Pneumomediastinum is a rare situation in which air is present in the mediastinum. Depending on the pathophysiology of pneumomediastinum, it is usually described as spontaneous and secondary. Spontaneous pneumomediastinum is the presence of air into the mediastinum without any obvious causative factor. The secondary pneumomediastinum may be induced by mechanical ventilation, thoracic trauma, and some invasive procedures. In this report, although the pneumomediastinum and bilateral tension pneumothorax occurred secondary to intubation and mechanical ventilation, the airway had no observable injury as viewed by bronchoscopy. The pathophysiology of these five cases could also be explained as spontaneous pneumomediastinum.
As reported, the pathophysiology of spontaneous pneumomediastinum is based on alveolar rupture due to a rapid increase of intrathoracic pressure, followed by air tracks through the pulmonary interstitium along the bronchovascular sheaths toward the pulmonary hila into the mediastinum, called the Macklin effect (5, 6). Because different amounts of time had elapsed between the chest CT scans and development of pneumomediastinum and tension pneumothorax, the air of bronchovascular sheaths removed by chest tube insertion, only the patient 2 and patient 5 exhibited the Macklin effect in CT image (Figure 3). Many conditions have been reported to be associated with pneumomediastinum, such as bronchial asthma, diabetic ketoacidosis, severe cough or vomiting, and other activities associated with the Valsalva maneuver (6). From the time of development, all the pneumomediastinum and pneumothorax occurred during the stage of recovery of spontaneous breathing shortly after the completion of surgery. Although the remifentanil was administrated continuously to avoid coughing, irregular respiratory rate and tidal volume might also be the causative factors of alveolar rupture.
In one patient, severe neurogenic pulmonary edema was detected before surgery, and the other four patients also presented varying degrees of pneumonia. All five patients were admitted with grade IV. One study found that pulmonary edema was unobserved in patients with grade I–III, and pulmonary edema appeared explicitly in grade IV or V patients with a prevalence of 31.3% (7). Another study found poor-grade subarachnoid hemorrhage to be an independent predictor of ALI and acute respiratory distress syndrome (8). For our five cases, the ALI during aSAH might be etiological basis of pneumomediastinum and bilateral tension pneumothorax.
The decision of whether to extubate the trachea at the end of the surgical procedure usually depends on the severity of aSAH, preexisting illness, and the preoperative level of consciousness. We consider that a shorter time period between chest CT scan and surgery could allow more accurate assessment of acute lung injury complicated from aSAH. The respiratory condition should also be checked immediately after surgery to re-evaluate the pulmonary condition in case there is an immediate need for management. If the patient had ALI either before or after surgery, paralysis and sedation should be performed until the lung injury improves.