Esophageal cancer staging: improved accuracy by endoscopic ultrasound of celiac lymph nodes Carolyn E.

Reed, Girish Mishra, Anand V. Sahai, Brenda J. Hoffman and Robert H. Hawes Ann Thorac Surg 1999;67:319-321

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://ats.ctsnetjournals.org/cgi/content/full/67/2/319

The Annals of Thoracic Surgery is the official journal of The Society of Thoracic Surgeons and the Southern Thoracic Surgical Association. Copyright 1999 by The Society of Thoracic Surgeons. Print ISSN: 0003-4975; eISSN: 1552-6259.

Downloaded from ats.ctsnetjournals.org by on July 6, 2007

Esophageal Cancer Staging: Improved Accuracy by Endoscopic Ultrasound of Celiac Lymph Nodes
Carolyn E. Reed, MD, Girish Mishra, MD, Anand V. Sahai, MD, Brenda J. Hoffman, MD, and Robert H. Hawes, MD
Department of Surgery and Division of Gastroenterology, Medical University of South Carolina, Charleston, South Carolina

Background. Clinical staging of esophageal cancer is required for optimal therapy but remains imprecise. Pathologic verication of involved lymph nodes could potentially direct treatment allocation. With the rising incidence of distal and gastroesophageal junction adenocarcinomas, assessment of the celiac axis lymph nodes (CLNs) becomes important because it is a common nodal drainage basin. Endoscopic ultrasound (EUS) permits evaluation of CLNs and biopsy by ne-needle aspiration. This study examined the usefulness of this staging tool. Methods. A consecutive series of 62 patients with esophageal cancer considered resectable by computed tomographic scan underwent EUS for T and N staging and were retrospectively studied. A CLN visualized by EUS as greater than 5 mm was considered positive. Fine-needle aspiration of the CLN was performed routinely. Endoscopic ultrasound and computed tomographic staging were compared on the basis of pathologic verication of CLNs. Results. It was possible to evaluate CLNs by EUS in 59 (95%) of 62 patients: positive in 19, negative in 40. In EUS-positive patients, ne-needle aspiration was posi-

tive in 15, falsely negative in 2, and not done in 2. By computed tomographic scan, CLNs were negative in 57 patients and positive in 2. The CLNs were positive in 23 of 54 patients eligible for CLN pathologic verication. All positive CLNs not identied by EUS (7 false-negative EUS) were microscopic foci in one or two nodes and were associated with T3 tumors. Sensitivity and specicity of EUS were 72% and 97%, respectively, compared with 8% and 100% for computed tomographic scan. When EUS identied CLNs, ne-needle aspiration conrmed positivity in 88% of cases. Conclusions. Endoscopic ultrasound with ne-needle aspiration is useful in the detection and conrmation of CLN metastasis. In T3 tumors of the distal esophagus, a negative EUS result does not substantiate absence of CLN disease. Endoscopic ultrasound with ne-needle aspiration may be important in guiding treatment for patients with distal adenocarcinoma and documenting disease before neoadjuvant therapy. (Ann Thorac Surg 1999;67:319 22) 1999 by The Society of Thoracic Surgeons needle aspiration (FNA). The purpose of the present study was to evaluate the usefulness of this staging tool in potentially resectable tumors.

he staging of esophageal cancer remains imprecise. Computed tomography (CT) is useful for evaluating metastatic disease but does not allow accurate assessment of depth of tumor invasion or regional lymph node status. Endoscopic ultrasound (EUS) is presently the most accurate tool for estimating depth of tumor [13] and compared with CT has improved accuracy in the identication of metastatic regional lymph nodes [15]. However, histologic verication of involved lymph nodes is needed to clinically stage esophageal cancer more conclusively. Such ability would improve the design and interpretation of clinical trials and potentially help allocate patients to a particular treatment modality (ie, neoadjuvant therapy). There is a rising prevalence of distal and gastroesophageal junction adenocarcinoma [6]. The celiac axis lymph nodes (CLNs) are considered by most surgeons to be a regional nodal basin of these tumors. Endoscopic ultrasound permits evaluation of CLNs and biopsy by nePresented at the Forty-fth Annual Meeting of the Southern Thoracic Surgical Association, Orlando, FL, Nov 1214, 1998. Address reprint requests to Dr Reed, Medical University of South Carolina, 171 Ashley Ave, Charleston, SC 29425 (e-mail: reedce@ musc.edu).

Patients and Methods Patients

Between January 1, 1994, and April 1, 1998, a consecutive series of 62 patients presenting with esophageal cancer and having no evidence of T4 or metastatic disease by CT scan were considered potential surgical candidates and underwent EUS staging at the Medical University of South Carolina. The status of the CLNs in this group was retrospectively reviewed. Of the 62 patients, 51 were male and 11 female. Tumor location was midesophageal in 13 patients, distal esophageal in 35, gastroesophageal junction in 13, and diffuse in 1 (mid and distal esophagus). Histologic studies revealed adenocarcinoma in 38 patients, squamous cell in 22, adenosquamous in 1, and carcinosarcoma in 1.

Endoscopic Ultrasound and Fine-Needle Aspiration

Endoscopic ultrasound was performed, after informed consent was obtained, in an outpatient endoscopy suite
0003-4975/99/$20.00 PII S0003-4975(99)00031-4

1999 by The Society of Thoracic Surgeons Published by Elsevier Science Inc

Downloaded from ats.ctsnetjournals.org by on July 6, 2007

320

REED ET AL EUS/FNA OF CELIAC NODES IN ESOPHAGEAL CANCER

Ann Thorac Surg 1999;67:319 22

using topical oropharyngeal anesthesia and intravenous sedation (midazolam and meperidine). All EUS examinations were initially performed at 7.5- and 12-MHz frequencies using a radial scanning echoendoscope (Olympus GIF-UM 20; Olympus American Inc, Lake Success, NY). Dilation was performed up to 42F to 45F to allow passage of the 13-mm echoendoscope using wire-guided Savary-Gilliard dilators (Wilson-Cook Medical, Inc, Winston-Salem, NC). An attempt to visualize the celiac axis was made in all cases by positioning the echoendoscope in the proximal stomach. In normal subjects, EUS does not detect CLNs. In the present study, a node greater than 5 mm was considered positive. Peritumoral nodes that were greater than 10 mm in the short-axis diameter, round and hypoechoic, and having sharp borders were recorded as abnormal. Fine-needle aspiration biopsy of CLNs was performed using either a mechanical sector scanning biopsy echoendoscope (Olympus GF UM 30P; Olympus America Inc) or a linear array sector scanning echoendoscope (FG 32-UA; Pentax, Orangeburg, NY). The instrument was placed in the stomach lumen opposite the identied CLN. The FNA needlesheath system, consisting of a 23- or 24-gauge needle (WilsonCook, Olympus America Inc or Pentax) was inserted through the working channel of the endoscope. The needle was advanced from the metal sheath through the wall of EUS normal mucosa of the stomach and guided into the target site using real-time ultrasound. The stylet was removed, and suction was applied with a 10-mL syringe while the needle was manipulated back and forth within the lymph node. After 1 minute, the suction was released, and the needle was retracted. The aspirate was placed on glass slides, preserved with Diff-Quik stain (American Scientic Products, McGraw Park, IL), and reviewed immediately by an on-site pathologist to ensure adequate specimen (ie, lymphocytes). At least three adequate samples were obtained.

Results
It was possible to evaluate CLNs by EUS in 59 (95%) of the 62 patients. In 3 patients, the obstructing tumors could not be dilated, and a catheter miniprobe (Olympus UM2R/3R, Olympus American Inc) was used to assess the proximal portion of the lesions. In the 59 remaining patients, CLNs were considered positive by EUS in 19 patients and negative in 40. In the EUS-positive patients, FNA was positive in 15 patients, falsely negative in 2, and not done in 2. There was no morbidity when FNA was added to EUS. By CT scan, CLNs were negative in 57 patients and positive in 2. The CLNs were found to be positive in 26 of 57 patients eligible for CLN evaluation. One patient did not undergo laparotomy secondary to progressive disease found at initial cervical exploration. In another patient, primary tumor directly invaded the celiac axis, and separate assessment of the nodes was not possible. Of the 26 patients with positive CLNs, 7 were not identied by EUS. These 7 patients had microscopic foci of tumor in one or two celiac nodes (approximately 15% of all detected CLNs), and all these patients had T3 tumors. No false-negative EUS ndings occurred in patients with T1 or T2 tumors. Thirty-three patients underwent preoperative adjuvant therapy, with most receiving two cycles of 5-uorouracil and cisplatin combined with concurrent radiotherapy. Three patients were EUS positive for involvement of CLNs and were found to have N0 disease at resection. Squamous cell carcinoma considered potentially resectable in 4 patients was declared unresectable after FNA of CLNs was positive (M1 disease). They were treated with chemoradiation or radiation. Three other patients who had FNA-positive CLNs chose to be treated by chemoradiotherapy alone. The calculated sensitivity and specicity of EUS in the 57 patients with pathologic verication of CLNs was 72% and 97%, respectively. The CT sensitivity was 8% and specicity 100%. The positive predictive value of EUS was 95%, and the negative predictive value was 82%. When EUS identied CLNs, FNA conrmed positivity in 88% of cases.

Pathologic Verication
Histologic or cytologic examinations of the CLNs was possible in all evaluable cases but 1 by FNA, surgical resection, or laparoscopy (1 patient). Progressive disease occurred in 1 patient, which was veried by cervical exploration, and no laparotomy was performed. At the time of operation, CLNs (lymph nodes at the base of the celiac axis and the proximal left gastric artery) were dissected en bloc with the specimen and labeled by the surgeon (C.R.). When small, the entire lymph node was submitted. Larger lymph nodes were bisected and the two portions examined.

Comment
The staging of cancer allows the physician to recommend treatment, anticipate the prognosis, and communicate and study the results of various treatment options. The staging of esophageal cancer has lacked precision in the identication of the T and N components. Computed tomography, the most frequently used staging tool, is helpful in evaluating adjacent organ involvement (T4 disease) and metastases, but is inaccurate in the assessment of nodal disease and depth of tumor invasion. Endoscopic ultrasound is presently the most accurate noninvasive method for staging the T and N components [15]. The ability to combine FNA with EUS potentially offers greater precision in staging by histologic verication of identied lymph nodes. Although not yet as

Analysis
The sensitivity and specicity of EUS for staging CLNs were calculated on the basis of pathologic verication and compared with CT scanning results. The negative and positive predictive values were assessed for EUS evaluation of CLNs.

Downloaded from ats.ctsnetjournals.org by on July 6, 2007

Ann Thorac Surg 1999;67:319 22

REED ET AL EUS/FNA OF CELIAC NODES IN ESOPHAGEAL CANCER

321

rened, but in a manner similar to the mediastinoscopic evaluation of nonsmall cell lung cancer to conrm stage IIIA disease, this procedure could then aid in the allocation of treatment and the interpretation of the results of clinical trials. At present, EUS with FNA of peritumoral lymph nodes cannot be performed because the needle would pass through the tumor in the esophageal wall. Abnormal periesophageal nodes distal or proximal to the tumor can be sampled. The CLNs are readily accessible for EUSguided FNA. With the rising incidence of distal adenocarcinoma, CLNs become an important site of metastatic drainage. As in the case at many centers, we considered distal adenocarcinomas with CLN involvement to be locally advanced (ie, N1 disease) and still potential candidates for resection. The usefulness of the evaluation of CLNs by EUS and biopsy by FNA was assessed in a group of patients being considered for operation. In the present series, the sensitivity of EUS for identication of positive CLNs was 72%, with a negative predictive value of 82%. As expected, CT was extremely poor in staging CLNs. It is doubtful that the sensitivity of EUS will ever be greater than 70% to 75% because microscopic foci of metastatic disease in normal lymph nodes ( 5 mm) will be missed. However, when CLNs are detected by EUS, FNA is highly successful in conrming metastasis. To increase the yield of identifying positive CLNs before treatment, laparoscopic evaluation has been proposed. Both Luketich and associates [7] and Krasna and colleagues [8] have used thoracoscopic and laparoscopic staging to assess lymph nodes. Laparoscopy accurately identied the disease status of CLNs in 16 of 17 patients described by Krasna and associates [8]. Luketich and colleagues [7] compared EUS with minimally invasive operation in 26 patients. Endoscopic ultrasound was considered positive for N1 disease in 12 of 21 patients, whereas laparoscopy and thoracoscopy revealed N1 disease in 18 of these 21 patients. However, only a linear array echoendoscope was used in the Luketich study. Most centers stage with a radial scanner, which permits complete visualization in a 360-degree eld. With the requirement of general anesthesia, an operative time often exceeding 3 hours, and the need for a 2- to 3-day hospital stay, laparoscopic and thoracoscopic staging are actually invasive staging tools. Luketich and colleagues [7] estimated total average charges at $20,000 to $25,000 for such staging. At our institution, EUS with FNA adds approximately 30 minutes to routine outpatient esophagoscopy, and total charges average $1,000. Positron emission tomography is the newest staging modality, and its usefulness in esophageal cancer is undened. At present, it is most useful in detecting distant metastatic disease. Peritumoral lymph nodes cannot easily be distinguished from the tumor. Celiac lymph nodes less than 10 mm in diameter containing metastatic tumor would fall below the sensitivity of positron emission tomography [9, 10]. Rice and associates [11] have shown that for patients with esophageal carcinoma, T status is an important

predictor of N disease. The likelihood of N1 disease was 77% for 219 patients with T3 disease. In our series, the 7 patients who had false-negative EUS assessment of CLNs had T3 tumors. One could argue that this select group might benet from laparoscopic assessment. However, 4 of these patients had peritumoral lymph nodes that were classied as positive. With the anticipated development of a sheathed aspiration needle, the need for more aggressive staging would be limited to very few patients. Ultimately, the usefulness of a staging tool will depend on its ability to affect treatment decisions. The best treatment for T3 or N1 disease, or both, remains controversial and requires further study. Pathologic verication of N1 disease has the potential of providing baseline information against which effectiveness of treatment can be measured. In conclusion, EUS with FNA is useful in the detection and conrmation of CLN metastasis. In T3 tumors of the distal esophagus, negative EUS results do not substantiate absence of CLN disease. Endoscopic ultrasound with FNA may prove to be critically important in guiding treatment of patients with distal adenocarcinomas and in documenting involved CLNs before preoperative neoadjuvant therapy.

References
1. Ziegler K, Sanft C, Zeitz M, et al. Evaluation of endosonography in TN staging of oesophageal cancer. Gut 1991;32: 1620. 2. Botet JF, Lightdale CJ, Zauber AG, et al. Preoperative staging of esophageal cancer: Comparison of endoscopic US and dynamic CT. Radiology 1991;181:41925. 3. Tio TL, Coene PPLO, Schouwink MH, Tytgat GNJ. Esophagogastric carcinoma: preoperative TNM classication with endosonography. Radiology 1989;173:41117. 4. Grimm H, Soehendra N, Hamper K, et al. Endosonography versus computed tomography for determination of locoregional spread in esophageal cancer: a prospective controlled study. In: Ferguson MK, Little AG, Skinner DB, eds. Diseases of the esophagus, vol 1: Malignant diseases. Armonk, NY: Futura, 1990:12534. 5. Vilgrain V, Mompoint D, Palazzo L, et al. Staging of esophageal carcinoma: comparison of results with endoscopic sonography and CT. AJR Am J Roentgenol 1990;155:277 81. 6. Blot WJ, Devesa SS, Kneller RW, Fraumeni JF. Rising incidence of adenocarcinoma of the esophagus and gastric cardia. JAMA 1991;265:128791. 7. Luketich JD, Schauer P, Landreneau R, et al. Minimally invasive surgical staging is superior to endoscopic ultrasound in detecting lymph node metastases in esophageal cancer. J Thorac Cardiovasc Surg 1997;114:81723. 8. Krasna MJ, Flowers JL, Attar S, McLaughlin J. Combined thoracoscopic/laparoscopic staging of esophageal cancer. J Thorac Cardiovasc Surg 1996;111:8007. 9. Block MI, Patterson GA, Sundaresan RS, et al. Improvement in staging of esophageal cancer with the addition of positron emission tomography. Ann Thorac Surg 1997;64:7707. 10. Luketich JD, Schauer PR, Meltzer CC, et al. Role of positron emission tomography in staging esophageal cancer. Ann Thorac Surg 1997;64:7659. 11. Rice TW, Zuccaro G, Adelstein DJ, Rybicki LA, Blackstone EH, Goldblum JR. Esophageal carcinoma: depth of tumor invasion is predictive of regional lymph node status. Ann Thorac Surg 1998;65:78792.

Downloaded from ats.ctsnetjournals.org by on July 6, 2007

Esophageal cancer staging: improved accuracy by endoscopic ultrasound of celiac lymph nodes Carolyn E. Reed, Girish Mishra, Anand V. Sahai, Brenda J. Hoffman and Robert H. Hawes Ann Thorac Surg 1999;67:319-321
Updated Information & Services Citations Permissions & Licensing including high-resolution figures, can be found at: http://ats.ctsnetjournals.org/cgi/content/full/67/2/319 This article has been cited by 12 HighWire-hosted articles: http://ats.ctsnetjournals.org/cgi/content/full/67/2/319#otherarticles Requests about reproducing this article in parts (figures, tables) or in its entirety should be submitted to: http://www.us.elsevierhealth.com/Licensing/permissions.jsp or email: healthpermissions@elsevier.com. For information about ordering reprints, please email: reprints@elsevier.com

Reprints

Downloaded from ats.ctsnetjournals.org by on July 6, 2007