Varus-valgus constrained insert with posterior-stabilized femoral components in complex primary total knee arthroplasties.

AIMS: Varus-valgus constrained (VVC) devices are typically used in revision settings, often with stems to mitigate the risk of aseptic loosening. However, in at least one system, the VVC insert is compatible with the primary posterior-stabilized (PS) femoral component, which may be an option in complex primary situations. We sought to determine the implant survivorship, radiological and clinical outcomes, and complications when this VVC insert was coupled with a PS femur without stems in complex primary total knee arthroplasties (TKAs).
METHODS: Through our institution's total joint registry, we identified 113 primary TKAs (103 patients) performed between 2007 and 2017 in which a VVC insert was coupled with a standard cemented PS femur without stems. Mean age was 68 years (SD 10), mean BMI was 32 kg/m2 (SD 7), and 59 patients (50%) were male. Mean follow-up was four years (2 to 10).
RESULTS: The five-year survivorship free from aseptic loosening was 100%. The five-year survivorship free from any revision was 99%, with the only revision performed for infection. The five-year survivorship free from reoperation was 93%. The most common reoperation was treatment for infection (n = 4; 4%), followed by manipulation under anaesthesia (MUA; n = 2; 2%). Survivorship free from any complication at five years was 90%, with superficial wound infection as the most frequent (n = 4; 4%). At most recent follow-up, two TKAs had non-progressive radiolucent lines about both the tibial and femoral components. Knee Society Scores improved from 53 preoperatively to 88 at latest follow-up (p < 0.001).
CONCLUSION: For complex primary TKA in occasional situations, coupling a VVC insert with a standard PS femur without stems proved reliable and durable at five years. Longer-term follow-up is required before recommending this technique more broadly. Cite this article: Bone Jt Open 2021;2(11):921-925.

Introduction

Varus-valgus constrained (VVC) devices were initially designed for use with dedicated revision femoral components and stems to limit the risk of aseptic loosening during revision total knee arthroplasty (TKA). However, VVC devices have become more commonly used in complex primary TKAs, often without the use of adjunctive stem fixation.

In at least one system, the VVC insert is compatible with the primary posterior-stabilized (PS) femoral component. Surgeons may be tempted to occasionally pair that VVC insert with a standard, non-stemmed PS femur in unique, complex primary situations (e.g. femur already cemented, abnormal anatomy). With such unconventional use, loosening of components is a concern.

There is a paucity of literature on the use of a PS femoral component without stems with a VVC insert in the setting of primary TKA. The aims of this study were to determine the implant survivorship, radiological and clinical outcomes, and complications when this VVC insert was coupled with a PS femur without stems in complex primary TKAs.

Methods

Through our institutional total joint registry, we identified 103 patients who underwent 113 primary TKAs performed between 2007 to 2017 in which a VVC insert was coupled with a standard cemented PS femur without stems or other adjuvant fixation (total stabilizing (TS) insert, Triathlon PS femur (Stryker, USA)). Of note, this is an on-label usage of this implant. During the study period, a total of 1,961 Stryker PS femora were used in primary TKAs. Thus, our cohort represents 5.8% of all Stryker PS femora used during the study period. Mean patient age at the time of TKA was 68 years (standard deviation (SD) 10). Mean BMI was 32 kg/m2 (SD 7), and 59 of the patients (50%) were male. Prior to study initiation, institutional review board approval was obtained.

All patients had a minimum potential follow-up of two years. Two patients (2%) died prior to reaching two years of follow-up, and 12 patients (12%) had less than two years of follow-up. Mean follow-up was four years (2 to 10) for the remaining patients. In this cohort, 53 patients (47%) had minimum potential follow-up of five years.

Use of a VVC device was used in the setting of a preoperative varus deformity on long leg radiographs in 67 cases (59%) with a mean of 11.6⁰ of varus (standard deviation (SD) 6.0⁰), a valgus deformity in 43 cases (38%) with a mean of 10.2⁰ of valgus (SD 5.5⁰; Figure 1), and little or no deformity in three cases (3%).

Fig. 1

a) Preoperative anteroposterior; and b) lateral radiographs of a patient with a valgus deformity. c) Posteroperative anteroposterior; and d) lateral radiographs depicting correction of the valgus deformity with the posterior-stabilized femoral component and varus-valgus insert.

Statistical analysis

Survivorship was assessed using the Kaplan-Meier method, and clinical outcomes were assessed using Knee Society Scores (KSSs).

Data are reported as a mean accompanied by a range for continuous variables and as a count with a percentage for categorical variables. Continuous variables were analyzed using an unpaired t-test. Survivorship was assessed using the Kaplan-Meier method; survivorship data are presented with 95% confidence intervals (CIs). Patients who underwent a revision, reoperation, or died were censored at the time of the event. A p-value of 0.05 was considered significant, and all statical tests were two-sided.

Results

Survivorship

The five-year survivorship free from aseptic loosening was 100%. The five-year survivorship free from any revision was 99% (95% CI 93% to 100%; Figure 2), and the one revision was performed for a chronic periprosthetic joint infection (PJI). This patient was treated with a two-stage exchange protocol.

Fig. 2

Kaplan-Meier survivorship curve depicting survivorship free from any revision.

The five-year survivorship free from any reoperation was 93% (95% CI 82% to 97%; Figure 3). The most common reoperation was for PJI (n = 4: two irrigation and debridements with polyethylene exchange, one two-stage exchange described above, and one irrigation and debridement without polyethylene exchange), followed by manipulation under anaesthesia (MUA; n = 2), and removal of heterotopic ossification (n = 1).

Fig. 3

Kaplan-Meier survivorship curve depicting survivorship free from any reoperation.

Complications

Survivorship free from any nonoperative complication at five years was 90% (95% CI 76% to 96%). A total of ten complications were identified, including superficial wound infection with no surgical intervention (n = 4), patellar clunk (n = 2), deep vein thrombosis (n = 2), haematoma (n = 1), and stress fracture of the distal femur (n = 1).

Radiological outcomes

At most recent follow-up, two unrevised TKAs (2%) had non-progressive radiolucent lines about both the tibial and femoral components with all others showing no radiological evidence of loosening.

Clinical outcomes

The mean KSS improved from 53 preoperatively (24 to 83) to 88 at latest follow-up (43 to 100; p < 0.001, unpaired t-test). Range of motion (ROM) improved from a mean of 105° (60° to 135°) preoperatively to a mean of 113° postoperatively (60° to 140°; p < 0.001, unpaired t-test).

Discussion

This study assessed the mid-term survivorship, clinical and radiological outcomes, and complications of a specific VVC insert paired with a standard, non-stemmed PS femoral component in complex primary TKAs. In this series, we observed no cases of aseptic loosening, with excellent survivorship free from any revision, reoperation, and nonoperative complications at five years.

VVC devices are commonly used in the revision TKA setting, often in combination with metaphyseal cones or sleeves and/or diaphyseal stems. The absence of adjuvant fixation and the use of a VVC insert raises concerns for increased rates of aseptic loosening. However, the present study found a survivorship free from aseptic loosening of 100% at five years. Similar results are reported in various other studies that used constrained implants in complex primary TKAs. One comparable study examined the outcomes of PS femoral implants with two different inserts, a PS insert (n = 244) and a VVC insert (n = 242), all without stemmed components. The authors reported one revision for aseptic tibial component loosening in the VVC insert group (0.4%) and no cases of aseptic loosening in the PS group at a mean follow-up of 3.25 years. Similarly, Crawford et al, in a series of 96 patients (103 TKAs), reported no cases of aseptic loosening at a mean follow-up of five years with mid-level constraint without stems in primary TKA. Anderson et al, in a series of 248 non-modular constrained condylar primary TKAs without stem extensions, reported one case of aseptic loosening, which required a femoral component revision and the addition of a diaphyseal stem.

Our study also demonstrates excellent survivorship free from any revision at five years at 99%, with the lone revision being performed for PJI. Other studies of VVC implants for primary TKAs report similar findings with Anderson et al. A systematic review by Avino et al on the use of VVCs in primary TKA estimated the need for implant revision to be 9% at 12 years postoperatively. Their findings were that VVC devices provided good clinical improvement, but long-term implantation may lead to more failures. It is hypothesized that these late failures were more likely due to the increased constraint causing mechanical stresses resulting in aseptic looseing. More long-term studies are required to assess survivorship and failures, and to delineate the potential protective role of adjuvant fixation.

Survivorship free from any reoperation at five years was 93% for this cohort, consistent with current literature on the use of constrained implants in primary TKA that have demonstrated a range of 2.5% to 10% of patients requiring a reoperation at early follow-up. The most common reoperation in our study was treatment of infection followed by MUA for treatment of arthrofibrosis. Several other studies also observed arthrofibrosis as a common complication in this setting. It is possible that patients with greater intraoperative laxity who receive VVC devices may be predisposed to arthrofibrosis secondary to subtle laxity not fully addressed by the VVC implant.

Radiological review demonstrated two knees with non-progressive radiolucent lines at most recent follow-up in our study. Anderson et al reported no radiological evidence of loosening or wear in their cohort of primary constrained arthroplasties. One study comparing the outcomes of primary VVC arthroplasties matched to PS arthroplasties noted no signs of implant loosening in either group at mid-term follow-up. Additionally, clinical outcomes (KSS and ROM) improved significantly in this cohort. This is comparable to the current literature on this topic. However, in studies that compared PS to VVC implants, no difference in clinical outcomes were detected. Thus, our study adds to the building body of literature suggesting that higher constraint inserts are proving clinically and radiologically durable at mid-term follow-up.

Our study is not without limitations. First, this is a retrospective review using data from a single institution. Second, this is a relatively small cohort of patients with intermediate follow-up; longitudinal surveillance studies will be necessary moving forward, especially for addressing the incidence of aseptic loosening. This is particularly important as the primary concern is eventual aseptic loosening from increased constraint with mismatched implants, combined with lack of adjuvant fixation. Third, this study examines a single implant design, which may limit the transferability of these results to other designs.

In conclusion, for complex primary TKAs, coupling a VVC insert with a standard PS femur without stems, proved reliable and durable at five years. It is particularly notable that aseptic loosening has not presented in any patients clinically or radiologically at mid-term follow-up. Longer-term follow-up is required before recommending this technique more broadly.

Take home message

- For complex primary total knee arthroplasty, in occasional situations, coupling a varus-valgus constrained insert with a standard posterior-stabilized femur without stems proved reliable and durable at five years.