Lung resection still achieves the best long-term results

Quality of Life Before and After Major Lung Resection for Lung Cancer: A Prospective Follow-Up Analysis Alessandro Brunelli, MD, Laura Socci, MD, Majed Refai, MD, Michele Salati, MD, Francesco Xiumé, MD, and Armando Sabbatini, MD Division of Thoracic Surgery, Umberto I Regional Hospital, Ancona, Italy Background. The objective of this study was to assess the preoperative and postoperative quality of life of candidates for major lung resection with lung cancer. Methods. In all, 156 consecutive patients (144 lobectomies, 12 pneumonectomies) were prospectively assessed by means of the Short Form-36 Health Survey, version 2, preoperatively and 1 month and 3 months after operation. Serial quality of life scales were compared by repeated measures analysis of variance. Results. In our series, most quality of life values were reduced compared with the general population. Compared with preoperative values, the physical composite scale was significantly reduced at 1 month (51 versus 45.1, p < 0.0001), and completely recovered at 3 months (51 versus 52.4, p 0.2), whereas the mental composite scale remained unchanged. All correlation coefficients between these values and forced expiratory volume in 1 second, carbon monoxide lung diffusion capacity, and height reached at stair-climbing test at each evaluation period were below 0.2. With the exception of pneumonectomy patients (who had a significantly lower physical composite scale [p 0.04]), no significant differences in both physical and mental values were noted in other high-risk subgroups of patients (elderly, coronary artery disease, poor pulmonary function) compared with lowerrisk counterparts. Conclusions. Candidates for lung resection with lung cancer had a worse preoperative quality of life compared with the general population. Quality-of-life measures had poor correlation with forced expiratory volume in 1 second, carbon monoxide lung diffusion capacity, and exercise test performance. Therefore, these functional variables cannot substitute for specific evaluation instruments. Finally, patients traditionally considered at higher risk for lung resection had postoperative physical and emotional quality of life scores similar to those observed in younger and fitter patients. (Ann Thorac Surg 2007;84:410 6) 2007 by The Society of Thoracic Surgeons Lung resection still achieves the best long-term results in patients with nonsmall-cell lung cancer. However, what price is the patient ready to pay for radical resection? For many patients, the risk of an impaired quality of life (QOL) after surgery is an important consideration when deciding whether to proceed with surgery. Some of them may be ready to accept the risk of immediate cardiopulmonary complications but are not prepared to accept significant postoperative functional debility [1]. Furthermore, whereas the surgeons main objectives are radical clearance of the tumor and longterm survival of the patient with low postoperative morbidity and mortality rates, the referring physician may perceive that the patient will be left physically and emotionally handicapped by surgery and may refer the patient to alternative, less efficacious treatments [2]. The potential benefit (survival) of an operation must be always weighed not only against perioperative morbidity and mortality, but also against residual health-related QOL measures. In fact, more comprehensive and honest Accepted for publication April 2, 2007. Address correspondence to Dr Brunelli, Via Santa Margherita 23, Ancona, 60124, Italy; e-mail: alexit_2000@yahoo.com. patient counseling should include reliable information not only about early outcomes (morbidity and mortality) and long-term survival, but also about the anticipated residual QOL. The importance of QOL measures is such that future analyses on quality of care evaluation will necessarily take into consideration these endpoints in addition to more traditional measures of performance (morbidity and mortality) [3]. Although the need exists for a more comprehensive understanding of the effects of thoracic surgery on patients functional and QOL outcomes, few studies have recently addressed this subject [1, 4 10]. Only three of them included a preoperative QOL assessment that allowed a comparison with postoperative measures [6, 8, 10], and only two reported norm-based data that may help to clarify what is the physical and emotional condition of patients submitted to lung resection in comparison with normal healthy subjects [6, 9]. Furthermore, most of the studies assessed postoperative QOL 6 months to 5 years after operation [4 6,9], and this may have excluded the patients not cured by the operation, for whom the knowledge of the residual QOL 2007 by The Society of Thoracic Surgeons 0003-4975/07/$32.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2007.04.019

