Lung Volume Reduction Surgery Using the NETT Selection Criteria

Transcription

1 Lung Volue Reduction Surgery Using the NETT Selection Criteria Mark E. Ginsburg, MD, Byron M. Thoashow, MD, Chun K. Yip, MD, Angela M. DiMango, MD, Roger A. Maxfield, MD, Matthew N. Bartels, MD, Patricia Jellen, MSN, Willia A. Bulan, MD, David Lederer, MD, Francis L. Brogan, RN, Lyall A. Gorenstein, MD, and Joshua R. Sonett, MD Departents of Surgery, Medicine, Radiology, Rehabilitation, and Nursing, Colubia University College of Physicians and Surgeons, New York, New York Background. The National Ephysea Treatent Trial (NETT) proved that lung volue reduction surgery (LVRS) was safe and effective in patients with certain clinical characteristics and using defined inclusion-exclusion criteria. Based on the selection criteria developed in that trial, we perfored bilateral LVRS on 49 patients during the period of February 2004 until May Methods. Forty-nine patients underwent lung volue reduction by either edian sternotoy (10) or videoassisted thoracoscopic surgery (39) selected according to NETT described paraeters. Preoperative characteristics were the following: ean ( SD) age years, preoperative FEV 1 (forced expiratory volue in the first second of expiration) 691 cc ( 159), % of predicted FEV ( 6.2), preoperative DLCO (diffusing capacity of lung for carbon onoxide) 7.6 ( 2.7), and % of predicted DLCO 27% ( 7.3). All patients had upper lobe predoinant disease and either low exercise capacity (n 23) or high exercise capacity (n 26) as defined by the NETT. Results. There was no operative or 90-day ortality. Median length of stay was 8 days (interquartile range 6 to 10). Two patients required reintubation and tracheostoy but were decannulated prior to discharge. The BODE index (body ass index, airflow obstruction, dyspnea, and exercise capacity), a ultidiensional predictor of survival in chronic obstructive pulonary disease, iproved 2.3 ( 1.5, p < ) (issing data: 5 of 42, 11.9%) and the FEV 1 iproved 286 cc ( 221, p < ), both 1 year after surgery. Probability of survival was 0.98 (95% CI [confidence interval] 0.94 to 1) at 1 year, and 0.95 (95% CI 0.88 to 1) at 3 years. Conclusions. Surgical lung volue reduction for ephysea can be perfored in patients using selection criteria developed by the NETT with very low surgical risk and excellent idter results. Surgical LVRS is the standard against which other nonsurgical treatents for advanced ephysea should be judged. (Ann Thorac Surg 2011;91: ) 2011 by The Society of Thoracic Surgeons The Centers for Disease Control and Prevention reported that in 2008 there were 3.8 illion noninstitutionalized adults diagnosed with ephysea [1]. Palliative treatent for ephysea includes inhaled corticosteroids, long-acting bronchodilators, oxygen therapy, and pulonary rehabilitation. In 1993, Dr Joel Cooper and colleagues [2] reintroduced lung volue reduction surgery (LVRS) as a possible surgical therapy for selected patients with a heterogeneous for of ephysea. The procedure was et with significant skepticis on the part of the pulonary counity. Concerns were raised relating to safety, patient selection, effectiveness, choice of surgical technique, and cost. Because of these concerns, general acceptance by the edical counity did not occur. Accepted for publication Jan 20, Presented at the Forty-sixth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 25 27, Address correspondence to Dr Sonett, Division of Cardiothoracic Surgery, Colubia University Medical Center, 161 Fort Washington Ave, New York, NY 10032; e-ail: eg18@colubia.edu. The National Ephysea Treatent Trial (NETT) was designed to answer the fundaental questions surrounding LVRS and to identify clinical paraeters that predicted success [3]. The study copared LVRS to axiu edical therapy and deonstrated that bilateral LVRS was safe and effective in patients with certain preoperative characteristics [4]. Furtherore, a significant survival advantage was observed in a specific subgroup of patients with upper lobe predoinant ephysea and a low exercise capacity [5]. As a result of the NETT findings, the Centers for Medicare and Medicaid Services (CMS) initiated coverage of LVRS in 2003 at designated centers of excellence. Despite the copelling results of the trial, adoption of the procedure by the edical counity has been exceedingly poor. The CMS reported [6] that only 119 Medicare beneficiaries underwent the LVRS in Our center has continued to actively evaluate and perfor LVRS and we report our experience utilizing the NETT inclusion-exclusion and CMS selection criteria during the period after the NETT by The Society of Thoracic Surgeons /$36.00 Published by Elsevier Inc doi: /j.athoracsur

