Results of surgical resection in patients over the age of 70 years with non small-cell lung cancer

European Journal of Cardio-thoracic Surgery 11 (1997) 432 439 Results of surgical resection in patients over the age of 70 years with non small-cell lung cancer U. Morandi, A. Stefani, M. Golinelli, C. Ruggiero, L. Brandi, A. Chiapponi, C. Santi, R. Lodi* Department of Cardiothoracic Surgery, Uni ersity of Modena, Largo del Pozzo 71, 41100 Modena, Italy Received 25 July 1995; revised 4 October 1995; accepted 13 September 1996 Abstract Methods: From January 1989 to October 1993, at the Department of Cardio-Thoracic Surgery of the University of Modena, 806 patients underwent thoracotomy and curative pulmonary resection for non small-cell lung cancer. Eighty five patients were 70 years old or older (mean 73.4 years, range 70 88). There were 78 males (91.7%) and 7 females (8.3%). This population was compared to 130 younger patients (under 70 years old), treated during the same period and with similar features with respect to the type of resection, sex, histology, grading and staging. Lobectomy was the procedure of preference in both groups. Results: As regards postoperative mortality and overall complications, no significant differences were noted between the two groups of patients (two younger patients died and 43.8% had postoperative complications; one patient of the older group died and 55.2% had postoperative complications), but in the older ones a higher incidence of cardiovascular complications was found (P 0.01). With respect to the long-term survival (follow-up 12 70 months), no significant difference was found between the two groups. Conclusion: Such findings show that pulmonary resection for bronchogenic cancer is feasible and justified in patients more than 70 years old, even if a higher incidence of cardiovascular complications may occur: a careful preoperative selection ought to be performed and lobectomy should be preferred. 1997 Elsevier Science B.V. Keywords: Lung cancer; Elderly patients; Surgical treatment 1. Introduction Lung cancer is the most frequent neoplasm observed in the males of Western countries and its incidence is growing gradually in females too [7]. Life expectancy is increasing, the number of patients 70 years with lung cancer is increasing also. Since surgery is the only potentially curative form of treatment, we ought to extend the surgical indications to elderly patients as much as possible [9]. These patients need a very careful assessment of surgical indications, * Corresponding author. Tel.: +39 59 422171; fax: +39 59 360159. clinical staging, cardiovascular and pulmonary function and a careful selection of the appropriate operative procedure. Such approach allows the right management of the surgical patients who are more than 70 years old [2,6,11]. In a previous paper, we [7] as well as other authors [11 13] reported high survival rates. This induced us to a further investigation into the problems of lung resection in elderly patients and to a review of our late experience (1989 1993). The purpose of this paper is to assess the risks and the results of surgery in the elderly population, by investigating postoperative complications and longterm survival. 1010-7940/97/$17.00 1997 Elsevier Science B.V. All rights reserved. PII S1010-7940(96)01084-6

U. Morandi et al. / European Journal of Cardio-thoracic Surgery 11 (1997) 432 439 433 2. Patients and methods From January 1 1989 to October 1 1993, at the Department of Cardio-Thoracic Surgery of the University of Modena, 806 patients underwent thoracotomy for primary lung cancer. Of these, 107 were 70 years: these were defined as elderly patients. In 22 of these (20%), we performed a thoracotomy without resection; eight patients (7%) underwent exploratory thoracotomy and 14 (13%) underwent diagnostic trans-axillary thoracotomy. The remaining 85 patients underwent curative pulmonary resection, so they were included in the present paper. There were 78 males (91.7%) and seven females (8.3%), whose mean age was 73.4 years (S.D. 3.4; range 70 88). These people were compared to a group of 130 younger patients, who were 70 years (mean age 60.2 years, S.D. 12.7; range 42 69) and underwent curative pulmonary resection during the same period at our department. This group of patients were chosen at random. 