Pelvic ring fractures are associated with high-energy trauma, usually occurring during road traffic collisions (RTCs). Their true incidence is unknown, with estimates ranging from one in 30 to one in 80 000 of the UK population (Bircher and Giannoudis, 2004). Unstable pelvic ring injuries are associated with high mortality and morbidity.
Pelvic binders or circumferential compression devices (PCCDs), also referred to as binders, are adjuncts used mainly in prehospital care to reduce deaths from life-threatening haemorrhage occurring in pelvic injuries (Spanjersberg et al, 2009). First commercially introduced in 1999 (Vermeulen et al, 1999), they work by reducing pelvic ring displacement; this not only minimises bleeding from fractured bone ends but also decreases the effective volume of the pelvis into which haemorrhage can discharge (Spanjersberg et al, 2009). This is especially true in anteroposterior compression (APC) or ‘open book’ injuries (Bottlang et al, 2002).
Biomechanical cadaveric studies have found that placement of the device at the level of the trochanters provides the greatest reduction in diastasis with the lowest force application (Bottlang et al, 2002; DeAngelis et al, 2008).
Despite the evidence supporting placement of the devices at the level of the greater trochanters, studies have shown that consistent placement in the field, often in high-stress situations, is poor (Bonner et al, 2011; Vaidya et al, 2016).
Placement of PCCDs in civilian populations in the UK has not been examined in the literature. The aim of this study is to investigate placement of binders in patients arriving at a single trauma centre in England and to assess whether they are being placed optimally at the level of the trochanters.
Methods
Major trauma cases attending the University Hospitals of Coventry and Warwickshire (UHCW) were identified from the Trauma Audit and Research Network (TARN) database. The time periods assessed were 6 months in total: from July to August in 2014 (234 patients) and from November 2015 to February 2016 (392 patients). The earlier sample was taken first, analysed and, when a potential problem with malpositioning of binders was identified, the sample size was increased to cover 6 months' worth of patients.
All patients' clinical imaging was reviewed using Insight PACS (Picture Archiving and Communication System, Insignia Medical Systems, Basingstoke, UK). Patients were identified as having a PCCD using the scout image from the computerised tomography (CT) trauma scan. Each of these was then independently analysed by two orthopaedic registrars to judge the position of PCCD, with resolution of any disagreements at a separate occasion.
Binder position was assessed using a standardised protocol according to the relationship of the PCCD to the trochanters, as per Bonner et al (2011). The centre of the buckle/straps visible on the imaging was assumed to be the centre of the binder. If the centre was within the bounds of the top of the greater trochanter and the bottom of the lesser trochanter, it was ‘centred’. Above the greater trochanter was deemed ‘high’ and below the lesser trochanter was ‘low’ (Figure 1).
Demographics such as age and sex were collected, along with PCCD type. Data on the mechanism of injury and actual injuries identified (including their presence, type and number) were collected in the patients sampled in 2015–2016. It was postulated that multiple injuries or a disrupted pelvis/acetabulum might make accurate application of a PCCD more difficult.
Results
In total, 89 (14%) of the 626 people presenting as major trauma patients to UHCW had a PCCD in situ on imaging. Of these, 21 were female and 68 were male (1:3.2). Their mean age was 39 years (range: 11–83 years).
Three different PCCDs were identified as being used by the prehospital teams: the SAM Pelvic Sling™ II (SAM Medical Products, Wilsonville, US); the T-POD® (Pyng Medical Corporation, British Columbia, Canada); and the Prometheus Pelvic Splint™ (Prometheus Medical, Hereford, UK).
Overall, 52 (59%) PCCDs were centred, 28 (31%) were too high and the remaining 8 (10%) were too low. Table 1 shows positioning by PCCD type.
| Prometheus Pelvic Splint | SAM Pelvic Sling II | TPOD | Overall | |
|---|---|---|---|---|
| High | 4 (26%) | 14 (38%) | 10 (27%) | 28 (31%) |
| Centred | 10 (67%) | 19 (51%) | 24 (65%) | 53 (59%) |
| Low | 1 (7%) | 4 (11%) | 3 (8%) | 8 (10%) |
A χ2 test, with the null hypothesis that differences between observed binder positions (by type) were independent of binder brand—i.e. that the observed difference between position of PCCD in each binder group was because of chance alone—was performed. The χ2 for the Prometheus Pelvic Splint was 0.488 (P=0.784), for the SAM Pelvic Sling II™, χ2 was 0.024 (P=0.622) and the TPOD had a χ2 of 0.132 (P=0.764). Therefore, although there was a trend for higher numbers of poorly positioned PCCDs in the SAM Pelvic Sling II group, this was not statistically significant (P>0.05).
