Thoracoscopic esophagectomy: Technique and initial results

Thoracoscopic esophagectomy: Technique and initial results

Session IV: Esophageal Disease Thoracoscopic Esophagectomy: Technique and Initial Results Dominique Gossot, MD, Pierre Fourquier, MD, and Michel Cele...

740KB Sizes 1 Downloads 82 Views

Session IV: Esophageal Disease

Thoracoscopic Esophagectomy: Technique and Initial Results Dominique Gossot, MD, Pierre Fourquier, MD, and Michel Celerier, MD Department of Surgery, Saint-Louis Hospital, Paris, France

To reduce the high morbidity rate associated with esophageal surgery, we have developed a technique of thoracoscopic esophagectomy. A feasibility study was first carried out in an animal model and a specific instrument was developed for this purpose. Esophagectomy using a right thoracoscopic approach was attempted in 15 patients, 13 males and 2 females whose average age was 48 years. Indications consisted of squamous cell carcinoma in 10 patients, adenocarcinoma in 1, and caustic stenosis in 4. We used a technique that consisted of double-lumen tracheal intubation and the creation of five ports. The whole esophagus was mobilized thoracoscopically and the esophagectomy was completed through the abdomen. The reconstruction was achieved using a gastric pull-through, and the anastomosis was made in the neck.

There were three failures: in 1 patient there was a large tumor, making the exposure unsafe, and, in 2 patients, incomplete lung collapse made exposure of the posterior mediastinum difficult. These 3 cases were converted into a thoracotomy. The thoracoscopic dissection was successful in the remaining 12 patients. The average time of the thoracoscopic stage was 125 minutes. The postoperative course was uneventful in 10 patients. Two patients had a left atelectasis. Although our series is limited, these initial results indicate that thoracoscopic esophagectomy is feasible. However, further evaluation of the technique is needed to assess its benefit in terms of respiratory morbidity.

D

plete, making exposure of the mediastinum quite difficult. Unless the esophageal stenosis is too tight, a nasogastric tube is introduced to help grasp the esophagus. The insertion of a flexible endoscope has also been proposed as a method for lifting the esophagus out of the vertebral gutter [3]. The patient is placed in the left lateral position and tilted a little forward to facilitate the posterior mediastinum exposure. The patient’s right arm must be pulled up to leave the axilla free for the eventual insertion of additional trocar tubes. The operating table’s bridge is adjusted to the position appropriate for an eventual thoracotomy, but is not elevated. We have found no advantage to arching the patient during the usual thoracoscopic procedures. A 10-mm, 0-degree Olympus (Lake Success, NY) telescope is introduced into the seventh intercostal space in the midaxillary line. Two lung retractors are introduced in the anterior axillary line, usually in the fourth and sixth intercostal space. At least one 10-mm and one 5-mm retractor are required. However, the use of two 10-mm retractors is more efficient. They are held either by an assistant or by an autostatic holder. Time must be taken to achieve correct exposure of the mediastinum. A gentle flattening of the lung with the retractor while the anesthesiologist aspirates the right bronchus helps keep the lung deflated. The esophageal dissection should be started only if a clear view of the mediastinum has been obtained. In the event of oozing or hemorrhage, hemostasis is very difficult to achieve if vision is hindered by the lung. Usually, after a certain

ecreasing the mortality and morbidity rates associated with esophageal surgery remains one of the important goals for surgeons. Avoiding the consequences of thoracotomy in fragile patients might be the first step toward this goal [l].Until recently, however, the transhiatal esophagectomy introduced by Orringer and Sloan [2] was the only alternative to thoracotomy, but it has disadvantages inherent in the blind dissection involved. With the development of minimally invasive surgery, various endoscopic techniques have been proposed for esophagectomy. We describe here the application of the thoracoscopic approach.

Operative Technique The operation is carried out in three stages. The first consists of the thoracoscopic mobilization of the esophagus. The second is the construction of a gastroplasty, which is carried out through a laparotomy. The third stage is the creation of the cervical anastomosis between the esophagus and the gastric pull-through. These last two stages are well known, and therefore only the thoracoscopic dissection is described here. The preparation and draping are done as for a usual open procedure. The double-lumen tracheal tube has to be placed perfectly, otherwise the lung collapse is incomPresented at The First International Symposium on Thoracoscopic Surgery, San Antonio, TX, Jan 22-23, 1993. Address reprint requests to Dr Gossot, Department of Surgery, Hbpital Saint-Louis, 1 Ave CI.Vellefaux, F-75010 Pans, France.

