Authors: Zhi‐Yuan Yao, Shu‐Yao Fan, Wei‐Qiang Zhao, Jie‐Feng Huang
Categories: Review Article, Classification system, Hip arthroplasty, Periprosthetic femoral fracture, The Unified classification system, The Vancouver classification system
Source: Orthopaedic Surgery
Doi: 10.1111/os.14149
Periprosthetic femoral fractures (PPFFs) following total hip arthroplasty (THA) present a significant clinical challenge due to their increasing incidence with an aging population and evolving surgical practices. Historically, classifications were primarily based on anatomical fracture location, the stability of the implant, and bone quality surrounding the implant. We critically analyzed 25 classification systems, highlighting the emergence and adaptations of key systems such as the Vancouver classification system (VCS) and the Unified classification system (UCS), which are lauded for their simplicity and effectiveness yet require further refinement. VCS, developed in 1995, categorizes fractures based on the site, implant stability, and bone quality, and remains widely used due to its robust applicability across different clinical settings. Introduced in 2014, UCS expands the VCS to encompass all periprosthetic fractures with additional fracture types, aiming for a universal application. Despite their widespread adoption, these systems exhibit shortcomings, including the incomplete inclusion of all PPFF types and the imprecise assessment of implant stability and surrounding bone loss. These gaps can result in misclassification and suboptimal treatment outcomes. This paper suggests the necessity for ongoing improvements in classification systems to include emerging fracture types and refined diagnostic criteria, ensuring that they remain relevant to contemporary orthopedic practices and continue to facilitate the precise tailoring of treatment to patient‐specific circumstances. This comprehensive historical review serves as a foundation for future innovations in classification systems, ultimately aiming to standardize PPFF treatment and improve patient prognosis.
Keywords: Classification system, Hip arthroplasty, Periprosthetic femoral fracture, The Unified classification system, The Vancouver classification system
Periprosthetic fracture has become a frequent complication of total hip and knee arthroplasties with rates up to 15% in major registries. ^1^ The most common fracture localization of the periprosthetic fracture is the femur, and periprosthetic femoral fracture (PPFF) after total hip arthroplasty (THA) has become a severe clinical disease which has been found to be much more common after uncemented hip replacement. ^2^ , ^3^ PPFF is the second major cause for revision THA, ^4^ and the third most common complication after THA following aseptic loosening and infection. ^5^ , ^6^ , ^7^
With the aging of the population, the incidence of PPFFs is increasing year by year. ^8^ , ^9^ , ^10^ , ^11^ , ^12^ , ^13^ , ^14^ With extending life expectancy increasing, the constantly expanding indications for THA and the decreasing age of patients undergoing THA, ^15^ , ^16^ the number of patients with PPFFs increases. ^17^ , ^18^ , ^19^ , ^20^ , ^21^ , ^22^ , ^23^ , ^24^ More young patients with end‐stage hip disease are willing to undergo THA, and tend to have a higher range of motion after surgery, which also increases the risk of PPFF. ^25^ , ^26^ The Mayo Clinic Joint Registry identified the overall incidence of the PPFF following THA was 4.1% from 1969 to 1999. ^27^ In 2016, the data further updated by the Mayo Clinic Joint Registry from 1969 to 2011 showed that the incidence of intra‐operative PPFFs was 1.7% and 20‐year probability of post‐operative PPFFs was 3.5% after primary THA, while the incidence of intra‐operative PPFFs was 12% and 20‐year probability of post‐operative PPFFs was 11% after revision THA. ^19^ , ^28^ According to the cases analysis reported from the Swedish National Hip Arthroplasty Register of periprosthetic fracture after THA, the accumulated incidence of PPFFs was 0.4% for the primary THA and 2.1% for the revision of THA. ^29^ According to the Australian Prosthesis Registry, the incidence of periprosthetic fractures associated with THA was 10%. ^30^ A multicenter retrospective cohort study in the UK revealed that patients with PPFFs were predominantly elderly, frail, female, and more likely to experience the unified classification system B type fractures (the fracture around the stem or just below it), with females being a significant predictor for PPFF type. ^31^
Previous studies have shown the incidence of postoperative complications of PPFF is high, and the long‐term function is not satisfactory. ^32^ , ^33^ , ^34^ Since PPFFs frequently occur in the elderly with severe bone loss, surgical complications are common, making treatment challenging and prognosis poor. ^35^ For instance, the rate of PPFF‐related infection, a typical surgical complication, stands at 7.3%. This commonly results in multiple re‐operations, possible non‐union, a decreased clinical function, and long‐term antibiotic treatment. ^36^ Such severe complications are costly to our society, ^37^ , ^38^ and result in considerable mortality. ^5^ , ^6^ , ^29^ , ^39^ , ^40^ , ^41^ , ^42^ , ^43^ , ^44^ , ^45^ , ^46^ , ^47^ , ^48^ , ^49^ , ^50^ , ^51^ The mortality rate within 1 year after surgery for PPFF is as high as 11%. ^52^ , ^53^ , ^54^ , ^55^ The peri‐implant periprosthetic survival analysis, a multicenter study from Spain, reported a 6.2% in‐hospital mortality rate for PPFFs. ^56^ Therefore, PPFF has become a hot issue in surgical treatment area with high rising burdens associated to the treatment in the near future. ^9^ , ^57^ , ^58^ , ^59^
In comparison with primary THA, ^60^ it is a significant clinical challenge for orthopedic surgeons to take aseptic loosening, bone loss and osteoporosis as well as general state of patients into consideration when treating PPFF. ^11^ , ^29^ , ^40^ , ^41^ , ^42^ , ^61^ , ^62^ , ^63^ In addition, each periprosthetic fracture poses a unique challenge to the orthopedic surgeons because of the variables including the relationship of the fracture to the implant, the specifics of the implant including wear, and the functional demands of the patient also should be taken into account. ^64^ To obtain an ideal prognosis, it is very important to establish reasonable treatment principles and providing patients with the individualized treatment. ^5^ Take history as a mirror, we can know what is going on. Understanding the history of classification systems for PPFF will allow us to better understand and grasp the treatment of PPFF. This study provides a literature review of reported classification systems for PPFF to provide reference for the classification and standardized treatment of PPFF in our clinical work and provide evidence for further improvement of PPFF classification system.
