Advertisement

Critical review of the classification and treatment of pediatric pelvic fractures

Open AccessPublished:July 06, 2022DOI:https://doi.org/10.1016/j.injury.2022.07.009

      Highlights

      • Currently there is no consensus about which fracture pattern should be considered stable or unstable in pediatric pelvic fractures, or if these unstable fractures should be treated operatively.
      • Most recent literature suggests that the functional outcome of pediatric pelvic fractures is related to the degree of pelvic asymmetry seen during follow-up.
      • This review shows that there is a large heterogeneity in which fracture patterns are considered to be unstable or in need of surgical fixation.
      • The results of this review imply that the current classification systems are inadequate in guiding treatment decisions. Fracture patterns might be wrongly classified as stable or treated non-operatively, resulting in poor functional outcomes.

      Abstract

      Background

      Historically most pediatric pelvic fractures were treated non-operatively because of the presumed potential of the pediatric pelvis to remodel and the subsequent increased fracture stability. Currently a wide variety of classifications in pediatric pelvic fractures is used to assess fracture stability and guide treatment, yet none have proven to be ideal since the structural behavior of the pediatric pelvis differs greatly from the adult pelvis. The aim of this review is to critically appraise the use of these different classification systems, fracture (in)stability, the treatment of pediatric pelvic fractures and how it reflects on long-term complications such as pelvic asymmetry and functional outcome.

      Methods

      A literature search was performed in Medline, Embase, Cochrane, PubMed, Google Scholar and references of the selected articles. Studies that reported on pain, leg length discrepancy (LLD), abnormal gait (GA), pelvic asymmetry, and functional outcomes of pediatric pelvic fractures were included.

      Results

      A total of six different classification systems were used, the most common were Tile (n= 9, 45%) and Torode and Zieg (n= 8, 40%). There was great disparity in treatment choice for the same type of fracture pattern, resulting in several pelvic ring fractures that were defined as unstable being treated non-operatively. Pelvic asymmetry is seen in rates up to 48% in non-operatively treated patients. In contrast, pelvic asymmetry in surgically fixated unstable pelvic fractures was rare, and these patients often showed excellent functional outcomes during follow-up.

      Conclusion

      There is a substantial heterogeneity in which fracture patterns are considered to be unstable or in need of surgical fixation. Functional outcomes seem to be correlated with the frequency of pelvic asymmetry and are likely due to an underestimation of the stability of the pelvic fracture. Taking into consideration the force that is necessary to cause a facture in the pediatric pelvis, a fracture of the pelvic ring alone could be suggestive for instability. The results of this review imply that the field of pediatric pelvic surgery is currently not grasping the full scope of the complexity of these fractures, and that there is a need for a pediatric pelvic classification system and evidence-based treatment guideline.

      Keywords

      Introduction

      Pediatric pelvic ring fractures are rare but potentially life-threatening injuries, especially open pelvic fractures [
      • Galano G.J.
      • et al.
      The most frequent traumatic orthopaedic injuries from a national pediatric inpatient population.
      ]. With an estimated incidence of one per 100.000 children per year, they only make up for 0.2% of all pediatric fractures [
      • Kruppa C.G.
      • et al.
      Pediatric pelvic ring injuries: how benign are they?.
      ].
      In the past, non-operative treatment was deemed sufficient to treat most pelvic fractures in children because of the presumed potential of the pediatric pelvis to remodel and the increased stability due to the thick periost despite disruption of the pelvic ring [
      • Gansslen A.
      • Heidari N.
      • Weinberg A.M.
      Fractures of the pelvis in children: a review of the literature.
      ]. While the non-operative treatment method is suitable for stable pelvic fractures, e.g. avulsion, iliac wing, or isolated pubic branch fracture, more recent studies show that children who are treated non-operatively for a pelvic fracture with disruption of the pelvic ring could suffer from pelvic asymmetry, especially in the case of plastic deformity [
      • Amorosa L.F.
      • Kloen P.
      • Helfet D.L.
      High-energy pediatric pelvic and acetabular fractures.
      ,
      • Guillaume J.M.
      • et al.
      Pelvic fractures in children (pelvic ring and acetabulum).
      ]. Nowadays, literature suggests that if a pelvic ring fracture in children is assessed as mechanically unstable, surgical stabilization of the fracture is indicated as it leads to less pelvic asymmetry and therefor better functional results in the long-term [
      • Schwarz N.
      • et al.
      Long-term results of unstable pelvic ring fractures in children.
      ].
      Currently a wide variety of classifications is used to assess the need for surgical fixation in pediatric pelvic fractures. One of the most frequently used classifications, the Torode and Zieg is intended as a guide to assess the likelihood of associated injuries and the expected outcome, e.g. survival, of the injured child. This demonstrates that it should not guide the indication for surgical fixation of the pelvic fracture as this is dependent on mechanical instability. Another frequently used classification, the Tile classification, focuses on biomechanical instability, but does not take skeletal maturity into consideration as it is based on the characteristics of the adult pelvis. Tile himself states that both anterior and posterior types of lesions must occur in all cases of major pelvic trauma. The pediatric pelvis however has the unique ability to break in a single fracture of the ring. This is due to the pediatric pelvis being more elastic due to the relatively high percentage of cartilage [
      • Gansslen A.
      • et al.
      Pelvic ring injuries in children. part I: epidemiology and primary evaluation. a review of the literature.
      ]. Furthermore, the pediatric pelvis can absorb extremely large amounts of energy before it breaks, in contrast to adults [
      • Amorosa L.F.
      • Kloen P.
      • Helfet D.L.
      High-energy pediatric pelvic and acetabular fractures.
      ,
      • Guillaume J.M.
      • et al.
      Pelvic fractures in children (pelvic ring and acetabulum).
      ,
      • Zwingmann J.
      • et al.
      Pelvic fractures in children results from the German pelvic trauma registry: a cohort study.
      ]. This suggests that the structural behavior of the pediatric pelvis is very different than that of the adult pelvis, so the Tile classification has limited usefulness in the pediatric population. Yet, this classification is still widely used in both research and clinical practice to determine the need for surgical fixation of pediatric pelvis fractures.
      The aim of this review is to provide insight into which classification systems for pediatric pelvic fractures are used in the literature, and how they relate to the provided treatment of these fractures and the long-term complications, such as pelvic asymmetry, leading to LLD and overall functional outcome.

      Material and methods

      Search strategy

      A literature search was performed by two authors (IZ, AM) of the English literature, published between 1990 and February 2022 using the databases of Embase, PubMed, Medline, Cochrane, and Google Scholar (Fig. 1). Keywords and index terms related to outcomes of pelvic ring fractures in children were used. The terms were kept broad and synonyms were used to maximize sensitivity. The reference lists of all relevant articles and narrative reviews were reviewed to find additional suitable publications. This review is fully compliant with the PRISMA guidelines [
      • Moher D.
      • et al.
      Preferred reporting items for systematic reviews and meta-analyses: the prisma statement.
      ].

      Eligibility criteria

      All prospective and retrospective cohort studies, case control studies and case series enrolling children (<16 years old) with operatively or non-operatively treated pelvic fractures and reported radiographic outcome, functional outcome, or complications at follow up were reviewed. Studies that focused on pelvic fractures but also described concomitant acetabular fractures were also included in this study. Case studies, studies with a mixed age population with no possible distinction between patients under the age of 16 and adults, studies that solely focused on acetabular or sacral fractures, and studies in languages besides English without translation available were excluded. Studies published before 1990 or with a loss to follow-up of >95% were also excluded.

      Data extraction

      The primary objective was to determine the radiographic and functional outcomes of pelvic ring fractures in children. The following outcomes were reviewed; pain, leg length discrepancy (LLD), restricted range of motion (RROM), abnormal gait (GA), pelvic asymmetry, functional outcomes, neurological deficits, and urogenital deficits. When an article used two different classifications in the same article to classify the pelvic fractures, the results of that article were included in the data analysis of both type of classifications.

