Fractures of the Femur After Hip Joint Replacement: The Vancouver Classification After 30 Years

The Original Vancouver Classification In 1995, at the invitation of the American Academy of Orthopaedic Surgeons (AAOS), we proposed a system for the analysis of periprosthetic fractures of the femur following the insertion of a stem, as well as an algorithm for the translation of important outcome determinants into a treatment plan (see Appendix)1. It was recognized that these fractures, first described in 1954 by Horwitz and Lenobel2, were on the rise and that the classification systems up to then were still evolving and were incomplete with respect to guiding a rational approach to management3–5. Our proposal was well-received and widely adopted by the orthopaedic community. It was generously coined the Vancouver Classification System (VCS) in recognition of the academic center from which it originated, and it subsequently performed well during the field testing of its observer reliability and validity6,7. For ease of access, the original instructional course lecture1 is reprinted in the Appendix of this update, compliments of the AAOS. In essence, the original 1995 version of the VCS identified 3 factors as paramount to the analysis of periprosthetic fractures: the location of the fracture in relation to the implant, the stability of the implant fixation, and the quality of the supporting bone. The identification of these factors prompted 3 fundamental questions, as follows: Question 1. What zone of the implant-bone segment is involved? The zone is classified using a simple “ABC” mnemonic: A if Apophyseal (greater trochanter = AG; lesser trochanter = AL). B if involving the Bed of the stem. C if Clear of the stem; well below the implant. Question 2. Is the implant still well-fixed or is it loose? (This question almost exclusively refers to Type-B fractures but can apply to the A and C categories as well, although uncommonly encountered.) B1 if yes. B2 if no. Question 3. Does the bone supporting the loose stem have sufficient integrity to support a new stem? Remains B2 if yes. Advances to B3 if no (i.e., because of considerable comminution or osteolysis). The basic types that were most commonly encountered in 1995 (AG, AL, B1, B2, B3, and C) are illustrated in Figures 1 through 6. More subtypes were recognized in time and added to the VCS through its incorporation within the more comprehensive Unified Classification System (UCS). The UCS was proposed in 20148, primarily to apply the VCS to other joints, such as the knee and shoulder, and to either side of the joint, such as the acetabulum and proximal femur in a hip replacement and the distal femur and proximal tibia in a knee replacement. The value of the UCS is that it brought the VCS to other joints.Fig. 1: Type AG: fracture of the greater trochanter. The stem is stable.Fig. 2: Type AL: avulsion fracture of the lesser trochanter. The stem is stable.Fig. 3: Type B1S: the fracture is below the apophysis, with an oblique configuration, and the fixation surface proximally is unaffected. The stem is stable.Fig. 4: Type B2: the fracture is below the apophysis and disrupts the fixation of the stem. The stem is loose.Fig. 5: Type B3: the fracture is below the apophysis. The stem is loose and there is severe bone loss.Fig. 6: In 1995, this fracture was referred to as Type C. This classification is now retired, and the fracture should be referred to as Type B1S.Within the context of 1995, when the majority of these injuries were Type B, the attraction of the initial VCS was its simplicity and easy transformation into a rational treatment plan, as follows: Type B1: Fix the fracture and keep the implant, unless there is another reason to exchange the stem. Type B2: Revise the stem to a longer stem and use it to fix the fracture. Type B3: Perform complex reconstruction to simultaneously address the fracture, the loose stem, and the deficient bone. Historically, in the 1990s and for some time after, this involved a variety of strategies: long stems, segmental replacement prostheses, and/or allograft techniques. A number of studies have reported the reliability of the VCS and the predictably satisfactory outcomes when the principles that flow from its application are applied9,10. What Has Changed in 30 Years Now, 30 years later, by way of an update, it is time to reflect on the VCS (and its offspring, the UCS) and to comment on its ongoing utility in the increasingly important field of periprosthetic fractures. We have chosen 6 subjects for commentary: the Type-AL fracture, the Type-D fracture, the cementless stem, the polished tapered cemented stem, the Type-C fracture, and the Type-B3 fracture. Type-AL Fracture and Its Pretender Type AL is an uncommon fracture, usually a traction injury or an avulsion through osteolysis-weakened bone, which may occur spontaneously or during revision hip replacement. It can be safely observed (Fig. 2). However, it is important to distinguish the Type-AL fracture from its pretender, the pseudo AL fracture, which involves a substantial portion of the underlying cortex11,12. The latter is, by definition, a Type-B fracture. Most commonly, it occurs during or soon after the insertion of a cementless stem, especially with so-called single-taper stems, as can be seen in Figure 7. The pseudo AL fracture represents an unrecognized intraoperative or early postoperative fracture. The stem fixation is usually compromised, and reoperation is required. It is, in fact, a Type-B2 fracture, and the prevalence of this fracture has led to the reduced popularity of single-taper stems in favor of collared triple-taper stems.Fig. 7: An example of a fracture that might be misinterpreted as Type AL, but, in reality, is Type B2 because the stem is loose and has subsided. As demonstrated by the staples, this is an early postoperative fracture where the stem has not had a chance to stabilize.Type-D Fracture, the Interprosthetic Fracture The prevalence of fractures affecting the femur following hip and ipsilateral knee replacement is rising (Fig. 8). When these fractures are added to the original “ABC” mnemonic as Type D, they represent a fracture that is Dividing the femur between 2 implant systems. The original mnemonic is therefore extended to “ABCD.”Fig. 8: Type D, the so-called interprosthetic fracture. It divides the femur supporting 2 joint replacements (hip and knee). Block-out analysis reveals it to be Type B1S for both the hip and knee. Neither arthroplasty prosthesis needs to be disturbed.Type-D fractures present their own set of challenges to the surgeon and warrant separate consideration. They are an extension of the VCS as a “new” fracture that is now included in the UCS8. For these fractures, the exercise of “block-out analysis” is advised. That is, block out the knee and ask, “What type of fracture is this for the hip?” Usually it is B1, B2, B3, or C. Next, block out the hip and ask the same question for the knee. A rational, principle-based treatment plan will flow from this exercise, which will integrate the analysis of both replaced joints. The closer the fracture is located to one joint or the other, the more its management will be influenced by the block-out analysis of that joint. Ultimately, the type of treatment required depends on whether the stem is loose or not (B1 or C versus B2), thus emphasizing the usefulness of the VCS, with the added nuance that a potential stress riser effect must be considered if both implants in the hip and knee have stems (Fig. 9).Fig. 9: Another example of Type D, the so-called interprosthetic fracture. Due to the presence of 2 stems within a centimeter or so of each other, as well as the fracture around both stems, treatment is more nuanced and attention needs to be paid to both stems, rendering internal fixation more difficult.The Cementless Stem, Our Modern Workhorse In 1995, the VCS was proposed in an orthopaedic world dominated by previous stem fixation with cement. It was generally an easy task to identify the loose stem (Type B2). There were exceptions to this, of course, as emphasized by an important study of Swedish National Hip Arthroplasty Register fractures by Lindahl et al.13. In another study, the advice of Stoffel et al.14 regarding careful analysis of radiographs and preoperative symptoms (e.g., pre-fracture pain, comminution, cement fracture, stem subsidence, extension of the fracture into the stem fixation surface) is worth noting. In that paper, Stoffel et al. provided a useful algorithm to allow the surgeon to more precisely determine the fixation of a cementless stem and, hence, whether the fracture is a B1 or B2 fracture. Similarly, Ninan et al.15 introduced the concept of the happy versus unhappy hip in the context of a periprosthetic fracture, thus stressing the importance of identifying the presence of pain before a fracture in order to help guide the surgeon to look for features of loosening, as was very eloquently articulated by Stoffel et al.14. Today, however, with the popularity of proximally coated cementless stems, it is easy to “overcall” the B2 fracture in a case in which the stem is still adequately fixed to the major proximal fracture fragment and reduction with fixation—using the distal part of the stem partly as an intramedullary rod—is a preferred option (Figs. 3 and 10). If the cementless stem has not subsided (an important observation) and the surgeon is in doubt about the robustness of its fixation, a computed tomography (CT) scan (Figs. 11-A and 11-B), coupled with a possible intraoperative evaluation, is valuable for assessing the stem-bone ingrowth interface.Fig. 10: The B1S fracture shown in Figure 3 has been reduced and fixed with a locking periprosthetic fracture plate. Figs. 11-A and 11-B: CT scans showing excellent osteointegration at the fixation interface of a cementless stem with a B1S fracture.Fig. 11-A: Coronal cut.Fig. 11-B: Axial cut. The Polished Tapered Cemented Stem It is important to separate out fractures of the femur around polished tapered cemented stems as a unique entity and to understand that subsidence of the