a b s t r a c t

Background: Severely traumatized patients undergo whole-body computed tomography (WCT) to detect lethal anatomical injuries. When checking the images, we have sometimes recognized minute gas (the vacuum phenomenon [VP]) near the traumatized lesions. Accordingly, we investigated the significance of the VP in patients with trauma.

Basic procedures: From April to October 2013, a medical record review was retrospectively performed for all patients with trauma. The exclusion criteria included an age more than 60 years, patients who did not receive the WCT, scan and those in cardiopulmonary arrest on arrival. The subjects were divided into 2 groups: a VP group (n = 19), which included patients who had the VP, and a control group (n = 49).

Main findings: There were no significant differences between the 2 groups with regard to age, the mechanism of injury, or the survival rate. In contrast, the ratio of women, the Injury Severity Score, and the duration of hospita- lization in the VP group were significantly higher than those in the control group. The greatest number of the VP was located at or near rib fractures, followed by joint spaces that experienced a traumatic impact.

Principal conclusion: This study demonstrated that the VP tended to be observed most often in severely trauma- tized female cases. The VP is observed at locations that experience a traumatic impact, so an analysis of the VP may be useful to elucidate the mechanism of injuries. The presence of traumatic VP itself does not influence the final outcome.

(C) 2014

Introduction

Severely traumatized patients undergo whole-body computed to- mography (WCT) or traumatic pan scans to detect lethal anatomical in- juries even when their vital signs are stable because this approach is considered to lead to a favorable outcome [1-3]. As a result, even mildly traumatized patients who have experienced a high-energy impact tends to undergo a WCT examination [4,5]. In our institute, we also per- form WCT for traumatized patients who have experienced a high- energy impact and patients with an unknown mechanism of injury who were unconscious or had dementia without a witness of the accident. We ourselves first check the WCT images to make the decision regarding the treatment. However, when checking the images, we have sometimes recognized minute gas near the traumatized lesions. Previous reports have revealed the presence of the vacuum phenome- non (VP) at joints or fractured sites [6,7]. Other pathologies associated with the VP, other than trauma, have been reported to include degene- rative joint disease, metastasis, abscess, osteomyelitis, multiple myeloma, surgical introduction of air, Joint effusion, or Schmorl’s nodes. The

? Support, presentation: none.

* Corresponding author. 410-2295, 1129 Nagaoka Izunokuni city Shizuoka, Japan. Tel.: +81 55 948 3111.

E-mail address: [email protected] (K. Omori).

mechanism responsible for the formation of the VP has been reported to be the following [6]: If an enclosed space is allowed to expand, the vol- ume within the enclosed space will increase. In the setting of expanding volume, the pressure within the space will fall. The solubility of the gas in the enclosed space will decrease as the pressure of the space decreases. Decreased solubility allows the gas to leave solution. The combination of lower nitrogen solubility and minimal metabolism of nitrogen by the body mainly accounts for the formation of the VP. Analyses of the clinical significance of the VP at specific local lesions have been reported [6-15]; however, the clinical significance of the VP on the WCT obtained from traumatized patients has not been reported. Accordingly, we herein in- vestigated the significance of the VP in patients with trauma evaluated by WCT in our institution.

Subjects and methods

This retrospective study protocol was approved by the review board of Juntendo Shizuoka Hospital, and all examinations were conducted according to the standards of good clinical practice and the Declaration of Helsinki.

The Department of Acute Critical Care Medicine, Juntendo Shizuoka Hospital, which is a 552-bed hospital in the Izu Peninsula in Shizuoka prefecture located near Tokyo, serves a population of approximately

http://dx.doi.org/10.1016/j.ajem.2014.11.055

0735-6757/(C) 2014

K. Omori et al. / American Journal of Emergency Medicine 33 (2015) 282-285 283

270 000. The helicopter parks at our hospital, and an emergency medi- cal system uses physician-staffed emergency helicopters in Eastern Shizuoka prefecture, which primarily transport patients in severe condi- tion from Izu Peninsula to our hospital. Our department mainly treats patients with severe trauma, cardiopulmonary arrest, unconsciousness, convulsions, intoxication, and unstable vital signs. When we treat trau- matized patients, we perform WCT scans that include the head to the pelvis; and additional examinations for extremities are performed when a fracture is suspected. Whether the study is plain plus enhanced or plain only is decided by the staff members on duty. The indications for WCT are described above in the introduction. The images of the plain studies were reviewed by manually changing the window and level of the CT images to differentiate fat from air and magnifying the CT images to detect minute VP [16].