Of the 392 patients in the 2015–2016 group, 22 had pelvic ring fractures. Of these, 10 (45%) had a PCCD in position. The pelvic ring fractures identified were mainly of the lateral compression type (LC) (19/22) and the remainder were APC (3/22) according to the Young and Burgess classification (Burgess et al, 1990). Of the patients who had both a pelvic fracture and a PCCD in situ, as with the overall findings, most incorrectly positioned PCCDs were placed too high. (Table 2).
| Pelvic ring fracture | Overall | |
|---|---|---|
| High | 2 (20%) | 28 (31%) |
| Centred | 7 (70%) | 53 (59%) |
| Low | 1 (10%) | 8 (10%) |
The majority of patients in the second, later cohort were involved in randomised controlled trials (RTCs), with the next most common mechanism of injury being a fall from a height (of >2 m and <2 m) and the remainder being sustained crush injuries. The position of the PCCD according to mechanism of injury is shown in Table 3.
| RTC | Fall >2 m | Fall <2 m | Crush | Overall | |
|---|---|---|---|---|---|
| High | 8 (26%) | 0 | 1 | 2 (67%) | 28 (31%) |
| Centred | 21 (68%) | 3 (60%) | 0 | 1 (33%) | 53 (59%) |
| Low | 2 (6%) | 2 (40%) | 0 | 0 | 8 (10%) |
Injury patterns were divided into:
The relationship between injury pattern and binder position is shown in Table 4.
| Pelvic ring fracture | Acetabular fracture | Head injury | Chest injury | Abdominal injury | Spinal injury | Long bone fracture | Overall | |
|---|---|---|---|---|---|---|---|---|
| High | 3 (27%) | 2 (50%) | 3 (23%) | 6 (31%) | 2 (33%) | 2 (12.5%) | 4 (31%) | 28 (31%) |
| Centred | 7 (64%) | 1 (25%) | 8 (62%) | 10 (53%) | 3 (50%) | 12 (75%) | 7 (54%) | 53 (59%) |
| Low | 1 (9%) | 1 (25%) | 2 (15%) | 3 (16%) | 1 (17%) | 2 (12.5%) | 2 (15%) | 8 (10%) |
The number of injuries identified compared to position of PCCD are shown in Table 5, but the numbers were too small to be of any significance.
| 1 | 2 | 3 | 4 | 5 | 6 | Overall | |
|---|---|---|---|---|---|---|---|
| High | 4 (25%) | 3 (25%) | 1 (17%) | 2 (40%) | N/A | 0 | 28 (31%) |
| Centred | 11 (29%) | 9 (75%) | 4 (66%) | 1 (20%) | N/A | 1 | 53 (59%) |
| Low | 1 (6%) | 0 | 1 (17%) | 2 (40%) | N/A | 0 | 8 (10%) |
Discussion
Results demonstrate that, compared with previous studies examining PCCD position, overall placement was found to be superior by almost 10%. The majority of incorrectly placed binders in both studies were positioned too high (Table 6).
| Present study | Bonner et al (2011) | |
|---|---|---|
| High | 28 (31%) | 65 (39%) |
| Centred | 53 (59%) | 83 (50%) |
| Low | 8 (10%) | 19 (11%) |
From this study, it does not appear that the presence of a pelvic fracture, mechanism of injury, injury pattern or number of associated injuries has any significant effect on PCCD malpositioning. It may be that overall patient numbers are too small to reach statistical significance when tested using a χ2 test. However, the percentage of patients with pelvic ring injuries who were given a PCCD was similar to that in published literature; this study showed 11/22 (50%) were given PCCDs compared with 47% (55/112) in the study by Vaidya et al (2016). It should be noted, however, that their sample was larger than the one in the current study.
