., H.L. Leistner, R.A. Epstein, M.A.F. Epstein, W.K.
Grodin, and R.B. Mellins. 00 -induced changes in ventilation and
ventilatory pattern in noai sleeping infants. J. Appl. Physiol.
48: 684-688, 1980.

46. Hagan, R., and A.G. Gulston. The newborn: Respiratory electro-
myograms and breathing. Aust. Paediatr. J. 17: 230-231, 1981.

47. Harding, R. State-related and developmental changes in laryngeal
function. Sleep Basel 3:307-322, 1980.

48. Harding, R., P. Johnson, and M.E. McClelland. Respiratory
function of the larynx in developing sheep and the influence of
sleep state. Respir. Rhysiol. 40: 165-179, 1980.

49. Haxhiu, M.A., J. Mitra, E. van lunteren, M.D. Goldman, and N.S.
Cherniack. Differential effects of chemical stimuli on the
activity of the diaphragm (D) and upper airway dilating muscles
(UAIIM). Fed. Proc. 42: 331, 1983.

50. Hendersan-Smart, D.J., and D.J.C. Read. Reduced lung volume
during behavioral active sleep in the newborn. J. Appl. Piysiol.
46: 1081-1085, 1979.

51. Heaxerson-eQart, D.J., P. Johnson, and M.E. McClelland.
Asynchronous respiratory activity of the diaphragm during
spontaneous breathing in the lamb. J. Riysiol. LInd. 327:
377-391, 1982.

52. Honma, Y., D. Wilkes, M.H. Bryan, and A.C. Bryan. Rib cage and
abdominal contributions to ventilator response to C2 in
infants. J. Appl. Physiol. 56: 1211-1216, 1984.

53. Hwang, J.C., W.M. St. John, and D. Bartlett, Jr. Respiratory-
related hypoglossal nerve activity: Influence of anesthetics. J.
Appl. Physiol. 55: 785-792, 1983.

54. Johnson, P., J.E. Fewell, L.M. Fedarko, and J.C. Wollner. The
vagal control of breathing in postnatal life: Implications for
sleep related respiratory failure. In: Sudden Infant Death
SyndraLe, edited by J.T. Tildon, L.M. Roeder, and A.
Steinscineider. landon: Academic, 1983, p. 467-490.

55. Kalapesi, Z., M. Durand, F.N. Leahy, D.B. Cates, M. MacCallum,
and H. Rigatto. Effect of periodic or regular respiratory pattern
on the ventilator response to low inhaled OD in preterm
infants. Am. Rev. Respir. Dis. 123: 8-11, 1981.

56. Knax, C.K. Characteristics of inflation and deflation reflexes
during expiration in the cat. J. Neurophysiol. 36: 284-295, 1973.

57. Knox, C.K. Reflex and central mechanisms controlling expiratory
duration. In: Central Nervous Control Mechanisms in Breathing,
edited by C. vaon Euler and H. lagercrantz. Oxford, UK: Pergamon,
1979, vol. 32, p. 203-216. (Wenner-Gren Ctr. Int. Synp. Ser.)

58. Koehler, R.C., and B. Bishop. Expiratory duration and abdcninal
muscle responses to elastic and resistive loading. J. Appl.
Riysiol. 46: 730-737, 1979.

59. KoSch, P.C., P.W. Davenport, J.A. Wozniak, and A.R. Stark. Reflex
control of expiratory duration in newborn infants. J. Appl.
Physiol. 58: 575-581, 1985.

60. Fosch, P.C., P.W. Davenport, J.A. Wozniak, and A.R. Stark. Reflex
control of inspiratory duration in newborn infants. J. Appl.
Physiol. 60: 2007-2014, 1986.

61. Kosch, P.C., A.A. Hutchison, J.A. Wozniak, W.A. Carlo, and A.R.
Stark. Posterior cricoarytenoid and diaphragm activities during
tidal breathing in neonates. J. Appl. Physiol. 64: 1968-1978,

62. RKsch, P.C., and A.R. Stark. Determination and homeostasis of
functional residual capacity (FRC) in infants. The Physiologist
22: 71, 1979.

63. Kosch, P.C., and A.R. Stark. Dynamic maintenance of end-
expiratory lung volume in full-term infants. J. Appl. thysiol.
57: 1126-1133, 1984.

64. Krauss, A.N., D.B. Klain, S. Waldman, and P.A.M. Auld.
Ventilatory response to carbon dioxide in newborn infants.
Pediat. Res. 9: 46-50, 1975.

65. Lopata, M., E. Onal, and A.S. Ginzburg. Respiratory muscle
function during CD rebreathing with inspiratory flow-resistive
loading. J. Appl. ?hysiol. 54: 475-482, 1983.

66. Lopes, J., N.L. Muller, M.H. Bryan, and A.C. Bryan. Importance of
inspiratory muscle tone in naintenanoe of FRC in the newborn. J.
Appl. Physiol. 51: 830-834, 1981.

