Pulmonary Barotrauma: Clinical Features

Pulmonary tissue damage

At the point of surfacing in a panic ascent situation, an explosive exhalation of expanded gases may be accompanied by a characteristic sudden, high-pitched cry. Although lung damage resulting from barotrauma can produce respiratory symptoms in the absence of any of the other associated complications, this seems rare in practice. Nevertheless, symptoms that may be seen include dyspnoea, cough and haemoptysis. Clearly, these symptoms may occur in association with any of the complications of PBT discussed later, but the symptoms of pulmonary tissue damage are not invariably present, and their absence should never be used to rule out any of the following diagnoses.

Mediastinal emphysema

As previously described, after alveolar rupture gas may escape into the interstitial pulmonary tissues and track along the loose tissue planes surrounding the airways and blood vessels into the hilar regions and thence into the mediastinum and neck (subcutaneous emphysema). It may also extend into the abdomen as a pneumoperitoneum. When the pleura is stripped off the heart and mediastinum, a pneumoprecordium may be misdiagnosed as a pneumopericardium (Figure 6.2).

Pulmonary barotrauma of ascent
Figure 6.2 Pulmonary barotrauma of ascent: chest x-ray film showing mediastinal emphysema causing the ‘tram track’ sign, from air stripping the pleura from the edge of the cardiac shadow.

CASE REPORT 6.1:RJN, a 19-year-old, was having his second dive in scuba equipment at a depth of 5 metres when he noted a slight pain in his chest. He then noted a restriction in his air supply and thought he had exhausted his gas. He opened his reserve valve and ascended to the surface. He was asymptomatic after the dive, but later, during physical training, he noted that he was breathing heavily and felt weak. A few minutes later he noted slight retrosternal chest pain. During lunch, he developed a fullness in his neck (a ‘tightness’) and dysphagia.

An hour and a half after the dive, he decided to see the doctor because he was not feeling well. It was then noted that his voice was altered in quality and that he had subcutaneous emphysema in both supraclavicular fossae, bilateral generalized crepitus over the chest and positive Hamman’s sign. Chest x-ray study showed gas in the upper mediastinum and neck. An electrocardiogram showed ischaemic changes in leads II, III and aVF.

He was treated with 100 per cent oxygen and improved rapidly.

Chest x-ray study and electrocardiogram were normal 6 days later. Subsequent lung function studies showed that pulmonary compliance was reduced below predicted values.

Diagnosis: Pulmonary barotrauma with mediastinal emphysema and coronary artery embolism.

Symptoms may appear rapidly in severe cases, or they may be delayed for several hours in lesser cases (Case Report 6.1 and Case Report 6.2). Delay may reflect that the symptoms are often ‘mild’ or that it takes time for gas to migrate to the sites where it provokes symptoms. Symptoms may include a voice change including hoarseness or a brassy monotone, a feeling of fullness in the throat, dyspnoea, dysphagia and retrosternal discomfort. In very rare severe cases syncope and shock are possible. The voice changes are described as ‘tinny’ and have been attributed to ‘submucosal emphysema’ of the upper airways and/or recurrent laryngeal nerve damage, although it is difficult to see how bubbles external to the nerve in a relatively compliant tissue space would achieve nerve damage.

CASE REPORT 6.2:TC, an experienced Navy clearance diver, developed epigastric discomfort toward the end of a 90-minute, 11-metre scuba work dive. The dive was otherwise unremarkable, although he had at times worked hard, and he made four controlled ascents during the dive to change his tools.

Approximately 15 minutes after leaving the water, he developed retrosternal chest pain, which increased in intensity over the next few hours. The pain extended from the epigastrium to the base of the throat. The pain was pleuritic in nature and aggravated by inspiration, coughing and movement. He was not dyspnoeic, and there was no associated cough or haemoptysis.

Examination was unremarkable; in particular there were no palpable subcutaneous emphysema and no positive neurological signs. He had no clinical evidence of pneumothorax.