It is interesting that binders were mainly applied in the absence of a pelvic injury, with only 26% of PCCDs fitted in the presence of a pelvic ring disruption. The data explaining the rationale of placement were not available retrospectively, but it is normal practice among most prehospital professionals to apply them in cases where there are clinical findings associated with pelvic injury (such as tenderness, pain, bruising and leg length discrepancy) or a high index of suspicion. Despite this, 55% of patients who did sustain pelvic fractures had no PCCD in place. This is a major concern, giving the life-saving effect PCCDs can have. When patients are subdivided into those with a confirmed pelvic injury, only 70% had the PCCD correctly placed. Given that PCCDs are most effective in APC-type injuries, and two-thirds of patients with these injuries had no PCCD, there may be an issue with the rationale for application. It must be noted, however, that none of the patients who had a pelvic injury but no PCCD died.
This study does raise an interesting question over ease of placement of different binder styles. To the authors' knowledge, no previous study has compared different binder types in terms of correct placement over trochanters. It would appear from the results that equal numbers of SAM Pelvic Sling IIs and TPOD PCCDs were used (37 each). Almost half (58%/18 patients) of SAM Pelvic Slings II were placed incorrectly (mostly too high) compared with more than one-third (35%/13 patients) of TPODs. It is difficult to compare these figures with the Prometheus Pelvic Splint as far fewer of these were applied, but the percentages incorrectly placed were more comparable to TPOD, with a third incorrectly placed (33%/5 patients).
This may be because of the inherent difference in the design of these PCCDs; the TPOD and the Prometheus Pelvic Splint are both wider (top to bottom) than the SAM Pelvic Sling II. The authors looked at whether the centre of the binder passed between the trochanters, hopefully reducing any observation bias from the width of the binder. They hypothesise that perhaps the SAM Pelvic Slings II may be harder to apply correctly over the trochanters in the field than the other two, but this would need to be independently verified. Confounding factors, such as regional differences regarding prehospital training, could also be responsible.
Having two registrars independently verifying PCCD placement was intended to minimise the effect of any bias in terms of assessing binder position on imaging. It is, however, possible that the binder's position may have changed before imaging. It is trust policy at UHCW to move a PCCD if it is obviously malpositioned. However, in the authors' experience, imaging is normally arranged swiftly after the primary survey so, unless a PCCD was very poorly positioned, it is unlikely many were changed. With retrospective analysis, this kind of repositioning would be impossible to rule out. It is not routine practice in the authors' trust to apply a binder in the emergency department (before scanning), unless there is haemodynamic instability and obvious signs of pelvic injury on primary survey. For this reason, the authors have assumed all PCCDs were applied by prehospital professionals. The trust keeps only the TPOD in stock, so all SAM Pelvic Sling IIs and Prometheus Pelvic Splints were applied before patients arrived at UHCW. Radiolucent PCCDs, such as sheets, are not usually used by the prehospital teams and could not have been assessed in this study.
Factors including patients' haemodynamic stability, level of consciousness (as measured by Glasgow Coma Scale) and body mass index (BMI) were not collected in this study. This is a potential limitation, as confusion or high BMI could make it more difficult to apply a PCCD over the trochanters.
Conclusions
The authors conclude that there has been a modest improvement in placement of PCCD since previous studies were carried out. However, 41% are still being positioned above or below the correct level.
This article highlights the need for further training for all health professionals involved in binder application. Perhaps simulation training, offering opportunities to practise placement in patients with varying BMIs or using actors playing distressed or unconscious patients may be helpful.
The present study did not have large enough numbers to associate mechanism of injury, pattern of injury or number and type of injuries with correct binder placement. Given that the patients who most need a correctly placed PCCD are haemodynamically unstable with disruption of the pelvic ring, the authors would hope that the presence of either of these is not linked to an increased likelihood of malpositioning.
Comparison between different commercially available PCCDs regarding accurate placement would be an interesting avenue for future research. Perhaps the most pertinent finding from this study is that most patients with a pelvic injury did not receive treatment with a PCCD before arrival in hospital. While none of these patients died, it raises an important question regarding the rationale for application—that practitioners should probably be applying PCCDs correctly and in more patients with trauma. The challenge is how to identify these patients more effectively and how to train staff to apply binders more accurately.