Failure to detect a pneumothorax may have serious complications. A case of a pneumothorax, which may have been overlooked if thoracic computed tomography had not been performed, is discussed.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
A 35 year old man was ejected from a sports car while racing at the UK Goodwood Revival Meeting 2000 (fig 1). At the scene, the patient was maintaining his own airway with normal bilateral air entry. He was tachycardic (110 beats per minute), blood pressure 130/80, with normal capillary refill time, and a Glasgow coma score of 15/15. The only obvious injury was bruising to the left thoracic-scapular area. Intravenous access was established and high flow oxygen was administered. The patient was immobolised on a spinal board and transferred to the local hospital.
On arrival, the primary survey remained unchanged. Secondary survey confirmed a painful right hemithorax and bruising of the left thoracic-scapular region with some overlying abrasions. The abdomen was soft and non-tender and no signs of injury were evident in the upper and lower limbs with no neurovascular deficit. Spinal examination was unremarkable. Chest radiographs revealed an intra-articular fracture of the left glenoid and left scapula and fractures of the right fifth, sixth, and seventh ribs. Radiographs of the cervical spine, pelvis, and thoracic and lumbar spines were normal. However, because of the markers of severe injury present and the possibility of an underlying pneumothorax, the chest radiograph was repeated. This did not reveal any new changes. The patient remained haemodynamically stable but in view of the high energy mechanism of injury, computed tomography of the chest and abdomen was undertaken.
The CT scan showed a right anterior pneumothorax and small left haemopneumothorax (fig 2), which in retrospect was not visible on the chest radiographs. Bilateral chest drains were inserted in the fifth intercostal space in the mid-axillary line and the patient transferred to the intensive therapy unit for 24 hours of close observation. The scapular injury was assessed by shoulder arthroscopy and treated expectantly. The patient remained stable and was discharged from hospital five days later.
The risk of serious injury is increased by more than 300% if the occupant is ejected from a vehicle.1 The incidence of significant chest injury is also increased in the presence of scapula fracture, reflecting the severity of trauma required to a cause this injury.2
Thoracic injuries are the cause of death in 25% of trauma fatalities and a major contributor in 50%.3 Despite this, pneumothorax may often be initially overlooked and not always be detected on plain chest radiography. In one study of 103 severely injured patients with blunt chest trauma, 67 had major chest injuries that had been missed on chest radiograph and of these 27 had pneumothoraces and 21 had haemothoraces. The thoracic CT scan was significantly more sensitive than routine chest radiography in detecting pneumothoraces.4 In another study of 90 trauma patients initial supine chest radiography failed to detect a pneumothorax in 35 patients 3; the diagnosis being made on thoracic CT scans within 24 hours of admission. A further study reported four cases of occult pneumothorax not visible on chest radiograph, which were later diagnosed on spiral CT scan. 5
Mechanical ventilation of a pneumothorax may cause tensioning and respiratory and cardiovascular deterioration. The use of nitrous oxide during anaesthesia may cause further expansion of pneumothoraces.6 This case reiterates the need for a high index of suspicion relating mechanism of injury to possible underlying occult chest injuries such as a pneumothorax. While routine thoracic computed tomography may not be advocated in all cases of blunt chest trauma, it should be undertaken whenever markers of severe injury or high energy mechanisms of injury are present or if the patient requires mechanical ventilation, in which case it may significantly change management.6 Early thoracic computed tomography is superior to routine chest radiography in diagnosing pneumothorax and haemothorax.4,7 If left undiagnosed, the consequences of mechanical ventilation, in the presence of a pneumothorax may be fatal. Blunt injury to the chest may also be associated with injury to abdominal viscera and a CT scan of the abdomen should be performed at the same time as the thoracic CT scan. With the advent of spiral computed tomography, the time spent in the CT scanner is not significantly increased, which otherwise would obviously have an inherent danger.
T P C Kane initiated the case report. M C Nuttall performed the literature search and wrote the report in conjunction with T P C Kane. R C Bowyer was the consultant surgeon responsible for the care of the patient and V Patel was the consultant orthopaedic surgeon who performed the shoulder arthroscopy. M C Nuttall acts as guarantor for the case report.