The main CT protocol was as follows: Most patients were examined using a 64-slice multidetector CT scanner (LightSpeed VCT; GE Healthcare Asia, Hino-shi, Tokyo). The emergency protocol consisted of 2 separate scans. The first was an axial scan of the cranium angulated between 0? and 30? (depending on the position of the patient). Second, a helical scan of the whole body was performed in the craniocaudal direction, ranging from the orbital cavity (to picture the cervical spine) to the bottom of the pelvis. image acquisition was conducted with the following parameters: tube voltage: 120 kV, collimation: 64 x 0.625, pitch: 1375, and noise index: 8.

From April to October 2013, a medical record review was retro- spectively performed for all patients with trauma. The exclusion criteria included an age more than 60 years, patients who did not re- ceive a WCT scan, and those in cardiopulmonary arrest on arrival. The reason for excluding the elderly subjects is that aged people tend to have the VP at degenerative joints [6,8]. The definition of the VP was the existence of air density at nonphysiological sites (physiological sites include the paranasal sinus, airway, lung, bowel, and bladder). We also excluded the gas in the intrathoracic cavity, intraabdominal cavity, disk space, and intravascular space because gas in these locations may be created by lung injury, a perfo- rated bowel, or aging or may be due to iatrogenic causes.

The subjects were divided into 2 groups: a VP group, which included patients who had the VP on the WCT scan, and a control group, which included patients who did not. The data regarding the Patient sex, age, mechanism of injury, Injury Severity Score, presence of hypertension or diabetes mellitus, number of contusional lacerated wounds at any site, duration of hospitalization, and survival rate were analyzed between the 2 groups. In the VP group, the existence of contusional lacerated wounds close to the site of VP and the number and site of the VP were also investigated.

The statistical analyses were performed using the unpaired Student t test, the paired Student t test, or the ?² test, as appropriate. A P value b .05 was considered to indicate a statistically significant difference. All data are presented as the means +- standard error.

Results

During the investigation period, a total of 166 patients were treated in the emergency department. Among these patients, 80 were more than 60 years old. Another 17 patients did not undergo the WCT. After excluding these 97 patients and 1 patient who was in cardiopulmonary arrest, a total of 68 patients were enrolled as subjects in this study, in- cluding 19 patients in the VP group and 49 patients in the control group. The background characteristics of the subjects are shown in Table 1.

There were no significant differences between the 2 groups with regard to age, the mechanism of injury, or the survival rate. All subjects were independent in activities of daily living, and the rates of hypertension and diabetes mellitus were not statistically different between the 2 groups. In contrast, the ratio of females, the Injury Severity Score, and the duration of hospitalization in the VP group were significantly higher than those in the control group. The frequency of contusional lacerated

Table 1

A comparison between the VP group and the control group

VP (n = 19) Control (n = 49) P

Sex (female/male) 7/12 7/42 .05

Age 35.9 +- 2.9 34.8 +- 2.0

Mechanism of injury

Traffic accident	14 (74%)	31 (63%)
Fall	4 (21%)	15 (31%)
Others	1 (5%)	3 (6%)
Injury Severity Score	12.1 +- 2.3	6.8 +- 0.7	b.01
Hypertension or diabetes mellitus	0	5 (10.2%)	NS
Contusional lacerated wound at anywhere	10 (52.6%)	14 (28.5%)	.08
Duration of hospitalization (d)	28.2 +- 7.8	10.2 +- 1.5	b.01
Survival rate (%)	100	100	NS

NS: not significant.

wounds at any sites of the body was greater in the VP group than in the control group; however, the difference was not statistically significant.

Table 2 shows the numbers and sites of the VP. The greatest number of the VP was located at or near rib fractures, followed by joint spaces. Representative cases are shown in Figs. 1 to 4. All of the patients with the VP in the joint space had bony factures that affected the joint. The other types of the VP except occurred at borders between fractures and Soft tissues.

Discussion

This is the first study to demonstrate that, among traumatized pa- tients younger than 60 years, the VP tended to be observed more often in severely traumatized female cases; and the finding of traumatic VP itself on the WCT scan did not influence the final outcome. There is currently controversy whether the VP is a potential finding in the early stage of a sacral insufficiency fracture of the sacroiliac joint [9-11]. The inclusion of subjects of different ages (elderly or not) may explain the differences among these reports. Although we excluded the elderly subjects to avoid including any apparent cases of the VP due to age-related degeneration, the VP may correlate well with the traumatic site. Regarding the predominance of the VP in women, the female tissues tend to be more fragile in comparison with those of men [17,18]; so the production of the VP was more frequently observed in women than men when the women experience a traumatic impact. Based on the nature of the retrospective study design, the medical records of the subjects unfortunately did not record whether the female patients were of childbearing age or menopausal, a difference related to potential alterations in bone mineralization associated with osteoporosis correlating with a tendency toward fractures.

Chronic microvascular diseases, such as atherosclerosis, are long- term comorbidities that increase with age and can interfere with the Blood supply to the joint [6]. Hypertension and diabetes mellitus are well-known risk factors of atherosclerosis. Reduced perfusion due to atherosclerosis may impair the clearance of nonmetabolized gas. How- ever, in the present study, the rates of hypertension and diabetes were not significantly different between the 2 groups. We also excluded el- derly subjects with potential chronic microvascular diseases. Accordingly, the presence of chronic microvascular disease did not affect the onset of VP in this study.

Table 2

The numbers and locations of the vacuum phenomena (in total)

Around a rib in the chest	9
Around a rib in the abdomen	5
In front of the sacroiliac joint	2
At a fracture of an extremity	2
In the intraSpinal canal	1
In the sternoclavicular joint	1
At the hip joint	1
At the Shoulder joint	1

284 K. Omori et al. / American Journal of Emergency Medicine 33 (2015) 282-285

Fig. 1. A and B, A case of the VP at a rib in the chest. This case shows the VP at a rib (A, white arrow). This case had rib fractures. The black arrow indicates the VP located posterior to the sternum (B). However, this lesion was not considered to be the VP because it was located in the thoracic cavity in this study.

The VP in this study tended to exist near bony fractures. It has been demonstrated that a hydrodynamic impact pressure is generated from the implosion of various individual acoustic cavitation bubbles near a rigid boundary [19], so the accumulation of such cavitation bubbles may contribute to the formation of a visible VP on the WCT scan. Mean- while, Gohil et al [6] suggested that VP is almost exclusively detectable in synovial joints because of changes in the solubility of nitrogen in re- lation to changes in volume/pressure; however, this phenomenon was seen most frequently close to the ribs in this study. One possible reason for the discrepancy in results between the 2 studies is variation in the targeted area and positioning of the extremities during CT. In general, according to our protocol, the patient was scanned from the head to the pelvis; and the extremities were not targeted because we did not aim to detect VP in the joints. In addition, traction of the extremities re- sulted in an increase in the volume of the joint, leading to greater pro- duction of intraarticular gas. However, we did not apply this maneuver during the CT examinations. Accordingly, the decreased number of joints evaluated without retraction using CT was associated with a decreased opportunity to detect VP in the joint. Another possible reason for the above discrepancy in results is differences in the charac- teristics of the targeted subjects. The subjects in this study generally ex- perienced a high-energy impact. An intact joint capsule is required to produce intraarticular gas. However, a high-energy impact may increase the frequency of breakage of the joint capsule, thus resulting in a decreased rate of detection of VP in the joint. In addition, Gohil et al

Fig. 2. A case of the VP at the rib in the abdomen. This case showed the VP at a rib (black arrows). This case had rib fractures. The images were reviewed by manually changing the window and level of the CT images to differentiate fat from air and magnifying the CT im- ages to detect minute VP.

reviewed relevant reports, which primarily focused on elderly patients or those with degenerative diseases of the joints, without a history of high-energy impacts. These differences in the background factors of the subjects may account for the differences in findings between the 2 studies.

