| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Gastroenterological and Surgical Sciences, Clinica Chirurgica III, University of Padova School of Medicine, Padova, Italy
b Istituto Oncologico Veneto (IOV-IRCCS), Padova, Italy
Accepted for publication August 9, 2007.
* Address correspondence to Dr Battaglia, Istituto Oncologico Veneto (IOV-IRCCS), University of Padova School of Medicine, Via Giustiniani 2, Padova, 35128, Italy (Email: giorgio.battaglia{at}unipd.it).
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentReferences |
|---|
Methods: The study population consisted of 55 patients with superficial carcinoma of the esophagus who underwent EUS (October 2002 to January 2007). Endoscopic ultrasound features were compared with findings from surgical specimens or samples obtained at mucosectomy.
Results: There were 33 patients with adenocarcinoma (60%), which developed on Barrett’s esophagus in 27 cases, 21 patients (38%) with squamous cell carcinoma, and 1 (2%) with lymphoepithelial-like carcinoma. All lesions were confirmed as T1 on pathology. Of the 22 (40%) T1m lesions on EUS, 19 (86%) were confirmed as T1m on pathology; of the 33 T1sm on EUS, 22 (66%) were confirmed as T1sm. Positive predictive value of EUS for invasion of the submucosa was 67%, negative predictive value 86%, sensitivity 88%, specificity 63%, and diagnostic accuracy 75%. The accuracy of EUS in evaluating lymph node metastases was 71%, with a negative predictive value of 84%. Endoscopic ultrasound accuracy in differentiating mucosal from submucosal lesions increased from the lower esophagus or gastroesophageal junction to the mid and upper esophagus (71%, 76%, and 100%, respectively; not significant). As for the histologic type, accuracy was 70% for adenocarcinoma and 81% for squamous cell carcinoma, (not significant); for lesions detected as type 0-IIa (13 patients), accuracy was 100%; for type 0-I lesions (23 patients), accuracy was 70% (p = 0.03).
Conclusions: Despite difficulties in differentiating mucosal from submucosal lesions, even with 20-MHz miniprobes, EUS remains an extremely valuable tool when nonsurgical treatments are considered. Its staging accuracy depends on site and macroscopic appearance of the neoplasm.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Several factors, however, have stimulated interest in alternative, nonsurgical treatment options: the natural history of high-grade dysplasia is variable: in some patients it remains stable for long periods, whereas in others it can evolve toward invasive disease in a relatively short time frame [2]. Esophagectomy suffers from high mortality and morbidity rates. Furthermore, many patients are not suitable candidates for surgery because of comorbidities.
The presence of lymph node metastases is strictly related to the depth of tumor infiltration in the esophageal wall. This means that it is negligible when the carcinoma is limited to the mucosa but reaches 20% to 30% in case of submucosal invasion. Distinguishing between superficial carcinoma (limited to the mucosa, T1m) and more advanced ones infiltrating the submucosa (T1sm and higher) is crucial to select patients for local endoscopic resection. Endoscopic ultrasound (EUS) provides a detailed image of the esophageal wall and is the most accurate means available to date for staging esophageal carcinoma, in terms of both depth of invasion (T stage) and presence or absence of lymph nodes (N stage).
The aims of this study were to assess the diagnostic accuracy of EUS in identifying the more superficial lesions (T1m and T1sm) in patients undergoing esophagectomy or local endoscopic resection, and to establish whether certain variables (site of neoplasm, histologic type, macroscopic appearance) can influence the diagnostic accuracy of EUS in distinguishing mucosal from submucosal lesions.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
The endoscopic features of the tumor were assessed in all patients (ie, site and macroscopic appearance; Table 1), as were the ultrasound findings (including EUS staging), the histopathologic findings, and, for patients undergoing operation, the pathologic staging.
|
The EUS features were compared with the surgical specimen or the sample obtained on mucosectomy. The accuracy of EUS in establishing the presence or absence of lymph node metastases was only assessed in patients who underwent surgery. None of the patients had previously received chemotherapy or chemoradiotherapy.
The study was approved by the Ethics Committee of the University of Padova Medical School. Given that individual patients could not be identified, the need for patient consent was waived.
Endoscopic Ultrasound
Endoscopic ultrasound was performed under conscious sedation with benzodiazepine in 31 patients (56%), whereas propofol was used in 19 patients (35%) and a local oropharyngeal anesthetic was used in 5 patients (9%). All EUS procedures were performed by the same expert endoscopist, using an operative (Olympus GIF 130T) or dual-channel (Olympus 2T160) endoscope to allow for the passage of the miniprobes and the injection of water directly with a pedal pump (pressure-controlled and temperature-controlled system), with 12- and 20-MHz miniprobes (Olympus UM-2R/3R, Tokyo, Japan), both in all patients.
Patients lay supine on their left side, with their shoulder raised in an anti-Trendelenburg position. In the event of proximal esophageal lesions, a soft silicone overtube was used to prevent any aspiration of water into the airways. No complications were recorded.
The integrity of the hyperechogenic third sonographic layer, or the interface between the second and third layer underneath the lesion, excluded an invasion of the submucosa. Lesions with these features were classified as T1m. Invasion of the submucosa (T1sm) was diagnosed when there was evidence of hypoechogenic destruction of the interface between the second and third sonographic layers or hypoechogenic focal thickening of the third layer.
Lymph node staging was completed in all cases. The lymph node status was assessed on the full length of the esophagus and also in the proximal and celiac gastric region. Lymph nodes were considered positive (N1) when two of the following criteria were met: lymph node size greater than 5 or 10 mm, hypoechogenic, with rounded shape and with clearly defined margins. According to Bhutani and colleagues [4], when all four of these variables are present, accuracy in predicting lymph node involvement is close to 85%. A lymph node was considered not involved (N0) if it was small, isoechogenic or hyperechogenic, with poorly defined margins, and not rounded in shape. Endoscopic ultrasound–guided fine-needle aspiration (FNA) biopsy was not performed in patients considered in the present analysis because it was not available until early 2007.
Patients were considered eligible for endoscopic treatment if the superficial lesion was up to a diameter of 20 mm in case of well-differentiated nondepressed lesions, and up to 10 mm in case of depressed lesions [3]. Treatment choice was also based on the patients’ conditions (severe comorbidities, refusal for surgery, clear lymphadenopathy).
Patients undergoing curative mucosectomy were followed up with endoscopy at 1 month (with biopsies on and around the scar, to make sure that there were no other foci of neoplastic disease), and then with endoscopy and EUS at 6, 12, 18, and 24 months.
Histopathologic Assessment
Surgical specimens and samples obtained on mucosectomy were analyzed according to a standard method: the tissues were fixed in 5% formalin and the area of tumor and adjacent normal tissue were embedded in paraffin. Sections 5 µm thick were cut from each block and stained with hematoxylin and eosin.
Based on the 6th edition of the TNM classification of malignant tumors [5], T1 carcinomas were further classified as T1a or T1m (limited to the mucosa or to the muscularis mucosae) and T1b or T1sm (extending to the submucosa).
Surgery and Local Endoscopic Resection
Thirty-eight patients (69%) underwent esophagectomy: the Ivor-Lewis or McKeown method was used in 35 patients (92%) means of a transthoracic approach, with standard lymphadenectomy extending to two levels, and esophagogastroplasty; the transhiatal approach was used in 3 patients (8%), with esophagojejunoplasty in 1 patient. Seventeen patients (31%) underwent endoscopic mucosectomy without lymphadenectomy.
Mucosectomy is defined as the resection of a fragment of the digestive tract wall, including the mucosa, the muscularis mucosae, and all or part of the submucosa [6]. The technique is called "suck and cut" and has four steps: to mark the lesion at the four cardinal points with a closed diathermal loop; to inject saline solution to detach the mucosa and submucosa from the muscularis propria; to aspirate the lesion inside a hood, while applying an elastic ring to the base of the lesion, which is resected with a diathermal loop. Alternatively, the lesion can be lifted, aspirated inside a hood and resected directly with a diathermal loop. At the end of the procedure, biopsies are taken from the fundus and at the margins of the mucosectomy.
Statistical Analysis
Data are expressed as median and interquartile ranges. Fisher’s exact test was used to compare categorical data. A probability value of less than 0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
|
All 55 patients staged as T1 at EUS were confirmed as T1 on pathology, with a 100% diagnostic accuracy of EUS in distinguishing T1 lesions from a more advanced stage of disease (T2 and beyond).
On EUS, 22 of the 55 patients (40%) had an uninterrupted third layer, suggesting no invasion of the submucosa (stage T1m; Fig 1). Signs of hypoechogenic disruption of the third layer (stage T1sm) were recorded in 33 of 55 patients (60%; Fig 2, center).
|
|
The accuracy of EUS in identifying invasion of the submucosa is shown in Table 3. Absence of submucosal invasion was confirmed in 19 of 22 patients (86%) staged T1m on EUS, whereas 22 of 33 patients (66%) showing submucosal lesions on EUS were confirmed T1sm on pathology. The positive predictive value of EUS for submucosal invasion was 67%, the negative predictive value was 86%, the sensitivity was 88%, and the specificity was 63%. Its diagnostic accuracy was 75%.
|
|
Macroscopic appearance was another important variable in case of type 0-IIa lesion (13 patients); the diagnostic accuracy was 100%, although this dropped to 70% for nodular lesions, type 0-I (in 23 patients; p = 0.03).
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Endoscopic ultrasound findings should consequently be interpreted with caution, especially in case of nodules or protruding lesions, which can negatively affect the reliability of the method (the diagnostic accuracy of EUS dropped in our series from 100% for type 0-IIa lesions to 70% for type 0-I lesions). Based on experience with EUS in a series of patients with Barrett’s esophagus and dysplasia, Falk and colleagues [10] suggested that nodularity of the mucosa contributed to lesion overstaging, probably as a result of the altered pattern of the sonographic layer from inflammatory changes.
Overstaging can be attributed to the following factors [11]: (1) peritumoral inflammation, which leads to wall thickening, making the different layers less easily distinguishable; (2) inappropriate positioning of the ultrasound transducer, causing a pseudothickening and a poor view of the layers (a problem overcome by the use of miniprobes); and (3) massive invasion of the submucosa, which can reduce the thickness of the hyperechogenic medial layer to the point where it becomes unrecognizable. The accuracy of EUS in differentiating mucosal from submucosal lesions drops progressively from the upper esophagus toward the gastroesophageal junction; this is attributable to technical problems: it is more difficult to create an acoustic interface at the gastroesophageal junction unless the instruments used are fitted with a balloon, but this poses problems of pressure applied to the mucosa.
Miniprobes that pass through the endoscopic operating channel have been developed to assess superficial lesions of the gastrointestinal tract. The potential advantages of these devices include a higher definition, owing to their high frequency and positioning under direct endoscopic guidance. The images obtained with high-frequency probes afford an excellent view of the wall, revealing up to nine layers in the esophagus [12]. Compared with standard EUS, miniprobes reveal a greater diagnostic accuracy for T staging [12] for the muscularis mucosae (100%), the submucosa (83%), and the muscularis propria (100%) [13]. Cases of superficial carcinoma appear as a focal thickening, restricted to the first two (T1m) or three layers (T1sm) [12]. In a Japanese study [14], 96 lesions were staged using miniprobes, revealing an accuracy of 93% in differentiating between mucosal and submucosal lesions. In our series, its diagnostic accuracy was only 75%, confirming the difficulty in distinguishing mucosal from submucosal lesions, as reported by May and associates [15], who calculated a staging accuracy of 79.6% (74 of 93 patients) for high-resolution EUS.
Endoscopic mucosectomy remains a valid preoperative staging method, which becomes a curative treatment in cases of intramucosal lesions included entirely in the resection margins. When there is evidence of submucosal involvement, however, mucosectomy cannot be considered a definitive treatment, and surgery or chemoradiotherapy (for unfit patients or for those refusing surgery) is recommended [16]. As for the accuracy of EUS in determining malignant adenopathy associated with esophageal carcinoma, this varies from 50% to 70% [17, 18], and miniprobes are less accurate for N-staging purposes than standard EUS.
In our study, the negative predictive value was 84%, and the diagnostic accuracy was 71%, as reported in the literature. According to Shami and coworkers [19], a lymph node is highly suspect when two of the following ultrasound criteria are met: a rounded shape, clearly defined margins, and hypoechogenicity. Size cannot be considered a criterion to rule out malignancy, however; in fact, two of five pathologic lymph nodes were less than 1 cm in size. Combining needle aspiration with EUS can increase its diagnostic accuracy: in 20% of patients, EUS and needle aspiration prompted a change in the treatment choice. The results obtained using this approach cannot be compared with the histopathologic analysis of the surgical specimen, however.
In conclusion, EUS is an extremely useful tool when considering emerging nonsurgical treatments that can be effective alternatives to esophagectomy in patients with superficial carcinoma of the esophagus who refuse surgery or are unsuitable candidates for esophagectomy because of comorbidities. So far, however, it remains extremely difficult to differentiate between mucosal and submucosal lesions, even using high-frequency miniprobes. In our study, we used 12- and 20-MHz miniprobes; 30-MHz miniprobes seem to improve accuracy, but further studies are needed to confirm this finding. The diagnostic accuracy of EUS depends on the site of the neoplasm: it increases progressively from the gastroesophageal junction toward the upper esophagus, where it is 100% accurate. It is also influenced by the macroscopic appearance of the lesion: it is greatest if the lesion is slightly elevated (0-IIa) and declines if the lesion is protruding or nodular (type 0-I; p = 0.03). Histologic type (squamous cell carcinoma or adenocarcinoma) does not significantly affect the accuracy of EUS in T staging, and is greater for squamous cell carcinoma. Finally, for best diagnostic accuracy, we recommend using miniprobes for T staging and 7.5- to 12-MHz probes in combination with fine needle aspiration in case of suspected lymph nodes for N staging.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
