Introduction

Cemented fixation is the gold standard in total knee arthroplasty (TKA), and successful long-term results have been reported in clinical trials (1,2). However, the lack of remodeling capacity of bone cement has raised questions regarding the long-term results of cemented TKA, especially in young and active patients due to the effect of third-body wear (3). Aseptic loosening is most commonly observed in tibial components in TKA, whether cemented or not (4,5). Since the idea of hybrid components emerged in the late 1980s (6), hybrid TKA has become a preferred method in some centers, and comparable success rates have been reported (7,8).

Whether cemented or not in unicondylar knee arthroplasty (UKA), the general view is that the tibial component is at greater risk of aseptic loosening. In addition, if there is pain in the first year postoperatively due to increased stress in the proximal tibia and subchondral bone marrow edema, it can be easily misdiagnosed as aseptic loosening (9,10).

It is generally accepted that uncemented UKA provides biological adherence to bone through bony ingrowth, with the associated theoretical advantages of eliminating cement complications and shortening operation and tourniquet time (11,12). In spite of these advantages of a cementless prosthesis, there is no clear consensus in the literature on the fixation method of the tibial component. Although there are comparative publications in the literature about cemented and cementless UKA, there are no comparative publications on cementless and hybrid (a combination of an uncemented femoral component and a cemented tibial component) UKA. The aim of this study was to compare the early clinical and radiological results of cementless and hybrid UKA.

Methods

After approval was obtained from University of Health Sciences Turkey, Dışkapı Yıldırım Beyazıt Training and Research Hospital Local Ethics Board (approval number: 56/07, date: 12.11.2018), this retrospective cohort study was initiated. A retrospective review was performed of 78 consecutive patients with a medial knee arthrosis treated with UKA between January 2014 and December 2018. Radiological data were collected from the picture archiving and communication system.

The clinical and radiological results were compared for patients treated with an Oxford hybrid UKA and uncemented UKA, by a single surgeon (H.A.) with a follow-up of at least 2 years between 2014 and 2017. The demographic properties of the patients are given in Table 1.

The indication for cement usage was given according to the bone hardness test in patients undergoing surgery with the indication of UKA (13). In this test, if the bone surface collapsed (thumb penetrates the bone tissue) when pressure was applied to the trabecular bone surface with the thumb after the tibia was cut, it was deemed unsuitable for the uncemented tibial component, and the cemented one was applied.

The patients’ clinical results were evaluated using the Oxford knee score, EuroQol-5 dimensions (EQ-5D-3L), EQ-visual analog scale, and the knee injury and osteoarthritis outcome scores (pain, symptom, daily life, sports, and quality of life). Joint range of motion was measured at the final visits. Radiological examination was performed by two independent orthopedic surgeons. Measurements were performed on the patients’ anteroposterior and lateral knee radiographs. The varus-valgus and flexion-extension angles of the tibial and femoral components were measured to determine component alignment on radiological evaluation (Figure 1, 2). The incidence of radiological lines at the tibial component-bone and femoral component-bone interfaces was also evaluated.

The bone contact points of the tibial component were divided into six regions according to the technique described by Pandit et al. (14) Zone 1: from the medial tibial plateau to the medial keel, Zone 2: from the medial plateau to the lateral keel, Zone 3: the medial vertical surface of the keel, Zone 4: the lower part of the keel, Zone 5: the lateral vertical surface of the keel, and Zone 6: the lateral to the keel of the medial plateau (Figure 3). Additionally, the bone contact points of the femoral component were measured by a similar technique (Figure 4).

During the follow-up period, insert dislocation occurred in two patients in both groups. All four patients underwent replacement with a 1 mm-thick insert.

Statistical Analysis

The statistical analysis was performed using SPSS for Windows (SPSS Inc, Chicago, Illinois). Odds ratios and means were compared between groups, with 95% confidence intervals. Results were considered statistically significant at p<0.05. A comparison of proportions method was used to calculate the sample size.

