| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
© 2003 European Association of Cardio-Thoracic Surgery
Invasive staging of superior mediastinum in non-small cell lung cancer patients with specific indications
a Department of Thoracic Surgery, Istanbul Medical School, Istanbul University, Istanbul, Turkey
* Corresponding author. Inönü Cad. Y Received January 8, 2003; received in revised form April 13, 2003; accepted May 30, 2003
This prospective study was done between February 2001 and December 2002 on 84 non-small cell lung cancer patients who were apparently operable. We selectively performed mediastinoscopy to 46 patients (54.76%, group 1) with the following indications: clinical T4 tumor, high operative risk, radiologically enlarged mediastinal lymph nodes, clinical T3 tumors with central location, radiologically identified mediastinal lymph nodes of any size with adeno or large cell carcinoma histology. Other 38 patients (45.23%, group 2) underwent thoracotomy without mediastinoscopy. Sensitivity, specificity, negative predictive value and positive predictive value of the indications were calculated. Cost analysis was done in the 84 patients and the results were compared with alternative mediastinal staging strategies (vs. routine, and vs. selectively to patients with radiologically positive mediastinal lymph nodes) if they had been applied to our population. Group 1 had higher selectivity to differentiate N2 patients (  ). Sensitivity, specificity, negative predictive value and positive predictive value of indications were calculated as: 0.85, 0.54, 0.92 and 0.36, respectively. Our approach was most economical in terms of total cost per patient and money spent unnecessarily per patient. Mediastinal evaluation in operable lung cancer patients should decrease the number of surgical procedures, N2 disease found at thoracotomy and cost.
Key Words: Lung cancer; Mediastinoscopy; Indications
Prognosis and treatment of non-small cell lung cancer (NSCLC) are dependent on tumor stage. It has been shown that absence of mediastinal lymph node metastasis is the key factor for prolonged survival [1,2]. Therefore, many clinical trials have been carried out to improve preoperative mediastinal lymph node staging. Mediastinoscopy is accepted to be the most accurate method of assessing mediastinal lymph node involvement [3,4]. Some surgeons prefer routine surgical mediastinal evaluation and some offer selective evaluation [5,6]. We adopted specific indications for invasive staging of mediastinum [7]. This study was performed to analyze the effectiveness of specific indications in a prospective design.
Eighty-four NSCLC patients were enrolled into this prospective study between February 2001 and December 2002, who were admitted to our department with curative intent. Preoperative staging included chest X-ray, bronchoscopy, abdominal ultrasonography, computerized tomography (CT) of the thorax and upper abdomen, CT of the brain, bone scintigraphy for all patients and 18-fluorodeoxyglucose positron emmission tomography (PET-FDG) in patients with suspected distant metastasis or a second mass in the lung. Patients were divided into two groups. Forty-six patients (group 1) had standard cervical mediastinoscopy before thoracotomy due to our specific indications, whereas the other 38 patients (group 2) had thoracotomy without mediastinoscopy since they did not carry the specific indications. Patients with N2 positive lymph nodes did not undergo thoracotomy. Patients who had mediastinal lymph nodes larger than 2 cm in the shortest axis with computerized thoracic tomography were evaluated with fiberoptic bronchoscopy and transtracheal biopsy and they (two patients) had mediastinoscopy in case of an unconfirmed result. All patients with specific indications (group 1) underwent standard cervical mediastinoscopic evaluation of both paratracheal and subcarinal lymph nodes (five stations 2R, 2L, 4R, 4L and seven according to American Thoracic Society Guidelines [8]). Patient who did not have N2 disease underwent thoracotomy and systematic lymph node sampling. Other patients who did not have specific indications (group 2) underwent thoracotomy and perioperative mediastinal staging. 2.1. Specific indications Specific indications for mediastinoscopy were defined as:
2.2. Perioperative mediastinal staging In both groups routine systematic sampling of the mediastinal lymph nodes was the standard procedure with evaluation of at least four mediastinal stations according to the drainage pattern of the primary tumor lobe. Systematic lymph node sampling procedure was carried out even in the absence of recognizable or palpable mediastinal lymph nodes by the dissection of the mediastinal lymph node region with all fatty tissue including the lymph nodes. If a metastatic lymph node was confirmed with frozen section analysis mediastinal dissection was the procedure of choice.2.3. Evaluation of groups In group 1, 46 patients underwent cervical mediastinoscopy with the previously defined indications (Table 1). Clinical staging of the both groups could be seen in Table 2. Thirteen patients were diagnosed to have N2 disease in the superior mediastinum. None of these patients proceeded with thoracotomy but referred to oncological treatment for either adjuvant or neoadjuvant treatment according to the invasion pattern of the lymph nodes. Four patients from this group had superior mediastinal N2 disease, missed by mediastinoscopy detected by mediastinal sampling at thoracotomy. A total of 17 patients with specific mediastinoscopy indications were found to have superior N2 disease. Two patients from this group were also diagnosed to have inferior mediastinal lymph node positivity, these positivities were ignored since the superior mediastinum is the point of view (Table 2). Pathological staging of the patients could be seen in Table 2.
In group 2, 38 patients underwent thoracotomy without invasive mediastinal staging. Three patients from this group were noticed to have superior mediastinal N2 disease during mediastinal sampling at thoracotomy and also four patients from this group were noticed to have inferior mediastinal N2 disease which were ignored due to aforementioned reason (Table 2). 2.4. Data analysis Forty-six patients selected for mediastinoscopy and patients who did not have mediastinoscopy, were compared with Fisher's exact test in terms of differentiating superior mediastinal disease. The patients who were diagnosed to have N2 positivity with thoracotomy after negative mediastinoscopy (false negative patients) and the patients who were not employed mediastinoscopy but found to have N2 positivity were compared with Fisher's exact test in terms of perioperative N2 positivity.The sensitivity, specificity, negative predictive value and positive predictive value of the indications for selective mediastinoscopy procedure were evaluated. Number of patients saved from an unnecessary mediastinoscopy: total number of patients–(number of patients subjected to mediastinoscopy+number of patients in whom superior mediastinal N2 disease was found at thoracotomy in the second group) (Table 3).
Patients subjected to unnecessary mediastinoscopy: number of patients who underwent mediastinoscopy–(number of N2 disease in mediastinoscopy+number of N2 disease at thoracotomy (false negative patients)) (Table 3). The analyses were done by assuming that our population was subjected to more commonly employed strategies. These commonly employed procedures were accepted to be routine mediastinoscopy and mediastinoscopy in case of a CT diagnosed superior mediastinal disease (Table 3). 2.5. Economical analysis We determined the cost per patient by prospective data collection. Istanbul Medical School prices for a specific procedure in Turkish Liras were converted to Euro in discharged month's exchange rate. The costs were calculated as total cost of thoracic surgical ward stay, biochemical analyzes, routine postoperative radiological evaluation, pathological analyzes including frozen section analyzes, professional fee for operations (surgeon, surgical assistant, anesthetist, operating room), consultations from other departments and disposable equipments used in operation or in ward. Intensive care unit and medicine and blood products were excluded. Total surgical cost was calculated and then determined per patient for only mediastinoscopy, only thoracotomy or both. Unnecessarily spent money was defined either as unnecessary mediastinoscopy or as N2 positive thoracotomy.
The difference in terms of N2 positivity between patients who were selected for mediastinoscopy and patients who were not selected was statistically significant ( ). The difference between patients who were diagnosed to have N2 positivity with thoracotomy after false negative mediastinoscopy and patients who were not employed mediastinoscopy but found to have N2 positivity (false negative indications) was not statistically significant ( ). The sensitivity, the specificity, negative predictive value and positive predictive value of the indications were 0.85, 0.54, 0.92 and 0.36, respectively. In the study group a total of 35 (41.66%) patients were saved from unnecessary mediastinoscopy. Number of patients who were subjected to unnecessary mediastinoscopy was 29 (34.52%) (Table 3). In the study group, surgical cost per patient was calculated as 2372.59 Euro and unnecessarily spent money was calculated as 303.19 Euro (Table 3).
Some large series have demonstrated striking differences between clinical and pathological staging indicating inaccurate clinical staging [6,10]. In order to differentiate between patients eligible for immediate thoracotomy and others eligible for neoadjuvant treatment modalities, accurate staging is essential. Mediastinoscopy combined with CT scanning of the mediastinum is believed to be the best way to evaluate N2 disease. Mediastinoscopy is commonly employed in the following settings: (a) routinely to all apparently operable lung cancer patients, (b) patients having mediastinal lymph nodes larger than 1 cm in the shortest axis and (c) patients with some definite indications [7]. In a study, specific indications were defined as (1) lymph nodes larger than 1 cm in the shortest diameter on CT, (2) presence of centrally located tumor or large peripheral tumor, (3) pulmonary lesion with documented adenocarcinoma histology, (4) marginally acceptable operative candidate for thoracotomy and (5) as part of an experimental protocol [7]. The offered indications were similar to our indications. We proposed the concept of specific indications in order to increase the number of preoperatively diagnosed N2 patients. By this way, we successfully selected 17 superior mediastinal N2 positive patients out of 20. Unfortunately four patients from mediastinoscopy group were missed and they were diagnosed at thoracotomy. These patients had minimal N2 disease and they were expected to have longer survival rates when compared to preoperatively diagnosed N2 patients. In the non-mediastinoscopy group we noticed three patients with superior mediastinal N2 disease. The rate of N2 positive patients without defined indications were almost the same with the false negative mediastinsocopy patients of the group 1 ( ). We can conclude from this data that the indications selectively spared the patients who did not need mediastinoscopy. In our study, thirty-five patients were saved from unnecessary mediastinoscopy compared with routine mediastinoscopy strategy. In routine mediastinoscopy strategy, the aim is to differentiate N2 patients. However, false negative rate of mediastinoscopy is still around 10% and slightly higher for subcarinal nodes sampled [11]. Therefore, it is expected to have false negativity of the procedure. The false negativity rate of our indications and false negativity rate of mediastinoscopy did not differ in our study groups. Proponents of routine mediastinoscopy cite the low complication rate of the procedure and the 10–15% false negative rate of CT scans in demonstrating nodal disease as the main reason to perform routine mediastinoscopy [12]. False negativity of the procedure itself is ignored with this approach. Furthermore the number of unnecessary mediastinoscopies would be expected to double in our study group if routine mediastinoscopy were the procedure of choice. Proponents of CT screening for mediastinal nodal invasion cite the results Canadian Lung Oncology Group's multiinstitutional study which supported the idea of performing mediastinoscopy on patients who had mediastinal lymph nodes larger than 1 cm in size on CT [3]. In our study group, if mediastinoscopy were employed for lymph nodes larger than 1 cm, the number of mediastinoscopies would be halved. The number of preoperatively detected N2 number would also decrease by almost 40%. Thoracotomy without a cure would increase in number in such a case. Researches on mediastinal disease evaluation has been ongoing. PET-FDG scan was noticed to enhance surgical staging with its high negative predictive value and increased sensitivity [13]. The sensitivity and negative predictive value of indications in our study group, are almost the same as this study which could be another alternative in mediastinal staging [13]. Cost effectiveness of alternative mediastinal staging strategies for NSCLC demonstrated that routine mediastinoscopy maximized quality adjusted life expectancy in patients with known NSCLC [14]. Interestingly, in economical analysis the highest cost per patient would be noticed if mediastinoscopy were to be employed to patients with mediastinal lymph nodes larger than 1 cm in CT. Additionally the unnecessarily spent money would be highest if mediastinoscopy were the routine procedure in NSCLC patients. In comparing the costs, groups did not differ too much per patient total cost prices with most apparent difference between our study group and lymph nodes larger than 1 cm group as 120 Euro. Our group differed from the routine mediastinoscopy group as 153 Euro per patient for unnecessary procedure. Chest surgeons should clarify the ideal approach for the evaluation of mediastinal disease as: decreased number of surgical procedures for mediastinal evaluation and decreased number of superior mediastinal disease found at thoracotomy. Economical analysis should also be considered when offering ideal approaches. Therefore, we offer to extend indications for mediastinoscopy to more than only physical appearance of lymph nodes on CT, but not to perform this invasive method to the patients who do not really need it.
Presented at the 10th Annual Meeting of the European Society of Thoracic Surgeons, Istanbul, Turkey, October 26–28, 2002. doi:10.1016/S1569-9293(03)00117-8
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
ld
Top
Abstract
1. Introduction