|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Department of Physiology and Biophysics (Drs. DiMarco and Kowalski), and the Department of Biomedical Engineering (Dr. Mortimer), Case Western Reserve University, Cleveland, OH; and the Department of Surgery (Dr. Onders and Mr. Ignagni), University Hospitals of Cleveland, Cleveland, OH.
Correspondence to: Anthony F. DiMarco, MD, FCCP, Department of Physiology and Biophysics, Case Western Reserve University, MetroHealth Medical Center, Rammelkamp Center for Education & Research, 2500 MetroHealth Dr, Cleveland, OH 44109-1998; e-mail: afd3{at}cwru.edu
| Abstract |
|---|
|
|
|---|
Objective: To assess the feasibility of laparascopic placement of IM diaphragm electrodes to achieve long-term ventilatory support in ventilator-dependent tetraplegic subjects.
Design, setting, and participants: Two IM diaphragm electrodes were placed laparoscopically in each hemidiaphragm in five subjects with ventilator-dependent tetraplegia. Studies were performed either on an outpatient basis or with a single overnight hospitalization. Ventilator-dependent tetraplegic subjects were identified in whom bilateral phrenic nerve function was present, as determined by phrenic nerve conduction studies. Following electrode placement, subjects participated in a conditioning program to improve the strength and endurance of the diaphragm over a period of 15 to 25 weeks. The duration of the study was variable depending on the time necessary to determine the maximum duration that individuals could be maintained without mechanical ventilation support.
Main outcome measures: Magnitude of inspired volume generation and duration of ventilatory support with bilateral diaphragm pacing alone.
Results: In four of the five subjects studied, initial bilateral diaphragm stimulation resulted in inspired volumes between 430 and 1,060 mL. Reconditioning of the diaphragm over several weeks resulted in substantial increases in inspired volumes to 1,100 to 1,240 mL. These subjects were comfortably maintained without mechanical ventilatory support for prolonged time periods by diaphragm pacing, by full-time ventilatory support in three subjects, and 20 h per day, in the fourth subject. No response to stimulation was observed in one subject, most likely secondary to denervation atrophy.
Conclusions: Diaphragm pacing in ventilator-dependent tetraplegic subjects can be successfully achieved via laparascopic placement of IM electrodes.
Key Words: diaphragm pacing laparoscopy spinal cord injury
| Introduction |
|---|
|
|
|---|
In this report, we describe a new method by which the phrenic nerves can be activated to provide long-term ventilatory support (ie, through laparoscopic placement of IM diaphragm electrodes). This technique has the potential to provide a less invasive and more convenient method of phrenic nerve stimulation with a lower cost in ventilator-dependent tetraplegic subjects.
| Materials and Methods |
|---|
|
|
|---|
The clinical data of the subjects who participated in this study are shown in Table 1 . All subjects were male, had experienced traumatic injury to the upper cervical spinal cord, and had required full-time ventilatory support since the time of the injury. The elapsed time since injury ranged from 1 to 8 years. At the time of recruitment into the study, the spontaneous vital capacities ranged between 280 and 890 mL. Each of the subjects was hyperventilated while receiving mechanical ventilation. In each case, ventilator adjustments were attempted to restore PCO2 to more normal values per subject tolerance. The ventilator settings and arterial blood gas measurements of each of the subjects are shown in Table 2 . Each of the subjects recruited for this study could tolerate at least 20 min without mechanical ventilation, but none could comfortably tolerate spontaneous breathing for > 1 to 2 h. Chest radiograph findings were normal in two subjects but demonstrated basilar atelectasis in the others. Each subject had normal bilateral phrenic nerve function, as determined by phrenic nerve conduction studies.
|
|
|
Using the coarse mapping technique, an initial permanent electrode18 was placed using a unique tool to implant the IM electrodes14 in each hemidiaphragm. This was followed by the placement of a second electrode in a location that also provided the recruitment of a major portion of the muscle, usually within 1 to 2 cm of the initial electrode location. Repeat visual assessment and determination of intraabdominal pressure was made following each electrode insertion. Wires from the electrodes were brought out through the epigastric port and tunneled subcutaneously to the right or left subclavicular region where they exited the chest wall, as described previously.1416
The monitoring of cardiac rhythm during maximum diaphragm stimulation was observed in the operating room in each subject. No arrhythmias were observed during electrical stimulation.
Muscle Reconditioning
Approximately 2 weeks following the placement of the diaphragm electrodes, inspired volume generation was evaluated with each of the four electrodes individually. Relationships between stimulus amplitude (pulse width, 0.1 ms) and inspired volume at constant stimulus frequency, and between stimulus frequency and inspired volume at constant stimulus amplitude also were determined. In each instance, the maximum or near-maximum amplitude was required to achieve the maximum inspired volume generation. Consequently, 24 or 25 mA (depending on the stimulator employed) was applied in each instance during long-term stimulation. Subsequently, the interaction between the two electrodes implanted in each hemidiaphragm was assessed. During the early portion of the reconditioning period, the stimulation of both electrodes within each hemidiaphragm generally resulted in greater inspired volumes compared to the stimulation of one electrode alone. In some instances, increases in pulse width to 0.150 ms resulted in greater inspired volumes, and this higher pulse width was employed. Pulse widths above this value were not applied during long-term stimulation due to safety concerns related to electrode corrosion. In each subject, the initial applied stimulus frequency was 20 Hz. During the conditioning period, attempts were made to gradually reduce the stimulus frequency to the lowest value that resulted in adequate ventilation. Inspiratory time (ie, train duration) was set at 1.1 s, and the respiratory rate was set between 10 and 12 breaths/min.
Initially, pacing was provided for 5 to 10 min each hour for 5 to 6 h per day and was gradually increased, as tolerated. When continuous pacing for 6 to 8 h was achieved, the number of hours per day of pacing was increased over time.
During muscle reconditioning, a multifunction monitor (model N-100; Nellcor; Hayward, CA) was used to monitor arterial oxygenation and end-tidal PCO2 via a finger probe at the tracheal opening. Arterial blood gases were sampled intermittently.
Tidal volume was monitored by the electrical integration of the flow signal at the tracheal opening using a pneumotachograph (model 3700; Hans Rudolph; Kansas City, MO) and also with a respirometer (Wright respirometer, model Mark 14; Ferraris Medical Ltd; Enfield, UK). Airway pressure was monitored intermittently during airway occlusion for a single breath using a differential pressure transducer (Validyne MP; Validyne). Measurements were recorded on an eight-channel strip chart recorder (model DASH8; Astro-Med; Warwick, RI).
In subject 3, no appreciable inspired volume could be achieved with diaphragm stimulation. This represented either a false-positive phrenic nerve conduction study or phrenic nerves that were inaccessible to IM diaphragm stimulation. Under direct observation, the appearance of the diaphragm in this subject appeared quite thin compared to the other subjects, suggesting denervation atrophy. The following results therefore represent data on the remaining four subjects.
With the exception of one subject in which electrodes were positioned several centimeters apart in the right hemidiaphragm, both IM diaphragm electrodes in each hemidiaphragm were placed within 1 to 2 cm of each other. Moreover, the stimulation of each of these electrodes appeared to result in the contraction of both the anterior and posterior portions of the hemidiaphragm. Following reconditioning in three subjects, the stimulation of two electrodes resulted in greater inspired volumes than either one alone, and therefore artificial ventilation was maintained with the stimulation of all four electrodes. In one subject, the stimulation of one electrode within each hemidiaphragm provided inspired volumes that were similar to that achieved with both electrodes, and, therefore, artificial ventilation was provided with one electrode in each hemidiaphragm.
| Results |
|---|
|
|
|---|
|
|
The relationship between stimulus frequency and inspired volume generation is shown for one subject in Figure 4 . During the initial stimulation, there were progressive increases in inspired volume generation with increasing stimulus frequency under each condition. Over the course of the reconditioning period, there were progressive increases in inspired volume generation under each condition at each stimulation frequency, such that each curve became shifted upward and to the left. Similar responses were observed in the other subjects.
|
|
|
| Discussion |
|---|
|
|
|---|
Comparison With Conventional Diaphragm Pacing
Based on previously published reports,69 the magnitude of inspired volumes generated by phrenic nerve stimulation via IM electrodes is comparable to that achieved with conventional PNP. The maximum airway and transdiaphragmatic pressures measured in one subject were also similar to values previously reported for conventional PNP.9 Importantly, end-tidal PCO2 measurements were maintained in the low normal range with long-term pacing parameters in each of the subjects. Taken together, these data indicate that the activation of the diaphragm by IM electrodes results in a comparable level of diaphragm activation when compared to direct phrenic nerve stimulation.
The subjects in the present study also reported clinical benefits of diaphragm pacing that were similar to those described with conventional PNP, including improved sense of smell, more comfortable breathing, improved speech, greater independence, and elimination of the fear of disconnection from the mechanical ventilator.716
Conventional PNP generally requires a thoracotomy,6 which is a major surgical procedure with associated risks, a required in-patient hospital stay, and high cost. In addition, the manipulation of the phrenic nerves, a procedure that carries some risk of nerve injury, is required. These disadvantages have limited the number of patients undergoing this procedure and present a significant obstacle to those patients undergoing this procedure. Laparoscopy, in contrast, is a less invasive procedure with obvious cosmetic benefits and is often performed on outpatient basis. Due to the investigational nature of the laparoscopic placement of diaphragm electrodes, most subjects in the present study were admitted to the hospital overnight for observation. With greater clinical experience, however, it is likely that this procedure also can be performed on an outpatient basis, as was done following the second surgical procedure in the initial subject.
