1 Introduction

Total knee arthroplasty (TKA) is a frequently performed procedure that reduces pain and increases mobility in patients with late-stage osteoarthritis. Despite the success of TKA in treating severe osteoarthritis, approximately 20–30% of patients with TKA are dissatisfied with their results.^1,2 TKA often results in the loss of the anterior and posterior cruciate ligaments. Posterior cruciate substituting, posterior stabilized, and bicruciate stabilized TKA result in loss of the anterior and posterior cruciate ligaments; posterior cruciate retaining TKA maintains the posterior cruciate ligament but sacrifices the anterior cruciate ligament. Sacrifice of the anterior cruciate ligament, which is critical to preventing anterior tibial displacement, disrupts normal knee kinematics and causes a loss of knee stability.^3–6 As a result, patients who frequently climb stairs, squat, and kneel tend to be particularly disappointed with TKA outcomes.^7,8 Patients’ dissatisfaction with TKA outcomes may partially relate to the loss of native knee kinematics.

The bicruciate stabilized (BCS) total knee system is a modern knee implant that replaces both the ACL and the PCL, but improves on several aspects of earlier TKA systems.⁹ In vivo kinematic analyses found that second generation BCS systems reduce anterior-posterior motion and reduce tibio-femoral axial rotation during deep-knee bends. The second generation BCS system also has reduced risk of reoperation and revision compared to the first generation device. However, a randomized controlled trial found that BCS TKA was not superior to posterior stabilized TKA.¹⁰ At two-year follow-up, knee flexion and Oxford Knee Scores (OKS) in the two groups were similar.

This cohort study was designed to assess patient outcomes following BCS TKA in different patient subgroups. The effect of body-mass index (BMI) on patient outcomes was of particular interest because the relationship between BMI and patient reported outcomes is unknown for the BCS implant.

2 Methods

We reviewed the medical records of patients who underwent a primary total knee arthroplasty (TKA) Journey 2 BCS knee implant from 2016 to 2017 at a single institution. Patients were included in the study if they had, at minimum, two-year follow-up. Patient reported outcome measures (PROMs) included the Knee Society Score (KSS), Knee Injury and Osteoarthritis Outcome Score for Joint Replacement (KOOS JR), Lower Extremity Activity Score (LEAS), and Visual Analogue Scale (VAS). Data on LEAS were excluded from analyses because there were 203 (69.5%) missing values.

The KSS was developed in 1989, and was updated in 2012.¹¹ The score primarily measures pain, alignment, and stability of the knee joint. The KSS was validated by both confirming internal reliability and by assessing for evidence of differential item functioning.¹² KSS can be applied across age, sex, implant types, and activity levels.

KOOS JR is a seven-item instrument that was established using Rasch analysis in 2016.¹³ The short-form version of KOOS was specifically designed to analyze outcomes directly relevant to TKA. KOOS JR assesses “knee health” and reflects aspects of patient pain, symptom severity, and activities of daily living.

The VAS is considered a sensitive, robust, and reproducible scale for assessing pain severity.¹⁴ The scale was developed as a continuous score because pain is thought to vary over a continuum. Severity of pain, as measured by VAS, ranges from 0 (no pain) to 10 (extreme pain).

In order to better understand the relationship between patient outcomes and body-mass index (BMI), BMI was divided into pre-specified subgroups for parts of the analysis, using the established BMI grading scale provided by the Centers for Disease Control and Prevention.¹⁵ The pre-specified subgroups were as follows: underweight, normal, or overweight (<30 kg/m²), obese (≥30 to <35 kg/m²), and severely obese (≥35 kg/m²). Paired t-tests were used to evaluate the pre-operative to post-operative changes in KSS and KOOS JR. For the pre-specified BMI subgroups, PROMs at baseline, post-operatively, and change in pre-operative to post-operative values were assessed by one-way analysis of variance (ANOVA). The relationship between BMI and PROMs was also assessed by linear regression. Spearman's rho was used in order to determine if BMI and PROMs were monotonically associated, even without a linear association. Finally, post hoc analyses were used to determine if older (>75 years) severely obese patients reported inferior outcomes relative to younger (<60 years) underweight, normal, or overweight patients. The two-sided significance cutoff was p < 0.05. All analyses were performed with the use of R software, version 3.5 (R Foundation for Statistical Computing, Vienna, Austria).