      Methodologic quality and risk of bias assessment

      We included twenty studies of which the critical appraisal and level of evidence was assessed, the detailed scores are displayed in appendix 1, and appendix 2. The methodological validity was assessed using the ROBINS-I tool [
      • Sterne J.A.
      • et al.
      ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions.
      ]. Some studies included avulsion fractures of the pelvis in their analysis. Since these injuries have a uniformly good prognosis and should be seen as a different entity than other pelvic fractures, these studies were considered to have a moderate risk of bias due to confounding. When pelvic instability was classified using a combination of radiographic and clinical examination this was considered to be a low risk of bias, when only radiographic files were used there were some concerns about risk of bias. Loss to follow-up percentage of <5% was scored as low risk of bias, 6-20% was scored as moderate risk of bias, and >20% was scored as serious risk of bias. When non-validated or self-reported subjective measurement tools were used to define outcomes, the study scored as moderate risk of bias, while validated questionnaires were considered to have a low risk of bias on the measurement outcome. No use of a measurement tool or method was considered to be a high risk of bias.

      Definitions

      For this analysis we labeled fracture stability conform the description or classification used in the articles, regardless of the authors own opinion on fracture stability. For example, if an article noted that they considered all Tile B fractures to be unstable, we included those numbers in the unstable category. If an article noted that they deemed some Tile B fractures stable and other unstable, depending on clinical findings, they were categorized accordingly.

      Results

      A total of 1421 article were identified through various sources and 20 of the articles met the eligibility criteria (Fig. 1). The most commonly used classification systems were Tile (n = 9, 45%) and Torode and Zieg (n = 8, 40%). A total of six different classification systems were used to guide treatment decisions, with multiple articles using two different classification systems to classify the same set of pelvic fractures.
      A total of 766 pediatric pelvic fractures were described in the 20 articles, of which 39% were defined as stable and 57% as unstable. One article describing 29 pediatric pelvic fractures did not define which fracture patterns they considered to be stable. A total of 504 (66%) pediatric pelvic fractures were treated non-operatively and 262 (34%) were treated with some type of internal or external surgical fixation.

      Tile

      Nine articles classified a total of 307 pediatric pelvic fractures using the Tile classification (Table 1). The majority were type B (50%) and type C (39%) fractures. Overall, 81% of the fractures were defined as unstable and 40% off all fractures were treated operatively (see Fig. 2). 44 (15%) patients had residual complaints of pain. Of the articles that described LLD and gait abnormalities, respectively 11% and 12% of the patients described residual complaints. Five articles described the number of patients that suffered from pelvic asymmetry during follow-up, which was 20 (12%) in total. Schwarz et al described the highest prevalence of pelvic asymmetry, namely 48%, which was seen in a cohort of non-operatively treated patients [
      • Schwarz N.
      • et al.
      Long-term results of unstable pelvic ring fractures in children.
      ]. Smith et al noted that pelvic asymmetry in the non-operatively treated pediatric pelvic fractures worsened over time, which was not objectified in the surgically fixated pelvic fractures [
      • Smith W.
      • et al.
      Clinical outcomes of unstable pelvic fractures in skeletally immature patients.
      ]. Functional outcome scores can be seen in Table 1. Blasier described only 1 (8%) fair or poor outcome in the operatively treated group, compared to 20% in the non-operatively treated group [
      • Blasier R.D.
      • et al.
      Disruption of the pelvic ring in pediatric patients.
      ].
      Table 1From left to right: first author and year of publication, number of pelvic fractures defined as stable, number of pelvic fractures defined as unstable, number of patients non-operatively treated per fracture type according to Tile, number of patients operatively treated per fracture type according to Tile, number and percentage of patients with pain complaints, number and percentage of patients with leg length discrepancy, number and percentage of patients with gait abnormalities, number and percentage of patients with pelvic asymmetry on radiographs, functional outcome measurements.
      Fig 2
      Fig. 2Treatment per fracture pattern according to the different classifications.

