|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Departments of Medicine (Dr. Juan) and Pharmacology (Drs. Rubio and Morcillo), Faculty of Medicine, University of Valencia, Valencia, Spain; Service of Pneumology (Dr. Ramón), Service of Anesthesia (Dr. Valia), and Research Foundation (Dr. Cortijo), University General Hospital, Valencia, Spain; and University Hospital Aintree (Dr. Calverley), Liverpool, UK.
Correspondence to: Gustavo Juan, MD, PhD, Departamento de Medicina, Facultad de Medicina, Avda. Blasco Ibáñez, 15, E-46010 Valencia, Spain; e-mail: Gustavo.Juan{at}uv.es
| Abstract |
|---|
|
|
|---|
Design: Open-label clinical trial without a control group.
Setting: University hospital.
Patients: The inclusion criteria were a diagnosis of emphysema, basal dyspnea index (Mahler scale)
3, FEV1
35%, and no indication for pneumoreduction or lung transplantation surgery.
Interventions: An epidural catheter was inserted at the thoracic level connected to a perfusion pump for administering methadone (6 mg/24 h). Assessments were made at baseline, 1 week, and 1 month after catheter insertion.
Measurements: Pulmonary function tests were performed, and determinations were made of arterial blood gas levels, respiratory control data, dyspnea quantification by Mahler transitional dyspnea index (TDI), and the Borg scale change with inspiratory resistive loading, 6-min walk (6MW) distance, and health-related quality of life using the Chronic Respiratory Disease Questionnaire.
Results: Of the nine patients studied, infection and catheter migration lead to suspension of treatment before the end of the study in two cases. A significant improvement in dyspnea occurred by 1 week: mean TDI, 3.77 (SD, 1.98) [p < 0.01]. After 1 month of treatment, there were significant improvements in the 6MW distance (mean, 35.33 m; SD, 17.03; p < 0.05), health-related quality of life (mean, 1.63; SD, 0.36; p < 0.05), and dyspnea (mean TDI, 5.33; SD, 2.16; p < 0.05). In addition, after 1 month, PaCO2 fell by 6.67 mm Hg (p < 0.05) and rapid shallow breathing index decreased from 38 to 27 (p < 0.05). These changes occurred without any alteration in the subjects ability to perceive or respond to inspiratory loading.
Conclusion: Epidural methadone perfusion at chest level can effectively palliate dyspnea and improve exercise capacity and quality of life in patients with advanced emphysema, without deterioration in respiratory control or lung function. These data suggest that modulation of spinal cord afferent signaling is an appropriate novel target for dyspnea control in chronic respiratory disease.
Key Words: dyspnea emphysema epidural methadone
| Introduction |
|---|
|
|
|---|
The sensation of dyspnea in spontaneously breathing patients with obstructive lung disease involves inputs from the respiratory muscles. Muscle spindles are abundant in the intercostal muscles, and afferent activity from them is involved in both spinal and supraspinal reflexes.9 These sensory receptors located in the chest wall muscles participate in the sensation of dyspnea, the afferent stimuli traveling in the spinal cord to the brain.10 In patients with pulmonary hyperinflation, one mechanism of dyspnea appears to be the sensation of inspiratory effort induced by stimuli originating from the accessory respiratory muscles that reach the brain through the posterior horns of the spinal cord.11 Despite these physiologic considerations, there have been no attempts to date to treat dyspnea at spinal cord level by modifying neural traffic in these afferent pathways.1 However, we have noted improvement in dyspnea intensity in the immediate postoperative period among patients subjected to lung volume reduction surgery receiving thoracic epidural perfusion of local anesthetic and opiates to control postoperative pain. The present study investigates whether epidural methadone perfusion at the thoracic level can mitigate dyspnea in patients with advanced emphysema, without causing associated respiratory depression.
| Materials and Methods |
|---|
|
|
|---|
35%); severe pulmonary hyperinflation seen on chest radiography12; basal dyspnea index (BDI) [Mahler scale]
3,13 and no indication for pneumoreduction or lung transplantation because of age and comorbidity. Patients were clinically stable for at least 3 months, and patients with vertebral fractures and/or thoracic radicular pain were excluded. The protocol was approved by the clinical research ethics committee of the Valencia University General Hospital (Valencia, Spain) and the General Subdirectorate for Drug Evaluation of the Spanish Ministry of Health and Consumer Affairs (protocol No. 98/510). Informed consent to participation in the study was obtained from each patient.
Study Design
An open-label clinical trial without a control group was used, with a planned treatment duration of 1 month. Measurements were made before and 1 month after treatment. Arterial blood gases and the dyspnea scaling were also analyzed after 1 week of treatment. All patients were clinically monitored every week in the outpatient clinic of the Pain Treatment Unit.
After selecting the patient and obtaining the basal measurements, an epidural catheter was inserted at the thoracic level connected to a patient-controlled analgesia perfusion pump (model 5800; Pharmacia Deltec; St. Paul, MN) for administering methadone. The procedure was performed by the Pain Treatment Unit with their usual technique for pain control by thoracic epidural analgesia. After insertion in the epidural space at the T4-T5 level, the catheter was tunneled in the subcutaneous tissue to the anterior abdominal wall, where it was secured using skin suture and connected to the perfusion pump. The patient remained hospitalized for 24 h and was instructed how to stop the pump and protect the connection. The domiciliary treatment unit visited the patient every week to reload the pump. The initial perfusion rate of methadone was 6 mg/24 h, followed by modification according to the dyspnea response or undesirable effects (mainly constipation and drowsiness).
Measurements
The followings parameters were recorded: (1) Pulmonary function testing, including airway resistance measurement (Compact Transfer; Erich Jaeger; Würzburg, Germany), according to the American Thoracic Society.14 (2) Arterial blood gas levels (Rapid 850; Chiron Diagnostic Corporation; East Walpole, MA). (3) Breathing pattern, determined with the patient comfortably seated and after stabilizing breathing; flow was measured with a heated pneumotachograph (Fleisch No. 2; Master Lab, Erich Jaeger). Automatic measurements were obtained of tidal volume (VT), duty cycle (TI/TTOT), and respiratory frequency (f). We also calculated the rapid shallow breathing index (RSBI) [f/VT].15 (4) Ventilatory effort during inspiration was determined by P0.1 after the start of inspiration.16 Closure of the inspiratory line during expiration was manually controlled, and the valve automatically opened after the first 100 ms of occluded inspiration. Five repeated measurements of P0.1 were averaged. P0.1 was measured without resistance and in response to flow-resistive loading during inspiration. (5) P0.1 response to flow-resistive loading breathing during inspiration was measured with a device similar to that described by Wiley and Zechman.17 The device consisted of a glass cylinder with various parallel outlets of different calibers. These outlets were in turn equipped with valves controlling aperture. This system of resistances was connected to the inspiratory inlet of a two-way breathing valve and then to the flowmeter (included in the Master-Lab system). Pressure drop across the system was measured as the difference in pressure between the system inlet (assumed to be atmospheric) and the midstream pressure at the airway opening. Airway opening pressure was measured with a differential pressure transducer (Validyne MP-45; Validyne Engineering; Northridge, CA). The pressure drop across each resistance was linear to at least 1 L/s. The apparatus dead space was 380 mL, and the background resistance of the measuring system was 0.58 cm H2O/L/s. The subjects inhaled by means of the two-way valve through the first side port for 3 min, and the inspiratory resistance was then added (range, 1.7 to 10.7 cm H2O/L/s). P0.1 was measured as described above with three different added resistances. (6) Inspiratory load increment perception capacity (Weber fraction) was studied with the same device as above. The subjects inhaled by means of the two-way valve through the first side port for 3 min. Next, the inspiratory resistance was altered in random order for one inhalation in every three to six breaths. The subject was instructed to raise a hand to indicate whether or not any increase in resistance was noticed. The operator sat behind the subject and manipulated the valves of the resistance device so that the subject could not use tactile sensations as clues to discriminate changes in resistance. Five different loads were chosen for the experiment, which took place over a period of 8 to 10 min. The probability of detecting a given load was calculated as the ratio of the number of correct identifications of that load to its total number of presentations. For each subject, the probability of detecting given changes in resistance was plotted against the actual changes in resistance. The best fit line through these points was determined by least-squares regression analysis, and the load change corresponding to 0.5 probability was calculated from the regression equation. This calculated load change was expressed as a fraction of the sum of airway resistance and background apparatus resistance. The ratio of the change in stimulus intensity to background stimulus intensity, known as the Weber fraction, is one way of characterizing the perception of various physical sensations, including those arising in the respiratory system.18 (7) Dyspnea quantification via Mahler scale using the interview/observation by the provider method,13 with BDI ranges between 0 and 12 and transitional dyspnea index (TDI) between 9 and + 9. (8) The Borg scale variation in response to increasing loads.19 To this effect, the patient breathed through the above-mentioned resistance circuit, with the application of two increasing resistances (range, 5.6 to 10.7) and was asked to score breathlessness on a one-meter vertical Borg category scale with descriptors, graded from 0 (none) to 10 (maximum), after breathing through the different resistances for 1 min. This parameter was calculated by dividing the Borg scale increment by the resistance increase causing it. (9) Quality of life was assessed by means of the Chronic Respiratory Diseases Questionnaire developed by Guyatt et al20 and translated into Spanish by Güell et al.21 This is a 20-item questionnaire evaluating four dimensions of illness: dyspnea, fatigue, emotional function, and mastery. (10) Six-minute walk (6MW) test.22
Statistical Analysis
The primary outcome measure was the TDI. Secondary outcome measure was the quality of life and 6MW. The parameters were analyzed comparing baseline and posttreatment values. Since the sample was small in size, the Wilcoxon signed-rank test was used; p < 0.05 was considered statistically significant. Data are presented as means (SD) and range. There were three patients who failed to complete the study. These patients were excluded from all comparisons with the exception of TDI after 1 week of treatment.
| Results |
|---|
|
|
|---|
All patients reported significantly improved dyspnea after 1 week of treatment (Fig 1 ). Analysis of the dyspnea subscales also yielded significant improvement of all of them after 1 week of treatment (Table 1 ). After 1 month, there was also a statistically significant improvement of dyspnea in the six patients who continued treatment (Fig 1), with similar improvements in the dyspnea subscales (Table 1).
|
|
|
|
|
As regards to the systemic adverse effects observed, four patients had constipation that was corrected with laxatives, while one subject presented drowsiness that was corrected by reducing the methadone dose to 5 mg/24 h. Only one patient (case 5) had an exacerbation of COPD while using the perfusion system. This problem occurred after 1 month of treatment and consisted of fever, yellowish expectoration, and increased dyspnea.
| Discussion |
|---|
|
|
|---|
Clinically important improvements in dyspnea as assessed by the TDI were evident by one week and were maintained throughout the study, changes which were substantially larger than those seen with bronchodilator treatment.2324 Although the change in quality of life did not exceed the clinically significant differences of 2.0 for the global Chronic Respiratory Disease Questionnaire score, the changes seen were still statistically significant and were > 0.5 in the dyspnea domain of this questionnaire, the area where most impact would be registered over a short time. Likewise, changes in 6MW distance were smaller than the proposed minimal clinically important difference of 54 m25 but were still of the order seen after lung volume reduction surgery,26 a procedure for which these patients were not eligible. The changes seen were better than those reported with sustained-release morphine, which did not change quality of life or walking distance relative to placebo but was associated with significant adverse effects.7 Abernethy et al27 reported significant improvements in dyspnea scores with sustained-released oral morphine in patients with refractory breathlessness, but the treatment duration in this study was only 4 days and there was no assessment of the impact on other physiologic variables.
