Combination of external fixation using digital six-axis fixator and internal fixation to treat severe complex knee deformity

Complex knee deformity with lower limb discrepancy (LLD) is always caused by congenital factors or injury during the period of epiphyseal development. They both affect the lower limb alignment and lead to hip, knee or other severe problems. Because of the deformity of lower limb alignment, the knee compartment is relatively stressed out and degeneration becomes faster [1]. When the degeneration reaches a certain extent, it will only be treated by knee replacement to recover. Distal femoral osteotomy (DFO) is the major surgical approach to treat knee valgus/varus and early osteoarthritis, which can treat malalignment and avoid total joint replacement [2]. LLD can be one of the associated symptoms of knee valgus/varus and can lead to the change of gait. Eventually, the spine will be involved [3]. For simple LLD, Ilizarov external fixator can provide a stable fixation and promote natural bone growth [4]. But for LLD patients accompanied with knee deformity, it is hard to treat by Ilizarov external fixator. Taylor spatial frame (TSF) is a promising method in multiplanar deformities correction and limb lengthening [5, 6]. However, the TSF is hard to operate and the ring is required to be as perpendicular to the tibia as possible. To provide an easy and practical external fixator in the correction of LLD with knee deformity, we designed a digital six-axis external fixator of shape letter Q. Due to the special shape and its designer Prof. Qiao, the fixator was named Q spatial fixator (QSF) (Fig. 1). Depending on the data obtained from computed tomography (CT) scan and the structural optimization, QSF is easier to operate and study, as well as more precise.

This study included 24 patients (25 knees) who were treated with the QSF and internal fixation at our department for complex knee deformity from January 2018 to February 2021. Under the assistance of a self-designed angle adjustable osteotomy guide (AAOG) [7], all patients underwent the distal femoral osteotomy (DFO) as well as the tibia and fibular osteotomy, and then, QSF was mounted to the tibia. Follow-up time after operation was > 6 months (3 months after removing QSF). There were 13 women and 11 men with a mean age of (31.8 ± 11.7) years (range 18–55 years). The mean body mass index (BMI) was (25.1 ± 3.4) kg/m² (range 19.8–31.2 kg/m²). Six patients had primary knee valgus, and 18 patients had primary knee varus. There were 9 cases of both femoral and tibia sides discrepancy and 15 cases of tibial sides discrepancy. Two patients underwent internal fixation and QSF fixation at both lower limbs. Five patients underwent internal fixation and QSF fixation at one side while internal fixation at another side. All patients did not take surgery before. The study protocol was approved by the Institutional Review Boards and the Ethics Committees. Before surgery, informed consent was obtained from all patients after a full explanation of the therapeutic procedure.

Lower limb discrepancy and alignment were measured by full limb length and hip–knee–ankle angle (HKA) (Fig. 2). HKA was the angle between the mechanical axis of the femur and tibia. To measure the full limb length, the center of the knee was determined first. The length between the center of femoral and the center of the knee was defined as the length of the femur. The length between the center of ankle and the center of the knee was defined as the length of the tibia. The sum of the femur length and the tibia length was defined as the lower limb discrepancy because the aimed HKA angle was 0° [8]. Indications for surgery included: Both femur and tibia sides of lower extremities have the deformities, HKA varus or valgus > 10°. Indication for DFO was: lateral distal femoral angle (LDFA) < 85°. Indications for QSF correction and tibial lengthening included: discrepancy of the lower limb length > 2 cm and medial proximal tibial angle (MPTA) > 5° after DFO simulation.

Related inspections and full-Length weight-bearing radiograph of lower limbs were completed after admission to get preoperative plan. The correction goal was LDFA = 90°. The operative plans were made based on Solidworks™ software. First, we got the height of the osteotomy and the angle of the wedge. Then, we simulated the DFO. Based on the simulated postoperative lower extremities X-ray in the standing position, we measured the difference of the length between the two lower extremities to decide whether the patient should have taken the tibial lengthening and to which extent should the length be extended.

The correction angle of MPTA and the plan of lengthening were decided after operation. Details are as follows: Postoperative spiral CTs were used to calculate the exact difference of HKA and the length of bilateral lower extremities again. Based on the target lower limb alignment and the target lower limb length, we calculated the limb length to be extended and the correction angle by the digital simulated reverse engineering [9] and got the correction plan which was called correction prescription (Table 1).

Mean discrepancy length of operative side was 2.39 ± 1.04 cm (range 0.9–4.4 cm) preoperatively. The mean difference of lower limb was 0.32 ± 0.13 cm (range 0.11–0.58 cm) postoperatively. The length of limb correction had significant difference (p < 0.05). The mean HKA was 15.28 ± 6.05° (range − 19.2°–29°) preoperatively and 1.95 ± 0.94° (range − 2.5° to 4.1°) postoperatively. The lower limb alignment was corrected significantly (p < 0.05). The mean time for the removal of the external fixator was 112.8 ± 17.9 days (range 83–147 days).

All patients were followed for more than 3 months after removing the external fixator and the total follow-up, which were totally more than 6 months. MAD decreased from (51.27 ± 20.72) mm (range 26.2–112.0) to (5.59 ± 3.68) mm (range 2–15.1 mm). MPTA increased from (73.23 ± 11.41°) (range 50.40–96.3) to (87.38 ± 1.68°) (range 82.20°–90°). LDFA increased from (79.10 ± 3.85°) (range 70.70–84.3) to (88.11 ± 1.25°) (range 85.60°–90.5°). AKSS-O improved from (71.15 ± 12.75) points (range 50–95points) to (94.81 ± 5.74) points (range 80–100 points) (t = 13.735, P = 0.000). AKSS-F was improved from (62.92 ± 11.38) points (range 35–77 points) to (90.27 ± 5.46) points (range 78–100 points) (t = − 16.201, P = 0.000). Tegner Activity Score improved from (2.81 ± 1.47) points (range 1–6 points) to (6.19 ± 1.79) points (range 2–9 points) (t = 17.558, P = 0.000). All differences had statistical significance when p < 0.05. The knee function and motor function were improved significantly (Table 2).