stem does not imply that the stem-cement construct has been destabilized and requires revision. As the stem is not bonded to the cement, and the fixation of the composite is dependent on the integrity of the cement-bone interface, an analysis of that interface is critical in guiding treatment, even if the stem itself has subsided (Figs. 12-A and 12-B). In select circumstances (adequate cement mantle, an intact bone-cement interface, and minimal or no comminution), other treatment strategies can be considered: namely, nonoperative management (rarely, in highly selected cases) or reduction with cerclage fixation and bridge plating (Figs. 13-A and 13-B) with or without cement-in-cement revision16,17 (Fig. 14). Figs. 12-A and 12-B: Anteroposterior (Fig. 12-A) and lateral (Fig. 12-B) views of a periprosthetic fracture with a polished tapered cemented stem. This is a Type-B1S fracture because the cement mantle is still fixed to the bone and the fracture can be anatomically reduced, thus restoring the integrity of the construct. This fracture was managed with open reduction and internal fixation (ORIF), as shown in Figure 13.Fig. 12-AFig. 12-B Figs. 13-A and 13-B: Anteroposterior views of the proximal femur (Fig. 13-A) and distal femur (Fig. 13-B) after ORIF and healing of the fracture that was shown in Figure 12.Fig. 13-AFig. 13-BFig. 14: A Type-B1S fracture with a polished tapered cemented stem treated with cement-in-cement revision and cerclage fixation. This could have also been treated with ORIF, as in Figs. 13.Is Type C Redundant? The techniques and technology of fracture fixation have advanced enormously in the past 30 years, such that there now is little, if any, difference in the surgical management of Type-B1 and C fractures. This has prompted the question of whether there is a need to retain Type C as a distinct entity. We would support the retirement of Type C and the incorporation of this fracture type into Type B. The majority will be Type B1. Transverse B1 Fractures? While much attention has been paid to the location of a B1 fracture, we propose that greater attention be paid to the orientation of the fracture, especially where internal fixation is contemplated. The 2 fracture patterns are those that are transverse and those that are spiral or oblique. Greater awareness of fracture orientation is important because of the notorious reputation of transverse fractures to proceed to nonunion and fixation failure unless initially treated with biplanar fixation or, in select cases, intramedullary fixation with a long-stem revision. Whether or not this proposal emerges as 2 subtypes that are based on fracture orientation remains for the orthopaedic community to decide. If it does, we would propose the use of “B1T” (T for transverse) (Fig. 15) and “B1S” (S for spiral or oblique) (Fig. 3).Fig. 15: An example of a Type-B1T fracture, with a transverse fracture at the tip of a polished tapered cemented stem.It follows that, while the fracture shown in Figure 6 would have been referred to as a Type-C fracture in 1995, we propose in 2025 that it be referred to as a Type-B1S fracture. The same applies to the interprosthetic fracture shown in Figure 9 (UCS Type D), which, after block-out analysis, would be Type B1S for the hip and the knee. These fractures would no longer be referred to as Type-C fractures, as that fracture type is now officially retired. We are not the first to try to add some granularity to the VCS. Huang et al.18 suggested adding another fracture pattern to describe a hip where the metallic stem is fractured. Uzoigwe et al.19 proposed a significant expansion of the VCS through the addition of 8 subclasses of the B2 fracture. However, such a change nullifies the obvious advantage of the VCS, which is its simplicity and clinical applicability. Type-B1T fractures pose unique difficulties in their management, primarily due to the risks of nonunion and fixation failure (Figs. 16-A and 16-B)20. In such fractures, a very large varus moment is applied onto the hip stem, and these forces need to be neutralized to allow the fracture to heal before the construct fails21,22. To allow healing, some authors have advocated revision total hip arthroplasty for these fractures, so that the fracture is bypassed with a longer revision stem. This is certainly an effective treatment strategy and would be the exception to the recommendation that Type-B1 fractures be treated with internal fixation. This unique fracture pattern warrants a discussion between the arthroplasty surgeon and trauma surgeon in order to individualize treatment on the basis of patient characteristics, unique needs, and difficulties in revision or internal fixation. The ultimate treatment strategy will depend on what is the least morbid operation for the patient, considering their functional needs and medical comorbidities. Figs. 16-A and 16-B: A Type-B1T fracture that was treated with a cortical onlay allograft, which failed (Fig. 16-A). It was subsequently treated with ORIF with 2 orthogonal plates to prevent hardware failure and nonunion (Fig. 16-B).Fig. 16-AFig. 16-B Is Type B3 Redundant? When the VCS was developed in 1995, revision techniques were rudimentary compared with the state of the art today. This is why a distinction was made between a fracture with a loose stem that required a revision with then-existing revision implants (Type B2) and a fracture needing a more complex reconstruction with subspecialized techniques (Type B3). These techniques, to mention a few, included distally locked stems, segmental proximal femoral replacements, and allograft-prosthesis composites. In 2025, there is increasing evidence that tapered fluted titanium stems (TFTSs) are so versatile that they may be used for the majority of Type-B2 and B3 fractures, with predictably good outcomes23,24. For this reason, it has been asked whether the B3 fracture subtype should be retired from the VCS. The authors are of 2 minds on this question. One author (C.P.D.) acknowledges that Type-B3 fractures have decreased in prevalence within the Type-B spectrum because of the success of TFTS technology and techniques but believes that the subtype should be preserved for use in cases of severe proximal bone loss (Fig. 5), even more so if the remaining femur is vacuous and the fracture is distal to the isthmus. Cemented proximal femoral replacement, allowing early full weight-bearing, is a valuable treatment option in this group of patients, who are commonly elderly and fragile, with low demand and limited lifespan remaining (Fig. 17). In addition, complex reconstruction with mesh containment and impaction grafting is still favored in some centers for severe bone loss, and it would seem prudent to retain the B3 designation for these cases as a group in order to facilitate outcome analysis. In contrast, the other author (B.A.M.) has not had to resort to any of these techniques in over 10 years, regardless of the complexity of the fracture, and thus votes to retire the B3 fracture.Fig. 17: An example of a proximal femoral replacement that was performed to treat a Type-B3 fracture. The osseous remnants, including the greater trochanter, were wrapped around the implant to increase stability.Because the term “Vancouver Classification” was developed organically over the years by the orthopaedic community, who found the VCS to be very useful for the management of periprosthetic fractures, the authors look forward to how our colleagues respond to our thoughts regarding the B3 fracture and whether it should be retired or not. For now, we will keep it in the updated VCS, with the caveats The VCS had the of time for 30 years and has found a favored in the and of orthopaedic It is to be that this will In addition, the VCS the UCS by way of the of its principles and algorithm to other joint replacements, regardless of the joint or bone using a very The observer reliability and utility of both systems have been and the 2 systems have demonstrated some features when namely, the simplicity of the VCS, and the comprehensive of the In essence, the UCS can be seen as an extension of the VCS, which was its and is to the VCS. The question that remains is whether of these classification systems is and, if whether such should be by the addition of more fracture the or retirement of previous or the of the principles of treatment that flow from the classification systems. more fracture subtypes will be suggested in such as those proposed by Huang et al.18 and Uzoigwe et We will it to the to if such are useful or not. Similarly, as it has been proposed that 2 of the Type C and Type B3, be retired, although the authors are in on the of on the of treatment, from its has that has not been to the of the transverse B1 fracture and its to proceed to It is that this be with a designation (T for transverse) to the need for fixation as if or long-stem intramedullary fixation for this fracture or oblique Type-B1 fractures are now as B1S fractures. The Vancouver Classification System in 2025 In of our and the made in treatment over the past 30 years, we would to propose the of the 1995 VCS, as follows: Type A of the into and Type B the into B1 if the construct is and B2 if it is not. Type C is now retired. Type the femur between a hip replacement and a knee as B1 or B2 after block-out analysis. The proposed B subtypes and principles of treatment would B1: Cementless stem with remaining ingrowth requires fracture fixation. B1: cemented implant with a fixed stem and intact cement requires fracture fixation. B1: Polished cemented stem with cement mantle and bone-cement fixation. of of the cement, fracture fixation, and possible cement-in-cement revision. Type-B1 fractures with a transverse at the tip of the stem. B1S: Type-B1 fractures with a spiral or oblique B2: of the an construct stem exchange and fracture fixation. B3: A loose stem with such severe bone loss that complex bone is chosen or segmental femoral is Appendix provided by the authors is with the version of this as a at with from of the femur after hip replacement.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.