Respiration Physiology, 80 (1990) 193-202 Elsevier
193
RESP 01649
Influence of lung volume and rib cage configuration on transdiaphragrnatic pressure during phrenic nerve stimulation in man Anne Mier, Conor Brophy, John Moxham and Malcolm Green Brompton Hospital, London, U.K. (Accepted 16 February 1990) Abstract. Transdiaphragmatic pressure was recorded during bilateral supramaximal percutaneous phrenic
nerve stimulation at 1 Hz (twitch Pdi) to investigate the effect of lung volume and rib cage configuration on diaphragm contractility in man. Stimulations were performed in 5 normal supine subjects at resting end expiration (FRC) and at lung volumes above and below FRC, during relaxation against a closed airway and during isovolume manoeuvres.Twitch Pdi at FRC was 24.4 cm H20. At lung volumes above FRC, twitch Pdi decreased by 7.04 + 3.2 cm HaO per litre of volume change. At lung volumes below FRC, twitch Pdi increased by 12.4 _+8.6 cm HzO per litre of volume change. When the diaphragm was lengthened during an isovolume manoeuvre at FRC, twitch Pdi increased. A similar relationship between lung volume and twitch Pdi was obtained during stimulations performed with abdominal binding. These results demonstrate that the pressure developed by the diaphragm during phrenic nerve stimulation is significantlyaffected both by increases and decreases in lung volume and by the rib cage configuration at which stimulation is performed.
Animal, man; Diaphragm, contractility, dependence on lung volume; Lung volume, influence on diaphragmatic contractility; Phrenic nerve, stimulation, transdiaphragmatic pressure; Pressure, esophageal, gastric, transdiaphragmatic; Rib cage configuration, influence on diaphragm contractility; Technique in respiratory physiology,phrenic nerve stimulation (electric)
Previous studies both in vitro and in vivo have d e m o n s t r a t e d that the force developed by the diaphragm d e p e n d s o n muscle length. The tension generated by the diaphragm has been noted to decrease as diaphragm length shortens a n d the length tension properties of the m a m m a l i a n diaphragm in vitro have been shown to be similar to those of intact limb muscles a n d to be similar between species (McCully a n d Faulkner, 1983). Other studies in cats a n d m a n have d e m o n s t r a t e d that twitch tension during submaximal unilateral phrenic nerve stimulation decreases as lung volume increases (Pengelly et al., 1971). The f o r c e - l e n g t h relationship of the diaphragm was further d e m o n s t r a t e d during
Correspondence to: A. Mier, Department of Medicine, Charing Cross Hospital, Fulham Palace Road, London W6, U.K. 0034-5687/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
194
A. MIER et al.
supramaximal phrenic nerve stimulation in anaesthetised dogs (Kim and Druz, 1976; Road et aL, 1986) and during maximal inspiratory efforts in man (Braun et al., 1982). Several studies have recently focused on the assessment of diaphragm contractility by recording transdiaphragrnatic pressures during phrenic nerve stimulation in man (twitch Pdi) (McKenzie and Gandevia, 1983; Bellemare et al., 1986). Since diaphragm muscle length changes with lung volume and rib cage configuration, it seems important to monitor the volume and configuration at which such phrenic nerve stimulation is performed. Hubmayr and coworkers have recently studied the effect of diaphragmatic shortening on twitch Pdi (Hubmayr et aL, 1989); however, they did not study the effect of diaphragmatic lengthening. We therefore aimed at determining how twitch Pdi was affected both by diaphragmatic shortening and lengthening, by increasing and decreasing lung volume above and below resting end expiration (FRC). We also studied the effects on twitch Pdi of an 'isovolume manoeuvre' at FRC which lengthened the diaphragm. Finally we performed reproducibility studies and then phrenic nerve stimulation in the presence of abdominal binding to determine the effect of lung volume change on twitch Pdi under more isometric conditions. In this way we aimed to determine whether the relationship between twitch Pdi and lung volume would be similar to the length-tension characteristics of the mammalian diaphragm and of the human diaphragm previously described during voluntary inspiratory efforts.
Methods
Studies were performed on 5 healthy subjects, 4 men and one woman, who gave no history of respiratory disease. All subjects were experienced in respiratory manoeuvres, and their phrenic nerves had been stimulated on many occasions so that they were able to relax well during the studies. Their average age was 36 yrs (range 29-43), their height 171 (157-179) cm, their weight 66 (48-83) kg and predicted vital capacity 4.36 (3.0-5.0) L. All subjects gave informed consent to the studies. Lung volume was recorded using an Ohio 840 dry spirometer. Changes in rib cage and abdominal anteroposterior dimensions were recorded with two pairs of previously calibrated, linearised magnetometers (N. H. Peterson, Boston, MA) placed in the midline in the 5th intercostal space and 2 cm above the umbilicus respectively (Mead et aL, 1967). Electromyograms (EMG) of the right and left hemidiaphragms were recorded with surface electrodes in the 6th and 7th intercostal spaces 2-3 cm from the costal margin on either side. Electromyogram signals were processed using an amplifier (Medelec PA 63) and were filtered below 16 and above 1600 Hz. Oesophageal and gastric pressures were recorded with balloon catheter systems (PK Morgan), positioned in the mid-oesophagus and stomach (Agostoni and Rahn, 1960) and filled with 0.5 and 1.5 ml of air, respectively. The balloon catheters were connected to differential pressure transducers (Validyne, MP-45-1 + 150 cm H20 ) and transdiaphragmatic pressure (Pdi) was obtained electronically by subtraction of oesophageal pressure from gastric pressure. Pdi at resting end expiration (FRC) was used as zero reference point.
PHRENIC NERVE STIMULATION
195
The right and left phrenic nerves were stimulated simultaneously at the posterior border of the sternomastoid muscles at the level of the cricoid cartilage (Newsom-Davis, 1967). Two pairs of bipolar surface stimulating electrodes with felt tips 0.5 cm in diameter (Medelec 53054) were used, attached to a dual Digitimer 3072 isolated stimulator producing rectangular wave impulses 0.1 msec in duration at a frequency of 1 Hz. The voltage of stimulation was increased on each side in turn until there was no further increase in the size of either the diaphragm muscle action potential or the twitch Pdi. The voltage was then increased by a further 10~o so that supramaximal stimulation was achieved (between 80 and 160 volts for each nerve). Optimal position of the stimulating electrodes in the neck was ensured by closely monitoring the size of the diaphragm muscle action potentials recorded on the E M G during stimulation. If the size of either E M G decreased, it was assumed that the stimulation electrode in the neck had moved and the run was abandoned. A series of 10 maximal sniffs was initially performed from FRC, without a noseclip and the best of 10 sniffs was recorded to assess diaphragm contractility and to confirm that all subjects had normal diaphragm strength (Miller et al., 1985). A full vital capacity manoeuvre from total lung capacity (TLC) to residual volume (RV) was recorded in the supine posture three times. Both phrenic nerves were then stimulated simultaneously in the supine posture while the subjects, wearing a noseclip, breathed through the spirometer. Stimulations were first performed at FRC to obtain at least 10 twitches and then whilst the subject relaxed with a closed glottis at different lung volumes up to 4.0 L (mean 2.46 + 0.95 L) above FRC and decreasing volumes down to 0.9 L (mean 0.49 + 0.27 L) below FRC. Between 3 and 5 twitches were obtained at each lung volume. At least 5 runs were obtained in each subject and a series of 10 twitches was recorded at FRC between each run to ensure that twitch Pdi at FRC remained constant and that the effect of lung volume changes was a true one. Subject 1 was studied on 3 separate days and subjects 2 and 3 on two separate days. In 2 subjects, (1 and 2) isovolume manoeuvres were also performed. The subject initially relaxed at FRC with closed glottis. Phrenic nerve stimulation was performed at FRC and then immediately while the subject performed an isovolume manoeuvre (abdomen sucked in and rib cage expanded, thereby lengthening the diaphragm) at FRC. In 2 subjects (1 and 3) studies were also performed with an abdominal binder so that stimulation could be performed at different lung volumes with the diaphragm under conditions that were as isometric as possible. The abdomen was tightly bound with firm inextensible bandages from the lower rib cage to the level of the symphysis pubis. Volume studies with an abdominal binder were performed on 2 occasions in subject 1 and on one occasion in subject 3. All signals were recorded on an eight channel recorder (Minograph 800, Siemens) and were stored on a magnetic FM tape recorder with a frequency response of up to 5000 Hz (Racal Store 7) for later play back and analysis. All results are expressed as mean + SD. Results of twitch Pdi are expressed as mean + SD at each lung volume. Stepwise polynomial regression analysis was performed using the Cricket Graph Data Analysis Program (Apple plus) to investigate the relationship between twitch Pdi and lung volume.
196
A. MIER etal.
Results Vital capacity (4.12 + 1.03 L) and sniffPdi (138 + 27 cm H 2 0 ; normal > 98 cm H 2 0 ) (Miller et aL, 1985) were normal in all 5 subjects. Phrenic nerve stimulation was well tolerated by all subjects and no problems were experienced with relaxing at different lung volumes with a closed glottis. Typical records of EMG, Poes, Pg and Pdi during phrenic nerve stimulation at FRC in one subject are shown in fig. 1. Phrenic nerve conduction time was normal (6.7 + 0.5 msec; Newsom-Davis, 1967). Bilateral twitch Pdi recorded at FRC ranged from 20.5 to 32.3 cm H 2 0 (24.4 + 4.7 cm H 2 0 ; normal > 10.1) (Mier et al., 1989). Twitch Pdi was reproducible when recorded at FRC between runs in each subject. Coefficient of variation of twitch Pdi at FRC was 6.74 + 1.54% within days in the 5 subjects. By contrast twitch Pdi altered significantly with changes in lung volume. Records obtained in one representative subject are shown in fig. 2. As lung volume increased above FRC, twitch Pdi decreased. As lung volume decreased below FRC, twitch Pdi increased. Almost no background diaphragm E M G activity was seen during stimulation at FRC, at increased and decreased lung volumes. This and careful observation confirmed that subjects were able to relax against a closed glottis and that the 'diaphragm was indeed relaxed during stimulation at different lung volumes. Results obtained in all 5 subjects are shown in fig. 3. The mean decrease in twitch Pdi at the highest lung volumes studied above FRC was 15.1 + 5.0 cm H20. This was equivalent to a reduction in twitch Pdi of 61.7 + 19.7% compared to values at FRC. Overall this represented a decrease in twitch Pdi of 7.04 + 3.2 cm H 2 0 per litre of volume change above FRC. The mean increase in twitch Pdi at lowest lung volumes below FRC was 5.2 + 3.9 cm H20. This was equivalent to an increase in twitch Pdi of 22.6 + 15.4% compared to values at FRC. Overall twitch Pdi increased by 12.4 + 8.6 cm H 2 0 per litre of volume change below FRC. When an isovolume manoeuvre was performed at FRC, the rib cage anteroposterior dimension recorded on the magnetometers increased while the abdominal dimension decreased (Mead and Loring, 1982) and twitch Pdi increased (fig. 3). The increase in Edl Edi
(L) (R)
~
/~.
~
'~-,.
"1 4001~v
Poes
~
[
lOcmH20
Pg Pdl i--.-4
0.1sec Fig. 1. Electromyograms (Edi) of left (L) and right (R) hemidiaphragms, oe~ophageal (Poes), gastric (Pg) and transdiaphragmatic (Pdi) pressures recorded during bilateral phrenic nerve stimulation at 1 Hz at FRC (Note: electrocardiogram also seen on top trace).
197
PHRENIC NERVE STIMULATION
/
Volume
(L)
/
RC AP -~/~._ ABDOAP EdI(L)
Edi(R)
I IL ~ -
-
-
-
-
-
-
-
-
.,,1. i. t .... _~JF~I~ -
..L_LLJ_LJ._L-~. L [ 1 J :I ~ - ,
1] I
I
l 1_ I I ] I [ I ±ll I
I-I
/ IJ/~
r
Io.scm
L ~_~~'-~-
I 1 i
I-r
I 1
J. l.. [_1 I t I
I
l~l
1 400Hv
I ....
~ ~ ~ I ~/~A , , . .II
rlocmH20
Pdi 1 sec Fig. 2. Lung volume, rib cage (RC AP) and abdominal (ABDO AP) anteroposterior dimensions, electromyograms (Edi) oflett (L) and right (R) hemidiaphragms and transdiaphragmatic pressures (Pdi) recorded during bilateral phrenic nerve stimulation at 1 Hz at increasing lung volumes, in one representative subject. (Note: (L) and (R) Edi are contaminated by activity in chest wall muscles during each inspiration).
Sublect1
Subject 2 30.
,.r
10'
lo
o
• 2
• I
• 0
~bl*ct3
Y
0
n u
..j"
t.. a
SubJe~S
10
o
V o I u m •
(I)
Fig. 3. Relationship between lung volume and twitch transdiaphragmatic pressure recorded during bilateral phrenic nerve stimulation in 5 subjects. (Arrow = stimulation during isovolume manoeuvre at FRC. Zero lung volume represents residual volume (RV); lung volume increases from right to left).
198
A. MIER etal.
,/d
Subject1 -r-
Subject 2
30
10
o 4
3
2
~
0
lO .
o
O'
Volume
4
3
2
1
0
111
Fig. 4. Reproducibility of twitch transdiaphragmatic pressure during bilateral phrenic nerve stimulation at increasing lung volumes above RV in 3 subjects on different days. (Zero lung volume represents RV. /x = day 1 as in Fig. 1; • = day2; © = day 3.)
twitch Pdi during the isovolume manoeuvre was 6~o of the relaxed FRC value in subject 1 and 23 ~/o of the relaxed FRC value in subject 2. The relationship between lung volume and twitch Pdi seen in the 3 subjects who were studied on 2 or 3 separate days is shown in fig. 4. A similar relationship was seen in 2 subjects on different days although the actual height of twitch Pdi varied between days in one of these subjects (left hand panel, fig. 4). A different pattern of pressure-volume relationship was seen in the third subject on 2 different days (right hand panel, fig. 4). Results of stimulations performed with an abdominal binder, are shown in fig. 5. The size of the diaphragm evoked muscle action potentials during phrenic nerve stimulation remained constant during stimulation at FRC confirming that the position of the stimulating electrodes in the neck remained optimal. The voltage of stimulation did not need to be altered in any subject once supramaximal stimulation had been achieved. The size of the evoked diaphragm muscle action potentials altered slightly (increase or decrease) at increased lung volumes in some subjects, suggesting an altered contact between the recording surface electrodes and the diaphragm muscle.
Subject3 30"20,
Subjecty
20-
"r
E(,I '112 n r-
1o-
10-
0
' 4
" 2
' 0
Volume
0
, 4
, 2
, 0
(I)
Fig. 5. Twitch transdiaphragmatic pressure during bilateral phrenic nerve stimulation in 2 subjects with abdominal binders. (Zero lung volume represents RV. Subject 1 studied on 2 separate occasions.)
PHRENIC NERVE STIMULATION
199
Discussion These studies demonstrate that transdiaphragrnatic pressure recorded during bilateral supramaximal phrenic nerve stimulation depends significantly on the lung volume at which stimulation is performed. Since twitch Pdi has been advocated as a reliable, objective indication of diaphragm contractility (Bellemare et al., 1986) it is important to control carefully for both lung volume and rib cage configuration, when attempting to compare results of twitch Pdi within a subject between different studies. We have shown that twitch Pdi not only decreases at increased lung volumes as shown by Hubmayr et al. (1989) but that twitch Pdi increases at decreased lung volumes. In addition to changing with lung volume, twitch Pdi also varied with rib cage configuration. When the diaphragm was lengthened by performing an isovolume manoeuvre at FRC, twitch Pdi increased, even though lung volume did not alter. This illustrates the importance of monitoring rib cage configuration when recording twitch Pdi to ensure that stimulations are being performed at a constant diaphragm length. We have previously shown that the height of twitch Pdi at FRC is reproducible in normal subjects (Mier et aL, 1989). In the present studies we similarly found that twitch Pdi performed at FRC returned to the control value between each volume run. This supported the fact that the reduction in twitch height found at increased lung volumes was a true effect of diaphragm shortening. We took care to ensure that stimulus maximality was maintained throughout the range of lung volumes studied. Stimulations were not performed near total lung capacity or residual volume since tensing of the neck muscles might have reduced contact between the stimulating electrodes and the phrenic nerves. Accordingly, we studied the effect of lung volume changes that were smaller and not at the extremes of vital capacity. Similarly studies were performed in the supine posture, in which subjects found it easiest to relax the neck muscles and in which it was easiest to locate and maintain reliable stimulation of the phrenic nerves. Other workers have used rigid needle electrodes to provide phrenic nerve stimulation (Aubier et al., 1985) but the position of needle electrodes in the neck may similarly be altered by movement of the soft tissue of the neck during deep inspiration and expiration. We recorded electromyograms of each hemidiaphragm with surface electrodes in an attempt to ensure that constant stimulation of the phrenic nerves was being maintained. Although a small amount of variability in the amplitude of diaphragm E M G was noted on inspiration, it seems likely that this was due to movement of the diaphragm causing an alteration in the relationship between the surface electrodes and the diaphragm (Grassino et al., 1976; Bellemare et al., 1986; Gandevia and McKenzie, 1986) rather than to a reduction in stimulation of the phrenic nerves due to movement of the stimulating electrodes. Alternatively, an oesophageal electrode could have been used to record diaphragm E M G activity but similar problems with alterations in amplitude of phrenically evoked compound muscle action potentials may occur with such an electrode because of shifts in diaphragm position. Indeed, it has recently been shown that surface electrodes can be used to record diaphragm activity without distortion caused by movement of the diaphragm at lung volumes from FRC to about 50~o of vital capacity (Lansing and Savelle, 1989).
200
A. MIER et al.
A recent paper reported the effect of lung volume changes on twitch Pdi, using fine wire electrodes implanted near the phrenic nerves (Hubmayr et aL, 1989). Bilateral twitch Pdi in this study (19 to 36 cm H20 ) was very similar to that found in our subjects and the relationship of change in twitch Pdi with lung volume was almost identical (8 cm H20/L). Similarly the coefficient of variation within a day (2 to 5 ~o) was similar to that in our subjects and in previous studies that we have also performed on normal subjects (5 ~o) (Mier et aL, 1989). However, we chose to use surface stimulating electrodes which are less invasive, easier and quicker to use; they can also be applied more routinely in the investigation of normal subjects and patients. The finding that our results were similar to those of Hubmayr and coworkers, helps to confirm that percutaneous phrenic nerve stimulation is indeed a reliable technique, in addition to being easier and less invasive to use. Since we used a non-invasive technique, we were able to perform studies on several different days in our subjects. Although we obtained similar results on different days in one subject (middle panel, fig. 4), the results in two further subjects were much less repeatable. We felt that this might have been due to the fact that stimulations were not performed under isometric conditions. We therefore repeated the studies in these 2 subjects, with abdominal binding, but again found that results were not very reproducible. We were unable to explain why the shape of the pressure-volume relationship in the first study in subject 3 differed from that in the other subjects, and why it differed so significantly between days. However, previous reports have shown that twitch Pdi can vary from day to day in normal subjects (Mier et al., 1989). Since Hubmayr et al. (1989) did not perform studies on more than 1 day, it is not possible to know whether the repeatability of their results was similar to ours. The pressure-volume curves that we obtained were quantitatively similar to the length-tension relationship of the mammalian diaphragm, demonstrated in previous studies. The force length curve of the isolated skeletal muscle preparation is quite uniform in shape (Close, 1972). When measured previously during static inspiratory efforts in man, the shape of the force length curve was found by Braun and coworkers, (Braun et al., 1982) to be different from that obtained by direct measurement in dog (Kim and Druz, 1976); in man a plateau was found at higher lung volumes. Braun and colleagues concluded that the difference in shape between the two studies was because diaphragmatic length in their subjects had been estimated from lung volume. We were similarly anxious that this indirect assessment of diaphragmatic length was less good than the determination of diaphragm length as measured on a chest radiograph. However, the individual lengths of the costal and crural parts of the diaphragm have been assessed using implanted piezoelectric transducers in dogs (Newman etal., 1984). During passive lung inflation, a nearly linear inverse relationship between lung volume and diaphragm length was found. Thus although changes in rib cage configuration can significantly affect diaphragm length, it seems that an indication of diaphragm muscle length can be obtained from lung volume. In addition to the length tension properties of the diaphragm muscle, the curvature of the diaphragm may also play a role in determining the height of twitch Pdi. Laplace's
PHRENIC NERVE STIMULATION
201
law, initially devised for a body of infinitely thin walls, states that the pressure developed is inversely related to the radius of curvature of that body. As the diaphragm flattens and shortens with increasing lung volume, its radius of curvature increases and hence the pressure generated by the diaphragm may decrease (Kim and Druz, 1976). It is not possible to know the relative importance of these two mechanisms from our studies, but the finding that twitch Pdi decreased when lung volume increased in the presence of the abdominal binding, suggests that the radius of curvaturdof the diaphragm may also have played a role in the decrease in twitch Pdi. It is well known that patients with chronic obstructive pulmonary disease have impaired diaphragmatic function and ventilatory abnormalities (Rochester et al., 1979; Decramer et al., 1980). Altered thoracic shape is thought to be responsible for the diaphragmatic flattening and shortening that occurs (Butler, 1976). The inspiratory force reserve of the respiratory muscles is reduced in such patients and especially in children, so that even small increases in breathing load might expose them to respiratory muscle fatigue and respiratory failure (Bellemare and Grassino, 1983). In this context the present studies in normal volunteers have confirmed that diaphragmatic contractility decreases when lung volume increases and diaphragm shortens. An adaptation to such chronic hyperinflation has been shown to develop in the hamster; a shift of the force-length relationship of the diaphragm occurs to the left so that maximal force can be developed at a shorter diaphragmatic fibre length (Farkas and Roussos, 1982). The extent to which such adaptation may occur in patients with obstructive pulmonary disease is not known. We have also shown that twitch Pdi increases when lung volume decreases and the diaphragm lengthens. Such an increase in diaphragmatic contractility may be relevant clinically in patients with restrictive lung disease where the diaphragm is elevated. Furthermore, respiratory muscle studies in such patients might falsely overestimate diaphragmatic strength. In conclusion, changes in both lung volume and rib cage configuration can affect transdiaphragmatic pressure developed during bilateral phrenic nerve stimulation. When Pdi is being measured, lung volume and rib cage configuration must be monitored carefully. Abdominal binding which achieves more isometric conditions, however, does not appear to be necessary. If follow up studies are being performed, comparison ought to be made only of those stimulations performed at the same lung volume and the same rib cage configuration. This is most likely to be at FRC. Acknowledgements. A.M. was supportedby the Medical Research Councilofthe U.K., and C.B. by NAPP
Laboratories.
References
Agostoni, E. and H. Rahn (1960). Abdominal and thoracic pressures at differentlung volumes.J. Appl. PhysioL 15: 1087-1092. Aubier, M., D. Murciano,J. Lecocguic,N. Viires and R. Pariente(1985).Bilateralphrenicnerve stimulation - a simple technique to assess diaphragm fatigue in humans. J. Appl. PhysioL 58: 58-64.
202
A. MIER et aL
Bellemare, F. and A. Grassino (1983). Force reserve of the diaphragm in patients with chronic obstructive pulmonary disease. J. AppL PhysioL 55: 8-15. Bellemare, F., B. Bigland-Ritchie and J.J. Woods (1986). Contractile properties of the human diaphragm in vivo. J. Appl. Physiol. 61: 1153-1161. Braun, N., N. Arora and D. Rochester (1982). Force length relationship of the normal human diaphragm. J. AppL Physiol. 53: 405-412. Butler, D. (1976). Diaphragmatic changes in emphysema. Am. Rev. Respir. DIS. 114: 155-159. Close, R. (1972). Dynamic properties of mammalian skeletal muscle. Physiol. Rev. 52: 129-183. Decramer, M., M. Demedts, F. Rochette and L. Billiet (1980). Maximal transrespiratory pressures in obstructive lung disease. Clin. Respir. Physiol. 16: 479-490. Farkas, G. A. and C. Roussos (1982). Adaptability of the hamster diaphragm to exercise and/or emphysema. J. Appl. Physiol. 53: 1263-1272. Gandevia, S.C. and D. McKenzie (1986). Human diaphragmatic EMG: changes with lung volume and posture during supramaximal phrenic stimulation. J. AppL Physiol. 60: 1420-1428. Grassino, A. E., W.A. Whitelaw and J. Milic-Emili (1976). Influence of lung volume and electrode position on EMG of the diaphragm. J. Appl. Physiol. 40: 971-975. Hubmayr, R., W. Litchy, P. Gay and S. Nelson (1989). Transdiaphragmatic twitch pressure: effects of lung volume and chest wall shape. Am. Rev. Respir. Dis. 139: 647-652. Kim, M. and W. Druz (1976). Mechanics of the canine diaphragm. J. Appl. Physiol. 41: 369-382. Lansing, R. and J. Savelle (1989). Chest surface recording of diaphragm potentials in man. Electroencephalogr. Clin. Neurophysiol. 72: 59-68. McCully, K.K. and J.A. Faulkner (1983). Length tension relationship of mammalian diaphragm muscles. J. AppL Physiol. 54: 1681-1686. McKenzie, D.K. and S.C. Gandevia (1983). Phrenic nerve conduction times and twitch pressures of the human diaphragm. J. Appl. PhysioL 58: 1496-1504. Mead, J., H. Peterson, G. Grimby and J. Mead (1967). Pulmonary ventilation measured from body surface movements. Science 156: 1383-1384. Mead, J. and S. Loring (1982). Analysis of volume displacement and length changes of the diaphragm during breathing. J. Appl. Physiol. 53: 750-755. Mier, A., C. Brophy, J. Moxham and M. Green (1989). Twitch pressures in the assessment of diaphragm weakness. Thorax 44: 990-996. Miller, J., J. Moxham and M. Green (1985). The maximal sniffin assessment of diaphragm function in man. Clin. Sci. 69: 91-96. Newman, S., J. Road, F. Bellemare, J.P. Clozel, C. Lavigne and A. Grassino (1984). Respiratory muscle length measurements by sonomicrometry. J. Appl. Physiol. 56: 753-764. Newsom-Davis, J. (1967). Phrenic nerve conduction in man. J. Neurol. Neurosurg. Psych. 30: 420-426. Pengelly, L.D., A. Alderson and J. Milic-Emili (1971). Mechanics of the diaphragm. J. AppL Physiol. 30: 797-805. Road, J., S. Newman, J. Derenne and A. Grassino (1986). In vivo length-force relationship of canine diaphragm. J. Appl. Physiol. 60: 63-70. Rochester, D.F., N. S. Arora and N. Braun (1979). The respiratory muscles in chronic obstructive pulmonary disease (COPD). Bull. Eur. Physiopathol. Respir. 15: 951.
193
RESP 01649
Influence of lung volume and rib cage configuration on transdiaphragrnatic pressure during phrenic nerve stimulation in man Anne Mier, Conor Brophy, John Moxham and Malcolm Green Brompton Hospital, London, U.K. (Accepted 16 February 1990) Abstract. Transdiaphragmatic pressure was recorded during bilateral supramaximal percutaneous phrenic
nerve stimulation at 1 Hz (twitch Pdi) to investigate the effect of lung volume and rib cage configuration on diaphragm contractility in man. Stimulations were performed in 5 normal supine subjects at resting end expiration (FRC) and at lung volumes above and below FRC, during relaxation against a closed airway and during isovolume manoeuvres.Twitch Pdi at FRC was 24.4 cm H20. At lung volumes above FRC, twitch Pdi decreased by 7.04 + 3.2 cm HaO per litre of volume change. At lung volumes below FRC, twitch Pdi increased by 12.4 _+8.6 cm HzO per litre of volume change. When the diaphragm was lengthened during an isovolume manoeuvre at FRC, twitch Pdi increased. A similar relationship between lung volume and twitch Pdi was obtained during stimulations performed with abdominal binding. These results demonstrate that the pressure developed by the diaphragm during phrenic nerve stimulation is significantlyaffected both by increases and decreases in lung volume and by the rib cage configuration at which stimulation is performed.
Animal, man; Diaphragm, contractility, dependence on lung volume; Lung volume, influence on diaphragmatic contractility; Phrenic nerve, stimulation, transdiaphragmatic pressure; Pressure, esophageal, gastric, transdiaphragmatic; Rib cage configuration, influence on diaphragm contractility; Technique in respiratory physiology,phrenic nerve stimulation (electric)
Previous studies both in vitro and in vivo have d e m o n s t r a t e d that the force developed by the diaphragm d e p e n d s o n muscle length. The tension generated by the diaphragm has been noted to decrease as diaphragm length shortens a n d the length tension properties of the m a m m a l i a n diaphragm in vitro have been shown to be similar to those of intact limb muscles a n d to be similar between species (McCully a n d Faulkner, 1983). Other studies in cats a n d m a n have d e m o n s t r a t e d that twitch tension during submaximal unilateral phrenic nerve stimulation decreases as lung volume increases (Pengelly et al., 1971). The f o r c e - l e n g t h relationship of the diaphragm was further d e m o n s t r a t e d during
Correspondence to: A. Mier, Department of Medicine, Charing Cross Hospital, Fulham Palace Road, London W6, U.K. 0034-5687/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
194
A. MIER et al.
supramaximal phrenic nerve stimulation in anaesthetised dogs (Kim and Druz, 1976; Road et aL, 1986) and during maximal inspiratory efforts in man (Braun et al., 1982). Several studies have recently focused on the assessment of diaphragm contractility by recording transdiaphragrnatic pressures during phrenic nerve stimulation in man (twitch Pdi) (McKenzie and Gandevia, 1983; Bellemare et al., 1986). Since diaphragm muscle length changes with lung volume and rib cage configuration, it seems important to monitor the volume and configuration at which such phrenic nerve stimulation is performed. Hubmayr and coworkers have recently studied the effect of diaphragmatic shortening on twitch Pdi (Hubmayr et aL, 1989); however, they did not study the effect of diaphragmatic lengthening. We therefore aimed at determining how twitch Pdi was affected both by diaphragmatic shortening and lengthening, by increasing and decreasing lung volume above and below resting end expiration (FRC). We also studied the effects on twitch Pdi of an 'isovolume manoeuvre' at FRC which lengthened the diaphragm. Finally we performed reproducibility studies and then phrenic nerve stimulation in the presence of abdominal binding to determine the effect of lung volume change on twitch Pdi under more isometric conditions. In this way we aimed to determine whether the relationship between twitch Pdi and lung volume would be similar to the length-tension characteristics of the mammalian diaphragm and of the human diaphragm previously described during voluntary inspiratory efforts.
Methods
Studies were performed on 5 healthy subjects, 4 men and one woman, who gave no history of respiratory disease. All subjects were experienced in respiratory manoeuvres, and their phrenic nerves had been stimulated on many occasions so that they were able to relax well during the studies. Their average age was 36 yrs (range 29-43), their height 171 (157-179) cm, their weight 66 (48-83) kg and predicted vital capacity 4.36 (3.0-5.0) L. All subjects gave informed consent to the studies. Lung volume was recorded using an Ohio 840 dry spirometer. Changes in rib cage and abdominal anteroposterior dimensions were recorded with two pairs of previously calibrated, linearised magnetometers (N. H. Peterson, Boston, MA) placed in the midline in the 5th intercostal space and 2 cm above the umbilicus respectively (Mead et aL, 1967). Electromyograms (EMG) of the right and left hemidiaphragms were recorded with surface electrodes in the 6th and 7th intercostal spaces 2-3 cm from the costal margin on either side. Electromyogram signals were processed using an amplifier (Medelec PA 63) and were filtered below 16 and above 1600 Hz. Oesophageal and gastric pressures were recorded with balloon catheter systems (PK Morgan), positioned in the mid-oesophagus and stomach (Agostoni and Rahn, 1960) and filled with 0.5 and 1.5 ml of air, respectively. The balloon catheters were connected to differential pressure transducers (Validyne, MP-45-1 + 150 cm H20 ) and transdiaphragmatic pressure (Pdi) was obtained electronically by subtraction of oesophageal pressure from gastric pressure. Pdi at resting end expiration (FRC) was used as zero reference point.
PHRENIC NERVE STIMULATION
195
The right and left phrenic nerves were stimulated simultaneously at the posterior border of the sternomastoid muscles at the level of the cricoid cartilage (Newsom-Davis, 1967). Two pairs of bipolar surface stimulating electrodes with felt tips 0.5 cm in diameter (Medelec 53054) were used, attached to a dual Digitimer 3072 isolated stimulator producing rectangular wave impulses 0.1 msec in duration at a frequency of 1 Hz. The voltage of stimulation was increased on each side in turn until there was no further increase in the size of either the diaphragm muscle action potential or the twitch Pdi. The voltage was then increased by a further 10~o so that supramaximal stimulation was achieved (between 80 and 160 volts for each nerve). Optimal position of the stimulating electrodes in the neck was ensured by closely monitoring the size of the diaphragm muscle action potentials recorded on the E M G during stimulation. If the size of either E M G decreased, it was assumed that the stimulation electrode in the neck had moved and the run was abandoned. A series of 10 maximal sniffs was initially performed from FRC, without a noseclip and the best of 10 sniffs was recorded to assess diaphragm contractility and to confirm that all subjects had normal diaphragm strength (Miller et al., 1985). A full vital capacity manoeuvre from total lung capacity (TLC) to residual volume (RV) was recorded in the supine posture three times. Both phrenic nerves were then stimulated simultaneously in the supine posture while the subjects, wearing a noseclip, breathed through the spirometer. Stimulations were first performed at FRC to obtain at least 10 twitches and then whilst the subject relaxed with a closed glottis at different lung volumes up to 4.0 L (mean 2.46 + 0.95 L) above FRC and decreasing volumes down to 0.9 L (mean 0.49 + 0.27 L) below FRC. Between 3 and 5 twitches were obtained at each lung volume. At least 5 runs were obtained in each subject and a series of 10 twitches was recorded at FRC between each run to ensure that twitch Pdi at FRC remained constant and that the effect of lung volume changes was a true one. Subject 1 was studied on 3 separate days and subjects 2 and 3 on two separate days. In 2 subjects, (1 and 2) isovolume manoeuvres were also performed. The subject initially relaxed at FRC with closed glottis. Phrenic nerve stimulation was performed at FRC and then immediately while the subject performed an isovolume manoeuvre (abdomen sucked in and rib cage expanded, thereby lengthening the diaphragm) at FRC. In 2 subjects (1 and 3) studies were also performed with an abdominal binder so that stimulation could be performed at different lung volumes with the diaphragm under conditions that were as isometric as possible. The abdomen was tightly bound with firm inextensible bandages from the lower rib cage to the level of the symphysis pubis. Volume studies with an abdominal binder were performed on 2 occasions in subject 1 and on one occasion in subject 3. All signals were recorded on an eight channel recorder (Minograph 800, Siemens) and were stored on a magnetic FM tape recorder with a frequency response of up to 5000 Hz (Racal Store 7) for later play back and analysis. All results are expressed as mean + SD. Results of twitch Pdi are expressed as mean + SD at each lung volume. Stepwise polynomial regression analysis was performed using the Cricket Graph Data Analysis Program (Apple plus) to investigate the relationship between twitch Pdi and lung volume.
196
A. MIER etal.
Results Vital capacity (4.12 + 1.03 L) and sniffPdi (138 + 27 cm H 2 0 ; normal > 98 cm H 2 0 ) (Miller et aL, 1985) were normal in all 5 subjects. Phrenic nerve stimulation was well tolerated by all subjects and no problems were experienced with relaxing at different lung volumes with a closed glottis. Typical records of EMG, Poes, Pg and Pdi during phrenic nerve stimulation at FRC in one subject are shown in fig. 1. Phrenic nerve conduction time was normal (6.7 + 0.5 msec; Newsom-Davis, 1967). Bilateral twitch Pdi recorded at FRC ranged from 20.5 to 32.3 cm H 2 0 (24.4 + 4.7 cm H 2 0 ; normal > 10.1) (Mier et al., 1989). Twitch Pdi was reproducible when recorded at FRC between runs in each subject. Coefficient of variation of twitch Pdi at FRC was 6.74 + 1.54% within days in the 5 subjects. By contrast twitch Pdi altered significantly with changes in lung volume. Records obtained in one representative subject are shown in fig. 2. As lung volume increased above FRC, twitch Pdi decreased. As lung volume decreased below FRC, twitch Pdi increased. Almost no background diaphragm E M G activity was seen during stimulation at FRC, at increased and decreased lung volumes. This and careful observation confirmed that subjects were able to relax against a closed glottis and that the 'diaphragm was indeed relaxed during stimulation at different lung volumes. Results obtained in all 5 subjects are shown in fig. 3. The mean decrease in twitch Pdi at the highest lung volumes studied above FRC was 15.1 + 5.0 cm H20. This was equivalent to a reduction in twitch Pdi of 61.7 + 19.7% compared to values at FRC. Overall this represented a decrease in twitch Pdi of 7.04 + 3.2 cm H 2 0 per litre of volume change above FRC. The mean increase in twitch Pdi at lowest lung volumes below FRC was 5.2 + 3.9 cm H20. This was equivalent to an increase in twitch Pdi of 22.6 + 15.4% compared to values at FRC. Overall twitch Pdi increased by 12.4 + 8.6 cm H 2 0 per litre of volume change below FRC. When an isovolume manoeuvre was performed at FRC, the rib cage anteroposterior dimension recorded on the magnetometers increased while the abdominal dimension decreased (Mead and Loring, 1982) and twitch Pdi increased (fig. 3). The increase in Edl Edi
(L) (R)
~
/~.
~
'~-,.
"1 4001~v
Poes
~
[
lOcmH20
Pg Pdl i--.-4
0.1sec Fig. 1. Electromyograms (Edi) of left (L) and right (R) hemidiaphragms, oe~ophageal (Poes), gastric (Pg) and transdiaphragmatic (Pdi) pressures recorded during bilateral phrenic nerve stimulation at 1 Hz at FRC (Note: electrocardiogram also seen on top trace).
197
PHRENIC NERVE STIMULATION
/
Volume
(L)
/
RC AP -~/~._ ABDOAP EdI(L)
Edi(R)
I IL ~ -
-
-
-
-
-
-
-
-
.,,1. i. t .... _~JF~I~ -
..L_LLJ_LJ._L-~. L [ 1 J :I ~ - ,
1] I
I
l 1_ I I ] I [ I ±ll I
I-I
/ IJ/~
r
Io.scm
L ~_~~'-~-
I 1 i
I-r
I 1
J. l.. [_1 I t I
I
l~l
1 400Hv
I ....
~ ~ ~ I ~/~A , , . .II
rlocmH20
Pdi 1 sec Fig. 2. Lung volume, rib cage (RC AP) and abdominal (ABDO AP) anteroposterior dimensions, electromyograms (Edi) oflett (L) and right (R) hemidiaphragms and transdiaphragmatic pressures (Pdi) recorded during bilateral phrenic nerve stimulation at 1 Hz at increasing lung volumes, in one representative subject. (Note: (L) and (R) Edi are contaminated by activity in chest wall muscles during each inspiration).
Sublect1
Subject 2 30.
,.r
10'
lo
o
• 2
• I
• 0
~bl*ct3
Y
0
n u
..j"
t.. a
SubJe~S
10
o
V o I u m •
(I)
Fig. 3. Relationship between lung volume and twitch transdiaphragmatic pressure recorded during bilateral phrenic nerve stimulation in 5 subjects. (Arrow = stimulation during isovolume manoeuvre at FRC. Zero lung volume represents residual volume (RV); lung volume increases from right to left).
198
A. MIER etal.
,/d
Subject1 -r-
Subject 2
30
10
o 4
3
2
~
0
lO .
o
O'
Volume
4
3
2
1
0
111
Fig. 4. Reproducibility of twitch transdiaphragmatic pressure during bilateral phrenic nerve stimulation at increasing lung volumes above RV in 3 subjects on different days. (Zero lung volume represents RV. /x = day 1 as in Fig. 1; • = day2; © = day 3.)
twitch Pdi during the isovolume manoeuvre was 6~o of the relaxed FRC value in subject 1 and 23 ~/o of the relaxed FRC value in subject 2. The relationship between lung volume and twitch Pdi seen in the 3 subjects who were studied on 2 or 3 separate days is shown in fig. 4. A similar relationship was seen in 2 subjects on different days although the actual height of twitch Pdi varied between days in one of these subjects (left hand panel, fig. 4). A different pattern of pressure-volume relationship was seen in the third subject on 2 different days (right hand panel, fig. 4). Results of stimulations performed with an abdominal binder, are shown in fig. 5. The size of the diaphragm evoked muscle action potentials during phrenic nerve stimulation remained constant during stimulation at FRC confirming that the position of the stimulating electrodes in the neck remained optimal. The voltage of stimulation did not need to be altered in any subject once supramaximal stimulation had been achieved. The size of the evoked diaphragm muscle action potentials altered slightly (increase or decrease) at increased lung volumes in some subjects, suggesting an altered contact between the recording surface electrodes and the diaphragm muscle.
Subject3 30"20,
Subjecty
20-
"r
E(,I '112 n r-
1o-
10-
0
' 4
" 2
' 0
Volume
0
, 4
, 2
, 0
(I)
Fig. 5. Twitch transdiaphragmatic pressure during bilateral phrenic nerve stimulation in 2 subjects with abdominal binders. (Zero lung volume represents RV. Subject 1 studied on 2 separate occasions.)
PHRENIC NERVE STIMULATION
199
Discussion These studies demonstrate that transdiaphragrnatic pressure recorded during bilateral supramaximal phrenic nerve stimulation depends significantly on the lung volume at which stimulation is performed. Since twitch Pdi has been advocated as a reliable, objective indication of diaphragm contractility (Bellemare et al., 1986) it is important to control carefully for both lung volume and rib cage configuration, when attempting to compare results of twitch Pdi within a subject between different studies. We have shown that twitch Pdi not only decreases at increased lung volumes as shown by Hubmayr et al. (1989) but that twitch Pdi increases at decreased lung volumes. In addition to changing with lung volume, twitch Pdi also varied with rib cage configuration. When the diaphragm was lengthened by performing an isovolume manoeuvre at FRC, twitch Pdi increased, even though lung volume did not alter. This illustrates the importance of monitoring rib cage configuration when recording twitch Pdi to ensure that stimulations are being performed at a constant diaphragm length. We have previously shown that the height of twitch Pdi at FRC is reproducible in normal subjects (Mier et aL, 1989). In the present studies we similarly found that twitch Pdi performed at FRC returned to the control value between each volume run. This supported the fact that the reduction in twitch height found at increased lung volumes was a true effect of diaphragm shortening. We took care to ensure that stimulus maximality was maintained throughout the range of lung volumes studied. Stimulations were not performed near total lung capacity or residual volume since tensing of the neck muscles might have reduced contact between the stimulating electrodes and the phrenic nerves. Accordingly, we studied the effect of lung volume changes that were smaller and not at the extremes of vital capacity. Similarly studies were performed in the supine posture, in which subjects found it easiest to relax the neck muscles and in which it was easiest to locate and maintain reliable stimulation of the phrenic nerves. Other workers have used rigid needle electrodes to provide phrenic nerve stimulation (Aubier et al., 1985) but the position of needle electrodes in the neck may similarly be altered by movement of the soft tissue of the neck during deep inspiration and expiration. We recorded electromyograms of each hemidiaphragm with surface electrodes in an attempt to ensure that constant stimulation of the phrenic nerves was being maintained. Although a small amount of variability in the amplitude of diaphragm E M G was noted on inspiration, it seems likely that this was due to movement of the diaphragm causing an alteration in the relationship between the surface electrodes and the diaphragm (Grassino et al., 1976; Bellemare et al., 1986; Gandevia and McKenzie, 1986) rather than to a reduction in stimulation of the phrenic nerves due to movement of the stimulating electrodes. Alternatively, an oesophageal electrode could have been used to record diaphragm E M G activity but similar problems with alterations in amplitude of phrenically evoked compound muscle action potentials may occur with such an electrode because of shifts in diaphragm position. Indeed, it has recently been shown that surface electrodes can be used to record diaphragm activity without distortion caused by movement of the diaphragm at lung volumes from FRC to about 50~o of vital capacity (Lansing and Savelle, 1989).
200
A. MIER et al.
A recent paper reported the effect of lung volume changes on twitch Pdi, using fine wire electrodes implanted near the phrenic nerves (Hubmayr et aL, 1989). Bilateral twitch Pdi in this study (19 to 36 cm H20 ) was very similar to that found in our subjects and the relationship of change in twitch Pdi with lung volume was almost identical (8 cm H20/L). Similarly the coefficient of variation within a day (2 to 5 ~o) was similar to that in our subjects and in previous studies that we have also performed on normal subjects (5 ~o) (Mier et aL, 1989). However, we chose to use surface stimulating electrodes which are less invasive, easier and quicker to use; they can also be applied more routinely in the investigation of normal subjects and patients. The finding that our results were similar to those of Hubmayr and coworkers, helps to confirm that percutaneous phrenic nerve stimulation is indeed a reliable technique, in addition to being easier and less invasive to use. Since we used a non-invasive technique, we were able to perform studies on several different days in our subjects. Although we obtained similar results on different days in one subject (middle panel, fig. 4), the results in two further subjects were much less repeatable. We felt that this might have been due to the fact that stimulations were not performed under isometric conditions. We therefore repeated the studies in these 2 subjects, with abdominal binding, but again found that results were not very reproducible. We were unable to explain why the shape of the pressure-volume relationship in the first study in subject 3 differed from that in the other subjects, and why it differed so significantly between days. However, previous reports have shown that twitch Pdi can vary from day to day in normal subjects (Mier et al., 1989). Since Hubmayr et al. (1989) did not perform studies on more than 1 day, it is not possible to know whether the repeatability of their results was similar to ours. The pressure-volume curves that we obtained were quantitatively similar to the length-tension relationship of the mammalian diaphragm, demonstrated in previous studies. The force length curve of the isolated skeletal muscle preparation is quite uniform in shape (Close, 1972). When measured previously during static inspiratory efforts in man, the shape of the force length curve was found by Braun and coworkers, (Braun et al., 1982) to be different from that obtained by direct measurement in dog (Kim and Druz, 1976); in man a plateau was found at higher lung volumes. Braun and colleagues concluded that the difference in shape between the two studies was because diaphragmatic length in their subjects had been estimated from lung volume. We were similarly anxious that this indirect assessment of diaphragmatic length was less good than the determination of diaphragm length as measured on a chest radiograph. However, the individual lengths of the costal and crural parts of the diaphragm have been assessed using implanted piezoelectric transducers in dogs (Newman etal., 1984). During passive lung inflation, a nearly linear inverse relationship between lung volume and diaphragm length was found. Thus although changes in rib cage configuration can significantly affect diaphragm length, it seems that an indication of diaphragm muscle length can be obtained from lung volume. In addition to the length tension properties of the diaphragm muscle, the curvature of the diaphragm may also play a role in determining the height of twitch Pdi. Laplace's
PHRENIC NERVE STIMULATION
201
law, initially devised for a body of infinitely thin walls, states that the pressure developed is inversely related to the radius of curvature of that body. As the diaphragm flattens and shortens with increasing lung volume, its radius of curvature increases and hence the pressure generated by the diaphragm may decrease (Kim and Druz, 1976). It is not possible to know the relative importance of these two mechanisms from our studies, but the finding that twitch Pdi decreased when lung volume increased in the presence of the abdominal binding, suggests that the radius of curvaturdof the diaphragm may also have played a role in the decrease in twitch Pdi. It is well known that patients with chronic obstructive pulmonary disease have impaired diaphragmatic function and ventilatory abnormalities (Rochester et al., 1979; Decramer et al., 1980). Altered thoracic shape is thought to be responsible for the diaphragmatic flattening and shortening that occurs (Butler, 1976). The inspiratory force reserve of the respiratory muscles is reduced in such patients and especially in children, so that even small increases in breathing load might expose them to respiratory muscle fatigue and respiratory failure (Bellemare and Grassino, 1983). In this context the present studies in normal volunteers have confirmed that diaphragmatic contractility decreases when lung volume increases and diaphragm shortens. An adaptation to such chronic hyperinflation has been shown to develop in the hamster; a shift of the force-length relationship of the diaphragm occurs to the left so that maximal force can be developed at a shorter diaphragmatic fibre length (Farkas and Roussos, 1982). The extent to which such adaptation may occur in patients with obstructive pulmonary disease is not known. We have also shown that twitch Pdi increases when lung volume decreases and the diaphragm lengthens. Such an increase in diaphragmatic contractility may be relevant clinically in patients with restrictive lung disease where the diaphragm is elevated. Furthermore, respiratory muscle studies in such patients might falsely overestimate diaphragmatic strength. In conclusion, changes in both lung volume and rib cage configuration can affect transdiaphragmatic pressure developed during bilateral phrenic nerve stimulation. When Pdi is being measured, lung volume and rib cage configuration must be monitored carefully. Abdominal binding which achieves more isometric conditions, however, does not appear to be necessary. If follow up studies are being performed, comparison ought to be made only of those stimulations performed at the same lung volume and the same rib cage configuration. This is most likely to be at FRC. Acknowledgements. A.M. was supportedby the Medical Research Councilofthe U.K., and C.B. by NAPP
Laboratories.
References
Agostoni, E. and H. Rahn (1960). Abdominal and thoracic pressures at differentlung volumes.J. Appl. PhysioL 15: 1087-1092. Aubier, M., D. Murciano,J. Lecocguic,N. Viires and R. Pariente(1985).Bilateralphrenicnerve stimulation - a simple technique to assess diaphragm fatigue in humans. J. Appl. PhysioL 58: 58-64.
202
A. MIER et aL
Bellemare, F. and A. Grassino (1983). Force reserve of the diaphragm in patients with chronic obstructive pulmonary disease. J. AppL PhysioL 55: 8-15. Bellemare, F., B. Bigland-Ritchie and J.J. Woods (1986). Contractile properties of the human diaphragm in vivo. J. Appl. Physiol. 61: 1153-1161. Braun, N., N. Arora and D. Rochester (1982). Force length relationship of the normal human diaphragm. J. AppL Physiol. 53: 405-412. Butler, D. (1976). Diaphragmatic changes in emphysema. Am. Rev. Respir. DIS. 114: 155-159. Close, R. (1972). Dynamic properties of mammalian skeletal muscle. Physiol. Rev. 52: 129-183. Decramer, M., M. Demedts, F. Rochette and L. Billiet (1980). Maximal transrespiratory pressures in obstructive lung disease. Clin. Respir. Physiol. 16: 479-490. Farkas, G. A. and C. Roussos (1982). Adaptability of the hamster diaphragm to exercise and/or emphysema. J. Appl. Physiol. 53: 1263-1272. Gandevia, S.C. and D. McKenzie (1986). Human diaphragmatic EMG: changes with lung volume and posture during supramaximal phrenic stimulation. J. AppL Physiol. 60: 1420-1428. Grassino, A. E., W.A. Whitelaw and J. Milic-Emili (1976). Influence of lung volume and electrode position on EMG of the diaphragm. J. Appl. Physiol. 40: 971-975. Hubmayr, R., W. Litchy, P. Gay and S. Nelson (1989). Transdiaphragmatic twitch pressure: effects of lung volume and chest wall shape. Am. Rev. Respir. Dis. 139: 647-652. Kim, M. and W. Druz (1976). Mechanics of the canine diaphragm. J. Appl. Physiol. 41: 369-382. Lansing, R. and J. Savelle (1989). Chest surface recording of diaphragm potentials in man. Electroencephalogr. Clin. Neurophysiol. 72: 59-68. McCully, K.K. and J.A. Faulkner (1983). Length tension relationship of mammalian diaphragm muscles. J. AppL Physiol. 54: 1681-1686. McKenzie, D.K. and S.C. Gandevia (1983). Phrenic nerve conduction times and twitch pressures of the human diaphragm. J. Appl. PhysioL 58: 1496-1504. Mead, J., H. Peterson, G. Grimby and J. Mead (1967). Pulmonary ventilation measured from body surface movements. Science 156: 1383-1384. Mead, J. and S. Loring (1982). Analysis of volume displacement and length changes of the diaphragm during breathing. J. Appl. Physiol. 53: 750-755. Mier, A., C. Brophy, J. Moxham and M. Green (1989). Twitch pressures in the assessment of diaphragm weakness. Thorax 44: 990-996. Miller, J., J. Moxham and M. Green (1985). The maximal sniffin assessment of diaphragm function in man. Clin. Sci. 69: 91-96. Newman, S., J. Road, F. Bellemare, J.P. Clozel, C. Lavigne and A. Grassino (1984). Respiratory muscle length measurements by sonomicrometry. J. Appl. Physiol. 56: 753-764. Newsom-Davis, J. (1967). Phrenic nerve conduction in man. J. Neurol. Neurosurg. Psych. 30: 420-426. Pengelly, L.D., A. Alderson and J. Milic-Emili (1971). Mechanics of the diaphragm. J. AppL Physiol. 30: 797-805. Road, J., S. Newman, J. Derenne and A. Grassino (1986). In vivo length-force relationship of canine diaphragm. J. Appl. Physiol. 60: 63-70. Rochester, D.F., N. S. Arora and N. Braun (1979). The respiratory muscles in chronic obstructive pulmonary disease (COPD). Bull. Eur. Physiopathol. Respir. 15: 951.
