Lung Volume Reduction Surgery
Is Less Really More?

HENRY E. FESSLER and ROBERT A. WISE

Division of Pulmonary and Critical Care Medicine, Johns Hopkins Medical Institutions, Baltimore, Maryland

	INTRODUCTION

TOP INTRODUCTION INTRODUCTION MORTALITY SHORT-TERM EFFECTS OF LVRS... DURATION OF BENEFIT CONCLUSION REFERENCES

"Less is more."

Ludwig Mies van der Rohe, architect, quoting Robert Browning

	INTRODUCTION

TOP INTRODUCTION INTRODUCTION MORTALITY SHORT-TERM EFFECTS OF LVRS... DURATION OF BENEFIT CONCLUSION REFERENCES

New pharmacotherapy or medical devices require extensive testing, mandated and reviewed by the Federal Drug Administration and by local institutional review boards. New surgical therapy, in contrast, may be tried on willing patients with comparatively little oversight. Apparently successful procedures are repeated, and case series are subjected to peer review, publication, and duplication or refinement by other surgeons. This system fosters innovation and creativity. However, it also risks the widespread application of procedures whose benefits compared with established therapy are untested. Coronary artery bypass grafting, carotid end-arterectomy, and intracranial/extracranial bypass are examples of procedures that were performed on thousands of patients before their indications, or lack thereof, were known. The pressure for premature dissemination of a surgical procedure is greatest when the surgery shows promise for a prevalent disease of high morbidity, for which available therapy is often unsatisfactory. This precisely describes the current status of the surgical treatment of emphysema.

In 1995, Cooper and colleagues reported the results of a procedure they termed "reduction pneumectomy" in 20 patients with severe emphysema (1). Cooper and coworkers resected the most severely involved regions of both lungs using an automated stapler through a median sternotomy. Their patients showed improvements in lung function, indices of dyspnea and health-related quality of life, and oxygen and corticosteroid use at 3 mo after surgery.

Based on these findings, other surgeons began offering a variety of similar procedures to broadly similar, severely disabled patients. Demand was fueled by hospital marketing efforts and reports in the lay press (2, 3). The initial study was soon joined by many others in the medical literature with uniformly positive findings (4). Renamed lung volume reduction surgery (LVRS), this group of procedures was acknowledged by the Health Care Financing Administration (HCFA) with a Medicare International Classification of Diseases, 9th Revision (ICD-9) billing code. In the next 3 mo, 722 such procedures were billed, exclusive of additional LVRS procedures billed under other codes or performed on patients covered by other insurers.

Some, however, counseled caution. A National Institutes of Health (NIH) workshop in September 1995 concluded that, although initial results were promising, LVRS was often being performed with insufficient evaluation, and that a randomized study "should be undertaken to evaluate the procedure critically" (9). The Center for Health Care Technology noted significant morbidity and mortality, and concluded that the available data did not permit "a logical and scientifically defensible conclusion regarding the risks and benefits of LVRS" (10). HCFA ruled in December 1995 that LVRS would not be covered by Medicare. Although patient volumes decreased, surgery has continued on patients whose insurers would cover the expense or who could pay out-of-pocket. The medical literature on the procedure has continued to swell. A Medline search of the topic reveals five articles in 1995, 41 in 1996 and 66 in six languages in 1997, and the duration of published follow-up has increased.

In the interim, HCFA has joined with the National Heart, Lung, and Blood Institute (NHLBI) to sponsor a randomized trial comparing LVRS to optimal medical therapy for emphysema, the National Emphysema Treatment Trial (NETT). This landmark multicenter study is unique in several respects. It will be a controlled trial of LVRS in which suitable candidates will be randomly assigned to management with or without surgery. It will set a standard for optimal medical therapy and pulmonary rehabilitation for the emphysema patient. It is the most complex, and possibly the most expensive clinical trial by the pulmonary community. It is also a unique collaboration between the NIH and Medicare which could serve as the model for evaluation of future innovative medical or surgical interventions. Similar but smaller studies are being conducted by Blue Cross/Blue Shield of Massachusetts and in Canada.

This high-profile study has attracted the participation of numerous prominent investigators and the press (11). Medicare coverage of LVRS was discussed before Congress, the NHLBI has convened consultants to review ethical issues of the study, physicians have been confronted by frustrated patients who suspect effective therapy is being withheld by Medicare for fiscal reasons. The Society of Thoracic Surgeons has an Internet site for discussion of LVRS and the NETT by patients and their families which has hundreds of postings, many expressing this frustration.

As this trial begins, it is appropriate to consider what we have learned about LVRS over the last few years. We will focus on only three of the many unanswered questions: First, how does the procedure affect mortality; second, how does the procedure affect short-term pulmonary function; and third, how does the resection affect the subsequent rate of change of lung function? Our goal is to distinguish what is known about LVRS from what is merely hoped for.

	MORTALITY

TOP INTRODUCTION INTRODUCTION MORTALITY SHORT-TERM EFFECTS OF LVRS... DURATION OF BENEFIT CONCLUSION REFERENCES

Mortality from Emphysema

It is impossible to interpret LVRS mortality without knowing the mortality of similar patients not receiving surgery. Unfortunately, there is little in the literature to address that question. Epidemiologic studies have largely been performed in patients with chronic obstructive pulmonary disease (COPD) (12). This includes patients not only with emphysema but also chronic bronchitis, bronchiectasis, and reactive airways disease. All of the latter patients would be excluded from LVRS. Patients are also included in these population-based studies with conditions such as coronary artery disease or previous chest surgery which would make them poor LVRS candidates (18), but could affect their mortality rates. In contrast, LVRS patients are typically screened to have emphysema as their exclusive pulmonary disorder, to lie within a window of severity, to be free of comorbidities, and no longer smoking (1, 6, 8, 9, 19).

Several features that are either selection or exclusion criteria for LVRS have been shown to affect mortality in patients with COPD. Common LVRS exclusions include extreme age and cor pulmonale (correlates of increased mortality [13]), and airways hyperreactivity (decreased mortality [16, 22]). Characteristics sought in LVRS patients include elevated TLC and residual volume (RV) and disabling dyspnea, all associated with increased COPD mortality (13, 16). Patients undergoing LVRS also receive a variety of nonsurgical interventions such as increased attention to their medical treatment, pulmonary rehabilitation, and dietary and psychological counseling. Thus, it is likely that mortality among LVRS candidates will be quite different from "historical controls," and further that it may be influenced by the nonsurgical aspects of their care.

With these caveats, it may nevertheless be enlightening to review the available mortality data in COPD to provide a broad comparison. Burrows and coworkers studied 300 obstructed patients with FEV₁ < 60% predicted in two series from Chicago and Tucson (12, 13). All subjects were symptomatic and had sought medical care for their lung disease, had "irreversible" obstruction, and were free from other progressive or fatal illnesses. Smokers were included, and there was no attempt to standardize therapy. FEV₁ averaged 33.5% predicted on entry in both series and mortality was about 10% per year over the first 5 yr.

In another study, Burrows and colleagues followed 120 white patients with an FEV₁ < 65% predicted and who had seen a physician for asthma, chronic bronchitis, emphysema, or "bronchial trouble." They further stratified this group clinically into those who appeared to have "asthmatic bronchitis" and those primarily with emphysema. In the latter group, mean FEV₁ was 47% predicted, more than the minimum usually used to identify LVRS candidates, and over half were current smokers. Annualized mortality was 5.3% (22).

In the Nocturnal Oxygen Therapy Trial, subjects had airway obstruction, an average FEV₁ 30% predicted, hypoxemia, and were free from other illnesses expected to affect mortality. One-year mortality in the group receiving continuous O₂ was 11.9% (17). In the Intermittent Positive Pressure Beathing Trial, 985 nonhypoxemic patients with COPD were followed for approximately 3 yr. FEV₁ averaged 41% predicted, and 40% were current smokers. When stratified by FEV₁ and age, those older than 60 yr with FEV₁ 30 to 39% predicted had about 10% per year mortality, increasing to almost twice that when initial FEV₁ was below 30% predicted (16).