2 Ann Thorac Surg GINSBURG ET AL 2011;91: LVRS USING NETT CRITERIA 1557 Material and Methods We perfored a retrospective study of all patients undergoing LVRS at Colubia University Medical Center between January 2004 and April Patients with severe ephysea referred to our institution during this period were evaluated for LVRS using the ethodology and selection criteria of NETT. Clinical evaluation was perfored as described in NETT [4] and included postbronchodilator spiroetry, roo air arterial blood gas analysis, plethysography, single-breath carbon onoxide diffusing capacity, chest X-ray, coputed toography of the chest, echocardiography, cardiac stress testing, post-rehabilitation cardiopulonary exercise testing using cycle ergoetry, quantitative ventilation-perfusion scanning, assessent of dyspnea (MRC [odified Medical Research Council] dyspnea scale), and 6-inute walk test. The BODE index (Table 1) was calculated as described by Celli and colleagues [7] using postrehabilitation data. After pulonary rehabilitation, a CMS-qualified thoracic surgeon evaluated patients eeting all CMS inclusion criteria with exclusion criteria. During the period of this study, 3 clinical trials in bronchoscopic LVRS were ongoing at our institution and eligible patients were offered entry into these trials. Patients were also considered for lung transplantation if edically appropriate. All patients accepted for LVRS underwent preoperative pulonary rehabilitation and were reevaluated after copleting that progra. All patients undergoing LVRS et NETT inclusionexclusion criteria and CMS requireents (Table 2). Follow-up was perfored at 6 onths, and then yearly. We confired deaths using the Social Security Death Index. The Colubia University Institutional Review Board approved this study. Table 2. NETT Inclusion-Exclusion Criteria Inclusion: BMI 31.1 kg/ 2 (en) or 32.3 kg/ 2 (woen) FEV 1 45% predicted (and 15% predicted if older than 70 years) TLC 100% predicted RV 150% predicted Arterial CO 2 60 Hg at rest on roo air Arterial O 2 45 Hg at rest on roo air Ability to walk 140 eters in 6 inutes Ability to pedal 3 inutes on a bicycle ergoeter without a load Abstinence fro soking for 6 onths Exclusion: Previous lung transplant; LVRS; lobectoy; or edian sternotoy Pulonary: PAsys 45 Hg, giant bullae, significant sputu production with recurrent infections, pleural disease that precludes surgery O 2 requireent 6 L to keep saturation 90% with exercise Cardiac significant arrhythia, MI or CHF within 6 onths and LVEF 0.45, uncontrolled hypertension BMI body ass index; CHF congestive heart failure; CO 2 carbon dioxide; FEV 1 forced expiratory volue in the first second of expiration; LVEF left ventricular ejection fraction; LVRS lung volue reduction surgery; MI yocardial infarction; O 2 oxygen; PAsys systolic pulonary artery pressure; RV residual volue; TLC total lung capacity. Surgical Care All patients had placeent of an epidural catheter preoperatively for pain anageent. Patients underwent either a edian sternotoy or a bilateral sequential video-assisted thoracoscopic surgery (VATS) approach (Fig 1). Surgical staplers with buttressed aterial and selective use of lung sealants were used to resect lung parenchya. Resection targets were identified by perfusion and coputed toographic scanning, and confired by intraoperative observations. The ideal aount of lung to resect has not been defined; however, we attepted to resect an aount of lung that left the resting volue at end inspiration adequate to fill the relaxed pleural space. Statistics Descriptive statistics are expressed as ean SD unless otherwise stated. Discrete ordinal variables were su- Table 1. BODE Index for COPD Variable Score BMI FEV 1 % predicted 65% 50 64% 36 49% 35% MRC dyspnea scale inute walk test BMI body ass index; BODE index body ass index, airflow obstruction, dyspnea, and exercise capacity; COPD chronic obstructive pulonary disease; FEV 1 forced expiratory volue in the first second of expiration; MRC odified Medical Research Council. Fig 1. Thoracoscopic lung volue reduction surgery technique.