2.1. Preoperati e functional assessment Pulmonary function was determined by spirometry and arterial blood gases, measurements at rest and on exercise in all cases. Cardiovascular function was assessed by ECG and echocardiography in every patient. An exercise test with a bicycle ergometer was an additional test for all elderly patients. They needed further assessment because of their higher risk of perioperative complications. In the presence of symptoms of angina, peripheral vascular disease, cerebrovascular disease, congestive heart failure or late myocardial infarction, even younger patients ( 70 years) were studied by an exercise test with bicycle ergometer. Bicycle ergometry testing was performed using stepwise increased loads, starting with 40 W and continuing with 20 W increments every second min, until 90% of the predicted maximal heart rate was obtained. The test was terminated 4 min after the 90% of the maximal cardiac rate was achieved or at the onset of limiting dyspnea, chest pain, fatigue or signs of myocardial ischemia and arrhythmias. A positive response to exercise testing required ST segment depression of at least 1 mm, angina during or after the exercise or both of these. When the FEV1 value was lower than 0.6 l, a conventional resection, that is pneumonectomy or (bi)lobectomy, was contraindicated. If the FEV1 value was greater than 1 l, the patient underwent conventional resection, whenever possible. Minimum criteria for pneumonectomy were a FEV1 greater than 2 l and a mean forced expiratory flow greater than 1.6 l. When the PaO 2 value was 60 mmhg and/or the PaCO 2 45 mmhg, the patient was considered inoperable for conventional resection. Patients with border-line values underwent conservative resections (wedge or segmentectomy), if they otherwise appeared to be operable. Perfusion lung scans were performed in border-line patients, with impaired pulmonary function but who would be candidates for conventional resection. With respect to cardiovascular function, surgical resection was contraindicated in the presence of ischemic heart disease in the third or fourth class CCS. In these cases, a pharmacologic therapy was administered to improve myocardium functions and reduce perioperative risks (Cycle ergometer test positive at heart rate 100 bpm). If pharmacologic therapy was unsuccessful in restoring myocardial function, the patient underwent coronarography and, when indicated, a revascularization was considered. 2.2. Clinical data and type of resection The data recorded from each patient of both groups included sex, performance status (according to Karnofski), histology and degree of differentiation of the tumor, postoperative staging and type of resection (Table 1). Preoperative staging was assessed in all patients with chest roentgenograms, CT scan of the chest, abdomen and pelvis, bronchoscopy. CT scan of the brain and radionuclide bone scan were performed only in symptomatic patients. When the evaluation of the mediastinal lymph nodes by CT scanning was positive for advanced N2 or N3 disease, we performed a mediastinoscopy. If the nodal involvement was confirmed by the mediastinal exploration, the patient underwent neoadiuvant chemotherapy. Once restaged, some of these patients could undergo curative pulmonary resection, but they were not included in this study (28/806 cases, 3.5%). The type of resection was dictated by the local extent of the tumor and the cardiorespiratory function. A complete mediastinal lymph node dissection, combined with the pulmonary resection, was performed in all cases. Intraoperative complications were carefully recorded in the anesthesiologist s chart, the postoperative ones in the clinical chart in our centre of intensive care. 2.3. Follow-up Most patients have been directly followed at our institution with periodic office visits. The information about the long-term survival of patients who have not been followed at our institution, were obtained either at the general practitioner s office or by telephone interviews with the patient or his/her relative. Follow-up was complete for all patients within December 1995; it ranged from 24 to 73 months.