Thus, the study by Burrows is the only one that focused on emphysema and included smokers and others far less ill than typical LVRS candidates. Studies of patients with COPD of about the same severity as LVRS candidates included smokers, differing age groups, differing interventions, and unknown degrees of other obstructive lung disease. Although an annual mortality of about 10% is common to several studies, it is uncertain how that would compare with the strictly defined subset of patients typically selected for LVRS. Perhaps the best comparison group to candidates for LVRS are those who qualify for lung transplantation for emphysema. In the United Network for Organ Sharing registry for patients awaiting lung transplantation, the annual mortality is about 7 to 8% (23).

Mortality from LVRS

Any major surgery in these fragile patients may cause the death of some. Thirty-day mortality, the classic surgical benchmark, may not capture all deaths attributable to LVRS, as some follow more prolonged hospitalization (6). Hospital mortality is a preferable statistic, but may still miss patients who die after transfer to chronic care facilities. By any such measure, LVRS could be expected to increase early mortality in emphysema patients.

Published operative mortality rates for LVRS have varied from zero to 19% (1, 4, 6, 8, 20, 24, 25). Variability may be due to small patient numbers, different definitions of surgical mortality, different patient selection, or differences in surgical technique, experience, or perioperative management. A few characteristics may identify patients who have a higher likelihood of perioperative death, including advanced age, severe hypercarbia, and extremely limited exercise capacity (6, 25). This suggests that patients dying in the perioperative period are not just a random sample of the operated patients, subject to the vagaries of infection, air leak, pulmonary embolism, or other misfortune. For equivalent FEV₁, age, hypercarbia, and extreme exercise limitation identify COPD patients more likely to die even without surgery (13, 16). The interpretation of survival statistics after initial surgical mortality can be biased toward improved postoperative survival when patients at highest risk are selected out by the surgical procedure, so-called survivor bias. The comparison of surgical and nonsurgical treatments, therefore, is not straightforward. Even with data acquired from randomized controlled clinical trials, evaluation of efficacy depends upon the time frame of the analysis, the value judgments of the physicians and patients, and the cost-utility of the procedure (26).

Among survivors of LVRS, FEV₁ is frequently increased. Because studies in COPD have shown low FEV₁ to be a strong predictor of mortality (13, 16), it is a logical hypothesis that LVRS decreases mortality in patients with emphysema. However, consideration must be given to survivor bias as well as potential benefits of the associated intensive medical management.

There are scant published data on 1-yr outcome after LVRS. Such longitudinal data are very sensitive to the problem of incomplete follow-up. Despite intense effort, postoperative outpatient follow-up is as low as 70% after LVRS (27). One organization which has complete vital status on its constituents is Medicare. HCFA recently reviewed the outcome of 722 patients who had undergone LVRS between October 1995 and January 1996, when a specific billing code existed. At 12 mo after surgery, 23% had died, and mortality was 28% at 18 mo. In addition, acute hospitalizations in this cohort increased from 197 averaging 11.8 d in the year before surgery to 304 averaging 18.7 d in the year after, exclusive of the surgery itself (HCFA Report to Congress, "LVRS and Medicare Coverage Policy: Implications of Recently Published Evidence," August 1998). The reasons for this high rate versus those reported in the medical literature are uncertain, but could certainly include a high mortality during the learning period at many hospitals with little LVRS experience or the tendency of hospitals with poor outcome not to report it in the medical literature.

	SHORT-TERM EFFECTS OF LVRS ON PULMONARY FUNCTION

TOP INTRODUCTION INTRODUCTION MORTALITY SHORT-TERM EFFECTS OF LVRS... DURATION OF BENEFIT CONCLUSION REFERENCES

There is little doubt that LVRS causes mean changes in spirometry and other pulmonary functions which are both statistically and clinically significant at 3 to 6 mo after surgery (1, 6, 8, 9, 19, 20, 25). While it is arguable which specific pulmonary function is most important, we will limit our consideration to the changes in FEV₁. This is perhaps the most widely used measure of the severity of airflow obstruction, and is the most consistently reported variable on patients who have undergone LVRS.

