Japanese Journal of Clinical Oncology Advance Access originally published online on January 25, 2006
Japanese Journal of Clinical Oncology 2006 36(2):76-79; doi:10.1093/jjco/hyi226
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2006 Foundation for Promotion of Cancer Research
Feasibility of Imprint Cytology for Evaluation of Mediastinal Lymph Nodes in Lung Cancer
1 Department of Thoracic Surgery, Heybeliada Chest Disease and Thoracic Surgery Centre and 2 Department of Pathology, Heybeliada Chest Disease and Thoracic Surgery Centre, Istanbul, Turkey
For reprints and all correspondence: Cagatay Tezel, Yazmaci Tahir sok 51/9 A Blok Bostanci, 34744 Istanbul, Turkey. E-mail: mdcagatay{at}hotmail.com
Received May 13, 2005; accepted December 5, 2005
 |
Abstract |
|---|
 TOP Abstract INTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Background: Intraoperative evaluation of mediastinal lymph nodes is a necessary step which helps us to decide whether or not to continue the operation of lung cancer. Imprint cytology (IC) can be used as an alternative method in staging. It is a more rapid and simpler procedure than frozen section (FS) analysis. Therefore, we compared the diagnostic accuracy of IC with permanent section on 1050 mediastinal lymph nodes.
Methods: A total of 255 non-small cell lung cancer patients who underwent surgical procedure between January 1995 and April 2004 were included. There were 236 males and 19 females with a mean age of 54.2 years (range 26–79 years). In order to obtain lymph node samples mediastinoscopy was performed in 232 (91%), anterior mediastinotomy in 50 (20%) and video-assisted thoracoscopic surgery in 16 (6.3%) patients. During final pathological diagnosis, both imprint and permanent section slides were compared.
Results: There were five false-positive and eight false-negative results. The sensitivity, specificity and the predictive values for positive and negative results were 93.1, 99.5, 95.6 and 99.1%, respectively. The overall efficiency was 98.8%.
Conclusions: The diagnostic IC is an accurate, reliable, simple and less time-consuming method for evaluation of mediastinal lymph nodes in lung cancer, compared with FS method.
Key Words: frozen section • lung cancer pathology and biology • imprint cytology • staging
|
INTRODUCTION |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
It is well known that at the time of the initial presentation of patients with non-small cell lung cancer (NSCLC), N-status is very important in obtaining an accurate staging determination and avoiding inappropriate surgical procedures (1–3). Preoperative evaluation of mediastinal lymph nodes via invasive staging methods, especially with mediastinoscopy, is commonly used in many thoracic surgery clinics. Some thoracic surgeons believe that single-stage operation combined with mediastinoscopy should be performed to reduce total operation and anesthesia time along with the operative risks and the hospital costs (4,5). In addition to cervical or extended mediastinoscopy, other invasive staging procedures such as video-assisted thoracoscopic surgery (VATS) or mediastinotomy can be combined with thoracotomy (6–8). In single-stage operations, intraoperative evaluation of related lymph nodes [generally by frozen section (FS)] is essential (4,5,9).
Accurate staging requires multiple biopsies that should be taken from all available lymph node stations. For the reason that each specimen must be separately embedded, sectioned and then stained, FS analysis is a time-consuming and labor-intensive procedure given the number of lymph node stations with the number of samples from each node (5,9–11). Moreover, the distance of FS laboratories to the operating room and technical inadequacy, such as limited cryostats, may contribute to substantial delays.
Imprint cytology (IC) can be an alternative method because it is a more rapid and simpler procedure than FS, which allows more extensive sampling of multiple nodal stations. In this prospective study, we compared the diagnostic efficacy of IC with permanent section via 1050 mediastinal lymph nodes obtained by invasive staging methods from 255 patients with NSCLC.
|
PATIENTS AND METHODS |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Data were prospectively collected from 255 patients with NSCLC who had undergone surgical procedures at Heybeliada Chest Surgery Center between January 1995 and April 2004. There were 236 men and 19 women with a mean age of 54.2 years (range 26–79 years). Out of 255 patients, cervical mediastinoscopy was performed for 232 (91%) as the initial step; other staging methods like anterior mediastinotomy (n = 50, 20%) or VATS (n = 16, 7%) were performed if a necessity has occurred with regard to tumor location and lymph node involvement on computed tomography (Table 1). When a suspicion of vascular invasion or involvement of station 5 and 6 lymph nodes in the left upper lobe tumor occurred, the initial step was replaced with anterior mediastinotomy. VATS was performed to identify the posterior subcarinal, paraesophageal and inferior pulmonary lymph node involvements in both sides.
|
One thousand and fifty lymph node samples were taken from 255 patients and were immediately sent to the pathology department. IC technique was used for all 1050 samples. While N2 disease was reported by IC, the next surgical procedure was cancelled, if the result was turned out to be ‘negative’, the next staging procedure or thoracotomy was performed.
IMPRINT CYTOLOGY TECHNIQUE
All lymph node fragments from the same station were put together and several touches were made on a glass slide. Usually one or two slides were prepared from each station. Slides were immediately fixed with 96% ethanol, and then were stained with standard Papanicolau (PAP) method. The average time for preparing the slide in our laboratory was 
2–2.5 min. The most important part of the process is rapid touch of the fragments on a glass slide and following immediate fixation to prevent artifacts related to drying.
PERMANENT-SECTION TECHNIQUE
After invasive staging procedure was completed, all tissues that had been used for imprint preparation were placed into 10% formalin, and subjected to a routine paraffin section staining with hematoxylin–eosin.
Both imprint and permanent section slides were reviewed and diagnoses were compared. Sensitivity, specificity, predictive value of a positive result (PVPR), predictive value of a negative result (PVNR) and efficiency of the test were calculated using standard methods.
|
RESULTS |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Imprint slides, which were prepared from both benign and metastatic lymph nodes, manifested a satisfactory appearance with cytologic detail. It was often possible to determine tumor cell type by IC for squamous cell carcinoma, adenocarcinoma or small cell carcinoma (Figs 1
–3).
|
|
|
There were 113 positive results including 108 true-positive and 5 false-positive lymph nodes. Among 937 negative results, 929 were true-negative and 8 were false-negative (Table 2). Four false-positive results were due to the difficulty of discriminating histiocyte clusters from tumor cells. Desquamated neoplastic cells in the staining solution remaining from a previous case, constituted the other reason for a positive result. According to the false-negative results, it was not possible to identify the metastasis due to the size of the focus (<1 mm diameter) in four of the slides. These tumor cells were revealed as follows: one adenocarcinoma, one large cell carcinoma and two bronchoalveolar cell carcinomas. Technical defaults due to the insufficient adhesion of tumor cells to the slides caused false-negativity in two slides. Finally, two slides of combined small cell carcinoma with variant of squamous cell carcinoma were the reason for the other false-negative results.
|
The sensitivity, specificity, PVPR, PVNR and overall efficiency of a total of 1050 lymph node samples were outlined in Table 3.
|
Owing to detection of N2 disease by IC in 37 (14.5%) patients, the surgical procedures were stopped. Among them, five patients have underwent resection after neoadjuvant therapy. At the time of the invasive staging procedure, unnecessary thoracotomies were prevented in 20 patients (8%) who were diagnosed with T4 disease. In the remaining 18 (7.2%) patients, the decision of inoperability or inresectability due to direct vascular or mediastinal invasion of the tumor was standing out as a possibility during the thoracotomy.
|
DISCUSSION |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Intraoperative histopathologic evaluation is frequently needed in lung cancer surgery, not only for detection of lymph node involvement but also for lesions or metastases previously seemed to be indeterminate. FS analysis is still the most useful technique for this purpose in many surgical oncology clinics. However, it is a labor intensive and time-consuming method. More rapid, simpler and efficient cytopathological methods should be desirable when multiple samples are submitted for analysis within a short period of time.
Thus, intraoperative cytology using fine-needle aspiration biopsy or touch IC have been preferred for many years in histopathologic diagnosis of lesions from different organs (11–14).
Lee reported the results of his study in which he used IC for tumor diagnosis on 522 cases in 1982. He found that the total diagnostic accuracy was 92.9% for benign and 97.2% for malignant tumors. Overall, the false-positive and false-negative rates were 0.8 and 4.8%, respectively (12). Tamiolakis et al. (15) achieved better results in his limited number of cases revealing rates of 99.1% sensitivity and 100% specificity.
Sometimes sinus histiocytosis is highly predominant in lymph nodes. These cells resemble the tumor cells and cause false-negative results. However, false-positive results may occur due to small size of the metastatic focus. Insufficient adhesion of the tumor cells to the slides in two of our cases caused false-negativity. Finally, two slides of small cell carcinoma, a variant of squamous cell, were the reason of the remaining false-positive results. Sometimes it proves to be difficult to distinguish malignant cells among lymphocytes.
A similar report was revealed by Ghandur-Mnaymneh and Paz (9), in which the accuracy was 99% for IC and 98.7% for FS on 300 lymph nodes obtained from metastatic carcinomas. They recommended that IC should be used instead of the FS in cancer staging procedures whereas many lymph nodes are submitted for immediate intraoperative examination for the determination of the extent of surgery.
Mair et al. (10) studied on 206 surgical specimens and compared the diagnostic accuracy and quality of FS and IC. The quality of cytologic preparations was significantly superior to the FS (P = 0.0001). There was no significant difference between the accuracy of the FS diagnoses compared with that achieved by cytologic examination (P = 0.35). Thus, they concluded that, IC manifested a superior quality of the cytologic preparation compared with FS along with more advantages such as major time saving without any loss of accuracy in diagnosis. As a preliminary diagnostic method, IC can be used. The diagnostic accuracy of IC in diagnosis of cancer was evaluated. When the core needle biopsy specimen was taken as a gold standard, the positive predictive value of the IC technique results was 100% in 90 specimens (14).
Although increasing number of studies have been reported on the use of IC in lung cancer including touch cytology from surgical margin of excised malignant tumor, chest wall, visceral pleura or bronchial stump; unfortunately, there are limited reports regarding the validity of IC in staging of lung cancer (16–18).
Invasive staging procedures like mediastinoscopy, VATS or mediastinotomy have been commonly used in many thoracic surgery clinics and FS analysis is generally preferred for the evaluation of mediastinal lymph nodes. However, the limited use of cytologic techniques for intraoperative evaluation of mediastinal lymph nodes may be explained with the option of separated operations instead of a single operation in chest surgery. Thus ‘one stage operation’, combining staging procedure and thoracotomy, should be used for lung cancer surgery to reduce the total operation and anesthesia time, operative risks, hospital stay and hospital costs.
Clarke et al. (5) reported one of the most encouraging studies about IC for staging of lung cancer in 1994. Either IC (Group 1, 192 samples from 38 patients) or FS analysis (Group 2, 168 samples from 36 patients) randomly evaluated in 74 patients who had undergone mediastinoscopy at the time of the operation. In both groups, there was only one false-negative without false-positive results. The overall efficiency, sensitivity and specificity were 99.2, 96.6 and 100%, respectively, for IC. The same results were 99.2, 94.1 and 100% for FS analysis, respectively. This study confirmed the usefulness of IC over FS analysis in staging of lung cancer.
Okuba et al. (18) recently published one of the few reports regarding the mediastinal nodes with lung cancer. The study highlights the value of IC by comparing the results of final histologic examination in 512 lymph nodes from 157 patients. The sensitivity, accuracy and negative predictive value of the examination were 95.7, 99.4 and 99.3%, respectively.
Since 1995, besides mediastinoscopy, staging VATS and anterior mediastinotomy have been performed with thoracotomy in our clinic. Our results revealed that IC in intraoperative staging of lung cancer showed an overall efficiency of 98.7%. The sensitivity (93.1%) and specificity (99.5%) of our study was also consistent with those reported in other studies comparing the diagnostic accuracy of IC with FS analysis.
Besides practicability, usefulness and accuracy, time is also very important in surgical pathology. Completion of a preparation and staining of an imprint slide takes 2 min while FS analysis needs at least 15 min. The average time for preparing an imprint slide in our pathology laboratory was 2–2.5 min.
Another advantage of IC is, allowing FS in difficult cases due to intact status of tissue remains. Although FS analysis has been used in our clinic since 1999, the entire 1050 lymph node samples were evaluated by IC intraoperatively.
In conclusion, IC is a simple, reliable and accurate method with less time-consuming and less labor-intensive features. Especially during the intraoperative evaluation of mediastinal lymph nodes in lung cancer, IC should be preferred over FS because it allows staging procedures and thoracotomy to be combined (‘single-stage operation’ feature).
|
References |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
1 Roberts JR, Blum MG, Arildsen R, Drinkwater DC, Christian KR, Powers TA, et al. Prospective comparison of radiologic, thoracoscopic, and pathologic staging in patients with early non-small cell lung cancer. Ann Thorac Surg 1999;68:1154–8.
2 Mountain CF, Dresler CM. Regional lymph node classification for lung cancer staging. Chest 1997;111:1718–23.
3 Reed CE, Silvestri GA. Invasive staging of the mediastinum. In: Shields TW, Locicero J, Ponn RB, editors. General Thoracic Surgery, 5th edn. Philadelphia: Lippincott Williams & Wilkins 2000;1306–9.
4 Gephardt GN, Rice TW. Utility of frozen-section evaluation of lymph nodes in the staging of bronchogenic carcinoma at mediastinoscopy and thoracotomy. J Thorac Cardiovasc Surg 1990;100:853–9.[Abstract]
5 Clarke MR, Landreneau RJ, Borochovitz D. Intraoperative imprint cytology for evaluation of mediastinal lymphadenopathy. Ann Thorac Surg 1994;57:1206–10.[Abstract]
6 Mouroux J, Venissac N, Alifano M. Combined video-assisted mediastinoscopy and video-assisted thoracoscopy in the management of lung cancer. Ann Thorac Surg 2001;72:1698–704.
7 Roviaro GC, Varoli F, Rebuffat C, Vergani C, Maciocco M, Scalambra SM, et al. Videthoracoscopic staging and treatment of lung cancer. Ann Thorac Surg 1995;59:971–4.
8 Vallieres E, Page A, Verdant A. Ambulatory mediastinoscopy and anterior mediastinotomy. Ann Thorac Surg 1991;52:1122–6.[Abstract]
9 Ghandur-Mnaymneh L, Paz J. The use of touch preparations (tissue imprints) in the rapid intraoperative diagnosis of metastatic lymph node disease in cancer staging procedures. Cancer 1985;56:339–44.[CrossRef][Web of Science][Medline]
10 Mair S, Lash RH, Suskin D, Mendelson GH. Intraoperative surgical specimen evaluation: frozen section analysis, cytologic examination, or both? A comparative study of 206 cases. Am J Clin Pathol 1991;96:8–14.[Medline]
11 Hasenburg A, Ledet SC, Ardaman T, Levy T, Kieback DG. Evaluation of lymph nodes in squamous cell carcinoma of the cervix: cytology versus frozen section histology. Int J Gynecol Cancer 1999;9:337–41.[Medline]
12 Lee TK. The value of imprint cytology in tumor diagnosis: a retrospective study of 522 cases in northern China. Acta Cytol 1982;26:169–71.[Medline]
13 Paulose RR, Shee CD, Abdelhadi IA, Khan MK. Accuracy of touch imprint cytology in diagnosing lung cancer. Cytopathology 2004;15:109–12.[CrossRef][Web of Science][Medline]
14 Terada Y, Matsunobe S, Nemoto T, Tsuda T, Shimizu Y. Feasibility of intraoperative cytodiagnosis of lung cancer. Chest 1990;98:1156–8.
15 Tamiolakis D, Mikroulis D, Lambropoulou M, Bitzikas G, Didilis V, Kotini A, et al. Imprint cytology for rapid evaluation of the lung and the mediastinal lesions. Minerva Med 2003;94:97–102.[Medline]
16 Sawabata N, Mori T, Iuchi K, Maeda H, Ohta M, Kuwahara O. Cytologic examination of surgical margin of excised malignant pulmonary tumor: methods and early result. J Thorac Cardiovasc Surg 1999;117:618–9.
17 Saito Y, Yamakawa Y, Kiriyama M, Fukai I, Kondo S, Kaji M, et al. Diagnosis of visceral pleural invasion by lung cancer using intraoperative touch cytology. Ann Thorac Surg 2002;73:1552–6.
18 Okuba K, Kato T, Hara A, Yoshimi N, Takeda K, Iwao F. Imprint cytology for detecting metastasis of lung cancer in mediastinal lymph nodes. Ann Thorac Surg 2004;78:1190–3.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||