Pulsion Intubation Versus Traction Intubation for Obstructing Carcinomas of the Esophagus

Transcription

1 Pulsion Intubation Versus Traction Intubation for Obstructing Carcinomas of the Esophagus Helmut W. Unruh, M.D., F.R.C.S.(C), and K. Michael Pagliero, M.B.B.S., F.R.C.S. ABSTRACT Eighty-eight patients were intubated to relieve the dysphagia of malignant esophageal obstruction. Because of advanced metastatic disease or poor general condition, 49 patients could not be operated on and were intubated endoscopically. Under radiographic control, the Nottingham Introducer was used to position Atkinson or modified Celestin tubes. In 39 other patients, palliative resection could not be done or liver metastasis was found at preliminary exploration with a view to esophagectomy. In these patients, Celestin tubes were inserted by the traction technique. The pulsion intubation group was older than the traction intubation group but comparable in other respects. Hospital stay was significantly shorter for the pulsion intubation group (8.4 versus 18.6 days). Hospital mortality rates were comparable (14.3% in the pulsion intubation group versus 23.1% in the traction intubation group), and survival did not differ (93 days in the pulsion intubation group versus 137 days in the traction intubation group). Overall complication rates were similar, but there were two significant differences: (I) wound infection or dehiscence was a major problem in the traction intubation group, occurring in 23% of the patients; and (2) tube obstruction or displacement occurred more frequently in the pulsion intubation group (18.3% versus 5.1%) but did not constitute a serious problem. Perforation occurred in 4 patients (8.2%) in the pulsion intubation group but was fatal in only 1. Pulsion intubation offers distinct advantages over traction intubation in that hospital stay is decreased and morbidity reduced. Palliation of dysphagia due to malignant obstruction of the esophagus is often required. Methods of palliation include resection, bypass, radiotherapy, dilation, and intubation. Intubation can be achieved by the traction or the pulsion method. Traction intubation requires laparotomy with gastrotomy. Pulsion intubation is achieved endoscopically under radiographic control. We compared the two methods of intubation with regard to hospital stay, early and late mortality, survival times, and complications. From the Department of Surgery, The University of Manitoba Health Sciences Centre. Winnipeg, Man, Canada, and the Thoracic Surgery Unit, Royal Devon and Exeter Hospital (Wonford), Exeter, England. Accepted for publication Jan 17, Address reprint requests to Mr. Pagliero, Thoracic Surgery Unit, Royal Devon and Exeter Hospital (Wonford), Barrack Rd, Exeter EX2 5DW. England. Material and Methods Between the years of 1977 and 1983, 88 patients with malignant esophageal obstruction were treated with esophageal intubation. In each instance, a direct attack on the tumor itself by resection, bypass, or radiotherapy had been ruled out and intubation was employed merely to palliate severe dysphagia and combat associated malaise and weight loss. Diagnosis was established in all patients by prior fiberoptic endoscopy and biopsy. Staging of the disease was accomplished by physical examination and biochemical investigation including liver function tests which, when abnormal, were followed up with radionuclide liver scan, ultrasonic liver scan, or computed axial tomography. In general, localized tumors of the esophagus were treated with resection. In a few instances in which the patient was elderly or there was an associated medical condition that precluded operation, or to satisfy the patient's wishes, the tumor was treated with radical radiotherapy primarily. The group of patients under discussion in this article includes those who were thought to be unsuitable for either method of treatment by virtue of their general condition or the presence of distant metastases. These 88 patients fell into two distinct groups. The first group included 49 patients in whom the factors that determined need for palliative esophageal intubation were obvious without laparotomy. Palliation was achieved in these patients with pulsion intubation with the aid of fiberoptic endoscopy. The second group consisted of 39 patients in whom clinical assessment had suggested that the primary tumor would be resectable. Subsequent laparotomy, however, had shown that advanced local or metastatic disease had made palliative resection impossible or futile. Therefore, these patients were palliated with traction intubation during laparotomy using Celestin tubes (11 (Fig l).* Pulsion intubation is performed under general anesthesia and radiographic control. After the tumor is viewed with the endoscope, a stainless steel guidewire is passed through the narrowed lumen into the stomach, where its position is confirmed by fluoroscopic examination. A 45F bougie is then passed over the guidewire to dilate the obstruction. The endoscope is reintroduced to visualize the lower extent of the tumor so that an Atkinsont [2] or Celestin tube of appropriate length can be selected. The Atkinson tube (Fig 2) is made of latex rub- 'Celestin tubes:medoc (Glos) Ltd, PO Box 1, Tetbury, Gloucester CU) 8TL. England. 'Key Med Inc., 400 Airport Executive Park, Spring Valley, NY lw

2 338 The Annals of Thoracic Surgery Vol 40 No 4 October 1985 Fig 2. Atkinson tube. Fig 1. Traction Celestin tube. ber with a ridge at its lower end that discourages upward dislocation after insertion. The collecting funnel is fairly stiff at its narrow end to discourage passage through the tumor but is progressively softer toward its rim so that it molds gently with the contours of the esophagus without any tendency for pressure necrosis. Alternatively, a Celestin pulsion tube may be used (Fig 3). It is similar to that used for traction intubation but tries to overcome the disadvantage of not being sutured in place by the addition of a flange at its lower end. Eighteen Celestin flanged latex rubber tubes and 31 Atkinson silicon rubber tubes were used. Both tubes had similar cross-sectional diameters (Celestin tubes: external diameter, 16 mm, internal diameter, 12 mm; Atkinson tubes: external diameter, 16 mm, internal diameter, 11 mm). The tube is mounted on the Nottingham Introducer so that the nose cone of the Introducer just protrudes from its lower end. The plastic nose cone (Fig 4A) is made to expand by advancing a conical metal olive so that it grips the tube firmly from within. The Nottingham Introducer apparatus* consists of an inner rod to which the plastic nose cone is attached and an outer tube to which the Key Med Inc., 400 Airport Executive Park, Spring Valley, NY Fig 3. Pulsion Celestin tube.

3 339 Unruh and Pagliero: Pulsion Versus Traction Intubation A B Fig 4. Nottingham Introducer. (A) The apparatus consists of an inner rod to which a plastic nose cone is attached and an outer tube to which a metal olive is attached. (6) Mechanism at the proximal end of the device enables advancing or retarding the olive. metal olive is attached. The metal olive can be advanced and retarded by the mechanism shown in Figure 4B at the proximal end. After the Atkinson tube is mounted on the Nottingham Introducer, a plastic ramrod is passed over the proximal end of the apparatus so that its end comes to lie within the cup of the Atkinson tube. The whole apparatus is then passed over the guidewire (Fig 5) until the tube is positioned appropriately within the lumen of the malignant tumor as determined by roentgenographic control. When the cup of the tube is sited immediately above the tumor, the nose cone is relaxed and the Introducer is withdrawn, maintaining the position of the tube with the ramrod. Once this is achieved, the ramrod and guidewire are removed. Finally, the satisfactory position is confirmed by repeating the endoscopy to ensure that the funnel is above the tumor and that the lower end is distal to the site of obstruction. Postoperatively, the head of each patient s bed was slightly elevated. A liquid diet was begun after recovery from anesthesia; this was rapidly increased to a regular diet. The patients were all instructed to chew their food Fig 5. Insertion of tube by the pulsion method.

4 340 The Annals of Thoracic Surgery Vol 40 No 4 October 1985 well and to drink adequate liquids to avoid bolus obstruction of the tube. They were also advised that severe symptomatic reflux of gastric contents was likely to occur, especially if the gastroesophageal junction had been traversed. They were therefore instructed to sleep with the head of the bed elevated and to avoid eating before going to bed. If required, antacids were prescribed for symptomatic relief. The second group of 39 patients all underwent laparotomy as a prelude to esophagectomy, and the extent of intraabdominal disease was carefully assessed. Lymph node involvement with metastatic carcinoma was not considered an absolute contraindication to palliative resection. Thus, all 39 patients had either liver metastasis or localized extensive tumor infiltration that precluded resection, and were therefore candidates for surgical palliation by traction intubation. A small high anterior gastrotomy was performed and a pilot bougie was passed from the mouth through the narrowed lumen of the tumor into the stomach. In a few patients, this proved impossible and it was necessary to pass a fine gumelastic bougie through the malignant stricture from below by means of the anterior gastrotomy; the elastic bougie was then used to guide the pilot bougie downward. A Celestin tube [l] (external diameter, 14 mm; internal diameter, 11 mm) was positioned by the traction technique and secured to the anterior gastric wall by a single suture tied over a small piece of felt. The gastrotomy was closed in standard fashion. The correct position of the tube was confirmed by fiberoptic endoscopy. After the return of bowel sounds, the postoperative management was similar to that described for the endoscopically inserted tubes. Table 1. Patient Data Variable No. of patients Age (yr) Sex Male Female Diagnosis Squamous cell carcinoma Adenocarcinoma Other Hospital stay (d) Hospital mortality Days until hospital mortality No. of deaths within 2 weeks of discharge Survival excluding early deaths 'Significance: p < hsignificance: p < Pulsion Intubation Group (14.3%) Traction lntubation Group " I 18.6b 9 (23.1%) Results The patient data are summarized in Table 1. The two groups were comparable with regard to size, sex ratio, and diagnosis. The patients in the pulsion intubation group were significantly older, however. This is explained by the selection process for the study, whereby the very elderly and individuals in poor general condition constituted a major part of the pulsion group. Hospital stay was significantly longer for the patients having traction intubation (18.6 days versus 8.4 days in the pulsion group) but hospital mortality, death within 2 weeks of discharge from the hospital, and survival in days were not significantly different. Among the 37 patients with pulsion intubation and the 30 with traction intubation who survived beyond the first 2 weeks after discharge, the average survival was 93 and 137 days, respectively. From Figure 6 it is apparent that in fact all but 1 patient had died within 12 months. The complications of both types of intubation are presented in Table 2. The total number of complications do not differ significantly between the two groups. Perforation recognized at the time of intubation or subsequently by the development of surgical emphysema occurred in 4 patients having pulsion intubation and was fatal in 1. One fatal perforation occurred in the traction intubation f 40- t- rn w m I- a io- W &--,.---*---, I Fig 6. The percentage of patients surviving is plotted against months of intubation. All patients having pulsion intubation died by 8 months, and only 1 patient having traction intubation survived longer than 12 months. group, but total perforation rates were not significantly different. The incidences of inability to intubate, development of aspiration pneumonia, and no symptomatic improvement were similar for both groups. Bolus obstruction of the tube required readmission to the hospital for bougienage, while tube displacement required repositioning or removal of the tube. The incidence of bolus obstruction was similar in both groups, but tube displacement occurred more frequently in the pulsion intubation group.

