You are leaving Medscape Education
Cancel Continue
Log in to save activities Your saved activities will show here so that you can easily access them whenever you're ready. Log in here CME & Education Log in to keep track of your credits.
 

 

New Data on Lung Volume Reduction Surgery and Outcomes From the National Emphysema Treatment Trial

Authors: Neil MacIntyre, MDFaculty and Disclosures

processing....

Introduction -- Lung Volume Reduction Surgery and the National Emphysema Treatment Trial

Lung volume reduction surgery (LVRS) is the technique of surgically (either with a sternotomy approach or a video-assisted thoracascopic technique) removing overinflated regions of emphysematous lungs. In patients with severe emphysema, LVRS may offer 2 benefits: (1) overinflated regions can be removed, thus allowing healthier regions to reexpand and be more functional; and (2) with less overinflation, the diaphragm will be in a more advantageous position to generate inspiratory pressures. A number of observational studies in the 1990s suggested that LVRS offered improved functional status in selected patients with severe emphysema. However, a careful review of the Medicare database in the late 1990s indicated that LVRS, when applied to a broader population of elderly patients, may in fact be detrimental to many of them.

To establish the safety and the functional outcomes of LVRS in Medicare patients, the National Institutes of Health, the Center for Medicare and Medicaid Services, and the Agency for Health Care Quality Research funded the National Emphysema Treatment Trial (NETT), a multicenter (17 sites), randomized trial comparing LVRS and optimal medical therapy. From 1997 to 2002, over 1200 patients were enrolled in the NETT and have been followed for as many as 6.8 years (average, 3.6 years) for mortality and for physiologic and functional outcomes.

An early interim analysis revealed a subgroup with a particularly poor surgical outcome (16% 30-day mortality).[1] This subgroup is characterized by forced expiratory volume in 1 second (FEV1) < 20% predicted and either a Dlco < 20% odds ratio homogeneous disease on chest computed tomography (CT). Analysis of outcomes in the remaining non-high-risk patients revealed an overall improvement in exercise tolerance and functional status (but not mortality) with LVRS.[2] However, prospectively defined subgroup analyses also revealed a clear mortality benefit (even accounting for early surgical mortality) for LVRS in patients with upper lobe predominant emphysema and poor exercise performance presurgery.[2]

There are clear opportunities to learn much more about emphysema and LVRS outcomes from the NETT database. To illustrate the richness of this database, there are 194,000 patient-encounter records, 1098 digitized CT scans, and over 13 million individual data items. Reviewed below are 2 analyses performed subsequent to the publication of the main outcome data with this database. Following these reviews is a brief commentary on the issue of assessing functional outcomes with the concept of a Minimal Clinically Important Difference (MCID).

Hemodynamic Consequences of LVRS

Emphysema and lung overinflation can compromise cardiovascular (CV) function in several ways. First, alveolar-capillary destruction can increase pulmonary vascular resistance (PVR) and load the right ventricle (RV); second, hypoxemia can further increase PVR and RV loading; third, hypoxemia can compromise myocardial oxygenation in the presence of coexistent coronary artery disease; fourth, hyperinflation increases intrathoracic pressure, which can reduce RV filling. LVRS conceptually could reverse many of these processes and thus improve CV function. However, it is also conceivable that further removal of pulmonary vasculature with LVRS may actually worsen PVR with attendant detrimental CV effects.

Analysis of the NETT data from 3 sites participating in a CV substudy has shed considerable light on this issue. In these 3 sites, patients received right heart catheterizations pre- and 6 months post LVRS. These patients also received all of the other NETT evaluation and follow-up procedures in addition to being randomized to medical vs surgical therapy. The NETT prerandomization demographics and functional assessments as well as the functional outcomes in these patients were all similar to patients from other NETT sites not in this CV substudy.

At baseline both the medical and the surgical patients in these 3 sites were well matched for lung impairment and CV function. For example, mean pulmonary artery pressures were 22-23 mm Hg, and normal cardiac indices were seen in both groups. The CV measurements from right heart catheterizations at 6 months post randomization were also very similar in both the surgical and the medical patients at these 3 sites. The only significant difference was a slightly lower pulmonary artery occlusion pressure in the surgical patients. Reductions in pulmonary artery occlusion pressure were correlated with other indices of lung function improvement. Thus, it appears that the major factors that improve exercise tolerance with LVRS are improvements in lung function, not CV function.

Predicting Perioperative Morbidity and Mortality

Older observational studies on LVRS reported perioperative mortality rates of 4% to 17% and varying rates of morbidities expected with LVRS (pneumonia, air leaks, and prolonged mechanical ventilation). These same reports also suggest that age, PCO2, FEV1, exercise tolerance, radiologic characteristics, and sex may all have some ability to predict these morbidities. However, because of small sample sizes and different entry criteria, it is difficult to generalize these results.

The perioperative mortality and morbidities have recently been analyzed in a NETT substudy to find correlations with a variety of potential preoperative predictors. The analysis was done on the 511 surgical patients who were not in the high-risk subgroup described above. It should be noted that the NETT surgeons believed these patients to be a particularly "fragile" group because they were elderly Medicare patients with high supplemental O2 usage and frequent oral steroid use.

Overall 90-day mortality in these NETT surgical patients was 5.5%. Only non-upper lobe disease predicted mortality in the multivariate analysis. Perioperative (up to 30 day) morbidities were frequent (59% had at least 1) with the most common being pneumonia (18%), reintubation (22%), prolonged mechanical ventilation requiring tracheostomy (8%), and arrhythmias requiring therapy (19%). In multivariate analyses, older age, low FEV1, and low Dlco predicted pulmonary morbidities; older age, O2 use, and non-upper lobe disease predicted CV morbidities.

Assessing Outcome Results

In assessing any medical or surgical intervention, ultimate clinical outcomes that are meaningful to a patient's mortality or quality of life should be the "gold standards." Unfortunately, in clinical trials, the readily available outcome data are usually short-term clinical or laboratory data. Although statistically significant improvements in 1 or more of these clinical or laboratory results may be detected in a large trial, the far more important question is whether these measured improvements translated into a meaningful outcome from the patient's perspective. Put another way, a key question following a "positive" clinical trial is whether the statistically positive results made the patient's life better.

The smallest measured improvement associated with a patient perceived benefit has been defined as the Minimal Clinically Important Difference MCID. Simply put, the MCID is the difference in a measurement that must be attained to produce a detectable benefit to the patient. Thus, an improvement in a parameter may be real (ie, statistically significant) but have no impact on the patient's life if the improvement is less than the MCID. An example may be that very small but statistically significant improvements in FEV1 detected in a very large trial of a new bronchodilator do not produce any meaningful impact on patients' lives if that FEV1 change is below the MCID. Defining "meaningful," of course, is a separate challenge, and although expert opinions are sometimes used, objective quality-of-life scores and hard data, such as mortality or healthcare utilization, are preferred.

The NETT provides a unique opportunity to assess MCID levels in chronic obstructive pulmonary disease (COPD) patients for a variety of physiologic measurements because both mortality and sophisticated quality-of-life assessments were also done. From these analyses, significant quality-of-life improvement required at least an improved 6-minute walk distance of 121 feet, a PO2 increase of 5 mm Hg, an FEV1 increase of .12 L, and a 5-W increase in a maximal exercise test. These observations underscore the importance of using a MCID "yardstick" in addition to statistical significance to evaluate clinical trial results.

Conclusion

The NETT has been an important contribution to our understanding of the pathophysiology and management of COPD. It is a large clinical trial that has demonstrated the role of LVRS in elderly emphysema patients, including important outcome benefits from LVRS in selected subgroups. It has been more than this, however. NETT has provided a variety of insights into the pathophysiology of emphysema. NETT has also demonstrated the value of several governmental agencies working together to support important clinical trials. And finally, NETT has given insight into the relationship of physiologic outcomes and functional outcomes that are meaningful to patients and payers.

References

  1. National Emphysema Treatment Trial Research Group. Subgroup at high risk for surgical mortality in patients receiving lung volume reduction surgery. N Engl J Med. 2001;345:1075-1083.
  2. National Emphysema Treatment Trial Research Group. A randomized trial comparing lung volume reduction surgery with medical therapy for severe emphysema. N Engl J Med. 2003;348:2059-2073.