3 1558 GINSBURG ET AL Ann Thorac Surg LVRS USING NETT CRITERIA 2011;91: Table 3. Patient Characteristics Age (years) 62.5 (ean, SD 6.6) Gender (M/F) FEV 1 (cc) 691 (ean, SD 159) FEV 1 (% predicted) 25.3 (ean, SD 6.2) Dlco 7.6 (ean, SD 2.7) Dlco (% predicted) 27 (ean, SD 7.3) Oxygen usage 23/37 (62%) Maxial exercise capacity (watts) 37.3 (ean, SD 15.4) NETT group 1 23 NETT group inute walk test (ft) 1359 (ean, SD 209) Body ass index 25.2 (ean, SD 3.7) MRC dyspnea index 2.7 (ean, SD 0.9) Dlco diffusing capacity of lung for carbon onoxide; FEV 1 forced expiratory volue in the first second of expiration; MRC Medical Research Council; NETT National Ephysea Treatent Trial. arized by edian, range, and interquartile range. Coparisons were done with paired, 2-tailed t tests for eans of norally distributed continuous variables. Kaplan- Meier estiation was used to report survival. Because prior studies of LVRS have been criticized for the failure to account for the proble of survivorship bias, we also iputed 1-year issing data with the worst 1-year values found in the entire cohort to evaluate results (worst rank score iputation). Patient Characteristics Bilateral LVRS was perfored on 49 patients during the study period. No patients underwent unilateral LVRS and no procedures were aborted. The ean age was 62.5 (SD 6.6) years. The surgical approach was a edian sternotoy in 10 patients and bilateral sequential VATS in 39. The VATS was used exclusively on all patients operated on since October 2005, and no open conversions were necessary. The preoperative clinical variables are suarized in Table 3. All patients were in NETT groups 1 or 2 (Table 4). The post-rehabilitation preoperative forced expiratory volue in the first second of expiration (FEV 1 ) was 681 cc (SD 159) consistent with severe ephysea. Oxygen use, either at rest or with exercise, was required in 23 of 37 (62%) of the patients. Maxial exercise capacity was 37.3 watts (ean, SD 15.4). Table 5. Coplications Results Frequency (n 49) Event No. % Prolonged air leak ( 7 days) Respiratory failure/tracheostoy 2 4 Pneuonia 2 4 C. diff 2 4 DVT 1 2 Reoperation 1 2 Deliriu 1 2 Epyea/eloesser 1 2 Readission 1 2 Heoptysis 1 2 No coplications C. diff Clostridiu difficile; DVT deep venous throbosis. Median length of hospital stay was 8 days (interquartile range 6 to 10 days). There was no operative, hospital, or 90-day ortality. Coplications (Table 5) were frequent, but ostly related to prolonged air leaks. Two patients required reintubation and tracheostoy but were decannulated prior to discharge. Four patients were discharged to inpatient rehabilitation facilities and subsequently discharged to hoe. Forty-five patients were discharged directly hoe. No patients were discharged on ventilator support or to nursing hoes. The Kaplan-Meier estiate of survival was 0.98 (95% confidence interval [CI] 0.94 to 1) at 1 year and 0.95 (95% CI 0.88 to 1) at 3 years (Fig 2). No patients have subsequently undergone lung transplantation. The post-bronchodilator FEV 1 iproved fro 25.0% predicted to 36.2% predicted 1 year after LVRS (p Table 4. National Ephysea Treatent Trial (NETT) Groups (Low Exercise 25 Watts for Feales, 40 Watts for Males; High Exercise 25 Watts for Feales, 40 Watts for Males) NETT Grouping 1 Upper lobe predoinant/low exercise capacity 2 Upper lobe predoinant/high exercise capacity 3 Non-upper lobe predoinant/low exercise capacity 4 Non-upper lobe predoinant/high exercise capacity Fig 2. Kaplan-Meier survival after lung volue reduction surgery. (CI confidence interval.)

4 Ann Thorac Surg GINSBURG ET AL 2011;91: LVRS USING NETT CRITERIA 1559 Table 6. BODE Coponents ). Likewise, the MRC dyspnea index decreased fro 2.7 to 0.9 (p ) at 1 year. There was a odest iproveent in the 6-inute walk test fro 1,359 feet to 1,495 feet (p 0.001). The BODE index, a ultidiensional easure coposed of the FEV 1, the body ass index, the odified Medical Research Council dyspnea score, and the 6-inute walk distance was calculated by suing the scores assigned to each of its four coponents. Higher scores signify ore severe disease and have been correlated with an increased risk of death fro both chronic obstructive pulonary disease (COPD) and all causes. The BODE index (body ass index, airflow obstruction, dyspnea, and exercise capacity) in our patients iproved 2.3 (ean, SD 1.5) at 1 year. An iproveent of at least 1 point occurred in 34 of 37 (92%) patients, and at least 2 points in 25 of 37 (68%) patients. All 4 individual coponents of the BODE index iproved significantly (Table 6), but only the dyspnea index and FEV 1 contributed to changes in the BODE index itself. Furtherore, the BODE index iproved ore in the NETT group 1 patients copared with NETT group 2 patients (2.87 vs 1.91, p 0.026). The iputed BODE index, using worst rank score iputation to account for issing data, iproved 1.95 (ean, SD 1.55, p 0.001) copared with baseline. Suppleental oxygen use declined substantially after LVRS preoperatively. Twenty-three of 37 (62%) patients followed for 1 year used oxygen either at rest or with exercise, as opposed to 15 of 37 (41%) 1 year postoperatively. Coent Preoperative 1 Year Change in BODE p Value BMI FEV DI MWT Missing data 5 of 42 patients. BMI body ass index; BODE body ass index, airflow obstruction, dyspnea, and exercise capacity; FEV 1 forced expiratory volue in the first second of expiration; 6MWT 6 inute walk test. Adoption of LVRS by the pulonary counity has been exceedingly poor. This is surprising in view of the ipressive results reported by the NETT. The reasons underlying this outcoe are not entirely evident. The NETT deonstrated a significant survival advantage in a well-characterized group of surgically treated patients with an acceptable surgical risk. Furtherore, the NETT showed that these results could be generalized to a broad spectru of institutions with experience in treating advanced lung disease. The results detailed in this report suggest that the use of the selection criteria developed by NETT and adopted by CMS results in a very low treatent risk and high probability of treatent benefit. These results reflect not only the highly rigid selection process iposed by CMS but also our growing experience with this operation and avoidance of issteps that led to coplications and poor results in the past. While we do not iply that siilar results could not be obtained in a broader patient population, we feel that there is inadequate underlying knowledge at the present tie to expand patient selection. We have continued to evolve our surgical technique. We initially perfored LVRS using a edian sternotoy and bovine pericardial-reinforced stapling as originally described by Cooper and colleagues [2]. However, our growing technical cofort and experience with VATS for any thoracic surgical procedures has allowed us to apply this technique cofortably to LVRS and is now our approach of choice. Since 2005, we have perfored all LVRS procedures using bilateral sequential VATS through either 2 or 3 incisions. All patients have had staple lines reinforced using a variety of preconstructed aterials. Surgical glues are used liberally at the discretion of the surgeon. Open conversion has not been required. We believe the ost evolution in our surgical technique over the past decade has been a trend toward ore conservative resections. We have intentionally reduced the volue of pulonary tissue that we reove so as to avoid coplications related to pleural space probles. While difficulties with parenchyal air leaks continue to occur, larger air leaks requiring reoperation and creating difficult pleural space issues are now far less coon. We believe this less aggressive surgical approach has contributed both to short-ter reduction in orbidity as well as to long-ter reduction in ortality, but with continued iproveents in overall cardiopulonary function. The BODE index has been proposed [7] as a surrogate endpoint to predict survival for patients with COPD and to study therapeutic interventions that ay offer benefit in ephysea. The usefulness of the BODE index is that it captures systeic consequences of the disease in addition to the degree of airflow obstruction. The utility of the BODE index to evaluate outcoes after LVRS is unknown. We [8] have previously reported on a saller group of LVRS patients using the BODE index to assess benefit. Martinez and colleagues [9] applied a odified BODE index to the NETT database and showed that it predicted both long-ter and short-ter ortality better than its coponents. Ifeld and colleagues [10] found that the postoperative BODE index predicted survival better than the preoperative BODE index after LVRS. The iproveent in BODE index seen in our series was clinically relevant. Martinez and colleagues [8] deonstrated that an increase in the BODE score of only 1 point proved predictive of an increase in ortality in post-lvrs patients. Analysis of the individual coponents of the Bode index revealed that the iproveent cae fro a reduction in airflow obstruction and a decrease in dyspnea. The iproveent was greater in NETT group 1 versus NETT group 2 patients consistent