434 U. Morandi et al. / European Journal of Cardio-thoracic Surgery 11 (1997) 432 439 Table 1 Clinical data Variable Elderly patients (n=85) Younger patients (n=130) P Sex ratio (M/F) 78/7 (11:1) 111/19 (6:1) 0.2 Histology Epidermoid carcinoma 51 (60%) 66 (51%) 0.4 Adenocarcinoma 29 (34%) 55 (42%) Large cell carcinoma 2 (2%) 5 (4%) Small call carcinoma 3 (4%) 4 (3%) Degree of differentiation Well differentiated 6 (7%) 18 (14%) Moderately differentiated 34 (40%) 53 (41%) Poorly differentiated 44 (52%) 56 (43%) Undifferentiated 1 (1%) 3 (2%) 0.2 Surgical staging I 48 (57%) 79 (61%) 0.7 II 12 (14%) 14 (11%) IIIA 19 (22%) 29 (22%) IIIB 6 (7%) 8 (6%) Performance status 90 44 (52%) 77 (59%) 90 41 (48%) 53 (41%) 0.3 Operation Pneumonectomy 11 (13%) 20 (15.4%) 0.06 Bilobectomy 6 (7%) 19 (14.6%) Lobectomy 58 (68%) 86 (66%) Conservative resections* 10 (12%) 5 (4%) *P 0.05. The data concerning the long-term survival were expressed with the survival curves (absolute survival). 2.4. Statistical analysis The descriptive analysis was expressed in terms of the frequency, mean and S.D. Frequencies were compared with the 2 test (Yates correction was done for small samples) and the Fischer s test. The probability of survival was calculated according to the Kaplan Meier method and resulting curves were compared by the log-rank test. A probability value of 0.05 was considered statistically significant. 3. Results Table 1 shows that there were no significant differences among elderly patients and younger ones with respect to sex, histology, degree of differentiation, pathological staging, performance status and type of resection. In both groups, epidermoid carcinoma was the most common cell type. Patients, 127 (59%) were classified as having stage I carcinomas, on the basis of histopathological examination. In most cases (171/215, 79%) a N2 involvement was not demonstrated. Lobectomy was the procedure of preference in both groups: 68% in the elderly patients, 66% in the younger ones. As regards the type of resection, we found that the difference was nearly significant (P=0.060), thus, we performed further investigation into each operative procedure in both groups. It showed that conservative resections (segmental and wedge resections) were significantly more frequent in patients 70 years old (P=0.050). In the younger group bilobectomy was performed twice (14.6 vs. 7%), although this difference was not statistically significant. These two groups were comparable with respect to all clinical-pathological features; the age distribution was just the main difference between them and likely responsible for the differences in postoperative course and long-term survival. 3.1. Postoperati e complications and mortality No intraoperative deaths occurred. Three patients died in the postoperative course (hospital mortality): one in the elderly group, two in those who were 70 years. The older patient (Stage IIIA), who had undergone left pneumonectomy, died after 17 days of respiratory insufficiency; the younger ones, both undergoing

U. Morandi et al. / European Journal of Cardio-thoracic Surgery 11 (1997) 432 439 435 Table 2 Incidence of respiratory complications (deaths included) Variable Elderly patients (n=85) Younger patients (n=130) Total (n=215) Respiratory complications 40 (47%) 57 (43.8%) 97 Major complications 12 (14.1%) 18 (13.8%) 30 Pneumonia 5 3 8 Pleural effusion 2 4 6 Prolonged bronchospasm 1 0 1 Atelectasis 2 5 7 Bronchial fistula 1 4 5 Subcutaneous emphysema 1 2 3 Minor compl. 28 (32.9%) 39 (30%) 67 Pleural effusion 14 16 30 Transient bronchospasm 1 0 1 Atelectasis 1 6 7 Prolonged air leak 4 12 16 Subcutaneous emphysema 7 3 10 Hemoptysis 1 2 3 bilobectomy (Stage I and Stage IIIA), died after 15 and 26 days of septic shock from empyema and congestive heart failure from myocardial infarction, respectively. Postoperative complications occurred in 104 patients (48.3%), but only 37 patients suffered from major complications (Tables 2 and 3). Pleural effusion, diagnosed on chest radiographs, was considered a major complication when affecting the patient s clinical status, so that a therapeutic thoracentesis was needed. Two younger patients developed an empyema, but they have had a carcinomatous abscess; this was considered a major complication, reported as a major