Effects of nasal cold receptors on pattern of breathing

Effects of nasal cold receptors on pattern of breathing K. R. BURGESS AND W. A. WHITELAW Department of Medicine, University of Calgary, Calgary, Alberta T2N lN4, Canada BURGESS, K. R., AND W. A. WHITELAW. Effects of nasal cold receptors on pattern of breathing. J. Appl. Physiol. 64(l): 371376,1988.-To investigate the effect on the pattern of breathing of cooling receptors in the nose, eight normal male volunteers underwent a steady-state CO, stimulation by nasal inhalation. The inhaled gas temperature was randomly switched between “warm” (32°C) and “cold” (2°C) at each of three levels of inspired CO,! fraction (FI&. Breathing cold air through the nose reduced the mean slope of the ventilatory response to CO, by 27% (P < 0.05) and the mean intercept at PCO~ of 45 Torr by 6.6 l/min (P < 0.01). This was due mainly to a reduction in tidal volume (VT). Analysis of the breathing pattern recorded at a high level of minute ventilation (VE) (end-tidal partial pressure of CO2 -52 Torr) showed a reduction of VE that was due almost entirely to a reduction in VT (P < 0.05) associated with a reduction in inspiratory time (TI) as a fraction of total respiratory cycle time (P < 0.05) but little change in VT/TI. In a separate experiment conducted with five subjects, there was no significant difference in inspired nasal resistance between warm and cold runs during CO,-stimulated breathing. The results confirm the previous observation that cold air breathed through the nose inhibits ventilation in normal subjects and show that this is not related to an increase in flow resistance. The reduction in ventilation is due to reduction in VT associated with shortening of the duty cycle. temperature receptors; control of breathing; sponse to carbon dioxide; cold air

ventilatory

re-

FREQUENCY OF INVOLUNTARY inspiratory muscle contractions that occur in conscious subjects breath holding at functional residual capacity has been shown to be reduced by the stimulus of cool air flowing through the nose and pharynx at rates comparable to those of quiet breathing (11). The effect was more marked at higher flow rates and lower temperature of the air and was absent when the air was fully humidified at 37°C. It was abolished by topical anesthesia of the mouth and pharynx. This experiment suggested that there are receptors in the mucosa of the upper airway that are stimulated by cold or by the drying effect of cool air and have an inhibitory action on respiration. Cold air breathed through the nose was also found to decrease the ventilatory response to CO2 in a rebreathing experiment (4), but the details of its effect on the pattern of breathing were not clear and possible effects of cold air on upper airway resistance were not excluded. The present experiments using steady-state inhalations of gas with elevated COfLconcentrations and switching the temperature of inhaled gas back and forth, from cold to warm, were designed to confirm the previous observations, clarify the effect of cold air breathed THE

through the nose on the pattern of breathing, and determine the changes in nasal resistance produced by cold air in this type of experiment. The details of the change in breathing pattern might then be compared with the known effects of stimulating upper airway receptors in animals or of anesthesia of upper airway receptors in humans (22,29) and with the recently described changes in breathing pattern that result when the channel for normal breathing is switched from nose to mouth (14, 17) . METHODS

In the first experiment, eight normal male volunteers aged 18-35 yr were studied in the sitting position. They were not experienced in such studies and were not informed about the hypothesis being tested. The apparatus is shown diagrammatically in Fig. 1. The subjects inspired COs in O2 from a reservoir bag through a lowresistance one-way valve in the front of a size 6 anesthesia mask, (Medishield, UK) and exhaled via another low-resistance one-way valve in the side of the mask. Minute ventilation (VE) was measured by a heated Fleisch no. 2 pneumotachograph mounted in the expiratory side of the circuit. Flow was electrically integrated to give volume (HP 8815A, Hewlett-Packard, MA). The flow signal was calibrated daily (Checkmate, Bourns Medical Systems, CA), and the volume signal was calibrated before and after each experiment using a standard 2liter syringe. COs was monitored continuously below the nares with a mass spectrometer (Medspect II, Chemtron, MI) calibrated before and after each experiment using room air and 10% CO2 in Ns. Inspired air temperature was measured by a transistorized probe situated between the switching valve and the inspiratory one-way valve on the face mask. The temperature probe was calibrated before each experiment against a mercury thermometer over the range 0-40°C Air temperature, flow, volume, and CO2 were all displayed on a strip-chart recorder for subsequent analysis (Hewlett-Packard 7758B, HP 779619). The mask was adjusted to obtain an airtight seal and the subject was instructed to breathe only through his nose. The mask did not touch the nose, and air could flow freely into the nares. The subject was then allowed to equilibrate at one inspired CO2 fraction (FI& and temperature for 3-5 min before recordings were made. Three levels of FI coz were used (-2,4, and 6% COz, the balance 02). In the warm limb of the circuit, fully saturated warm gas at ~32°C was produced by passing gas from a gas mixer (Bird, CA) through a heater/humidifer

0161 7567/B $1.50 Copyright 0 1988 the American Physiological Society

371

372

NASAL

COLD AND PATTERN

OF BREATHING

Humidifier-Heater

1

f?eservoir)

Coolant

FIG. 1. Diagram of breathing circuit. See text for explanation.

ETco,, end-tidal Cog.

(Cascade 1, Puritan Bennett, CA) into a reservoir bag. In the cold limb of the circuit, cold dry air was produced by passing gas from the mixer through a low-resistance heat exchanger 1consisting of two concentric coils of 0.25 in. copper tube within an insulated PCV pipe. The gas passed through the PCV pipe and lost heat into the copper tubes, which were cooled by a slurry of ethanol and frozen CO2 pumped through them from a Dewar reservoir by a variable-speed roller pump (Sarns, MI). Coolant flow was varied to keep inspired air temperature at -OOC. The subject was then switched at the same inspired CO2 concentration to the other temperature and further recordings were made after ventilation and end-tidal PCO~ had reached a plateau (this took between 1 and 5 min, being faster at higher levels of VE). The level of the inspired CO2 was then adjusted, and the procedure was repeated to obtain data under six conditions (two temperatures at each of three levels of FI&. After equilibration at each temperature and FI~O, level, the data were recorded at a faster chart speed and 10 consecutive breaths were analyzed. The breaths were averaged to yield VE, tidal volume (VT), mean inspiratory flow rate (VT/TI), and inspiratory time as a fraction of total respiratory cycle time (TI/TT). The response to CO2 was analyzed bya first deriving the slope and intercept for regressions of \jE, VT, and respiratory frequency (f) against COa for each subject by leastsquares linear regression using all data points. The mean results were then compared by a Wilcoxon sign-rank test. To determine whether changes in nasal resistance were important under the conditions of the experiment, a second set of trials was done on five adult male subjects, two of whom had participated in the main experiment. The equipment and protocol were identical to the main experiment except that flow rate was measured by a pneumotachograph in the inspiratory line, immediately upstream from the inspiratory one-way valve. The pneu-

motachograph grid was heated during warm runs but not during the cold. The error due to the increase in volume and change in viscosity as the gas passed through the grid was calculated to be only -1% (21,25), since inspired gas was preheated to 32°C before reaching the pneumotachograph. Nasal resistance was measured by the technique of posterior rhinomanometry (18). Transnasal pressure was taken as the difference between mask pressure and pharyngeal pressure The subjects held between their lips a blocked-off mouthpiece that was connected to a pressure transducer (-1-50 cmHzO, Validyne Engineering, CA) to monitor pharyngeal pressure (Pph). Inspiratory nasal resistance was calculated as the average for five breaths of mask pressure (Pmask) - Pph/flow (V) for both warm and cold air breathing. Because the pressure-flow curve of the nose is not linear, measurements at identical flow rates of warm and cold were used for comparison. The resistances of the warm and cold limbs of the breathing circuit were measured separately. At a peak inspiratory flow of 1 l/s, resistance was 1.3 cmHaO .1-l. s in the warm limb and 1.4 cmHzOJ1s in the cold limb. At a peak inspiratory flow of 2 l/s, resistance was 3.7 cmHsO 4-l .s in the warm limb and 3.5 cmHpOl-los in the cold limb of the circuit. RESULTS

All the subjects were aware of the temperature of inspired gas and found the cool air fresh and pleasant. The data from the main experiment appear in Tables 1, 2, and 3. Examples of the results from a representative subject are shown in Figs. 2 and 3. Table 1 shows the individual values of end-tidal Pco~, VE, VT, and f at each temperature and inspired CO2 level. Comparisons between warm and cold air breathing at each-level of CO2 were made by the Wilcoxon sign-rank test. At the lowest F1c4, cold reduced VT by 18%, but tended to be higher and there was not a significant change in ventilation or

NASAL

COLD

AND

PATTERN

.** .*.** p”’ ..’ t/

OT”“1 30

40 pETco

FIG. 2. Example response to CO, PETE*,, end-tidal

60

50 (mmHg) 2

of effect of inhaled gas temperature on ventilatory during nasal breathing. VE, minute ventilation; Pco~. SE’s were smaller than symbols.

PO

o-o Cold VT o==*m-*a Warm VT Cold f

OF

BREATHING

373

A subset of values at PCO~ close to 52 Torr was analyzed by the Wilcoxon sign-rank test to find &fferences in details of the breathing pattern. This included the high FI~O, data from all subjects except stijects 3 and 7, and the medium FI coz from the latter. This selection allowed the patterns to be compared when VT’S were large and differences easier to discern, but kept the PCO~ within a narrow range, to avoid seeing possible effects of COa alone. The results showed a significant reduction in VE (P < 0.01) and VT (P < 0.01) and also in Tr/TT (P < 0.05) during cold gas breathing. Figure 4 graphically illustrates the change in breathing pattern using the “average spirogram” (16). (The set of data from the highest ventilation for all subjects was also compared and gave identical conclusions.) The effects of inspired air temperature on nasal resistance during CO,-stimulated nasal breathing appear in Table 3. Resistance was calculated as the transnasal pressure difference in centimeters of water, divided by inspired flow rate, in liters per second. Resistance during breaths with warm air was matched with resistance measured during breaths with cold ar that had the same inspired flow rate. The inspired flow rate during which resistance was measured averaged 1.18 l/s. There was no significant difference in inspiratory nasal resistance between warm and cold gas breathing in the five subjects studied (paired t test). DISCUSSION

sb P ETCO,

60

(mmHg) Subject

FIG. 3. Example of effect of inhaled gas temperature (VT) and respiratory frequency (f) during CO,-stimulated ing. PETIT,, end-tidal Pco~. -, SE, not shown when than symbols.

5

on tidal volume nasal breaththey are smaller

end-tidal Pco~. At the medium level of F1C02,end-tidal PCO~was higher breathing cold air, but VT was lower and VE was not different from breathing warm air. At the highest FI cop, the ch.ange in end-tidal PCO~was not significant, but VT and VE were lower in the cold Table 2 shows that the average slope of the ventilatory response to COZ was 27% less during nasal cold gas breathing compared with warm (1.45 1 min-’ Tori? cold vs. 2.011 mine1 Torr-l warm, P c 0.05) and the intercept at a PCOz = 45 Torr was lower. The reduction in slope of the VT response to CO2 was not significant, but the mean intercept at PC02 = 45 Torr was reduced by 0.37 or 23%. The means of the slopes and intercepts of the frequency response lines were not significantly different between the two temperatures. Although there were definite changes in slope between warm and cold air in many subjects, these changes were not systematic through the group. l

l

l

l

In these eight subjects the ventilatory response to CO2 was reduced by the nasal breathing of cold dry gas (from bottled gas sources having a low absolute humidity) compared with warm saturated gas. The reduction in slope was 27% (P < 0.05), which is comparable to that previously found in a different group of subjects by a rebreathing technique (4), and the values of ventilation at a PCO~of 45 Torr were also lower with cold air. The inhibition of ventilation during cold ges breathing was most evident at the higher levels of VE, and this was confirmed statistically by separate analysis of two sets of data from each subject at a PCO~ near 52 T.orr (see Table 4). In our previous study the reduction in VE could not be correlated with reductions in VT, f, TI, or VT/TI. In this experiment we have shown that the reduction in VE by nasal cold air breathing was due to a reduction in VT, without a significant change in frequency, but with a reduction in TI/TT. As cold air flows in through the nose, the mucosa is cooled by evaporation and heat transfer to the air. The rate of cooling increases with higher flow rates and lower temperature and lower water content of inspired gas. With higher flow rates the distal mucosa is cooled more quickly to a lower temperature, and cool air penetrates deeper into the nose, increasing the number of receptors exposed to low temperature. When flow rates are high and inspired air is cold and dry, exposed mucosa and its receptors may be desiccated as well as cooled. During inspiration of cool gas there is a temperature gradient from the air passage across the mucosal surface to the warm perfused submucosa. The location of temperature

374

NASAL

COLD AND PATTERN

OF BREATHING

1. Data summary

TABLE

Low Is*

Subj No.

Temp

1 2 3 4 5 6 7 8

Mean

. VE, l/min

pcol

Medium

VT, liters

f, breaths/min

pcoz

FIG%

.

VE9

VT,

I/min

liters

W C W C W C W C W C W C W C W C

34 35 35 35 40 41 40 42 42 41 39 36 41 43 38 40

21.8 20.6 25.5 18.4 16.0 16.6 12.6 9.0 11.6 12.8 9.9 11.4 14.3 14.5 18.7 17.8

1.51 1.01 1.60 1.20 1.11 1.07 0.83 0.60 0.80 0.70 0.92 0.93 0.83 0.79 1.60 1.20

14.5 20.5 15.8 15.4 14.4 15.5 15.0 15.1 14.5 17.5 10.8 12.4 17.1 18.4 12.4 14.6

40 40 40 40 51 53 46 48 45 47 43 44 50 53 45 46

27.8 32.1 31.6 27.1 34.0 22.4 23.7 18.3 18.7 14.4 13.0 12.4 28.2 25.0 26.7 24.8

2.12 1.57 1.67 1.54 1.89 1.32 1.70 1.21 1.26 0.92 1.30 1.07 1.60 1.28 2.10 1.70

W C

38.6 39.2

16.3 15.1

NS

NS

1.15 0.94 co.05

14.3 16.2 NS

45.0 46.9