67. Martin, R.J., W.A. Carlo, S.S. Robertson, W.R. Day, and E.N.
Bruce. Biphasic response of respiratory frequency to hypercapnea
in preterm infants. Pediat. Res. 19: 791-796, 1985.

68. McCaffrey, T.V., and E.B. Kern. Laryngeal regulation of airway
resistance. II. Pulmonary receptor reflexes. Ann. Otol. Rhinol.
Laryngol. 89: 462-466, 1980.

69. Miller, H.C. Effect of high coxmentration of carbon dioxide and
oxygen on the respiration of fullterm infants. Pediatrics 14:
104-113, 1954.

70. Moriette, G., P. Van Reempts, M. Moore, D. Cates, and H. Rigatto.
The effect of rebreathing 00D on ventilation and diaphragmtic
electrumyography in newborn Infants. Respir. Rhysiol. 62:
387-397, 1985.

71. Martola, J.P., D. Magnante, and M. Saetta. Expiratory pattern of
newborn mammals. J. Appl. Physiol. 58: 528-533, 1985.

72. Mortola, J.P., J. Milic-Emili, A. Noworaj, B. Smith, G. Fox, and
S. Weeks. Muscle pressure and flow during expiration in infants.
Am. Rev. Respir. Dis. 129: 49-53, 1984.

73. O'Brien, M.J. Respiratory EMS findings in relation to periodic
breathing in infants. Early Human Dev. 11: 43-60, 1985.

74. Oliven, A., E.C. Deal, Jr., S.G. Kelsen, and N.S. Cherniack.
Effects of hypercapnia on inspiratory and expiratory muscle
activity during expiration. J. Appl. Physiol. 59: 1560-1565,

75. Olson, L.G., J.M. Fouke, P.L. Hoekje, and K.P. Strohl. A
bianuchanical view of upper airway function. In: Respiratory
Function of the Upper Airway, edited by O.P. Mathew and G.
Sant'Ambrogio. New York: Dekker, 1988, p. 374-375.

76. Onal, E., M. Inpata, and T.D. O'Comor. Diaphragmatic and
genioglossal electrmanyogram responses to C00 rebreathing in
humans. J. Appl. Physiol. 50: 1052-1055, 1981.

77. Patrick, G.B., K.P. Strohl, S.B. Rubin, and M.D. Altose. UgIer
airway and diaphragm muscle responses to chemical stimulation and
loading. J. Appl. Physiol. 53: 1133-1137, 1982.

78. Pengelly, L.D., A.M. Tarshis, and A.S. Rebuck. Contribution of
rib cage and abdoaen-diaphragm to tidal volume during 002
rebreathing. J. Appl. Itysiol. 46: 709-715, 1979.

79. Prechtl, H.F. The behavioral states of the newborn infant. Brain
Res. 76: 185-212, 1974.

80. Radvanyi-Bouvet, M.F., M. Manset-Ocuohard, F. Mmrel-Kahn, and C.
Dreyfus-Brisac. Expiratory patterns during sleep in normal full-
term and premature neonates. Biol. Neonate 41: 74-84, 1982.

81. Reamers, J.E., H. Gautier, and D. Bartlett, Jr. Factors
controlling expiratory flow and duration: In: loaded Breathing,
edited by L.D. Pengelly, A.S. Rebuck, and E.J.M. Campbell. Don
Mills, Ont.: Iongman, p. 122-129, 1974.

82. Richter, D.W. Generation and maintenance of the respiratory
rhythm. J. Exp. Biol. 100: 93-107, 1982.

83. Richter, D.W., and D. Ballantyne. Central rhythm generation by
inspiratory neurones in the anesthetised cat. Neuroeci. Lett.
Suppl. 7: S117, 1981.

84. Rigatto, H. Control of ventilation in the newborn. Ann. Rev.
Rhysiol. 46: 661-674, 1984.

85. Rigatto, H., J.P. Brady, and R.T. Verduzoo. Cemoreceptor
reflexes in preterm infants: II. The effect of gestational and
postnatal age on the ventilatory response to inhaled carbon
dioxide. Pediatrics 55: 614-620, 1975.

86. Rigatto, H., U. Desai, F. Leahy, Z. Kalapesi, and D. Cates. The
effect of 2% C00, 100% 0?, theophylline and 15% 0 on
inspiratoryy drve' and 'effective timing' in pre infants.
Early Hum. Dev. 5: 63-70, 1981.

87. Rigatto, H., Z. Kalapesi, F,N, Leahy, M. Durand, M. MacCallum,
and D. Cates. Chemical control of respiratory frequency and tidal
volume during sleep in preterm infants. Respir. 1hysiol. 41:
117-125, 1980.

88. Roberts, J.L., W.R. Reed, O.P. Mathew, and B.T. Thach. Control of
respiratory activity of an upper airway dilating muscle
(genioglossus) in infants. Pediat. Res. 16: 305A, 1982.

89. St. John, W.M., D. Bartlett, Jr., K.V. Knuth, and J.C. Hwang.
Brain stem genesis of automatic ventilatory patterns independent
of spinal mechanisms. J. Appl. Physiol. 51: 204-210, 1981.

