|
Back to 85th Annual Meeting
Back to Program Outline
34. Predictors of Operative Morbidity and Mortality in the National Emphysema Treatment Trial
Keith S Naunheim, Douglas E Wood, Mark J Krasna, Mark E Ginsburg, Malcolm M Decamp, Robert J Mckenna, Gerard J Criner, Eric A Hoffman, Alice Sternberg, Claude Deschamps; St. Louis, MO; Seattle, WA; Baltimore, MD; New York, NY; Boston, MA; Los Angeles, CA; Philadelphia, PA; Iowa City, IA; , MD; Rochester, MN
Objective:To identify predictors of operative mortality (OM), pulmonary morbidity (PM) and cardiovascular morbidity (CM) in patients undergoing lung volume reduction (LVR)
Methods:Univariate and stepwise regression analysis were performed on 23 variables describing demographics (age, gender, race) physical condition (BMI, O2 usage, steroid usage, creatinine), pulmonary function measures (FEV1, RV, RV/TLC, DLCO, PaCO2, Ve/VCO2), radiologic measures (heterogeneity, upper lobe distribution, hyperinflation, perfusion ratio) as assessed by radiologists (RAD) and by computerized analysis (CA) of CT scans, exercise capacity (6 minute walk, maximum exercise capacity), dyspnea (University of California San Diego respiratory scores) and quality of life. Three endpoints were analyzed: 1) operative mortality defined as death at 90 days 2) major pulmonary morbidity defined as tracheostomy, failure to wean, reintubation, pneumonia, ventilator > days 3) major cardiovascular morbidity defined as myocardial infarction, pulmonary embolus, or arrhythmia requiring treatment.
Results:There were 511 patients (pts)in the non high-risk group of the National Emphysema Treatment Trial (NETT) who underwent LVR. Operative mortality (OM) was 5.5%(28 pts) while major pulmonary (PM) and cardiovascular morbidity (CM) were 30.1%(154 pts) and 19.4%(99 pts) respectively. Independent predictors for these endpoints are depicted in the table.
Conclusions:Although LVR can be performed in selected patients with acceptable mortality, the incidence of major cardiopulmonary morbidity remains high. The lone predictor for operative mortality of LVR in non high-risk NETT patients was the absence of upper lobe predominant disease as assessed by the radiologist. Elderly patients with more severe airway obstruction and impaired gas exchange can be expected to sustain major pulmonary complications. Advanced patient age was also predictive of cardiovacular complications. When assessing morbidity, the computer assisted chest CT analysis proved useful only in predicting cardiovascular complications.
Independent Predictores of Morbidity
|
|
RO* |
P |
| OM |
Non upper lobe predominance-RAD (yes/no) |
3.71 |
0.004 |
|
|
|
|
| PM |
Age in years (per yr of age) |
1.05 |
0.01 |
|
O2 use - both rest and exercise (yes/no) |
1.85 |
0.009 |
|
FEV1% predicted (per unit change) |
0.95 |
0.01 |
|
|
|
|
| CM |
Age in years (per yr of age) |
1.06 |
0.01 |
|
Non upper lobe predominance-CA (yes/no) |
2.33 |
0.001 |
|
*RO = relative odds for those with vs without dichotomous parameters OR incremental odds for each unit change in continuous parameters |
|
|
Back to 85th Annual Meeting
Back to Program Outline
|
