|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Second Department of Propedeutic Surgery, Division of Thoracic Surgery Athens University School of Medicine (Dr. Dosios); and Division of Thoracic Surgery, Henry Dunant Hospital, Athens, Greece (Dr. Theakos).
Correspondence to: Theodosios Dosios, MD, Second Department of Propedeutic Surgery, Division of Thoracic Surgery Athens University School of Medicine, 2 Chatzigianni Mexi Str, 11528, Athens, Greece; e-mail: dosiosth{at}internet.gr
| Abstract |
|---|
|
|
|---|
Materials and methods: During the last 28 years, 39 consecutive patients with SVCO underwent biopsy of mediastinal lesions by CMDS (n = 18) or AMDT (n = 19) or both these techniques (n = 2). The medical records of all patients were reviewed, and demographic data, operative notes, perioperative complications, outcome, and histologic diagnoses were examined. The findings were compared with those of 367 patients without SVCO who underwent biopsy of mediastinal lesions during the same period of time. An up-to-date English-language literature search was performed.
Results: The sensitivity of CMDS and/or AMDT in detecting malignancies in 39 patients with SVCO was 97.4%, specificity was 100%, and diagnostic accuracy was 97.4%. There was no in-hospital mortality, while morbidity consisted of five major complications and one minor complication, including two major hemorrhages and two airway obstructions. These patients, compared to those without SVCO, showed significantly higher postoperative morbidity (p < 0.001) and had a higher rate of malignancy (p < 0.001). Among 280 patients of the literature review, major hemorrhage was recorded in eight cases and airway obstruction in none.
Conclusions: CMDS and AMDT are effective methods to establish a histologic diagnosis in patients with SVCO. Although their mortality is negligible, they are accompanied by a significantly higher morbidity compared to patients without SVCO. Airway obstruction is a life-threatening complication that can occur in these patients. In our series, patients with SVCO had a higher rate of malignancy compared to patients without SVCO.
Key Words: airway obstruction anterior mediastinotomy biopsy cervical mediastinoscopy hemorrhage superior vena cava obstruction
| Introduction |
|---|
|
|
|---|
Cervical mediastinoscopy (CMDS) and anterior mediastinotomy (AMDT) have been advocated as the most reliable techniques in establishing a histologic diagnosis in patients with SVCO.789 However, because these techniques necessitate dissection and biopsy of tissue in the vicinity of vital organs of the superior mediastinum and in the presence of venous stasis, they entail the risk for hemorrhage and other complications. The purpose of this study is to evaluate the safety and diagnostic efficacy of CMDS and AMDT in patients with SVCO by reviewing our experience and the pertinent English-language literature.
| Materials and Methods |
|---|
|
|
|---|
The diagnosis of SVCO was made on clinical grounds in all patients. Prior to operation, patients underwent a thorough physical examination for detection of peripheral lymph nodes accessible for biopsy with minor procedures, and routine laboratory investigation including CT of the thorax. In a few patients, we chose MRI as a third-line examination. None of the patients underwent fine-needle aspiration cytology or core biopsy.
The medical records of all patients were reviewed, and demographic data, operative notes, perioperative complications, outcome, and histologic diagnoses were examined. The findings were compared with those of patients without SVCO who underwent biopsy of mediastinal lesions during the same period of time. In this article, major hemorrhage is defined as bleeding that required exploration through a sternotomy or thoracotomy for definitive control.
Surgical Technique
All patients underwent open biopsy of mediastinal lesions without any previous medical treatment for SVCO syndrome. The position of the patient on the operating table was with the trunk slightly elevated and the neck overextended. All operations were performed under general endotracheal anesthesia. The operative plan was based on chest CT findings. Patients with paratracheal and pretracheal lesions were submitted to CMDS, while those with lesions in front of the great vessels of the anterior-superior mediastinum underwent AMDT. The operative technique did not substantially differ between patients with SVCO and those without SVCO. However, patients with SVCO had a more careful hemostasis and a meticulous dissection. It is important to note that in performing CMDS on such patients, we remained strictly within the midline when dissecting to the trachea in the neck. Sometimes the tumor was easily palpable at the thoracic inlet and CMDS was a very limited procedure. Biopsy samples were submitted to frozen section in all patients. The wound was closed after confirmation by the frozen-section biopsy that adequate diagnostic tissue was included in the sample(s).
Literature Search
We utilized the MEDLINE database to perform an up-to-date English-language literature search using the terms "superior vena cava obstruction," "cervical mediastinoscopy," and "anterior mediastinotomy."
Statistical Methods
Statistical analysis was performed using the
2 test with Yates correction factor and Student t test. Statistical significance was assumed at p < 0.05.
| Results |
|---|
|
|
|---|
Table 1 presents the comparative data between patients with SVCO and those without SVCO regarding demographic information, operations performed, mortality, morbidity, and the malignant/benign diagnoses ratio. Patients with SVCO compared with those without SVCO showed significantly higher postoperative morbidity (p < 0.001), required exploration for management of intraoperative complications more frequently (p < 0.001), and had a higher rate of malignancy (p < 0.001).
|
|
Overall morbidity was 8.2%. Major hemorrhage was the main complication of concern in most of the publications of this review, although its definition was unclear in some of them. Major hemorrhage was recorded in 8 of 319 patients (range, 0 to 12.5%). Major hemorrhage was arterial in three patients, venous in four patients, and undetermined in one patient. Other complications included bleeding controlled by tamponade (n = 5), wound infection (n = 5), episodes of airway obstruction, cardiac arrhythmias, and temporary arterial hypotension (Table 3 ).
|
| Discussion |
|---|
|
|
|---|
Histologic diagnosis of the mediastinal lesion can be obtained in some patients with SVCO by lesser invasive techniques, such as bronchoscopy, peripheral lymph node biopsy, and even transvenous biopsy.1618 However, the diagnostic yield of these methods is low.17 During the last 20 years, transthoracic needle and core biopsy have been used for the same purpose with controversial results. By using these diagnostic modalities on patients with mediastinal masses or lymphadenopathy but no signs of SVCO, a diagnostic yield of 71.3 to 100% has been reported by several authors,1920212223 who affirm that these techniques constitute reliable and sensitive diagnostic tools that provide a precise diagnosis and obviate the need for exploratory thoracic surgery in the majority of mediastinal lesions. In patients with SVCO, Ko et al24 from China obtained a diagnostic yield of 83.1% with ultrasound-guided transthoracic needle aspiration biopsy on 30 undiagnosed patients between 1989 and 1993, and concluded that this approach appears to be safe, effective, and rapid for obtaining an accurate histologic diagnosis. However, most recent publications favor CMDS and/or AMDT in preference to needle biopsy in patients with SVCO. Jahangiri and Goldstraw8 of Brompton Hospital in London reported that they do not perform needle biopsy in the presence of SVCO because their radiologists are reluctant to do so and their pathologists find it difficult to differentiate lymphomas from small cell carcinoma. Porte et al17 from France concluded that CT-guided biopsy is useful for the diagnosis of carcinoma and thymoma but is not sufficiently accurate for precise individualization of NHL, which requires larger samples for immunohistochemical and cytogenetic studies. Even when CT-guided biopsy permits the diagnosis of NHL, a surgical biopsy is required for its characterization so that further treatment can be planned in as appropriate a fashion as possible. In cases of suspected NHL compressing the trachea or the main bronchi to a point at which general anesthesia with tracheal intubation would be extremely hazardous, they advocate anterior mediastinotomy under local anesthesia rather than CT-guided biopsy. Watanabe et al25 from Japan found that diagnostic accuracy was significantly higher in mediastinotomy biopsies than in needle biopsies. Finally, Massie et al,26 who sampled the mediastinal lesions of 55 children in Australia, concluded that despite recent advances in imaging technology and biopsy techniques, full histologic examination is required to exclude malignancy. We advocate the use of less invasive techniques in selected patients prior to open biopsy, especially when lung cancer is suspected. When other than lung cancer is the presumptive diagnosis, CMDS and/or AMDT are the procedures of choice because histologic and immunohistochemical diagnoses are required that necessitate an amount of tissue that can be obtained only by open biopsy.27
In this collective series of 319 patients with SVCO who underwent CMDS and/or AMDT, there was no mortality related to the procedures, while morbidity was 8.2%. The most severe of the complications was "major hemorrhage," which was recorded at a rate ranging from 0 to 12.5%. This incidence is significantly higher than that of 0 to 0.3% reported in patients without SVCO.28 Obviously, prevention is the best treatment for this potentially mortal complication. To avoid hemorrhage, careful hemostasis and meticulous dissection is recommended. In performing CMDS, the surgeon should remain strictly within the midline when dissecting to the trachea in the neck. After entering the mediastinum, when in doubt, needle aspiration should be performed to preclude biopsy of a vessel. Excessive traction of the tissue to be sampled should be avoided. If the anatomy of the mediastinum has been distorted to such an extent that the vessels and trachea cannot be discerned with confidence, it would be preferable to abandon further attempts at biopsy. If CMDS does not provide an adequate tissue sample, AMDT can be performed in the same surgical session. If in spite of all these preventive measures hemorrhage occurs, control can be achieved by exploration. The incision for most appropriate exposure is determined by the site of the injured vessel and the status of the mediastinum.
Central airway obstruction due to compression of the trachea was a serious complication that arose in two young patients of our series during induction to anesthesia. It is interesting that no such complication was recorded in the 13 publications of this review, although Mineo and associates16 reported "episodes of mild bronchospasm, which were controlled by steroid injection." However, Ferrari and Bedford29 encountered central airway problems in 9 of 44 children with anterior mediastinal masses treated in the Memorial Sloan Kettering Cancer Center: 2 children experienced difficulty on induction of anesthesia and required tracheal intubation with a rigid bronchoscope, airway obstruction developed in 2 children during anesthetic maintenance that was corrected by a change in patient position, 4 children could not be extubated, and 1 child required reintubation in the immediate postoperative period. Similar complications have been reported by others.3031 To circumvent this complication, AMDT under local anesthesia has been advocated as already mentioned.17 We consider that central airway obstruction is an underestimated life-threatening complication that can occur mainly in young patients with anterior-superior mediastinal masses. For this reason, thoracic surgeons should be aware of its diagnosis and treatment. Other complications include laceration of the trachea, recurrent laryngeal nerve injury, cardiac arrythmias, temporary arterial hypotension,16 and wound infection.17
Although there is wide agreement that intrathoracic malignancy has become the commonest cause of SVCO in recent years,15 significant disagreement exists regarding the histology of this malignancy. Ahmann2 reviewed 1,986 patients with SVCO published in the English-language literature from 1934 to 1983 and found that SCLC was the most common cause of SVCO. Callejas et al14 reached the same conclusion in 1991. However, Porte et al17 found that among the patients referred to surgeons, NHL is the most frequent etiology of SVCO together with SCLC. Jahangiri and Goldstraw8 reviewed 34 patients with SVCO and found that 26 of them had lung cancer, including only 5 patients (14.7%) with SCLC. In our series, malignancy was the etiology of SVCO in all patients, although this malignancy was not revealed by CMDS in one patient. The most frequent malignant disease was lymphoma, followed by lung cancer. The great discrepancy regarding the etiology of SVCO seems to be due to patient selection.
The limitations of the present publication are as follows: (1) the relatively long duration of the study period, and (2) the relatively small number of patients submitted to operation each year in our series.
In conclusion, the results of this series and the literature review show that CMDS and AMDT are effective methods to establish histologic diagnosis in patients with SVCO when less invasive procedures have failed. Although mortality is negligible, patients with SVCO compared to those without SVCO undergoing CMDS or AMDT show significantly higher morbidity. Central airway obstruction is a serious complication that could occur mainly in young patients with anterior-superior mediastinal masses. Thoracic surgeons should be aware of diagnosis and treatment of this life-threatening complication, which appears to be underestimated. In this series, the rate of malignant diagnosis was significantly higher among patients with SVCO than those without SVCO.
| Footnotes |
|---|
Received for publication December 23, 2004. Accepted for publication March 28, 2005.
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |