Auto–Positive End-expiratory Pressure Masquerading as Loss of Lung Separation during Thoracoscopy

A 72-yr-old woman diagnosed with a 2 × 3-cm right upper lobe lung mass was scheduled for thoracoscopic biopsy and possible wedge resection. She was 150 cm tall and weighed 67.3 kg. Her medical history was significant for hypertension, coronary artery disease, hiatal hernia, chronic renal insufficiency, and chronic obstructive pulmonary disease. Her medications included atenolol, albuterol, ipratropium, montelukast, furosemide, atorvastatin, and famotidine. Physical examination revealed a blood pressure of 141/72 mmHg, a heart rate of 71 beats/min, a respiratory rate of 20 breaths/min, clear lungs, and a normal airway. Preoperative pulmonary function testing demonstrated a forced expiratory volume in 1 s of 0.45 l (44% of predicted), a forced vital capacity of 1.02 l (92% of predicted), a total lung capacity of 3.81 l (97% of predicted), a functional residual capacity of 3.19 l (139% of predicted), and a residual volume of 2.37 l (147% of predicted). Arterial blood gas on room air revealed a pH of 7.41, a partial pressure of carbon dioxide of 43.5 mmHg, and a partial pressure of oxygen of 64.9 mmHg.

General anesthesia was induced with 100 mg propofol, 50 μg fentanyl, and 14 mg cisatracurium. Bag mask ventilation was easy, and inhalational anesthesia was deepened with sevoflurane, which was continued throughout the case in end-tidal concentrations between 1.0% and 1.8%. The patient was intubated with an 8.0-mm-ID single-lumen endotracheal tube to facilitate diagnostic bronchoscopy by the surgeon. The tube was then replaced with a 37-French left-sided double-lumen endotracheal tube (Broncho-Cath; Mallinckrodt Inc., St. Louis, MO). The size was determined by measuring the left mainstem bronchus on the chest computerized tomography scout film. The position of the double-lumen endotracheal tube was confirmed with a fiberoptic bronchoscope, and the patient was repositioned in the left lateral decubitus position. After confirming excellent double-lumen endotracheal tube position through both lumens with the fiberoptic bronchoscope (unimpeded view of the left upper and lower lobe orifices as well as a completely subcarinal bronchial cuff), lung separation was achieved and confirmed using the positive-pressure underwater bubble test7with 1.5 ml air in the bronchial cuff. Using an Aestiva/5 anesthesia machine (Datex-Ohmeda, Helsinki, Finland) in volume control mode, the patient’s lungs were ventilated with 100% oxygen and a tidal volume of 400 ml, a respiratory rate of 6 breaths/min, and an inspiratory:expiratory ratio of 1:2.5 with a resultant peak airway pressure of 40 cm H²O and end-tidal carbon dioxide readings of approximately 47 mmHg. Approximately 20 min after beginning thoracoscopy, the surgeon remarked that the right lung was increasingly impeding surgical exposure. He was concerned that lung separation may have been lost. The patient’s vital signs, capnograph, and airway pressures were unchanged. Adequate lung separation was immediately reconfirmed with the positive-pressure test used earlier.7Because of the patient’s baseline irreversible airflow obstruction, the hypothesis that significant auto-PEEP might be present in the dependent, ventilated lung was considered. Mechanical ventilation was discontinued, and the patient was allowed to exhale for approximately 60 s (fig. 1). At this time, the quality of thoracoscopic visualization in the right hemithorax had returned to its previous level. The ventilator was adjusted to increase the inspiratory:expiratory ratio to 1:3.5, allowing the patient more expiratory time, and the surgery concluded uneventfully.

We believe that this patient with expiratory airflow obstruction developed auto-PEEP in the dependent lung during one-lung ventilation. Auto-PEEP during one-lung ventilation is well described1–4in patients with expiratory airflow obstruction. This patient’s preoperative pulmonary function testing values demonstrated expiratory flow limitation, and we initiated one-lung mechanical ventilation with this in mind. The initial respiratory rate of 6 breaths/min and the inspiratory:expiratory ratio of 1:2.5 resulted in an inspiratory time of 2.84 s and an expiratory time of 7.14 s. Normally, significant amounts of auto-PEEP in the lungs increase airway and intrathoracic pressures and cause a decrease in venous return, cardiac output, and blood pressure. In the current patient, the operative hemithorax was open to atmosphere through trocar sites and served as a “pop off,” allowing the ventilated lung to expand unencumbered and to shift the mediastinum and the deflated right lung into the right hemithorax.

It is difficult to detect auto-PEEP with most anesthesia machines because the proximal position at which airway pressure is measured may not reflect or detect the increased alveolar pressure.1Auto-PEEP can be measured accurately using the end-expiratory occlusion method in a ventilator capable of producing an end-expiratory hold.1–4However, auto-PEEP can be expeditiously recognized and definitely treated using the apnea test described by Myles et al. ,5,8which consists of disconnecting the patient from the ventilator and allowing the lungs to deflate passively to functional residual capacity. As with our patient, the expiratory flow may continue for a considerable time if exhalation is severely obstructed (fig. 1). Auto-PEEP is best managed by using low respiratory rates, long expiratory times, and permissive hypercapnea.9 

A number of techniques have been described to confirm and in fact quantify lung separation.7,10–12We routinely use the positive-pressure or bubble test, which was particularly useful in this setting. The most obvious reason that the operative hemithorax becomes filled with lung tissue after initial deflation is that some ventilation reaches that lung when lung separation is incomplete. A common response (with or without confirming adequate double-lumen endotracheal tube position) is to inject more air into the bronchial cuff. Inflation of more air into the bronchial cuff than is needed to achieve an airtight seal may predispose to cuff herniation, bronchial mucosal ischemia, or even rupture.10,13,14Confirming adequate lung separation definitively using the bubble test enabled us to recognize auto-PEEP quickly and to treat it effectively.