|
|
||||||||
a Service de Chirurgie Thoracique, Hôpitaux Universitaires de Strasbourg, Strasbourg, France
b Département de Biostatistiques, Hôpitaux Universitaires de Strasbourg, Strasbourg, France
c Département de Pneumologie, Hôpitaux Universitaires de Strasbourg, Strasbourg, France
Accepted for publication July 15, 2009.
* Address correspondence to Dr Massard, Service de Chirurgie Thoracique Hôpitaux Universitaires de Strasbourg, 1 place de l'Hopital, Strasbourg, 67091 France (Email: gilbert.massard{at}chru-strasbourg.fr).
| Abstract |
|---|
|
|
|---|
Methods: The records of 323 pneumonectomy patients were retrospectively reviewed. Multiple demographic data were collected. End points were operative mortality at 30 and at 90 days, major procedurally related complications, and cardiovascular events. Univariate and multivariate statistical analyses were performed.
Results: Smoking habits, chronic obstructive pulmonary disease (COPD) status, induction chemotherapy status, diabetes, and obesity had no statistical influence on short-term outcomes. After right pneumonectomy, 30-day mortality (p = 0.045) and the incidence of bronchopleural fistulas (p = 0.009) were increased. Multivariate analysis for postoperative bronchopleural fistulas discovered that right pneumonectomies are the sole risk factor (p = 0.015). Univariate analysis for postoperative atrial fibrillation showed that male gender, age 70 and older, hypertension, and dyslipidemia are risk factors. Multivariate analysis found no definite risk factor. Patients with coronary artery disease had more postoperative cardiovascular events (p = 0.003). Among patients free of coronary artery disease, COPD led to an increased 90-day mortality rate (p = 0.028).
Conclusions: Patients with right pneumonectomies are at increased risk. Postoperative cardiovascular events are more frequent in coronary artery disease patients. COPD is a risk factor in patients free of coronary disease.
| Introduction |
|---|
|
|
|---|
| Patients and Methods |
|---|
|
|
|---|
Data Collection
Between January 1999 and July 2005, 341 patients underwent a pneumonectomy, and their records were retrospectively reviewed. Requested data were complete in all but 8 patients, who were excluded from the study. The records of the remaining 323 patients were screened for demographic data and postoperative short-term results. Demographic data collected were age, sex, side of operation, body mass index, smoking habits, chronic obstructive pulmonary disease (COPD), induction chemotherapy, coronary artery disease, hypertension, diabetes, dyslipidemia, and peripheral artery disease (PAD).
A patient was considered to have COPD when the forced vital capacity in 1 second (FEV1) was less than 80% of predicted values or when FEV1/forced vital capacity (FVC) was less than 70%, or both. A patient was considered obese when the body mass index was 30 kg/m2 or more.
End points were operative mortality at 30 and at 90 days, major procedurally related complications such as empyema, bronchopleural fistulas (BPFs), and acute respiratory distress syndrome (ARDS), as well as cardiovascular events such as the occurrence of an atrial fibrillation (AF), a myocardial infarction, or a cerebrovascular event. A series of potential risk factors for mortality and morbidity were tested and are listed in Table 1.
|
Complete lymph node resection was performed in all patients with a neoplastic disease. On the right side it included upper mediastinal nodes (levels 2, 4 and 10), subcarinal nodes (levels 7 and 8), and pulmonary ligament (level 9); on the left side we dissected the paraaortic nodes (levels 5 and 6), subaortic nodes (levels 10 and 4), subcarinal nodes, and pulmonary ligament.
The bronchial suture was preferentially made with a stapling device. We routinely covered the bronchial stump after right-sided pneumonectomy; our preferred technique is a pericardial fat pad placed below the superior vena cava. The chest was systematically drained with a single tube connected to a balanced drainage system and kept in place for 48 hours. Patients stayed in the surveillance unit for at least 48 hours before being transferred to the ward.
Statistical Analysis
A biostatistician of the Strasbourg University Hospital Department of Biostatistics controlled the overall statistical analyses and performed the multivariate analyses. Demographic data and results were calculated using SPSS 11.5 software (SPSS Inc, Chicago, IL). The Pearson
2 test or the Fisher exact test (when needed) were used to compare proportions, and the t test was used to compare means. Data were reported as mean ± standard deviation or as proportions. Multiple univariate analyses were performed, and all compared variables with a value of p < 0.1 were included in multivariate analyses, using logistic regression statistics. A value of p < 0.05 was considered as significant.
| Results |
|---|
|
|
|---|
Demographic and comorbidity data are summarized in Table 2. The mean patient age was 60.54 ± 9.99 years and 52 were women (16.1%). COPD was present in 162 patients (50.2%), 66 (20.4%) had undergone induction chemotherapy, 56 (17.3%) were obese, 148 (45.8%) were weaned smokers, 38 (11.8%) had a history of coronary artery disease, 122 (37.8%) were hypertensive, 37 (11.5%) were diabetic, 69 (21.4%) had dyslipidemia, and 33 (10.2%) had PAD.
|
|
|
After right pneumonectomy, 30-day mortality was increased (8.3% vs 3.4% for left; p = 0.045), as well as the incidence of BPFs (7.6% vs 1.7% for left; p = 0.009).