In the current study, the frequency of contusional lacerated wounds tended to be greater in the VP group than in the control group. As the rate of the Injury Severity Score was also greater in the VP group than in the control group, the increased prevalence of high-energy impacts may have resulted in the increased rate of contusional lacerated wounds noted in the VP group. Only 2 subjects in the VP group had contusional lacerated wounds close to the site of VP, whereas the remaining 8 sub- jects had contusional lacerated wounds located a long distance from the site of VP (eg, the VP was detected in the chest, and contusional lacerated wounds were observed on both legs). Although the 2 sub- jects who exhibited VP close to their wounds were clinically judged to not have a history of direct blunt trauma leading to tearing, the potential entrance of micro gas bubbles from the wounds cannot be completely excluded.

The diagnostic accuracy for changes from active bleeding, detected only using contrast medium in patients in a state of trauma. However, negative studies have also been reported [20]; and the use of contrast medium is associated with a risk of anaphylaxis. In our institute, the emergency department physicians are in the CT room and check the screen image of Plain CT immediately after the CT examination. When a massive or moderate amount of intrathoracic fluid, mediastinal

Fig. 3. A case of the VP at the pelvis. This case showed the VP in front of the sacroiliac joint at the S3 level (black arrow). The patient had multiple rib fractures without pelvic factures, although he reported hip pain.

K. Omori et al. / American Journal of Emergency Medicine 33 (2015) 282-285 285

Fig. 4. A case of the VP in a hip joint. This case showed the VP at a hip joint (white arrow). This case had a fracture of the femoral head.

hemorrhage, any amount of fluid on the abdomen (such as the triangle sign [21]), or a pelvic fracture was detected, then contrast medium was administrated. If there was no significant sign of visceral injury or pelvic fracture, an enhanced study was not performed even though we had already obtained informed consent for the use of contrast medium. As the Japan Advanced trauma evaluation and Care program based on the Advanced Trauma Life Support did not make any comments about WCT with or without contrast medium, there is no precedent due to the lack of use of contrast medium in CT examinations under a traumatic emergency state that has resulted in a medicolegal case in Japan. As the literature provides limited evidence whether immediate total-body CT with an enhanced study leads to better clinical outcome than conven- tional radiographic imaging supplemented with selective CT scanning in trauma patients, a multicenter, randomized controlled trial of immediate total-body CT scanning in trauma patients is now ongoing [22].

Concerning the clinical significance of the VP, the traumatic VP in itself found on the WCT scan does not influence the final outcome because all the subjects with the VP survived without any clinical com- plications induced by the VP. However, the VP might be induced by serious pathology, such as infection; so physicians should pay close attention to rule out the development of such disease. In addition, the VP was observed at locations of traumatic impact; so an analysis of the VP can be useful to elucidate the mechanism of injuries.

Some of the limitations associated with this study include its retro- spective design and the small number of cases. Therefore, future pro- spective studies involving a larger number of patients are needed to further examine this issue.

Conclusion

Among traumatized patients younger than 60 years, the VP tended to be observed most often in severely traumatized female cases. The

VP is observed at locations that experience a traumatic impact, so an analysis of the VP may be useful to elucidate the mechanism of injuries. The presence of traumatic VP itself does not influence the final outcome.