Results

The clinical results of both the cementless and hybrid groups are given in Table 1. Although there was a proportional difference in clinical results in both groups, there was no statistically significant difference between the two groups.

When the femoral components were evaluated in terms of alignment, in the cementless group, the mean varus-valgus angle was 5.6±2.1 (-6.3-11.0), and the mean flexion-extension angle was 9.4±3.3 (6.5-13.3); in the hybrid group, the mean varus-valgus angle was 7.1±3.3 (-8.4-13.2), and the mean flexion-extension angle was 6.8±4.4 (-5.6-12.2). When the tibial components were evaluated in terms of alignment, the mean varus-valgus angle in the cementless group was -0.5±2.3 (-1.4-4.9), and the mean flexion-extension angle was 4.5±2.7 (-2.6-6.8), and in the hybrid group, the mean varus-valgus angle was -1.1±3.4 (-2.1-5.5), and the mean flexion-extension angle was 4.8±2.5 (0.1-9.3). There was no statistically significant difference between the varus-valgus and flexion-extension angles of the femoral and tibial components in both groups (p>0.05).

The radiolucent (RL) areas in each zone, the total number of RL areas, and the numbers of complete and partial RL areas in the patients were compared between the groups (Table 22). When the incidence of RL lines was evaluated in both groups, no complete RL line was observed in any prosthesis. When evaluations were performed of the tibial components alone, a partial RL area was detected in 7 (15.2%) of the cementless group and 12 (35.2%) of the hybrid group. The incidence of the RL line at the tibial component interface was higher in the hybrid group than in the cementless group, but the difference was not statistically significant (p=0.069). When partial RL involvement was evaluated according to the region, a total of 13 regions in the cementless group and a total of 32 regions in the hybrid group were detected. This difference was statistically significant (p=0.025). When all the tibial components were considered in the cementless group, the highest rates of partial RL region involvement were determined in Zone 3 and Zone 5, respectively. The total of all partial RL regions was 13. In the hybrid group, the highest rates of partial RL region involvement in the tibial components were seen in zones 1, 5, and 2, respectively.

When the femoral components were evaluated among themselves, a partial RL area was detected in one patient of the cementless group and in two patients of the hybrid group. There was no statistically significant difference in the incidence of the partial RL line at the femoral component interface (p=0.691).

Discussion

There is no comparative study in the literature regarding hybrid vs all uncemented or all cementless UKA. Therefore, the results of this study could only be compared with the results of hybrid vs cementless fixation in TKA.

Cement application is the most widely used fixation method in TKA compared with the cementless and hybrid options. However, there is confusion as to which is the optimum fixation method. In view of the potential late loosening and third-body wear effects of cement, cementless implants have been developed as an alternative method that allows biological fixation and has potential advantages such as the preservation of bone stock for revisions (3,15,16) However, some studies have reported that cementless fixation leads to instability and loosening, especially in the tibial component (17,18). To avoid these problems in the cementless tibial component, the idea of hybrid application was proposed and successful medium- and long-term results of hybrid TKA have been reported (19-22). Behery et al. (23) found that 76 cases of aseptic loosening comprised 10% with cementless prostheses and 0% cemented. Early weight-bearing and obesity were reported to cause loosening. Another interesting point of that study was that all the loosened tibial components were seen in patients with a cruciate retaining prosthesis. The authors attributed this to the different femoral rollback kinematics in the two design types, with a different pattern of tibial weight-bearing, which could be implicated in micromotion and failure of osseointegration. In the present study, aseptic loosening was not observed in both cementless and hybrid UKA groups. The probable reason for the lack of aseptic loosening in both groups may be the inadequate follow-up period. However, the higher number of tibial RL lines seen in the hybrid group may be a predictor of future aseptic loosening.

In UKA, the fixation evolution of the components undergoes a process similar to that of TKA. Cementless fixation of tibial and femoral components has been accepted as a widely used method in recent years, whereas cemented prosthesis had previously been the generally accepted method (24). As in TKA, the femoral component in UKA is at a lower risk of loosening than the tibial component. However, there are different views on cementless application of the tibial component in patients with high-level activity expectations, obese patients, and patients with osteoporosis. The design group has argued that the use of cementless application in osteoporotic patients is not a contraindication and does not include any indications different from those for cemented versions (25). In a study of 12 cadaveric knees, Jaeger et al. (26) emphasized that bone quality is important in implant choice, and if bone quality is poor, it will lead to subsidence of the implant. Similarly, in the present study, there was no significant difference between the two groups which had patients with similar demographical properties in terms of RL lines seen in the femoral components.

Stempin et al. (13) stated that they made the decision to use cement intraoperatively by observing the quality of bone visually and by applying the bone hardness test. Accordingly, after the tibial cut is made, pressure is applied to the bone surface with the thumb (index finger in small knees), and a slight deflection is observed on the bone surface. If the resected surface collapses, the stiffness of the bone is considered to be insufficient to provide primary stability of the implant and a cemented component is used (13). In contrast, Campi et al. (11) reported that the problem of bone quality will not affect the results. They stated that as the forces primarily transmitted are compressive, the implant works well with cementless fixation, and neither bone density nor patient age affects the success or failure of a cementless fixation (27). The decision to apply the tibial component with or without cement was also made in this present study according to the bone hardness test.

In the literature, no significant difference has been reported between the functional outcomes of cemented and cementless UKA (28).

Likewise, in the present study, there was no significant difference between cementless and hybrid UKA with respect to functional outcomes.

In the current series, as there was no fracture due to press-fit implantation, the risk reduction can be explained by strict adherence to the technique described by Campi et al. (11), including adequate clearing of peg and keel slots, avoidance of damage to the posterior cortical bone, and delicate impaction using a small hammer to avoid causing a fracture.

Another reason for the success of the current series may be that all operations were performed by a single surgeon and the team is experienced in UKA. The learning curve for UKA is considered to be long, so experience provides minimization of surgical errors. Therefore, the postoperative results can be evaluated without the bias of technical errors.

Study Limitation

There are a few limitations associated with this study, primarily its retrospective nature and the limited sample size. For arthroplasty cases, a 2-year follow-up period may not be sufficient. The osteolysis and UKA loosening in this early period may be due to undetected infection, inappropriate surgical technique, and/or the choice of fixation (cemented or cementless). The results of this study describe a single surgeon’s experience and, therefore, may not be generalizable.

The most important feature of this study is that it is the first study in the literature to compare the clinical and radiological results of cementless and hybrid UKA.

Conclusion

The most important finding of the study was that cementless tibial components showed significantly less RL than components of the hybrid group. Cementless and hybrid UKA have been used as a popular method in recent years and are preferred by many surgeons, and now it may be possible to predict what problems might be encountered, especially in obese and osteoporotic patients in long-term follow-up.

Ethics

Ethics Committee Approval: This study approval was obtained from University of Health Sciences Turkey, Dışkapı Yıldırım Beyazıt Training and Research Hospital Local Ethics Board (approval number: 56/07, date: 12.11.2018).

Informed Consent: Retrospective study.

Peer-review: Externally peer-reviewed.

Authorship Contributions: Surgical and Medical Practices - H.A., M.A.; Concept - H.A., H.A.A., M.A., K.B., H.B.Ç.; Design - H.A., H.A.A., M.A., K.B., H.B.Ç.; Data Collection or Processing - H.A., H.A.A., M.A., K.B., H.B.Ç.; Analysis or Interpretation - H.A., H.A.A., M.A., K.B., H.B.Ç.; Literature Search - H.A., H.A.A., M.A., K.B., H.B.Ç.; Writing - H.A., H.A.A., M.A., K.B., H.B.Ç.

Conflict of Interest: No conflict of interest was declared by the authors.

Financial Disclosure: The authors declared that this study received no financial support.