Ann Thorac Surg BRUNELLI ET AL 2007;84:410 6 LUNG RESECTION AND QUALITY OF LIFE Abbreviations and Acronyms QOL quality of life SF36v2 Short Form 36-Item Health Survey version 2 FEV1 forced expiratory volume in 1 second Dlco carbon monoxide lung diffusion capacity ppofev1 predicted postoperative FEV1 ppodlco predicted postoperative Dlco PCS physical composite scale MCS mental composite scale in the months after surgery may be equally important. Only two studies measured QOL within 3 months after operation [8, 10], presumably reporting a more realistic picture of the physical and emotional status of the entire cohort of patients undergoing lung resection. However, none of these studies used the Short Form 36-Item Health Survey, version 2 (SF36v2). Thus, the objectives of this prospective study were the following: (1) to assess the postoperative QOL in a consecutive series of patients undergoing major lung resection for lung cancer, by using the SF36v2 preoperatively and at 1 and 3 months after operation (the revised and updated version of SF36v2 has never been used before in lung resection patients and has several advantages over the previous version); (2) to assess the correlation of QOL measures with functional determinants (forced expiratory volume in 1 second [FEV1], carbon monoxide lung diffusion capacity [Dlco], and exercise tolerance) evaluated at the same period; and (3) to verify whether residual QOL differed between patients grouped by age, chronic obstructive pulmonary disease status, sex, presence of coronary artery disease, extent of resection, pulmonary risk, and exercise tolerance. Patients and Methods Population One hundred ninety-one consecutive patients underwent major lung resection for nonsmall-cell lung cancer at our unit from July 2004 through September 2006 and were prospectively enrolled in this study. The study was approved by the local Institutional Review Board of the hospital, and all patients gave their informed consent to participate in the study. Postoperative early mortality was 2.6% (5 cases). Of the 186 patients surviving the operation, a total of 156 patients had complete cardiopulmonary (FEV1, Dlco, stair climbing test) and QOL assessment preoperatively and at 1 and 3 months postoperatively (144 lobectomies, 12 pneumonectomies). Thirty patients (15.7%) dropped out for several reasons (lung cancer recurrence, current chemotherapy, refusal). Follow-Up and Quality of Life Measure Patients were evaluated by pulmonary function test and symptom-limited stair climbing test before the operation 411 (usually 2 to 3 days before surgery) and 1 and 3 months postoperatively. In addition, QOL was measured by administration of the SF36v2, which is a generic QOL instrument that, through a 36-item short form survey, assesses eight health concepts (physical functioning, role limitation caused by physical problems, bodily pain, general health perception, vitality, social functioning, role limitation caused by emotional problems, mental health). Scores standardized to norms and weighted averages are used to create summary physical component scale (PCS) and mental component scale (MCS) scores on a standard scale [11]. The new SF36v2 offers significant measurement improvements over the original SF-36 Health Survey, including corrected deficiencies in the original SF-36 Health Survey, updated population norms, norm-based scoring for all eight scales that makes possible to meaningfully compare scores for eight scale profiles and PCS and MCS on the same graph, substantially increased precision and range of the rolefunctioning scales, significant enhancements in item wording and format, and less biased estimates of missing responses. The norm-based scoring algorithms introduced for all eight scales employ a linear score transformation, which scores scales that have a mean of 50 and a standard deviation of 10 in the 1998 US general population. The differences in scale scores can more clearly reflect the impact of the disease or treatment. In fact, any score less than 50 falls below the general population mean, and each point represents 1/10th of a standard deviation. General population norms serve as reference points or benchmarks for applications such as estimating disease burden and evaluating treatment benefit. After completion of the SF36v2 surveys, the patients performed pulmonary function tests and exercise test (symptom-limited stair climbing test) at each evaluation period. Pulmonary function tests were performed according to the American Thoracic Society criteria. The Dlco was measured by the single-breath method. Results of spirometry were collected after bronchodilator administration and were expressed as percentage of predicted for age, sex, and height according to the European Community for Steel and Coal prediction equations [12]. Thoracotomy chest pain at the time of repeat pulmonary function testing was assessed and, if any, controlled by administration of oral analgesics. In all cases, the visual analog scale scores before the pulmonary function tests and repeat exercise tests were kept below 2 (on a scale from 0 to 10). The stair climbing test was performed as a symptomlimited exercise. Our hospital has 16 flights of stairs, each flight having 11 steps. Each step is 0.155 m in height. The patients were asked to climb, at a pace of their own choice, the maximum number of steps and to stop only for exhaustion, limiting dyspnea, leg fatigue, or chest pain. During the exercise, pulse rate and capillary oxygen saturation were monitored by means of a portable pulse oximeter. GENERAL THORACIC

412 BRUNELLI ET AL Ann Thorac Surg LUNG RESECTION AND QUALITY OF LIFE 2007;84:410 6 Operative and Postoperative Management Patients were operated on by qualified thoracic surgeons and were managed in a dedicated thoracic surgery unit. Criteria for inoperability were predicted postoperative FEV1 (ppofev1) and predicted postoperative Dlco (ppodlco) less than 30% of predicted in association with insufficient exercise tolerance (height at preoperative stair climbing test less than 12 m or VO 2 max measured at cycle-ergometry less than 10 ml kg 1 min 1 ). As a rule, lung resections were performed through a musclesparing lateral thoracotomy. Postoperative management included chest physiotherapy, early as possible mobilization, antibiotic and antithrombotic prohylaxis, and thoracotomy chest pain control by continuous intravenous infusion of ketorolac and tramadol to keep the visual analog scale score below 3 or 4 in the first 72 hours (on a scale from 0 to 10, assessed twice daily). No formal preadmission or postdischarge physiotherapy or psychological supportive programs were administered. Table 1. Characteristics of the Patients in the Study (n 156) Variables Value Age 65.9 (10.7) Sex, males (n, %) 123 (79%) FEV1% 89.4 (18.3) Dlco% 80.6 (17.6) FEV1/FVC ratio 0.71 (0.09) ppofev1% 70.8 (16.3) ppodlco% 63.9 (15.5) Pneumonectomy (n, %) 12 (7.7%) Smoking pack-years 38.2 (28.6) Neoadjuvant chemotherapy (n, %) 14 (9%) Coronary artery disease (n, %) 18 (12%) Results are expressed as means SD unless otherwise specified. Dlco diffusion capacity of lung for carbon monoxide; FEV1 forced expiratory volume in 1 second; FVC forced vital capacity; ppo predicted postoperative. Fig 1. Box plot of the physical composite scale (open boxes) and mental composite scale (gray boxes) at different evaluation periods. Closed circles represent outside values. Statistical Analysis Normal distribution of SF36v2 scales was tested by the Shapiro-Wilk normality test. Plots of median SF36v2 scales and their confidence intervals were used to explore changes over time. To assess the significance of changes over time, preoperative and repeat postoperative (1 and 3 months) SF36v2 scales were compared by means of repeated measures analyses of variances with adjusted (Games-Howell post-hoc test) pairwise comparisons [13]. Subgroups (stratified by age, chronic obstructive pulmonary disease status, sex, presence of coronary artery disease, Dlco 70%, ppofev1 or ppodlco 40%, extent of operation) comparisons of physical and mental composite scales at 3 months were performed by means of the Mann-Whitney test. In addition, we explored the relationship between SF36v2 physical and mental composite scales and other functional determinants (FEV1, Dlco, exercise tolerance) measured at different evaluation periods (preoperatively and 1 and 3 months postoperatively) using correlation coefficients. All the statistical tests were two-tailed, with a significance level of p 0.05, and were performed on the statistical software Stata 8.2 (Stata Corp, College Station, Texas). Results The characteristics of the patients enrolled in the study are shown in Table 1. Twenty-two patients were treated by neoadjuvant chemotherapy before operation. These patients had a similar preoperative MCS score (47.5 versus 46.8, p 0.7), but significantly lower PCS score (46.3 versus 51.8, p 0.003) compared with those not treated. Figure 1 shows the box plot of the PCS and MCS scores at different evaluation periods. In particular, compared with preoperative values, the PCS was significantly reduced at 1 month (51 versus 45.1, p 0.0001), but completely recovered at 3 months (51 versus 52.4, p 0.2), whereas the MCS was not different among the three evaluation periods. Table 2 shows the results of the repeated measures analyses of variances comparing all the SF36v2 scales at different evaluation times. Compared with the general population means, candidates for lung resection with lung cancer had reduced preoperative values of role limitation caused by physical problems, general health perception, social functioning, role limitation caused by emotional problems, and mental health perception scales. All these scales, with the exception of general health, remained below 50 (general population mean) even 3 months after operation. With the exception of the physical functioning, social functioning, and mental health scales, the time factor had a significant influence on the changes of all the other physical and emotional scales. In general, compared with preoperative values, most of the SF36v2 scales were significantly reduced at 1 month after operation and returned to preoperative values at 3 months. Exceptions were the physical functioning scale, which remained

Ann Thorac Surg BRUNELLI ET AL 2007;84:410 6 LUNG RESECTION AND QUALITY OF LIFE Table 2. Repeated Measurements Analysis of Variance (ANOVA) of the SF36v2 Domains at Different Observation Periods Domains Preoperative 1 Month 3 Months ANOVA p Value Physical composite scale 51 (7) 45.1 (7.7) a 52.4 (6.9) b 0.0001 Mental composite scale 49.4 (9.6) 48.4 (11.3) 48 (12.8) 0.05 Physical functioning 51.8 (6.7) 48.3 (7.9) a 49.9 (8.6) a,b 0.2 Role physical 46.8 (11.3) 37.7 (11.3) a 46.5 (10.5) b 0.0001 Bodily pain 55.6 (9.7) 44.4 (10.5) a 52.2 (9.6) a,b 0.0001 General health 48.2 (9.2) 48.5 (8.6) 50.9 (7.9) a,b 0.05 Vitality 57 (9.8) 55.4 (8.9) a 57.1 (7.6) 0.06 Social functioning 48.9 (10.9) 45.7 (12.4) 49.3 (9.8) b 0.09 Role emotional 45.9 (14.2) 42.3 (14.6) a 47 (11.7) b 0.03 Mental health 47.8 (10.7) 48.4 (9.5) 48.2 (10.2) 0.08 413 GENERAL THORACIC a p 0.05 compared with preoperative values. b p 0.05 compared with 1-month values. Results are expressed as means SD. SF36v2 Short-Form 36-Item Health Survey, version 2. below preoperative value at 3 months, the general health perception, which improved at 3 months compared with preoperative value, and the mental health perception, which remained unchanged at all three evaluation periods (Figs 2 and 3). Table 3 shows the frequency of patients with PCS and MCS scores below norms ( 50) at each evaluation period. All correlation coefficients between PCS and MCS scores and FEV1, Dlco, and height reached at stair climbing test at each evaluation period were below 0.2, indicating poor correlation. Table 4 shows the results of the comparisons of the SF36v2 PCS and MCS scores 3 months after operation between different groups of patients stratified by age, chronic obstructive pulmonary disease status, sex, presence of coronary artery disease, pulmonary high risk (ppofev1 40% or ppodlco 40%), Dlco 70%, and extent of operation. Patients who underwent pneumonectomy had a significantly lower PCS (p 0.04) but similar MCS (p 0.2) compared with patients who underwent lobectomy at 3 months. No significant differences of PCS and MCS were noted between the other groups of patients. Comment The patients perspective about the surgical risk of lung resection may differ from that of the surgeons. What patients fear most is not an increased risk of perioperative major morbidity or mortality, but to be left physically and mentally handicapped and not be able anymore to resume an acceptable daily lifestyle [1]. Therefore, reliable information about the anticipated residual QOL should always complement the traditional data concern- Fig 2. Mean values of Short Form 36-Item, version 2, physical domains at different evaluation times. (Open bars physical functioning; gray bars role, physical; hatched bars bodily pain; black bars vitality; checkered bars physical composite scale.) Fig 3. Mean values of Short Form 36-Item, version 2, emotional domains at different evaluation times. (Open bars general health; gray bars social functioning; hatched bars role, emotional; black bars mental health; checkered bars mental composite scale.)

414 BRUNELLI ET AL Ann Thorac Surg LUNG RESECTION AND QUALITY OF LIFE 2007;84:410 6 Table 3. Distribution of Patients With Physical Composite Score (PCS) and Mental Composite Score (MCS) Below Norms at Each Evaluation Period (156 Patients) Preoperative 1 Month 3 Months PCS 50 (n, %) 67 (43%) 103 (66%) 68 (44%) MCS 50 (n, %) 81 (52%) 68 (44%) 66 (42%) ing early outcome and long-term survival whenever discussing the operation with the patients. Quality of life is not only an important consideration for those patients achieving long-term survival after curative resection, but even more for those who will experience recurrence months or years after operation. What is the price the patients are ready to pay for a treatment that, although the most efficacious among those available, can not guarantee long-term survival? Although a need exists for a more comprehensive understanding of the effects of thoracic surgery on patients functional and QOL outcomes, few studies have recently addressed this subject [1, 4 10]. The objective of the present study was to assess the QOL of patients before and after major lung resection for nonsmall-cell lung cancer by using the Short Form-36, version 2 questionnaire. The SF-36 has well-established reliability and has been reported to be sensitive to changes at 1, 6, and 12 months after thoracic surgery for nonsmall-cell lung cancer [14]. The new SF36v2 Health Survey offers significant measurement improvements over the original SF-36 Health Survey. One of the most important improvements is the introduction of normbased scores for all eight scales that makes possible to meaningfully compare scores for eight scale profiles and PCS and MCS. In this form, the differences in scale scores can more clearly reflect the impact of the disease or treatment. Our study has used this revised version of SF-36 for assessing QOL in lung resection patients. As all our patients were operated on for lung cancer, we chose the 3-month period as the latest evaluation time with the main intent to limit the drop-out rate. Indeed, at 3 months, 16% of patients operated on during the period of the study dropped out for several reasons (recurrence, adjuvant chemotherapy, refusal to show at follow-up). Concerns over drop-out rates have been reported in other studies [15, 16], which reported drop-out rates as high as 50% at 6 to 12 months after operation. Patients who fail to present at follow-up should presumably be considered the ones in the worse conditions, and prolonging the last evaluation time (ie, 6 or 12 months) could affect the results for a cream-skimming effect. Longerterm evaluations including the survivors of lung cancer operations (cured) will be certainly needed to assess the quality of life in this restricted cohort of patients and to see, from their perspective, the price they paid for cure. Nevertheless, in this study, we wanted to include the majority of patients undergoing major lung resections to have a wider picture of their postoperative physical and emotional status and correlate these measures with functional determinants. We found that candidates for lung resection with lung cancer had many physical and emotional preoperative QOL scales (role limitation caused by physical problems, general health perception, social functioning, role limitation caused by emotional problems, mental health) reduced compared with the general population, confirming previous investigations [6, 9]. In general, these reduced scales remained below norms even 3 months after operation. In fact, 40% to 50% of patients had PCS and MCS scores below norms before and 3 months after the operation. These patients may be the ones to benefit the most of specific program of physical rehabilitation, information, and psychological and social support. Nevertheless, we found that most of the QOL scales returned to preoperative levels by 3 months. This is in contrast with the findings of Handy and colleagues [6], who reported values of physical functioning, role limitation caused by physical problems, bodily pain, and mental health lower than preoperative scales at 6 months, but confirmed the results of Win and colleagues [8], who showed that global quality of life and several functioning scales had returned to preoperative values by 3 months. Many studies have examined the residual cardiorespiratory function after lung resection. The unspoken leap of logic in such literature is that decreased functional capacity, as assessed by traditional pulmonary function test and exercise tests, translates to impaired QOL. This assumption would be acceptable only in case QOL scales had a high correlation with traditional functional measures. Unfortunately, we found that physical and mental composite scales had a poor correlation with FEV1, Dlco, Table 4. Results of Comparisons of SF36v2 Physical Composite Scale (PCS) and Mental Composite Scale (MCS) Scores 3 Months After Operation Among Different Subgroups of High-Risk Patients PCS p Value MCS p Value Age 70 years (65) 50.5 (7) 0.8 50.5 (9.3) 0.4 Young (91) 50.9 (7.4) 49.4 (9.5) COPD (29) 51.7 (7) 0.5 51.6 (8.9) 0.3 Non-COPD (127) 50.5 (7.3) 49.5 (9.5) Pulmonary risk (12) 48.8 (7.7) 0.3 49.6 (10.1) 0.9 Nonpulmonary risk 51 (7) 49.9 (9.5) (134) Dlco 70% (41) 50.8 (6.4) 0.9 49.6 (9.8) 0.8 Dlco 70% (115) 50.7 (7.5) 49.9 (9.4) Male (123) 50.7 (7.2) 0.9 50.2 (9.4) 0.4 Female (33) 50.7 (7.2) 48.6 (9.5) Pneumonectomy (12) 46.8 (6.5) 0.04 52.8 (8.9) 0.2 Lobectomy (144) 51.1 (7.2) 49.6 (9.5) CAD (18) 50.3 (8.4) 0.8 52 (7) 0.3 Non-CAD (138) 50.8 (7.1) 49.6 (9.7) Results are expressed as means SD. Chronic obstructive pulmonary disease (COPD) by GOLD criteria: FEV1 80% FEV1/FVC ratio 0.7. Pulmonary risk: ppofev1 40% or ppodlco% 40%. CAD coronary artery disease; SF36v2 Short-Form 36-Item Health Survey, version 2. Other abbreviations as in Table 1.

Ann Thorac Surg BRUNELLI ET AL 2007;84:410 6 LUNG RESECTION AND QUALITY OF LIFE and stair climbing performance at each evaluation period. Thus, pulmonary function tests and exercise tests can not be taken as surrogates for QOL evaluation, as discrepancies may exist between patients perception about their residual physical and emotional status and objective functional measures. An ad-hoc instrument should always be used for QOL evaluation. We then examined whether residual quality of life 3 months after operation was different in groups of patients considered at increased risk compared with lower risk ones. Only those patients undergoing pneumonectomy had lower residual PCS score (but similar MCS score) compared with patients undergoing lobectomy, confirming another study [9]. On the other hand, patients older than 70 years of age, with chronic obstructive pulmonary disease, Dlco lower than 70%, ppofev1 or ppodlco lower than 40%, and with a coronary artery disease had similar residual PCS and MCS scores at 3 months compared with their lower-risk counterparts. These findings may have great importance during patients counseling before the operation. Patients concerns about their residual QOL should be regarded at least on equal footing with their concerns about perioperative morbidity and mortality and long-term survival. Particularly for patients deemed at higher risk for postoperative major cardiopulmonary complications, the information that residual QOL will be similar to the one experienced by younger and fitter patients may help in their decision to proceed with surgery. This information will be of paramount importance even for the referring physicians, who will be less reluctant to refer high-risk lung cancer patients to surgery. This study has potential limitations. The first limitation is one common to most of the follow-up analyses and concerns the dropped-out patients. As these patients could have been those in the worst functional status, their inclusion in the analysis could have perhaps changed the results, and that should be taken into account when interpreting the results. Second, a certain proportion of our patients had adjuvant chemotherapy. As chemotherapy has been proven to impair the gas exchange [17], and may have a detrimental effect both on the physical and emotional status, the inclusion of these patients could have influenced the results. As most of our patients started chemotherapy 4 to 6 weeks after operation, the problem refers mainly to the last evaluation time (3 months). However, only 19 of the 156 patients studied at 3 months underwent adjuvant chemotherapy. We selected to include the 19 patients under chemotherapy after a preliminary analysis that did not show differences in SF36v2 scales at 3 months compared with the other patients. Quality of life measures reflect the patients perspective and may be affected by several external factors with an emotional impact such as the radicality of the procedure, the degree of satisfaction of received care, and the social support at home, among others. Further analyses are needed to investigate the influence of these factors on the postoperative residual QOL. In conclusion, we were able to show that candidates for lung resection with lung cancer had reduced preoperative QOL measures compared with general population. Although these measures return to preoperative values by 3 months, many scales still remained below norms. Supportive physical rehabilitation and psychological programs seem indicated for these patients to facilitate and protect their postoperative recovery. We also showed that QOL measures did not correlate well with pulmonary function test and exercise test performance. Therefore, these functional measures can not substitute for specific QOL evaluation instruments. Finally, patients traditionally considered at higher risk for lung resection showed residual physical and emotional status comparable to that observed among younger and fitter patients. The use of simple questionnaires such as the SF36v2 for assessing the preoperative and postoperative QOL of lung resection candidates should, in our view, be systematically used to provide patients with a reliable estimate of their residual QOL and possibly plan supportive physical and psychological programs. The results generated in this study need to be confirmed by future prospective investigations, even on a multicentric basis, including also longer-term evaluations. References 415 1. Cykert S, Kissling G, Hansen CJ. Patient preferences regarding possible outcomes of lung resection. What outcomes should preoperative evaluations target? Chest 2000;117:1551 9. 2. McManus K. Concerns of poor quality of life should not deprive patients of the opportunity of curative surgery. Thorax 2003;58:189. 3. Berrisford R, Brunelli A, Rocco G, Treasure T, Utley M. The European Thoracic Surgery Database project: modelling the risk of in-hospital death following lung resection. Eur J Cardiothorac Surg 2005;28:306 11. 4. Zieren HU, Muller JM, Hamburger U, Pichlmaier H. Quality of life after surgical therapy of bronchogenic carcinoma. Eur J Cardiothorac Surg 1996;10:233 7. 5. Sarna L, Padilla G, Holmes C, Tashkin D, Brecht ML, Evangelista L. Quality of life of long term survivors of non-small-cell lung cancer. J Clin Oncol 2002;20:2920 9. 6. Handy JR, Asaph JW, Skokan L, et al. What happens to patients undergoing lung resection? Outcomes and quality of life before and after surgery. Chest 2002;122:21 30. 7. Uchitomi Y, Mikami I, Nagai K, Nishiwaki Y, Akechi T, Okamura H. Depression and psychological distress in patients during the year after curative resection of non-smallcell lung resection. J Clin Oncol 2003;21:69 77. 8. Win T, Sharples L, Wells FC, Ritchie AJ, Munday H, La Roche CM. Effect of lung cancer surgery on quality of life. Thorax 2005;60:234 8. 9. Myrdal G, Valtysdottir S, Lambe M, Stahle E. Quality of life following lung cancer surgery. Thorax 2003;58:194 7. 10. Barlesi F, Doddoli C, Loundou A, Pillet E, Thomas P, Auquier P. Preoperative psychological global well index (PG- WBI) predicts postoperative quality of life for patients with non-small cell lung cancer managed with thoracic surgery. Eur J Cardiothorac Surg 2006;30:548 53. 11. Hays RD, Morales LS. The RAND-36 measures of health related quality of life. Ann Med 2001;33:350 7. 12. Quanjer PhH, Tammeling GJ, Cotes JE, Pedersen OF, Peslin R, Yernault JC. Lung volumes and forced ventilatory flows. Report Working Party. Standardization of lung function tests. European Community for Steel and Coal. Official GENERAL THORACIC

416 BRUNELLI ET AL Ann Thorac Surg LUNG RESECTION AND QUALITY OF LIFE 2007;84:410 6 statement of the European Respiratory Society. Eur Respir J 1993;6(Suppl 16):5 40. 13. Steel RGC, Torrie JH. Multiple comparisons. In: Steel RGC, Torrie JH, eds. Principles and procedures of statistics: a biometrical approach. 2nd ed. New York: McGraw-Hill, 1980:173 94. 14. Mangione CM, Goldman L, Orav EJ, et al. Health-related quality of life after elective surgery: measurement of longitudinal changes. J Gen Intern Med 1997;12:686 97. 15. Markos J, Mullan BP, Hillman DR, et al. Pre-operative assessment as a predictor of morbidity and mortality after lung resection. Am Rev Respir Dis 1989;139:902 10. 16. Bolliger CT, Jordan P, Soler M, et al. Pulmonary function and exercise capacity after lung resection. Eur Respir J 1996;9: 415 21. 17. Leo F, Solli P, Spaggiari L, et al. Respiratory function changes after chemotherapy: an additional risk for postoperative respiratory complications. Ann Thorac Surg 2004;77:260 5. INVITED COMMENTARY Traditionally, surgical mortality and morbidity studies have been limited to in hospital or 30-day outcomes. Brunelli and colleagues [1] article addresses important issues that only recently have become apparent to most surgeons (ie, long-term concerns beyond immediate postsurgical mortality and complications are quite significant). Not only is this information important to patients [2], but it is also information that healthcare policymakers need. This report is well conceived with good statistical analyses and uses the most updated methods of quality of life analyses to provide excellent outcome data. However, as with many of these studies, this report is based on a relatively small patient population, particularly with patients undergoing more extensive resections such as pneumonectomies and bi-lobectomies, and with dropouts of patients from follow-up for whatever reason, which poses a problem for this type of small study. The author s presumption that follow-ups longer than 3 months introduced an important selection bias is true; however longer follow-up information is quite relevant to the patient. The role of surgical resection for lung cancer is to achieve a cure, and cancer recurrence is a failure of that surgical treatment option. In evaluating surgically related long-term quality of life, the presumption is that the patient will be cured of his cancer by the operation. What the patient wishes to know is what will his quality of life be like if his cancer is cured? Because most resected lung cancers that are not cured surgically will recur within 2 years, a 2-year follow-up quality of life study may provide the most useful information for the patient. Interestingly, in an earlier paper by Brunelli and colleagues [3], they demonstrated that pulmonary function was continuing to improve at the 3-month follow-up interval, and the actual and predicted postoperative pulmonary function in their graphs had not leveled, suggesting that there could be further improvement of pulmonary function with time. James W. Asaph, MD Earle A. Chiles Research Institute Providence Portland Medical Center 4401 SW Westdale Dr Portland, OR 97221-3158 e-mail: mrnous@aol.com References 1. Brunelli A, Socci L, Refai M, Salati M, Xiumé F, Sabbatini A. Quality of life before and after major lung resection for lung cancer: a prospective follow-up analysis. Ann Thorac Surg 2007;84:410 6. 2. Cykert S, Kissling G, Hansen CJ. Patient preferences regarding possible outcomes after lung resection: what outcomes should preoperative evaluations target? Chest 2000;117: 1551 9. 3. Brunelli A, Refai M, Salati M, Xiume F, Sabbatini A. Predicted versus observed FEV-1 and DLCO after major lung resection: a prospective evaluation at different postoperative periods. Ann Thorac Surg 2007;83:1134 9. 2007 by The Society of Thoracic Surgeons 0003-4975/07/$32.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2007.04.055