|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Departments of Medicine (Ms. Appleton, and Drs. Adams and Ruffin) and Surgery (Mr. Peacock), University of Adelaide, The Queen Elizabeth Hospital Campus, Woodville, SA, Australia; private practice (Dr. Porter), Adelaide, SA, Australia.
Correspondence to: Robert Adams, MBBS, MD, Department of Medicine, The Queen Elizabeth Hospital, 28 Woodville Rd, Woodville, SA, Australia 5011; e-mail: robert.adams{at}nwahs.sa.gov.au
| Abstract |
|---|
|
|
|---|
Design: Retrospective observational study.
Setting: Academic medical center
Methods: Telephone and postal surveys were used to obtain patient dyspnea scores and HRQL scores. Hospital databases and state registries were searched to determine patient survival and pulmonary function.
Results: Of 54 patients undergoing LVRS, 29 patients (18 men and 11 women) were available for follow-up, which ranged from 36 to 66 months (mean ± SE, 51 ± 1.5 months). There was significant sustained improvement in modified Medical Research Council scores compared to pre-LVRS: 2.19 ± 0.19 vs 2.88 ± 0.14 (p = 0.0000). Eleven of 22 patients demonstrated an increase in all three Mahler baseline dyspnea index grades of at least one level. Baseline body mass index (BMI) and post-LVRS length of stay (LOS) were significantly associated with survival: survivor vs deceased baseline BMI, 24.2 ± 0.6 vs 21.4 ± 0.5 (p = 0.002), and post-LVRS LOS, 15.4 ± 1.7 days vs 28.7 ± 5.3 days (p = 0.015). Compared to pre-LVRS, 20 patients with mean follow-up time of 45 months demonstrated significant sustained improvements in FEV1 percentage of predicted (31.4 ± 2.1% vs 39.8 ± 3.5%, p = 0.038), total lung capacity percentage of predicted (136 ± 4% vs 122 ± 3%, p = 0.0004), and residual volume percentage of predicted (237 ± 14% vs 172 ± 11%, p = 0.0001). Patient HRQL measured using the Dartmouth Primary Care Co-operative Quality of Life questionnaire was more favorable than that reported in aged-care settings. Caregiver burden scale scores indicate caring for a recipient of LVRS carries similar burden to that for caring for individuals with other chronic illnesses.
Conclusions: In this population, a majority of the LVRS patients survived for ≥ 3 years. Among survivors, dyspnea and lung function benefits were seen. Baseline BMI and postoperative LOS were significantly associated with survival.
Key Words: emphysema lung diseases, obstructive lung volume reduction surgery
| Introduction |
|---|
|
|
|---|
Few studies of LVRS have considered HRQL,6 7 8 particularly disease-specific quality of life.9 10 These end points are important because care in late-stage COPD can be considered as essentially palliative.
We have previously reported11 the short-term effectiveness of LVRS in a cohort of 54 patients, with a median follow-up of 788 days, and demonstrated significant improvements in lung function and 6-min walk distance, and an appreciable mortality rate of 14.8% at 90 days after surgery. These results were consistent with previous reports.12 13 14
The aim of this study was to determine longer-term mortality data after LVRS, and to examine what factors are associated with longer survival. We also sought to determine what qualitative benefits are sustained after a follow-up period in the range of 3 to 5.5 years, in terms of perceived dyspnea and HRQL of patients and their caregivers.
| Materials and Methods |
|---|
|
|
|---|
|
Study Design
This study was conducted using a telephone survey. Survival status was ascertained by searching the institutional patient record database and the state death registry. Letters explaining the purpose of the study were mailed to all subjects, inviting their participation. All patients were telephoned over a 3-week period to obtain informed consent, and dyspnea was graded at that time. HRQL questionnaires were mailed out for self-administration by patients and their caregivers, where present.
Dyspnea
Dyspnea was graded using two scales. Firstly, the modified Medical Research Council (MRC) dyspnea scale was used. This is scored on a 5-point scale from 0 to 4, where grade 0 describes the subject as "not troubled by shortness of breath except with strenuous exercise," and grade 4 describes the subject as "too breathless to leave the house or breathless when dressing or undressing."15
The second scale used was the Mahler baseline dyspnea index (BDI). The Mahler BDI is a multidimensional grading of dyspnea, using the categories of "functional impairment," "magnitude of task," and "magnitude of effort" on a scale of 0 to 4, where 4 represents best and 0 represents worst.16
The BDI was used to determine current dyspnea (36 to 65 months after LVRS) in preference to the transition dyspnea index because there was a substantial risk of subject recall bias in evaluating change compared to shortly after the procedure. All assessments of patient dyspnea were conducted by the same person to eliminate interobserver error.
Post-LVRS HRQL
HRQL was assessed using the Dartmouth Primary Care Co-operative Quality of Life questionnaire (COOP),17
modified for use in COPD, as previously reported.18
The modifications were made so as to avoid floor effects and permit potential deteriorations in condition to be detected. Two changes were made to the original questionnaire: firstly, a supplementary item specifically related to dyspnea was added (When do you become short of breath? Extreme effort [walking up a steep hill/jog at slow pace], moderate effort [walking up stairs/easy digging in garden], little effort [walking on a level surface at regular pace], very little effort [walk at slow pace/washing dishes], and at rest [sitting or lying down]). Secondly, in order to more appropriately relate to the level of activity of patients with severe COPD, the item regarding physical condition (What is the most strenuous level of activity you can do for 2 min?) was altered by deleting the "heavy" category (jog at fast pace) and adding an "extremely light" category (eating, getting out of bed). The remaining eight items refer to the emotional and overall condition and how these impact on the ability to perform daily and social activities, and the perception of pain and change in condition. The COOP questionnaire features a simple graphical layout, has documented reliability and validity in a range of settings including respiratory disease,19
20
and was selected in view of the level of morbidity of the subjects to be enrolled. High COOP scores indicate lower perceptions of HRQL.
Caregiver Burden
To determine the caregiver burden associated with caring for a patient who has undergone LVRS, the caregivers completed the Caregiver Burden Scale, which has proven validity and reliability in chronic illness.21
Pulmonary Function
Lung function data were obtained on survivors from The Queen Elizabeth Hospitals Pulmonary Function Laboratory database. Patient data were only included in the analysis if the follow-up time was > 2 years. The data of four subjects were excluded because the follow-up on lung function data was available for < 2 years.
Statistical Analysis
Data are provided as mean ± SE. Differences between pre-LVRS and follow-up respiratory parameters and outcomes were analyzed using paired and unpaired t tests. Preoperative and postoperative dichotomous variables and outcomes that were used as predictors of mortality were analyzed for statistical significance using the Pearson
2 test. All p values were considered significant if < 0.05. Kaplan-Meier survival analysis was performed to determine survival probability, and Cox proportional hazards models were used to determine preoperative and postoperative factors influencing long-term survival. Institutional ethics committee approval was obtained for this study.
| Results |
|---|
|
|
|---|
|
Twenty-nine of 30 surviving patients who underwent LVRS were available for follow-up with the survey. Median duration of survivor follow-up was 49 months (range, 36 to 66 months).
Table 3 shows the mean modified MRC scale and Mahler BDI scores for all patients prior to LVRS and for the survivors at baseline and follow-up. There was a statistically significant reduction in dyspnea for survivors at follow-up compared to baseline (p ≤ 0.01). Among survivors, 58% (15 of 26 patients) improved at least one level on the modified MRC scale, and 50% (11 of 22 patients) improved by more than one level in all three scales of the Mahler BDI, compared to their baseline values. When dyspnea benefits were considered as a proportion of all subjects undergoing LVRS and not just the survivor fraction, 27.8% and 20.4% of all subjects experienced ongoing reductions in dyspnea according to the MRC scale and Mahler BDI scores, respectively.
|
|
Table 5
shows results for patient HRQL as measured by the COOP charts, and for caregiver burden, as measured using the Caregiver Burden Scale. Patient COOP scores and Caregiver Burden Scale scores were not significantly correlated with each other. No significant relationship was demonstrated between the level of patient support and postsurgical mortality (Pearson
2 = 1.39, p = 0.71).
|
|
|
| Discussion |
|---|
|
|
|---|
Given the limitations inherent in a retrospective analysis of a case series and absence of adequate presurgical HRQL measures for follow-up analysis, we have demonstrated reductions in dyspnea comparable to previous reports. Gelb et al22 reported a reduction in dyspnea grade ≥ 1 at 3-year, 4-year, and 5-year follow-up in 46%, 27%, and 15% of subjects, respectively. Approximately 30% of our patients who had dyspnea relief did not have an increase of > 200 mL of FEV1 after surgery. Post-LVRS improvements in dyspnea have been related to increased lung elastic recoil,23 24 reduced hyperinflation,24 improved respiratory muscle function,24 and reduced mechanical restraints on tidal volume,25 all of which may be independent of FEV1.26
Our results show that a lower BMI has an adverse impact on longer-term survival after LVRS. This was not due to any effect of patients with very poor nutritional status, as patients with a BMI < 18 were not eligible for surgery. In particular, a post hoc analysis indicated that a BMI of between 25 and 30 was significantly associated with improved survival. We also observed a trend for a low BMI (18 to <20) to be associated with a prolonged after surgery LOS. Nutritional status, specifically depletion, has been linked to morbidity27 and mortality28 in patients with COPD, and with adverse LVRS outcomes in the early postsurgical period.29 30 A study30 with 6-month follow-up of 23 patients has demonstrated significantly higher mean BMI and fat-free mass index in patients without major complications. The significantly increased mortality among those with both low and normal BMI, compared with those with BMI ≥ 25 to 30, may be related to masking of fat-free muscle depletion in the apparently normal weight patients31 32 as a contributing factor. This may need to be considered when planning and selecting candidates for LVRS.
We were unable to confirm the high-risk baseline factors for mortality of FEV1 ≤ 20% predicted and DLCO ≤ 20% predicted or a homogeneous pattern of emphysema recently reported on preliminary data by the National Emphysema Treatment Trial (NETT) Research Group.33 We did detect a trend toward a higher DLCO predicting survival. However, only 5 of 54 patients in our uncontrolled case series had an FEV1 ≤ 20% at baseline, and only one subject had a DLCO at baseline of ≤ 20%. The NETT investigators also reported a much higher 30-day mortality rate than our study (16% vs 5.5%). It is of note that with less severely affected patients than in the NETT, we were able to demonstrate substantial ongoing functional benefits from LVRS.
There is clear variation in function and quality of life in our patient group. Mean COOP scores 3 to 5.5 years after LVRS were better than those reported in a trial of hospital at home care vs standard hospital care for patients with COPD34 and in a group of elderly patients in residential care,35 even though an extra item pertaining to breathing was included in our "modified" COOP questionnaire. A significant limitation of this study is that we have inadequate pre-LVRS COOP data and follow-up lung function data to make statistically valid pre-LVRS and post-LVRS comparisons for these variables. Few studies have reported HRQL outcomes after LVRS. A study by Leyenson et al36 showed short-term improvements in HRQL were not correlated with routine measures of pulmonary function, gas exchange, and exercise tolerance. Another study37 showed that improvements in Medical Outcomes Study Short Form-36 scores at 6 months were significantly associated with improvements in FEV1. An increased level of performance of daily activities may maintain exercise tolerance and prevent physical deconditioning in the presence of gradually declining objective lung function benefits.
There was no significant correlation between overall patient HRQL and caregiver burden, nor was there a significant independent relationship between care burden scores and the COOP social support item. Caregiver burden was lower for our patients caregivers than that reported by those caring for people with rheumatoid arthritis38 and for people 3 years after stroke.21 Although interpretation of this data is limited by the lack of preoperative comparison data, it suggests that LVRS is unlikely to impose a significant extra burden on most caregivers.
The number of subjects with sustained increases in lung function at follow-up compared to pre-LVRS is consistent with that previously reported. Gelb et al1 39 demonstrated that increases in FEV1 > 200 mL and/or FVC > 400 mL occurred in 46% of patients at 2 years and in 35% of subjects at 3 years, and Flaherty et al2 reported FEV1 improvements > 200 mL in 29% of patients at 3 years.
Post-LVRS survival rates vary across studies. Our follow-up has shown a survival rate of 67% at 3 years after LVRS, 63% at 4 years, and 56% at 5 years. This compares favorably with reported 3-year, 4-year, and 5-year survival rates of 69%, 54%, and 42%, respectively, by Gelb et al.26 In contrast, our survival rates compare unfavorably with other reports of 5-year survival of 74%3 and the 3-year survival of 82%.2 These differences may be explained by the younger age,2 3 less dyspnea,2 and better exercise tolerance3 at baseline in these study subjects, compared to our study subjects.
It is arguable, however, that analysis of outcomes in the survivor fraction exaggerates the true benefits of LVRS. When lung function and dyspnea benefits were considered as a proportion of all of those subjects who underwent the procedure, the benefits were reduced and, at best, would seem marginal. Significant increases in FEV1 and FVC after 3 years were demonstrable in 11% and 13% of all subjects, respectively. Similarly, 28% and 20% of all subjects experienced ongoing dyspnea relief as measured with the MRC scale and Mahler BDI. However, in a discussion of survival benefits associated with LVRS, it is necessary to consider the prognosis for people with COPD who receive optimal medical management only. Varying survival rates are reported in the literature. In the study by Almagro et al40 in 135 elderly patients (mean age, 72.2 ± 9.3 years) who were hospitalized for acute exacerbations of COPD, the survival rate was 64% at 2 years. Given that people undergoing LVRS are a highly selected group of patients who usually undergo presurgical and postsurgical respiratory rehabilitation programs that in themselves are known to be beneficial,41 a direct comparison of our 3-year to 5-year survival rates and associated benefits in outcome measures with those reported in the literature for medical management alone possibly is not valid. Meyers et al4 compared 3-year survival rates for patients who were eligible for but were denied LVRS and proceeded to receive medical care, and those that underwent LVRS. They reported 3-year survival rates of 64% in the medical group, and 82% for the LVRS group, suggesting that the survival benefit conferred by LVRS is indeed meaningful. Indirect measures, such as rates of hospital admissions for exacerbations of COPD, in addition to the direct benefits of LVRS should be considered in future studies because these also have significant impacts on patients and health-care systems.
In summary, we have demonstrated sustained improvements in perceived dyspnea and lung function from LVRS that are preserved beyond 3 years. These improvements appear to be ongoing among those with up to 5.5-year survival. The HRQL of survivors is comparable to that experienced by the "healthy" elderly in the community. LVRS does not appear to impose an onerous burden on patient caregivers. Importantly, we have identified BMI as a predictor of longer postsurgical survival. This may have implications for selection of LVRS candidates and for preoperative management.
| Acknowledgements |
|---|
| Footnotes |
|---|
This study was conducted entirely at The Queen Elizabeth Hospital.
Received for publication March 19, 2002. Accepted for publication September 27, 2002.
| References |
|---|
|
|
|---|
1-antitrypsin deficiency versus smokers emphysema. Eur Respir J 1998;12,1028-1032[Abstract]
This article has been cited by other articles:
![]() |
Q.-Y. Tan, R.-W. Wang, Y.-G. Jiang, S.-Z. Fan, M. K.Y. Hsin, T.-Q. Gong, J.-H. Zhou, and Y.-P. Zhao Lung Volume Reduction Surgery Allows Esophageal Tumor Resection in Selected Esophageal Carcinoma With Severe Emphysema Ann. Thorac. Surg., November 1, 2006; 82(5): 1849 - 1856. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. B. Gorman, D. K. McKenzie, J. E. Butler, J. F. Tolman, and S. C. Gandevia Diaphragm Length and Neural Drive after Lung Volume Reduction Surgery Am. J. Respir. Crit. Care Med., November 15, 2005; 172(10): 1259 - 1266. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. D. Yusen and B. Littenberg Integrating Survival and Quality of Life Data in Clinical Trials of Lung Disease: The Case of Lung Volume Reduction Surgery Chest, April 1, 2005; 127(4): 1094 - 1096. [Full Text] [PDF] |
||||
![]() |
F. Laghi, A. Jubran, A. Topeli, P. J. Fahey, E. R. Garrity Jr, D. J. de Pinto, and M. J. Tobin Effect of Lung Volume Reduction Surgery on Diaphragmatic Neuromechanical Coupling At 2 Years Chest, June 1, 2004; 125(6): 2188 - 2195. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |