| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL ARTICLE |
General Thoracic Surgery, Unit Assaf Harofeh Medical Center, Zerifin, Israel
Ilan Bar, MD, Tel: +972 8 9779822, Fax: +972 8 9778149, Email: fredricag{at}asaf.health.gov.il, General Thoracic Surgery Unit, Assaf Harofeh Medical Center, Zerifin 70300, Israel.
ABSTRACT
Accurate preoperative staging of the mediastinum is important in the treatment of non-small-cell lung cancer. Enlarged mediastinal lymph nodes on chest computed tomography are positive for malignancy on mediastinoscopy in only half of these patients. After negative mediastinoscopy, some positive nodes are found at thoracotomy. The aim of this study was to attempt to remove all lymph nodes accessible by cervical mediastinoscopic lymphadenectomy and reevaluate the same mediastinal stations at thoracotomy for missed lymph nodes. Between 1999 and 2003, 30 patients with operable non-small-cell lung cancer and enlarged mediastinal lymph nodes (>1 cm in diameter on computed tomography) that were negative on cervical mediastinoscopy underwent pulmonary resection with complete lymph node dissection. The total number of lymph nodes dissected in these 30 patients was 329 (143 at mediastinoscopy and 186 at thoracotomy); the mean numbers of nodes dissected were 4.8 at mediastinoscopy and 6.2 at thoracotomy. Ten (6.5%) residual lymph nodes were detected at thoracotomy in mediastinal stations R4, L4, and 7. The low number of missed lymph nodes demonstrates the accuracy of the technique of cervical mediastinoscopic lymphadenectomy.
Key Words: Dissection • Lung Neoplasms • Lymph Nodes • Lymph Node Excision • Neoplasm Staging
INTRODUCTION
Survival of non-small-cell lung cancer (NSCLC) patients depends on the stage of the disease. Involvement of mediastinal lymph nodes (MLN) is a negative prognostic factor for long-term survival. Enlarged MLN (>1.0–1.5 cm in diameter) on computed tomography (CT) are positive in only half of these patients. Cervical mediastinoscopy plays an important role in sampling these nodes. Systematic MLN dissection (lymphadenectomy) is one of the staging procedures used in NSCLC patients; those with positive results are referred for induction chemotherapy, while those with negative results can undergo curative resection. We perform cervical mediastinoscopic lymphadenectomy to dissect most of the lymphatic tissue, instead of taking biopsies. In our opinion, this method may be helpful in the staging of NSCLC, together with other invasive methods. Pulmonary resection with complete lymph node dissection (CLND) in negative cases results in more accurate staging and aids in directing the next stage of therapy. A comparison of the number of lymph nodes removed by mediastinoscopic lymphadenectomy and by thoracotomy (with CLND) from the stations assessed by cervical mediastinoscopy (2, 4, and 7) was the aim of this study.
PATIENTS AND METHODS
Institutional review board approval was obtained for the study, and the board waived informed consent because this was a retrospective study. Thirty NSCLC patients (21 male and 9 female; median age, 67.3 years; age range, 46 to 83 years) were enrolled. They had all undergone cervical mediastinoscopic lymphadenectomy between 1998 and 2003, which was negative for malignancy, followed by post-mediastinoscopy thoracotomy with lung resection and CLND in the General Thoracic Surgery Unit at Assaf Harofeh Medical Center. All patients were asymptomatic for pulmonary and extrapulmonary disease, had solitary pulmonary lesions, only mediastinal lymphadenopathy on CT scan (1.0–2.0 cm in diameter) without lymph node enlargement in the aortopulmonary window, and negative lymph nodes harvested by cervical mediastinoscopic lymphadenectomy. Preoperative chest radiography, bronchoscopy, and CT of the thorax were undertaken in all cases. Brain CT and bone scintigraphy were performed in patients due to undergo pneumonectomy. Positron-emission tomography was not used in our institution during the period of this study. The mediastinum was scanned at 5-mm intervals with a Siemens scanner, and surgery was delayed no more than 30 days from the CT scan.
All pathologic and operative results were examined, and the tumors were classified and staged according to the TNM system and International Union Against Cancer classification. The Naruke map was used to indicate lymph node location during mediastinoscopy and thoracotomy. Lymph nodes at stations 1–9 were considered to be mediastinal nodes (N2 disease), and nodes at stations 10–13 were regarded as hilar lymph nodes (N1 disease). Lymph nodes that were completely dissected by cervical mediastinoscopic lymphadenectomy (stations 2 and 4 bilaterally, and station 7) were pathologically assessed. At thoracotomy, stations 2, 4, and 7 were reassessed for residual lymph nodes, and dissection of lymph nodes at other stations unsuitable for mediastinoscopic lymphadenectomy was also carried out (stations 8–13 ipsilaterally, and 5 and 6 from the left). Lymph nodes dissected at mediastinoscopy underwent frozen section study and were sent for final examination, while those dissected at thoracotomy were fixed in formalin and sent for routine histopathology. Frozen sections obtained by the standard procedure of 4–5 µm sections embedded in Tissue-Tek optimal cutting temperature embedding medium enabled optimal assessment of MLN by highly experienced pathologists. The 2 procedures (mediastinoscopy and thoracotomy) were carried out in a single session. The following clinical and pathologic parameters were studied in each case: age, sex, side and type of surgery, histological type, pleural and vascular invasion, stage, tumor size (T status), number of stations, and number of lymph nodes dissected (Tables 1
and 2).
|
|
RESULTS
All 30 NSCLC patients included in the study were staged before mediastinoscopy as T1–T2 NM0 tumors, based on clinical and radiological findings (CT of the chest, brain and bone scintigrams). Negative pathologic results of lymph nodes assessed by mediastinoscopic lymphadenectomy comprised 9 patients with T1N0 disease (30%) and 21 with T2N0 disease (70%). The pathologic nature, number, and station of lymph nodes dissected by mediastinoscopy and thoracotomy can be seen in Table 2
. All 143 lymph nodes removed by mediastinoscopic lymphadenectomy were negative for malignancy in frozen sections, and 1 lymph node was changed to positive on final pathologic examination (which upstaged one patient to N2 disease postoperatively). Based on pathologic characteristics of the tumor (T status) and lymph nodes (N status) after thoracotomy, the T status of 3 patients changed (one to T3 due to parietal pleural invasion, and 2 to T2 due to visceral pleural invasion), and the N status changed in 5 patients (4 were upstaged to N1 disease due to positive lymph nodes at stations R11, R12, L10, and L12, and one was upstaged to N2 disease due to positive lymph nodes at station L5; Tables 1 & 2). Ten of the 186 lymph nodes dissected during thoracotomy were positive. According to the final pathological examination, the stages of patients after thoracotomy were: T1N0M0 in 5; T1N1 M0 in 1; T2N0M0 in 19; T2N1M0 in 2; T2N2M0 in 2; and T3N1M0 in 1 (T1 tumors were 10–30 mm, with a mean of 25.2 mm; T2 tumors were 32–90 mm, with a mean of 52.3 mm). As a result of both surgical procedures, only 2 (6.6%) patients were upstaged to N2 disease, 1 after each procedure: 1 with the primary tumor in the right upper lobe, due to a frozen section misinterpretation; and the other with the primary tumor in the left upper lobe, due to a positive lymph node at station L5 not assessed by mediastinoscopic lymphadenectomy). The total number of lymph nodes dissected was 329 (143 by mediastinoscopic lymphadenectomy and 186 by CLND), the mean number of lymph nodes dissected in each patient was 4.8 by mediastinoscopic lymphadenectomy and 6.2 by CLND (mean number of lymph nodes dissected by both procedures, 10.9). One hundred and fifty-three MLN were dissected by mediastinoscopy and thoracotomy from stations R4, L4 and 7 (143 during mediastinoscopic lymphadenectomy and 10 during CLND at thoracotomy). Ten (6.5%) residual lymph nodes from stations 2, 4, and 7 were not detected during mediastinoscopy (2 from R4, 1 from L4 and 7 from station 7). Thus, the dissection of lymph nodes at the stations assessed by mediastinoscopic lymphadenectomy was almost complete for stations 2 and 4 (100% for R2 and L2; 96.4 % for R4 and 97.8% for L4) but less (69.6%) for station 7. Ten (5.4%) lymph nodes dissected during CLND at thoracotomy were metastatic and restaged 4 patients from N0 to N1 disease (stations R11, R12, L10, L12) and one to N2 disease (station L5). The final surgical-pathologic TMN stage of the 30 patients after pulmonary resection and CLND can be seen in Table 3; 10 (33%) patients were upstaged. There were no complications during or after mediastinoscopic lymphadenectomy, and the mean period of hospitalization was 10.4 days for all 30 patients (8.5 days after pneumonectomy and 12.1 days after lobectomy). Post-pulmonary resection complications were atrial fibrillation in 6 (20%) cases, pneumonia in 3 (10%), lobar atelectasis in 2 (6.7%), and leak after lobectomy in 2 (6.7%). Of 21 patients who underwent lobectomy, 11 were alive after mean follow-up of 7.1 years, and 10 died at a mean of 3.2 years. Of 9 patients who underwent pneumonectomy, 1 was alive after 8.5 years, and 8 died at a mean of 3.1 years.
|
Surgical resection remains the cornerstone of NSCLC patient management. Among other factors, the prognosis depends on metastases to the lymph nodes, especially N2 and N3. Preoperative staging resulting in the detection of positive N2 lymph nodes using mediastinoscopy or other techniques renders surgical resection unnecessary, and these patients should be referred for neoadjuvant therapy, and thereafter reconsidered for surgical treatment.
The presence of small MLN (<1.0 cm in diameter) on CT is generally accepted as proof of the absence of MLN involvement by metastasis, making it unnecessary for the surgeon to perform preoperative mediastinal surgical exploration.1 However, even with the new generation of CT scanners, the false negative rate in evaluating small MLN is as high as 20%.1 Enlarged MLN (>1.0–1.5 cm in diameter) may be benign in up to 40% of cases, while normal looking ones may be malignant in almost 25% of cases.2 Mediastinal lymph node removal, whether defined by complete dissection or sampling, has been well accepted as part of the standard staging procedure in lung cancer surgery.3 Dissection should also be carried out because of the frequent incidence of skip metastases (hidden N2 disease).4 Prenzel and colleagues5 found that 33% of patients were upstaged postoperatively because of skip metastases. Benoit and colleagues6 also demonstrated approximately 30% skip metastasis in MLN in 142 N2 resected cases. Other studies have revealed that detection of micrometastatic tumors in node-negative lung cancer patients provides a more accurate assessment of tumor staging and has prognostic implications.7 Systematic MLN dissection at thoracotomy is also an accurate intraoperative staging procedure, and it may increase the probability of micrometastatic lymph node clearance when nonsystematic lymph node sampling is thought to be inadequate for accurate staging.8 Wu and colleagues3 suggested that removal of less than 15 lymph nodes during thoracotomy could be interpreted as inadequate for proper staging, and could adversely affect survival. They demonstrated a mean of 11 ± 7.8 (range, 0–45) lymph nodes removed in 321 patients with 1 A/1B NSCLC; <15 MLN in 244 (74%) and>15 in 70 (24%). Naruke and colleagues9 recommended performing sentinel lymph node sampling first; if the nodes prove to be negative, CLND might be omitted, but if they are positive, CLND is required for curative resection.
As an assessment procedure, MLN dissection can be performed using various methods. Mediastinoscopy and its variants are widely used (cervical mediastinoscopy, extended cervical mediastinoscopy, anterior or parasternal mediastinoscopy). Cervical mediastinoscopy can assess lymph nodes at levels 2 and 4 (from left and right) and 7, and this procedure is still considered the gold standard for staging NSCLC patients, with high sensitivity, specificity, and accuracy in diagnosing MLN involvement.10 Staging procedures such as endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) and endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) are also suitable for enlarged MLN. Several case series using EBUS-TBNA for MLN staging in lung cancer patients have reported a high yield, ranging from 89% to 98% (mean, 94.5%).11 Consideration of EUS-FNA in the staging of lung cancer is also acceptable, but the success of this procedure depends on the size of the lymph nodes (sensitivity and negative predictive value in detecting malignancy in enlarged MLN: 96% and 67%, respectively), or on the location of the primary tumor and the extent and location of mediastinal disease (EUS-FNA sensitivity: 91.7%, 78.1%, and 43.8% for bulky disease, enlarged MLN, and normal lymph nodes on CT, respectively).12,13 Moreover, a recent report showed 100% sensitivity and specificity when EUS-FNA and EBUS-TBNA were used together for staging of the mediastinum in lung cancer patients.14
Video-assisted mediastinal lymphadenectomy, a method for systematic MLN dissection, is also widely used as a staging procedure in NSCLC. Leschber and colleagues15 compared the results of video-assisted mediastinoscopic lymphadenectomy with final lymph node staging at subsequent thoracotomy in 25 patients (8.6 lymph nodes per patient were removed; range, 4–14). Lymph node dissection at stations 2R, 2L, 4R, 7, and 4L was achieved in 96%, 28%, 98%, 100%, and 100%, respectively. While lymphadenectomy could be performed without problems in the right paratracheal area, dissection remained difficult in the subcarinal region due to anatomical reasons.15 Using the bimanual dissection technique combined with video-assisted mediastinal lymphadenectomy, Hürtgen and colleagues16 harvested significantly more nodes (mean 20.7, range 5–60) by video-assisted mediastinal lymphadenectomy compared to open lymphadenectomy (mean, 14.3) in 40 NSCLC patients. A more recent study by Leschber and colleagues17 also demonstrated that video mediastinoscopy routinely yielded more lymph nodes, with a tendency for improved accuracy (87.9%) and negative predictive value (83%) compared to cervical mediastinoscopy (accuracy 83.8%, negative predictive value 81%).
Transcervical extended mediastinal lymphadenopathy, a more aggressive technique of MLN staging described by Zielinski,18 demonstrated 94.1% sensitivity, 100% specificity, 98% accuracy, 97.2% negative predictive value, and 100% positive predictive value in the discovery of N2–3 nodes. However, it is important to define the need for different invasive staging procedures and to select the best for each individual patient. Detterbeck and colleagues19 gave some recommendations on this issue. In our study, we evaluated the role of mediastinal lymphadenopathy as a technique of staging, and its potential reduction of undetected N2 disease. We found that after cervical mediastinal lymphadenopathy the percentage of missed lymph nodes found at thoracotomy was low for stations 2 and 4, and higher for station 7, possibly due to difficulties in completely dissecting MLN at station 7 because of its proximity to the right pulmonary artery. Witte and colleagues20 also noted that remnant lymph nodes were more often located in the subcarinal station than the tracheobronchial, paratracheal, and pretracheal regions. Our finding that only 6.5% of MLN were missed (stations R4, L4 and 7) demonstrates the usefulness of cervical mediastinal lymphadenopathy in reducing positive lymph node metastasis at thoracotomy.
It was concluded that the accuracy of mediastinal lymphadenopathy is high because it enables the removal of nearly all MLN at stations 2, 4, and 7. Operative techniques for complete dissection of nodes at all stations, and especially station 7, must be improved in the future. Better visualization of the subcarinal station, which is the furthest station from the surgeon using a conventional mediastinoscope, may be achieved with video mediastinoscopic techniques. To avoid false negative results on frozen section, it is now our policy to wait for formal pathologic reports before proceeding to thoracotomy. Immediately after the introduction of positron-emission tomography in our country, we began using this effective staging method widely, before performing EBUS-TBNA, EUS-FNA or more aggressive procedures (mediastinal lymphadenopathy or video-assisted thoracoscopic surgery for dissection of MLN from anterior stations 5 and 6). Despite the limitations of this retrospective descriptive study, we recommend performing mediastinal lymphadenopathy as an additional staging procedure in NSCLC patients when other invasive or noninvasive staging methods are insufficiently informative.
REFERENCES
Asian Cardiovasc Thorac Ann 2009;
17:357-361
© 2009 by SAGE Publications
DOI: 10.1177/0218492309338097
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
