|Home | About | Journals | Submit | Contact Us | Français|
In a patient urgently requiring anaesthesia for surgical investigation of thoracic injury, rapid-sequence induction with one-lung ventilation may be the best option.
A man aged 18 was brought to the accident and emergency department after being stabbed in the back. The knife had been left in situ. On arrival his pulse rate was 130/min blood pressure 114/45 mmHg and respiration rate 16/min. Air entry to the lungs was equal and oxygen saturation was 99% with the patient on oxygen 15 L/min. A kitchen-knife handle was protruding from the right posterior thoracic wall in the paravertebral region at the level of T6. There was said to have been little blood loss at the scene of the injury. Primary and secondary surveys, performed with the patient sitting, identified no further injuries. During his initial assessment he was given 500 mL crystalloid, 1500 mg cefuroxime, 3 mg morphine and 10 mg metaclopramide intravenously. The heart rate fell to 95/min and all other indices remained stable. Chest X-rays were taken (Figure 1). It was decided that the patient should undergo emergency thoracoscopy to exclude vascular injury, ascertain the position of the knife and remove it under direct vision. The patient had normal dentition, a non-protruding mandible, full range of neck and jaw movements and a Mallampati1 score of 1. He had eaten immediately before the assault. After transfer to the operating theatre the right radial artery was cannulated and invasive blood pressure monitoring was started. He was preoxygenated in the left lateral position for 5 min (Figure 2). With the patient supported by theatre personnel in the left lateral position, a modified rapid sequence induction was performed with 20 mg etomidate and 70 mg rocuronium. Direct laryngoscopy revealed a Cormack and Lehane2 grade 1 view, and a size 37 mm left double-lumen endobronchial tube (Bronchocath) was inserted. Two-handed cricoid pressure, as originally described by Sellick3, was maintained until tracheal cuff inflation and satisfactory tube placement had been confirmed by ausculation. Correct endobronchial placement was further verified by the use of a flexible fibreoptic bronchoscope. Anaesthesia was maintained with isoflurane, air and oxygen, plus 2 μg/kg fentanyl. Ventilation of the left lung continued whilst the right lung was deflated. On video-assisted thoracoscopy the knife was found not to have penetrated any major vascular structures or the oesophagus. It was removed without incident. Afterwards a chest drain was inserted and the patient was extubated. He was discharged home the following evening.
For a patient suspected of having a full stomach, rapid-sequence induction of anaesthesia, with preoxygenation, application of cricoid pressure and avoidance of manual inflation of the lungs, is the recommended way to limit the risk of aspiration of stomach contents4. This patient presented a particular challenge because, as well as having a full stomach, he was unable to lie supine, had a possibility of a mediastinal or thoracic injury, and one-lung ventilation was preferable for the planned procedure. Because of the nature of the injury and the lack of a pneumothorax on chest radiography a bronchopleural fistula was thought unlikely.
Rapid-sequence induction is possible in the lateral position, although intubation may take longer, mainly because of operator unfamiliarity5. Placement of a double-lumen tube has been described with a bougie6. One-lung ventilation can be provided with a single-lumen endotracheal tube, either by the use of an endobronchial blocker or by endobronchial placement of a single-lumen tube, although intubation of the left main bronchus can be difficult. The use of an endobronchial blocker has been described during rapid-sequence induction7.
One-lung ventilation is not an absolute requirement for thoracic surgery but improves surgical access for video-assisted thoracoscopy. The left lateral position was perceived to aid the positioning of a left-sided double-lumen tube.
Before proceeding, we discussed several other options for this patient. A gaseous induction was considered, to maintain negative-pressure ventilation in case of a bronchopulmonary fistula, but the risk of aspiration was deemed greater. Other suggestions were the use of a Montreal mattress or a Toronto frame to support the patient in the supine position with the knife remaining in his back. We used rocuronium instead of suxamethonium, to allow more time for manipulation of the airway.
We thank Michael Whitehorn, chief clinical perfusionist, King's College Hospital, for the images.