pleural effusion in Table 1. Also the five patients in which a bronchial fistula occurred developed an empyema. Lung atelectasis was considered a major complication if it necessitated a bronchoscopy to re-expand the lobe or lung affected. We considered abnormal a persistent air leak beyond 6 days and we classified that as a minor complication. Subcutaneous emphysema was considered a major complication only when it needed subcutaneous needles, thoracic or cervical incisions or a chest tube drainage system to decompress the area. With respect to cardiac arrhythmias, we classified as major complications only the ventricular ones. Postoperative complications were more frequent in elderly patients, though the difference between the two groups was not statistically significant. Overall morbidity rates were 55.2% in patients who were 70 years and 43.8% in the patients 70 years; it means that 47 patients out of 85 and 57 out of 130, respectively, suffered from complications (P 0.05). Relating morbidity to surgical staging in the older group, no significant differences were noted: the morbidity rates were 60% for patients with Stage IIIA IIIB disease, 41% for patients having Stage II disease and 56% for the ones with Stage I disease. The incidence of cardiovascular complications was significantly higher in the elderly patients: 25/85 (29.4%) compared to 18/130 (13.8%), P 0.01. It was mostly the higher incidence of the minor complications to determine this difference, 21/85 (24.7%) vs 15/130 (11.5%), P=0.019. In the older group, the incidence of major cardiovascular complications was twice higher than in the younger, though this difference was not statistically significant (P=0.27). With respect to respiratory and extrathoracic complications, no significant difference was noted among elderly and younger patients: respectively 40/85 (47%) vs. 57/ 130 (43.8%) and 4/85 (4.7%) vs. 4/130 (3%). Four out of 85 elderly patients suffered from extrathoracic complications: hemorrhagic cystitis (2), gastrointestinal bleeding (1) and TIA (1); in the younger ones, we observed two cases of acute renal failure, a hemorrhagic cystitis and a gastrointestinal bleeding. None of these was fatal, they all resolved with adequate medical treatment. Relating morbidity to the type of resection, there were no significant differences in the postoperative course of patients 70 years compared with patients who were 70 years (Table 4): the overall morbidity rates were 40% for conservative resection in both groups (P= 0.713), 54.6 and 46.6% respectively for (bi)lobectomy (P=0.330). An increase of overall morbidity rate for pneumonectomy was noted in the elderly patients compared to the younger ones: 8/11, 72.7% and 6/20, 30%, respectively. The difference was nearly significant: P= 0.055. On the other hand, relating morbidity to the type of resection in every single population separately, no significant differences were found: P=0.179 in the elderly patients, P=0.132 in the younger patients.

436 U. Morandi et al. / European Journal of Cardio-thoracic Surgery 11 (1997) 432 439 Table 3 Incidence of cardiovascular complications (deaths included) Variable Elderly patients (n=85) Younger patients (n=130) Total Cardiovascular complications* 25 (29.4%) 18 (13.8%) 43 Major complications 4 (4.7%) 3 (2.3%) 7 Ventricular arrhythmias 1 1 2 Prolonged hypertension 1 0 1 Angina pectoris 1 0 1 Congestive heart failure (myocardial infarction) 1 2 3 Minor complications** 21 (24.7%) 15 (11.5%) 36 Atrial fibrillation or flutter 16 13 29 Supraventricular extrasystolia 2 0 2 Paroxysmal supraventricular tachycardia 1 1 2 Transient hypertension 2 1 3 *P 0.01. **0.05 P 0.01. 3.2. Follow-up Up to December 1995, 123 (57%) out of 215 patients undergoing pulmonary resection between 1989 and 1993, died. The overall mortality rates were 56% (48/85) in patients who were more than 70 years and 58% (75/130) in the younger ones. With respect to the long-term survival, the curves analysis demonstrates no significant differences between the two groups (Fig. 1): 1 year survival was 75% in the younger patients and 72% in the older ones; 64 and 61%, respectively, at 2 years; 50 and 47% at 3 years; 41 and 37% at 4 years; 35 and 28% at 5 years (P=0.54). We related overall long-term survival to the surgical staging: in patients with Stage I disease, 1 year survival was 89% and 5 year survival 51%; Stage II 75 and 13%, respectively; Stage IIIA 1 year survival 46% and 4 year survival 8%; Stage IIIB 1 year survival 30% and 3 year survival 10%. The difference is significant (P 0.01). Relating long-term survival to staging, degree of differentiation, histology, sex and type of resection, no differences were found between the two populations: P=0.779, 0.472, 0.505, 0.543 and 0.468, respectively. Among the 37 old patients still living up to December 1995, 6 (16%) relapsed: 4 had locoregional recurrences and two had distant metastases, to the liver (1) and bone (1). Recurrent disease occurred in six (11%) of the 55 younger patients who were still alive: three had locoregional recurrences and three had metastases, to supraclavicular and laterocervical lymph nodes (2) and liver (1). Recurrent malignant disease was determined to be the cause of death in 107 patients (86.9%). The causes of death deemed to be unrelated to cancer were mostly myocardial infarction and respiratory insufficiency. Of 16 patients who died of unrelated causes, 10 were 70 years (20.8%) and six 70 years (8%); the difference was nearly significant (P=0.073). In the elderly, death of unrelated causes is higher, thus we analyzed survival curves considering only related deaths (unrelated ones were considered as lost to follow-up). No differences were found between older patients and the younger ones: 1 year survival was 77 and 76% respectively, 5 year survival was 39% in both groups (P=0.79). 4. Discussion There is general agreement that surgical intervention should be the treatment of choice for elderly patients with bronchogenic carcinoma, provided that the indication is appropriate and that the selection of patients is adequate [2,6,9,13,16]. In recent studies the long-term survival of septuagenarian patients has been shown to be comparable to that of younger patients [1,3,6,13,16]. The results of our own study confirm these data: we found no difference in long-term survival between the two age groups. Nevertheless, the probabilities of long-term survival ought to be weighed against the risks of such a surgical treatment in elderly patients. When initial reports appeared, high postoperative mortality rates, close to 20%, led to consider this surgery too dangerous for septuagenarians; thus, advanced age was a contraindication to surgical approach [15]. Lower mortality rates have been reported in recent studies, but still mortality was significantly higher than the one seen in the younger patients. Thomas et al. reported a postoperative mortality rate of 13% where pneumonia and myocardial infarction were the main causes of death [13]. Borelly et al. recorded a mortality rate of 19% and death was mostly due to pulmonary complication and ischemic heart disease. Lower mortality rates (3 7%) in this age-group have been recorded by some authors

U. Morandi et al. / European Journal of Cardio-thoracic Surgery 11 (1997) 432 439 437 Table 4 Postoperative complications by age and operative procedure Pneumonectomy* Lobectomy bilobectomy Conservative resection Total 70 6/20 (30%) 49/105 (47%) 2/5 (40%) 57/130 (44%) 70 8/11 (73%) 35/64 (55%) 4/10 (40%) 47/85 (55%) Total 14/31 (45%) 84/169 (50%) 6/15 (40%) 104/215 (48%) *P=0.055. [3 6] and our results are comparable with this report (only a patient died in the postoperative period, 1.17%). Since most authors perform a careful preoperative functional assessment in the elderly patients, these differences should be partly attribute to the use of lungsparing operations. Thus, 51% of the older patients in the series of Thomas underwent pneumonectomy and 31% of Borelly s, as opposed to 19% in the series of Ginsberg (5), 13% of our series (11/85), 10% of Breyer s (3) and 6% of Ishida s (6). As previously reported [2,6,11,13], postoperative complications occurred in over one-third of the patients (48.3%). Of the 85 elderly patients, 47 suffered from postoperative complications; such a morbidity rate (55.2%) seems to be high, just because we included in our series even slight complications (pleural effusion, transient atrial fibrillation), that may explain the apparently high morbidity rate. As in other reported series [10,13,14], cardiovascular complications were significantly more frequent in older patients than in younger ones: 29.4 compared to 13.8%. It was mostly the higher incidence of the minor ones, such as supraventricular arrhythmias, that determined this difference. However, major cardiovascular complication too are much more frequent in the elderly group. Despite the careful preoperative cardiovascular evaluation (routine echocardiography and cycle ergometer exercise test), the choice of an aggressive surgical approach in the elderly patients still seems to be difficult. This may partly be due to the great incidence of coexisting diseases and to the restricted cardiac and pulmonary reserve because of advanced age. Anyhow, even the overall morbidity rate was higher in the elderly group of patients (55.2 compared to 43.8% in the younger ones), although this difference was not statistically significant; otherwise, as regards the incidence of respiratory complications, no significant difference was noted. If we reduced the minimal values for operability in elderly patients, the risk of postoperative complications would likely decrease. Nevertheless, because the only significant difference was about minor cardiovascular complications, we would exclude from surgery a group of patients just to avoid such slight complications. In our paper we excluded patients who underwent surgery after neoadiuvant chemotherapy, because this represents a further factor affecting morbidity and mortality. A specific study about these relationships would be interesting and more suitable. In the younger group the proportion of women is twice as high than in the older: 14.6% vs. 8.2%. This difference, previously reported [13], ought to be emphasized, though it was not significant: a further evaluation of the relationship among bronchogenic cancer, sex and age would be interesting. With respect to the type of resection performed, conservative resections were significantly more frequent in patients who were 70 years. This is likely because of the poorer cardiopulmonary function in these patients, that led us to limit the extent of pulmonary resection. Relating morbidity to the type of resection, we observed an increase of the morbidity rate in the elderly patients only for pneumonectomy: postoperative complications in patients who were 70 years who underwent pneumonectomy were more frequent than in the younger ones and the difference was almost significant. It is common knowledge that the mortality and morbidity rates are significantly higher in patients undergoing pneumonectomy and the 5 year survival rate is lower. These risks are increased in the elderly patients. On the contrary, conservative operations result in a lower mortality and morbidity [3,5,6,13], but this approach may lead to a greater incidence of local recurrence and to a reduced survival, because of its poor carcinologic value [4]. In our series, no postoperative deaths occurred in the patients who underwent conservative resection (10 70 years, 5 70 years); the morbidity rate was comparatively low, 40% in both groups; we observed only one major complication, a pneumonia in a patient who was 70 years. Otherwise, only 26% of patients undergoing conservative resection (4 out of 15) survived more than 3 years; there were two elderly patients (20%) and two younger ones. These are low survival rates, compared to the mean values reported in our series. Whenever possible, lobectomy is the preferable operative procedure [8,9,16]: it permits lung tissue preservation without compromising the adequacy of the removal of the tumor. Thus, survival rates are higher than for conservative resection and morbidity rates are lower than for pneumonectomy. The incidence of the causes of death unrelated to cancer was higher in the elderly patients than in the

438 U. Morandi et al. / European Journal of Cardio-thoracic Surgery 11 (1997) 432 439 Fig. 1. Survival curves in elderly and younger patients. younger ones (P=0.078). It may be due just to advanced age, limiting the life time, independently of the surgical treatment. We also investigated long-term survival without unrelated deaths, but no difference was noted between the two groups of patients. The major findings in the present paper were: (1) The incidence of postoperative complications and the long-term survival in the older patients were similar to those in the younger ones, though the cardiovascular complications were more frequent in the elderly patients. (2) The choice of an appropriate and selective operative procedure is essential. Pneumonectomies have been demonstrated to carry a higher risk than lesser resections, otherwise conservative resections seem to have a real carcinologic value only in very selected cases. (3) Finally, a great importance assumes the preoperative evaluation of cardiovascular and pulmonary function in the elderly patients, to select those who underwent a surgical treatment and to choose an optimal operative procedure for the individual patient. We conclude, therefore, that, since surgery remains the most effective treatment to a patient with lung carcinoma, advanced age per se is not a contraindication to the curative pulmonary resection. In fact, even if the incidence of minor complications was higher in elderly patients, good results have been obtained as regards long-term survival. Since benefits are more than risks, we think that pulmonary resection is justified in these patients. A careful preoperative assessment ought to be performed and lobectomy should be preferred, in order to decrease the high risk factors and to reduce the incidence of the major postoperative complications. This can allow more patients to undergo surgery safely, even if they are 70 or 80 years old. References [1] Berggren H, Ekroth R, Malmberg R, Naucler J, William Olssen G. Hospital mortality and long-term survival in relation to preoperative function in elderly patients with bronchogenic carcinoma. Ann Thorac Surg 1984;38:633 636. [2] Borelly J, Grosdidier G, Sibille P. L exérèse du néoplasme bronchique chez le sujets de 70 ans et plus. A propos d une série de 193 exérèse. Ann Chir 1992;46:130 133. [3] Breyer RH, Zippe C, Pharr WF, Jensik RJ, Kittle CF, Faber LP. Thoracotomy in patients over age 70 years. 10 years experience. J Thorac Cardiovasc Surg 1981;81:187 193. [4] Errett LE, Wilson J, Chu-Jeng Chiu R, Munro DD. Wedge resection as an alternative procedure for peripheral bronchogenic carcinoma in poor-risk patients. J Thorac Cardiovasc Surg 1995;90:656 661. [5] Ginsberg RJ, Hill LD, Eagan RT et al. Modern 30 day operative mortality for surgical resection in lung cancer. J Thorac Cardiovasc Surg 1983;86:654 658. [6] Ishida T, Yokoyama H, Kaneko S, Sugio K, Sugimachi K. Long-term results of operations for non-small cell lung cancer in the elderly. Ann Thorac Surg 1990;50:919 922. [7] Lodi R, Morandi U, Smerieri A et al. La terapia chirurgica del cancro broncopolmonare. Analisi di una casistica selezionata del Centro di Chirurgia Toracica di Modena (328 osservazioni). In: Lodi R, Pisaneschi M, Morandi U, editor. Il cancro broncopolmonare, Tabiano, monography, Parma, 1987:309 322. [8] Miller JI, Hatcher CR. Limited resection of bronchogenic carcinoma in the patients with marked impairment of pulmonary function. Thoracic Surg 1987;44:340 343. [9] Morandi U, Fontana G, Lavini C et al. Il cancro broncopolmonare nella popolazione anziana. Risultati della terapia chirurgica e follow-up a distanza. Abstracts XXI Congresso della Società Italiana di Chirurgia Toracica, 1988. [10] Nagasaki F, Flehinger BJ, Martini N. Complications of surgery in the treatment of carcinoma of the lung. Chest 1982;82:25 29. [11] Naunheim KS, Kesler KA, D Orazio SA, Fiore AC, Judd DR. Lung cancer surgery in the octogenarian. Eur J Cardio-thorac Surg 1994;8:453 456.

U. Morandi et al. / European Journal of Cardio-thoracic Surgery 11 (1997) 432 439 439 [12] Richelme H, Bereder JM, Mouroux J, Bernard JL, Benchimol D. La chirurgie d exérèse du cancer bronchique après 70 ans est-elle legitime? Chirurgie 1990;116:385 394. [13] Thomas P, Sielezneff I, Ragni J, Giudicelli R, Fuentes P. Is lung cancer resection justified in patients aged over 70 years?. Eur J Cardio-thorac Surg 1993;7:246 251. [14] Von Knorring J, Lepantalo M, Lindgren L, Lindfors O. Cardiac arrhythmias and myocardial ischemia after thoracotomy for lung cancer. Ann Thorac Surg 1992;53:642 647. [15] Weiss W. Operative mortality and five years survival rate in patients with bronchogenic carcinoma. Am J Surg 1974;128:799 804. [16] Yellin A, Benfield JR. Surgery for bronchogenic carcinoma in the elderly. Am Rev Respir Dis 1985;131:197 198...