Abstract
A 10-year-old boy struck a car bonnet following which his left shoulder got pinned under the wheel. No life-threatening injuries were identified. However, the patient sustained extensive abrasions to the back and the left shoulder, a closed deformity of the left clavicle and a swollen, but stable right knee. The patient was neurovascularly intact globally and all joints had a full range of motion. Plain radiographs suggested a possible greenstick fracture of the left clavicle, but also free gas within the left glenohumeral joint. Concern was raised of an unidentified open injury to the joint. CT was supportive of the finding of gas within the left glenohumeral joint, but ruled out the possible greenstick fracture as a spurious finding. There were no other injuries. The gas was within the left glenohumeral joint and was consistent with vacuum phenomenon. The injury was treated expectantly and the child made a full recovery.
Keywords: radiology (diagnostics), orthopaedics, paediatrics
Background
Gas collection within a closed joint space is known as the vacuum phenomenon.1–3 The vacuum phenomenon is usually observed in intervertebral discs on routine spinal CT. Case reports exist reporting vacuum phenomenon in association with fractures of the mandibular condyle,4 the knee5 and the pubic bone6, it has been described as a sign of delayed or non-united fractures.6 7 The phenomenon has also been reported in association with infections,8 dislocated joints,9 in the ankle10 and in spinal discs and intervertebral spaces.5 6 8 11 A number of reports exist of vacuum phenomenon in the shoulder,12–14 but none within the English language literature. Furthermore, vacuum phenomenon in children is extremely rare; indeed, to our knowledge it has only been discussed in two papers, neither of which are readily available in English.15 16 The finding of vacuum phenomenon on its own, without any other clinical pathology may not lead to any complications. However, it is an important clue for occult penetrating injuries, subtle fractures, recent dislocation/relocation injuries and trauma to the joint which may otherwise go unnoticed. Additionally, awareness of such a phenomenon can prevent clinicians from misdiagnosing chondrocalcinosis or intra-articular loose bodies.7 9 17 We report the unusual case of vacuum phenomenon in the shoulder of a child following trauma.
Case presentation
A 10-year-old boy struck a car bonnet before being pinned under a wheel by his left shoulder. In the Accident and Emergency Department, full Advanced Trauma Life Support (ATLS) protocols were followed. The patient remained stable and conscious throughout the primary survey and no life-threatening injuries were identified. Secondary survey identified pain and extensive abrasions to the back and the left shoulder, a closed deformity of the left clavicle and a swollen, but stable right knee.
Investigations
Plain radiography suggested superior migration of the humeral head, a possible greenstick fracture or plastic deformity of the left clavicle and also free gas within the left glenohumeral joint (figure 1). Repeated examination revealed superficial abrasions to the back and the left shoulder with no conclusive penetrating injury. The left arm was neurovascularly intact and movements of the shoulder, although laboured and tender, had a full range of motion. Due to the pain in the shoulder, deformity of the left clavicle, abrasions to the shoulder and back and initial radiological findings, concern was raised of a possible subtle and unidentified open injury to the joint. The patient was transferred to radiology for trauma CT. Review of the images (figure 2) was supportive of the finding of gas within the left glenohumeral joint, but ruled out any bony injury to the shoulder (suggesting the suspected greenstick fracture of the clavicle and superior migration of the humeral head seen in the radiographs were spurious findings). There was no visible pneumothorax or thoracic injury either. The imaging was further discussed at the following morning’s departmental orthopaedic trauma meeting, in order to confirm the spurious findings.
Figure 1.

AP radiograph of the left shoulder demonstrating lucency in the left glenohumeral joint consistent with free gas (identified by arrow). AP, anteroposterior.
Figure 2.

Cross-sectional mediastinal-weighted CT of the chest demonstrating free gas in the left glenohumeral joint (identified by arrow).
Differential diagnosis
The gas within the left glenohumeral joint was felt to be consistent with vacuum phenomenon of the shoulder rather than an unidentified open injury as previously thought.
Another possible differential for the gas seen on the initial radiograph was of surgical emphysema. However, while there were some superficial abrasions, there was no penetrating breach to the skin, making surgical emphysema unlikely. Additionally, the gas shadow was in the glenohumeral joint (confirmed by CT) as opposed to the subcutaneous tissue typical of surgical emphysema. This was also confirmed at the following morning’s departmental orthopaedic trauma meeting.