      Torode and Zieg

      The Torode and Zieg classification was used by eight articles (Table 2). More than half (n= 156, 53%) of the pediatric pelvic fractures was classified as type I/II/III, and 47% (n= 140) as type IV. 60% of the fractures were defined as unstable, yet only 23% of all patients underwent surgical fixation of their pelvic fracture (see Fig. 2). Pain and gait abnormality were the most common long-term complications seen in these studies, with 14% and 12% respectively. Pelvic asymmetry was described in three studies, with only the study of Rieger showing any patients with pelvic asymmetry (n = 2, 7%) [
      • Rieger H.
      • Brug E.
      Fractures of the pelvis in children.
      ]. Functional outcome scores varied greatly. Garvin et al described a small cohort of primarily non-operatively treated pediatric patients with pelvic fractures and found an incidence of 38% of persistent morbidity [
      • Garvin K.L.
      • et al.
      Pediatric pelvic ring fractures.
      ]. Furthermore, the study of Upperman et al showed great differences in FIM scores between stable and unstable fractures, with complete dependence during locomotion of 36% and 75% respectively [
      • Upperman J.S.
      • et al.
      Early functional outcome in children with pelvic fractures.
      ]. Notably, these results were in a cohort of patients that was treated non-operatively in 90% of the cases.
      Table 2From left to right: first author and year of publication, number of pelvic fractures defined as stable, number of pelvic fractures defined as unstable, number of patients non-operatively treated per fracture type according to Torode and Zieg, number of patients operatively treated per fracture type according to Torode and Zieg, number and percentage of patients with pain complaints, number and percentage of patients with leg length discrepancy, number and percentage of patients with gait abnormalities, number and percentage of patients with pelvic asymmetry on radiographs, functional outcome measurements.

      AO/OTA

      Four articles used the AO/OTA classification to guide treatment choices in a total of 126 pediatric patients with a pelvic fracture (Table 3). The vast majority of these fractures (87%) were defined as unstable, and all these unstable fractures were treated operatively (see Fig. 2). Kruppa et al described a cohort of patients with both stable and unstable pelvic fractures, of which half was operatively fixated [
      • Kruppa C.G.
      • et al.
      Pediatric pelvic ring injuries: how benign are they?.
      ]. Radiographic follow-up of these patients showed relatively high rates of pain complaints (39%) and pelvic asymmetry (46%), yet all patients returned to full activities of daily living. Both Guimarães et al and Scolaro et al noted no pelvic asymmetry during follow-up in a cohort of surgically fixated unstable pelvic fractures, with excellent functional outcomes during follow-up and a 100% return to school and desired activities rate [
      • Guimaraes J.A.
      • et al.
      Surgical treatment for unstable pelvic fractures in skeletally immature patients.
      ,
      • Scolaro J.A.
      • Firoozabadi R.
      • Routt M.L.C.
      Treatment of pediatric and adolescent pelvic ring injuries with percutaneous screw placement.
      ].
      Table 3From left to right: first author and year of publication, number of pelvic fractures defined as stable, number of pelvic fractures defined as unstable, number of patients non-operatively treated per fracture type according to AO/OTA, number of patients operatively treated per fracture type according to AO/OTA, number and percentage of patients with pain complaints, number and percentage of patients with leg length discrepancy, number and percentage of patients with gait abnormalities, number and percentage of patients with pelvic asymmetry on radiographs, functional outcome measurements

      Young and Burgess

      Only one study used the Young and Burgess classification to describe a total of 18 pediatric pelvic fractures (see Table 4). They defined 8 (44%) fractures as unstable, regardless of classification, and all of them were operatively treated (see Fig 2). Incomplete fracture reduction resulting in leg length discrepancy was seen in up to 50% of all patients, with high rates of pain (50%) and restricted range of motion (67%). Furthermore, gross pelvic asymmetry was seen in 17% of the pelvic fractures. This resulted in 34% of the patients experiencing limited or complete inability to participate in sports.
      Table 4From left to right: first author and year of publication, number of pelvic fractures defined as stable, number of pelvic fractures defined as unstable, number of patients non-operatively treated per fracture type according to Young and Burgess/Key and Conwell/anatomic distribution, number of patients operatively treated per fracture type according to Young and Burgess/Key and Conwell/anatomic distribution, number and percentage of patients with pain complaints, number and percentage of patients with leg length discrepancy, number and percentage of patients with gait abnormalities, number and percentage of patients with pelvic asymmetry on radiographs, functional outcome measurements. SBE = Single bony element, MBE = Multi bony elements, MF = Malgaigne fracture, TF = Triradiate fracture.

      Key and Conwell

      The Key and Conwell classification was used in one article to describe 120 pediatric pelvic fractures (see Table 4). Almost all fractures were defined as stable (94%) and treated non-operatively. Chia et al only described leg length discrepancy as long-term outcome which was present in 5 (4%) patients. Remarkably, 3 of the patients with LLD were classified as isolated fractures not involving the pelvic ring (type I).