0 1993 by The Society of Thoracic Surgeons

(Ann Thorac Surg 1993;56:667-70)

0003-4975/93/$6.00

668

THORACOSCOFY GOSSOT ET AL THORACOSCOFIC ESOPHAGECTOMY

amount of time, the lung remains retracted by itself, sometimes without the need for retractors. Once exposure of the posterior mediastinum is correct, ie, when the esophagus can be clearly identified, two additional 10-mm ports are introduced, one in the fifth intercostal space in the posterior axillary line, or even more posteriorly to accommodate an esophageal grasping forceps, and one in the sixth intercostal space in the anterior axillary line for the introduction of dissecting instruments and a clip applier. It is more convenient to use only 10-mm trocar tubes, allowing the surgeon the opportunity to easily change the position of the instruments or the telescope during the procedure. A grasping forceps is introduced through the posterior trocar tube. The mediastinal pleura is grasped, pulled upward, and opened with scissors. The pleural incision is then continued from the bottom to the top, with or without cautery, depending on the degree of mediastinal inflammation. The lateral sides of the esophagus are loosened using blunt-tipped scissors or a dissector. During the dissection, all nodes encountered are removed. However, until now, we have not performed an extensive lymphadenectomy . Once they are partly freed, the esophagus and the surrounding mediastinal tissues are grasped with an esophageal forceps, which has been specially designed for this purpose (Olympus), and pulled upward and backward. The forceps can be shifted along the length of the esophagus, and this is thus more convenient than a loop passed around the esophageal body, which cannot be moved easily. Variable-curvature dissecting spatulas, as described by Cushieri and associates [3], are also useful for dissection of the mediastinal side of the esophagus. Hemostasis of bleeding esophageal vessels is achieved with cautery and clips. The use of curved scissors and curved forceps is very helpful for the dissection of the mediastinal side of the esophagus. This stage of the procedure is sometimes made difficult by a minor but permanent oozing, which requires frequent suction. To avoid an awkward inflation of the lung during suction, the aspiration periods must be short. Keeping one or two trocar tubes empty is also an efficient solution, allowing a permanent suction device to be kept in place. Once the esophagus has been mobilized up to its upper third, the azygos vein must be divided. Although some authors do not find it necessary to divide the vein [3], we believe that this maneuver makes the freeing up of the upper third of the esophagus easier and safer. Before starting this new step, one must make sure that the instruments can easily reach the upper part of the chest so they can be maneuvered safely, and that the vision is not too tangential. This may pose a problem in endomorphic patients. In this event, the 0-degree telescope must be exchanged for a 30-degree one, or it must be introduced into the axilla, thus affording a direct vision of the azygos. The blunt-tipped scissors are gently slipped under the mediastinal pleura, which is divided up to the top of the chest. The back side of the vein is dissected using scissors and a dissector. Before engaging the endostapler, one must be sure that

Ann Thorac Surg 1993;56:667-70

the dissector tip is visible at the superior edge of the vein and that 2 cm of the azygos is totally dissected. The vein is then divided using the vascular endostapler, which must be introduced perpendicular to the vein. To pass the stapler smoothly behind the azygos, the vein can be lifted up either with the dissector or with a loop. Once sure that the vein is completely between the stapler’s jaws, the stapler can be fired. The division of the azygos vein allows one to end the dissection of the upper third of the esophagus. Using the instruments in the locations already described, only the upper two-thirds of the esophagus can be dissected. For dissection of the lower third, the telescope and the instrument direction must be shifted 180 degrees. For this stage, the operator should move to the patient’s front, making maneuvering of the instruments easier. The esophageal dissection is then continued downward to the diaphragm. The dissection of the lower third can be left incomplete if the esophagoplasty stage is conducted through a laparotomy, because the lower part of the esophagus is easy to mobilize through a laparotomy. However, if the esophagoplasty is performed laparoscopically, it is easier to dissect the esophagus as far as possible using thoracoscopy. Once the esophagus is totally free, the esophagectomy can be ended, either through the cervicotomy or through the abdomen. A chest tube is placed into the esophageal bed. The patient is then repositioned supine for the laparotomy and cervicotomy. The esophageal reconstruction is made using either colon or stomach, depending on the surgeon’s choice and the respective possibilities. A gastroplasty can be performed by means of laparoscopy, as described by Dallemagne and associates [4]. However, this technique is time-consuming and its benefit, compared with the open technique, has not yet been demonstrated.