Treatment
The patient’s injuries were treated expectantly. No immobilisation was advised since there was no fracture and the patient had a full range of motion. Analgesia was prescribed to take as required and safety netting advice was given to the patient’s parents (to come back to the emergency department if there were any new/worsening symptoms) and advice was given to avoid strenuous activities such as sports for 4 weeks.
Outcome and follow-up
The patient was routinely followed up in the fracture clinic for 2 weeks. At this time, he had a full range of motion, minimal tenderness and slight bruising. No further imaging was warranted due to the age of the child, lack of clinical symptoms and the benign nature of vacuum phenomenon. The patient was followed up again at 6 weeks. There was no evidence of shoulder instability, and as such, physical therapy was not indicated. The patient subsequently returned to competitive tennis and football with no difficulty.
Discussion
The vacuum phenomenon presents with a radiolucent area seen in synovial joints, vertebrae and intervertebral discs.3 Current understanding is that the gas, of which 90% to 92% is nitrogen,18 is produced from surrounding soft tissues and accumulates within an enclosed space. The mechanism of gas production depends on the negative pressure within the joint for dissolved nitrogen to come out of solution.18 The process is similar to that which occurs in decompression sickness, colloquially referred to as ‘the bends’. Distraction of opposing articular surfaces accounts for the phenomenon in synovial joints,3 whereas degenerative processes cause the phenomenon in intervertebral discs, with the phenomenon seen in approximately 20% of elderly patients with degenerative spinal disease.3 11 18
In the absence of fracture, pain or functional deficit, we extrapolate that negative pressure was generated within the joint as a result of prolonged distraction while trapped under the car wheel. It is uncertain as to whether the joint may have been dislocated and relocated in the injury, but this diagnosis is certainly suggestive of significant trauma. Minowa et al 4 hypothesise (in non-open condylar fractures) that rapid dilatation of a joint capsule creates a negative pressure which in turn attracts gas from the articular surfaces only when the joint capsule is not ruptured. Ford et al and Malghem et al concur that the intensity of the distraction applied to the joint and the physiological factors governing the gasses in the soft tissues, such as partial pressures and diffusion gradients, intertwine to determine the appearance of the gas within the joint and how long it remains before resorption.11 18
This case illustrates how radiology can reveal histories of injuries and provide evidence of trauma to joints where there is little to find on clinical examination and no bony injury. In traumatic injuries in this child, the presence of the vacuum phenomenon might serve to provide evidence to clinicians that a joint such as the shoulder has undergone a significant trauma. This might be a useful diagnostic marker where an injury had hitherto not been clinically evident, but was suspected. The patient’s symptoms of pain, abrasions to the skin, gas and possible fracture seen on plain radiographs warranted further imaging with CT, which ruled out several causes for vacuum phenomenon (such as fracture or infection), but suggested other potential causes (such as recent dislocation). While it may be deemed that the CT was unnecessary, we believe that it ruled out significant injury which was initially suspected and educated us about the vacuum phenomenon steering us to an appropriate conservative management strategy rather than overtreating the patient.
There is also an argument that MRI would have provided more information regarding soft tissue injury. However, CT imaging is our unit’s preferential ATLS screening modality. Perhaps in the future, MRI could be utilised as an alternative or additional modality in similar situations.
Learning points.
Those assessing children who have been involved in trauma should be aware of the vacuum phenomenon.
Recognition of this diagnosis and associated mechanism may trigger appropriate further examination and investigation
Knowledge of this diagnosis should also avoid unnecessary clinical concern regarding potential open injury.
Unawareness of vacuum phenomenon or inability to identify it on imaging can lead to either missing the sign altogether, leading to undertreatment of occult injuries/conditions or attributing the sign to an alternative misdiagnosis, which may lead to overtreating the patients.
Footnotes
Contributors: IK and PT treated the patient in the case and wrote the first draft of the manuscript in equal parts. IK obtained consent to write the manuscript from the patient’s mother and the patient. PR reviewed and revised the manuscript in it’s final form and compiled the figures supplementing the case report. All three authors have made a substantial contribution to drafting this paper and have read and approved this submission, which has been made as per the BMJ Case Reports instructions for authors.
Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests: None declared.
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Parental/guardian consent obtained.
References
- 1. Fuiks DM, Grayson CE. Vacuum pneumoarthrography and the spontaneous occurrence of gas in the joint spaces. J Bone Joint Surg Am 1950;32 A:933–8. [PubMed] [Google Scholar]
- 2. Knutsson F. The Vacuum Phenomenon in the Intervertebral Discs. Acta radiol 1942;23:173–9. 10.3109/00016924209134912 [DOI] [Google Scholar]
- 3. Resnick D, Niwayama G, Guerra J, et al. Spinal vacuum phenomena: anatomical study and review. Radiology 1981;139:341–8. 10.1148/radiology.139.2.7220878 [DOI] [PubMed] [Google Scholar]
- 4. Minowa K, Sakakibara N, Inoue N, et al. Vacuum phenomenon in non-open fracture of the mandibular condyle: report of three cases. Dentomaxillofac Radiol 2008;37:361–4. 10.1259/dmfr/53043000 [DOI] [PubMed] [Google Scholar]
- 5. Wright DM, Sochart DH. Spontaneous vacuum phenomenon in the lateral compartment of the knee associated with a lateral tibial plateau fracture. Knee 2006;13:42–4. 10.1016/j.knee.2005.05.005 [DOI] [PubMed] [Google Scholar]
- 6. Zibis AH, Karantanas AH. Vacuum phenomenon in posttraumatic nonunion of pubic bone fracture. AJR Am J Roentgenol 1999;172:251 10.2214/ajr.172.1.9888787 [DOI] [PubMed] [Google Scholar]
- 7. Stallenberg B, Madani A, Burny F, et al. The vacuum phenomenon: a CT sign of nonunited fracture. AJR Am J Roentgenol 2001;176:1161–4. 10.2214/ajr.176.5.1761161 [DOI] [PubMed] [Google Scholar]
- 8. Bielecki DK, Sartoris D, Resnick D, et al. Intraosseous and intradiscal gas in association with spinal infection: report of three cases. AJR Am J Roentgenol 1986;147:83–6. 10.2214/ajr.147.1.83 [DOI] [PubMed] [Google Scholar]
- 9. Ahmad R, Annamalai S, Radford M, et al. Vacuum phenomenon in a dislocated joint. Emerg Med J 2007;24:862 10.1136/emj.2006.043992 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Lee TH, Wapner KL, Mayer DP, et al. Computed tomographic demonstration of the vacuum phenomenon in the subtalar and tibiotalar joints. Foot Ankle Int 1994;15:382–5. 10.1177/107110079401500707 [DOI] [PubMed] [Google Scholar]
- 11. Malghem J, Maldague B, Labaisse MA, et al. Intravertebral vacuum cleft: changes in content after supine positioning. Radiology 1993;187:483–7. 10.1148/radiology.187.2.8475295 [DOI] [PubMed] [Google Scholar]
- 12. Delahaye RP, Weber J, Barres J. [The vacuum phenomenon in articulations of the shoulder and the hip (apropos of 15 cases)]. J Radiol Electrol Med Nucl 1961;42:541–5. (Article in French). [PubMed] [Google Scholar]
- 13. Gandin J, Laurens G, Bobo G. [The articular vacuum phenomenon after shoulder injury (apropos of 35 cases)]. Rev Chir Orthop Reparatrice Appar Mot 1962;48:545–50. (Article in French). [PubMed] [Google Scholar]
- 14. Virtama P. The vacuum phenomenon in knee- and shoulder-joints. Ann Chir Gynaecol Fenn 1957;46:51–6. (Article in Finnish). [PubMed] [Google Scholar]
- 15. Gyarmati J, Oláh J. [The vacuum phenomenon in childhood and adult age]. Radiol Diagn 1968;9:357–62. (Article in German). [PubMed] [Google Scholar]
- 16. Iordache S, Rath E, Atar D, et al. [Vacuum phenomenon in the hip joint: diagnostic value]. Harefuah 1998;135(3-4):108–9. 67 (Article in Hebrew). [PubMed] [Google Scholar]
- 17. Patten RM. Vacuum phenomenon: a potential pitfall in the interpretation of gradient-recalled-echo MR images of the shoulder. AJR Am J Roentgenol 1994;162:1383–6. 10.2214/ajr.162.6.8192004 [DOI] [PubMed] [Google Scholar]
- 18. Ford LT, Gilula LA, Murphy WA, et al. Analysis of gas in vacuum lumbar disc. AJR Am J Roentgenol 1977;128:1056–7. 10.2214/ajr.128.6.1056 [DOI] [PubMed] [Google Scholar]