3 Results

Patient-level characteristics of the 292 patients with the minimum of 2-year follow-up are shown in Table 1. The mean (± standard deviation) age of participants was 64.8 ± 7.7 years, 45.9% were male, and mean BMI was 32.3 ± 6.4 kg/m². Surgeries were more common on the left (49.0%) than the right (45.9%), and 14 (4.8%) surgeries were bilateral. There were 116 (39.7%) patients who were underweight, normal, or overweight; 88 (30.1%) were obese; and 88 (30.1%) were severely obese. Representative radiographs of patients with different BMIs are shown in Fiure 1. Overall, there were no noticeable differences amongst radiographic features for different BMI categories. Mean KSS prior to surgery was 69.2 ± 7.7, and KOOS JR was 71.5 ± 6.9. Post-operative KSS, KOOS JR, and VAS were 84.1 ± 3.8, 88.6 ± 7.6, and 1.5 ± 1.2, respectively. Both KSS and KOOS JR improved significantly after BCS TKA (p < 0.001).

While there was minimal (<1%) missing data on age, BMI, sex, and laterality of surgery, there was substantial missing data on post-operative KSS (n = 101, 34.6%). Patients missing data on post-operative KSS had a mean BMI of 33.0 kg/m², mean age of 65.2 years, and 43.5% were male. The only other measure with >10% missing data was pre-operative KSS (n = 39, 13.3%). Overall, comparisons between patients missing data and those without missing data suggested that there were no significant differences between the two groups (see Fig. 1).

Close

Fig. 1 Shown are post-operative radiographs for patients in different BMI categories. (A) and (B) show AP and lateral radiographs for a patient with a BMI of 25 kg/m² (C) and (D) show AP and lateral radiographs for a patient with a BMI of 32 kg/m². Finally, (E) and (F) show AP and lateral radiographs for a patient with a BMI of 50 kg/m².

Export to PPT

Next, we analyzed differences in PROMs based on the predefined BMI subgroups using one-way ANOVA. At baseline, there were no differences in KSS (p = 0.13) or KOOS JR (p = 0.54) (Fig. 2). Post-operative KSS, KOOS JR, and VAS were also similar across BMI subgroups (p > 0.40 for all comparisons). Finally, there was no evidence that changes between pre-operative and post-operative KSS (p = 0.21) and KOOS JR (p = 0.62) differed by BMI subgroup.

Close

Fig. 2 There were no statistically significant differences amongst the body-mass index subgroups with KSS (A) or with KOOS JR (B). The KSS and KOOS JR differences were calculated by subtracting pre-operative scores from post-operative scores.

Export to PPT

We also assessed the linear and monotonic relationship between BMI and PROMs by maintaining BMI as a continuous variable. There was no evidence of a linear association between post-operative KSS and BMI (p = 0.77), or between BMI and post-operative KOOS JR (p = 0.09). There was also no evidence that BMI was associated with the PROM difference, measured as pre-to post-operative changes, for KSS (p = 0.17) or KOOS JR (p = 0.76). The multiple-R² for the difference in PROMs was 0.01 for KSS and <0.001 for KOOS JR, consistent with a weak correlation. Spearman's rho for the difference in PROMs was 0.10 for KSS and 0.01 for KOOS JR.

Finally, we conducted post hoc analyses to determine if extremes of age and BMI were associated with poorer outcomes. We hypothesized that older patients (>75 years) with severe obesity would have superior results, when comparing pre-to post-operative PROMs, relative to younger patients (<60 years) in the underweight, normal, or overweight group. However, there was no evidence of any difference in KSS (p = 0.55) or in KOOS JR (p = 0.47) between the two groups.

4 Discussion

In this study, we found that, for patients undergoing BCS TKA, BMI was not associated with patient outcomes, as measured by KSS, KOOS JR and VAS. All comparisons at the pre-operative stage and the post-operative stage, along with comparisons of the pre-to post-operative differences, found no evidence of statistically significant differences amongst PROMs. While the BCS design has typically been thought to primarily benefit younger and more active patients, the results from this study suggest that BMI and age do not influence PROMs following BCS TKA. A kinematic implant system can be favorable for patients across all demographics.

Large registry studies of TKAs suggest that improvements in PROMs are similar, irrespective of patient BMI. A study of over 13,000 patients who underwent TKA in the United Kingdom found that patients of different BMI classes reported similar results on the Oxford Knee Score, EuroQol 5D index, and the EuroQol 5D Visual Analogue Scale.¹⁶ Although, the authors did note that the incidence of wound complications was significantly greater in the group of patients with the highest BMI. In a study of over 3,300 primary TKA patients in Sweden, Overgaard et al. similarly found that BMI was not associated with PROMs, including KOOS and EuroQol VAS, reported one year after surgery.¹⁷ The proportion of patients satisfied with their TKA procedure ranged from 80 to 83% across different BMI categories. While both studies examined the relationship between BMI and PROMs in a large number of patients, neither study specifically investigated BCS TKA, as was done in this study.

In a systematic review of the literature that incorporated data from nine separate studies, Boyce et al. found that following TKA the Knee Society Objective Score (KSOS) and Knee Society Functional Score (KSFS) were poorer in morbidly obese patients (BMI ≥40 kg/m²) both pre- and post-operatively.¹⁸ In morbidly obese patients, there were also increased rates of hard outcome measures including revision and complications, such as infection. However, there was no evidence of any difference in the improvement of scores, with either KSOS or KSFS. The comparisons of pre-to post-operative improvement in PROMs amongst different BMI groups in the systematic review are in concordance with the results from our study. Of note, however, in our analyses we were unable to assess hard patient outcomes including revision and complication rates with the BCS system.

Several studies have specifically investigated PROMs and revision rates of BCS systems. Christen et al. found that bicruciate TKA resulted in improvement of all KOOS subscores, along with Knee Score (KS), and Knee Functional Score (KFS).¹⁹ A separate report of over 2,000 s generation BCS total knees found that the revision rate (3.1–3.6 revisions per 100 TKAs) was similar to that of posterior stabilized controls.²⁰ The most common indications for revision were infection (33%), mechanical loosening (21%), fracture of bone around the joint (16%), and instability (15%). Nearly half of revisions, 30 of 67 (44.7%), involved isolated tibial insert exchange.

To our knowledge, this was the first study to focus on the effects of different BMIs on PROMs in patients who underwent BCS TKA. Future studies are warranted in order to determine how the effects of BMI on patient outcomes compare across different knee replacement systems. The authors of this study hypothesize that BMI is not correlated with PROMs because kinematics and function of the knee are restored to a similar extent in obese and non-obese patients. However, increased rates of complications in obese patients must be taken into consideration when advising surgery, despite similarities in pain, symptom severity, and activities of daily living amongst patients with different BMIs.

In their study comparing patients who received a TKA with either a BCS or PS knee system, Mugnai et al. concluded that not only were KOOS scores higher at a mean follow up of 29 months for the BCS group, but ROM was greater as well.²¹ This indicates that the kinematic implant system is beneficial to patients not only in terms of reported outcomes, but clinical functionality as well. This topic warrants further investigation. A potential avenue of research to be explored is whether this study's BCS cohort had better functional outcomes across different BMI groups.

There were several limitations to this study. First, there was considerable missing data for post-operative KSS. While we did analyze post-operative KSS, despite more than 30% missing data, there is a concern of selection bias due to the missing data. However, analyses of baseline characteristics amongst patients with and without missing data suggested that the two patient groups were similar. Second, while our study did investigate patient outcomes using PROMs, we were unable to assess hard outcomes including complication and revision rates. Because complications and revisions are outcomes that involve considerable financial costs and substantial health risks, they are particularly important to track following TKA.

In conclusion, we found that, amongst patients undergoing primary BCS TKA, PROMs were similar across different BMIs. PROMs were similar both at baseline and during the post-operative period. The results from this study suggest that BCS TKA is an effective implant system for patients, regardless of age or BMI. The results from this study should help reassure surgeons that PROMS are not inferior for obese patients with BCS TKA implants.

Financial disclosures

The authors have financial disclosures with relation to Canary Medical, other from OrthAlign, personal fees from Conformis, other from Osso VR, other from Parvizi Surgical Innovations, personal fees and other from Smith & Nephew, personal fees from Stryker, personal fees from Surgical Care Affiliates, outside the submitted work.

Ethical clearance

This study was approved by the review board at the Hospital for Special Surgery.

Authors’ contributions

David Kolin: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Resources; Validation; Visualization; Writing – original draft; Writing – review & editing.

Kaitlin Carroll: Conceptualization; Formal analysis; Investigation; Writing – review & editing.

Michael Ast: Conceptualization; Formal analysis; Investigation; Writing – review & editing.

David Mayman: Conceptualization; Formal analysis; Investigation; Writing – review & editing.

Steven Haas: Conceptualization; Formal analysis; Investigation; Writing – review & editing.

Fred Cushner: Conceptualization; Formal analysis; Investigation; Writing – review & editing.