- Email: [email protected]
The Effects of Acute Hypoxia and Hypercapnia on Pulmonary Mechanics in Normal Subjects and Patients with Chronic Pulmonary Disease
The Effects of Acute Hypoxia and Hypercapnia on Pulmonary Mechanics in Normal Subjects and Patients with Chronic Pulmonary Disease* Polly E. lbrsons, M.D.; Michael M. Grunstein, M.D.; and Enrique Fernandez, M.D.
The inftuence of progressive hypoxia and hypercapnia on respiratory mechanics was evaluated in 26 subjects (six normal subjects, seven asthmatic subjects, seven patients with IPD, and six patients with COPD). During separate rebreathing runs of progressive isocapnic hypoxia and normoxic hypercapnia, breath-to-breath changes in lb. and Cdyn were determined. In 6ve of the six normal subjects, seven of the seven asthmatic subjects, and six of the seven subjects with IPD, lb. decreased with both progressive hypoxia and hypercapnia without a change in Cdyn. In the
the effects of acute chemostimulation on W hile respiratory drive and the pattern of breathing in normal subjects and patients with pulmonary disease have been intensively studied, the influence of progressive hypoxia and hypercapnia on respiratory mechanics has not been systematically evaluated. The relatively few studies addressing this issue conducted in normal subjects have yielded conflicting results. In normal subjects exposed to acute hypoxia (Sa02 of 60 to 70 percent), Sterling1 demonstrated significant increases in both Raw and TGV. In contrast, MilicEmili and Petit2 evaluated pulmonary function in normal subjects in response to 10 percent oxygen and found no change in either RL or Cdyn. These data were subsequently confirmed by studies by Saunders et al3 and Goldstein et al. 4 Two studies in dogs suggest that acute hypoxia causes profound bronchoconstriction, as documented by decreased Raw.M The literature regarding the effect of acute hypercapnia on pulmonary mechanics is similarly confusing. Sterling7 demonstrated a decrease in airway specific conductance in normal subjects in response to acute hypercapnia. Nastro et al8 and Ahmed et al9 demonstrated an increase in specific conductance in normal subjects in response to hypercapnia. Furthermore, they documented an increase in FRC and TGV in response to breathing 6 percent carbon dioxide. Two
*From the Department of Medicine, National Jewish Center for Immunology and Respiratory Medicine, Denver, and the Department of Pulmonary Medicine, Children's Hospital, Philadelphia. Manuscript received January 22; revision accepted November 4.
98
patients with COPD, the effects of hypoxia and hypercapnia on lb. and Cdyn were variable. Compared to normal subjects, the changes in lb. during hypoxia and hypercapnia were not signi6cantly different in the asthmatic subjects and the patients with IPD. These data provide evidence that acute progressive hypoxia and hypercapnia are associated with signi6cant changes in Raw in both normal subjects and patients with chronic pulmonary disease. (Cheat 1989; 96:96-101)
other groups found no change in Raw in normal subjects exposed to hypercapnia. 10•11 To further complicate the issue, hypercapnia has been shown to cause bronchoconstriction in intact dogs, 5.6 whereas hypercapnic ventilation of isolated perfused canine lungs has resulted in bronchodilation. 7 These contradictions regarding the alterations in pulmonary mechanics in response to either acute hypoxia or acute hypercapnia cannot be readily explained. Moreover, the effects of acute hypoxia and hypercapnia on dynamic respiratory mechanics in patients with pulmonary disease remain to be identified. In an attempt to evaluate the pulmonary mechanics in these situations, we addressed a new issue: do patients with chronic pulmonary disease respond differently to acute hypoxia or acute hypercapnia (or both) than normal subjects? Our present study was designed to quantitatively assess the separate effects of acute progressive isocapnic hypoxia and acute progressive normoxic hypercapnia on pulmonary mechanics in normal subjects as well as patients with asthma, chronic obstructive pulmonary disease (COPD), and interstitial pulmonary disease (IPD). MATERIALS AND METHODS
Six healthy nonsmoking subjects with normal pulmonary function served as control subjects. The three groups of patients consisted of seven asthmatic subjects, six patients with COPD, and seven patients with IPD. Baseline measurements of pulmonary function, including VC, FEY, TGV, RV, and Raw, were measured with the subjects seated in a pressure-compensated body plethysmograph breathing through a heated No. 3 Fleisch pneumotachygraph. Volume was measured with a Krogh spirometer attached to the plethysmograph. The plethysmographic volume signal was pressure
Elfecls of Acute Hypoxia and Hypercapnia on Pulmonary Mechanics (Parsons, Glunsteln, Femendez)
Table 1-Ptdmonary Function• Group
n
Age, yr
FEV,L (percent predicted)
Normal Asthma IPD COPD
6 7 7 6
32±4 48±5 36±4 52±5
3.77±0.31 (101) 2.28±0.37 (64) 2.70±0.2 (73) 1.67±0.38 (48)
FVC,L (percent predicted) 4.72±0.50 3.28±0.50 3.34±0.26 3.07 ± 0.53
TGV,L (percent predicted)
FEV,IFVC%
(100) (69) (71) (62)
80±2 70±4 81±2 54±6
3.00±0.34 3.78±0.34 2.34±0.10 5.77±0.27
(92) (114) (71) (149)
Raw, em H 10/L/s
1.94±0.23 2.85±0.41 2.36±0.39 4.11 ±0.88
(119) (286) (146) (218)
*Values are means± SEM. Numbers in parentheses represent mean percent of predicted normal values. compensated and had a frequency response which was linear to 12 Hz. The pulmonary functional characteristics of the four groups of subjects are outlined in Table 1. All studies were performed in Denver (elevation, 1,585 m) when the patients were in a baseline stable clinical status. After obtaining consent for the procedure, each subject was seated comfortably with his eyes closed and allowed to listen to soothing music provided through headphones. The subjects wore nose clips and breathed via a mouthpiece attached to a pneumotachograph (Fleisch No. 3), which was connected to a low-resistance (1.3 em H 10/L/s) rebreathing circuit partitioned into separate inspiratory and expiratory limbs. After five minutes of stabilization on the circuit, progressive normoxic hypercapnia was produced by the method of Read,•• wherein subjects rebreathed from a bag initially containing a mixture of 6 percent carbon dioxide, 40 percent oxygen, and the balance nitrogen. The Sa01 was continuously monitored throughout the carbon dioxide rebreathing trial with an ear oximeter (HewlettPackard 47201A) and remained at greater than 92 percent. Each rebreathing trial lasted between four and six minutes. Progressive isocapnic hypoxia was produced by rebreathing from a bag initially containing room air in a volume equal to the patient's VC plus 1 L. The PETC01 was continuously monitored at the mouthpiece via a mass spectrometric gas analyzer (Perkin-Elmer 1100 Medical Gas Analyzer) and maintained within 2 mm Hg by means of a carbon dioxide absorber attached to the expiration limb of the rebreathing circuit. Volume was derived by integration of the airflow signal. Transpulmonary pressure was monitored by means of a differential pressure transducer (Statham PM 5 ETC), attached at one end to monitor mouth pressure and at the other end to monitor intraesophageal pressure via a catheter attached to an esophageal balloon placed in the lower third of the esophagus. The frequency response of the balloon-catheter system was linear up to 10 Hz. The measurement of Pr was related to the airflow at the mouth to obtain breath-to-breath determinations of RL. The RL was calculated by determining the changes in Pr and flow (~ V) at isovolumes (middle of inspiration and middle of expiration) so that RL (in liters per 13 • 14 The Cdyn was determined by continuously second)=~Pr/~V. recording the Pr and volume and recording the changes in Pr and volume (V) from the beginning of an inspiration to the end of inspiration (times of zero airflow) so that Cdyn (in liters per centimeter of water)=~ VI~Pr.••·•• The airflow, volume, Pr, PETC01 , and Sa01 signals were displayed on a multichannel pen recorder (Gould). In addition to the previously mentioned measurements, in order to determine any change in TGV during the course of rebreathing, three patients with asthma and two normal subjects underwent progressive isocapnic hypoxic challenges while seated in a body plethysmograph with the rebreathing circuit placed outside the plethysmograph and connected to the mouthpiece via a port in the plethysmograph. Serial measurements of TGV were made during the isocapnic hypoxic challenge. Relationships between breath-to-breath changes in RL and Cdyn vs changes in Sa01 and PETC01 were determined for each subject. The individual and grouped data were compared using Student's ttest for paired and unpaired analysis, where appropriate. Slopes of
relationships were compared using the Mann-Whitney (rank sums) analysis.
REsuLTS In all of the normal subjects, RL decreased in response to progressive isocapnic hypoxia, as exemplified by subject 5 in Figure 1. It can be seen that the relation is closely approximated by a linear function providing a regression slope of +0.1035 em H 20/l/s/ percent Sa02 (correlation coefficient= 0.81). Quantitatively similar results were obtained in the other normal subjects studied. The mean slope of the linear regression ofRL vs Sa02 was + 0.103 ± 0.053 em H 20/ Us/percent Sa02 for the normal subjects (Fig 2). Similarly, in all of the asthmatic subjects and six of the seven patients with IPD, RL decreased in response to acute progressive isocapnic hypoxia. The mean slope of the linear regression of RL vs Sa02 was Hypoxia
~
4
0N
r•0.81
l:
E
.....u....
•
2
a:
0~,~~----~----~--~----~--~
,.z
0.6
•
0.4
r=0.32
•
0
0
0.2
,
•
• • • ••••
I
I
70
80 Sa02
90
FIGURE lA (top). Relationship between RL (in centimeters of H 10 per liter per second) and hypoxia for representative normal subject 5. Slope of relationship is +0.1035 (r=0.81). B (bottom). Relationship between Cdyn (in centimeters of H.O per liter) and hypoxia for same subject (r=0.32). CHEST I 96 I 1 I JULY, 1989
97
+0.4
q,"' +0.2
en
0 >_.
a:
j Normal
Asthma
ILD
2. Data are grouped by disease. Bars represent mean slope for relationship of RL (in liters per second) vs Sa01 (percent) for each group. Error bars are SEM. FIGURE
+0.192±0.070 em H 20/Us/percent Sa02 for the asthmatic subjects and +0.0936±0.0399 em H20/U s/percent Sa02 for the subjects with IPD (Fig 2). The response to hypoxia of the patients with COPD was variable. In each of the normal and the asthmatic subjects and in three of the subjects with IPD, the decrease in RL was statistically significant (p<0.05), as can be seen in 'Thble 2; however, there was no significant difference in the responsiveness between the groups. The Cdyn was calculated for each subject. As can be seen from the data of a representative normal subject (subject 5) {Fig 1), Cdyn remained unchanged during progressive hypoxia despite an increase in respiratory frequency. This stability of Cdyn {in the face of an increased respiratory rate) was seen in all normal and asthmatic subjects and all subjects with IPD {'Thble 2). The RL also decreased in response to acute progressive norm oxic hypercapnia in five of the six normal subjects as exemplified by subject 5 in Figure 3. This relationship was also best approximated by a linear function. The regression slope of this subject was -0.2017 em H20/mVs/mm Hg PEC02 • The mean
Table 2-Slopea ofGraphl of Pulmonary Function* Group and Subject Normal Subject 1 Subject 2 Subject 3 Subject 4 Subject 5 Subject 6 Asthma Subject 7 Subject 8 Subject~
Subject 10 Subject 11 Subject 12 Subject 13 IPD Subject 14 Subject 15 Subject 16 Subject 17 Subject 18 Subject 19 Subject 20 COPD Subject 21 Subject 22 Subject 23 Subject 24 Subject25 Subject26
Slope of RL vs Sa01 , Us/percent
Slope of Cdyn VS sao., em H 10/Lipercent
Slope of RL vs PEC01 , Us/mmHg
Slope ofCdyn vs PEC01 , em H10/IJmm Hg
+0.0618 +Q.1690 +0.0501 +0.1497 +0.1035 +0.0654
(0.81)t (0.93)t (0.66)t (O.SO)t (0.81)t (0.78)t
-0.0088 -0.0127 +0.0013 +0.0016 -0.0061 +0.0009
(0.45) (0.16) (0.20) (0.21) (0.32) (0.11)
-0.1039 +0.0226 -0.0158 -0.2813 -0.2017 - 0.4700
+0.2516 +0.1069 + 0.0906 +0.0897 +0.1022 +0.5893 +0.1140
(0.62)t (0.74)t (0. 76)t (0.64)t (0.67)t (0.87)t (0.69)t
-0.0050 -0.0006 +0.0055 -0.0019 +0.0013 - 0.0034 + 0.0013
(0.49) (0.29) (0.27) (0.29) (0.33) (0.30) (0.33)
-0.5775 (0.93)t
+0.0018 (0.86)t
-0.1683 (0.81)t -0.1975 (0.84)t -0.5382 (0.60)t
-0.0025 (0.15) -0.0051 (0.44) -0.0018 (0.09)
+0.1000 +0.0093 -0.0390 +0.2368 +0.2286 +0.0834 +0.0362
(0.99)t (0.24) (0.18) (O.OO)t (O.BS)t (0.26) (0.47)
-0.0020 +0.0073 -0.0011 -0.0009 +0.0005 +0.0053 -0.0011
(0.37) (0.21) (0.11) (0.08) (0.09) (0.88)t (0.40)
-0.0643 (0.63)
+0.0038 (0.54)
-0.0689 -0.4609 +0.1041 -0.5915 -0.4lll
-0.0004 + 0.0038 - 0.0036 -0.0033 +0.0023
-0.6409 +0.1324 +0.0349 -0.0471 +0.0746 +0.0378
(0.62)t (0.67)t (0.36) (0.11) (0.53)t (0.53)
-0.0060 +0.0009 -0.0011 -0.0078 -0.0010 +0.0023
(0.09) (0.12) (0.37) (0.64) (0.04) (0.37)
(0.93)t (0.12) (0.61) (0.77)t (0.83)t (0.56)t
(0.60) (0.93)t (0.61) (0.92)t (0.87)t
-0.0037 -0.0022 -0.0035 +0.0078 -0.0055 -0.0053
(0.23) (0.15) (0.22) (0.54) (0.18) (0.51)
(0.09) (0.47) (0.49) (0.40) (0.36)
-0.0788 (0.53)
-0.0111 (0.42)
+0.0607 (0.51)
-0.0022 (0.17)
*Numbers within parentheses are r values. tp
98
Effecls of Acute Hypoxia and Hypercapnia on Pulmonary Mechanics (Pataons, Gtuns181n, Femandez)
Hypercapnia
s ~
plethysmograph, and serial measurements of TGV were determined as the subject's oxygen saturation was decreased from 92 percent to 75 to 78 percent using the previously described rebreathing circuit. There was no significant change in the TGV in response to progressive isocapnic hypoxia in any of the subjects studied (Fig 5).
4
2
•
E
u
~
a:
r=0.830
0~,~------~--------~--------~
0.4
,..z
a
0 0.2
. . ., ..
r=0.38
••• ••
0~~------~------~~------~
0
40
•
45
50
PnC02 FIGURE 3A (top). Relationship between RL (in centimeters of H1 0 per liter per second) and hypercapnia (in millimeters of mercury) for single normal subject 5. Slope of this relationship is -0.2017 (r=O.B3). B (bottom). Relationship between Cdyn (in centimeters of H10 per liter) and PEC02 (in millimeters of mercury) for same subject (r=0.36).
slope of the linear regression of RL vs PEC02 was -0.112 for the normal subjects studied. The RL also decreased linearly in response to hypercapnia in the subjects with asthma and IPD. The mean slope of these relationships was - 0.3703 for the asthmatic subjects and - 0.2488 for the subjects with IPD (Fig 4). The response was statistically significant (p<0.05; Mann-Whitney) in three of the six normal subjects, in three of the four asthmatic subjects, and in three of the five subjects with IPD ('Thble 2). The degree of responsiveness of the subjects with asthma and IPD was not different from that of the normal subjects. The patients with COPD experienced severe dyspnea in response to hypercapnia. Only two patients with COPD were able to tolerate the complete hypercapnic protocol, so no conclusions regarding that group can be drawn. The Cdyn did not change in response to acute progressive hypercapnia, as can be seen by the data from normal subject 5 (Fig 3). The data from each individual subject is shown in Thble 2. Previous investigators have suggested that TGV (or FRC) may be affected by hypoxia and could, therefore, account for the observed change in RL. 3•16.1 7 To evaluate whether TGV varied during our challenges, we studied three of the asthmatic subjects and two of the normal subjects. Each subject was seated in a volume body
DISCUSSION
We found that our normal nonsmoking subjects significantly decreased their RL in response to both acute progressive isocapnic hypoxia and acute progressive normoxic hypercapnia. The majority of the patients with asthma and IPD whom we studied also responded to those two stimuli with a decrease in RL. The changes in Raw in response to hypoxia and hypercapnia were not associated with changes in measured Cdyn despite an increase in respiratory frequency. This apparent lack of a frequency-dependent component to the serial measurements of Cdyn suggested that the changes in RL may not have been due primarily to changes in the small airways; however, an alternative explanation was that Cdyn was dependent on respiratory frequency but tha,t both hypoxia and hypercapnia caused sufficient bronchodilation of
-0.6
~
~
--
N
0
o....
-0.4
~
--
~
~_. a:
i
~
-0.2
~
'-
T
0 Normal
Asthma
ILD
FIGURE 4. Data are grouped by disease. Each bar represents mean slope of RL (in liters per second) vs PEC01 (in millimeters of mercury) for group. Error bars represent SEMs. CHEST I 96 I 1 I JULY. 1989
89
5 (J
-j ...~
~
~~
~ ~
4 l:l.
i
A
AADJt. A
• • fP• • •
3
0
0
0
cP
0 normal
80
85
90
95
0 1 Salu'atlon (")
100
FIGURE 5. Data on TGV measured at oxygen saturations of 80 to 100 percent in two normal and three asthmatic subjects.
the peripheral airways to preserve the measurements ofCdyn. The literature suggests that both large and small airways are affected by acute hypoxia and hypercapnia. Previous investigators have shown that the larynx and large airways account for a significant portion of the total RL measured in humans. 18•19 Furthermore, the human larynx has been shown to provide a variable controlled resistance which participates in the regulation of airflow. m This observation was elaborated on by England and associates, 21 who observed vocal cord movements during hypoxia and hypercapnia. Under the influence of both of these stimuli, the narrowing of the glottic aperture normally seen during expiration was minimal. 21 Similar observations have been made in animals. Bartlett et al22 showed that up to 50 percent of the respiratory resistance could be accounted for by the upper airway in cats. Both hypoxia and hypercapnia22 •23 significantly decreased laryngeal resistance in these animals, suggesting that the larynx plays a major role in determining respiratory patterns both at rest and in response to stimuli. There is evidence in dogs that the response may not be limited to the larynx. Strohl and Fouke24 showed that even after resection of the recurrent laryngeal nerve, there was an increase in negative pressure in the upper airway in response to hypercapnia, and they speculated that other regions of the upper airway participated in respiratory control. Other data suggest that the small airways may play a significant role in the changes in pulmonary mechanics in response to acute hypoxia and acute hyper100
capnia. Duane et al25 evaluated the response of isolated strips of distal pulmonary parenchyma to hypoxia and found a significant decrease in resistance. The response to hypocapnia was also evaluated and was found to be associated with an increase in resistance (suggesting that hypercapnia would produce a decrease in resistance). Thus, small airways and the larynx appear to both respond to hypoxia and hypercapnia with a decrease in resistance. In conclusion, normal subjects and patients with chronic pulmonary disease responded to acute progressive isocapnic hypoxia and acute progressive normoxie hypercapnia with a decrease in RL. There was a suggestion that the change in resistance was greater for asthmatic subjects than for subjects with IPD or normal subjects. These changes in the function of the airways in response to hypoxia and hypercapnia may be relevant to evaluate mechanisms of breathing and respiratory control and need to be considered in future studies of pulmonary mechanics. REFERENCES 1 Sterling GM. The mechanisms of bronchoconstriction due to hypocapnia in man. Clin Sci 1968; 34:277-85 2 Milic-Emili J, Petit JM. Effetes de Ia teneur en oxygene de I'air inspire sur les proprietes mechaniques des poumons chez l'homme normal. J Physiol (Paris) 1960; 52:175-77 3 Saunders NA, Betts MF, Pengelly LD, Rebuck AS. Changes in lung mechanics induced by acute isocapnic hypoxia. J Appl Physiol1977; 43:413-19 4 Goldstein RS, Zamel N, Rebuck AS. Absence of effects of hypoxia on small airway function in human. J Appl Physiol1979; 47:251-56 5 Nadel JA, Widdicombe JG. Effect of changes in blood gas tensions and carotid sinus pressure on tracheal wlume and total lung resistance to airflow. J Physiol1962; 163:13-33 6 Green M, Widdicombe JG. The effects of ventilation of dogs with different gas mixtures on airway calibre and lung mechanics. J Physiol1966; 186:363-81 7 Sterling GM. The mechanism of decreased specific airway conductance in man during hypercapnia caused by inhalation of 7% C01 • Clin Sci 1969; 37:539-48 8 Nastro JA, Weiss M, Soorani J, Lyons HA. The effects of carbon dioxide on airway resistance. Fed Proc 1960; 28:460 9 Ahmed S, Weiss M, Lyons HA. The effect ofC01 breathing on bronchomotor tone. Physiologist 1967; 10:107 10 Butler J, Caro CG, Alcala R, DuBois AF. Physiological factors affecting airway resistance in normal subjects and in patients with obstructive respiratory disease. J Clin Invest 1960; 39:58491 11 Jaeger MJ, Otis AB. Measurement of airway resistance with a wlume displacement body plethysmograph. J Appl Physiol 1964; 19:813-20 12 Read DJ. A clinical method for assessing the ventilatory response to carbon dioxide. Aust Ann Med 1966; 16:20-32 13 lngrani RH, Pedley TJ. Pressure flow relationships in the lungs. In: Macklem PT, Mead J, eds. Handbook of physiology: the respiratory system II. Bethesda, Md: American Physiological Society, 1986:277-93 14 Chemiack RM. Pulmonary function testing. Philadelphia: WB Saunders Co, 1977:27-29 15 Anthonisen NR. Tests of mechanical function. In: Macklem PT, Mead J, eds. Handbook of physiology: the respiratory system
Eflecls of Acute Hypoxia and Hypercapnia on Pulmonary Mechanics (Parsons, Grunsteln, Fernandez)
16
17 18 19
20
III. Bethesda, Md: American Physiological Society, 1986:75384 Garfinkel F, Fitzgerald R. The effects of hypercapnia, hypoxia and hyperoxia on FRC and the P 0.1 in humans. Physiologist 1975; 18:223 Kellogg RH, Mines AH. Acute hypoxia fails to affect FRC in man. Physiologist 1975; 18:275 Hyatt RE, Wilcox RE. Extrathoracic airway resistance in man. J Appl Physiol1961; 16:326-30 Ferris BG Jr, Mead J, Opie LH. Partitioning of respiratory flow resistance in man. J Appl Physiol1964; 19:653-58 England SJ, Bartlett JR, Daubenspeck JA. Influence of human vocal cord movements on airflow and resistance during eupnea.
Advances
J Appl Physiol1982; 52:773-79 21 England SJ, Bartlett D, Knuth SL. Comparison of human vocal cord movements during isocapnic hypoxia and hypercapnia. J Appl Physiol1982; 53:81-6 22 Bartlett D, Remmers JE, Gautier H. Laryngeal regulation of respiratory airflow. Respir Physiol1973; 18:194-204 23 Bartlett D Jr. Effects of hypercapnia and hypoxia on laryngeal resistance to airflow. Respir Physiol1979; 37:293-302 24 Strohl KP, Fouke JM. Dilating forces on the upper airway of anesthetized dogs. J Appl Physiol1985; 58:452-58 25 Duane SF, Weir EK, Stewart RM, Niewaehner DE. Distal airway responses to changes in oxygen and carbon dioxide tensions. Respir Physiol1979; 38:303-11
in Diagnosis and Treatment of Allergic Diseases
The Division of Allergy and Immunology, Department of Medicine, University of California, San Francisco, will present this program September 21-22 at the Stanford Court Hotel, San Francisco. For information, contact Postgraduate Programs, Department of Medicine, 521 Parnassus Avenue, University of California, San Francisco 94143-0656 (415:476-5208).
Applied Duplex Ultrasound The Department of Radiology, University of Alabama, Birmingham School of Medicine, will present this course September 23-25 at the Wynfrey Hotel Galleria Plaza, Birmingham. For information, contact: Dawne Ryals, Ryals & Associates, PO Box 920113, Norcross, Georgia 30092-0113 (404:641-9773).
CHEST I 98 I 1 I JULY, 1989
101
The inftuence of progressive hypoxia and hypercapnia on respiratory mechanics was evaluated in 26 subjects (six normal subjects, seven asthmatic subjects, seven patients with IPD, and six patients with COPD). During separate rebreathing runs of progressive isocapnic hypoxia and normoxic hypercapnia, breath-to-breath changes in lb. and Cdyn were determined. In 6ve of the six normal subjects, seven of the seven asthmatic subjects, and six of the seven subjects with IPD, lb. decreased with both progressive hypoxia and hypercapnia without a change in Cdyn. In the
the effects of acute chemostimulation on W hile respiratory drive and the pattern of breathing in normal subjects and patients with pulmonary disease have been intensively studied, the influence of progressive hypoxia and hypercapnia on respiratory mechanics has not been systematically evaluated. The relatively few studies addressing this issue conducted in normal subjects have yielded conflicting results. In normal subjects exposed to acute hypoxia (Sa02 of 60 to 70 percent), Sterling1 demonstrated significant increases in both Raw and TGV. In contrast, MilicEmili and Petit2 evaluated pulmonary function in normal subjects in response to 10 percent oxygen and found no change in either RL or Cdyn. These data were subsequently confirmed by studies by Saunders et al3 and Goldstein et al. 4 Two studies in dogs suggest that acute hypoxia causes profound bronchoconstriction, as documented by decreased Raw.M The literature regarding the effect of acute hypercapnia on pulmonary mechanics is similarly confusing. Sterling7 demonstrated a decrease in airway specific conductance in normal subjects in response to acute hypercapnia. Nastro et al8 and Ahmed et al9 demonstrated an increase in specific conductance in normal subjects in response to hypercapnia. Furthermore, they documented an increase in FRC and TGV in response to breathing 6 percent carbon dioxide. Two
*From the Department of Medicine, National Jewish Center for Immunology and Respiratory Medicine, Denver, and the Department of Pulmonary Medicine, Children's Hospital, Philadelphia. Manuscript received January 22; revision accepted November 4.
98
patients with COPD, the effects of hypoxia and hypercapnia on lb. and Cdyn were variable. Compared to normal subjects, the changes in lb. during hypoxia and hypercapnia were not signi6cantly different in the asthmatic subjects and the patients with IPD. These data provide evidence that acute progressive hypoxia and hypercapnia are associated with signi6cant changes in Raw in both normal subjects and patients with chronic pulmonary disease. (Cheat 1989; 96:96-101)
other groups found no change in Raw in normal subjects exposed to hypercapnia. 10•11 To further complicate the issue, hypercapnia has been shown to cause bronchoconstriction in intact dogs, 5.6 whereas hypercapnic ventilation of isolated perfused canine lungs has resulted in bronchodilation. 7 These contradictions regarding the alterations in pulmonary mechanics in response to either acute hypoxia or acute hypercapnia cannot be readily explained. Moreover, the effects of acute hypoxia and hypercapnia on dynamic respiratory mechanics in patients with pulmonary disease remain to be identified. In an attempt to evaluate the pulmonary mechanics in these situations, we addressed a new issue: do patients with chronic pulmonary disease respond differently to acute hypoxia or acute hypercapnia (or both) than normal subjects? Our present study was designed to quantitatively assess the separate effects of acute progressive isocapnic hypoxia and acute progressive normoxic hypercapnia on pulmonary mechanics in normal subjects as well as patients with asthma, chronic obstructive pulmonary disease (COPD), and interstitial pulmonary disease (IPD). MATERIALS AND METHODS
Six healthy nonsmoking subjects with normal pulmonary function served as control subjects. The three groups of patients consisted of seven asthmatic subjects, six patients with COPD, and seven patients with IPD. Baseline measurements of pulmonary function, including VC, FEY, TGV, RV, and Raw, were measured with the subjects seated in a pressure-compensated body plethysmograph breathing through a heated No. 3 Fleisch pneumotachygraph. Volume was measured with a Krogh spirometer attached to the plethysmograph. The plethysmographic volume signal was pressure
Elfecls of Acute Hypoxia and Hypercapnia on Pulmonary Mechanics (Parsons, Glunsteln, Femendez)
Table 1-Ptdmonary Function• Group
n
Age, yr
FEV,L (percent predicted)
Normal Asthma IPD COPD
6 7 7 6
32±4 48±5 36±4 52±5
3.77±0.31 (101) 2.28±0.37 (64) 2.70±0.2 (73) 1.67±0.38 (48)
FVC,L (percent predicted) 4.72±0.50 3.28±0.50 3.34±0.26 3.07 ± 0.53
TGV,L (percent predicted)
FEV,IFVC%
(100) (69) (71) (62)
80±2 70±4 81±2 54±6
3.00±0.34 3.78±0.34 2.34±0.10 5.77±0.27
(92) (114) (71) (149)
Raw, em H 10/L/s
1.94±0.23 2.85±0.41 2.36±0.39 4.11 ±0.88
(119) (286) (146) (218)
*Values are means± SEM. Numbers in parentheses represent mean percent of predicted normal values. compensated and had a frequency response which was linear to 12 Hz. The pulmonary functional characteristics of the four groups of subjects are outlined in Table 1. All studies were performed in Denver (elevation, 1,585 m) when the patients were in a baseline stable clinical status. After obtaining consent for the procedure, each subject was seated comfortably with his eyes closed and allowed to listen to soothing music provided through headphones. The subjects wore nose clips and breathed via a mouthpiece attached to a pneumotachograph (Fleisch No. 3), which was connected to a low-resistance (1.3 em H 10/L/s) rebreathing circuit partitioned into separate inspiratory and expiratory limbs. After five minutes of stabilization on the circuit, progressive normoxic hypercapnia was produced by the method of Read,•• wherein subjects rebreathed from a bag initially containing a mixture of 6 percent carbon dioxide, 40 percent oxygen, and the balance nitrogen. The Sa01 was continuously monitored throughout the carbon dioxide rebreathing trial with an ear oximeter (HewlettPackard 47201A) and remained at greater than 92 percent. Each rebreathing trial lasted between four and six minutes. Progressive isocapnic hypoxia was produced by rebreathing from a bag initially containing room air in a volume equal to the patient's VC plus 1 L. The PETC01 was continuously monitored at the mouthpiece via a mass spectrometric gas analyzer (Perkin-Elmer 1100 Medical Gas Analyzer) and maintained within 2 mm Hg by means of a carbon dioxide absorber attached to the expiration limb of the rebreathing circuit. Volume was derived by integration of the airflow signal. Transpulmonary pressure was monitored by means of a differential pressure transducer (Statham PM 5 ETC), attached at one end to monitor mouth pressure and at the other end to monitor intraesophageal pressure via a catheter attached to an esophageal balloon placed in the lower third of the esophagus. The frequency response of the balloon-catheter system was linear up to 10 Hz. The measurement of Pr was related to the airflow at the mouth to obtain breath-to-breath determinations of RL. The RL was calculated by determining the changes in Pr and flow (~ V) at isovolumes (middle of inspiration and middle of expiration) so that RL (in liters per 13 • 14 The Cdyn was determined by continuously second)=~Pr/~V. recording the Pr and volume and recording the changes in Pr and volume (V) from the beginning of an inspiration to the end of inspiration (times of zero airflow) so that Cdyn (in liters per centimeter of water)=~ VI~Pr.••·•• The airflow, volume, Pr, PETC01 , and Sa01 signals were displayed on a multichannel pen recorder (Gould). In addition to the previously mentioned measurements, in order to determine any change in TGV during the course of rebreathing, three patients with asthma and two normal subjects underwent progressive isocapnic hypoxic challenges while seated in a body plethysmograph with the rebreathing circuit placed outside the plethysmograph and connected to the mouthpiece via a port in the plethysmograph. Serial measurements of TGV were made during the isocapnic hypoxic challenge. Relationships between breath-to-breath changes in RL and Cdyn vs changes in Sa01 and PETC01 were determined for each subject. The individual and grouped data were compared using Student's ttest for paired and unpaired analysis, where appropriate. Slopes of
relationships were compared using the Mann-Whitney (rank sums) analysis.
REsuLTS In all of the normal subjects, RL decreased in response to progressive isocapnic hypoxia, as exemplified by subject 5 in Figure 1. It can be seen that the relation is closely approximated by a linear function providing a regression slope of +0.1035 em H 20/l/s/ percent Sa02 (correlation coefficient= 0.81). Quantitatively similar results were obtained in the other normal subjects studied. The mean slope of the linear regression ofRL vs Sa02 was + 0.103 ± 0.053 em H 20/ Us/percent Sa02 for the normal subjects (Fig 2). Similarly, in all of the asthmatic subjects and six of the seven patients with IPD, RL decreased in response to acute progressive isocapnic hypoxia. The mean slope of the linear regression of RL vs Sa02 was Hypoxia
~
4
0N
r•0.81
l:
E
.....u....
•
2
a:
0~,~~----~----~--~----~--~
,.z
0.6
•
0.4
r=0.32
•
0
0
0.2
,
•
• • • ••••
I
I
70
80 Sa02
90
FIGURE lA (top). Relationship between RL (in centimeters of H 10 per liter per second) and hypoxia for representative normal subject 5. Slope of relationship is +0.1035 (r=0.81). B (bottom). Relationship between Cdyn (in centimeters of H.O per liter) and hypoxia for same subject (r=0.32). CHEST I 96 I 1 I JULY, 1989
97
+0.4
q,"' +0.2
en
0 >_.
a:
j Normal
Asthma
ILD
2. Data are grouped by disease. Bars represent mean slope for relationship of RL (in liters per second) vs Sa01 (percent) for each group. Error bars are SEM. FIGURE
+0.192±0.070 em H 20/Us/percent Sa02 for the asthmatic subjects and +0.0936±0.0399 em H20/U s/percent Sa02 for the subjects with IPD (Fig 2). The response to hypoxia of the patients with COPD was variable. In each of the normal and the asthmatic subjects and in three of the subjects with IPD, the decrease in RL was statistically significant (p<0.05), as can be seen in 'Thble 2; however, there was no significant difference in the responsiveness between the groups. The Cdyn was calculated for each subject. As can be seen from the data of a representative normal subject (subject 5) {Fig 1), Cdyn remained unchanged during progressive hypoxia despite an increase in respiratory frequency. This stability of Cdyn {in the face of an increased respiratory rate) was seen in all normal and asthmatic subjects and all subjects with IPD {'Thble 2). The RL also decreased in response to acute progressive norm oxic hypercapnia in five of the six normal subjects as exemplified by subject 5 in Figure 3. This relationship was also best approximated by a linear function. The regression slope of this subject was -0.2017 em H20/mVs/mm Hg PEC02 • The mean
Table 2-Slopea ofGraphl of Pulmonary Function* Group and Subject Normal Subject 1 Subject 2 Subject 3 Subject 4 Subject 5 Subject 6 Asthma Subject 7 Subject 8 Subject~
Subject 10 Subject 11 Subject 12 Subject 13 IPD Subject 14 Subject 15 Subject 16 Subject 17 Subject 18 Subject 19 Subject 20 COPD Subject 21 Subject 22 Subject 23 Subject 24 Subject25 Subject26
Slope of RL vs Sa01 , Us/percent
Slope of Cdyn VS sao., em H 10/Lipercent
Slope of RL vs PEC01 , Us/mmHg
Slope ofCdyn vs PEC01 , em H10/IJmm Hg
+0.0618 +Q.1690 +0.0501 +0.1497 +0.1035 +0.0654
(0.81)t (0.93)t (0.66)t (O.SO)t (0.81)t (0.78)t
-0.0088 -0.0127 +0.0013 +0.0016 -0.0061 +0.0009
(0.45) (0.16) (0.20) (0.21) (0.32) (0.11)
-0.1039 +0.0226 -0.0158 -0.2813 -0.2017 - 0.4700
+0.2516 +0.1069 + 0.0906 +0.0897 +0.1022 +0.5893 +0.1140
(0.62)t (0.74)t (0. 76)t (0.64)t (0.67)t (0.87)t (0.69)t
-0.0050 -0.0006 +0.0055 -0.0019 +0.0013 - 0.0034 + 0.0013
(0.49) (0.29) (0.27) (0.29) (0.33) (0.30) (0.33)
-0.5775 (0.93)t
+0.0018 (0.86)t
-0.1683 (0.81)t -0.1975 (0.84)t -0.5382 (0.60)t
-0.0025 (0.15) -0.0051 (0.44) -0.0018 (0.09)
+0.1000 +0.0093 -0.0390 +0.2368 +0.2286 +0.0834 +0.0362
(0.99)t (0.24) (0.18) (O.OO)t (O.BS)t (0.26) (0.47)
-0.0020 +0.0073 -0.0011 -0.0009 +0.0005 +0.0053 -0.0011
(0.37) (0.21) (0.11) (0.08) (0.09) (0.88)t (0.40)
-0.0643 (0.63)
+0.0038 (0.54)
-0.0689 -0.4609 +0.1041 -0.5915 -0.4lll
-0.0004 + 0.0038 - 0.0036 -0.0033 +0.0023
-0.6409 +0.1324 +0.0349 -0.0471 +0.0746 +0.0378
(0.62)t (0.67)t (0.36) (0.11) (0.53)t (0.53)
-0.0060 +0.0009 -0.0011 -0.0078 -0.0010 +0.0023
(0.09) (0.12) (0.37) (0.64) (0.04) (0.37)
(0.93)t (0.12) (0.61) (0.77)t (0.83)t (0.56)t
(0.60) (0.93)t (0.61) (0.92)t (0.87)t
-0.0037 -0.0022 -0.0035 +0.0078 -0.0055 -0.0053
(0.23) (0.15) (0.22) (0.54) (0.18) (0.51)
(0.09) (0.47) (0.49) (0.40) (0.36)
-0.0788 (0.53)
-0.0111 (0.42)
+0.0607 (0.51)
-0.0022 (0.17)
*Numbers within parentheses are r values. tp
98
Effecls of Acute Hypoxia and Hypercapnia on Pulmonary Mechanics (Pataons, Gtuns181n, Femandez)
Hypercapnia
s ~
plethysmograph, and serial measurements of TGV were determined as the subject's oxygen saturation was decreased from 92 percent to 75 to 78 percent using the previously described rebreathing circuit. There was no significant change in the TGV in response to progressive isocapnic hypoxia in any of the subjects studied (Fig 5).
4
2
•
E
u
~
a:
r=0.830
0~,~------~--------~--------~
0.4
,..z
a
0 0.2
. . ., ..
r=0.38
••• ••
0~~------~------~~------~
0
40
•
45
50
PnC02 FIGURE 3A (top). Relationship between RL (in centimeters of H1 0 per liter per second) and hypercapnia (in millimeters of mercury) for single normal subject 5. Slope of this relationship is -0.2017 (r=O.B3). B (bottom). Relationship between Cdyn (in centimeters of H10 per liter) and PEC02 (in millimeters of mercury) for same subject (r=0.36).
slope of the linear regression of RL vs PEC02 was -0.112 for the normal subjects studied. The RL also decreased linearly in response to hypercapnia in the subjects with asthma and IPD. The mean slope of these relationships was - 0.3703 for the asthmatic subjects and - 0.2488 for the subjects with IPD (Fig 4). The response was statistically significant (p<0.05; Mann-Whitney) in three of the six normal subjects, in three of the four asthmatic subjects, and in three of the five subjects with IPD ('Thble 2). The degree of responsiveness of the subjects with asthma and IPD was not different from that of the normal subjects. The patients with COPD experienced severe dyspnea in response to hypercapnia. Only two patients with COPD were able to tolerate the complete hypercapnic protocol, so no conclusions regarding that group can be drawn. The Cdyn did not change in response to acute progressive hypercapnia, as can be seen by the data from normal subject 5 (Fig 3). The data from each individual subject is shown in Thble 2. Previous investigators have suggested that TGV (or FRC) may be affected by hypoxia and could, therefore, account for the observed change in RL. 3•16.1 7 To evaluate whether TGV varied during our challenges, we studied three of the asthmatic subjects and two of the normal subjects. Each subject was seated in a volume body
DISCUSSION
We found that our normal nonsmoking subjects significantly decreased their RL in response to both acute progressive isocapnic hypoxia and acute progressive normoxic hypercapnia. The majority of the patients with asthma and IPD whom we studied also responded to those two stimuli with a decrease in RL. The changes in Raw in response to hypoxia and hypercapnia were not associated with changes in measured Cdyn despite an increase in respiratory frequency. This apparent lack of a frequency-dependent component to the serial measurements of Cdyn suggested that the changes in RL may not have been due primarily to changes in the small airways; however, an alternative explanation was that Cdyn was dependent on respiratory frequency but tha,t both hypoxia and hypercapnia caused sufficient bronchodilation of
-0.6
~
~
--
N
0
o....
-0.4
~
--
~
~_. a:
i
~
-0.2
~
'-
T
0 Normal
Asthma
ILD
FIGURE 4. Data are grouped by disease. Each bar represents mean slope of RL (in liters per second) vs PEC01 (in millimeters of mercury) for group. Error bars represent SEMs. CHEST I 96 I 1 I JULY. 1989
89
5 (J
-j ...~
~
~~
~ ~
4 l:l.
i
A
AADJt. A
• • fP• • •
3
0
0
0
cP
0 normal
80
85
90
95
0 1 Salu'atlon (")
100
FIGURE 5. Data on TGV measured at oxygen saturations of 80 to 100 percent in two normal and three asthmatic subjects.
the peripheral airways to preserve the measurements ofCdyn. The literature suggests that both large and small airways are affected by acute hypoxia and hypercapnia. Previous investigators have shown that the larynx and large airways account for a significant portion of the total RL measured in humans. 18•19 Furthermore, the human larynx has been shown to provide a variable controlled resistance which participates in the regulation of airflow. m This observation was elaborated on by England and associates, 21 who observed vocal cord movements during hypoxia and hypercapnia. Under the influence of both of these stimuli, the narrowing of the glottic aperture normally seen during expiration was minimal. 21 Similar observations have been made in animals. Bartlett et al22 showed that up to 50 percent of the respiratory resistance could be accounted for by the upper airway in cats. Both hypoxia and hypercapnia22 •23 significantly decreased laryngeal resistance in these animals, suggesting that the larynx plays a major role in determining respiratory patterns both at rest and in response to stimuli. There is evidence in dogs that the response may not be limited to the larynx. Strohl and Fouke24 showed that even after resection of the recurrent laryngeal nerve, there was an increase in negative pressure in the upper airway in response to hypercapnia, and they speculated that other regions of the upper airway participated in respiratory control. Other data suggest that the small airways may play a significant role in the changes in pulmonary mechanics in response to acute hypoxia and acute hyper100
capnia. Duane et al25 evaluated the response of isolated strips of distal pulmonary parenchyma to hypoxia and found a significant decrease in resistance. The response to hypocapnia was also evaluated and was found to be associated with an increase in resistance (suggesting that hypercapnia would produce a decrease in resistance). Thus, small airways and the larynx appear to both respond to hypoxia and hypercapnia with a decrease in resistance. In conclusion, normal subjects and patients with chronic pulmonary disease responded to acute progressive isocapnic hypoxia and acute progressive normoxie hypercapnia with a decrease in RL. There was a suggestion that the change in resistance was greater for asthmatic subjects than for subjects with IPD or normal subjects. These changes in the function of the airways in response to hypoxia and hypercapnia may be relevant to evaluate mechanisms of breathing and respiratory control and need to be considered in future studies of pulmonary mechanics. REFERENCES 1 Sterling GM. The mechanisms of bronchoconstriction due to hypocapnia in man. Clin Sci 1968; 34:277-85 2 Milic-Emili J, Petit JM. Effetes de Ia teneur en oxygene de I'air inspire sur les proprietes mechaniques des poumons chez l'homme normal. J Physiol (Paris) 1960; 52:175-77 3 Saunders NA, Betts MF, Pengelly LD, Rebuck AS. Changes in lung mechanics induced by acute isocapnic hypoxia. J Appl Physiol1977; 43:413-19 4 Goldstein RS, Zamel N, Rebuck AS. Absence of effects of hypoxia on small airway function in human. J Appl Physiol1979; 47:251-56 5 Nadel JA, Widdicombe JG. Effect of changes in blood gas tensions and carotid sinus pressure on tracheal wlume and total lung resistance to airflow. J Physiol1962; 163:13-33 6 Green M, Widdicombe JG. The effects of ventilation of dogs with different gas mixtures on airway calibre and lung mechanics. J Physiol1966; 186:363-81 7 Sterling GM. The mechanism of decreased specific airway conductance in man during hypercapnia caused by inhalation of 7% C01 • Clin Sci 1969; 37:539-48 8 Nastro JA, Weiss M, Soorani J, Lyons HA. The effects of carbon dioxide on airway resistance. Fed Proc 1960; 28:460 9 Ahmed S, Weiss M, Lyons HA. The effect ofC01 breathing on bronchomotor tone. Physiologist 1967; 10:107 10 Butler J, Caro CG, Alcala R, DuBois AF. Physiological factors affecting airway resistance in normal subjects and in patients with obstructive respiratory disease. J Clin Invest 1960; 39:58491 11 Jaeger MJ, Otis AB. Measurement of airway resistance with a wlume displacement body plethysmograph. J Appl Physiol 1964; 19:813-20 12 Read DJ. A clinical method for assessing the ventilatory response to carbon dioxide. Aust Ann Med 1966; 16:20-32 13 lngrani RH, Pedley TJ. Pressure flow relationships in the lungs. In: Macklem PT, Mead J, eds. Handbook of physiology: the respiratory system II. Bethesda, Md: American Physiological Society, 1986:277-93 14 Chemiack RM. Pulmonary function testing. Philadelphia: WB Saunders Co, 1977:27-29 15 Anthonisen NR. Tests of mechanical function. In: Macklem PT, Mead J, eds. Handbook of physiology: the respiratory system
Eflecls of Acute Hypoxia and Hypercapnia on Pulmonary Mechanics (Parsons, Grunsteln, Fernandez)
16
17 18 19
20
III. Bethesda, Md: American Physiological Society, 1986:75384 Garfinkel F, Fitzgerald R. The effects of hypercapnia, hypoxia and hyperoxia on FRC and the P 0.1 in humans. Physiologist 1975; 18:223 Kellogg RH, Mines AH. Acute hypoxia fails to affect FRC in man. Physiologist 1975; 18:275 Hyatt RE, Wilcox RE. Extrathoracic airway resistance in man. J Appl Physiol1961; 16:326-30 Ferris BG Jr, Mead J, Opie LH. Partitioning of respiratory flow resistance in man. J Appl Physiol1964; 19:653-58 England SJ, Bartlett JR, Daubenspeck JA. Influence of human vocal cord movements on airflow and resistance during eupnea.
Advances
J Appl Physiol1982; 52:773-79 21 England SJ, Bartlett D, Knuth SL. Comparison of human vocal cord movements during isocapnic hypoxia and hypercapnia. J Appl Physiol1982; 53:81-6 22 Bartlett D, Remmers JE, Gautier H. Laryngeal regulation of respiratory airflow. Respir Physiol1973; 18:194-204 23 Bartlett D Jr. Effects of hypercapnia and hypoxia on laryngeal resistance to airflow. Respir Physiol1979; 37:293-302 24 Strohl KP, Fouke JM. Dilating forces on the upper airway of anesthetized dogs. J Appl Physiol1985; 58:452-58 25 Duane SF, Weir EK, Stewart RM, Niewaehner DE. Distal airway responses to changes in oxygen and carbon dioxide tensions. Respir Physiol1979; 38:303-11
in Diagnosis and Treatment of Allergic Diseases
The Division of Allergy and Immunology, Department of Medicine, University of California, San Francisco, will present this program September 21-22 at the Stanford Court Hotel, San Francisco. For information, contact Postgraduate Programs, Department of Medicine, 521 Parnassus Avenue, University of California, San Francisco 94143-0656 (415:476-5208).
Applied Duplex Ultrasound The Department of Radiology, University of Alabama, Birmingham School of Medicine, will present this course September 23-25 at the Wynfrey Hotel Galleria Plaza, Birmingham. For information, contact: Dawne Ryals, Ryals & Associates, PO Box 920113, Norcross, Georgia 30092-0113 (404:641-9773).
CHEST I 98 I 1 I JULY, 1989
101