COPD had no influence on early postoperative outcomes, the 90-day mortality rate for COPD patients of 13.6% approached statistical significance compared with the 7.5% rate for patients free of COPD (p = 0.053).
There was no difference in early outcomes of pneumonectomies while testing the influence of induction chemotherapy. The rates of 30-day and 90-day mortality were identical. Even though no statistical difference was noted in occurrence of AF, AF was likely to occur in patients having not received induction chemotherapy (4.5% vs 11.3%; p = 0.073).
Obesity did not influence short-term outcomes.
Patients who were known to have preoperative CAD had more postoperative cardiovascular events than those who had no known preoperative CAD (13.2% vs 1.8%; p = 0.003). The difference in 90-day mortality rates when considering CAD tended to statistical significance (18.4% vs 9.5%; p = 0.086). Otherwise, CAD had no statistical influence on postoperative outcomes.
Smoking status did not have any influence on early postoperative results. Hypertensive patients had more postoperative AF than nonhypertensive ones (14.8% vs 7.0%; p = 0.020). Other outcomes were not influenced by hypertension. Diabetes had no adverse effects on short-term results after pneumonectomies. The only adverse effect of dyslipidemia was the occurrence of AF in the early postoperative period (17.4% vs 7.9%; p = 0.021). Patients with PAD had no worse postoperative outcomes than those without PAD.
Bivariate Analysis
The bivariate analysis associating COPD and CAD showed that the 90-day mortality rate was higher in patients free from CAD if there was a concomitant COPD than in the absence of COPD (13.0% vs 5.8%; p = 0.028). COPD did not influence outcomes in patients with a preexistent CAD. Non-CAD non-COPD patients have the lowest 30-day and 90-day mortality rates (Table 5).
|
|
|
| Comment |
|---|
|
|
|---|
Four authors have reported risk factors in pneumonectomies. Licker and colleagues [10], in a series of 193 pneumonectomies, found an increased 30-day mortality in male gender and an increased cardiac morbidity in patients aged older than 70 years. An extended resection had a negative outcome on respiratory morbidity, whereas epidural analgesia had a protective influence.
Darling and colleagues [11], in a series of 187 pneumonectomies, found that factors associated with increased mortality were hand-sewn bronchial stump closure, a history of smoking, and BPF; the latter was proved by univariate and multivariate analyses. They also found that right pneumonectomies were associated with a higher mortality rate than left, mainly due to the increased risk of BPFs.
Leo and colleagues [12], in a series of 202 pneumonectomies, proved that induction chemotherapy did not influence the mortality rate. Their univariate and multivariate analyses showed higher postoperative complications in patients aged older than 70 years, in patients who had received induction chemotherapy, and in those who had a lower diffusion capacity adjusted for alveolar volume.
López Pujol and colleagues [13], in a series of 266 pneumonectomies with similar demographic data comparing with the present study, found a similar 30-day mortality rate. Their univariate analyses showed the following mortality risk factors: age older than 70 years, diabetes, induction chemotherapy, FEV1 of less than 1800 mL, right pneumonectomy, extended pneumonectomy, absence of bronchial stump coverage, cardiac complications, lung complications, and digestive complications. They did not investigate morbidity risk factors.
A fifth study in 2001 by Joo and colleagues [14] dealt with the same subject in relatively restricted cohort (n = 105) but was not included in the following analysis. This team found that respiratory failure, sepsis, and male sex were predictors of postoperative mortality.
Our study included 323 pneumonectomy patients in an attempt to find the risk factors that influenced the early postoperative outcomes. The overall 30-day mortality rate was 5.6%; it almost doubled on postoperative day 90 to reach 10.5%. These rates are slightly better than those by Licker and colleagues, and are similar to those of Darling, Leo, and López Pujol and their colleagues. The overall complications rate was 39.3%, identical to those of Darling, Leo, and López Pujol, and slightly better than the one of Licker. Tables 8 and 9
summarize mortality and morbidity risk factors described in the literature and in the present study.
|
|
Licker and colleagues agreed with us on the absence of influence of CAD on postoperative mortality, but we disagreed on the influence of CAD on postoperative morbidity. We found increased cardiovascular events in patients with known CAD, but they found no influence of CAD on cardiac and respiratory morbidity (Tables 8 and 9).
The bivariate analysis we performed after combining CAD and COPD, due to the 90-day mortality rates of these two variables that approached statistical significance, showed best results in non-CAD non-COPD patients (Table 5). This might be because these patients had the lowest major comorbidities. It also showed that among non-CAD patients, those without COPD have a statistically better 90-day survival than the other subsets.
Unsurprisingly, like most other reports, univariate analyses found out that right pneumonectomies are associated with increased 30-day mortality, a fact proved by López Pujol and colleagues but which Darling and colleagues failed to prove.
Right pneumonectomy was also associated with an increased incidence of postoperative BPFs, which Darling and colleagues proved (Tables 8 and 9). BPFs were only influenced by right pneumonectomies in the univariate analysis, which was proved by the multivariate analysis (Table 7). BPFs are more likely to occur after right pneumonectomies because right bronchial stumps, in the absence of a protective flap or in the presence of a defective one, swim freely in the right pleural cavity fluid, whereas left bronchial stumps are relatively protected by the pericardium and the esophagus. In addition to the deleterious medical effect of the BPF on the patient, one must also consider the economic aspect of this complication. Patients with BPF had to be hospitalized about 10 days more than those without BPF, with a mean supplemental hospital expenses of about
10,000 per patient. For these reasons, coverage of the bronchial stump, especially in right pneumonectomies, should become routine.
Many authors treated cardiac dysrhythmias (including AF) after pulmonary resections, but few treated cardiac dysrhythmias exclusively after pneumonectomies. Because cardiac dysrhythmia is not the main aim of this study, we will only cite three selected reports
According to univariate analyses in our series, postoperative AF predominated in men, in patients aged 70 years and older, in hypertensive patients, and in dyslipidemic patients. Multivariate analysis did not retain any of these risk factors as an independent factor, even though male gender approached statistical significance (Table 6). As we already have noted, a multitude of risk factors have been described, but the data are not unanimous. Therefore, no exact risk factor or a single mechanism of physiopathology for AF can be retained.
AF is a relatively benign complication, is easy to treat, and is almost always reversible. We found no adverse effects of AF in our patients who had this complication. AF did not result in prolonged postoperative hospitalization. Nevertheless, this does not mean that we should not try to prevent AF. Many prophylactic protocols have been described, based on β-blockers such as metoprolol [18], calcium channel blockers such as diltiazem [19, 20], and class III antiarrhythmic drugs such as amiodarone [21], with encouraging results. Amar and colleagues [22] showed that the use of preoperative statins statistically reduced the incidence of postoperative AF. This finding might go in parallel with our univariate analysis that found that the risk of postoperative AF is increased in patients with known history of dyslipidemia. This could be a leading hypothesis for a future prospective study to clarify the physiopathology of postoperative AF occurrence after pneumonectomies.
Of note in our series, AF was less likely to occur in patients who had received induction chemotherapy than those who had not (4.5% vs 11.3% respectively). One might think that chemotherapy regimens might have some protective effect from AF, probably by reducing inflammatory status. In our opinion, AF should be systematically prevented before scheduled pneumonectomies, especially in men.
This original study is retrospective and has the limitations of all retrospective studies. Even though patients were referred to our department from different medical facilities, they were operated on in a single center. We think that the actual study may serve as a basis for multiple prospective investigations trying to prevent postoperative complications of pneumonectomy by a preoperative correction of modifiable risk factors.
In conclusion, the exhaustive statistical study on our series of patients showed that COPD, induction chemotherapy, obesity, smoking habits, and diabetes did not influence the short-term mortality or morbidity after pneumonectomies; right pneumonectomies were associated with higher 30-day mortality rate, and higher incidence of BPF; right pneumonectomies present an independent risk factors for the occurrence of BPF; non-CAD non-COPD patients have the best postoperative survival; and no independent factor for the occurrence of postoperative AF was identified.
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
A. W. Kim, D. J. Boffa, Z. Wang, and F. C. Detterbeck An analysis, systematic review, and meta-analysis of the perioperative mortality after neoadjuvant therapy and pneumonectomy for non-small cell lung cancer J. Thorac. Cardiovasc. Surg., January 1, 2012; 143(1): 55 - 63. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. G. Fernandez, S. D. Force, A. Pickens, P. D. Kilgo, T. Luu, and D. L. Miller Impact of Laterality on Early and Late Survival After Pneumonectomy Ann. Thorac. Surg., July 1, 2011; 92(1): 244 - 249. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Paul, F. Mirza, J. L. Port, P. C. Lee, B. M. Stiles, A. L. Kansler, and N. K. Altorki Survival of patients with clinical stage IIIA non-small cell lung cancer after induction therapy: Age, mediastinal downstaging, and extent of pulmonary resection as independent predictors J. Thorac. Cardiovasc. Surg., January 1, 2011; 141(1): 48 - 58. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Shapiro, S. J. Swanson, C. D. Wright, C. Chin, S. Sheng, J. Wisnivesky, and T. S. Weiser Predictors of Major Morbidity and Mortality After Pneumonectomy Utilizing The Society for Thoracic Surgeons General Thoracic Surgery Database Ann. Thorac. Surg., September 1, 2010; 90(3): 927 - 935. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. P. van der Kaaij and J. Kluin Invited Commentary Ann. Thorac. Surg., December 1, 2009; 88(6): 1743 - 1744. [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 |