Obvious continuous callus was formed at the osteotomy site. None of the patients said it had an obvious negative impact on quality of life. There were no complications such as nonunion and pin sites infection. No infection was occurred with only several cases which had small amount of pin sites secretors. For these cases, it was recovered naturally after removing the QSF with regular disinfection. Re-examination was done 8 weeks after the removal of the external fixator. All patients returned to normal lives and no fractures occurred. Typical cases are shown in Figs. 3 and 4.

The abnormal excessive pressure imposed on the knee compartment is associated with the initiation and progression of osteoarthritis [10]. Knee valgus/varus deformity combined with LLD is common and hard to treat. Because it is a complex and multiplanar deformity, only internal fixation is unable to correct this complex deformity. While lengthening the limb, the rotation of the affected limb also needs to be corrected. Compared with traditional treatments, hexapod fixator can distract bone progressively to correct deformities from different planes and has been widely used in deformity correction [11, 12]. Based on the hexapod principle, we developed a digital six-axis fixator. The results of our study demonstrated that the combination of our self-designed QSF and internal fixator was an effective treatment in patients with severe knee deformities combined with LLD. Lower limb length and alignment of patients were corrected to normal range. All patients had significant improvement in their AKSS-O, AKSS-F and Tegner Activity Score.

Distal femoral osteotomy (DFO) is a surgical method to treat osteoarthritis and knee deformities. Particularly, when valgus deformity is > 12° or joint line obliquity > 10°, DFO is the best choice for correction [13]. Though total knee arthroplasty has a good correction effect and long-term result, its high cost and great harm are an obstacle for many patients [14]. Also, knee arthroplasty is not suitable for young patients. In our research, the average age of our patients was only 33 years old and the oldest patient was 55 years old. Therefore, we used DFO for the deformity correction. The correction accuracy is significantly associated with the outcome and survival time for patients. Felson et al. demonstrated that the outcome of valgus alignment depends on the valgus mechanical axis. When mechanical axis ranges from 1.1° to 3° valgus, the risk of OA progression increases. And valgus alignment > 3° is associated with OA incidence [15]. Some studies showed that the accuracy of internal fixation only was not perfect. Overcorrection and undercorrection happened sometimes [16]. Internal fixation also has the risk of rotational malalignment [17]. External fixation can compensate the correct deviation of internal fixation by changing MPTA. But conventional Ilizarov external fixator is not proper for multiplanar deformities correction because it was major for lengthening and axial correction. Hexapod external fixator has been proven to be effective in multiplanar deformities correction with high precision [18, 19].

LLD is very common in public. 90% of the normal people are different in lower limb length, and the average discrepancy was found to be 5.2 mm [20]. Most time the length discrepancy does not affect the function and gait. It will be compensated by mild passive structural changes. But those compensations are not enough for the significant length discrepancy [21]. Gait analysis indicates that when the length discrepancy is > 1 cm, the gait will be asymmetry [22]. A cohort study indicated that the length discrepancy > 2 cm was associated with the knee OA progression and patients will have clinical symptoms in standing [23]. Conservative treatment like shoe lift is recommended by some guidelines [22]. But a systematic review showed that the effect of shoe lift to improve function and relieve pain is uncertain because the evidence of quality was low [24]. Therefore, 2 cm is a generally accepted indication for surgery of LLD [25, 26]. External fixation has long been the major method to treat LLD since the invention of Ilizarov external fixator [27]. For patients have severe knee deformity with LDD, hexapod external fixator can treat both LDD and deformity. In our study, the least discrepancy was only 0.9 mm. The only discrepancy did not have to correct. But the MPTA of this patient was 74.2°, and the deviation may be caused by the metaphyseal deformity instead of the bone development. During the process of correction, an external fixator can adjust the length at the same time to make the limbs tend to be equal in length. Infection risk should also be taken into consideration in the application of internal fixation with QSF. In our study, no infections occurred and only some cases had a small amount of pin sites secretors. Nolte et al. revealed there was no difference between pin sites debridement and simply disinfection in external fixation [28]. Therefore, patients with pin sites secretors were asked to simply disinfection regularly. After removed the pin, they were recovered naturally.

The reasons for the indications of our study (LLD and MPTA difference) were: those who had length discrepancy without MPTA difference, DFO was not essential, and Ilizarov external fixator was effective in limb lengthening [29], for those who had MPTA difference without LLD, and internal fixation was sufficient to meet correction needs. All the patients took osteotomy at the femoral sites. Because the distal lateral femoral angle (DLFA) of some patients was small, osteotomy at the tibia site led to the joint line oblique while at the femoral site did not alter it [30].

The limitation of our study is the small number of patients. Although all our patients achieved satisfactory results, we did not observe any complications like nonunion and needle tract infections of the external fixator [31]. We should enlarge our samples and assess the safety of both our surgery method and external fixator. The data for correction were from CT scans and computer analyses before surgery. To save cost, we choose X-ray to measure the limb length after surgery and in re-examination. So the real results were better than measured by X-ray.

Knee valgus combined with LLD can be treated by QSF with internal fixation without the need for total knee arthroplasty. Lower limb malalignment and discrepancy can be corrected precisely and effectively by this approach. Patients can walk soon after the procedure and avoid undergoing knee arthroplasty in the advanced stage of OA.