      Other

      Nierenberg et al classified the pelvic fractures in their case series according to anatomic distribution (see Table 4). Thirty percent of the fractures were considered to be unstable, and all fractures were treated non-operatively. This resulted in a 20% rate of radiographic pelvic asymmetry during follow-up. However, functional outcome was excellent or good in 95% of the cases.

      Discussion

      In this review there was great variety in the type of fractures that were considered to be stable or unstable between studies, even when comparing studies that used the same type of classification system. This shows that there is currently no consensus on which fracture patterns are considered stable or unstable in the pediatric pelvis. While several articles state that unstable ring fractures in children are rare, this statement seems controversial taking the lack of an appropriate classification system and the biomechanics of the pediatric pelvis into consideration. Biomechanical studies show that forces up to 6000 N are needed to cause a pelvic fracture in children up to 14 years old. In one-year-old children up to 10.000 N did not result in a fracture, but in plastic deformation [
      • Stuhler T.
      • et al.
      [Pelvic fractures in children: clinic, late results, biomechanic (author's transl)].
      ]. In addition, experimental cadaveric studies show that lateral compression of the pelvis of up to 55% is tolerated in children before a fracture occurs [
      • Ouyang J.
      • et al.
      Experimental cadaveric study of lateral impact of the pelvis in children.
      ]. So, the prevalence of one or multiple fractures in a pediatric pelvis is itself a signal of the immense force the pelvis had endured and the displacement that has occurred before the breakage.
      Children have tense ligaments and increased laxity in the joints, which may result in little dislocation and a stable appearance of the fractures on initial imaging, even if there is more than one fracture through the pelvic ring or significant displacement has occurred during trauma [
      • Ouyang J.
      • et al.
      Experimental cadaveric study of lateral impact of the pelvis in children.
      ]. During follow-up, increasing displacement can be seen, retrospectively indicating instability and resulting in pelvic asymmetry. Therefore, lack of displacement in a pediatric pelvic fracture during initial presentation should not be seen as evidence for stability [
      • Schwarz N.
      • et al.
      Long-term results of unstable pelvic ring fractures in children.
      ]. Furthermore, Keshishyan et al showed that radiographic signs of pediatric pelvic fractures are limited [
      • Keshishyan R.A.
      • et al.
      Pelvic polyfractures in children. radiographic diagnosis and treatment.
      ]. When comparing post-mortem and radiographic data only 20% of fractures of the sacrum and 14% of SI-disruptions had convincing radiographic signs, which lead to high rates of discrepancy between clinical and post-mortem diagnoses. In our review we also found some signs of cases being misdiagnosed, for example Chia et al describes three patients with LLD that were classified as isolated fractures not involving the pelvic ring (type I) and treated non-operatively [
      • Chia J.P.
      • et al.
      Pelvic fractures and associated injuries in children.
      ]. Since these pelvic fractures caused LLD, it suggests that these fractures could be misclassified and/or undertreated, causing long-term complications.
      Signs of misclassification can also be found when comparing the number of associated injuries, urogenital and neurologic dysfunctions with the number of pelvic fractures that is classified as unstable. For example, Chia et al classify only seven pelvic fractures as unstable, while there are 20 patients with urogenital injuries [
      • Chia J.P.
      • et al.
      Pelvic fractures and associated injuries in children.
      ]. The sheering force causing these urogenital injuries only occurs in unstable pelvic fracture patterns, which suggests that some fracture patterns have been classified as stable, even though they might have been unstable.
      Many articles state that some degree of pelvic asymmetry in children can be accepted since the pediatric pelvis is capable of remodeling [
      • Garvin K.L.
      • et al.
      Pediatric pelvic ring fractures.
      ,
      • Bryan W.J.
      • Tullos H.S.
      Pediatric pelvic fractures: review of 52 patients.
      ,
      • Musemeche C.A.
      • et al.
      Selective management of pediatric pelvic fractures: a conservative approach.
      ,
      • Nierenberg G.V.G.
      • Bialik V.
      • Stein H
      Pelvic fractures in children: a follow up in 20 children treated conservatively.
      ,
      • Quinby Jr, W.C.
      Fractures of the pelvis and associated injuries in children.
      ]. In contrast to popular belief, Guimarães et al and Smith et al showed that the pelvic ring in fact did not remodel in any pediatric patient, regardless of the age at time of injury [
      • Smith W.
      • et al.
      Clinical outcomes of unstable pelvic fractures in skeletally immature patients.
      ,
      • Guimaraes J.A.
      • et al.
      Surgical treatment for unstable pelvic fractures in skeletally immature patients.
      ]. Thus, any form of displacement in pediatric pelvic fractures could become more problematic overtime. In our review, the incidence of pelvic asymmetry was seen in up to 48% in non-operatively treated patients. Several studies have shown that the functional outcome of pelvic ring fractures in children seems to depend mainly on the degree of asymmetry in the pelvic ring, therefore anatomical reduction and surgical fixation of an unstable pelvic fracture in a child seems to be the best treatment option to ensure a good functional outcome [
      • Schwarz N.
      • et al.
      Long-term results of unstable pelvic ring fractures in children.
      ,
      • Smith W.
      • et al.
      Clinical outcomes of unstable pelvic fractures in skeletally immature patients.
      ,
      • Hargitai E.
      • et al.
      Unstable pelvic fractures in children.
      ]. Keshishyan et al showed that when treating pelvic ring fractures non-operatively, pelvic symmetry was achieved in only some of the patients in their series and the results of this therapy were less than satisfactory for 50% of their cases [
      • Keshishyan R.A.
      • et al.
      Pelvic polyfractures in children. radiographic diagnosis and treatment.
      ]. In our review similar results can be seen, and some unsatisfactory outcomes found in our review may be partially attributable due to misdiagnosed unstable injuries, treated as if they were stable. Furthermore, Smith et al demonstrated that there was less pelvic asymmetry when treating pediatric pelvic ring fractures with combined anterior and posterior fixation compared to using only external fixation [
      • Smith W.
      • et al.
      Clinical outcomes of unstable pelvic fractures in skeletally immature patients.
      ]. This suggests that even with minimal radiographic posterior displacement there is instability of the ring, so only providing the pelvis with anterior stability does not seem to be sufficient. The discrepancy between minimal radiographic displacement and degree of instability in adults has previously been described by Sagi et al [
      • Sagi H.C.
      • Coniglione F.M.
      • Stanford J.H.
      Examination under anesthetic for occult pelvic ring instability.
      ]. This study showed that performing dynamic stress tests under anesthesia reveals occult instability in 50% of presumed APC-1 injuries, 39% of APC-2 injuries, and 37% of LC-1 injuries reflecting the often dynamic component in pelvic ring instability.
      Since pelvic fractures in children are such a rare entity most surgeons have little experience treating them which, in our experience, leads to surgeons often underestimating the injury. Also, Torode and Zieg is still frequently used in order to assess fracture stability even though this classification system is not intended for evaluating mechanical instability. Surprisingly, this review shows that the classification most recommended by pelvic experts worldwide in adults, the Young and Burgess classification, is seldom used when it comes to pediatric pelvic fractures. The Young and Burgess classification categorizes pelvic ring injuries taking the type and direction of force causing the fracture into consideration. A pediatric modification of the Young and Burgess classification might be beneficial to improve the correlation between classification and mechanical instability, to better guide treatment in the future.
      Not properly treating these injuries is especially problematic in children since they have to live with the poor functional outcome for many years to come. This review shows that we, orthopaedic trauma surgeons, need to be more aware of the gap of knowledge in the field of pediatric pelvic surgery, and we need to start by correctly classifying these injuries, creating more insight into when a fracture is stable or unstable, and implementing an evidence based or expert-opinion-based treatment guideline in order to improve the long-term outcome for these severely injured children.