5 1560 GINSBURG ET AL Ann Thorac Surg LVRS USING NETT CRITERIA 2011;91: with the results obtained in the NETT. It is interesting that NETT group 2 patients in this series also had a significant iproveent in BODE, suggesting a future ortality reduction. The use of the BODE index to evaluate outcoes in LVRS has potential probles. In our series, neither the body ass index nor the 6-inute walk contributed to the changes in the BODE index. The LVRS patients are a highly select cohort that undergoes substantial pulonary rehabilitation prior to surgery. Surgical selection is, to soe degree, based on their success in copleting this exercise progra. Therefore, further significant iproveent in exercise capacity that would result in changes in the BODE index would be unlikely. The clinical relationship of surgical LVRS to bronchoscopic LVRS is unclear. Pilot studies of bronchoscopic LVR [11 13] have been proising. However, randoized studies [14, 15] of bronchoscopic LVR have not, to date, been fruitful. No copleted randoize study has shown a clinically eaning benefit with these techniques. It will be iportant, given the good results and low risk associated with surgical LVRS deonstrated here, not to accept arginal iproveent fro a technique which offers little benefit in ters of procedural risk in siilar patients. Bronchoscopic LVR, if proven to be effective, ay show its greatest proise in patients unwilling or unable to undergo surgical LVR, and in patients who do not eet current selection criteria for LVRS. The statistical strength of previous studies of LVRS has been criticized because of incoplete follow-up data and the concern that survivorship bias ay favorably influence results. We were fortunate to have no surgically related ortality, to have excellent long-ter survival, and to have coplete clinical follow-up. Nonetheless, we perfored a worst rank score iputation, replacing issing BODE coponent data with worst in cohort results. This analysis deonstrated that LVRS still resulted in a clinically eaningful and statistically significant iproveent in the BODE index at 1 year. A further criticis of previous LVRS studies has been that a few patients with large iproveents in study endpoints accounted for the overall iproveent in results when expressed as ean changes. In our experience, the nuber of patients receiving at least a 1 point iproveent in BODE index was 92%, indicating that alost the entire cohort of patients were able to benefit fro LVRS. In conclusion, our experience has shown that patients with upper lobe predoinant ephysea eeting NETT-defined selection criteria and undergoing bilateral LVRS have a very low surgical risk with substantial iproveent in their easured pulonary functions and subjective dyspnea scores. These easured iproveents in clinical variables, as suarized by BODE index, predict ortality reduction fro COPD as has been docuented in the NETT trial. This experience further validates the results of NETT. Lung volue reduction surgery should be considered the treatent of choice in this patient population and the gold standard against which new technologies should be judged. We believe that thoughtful expansion of the procedure based on discoverable inforation concerning preoperative predictors of benefit should be pursued. Further analysis of the NETT database ay be fruitful and should be encouraged. Evolution of this proven therapy and steady iproveent in clinical results should allow greater acceptance of LVRS by the COPD counity. References 1. Suary Health Statistics for U.S, Adults: National Health Interview Survey U.S. Departent of Health and Huan Services, Centers for Disease Control and Prevention, National Center for Health Statistics. 2. Cooper JD, Trulock EP, Triantafillou AN, et al. Bilateral pneuectoy (volue reduction) for chronic obstructive pulonary disease. J Thorac Cardiovasc Surg 1995;109: Rationale and design of The National Ephysea Treatent Trial: a prospective randoized trial of lung volue reduction surgery. The National Ephysea Treatent Trial Research Group. Chest 1999;116: Edwards MA, Hazelrigg S, Naunhei KS. The National Ephysea Treatent Trial: suary and update. Thorac Surg Clin 2009;19: Naunhei KS, Wood DE, Mohsenifar Z, et al. Long-ter follow-up of patients receiving lung-volue-reduction surgery versus edical therapy for severe ephysea by the National Ephysea Treatent Trial Research Group. Ann Thorac Surg 2006;82: Medicare Part B Physician/Supplier BESS Data for Calendar Year Available at: NonIdentifiableDataFiles/04_5_PartBCarrierSuaryDataFile. asp. Accessed Septeber Celli BR, Cote CG, Marin JM, et al. The body-ass index, airflow obstruction, dyspnea, and exercise capacity index in chronic obstructive pulonary disease. N Engl J Med 2004; 350: Lederer DJ, Thoashow BM, Ginsburg ME, et al. Lungvolue reduction surgery for pulonary ephysea: Iproveent in body ass index, airflow obstruction, dyspnea, and exercise capacity index after 1 year. J Thorac Cardiovasc Surg 2007;133: Martinez FJ, Han MK, Andrei AC, et al. Longitudinal change in the BODE index predicts ortality in severe ephysea. A J Respir Crit Care Med 2008;178: Ifeld S, Bloch KE, Weder W, Russi EW. The BODE index after lung volue reduction surgery correlates with survival. Chest 2006;129: Wan IY, Toa TP, Geddes DM, et al. Bronchoscopic lung volue reduction for end-stage ephysea: report on the first 98 patients. Chest 2006;129: Steran DH, Mehta AC, Wood DE, et al. A ulticenter pilot study of a bronchial valve for the treatent of severe ephysea. Respiration 2010;79: Lausberg HF, Chino K, Patterson GA, Meyers BF, Toeniskoetter PD, Cooper JD. Bronchial fenestration iproves expiratory flow in ephyseatous huan lungs. Ann Thorac Surg 2003;75: Herth FJ, Gopelann D, Ernst A, Eberhardt R. Endoscopic lung volue reduction. Respiration 2010;79: Toa TP. Lessons fro a valve trial. Respiration 2010;79:

6 Ann Thorac Surg GINSBURG ET AL 2011;91: LVRS USING NETT CRITERIA 1561 DISCUSSION DR DAVID H. HARPOLE, JR. (Durha, NC): Just as a coent, this is a sobering effect of what has happened with this procedure. This is a ajor center and one of the international leaders of LVRS [lung volue reduction surgery], and I guess I counting about 9 a year that you re doing. In our institution we certainly do about 8 to 10 a year as well. It s aazing how the enthusias of this procedure has been quelled by our pulonary colleagues. DR SCOTT J. SWANSON (Boston, MA): Were you using reinforced strips on the staple lines or not? DR GINSBURG: We have reinforced all of the staple lines with a variety of products. DR SWANSON: Is there one that you favor over any other? DR GINSBURG: I don t think so. Dr Harpole reebers the early days when we had to hand sew the bovine patches and each staple line would take forever to do. There are now a nuber of coercially available reinforceents that coe preloaded. It akes a big difference. DR SWANSON: So you think they are pretty equivalent? DR GINSBURG: Pretty equivalent, yes. DR AKIF TURNA (Istanbul, Turkey): What percentage of patients copleted a pulonary rehabilitation progra before the operation? DR GINSBURG: We think this is very uch a palliative operation. All patients, when we evaluate the, we will axiize their edical therapy, which we think includes pulonary rehabilitation, and reevaluate the after pulonary rehabilitation to see if the surgery is really indicated. So, all patients have pulonary rehabilitation. DR MALCOLM M. DECAMP, JR. (Chicago, IL): Excellent review, Mark. Unfortunately, we re preaching to the choir and not to the people who need to hear this. I have one question for you. During this sae tie frae, you participated in all the relevant bronchoscopic approaches to volue reduction. How did you select patients for either one of those trials versus your excellent progra in surgical lung volue reduction? DR GINSBURG: I think we really believe that surgical lung reduction for a certain select, highly heterogeneous group is the best way to go. On the other hand, we, as you, see a lot of patients who are really not great candidates for surgical lung reduction but were eligible for these trials, and we offered nearly everyone that was eligible entry into the trials. DR DECAMP: And did patients express a bias on their end? I know your pulonary colleagues have pretty pro-surgery views about this, but did you have pushback fro the patients, saying, Why don t we try one of these bronchoscopic interventions? DR GINSBURG: A lot of the patients cae to us requesting the bronchoscopic approach. I think the local pulonologists have really been pushed in that direction. To date, as you know, none of the trials in bronchoscopic lung reduction have worked out, but a lot of the patients really cae hoping to get bronchoscopic LVRS. DR DECAMP: Have any of those bronchoscopic patients coe back now for surgery? DR GINSBURG: We have done one so far, a patient fro the VENT [Valve for Ephysea Palliation Trial] trial, and we could not reove the valves at the tie of surgery. DR SWANSON: Since it s beneficial, what do you think we should be doing to try to get the volue of this up? Do you have any ideas on that? DR GINSBURG: It s unclear what the real nubers should be. I was speaking to Dr Cooper yesterday and we both cae up with a couple of thousand patients per year as probably a reasonable nuber. According to the CDC [Centers for Disease Control], there are 1.8 illion patients with ephysea in the US, but the real nuber of patients who ight really benefit fro LVRS, if you re very selective, is probably not a very large nuber; it ay be 2,000 or 3,000 in the United States. I think LVRS has a big PR proble, and if you speak to the local pulonologists, they will tell you they either don t believe it works or it s too difficult to select patients. I think part of the proble is related to soe bad results in the 90s, and I think we have to reacquaint the to LVRS by reporting good results, and deonstrating to the that this operation is very effective treatent for ephysea and a very safe treatent for a select group of patients.