Diaphragm pacing

Diaphragm pacing

Diaphragm pacing Evaluation of current waveforms for effective ventilation To evaluate the effectivenss of the configuration of the stimulating wavefo...

1MB Sizes 0 Downloads 116 Views

Diaphragm pacing Evaluation of current waveforms for effective ventilation To evaluate the effectivenss of the configuration of the stimulating waveform on diaphragm pacing, we evaluated several different current forms: UDC-bipolar, UDC-monopolar cathodal, UDC-monopolar anodal, and ABDC. During stimulation with a pulse interval of 37 msec., a decrease in tidal volume was observed during the initial 3 hours with UDC-bipolar and UDC-monopolar anodal waveforms. Both UDC-monopolar cathodal and ABDC stimulation maintained the initial effectiveness for 6 hours. The decrease in tiddl volume of UDC-monopolar anodal closely paralleled that of UDC-bipolar stimulation. Decreasing the pulse interval to 20 msec. caused a decrease in tidal volume with both UDC-monopolar cathodal and ABDC waveforms. Arterial oxygen tension (Pao.) in these experiments decreased to about 60 mm. Hg soon after the onset of unilateral diaphragm pacing. The concomitant decrease in tidal volume seen with UDC-bipolar stimulation could be avoided through the administration of oxygen to keep the animal's Pao, about 100 mm. Hg. The amplitude of the evoked diaphragmatic action potentials decreased significantly under hypoxemia and returned to normal with hyperoxygenation. From these short-term experiments, our findings indicate that waveform configuration does influence the time of onset of diaphragm fatigue due to either an electrochemical disturbance at the electrode-nerve interface or exhaustion of acetylcholine in the neuromuscular junction. Further, hypoxemia accelerates the occurrence offatigue.

Toshihiro Kaneyuki, M.D., James F. Hogan, B.S.E.E., William W. L. Glenn, M.D., and Wade G. Holcomb, H.E.E., New Haven, Conn.

Electrical stimulation of the phrenic nerve (diaphragm pacing) is an attractive method for prolonged ventilatory assistance because it overcomes several physiological drawbacks of positive-pressure respiration.P": 12. 13. 16. 19.21 However, the problem remains of how to extend the useful period of effective ventilation. Diminished excitability of the nerve may occur secondary to polarization or electrolysis at the electrodenerve interface. 6, 19 The electrochemical effects may be significantly decreased if an appropriate combination of stimulation parameters and electrode materials can be found. Many techniques for prolonging the time of effective nerve stimulation have been attempted: reversing the direction of current flow through the nerve electrode after termination of the stimulating current pulse,

resulting in "zero net flow" 9, 11; combining biphasic current flow and platinum electrode-" 19; using bipolar stimulation which reverses the direction of current flow through the nerve with each excitation pulse>: 15; periodically switching the current flow among four electrodes placed radially around the nerve 7; reducing the total coulombs applied to the nerve by reducing pulse duration, pulse repetitive frequency, and inspiration duration." 5, 15, 18 The present study has been designed to evaluate in short-term experiments the effects of various current waveforms and electrode combinations by comparison of their ability to provide ventilatory support. Four current waveforms were tested: biphasic (UDC) bipolar, biphasic (UDC) monopolar anodal, biphasic (UDC) monopolar cathodal, and alternating biphasic bipolar (ABDC).*

From the Department of Surgery, Section of Cardiothoracic Surgery, Yale University School of Medicine, 333 Cedar St., New Haven, Conn. 06510. Supported by U. S. Public Health Service Grant HL0465I.

Materials and methods

Received for publication Feb. 2, 1977. Accepted for publication Feb. 25, 1977.

*UDC = unidirectional current; ABDC = alternating biphasic directional current.

Adult mongrel dogs weighing 10 to 20 kilograms were sedated with pentobarbital (30 mg. per kilogram).

109

The Journal of

1 10

Thoracic and Cardiovascular Surgery

Kaneyuki et al.

Cathodal Anodal

125

llJ

100

L..-."..._--y

~

::J ...J

o

> 75 ...J

« o

I-

u.

50

o

l/)

llJ

~

«

25

:I:

u

o

L-_----l._ _-'--_ _..L-_ _L.-_--'-_ _--'-_

Fig. 1. Average tidal volume changes following unilateral phrenic nerve stimulation with UDC-bipolar, UDCmonopolar anodal, UDC-monopolar cathodal, and ABDC waveforms. Bars represent standard deviations. UDC, Unidirectional current. ABDC, Alternating biphasic directional current. They were placed on the operating table in the left lateral recumbent position and ventilated with a positive-pressure respirator with room air. The femoral artery and vein were cannulated for blood pressure monitoring and arterial blood sampling, as well as for administration of fluids and anesthetics. The electrocardiogram (Lead II) was also monitored continuously. The right phrenic nerve in the neck was explored, with meticulous care taken to prevent injury. The 5 mm. electrode cuff was then secured around the nerve. The operative field was kept from drying by frequent flooding with saline solution. The stimulator used in this study was a radiofrequency coupled unit. The system consists of an external power source (transmitter) and a receiver circuit with electrodes connected to a nerve cuff. The stimulating pulses are coupled to the receiver via a 2.0 MHz carrier from an antenna. The constant current output from the receiver can be adjusted by the transmitter to provide an output up to 10 mao The current pulses applied to the nerve in our study were 0.15 rnsec. in duration, with a pulse interval of either 37 msec. or 20 msec. for an inspiration period of 1.3 seconds. The diaphragm was paced at a rate of 20 breaths per minute throughout the experiments. Following determination of the initial threshold current and the current required for maximal diaphragm contractions, the current was set at 1.5 mao This current

amplitude was, in all experiments, greater than the amount required for maximal contraction of the diaphragm and was maintained throughout the experiments. The experiments were divided into the following three series (Table I): Series I. Evaluation of the different current waveforms of nerve stimulation in accordance with the waning of effectiveness as ascertained by a decrease in tidal volume. The current waveforms used in this series were (I) a biphasic unidirectional current with bipolar electrodes, with cathodal current applied to the distal electrode (UDC-bipolar); (2) a biphasic unidirectional current with a cathodal nerve electrode and an indifferent anodal plate (Medtronic, Model 6983 indifferent lead) placed subcutaneously in the ipsilateral chest wall (UDC-monopolar cathodal); (3) the same current as in (2), with an anodal nerve electrode and an indifferent cathodal plate (UDC-monopolar anodal); (4) an alternating bidirectional current provided by an additional circuit delivering reversed pulses at every stimulating event (ABDC). The pulse interval used in this series was 37 msec. (approximately 25 Hz) for the inspiration duration of 1.3 seconds. In another group (UDC-bipolar with oxygen), the effect of oxygen on the onset of fatigue was examined by supplying 100 percent oxygen to the animal through the endotracheal tube at a rate of 2 L. per minute; the identical procedure used in UDC-bipolar stimulation was employed. Each group consisted of five animals, and spontaneous breathing was suppressed by carefully administered doses of barbiturates. No experiment was continued for more than 6 hours to ensure stability of the animal's condition. Throughout each experiment the evoked diaphragm action potentials were measured via a pair of electrodes* attached at the eighth intercostal space on the ipsilateral side. Series II. Comparison between UDC-monopolar cathodal and ABDC, using pulse trains with 20 msec. intervals (50 Hz) to hasten the signs of fatigue. The best current for diaphragm pacing was determined by making comparisons according to the method described under Series I. Each group consisted of five animals, and the experiments lasted 3 hours. The pacing rate and inspiration duration were the same as in the Series I experiments. *Flexon, No. 2597-63; Davis & Geck, American Cyanamid Company, Pearl River, N. Y.

Volume 74

Diaphragm pacing

Number 1 July, 1977

120

0--0

P02. e-e PC02: UDC-Bipolar

-~

P02'

'<1---"l

P02' ........... PC02: UDC-Monopolar Anodal

0--0

P02' - - - PC02 : ABDC

.to--.to

111

PC02: UDC-Monopolar Cathodal

100

80 Cl'

~

E

--

60

E 40 20

o

o

(30")

10

Fig. 2. Blood gas alterations following unilateral diaphragm pacing using four different modes of stimulation.

Table I. Experimental modes and results Pulse interval (msec.)

Time of stimulation (hr.)

