Operative management of fragility fractures of the pelvis – a systematic review

The rising incidence of osteoporosis has brought significant challenges regarding the management of associated low energy injuries. In the past, high energy pelvic fractures associated with major trauma have predominated, however the incidence of low energy fragility fractures of the pelvis (FFP) is increasing and predicted to continue rising in the future [1‐3].

High energy injuries often involve both the bony pelvis and associated ligaments and soft tissues. In contrast, low energy FFP are more commonly characterised by injuries affecting only the weaker osteoporotic bone. This observation is the basis for Rommen’s classification which divides FFP into four groups of increasing instability ranging from isolated pubic rami injuries to complex bilateral displaced sacral injuries [4].

Historically the management of FFP in the elderly has involved a brief period of bed rest and analgesia followed by return to mobilisation as pain allowed [5, 6]. Despite this, significant pain caused by the injury [7, 8] often leads to prolonged immobility [9]. The consequences of this in a vulnerable population are significant. The mortality rate is similar to hip fractures, there is significant associated morbidity, prolonged inpatient stay and patients often require temporary or permanent admission to nursing homes [3, 10‐15]. As with other debilitating osteoporotic fractures, this injury carries with it a significant financial burden [16] and results in a significant reduction in patient reported quality of life [17].

Some centres now advocate surgical management of these injuries to improve outcomes [18, 19]. The aim of this systematic review is to evaluate evidence for surgical management of FFP. This includes fracture patterns encountered as well as fixation methods and timing of interventions. We also aimed to identify whether these interventions improve pain and other outcomes such as mobility, length of stay, quality of life and mortality compared to conservatively managed patients.

This systematic review was registered with the International prospective register of systematic reviews (PROSPERO) Centre for Reviews and Dissemination, University of York and conducted in accordance with PRISMA guidelines [20]. A clinical librarian searched the followed databases; NHS Evidence, TRIP, the Cochrane Database of Systematic Reviews, MEDLINE and EMBASE. Bibliographic database searches were conducted using the NHS Evidence Healthcare Databases Advanced Search platform. The final search was conducted on 4th May 2020. Relevant natural language and controlled vocabulary terms were selected and combined, and final results reviewed. There were no language restrictions in the initial search, however studies not written in English were excluded.

All studies with fragility fractures of the pelvic ring including sacral insufficiency fractures were included in patients 60 years or older. Studies were excluded if they contained patients with high-energy injuries (defined as a fall greater than from standing height), or a pathological fracture from a cause other than osteoporosis. We also excluded patients managed with sacroplasty alone although included patients with augmentation of sacroiliac joint (SIJ) screws with bone cement or other void filler. Acetabular fractures and isolated iliac crest fractures were excluded as were studies with fewer than ten patients.

Abstracts were analysed by two authors (DW and JK). Relevant abstracts were selected for full text review and inclusion where relevant. Any disagreement on studies to include was decided by discussion and with the senior author (MR). A flowchart of the study selection can be seen in Fig. 1. Randomised and non-randomized studies were assessed for bias using The Cochrane Collaboration’s tool for assessing risk of bias [21, 22] and case series assessed by the Joanna Briggs Institute checklist for case series [23]. Studies were assessed by two authors (DW and JK) and any disagreements resolved by discussion with a third (MR). A narrative synthesis was then performed using these tools. Main outcomes were pain scores, quality of life, mobility, length of stay and mortality. Secondary outcomes were complications including re-operations, failure of fixation, neurological deficits, and infection. No meta-analysis was planned.

Seventeen studies were identified with 766 patients, of which 463 were managed operatively. One randomised study and three studies with non-operative comparison groups were identified. The remaining studies were case series. A summary of the results is presented in Table 1.

Traditionally FFP have been managed non-operatively, often with a period of bed rest with analgesia and then mobilisation [5, 40]. Knowledge regarding the detrimental effect of bed rest and immobility in the elderly population has been present for decades [41]. Even short periods of immobilisation or decreased mobility lead to prolonged loss of muscle strength [42]. As a consequence there is reduced capacity to perform activities of daily living leading to loss of independence [43, 44]. Mortality from non-operatively managed pelvic fractures is also similar to matched patients treated for hip fracture [45].

Elderly patients are unable to comply with restricted weightbearing regimes [46] – any management strategy should facilitate immediate full weight bearing with sufficient control of pain. This review highlights that whilst the quality of evidence for surgical fixation of FFP is poor, there is consistent data to support the idea that the majority of fracture patterns can be stabilised through minimally invasive methods, with a consequent reduction in pain. Whether this assertion is correct or if it translates to outcomes with regards mobility, length of stay, quality of life and mortality needs further investigation.

A heterogenous selection of fracture classifications were used by the studies, which made comparisons difficult. FFP often do not fit well into the OTA, Tile or Young and Burgess classifications as elderly patients tend to have bony injuries rather than ligamentous involvement [4, 8]. Recognising this, the Rommen’s classification was developed specifically for these injuries and may be more useful in classifying these injuries in the future [47].