Some studies19 have demonstrated that phrenic nerve electrodes can also be placed thoracoscopically in children for management of congenital central hypoventilation syndrome. While obviating the need for a thoracotomy, this technique still requires the manipulation of the phrenic nerves and is technically quite challenging.19 It is also not clear whether this procedure can be performed successfully in adults. Since laparascopic IM electrode placement, as described in the present study, does not involve direct contact with the phrenic nerve, the risk of nerve injury is virtually eliminated.
Diaphragm Mapping and Determination of Electrode Insertion Sites
Uniform diaphragm activation requires the placement of IM electrodes near the phrenic nerve motor points (ie, the area of the muscle contained within the space defined by the entrance points of the phrenic nerves into the diaphragm).1015 An important finding of the present study was that the coarse mapping procedure, performed with a suction electrode in the general vicinity of the motor point (based on previous anatomic evaluations), and the visualization of the contraction of both anterior and posterior portions of the diaphragm were deemed sufficient to determine the optimal sites for permanent electrode placement. Stimulation in these same regions also resulted in the largest increases in intraabdominal pressure. Therefore, performance of the previously described fine mapping technique involving a complex mathematical analysis was discontinued.
It is important to note that the measurements of intraabdominal pressure during laparoscopy have significant limitations. During the course of these studies, it was often observed that repeat stimulation at the same site on the diaphragmatic surface often resulted in widely varying pressure levels. Small variations in the degree of abdominal insufflation, which can occur during the course of the procedure, most likely altered diaphragm length, which is an important determinant of force generation.20
Interaction Between Left and Right Hemidiaphragm Contraction
In each subject, inspired volumes resulting from bilateral diaphragm contraction were substantially greater than the arithmetic sum of left and right hemidiaphragm contraction alone, indicating a synergistic effect. This occurred both during maximum diaphragm stimulation and also with lower stimulus frequency, long-term pacing parameters. This same effect also was observed qualitatively by Glenn et al6 with conventional PNP. The mechanism of this phenomenon most likely relates to the retraction of the flaccid hemidiaphragm into the thorax as a result of the negative pressure generated by the contraction of the opposite hemidiaphragm. The greater the compliance of the noncontracting hemidiaphragm, the greater the degree of synergism. The clinical correlate of this phenomenon includes patients with unilateral diaphragm paralysis who experience dyspnea that is related to the retraction of the paralyzed diaphragm into the thorax and derive significant benefit from diaphragm plication.21
Side Effects
While much less invasive than a thoracotomy, laparoscopic surgery also has some associated risks.212223 In addition to complications common to all surgical procedures, laparoscopy may be associated with the development of pneumothorax and subcutaneous emphysema. Pneumothorax is thought to develop from the movement of gas from the peritoneal cavity to the pleural space via the mediastinum through tissue planes or congenital diaphragmatic defects and usually resolves spontaneously.23 One of the subjects in the present study had a pneumothorax that may have developed by one of the above-mentioned mechanisms or by air tracking along openings in the diaphragmatic surface created by electrode placement. Although this subject was asymptomatic and had normal levels of oxygenation, the pneumothorax was evacuated by chest tube drainage, since the pneumothorax was slow to resolve and the subject was ventilator-dependent.
One subject developed right shoulder pain during the maximum stimulation of a single electrode. This most likely occurred as a consequence of the stimulation of phrenic nerve afferents. His symptoms were completely alleviated by a modest reduction in stimulus current. Another subject developed hay fever symptoms, which were a common problem prior to his injury but had not been present when he was receiving mechanical ventilation. The restoration of nasal airflow during diaphragmatic pacing evidently resulted in the recurrence of symptoms. One subject had intermittent aspiration of food during meals, which most likely was related to the large negative airway pressure generated during contraction of the diaphragm. This problem was eliminated by use of a Passy-Muir valve during meals. This device served to reduced the magnitude of negative pressure in the oropharynx. Other potential long-term effects of prolonged pacing include electrode dislodgement and electrode breakage, either of which could result in inadequate inspired volume generation.
Future Developments
The current system requires electrode wires that exit the skin and are connected to an external power generator. These wires carry a small risk of infection and represent a significant inconvenience. Current investigations are underway to develop a totally implantable system such as the radiofrequency-powered pulse generators, which are currently utilized with conventional PNP and combined intercostal and diaphragm pacing systems.
Combined intercostal and diaphragm pacing systems are also successful in maintaining long-term ventilatory support in patients with only a single functional phrenic nerve.24 Rather than direct PNP, it is possible that combined intercostal and unilateral phrenic nerve stimulation also can be achieved by IM diaphragm pacing, eliminating the need for a thoracotomy and for manipulation of the phrenic nerves in these subjects as well.
| Footnotes |
|---|
Abbreviation: PNP = phrenic nerve pacing
Received for publication March 29, 2004. Accepted for publication August 25, 2004.
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |