Lung function 12 months following emphysema resection.

OBJECTIVE
To investigate the mechanism of airflow limitation before and 6 and 12 months after targeted emphysematous resection in 10 male patients aged 67 +/- 8 years (mean +/- SD) with very severe COPD undergoing bilateral thoracoscopic stapling techniques.


DESIGN
Lung function, including static lung elastic recoil, was measured 2 weeks before and 6 and 12 months after surgery.


RESULTS
Twelve months after surgery, there was a significant (p < 0.001) reduction in total lung capacity (TLC), 9.5 +/- 0.3 L (mean +/- SEM) to 8.5 +/- 0.3 L, functional residual capacity, and residual volume. Airway conductance and FEV1, 0.71 +/- 0.1 L (mean +/- SEM) to 0.95 +/- 0.1 L, improved significantly (p < 0.01). Lung elastic recoil increased markedly at TLC from 11.7 +/- 0.7 cm H2O (mean +/- SEM) to 15.0 +/- 1.0 cm H2O (p < 0.01) as did maximum expiratory airflow in every patient. However, when compared with data obtained in each patient at 6 months, lung volumes are significantly increased, and expiratory airflow and lung elastic recoil pressures are significantly reduced (p < or = 0.05). Analysis of maximum expiratory flow-static elastic recoil pressure curve indicates conductance of the S airway segment (Gs) increased from 0.20 +/- 0.03 L/s/cm H2O (mean +/- SEM) to 0.28 +/- 0.04 L/s/cm H2O (p < 0.02), and critical transmural pressure in the collapsible segment (Ptm') decreased from 3.2 +/- 0.2 cm H2O (mean +/- SEM) to 2.5 +/- 0.2 cm H2O (p < 0.01).


CONCLUSION
The improvement in maximal expiratory airflow can be attributed primarily to increased lung elastic recoil and its secondary effect on enlarging airway diameter causing increased airway conductance, increased Gs, and decreased Ptm'. The improvement in lung function and elastic recoil peaks at 6 months.

previous reports of surgical intervention to improve lung function in patients with emphysema have noted increased morbidity and/or mortality, and physiologic im~rovement has been variable and transient :51 year. 1 -Recently, there has been renewed interest in thoracic surgical procedures that attempt to provide relief for markedly dyspneic patients suffering from severe generalized emphysema. 7 -15 Unilateral 7 -'9,ll.l 2 or bilat-*From the P!llmonary Division, Department of Medicine, Lakewood (Califomia) Regional Medical Center (Drs. Gelb 1 -15 video-assisted thoracoscopic 7 -9 · 11 • 12 .1 4 .1 5 or median stemotomy 7 .l 0 .1 3 incisions are made and visually the most emphysematous areas are excised using stapling 7 .l 0 -15 or laser techniques 8 · 9 or both. Results 1 'll indicate vmiable clinical relief of dyspnea and improvement in lun~ fun;tion in selected patien!s followed up 6 months, 8 -r2.14,1<:> 1 year, 13 and 3 years. 1 We and others have reported previously that the increased expiratory airflow and airway conductance following bullectomy in isolated bullous lung disease 5 .l 7 and bullous emphysema 1 · 2 · 4 -6 could be accounted for by the measured improvement in lung elastic recoil. 1 · 2 · 4 · 6 .1 7 Recent short-term (3 to 6 months) studies have noted the mechanism(s) of improvement in lung function following unilateral 11 and bilateral 14 .l 5 surgical removal of targeted emphysematous tissue lung volume reduction surgery [LVRS]) in markedly symptomatic patients with far-advanced COPD could be accounted for by the increase in lung elastic recoil.
The present study evaluates lung function and elas-CHEST / 110 / 6 / DECEMBER, 1996 tic recoil at 6 and 12 months following LVRS for generalized emphysema.

Patient Selection
We consecutively studied 14 patients (12 men) aged 67::'::8 years (mean::'::SD ). However, four patients refused to undergo repeated esophageal b alloon studies at 12 months. Therefore we report complete data o n lO male patients. As previously described, 14 · 15 all the patients who underwent the r rocedure were markedly symptomatic with grade 2::3 dyspnea, 1 with severe Hxed expiratory obstruction that h ad n ot improved d espite appropriate therapeutic interventions, including oxygen, antibiotics, aerosol and systemic bronchodilators, cmticosteroids, and pulmona~y rehabilitation. In addition, high-resolution, thin-section CT of the lungs demonstrated emphysema scores 19 · 20 2::60 with heterogeneous distribution, ie, predominant emphysematous destruction of upper t o middle lung fi elds with relative preservation of n ormal lung tissue in the l ower lung fields . Standard nuclear medicine ventilation and perfusion lung scans demonstrated similar heterogeneous distribution.

Operative Technique
As previously described, 8 · 12 .l 4 • 15 after obtaining informed consent, and approval of the Institutional Human Investigation Committee at Chapman Medical Center, patients underwent sequential bilateral video-assisted thoracoscopic surge1y a t the same o perative sitting under vecuronium paralysis and isoflurane general anesthesia with fraction of inspired oxygen of 1.0 using a le ft-sided 39F double-lumen endotracheal tube ( Mallincrodt Anesthesia; St. Louis) . After single dependent lung ventilation had been a chieved, the contralateral upside deflated lung was examined. Visually, the most distended, destroyed, emphysematous a reas previously t argeted b y the preoperative CT lung scan in the upper and middle lung fields were excised and linear staple lines were reinforced with bovine pericardium 21 (Peri-Strips; Bio-Vascular Inc; St. Paul, Minn) or bovine collagen ( In stat; Johnson and Johnson; New Bruns·wick, NJ) to minimize air l eaks. It was estimated that the excised lung volume was approximately 15 to 20% of each lung. A ctual weight of each resected lung was 30 to 90 g. Following lung excision, apical pleural tents and! or t alc pleurodesis were not required. Operative time ranged from 1 t o 2 h .

Lung Function Studies
As previously described, 14 ,1 5 outpatient lung function studies were performed after obtaining in formed consent. These included functional residual capacity (FRC) measured by plethysmographic techniques, 22 timed spirometly, and single-breath diffusing capacity in accordance \vith Ame1ican Thoracic Society recomm enda-tions23·24 and values were compared with predictions. 25 · 27 All patients were considered to have ftxed airflow limitation since the F EV1 following three inhalations o f aerosolized albuterol (670 pg) improved <12% and/or <200 mL. 23 Maximum insr iratory and expiratory flow volume curves, thoracic gas volume, 2 and airway re-sistance28 were all measured in a pressure-corrected v olume plethysmograph (model6200; Sensormedics Inc; Yorba Buena, Calif) and compared with predicted values. 22 • 28 • 29 The r eciprocal of airway resistance 28 is conductance and it was divided b y the t horaci c gas volume at which it was m easured and specific conductance (SGaw) calculated. Normal values are >0.12 Us/em H20 /L. 30 Residual volume was calculated b y subtracting vital capacity from t otal lung capacity (TLC). Studies were obtained within 2 weeks prior to and repeated within 5 to 6 months and again 12 months after surge1y .

Lung Elastic Recoil Pressures
As previously noted, 14 · 15 static lung elastic r ecoil curves w ere obtained in all patients in the plethysmograph in the sitting position after positioning a n intraesophageal balloon inflated \vith 0.5 mL air in the l ower third of the esophagus using previously described techniques 31 • 32 to best r eflect pleural pressure. After at l east tvvo inspirations to TLC, static transpulmonary (airway-esophageal) pressures were recorded following stepwise interruption of exhalation against a closed shutter at a given lung volume for at least 3 s . A minimum of Hve curves were obtained in each p atient and a line of b est visual Ht of the pooled d ata was drawn. Studies were obtained within 2 weeks prior to and repeated within 5 to 6 months and again 12 months after surgery.

Mechanism of Expiratory Airflow Limitation
As previously noted, 15 to determine the mechanism of airflow limitation, we had t o evaluate the d1iving pressure for expiratory airflow (elastic r ecoil) and the airway caliber. We constructed maximum expiratory flow-static lung elastic recoil pressure c urves (MFSR) 33 by plotting maxi mum expiratory airflow (U s) obtained from the maximum expiratmy flow volume curve against the corresponding s tatic transpulmonmy pressure (em HzO) at the same lung volume and compared \vith previous s tandards. 29 The slope of the MFSR curve bet ween 50% and 30% of the FVC was calculated and represents the conductance of the S airway segment (Gs) according to Pride et al. 34 We have used this model previously to determine the mechanism of expiratory airflow limitation in normal subjects 29 and in patients with chronic obstructive lung disease. l5, 35.36 Normal values 29 · 35 previously obtained in seven subjects aged 61 to 74 years for Gs, 0.6::' :: 0.1 U s/em H 2 0 (mean ::' :: SD), and critical transmural pressure in the collapsible segment (Ptm' ), 1.73::' :: 0.41 em H20 (mean ::' :: SO), are similar to those of Leaver and cowork-ers36 and Gibson et ai. 37 Airway conductance as meas ured in the plethysmograph was plotted against the corresponding s tatic transpulmonary pressure a t the same lung volume and compared with previous values obtained in seven n ormal subjects aged 61 t o 74 years. 29

Statistical M ethods
Compa1ison of the difference between p atients before and after surgery was determined u sing t wo-tailed paired t test with values s Q.0. 5 being significant. The degree of linear association between two continuous vaJiables was assessed using Spearman correlation coefficients based on ranks.

R ESU LTS
Results of lung function studies appear in Table 1 and Figures 1-4. The average hospital staywas 10.9 ::±: 1.1 days (mean::±:SD). D~spnea was improved in every patient by 2:1 grade. 1 Oxygen dependence was eliminated in seven of the 10 patients.
Preoperative and 6-month postoperative lung function studies of the four patients who refused to undergo additional esophageal balloon studies at 12 months postoperation have been previously reported 15 and were not significantly different (p>0.05) from the patients described herein at comparable time periods. Additionally, their clinical course and spirometry (data not shown) at 12 months postsurgery is similar to the patients in the present report. Spirometric studies were available in six patients up to 1 year prior to sur-  *VC=vital capacity; RV=residual volume; Dco=diffusing capacity (mUminlmm Hg); VA=alveolar lung volume ( L); PsTAT=static lung e l astic recoil pressure (em H20); Gaw=aiJway conductance (Us/em H20); Gs=conductance s airway segment. ge1y, and results were similar for FYC and FEY 1 when compared with preoperative baseline values despite aggressive therapeutic intervention, including physical rehabilitation. Six-month postoperative data for cases 1, 2, 4, and 6 through 10 have been repmted previously. 15 There was improvement in both static and dynamic lung function studies. Twelve months following surgery, TLC, residual volume, and the FRC decreased significantly (p< 0.001 ) ( Table 1). Total lung capacity decreased from 9.5±0.3 L (mean±SEM) (148±5.0% predicted) to 8.5±0.3 L (133±4.0% predicted) (p<0.001 ). However, the values at 12 months are significantly (p:50.03) greater than the values at 6 months (Table 1), indicating increasing hyperinflation.

Analysis of the Static Lung Elastic Recoil Pressure Curves
Twelve months postoperatively, there was an increase in lung elastic recoil pressure at the same lung volume when compared with preoperative values with a shift in the curve to the right (Fig 2) . There was no significant (p=0.2) change in lung compliance at FRC plus 0.4 L; preoperative value was 0.320±0.03 Ucm H20 (mean±SEM) and postoperative value was 0.280±0.03 Ucm H20. At 12 months postsurgery, there was a significant (p <O.Ol ) increase in static lung elastic recoil pressure at TLC; preoperative value was 11.0±0.7 em H20 (mean±SE M) and postoperative value was 15.0± 1.0 em H20 . However, this is significantly reduced (p=0.03) when compared with values obtained 5 to 6 months after surgery, 17.0±0.9 em H20. There was also a significant increase in static lung CHEST / 110 / 6 / DECEMBER, 1996 1409     15 Results of maximum ex'Piratory and inspiratory flow volume loops 2 weeks before and 6 and 12 months after LVRS for emphysema. In each of 10 patients, the loops initially +---t--I-{F=-1~"'t·rt----l shifted to a l ower lung volume fol-2 lowing surge1y. After 12 months, the l oops in eight patients have shifted to a higher lung volume. Preoperative and postoperative values. for FEV 1 (liter) and (percent predicted) are included. For comparison with normal data, maximum expiratory flow at 80% TLC (4.6 L) is5.8::<:: 1.5 Us (range). 29 The preoperative loop in each patient is at the highest lung volume. - § . . . . . -· · · · · · · · · · · · · · · · · · · ·   5 Results of static lung e l astic recoil pressure curves in each of 10 patients obtained 2 weeks before and 6 and 12 months after LVRS for emphysema. A ctual lung volume is pl otted. Separate curves were obtained in each p atient at l e ast five times and data pooled and une of best visual fit are described. Nom1al data (mean±2 SDs) were obtained in seven h ealthy subjects aged 61 to 74 years. 29 Initially following surgery, the curves are shifted to the right in each patient, and after 12 months, the curves in every patient demonstrate subsequent loss of lung elastic recoil.

Analysis of the MFSR Curve
Preoperatively, the intercept (Ptm 1 ) was shifted toward higher pressures in all patients, 3.2±0.2 em H20 (mean±SEM) and the Gs was markedly reduced, 0.20±0.03 Us/em H20 (mean±SEM), when compared with normal values (Fig 3). 29 This suggests maximum expiratory airflow was severely reduced not only due to loss of lung elastic recoil, but also due to intrinsic narrowing and/or collapse of the airways.
Twelve months following surgery, with increased lung elastic recoil, airflow increased and the intercept on the pressure axis at zero maximum flow (Ptm 1 ) of the MFSR slope shifted toward lower pressures, 2.5±0.2 em H 2 0 (mean±SEM), and the Gs (MFSR slope) improved in every patient to 0.28±0.04 Us/em H20 (mean±SEM). The decrease in Ptm 1 and increase in Gs was significant (p<0.01) and suggests that ainvay collapse was partially relieved due to increased elastic recoil with better airway traction and support and enlarged airway caliber. The values for Ptm 1 and Gs are similar at 6 and 12 months postsurgery.
Following surgery, persistent airflow limitation in four patients (cases 1, 4, 7, and 10) could be accounted for almost completely by loss of lung elastic recoil. In the remaining patients, persistent airflow limitation in the S segment could be accounted for in part by loss of lung elastic recoil and reduced caliber of the airway lumen due to intrinsic disease and/or ainvay collapse due to reduced ainvay traction and/or compression.

Analysis of the Airway Conductance Lung Elastic Recoil Pressure Curve
Twelve months following surgery, despite the reduction in lung volume at FRC, there was an increase in airway conductance that could be accounted for by the mean increase in lung elastic recoil (1.7 em) at FRC (Fig 4). However, airway conductance remains borderline normal in relation to elastic recoil due to marked decrease in traction support and/or intrinsic narrowing of the airway lumen and/or bronchial compression.

Analysis of Predictors of Postoperative Increase in FEV1
After 12 months, the postoperative increase in FEV 1 was poorly correlated with baseline preoperative static lung elastic recoil at TLC (r=0.3, p=0.3), static elastic recoil at TLC/TLC (coefficient of retraction 38 ) (r=0.28, p=0.3), and Gs (r=0.4, p=0.2). Multiple linear regres-1412 sion analysis revealed moderately good correlation between postoperative increase in FEV1 with postoperative increase in coefficient of retraction and Gs at 6 months (r=0.70, p=0.05) 15 and (r=0.70, p=O.lO) at 12 months. DISCUSSION Results in the present study extend our earlier short-term (6 months) observations.l 4 .l 5 Marked clinical and physiologic improvement persists in lung function at 1 year following bilateral targeted excision of severely emphysematous areas in patients with very severe expiratory airflow limitation. As previously noted, 14 .1 5 the improvement in lung function can be accounted for primarily by increased elastic recoil properties of the lung with greater driving pressure and to secondary increases in airway diameter.
The magnitude of the 12-month improvement in lung function in the present study appears similar to that reported by Gaissert et atl 3 after bilateral emphysema resection using a median sternotomy incision.
We believe the decrease in lung volumes and increased expiratory airflow and ainvay conductance are accounted for by the increase in lung elastic recoil that was measured in each patient. These changes reach their greatest magnitude at 5 to 6 months following excision of the most severely emphysematous lung tissue. After 12 months, the lung elastic recoil pressures in the remaining lung are submaximal. Moreover, the 6-and 12-month changes in lung volume and expiratory airflow reflect the initial increase and then loss of lung elastic recoil (Table 1 and Figs 1 and 2).
Results of MFSR curves 12 months after emphysematous lung resection demonstrate a parallel slope to that observed at 6 months after surge1y. The increased slope of the MFSR curves together with the decrease in Ptm 1 when compared with preoperative values suggests greater ainvay diameter and stability \vith less collapse in the S segment and can be attributed to the increase in lung elastic recoil. 15 Atrophy and/or compression of the airway wall and/or increased bronchomotor tone and/or intrinsic small airways disease could also lead to increased airway collapse (increased Ptm 1 ) and decreased slope of the MFSR curve. We suspect that the increase in the slope of the MFSR curve and decrease in Ptm 1 may be attributed to increases in lung elastic recoil with better traction around the ai1ways, thereby increasing airway diameter. Alternatively, the failure to increase the slope of the MFSR curve postoperatively probably reflects persistent intrinsic airway disease and/or airway comgression.
As previously discussed, our studies in asymptomatic patients with mild physiologic abnormalities yet moderately advanced macroscopic anatomic emphy-  15 Results of maximum expiratmy flow-static lung elastic recoil pressure cmve in 10 patients obtained 2 weeks before and 6 and 12 months after LVRS for emphysema. P1ior to surgery, airflow limitation is accounted for by a decrease in lung elastic recoil pressure as well as decreased airway caliber due to impaired ain..,ay traction and/or bronchial compression and/or intrinsic airway obstruction (decreased Gs). Initially after surgery, there was a modest to large (four patients) i ncrease in Gs and lung elastic recoil with a decrease in Ptm' suggesting that increased flow was predominantly accounted for by increased lung elastic recoil and secondarily to increased airway caliber and stability probably due to increased traction around tl1e airway. Even after surgery, the Gs is abnormal in six patients suggesting maximal flow is reduced due to decreased intraluminal airway caliber in addition to loss of lung elastic recoil. In four patients (cases 1, 4, 7, and 10), airflow limitation is almost completely accounted for by loss oflung elastic recoil with near normal Gs even at 12 montl1s after surgety. Changes in Gs and Ptm' are similar at 6 and 12 montl1s. P stat (l )em H20 FIGURE 4. Using previous format, 15 results of aiJWay conductance (Caw) vs static lung elastic recoil relationship. Mean data are shown for 10 patients 2 weeks before and 6 and 12 months after LVRS for emphysema. Normal data range is obtained from previous studies 29 in seven normal subjects aged 61 to 74 yrs. Following lung resection despite the marked reduction in lung volume, aiJWay conductance is increased due to increased lung elastic recoil. However, even 12 months after surgery, Gaw remains borderline normal predominantly due to loss of lung elastic recoil with decrease in ai1way traction and/or intraluminal airway obstruction.
sema demonstrated that airflow limitation was primarily due to loss of lung elastic recoil 35 · 39 and the slope of the MFSR curve ( Gs) was reduced in only one offive patients. 35 The reduced Gs was attributed to concurrent small airway abnormalities. 35 Even in patients with severe expiratory airflow limitation associated with a1-antitrypsin deficiency, Black et al 40 found that decreases in lung elastic recoil could account for abnormal flow and pulmonary conductance in only five of 10 patients. In the other five patients, concomitant intrinsic airways disease and/or increased bronchial compression were suspected. They 40 noted that a decreased MFSR slope was characteristic of COPD patients without a 1 -antitrypsin deficiency. The improvement in lung function attributable to increased lung elastic recoil followin? LVRS in emphysema in this and our previous studies 4 .1 5 is similar to increased lung elastic recoil measured following bullectomy in bullous emphysema 1 · 2 • 4 · 6 and in isolated bullous lung disease. 17 The study by Sciurba et al 11 has also noted improvement, but oflesser magnitude in lung elastic recoil and lung function 6 months following unilateral excision of targeted emphysema using combined laser and/or stapling techniques.
In summary, we have demonstrated persistent clinical improvement in lung function 12 months after bilateral LVRS for emphysema. There is an increase in lung elastic recoil and improved lung function that peaks at 5 to 6 months. There is a parallel decrease in lung volumes and increased expiratory airflow that peaks at 5 to 6 months. Longer-term follow-up will be needed to assess the impact on lung function over time. Proper selection of patients is crucial to optimize potential benefits.