Chest x-ray study revealed the presence of surgical emphysema in the neck and superior mediastinum. No pneumothorax was seen and the lung fields were clear. A computed tomography (CT) scan of the chest was reported as showing ‘air in the mediastinum. Inferiorly, this is seen around the oesophagus in the retrocardiac recess. Superiorly, it is seen surrounding the descending aorta at the level of the carina. It also extends along the major branches of the aortic arch adjacent to the trachea and oesophagus superiorly into the base of the neck on both sides. The spread of air appears to be mainly along the major vessels of the aortic arch into the base of the neck’.

He was treated with 100 per cent oxygen and bed rest, with complete resolution of his symptoms. He was considered permanently medically unfit to dive.

Diagnosis: Pulmonary barotrauma with mediastinal emphysema.

Clinical signs include subcutaneous emphysema of neck and upper chest wall, i.e. crepitus under the skin (described as the sensation of egg-shell crackling, by divers), decreased cardiac dullness to percussion, faint heart sounds, left recurrent laryngeal nerve paresis and in severe cases cyanosis, tachycardia and hypotension. Precordial emphysema may be palpable and produce Hamman’s sign – crepitus related to heart sounds that can sometimes be heard at a distance from the patient. An extension of the mediastinal gas into the tissues between the pleura and the pericardium, rather than gas in the pericardial sac, has occasionally produced cardiac tamponade with its classic clinical signs. There may be radiological evidence of an enlarged mediastinum with air tracking along the cardiac border or in the neck.


If the visceral pleura ruptures, air enters the pleural cavity and expands during any subsequent ascent. It may be accompanied by haemorrhage, forming a haemopneumothorax. The pneumothorax may be unilateral or bilateral, the latter being more common following dramatic emergency ascents.

Pneumothorax from diving has the same clinical features and management as pneu-mothorax from other causes.

Symptoms usually have a rapid onset and include sudden retrosternal or unilateral (sometimes pleuritic) pain, with dyspnoea and tachypnoea. Clinical signs may be absent, or they may include diminished chest wall movements, diminished breath sounds and hyper-resonance on the affected side, tracheal deviation toward the unaffected side with a tension pneumothorax, signs of shock and x-ray evidence of pneumothorax (Figure 6.3).

CASE REPORT 6.3:AI was a relatively inexperienced diver, 19 years old and in good health. He was performing a free ascent from 10 metres. On reaching the surface, he gave a gasp, his eyes rolled upward and then he floated motionless. While he was being rescued from the water it was noted that blood and mucus were coming from his mouth and that he was unconscious. Resuscitation was commenced immediately, using oxygen. He was noted to be groaning at this time but soon after appeared dead. Resuscitation was continued while he was rushed to the nearest recompression chamber. Thirty minutes after the dive he was compressed to 50 metres but with no response. Autopsy verified the presence of pulmonary barotrauma (PBT).

Diagnosis: air embolism resulting from PBT of ascent.

Arterial gas embolism

This dangerous condition is the result of gas passing from the ruptured alveoli into the pulmonary veins and thence into the systemic circulation, where it can cause vascular damage or obstruction, hypoxia, infarction and activation of an inflammatory cascade (see earlier).

Most of the clinical series refer to the brain (CAGE) as the dominant site of disease. Onset typically occurs immediately on surfacing or very soon afterward. In one large series5 of CAGE, the longest interval to onset of symptoms and signs was 8 minutes in a single case, with all other divers showing evidence of CAGE within 5 minutes of completing the dive. There were no cases occurring in excess of 10 minutes.

Serious neurological symptoms consistent with cerebral involvement, which develop immediately after ascent, must be regarded as air embolism and treated accordingly until a definitive diagnosis has been made.


The manifestations of CAGE may include the following:

  • Loss of consciousness and other neurological abnormalities such as confusion, aphasia, visual disturbances, paraesthesiae or sensory abnormalities, vertigo, convulsions and varying degrees of paresis, which is usually lateralized (Case Report 6.3). Paraplegia with a sensory level is more likely to be caused by spinal decompression sickness (DCS) (see Chapter 10) than by CAGE.
  • Cardiac-type chest pain and/or abnormal electrocardiograms (ischaemic myocardium, dysrhythmias).

In a series of 88 cases of CAGE7, mainly from free ascent practices, 34 per cent of the divers suffered loss of consciousness within seconds of surfacing, 23 per cent had become confused, disoriented or uncoordinated after emerging from the water, and 17 per cent had presented with paresis (6 cases with upper monoparesis and 6 with hemiparesis).

In another series presented by Pearson5 that included scuba divers without access to immediate recompression, 15 per cent had complete spontaneous remission within 4 hours, and 53 per cent had some spontaneous improvement before therapy; 77 per cent with coma improved to some degree before treatment. These spontaneous improvements were not always sustained, and 15 per cent of the divers died. It seems clear that divers who exhibit symptoms of CAGE may show partial or even complete recovery within minutes or hours of the incident. As discussed earlier, this may reflect redistribution of the embolus through the cerebral vasculature. Even those divers who become comatose may improve to a variable degree after the initial episode. Unfortunately, such recovery is unreliable. It may not occur or it may not be sustained. Recurrence of symptoms has an ominous prognostic significance.


Focal cerebral symptoms and signs (including unconsciousness) arising immediately after ascent from a compressed gas dive should always be considered most likely caused by PBT and CAGE, especially where the time and depth exposure would normally be considered ‘unprovocative’ for DCS (see Chapters 10 and 11). As previously mentioned, an absence of signs of the presumed barotraumatic injury to the lung (e.g. haemoptysis) is surprisingly common and should not influence the diagnosis.

The differential diagnosis for rapid-onset neurological symptoms after a dive that could be considered provocative for DCS is more problematic, but there are several relevant points. First, the principal competing diagnoses are CAGE and DCS, and distinguishing between them is unimportant from a management point of view. The management is virtually identical (see later). Second, it remains uncertain whether the venous bubbles formed from dissolved gas after decompression and ‘arterialized’ across a right-to-left shunt (see Chapter 10) are large enough to cause the stroke-like syndromes seen after PBT and CAGE. In addition, bubbles are unlikely to form from dissolved gas in the brain tissue itself (see Chapter 10). Third, for the purposes of diagnosis, emphasis should be placed on the putative organ involvement. Manifestations best explained by cerebral involvement (unconsciousness, lateralizing signs, loss of vision, aphasia) are most likely to result from PBT and CAGE (especially if there are concomitant symptoms of PBT), and manifestations best explained by spinal involvement (paraplegia, quadriplegia, loss of anal or bladder tone) are most likely caused by DCS. Confusingly, the two diagnoses may coexist and even interact. Thus, arterial bubbles from PBT may enter tissue micro-vessels and grow as a result of inward diffusion of supersaturated tissue inert gas (see Chapter 10). This mechanism has sometimes been referred to as type III DCS.

Another diagnosis that may cause confusion with CAGE is a haemorrhagic or thromboembolic cerebrovascular accident (CVA) occurring coincidentally with ascent from diving. Such events do occur but are extremely rare, and it is far more likely that cerebral symptoms occurring after a dive are the result of a diving disorder. Indeed, the principal reason for mentioning this differential diagnosis is the frequent inappropriate attribution of CAGE to a CVA when divers are taken to peripheral hospitals staffed by doctors unfamiliar with diving medicine. The same problem arises in cases of DCS.

Previously it was considered important to differentiate between CAGE and DCS because the recommended recompression regimen was different. DCS was treated with a 2.8-ATA oxygen table, e.g. US Navy (USN) Table 6, whereas CAGE was treated using USN Table 6A (which includes an initial deep excursion to 6 ATA). Several animal studies were not able to show an advantage in the initial deep excursion, and most centres now manage patients with CAGE and those with DCS identically (see Chapter 13). Consequently, recompression has become the priority rather than establishing the ‘correct’ diagnosis. It is still considered relevant subsequently to assess the likelihood of whether PBT occurred because this diagnosis has implications for future risk in diving.