LAUSR

Idiopathic pulmonary fibrosis (IPF) is a progressive life-threatening disease. Acute exacerbation of IPF is occasionally observed after thoracic surgery. Prolonged ventilation at a high tidal volume with a high FiO2 might be an intraoperative risk factor that triggers acute exacerbation of IPF. Moreover, a sustained FiO2 of 0.6 also may cause acute lung injury. As the mortality rate of patients with acute exacerbation of IPF is high, it is necessary to manage FiO2 carefully in such patients. On the other hand, a high FiO2 may be required to prevent hypoxaemia during one-lung ventilation (OLV). In this case report we describe the use of near-infrared spectroscopy (NIRS) to monitor brain oxygenation while avoiding a high FiO2 during OLV in a patient with IPF. A 75-year-old man was scheduled for left upper lobectomy for lung cancer. His medical history included IPF and chronic obstructive pulmonary disease. He had been receiving at-home oxygen therapy (2 l/min by nasal prongs) because he occasionally developed hypoxaemia (percutaneous oxygen saturation [SpO2]: 85%– 90%) during normal daily activities. His preoperative chest computed tomography scan showed severe emphysematous changes and honeycombing in both lung bases. The left upper lobe had a clear nodular shadow. His laboratory data revealed haemoglobin of 146 g/l. Pulmonary function tests indicated a vital capacity of 2.9 l (104% predicted), a forced expiratory volume in one second of 1.7 l (60% predicted) and carbon monoxide diffusing capacity of 9.8ml/min/ mmHg (56% predicted). In order to avoid acute postoperative exacerbation of his IPF, we considered that as low an FiO2 as possible should be administered. Therefore, we planned to start OLV with an FiO2 of 0.5. We also planned combined general and epidural anaesthesia with monitoring of regional cerebral oxygen saturation (rSO2) using NIRS. If there was no decrease in rSO2 values by more than 20% from baseline values or no decrease in absolute values to below 50%, we planned to leave the FiO2 unchanged regardless of oxygen desaturation. Additionally, we aimed to maintain mean arterial pressure at 65mmHg or higher. After his vital signs (blood pressure, heart rate, SpO2) were assessed, the NIRS sensors, connected to a regional oximetry system and to a monitor that indicated depth of sedation using Root with O3R and SedLineR (Masimo, Irvine, CA, USA), were placed on his forehead. This device simultaneously shows the values of rSO2 and depth of sedation. Baseline rSO2 values using 2 l/min of O2 by nasal prongs were 61% on both sides. General anaesthesia was induced and maintained with propofol, fentanyl, and desflurane, and muscle relaxation was achieved with rocuronium. His trachea was intubated with a 37 Fr left-sided double-lumen tube. FiO2 and tidal volume at 30 minutes after commencing OLV were 0.51 and 320ml (5.2ml/kg), respectively. Arterial blood gas analysis at that time indicated a pH of 7.18, a partial pressure of arterial oxygen of 60mmHg, an arterial oxygen saturation (SaO2) of 82% and a partial pressure of arterial carbon dioxide (PaCO2) of 79mmHg. We permitted the temporary hypercapnia to avoid larger tidal volumes. Despite the decrease in SpO2, the rSO2 values remained at approximately 60% on both sides. FiO2 during OLV was maintained between 0.45 and 0.51. After completion of the lobectomy and return to twolung ventilation the patient’s oxygenation returned to pre-OLV levels. He was extubated in the operating room. Perioperatively, there was no increase in lactic acid levels. Further, there were neither neurological