| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 2008;136:597-604
© 2008 The American Association for Thoracic Surgery
General Thoracic Surgery |
a Thoracic Surgery, St Luke's Health Network, Bethlehem, Pennsylvania
b Division of Thoracic Surgery, Duke University Medical Center, Durham, North Carolina
c Cancer Statistical Center, Duke University, Durham, North Carolina
d GlaxoSmithKline, Research Triangle Park, North Carolina
Received for publication May 15, 2007; revisions received October 17, 2007; accepted for publication February 19, 2008. * Address for reprints: William R. Burfeind Jr, MD, 701 Ostrum Street, Suite 201, Bethlehem, PA 18015. (Email: burfeiw{at}slhn.org).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
Methods: By using a validated quality of life instrument, 422 patients were prospectively assessed preoperatively and 3, 6, and 12 months after lobectomy. Outcomes were analyzed with respect to age (group 1: <70 years and group 2: 
70 years). The outcome domains of physical functioning, role functioning, emotional functioning, cognitive functioning, social functioning, global health, and pain in the chest were analyzed using a mixed model. The trend in quality of life was determined according to age. The Kaplan–Meier method was used for analysis of overall survival.
Results: The mean age was 60.1 years in group 1 (N = 256) and 74.7 years in group 2 (N = 166). Baseline demographics and quality of life were similar except that group 2 had better emotional functioning scores and worse pain in the chest scores. Postoperatively, both groups demonstrated significant decreases in quality of life at 3 months. However, at 6 and 12 months, all domains had returned to baseline except physical functioning, which remained below baseline in group 2. Emotional functioning improved postoperatively for both groups. Overall survival at 5 years was not different between groups.
Conclusion: By using a validated quality of life assessment tool with measurements at baseline and serially after resection in a large patient population, this analysis quantifies the degree of impairment of quality of life after lobectomy and documents time to full recovery for both age groups.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
Yet, successful operations for stages I and II lung cancer are associated with a 5-year survival of only 40% to 70%.2 Thus, treatment is often palliative rather than curative. Most studies published to date are concerned primarily with traditional outcomes, such as operative morbidity and mortality or long-term survival. The complex interplay among the treatment's benefits, risks, and the expectations held by patients with lung cancer serve to focus attention on the importance of health-related quality of life (QOL).
Information on QOL after operations for lung cancer surgery is sparse. Between 1994 and 2004, only 9 published articles were found that described QOL after lung cancer operations.3-11 Of the 9 studies, only 3 included preoperative assessments, and although most used a validated QOL instrument, only 2 used a cancer-specific instrument.8,10 Clearly, a prospective, longitudinal study that follows a large patient population and that uses a cancer-specific instrument is needed to more accurately portray QOL after lung cancer operations.
Lung cancer disproportionately affects the elderly, with a median age at diagnosis of 70 years.12 Prospective analyses of QOL in elderly patients after lobectomy are absent, yet surgeons may recommend operations with less proven oncologic efficacy to these patients on the basis of the belief that lobectomy is poorly tolerated. Surgical decision making in elderly patients with lung cancer is better informed when the benefits of lobectomy to survival and QOL are understood.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
Patients and Quality of Life Assessments
Patients with suspected lung cancer, who were seen in the Duke Thoracic Oncology Program, were asked to complete a QOL questionnaire preoperatively and then postoperatively at 3, 6, and 12 months. A trained clinical research assistant administered the questionnaire on a touch-screen computer, and attempts were made to administer the questionnaire before meeting with the clinician. Only patients who completed the baseline QOL assessment and at least 1 postoperative QOL assessment, and had undergone an anatomic lobectomy were included in the study. For purposes of analysis, patients in group 1 were defined as less than 70 years of age and patients in group 2 were defined as 70 years or older.
Quality of Life Measures
QOL was assessed using the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC-QLQ-C30). The QLQ-C30 is a validated instrument for measuring QOL in patients with cancer and consists of 5 functional scales, 3 symptom scales, and a global health status/QOL scale (sample questions can be seen at http://groups.eortc.be/qol/downloads/modules/specimen_20qlq_c30.pdf).13 In addition, 2 symptom scales for pain in the chest and pain in the arm/shoulder, as well as a dyspnea symptom scale, from the validated lung cancer module (QLQ-LC13), were used. Both questionnaires are designed for self-completion by the patient. For the functional scales (ie, those exploring physical, role, emotional, cognitive, and social functioning) and global health status, a higher value corresponds to a higher level of function or high QOL. Conversely, a higher score in a symptom scale corresponds to a higher level of symptoms or problems.
Statistical Considerations
The baseline characteristics of group 1 and group 2 were compared using the Fisher's exact test for dichotomous outcomes, and a 2-tailed t test was used for continuous variables. QOL scores collected from the EORTC QLC-C30 were linearly transformed to a scale of 0 to 100 in accordance with the EORTC guidelines.14 The major objective of this study was to determine how QOL changed longitudinally and to compare differences between group 1 and group 2. This comparison was made by fitting a linear mixed model for repeated measures with restricted maximum likelihood estimation method for each EORTC functional scale, the global QOL scale, as well as 2 pain scales and the dyspnea scale (SAS PROC MIXED, SAS Institute Inc, Cary, NC). Effects modeled included the patient's age group, survey time (preoperatively, 3, 6, or 12 months), and interaction between age group and survey time. Of primary interest in these analyses was the interaction, a measure of whether the effect of time on outcomes was the same between the 2 groups.
Because 20% to 38% of data were missing, sensitivity analyses were conducted to examine the effect of missing data on inferences. The model was reanalyzed with data imputed to replace the missing data. A Monte Carlo Markov Chain approach to multiple imputations was used with SAS PROC MI. The correlation of the 4 assessments of each outcome, marital status, living situation, sex, stage, education, and video-assisted thoracic surgery use, were used in estimating 5 complete datasets or imputations. A mixed linear model (SAS PROC MIXED) was fit to each dataset with factors describing the effect of time, age group, and their interaction. The results for the interaction were combined across models using SAS PROC MIANALYZE.
The survival function was estimated using the Kaplan–Meier method. Overall survival was calculated from the date of surgery to the date of death from any cause or censored at June 1, 2006. Survival was expressed as a proportion, and survival curves were compared using the log-rank statistic. SAS version 9.1 (SAS Institute Inc) was used for all analyses.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
A total of 422 patients completed at least 1 postoperative survey and were included in this study. Demographics from the eligible patients are listed in Table 1 . As indicated, the majority of patients were white and married, and had stage 1 or 2 NSCLC. The 2 groups were well balanced with respect to patient characteristics.
|
|
|
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
The multidimensional nature of the QOL assessment raises some important issues of interpretation. Multiplicity of comparisons and interdependency of responses at successive times make it necessary to focus on the whole pattern of results. The exclusive use of patients with complete data provides unbiased, albeit inefficient, results, if the data that are missing occurs completely at random. Alternatively, it provides biased results, if data that are missing occurs not completely at random (the more common form of missingness). In this study we used a linear mixed-effect model to analyze all available data and to reconcile the fact that intrasubject QOL measurements are not independent (repeated measures).15 On the basis of this analysis, both groups demonstrated significant reductions in physical, role, social, and global QOL domains, as well as increased pain involving the chest and arm/shoulder, and increased dyspnea at 3 months postoperatively. All parameters, except dyspnea, had recovered to baseline values by 6 months. Both groups demonstrated a similar decrement in QOL with a parallel return to baseline. The one notable exception was in the domain of physical functioning, which had returned to baseline by 6 months and decreased below its preoperative value in patients 70 years or older by 12 months. It is unclear whether physical functioning remained reset at this lower level or continued to deteriorate over time in these patients.
Patients aged 70 years or more had less impairment in emotional functioning preoperatively. This domain remained at its baseline value in both groups through the first 3 months after operation, but then demonstrated a significant increase by 6 months. This trend persisted through the end of the study. This reduction in emotional functioning preoperatively may be explained in part by the psychologic impact related to the recent cancer diagnosis. Not surprisingly, worsened emotional functioning seems to affect younger patients more than the older patients. In a similar pattern, older patients seem to perceive pain as being less severe both at baseline and postoperatively when compared with younger patients.
The observation that QOL decreases over the first 3 to 6 months postoperatively and then returns to its preoperative level has been reported by others. Dales and colleagues9 reported on 117 patients who underwent thoracotomy for presumed lung cancer (NSCLC in 91, other pathology in 26). QOL was measured with the Sickness Impact Profile (SIP) and Spitzer QL-Index preoperatively and at 1, 3, 6, and 9 months postoperatively. Dales and colleagues found that QOL and dyspnea significantly worsened at 1 and 3 months postoperatively but returned to baseline at 6 and 9 months. Zieren and colleagues10 reported similar findings using the Spitzer QL-Index and an earlier version of the EORTC QLQ (QLQ-CLC36). They examined 2 cohorts of patients who had undergone curative operations for lung cancer; 1 cohort completed a single QOL survey 12 months postoperatively (N = 52) and another was surveyed preoperatively and then again at discharge (median 11 days) and 3, 6, 9, and 12 months postoperatively (N = 20). In the latter cohort, patients demonstrated significant decreases in physical, role, and global domains of QOL, and increases in disease symptoms, such as dyspnea, at discharge. These differences, however, had disappeared at the 3-month survey.
Handy and colleagues6 reported on 139 patients undergoing lung resections who were evaluated with the short-form health survey [36 items] and the Powers' QOL index before operation and then again at 6 months after operation. Preoperative QOL in patients undergoing operation for lung cancer was significantly worse than in healthy controls. Handy and colleagues found that at 6 months, subscales for physical functioning, role-physical, bodily pain, and mental health on the short-form health survey [36 items] were significantly worse than baseline. Because the data collected were truncated at 6 months, however, it is uncertain whether these findings persisted at 12 months. In addition, Handy and colleagues found that emotional functioning, although not significantly worsened, failed to improve with time. An unanswered question is whether the difference in their findings compared with those of the present study can be attributed to a different patient population or differences in QOL instruments.
Compliance with the QOL assessment, defined as the number of assessments actually completed as a proportion of those expected, ranged from 62% to 80%.16 The most commonly missed survey time point was the 3-month survey with 28% of group 1 and 38% of group 2 missing this time point. Missing data were rarely the result of patient refusal or neglect, because a trained clinical research nurse guided each patient through their surveys. A more common reason was a failure of the patient to return for clinic follow-up within the prescribed time frame. Compliance with the QOL assessment is important because the failure to return for clinic follow-up may be influenced by the patient's QOL at that particular time point.17 Because of the retrospective nature of this study, it is not possible to determine whether a patient missed the assessment because of postoperative disability, with associated poor QOL, or because he or she was recovering exceptionally well and chose to skip a clinic visit. Moreover, because of our large referral area, some patients may have chosen to receive part of their cancer care with their local area physician. Certainly, the larger percentage of group 2 patients who missed their 3-month survey could introduce significant bias if these patients did so because of increased postoperative disability.
Missing data are common in QOL studies, and they are usually not missing completely at random.18 Thus, to account for missing data, appropriate statistical modeling (linear mixed model) was used. Sensitivity analysis was performed by replacing the missing data with multiply imputed values that were generated from individual baseline characteristics and previous QOL values.19 Furthermore, the same linear mixed model was used to analyze each of the 5 imputed datasets, and SAS PROC MIANALYZE was applied to combine the results across models. Although it is impossible for any model to completely remove all possible bias, the findings from this analysis confirmed the original results; no difference in QOL was observed between groups.
Overall survival for the 2 groups was not different, whereas median survival time for the 2 groups appeared different because of a separation in survival curves at approximately 60 months. The median survival was 73 months for group 1 and 56.7 months for group 2.
The most important limitation to the generalizability of this study's findings is the selection bias inherent in its retrospective design. The study cohort represented patients who were motivated and willing to complete QOL surveys, preoperatively and postoperatively. This study delineated the time course of change in QOL after lobectomy, but its conclusions obviously do not help a surgeon choose which particular elderly patient will survive a major lung cancer operation. Indeed, although not statistically significant, more patients in group 2 died within 90 days of completing the preoperative survey than in group 1 (12 patients in group 2 vs 8 patients in group 1). Another limitation to this study's generalizability is the relatively high percentage of women who were enrolled (55.9% in group 1 and 47.6% in group 2). This proportion is slightly higher than the general pool of patients with lung cancer and may indicate that women generally are more willing to complete QOL surveys.
To summarize the main findings, this study delineates the magnitude of the change in QOL experienced by younger and older patients undergoing anatomic lobectomy and shows that they have similar decrements to 3 months. This study also documents that the time course to recovery of QOL in both groups is generally less than 6 months.
|
Conclusions |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Moller and U. Sartipy Long-term health-related quality of life following surgery for lung cancer Eur J Cardiothorac Surg, February 1, 2012; 41(2): 362 - 367. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Amer, A.-Z. Khan, H. Vohra, and R. Saad Is it safe to include octogenarians at the start of a video-assisted thoracic surgery lobectomy programme? Eur J Cardiothorac Surg, February 1, 2012; 41(2): 346 - 352. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Chambers, T. Routledge, J. Pilling, and M. Scarci In elderly patients with lung cancer is resection justified in terms of morbidity, mortality and residual quality of life? Interact CardioVasc Thorac Surg, June 1, 2010; 10(6): 1015 - 1021. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Endoh, S. Tanaka, T. Yajima, T. Ito, K. Tajima, A. Mogi, Y. Shitara, and H. Kuwano Pulmonary function after pulmonary resection by posterior thoracotomy, anterior thoracotomy or video-assisted surgery Eur J Cardiothorac Surg, May 1, 2010; 37(5): 1209 - 1214. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Pompili, A. Brunelli, M. Refai, F. Xiume, and A. Sabbatini Does chronic obstructive pulmonary disease affect postoperative quality of life in patients undergoing lobectomy for lung cancer? A case-matched study Eur J Cardiothorac Surg, March 1, 2010; 37(3): 525 - 530. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Sartipy Prospective population-based study comparing quality of life after pneumonectomy and lobectomy Eur J Cardiothorac Surg, December 1, 2009; 36(6): 1069 - 1074. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. F. Berry, J. Hanna, B. C. Tong, W. R. Burfeind Jr, D. H. Harpole, T. A. D'Amico, and M. W. Onaitis Risk Factors for Morbidity After Lobectomy for Lung Cancer in Elderly Patients Ann. Thorac. Surg., October 1, 2009; 88(4): 1093 - 1099. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. R. Villamizar, M. D. Darrabie, W. R. Burfeind, R. P. Petersen, M. W. Onaitis, E. Toloza, D. H. Harpole, and T. A. D'Amico Thoracoscopic lobectomy is associated with lower morbidity compared with thoracotomy. J. Thorac. Cardiovasc. Surg., August 1, 2009; 138(2): 419 - 425. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Brunelli, A. Charloux, C. T. Bolliger, G. Rocco, J-P. Sculier, G. Varela, M. Licker, M. K. Ferguson, C. Faivre-Finn, R. M. Huber, et al. ERS/ESTS clinical guidelines on fitness for radical therapy in lung cancer patients (surgery and chemo-radiotherapy) Eur. Respir. J., July 1, 2009; 34(1): 17 - 41. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. K. Ferguson, C. M. Parma, A. D. Celauro, and W. T. Vigneswaran Quality of Life and Mood in Older Patients After Major Lung Resection Ann. Thorac. Surg., April 1, 2009; 87(4): 1007 - 1013. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
