Portal de Periódicos da CAPES

Esophageal Rupture Associated with the Use of the Combitube[trademark symbol]

1997; Lippincott Williams & Wilkins; Volume: 85; Issue: 4 Linguagem: Inglês

10.1097/00000539-199710000-00040

1526-7598

H-J Klein, M. E. R. Williamson, H M Sue-Ling, M. Vucevic, Audrey Quinn,

Restraint-Related Deaths

In many parts of the world, the Combitube[trademark symbol] (Sheridan Catheter Corp., Argyle, NY) is used for difficult airway management [1,2] and for aiding airway management during cardiorespiratory arrest, including those managed by nonanesthetic personnel [3]. Similarly, in the United Kingdom, the laryngeal mask airway (LMA) is a useful device for airway management by nonanesthetic staff during cardiac arrest [4]. The LMA consists of a flexible curved tube opening at the distal end into the lumen of an elliptical mask that has an outer inflatable rim. Proximally, the tube is joined to a standard 15-mm connector [5]. The LMA fills a niche between the face mask and the endotracheal tube in terms of anatomical location and invasiveness. Once placed, it sits with its tip in the hypopharynx, between the respiratory and gastrointestinal tracts, and forms a circumferential lowpressure seal around the glottis [5]. Because the LMA does not protect from aspiration of gastric contents [5], the Combitube[trademark symbol] esophageal airway may offer an advantage over the LMA during cardiopulmonary resuscitation. We assessed the ease of insertion and effectiveness of ventilation of the Combitube[trademark symbol] in comparison with the laryngeal mask when used by anesthetic staff in a controlled, randomized, prospective study. Because the Combitube[trademark symbol] is not widely used in the United Kingdom, our first step was to familiarize ourselves with inserting the airway. The incident in this case report occurred during our familiarization with the Combitube[trademark symbol]. This trial had been approved by our local research ethics committee. Case Report After eight uncomplicated placements of the Combitube[trademark symbol] during elective anesthesia by the same operator, our ninth patient was a 60-yr-old, 80-kg man scheduled for a routine check cystoscopy for bladder tumor. We obtained written, informed consent from the patient for insertion of the Combitube[trademark symbol] under general anesthesia. The patient's medical history revealed a noninvasive, papillary, transitional cell carcinoma of the bladder 4 yr previously. The patient had undergone 12 routine check cystoscopies for the screening and treatment of recurrent tumors. Eight of these general anesthetics had been performed using a LMA, three had been performed using an endotracheal tube, and one had been performed under spinal anesthesia. The patient was otherwise healthy and denied any other medical problems. Although he did not have symptoms of dyspepsia or reflux, he was taking ranitidine 150 mg bid. A standard anesthetic was administered as described per protocol. Usual monitoring was used. Premedication was not administered. Anesthesia was induced with fentanyl 1.5 micro g/kg, propofol 1.5-2.5 mg/kg, and mivacurium 0.15 mg/kg intravenously followed by 70% N2 O in oxygen with 1%-2% isoflurane to maintain anesthesia. After induction of anesthesia, manual ventilation was performed with bag and mask for approximately 3 min to allow time for satisfactory muscle relaxation before inserting the Combitube[trademark symbol]. The patient's mouth was opened with the operator's left hand. The mandible and tongue were moved forward and downward, and the tube was blindly advanced to the pharynx using the right hand. Three attempts were required to place the tube. The first and second attempt were abandoned because increased resistance was felt after advancing the tube 10-12 cm from the teeth. It was presumed that the tube was caught in pharyngeal structures. The patient's head position was changed, and the neck was further flexed [6]. On the third attempt, the tube was advanced easily until the marking lines were positioned between the front teeth. After inflation of the two pilot balloons (100 mL in proximal pharyngeal balloon and 15 mL in distal balloon, resulting in self-placement of the tube), it was impossible to ventilate the lungs through either lumen of the Combitube[trademark symbol]. Clear secretions (estimated 10 mL) then appeared in the esophageal lumen of the tube and were suctioned via a 8-French gauge suction catheter. The balloons were deflated, the tube was withdrawn, and manual ventilation of the lungs commenced easily with a bag and mask. We then inserted a LMA, but ventilation of the lungs still proved difficult, so bag-and-mask ventilation was reinstated. At this stage, approximately 10 min postinduction, the myoneural block was inadequate; therefore, 0.5 mg/kg atracurium was given, and a cuffed oral endotracheal tube was easily inserted by direct laryngoscopy. Positive pressure ventilation of the lungs proved easy, and the cystoscopy was performed uneventfully. Throughout these procedures, the patient remained well oxygenated and hemodynamically stable. At the end of surgery, neuromuscular block was antagonized using a combination of glycopyrrolate 0.5 mg and neostigmine 2.5 mg. On emerging from anesthesia, the patient complained of severe chest pain, and his arterial blood pressure decreased to 85/48 mm Hg. The oxygen saturation decreased to 92%, and he quickly developed surgical emphysema at the left lateral aspect of his neck. The oxygen flow rate was increased to 10 L/min, maintaining oxygen saturation >95%, an IV infusion of normal saline was started, and boluses of IV morphine were titrated for analgesia (10 mg given). The electrocardiogram was normal, but the chest radiogram showed mediastinal air. A water-soluble contrast swallow was arranged, which showed a contrast dye leak into the mediastinum, but the site of leakage was unclear. At this point, IV cefotaxime 1 g and metronidazole 500 mg and ranitidine 50 mg IV were administered. The patient was reanesthetized, and an esophagoscopic examination confirmed the presence of an esophageal perforation in the mid-esophagus. Invasive monitoring (central venous pressure, arterial line), a double-lumen endotracheal tube, and a thoracic epidural catheter were then inserted before a right lateral thoracotomy was performed. After mobilizing the esophagus, a 5-cm longitudinal tear with local mediastinal soiling was seen in the left posterolateral esophagus starting at 25 cm from the teeth. Otherwise, the esophagus looked normal, with no evidence of preexisting abnormality, and it was repaired in a longitudinal plane, followed by thorough mediastinal irrigation. Right basal and apical chest drains were inserted. Ventilation was controlled overnight. The patient's postoperative pain was well controlled with an epidural infusion of 0.1% bupivacaine and fentanyl 2 micro g/mL: IV antibiotics were continued, and omeprazole 42.6 mg once daily was begun. Total parenteral nutrition was started because he was allowed nothing by mouth until Day 7, when a gastrograffin swallow showed no evidence of leak from the site of the repair. His postoperative course was uneventful. Oral fluids were introduced on the 10th postoperative day, and he was discharged home 16 days after his operation. Discussion The esophagotracheal double-lumen Combitube[trademark symbol] is a development of the single-lumen Gordon-DonMichael esophageal obturator airway [7]. It is inserted blindly through the mouth, which results in placement in either the trachea or esophagus. Its position is confirmed after inflation of two pilot balloons by the ability to ventilate the lungs through one of the tubes' lumina. It is designed to be shorter than that of the obturator airway to avoid trauma to the esophagus. Contraindications to its use include active laryngeal or pharyngeal reflexes, known esophageal trauma, esophageal pathology, or ingestion of corrosive agents. A literature review of the complications of the obturator airway in 1980 reported at least 19 cases of esophageal perforations [8]. In their own retrospective analysis of 1058 cardiac arrests managed with the obturator airway, the authors documented four cases of perforation (only one survivor) and suggested that this number of cases reflects a small proportion of the actual incidence, because no postmortems had been routinely performed after use of the obturator airway in cardiac arrest in any of the studies. The mechanism of esophageal rupture by the esophageal obturator airway is not entirely clear, and several hypotheses have been proposed, including direct trauma to the esophageal mucosa by the tube itself during insertion or by a suction catheter or buildup of pressure in the stomach and esophagus against an obstruction, resulting in wall rupture. Severe retching against a tube in the esophagus could occur in patients because of incomplete myoneural blockade. In addition, increased pressure in the esophagus caused by excessive inflation of the distal balloon could result in splitting of the longitudinal esophageal smooth muscle. We believe the first hypothesis to be the most likely, but we would have expected the tear to be more proximal, because the length of the Combitube[trademark symbol] measured at the black line, which was correctly positioned at the teeth, was at the most 22 cm. The tear, however, started at 25 cm. An analysis of the reported cases of esophageal rupture by the obturator airway [8] showed that most occurred distal to the tip of the obturator tube in the distal third of the esophagus posteriorly and laterally. This area is the weakest part of the esophagus [9,10]. The lacerations are classically longitudinal due to the dominance of longitudinal muscles over the circular layer. We believe it unlikely that the 5-cm longitudinal esophageal tear could have been caused by the suction catheter. The second hypothesis may also be relevant in this case. Incomplete myoneural block is a possibility because although we administered mivacurium, we did not monitor the depth of myoneural block. Additionally, in this case, the presence of gastric secretions, which are incompressible, would have further increased the wall tension according to Laplace's law (P = 2T/R) where k depends on geometric shape and predisposed the esophagus to rupture. Our patient was known to have a hiatal hernia, which had been demonstrated on a barium meal examination 4 yr earlier. However, when he consented to the procedure, the patient did not have any active symptoms of reflux or esophageal disease. Known esophageal disease is a contraindication to the use of the Combitube[trademark symbol] [6]. In retrospect, it is debatable whether this patient should have been selected for inclusion into the trial, but the incidence of recognizable hiatus hernia increases with age; it is as great as 30% in the general population and 69% in those more than 70 yr of age. We propose that this case of esophageal rupture resulted from direct trauma by the Combitube[trademark symbol] airway or from increased intraluminal pressure distal to the tube, with poor patient selection as a contributory factor. This case highlights a serious complication of the new esophageal airway instrument, despite design alterations from the original esophageal obturator intended to reduce the incidence of esophageal perforation. The incidence of esophageal rupture by the Combitube[trademark symbol] has not been fully evaluated. We suggest that in cases in which the Combitube[trademark symbol] is to be used, the myoneural block should be monitored, patients should be carefully selected, and all personnel should be familiar with the use and complications of this device.