INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Lung volume reduction surgery (LVRS) has been advocated in select patients with severe, nonbullous, diffuse emphysema (1). Cooper and colleagues (1) reported the results of this surgery in 20 patients with severe chronic obstructive pulmonary disease (COPD) (mean FEV₁, 0.77 L) and hyperinflation (TLC, 8.5 L). These authors found that 6 mo after surgery mean FEV₁ increased by 82% (0.77 to 1.4 L), arterial PO₂ increased from 64 to 72 mm Hg, and 6-min walk distance increased by 33%. Other investigators have confirmed the beneficial effects of bilateral volume reduction (2) and, to a lesser extent, unilateral volume reduction (3), on improving spirometry and gas exchange, reducing residual volume, and enhancing exercise performance (8).

Several investigators have reported conflicting results on the effect of LVRS on respiratory muscle function (8). Although some have shown that LVRS increases diaphragm strength (12), and alters respiratory muscle recruitment during exercise (10), others have shown no significant effect (11). Moreover, the effect of LVRS versus pulmonary rehabilitation on lung function, exercise capacity, or respiratory muscle strength has not been carefully delineated in any of these earlier reports (1).

Herein, we examine the effects of LVRS versus pulmonary rehabilitation on diaphragm strength in patients with severe, nonbullous diffuse emphysema.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patient Selection

Thirty-seven patients were evaluated for LVRS. Twenty-four patients underwent 8 wk of intensive outpatient pulmonary rehabilitation. A total of 20 patients, including seven of the previous 24 patients who underwent 8 wk of pulmonary rehabilitation, met LVRS criteria shown in Table 1. After being considered an acceptable candidate, patients had the protocol and LVRS explained to them in detail (Figure 1). The study was approved by our Institutional Review Board for Human Research (Temple University School of Medicine, Philadelphia, PA).

View larger version (24K): [in this window] [in a new window]   Figure 1.   Outline of protocol for testing before and after pulmonary rehabilitation and lung volume reduction surgery.

Physiologic Measurements

Before and 8 wk after rehabilitation, 24 subjects performed a variety of pulmonary function studies (e.g., spirometry before and after bronchodilator administration, lung volumes measured by helium dilution and body plethysmography, diffusion capacity), incremental symptom-limited maximum exercise test, 6-min walk test, measurement of volitional maximum static inspiratory and expiratory mouth pressures, and transdiaphragmatic pressures measured during maximum static inspiratory efforts and during bilateral supramaximal twitch electrophrenic stimulation. Thirteen patients had similar physiologic measurements before and 3 mo after LVRS. In addition, seven patients had the preceding measurements made before and after 8 wk of outpatient pulmonary rehabilitation and then 3 mo post-LVRS.

Pulmonary function testing. Pulmonary function testing was performed (13) with a System 6200 Autobox DL plethysmograph (SensorMedics Corp., Yorba Linda, CA), using American Thoracic Society guidelines. Vital capacity (FVC), forced expiratory volume in 1 s (FEV₁), and FEV₁/FVC were measured. Airway resistance (Raw) and thoracic gas volumes were measured in a body plethysmograph. Functional residual capacity (FRC) was also measured by helium dilution technique. FRC_{trapped gas} is the difference between FRC measurements made by body plethysmography and helium dilution.

Diffusion capacity for carbon monoxide (DL_CO) was measured by the single-breath technique. It is reported as the carbon monoxide-diffusing capacity per unit of alveolar volume (DL_CO /VA).

Maximum voluntary ventilation (MVV) was measured with the subjects seated upright and instructed to breathe maximally in a deep and rapid manner for a sustained period of 12 s. At least two trials were performed, with the highest value reported.

All data reported are postbronchodilator results presented in absolute numbers and as a percentage of normal predicted values (14).

Exercise testing. Patients underwent incremental, maximal treadmill exercise (Precor, 9.4 sp; Precor, Bothell, WA) starting at 0% incline and 1 mph. Incline increased by 3% and speed by 1.5 mph every 3 min until symptom-limited maximum. Oxygen uptake (O₂), carbon dioxide production (VCO₂ ), minute ventilation (E), tidal volume (VT), and respiratory rate (f_b) were also recorded by a metabolic cart (SensorMedics 2900). Transcutaneous oxygen saturation (Nellcor N-200, Nellcor, Chula Vista, CA) and multiple-lead EKG (ECG Horizon; SensorMedics) were continuously recorded. Patients requiring oxygen with exercise used the same level of inspired oxygen at each exercise evaluation. At each exercise test conclusion, dyspnea was rated using a visual analog scale from 0 (no breathlessness) to 10 (severe breathlessness) (15). On a different day, a walk distance test was measured, with subjects encouraged to ambulate in a 100-ft corridor their maximum distance in 6 min (16).

Respiratory muscle pressures. Mouth pressures were measured using a previously reported technique (17). PI_max was measured from functional residual capacity (FRC), whereas PE_max was measured near total lung capacity (TLC). Tests were repeated until at least three attempts varied less than 5%, the average of three tests were then reported.

Transdiaphragmatic pressure measurement. Following topical anesthesia (4% lidocaine), two thin-walled balloon-tipped catheters were placed via the nares, one into the lower esophagus (esophageal pressure, Pes) and the other into the stomach (gastric pressure, Pga) (18). Both catheters were connected to pressure transducers (range, ± 100 cm H₂O; Validyne, Northridge, CA). Transdiaphragmatic pressure (Pdi) was displayed as the electronic subtraction of Pes from Pga. A plaster cast was then placed over the anterior abdomen to minimize outward displacement of the abdominal wall during electrophrenic stimulation. Voluntary Pdi was measured against an occluded airway using a combined expulsive-Mueller maneuver (Pdi_{max combined}) during visual oscilloscopic feedback with subjects seated upright in a high-backed chair (19). In addition, maximum transdiaphragmatic pressures were measured during a maximum sniff maneuver (Pdi_{max sniff}) (20) and uncoached maximum inspiratory effort (Pdi_max). The average of three values of Pdi_max during each separate maneuver, all within 5% of each other, are reported.

Electrophrenic stimulation. Compound diaphragm action potentials (CDAPs) were measured bilaterally by a pair of 3-mm EMG surface electrodes placed 2 mm apart in the seventh intercostal space in the anterior axillary line. The site for optimum phrenic nerve stimulation was located by using anatomic landmarks (21). Stimulus voltage was incrementally increased until there was no further increase in CDAP amplitude. Once maximum stimulus voltage was achieved, it was further increased by 20% to ensure supramaximal diaphragm activation.

A modified neck brace housing the right and left phrenic nerve stimulus probes was used to ensure consistency in phrenic nerve stimulation. The phrenic nerves were stimulated transcutaneously (S88 stimulator; Grass, Quincy, MA) with 100-140 V (approximately 30 mA), 0.1 ms in duration, to produce diaphragm twitch pressures (Pdi_twitch). There was an approximately 20 to 30 min interval between the end of maximum voluntary static maneuvers and the onset of Pdi_twitch testing.

Pdi_twitch at FRC. With the subject seated with upright posture, bilateral phrenic nerve stimulation was delivered at functional residual capacity (FRC) after closure of an in-line three-way valve at end expiration. Pes was continuously monitored to ensure that end-expiratory lung volume had returned to a consistent baseline before valve closure. Six to 15 consecutive twitches (each twitch separated by at least a 3-s pause) were delivered at FRC. Twitches analyzed were those considered acceptable after ensuring that the FRC was at baseline and twitch morphology was consistent. Three values, all within 5%, were averaged and reported as Pdi_twitch.

Surgical Technique

Lung resections were performed via median sternotomy and bilateral stapling (n = 19 patients) or unilateral thorascopy with stapling (1 patient). The goal for resection was to remove 20-40% of the volume of each lung, guided by the visual judgment of the same surgeon. High- resolution computed tomography of the chest and quantitative ventilation-perfusion scans were used preoperatively to target resection of lung regions with the worst emphysema, poorest perfusion, and greatest gas trapping.

Pulmonary Rehabilitation

Pulmonary rehabilitation consisted of twenty-four 2-h sessions over an 8-wk period. Rehabilitation included education, physical and respiratory care instruction; psychosocial support; and supervised exercise training by an exercise physiologist. After baseline exercise tests, all subjects received an individualized exercise prescription based on symptom-limited maximum. Patients used a motor-driven treadmill, performed arm cycling, and lifted arm and leg weights under supervision. The intensity of the program was increased on an individual basis.

Data Analysis

All data are expressed as mean ± SD except where otherwise noted. Student paired two-tailed t tests were used to compare data before and after rehabilitation and before and after LVRS. Stepwise and multiple linear regressions were used to evaluate changes associated with increases in diaphragm strength. All statistical analyses were conducted using a commercially available computer software program (Sigmastat, version 2.0; Jandel, San Rafael, CA). p Value < 0.05 was considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patient Characteristics

Baseline demographic characteristics are shown in Table 2. Baseline physiologic data are shown in Table 3. All subjects had severe airflow obstruction (FEV₁, 0.69 ± 0.21 L), hyperinflation, and air trapping (TLC, 7.1 ± 1.7 L; RV, 4.7 ± 1.4 L), and decreased exercise performance (O_2max 12.4 ± 3.2 ml/kg/min).

Physiologic Data before and after 8 wk of Pulmonary Rehabilitation

Spirometry and lung volumes. Table 4 shows spirometry, lung volumes, and diffusing capacity before and 8 wk after rehabilitation in 24 subjects. FVC, FEV₁, and FEV₁/FVC were not different before and after rehabilitation, TLC, RV, FRC_{trapped gas}, and FRC were either slightly increased or unchanged after rehabilitation. Raw was significantly decreased but DL_CO/VA remained unchanged.

Gas exchange and exercise performance. Table 5 shows gas exchange and cardiopulmonary exercise studies before and after rehabilitation. Oxygenation and PCO₂ were not different following rehabilitation. O_2max, E_max, VT_max, and maximum respiratory rate were also not different postrehabilitation during maximum exercise. Total exercise time was greater postrehabilitation (6.9 ± 1.7 versus 5.9 ± 1.6 min, p < 0.002), but the 6-min walk distance was unchanged.

Respiratory muscle strength. Table 6 shows maximum mouth and transdiaphragmatic pressures, and Pdi_twitch before and after pulmonary rehabilitation. Respiratory muscle function did not change following rehabilitation. Pdi_max, Pdi_twitch, PI_max, Pdi_{max sniff}, Pdi_{max combined}, PE_max, and Pdi TTI were similar before and after rehabilitation.

Physiologic Data before and after LVRS

Spirometry and lung volume. Table 7 shows spirometry, lung volumes, and diffusing capacity before and 3 mo after LVRS in all 20 subjects. Before surgery, subjects were severely obstructed and moderately to severely hyperinflated. Three months following LVRS, FVC increased 21%, FEV₁ increased 34%, TLC decreased 13%, FRC decreased 23%, and FRC_{trapped gas} and residual volume decreased by 57 and 28%, respectively. Raw also decreased 34% postoperatively.

Gas exchange and exercise performance. Table 8 shows gas exchange and cardiopulmonary exercise studies before and after LVRS in all patients. Postoperatively, there were no significant changes in oxygenation; however, carbon dioxide tensions were lower (44 ± 6 versus 48 ± 6 mm Hg, p < 0.003). After LVRS, 6-min walk distance (343 ± 79 versus 250 ± 89 m, p < 0.001), total exercise time (9.2 ± 1.9 versus 6.9 ± 2.7 min, p < 0.001) and O_2max (15.7 ± 4 versus 11.5 ± 3 ml/kg/ min, p < 0.001) all significantly increased. Following surgery, minute ventilation (29 ± 8 versus 21 ± 6 L/min, p < 0.001) during maximum exercise was greater, and was associated with a higher tidal volume (1.0 ± 0.33 versus 0.84 ± 0.25 L, p < 0.001) and lower respiratory rate (28 ± 6 versus 32 ± 7 breaths/min, p < 0.02) at peak exercise.

Respiratory muscle strength. Table 9 shows maximum mouth and transdiaphragmatic pressures, and Pdi_twitch before and after LVRS. Following surgery, PI_max, Pdi_{max combined}, Pdi_{max sniff}, Pdi_max, and Pdi_twitch were all greater. PE_max was not different. Tension time index for the diaphragm (Pdi/Pdi_max × TI/Ttot), however, was lower.

Body weight. In 20 patients who underwent LVRS, there was no significant difference in body weight before and after surgery (148 ± 23 versus 151 ± 24% IBW, p < 0.31).

Determinants of Increases in Diaphragm Strength Post-LVRS

In single regression analysis, increases in Pdi_{max combined}, Pdi_max, and Pdi_{max sniff} were not found to correlate with the reductions in FRC (r = 0.16, p = 0.57), RV (r = 0.22, p = 0.44), RV/TLC (r = 0.29, p = 0.28), FRC_{trapped gas} (r = 0.26, p = 0.35), or reductions in Pa_CO2 (r = 0.04, p = 0.88). Similarly, no correlation was found with these parameters and the measured increases in Pdi_sniff, Pdi_twitch, and PI_max (Table 10).

Stepwise multiple regression found that postoperative reductions in RV and FRC_{trapped gas} were strong determinants of the postoperative increases in PI_max (r = 0.67, p < 0.03). However, RV and trapped gas at FRC were not predictive of postoperative increases in Pdi_max (r = 0.33, p = 0.49), Pdi_twitch (r = 0.44, p = 0.47), and Pdi_{max sniff} (r = 0.42, p = 0.38).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Our data confirm that LVRS in patients with severe nonbullous diffuse emphysema improves spirometry and exercise tolerance and reduces lung volume (1). Moreover, our data show that LVRS improves diaphragm strength, lowers inspiratory muscle workload during ventilation, and affords emphysema patients a more comfortable breathing pattern (less rapid and shallow) during maximum exercise. These data clearly demonstrate that LVRS improves diaphragm, lung, and airway mechanics, in contrast to pulmonary rehabilitation alone.

The adverse effects of hyperinflation on diaphragm mechanics have been reported by others (22). These effects include diaphragm precontraction length foreshortening (22- 25), reduced radius of diaphragm curvature (23), impaired diaphragm blood flow (26), decreased diaphragm insertional rib cage action (27), increased internal elastic load (22), and a decrease in the area of apposition of the costal diaphragm with the chest wall (28, 29). As a result, hyperinflation reduces the diaphragm force-generating capacity and limits the ability of the patient with COPD to tolerate increased ventilatory workloads during exercise, or when complicating medical conditions occur. A treatment modality such as LVRS, that simultaneously decreases ventilatory workload and improves respiratory pump action, has the combined advantage of diminishing ventilatory workload while simultaneously improving maximum breathing capacity.

Several studies have reported the effects of LVRS on respiratory muscle recruitment during exercise and on inspiratory muscle strength. Bloch and coworkers (10) found in 19 patients with severe emphysema, who underwent bilateral or unilateral LVRS, that abdominal paradoxical motion monitored by respiratory inductive plethysmography decreased significantly during restful breathing post-LVRS. Benditt and colleagues (9) examined breathing pattern and respiratory muscle recruitment before and after LVRS by examining changes in pleural and gastric pressures. Following LVRS, they found a reduction in end-expiratory esophageal and gastric pressures at rest, and at isoexercise observed a rightward shift in the slope of the esophageal versus gastric pressure plot, suggesting increased use of the diaphragm. Although both studies suggested an improvement in diaphragm strength was responsible for the changes in breathing pattern post-LVRS, no measurements were reported.

Studies examining the effects of LVRS on respiratory muscle strength have reported conflicting results. Teschler and colleagues (12) reported the effects of LVRS on inspiratory muscle strength in 17 severely obstructed and hyperinflated patients with COPD (FEV₁, 0.82 ± 0.07 L; RV, 337 ± 31%). Twelve of the 17 patients had unilateral LVRS. The investigators found that mean PI_max increased by 52% and mean Pdi_{max sniff} increased by 28%, 1 mo postoperatively. In contrast, Martinez and co-workers (8) measured maximum mouth and transdiaphragmatic pressures in 17 subjects before and after bilateral LVRS. They found a 21% increase in PI_max after LVRS, but no significant change in Pdi_{max sniff}. Keller and colleagues (11) found no effect of unilateral LVRS on PI_max in 25 subjects despite significant changes in spirometry, lung volume, 6-min walk test, and ventilatory function during maximum exercise testing.

Why LVRS has shown conflicting results on respiratory muscle strength in the preceding reports is unclear. Several explanations could include the small numbers of patients studied, variability in the techniques used to measure respiratory muscle strength, the effect that rehabilitation may have had on patient well-being and muscle strength, and the effect of studying respiratory and exercise mechanics at different time points post-LVRS.

Our study used a variety of techniques in measuring transdiaphragmatic pressure (both volitional and nonvolitional techniques) and found highly significant increases in maximum mouth and transdiaphragmatic pressures post-LVRS. Moreover, in contrast to earlier investigations, we are confident that these changes were not due to other factors, such as pulmonary rehabilitation, adjustments in medications, or other medical interventions. Patients who received pulmonary rehabilitation in our study (the control group) showed no improvements in maximum mouth or transdiaphragmatic pressures despite comparable medical care by the same clinicians.

Although it has been hypothesized that LVRS improves diaphragm function by reducing end-expiratory lung volume, we failed to demonstrate correlations between changes in transdiaphragmatic pressures and reductions in lung volume. Increases in PI_max post-LVRS tended to correlate with reductions in residual volume (r = 0.44, p = 0.08) and multiple linear regression with RV, FRC_{trapped gas} and PI_max showed significance (r = 0.67 and p = 0.03). It is unclear why PI_max showed a correlation with a reduction in RV while transdiaphragmatic pressure did not. The larger number of subjects studied with PI_max, and the lesser degree of variability in its measurement than in measurements of transdiaphragmatic pressure between subjects, may be potential factors.

Besides a reduction in lung volume, other factors, such as electrolyte abnormalities, or a learning effect in the performance of repeated respiratory tasks, could have been partially responsible for the improvements in diaphragm strength that we found post-LVRS. However, none of our patients had electrolyte imbalances before or after surgery and our control group of rehabilitation patients performed a similar number of repeated ventilatory tasks.

In addition to a reduction in end-expiratory lung volume, a decrease in the arterial partial pressure for carbon dioxide, an increase in body weight, and a change in thoracoabdominal configuration postoperatively all could have been important factors contributing to increases in diaphragm strength. Hypercapnia has been reported to affect diaphragm contractility adversely (30). Autopsy studies have shown that a reduction in body weight is associated with a significant loss in diaphragm mass, thickness, and area (31), and these changes have been shown to decrease maximum respiratory pressures and reduce respiratory reserve (32). In our study, patient body weight did not change post-LVRS. The fact that PE_max was not increased after LVRS suggests that systemic factors, such as electrolyte changes or a reduction in hypercapnia, did not substantially affect global measurements of respiratory muscle strength.

A significant reduction in corticosteroid use postoperatively also could have had a beneficial impact on diaphragm strength. Several human (33) and animal (34) studies have shown the detrimental effects of steroids on diaphragm structure and function. Whether the reduction in postoperative steroid use accounts for some of the increase in diaphragm strength that we observed postoperatively is unclear from our results and cannot be discounted.

One limitation to our study is that diaphragm strength was measured only 3 mo after surgery, a time at which maximum postoperative changes in diaphragm length may not yet have occurred. Similowski and colleagues (35) have shown that adaptive changes may occur in the diaphragm in chronically hyperinflated patients with COPD. In eight well-nourished, chronically hyperinflated, but stable patients with COPD (FEV₁, 1.06 ± 0.4 L; TLC, 7.85 ± 1.13 L [mean ± SD]), they demonstrated that while Pdi_twitch at FRC was lower than that found in normal control subjects, at increased comparable lung volumes (TLC), patients with COPD tended to have greater Pdi_twitch values.

In our study, measurement of diaphragm strength at one time point (3 mo postoperatively) may not have reflected final diaphragm muscle adaptation to its new precontraction length. We may have underestimated the effects of lung reduction surgery on increasing diaphragm force generation if the diaphragm is still operating on the ascending limb of the length- tension curve and undergoing further lengthening. It is hoped that future studies will evaluate the effects of surgical reductions in lung volume on diaphragm force generation in sequential fashion, so as to determine the maximum effects of acute lung reductions on diaphragm force generation.

Patients receiving comprehensive outpatient pulmonary rehabilitation in our study failed to show any significant changes in lung function, as has been previously reported by others. Similar to a large prospective, randomized controlled trial conducted by Ries and collegues (36), we found significant increases in treadmill endurance time during maximum exercise testing. However, the increase in endurance we report is much less than that reported by Ries and coworkers, whose patients were less obstructed and hyperinflated (FEV₁, 1.21 ± 0.55 L, RV/ TLC, 60%) suggesting that their patients were able to tolerate more intensive rehabilitation that resulted in greater improvements.

In summary, LVRS results in significant improvements in spirometry, exercise performance, and diaphragm strength in patients with severe, diffuse, nonbullous emphysema and hyperinflation. Future studies are needed to confirm our results and to determine the long-term stability of these favorable changes over time.