Lung Function, Breathing Pattern, and Gas Exchange in Interstitial Lung Disease Thorax: First Published As 10.1136/Thx.47.2.93 on 1 February 1992

Thorax 1992;47:93-97 93

Lung function, breathing pattern, and gas exchange in interstitial lung disease Thorax: first published as 10.1136/thx.47.2.93 on 1 February 1992. Downloaded from

S Javaheri, L Sicilian

Abstract The relation between lung function test values, Background The aim of this study was breathing pattern, and gas exchange has been to determine the relation between the studied in patients with chronic obstructive severity of abnormalities in ventilatory lung disease. Carbon dioxide retention in these function tests and tidal breathing pattern patients is usually associated with a forced and gas exchange indices in interstitial expiratory volume in one second (FEVy) ofless lung disease. than 1 0 litrelA, and the same is true in Methods Pulmonary function, ventila- asthma.7 tion, carbon dioxide production, oxygen The pattern of breathing may contribute to consumption, arterial blood gas tensions, carbon dioxide retention. Hypercapnic and pH were measured during resting patients with chronic obstructive lung disease steady state conditions in 60 patients with breathe faster and more shallowly than eucap- proved interstitial lung disease. Patients nic patients,48causing an increase in the ratio of were categorised by forced vital capacity dead space to tidal volume (VD/VT) and in dead (FVC) (percentage ofpredicted values) as space ventilation (VD) with a decrease in having a mild, moderate, or severe res- alveolar ventilation (VA); this results in hyper- trictive defect with means (SD) of 71% capnia. (4%), 57% (4%), and 41% (7%) of predic- Few studies have examined the pattern of ted values, respectively. breathing in diffuse interstitial lung disease."' Results FVC varied from 29% to 79% of Although it is commonly believed that patients predicted values and from 0 99 1 to 4-32 1. with interstitial lung disease breathe more The two measurements ofFVC correlated rapidly and shallowly than normal subjects, strongly with most static lung volumes previous studies have reported considerable and with transfer factor for carbon mon- variation in both breathing frequency and tidal http://thorax.bmj.com/ oxide. Mean respiratory rates (per min- volume. In the original study ofLourenco et al, ute) and tidal volumes (ml) were 17 (4) for example, the respiratory rate varied from 13 and 484 (131), 20 (4) and 460 (139), and 23 to 40 breaths/min, and tidal volume from 306 to (5) and 377 (109) in mild, moderate, and 612 ml.9 In a more recent study Renzi et al severe restrictive defects, respectively. showed that changes in respiratory rate and FVC correlated negatively with res- tidal volume were related to stiffness (elas- piratory rate and positively with tidal tance) of the lungs in 12 patients with inter- volume. Arterial carbon dioxide tension stitial lung disease and suggested that varia- on October 1, 2021 by guest. Protected copyright. ranged from 30 to 49 mm Hg; only two tions in breathing pattern were probably patients were hypercapnic. Mean arterial related to differences in the severity of the oxygen tensions were not significantly disease. 3 different among the three groups, and The main aim of this study was to determine Pulmonary Section, there were no significant correlations bet- the relation between the pattern of breathing Veterans Affairs ween forced expiratory volume in one Medical Center and and gas exchange during steady state breathing Department of second or FVC and arterial carbon diox- at rest and ventilatory lung function in a large Medicine, University ide tension or carbon dioxide production. group ofcarefully selected patients with proved of Cincinnati College Conclusion Low values of FVC were of Medicine, interstitial lung disease. Cincinnati, Ohio; associated with increased respiratory Pulmonary Section, rate and decreased tidal volume; this Methods New England Medical pattern of breathing mimics external We reviewed the results ofpulmonary function Center Hospitals, Tuft elastic University School of loading, suggesting that mech- tests in patients with a restrictive lung defect Medicine, Boston; and anoreceptors may contribute to the rapid (forced vital capacity < 80% predicted) and an Thoracic Services, and shallow pattern ofbreathing in inter- established diagnosis of interstitial lung dis- University Hospital, stitial lung disease. Hypercapnia seems to Boston University ease. The patients were referred for diagnosis Medical Center, be rare in interstitial lung disease even and treatment of interstitial lung disease. The Boston, Massachusetts when functional impairment is severe diagnosis was made by open lung biopsy or S Javaheri and tidal volume is small. The increased L from the presence of specific historical findings Sicilian respiratory rate is important in main- such as occupational exposure. Methods of Address for reprints: Dr S Javaheri, taining adequate ventilation. In the face diagnosis have been reported in detail.'4 15 Pulmonary Section (II IF), of a severe restrictive defect carbon diox- We excluded all patients with (a) a ratio of VA Medical Center, Cincinnati, Ohio 45220, ide production did not increase, which FEV, to forced vital capacity (FVC) of < 75%; USA. also contributed to the maintenance of (b) neuromuscular disorders; (c) kypho- Accepted 10 September 1991 eucapnia. scoliosis, (d) pulmonary emboli; (e) lung car- 94 Javaheri, Sicilian

cinoma; (f) acute respiratory alkalosis or moderate (group 2) and severe (group 3) res- metabolic acidosis or alkalosis'6'7; and (g) trictive defects were compared with those hav- obesity. Obesity was defined as a body mass ing a mild restrictive defect (group 1) (two index (body weight (kg)/(height(m))2) greater comparisons), and p < 0-025 was considered to Thorax: first published as 10.1136/thx.47.2.93 on 1 February 1992. Downloaded from than 27-8 and 27-3 in men and women, respec- be significant. Data are presented as means tively.'8 Finally, the respiratory exchange ratio with standard deviations in parentheses. had to be 0-83 (0.05). We also determined Pearson linear correlations for certain variables and then considered a MEASUREMENT OF LUNG FUNCTION p value below 0-05 to be significant. The SAS The technical details, methods of analysis, computer software ofthe University of Cincin- precautions, and calibrations used in our nati was used for the calculations. laboratory have been outlined previously.'9 Spirometric measurements were obtained with a Stead-Wells spirometer while the Results patient was seated. Functional residual Sixty patients met the criteria, with 27, 18, and capacity (FRC) was measured by helium 15 in groups 1, 2, and 3, respectively. The equilibration, and residual volume (RV) and major diagnoses were idiopathic pulmonary total lung capacity (TLC) were calculated. fibrosis (25 patients), sarcoidosis (11), asbes- The single breath carbon monoxide transfer tosis (eight), collagen vascular disease (six), factor (TLCO) was measured as described.20 berylliosis and silicosis (three), and chronic Ventilation and arterial blood gas tensions and eosinophilic pneumonia and eosinophilic pH were determined with the patient supine granuloma (four). The distribution of disease and comfortable. An arterial cannula was was similar among the three groups. placed in the radial or brachial artery. The Morphometric data are shown in table 1. patient then breathed through a mouthpiece Although mean values for height, weight, and connected to a two way valve (Hans-Rudolf, body surface area were lower in group 3 than in Kansas City, Missouri) into a Douglas bag. groups 1 and 2 the differences were not sig- The respiratory rate and tidal carbon dioxide nificant. tension were monitored with an infrared FVC ranged from 79% to 29% of predicted analyser or a mass spectrometer (1100 Medical values. There were significant differences in Gas Analyzer, Perkin Elmer, Pomona, Califor- most spirometric and lung volume nia). When end tidal carbon dioxide was stable, measurements between the groups (table 2). after an initial wash out, expired air was FVC, FEV,, and TLC were significantly lower collected for two minutes. The volume of in groups 2 and 3 when compared with group 1 expired air was determined in a Tissot whether expressed as percentages of predicted

spirometer and oxygen and carbon dioxide values or as absolute values (table 2). The only http://thorax.bmj.com/ concentrations in expired air by a mass spec- measurements that did not differ significantly trometer. An arterial blood sample was between the three groups were small differen- obtained for determination ofarterial blood gas ces in residual volume (RV) (and RV percen- tensions and pH. A calculator was used to tages of predicted values) and FEV, to FVC calculate plasma bicarbonate concentration ratio. (Severinghaus Calculator, Radiometer, Although arterial carbon dioxide tension Copenhagen). (Paco2) varied from 30 mm Hg (4 0 kPa) to From these measurements, minute ventila- 49 mm in individual patients,

Hg (6-5 kPa) on October 1, 2021 by guest. Protected copyright. tion (VE), respiratory rate, tidal volume, dead mean values of Paco2 (table 3) did not differ space (VD), alveolar ventilation (VA), oxygen between the three groups. Mean values for consumption (Vo2), carbon dioxide production carbon dioxide production and oxygen con- (Vco2), and respiratory exchange ratio were sumption (per surface area) also did not differ calculated. significantly in the three groups (table 3). Normal values for the spirometric data were Although minute and alveolar ventilation taken from Kory et al,2 and regression equa- (per surface area) did not differ between the tions for TLCO from Gaensler and Smith.20 three groups (table 3), breathing pattern varied considerably with a significant and progressive ANALYSIS rise in respiratory rate and a progressive fall in Patients were divided into three groups on the tidal volume. basis of the FVC representing mild, moderate, and severe restrictive defect: 79% to >65%, CORRELATIONS BETWEEN VARIABLES 65% to > 50%, and < 50% ofpredicted values, FVC correlated significantly with other lung respectively. The groups were compared by volumes, the highest correlation occurring with analysis of variance and multiple t tests adjus- inspiratory capacity and TLC, and the lowest ted by the Bonferroni method. Patients with with RV (table 4). Similar correlations were

Table I Demographic data on three groups ofpatients with interstitial lung diseases categorised according to percentage ofpredictedforced vital capacity. Values are means (SD) Age (years) Height (cm) Weight (kg) BSA (m2) BMI (kg/M2) Group 1 (n = 27) 46-2 (17-2) 170 9 (11-2) 70 0 (13-4) 1 81 (0 22) 23-9 (3-9) Group2(n = 18) 47-2(17-1) 167-1 (9-9) 65-1(10-5) 173(018) 232(25) Group 3 (n = 15) 39-7 (17 0) 165-4 (10-0) 59-7 (13-0) 1-65 (0 21) 21 7 (3-0)

BSA = body surface area; BMI = body mass index. Lungfunction, breathing pattern, and gas exchange in interstitial lung disease 95

Table 2 Mean (SD) lung volumes in three groups ofpatients with interstitial lung Discussion diseases categorised according to percentage ofpredictedforced vital capacity To obtain reliable measures of the pattern of Group I Group 2 Group 3 breathing, ventilation, and arterial blood gas

Lung volume (n = 27) (n =18) (n = 15) tensions in patients with pulmonary disease, Thorax: first published as 10.1136/thx.47.2.93 on 1 February 1992. Downloaded from steady state criteria must be met.4 Applying a Fv litres 2-88 (0-76) 2-17 (0-53)* 1-54 (0-32)* % predicted 71 (4) 57 (5)* 41 (7)* noseclip and mouthpiece22 and sampling FEV Jl/s 2-42 (0-68) 1-82 (042)* 1-35 (031)* arterial blood may change the pattern of ' % predicted 73 (8) 59 (8)* 41 (7)* breathing. In our study several precautions FEV,/FVC (%) 84 (6) 84 (5) 87 (9) were taken to ensure the presence of a steady TLC litres 4-31 (1-01) 3-38 (0-86)* 2-54 (0-39)* T % predicted 91 (11) 74 (13)* 57 (12)* state. Measurements ofventilation were started FRC (1) 2-18 (0-60) 1-86 (0-53) 1-48 (039)* only after the patient had already breathed for RV litres 1-30 (0-44) 1-17 (0-41) 0-96 (0 40) V\%predicted 108 (31) 103 (43) 88 (36) several minutes through the mouthpiece and IC (1) 2-12 (0-69) 1-52 (059)* 1-06 (0.34)* shown stable end tidal carbon dioxide concen- RV/TLC (%) 31 (9) 35 (10) 38 (12) trations. Normal values of respiratory exchange ratio were required. Arterial blood *p 0-025 v group 1. FVC = forced vital capacity; TLC = total lung capacity; FRC = functional residual was obtained through an indwelling cannula, capacity; RV = residual volume; IC = inspiratory capacity. and the strong linear correlation (r = 0-85) between Paco, and bicarbonate concentration Table 3 Mean (SD) gas exchange and ventilation values in three groups ofpatients with a slope of 0-58 indicates that changes in with interstitial lung diseases categorised according to percentage ofpredictedforced Paco2 were long term. vital capacity We studied 60 patients, 11 of whom had a severe restrictive defect with an FVC below Group I Group 2 Group 3 1-5 1. Among these 60 patients two had a high Pao2 (mm Hg) 76 (12) 71 (14) 74 (17) Paco2 of 44 mm Hg (5-9 kPa), one had Paco2 (mm Hg) 37 (4) 36 (3) 39 (5) 45 mm Hg (6-0 kPa) and one 49 mm Hg [H+] (nEq/l) 39-0 (2-4) 37-6 (1-8) 38-0 (2-0) LHC03-] (mEq/l) 23 (2) 24 (2) 25 (3) (6-5 kPa). Chronic hypercapnia therefore Vo /BSA (ml/minIm') 135 (17) 147 (28) 142 (21) seems to be rare in ambulatory patients with Vo2/BSA (ml/min/m') 109 (14) 120 (24) 115 (21) interstitial lung disease. Severe mismatch RER 0-81 (0-03) 0-82 (0-04) 0-81 (0-04) V/0 impairs carbon dioxide elimination,2324 but our TC (ml/min/mmHg) 17 (7) 14 (7) 12 (6) LCO(% predicted) 64 (20) 54 (26) 49 (34)* patients were generally capable of increasing RR (min) 17 (4) 20 (4)** 23 (5)*** their minute ventilation and, hence, alveolar VT (ml) 484 (131) 460 (139) 377 (109)*** to VD (ml/kg) 2-4 (1-1) 2-7 (1-2) 2-7 (0-8) ventilation, prevent hypercapnia. In addi- VD/VT (%) 34 (12) 37 (10) 42 (7) tion, Vco2 (table 2), another determinant of VE/BSA (1/min/m2) 4-4 (0-8) 5-2 (1-5) 5-1 (0 9) Paco2, remained relatively constant even in VA/BSA (1/min/m2) 2-8 (0-3) 3-2 (0 4) 3-0 (0-4) patients with marked impairment of FVC. *p < 0-025;** p ( 0-017;*** p 0-01 (all v group 1). Several factos probably help to maintain VE : http://thorax.bmj.com/ BSA = body surface area; Vo2 = oxygen consumption; VCo2 = carbon dioxide and VA in these patients. One factor, not looked production; TLCO = transfer factor for carbon monoxide; RER = respiratory exchange at in this study, is the low to mid position ofthe ratio; VE = minute ventilation; VA = alveolar ventilation. Vo2, Vco2, and TLCO are at STPD. Ventilatory values are at BTPS. chest wall in patients with a low FVC. This Conversion to SI unit: 1 mm Hg - 0-133 kPa. position places the inspiratory muscles at a mechanical advantage. Two factors identified in this study were the low values of Pao2 and found between FVC as a percentage of predic- high respiratory rate. We found a significant ted values and other predicted values (data not negative correlation between Pao2 and minute shown). ventilation (table 5), perhaps suggesting that on October 1, 2021 by guest. Protected copyright. Both FVC and FVC % predicted correlated hypoxaemia may augment resting ventilation with respiratory rate and tidal volume (figures 1 in patients with more severe interstitial lung and 2). disease. A major factor, however, contributing The correlations between both Paco2 and to maintenance of ventilation was the increase Pao2 and other variables are summarised in in respiratory rate. table 5. In addition, there were no significant Our results show that in interstitial lung correlations between Paco2 and VD, VD/VT disease breathing pattern changes, lower values ratio, body weight, FEVI, FEV1 % predicted, of FVC being associated with an increased FVC, and FVC % predicted. However, Pao2 respiratory rate and decreased tidal volume. correlated significantly with FVC (r = 0-28, Previous studies have reported considerable p = 0-035) and TLCO (r = 0-39, p = 0-003). variation in respiratory rate and tidal volume in There was no correlation between Vo2 and patients with interstitial lung disease, though Vco2 (or Vo2 and Vco2 in relation to body the number of patients was usually small and surface area) and FEV, FVC, FEVI/FVC the variation reported is probably related to the ratio, RV/TLC ratio, or their respective differences in severity of restriction between predicted values. studies. In our study respiratory rate varied

Table 4 Pearson correlations offorced vital capacity versus other lungfunction values in interstitial lung disease FEV, FEVJ/FVC TLC RV RV/TLC IC ERV FRC TLCO r 0-98 -0-2 0-90 0-26 -0-52 0-89 0-53 0-58 0-52 p 0-0001 0-2 0-0001 0-04 0-0001 0-0001 0-0001 0-0001 0-0001 ERV = expiratory reserve volume; TLCO = single breath transfer factor for carbon monoxide; other abbreviations as in table 3. 96 Javaheri, Sicilian

Table S Pearson correlations ofPacoo2 and Pao2 versus results ofventilatory andgas exchange tests in interstitial lung disease VEIBSA VAIBSA RR VT VCO2/BSA [HCO3; [H+ ] TLCO (I/minIm2) (l/minIm2) (min) (ml) (l/minIm2) (mEq/l) (nEq/l) (ml/min/mmHg) Thorax: first published as 10.1136/thx.47.2.93 on 1 February 1992. Downloaded from Paco2 r -0-2 -0-29 -0-2 01 0-2 0-85 01 0-1 p 0-1 0-02 0-2 0-4 0-2 0 0001 0-4 0-2 Pao2 r -0-41 -0-1 -0-04 -0-27 -0-2 -0 1 0-01 0-39 p 0-001 0-4 0-74 0 04 0-2 0-3 0 9 0-003 VE = minute ventilation; VA = alveolar ventilation; BSA = body surface area; RR = respiratory rate; VT = tidal volume; TLCO = single breath transfer factor for carbon monoxide; Vco2 = carbon dioxide.

from eight to 33 per minute and tidal volume have been reviewed by several authors."' from 214 to 829 ml in individual subjects and They include the mechanical effects of were often normal in patients with minimally increased lung elastance, perceived as increas- abnormal FVC values (group 1). When rapid ing load by mechanoreceptors, and stimulation and shallow breathing is clinically obvious the of intrapulmonary receptors-for example, patients characteristically have severe mechan- rapidly adapting and J receptors. ical impairment. External elastic loading of the respiratory The mechanisms mediating the rapid and system (breathing from a rigid chamber) has shallow breathing in interstitial lung disease been used to study the role of mechanoreceptors in determining the pattern of breathing. External elastic loading is thought to represent x Group 1 the mechanical analogue of diffuse interstitial 35 - + Group 2 * 3 lung disease (internal elastic loading). Several Group studies have shown that normal subjects 30 - breathe shallowly and rapidly in response to an external load and that respiratory rate increases - progressively and tidal volume falls as the E 25- magnitude of the external elastic load a) * ~~~~~~~~x increases.2"" Our results show similar * +~~~~~~~~~~~e x differences in the pattern of breathing in rela- > 20- +*~~~~~~x 0 + ~ tion to lower values of FVC, which are consis- x * + ~~+~~~x I tent with those of studies on external elastic x ._L x + U) 15- load. http://thorax.bmj.com/ x x x Intrinsic elastic loading does not seem to account fully for the pattern ofbreathing in our = 10 - r 043 patients. Despite a significant correlation bet- p = 0-0005 x ween FVC and respiratory rate and tidal volume, the results show a considerable scatter. 5 * ~ ~~~~~~~~~~x Several factors, such as stimulation of both 05 1.0 15 20 25 30 3-5 4-0 4-5 mechanoreceptors and other intrapulmonary Forced vital capacity (I) receptors,2528 may have contributed to the Figure 1 Linear regression betweenforced vital capacity (FVC,) and respiratory rate pattern of breathing we observed. Intrapul- on October 1, 2021 by guest. Protected copyright. (RR). monary receptors may be stimulated by inflam- matory mediators29'6 released from cells which x Group 1 accumulate in the lung parenchyma of patients + Group 2 with interstitial lung disease,34 and studies by * et 900 - Group 3 Guz et al in humans27 28 and by Phillipson al in dogs35 indicate that stimulation of these 800 - receptors contributes to tachypnoea. r = 0-45 x Irrespective of the mechanisms involved, in p = 0-0003 severe interstitial disease a 700 - the face of lung raised respiratory rate is important in main- + x x xx taining adequate ventilation because tidal ,E 600- x x x volume is small and gas exchange impaired. E This pattern of breathing is appropriate to the 500- x 0 +* x x xx mechanical properties ofthe respiratory system + x x m interstitial lung disease.'6 It enables patients 'a 400- to maintain ventilation with a smaller ins- x x muscle force and less mechanical work x piratory 300 - + + x per breath,'6 which should minimise the work of breathing. Mean values of Vco2 and Vo2 in 200 - relation to body surface area did not differ in our three groups of patients. Indices of 100 -r I metabolic rate did not correlate with ven- 0.**5 1.0 1 5 20 25 30 3 5 4 0 4 5 tilatory impairment in contrast to the findings Forced vital capacity (I) in patients with chronic obstructive lung dis- Figure 2 Linear regression betweenforced vital capacity and t;idalvolume (VT). ease.'7 In normal subjects the increase in Lungfunction, breathing pattern, andgas exchange in interstitial lung disease 97

11 Young RC, Carr C, Shelton TG, Mann M, Ferrin A, Laurey oxygen consumption caused by an added in- FJR, Harden KA. Sarcoidosis: relationship between spiratory external elastic load is less than that changes in lung structure and function. Am Rev Respir Dis 1966;95:224-38. seen with a resistive load,38 an effect attributed 12 Marshall R, Karlish AJ. Lung function in sarcoidosis.

to preferential recruitment of faster and less Thorax 1971;26:402-5. Thorax: first published as 10.1136/thx.47.2.93 on 1 February 1992. Downloaded from 13 Renzi G, Milic-Emili J, Grassino AE. The pattern of efficient muscle fibres during breathing with a breathing in diffuse lung fibrosis. Bull Eur Physiopathol resistive load. Respir 1982;18:461-72. 14 Gaensler EA, Carrington CB, Coutu RE, Fitzgerald MX. From our results in 60 patients with inter- Radiographic, physiologic, pathologic correlations in stitial lung disease we draw several con- interstitial pneumonias. Progress in Respiration Research 1974;8:223-41. clusions. With respect to gas exchange, FVC 15 Gaensler EA, Carrington CB, Coutu RE, Tomasian A, correlates with impairment in TLCO more Hoffman L, Smith AA. Pathological, physiological and radiological correlations in the pneumoconioses. Ann NY strongly than with hypoxaemia. Changes in Acad Sci 1972;200:574-607. Paco2 do not relate to severity of impairment of 16 Goldberg M, Green SB, Moss ML, Marbach CB, Garfinkel D. Computer-based instruction and diagnosis ofacid-base any spirometric values-for example, FEV1 or disorders. JAMA 1973;223:269-75. FVC in absolute terms or as percentage of 17 Javaheri S, Kazemi H. Metabolic alkalosis and hypoventila- tion in humans. Am Rev Respir Dis 1987;136:1011-6. predicted values. Low values of FVC are 18 Itallie TBV. Health implications of overweight and obesity associated with higher respiratory rates and in the United States. Ann Intern Med 1985;103:983-8. 19 Gaensler EA, Macklem P, Cherniak R, Permutt S, Ferris B. lower tidal volumes; in the face of severe Pulmonary function in epidemiology standardization impairment in gas exchange, changes in res- project. Am Rev Respir Dis 1978;188:7-112. 20 Gaensler EA, Smith AA. Attachment for automated single piratory rate are important for maintaining breath diffusing capacity measurements. Chest 1973; adequate ventilation and eucapnia. This pat- 63:136-45. 21 Kori RC, Callahan R, Boren HG, Syner JC. The Veterans tern of breathing minimises the work of breath- Administration-Army Cooperative Study of Pulmonary ing, and carbon dioxide production and oxygen Function. I. Clinical spirometry in normal man. Am JMed 1961;30:243-58. consumption remain relatively constant even 22 Gilbert R, Auchincloss JH Jr, Brodsky J, Bowden W. when mechanical impairment is severe. The Changes in tidal volume, frequency and ventilation induced by their measurement. J Appl Physiol lack of a rise in carbon dioxide production in 1972;33:252-4. the face of increasing functional impairment 23 West JB. Causes of carbon dioxide retention in lung disease. N Engl J Med 1971;284:1232-6. (increasing mechanical and gas exchange 24 West JB. Assessing pulmonary gas exchange. N Engl J Med defects) is important in preventing the 1987;316:1336-8. 25 Kornbluth RS, Turino GM. Respiratory control in diffuse development of hypercapnia in interstitial lung interstitial lung disease and diseases of the pulmonary disease. vasculature. Clin Chest Med 1980;1:91-102. 26 Rebuck AS, Stulsky AS. Control of breathing in diseases of the respiratory tract and lungs. In: Fishman AP, ed. We thank Dr E A Gaensler for allowing us to Handbook of physiology. Vol 2. The respiratory system. Bethesda, Maryland: American Physiological Society, report the data collected in his laboratory. The 1986;771-91. authors also thank Mrs Saundra K Eversole for 27 Guz A, Noble MIM, Eisele JH, Trenchard D. Experimental results of vagal block in cardiopulmonary disease. In: her excellent secretarial help, Mr Robert J Porter R, ed. Breathing: Hering-Breuer centenary sym- Mohrman for his careful review of the manu- posium. London: Churchill, 1970;17-40. 28 Guz A, Noble MIM, Eisele JH, Trenchard D. Experimental http://thorax.bmj.com/ script, and Dr P Gartside for his help in the results of vagal bock in cardiopulmonary disease. In: statistical analysis. Porter R, ed. Breathing: Hering-Breuer centenary sym- posium. London: Churchill, 1970;315-36. 29 Freedman S, Campbell EJM. The ability of normal subjects to tolerate added inspiratory loads. Respir Physiol 1970;10:213-35. 30 Bland S, Lazerou L, Dyck G, Cherniack RM. Influence of 1 Burrows B, Sakesena FB, Diener CF. Carbon dioxide the "chest wall" on respiratory rate and depth. Respir tension and ventilatory mechanics in chronic obstructive Physiol 1967;3:47-54. pulmonary disease. Ann Intern Med 1966;65:685-700. 31 Agostoni E, D'Angelo E, Piolini M. Breathing pattern in 2 Howell JBL. Breathlessness in pulmonary disease. In: men during inspiratory elastic loads. Respir Physiol

Howell JBL, Campbell EJM, eds. Breathlessness. Oxford: 1978;34:279-393. on October 1, 2021 by guest. Protected copyright. Blackwell Scientific Publications, 1966;165-77. 32 Pope H, Holloway R, Campbell EJM. The effects of elastic 3 Javaheri S. Premorbid hypercapnic ventilatory response and and resistive loading of inspiration on the breathing of Paco, in emphysema. Pfluigers Arch 1987;408:S29. conscious man. Respir Physiol 1968;4:363-72. 4 Javaheri S, Blum J, Kazemi H. Pattern of breathing and 33 Remmers JE, Marttila I. Action of intercostal muscle carbon dioxide retention in chronic obstructive lung afferents on the respiratory rhythm of anesthetized cats. disease. Am J Med 1982;71:228-34. Respir Physiol 1975;24:31-41. 5 McFadden ER Jr, Lyons HA. Arterial-blood gas tension in 34 Turner-Warwick M, Haslan PL. The value of serial bron- asthma. N Engl J Med 1968;278:1027-32. choalveolar lavages in assessing the clinical progess of 6 Rebuck AS, Read J. Assessment and management of severe patients with cryptogenic fibrosing alveolitis. Am Rev asthma. Am JMed 1971;51:788-98. Respir Dis 1987;135:26-34. 7 Kelsen SG, Kelsen DP, Fleegler BF, Jones RC, Rodman T. 35 Phillipson EA, Murphy E, Kozar LF, Schultze RK. Role of Emergency room assessment and treatment of patients vagal stimuli in exercise ventilation in dogs with with acute asthma. Am J Med 1978;64:622-8. experimental pneumonitis. J Appl Physiol 1975;39:76-85. 8 Sorli J, Grassino A, Lorange G, Milic-Emili J. Control of 36 Mead J. Control of respiratory frequency. J Appl Physiol breathing in patients with chronic obstructive lung dis- 1960;15:325-36. ease. Clinical Science and Molecular Medicine 37 Donahoe M, Rogers RM, Wilson DO, Pennock BE. Oxygen 1978;54:295-304. consumption of the respiratory muscles in normal and in 9 Lourenco RV, Turino GM, Davidson LAG, Fishman AP. malnourished patients with chronic obstructive pulmon- The regulation of ventilation in diffuse pulmonary ary disease. Am Rev Respir Dis 1989;140:385-91. fibrosis. Am JMed 1965;38:199-216. 38 Cala SJ, Wilcox P, Edyvean J, Rynn M, Engel LA. Oxygen 10 Snider GL, Doctor RL. The mechanics of ventilation in cost of inspiratory loading: resistive v elastic. J Appl sarcoidosis. Am Rev Respir Dis 1964;89:897-908. Physiol 1991;70:1983-90.