5 341 Unruh and Pagliero: Pulsion Versus Traction Intubation TaOle 2. Complications of lntubafiona Pulsion Traction lntubation lntubation Group Group Complication (N = 49) (N = 39) Perforation 4 (8.2) 1 (2.6) Nonfatal 3 0 Fatal 1 1 Failed insertion 3 (6.1) 1 (2.6) Wound-related problems Infection 0 7 (17.9)' Dehiscence 0 2 (5.1) Aspiration pneumonia 3 (6.1) 1 (2.6) Tube bolus obstruction 3 (6.1) 2 (5.1) Tube displacement 6 (12.2)b 0 No symptomatic improvement 2 (4.1) 1 (2.6) Overall complication rate 21 (42.9) 15 (38.5) "Values in parentheses are percentages hsignificance: p < 'Significance: 1' < There were 7 wound infections and 2 wound disruptions in the traction intubation group. Thus, the incidence of wound complications was 23%. More than 60% of patients with carcinoma of the esophagus will prove to have inoperable disease, and only 18% will survive one year 131. Nevertheless, palliation of the distressing symptom of dysphagia is often required. Resection gives the best palliation for dysphagia but may not be appropriate in a certain proportion of patients for reasons of poor general condition, advanced metastasis, or extensive local infiltration. Radiotherapy is effective in relieving dysphagia in only half the patients after a 4- to 6-week course of treatment and is often only temporary 141. Chemotherapy has no role in relieving dysphagia. Retrosternal gastric bypass for unresectable carcinomas [5] was compared with pulsion intubation by Angorn and Haffejee [6]. They found bypass to have a higher complication rate and significantly more postoperative catabolism. Therefore, bypass was not favored since the life expectancy of these patients is often very limited. Simple dilation may be the only procedure tolerated by a very sick patient, but symptomatic relief is unpredictable and often limited [7, 81. All of our patients had malignant esophageal obstruction with severe dysphagia requiring palliation. Eightyone patients had primary carcinoma of the esophagus or cardia, and the other 7 had obstruction due to other malignancies. In 4 of these patients, obstruction was caused by breast carcinoma metastatic to periesophageal lymph nodes. The other 3 cases of obstruction were due to bronchial carcinoma that had involved the esophagus by direct infiltration. One of the 3 patients had a malignant tracheoesophageal fistula that was successfully controlled by traction intubation, and the patient went on to survive 9 months. The patients having pulsion intubation were significantly older than those having traction intubation. This reflects the selection process in that the very elderly and more infirm were chosen for the pulsion group in addition to those with preoperatively proven hepatic metastases. Despite this, hospital mortality figures were similar. There were 5 deaths within 2 weeks of discharge from the hospital among the patients having pulsion intubation compared with none in those having traction intubation. This does not, however, constitute a statistically significant difference, and is also most likely a reflection of the selection process. The survival times of the patients who lived beyond 2 weeks after discharge did not differ significantly despite the older age and poorer general condition of those intubated endoscopically. However, the duration of stay in the hospital was significantly shorter for the patients having pulsion intubation. This not only has important implications with regard to hospital costs but is also vitally important in view of the limited life expectancy for these patients. Failure to insert a tube at endoscopy is due to inability to pass the guidewire through the tumor, which occurred in 3 patients. In 1 patient undergoing traction intubation, it proved impossible to negotiate the stricture with a bougie. The results of esophageal intubation are very rewarding. Only 2 patients having pulsion intubation and 1 patient having traction intubation thought there was no symptomatic improvement despite patent tubes. Bolus obstruction by food is a relatively infrequent problem and can usually be resolved if the patient drinks aerated fluids. If hospital admission is necessary, the first procedure is to pass a nasogastric tube. Failing this, endoscopy is required (91. Should the tube become displaced, its removal and replacement is a simple matter [lo]. The guidewire is passed through the tube, and the Introducer is positioned in the tube with radiographic assistance. Once secure, the tube can be removed, resited, or replaced. The perforation rate with endoscopic intubation was 8.2%. This is comparable with that currently reported for Britain (111 and other parts of Europe [12]. Perforation is not synonymous with disaster. Often this problem manifests itself only as a slight amount of surgical emphysema, as it did in 3 of 4 of our patients with pulsion intubation, and is manageable with conservative measures because the tube usually bypasses the breach in the esophageal wall. Wound infection or dehiscence occurred in 23% of the patients having traction intubation. Such complications undoubtedly contributed to the longer hospital stay of these patients and are therefore a major disadvantage of this treatment. In addition, Angorn [ 131 demonstrated rapid improvement in nutritional status following endoscopic intubation. One would expect improvement to be much delayed after a major surgical procedure. Many patients experienced some degree of symptomatic reflux, but this was manageable by conservative measures, including antacids, and no patient required hospital readmission for this reason alone. However, the

6 342 The Annals of Thoracic Surgery Vol 40 No 4 October cases of aspiration pneumonia were caused by reflux. It is also likely that reflux was at least partially contributory in some, if not all, of the 3 patients who failed to achieve symptomatic improvement with intubation, since obstruction of the tubes was not noted at endos- COPY. Palliative intubation is highly effective in relieving the dysphagia of malignant esophageal obstruction. Because the endoscopically placed tube is not sutured distally to the stomach as is the surgically positioned tube, tube displacement occurs more frequently with pulsion intubation but does not pose major problems for the patient [9, 101. A high incidence of wound infection and dehiscence and a delay in nutritional status improvement (131 detract from the benefits of traction intubation. In addition, the hospital stay for patients having pulsion intubation is shorter, with hospital mortality and subsequent survival comparable to those for traction intubation. Therefore, we advocate pulsion intubation as an effective method for palliation in selected patients with inoperable disease. There will continue to be a group of patients in whom laparotomy is essential for staging and, thus, a continued need for traction intubation. Our results, however, indicate strongly that when a decision to intubate has been made before operation, the procedure should be accomplished by the pulsion method rather than by traction. References 1. Celestin LR: Permanent intubation in inoperable cancer of the esophagus and cardia. Ann R Coll Surg Engl 25:165, Atkinson M, Ferguson R, Parker GC: Tube introducer and modified Celestin tube for use in palliative intubation of oesophagogastric neoplasm at fibreoptic endoscopy. Gut 19:669, Earlam R, Cunha-Melo JR: Oesophageal squamous carcinoma: I. A critical review of surgery. Br J Surg 67381, Wilson SE, Plested WG, Carey JS: Esophagogastrectomy versus radiation therapy for midesophageal carcinoma. Ann Thorac Surg 10:195, Orringer MB, Sloan H: Substernal gastric bypass of the excluded thoracic esophagus for the palliation of esophageal carcinoma. J Thorac Cardiovasc Surg , Angorn IB, Haffejee AA: Pulsion intubation v. retrosternal gastric bypass for palliation of unresectable carcinoma of the upper thoracic esophagus. Br J Surg 70335, Belsey RHR: Palliative management of esophageal carcinoma. Ann Surg 139:789, Heit HA, Johnson LF, Siege1 SR, Boyce HW: Palliative dilatation for dysphagia in esophageal carcinoma. Ann Intern Med 89:629, Earlam R, Cunha-Melo JR Malignant esophageal strictures: a review of techniques for palliative intubation. Br J Surg 69:61, Ogilvie AL, Dronfield MW, Ferguson R, Atkinson M: Palliative intubation of the oesophagogastric neoplasms at fibreoptic endoscopy. Gut 23:1060, Bennett JR Intubation of gastro-oesophageal malignancies: a survey of current practice in Britain, Gut 22:336, Tytgat GN: Endoscopic methods of treatment of gastrointestinal and biliary stenosis. Endoscopy 1980 [Suppl], pp Angorn IB. Intubation in the treatment of carcinoma of the esophagus. World J Surg 5:535, 1981