37 37 37 37 37

3 6 3 6 6

58.6 92.8 63.8 97.4 88.6

Cathodal

20 20

3 3

63.4 ± 10.3 75.4 ± 9.8

Cathodal

20

No. of animals

Waveform

Electrode

Polarity

5 5 5 5 5*

UDC-biphasic UDC-biphasic UDC-biphasic ABDC UDC-biphasic

Bipolar Monopolar Monopolar Bipolar Bipolar

Cathodal-caudad Cathodal Anodal

Series II

5 5

UDC-biphasic ABDC

Monopolar Bipolar

Series III

8

UDC-biphasic

Monopolar

Group

Series I

Cathodal-caudad

Decrease in tidal volume at termination of experiment (%)

± ± ± ± ±

9.6 5.6 9.4 7.1 7.7

(Measurement of MAP under hypoxemia and hyperoxia) Legend: MAP, Muscle action potentials of the diaphragm. UDC, Unidirectional current. ABDC, Alternating bidirectional current. 'Oxygen was supplied to animals during diaphragm pacing at a rate of 2 L. per minute.

Series III. Measurement of the diaphragm action potentials under acutely induced hypoxemia and hyperoxia. Correlation of contractility of the diaphragm muscle with arterial oxygen tension (Pao.) was studied by subjecting eight dogs to a small upper medial laparotomy with the use of positive-pressure respiration. A pair of recording electrodes (Tefloncoated platinum-iridium lead wire, Mediwire No. 30.OIR) were secured to the anterior muscular portion of the ipsilateral diaphragm. The nerve was stimulated as in Series II experiments. Hypoxemia and hyperoxia were achieved by altering the tidal volume, the frequency of ventilations, or the oxygen supply to the

respirator. Threshold current, maximal diaphragm response level, and the amplitude of the action potentials were recorded at hourly intervals. A dual-channel oscilloscope equipped with a high-gain different preamplifier* was used to measure muscle action potential. Arterial blood samples were taken anaerobically every 30 minutes, and Po 2 , Pcoj, and pH were measured at 37° C. with a blood gas analyzer. t Serial tidal volumes were observed at 15 minute intervals with a Fleish *Tektronix, Inc., Portland, Ore.; Type 564 oscilloscope with camera and Type 3A3 amplifier. tComing Scientific Instruments, Medfield, Mass.; Digital 160.

The Journal of Thoracic and Cardiovascular Surgery

Kaneyuki et at.

1 12

0--<> P02 ••- . C02: UDC-Bipolor with 02

0--0 UDC-Bipolar with 02

0--0

__ UDC-Bipolor without 02

125

120

:.l!

° ~

P02. __ C02: UDC-Bipolar without 02

100 ~--¢---4---~---e~

100

:::>

...J

o > 75 ...J «

80 til

o

I-

u,

J:

E 60

50

E

o

40

Ul

w

~

«

25

20

J:

o

OL------'------'-------'----'-----"---""O"-



20



TIME (hrs)







°

o

(30')

I 1°

I





TIME (hrs)







Fig. 3. Comparison between UDC-bipolar stimulation with and without oxygenin accordance with tidal volume changes. Bars represent standard deviations.

Fig. 4. Blood gas alterations of UDC-bipolar stimulation with and without oxygen.

pneumotachygraph and an electronic integrator, constructed in our laboratory. Experimental data were analyzed by Student's t test.

creased with UDC-bipolar stimulation to 58.6 ± 9.6 percent of the initial value and with UDC-monopolar anodal to 63.8 ± 9.4 percent. UDC-monopolar cathodal and ABDC maintained the tidal volumes constant for 6 hours (Fig. 1, Table I). The decrease in tidal volume with UDC-monopolar anodal stimulation was very close to that of UDC-bipolar, and the difference between UDC-bipolar and the other two groups was significant (p < 0.05). Paoz decreased to nearly 60 mm. Hg within 30 minutes of initiation of diaphragm pacing, but Pacoz remained below 40 mm. Hg (Fig. 2). There were no significant differences in blood gas levels among the four groups. 4. A decrease in tidal volume on stimulation was prevented by the administration of oxygen. Fig. 3 compares UDC-bipolar with oxygen and UDC-bipolar without oxygen; the difference was significant (p < 0.05). There was significant difference between the Paoz levels (p < 0.05) of animals receiving oxygen by endotracheal tube and those receiving room air. The Pacoz increased slowly in both groups (Fig. 4). 5. Function of the phrenic nerve and diaphragm was evaluated by recording the evoked diaphragm action potentials with surface electrodes. Conduction timefrom stimulation of the nerve to onset of the diaphragmatic contractions-ranged from 3.6 to 5.4 msec. These values remained unchanged throughout the experiments, and variation in the conduction time was due mainly to the difference in size of the animals. The amplitude of the action potentials was an unreli-

