| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
© 2003 European Association of Cardio-Thoracic Surgery
Discordance between predicted postoperative forced expiratory volumes in one second (ppoFEV1) calculated before and after resection of bronchogenic carcinomaSection of Thoracic Surgery, Salamanca University Hospital, 37007 Salamanca, Spain
* Corresponding author. Tel./fax: +34-923-291-383 Received September 30, 2002; received in revised form November 25, 2002; accepted December 17, 2002
The aim of this study is to evaluate the concordance between predicted postoperative forced expiratory volumes in 1s (ppoFEV1) calculated on the basis of data known before surgery with ppoFEV1 calculated after completing surgical procedure. We have prospectively studied 66 consecutive patients (55 cases scheduled for lobectomy and 11 for pneumonectomy) operated on for bronchial carcinoma. According to location, 33 tumours were classified as central and 33 as peripheral. In all cases, ppoFEV1 was calculated twice: first (ppoFEV1-A) according to the scheduled surgical procedure; second (ppoFEV1-B) according to the procedure eventually performed. At operation, 43 lobectomies (65.2%) and 23 pneumonectomies (34.8%) were performed. Differences between ppoFEV1 A and B were found in 18 cases (12 central tumours). In three of them (4.5% of 66 cases), ppoFEV1-B was under 40%. Pearson coefficient was 0.85  for the whole series of cases; 0.83 for central and 0.87 for peripheral tumours. On multiple regression analysis, R2 was 0.76 and ppoFEV1-A had the highest influence on the dependent variable. We have found that: (1) there is no perfect correlation between ppoFEV1 calculated with data known before and after surgery; (2) discrepancies are most important in centrally located tumours and (3) in 4.5% of cases, discrepancies could have influenced the preoperative risk estimation.
Key Words: Lung carcinoma; Pulmonary resection; Predicted postoperative pulmonary function; Risk prediction
Lung resection remains the best therapeutic option for localised bronchial carcinoma. Unfortunately, chronic obstructive pulmonary disease is frequently associated to bronchial cancer [1] increasing the risk of lung parenchyma resection. Prediction of postoperative FEV1 has been proved to be a valuable tool to assess surgical risk [2] and its calculation is recommended as a routine test before lung cancer surgery. Patients having a predicted postoperative forced expiratory volumes in 1s (ppoFEV1) under 40% of predicted value are considered high-risk cases [3]. Although individual surgical decisions are taken on the basis of data known before surgical procedure, in some cases, bronchial or vascular extraluminal involvement by primary tumour or unexpected metastatic lymph nodes can change the surgical attitude modifying the estimated operative risk. This investigation was undertaken to quantify changes between ppoFEV1 calculated with data known before surgery and true ppoFEV1 calculated after completing the surgical procedure.
From December 1st 2001 to June 30th 2002, 66 patients underwent lung resection (lobectomy or pneumonectomy) for bronchogenic carcinoma at our unit. Selection criteria for operation consisted in the absence of major co-morbidity refractory to medical therapy, PO2 at rest over 50mmHg, PCO2 under 46mmHg and calculation of ppoFEV1 as absolute values and percent of estimated values for sex, age and height (see below). In 65 cases (one patient was unable to perform spirometry), ppoFEV1 according to the number of non-obstructed pulmonary segments to be resected [2], was calculated twice. The first calculation (ppoFEV1-A) was preoperative and based on the number of non-obstructed pulmonary segments to be resected on fibreoptic bronchoscopy findings. Patients' charts (including computed tomography (CT) scan complete pictures) were preoperatively reviewed by the authors and a surgical indication of lobectomy or pneumonectomy was settled on. Preoperative estimation on the extent of surgery was made on the basis of vascular or bronchial tumour invasion evidenced by CT scan image and endoscopy findings. No patient in whom a ppoFEV1 under 30% of the normal value was calculated at this time was operated on. According to tumour location on CT scan images, cases were divided in two groups: central (situated in the one-third proximal to the pulmonary hylum) and peripheral tumours. Classification was performed by two independent observers (GV and NN) and kappa agreement coefficient was calculated. Surgery was performed in all cases by one of the authors. If possible, a lung parenchyma sparing resection was performed by bronchoplastic techniques. After completion of the procedure, ppoFEV1 was calculated again (ppoFEV1-B) having conclusive information about the number of ventilated and resected pulmonary segments. For the whole series and for each group of central and peripheral tumours, estimated and performed procedures were arranged on a 2x2 table and uncertainty coefficients were calculated. Differences between ppoFEV1-A and ppoFEV1-B were calculated and mean differences between central and peripheral tumours estimated by unpaired t-test. Pearson correlation coefficient was calculated with the data on ppoFEV1-A and ppoFEV1-B (complete series and by groups) to demonstrate that linear correlation existed between both variables. Finally, a multiple linear regression was performed including ppoFEV1-A, location of the tumour and scheduled resection as independent variables and ppoFEV1-B as dependent variable. Data processing was performed by SPSS 10.0 software.
Fifty-five cases were scheduled for lobectomy and 11 for pneumonectomy. At operation, 43 lobectomies (65.1%) and 23 pneumonectomies (34.8%) were performed. In all cases scheduled for lobectomy in whom eventually a pneumonectomy was performed, the decision was taken due to intraoperative findings of bronchial or mediastinal neoplastic involvement. Bronchoplastic techniques were possible in 12.1% of cases (seven sleeve lobectomies and one sleeve right pneumonectomy). Surgical mortality (30 days) was 1.5% (one case died of disseminated malignancy undisclosed before lung resection).
According to location, 33 tumours were classified as central and 33 as peripheral (kappa agreement coefficient: 0.79). Differences between ppoFEV1 A and B were found in 18 cases (12 central and six peripheral) and only in three cases ppoFEV-B was higher than ppoFEV1-A (Table 1). In three cases (4.5% of the whole series), ppoFEV1-B value was under 40%. No mean differences between central and peripheral cases were found on unpaired t-test
A triple scatter plot of ppoFEV1-A and ppoFEV1-B values is presented in Fig. 1. Pearson coefficient was 0.85 for the whole series of cases, 0.83 for central and 0.87 for peripheral tumours. On multiple regression analysis (Table 3), R2 was 0.76 and ppoFEV1-A had the highest influence on the dependent variable.
In spite of significant progress in surgical and anaesthetic techniques, pneumonectomy still represents a high risk procedure and an independent predictor of death after lung resection [4]. In recent publications, operative mortality after pneumonectomy in most series varies between 5 and 12% [5–8], with up to 22% being reported in elderly patients [9]. Several authors recommend to avoid pneumonectomy whenever possible, by using lung-sparing operations [10,11]. We have performed a bronchoplastic procedure in 12.1% of our cases, a rate comparable to other published data (19.2% in 422 cases reported by Lausberg et al. [12] and around 10% in 459 cases reported by Martin-Ucar et al. [11]). Nevertheless, in our experience, the pneumonectomy/sleeve resection rate of 3/1 is much higher than in Lausberg et al. [12] report (1/2 rate). Obviously, surgical decisions regarding operability are taken using data known before surgery. Among the most cited risk factors, ppoFEV1 [3] is usually calculated on the basis of non-obstructed pulmonary segments to be resected [2]. A ppoFEV1 under 40% is considered an indication for more extensive preoperative assessment due to increased risk [3]. Long-term accuracy of ppoFEV1 calculation after pneumonectomy has been discussed in the literature [13] but we are not aware of articles dealing with concordance between pre- and postoperative estimation of remaining pulmonary function. In our series, there is a strong correlation between ppoFEV1-A and B (higher in peripheral tumours) but we have found discrepancies in 18 cases (27.6%) and in three of them (4.5%) differences were large enough to influence individual risk estimation. Since there is no mortality due to cardio-respiratory problems in this series, we can not evaluate whether a more accurate estimation of postoperative lung function would have improved surgical results or not. In this report, we have used the uncertainty coefficient, which is a directional measure to quantify how much the knowledge of an independent variable (in this case, the scheduled procedure) helps to predict another dependent variable (the eventually performed procedure). Uncertainty coefficient ranges from 0 to 1, where 0 means that knowledge of a variable does not help in predicting the dependent variable. As expected, in most of our cases, the changes in ppoFEV1 calculation were due to the necessity of performing a pneumonectomy, mainly in centrally located neoplasms; hence the uncertainty coefficient is lower in this subset of cases. As can be observed in Table 2, although only nine out of 33 cases (27.3%) with central neoplasm were scheduled for pneumonectomy, complete lung resection was eventually performed in more than 50% of the patients. Hence, in any central tumour, a pneumonectomy has to be anticipated and preoperative prediction of tolerance is required. It has been reported that clinical and pathological T classifications are different in around 11% of operated bronchial neoplasms [14] with a tendency to downstaging. Since according to tumour node metastases (TNM) classification, many centrally located neoplasms can be considered as T2 tumours after pneumonectomy, we believe that cited downstaging is not contradictory with our experience. On the other hand, it is well known that in a percentage of peripheral lung neoplasms considered clinical stage I (14% according to Takizawa et al. [15]) lymph node metastasis can intraoperatively be found both at the hylum or the mediastinum and in these cases a larger pulmonary resection can be mandatory. In conclusion, we have shown that: (1) there is no perfect correlation between ppoFEV1 calculated with data known before and after surgery; (2) discrepancies are most important in centrally located tumours and (3) in a small percent of cases, discrepancies could have influenced the preoperative risk estimation. From our data, we can state the hypothesis that the addition of other clinical variables, such as is reported here, to ppoFEV1 estimation could improve the calculation accuracy and individual risk estimation. Undoubtedly, a larger series is needed before conclusive results can be drawn.
ICVTS on-line discussion Author: Dr. Francesco Petrella, University of Bologna "S.Orsola - Malpighi" Hospital, Department of General and Thoracic Surgery, Via Magini, 17, Bologna, Italy Date: 6-Feb-2003 11:44 Message: I read with interest the manuscript of Varela and co-workers dealing with the value of predicted FEV 1 in patients undergoing pulmonary resection for lung cancer. This is a very important topic for thoracic surgeons and I congratulate the Authors on their interesting work. I'd like to know something more about patient's recruitment: 1.)How long after surgery was the second spirometry performed? How many times do you need to avoid interferences on the spirometric values due to post operative pain and chest wall modifications due to thoracotomy? 2.)Was any patient submitted to radio or chemotherapy after surgey? How much does post operative chemo radiotherapy interfere with spirometric values? 3.)Did you also select patients with big lesions (T > 3 cm) or with lesions causing lobar or pulmonary atelectasias? Can you select a patient with functional lobectomy or pneumonectomy? 4.)Did you perform pre operative lung scintigraphy? Do you think it is possible to select a patient with a lobe to be resected receiving more than 30% of the perfusion? I thank the Authors for their time and I compliment them on the interesting work proposed on our platform. Response Author: Dr. Gonzalo Varela, Head Thoracic Surgeon, Salamanca University Hospital, Thoracic Surgery, P. San Vicente 58, Salamanca 37007, Spain Date: 10-Feb-2003 12:45 Message: Answering your questions, we have not performed a second spirometry after lung resection, but re-calculated ppoFEV1% value on the basis of a more accurate knowledge of the amount of functional tissue resected. Of course, we select patients with pulmonary atelectasis and large tumours for surgery. As we all know, large tumours can spread through lung fissures or pulmonary vessels and in these cases the surgeon has to perform a resection larger than previously scheduled. Currently, we do not perform preoperative pulmonary scintigraphy, since it has been demonstrated that ppoFEV1 calculation according to the number of pulmonary segments to be resected is a simple and accurate procedure. In my mind, 30% is quite a lot of perfusion for a single lobe; nevertheless the maximal extent of resection depends on the remaining pulmonary function. Author: Mr. Andrew D. Muir, Registrar in Thoracic Surgery, Norfolk and Norwich University Hospital Trust, Thoracic Surgery, Colney Lane, Colney, Norwich, UK Date: 07-Feb-2003 04:27 Message: These are excellent operative mortality figures. Prediction of post-operative lung function is a fundamental component in the pre-operative workup of patients with bronchogenic carcinoma for lung resection. The authors had a pre-operative plan of pneumonectomy for one sixth of their patients, however over one third eventually had pneumonectomy. This highlights the frequent operative finding of a tumour which has extended further than expected despite all relevant staging tools being utilised. With central tumours, the authors found that one third (11 of 33) of patients had significantly worse predicted lung function following the actual procedure, whilst only one patient had better predicted FEV1. When embarking on lung resection, surgeons should always have in their mind, especially with central tumours, that the surgical procedure will often be more extensive than hoped, and that the patient should be fit for that more extensive resection also. Unfortunately the authors do not report the actual FEV1 figures taken post-operatively. It is well known that lung function takes time to recover following thoracotomy, but it would be of interest to see if there is any correlation between actual FEV1 in the immediate post-operative phase, predicted FEV1, and morbidity and mortality rates. An inability to cough effectively due to poor lung function will increase the likelihood of pneumonia. Secondly, following pulmonary resection, remaining lung segments compensate to some degree. Predictions based on pre-operative data seem to tend to overestimate the post-operative lung function, however, greater resections leave greater potential for residual lung compensation. A comparison of pre-operative predicted FEV1, and the actual FEV1 following a reasonable recovery period may be of greater use in risk prediction. Author: Dr. Christophoros Foroulis, Larissa University Hospital, Cardio-thoracic Surgery, 35 Ioustinianou Street, Larissa, 41223, Greece Date: 25-Feb-2003 09:57 Message: The manuscript of Varela and coworkers covers the very important topic of the preoperative prediction of postoperative FEV1, in patients undergoing major lung resection for bronchogenic carcinoma. The topic is always interesting for the thoracic surgeon. Reading the manuscript, I have the following comments, concerning mainly the methods section: 1) The characterization of a tumor as central or peripheral was based on CT findings. Indeed, tumors located on the lingula or the middle lobe may have the characteristics of central tumors (situated in the one-third proximal to the pulmonary hilum) and could be classified as central tumors, despite the fact that the tumor is located in a segmental or lobar bronchus at bronchoscopy. I believe that bronchoscopic findings are important information for characterization of a lung tumor as central or peripheral. 2) When was postoperative spirometry performed? If bronchoscopy was performed during the 30-days postoperative period, further lung function recovery should be expected, mainly in complicated cases (i.e. postoperative atelectasis). 3) I would like to know also, where discrepancies between preoperative estimated FEV1 and postoperative FEV1 were observed (18 patients / Table 1 of the manuscript): did it concern post-lobectomy or post-pneumonectomy cases? Response Author: Dr. Gonzalo Varela, Head Thoracic Surgeon, Salamanca University Hospital, Thoracic Surgery, P. San Vicente 58, Salamanca 37007, Spain Date: 28-Feb-2003 02:12 Message: As pointed out, endoscopic findings are relevant to classify the tumour as central but not just in middle lobe and lingula tumours. On the other hand, entrabronchial invasion must be assessed by image techniques. We'll keep in mind your remark for further analysis.. The aim of this study was not to evaluate pulmonary function in the follow up; so, we have neither shown data on postoperative spirometries nor bronchoscopies.. Highest ppoFEV1 discrepancies were observed in patients scheduled for lobectomy in whom a pneumonectomy was eventually performed due to extrabronchial invasion of structures. These cases are (on Table 1): 3, 4, 6, 8, 9-11, 12, 14 and 16-18. doi:10.1016/S1569-9293(02)00121-4
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
Top
Abstract
1. Introduction
. 
and 0.55 
for peripheral neoplasms. 
