Introduction

High tibial osteotomy (HTO) is recognised as a successful joint preservation surgery for knee osteoarthritis (OA) [14, 25]. The purpose of the HTO procedure is to realign the lower limb to shift the load distribution from the medial to the lateral compartment, thereby decreasing OA symptoms in the medial compartment of the knee [1, 4, 24]. However, opening wedge HTO (OWHTO) has been reported to induce unfavourable changes in patellofemoral mechanics [8, 28] due to a decrease in patellar height after OWHTO [3]. Consequently, the progression of patellofemoral OA after OWHTO has been widely discussed [9, 26]. Recently, HTO with a descending cut at the tibial tuberosity has been introduced for patellofemoral OA, because this procedure does not alter patellar height [16, 20]. While it might be true that this surgical technique does not reduce patellar height, patellofemoral joint congruity might not be improved in cases of symptomatic patellofemoral OA. Closing wedge HTO (CWHTO) is another procedure to address varus knee OA. Patellar height after CWHTO is elevated relative to the tibia [30], which creates a better patellofemoral joint congruity than OWHTO. However, CWHTO is known to result in several issues, such as leg shortening and step-off at the lateral cortex [31]. To reduce these issues, hybrid HTO, which was a novel surgical procedure that combines OWHTO and CWHTO, was established [29]. The hinge point is set at approximately two-thirds of the lateral distance along the proximal osteotomy line to combine the lateral closing and medial opening wedge osteotomy techniques. This is thought to maintain the appropriate positioning of the tibial tuberosity relative to the patellofemoral joint [29].

Although patellofemoral joint congruity is an important point to consider in improving clinical outcomes with HTO, it has not been assessed in any studies, including changes in the tibial tuberosity–trochlear groove (TT–TG) distance and joint space narrowing after hybrid HTO and OWHTO. Therefore, the purpose of this study was to compare changes in patellar height after hybrid HTO and OWHTO based on the hypothesis that hybrid HTO is preferable to OWHTO for the treatment of varus knees combined with patellofemoral osteoarthritis.

Materials and methods

One hundred and four knees that had undergone HTO from 2012 to 2016 were initially enrolled in this study. Pre-operative digital planning software (TraumaCaD, BRAINLAB, Feldkirchen, Germany) was used. First, it was planned to pass the post-operative mechanical axis through the lateral eminence of the proximal tibia in the picture archiving and communication system (PACS) [13]. Double-level osteotomy was selected when the planning indicated that the post-operative medial proximal tibial angle (MPTA) would have been greater than 95°. When the projected MPTA was less than 95°, OWHTO was indicated when the severity of patellofemoral OA was less than stage III, and hybrid HTO was selected when the severity was > stage III [22] or if patients complained of patellofemoral pain and crepitus with OA. As a result, 77 knees (43 males and 34 females) that had undergone OWHTO and 24 knees (13 males and 11 females) that had undergone hybrid HTO were identified. Three knees (1 male and 2 females) had undergone double-level HTO. In our final analysis, 24 knees that had undergone OWHTO were matched by age and gender with 24 knees that had undergone hybrid HTO. Patellofemoral joint congruity between the two groups was compared. Both groups consisted of 13 males and 11 females, while the average patient age was 66.9 (range 44–78) years in the hybrid HTO group and 66.0 (range 44–76) years in the OWHTO group.

Diagnostic and radiographic measurements

Coronal knee alignment after hybrid HTO and OWHTO was evaluated by the percent of mechanical axis (%MA); femorotibial angle (FTA), which is the lateral angle between the centreline of the femur and the tibia on the coronal radiograph in the standing position [15]; and MPTA, whereas sagittal alignment was assessed by the posterior tibial slope [18]. Patellar height was evaluated radiographically using the Caton–Deschamps index [5, 6] and modified Miura–Kawamura index [16]. These measurements are depicted in lateral radiographs in 30° of knee flexion, as shown in Fig. 1a–d. The TT–TG distance was assessed using three-dimensional CT (Fig. 2a–d). Patellar tilt and joint space narrowing, which were measured from the centre of the medial and lateral facets to the patella by drawing a perpendicular line, were also evaluated (Fig. 3a–d) [21]. Clinical outcomes, including anterior knee pain, were assessed using the Kujala anterior knee pain scale [10, 17]. All parameters were measured on radiographs which were taken pre-operatively and at the end of the follow-up period, and were evaluated by two orthopedic surgeons who were blinded to the patients’ information. The mean follow-up duration was 31.0 (range 21–48) months.

Fig. 1
figure 1

Patellar height measurement with hybrid and opening wedge HTO. Pre- and post-operative patellar heights are evaluated with the Caton–Deschamps index (B/A) (a, b), and modified Miura–Kawamura index (C/A) (c, d)

Full size image
Fig. 2
figure 2

Tibial tuberosity–trochlear groove (TT–TG) distance measurement with hybrid and opening wedge HTO. Pre- and post-operative patellar height are shown in hybrid HTO (a, b) and OWHTO (c, d)

Full size image
Fig. 3
figure 3

Patellofemoral joint space measurement. a–d Medial and lateral joint space was measured from the centre of the joint surface to the patella by drawing a perpendicular line. Pre- and post-operative joint space width for hybrid HTO (a, b) and OWHTO (c, d) is shown

Full size image

Surgical procedure

Biplane OWHTO was performed to correct the mechanical axis based on pre-operative planning [19] and the osteotomy was fixed with a locking plate. The surgical procedure for hybrid HTO was performed as previously reported [29]. Briefly, an approximately 10-mm length of the distal third of the fibula was removed and the proximal lateral tibia was exposed by separation of the attachment of the tibialis anterior muscle. Under fluoroscopy, hybrid HTO, a 33% medial opening combined with a 67% lateral closed wedge, was performed. A locking plate was used for fixation of the osteotomy site. Rehabilitation was initiated on post-operative day 1 with 50% weight-bearing allowed initially and full weight-bearing permitted starting on day 14 for both osteotomy procedures.

This study was performed in accordance with a protocol and was approved by the Institutional Review Board of Osaka Medical College (No. Rin-217).

Statistical analysis

A power analysis was performed for the primary comparison of interest (post-operative TT–TG distance and the Caton–Deschamps index between hybrid HTO and OWHTO). The power analysis was set up as follows: α = 0.05 and power = 0.9. As a result, a minimum of 32 knees for TT–TG distance and 24 knees for the Caton–Deschamps index were required for the total sample size based on G*power (version 3.1.9.2) [7]. On the basis of the power analysis, a sample size of 48 was considered adequate. A paired t test was performed to compare the pre- and post-operative radiographic results using JMP Pro (version 11.2.0, SAS, Cary, NC, USA). P values < 0.05 were considered statistically significant.

Results

The mean (± standard deviation) body mass index was 24.5 ± 4.4 kg/m2 for those who underwent hybrid HTO and 24.0 ± 2.4 kg/m2 for those who underwent OWHTO. Four patients had Kellgren–Lawrence grade II OA, 10 had grade III OA, and 10 had grade IV OA in the hybrid HTO group, while 3 had grade II OA, 12 had grade III OA, and 9 had grade IV OA in the OWHTO group. There were no significant differences in baseline characteristics between the two groups (Table 1). However, patellofemoral OA was more severe in those that underwent hybrid HTO than OWHTO (Table 1, p = 0.003). The average correction angle for hybrid HTO was 11.0 ± 2.9°, which was not significantly different than the average correction angle for OWHTO (10.6 ± 2.6°, Table 2). Regarding the intra- and inter-observer reliability of these parameters, pre- and post-operative %MA, FTA, MPTA, and posterior tibial slope were not significantly different between the hybrid HTO and OWHTO groups (Table 2). The Caton–Deschamps and modified Miura–Kawamura indices decreased significantly from 0.91 ± 0.12 to 0.73 ± 0.15 and from 1.06 ± 0.09 to 0.84 ± 0.07 after OWHTO, respectively (Table 3, p < 0.01), whereas no difference was seen after hybrid HTO (Table 3). Post-operatively, these indices were significantly different between the two groups (Table 3, p < 0.001). TT–TG distance after hybrid HTO decreased significantly from 11.4 ± 2.2 to 7.4 ± 3.0 (p < 0.01), whereas no difference was noted following OWHTO (Table 4). Although pre- and post-operative patellar tilt were not altered in either group, post-operative medial joint space at the patellofemoral joint was significantly increased with hybrid HTO compared to OWHTO (Table 5, p = 0.035). The pre-operative Kujala scores were significantly lower in the hybrid HTO group than in the OWHTO group (Table 5, p = 0.007); however, outcomes improved in both groups after HTO, and there were no significant differences between the groups post-operatively (Table 5). All radiographic parameters demonstrated good (0.5–0.75)-to-excellent (> 0.75) intra- and inter-observer reliability (Table 6).

Table 1 Patient characteristics
Full size table
Table 2 Changes in knee alignment after hybrid HTO and OWHTO
Full size table
Table 3 Changes in patellar height after hybrid HTO and OWHTO
Full size table
Table 4 Changes in TT-TG distance after hybrid HTO and OWHTO
Full size table
Table 5 Changes in patellofemoral joint congruity with hybrid HTO and OWHTO
Full size table
Table 6 Intra- and Inter-observcr reliability of radiographic measurement
Full size table

Discussion

The most important finding of this study is that hybrid HTO is the better surgical procedure for improving patellofemoral joint congruity with regards to patellar height and reduction of the TT–TG distance as compared with OWHTO. Hybrid HTO has the potential to improve the pre-operative medial and patellofemoral OA symptoms.

The advantages of hybrid HTO, which are first reported by Takeuchi et al. [29], are the removal of a smaller wedge of bone from the lateral osteotomy site and the preservation of pre-operative leg length compared with CWHTO. Patellar height might be diminished after hybrid HTO, because the insertion of the patellar tendon is elevated anteriorly and repositioned proximally during the hybrid HTO procedure.

Several reports have described changes in post-operative patellar height after HTO procedures [12, 18, 23] and the degree of change is correlated with the magnitude of the correction angle [2, 30]. Gaasbeek et al. compared patellar height before and after OWHTO and CWHTO and reported that the patella descended after OWHTO and ascended after CWHTO. Moreover, they reported that a larger correction angle resulted in greater alterations in patellar height after both [8]. The current study was able to compare the correction angle between two types of procedures, because the two procedure groups were matched based on patient characteristics. Consequently, it was clear that patellar height was significantly decreased after OWHTO, whereas no change occurred after hybrid HTO.

A descending cut of the tibial tuberosity with OWHTO might be another option for maintaining patellar height [16, 20]. Descending OWHTO preserves, whereas ascending OWHTO decreases patellar height according to the Caton–Deschamps index [16]. This has been described in both coronal and sagittal alignments as evaluated by several parameters; however, TT–TG distance has not been evaluated after HTO. TT–TG distance has already been described in varus OA and correlated with the severity of patellofemoral OA [22]. Earlier findings indicated that lateralization of the tibial tuberosity resulted in the deterioration of patellofemoral OA and that hybrid HTO is able to transfer the tibial tuberosity in a direction of medialisation, reducing patellofemoral contact pressures [27], unlike descending OWHTO. Modifications of the OWHTO technique might be necessary in patients with symptomatic patellofemoral OA. Hybrid HTO can correct the medialisation of the tibial tuberosity and this indicates that hybrid HTO is the better surgical option for varus knees with symptomatic patellofemoral OA.

Several parameters have been used to evaluate patellar height after HTO [12, 18]. The method that we chose, the Caton–Deschamps index, was the first used to assess patellar height after HTO by Tigani et al. [30]. In addition, Amzallag et al. reported patellar heights after HTO and determined that the Caton–Deschamps index was more reproducible [2]. The modified Miura–Kawamura index has been recently used to focus on evaluating patellar height after HTO, because it is not affected by tibial factors [18]. In the current study, the Caton–Deschamps and modified Miura–Kawamura indices showed similar changes, because alteration of the posterior tibial slope was similar between hybrid HTO and OWHTO. Both parameters might be useful if the pre- and post-operative posterior tibial slope is well controlled.

The grade of patellofemoral OA in the hybrid HTO group was more severe than that in the OWHTO group, and the pre-operative Kujala score was lower in the hybrid HTO group than in the OWHTO group. However, post-operative scores between the groups were not different. These findings indicate that hybrid HTO might be a potential therapy for patellofemoral OA rather than OWHTO.

Femoral anteversion is associated with trochlear morphology and tibiofemoral contact pressures [11, 32]. Although hybrid HTO has the potential to address symptomatic patellofemoral OA in varus knees, hybrid HTO should not be considered for symptomatic patellofemoral OA in varus knees that are possibly affected by femoral anteversion.

Conclusion

Improved patellofemoral joint congruity is considered an advantage of hybrid HTO over OWHTO. Hybrid high tibial osteotomy is superior to medial opening high tibial osteotomy in patients with symptomatic varus knee combined with patellofemoral OA.