The decision to pursue operative intervention in most of the studies was made after failure of conservative measures rather than based on fracture morphology. Whilst some unstable fractures may mandate surgical intervention, the commonest minimally displaced posterior injuries and even some more extensive and displaced injuries may be managed successfully with conservative measures. Conservative management was normally instituted for a period of 3–7 days although it did range up to 6 months. Given that relatively short periods of immobility can lead to significant morbidity we would suggest considering intervention after 48 h of failed conservative management and ideally within 1 week to avoid complications associated with prolonged immobility.

With regard to fixation techniques, when stabilising the posterior ring, the majority of authors utilised percutaneous posterior fixation where possible. Biomechanical studies have shown that two SIJ screws, a single augmented SIJ screw and a single transacral screw offer similar levels of stability and are all superior to single SIJ screw fixation in osteoporotic models [48‐52]. There is some clinical evidence identified in this review that highlights a significantly increased rate of screw loosening when one unaugmented SIJ screw was used in isolation [29]. This finding was independent of whether or not the anterior ring was stabilised. We would suggest that a single un-augmented SIJ screw may not provide sufficient stability in the osteoporotic patient to facilitate early full weightbearing and adequate pain relief. Either multiple screws, longer transacral screws, or screw augmentation should be considered. Surgeons should be mindful of the potential additional risks and benefits of each option.

Supplementary anterior fixation was sporadic if present and there was no correlation between lack of anterior fixation and overall fixation failure. Tile noted that the posterior ring contributes around 60% to pelvic stability [53] and Matta confirmed that even in unstable pelvic injuries rami fractures did not require stabilization by internal or external fixation when the posterior ring was stabilised [54]. The data presented here would support the assertion that posterior fixation in these injuries is more critical than anterior fixation. Stabilising the anterior ring contributes to overall pelvic stability and this may understandably be desirable to surgeons managing osteoporotic FFP. Percutaneous screw fixation of the ramus is an attractive option in minimally displaced fractures, however there is some biomechanical evidence in osteoporotic bone that plate fixation is superior to percutaneous retrograde screw fixation, the trade-off being that this requires an open approach [55]. External fixation is a quick and relatively easy technique but its use in the osteoporotic patient raises concerns with regard to pin-site infection, loosening and patient acceptance [4]. The Infix is a newer development and current trials are ongoing to identify whether it is a suitable method for stabilising osteoporotic type 1 lateral compression fractures [56], however its use is also not without complications [57]. A large study of the German pelvic database comparing Tile B and C type pelvic fractures with the anterior component involving the obturator foramen showed a higher rate of complications from more extensive anterior and posterior surgery compared to posterior stabilization alone with no difference in fixation failure or mortality [58]. When electing to stabilise FFP, surgeons should be mindful that supplementary anterior fixation may not be necessary with a robust posterior fixation. If electing to stabilise the anterior ring, surgeons should be mindful of potential complications associated with the chosen method. Anterior fixation alone may be appropriate depending on the strength of fixation and degree of posterior instability [7].

Observed complications from surgical interventions were low. Most studies used either computed tomography (CT) guidance or fluoroscopy to insert screws. Incorrect screw placement was noted with both techniques but was infrequent and any iatrogenic neurological deficits resolved after removal and repositioning [6, 14, 19, 29].

We noted a previous systematic review of four studies on the effectiveness of surgical fixation of osteoporotic LC1 fractures. This found insufficient evidence to support guidance on the most effective treatment for patients sustaining this injury however where reported, mobility and function did improve after surgery [59]. This systematic review is a more comprehensive overview of 17 studies encompassing all types of FFP.

This systematic review does have some weaknesses. Firstly there was a lack of good quality randomised studies on which to base conclusions and most studies suffered from significant risk of bias. There was also significant heterogeneity in fracture classification and outcome measures. This made appropriate data pooling and analysis between studies difficult. We also excluded trials with less than 10 patients and excluded papers where an English translation could not be obtained. This only resulted in the loss of a small number of cases and therefore was unlikely to change the overall impression of the review.

This systematic review set out to identify the evidence for surgical management of FFP. Overall the quality of evidence was low. Of the 17 studies identified only one was randomised but with high risk of bias. Three studies were identified with a non-operative comparator group but all suffered from significant bias. Despite this, consistent improvement in pain and mobility was noted with stabilisation of the pelvis, which was most often performed percutaneously to the posterior ring. Anterior fixation of the pelvic ring was often absent and variable techniques were used when present. More than one SIJ screw posteriorly, longer transacral screws or screw augmentation offer more robust fixation than single SIJ screws for stabilising the pelvis, especially in the context of osteoporosis. Surgeons should consider operative stabilisation of FFP after a brief period of conservative management to avoid morbidity associated with immobility.