JOURNALOF APPLIED Vol. 38, No. 6, June

PHYSIOLOGY 1975. Printed

in U.S.A.

Effect of resistive on ventilatory

loading

response

A. S. REBUCK AND E. F. JUNIPER Department of Medicine, McMaster lhiveysi&

to hypoxia

Hamilton,

REBUCK, A. S ., AND E. F. JUNIPER. Effect of resistiue loading on ventilatory response to hypoxia. J. Appl. Physiol. 38(6) : 965-968. 1975.-Ventilatory responses to hypoxia, with and without an inspiratory resistive load, were measured in eight normal subjects, using a rebreathing technique. During the studies, the end-tidal PCOZ was kept constant at mixed venous level (Pvco,) by drawing expired gas through a variable COT-absorbing bypass. The initial bag 02 concentration was 24 y0 and rebreathing was continued until the 02 concentration in the bag fell to 6 y0 or the subject’s arterial oxygen saturation (Sao,), monitored continuously by ear oximetry, fell to 70y0. Studies with and without the load were performed in a formally randomized order for each subject. Linear regressions for rise in ventilation against fall in Sao, were calculated. The range of unloaded responses was 0.78-3.59 l/min per 1 y0 fall in Sa o2 and loaded responses 0.37-1.68 l/min per 1 y0 fall in Sao,. In each subject, the slope of the response curve during loading fell by an almost constant fraction of the unloaded response, such that the ratio of loaded to unloaded slope in all subjects ranged from 0.41 to 0.48. However, the extrapolated intercept of the response curve on the SaOz axis did not alter significantly indicating that the Pco 2 did not alter between experiments. These results suggest that the change in ventilatory response to hypoxia during inspiratory resistive loading is related to the mechanical load applied, with the loaded slope being directly proportional to the unloaded one.

ventilation

I T IS ALMOST 20 YEARS since Cherniack and Snidal (3) first attempted to simulate the flow-limited breathing pattern of patients with airways obstruction by applying resistive loads to normal subjects. In their study, and in many of the subsequent investigations by others (4, 6, 15), chemical drive to breathing was assessed by measuring the ventilatory response to increasing levels of carbon dioxide. Patients with lung disease such as asthma (19) and emphysema (12) are more frequently hypoxernic than hypercapnic. The effect of resistive loading on hypoxic ventilation in man would therefore be of cpnsiderable clinical and physiological interest, but there are surprisingly few studies in the literature. In this paper, we exarnine the effect of a resistive load on hypoxic ventilatory response in normal subjects since we wanted to know whether change in responsiveness is related to the unloaded response. We have used a rebreathing technique which provides a relationship between ventilation and isocapnic hypoxia that is linear in form when oxygen saturation is taken as the independent variable ( 17).

Ontario,

SUBJECTS

Canada

AND

METHODS

Ventilatory response to hypoxia was studied in eight subjects with and without an inspiratory resistive load. Subjects. The subjects, seven males and one female, whose ages ranged from 22 to 49 yr, were physically fit, and one (subj 4) was in athletic training. All were nonsmokers and had normal spirometric function (Table 1). None had been born or had lived at high altitude. Although all were familiar with the techniques of respiratory physiology and were informed of the protocol of the study, they were not informed about the results of any study until all had been completed. Methods. Details of the experimental circuit have been described previously (18). The subjects rebreathed from a 6-liter bag in which the PCO~was kept constant using a variable COZ-absorbing bypass. Initially the bag was filled with a gas mixture which had an OX concentration of 22-24 76, 8 7% CO 2, and balance N2. Tidal gas was sarnpled at the mouth; after passing through a Godart Uras Capnograph and Servomex 0 2 analyzer (type OA-150), it was returned to the rebreathing bag. The 02 and CO2 analyzers were calibrated with gases that had been analyzed previously using the Lloyd-Haldane technique. Oxygen saturation of arterialized earlobe blood was measured with a Waters oximeter (type XP-350), calibrated and set on the ear as described previously (17). The ear oximeter was calibrated for each subject using samples of arterialized capillary blood taken from the other ear during rebreathing. Oxygen saturation in the blood samples was measured using a Radiometer (type OSMl) oxygen saturation meter. The resistive load was formed by a rigid tube, closed at one end, with several large perforations in its wall, wrapped in absorbent paper. The resistance chosen was cornparable with those used by others during CO2 studies (20 cmH20/1 per s) and had linear pressure/flow characteristics up to a flow rate of 3.0 l/s. The resistance was placed on the inspiratory side of an Otis-McKerrow valve, between the valve and the gasmeter, to avoid pressure being exerted on the gasmeter (Fig. 1). For unloaded studies, the resistance of the circuit was 0.5 cmH20/1 per s at a flow rate of 5 l/s. Studies of ventilatory response to decreases in arterial oxygen saturation, with and without loading, were performed in a formally randomized order for each of the subjects. Each study in any given subject was conducted in a single day, and at least 30 min was allowed between

965 Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.237.122.245) on January 16, 2019.

966

A. S. REBUCK

TABLE 1. Physical characteristics, uentilatory function, and mixed uenous PCO~ in subjects studied Subj No.

Age, yr

Sex

Ht, cm

Wt, kg

FE&, liters

VC, liters

22 27 24 30 23 25 49 33

F M R/1: WI M M M WI

165 185 185 168 183 175 182 183

62 78 79 75 92 77 82 85

3.8 6.1 4.5 4.4 6.0 4.8 4.7 4.2

4.1 7.8 5.7 5.2 7.4 6.0 6.0 6.9

PCo,* mmHg 45 46 58 52 45 49 45 44

AND

E. F. JUNIPER

the established plateau using the variable CO z-absorbing bypass. The oxygen concentration in the bag was allowed to fall as a result of the subject’s oxygen consumption. The experiment was terminated when the oximeter reading indicated 70 % saturation, or when the 02 analyzer indicated a concentration of 6 % 02 in the sampled gas (whichever occurred first). All gas volumes were corrected to BTPS, and the slopes of response curves were calculated using the least-squares regression method. The slopes and intercepts of the unloaded and loaded experiments were compared using a two-tailed paired Student t-test. RESULTS

PUMP

.ERROW

1 I

I I

I

I

I’ I

I ’

Plots of ventilation in response to decreases in oxygen saturation (AVE/ASa& were best fitted to a linear regression. The range of AVE/ASa,z for the unloaded experiments was 0.78-3.59 l/min per 1 % fall in Sao2 (mean =t SEMI, 1.64 =t 0.325). Ventilatory response to hypoxia decreased with resistive loading (Fig. 2); during loaded breathing, the range of ventilatory response to hypoxia was 0.37-1.68 l/min per 1 % fall in Sao2 0.73 =t 0.152) (Table 2). The change in slope of the individual response

RESISTIVE -LOAD

0 GAS VOLUME METER

1. Diagram response to

FIG.

tory

of rebreathing hypoxia. An

jJ

circuit inspiratory

used for resistive

measuring load has

experiments. Subjects were asked not to take tea or coffee within 2 h before any experiment and were also asked to empty their bladders before starting any study. After subjects were seated comfortably for 10 min and the readings of the oximeter were stable, they began breathing room air through the mouthpiece while endtidal Pco:! was monitored. When a stable end-tidal Pco:! had been achieved (& 1 mmHg), which usually took 5 min, the subject was switched into the rebreathing bag at the end of a normal expiration. The subject was then asked to take three large breaths to facilitate mixing between the lungs and bag. After 15-20 s from the start of rebreathing, when an end-tidal (mixed venous) Pco2 plateau had been attained. the PCO~ was held constant at a level equal to

LOADED

0

UNLOADED

1 70

I

I 80

1

I 90

1

1 100

Sa02

ventilabeen in-

corporated.

+

2. Ventilatory response to falls in oxygen saturation with (+) without (0) inspiratory resistive loading in one subject. Isocapnic was identical in each procedure.

FIG.

and level

TABLE 2. Calculated slopes of linear regressions of ventilatory res-onse to hypoxia with and without resistive loading Ventilatory Subj No.

1 2 3 4 5 6 7 8

Unloaded, l,/min per 1% fall in Saoz 0.78 0.90 0.98 1.29 1.73 1.86 2.06 3.59

Response

to Hypoxia l/min

Loaded, per 1% fall in Sac2 0.37 0.39 0.43 0.59 0.71 0.74 0.92 1.68

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.237.122.245) on January 16, 2019.

RESISTIVE

LOADING

DURING

967

HYPOXIA

curves was significant (P < O.Ol), but the extrapolated intercept in the Saoa axis did not change significantly (P > 0.2) in both instances using a two-tailed paired Student t-test. In all subjects, the ratio between the loaded and unloaded slope of the ventilatory response to hypoxia fell within the range 0.41-0.48. The best estimate of correlation of these data (Fig. 3), when the loaded slopes (r> were plotted against the unloaded slopes (x), was 0.993, the lower boundary for a 95 70 one-sided confidence interval being 0.979. The estimate of they intercept, when x is zero, is 0.039, which is not significantly different from zero (0.05 < P < 0.10). The fall in ventilation during loading occurred with decreases in both tidal volume and frequency (Table 3). These variables were compared at three levels of saturation, 90 %, 85 oio, and 80 %, and in all cases there was a significant fall in both tidal volume and frequency with loading, using a one-tailed two-sample t-test. DISCUSSION

This study has shown that resistive loading ventilatory response to hypercapnic isocapnic healthy I subjects. In the eight subjects studied s (y2.0 IUO

y = 0.464 r = 0.993

P m a*

I

I

x-O.039

I

I

I

1.0

A$ FIG. response text.

/ASa

3. Relationship to hypoxia

in

reduces the hypoxia in there was a

I

I

2.0 (UNLOADED)

between subjects

I

I 4.0

3.0 L./min./I%

loaded studied.

Fall

in

SaD2

and unloaded ventilatory For further explanation,

see

TABLE 3. Group values for changes in tidal volume and frequency during increasing hypoxia with and without resistiue loading in subjects studied Arterial 02 Saturation

Tidal ml

volume,

90 85 80

Frequency, breaths/min

Values

3,051 3,243 3,346

90 85 80 are

means

+

Unloaded

13.7 15.3 17.4 SD;

n =

Loaded

=t =t =t

745 748 714

=t =t rt

3.6 3.3 4.4

2,476 2,444 2,604

11.6 12.2

Signif,

P

=t =t h

716 594 1,052