Initial Results We have attempted 15 esophagectomies using a right thoracoscopic approach in 12 males and 3 females whose average age was 48 years. Indications for its use were squamous cell carcinoma in 10 patients, adenocarcinoma in 1, and caustic stenosis in 4. The whole esophagus was mobilized thoracoscopically and the esophagectomy was completed through the abdomen. The reconstruction was achieved using a gastric pull-through performed through a laparotomy. The anastomosis was made in the neck. There were 3 failures. In 1 patient, a large tumor made a thoracoscopic dissection unsafe because the dissection plane between the esophagus and the left bronchus was difficult to find. In 2 patients, incomplete lung collapse made exposure of the posterior mediastinum difficult. In these 3 cases, the procedure was converted to a thoracotomy. The thoracoscopic dissection was successful in the remaining 12 patients. The average time of the thoracoscopic dissection was 125 minutes and the average blood loss was 200 mL. The postoperative course was uneventful in 10 of these 12 patients and their mean hospital stay was 8 days. Two patients had a left atelectasis requiring

Ann Thorac Surg 1993;56:667-70

prolonged ventilation. Their hospital stay was 21 and 33 days.

Comment Until recently, there were only two ways to perform an esophagectomy, either through a thoracotomy or without thoracotomy (EWT) using blunt dissection. The respiratory morbidity associated with an open esophagectomy is high, ranging from 6% to 10%. This high morbidity is partly responsible for the 6% to 15% mortality rate observed for esophagectomy. Many techniques of EWT have been described since the publication of the report of Orringer and Sloan [2]. In some particular indications, such as caustic necrosis, a blunt stripping can be performed easily without the danger or risk of hemorrhage [6]. However, in most cases, this blind technique does not allow sufficient control and the middle third of the esophagus remains hidden, even when using a large phrenotO ~ 171. Y Although the blood vessels to the esophagus are small [8], there is an indisputable hazard of hemorrhage. In Orringer’s report, the average blood loss was 900 mL [9]. In a review of the literature, Liebermann-Meffert and co-workers [8] noted a fatal hemorrhage rate of 1.6% (most often resulting from an injury to the azygos vein) and a tracheal tear rate of 3%.Above all, the advantage in terms of respiratory morbidity has not been demonstrated. In a series of 304 patients who underwent EWT collected by Perrachia and Bardini [lo], the pulmonary complications consisted of a tracheobronchial tear in 1.5% and pleural effusion in 17.8%; the operative mortality rate was 10.3%. Shahian and associates [ l l ] have reported a respiratory morbidity rate that was higher (although statistically not significant) after EWT than after open esophagectomy. For 3 years, Buess and colleagues [12, 131 have worked to improve the technique of EWT by the use of an “endoscopic microsurgical dissection of the esophagus” performed using mediastinoscopy. In their technique, the operating mediastinoscope is introduced into the posterior mediastinum through a left cervicotomy. The mediastinoscope has a central working channel and its optic offers a two to four times enlarged image, thus allowing accurate dissection of vessels and nerves. The esophagus is dissected downward and removed transhiatally through a laparotomy. After having shown this method to be efficient and safe in an animal model [12], Buess and associates [13] reported on a series of 17 cases in which there was no operative mortality and minimal blood loss (<200 mL). Endoscopic esophagectomy is in its first stages of development and it is still too early to know the benefits in terms of morbidity and survival. Some questions need to be answered in the near future.

Will Respiratoy Morbidity Be Decreased by Avoiding Thoracotomy ? In the patients in our series, the respiratory complications have not totally disappeared. We have had 2 patients with

THORACOSCOPY GOSSOT ET AL THORACOSCOPIC ESOPHAGECTOMY

669

serious left atelectasis. This may be due to the prolonged one-lung ventilation. Although some authors have claimed that one-lung ventilation is not mandatory, we find it very helpful to have the right lung collapsed. However, one-lung ventilation may give rise to respiratory disorders such as atelectasis, not only in the upper lung but also in the dependent lung because of its lateral position [14]. It has been proposed that positive endexpiratory pressure should be added to the dependent lung, but this can promote an increase in the shunt flow by diverting the blood flow to the upper lung. This drawback can be partly diminished by adding continuous positive airway pressure to the nondependent (upper) lung. However, continuous positive airway pressure induces a lung inflation that can be as troublesome as two-lung ventilation. On the other hand, the mediastinoscopic approach is probably less aggressive with regard to respiratory function, because the patient remains supine and there is no need for selective ventilation [13].

What Are the Respective lndications for the Mediastinoscopic and Thoracoscopic Approach? The mediastinoscopic technique has been demonstrated to be a safe method, especially with regard to the control of perioperative hemorrhage [13]. However, it does not allow a view of the surrounding mediastinal structures, thus making tumor removal almost impossible. Buess and colleagues [13] justify their approach by the fact that the surgical resection of esophageal cancer is primarily palliative. This opinion is shared by other authors [15], but some surgeons have shown a significant benefit conferred by en bloc esophagectomy for early stage (NO and N1) carcinomas [161. Although extensive node dissection has been shown to be feasible through thoracoscopy in animal models [17] as well as in patients, it is not yet known whether such a dissection can really be safely performed.

Is There a Place for Thoracoscopy in Preoperative Staging? As far as lymph node involvement is concerned, no ideal preoperative examination has yet been devised, though endosonography has exhibited a better accuracy than computed tomography in this regard [18]. Tio and coworkers [19] have demonstrated that endoscopic ultrasonography can detect periesophageal nodes in most cases (95%), but the specificity is poor, in that 50% of the nodes prove to be only inflammatory nodes. These investigators have suggested that sonographically guided biopsies be performed to enhance the accuracy of endosonography. Thoracoscopic node picking might yield the best results. Laparoscopic staging has already been shown to be very effective in the management of cancer of the cardia [20], but the procedure is probably more invasive than endosonography. Furthermore, accomplishing complete and precise staging would require a large mobilization of the esophagus. However, thoracoscopy may be indicated in some cases in which the tumoral stenosis is too tight to allow passage of the sonographic probe, or in the event of doubt.

670

THORACOSCOPY GOSSOT ET AL THORACOSCOPIC ESOPHAGECTOMY

Ann Thorac Surg 1993;56:667-70

References 1. Hallfeldt KKJ, Knoeffel WTK, Thetter 0, et al. Respiratory function after thoracic operations. Ann Thorac Surg 1990;50: 6844. 2. Orringer MB, Sloan H. Esophagectomy without thoracotomy. J Thorac Cardiovasc Surg 1978;76643-54. 3. Cushieri A, Shimi S, Banting S. Endoscopic oesophagectomy

through a right thoracoscopic approach. J R Coll Surg Edinb

12. 13. 14.

1992;377-11.

4. Dallemagne B, Weerts JM, Jehaes C, et al. Subtotal esoph-

agectomy by thoracoscopy and laparoscopy. Min Invasive Ther 1992;1:183-5. 5. Akiyama H, Tsurumaru M, Ono Y, Udagawa H, Matsuda M, Kajiyama Y. Transoral esophagectomy. Surg Gynecol Obstet 1991;173:399-400.

6. Gossot D, Sarfati E, Celerier M. Early blunt esophagectomy

in severe caustic burns of the upper digestive tract. Report of 29 cases. J Thorac Cardiovasc Surg 1987;94:188-91. 7. Goldberg M, Freeman J, Gullane PJ, et al. Transhiatal esophagectomy with gastric transposition for pharyngeal malignant disease. J Thorac Cardiovasc Surg 1989;97:32733. 8. Liebermann-Meffert DAI, Luescher URS, Neff UR, et al. Esophagectomy without thoracotomy: is there a risk of intramediastinal bleeding? Ann Surg 1987;206:184-92. 9. Orringer MB, Orringer JS. Esophagectomy without thoracotomy: a dangerous operation? J Thorac Cardiovasc Surg 1983;85:72-80.

10. Peracchia A, Bardini R. Total esophagectomy without thora-

cotomy: results of a European questionnaire (GEEMO). Int Surg 1986;71:171-5. 11. Shahian DM, Neptune WB, Ellis FH, et al. Transthoracic

15.

versus extrathoracic esophagectomy: mortality, morbidity and long-term survival. Ann Thorac Surg 1986;41:237-46. Kipfmiiller K, Naruhn M, Melzer A, et al. Endoscopic microsurgical dissection of the esophagus. Results in an animal model. Surg Endosc 1989;3:63-9. Buess GF, Becker HD, Naruhn MB, et al. Endoscopic esophagectomy without thoracotomy. Probl Gen Surg 1991;8:47846. Barker S, Trivedi N. Anesthesia for thoracoscopy. In: Gossot D, KIeinmann P, Levi JF, eds. Surgical thoracoscopy. Paris: Springer-Verlag, 199219-24. Barbier PA, Becker CD, Wagner HE. Esophageal carcinoma: patient selection for transhiatal esophagectomy. A prospective analysis of 50 consecutive cases. World J Surg 1988;12:

263-9. 16. Skinner DB, Ferguson MK, Soriano A, et al. Selection of

operation for esophageal cancer based on staging. Ann Surg

1986;204:391-400. 17. Gossot D, Ghnassia MD, Debiolles H, et al. Thoracoscopic

dissection of the esophagus: an experimental study. Surg Endosc 1992;6:5941. 18. Ziegler K, Sanft C, Semsch 8, et al. Endosonography is superior to computed tomography in staging tumors of the esophagus and the cardia [Abstract]. Gastroenterology 1988; 94zA51-7. 19. Tio TL, Coene PPL, Luiken GJ, et al. Endosonography in the

clinical staging of oesophago-gastric carcinoma. Gastrointest Endosc 1990;36:2-10. 20. Watt I, Stewart I, Anderson D, et al. Laparoscopy, ultrasound and computed tomography in cancer of the esophagus and gastric cardia: a prospective comparison for detecting intra-abdominal metastases. Br J Surg 1989;761036-9.