90. Sherrey, J.H., and D. Megirian. Spontaneous and reflexly evoked
laryngeal abductor and adductor muscle activity of cat. Exp.
Neurol. 43: 487-498, 1974.

91. Sherrey, J.H., and D. Megirian. Respiratory BMG activity of the
posterior criooarytenoid, cricothyroid and diaphragm muscles
during sleep. Respir. Physiol. 39: 355-365, 1980.

92. Siafakas, N.M., M. Bonara, B. Duron, H. Gautier, and J.
Milic-Emili. Dose effect of pentobarbitol sodium on control of
breathing in cats. J. Appl. Physiol. 55: 1582-1592, 1983.

93. Smith, C.A., D.M. Ainsworth, K.S. Henderson, and J.A. Dehpsey.
Differential timing of respiratory muscles in response to
chemical stimuli in awake dogs. J. Appl. Ihysiol. 66: 392-399,

94. Stahlman, M., and C. Sexton. Ventilation control in the newborn:
Carbon dioxide tension and output. Am. J. Dis. Child. 101:
216-227, 1961.

95. Stark, A.R., B.A. Cohlan, T.B. Waggener, I.D. Frantz III, and
P.C. Kosch. Regulation of end-expiratory lung volume during sleep
in premature infants. J. Appl. Physiol. 62: 1117-1123, 1987.

96. Strohl, K.P., C.F. O'Cain, and A.S. Slutsky. Alae nasi activation
and nasal resistance in healthy subjects. J. Appl. Physiol. 52:
1432-1437, 1982.

97. Taeusch, H.W., S. Carson, I.D. Frantz, and J. Milic-Emili.
Respiratory regulation after elastic loading and COD rebreathing
in normal term infants. J. Pediatr. 88: 101-111, 1936.

98. Thach, B.T., I.F. Abroms, I.D. Frantz III, A. Sotrel, E.N. Bruce,
and M.D. Goldman. Intercostal muscle reflexes and sleep breathing
patterns in the human infant. J. Appl. Physiol. 48: 139-146,

99. Tripperbach, T. Effects of drugs on the respiratory control
system in the perinatal period and during postnatal deelopnt.
trmnnac. Other. 20: 307-340, 1983.

100. Trippenbach, T., and G. Kelly. Ihrenic activity and interdostal
muscle EM during inspiratory loading in newborn kittens. J.
Appl. thysiol. 54: 496-501, 1983.

101. Trippenbach, T., R. Zinman, R. Moses, and L.M. Murphy.
Differences and similarities in the control of breathing pattern
in the adult and neonate. In: Central Nervous Control Meadanisms
in Breathing, edited by C. van Euler and H. Lagercrantz. Oxford,
UK: Pergamon, Vol. 32, p. 53-69, 1979.

102. Wlitelaw, W.A., J.P. Derenne, J. Couture, and J. Milic-Emili.
Adaptation of anesthetized men to breathing through an
inspiratory resistor. J. Appl. Physiol. 41: 285-291, 1976.

103. Younes, M. and M. Youssef. Effect of five human anesthetics on
respiratory control in cats. J. Appl. Iiysiol. 44: 596-606, 1978.

104. Zuperku, E.J., F.A. Hcpp, and J.P. Kampine. Central integration
of pulmonary stretch receptor input in the control of expiration.
J. Appl. Thysiol. 52: 1296-1315, 1982.


John Alexander Wozniak was born in St. Louis, Missouri, on

August 27, 1957. He grew up in Mt. Prospect, Illinois, and graduated

in 1975 from John Hersey High School, Arlington Heights, Illinois. He

received a B.S. degree in electrical engineering from the

Massachusetts Institute of Technology, Cambridge, Massachusetts, in

1979. While at M.I.T., he was a member of Delta Tau Delta fraternity

and captain of the basketball team. He accepted a position as a

biomedical engineer in the College of Veterinary Medicine at the

University of Florida in November, 1979. He was accepted as a graduate

student in the College of Engineering in 1980 and received a Master of

Engineering degree in Electrical Engineering from the University of

Florida in 1986. He was admitted to the Department of Physiology in

the College of Medicine at the University of Florida in 1983. He was

married to Brenda Franey on March 22, 1981. They have two children;

Briana Noelle Wozniak, born December 28, 1985, and Alexis Nicole

Wozniak, born October 31, 1988.

I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.

ip sch, Cairman
Professor of Physiology

I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.

Marc Jeger
Professor of Vbysology

I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.

Wendell Stainsby v
Professor of Physiology

I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.

Paul Davenport (I
Associate Professor of Physiology

I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.

Assoc ate Neuroscience

This dissertation was submitted to the Graduate Faculty of the
College of Medicine and to the Graduate School and was accepted as
partial fulfillment of the requirements for the degree of Doctor of

Decezzber, 1989 _____
Dean, College of Medicine

Dean, Graduate School

I3 1262 08557 0041
3 1262 08557 0041

Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd