| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL ARTICLE |
Division of Thoracic Surgery Cardiothoracic & Vascular Department
1 Epidemiology and Biostatistics Department Santa Maria Nuova Hospital Reggio Emilia, Italy
Valerio Annessi, MD Tel: +39 0522 296929 Fax: +39 0522 296266 Email: joanna{at}ppukm.ukm.my, Operative Unit of Thoracic Surgery, Cardiothoracic & Vascular Department, Santa Maria Nuova Hospital, Viale Risorgimento 80, 42100, Reggio Emilia, Italy.
ABSTRACT
The lengthening of life expectancy has led to more surgical procedures in elderly patients. The aim of this work was to determine whether age >70 years is a risk factor for pneumonectomy. All cases of pneumonectomy from January 1999 to December 2006 were retrospectively reviewed. The 40 patients aged >70 years were compared with a group of 70 patients aged 40–68 years matched for sex, physical status, respiratory function, side of pneumonectomy, and pathologic stage. Postoperatively, significantly more older patients had atrial fibrillation (24% vs. 5.6%). There was a low incidence of respiratory complications in both groups, and reduced respiratory function did not increase respiratory morbidity. Thirty-day mortality was not significantly different (2.5% in older vs. 1.4% in younger patients), but long-term mortality rates evaluated at December 31, 2007 were 50% for those aged <70 years (35 patients) and 72.5% for the older group. Although age is a risk factor for morbidity and mortality in pneumonectomy, the risk is acceptable.
Key Words: Age Factors • Carcinoma • Non-Small-Cell Lung • Pneumonectomy • Risk Factors • Treatment Outcome
INTRODUCTION
Lung cancer remains a fatal disease for which surgery is the most effective treatment. Only 30% of patients are eligible for surgery; anatomic lobectomy or pneumonectomy with mediastinal lymph node dissection is the accepted standard of treatment. The lengthening of life expectancy has led to more surgical procedures in older patients, and a higher proportion of older patients require pneumonectomy for cancer. Recently, the Lung Cancer Group reported a 5.9% mortality rate after pneumonectomy over the age of 70 years.1 The aim of this study was to compare 2 groups of patients (
70 years and 
69 years) who underwent pneumonectomy to establish whether age is a crucial factor in post-pneumonectomy morbidity and mortality.
PATIENTS AND METHODS
The data of 850 consecutive patients who underwent pulmonary resection for lung cancer and pulmonary metastasis from January 1, 1999 to December 31, 2006 were reviewed, the follow-up time was 482.5 days (range, 9–2,643 days). The procedures included 110 pneumonectomies for lung cancer, 410 lobectomies for lung cancer, and 340 segmental or wedge resections (115 for lung cancer, 225 for pulmonary metastasis). A study group comprising all 40 patients aged 70–79 years was compared with a group of 70 patients aged 40–68 years matched for sex, American Society of Anesthesiologists physical status classification score, respiratory function, side of pneumonectomy, and pathologic stage (Table 1
).2 All patients were evaluated by chest radiography, computed tomography of the chest and brain, and fluorodeoxyglucose positron-emission tomography. Functional assessment was based on blood tests, pulmonary function determined by spirometry and arterial blood gas measurements, postoperative forced expiratory volume in the 1st sec (FEV1), ventilation-perfusion scintigraphy, as well as electrocardiography and echocardiography to detect myocardial disease. In the older group, the presence of unstable cardiac disease was considered an absolute contraindication to pneumonectomy; diagnosis was performed using a shuttle exercise test, O2 consumption evaluation, and coronary angiography when necessary. No pharmacologic prophylaxis for supraventricular dysrhythmias was undertaken. As in previous reports, a predicted post-operative FEV1 < 40% was considered a contraindication to pneumonectomy whatever the age;3–5 all patients excluded had a predicted postoperative FEV1 < 800 mL based on ventilation-perfusion scintigraphy. Diffusing capacity of the lung for carbon monoxide (DLCO), corrected for alveolar volume and expressed as a percent of predicted, has been a standard pulmonary function test in our institution for preoperative evaluation of patients for lung resection since 2003.
|
The following were considered risk factors for morbidity and mortality: hypertension, defined as systolic blood pressure >160 mm Hg or diastolic blood pressure >95 mm Hg (25 patients, 23%); diabetes mellitus, defined as fasting blood sugar >140 mg dL–1 (7 patients, 6%); abnormal electrocardiogram with pathologic Q waves, significant ST depression, left ventricular hypertrophy, complete bundle branch block, or ventricular premature contractions (7 patients, 6%). American Society of Anesthesiologists scoring which divides preoperative anesthetic risk assessment into 4 classes stage of disease, age, type of intervention, and neoadjuvant chemotherapy was carried out in patients with positive mediastinoscopy for N2 stage. The patients’ management was discussed during the weekly multidisciplinary meeting. Early death was defined as any death occurring during hospital stay or in the first 30 days after surgery.
Continuous variables are described as means with 95% confidence intervals (CI). Percentage values are reported for category variables. Associations between risk factors and postoperative events were calculated using the chi-squared test and Fisher’s exact test. A p value 0.05 was chosen as the level of significance in this study. Some results are presented with an odds ratio and the associated 95%CI. CI not containing unity were considered statistically significant. Survival curves were obtained using the Kaplan-Meier method and compared between groups by the log-rank test. All data were processed using SPSS for Windows, release 14 (SPSS, Inc., Chicago, IL, USA).
RESULTS
The mean hospitalization was 9.5 days in patients <70-years old, and 10.5 days in older patients. The histologic cell types in each group are given in Table 1
. There was no difference in cell differentiation between groups. Mediastinoscopy was carried out in 8 (20%) older patients and 14 (20%) younger patients for suspected stage IIIa tumors; however, the pathologic stage was IIIa in 6/8 and 13/14, respectively. These patients underwent neoadjuvant chemotherapy. TNM classifications were similar in both groups. Adjuvant chemotherapy was given to 3 older patients and no younger patient with postoperative pathologic stage IIIa. The mean preoperative FEV1 in the older group was 83.7%; 8 (20%) patients had FEV1 < 70%. In younger patients, the mean FEV1 was 78.9%; 21 (30%) had FEV1 < 70%. The DLCO was measured in 22 older and 33 younger patients: mean DLCO in the older group was 83%; 9/22 (41%) patients had DLCO < 70%. In younger patients, mean DLCO was 71%; 13/33 (39%) had DLCO < 70%. Compared to younger patients, there was a higher incidence of cardiac complications in the older group, in particular, atrial fibrillation (AF) was more frequent (24% vs. 5.6%; odds ratio =5.6, p =0.013). No older patient had a history of dysrhythmias or was receiving antiarrhythmic medication before the operation. Hypertension, diabetes, reduced respiratory function, and sex were not risk factors for postoperative AF. DLCO appeared to be a risk factor for AF in the older group (Table 2). There was no case of myocardial ischemia, myocardial necrosis, or other type of dysrhythmia in the postoperative period in either group. Complications according to respiratory function are given in Table 3. Two patients <70-years old were treated successfully for transitional atelectasia with bronchial aspiration, and one 79-year-old patient presented with pulmonary edema on the 2nd postoperative day and developed adult respiratory distress syndrome that resulted in death on the 7th postoperative day. The low incidence of respiratory morbidity did not allow statistical analysis, but reduced respiratory function (FEV1 < 70%) did not appear to increase respiratory morbidity. There were 5 (7.1%) cases of bronchopleural fistula in younger patients and none in older patients; 2 were undergoing neoadjuvant chemotherapy and radiation therapy, and 3 had a suppurative cancer. There was no correlation between age and bronchopleural fistula. A single case of postoperative hemorrhage occurred in the older age group.
|
|
). The mortality rates evaluated at December 31, 2007 were 50% (35 patients) for those aged <70 years and 72.5% (29 patients) for the older group (p = 0.04). The overall risk of death was 2.6 times greater for the group aged >70 years. The mean survival time for younger patients was 873 days (95%CI: 336–1,409), whereas for older patients, it was 531 days (95%CI: 131–930; log rank =0.03; Figure 1). In older patients, 24/29 (83%) deaths were related to lung cancer, 5 patients died for reasons not ascribable to surgery or lung cancer. In the younger age group, 33/35 (94%) deaths were from lung cancer metastasis. The 5-year survival was 12% in older patients, and 32% in the younger group (Figure 1); the overall survival of all 110 patients after 5 years was 25%. Analysis of mean survival in older patients according to FEV1 < 70% vs.>70% did not show a significant difference (log rank =0.6; Figure 2).
|
|
|
This study found that patients aged >70 years had a significantly higher incidence of AF in the postoperative period. This complication was not related to other risks factors. In some reports, AF appeared to be a prognostic factor for increased mortality, emphasizing the importance of early diagnosis and treatment.6,7 Whereas in other studies, including ours, this finding was not confirmed.8 However, there is general agreement that AF extends hospital stay. This may indirectly contribute to increased mortality. In our experience, postoperative intensive care with immediate treatment of supraventricular arrhythmias prevented further severe complications. We used mainly amiodarone and/or digitalis and these were effective in restoring sinus rhythm in 90% of patients. In cases of failure of this therapy, anticoagulants were given to prevent thromboembolism. The higher incidence of AF in the older group seemed to be unrelated to a history of significant cardiac disease because manifest cardiac risk factors were excluded in the preoperative management. There were more intrapericardial pneumonectomies in these patients (17.5% vs. 11.4%); this procedure can cause postoperative atrial irritation. Furthermore, left atrial dilatation is a hypothetical consequence in patients undergoing pneumonectomy, but this factor should be not related to age. Our findings infer that age is a true risk factor for postoperative AF.
Besides AF, the absence of serious adverse cardiac events in both groups could be explained by the application of American Heart Association/American College of Cardiology guidelines, and a low prevalence of coronary artery disease (10% of surgical candidates). In the older group, we excluded patients with ischemic heart disease and applied appropriate perioperative control of pain, hypovolemia, hypervolemia, and anemia.9 The absence of pneumonia and low incidence of atelectasis in both groups is also relevant. We found higher surgical and respiratory morbidity in younger patients, but we suggest this was related to the small numbers of patients studied.
Continuous epidural analgesia was used in 85% of patients in this study. It has been reported that fewer respiratory complications occur in patients receiving continuous epidural analgesia, compared to those treated with morphine.10 Prevention of atelectasis and secondary infections has been attributed to better preservation of the functional residual volume and efficient mucociliary clearance.10 Undoubtedly, the most severe complication following pneumonectomy is bronchopleural fistula. This complication has been reported to occur in a as few as 0.6% to as many as 30% of patients.7,8 Bronchopleural fistula occurred only in the younger group only (7.1% of this group, 4.5% of all 110 patients studied) and had a significant impact on mortality. We believe that the most important factor leading to this complication was not age nor closure of the bronchial stump, which was uniform in both groups, but the presence of underlying disease, such as chronic obstructive pulmonary disease, lung infection, or neoadjuvant chemotherapy and radiotherapy.
Our 30-day operative mortality rates after pneumonectomy of 2.5% in the older group and 1.4% in younger patients are low compared to 6.2% reported by Ginsberg and colleagues,1 6.9% by Deneffe and colleagues,11 and 6.8% by Kadri and Dussek.12 Perioperative mortality following pneumonectomy varies greatly, ranging from 3.0% to 25% in published series.13 The finding that operative mortality increases with age has been well reported.1,11 However, others have found that the risk of major complications or operative death was not related to age.12,14 Harvey and colleagues15 in a study of 370 patients with non-small-cell lung cancer treated surgically, reported no significantly increased operative mortality until 80 years of age: the rates were 1.4% for patients <70 years, 1.6% for 70–79 years, and 17.6% at 80 years or older. We found that the overall risk of death was 2.6 times greater for the group aged >70 years. The long-term survival for individual cancer stages in the 2 groups was not analyzed because the number of patients in each stage was low. Our data seem to confirm that age is a risk factor for morbidity and mortality after pneumonectomy, but some risks are acceptable or unavoidable, so age is not a contraindication to pneumonectomy.
Olsen and colleagues16 suggested that postoperative FEV1 < 800 mL contraindicates pneumonectomy and lobectomy. On the other hand, several reports recommend that postoperative FEV1 should be expressed as a percentage of predicted value.5 Thus postoperative predicted FEV1>40% would not contraindicate a pneumonectomy, whereas postoperative predicted FEV1 < 40% indicates a high operative risk, and FEV1 < 30% a prohibitive risk. Leo and colleagues17 reported 30-day mortality of 11.4% for patients aged 70 years or older, and 4.3% for those < 70 years, using 30% as the cut-off value of postoperative predicted FEV1. Concerning pulmonary function, we adopted preoperative restrictive criteria for selection of patients for pneumonectomy, the mean preoperative FEV1 was in 83.7% older patients and 78.9% in younger patients; those with postoperative predicted FEV1 < 40% were excluded from surgery. Brunelli and colleagues18 described the usefulness of DLCO in predicting complications in patients with and without abnormal spirometry. They found that almost 45% of those with normal spirometry had DLCO < 80% of predicted. In patients with a normal FEV1, a postoperative predicted DLCO < 40% was an independent and important predictor of postoperative cardiopulmonary complications. Since 2003, we have added DLCO to further improve the selection of patients for pneumonectomy. The numbers are still small, but preliminary data indicate a correlation between postoperative morbidity (particularly AF in the older group) and DLCO < 70% (p =0.04).
In contrast to other reports, we found that old age and low preoperative FEV1 were not predictive of mortality and morbidity.8 These differences may be explained by our cautious selection of surgical candidates and by the application of analysis that identified risk factors unrelated to chronologic age. We excluded patients with ischemic heart disease or calculated postoperative FEV1 < 40% because they have high risks of cardiac complications, acute respiratory failure, and incapacitating dyspnea.19 Alternatively, risk stratification based on DLCO could have been used to detect patients unable to tolerate pneumonectomy.20 In this retrospective study, we observed risks of significant selection bias despite efforts to achieve homogeneous groups of patients. Histology showed more adenocarcinoma in the younger group. None of our older patients underwent adjuvant chemotherapy, despite stage, because the oncologist avoided chemotherapy due to the high risk of complications. The age range in the older age group was narrower than in the younger group. Another clinical feature that differed between groups was the percentage of right pneumonectomies. The effect of patient selection and other confounding factors limit the conclusions of this study. Thus prospective randomized trials are needed to definitively address this issue. We have shown that pneumonectomy is justified in older patients, but only with careful selection to achieve acceptable morbidity and mortality.
REFERENCES
Asian Cardiovasc Thorac Ann 2009;
17:272-277
© 2009 by SAGE Publications
DOI: 10.1177/0218492309104772
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