References

1. Huber-Wagner S, Lefering R, Qvick LM, Korner M, Kay MV, Pfeifer KJ, et al. Effect of whole-body CT during trauma resuscitation on survival: a retrospective, multicentre study. Lancet 2009;373:1455-61.
2. Wada D, Nakamori Y, Yamakawa K, Yoshikawa Y, Kiguchi T, Tasaki O, et al. Impact on survival of whole-body computed tomography before emergency Bleeding control in patients with severe blunt trauma. Crit Care 2013;17:R178. http://dx. doi.org/10.1186/cc12861.
3. Huber-Wagner S, Biberthaler P, Haberle S, Wierer M, Dobritz M, Rummeny E, et al. Whole-body CT in haemodynamically unstable severely injured patients-a retro- spective, multicentre study. PLoS One 2013;8:e68880. http://dx.doi.org/10.1371/ journal.pone.0068880.
4. Adiotomre A, Chopra A, Kirwadi A, Kotnis N. Results from the first year as a major trauma radiology unit in the UK. Clin Radiol 2014;69:812-21.
5. Venugopal K, Kinghorn AF, Emordi CE, Atkinson PR, Kendall RJ. An evaluation of the use of whole-body computed tomography in trauma patients at a United Kingdom trauma centre. Eur J Emerg Med 2012;19:193-5.
6. Gohil I, Vilensky JA, Weber EC. Vacuum phenomenon: clinical relevance. Clin Anat 2014;27:455-62.
7. Ito H, Yoshikawa T, Hayashi N, Ohtomo K. MDCT demonstration of intraarticular gas in the Glenohumeral joint and sternoclavicular joint with reference to arm position. Radiat Med 2008;26:422-6.
8. Obinata M, Omori K, Ishikawa K, Osaka H, Oode Y, Yanagawa Y. Significance of pneumorrhachis detected by single-pass whole-body computed tomography in pa- tients with trauma. J Emerg Trauma Shock in press.
9. Stabler A, Beck R, Bartl R, Schmidt D, Reiser M. Vacuum phenomena in insufficiency fractures of the sacrum. Skeletal Radiol 1995;24:31-5.
10. Peh WC, Ooi GC. Vacuum phenomena in the sacroiliac joints and in association with sacral Insufficiency fractures. Incidence and significance. Spine (Phila Pa 1976) 1997; 22:2005-8.
11. Sze MJ, Mulligan MJ. Reliability of vacuum phenomenon in the sacroiliac joint as a sign of traumatic injury. Emerg Radiol 2002;9:100-2.
12. Patten RM, Dobbins J, Gunberg SR. Gas in the sternoclavicular joints of patients with blunt chest trauma: significance and frequency of CT findings. AJR Am J Roentgenol 1999;172:1633-5.
13. Abraham DJ, Vaccaro AR, Albert TJ, Cotler JM. Gas in the spinal canal associated with injury of the cervical spinal cord: a Diagnostic dilemma. A case report. J Bone Joint Surg Am 1997;79:591-3.
14. Fairbairn KJ, Mulligan ME, Murphey MD, Resnik CS. Gas bubbles in the hip joint on CT: an indication of recent dislocation. AJR Am J Roentgenol 1995;164: 931-4.
15. Nakayama E, Tabata O, Oobu K, Kanda S. Gas phenomenon in the superior space of the temporomandibular joint: report of a case. J Oral Maxillofac Surg 2004;62: 107-11.
16. Yanagawa Y, Fukazawa S, Imamura T. Importance of magnifying CT images to detect minute Free air in patients with bluntbowel injury. Emerg Med J 2011;28:196.
17. Hogler W, Blimkie CJ, Cowell CT, Inglis D, Rauch F, Kemp AF, et al. Sex-specific deve- lopmental changes in muscle size and bone geometry at the femoral shaft. Bone 2008;42:982-9.
18. Miller AE, MacDougall JD, Tarnopolsky MA, Sale DG. gender differences in strength and muscle fiber characteristics. Eur J Appl Physiol Occup Physiol 1993;66:254-62.
19. Tzanakis I, Eskin DG, Georgoulas A, Fytanidis DK. Incubation pit analysis and calcula- tion of the hydrodynamic impact pressure from the implosion of an acoustic cavitation bubble. Ultrason Sonochem 2014;21:866-78.
20. Sampson MA, Colquhoun KB, Hennessy NL. Computed tomography whole body imaging in multi-trauma: 7 years experience. Clin Radiol 2006;61:365-9.
21. Rust RJ, Kopecky KK, Holden RW. The triangle sign: a CT sign of intraperitoneal fluid.

    Gastrointest Radiol 1984;9:107-13.

    Sierink JC, Saltzherr TP, Beenen LF, Luitse JS, Hollmann MW, Reitsma JB, et al. A multicenter, randomized controlled trial of immediate total-body CT scanning in trauma patients (REACT-2). BMC Emerg Med 2012;12:4. http://dx.doi.org/10. 1186/1471-227X-12-4.