Results Series I. Evaluation of the various applied current waveforms. 1. Initial threshold currents ranged from 50 to 600 /La, and the maximal diaphragm response was achieved by increasing the current 60 to 700 /La above the threshold values. Measured at 30 to 60 minute intervals during the first 1 to 3 hours of stimulation, both the threshold and the maximal response levels increased slightly from 160 to 800 /La and 340 to 1,300 /La, respectively. After this time, current requirements either remained stable or decreased slightly throughout the experiment. There was no observed difference between the threshold and maximal response levels of the four current waveforms tested. 2. Initial tidal volumes were 157 ± 31.4 ml. (mean ± S.D.) with UDC-bipolar stimulation, 116 ± 17.0 ml. with UDC-monopolar cathodal, 121 ± 17.8 ml. with UDC-monopolar anodal, 141 ± 47.7 ml. with ABDC, and 114 ± 34.6 ml. with UDC-bipolar stimulation with oxygen. The tidal volume variations corresponded to the weight of the animals; there were no statistical differences among these groups. 3. Effective contractions of the unilateral diaphragm were obtained by supramaximal stimulation of the ipsilateral phrenic nerve with each of the four current waveforms. Over the first 3 hours, tidal volumes de-

Volume 74

able measure of the diaphragm contractility, since we found no correlation between the amplitudes and the tidal volumes. Series II. Comparison between UDC-monopolar cathodal and ABDC stimulation using a pulse interval of 20 msec. Over the 3 hour period of stimulation, tidal volume decreased with ABDC stimulation to 75.4 ± 9.8 percent of the initial value and with UDC-monopolar cathodal to 63.4 ± 10.3 percent of the initial value. (Table I). Even though ABDC appeared to be superior to the other current waveforms, the difference between them was insignificant. Alterations in blood gases were similar to those of the Series I experiment. Series III. Measurements of the muscle action potentials of the diaphragm with induced hypoxemia and hyperoxia. The amplitude of the action potentials decreased to 81.0 ± 12.8 percent of the control value under hypoxemia (Fig. 5), and this value was statistically significant (p < 0.01). On hyperoxygenation, the amplitude recovered to normal. Discussion Fig. I shows that UDC-bipolar stimulation resulted in a greater reduction in tidal volume than did UDCmonopolar cathodal or ABDC. Moreover, UDC-monopolar anodal stimulation caused a decrease in tidal volume similar to that of UDC-bipolar stimulation. Repeated electrical stimulation to the nerve with a monophasic bipolar current results in polarization and electrolysis at the electrode-nerve interface, with a subsequent increase in electrode impedance; excitability of the nerve either decreases or disappears. The biphasic current used in this study was designed to neutralize the polarization potentials. However, Tanae and collaborators!" have confirmed that this current was not sufficient to eliminate the effect of polarization presumably due to an asymmetric electrochemical situation developed at the electrode-nerve interface. Neuromuscular stimulation threshold values are classically lower at the cathodal terminal owing to membrane hypopolarization necessary for neural transmission. 10 The benefits of ABDC waveform could be related to stimulation at lower current levels with the cathodal pulse; the cathodal ABDC pulse was always related to the higher-amplitude muscle action potential and to a time-sharing process which effectively allows a resting period to be double that of the pulse interval (Fig. 6). This finding agrees with the conclusions of Mihailovic and Delgado!' that monopolar stimulation of a nerve with negative pulses invariably requires lower inten-

I I3

Diaphragm pacing

Number 1 July, 1977

150

125

"II! 100 75

50 0--0

e_

150

o

x--x pH

100

:J:

E

E

P02 PC0 2

50 L_-~~:b-'::::

.--_---1 ~j :::: I 7.30

o

:J: 0.

7.20 Control

Hypoxemia Hyperoxemia

Fig. S. Correlation of the amplitude of the muscle action potentials to hypoxemia and hyperoxia. Lower column shows arterial P02 , Pc0 2 , and pH values.

sines than when positive pulses are used and that stimulation with bidirectional pulses yields threshold values similar to those obtained with negative pulses. The results just presented imply that ABDC stimulation has the following advantages: It allows a rest to some nerve fibers during diaphragm pacing, and it neutralizes or divides the noxious effects of anodal current between the two electrodes. However, we are unable to show a significant advantage of ABDC stimulation of the phrenic nerve over UDC-biphasic in short-term experiments (Table I). It is of great interest that when the Flo. was increased so as to maintain the P0 2 over 100 mm. Hg, nerve fatigue during diaphragm pacing was prevented. In this study, when the dogs were ventilated with room air, the Pa02 level diminished to nearly 60 mm. Hg due in part (1) to the development of basilar atelectasis and to the structural feature of the animal's thorax that causes paradoxical movements during tetanic contractions of the unilateral diaphragm" 12. 13. 19. 21 and (2) to failure of stimulation of all of the phrenic nerve fibers. The phrenic nerve in the dog originates from the fourth to the seventh cervical vertebrae;" and it is possible that some nerve fibers, especially those from seventh cervi-

The Journal of Thoracic and Cardiovascular Surgery

Kaneyuki et al.

I I4

;;;;;;;;;;;;;;;

(~, )

I

I

~

.

..,

,.

-----

EI

r

I

u

0

I

..,

-

I

all ,

---1

'.11 I' lJ

I

-

H+~ ~

, l1

~I

n

H++

~

..- .... Ii'

;.1 rJ + I

ill ::II

~

II'

..

++~

! I

I,'J 1.1

Fig. 6. Evoked diaphragmatic action potentials delivered by ABDC monopolar stimulation. To evaluate the difference in threshold to anodal and cathodal current, we applied ABDC monopolar stimulation to the phrenie nerve. A, The muscle action potentials evoked by either anodal or cathodal current pulses were equal when applied current was enough to excite all the nerve fibers (amount of applied current was 500 /La in this case). B. When the current was reduced to 300 /La, anodal pulses failed to develop the muscle action potentials, while cathodal pulses still continued to excite the nerve. This implies that the anodal current threshold is much higher than cathodal.

cal vertebra, can be excluded from the nerve cuff placed around the nerve in the neck. The threshold of nerve fibers to electrical stimulation increases during anoxemia," and prolongation of the refractory period has been demonstrated as the result of tetanic contractions. 2. 6. 20 From measuring end-plate potentials of rat muscles, Krujevic and Miledi" reported that repetitive indirect stimuli led to a presynaptic block of nerve conduction. They inferred that the intramuscular portion of the nerve is easily subjected to a hypoxic state during tetanic contractions, and that even quite small changes in Pao. have a pronounced effect on nerve conduction. Stephens and Taylor!" demonstrated that, in maximal voluntary contractions, first the neuromuscular junction exhibited fatigue and then the contractile element, particularly when the blood supply was obstructed. Moreover, Bennett and Mcl.achlandemonstrated that repetitive electrical stimuli applied to the isolated sympathetic trunk of guinea pigs led to a decline in acetylcholine output. The acetylcholine declined to approximately 40 percent of the initial level during the first 5 to 15 minutes of stimulation and stayed at this level for a period of up to an hour. Acetylcholine is synthesized from choline and acetyl coenzyme A under the action of the enzyme choline acetyltransferase. For the biosynthesis of acetyl coenzyme A, adenosine triphosphate, as well as acetate (or citrate) and coenzyme A, is necessary. Therefore, if

repetitive stimuli were applied under hypoxic situations, acetylcholine resynthesis would decrease profoundly, and fatigue in the synapse (transmission failure) would develop more rapidly and to a greater extent. In our Series III experiments, the amplitude of the muscle action potentials decreased significantly under hypoxemia and was returned to normal by hyperoxygenation.

2

3

4

5

REFERENCES Bennett, M. R., and McLachlan, E. M.: An Electrophysiologieal Analysis of the Synthesis of Acetylcholine in Preganglionic Nerve Terminals, J. Physiol. 222: 669, 1972. Bugnard, L., and Hill, A. V.: The Effect of Frequency of Excitation on the Total Electric Response of Medullated , Nerve, 1. Physiol. 83: 396, 1935. Daggett, W. M., Shanahan, E. A., Kazemi, H., Morgan, A. P., and Austin, W. G.: Intracaval Electrophrenic Stimulation. II. Studies on Pulmonary Mechanics, Surface Tension, Urine Flow, and Bilateral Phrenic Nerve Stimulation, 1. THORAc. CARDIOVASC. SURG. 60: 98, 1970. Esher, D. J. W., Ashley, W., Ertag, W., Parker, B., Furman, S., and Robinson, G.: Clinical Control of Respiration by Transvenous Phrenic Pacing, Trans. Am. Soc. Artif. Intern. Organs 14: 192, 1968. Glenn, W. W. L., Holcomb, W. G., Gee, J. B., and Rath, R.: Central Hypoventilation: Long-Term Ven-

Volume 74 Number 1

Diaphragm pacing

115

July, 1977

6

7

8

9

10

II

12

13

tilatory Assistance by Radiofrequency Electrophrenic Respiration, Ann. Surg. 172: 755, 1970. Heinbecker, P.: Effect of Anoxemia, Carbon Dioxide and Lactic Acid on Electrical Phenomena of Myelinated Fibers of the Peripheral Nervous System, Am. J. Physio!. 89: 58, 1929. Holle, J., Moritz, E., Thoma, H., and Lishka, A.: Die Karusselstimulation, ein neue Methode zur elektrophrenischen Langzeitbeatmung, Wien. Klin. Wochenschr. 86: 23, 1974. Krujevic, K., and Miledi, R.: Presynaptic Failure of Neuromuscular Propagation in Rats, J. Physio!. 149: I, 1959. Lilly, J. c.. Hughes, J. R., Alvord, E. c., and Galkin, T. W.: Brief Noninjurious Electric Waveform for Stimulation of Brain, Science 121: 468,1955. Mendell, L. M., and Wall, P. D.: Presynaptic Hyperpolarization: A Role for Fine Afferent Fibers, J. Physiol. 172: 274, 1964. Mihailovic, L., and Delgado, J. M. R.: Electrical Stimulation of Monkey Brain With Various Frequencies and Pulse Durations, J. Neurophysio!. 19: 21, 1956. Minh, V. D., Jurihara, N., Friedman, P. 1., and Moser, K. M.: Reversal of the Pleural Pressure Gradient During Electrophrenic Stimulation, J. App!. Physio!. 37: 496, 1974. Minh, V. D., Friedman, P. J., Kurihara, N., and Moser, K. M.: Ipsilateral Transpulmonary Pressures During Unilateral Electrophrenic Respiration, J. App!. Physiol. 37: 505, 1974.

14 Rohlicek, V.: A Device for Polarity Alteration of Pulses for Biological Stimulation, Med. Electron. BioI. Eng. 2: 439, 1964. 15 Sato, G., Glenn, W. W. L., Holcomb, W. G., and Wuench, D.: Further Experience With Electrical Stimulation of the Phrenic Nerve: Electrically Induced Fatigue, Surgery 68: 817, 1970. 16 Stemmer, E. A., Crawford, D. W., List, 1. W., Heber, R. E., and Connolly, J. E.: Diaphragmatic Pacing in the Treatment of Hypoventilation Syndrome, J. THORAc. CARDIOVASC. SURG. 54: 649, 1967. 17 Stephens, J. A., and Taylor, A.: Fatigue of Maintained Voluntary Muscle Contraction in Man, 1. Physio!. 220: 1,1972. 18 Tanae, H., Holcomb, W. G., Yasuda, R., Hogan, J. F., and Glenn, W. W. L.: Electrical Nerve Fatigue: Advantages of an Alternating Bidirectional Waveform, J. Surg. Res. 15: 14, 1973. 19 Van Heeckeren, D. W., and Glenn, W. W. L.: Electrophrenic Respiration by Radiofrequency Induction, J. THORAc. CARDIOVASC. SURG. 52: 655, 1966. 20 Von Briicke, E. T., Early, M., and Forbes, A.: Fatigue and Refractoriness in Nerve, J. Neurophysiol. 4: 456, 1941. 21 Wanner, A., and Sackner, M. A.: Transvenous Phrenic Nerve Stimulation in Anesthetized Dog, 1. Appl. Physio!. 34: 483, 1973.