A comprehensive literature search was conducted using PubMed and the Cochrane Library, employing the keywords “periprosthetic femoral fracture,” “periprosthetic fracture,” and “peri‐implant fracture.” The search encompassed literature available until March 10, 2024. Additionally, the review included a search of published reports from the American Association of Orthopedic Surgeons (AAOS) annual meeting proceedings spanning the years 1990 to 2023.
The inclusion criteria specified studies discussing the classification systems or new types of PPFF. The exclusion criteria (i) duplicate articles; (ii) articles without full‐text access; (iii) irrelevant titles, abstracts or full texts; and (iv) editorials, letters, and commentaries.
During the literature selection process, we conducted a quality assessment to ensure the reliability and relevance of the review findings, adhering to established inclusion and exclusion criteria. Evaluation metrics included study design, methodological rigor, the appropriateness and accuracy of outcome measures, and reproducibility. Furthermore, references of all relevant articles were manually cross‐referenced to identify potential additional studies. Ultimately, 25 articles were included (Figure 1).
FIGURE 1 A brief flow chart of selection process.
Based on the literature search and quality assessment, we identified 25 classification systems or new types for PPFF (Table 1) (Figures 2 and 3). These systems were developed by various orthopedic researchers to guide treatment and improve prognosis. ^5^ , ^65^ , ^66^ , ^67^ , ^68^ , ^69^ , ^70^ , ^71^ , ^72^ , ^73^ , ^74^ , ^75^ , ^76^ , ^77^ , ^78^ The classification systems vary, with some focusing on the anatomical location of the fracture, while others emphasize the stability of the implant and the surrounding bone stock. We describe the history and evolution of classification systems for PPFF in the following (i) the classification systems for PPFF at the early stage; (ii) the milestone of classification systems for PPFF—the Vancouver classification system (VCS); (iii) the modification for VCS—the unified classification system (UCS); (iv) disadvantages of the VCS and UCS; (v) other classification systems supplementary to VCS and UCS; and (vi) the modified version of the UCS for PPFF.
FIGURE 2 Timeline of the overall development of classification systems for periprosthetic femoral fracture.
FIGURE 3 Current periprosthetic femoral fracture classification systems. (A) Parrish and Jones. ^65^ (B) Whittaker et al. ^66^ (C) Johansson et al. ^67^ (D) Bethea et al. ^68^ ; (E) Cooke and Newman. ^69^ (F) Jensen et al. ^70^ (G) Roffman and Mendes ^71^ ; (H) Schwartz et al. ^79^ (I) AAOS. ^72^ (J) Mont and Maar. ^73^ (K) Duncan and Masri (VCS). ^75^ (L) Beals and Tower. ^74^ (M) Ninan et al. ^111^ (N) Van Houwelingen and Duncan. ^77^ (O) Grammatopoulos et al. ^110^ (P) Phillips et al. ^115^ (Q) Capello et al. ^76^ (R) Duncan and Haddad (UCS). ^5^ (S) Baba et al. ^43^ (T) Huang et al. ^78^ (U) Huang et al. ^119^ (V) Karam et al. ^116^ (W) Fan et al. ^107^ (X) Malhotra et al. ^112^ (Y) González‐Martín et al. ^118^
Based on the anatomical location of the fracture, Parrish and Jones classified PPFF into four fractures in the trochanteric area, fractures in the proximal portion of the shaft, fractures in the mid‐shaft, and fractures in the distal part of the shaft of the femur. ^65^ Whittaker et al. classified PPFF into three types according to the relationship of the stem of the prosthesis with the fracture line and the degree of Type I, at the intertrochanteric level about the middle portion of the stem; Type II is an oblique or spiral fracture about the stem tip; and Type III, fracture is severely displaced with the distal fragment dislodged from the stem of the prosthesis. ^66^ Based on the location relative to the femoral component of the total hip prosthesis, Johansson et al. classified PPFF into three Type I, fractures were those in which the fracture occurred proximal to the tip of the prosthesis; Type II, fractures were those in which the fracture line extended from the proximal portion of the femoral shaft to beyond the distal tip of the prosthesis; and Type III, fractures were those in which the fracture was entirely distal to the tip of the prosthesis. ^67^ According to the anatomical location of the fracture, Bethea et al. extended the classification to three Type A, those that fractured at the tip of the femoral component, either transversely or with a distal spiral, were designated; Type B, spiral fractures around the femoral component were designated; and Type C, fractures proximal to the tip of the stem with comminution around the stem were designated. ^68^ Based on the anatomical location of the fracture, Cooke and Newman classified PPFF into four types and proposed treatment ^69^ Type 1, comminuted fractures around the stem of the prosthesis. Treatment early revisions; Type 2, oblique or spiral fractures around the shaft of the prosthesis. Treatment conservative treatment or revision operation; Type 3, fractures are transverse at the level of the tip of the prosthesis and are unstable. Treatment primary internal fixation; and Type 4, fractures are entirely distal to the prosthesis. Treatment internal fixation. Jensen et al. emphasized the stability of the post‐injury prosthesis as a predictor of prosthetic revision and classified PPFF into three Type 1, located around the proximal two thirds of the prosthetic femoral stem; Type 2, extending proximally and distally from the area of the femoral stem tip; and Type 3, extending distally from the area of the femoral stem tip. ^70^ Roffman and Mendes classified PPFF according to the stability of the prosthesis after THA. ^71^ Based on the fracture site and morphology, PPFFs were classified into five types by Schwartz et al.: Type A, involving the greater trochanter; Type B, longitudinal fracture of proximal femur; Type C, longitudinal fracture of distal femur; Type D, complete metaphyseal fracture; and Type E, complete fracture in distal femur prosthesis. ^79^ AAOS put forward a classification system according to the anatomical location of the Type I, proximal to the intertrochanteric line; Type II, vertical splitting fracture, not exceeding the inferior margin of the lesser trochanter; Type III, fracture extends below the inferior margin of the lesser trochanter, but not exceeding the lower 1/3 junction of the prosthesis; Type IV, fracture spanning the prosthesis tip; Type IVa, spiral fracture; Type IVb, transverse fracture; Type V, severely comminuted Type‐III and Type‐IV fracture, extends below the stem tip; and Type VI, fracture distal to the stem tip. ^72^ According to the anatomical location of the fracture, Mont and Maar divided PPFF into six Type 1, intertrochanteric; Type 2, proximal femur; Type 3, spanning the prosthesis tip; Type 4, distal to the prosthesis tip; Type 5, comminuted, blowout; and Type 6, supracondylar. ^73^ According to the anatomical location of the fracture, Beals and Tower classified PPFF into four Type I, fracture in the trochanteric region; Type II, proximal metaphyseal/disphyseal fractures that do not involve the stem tip; Type IIIA, proximal diaphyseal fractures at the stem tip with less than 25% disruption of the prosthetic interface; Type IIIB, proximal diaphyseal fractures at the stem tip with greater than 25% disruption of the prosthetic interface; Type IIIC, supracondylar fracture at the tip of a long femoral stem; and Type IV, supracondylar fracture distant to the stem tip. ^74^
These classifications did not take into account all three factors of the site of fracture, the stability of implant and surrounding bone stock at the same time, so their guiding significance for the treatment of PPFF was limited.
Duncan and Masri proposed the Vancouver classification system (VCS) in 1995, which was established according to the site of fractures, the stability of the femoral implant and the surrounding bone quality of the proximal femur.
^75^
VCS is the most widely accepted classification system nowadays.
^69^
,
^80^
It classified PPFF into three A, B and C, of which Type A includes two subtypes and Type B includes three subtypes (Table 2, Figure 4). Type A is the fracture in trochanteric region. Type A is further subdivided into Type AG involving the greater trochanter and Type AL involving the lesser trochanter. Type B is the fracture around the stem or just below it. Type B is further subdivided into B1 (Well‐fixed stem), B2 (loose stem with good proximal bone stock), and B3 (loose stem with poor‐quality bone stock). Type C is the fracture occurring well below the tip of the stem.
FIGURE 4 The Vancouver classification system for periprosthetic femoral fracture. For definitions of the fracture types, see Table 2.
According to VCS, the treatment regimens for each type as follows (Table 2): ^8^ , ^26^ , ^46^ , ^59^ , ^81^ , ^82^ , ^83^ , ^84^ , ^85^ , ^86^ , ^87^ Type A, conservative treatment is recommended if the fracture displacement is small, and internal fixation is recommended if the fracture displacement is large and associated with severe pain and dysfunction; Type B1, open reduction and internal fixation (ORIF) or THA revision with a long stem; Type B2, THA revision with ORIF; Type B3, THA revision with ORIF and management of bone defects such as bone graft; Type C, ORIF.
The clinical effectiveness of the VCS for PPFF has been extensively validated through several studies. Studies have consistently demonstrated the VCS's robust interobserver and intraobserver reliability and its validity across different clinical settings and populations. Naqvi et al. reported substantial agreement among consultants (κ = 0.69) and trainees (κ = 0.61) in using the VCS, with even higher intraobserver reliability ranging from 0.74 to 0.90. ^40^ The system also showed strong validity, aligning well with intraoperative findings with an 81% agreement rate and a κ value of 0.68. The study noted a potential issue with the underdiagnosis of implant instability when relying solely on preoperative radiographs, suggesting the necessity for intraoperative assessments to ensure accurate treatment planning. Brady et al. further corroborated these findings by assessing the VCS's reliability and validity, showing that the VCS helps in accurately reflecting the surgical and clinical realities and can be applied consistently across different levels of clinical expertise. ^42^ Moreover, Rayan et al. extended the validation of the VCS into a European context, confirming its applicability and reliability across different geographic and clinical environments. ^45^ This cross‐continental validation supports the universal applicability of the VCS, ensuring that it holds under various healthcare systems with different levels of resources and expertise. However, Lee et al. identified limitations within the context of cementless femoral stems, especially in differentiating stable (B1) from unstable (B2 and B3) fractures. ^88^ The interobserver reliability showed moderate agreement (κ = 0.45), indicating a potential need for refinement of the VCS when used with newer implant technologies.
The VCS provides a structured framework for classifying PPFF, guiding clinical decisions based on the site of fractures, implant stability and bone quality. This system not only facilitates clear communication among clinicians but also helps in planning surgical interventions and predicting outcomes. It promotes standardized care across different clinical settings and by various medical professionals. while the VCS is invaluable for managing PPFF, its moderate κ values in certain contexts highlight the necessity for ongoing evaluation and adjustment to ensure its continued relevance in all surgical scenarios.
In 2014, Duncan and Haddad proposed UCS for all‐site periprosthetic fractures. ^5^ The UCS aims to reach the following ^89^ (i) by adding Type D, Type E and Type F, the VCS was expanded and updated; (ii) so regardless of the which bone is broken and the joint involved, the groupings and treatment principles can cover all periprosthetic fractures; and (iii) after the popularization, the classification can eliminate language barriers and we may communicate with simplicity and clarity. For the PPFF, the UCS classification added relatively rare and new types of the fracture (Type D, Type E, Type F) and expanded the VCS (Table 3, Figure 5). Type D is the fracture of the femoral shaft between the hip and knee replacements. Type E is the fracture involving both the acetabulum and femur after hip replacement. Type F is the fracture involving only the acetabulum after hip replacement.
FIGURE 5 The unified classification system for periprosthetic femoral fracture. For definitions of the fracture types, see Table 3.
The treatment principles of Type A, B and C in UCS were the same as those in VCS. Most recently, several studies revealed that revision surgery was the most common approach for Type B2/B3 fractures, while simple fixation prevailed as the predominant strategy for Type B1/C fractures. ^90^ , ^91^ , ^92^ Furthermore, Thomas et al. reported that monoblock tapered stems can be an acceptable modality in the treatment of Type B2 fractures, while Capone et al. indicated that for Type B2 and B3 fractures, revision with modular stems is preferable to monoblock stems. ^91^ , ^93^ A multicenter study has indicated that for Type B fractures around cemented polished taper‐slip femoral components, ORIF is associated with lower reoperation rates, shorter waiting times for surgery, reduced transfusion needs, and lesser demands for critical care compared to revision surgery. ^94^ Conversely, research by Scalici et al. found that patients with Type B and C fractures undergoing revision surgery had better functional outcomes than those treated with ORIF. ^92^ Therefore, there is no definitive operative technique for all UCS B fractures at present. Meanwhile, several studies also applied the principles of treatments for new types as follows (Table 3): ^5^ , ^89^ , ^95^ , ^96^ , ^97^ , ^98^ , ^99^ Type D, logical individual treatments (revision of one, both or neither joint replacement and ORIF); Type E, logical individual treatments according to the “block‐out” analysis in relation to each component of the joint replacement; Type F, logical individual treatments (non‐operative approach with protected weight‐bearing, delayed and relatively straightforward conversion to THA, initial displacement).
The clinical effectiveness of the UCS for PPFF has been scrutinized across various studies. The UCS, aimed at providing a comprehensive framework for all periprosthetic fractures, demonstrates high reliability and validity across different clinical settings and fracture types. Vioreanu et al. validated its high interobserver and intraobserver reliability, suggesting that it could be consistently applied by different clinicians across international settings. ^100^ Huang et al. confirmed the reliability of the UCS, observing substantial agreement in clinical assessments. ^101^ Specifically, for 23 Type B cases, there was 79.71% agreement with intraoperative findings and a mean κ value of 0.694 for validity, indicating substantial agreement. De Meo et al. reported the UCS showed a slightly higher validity (κ = 0.64) compared to the VCS (κ = 0.56), suggesting its enhanced ability to match intraoperative findings. ^102^ Nevertheless, Jain et al. pointed out the UCS has moderate reliability but only fair validity when applied to PPFF around cemented polished taper‐slip (PTS) stems. ^103^ The findings indicate that UCS may not adequately define PTS stem loosening, indicating the need for potential refinements in its use. Moreover, the lack of perfect agreement in the κ values across these studies further confirms the necessity for ongoing improvements to the UCS. Furthermore, Schopper et al. concluded that despite the UCS's advancements, the VCS remains more frequently used in the literature and clinical practice, likely due to its established efficacy and simpler application. ^104^ In summary, the UCS possesses the capability to improve surgical planning and outcome prediction, while also necessitating ongoing adjustments and validations to ensure its clinical effectiveness.
Previous studies have shown that there were two following limitations during the clinical practice of the VCS and UCS: (i) not all types of PPFF are included; and (ii) there are individual variations in the evaluation of the stability of femoral prosthesis and the judgment of bone loss in proximal femur.
First, some fracture types cannot be grouped into either of the above two classifications, so doctors and researchers often encounter confusion in the choice of treatment options, or even choose the wrong treatment. For example, Mallory et al.,
^105^
Van Houwelingen and Duncan,
^77^
and Capello et al.
^76^
reported that the pseudo ALT periprosthetic fractures including a segment of the proximal medial femoral cortex, which is actually the new B2 fracture. Egrise et al. reported the clinical and radiological results and identified risk factors of this type of fractures.
^106^
Fan et al. reported the variant AGT periprosthetic fracture (a fracture of the greater trochanter with lateral cortical extension) and femoral stem destabilization, which is also actually the new B2 fracture.
^107^
These fracture patterns have not been previously described in the VCS and the UCS. Without a correct type differentiation, the treatment will surely become a failure. In addition, PPFF accompanied by stem fracture after hip arthroplasty were not classifiable under the original VCS or the UCS.
Second, Orthopedic surgeons have different opinions about the stability of femoral prosthesis in Type B fracture and the presence of bone loss in the proximal femur, which even may bring about inaccurate treatments. ^108^ , ^109^ It has been pointed out definitely in the verification articles of the VCS and UCS. ^40^ , ^42^ , ^45^ , ^100^ , ^101^ Grammatopoulos et al. retrospective reviewed the radiographs of some patients who were considered to have a Vancouver B1 fracture, found some subtle features, such as subsidence of the stem into the centralizer, that were characteristic of a B2 fracture pattern. ^110^ If we cannot distinguish this fracture pattern from Vancouver B1, it may lead to treatment failure.
Ninan et al. presented the Coventry classification system, which grouped the periprosthetic fractures into “happy hips” and “unhappy hips” based on stability of the stem. According to the therapeutic principle of VCS, fracture fixation alone is required in the “happy hips,” and revision of the prosthesis is needed in the “unhappy hips.” ^111^
Capello et al. presented their modified version of VCS.
^76^
Type TG: same as type AG of VCS. Type TL: same as type AL of VCS. Type A1: fracture of medial cortex that includes the residual neck, calcar, and lesser trochanter and is displaced medially, with a well‐fixed stem. Type A2: fracture of medial cortex that includes the residual neck, calcar, and lesser trochanter and is displaced medially, with a loose stem. Type B1: same as type B1 of VCS; Type B2: same as type B2 of VCS. Type B3: same as type B3 of VCS. Type C: same as type C of VCS. This study reported the new A category which represented the pseudo ALT periprosthetic fractures supplementing the VCS. The new A category was described to encompass the “clamshell” pattern supplementing the B type of VCS with A1 nominating this fracture with a stable stem and A2 nominating this fracture with an unstable stem. The “clamshell” fracture originates at the medial base of the greater trochanter and ends at the distal medial cortex of the lesser trochanter, and the lateral cortex is intact. Similarly, Mallory et al.,
^105^
Van Houwelingen and Duncan
^77^
also reported the periprosthetic fracture of the lesser trochanter including a segment of the proximal medial femoral cortex which is not the traditional Type AL but really the Type B2 fracture.
Huang et al. proposed PPFF accompanied by stem fracture supplementing VCS and UCS, and provided corresponding treatments. ^78^ They were divided into two types (Type A and Type B), and Type B has two subtypes. Type A: PPFF accompanied by stem fracture after hip arthroplasty in which the proximal portion of the fractured femoral prosthesis is stable. Treatment remove the distal portion of the fractured femoral prosthesis, followed by open reduction and internal fixation (ORIF). If the patient is in poor physical condition and a future revision is not possible, ORIF alone is performed without removal of the distal portion. Type B: PPFF accompanied by stem fracture after hip arthroplasty in which the proximal portion of the fractured femoral prosthesis is unstable. Type B is divided into two subtypes. Type B1: the proximal portion of the fractured femoral prosthesis is loose and the surrounding bone quality is good. Treatment revision of THA is performed with a longer stem. Type B2: the proximal portion of the fractured femoral prosthesis is loose and the bone bed is of poor quality. Treatment revision of THA with a particular stem. The study also identified three cases of periprosthetic femoral fractures accompanied by stem fracture after hip arthroplasty through retrospective analysis, which were not classifiable under the VCS and UCS. This validated the clinical relevance and effectiveness of the new fracture pattern classification. Similarly, Malhotra et al. also presented the unusual fractures through the femoral shaft and the femoral prosthesis and discussed the treatment of removal of the broken implants and subsequent revision replacement surgery. ^112^
For remedying the defects of the VCS and UCS in evaluating the stability of femoral prosthesis and overcome PPFF failures by objective evaluation, Baba et al. presented new classification for PPFF. ^43^ Type cementless stem. Type 1A: the main fractured region involves the porous‐coated area which is the region firmly bonding the stem to bone and is necessary to stabilize the stem, so the stem is likely to be unstable when this region is fractured. Type 1B: the main fractured region is outside the porous‐coated area. Since the stem is firmly bonded to bone in the porous‐coated region, the stem is stable when the fractured region is distal to it. Type cemented stem. Type 2A: The main fractured region involves the stem. Fixation of bone and cement may be broken, and the stem is likely to be unstable. Type 2B: the main fractured region is distal to the stem. The stem is stable because fixation of bone and cement are not broken, and fracture occurs distal to it. The reliability and validity of the Baba classification was improved when plain radiograms, CT imaging and implant information were given. ^113^ When supplemented with additional diagnostic information, this classification system exhibited superior interobserver reliability (κ = 0.94) compared to the VCS, and demonstrated substantial intraobserver reliability among experts (κ = 0.81). Additionally, the Baba classification achieved a 95% concordance with intraoperative findings, underscoring its clinical validity and utility in accurately guiding surgical interventions. They also presented treatment regimens for each type as Types 1A and 2A, intra‐operative stem stability test is necessary, the combination of stem revision and ORIF when the stem is unstable and ORIF alone when the stem loosening is not confirmed; Types 1B and 2B, revision with a long stem or ORIF if it is the implant‐tip fracture of a cement stem and ORIF alone if not. Stoffel et al. introduced an algorithmic approach to identify loose stems around proximal femoral periprosthetic fractures using patient history, stem design, and plain radiographs, which can help to detect stem loosening and improve curative effect. ^114^
Grammatopoulos et al. described the ‘spiral’ fracture pattern ^110^ and Phillips et al. described the comminuted ‘burst’ pattern ^115^ with cemented stems in B2 type of VCS. The “spiral” fracture is often in association with a separate wedge fragment and significant comminution. The comminuted “burst” pattern often splits along the cement mantle, similar to an “ax.” Based on the spiral, comminuted “burst,” and the previously described clamshell pattern ^76^ in B2 type of VCS, Karam et al. reported the newly observed “reverse clamshell” in B2 type. ^116^ This kind of fractures has an intact medial cortex, and originates from the medial calcar and passes through the lateral cortical outlet. All these fracture patterns belong to B2 fractures which indicate the implant instability, and the treatment regimen still refers to the treatment principle of B2 in VCS.
González‐Martín et al. divided the proximal femur into three zones (medial, lateral and distal) according to the Gruen system ^117^ and propose a sub‐classification of B2 type in VCS to B2.1 (1 fractured zone) and B2.2 (≥2 fractured zones). ^118^ They suggested B2.1 type treated via ORIF had a lower risk of complication than B2.2 type.
These classification systems supplementary to VCS and UCS aim to address specific scenarios and shortcomings in the VCS and UCS. Each system introduces unique categories and treatment strategies based on fracture characteristics, implant stability, bone quality or implant design, enhancing the diagnostic and therapeutic precision for PPFF. These supplementary systems collectively enrich the original frameworks, offering a more nuanced approach to managing the diverse and complex scenarios encountered in PPFF treatment.
In 2018, based on previous study in 2015, ^78^ Huang et al. further modified the application of UCS in the femur side through literature search and case collection and provided corresponding treatment methods ^119^ (Table 4, Figure 6). The modifications of UCS are as (1) add two new B2 B2PALT/B2PAGT (the pseudo ALT/AGT: fracture in trochanter region including a segment of the proximal medial/lateral femoral cortex); (2) add a new FS category to encompass stem fracture alone or accompanied by PPFF, with FSO designating stem fracture alone, FS1 designating this fracture with the proximal portion of the fractured femoral prosthesis being stable, FS2 designating this fracture with the proximal portion of the fractured femoral prosthesis being loose and the surrounding bone quality being good, and FS3 designating this fracture with the proximal portion of the fractured femoral prosthesis being loose and the bone bed being of poor quality; and (3) delete Type F which does not apply to the femur.
FIGURE 6 The modified version of the unified classification system for periprosthetic femoral fracture. For definitions of the fracture types, see Table 4.
The modified version of the UCS also applied the treatment regimens for the new types as follows (Table 4): Type B2PALT and B2PAGT, revision THA with a longer stem, along with ORIF of the fracture; Type FS (FSO, FS1, FS2, FS3), removal of the stem, and revision THA with logical individual treatment. In addition, it mentioned that if fracture is at the tip of the stem in B1 type, the cortical allograft struts were recommended for use. And a new study further confirmed that the longer stem revision and internal fixation (LSRIF) which obtained the excellent clinical outcome is the main treatment for Type B2PALT fractures, while simple ORIF is an option for elderly patients in poor condition. ^120^
This new modification compensated for the deficiency of some fracture types in classic UCS and VCS and refined the application on the femur side. Fan et al. evaluated the clinical effectiveness of this modified UCS for PPFF. ^121^ The study revealed almost perfect agreement among consultants (κ = 0.882) and substantial agreement among trainees (κ = 0.776) for interobserver reliability. Intraobserver reliability showed substantial to almost perfect agreement, with κ values ranging from 0.701 to 0.972. Validity assessments involving 299 type B cases demonstrated 89.854% agreement with intraoperative findings (κ = 0.849), indicating almost perfect agreement. These findings affirm the modified UCS's high reliability and validity in classifying PPFF, enhancing its utility in precisely evaluating implant stability and informing treatment decisions for PPFF. They also suggested that the trainee surgeons mastered the original VCS and subsequent UCS, while consultants mastered the modified UCS. ^102^
As mentioned above, Huang et al. suggested the cortical allograft struts for fracture at the tip of the stem in B1 type on the basis of the stable implant. ^119^ Fractures at the tip of the stem have also been reported in other studies. Yasen et al. recommended to use the cortical allograft struts with ORIF when the implant is stable, and a long stem revision with ORIF when the implant is loose for the treatment of fractures at the tip of the stem. ^59^ Baba et al. indicated that the therapeutic policy for implant‐tip fracture was controversial, revision with a long stem or ORIF may be chosen by different surgeons. ^43^ Woo et al. adopted a long‐stem femoral prosthesis augmented with autologous bone grafts and a new cable‐plate construct for the fracture at the tip of the stem, and fracture healing was achieved. ^97^ Hence, the fracture at the tip of the stem is a special pattern of PPFF which is difficult to treat because of the lack of rotational stability and may not be the common B1 type. Current classification systems do not mention the requirement for special classification of the fracture at the tip of the stem. Whether it requires special classification perhaps remains to be further analyzed in future studies.
The ultimate goal of the classification systems for PPFF is to guide treatment and improve prognosis. In order to achieve this goal, the classification system must reflect the conditions of injury comprehensively, be reliable and effective, and be easy to master. The VCS and UCS for PPFF has been shown to be reliable, valid, and easy to master, but there is still room for further improvement.
The classification systems at the early stage have significantly enhanced our understanding and management of PPFF. ^65^ , ^66^ , ^67^ , ^68^ , ^69^ , ^70^ , ^71^ , ^72^ , ^73^ , ^74^ , ^79^ Developed over decades, these systems provide a structured approach to categorizing fractures based on their anatomical location or implant stability, aiding in clinical decision‐making. They have several advantages. First, these systems innovatively categorized PPFF according to specific anatomical details, improving diagnostic precision by clearly identifying fracture locations—critical for planning targeted interventions. Additionally, by delineating fractures into distinct categories based on location or implant stability, these early systems enabled the development of standardized treatment protocols—from conservative management to revision surgery—tailored to each fracture type. This standardization has streamlined treatment processes and established care benchmarks applicable across various healthcare settings, thus enhancing consistency and optimizing outcomes. Furthermore, the structured categorization fostered by these systems has bolstered clinical research and communication within the orthopedic community by standardizing study inclusion criteria, facilitating outcome comparisons across different patient groups, and improving the efficiency of findings dissemination. These systems also paved the way for future advances in classification systems by highlighting the early models' limitations, allowing new systems to incorporate additional crucial factors like bone quality.
The classification systems at the early stage, which focused solely on anatomical location or implant stability, provided foundational frameworks for understanding and managing PPFF but had notable disadvantages. The classification systems based only on the anatomical location of the fracture fail to consider the implant stability and the surrounding bone quality. ^65^ , ^66^ , ^67^ , ^68^ , ^69^ , ^70^ , ^72^ , ^73^ , ^74^ , ^79^ First, inadequate evaluation of implant stability can lead to unsuitable treatment choices, as the system fails to effectively navigate between conservative management, fracture fixation, or revision surgery needs. Inadequate initial surgical repairs often fail to provide long‐term stability and functionality, significantly increasing the likelihood of requiring secondary revision surgery. Furthermore, implant instability can lead to continuous micro‐movements at the fracture site, which not only delays healing but may also cause further damage to the bone‐prosthesis interface. This can reduce the functional lifespan of the implant and potentially lead to a cascade of mechanical failures. In addition, Bone defects add complexity to the surgical repair of PPFF, often rendering standard fixation techniques like plates, screws, or intramedullary rods insufficient. As a result, revision surgeries and bone grafting frequently become necessary components of the treatment plan. To manage the instability caused by bone defects, enhanced fixation methods using longer or more robust prosthetic stems may be required to provide stability across the fracture site. Ignoring the surrounding bone quality can result in poor outcomes such as loosening of the implant, nonunion or malunion of the fracture, increased pain, and limited mobility. Alternatively, classification systems that are solely focused on implant stability often fall short by overlooking both the precise fracture morphology and location, which are crucial for selecting the appropriate surgical strategy and ensuring biomechanical integrity, as well as ignoring the surrounding bone quality, leading to potentially detrimental effects. ^71^
The VCS and UCS are pivotal in the classification of PPFF. The advantages of these two classification systems lie in the simultaneous consideration of the site of fractures, the stability of the femoral implant and the surrounding bone quality, thus making up for the shortcomings of the classification systems at the early stage. ^69^ , ^75^ , ^80^ The VCS facilitates precise intervention strategies by distinguishing between fracture types A, B, and C, each subdivided further to reflect different clinical scenarios. The UCS expands on the VCS by including additional fracture types (D, E, F), thereby covering all periprosthetic fractures and aiming to simplify global communication within the orthopedic community. ^5^ , ^89^ Their robustness is confirmed by studies showing strong interobserver and intraobserver reliability, as well as alignment with intraoperative findings, thereby validating its effectiveness across varied clinical environments. ^40^ , ^42^ , ^45^ , ^100^ , ^101^ , ^102^ Despite their extensive applications, both VCS and UCS have limitations that may affect their clinical utility. The VCS and UCS are sometimes criticized for not encompassing all PPFF types, which can lead to potential misclassifications and inappropriate treatment choices. For instance, certain fracture patterns reported in the literature are not adequately categorized by these systems, leading to clinical ambiguities and possible treatment failures. ^73^ , ^76^ , ^105^ , ^106^ , ^107^ Additionally, the clinical effectiveness of these systems in differentiating between stable and unstable fractures, particularly in the context of the application of cementless femoral stems or cemented PTS stems, has shown only moderate agreement, suggesting a need for refinement. ^88^ , ^103^ Moreover, discrepancies in the assessment of implant stability and bone quality highlight the systems' limitations in uniformly guiding surgical interventions across all cases. ^108^ , ^109^ , ^110^ Continuous validation and adjustment, informed by clinical outcomes and emerging research, will enhance their reliability, validity and applicability in all surgical scenarios.
The introduction and adaptation of various classification systems supplementary to the VCS and UCS have significantly enhanced the management of PPFF. These systems provide tailored approaches to diverse fracture scenarios, showcasing their advantages. The Coventry classification system introduced by Ninan et al. categorizes fractures based on stem stability, simplifying treatment decisions between “happy hips” that require fracture fixation alone and “unhappy hips” that need prosthesis revision. ^111^ Capello et al. introduced a modified version of the VCS to address pseudo ALT fractures, offering a nuanced view that enhances diagnostic accuracy and treatment precision. This inclusion of specific fracture types such as the “clamshell” pattern further refines surgical planning. ^76^ The classifications proposed by Huang et al. provide a framework for dealing with stem fractures, a scenario inadequately covered by previous systems, thus enhancing treatment specificity and potentially improving outcomes. ^78^ Moreover, the inclusion of unusual fractures through the femoral shaft and the femoral prosthesis by Malhotra et al. highlighted the need for removal of broken implants and revision surgery, emphasizing the classification's adaptability to diverse clinical scenarios. ^112^ The Baba classification system addresses the limitations of previous systems by focusing on implant design and the physical integrity of stems, which is critical for setting appropriate therapeutic strategies. ^43^ This system's ability to differentiate between the stability of cementless and cemented stems based on the location of the fracture relative to the porous‐coated area enhances treatment specificity. The classification by González‐Martín et al. introduces a sub‐classification for B2 fractures, facilitating more targeted treatment options based on fracture zone risks, which could reduce complications associated with over or under‐treatment. ^118^ While these adaptations expand the applicability of the VCS and UCS, they have several disadvantages, such as complicating the clinical decision‐making process. Each new sub‐classification introduces additional criteria that clinicians must consider, which can vary significantly in different clinical settings and require specific diagnostic tools not universally available. For instance, the system by Baba et al. enhances the detail by differentiating fractures based on the location of the break relative to the implant's porous‐coated area, which is crucial for determining the stability of cementless stems. ^43^ However, this requires precise imaging and can lead to variability in treatment approaches. ^113^ Huang et al. proposed classifications for PPFF with stem fractures, which were not previously categorized, highlighting the need for specific treatments. ^78^ While these additions aim to cover the broad spectrum of PPFF scenarios, they also require precise diagnostic capabilities and may increase the cognitive load on clinicians, potentially complicating decision‐making processes. Additionally, these classification systems may rely on advanced imaging for precise application, which may not be applicable to all clinical settings, especially those with limited resources. These systems collectively expand the granularity with which PPFF can be assessed and treated, providing frameworks that potentially improve patient outcomes by allowing for more customized treatment plans. However, their effectiveness depends on the clinical setting and available resources, necessitating ongoing evaluation and adaptation to ensure they remain relevant and beneficial across all orthopedic practices.
The modified UCS introduced by Huang et al. ^119^ presents several distinct advantages. By integrating new subtypes, such as B2PALT/B2PAGT for specific trochanteric fractures, and the FS category for stem fractures, this system offers a more granular and precise classification. This level of specificity aids in accurately assessing the stability of the implant and the quality of the surrounding bone, which is crucial for determining appropriate treatment strategies. Such detail‐oriented categorization helps to standardize treatment approaches, potentially reducing variability in clinical outcomes and improving prognostic accuracy. Additionally, the expansion of fracture types beyond the scope of the original VCS and UCS addresses previous limitations, allowing for a more comprehensive evaluation of complex fracture scenarios. However, the modified UCS also presents certain disadvantages. The complexity introduced by the additional fracture subtypes and categories may pose a steep learning curve, particularly for less experienced clinicians, potentially leading to inconsistencies in fracture classification and treatment. This complexity could impede the widespread adoption of the system, as thorough training and familiarity with the classification are required to utilize it effectively. Moreover, while the system enhances the granularity of classification, it may rely on detailed clinical data and imaging not always available in all healthcare settings, limiting its practical applicability in less resourced environments. These factors suggest that while the modified UCS offers substantial improvements in the classification and management of PPFF, its application may be constrained by the specificities of clinical practice settings and the expertise of the medical personnel.
The history of classification systems for PPFF following THA reveals a nuanced evolution, shaped by the growing complexities of patient demographics, surgical advancements and a deepening understanding of patient outcomes over several decades. Most classification systems at the early stage primarily focused on the anatomical location of fractures. These foundational frameworks established the basis for further innovations by identifying crucial areas requiring enhancement, notably the integration of evaluations concerning the surrounding bone quality and the implant stability.
The introduction of the VCS marked a significant milestone by offering a more comprehensive framework that considered not just the location of the fracture but also the implant stability and the bone quality. Its widespread adoption underscored the system's utility in facilitating communication among clinicians and standardizing treatment approaches. Furthermore, the UCS was developed to create a more inclusive framework by expanding the categories of the VCS, aiming to standardize the classification of all periprosthetic fractures regardless of their location. This was a step towards universal applicability, intending to simplify global communications. However, the VCS and UCS occasionally fail to encompass all fracture types, and there are individual variations in the evaluation of the stability of femoral prosthesis and the judgment of bone loss, which may lead to misclassification and suboptimal treatment strategies.
The evolution of classification systems for PPFF has paralleled advances in orthopedic surgery and a deeper understanding of biomechanics and patient‐specific factors. The integration of emerging technologies and interdisciplinary approaches promises to refine these systems further, enhancing both diagnostic precision and treatment efficacy. Looking forward, several improvements are expected to further improve these (i) the integration of advanced imaging techniques, such as high‐resolution MRI and 3D computed tomography, can provide more detailed insights into fracture morphology and the biomechanical environment around the implant. These technologies could help refine fracture classification by revealing subtle features that current systems may overlook, such as micro‐cracks in the bone or early signs of implant loosening; (ii) incorporating biomechanical modeling and simulation into classification systems could significantly improve the predictive accuracy of treatment outcomes. By simulating different scenarios of implant and bone interaction, doctors could predict potential complications and tailor interventions more precisely to individual patient anatomy; (iii) AI and machine learning offer robust tools for analyzing large datasets to predict outcomes, personalize treatment plans and identify previously unrecognized patterns in fracture presentation. By applying these technologies, future classification systems could dynamically incorporate real‐time data from clinical outcomes to update and refine prognostic models continuously.; (iv) enhancing collaboration between engineers, biologists, computer scientists, and clinicians could foster the development of integrative classification systems that reflect both clinical realities and theoretical advancements. This would not only enhance the functionality of these systems but also ensure they are adaptable to the rapid pace of technological change; and (v) there is a pressing need to ensure that new classification systems are validated across diverse populations and healthcare settings. This would involve multicenter studies to test the applicability and reliability of revised classifications, ensuring they are robust and universally applicable.
In summary, while the VCS and UCS have set a strong foundation, the dynamic nature of medical science necessitates their continuous evolution. The integration of advanced technologies and multidisciplinary approaches will likely play a critical role in the next generation of classification systems, ensuring they are capable of addressing the complex and varied needs of the global patient population. These advancements will further enhance our ability to deliver personalized, precise, and effective treatments to patients with PPFF.
Z.Y.Y: designed the study, wrote the text, prepared the tables. S.Y.F: searched the literature, organized the data. W.Q.Z: searched the literature, analyzed and confirmed the data. J.F.H: designed the study, wrote the text, prepared the figures, approved the final version for submission.
The authors declare no conflict of interest.
This study was financially supported by Zhejiang Traditional Chinese Medical Science and Technology Planning Project (No. 2023ZL400) and Scientific Research Project of Zhejiang Education Department (No. Y201942372).
The study was approved by the ethics committee of the First Affiliated Hospital of Zhejiang Chinese Medical University (2016‐K‐143‐01).
All authors listed meet the authorship criteria according to the latest guidelines of the International Committee of Medical Journal Editors, and all authors are in agreement with the manuscript.