A Persistent Air Leak, a Chest Tube, and a Predecessor's Note: Nitrous Oxide in Thoracic Surgery

No - I will not use nitrous oxide, and I will say so explicitly in writing.

The physics/physiology of the objection:

• N2O is roughly 30-fold more soluble in blood than nitrogen (blood:gas partition coefficients on the order of 0.47 vs 0.015). This enormous difference in solubility - not in diffusivity - drives the behavior of any gas-filled space.
• When 60–70% N2O is inspired, the partial pressure gradient from blood into the pleural space is steep. N2O enters the space far faster than nitrogen can be carried away by the relatively poorly perfused pleura and by the leak.
• The volume of the space therefore tends to increase until the partial pressures equilibrate. The result is expansion of a pneumothorax, or pressurization of a space whose vent is too small to keep up.

Why "it leaks out anyway" is wrong:

• The leak is through a small orifice and its conductance is fixed; N2O inflow is not. Expansion occurs whenever inflow exceeds outflow, and the margin is unpredictable from the bedside.
• Decisive point: the pleural space does not stay open. At the moment of pleurodesis and chest closure the space becomes functionally closed, and any N2O still on board then expands into an unvented space.

Plan and communication:

• Use air/O2; if a third gas is needed, use medical air. There is no outcome here that requires N2O.
• Tell the surgeon directly and contemporaneously, and document: "N2O contraindicated - persistent air leak/bullous disease; expansion of intrapleural gas and risk at chest closure." A correction note is appropriate and professional; it is not a personal attack on the covering anesthesiologist.

Lung isolation:

• A left-sided double-lumen tube (approx. 37 Fr for this height/weight) under fiberoptic confirmation is my default. Against a bronchial blocker: a blocker is easy to dislodge, is harder to suction through, and cannot be used for CPAP/recruitment of the operative lung. Its main advantage - avoiding a large tube in a bad airway - does not clearly apply here.
• With a large air leak, isolation is not optional: it protects the dependent, ventilated lung from contamination and from positive-pressure delivery into bullous, leaking tissue.

Ventilation goals:

• Lung-protective: tidal volume ~5–6 mL/kg predicted body weight during one-lung ventilation; PEEP 5 cmH2O to the dependent lung; keep plateau pressure < 25–30 cmH2O; accept permissive hypercapnia.
• FiO2 titrated to SpO2 ≥ 90% rather than 1.0 throughout, to limit absorption atelectasis and oxygen toxicity.
• Avoid high peak pressures and large tidal volumes in the ventilated lung - this patient's remaining lung is also diseased, and barotrauma in the contralateral side is my principal fear.

Monitoring and preparation:

• Arterial line before induction (beat-to-beat pressure, and serial ABGs for hypercapnia).
• Two suction setups, a spare chest tube tray, and the surgeon in the room before induction.
• Anticipate hypotension with CO2 insufflation and mediastinal shift; agree on an insufflation pressure ceiling (usually ≤ 10 mmHg) with the surgeon preoperatively.

Differential, most likely/urgent first:

1. Tube malposition - the DLT has moved (patient repositioned for lateral decubitus). Both lungs may now be ventilated, or the right upper lobe orifice is obstructing, so a leaking, bullous right lung is being positive-pressure ventilated.
2. Tension pneumothorax on the ventilated (left) side, or rapid expansion of the right-sided air space - the classic cause of rising airway pressure, desaturation, hypotension, and rising CO2 all at once.
3. Bronchospasm / mucus plugging (he is a wheezer with a 0.9 L FEV1).
4. Surgical/mechanical - excessive CO2 insufflation pressure, or the surgeon compressing mediastinal structures or the pulmonary artery.
5. Less likely but lethal: anaphylaxis, pulmonary embolism, or an acute coronary event in a patient with a hypertrophied, poorly tolerant ventricle.

First five actions:

1. Announce the problem; FiO2 to 1.0; tell the surgeon to stop insufflation, and to look in the chest.
2. Hand-ventilate with a low flow to feel compliance and to exclude a machine/valve problem; disconnect briefly to relieve any built-up pressure.
3. Bronchoscope the tube immediately and re-confirm position; suction for mucus.
4. Examine the field with the surgeon; if tension is possible, decompress (needle or open) without waiting for a radiograph - in a VATS case the surgeon can look directly.
5. Support perfusion with small aliquots of phenylephrine and a fluid bolus while the cause is corrected; recheck an ABG and reassess after every step.

Note: EtCO2 rising during one-lung ventilation is expected to some degree; what makes this an emergency is the accompanying pressure rise and hypotension.