      Limitations and strengths

      There are several limitations to our study. First, the findings of this review are limited by the retrospective design of the included studies which makes them prone to selection bias and the subjects are prone to confounding. Furthermore, follow-up times in the included studies varied greatly. The relative short follow up in some studies is problematic since some dysfunctions either improve or worsen over time
      The included studies are very heterogenic which makes comparing the outcomes of the studies extremely difficult. The fact that some studies included avulsion fractures in their results, a very benign subgroup of pelvic fractures with little to no associated injuries and long-term complications, leads to distorted results that can cause an underestimation of the severity of pelvic injuries in general. Another contributing factor might be the lack of an appropriate tool to measure functional outcome in paediatric patients who suffered a pelvic fractures. The only paediatric outcome score used in the articles we analysed was the weeFIM. To our knowledge there is no specific outcome score designed for paediatric pelvic fractures. The weeFIM has been validated for children with acquired or congenital disabilities, but might not be suitable when evaluating functional outcome specifically in paediatric patients who have suffered a pelvic fracture. Future research to develop a more tailored questionnaire would be advised so we can more accurately evaluate the functional outcome of these children.
      Furthermore, the complexity and the high incidence of concomitant injuries make it difficult to determine which injury is the cause of the poor outcomes; the pelvic fracture itself or its concomitant injuries.
      Future studies are needed to determine long-term outcomes per fracture and treatment type. Since these injuries are rare and complex, prospective studies are near impossible to conduct. Therefore, it might be necessary to create multinational expert groups to unify the treatment strategy for these injuries. Since these pelvic fractures occur in severely injured children with high chances of long-term problems with high impact on the child's future, centralization of pediatric trauma might be a very beneficial next step.

      Recommendations

      After evaluating the available literature, we can conclude that most authors agree on the following statements; fractures not compromising the stability of the pelvic ring, e.g. avulsion and isolated iliac wing fractures, rarely need operative stabilization. Unstable pelvic fractures with severe displacement benefit from surgical fixation. If surgical fixation is indicated, combined anterior and posterior fixation is advised since it seems to lead to better functional outcomes compared to using only external fixation in unstable fracture patterns. The indication for surgical fixation should not be influenced by the possibility of remodeling of the immature pelvis since this often does not happen, and residual displacement can lead to long-term dysfunction.
      We recommend that adult classifications of pelvic fractures should not be used to guide treatment in children who have suffered a pelvic fracture. Since the paediatric pelvis can absorb a greater amount of energy before it breaks than an adult's pelvis, and tense ligaments result in less dislocation on initial imaging, the mere occurrence of a fracture in the paediatric pelvis is indicative for some degree of pelvic instability.

      Conclusion

      Currently a wide variety of classifications is used to guide treatment decisions in pediatric pelvic fractures. None of the current classifications take the unique biomechanics of the pediatric pelvis into consideration. Furthermore, there is a large heterogeneity in which fracture patterns are considered to be stable or unstable, even within articles using the same classification system. Functional outcomes vary greatly within different cohorts of patients, but overall functional outcomes seem to be correlated with the frequency of pelvic asymmetry, which can partially be attributed to an underestimation of the fracture's instability. Taking into consideration the amount of force that is necessary to cause a fracture in the pediatric pelvis, an obvious fracture alone could be highly suggestive for instability. The results of this review imply that the field of pediatric pelvic surgery is currently not grasping the full scope of the complexity of these fractures, and that a tailored classification system taking the biomechanics of the pediatric pelvis into consideration should be developed. Furthermore, there is a need for an evidence-based treatment guideline to ensure that all children receive adequate care and to minimize long-term dysfunction.

      Authors contributions

      Conceptualization (AM, DE). Literature search and data curation (AM, IZ). Formal analysis and writing original draft (AM, IZ). Editing and visualization (AM, DE). Reviewing and final approval of the manuscript (AM, DE, FB, IZ).

      Funding

      The authors received no specific funding for this work.

      Availability of data and materials

      All data generated or analyzed during this study are included in this published article and its supplementary information files.

      Declaration of Competing Interest

      The authors have no competing interests.

      Appendix 1 Risk of bias included articles “traffic light” plots

      Appendix 2 Critical appraisal overall included articles

      References

        • Galano G.J.
        • et al.
        The most frequent traumatic orthopaedic injuries from a national pediatric inpatient population.
        J Pediatr Orthop. 2005; 25: 39-44
        • Kruppa C.G.
        • et al.
        Pediatric pelvic ring injuries: how benign are they?.
        Injury. 2016; 47: 2228-2234
        • Gansslen A.
        • Heidari N.
        • Weinberg A.M.
        Fractures of the pelvis in children: a review of the literature.
        Eur J Orthop Surg Traumatol. 2013; 23: 847-861
        • Amorosa L.F.
        • Kloen P.
        • Helfet D.L.
        High-energy pediatric pelvic and acetabular fractures.
        Orthop Clin North Am. 2014; 45: 483-500
        • Guillaume J.M.
        • et al.
        Pelvic fractures in children (pelvic ring and acetabulum).
        Orthop Traumatol Surg Res. 2020; 106: S125-S133
        • Schwarz N.
        • et al.
        Long-term results of unstable pelvic ring fractures in children.
        Injury. 1998; 29: 431-433
        • Gansslen A.
        • et al.
        Pelvic ring injuries in children. part I: epidemiology and primary evaluation. a review of the literature.
        Acta Chir Orthop Traumatol Cech. 2012; 79: 493-498
        • Zwingmann J.
        • et al.
        Pelvic fractures in children results from the German pelvic trauma registry: a cohort study.
        Medicine. 2015; 94: e2325
        • Moher D.
        • et al.
        Preferred reporting items for systematic reviews and meta-analyses: the prisma statement.
        Int J Surg. 2010; 8: 336-341
        • Sterne J.A.
        • et al.
        ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions.
        BMJ. 2016; 355: i4919
        • Smith W.
        • et al.
        Clinical outcomes of unstable pelvic fractures in skeletally immature patients.
        J Bone Joint Surg Am. 2005; 87: 2423-2431
        • Blasier R.D.
        • et al.
        Disruption of the pelvic ring in pediatric patients.
        Clin Orthop Relat Res. 2000; : 87-95
        • Rieger H.
        • Brug E.
        Fractures of the pelvis in children.
        Clin Orthop Relat Res. 1997; : 226-239
        • Garvin K.L.
        • et al.
        Pediatric pelvic ring fractures.
        J Pediatr Orthop. 1990; 10: 577-582
        • Upperman J.S.
        • et al.
        Early functional outcome in children with pelvic fractures.
        J Pediatr Surg. 2000; 35: 1002-1005
        • Guimaraes J.A.
        • et al.
        Surgical treatment for unstable pelvic fractures in skeletally immature patients.
        Injury. 2014; 45 (Suppl): S40-S45
        • Scolaro J.A.
        • Firoozabadi R.
        • Routt M.L.C.
        Treatment of pediatric and adolescent pelvic ring injuries with percutaneous screw placement.
        J Pediatr Orthop. 2018; 38: 133-137
        • Stuhler T.
        • et al.
        [Pelvic fractures in children: clinic, late results, biomechanic (author's transl)].
        Arch Orthop Unfallchir. 1977; 90: 187-198
        • Ouyang J.
        • et al.
        Experimental cadaveric study of lateral impact of the pelvis in children.
        Di Yi Jun Yi Da Xue Xue Bao. 2003; 23 (408): 397-401
        • Keshishyan R.A.
        • et al.
        Pelvic polyfractures in children. radiographic diagnosis and treatment.
        Clin Orthop Relat Res. 1995; : 28-33
        • Chia J.P.
        • et al.
        Pelvic fractures and associated injuries in children.
        J Trauma. 2004; 56: 83-88
        • Bryan W.J.
        • Tullos H.S.
        Pediatric pelvic fractures: review of 52 patients.
        J Trauma. 1979; 19: 799-805
        • Musemeche C.A.
        • et al.
        Selective management of pediatric pelvic fractures: a conservative approach.
        J Pediatr Surg. 1987; 22: 538-540
        • Nierenberg G.V.G.
        • Bialik V.
        • Stein H
        Pelvic fractures in children: a follow up in 20 children treated conservatively.
        J Pediatr Orthop. 1993; 1: 140-142
        • Quinby Jr, W.C.
        Fractures of the pelvis and associated injuries in children.
        J Pediatr Surg. 1966; 1: 353-364
        • Hargitai E.
        • et al.
        Unstable pelvic fractures in children.
        Acta Chir Hung. 1998; 37: 77-83
        • Sagi H.C.
        • Coniglione F.M.
        • Stanford J.H.
        Examination under anesthetic for occult pelvic ring instability.
        J Orthop Trauma. 2011; 25: 529-536