The first series by Cooper and coworkers reported an 82% mean improvement in FEV₁ in 20 patients (1). All subsequent reports have been less striking, with mean increases of 25 to 58% measured 3 to 6 mo after bilateral surgery (1, 4, 6, 7, 20, 24, 25, 28, 29). It is often not specified whether the preoperative measurements have been made after medical therapy was optimized or are pre- or postbronchodilator. Nevertheless, these are important changes in patients whose preoperative FEV₁ has averaged about 0.6 to 0.8 L, or 20 to 30% predicted.

The focus on mean values, however, can obscure important information contained in the variance about the mean. In many studies, the standard deviation around the mean percent improvement is not provided (4, 7, 24). However, consideration of a few examples will be illustrative. In 56 patients reported by Kotloff and coworkers, FEV₁ improved by 41.4 ± 37.3% after bilateral LVRS by median sternotomy. The frequency histogram of their results (their Figure 1) indicates approximately 30% of patients with less than 20% improvement in FEV₁ (6). Argenziano and coworkers reported results in 51 patients after bilateral LVRS. FEV₁ increased 58 ± 63% (28). If we assume that these outcomes were normally distributed, the fraction of patients with minimal improvement can be calculated from statistical tables. In about 23% of these patients, FEV₁ improved by less than 12%, the American Thoracic Society (ATS) definition of a bronchodilator response. In 18% of patients, FEV₁ would have been reduced. In another series in which patients were stratified by radiologic distribution of emphysema, 28% of 32 patients with diffuse or lower lobe emphysema and 12% of 106 patients with predominantly upper lobe emphysema improved FEV₁ by less than 12% (29). Martinez and coworkers reported that in six of 17 patients, FEV₁ increased by less than 20% (7). Half of a series of 29 patients reported by Ingenito and coworkers failed to increase their FEV₁ by more than 12% (30). The most detailed pulmonary function results are provided for each of 25 patients studied by Kellar and coworkers. Eight increased FEV₁ by less than 15% (5). Thus, it appears that about 20% to as much as 50% of patients show little to no improvement in FEV₁.

View larger version (22K): [in this window] [in a new window]   Figure 1.   The annual loss of one second forced expiratory volume (FEV1 slope) in patients with COPD or following LVRS for emphysema. Numbers in parentheses refer to the reference; JHH indicates experience from Johns Hopkins Hospital.

While acknowledging that changes in FEV₁ do not always correspond to changes in dyspnea, one of the goals of the NETT is to help refine patient selection criteria to narrow this range of potential outcomes. This range is often more important to patients than the mean improvement. Most patients considering LVRS are profoundly symptomatic. While describing the risks of the procedure, we have often been told by our patients that they do not care very much about the risk of dying from the operation. Most are willing to assume a substantial risk of death for relief from their dyspnea. However, few realize that their chance of failing to improve, or even worsening their lung function, exceeds their risk of not surviving to hospital discharge. They are less enthusiastic when they learn this, and for that very reason this is an essential element of preoperative informed consent.

	DURATION OF BENEFIT

TOP INTRODUCTION INTRODUCTION MORTALITY SHORT-TERM EFFECTS OF LVRS... DURATION OF BENEFIT CONCLUSION REFERENCES

For patients considering LVRS, a pressing concern is how long they might expect any improvements to last. There are extremely limited data on this issue. Again, we will focus on the FEV₁. It has long been recognized that FEV₁ declines with age, and that it declines more rapidly in patients with COPD and in smokers than in nonsmokers. The annual decline ("FEV₁ slope") in healthy nonsmokers is about 20 to 40 ml/yr (14, 31). In patients with COPD, the most extensive data come from the recent Lung Health Study. In that 5-yr study of patients with early COPD, FEV₁ slope is reported by smoking status for the 1,962 patients completing a smoking cessation program. FEV₁ declined 63 ml/yr in continued smokers and 34 ml/yr in those who successfully quit (15). This agrees with earlier figures from Fletcher and Peto, who found the annual decrease ranged from 37 ml in ex-smokers to 80 ml in heavy smokers with mild obstruction (14). Other studies in patients with COPD show rates of decline of about 50 ml/yr, without stratifying for smoking status (12). For comparison to LVRS series, these figures suffer the same limitations as mortality, namely that the patients are not entirely comparable.

In the few studies that provide 12 mo or longer follow-up after LVRS, FEV₁ generally peaks at about 6 mo but remains improved over baseline for at least 1 yr. The loss of lung function after 6 mo, however, greatly exceeds rates reported in COPD, even among continued smokers. Brenner and coworkers found an average loss of FEV₁ of 163 ml/yr in 180 patients who had undergone a variety of LVRS procedures. The most rapid loss (255 ml/yr) occurred in patients who had undergone bilateral thoracoscopic LVRS, the procedure producing the greatest improvement in FEV₁ (390 ± 30 ml) (21). In a cohort of 56 patients, Cooper and coworkers reported FEV₁ improving from 0.69 L preoperatively to 1.1 L at 6 mo and 1.0 L at 12 mo (standard deviation and statistical significance not reported) (4). Thus, the annualized decline based on the change from 6 to 12 mo is 200 ml. In a more recent 3-yr follow-up report of 25 of their early patients, the FEV₁ slope for the last 2 yr is 120 ml/yr (32). In one abstract, Ingenito and coworkers reported a decline in FEV₁ from 0.75 L at 6 to 12 mo to 0.61 L at 24 mo postoperatively in nine patients (p = 0.046) (33). This yields a FEV₁ slope of 93 to 140 ml/yr, depending on whether the initial measurement was at 6 or 12 mo. These researchers also comment that the patients showing the greatest early improvement in FEV₁ manifested the greatest loss over time. In 13 patients, Cordova and coworkers reported an FEV₁ loss of 140 ml/yr (34). Gelb and coworkers performed detailed lung mechanics measurements in 10 patients preoperatively, and at 6 and 12 mo. In this group, the annualized fall in FEV₁ beyond 6 mo was 400 ml/yr (35). At Johns Hopkins Hospital, 15 patients have had follow-up spirometry up to 32 mo after bilateral LVRS, in addition to at least one earlier postoperative study. The average interval between pulmonary function testing was 8 mo. The annualized rate of FEV₁ decline was 198 ml. Thus, early data suggest that LVRS may accelerate, perhaps markedly, the functional deterioration of the lung left behind (Figure 1).

There are speculative explanations for the accelerated FEV₁ decline after LVRS. The lung parenchyma is subjected to stress during daily respiratory activity. This may contribute to the normal, age-related increase in alveolar size and decline in lung elasticity. When alveolar walls are damaged by smoking or other insults, the rate of stress-induced failure would be increased and emphysema could worsen. LVRS, by increasing overall lung recoil, would also increase stress on individual alveolar elements during tidal breathing. This could increase their likelihood of failure and lead to more rapid development of emphysema in the remaining lung.

These estimates of FEV₁ slope are useful, because they allow some estimate of the time it will take for FEV₁ to reach its preoperative value. However, it is probably incorrect to assume that the postoperative decline will be linear. It will also be important to learn if the FEV₁ slope returns to its preoperative rate when the FEV₁ has reached its preoperative value. If not, then late morbidity and mortality could be increased, even in patients who achieve an early benefit. Finally, despite selection of a fairly homogenous group of patients, there is a wide range of short-term changes in FEV₁ after LVRS. Thus, it is likely that there will also be a wide range of postoperative FEV₁ slopes. Just as investigators are now searching for characteristics that will identify patients likely to have a good short-term outcome, it will be necessary to characterize those whose improvement is likely to endure.

	CONCLUSION

TOP INTRODUCTION INTRODUCTION MORTALITY SHORT-TERM EFFECTS OF LVRS... DURATION OF BENEFIT CONCLUSION REFERENCES

Over the past 2 years, LVRS has been widely practiced. It is supported by numerous case series. It has been aggressively marketed by many hospitals, and extensively reported in the lay and scientific press. Although the procedure shows great promise and justifies great interest, we believe that the ultimate benefit and appropriate patient selection criteria are unknown.

There have been no controlled, randomized trials of LVRS. We have shown that the use of historical controls is inappropriate because historical controls differ from LVRS candidates in several important characteristics that influence mortality. LVRS can be expected to alter natural history mortality rates in complex ways which are not easily predicted. These problems become even greater when one considers less precise outcome variables such as exercise capacity or symptoms, which are subject to motivational or placebo effects.

In groups of patients, LVRS causes mean increases in FEV₁ that are not likely a result of accompanying improvements in medical management alone. However, such mean changes do not convey the large fraction of patients who gain little, or whose function deteriorates after surgery. Awareness of that risk is essential to patient decision-making and informed consent.

There is little information about long-term duration of benefit or effects of surgery on the remaining lung tissue. It is likely that these effects will show a wide range of variability, but the existing evidence suggests that the benefits will not be sustained. After only 3 years or less of experience, it is premature to conclude that LVRS is effective therapy for a chronic illness of decades' duration. The results of ongoing clinical trials will be necessary to evaluate the effectiveness of LVRS and characteristics of patients most likely to benefit from the procedure. However, because of the surgical mortality, and the variability of both the FEV₁ response and its duration, it is likely that the ultimate evaluation of this procedure will also require judgments based upon risk aversion, quality of life, and functional status.

	Footnotes

Correspondence and requests for reprints should be addressed to Henry E. Fessler, M.D., Pulmonary and Critical Care Medicine, Johns Hopkins HospitalBlalock 910, 600 North Wolfe St., Baltimore, MD 21287. E-mail: hfessler{at}welchlink.welch.jhu.edu

(Received in original form August 13, 1998 and in revised form October 22, 1998).

Acknowledgments: The authors are participants in the National Emphysema Treatment Trial. This manuscript reflects their opinion, and not the official opinion of the NETT investigators. The authors are grateful for the expert secretarial assistance of Ms. Cara Zbylut.

	References

TOP INTRODUCTION INTRODUCTION MORTALITY SHORT-TERM EFFECTS OF LVRS... DURATION OF BENEFIT CONCLUSION REFERENCES

1. Cooper, J. D., E. P. Trulock, A. N. Triantafillou, G. A. Patterson, M. Pohl, P. Deloney, R. Sundaresan, and C. Roper. 1995. Bilateral pneumectomy (volume reduction) for chronic obstructive pulmonary disease. J. Thorac. Cardiovasc. Surg. 109: 106-119 [Abstract/Free Full Text].

2. Surgery may offer emphysema relief. USA Today. May 14, 1996.

3. Lung reduction surgery helps severe emphysema. Associated Press. April 25, 1996.

4. Cooper, J. D., G. A. Patterson, R. S. Sundaresan, E. Trulock, R. Yusen, M. Pohl, and S. Lefrak. 1996. Results of 150 consecutive bilateral lung volume reduction procedures in patients with severe emphysema. J. Thorac. Cardiovasc. Surg. 112: 1319-1330 [Abstract/Free Full Text].

5. Keller, C. A., G. Ruppel, A. Hibbett, J. Osterloh, and K. S. Naunheim. 1997. Thoracoscopic lung volume reduction surgery reduces dyspnea and improves exercise capacity in patients with emphysema. Am. J. Respir. Crit. Care Med. 156: 60-67 [Abstract/Free Full Text].

6. Kotloff, R. M., G. Tino, J. E. Bavaria, H. Palevsky, J. Hansen-Flaschen, P. Wahl, and L. Kaise. 1996. Bilateral lung volume reduction surgery for advanced emphysema. Chest 110: 1399-1406 [Abstract/Free Full Text].

7. Martinez, F. J., M. Montes de Oca, R. I. Whyte, J. Stetz, S. E. Gay, and B. R. Celli. 1997. Lung-volume reduction improves dyspnea, dynamic hyperinflation, and respiratory muscle function. Am. J. Respir. Crit. Care Med. 155: 1984-1990 [Abstract].

8. McKenna, R. J., M. Brenner, A. F. Gelb, M. Mullin, N. Singh, H. Peters, J. Panzera, J. Calmese, and M. Schein. 1996. A randomized, prospective trial of stapled lung reduction versus laser bullectomy for diffuse emphysema. J. Thorac. Cardiovasc. Surg. 111: 317-322 [Abstract/Free Full Text].

9. Weinmann, G. G., and R. Hyatt. 1996. Evaluation and research in lung volume reduction surgery. Am. J. Respir. Crit. Care Med. 154: 1913-1918 [Medline].

10. Holohan, T. V., and H. Handelsman. 1996. Lung-volume reduction surgery for end-stage chronic obstructive pulmonary disease. Health Technol. Assess. 10: 1-30 .

11. Gentry, C. Why Medicare covers a new lung surgery for just a few patients. Wall Street J. June 29, 1998.

12. Diener, C. F., and B. Burrows. 1975. Further observations on the course and prognosis of chronic obstructive lung disease. Am. Rev. Respir. Dis. 111: 719-724 [Medline].

13. Traver, G. A., M. G. Cline, and B. Burrows. 1979. Predictors of mortality in chronic obstructive pulmonary disease. Am. Rev. Respir. Dis. 119: 895-902 [Medline].

14. Fletcher, C., and R. Peto. 1977. The natural history of chronic airflow obstruction. B.M.J. 1: 1645-1648 .

15. Anthonisen, N. R., J. E. Connett, J. P. Kiley, M. Attose, W. Bailey, A. S. Buist, W. Conway, P. Enright, R. Kanner, P. O'Hara, G. Owens, P. Scanlon, D. Tashkin, and R. Wise. 1994. Effects of smoking intervention and the use of an inhaled anticholinergic bronchodilator on the rate of decline of FEV₁. J.A.M.A. 272: 1497-1505 [Abstract/Free Full Text].

16. Anthonisen, N. R., E. C. Wright, J. E. Hodgkin, and the IPPB Trial Group. 1986. Prognosis in chronic obstructive pulmonary disease. Am. Rev. Respir. Dis. 133: 14-20 [Medline].

17. Nocturnal Oxygen Therapy Trial Group. 1980. Continuous or nocturnal oxygen therapy in hypoxemic chronic obstructive lung disease. Ann. Intern. Med. 93: 391-398 .

18. Thurnheer, R., J. Muntwyler, U. Stammberger, K. Bloch, A. Zollinger, W. Weder, and E. Russi. 1997. Coronary artery disease in patients undergoing lung volume reduction surgery for emphysema. Chest 112: 122-128 [Abstract/Free Full Text].

19. Brenner, M., R. McKenna, A. Gelb, K. Osann, M. Schein, J. Panzera, H. Wong, M. Berns, and A. Wilson. 1997. Objective predictors of response for staple versus laser emphysematous lung reduction. Am. J. Respir. Crit. Care Med. 155: 1295-1301 [Abstract].

20. Argenziano, M., B. Thomashow, P. A. Jellen, E. Rose, K. Steinglass, M. Ginsburg, and L. Gorenstein. 1997. Functional comparison of unilateral versus bilateral lung volume reduction surgery. Ann. Thorac. Surg. 64: 321-327 [Abstract/Free Full Text].

21. Brenner, M., R. J. McKenna, A. F. Gelb, R. J. Fischel, and A. F. Wilson. 1998. Rate of FEV₁ changes following lung volume reduction surgery. Chest 113: 652-659 [Abstract/Free Full Text].

22. Burrows, B., J. W. Bloom, G. A. Traver, and M. G. Cline. 1987. The course and prognosis of different forms of chronic airways obstruction in a sample from the general population. N. Engl. J. Med. 317: 1309-1314 [Abstract].

23. Hosenpud, J. D., L. E. Bennett, B. M. Keck, E. B. Edwards, and R. J. Novick. 1998. Effect of diagnosis on survival benefit of lung transplantation for end-stage lung disease. Lancet 351: 24-37 [Medline].

24. McKenna, R. J., M. Brenner, R. J. Fischel, and A. F. Gelb. 1996. Should lung volume reduction for emphysema be unilateral or bilateral. J. Thorac. Cardiovasc. Surg. 112: 1331-1339 [Abstract/Free Full Text].

25. Szekely, L. A., D. A. Oelberg, C. Wright, D. Johnson, J. Wain, B. Trotman-Dickenson, J. Shepard, D. Kanarek, D. Systrom, and L. Ginns. 1997. Preoperative predictors of operative morbidity and mortality in COPD patients undergoing bilateral lung volume reduction surgery. Chest 111: 550-558 [Abstract/Free Full Text].

26. Howard, G., L. E. Chambless, and R. A. Kronman. 1997. Assessing differences in clinical trials comparing surgical vs nonsurgical therapy: using common (statistical) sense. J.A.M.A. 278: 1432-1436 [Abstract/Free Full Text].

27. Butler, C. W., P. Snyder, S. Lewis, D. Wood, E. Vallieres, R. Albert, and J. Benditt. 1998. Patient-reported reasons for failure to follow-up after a lung volume reduction surgery (LVRS) quality of life study (QOL) (abstract). Am. J. Respir. Crit. Care Med. 157: A497 .

28. Argenziano, M., N. Moazami, B. Thomashow, P. Jellen, L. Gorenstein, E. Rose, A. Weinberg, K. Steinglass, and M. Ginsburg. 1996. Extended indications for lung volume reduction surgery in advanced emphysema. Ann. Thorac. Surg. 62: 1588-1597 [Abstract/Free Full Text].

29. McKenna, R. J., M. Brenner, R. J. Fischel, N. Singh, B. Yoong, A. Gelb, and K. Osann. 1997. Patient selection criteria for lung volume reduction surgery. J. Thorac. Cardiovasc. Surg. 114: 957-967 [Abstract/Free Full Text].

30. Ingenito, E. P., R. B. Evans, S. H. Loring, D. Kaczka, J. Rodenhouse, S. Body, D. Sugarbaker, S. Mentzer, M. DeCamp, and J. Reilly. 1998. Relation between preoperative inspiratory lung resistance and the outcome of lung-volume-reduction surgery for emphysema. N. Engl. J. Med. 338: 1181-1185 [Abstract/Free Full Text].

31. Burrows, B., M. D. Lebowitz, A. E. Camilli, and R. J. Knudson. 1986. Longitudinal changes in forced expiratory volume in one second in adults. Am. Rev. Respir. Dis. 133: 974-980 [Medline].

32. Yusen, R. D., M. S. Pohl, V. Richardson, D. Biggar, G. Patterson, E. Trulock, S. Lefrak, and J. Cooper. 1998. 3-year results after lung volume reduction surgery (abstract). Am. J. Respir. Crit. Care Med. 157: A335 .

33. Ingenito, E. P., S. H. Loring, J. D. Rodenhouse, M. Moy, M. Connolly, S. Mentzer, and J. Reilly. 1998. Persistence of spirometry and lung volume response to LVRS (abstract). Am. J. Respir. Crit. Care Med. 157: A497 .

34. Cordova, F., G. O'Brien, S. Furukawa, A. Kuzma, J. Travaline, and G. Criner. 1997. Stability of improvements in exercise performance and quality of life following bilateral lung volume reduction surgery in severe COPD. Chest 112: 907-915 [Abstract/Free Full Text].

35. Gelb, A. F., M. Brenner, R. J. McKenna, N. Zamel, R. Fischel, and J. Epstein. 1996. Lung function 12 months following emphysema resection. Chest 110: 1407-1415 [Abstract/Free Full Text].