There was evidence of individual variation in response between patients. Patient 5 discontinued treatment immediately on completion of the protocol due to lack of efficacy, while patient 1 continued the therapy for 18 months. This type of variation has been noted previously, although we could not identify any obvious differences between the patients in our study to explain it.728 Although some patients experienced systemic side effects with methadone, these were easily controlled by dose reduction and the use of laxatives. Much more troublesome were problems with catheter site sepsis (two patients) and migration of the catheter (four patients), which led to treatment stopping at 2 months and 4 months in our remaining patients. This reflects our choice of an "open" rather than a "closed" opiate delivery system. We elected to use the open system, as we were concerned about the possibility of respiratory depression and wished to discontinue treatment rapidly if this occurred. In practice, this was not a problem, and in future studies a closed delivery system with its reduced risks of sepsis would appear preferable.
Our study included a detailed assessment of the patients ability to perceive and acquire a ventilatory response to inspiratory loading. Although inputs from muscle spindles have long been considered important in resistive load detection in man,29 we found no difference in the overall ability of our patients to scale the magnitude of inspiratory sensations while being treated with epidural methadone. The Weber fraction in our patients was lower than that in healthy control subjects and patients with mild-to-moderate COPD, a finding noted previously in individuals with significantly increased respiratory system impedance.29 We did not see any impairment in the mouth occlusion pressure (P0.1) response to inspiratory loading as judged by the slope of the P0.1-load relationship. There was a significant reduction in f, but no significant change in VT and P0.1 at rest. This led to a fall in the RSBI, a measure proposed as a guide to successful weaning in patients receiving mechanical ventilation,30 and a marker of the respiratory system response to increased mechanical loading. The reductions seen are compatible with a reduced afferent stimulus to ventilation. However, the changes we saw were not accompanied by any deterioration in gas exchange that usually accompanies a reduction in ventilatory drive in the absence of a change in lung mechanics.
The overall pattern of response to epidural methadone is compatible with a change in the operating lung volumes in these hyperinflated COPD patients, resembling the effects of acute bronchodilator treatment.31 We did not see any change in FEV1 and FRC in our population. However, we may speculate that reduction in the reflex activation of the intercostal muscles due to an interruption in the local spinal reflexes that are mediated at the level of the posterior horn cells could have increased the compliance of the chest wall muscles and allowed the end-expiratory lung volume to fall. Alternatively, the fall in f may have been sufficient to promote better lung emptying and a fall in operating lung volume. The absence of any change in the ability to perceive loading and the fall in arterial carbon dioxide tension favors this mechanical explanation, but future studies will be needed in a larger number of individuals to determine whether this hypothesis is correct. One positive conclusion from our study is that sustained use of epidural methadone did not produce significant impairments in ventilatory control in these patients with advanced COPD at least while being awake.
Our study was limited by the need to identify clinically stable patients who were already receiving optimal treatment but were still symptomatic from their COPD. We avoided individuals with resting hypercapnia or those with a cough productive of significant volumes of sputum, and so our findings should be viewed with caution. We chose methadone as the treatment opioid because of its lipid solubility, which fixes the drug within the epidural space close to the posterior medullary horn cells with little migration either cephalad or caudad.3233 All our patients were ambulatory, highly motivated, with good family support. Also, they were familiar with corridor walking testing before inclusion in the study having undergone pulmonary rehabilitation previously and all were clinically stable for several months before treatment began. We did not include a placebo limb in the present study, as we wished to establish whether treatment was effective and safe over an extended period before approaching the ethically difficult issue of a "placebo" epidural catheter placement. Although a placebo effect could contribute to our results, the magnitude and duration of the effect make this less likely.
In summary, the current study has identified a potentially substantial effect on breathlessness and patient well-being if afferent neural traffic is modulated by the use of an epidural opiate. Since this is pilot study, recommendations for clinical practice are premature, but this treatment was acceptable in the short term (over 4 weeks), and its lack of effect on ventilatory control and respiratory perception raises both clinical and physiologic issues about the modulation of breathlessness in COPD, which will be addressed in future randomized controlled studies.
| Acknowledgements |
|---|
| Footnotes |
|---|
This study was supported in part by grants from Laboratorios (Dr. Esteve; Barcelona, Spain) and University General Hospital (Valencia, Spain).
Received for publication November 25, 2004. Accepted for publication May 30, 2005.
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |