Developmental Changes in the Ventilatory ...

Developmental Changes in the Ventilatory Response of the Newborn to Added Airway Resistance 1 - 4

WILLIAM A. LAFRAMBOISE, THOMAS A. STANDAERT, ROBERT D. GUTHRIE, and DAVID E. WOODRUM

Introduction

There has been considerable interest in the ventilatory response of adult humans and experimental animals to an externally applied resistiveload (1-6). In contrast, there is little information available on the response of the newborn to such a challenge. Yet this is an important issue because the relativelynarrow airways of the neonate increase its susceptibility to a flow-resistive impairment. Moreover, the newborn breathes predominantly through its nasal passages and, unlike the adult, switches to oral breathing only with considerable difficulty should the nasopharyngeal pathway become inflamed or obstructed (7). To further compound these problems, there is evidence which indicates that the chemosensory and mechanoreflex systems, critical for stabilizing ventilation, are immature at birth and require a postnatal transition period before they reach maturity (8-10). Thus, the newborn may be highly vulnerable to a flow-resistive challenge at a time when it is least able to respond to such a challenge. We proposed to study the response of the newborn to added flow resistances by using the infant monkey, Macaca nemestrina, as a homologue of the human neonate. Both species exhibit similar postnatal potentiation of their ventilatory sensitivity to decreasing oxygen or increasing carbon dioxide blood levels (9-12). Reflexes mediating respiratory mechanical stability also mature over the same developmental time span (8, 13). We theorized that immaturity of these chemosensory and mechanical reflexes in the immediate neonatal period could compromise the ability of the newborn to detect and/or compensate for a flowresistive impairment. Specifically, we hypothesized that the newborn monkey would be less able to maintain its normal minute ventilation as resistanceswere added external to its airway in contrast to the older infant whom we expected to

SUMMARY Postnatal development of the steady-state response to inspiratory resistive loading was studied In eight 48·hour-old and seven 24-day-old tracheostomlzed monkeys. The newborn subiects did not maintain minute ventilation (VI) with Increasing loads of from 2 to 6 times baseline respiratory resistance, whereas the older subjects kept VI constant when challenged by the same added resistances. The response patterns In both groups ware characterized by a prolongation of TI and TllTtot, a reduction of respiratory frequency, and increases In airway occlusion pressure and respiratory work output. Apart from VI, tidal volume (VT)was the only other ventilatory variable that differed significantly between age groups during loading. Arterial CO. and O. did not change from baseline in either group during loading, Indicating that both age groups defended blood gas values equally wall. The Increases in occlusion pressures, inspiratory work output, and the maintenance of Paeo. in the newborns indicated the presence of load compensatory mechanisms deAM REV RESPIR DIS 1987; 136:1075-1083 spite the fact that VI was not strictly defended.

defend ventilation more successfully through the same range of resistance. Methods

Subjects Fifteen Macaca nemestrina infant monkeys were delivered via cesarean section at the same gestational age (0.95 of term). A cannula was placed in an umbilical artery at delivery, and blood gases were monitored to determine if supplemental oxygen was necessary for maintenance of an adequate Pao, (Pao2 > 50 mm Hg). Because radical circulatory and mechanical changes occur in the immediate postnatal transition period, no studies were undertaken at this time to allow for lung fluid clearance and elimination of major fetal shunts (14). By 24 to 48 h of age, all infants exhibited a regular respiratory pattern and normal arterial gases while breathing room air, indicating that they had established stable airways and resting lung volume and were without persistent fetal circulation. They were studied at this time as one of the "newborn" subjects (n = 8), otherwise, the arterial cannula was removed and they were assigned to the "mature" infant group (n = 7) for study at 21 to 24 days of age. All infants were reared in a primate nursery according to standard protocol prior to and after completion of this study (15).

Physiologic Measurements The details of our system for studying these infants have been previously published (13). Briefly, they were studied lying supine on a sponge rubber mattress while lightly restrained with thin velcro straps across each

arm and leg. A flexible tracheostomy tube (0.25 cm H 20·L- I·min at 3 L/min) was inserted on Day 2 after infiltrating the incision site with chloroprocaine HCl, a topical anesthetic. It was necessary to use ketamine (10 mg/kg) in the older monkeys to achieve appropriate sedation for this surgical procedure. Repeat studies were not performed in the newborns because of concern that the tracheostomy repair potentially created a chronic internal resistive load that might alter normal development of the load response under investigation. Measurement of chest wall compliance was made by hyperventilating the infants to apnea while they were still under the effects of the anesthesia. The volume of gas required to hold the respiratory system at 30 em H 20 airway pressure was divided by the corresponding esophageal pressure to give an estimate of chest wall compliance. They were then connected via the tracheostomy tube to a low resistance, nonrebreathing valve (0.08 cm H 20. L-I·min at 3 L/min) and allowed a mini(Received in original form February 13, 1986 and in revised form April 24, 1987) I From the Division of Neonatal and Respiratory Diseases, University of Washington, Seattle, Washington. • Supported by Grants HL-19187 and RR-00166 from the National Institutes of Health. 3 Presented in part at the Annual Meeting of the American Physiological Society, August 1983. 4 Requests for reprints should be addressed to William A. LaFramboise, Department of Pediatrics, Magee-WomensHospital, Forbes Avenue and Halket Street, Pittsburgh, PA 15213.

1075

1076

mum of 2 h to recover from the anesthesia prior to commencement of the loading protocol. The inspiratory port of the nonrebreathing valve could be occluded while the subject was freely exhaling. Airway pressures generated during an occluded inspiration were monitored from a pressure transducer (P-50; Gould Instruments, Cleveland, OH) mounted on a sidearm of the tracheostomy tube. Arterial blood pressure was monitored from the catheter inserted into the umbilical artery in the newborns or the femoral artery in the older infants. These sites were also used for blood gas sampling; 0.5-ml samples were analyzed using an IL-713analyzer (Instrument Laboratories, Lexington, MA). Temperature was monitored and servo-controlled to 37° C using a radiant heat source triggered by an abdominal skin sensor. Volumetric ventilatory data were obtained by integration of flow changes at the inspiratory port of the nonrebreathing valve as the subjects inspired from a constant flow of air (Hewlett-Packard 8811Arespiratory integrator; Hewlett-Packard, Waltham, MA). These flow changes were determined by a hot-wire anemometer of negligible resistance located downstream in the inspiratory background flow (16). Timing data were calculated from the strip charts, which could be operated at speeds of as great as 100 mm/s when necessary for maximal resolution. Pulmonary mechanics measurements were made throughout the study using inspiratory flow and esophageal pressure obtained from a water-filled catheter with an ID of 1.5 mm. The catheter tip was located so as to obtain the greatest pressure excursions in phase with inspiratory flow but with minimal interference from cardiac pulsations. This was consistently found in the distal esophagus approximately 1 em above the cardiac sphincter, a location that we documented by fluoroscopy as well as direct measurement. Accuracy of esophageal pressure measurements was further substantiated by comparison of pressure magnitude and phase with airway pressure values obtained during inspiratory occlusion maneuvers (17). Dynamic lung compliance was calculated by dividing the inspired volume by the elastic pressure differential required to inspire that volume. This pressure difference was obtained by transecting the esophageal pressure trace from the zero flow point at end-expiration to the zero flow point at end-inspiration. The area of the esophageal pressure trace exceeding this elastic recoil line was considered to be the resistive pressure increment necessary to generate flow through the respiratory system during inspiration (18). Pulmonary resistance was calculated by integrating this resistivepressure increment by planimetry and then dividing that value by the concomitant integral of inspiratory flow (VT). A minimum of 50 such determinations was made during unloaded breathing, and the resistance of the valveand tracheal tube weresubtracted to give the average pulmonary inspiratory resistance

LAFRAMBOISE, STANOAERT, GUTHRIE, AND WOODRUM

of each subject. Resistance was calculated in this manner to offset the tendency for increasing lung volume to decrease resistance within a breath and so that resistive loads could be calibrated as a multiple of inspiratory resistance alone since inspiration was the only phase to which the loads were applied. The pulmonary work of breathing during inspiration was calculated by substituting the pulmonary elastance (reciprocal of dynamic lung compliance) and inspiratory resistance obtained during periods of baseline ventilation along with tidal volume and respiratory rate into the formula of Otis and coworkers (3) for a sinusoidal inspiratory flow. The work of moving the chest wall for the various ventilatory conditions was added to the pulmonary work data to obtain total internal inspiratory work (see Appendix). The work performed in moving air through the external encumbrances of the measurement system, e.g., the tracheal tube, the nonrebreathing valve, and the inspiratory resistors, represents the external work component of total inspiratory work. Because of the low resistance of the valve, external work in the unloaded condition was nearly zero. External work performed during loading was obtained as the integral of the product of inspiratory air flow and airway pressure measured at the tracheal tube opening. Calculations were based on whether the particular flow or pressure profiles most closely approximated a sinusoidal, triangular, or square waveform (see Appendix). These calculations were checked by randomly digitizing various of the waveforms for 4 infants and generating the integrals on a Minc-23 computer (Digital Electronics, Maynard, MA). At no time did we err by greater than 90/0 in calculating the integrals as compared to computer calculation.

Protocol The inspiratory resistors used in this study were stainless steel tubes placed in line with the inspiratory port of the nonrebreathing valve. These tubes provided a nearly linear resistance to breathing in the range of flows achieved by our subjects and were chosen such that the total inspiratory resistance of a subject receiving a given load was approximately 2, 4, or 6 times the unloaded inspiratory resistance. This range was chosen because a preliminary study verified that these subjects would attain stable ventilatory and sleep states while breathing through resistances of this magnitude. Interanimal variability in inspiratory resistance was large enough that the group means were not statistically different with age (Day 2: 0.11 ± 0.07 cm H 20/mIls versus Day 24: 0.13 ± 0.05 em H 20/mIls). The loads averaged 0.12 ± 0.03, 0.29 ± 0.03, and 0.61 ± 0.10 em H 20/mIls on Day 2 and 0.16 ± 0.04, 0.37 ± 0.13, and 0.67 ± 0.07 cm H 20/mIls in the older monkeys, with the individual added resistances being determined to a degree by the peak inspiratory flow rates generated by each subject. These resistances

will be referred to as loads A, B, and C for ease of presentation. The subjects werestudied without anesthesia except for topical administration of chloroprocaine HCI to the incision site(s)as necessary. Control and loading runs were undertaken only while the infants slept and did not display characteristics of rapid-eye-movement (REM) sleep. The inspiratory port could be immediately switched to the added resistance by simultaneously clamping and unclamping opposite ends of a Y tube connecting the nonrebreathing valve to the background flow. Upon imposition of the load, I min was allowed for adjustment on the part of the subject, and then ventilation was followed sequentially each minute for a minimum of 10 and a maximum of 20 min. Occlusions were performed at the end of each 5-min interval to obtain an index ofload compensation (19). If the infants aroused or exhibited obvious REM sleep, data were not collected.

Statistical Analysis Volumetric and timing components of ventilation, occluded and unoccluded airway pressures, respiratory mechanics, and work data were tested by analysis of covariance to determine the effect of loading. The linear relationship of the dependent variables to added resistance was calculated within each age group. The slope of these relationships was then compared to zero to establish whether the added resistances had an effect on a given dependent variable. In order to discriminate any effect of postnatal maturation, we tested the slopes of the 2 age group relationships for differences by comparison to the t distribution. A p < 0.05 was considered to represent a statistically significant difference. Because of maturational changes in VI, VT, P 0.2, lung and chest wall compliance, and blood gas values, certain of the variables collected during resistive loading are presented as a difference from baseline to focus on the effects of the loading regimen with age apart from baseline changes. In these cases, p values cited are from statistical tests performed on the original data. Results Ventilatory and Mechanical Response

The infants rarely aroused when switched to an inspiratory resistiveload regardless of age, and several continued to sleep throughout the 20-min stimulus period. Ventilatory and mechanical data reached a steady state within the first 2 min of loading and remained constant until either arousal, a sleep state change, or the arbitrary time limit caused termination of the run. We chose to focus on the values obtained at 10 min of loading as a representative steady state because data were available during NREM sleep for all subjects at this time, and those infants who tolerated the loads further did not deviate from this steady state. The effect

1077

DEVELOPMENTAL RESPONSE TO RESISTIVE LOADING IN THE NEWBORN

of added resistance on ventilation and with increased inspiratory muscle duty its components and arterial blood gases time (Tr/Ttot) and reduced breathing frewith air as the inspired gas are displayed quency (table 1). In figure 2, the changes in respiratory timing at either age can be in table 1. There was a significant fall in ventila- seen to be similar in magnitude as the tion as progressively larger resistances plotted points for TIand frequency overwere added to the airways of the 2-day- lap one another. Tidal volume, on the old infants such that the largest load other hand, increased in 16of the 21loadresulted in a VI approximately 7511/0 of ing trials in the older infants while the baseline value (figure 1). The older decreasing in 15of the 24trials performed subjects demonstrated a more variable in the 2-day-old infants, resulting in a sigresponse, with some subjects consistently nificant age effect (Day 2: slope ± SEM dropping ventilation and others match- = -0.321 ± 0.237 versus Day 24: 0.515 ing or exceeding baseline values during ± 0.346; group difference = p < 0.05). loading. Consequently, the regression Thus, both groups utilized a common equation for the older group described strategy of increasing TIto gain inspired a null effect of the stimulus regimen. volume in offsetting the decline in respiThere was a statistically significant group ratory rate, but the older infants were difference (p < 0.025) between the 2 age more successful in accomplishing this obgroups, indicating the older subjects jective. defended VI during loading, whereas the Tidal volume was the only ventilatory newborns did not (figure 1 and table 1). parameter apart from minute ventilation Despite these differences in the ventila- itself that was statistically different betory output of the test groups, at no time tween age groups during loading. The were arterial blood gas values altered other volumetric component of minute ventilation, VT/TI, decreased signifistatistically from baseline (table 1). In general, the ventilatory pattern with cantly from baseline in both groups, but which the 2 groups responded to resis- we were unable to discern an age effect. tive loading bore marked similarities. This was somewhat surprising in that VI Both groups significantly prolonged in- is the product ofVT/TI and Tt/Ttot, and, spiration (TI) on each load associated therefore, a greater fall in flow would be

anticipated in the younger infants during loading given similar timing changes as occurred in the older infants. Because we measured VT and TI with considerable precision (VT: 1 ± 0.1 ml to 10 ± 0.2 ml; TI accurate to 0.010 s) the lack of a statistical group effect likely reflects the interanimal variability that was inherent in each age group and increased with increasing resistive loads. In this regard, it should be considered that the mean VT/TI fell by 8, 21, and 2811/0 on loads A, B, and C in the younger infants while decreasing 8, 9, and 1311/0 in the older infants, indicating a stronger tendency for the mature subjects to defend this variable. Indices of inspiratory drive during loading wereelevated in both age groups, although to a greater degree in the older subjects. It can be seen in figure 3 that occluded airway pressures (P 0 2) increased significantly during loading at both ages, with the magnitude of the increase being greater in the older infants. Peak airway pressures also uniformly increased in all subjects at both ages during loading (figure 4). Again, the magnitude of the increase was greater in the older infants than in the neonates. The effect of resistive loading on re-

TABLE 1 VENTILATION, VOLUME AND TIMING COMPONENTS, AND ARTERIAL BLOOD GAS VALUES Day 24

Day 2 Baseline

A

B

Baseline

C

VI, ml/min

140 ± 24 127 ± 39 Difference within: p < 0.001

118 ± 28

2.00 ± 0.31

1.80 ± 0.44

Group difference: p 1.73 ± 0.50

< 0.025

VT, ml

3.45 ± 0.69

Group difference: p 3.18 ± 0.95

< 0.05

4.04 ± 1.09 Difference within: p < 0.005 0.52 ± 0.05

0.56 ± 0.05

1.98 ± 0.65

107 ± 31

278 ± 83

3.90 ± 0.48

Difference within: NS

4.39 ± 0.57

VT/TI, mils

0.51 ± 0.05 Difference within: p < 0.010

TI/Tlot

70 ± 15

65 ± 16 Difference within: p < 0.05

n, s

0.46 ± 0.08

0.49 ± 0.11 Difference within: p < 0.001

PaOz, mm Hg

84 11

89 15

66 ± 18

0.53 ± 0.12

8.38 ± 1.87

Group difference: NS 0.56 ± 0.08 0.55 ± 0.07 Group difference: NS 62 ± 14

71 ± 16

Group difference: NS 0.48 ± 0.09 0.55 ± 0.09

84

Group difference: NS 83

13

12

92 5


B

A

C

276 ± 110

278 ± 111

4.13 ± 0.63 4.22 ± 1.07 Difference within: NS

4.30 ± 0.85

7.74 ± 2.08 Difference within: p

7.33 ± 2.45

< 0.025


< 0.001

63 ± 17 Difference within: p

< 0.05


< 0.001

262 ± 125 Difference within: NS

7.61 ± 2.46

0.60 ± 0.09

65 ± 19

0.58 ± 0.12

96 9

84 13

0.62 ± 0.08

64 ± 19

0.62 ± 0.14

87 13

Difference within: NS Group difference: NS

Paco,. mm

Hg

39

38

8

8

39 7

39 8


32 2

32 2

33 2

33 1

Difference within: NS Group difference: NS

Definitionof abbreviations:itl = minute ventilation; VT = tidal volume; VTNI = mean inspiratory flow; Tlfflot = inspiratory duty cycle; f = respiratory frequency in breaths per minute; TI = inspiratory time; NS = no statistical change. • Values are mean ± SO and were obtained for n = 8 on Day 2 and n = 7 on Day 24. Statistical results of comparing the effect of the loading tests on a given group are reported as the difference within. A sign~icant value indicates that a change from baseline occurred during loading. The group difference indicates a maturational effect, i.e., the responses of the 2 groups were different from one another.

1078

LAFRAMBOISE, STANDAERT, GUTHRIE, AND WOODRUM

Day 2

Day 24

+ 100

• •

-

+50

....

e

.~

::>

•*

E

~

000

:l

:l

i=

elevated in each of the newborns during loading (table 2 and figure 5). The older subjects did not alter internal elastic or resistive work with loaded breathing as expected at a constant VI, but total respiratory work was significantly elevated above baseline because of an increase in the external work element that was considerably greater than the increase

100

.10

Ql

~

was associated with a significant fall in both internal elastic and resistive work per minute (table 2). Nevertheless, the total respiratory work increased significantlyas inspiratory airway pressures rose during the loading challenge. These airway pressure changes are an element of the external work component of the total work output, which was consistently

o

0

0

0IlJlIl

40

.3

x x 20

1.2.3A5.6.78910

Ti-Ioaded (sec)

20

40

60

80

Frequen::y -loaded

100

3

\fr-

loaded

Fig. 2. Effect of loading on volume and timing for 2-day-old infants (crosses) and 3-week-old SUbjects (circles). Results are pooled for all load levels. A value on the line of identity indicates no change during loading. VT was defended or increased in the mature subjects, whereas the trend was for loading to reduce VT in the younger infants (group difference: p < 0.05). 11 and respiratory frequency changed similarly at each age.

1079

DEVELOPMENTAL RESPONSE TO RESISTIVE LOADING IN THE NEWBORN

12.0

12.0

11.0

11.0

10.0

10.0

9.0

9.0

4000

6 J:'"

6

N

E S

J:

..•

E

S

~

OJ

rE

0."