Total Knee Arthroplasty in Patients with Rheumatoid Arthritis, Severe Osteoporosis, and Genu Valgum

Background: Rheumatoid arthritis (RA) is a systemic inammatory disease characterized by symmetric, relapsing, or chronic destructive synovitis. The aim of this retrospective study was to assess the clinical benets of TKA in patients with severe osteoporosis and moderate genu valgum. Methods: This was a retrospective study of patients with RA, osteoporosis, and genu valgus treated using TKA at the Fourth Aliated Hospital of Harbin Medical University between 05/2016 and 05/2019. Posterior stable prostheses were used in all patients. The Knee Society Scoring System (KSS) was used. Results: There were eight males with 10 knees involved and 24 females with 34 knees involved. Healing by rst intention was achieved in all patients. All patients received follow-up, with the mean follow up time of 2.9 (1-4) years. At the last follow-up, the KSS clinical score was improved from 32.1±9.8 to 82.0±5.8, (P<0.001). The KSS functional score of the patients was improved from 38.3±9.3 to 88.1±10.7 (P<0.001). The exion-extension ROM of the knee joint was improved from 81.8°±27.9° to 105.8°±17.0° (P<0.001). The femorotibial angle (FTA) was changed from 23.1°±2.8° to 6.8°±1.9° (P<0.001). Common peroneal nerve paralysis and periarticular infection each occurred in one patient. Conclusion: TKA via the medial parapatellar approach with proper osteotomy and soft tissue balancing could restore the knee function in patients with RA, severe osteoporosis, and genu valgum.

In addition, patients with RA also generally suffer from severe osteoporosis [5,6], which in turn may lead to severe pain and functional limitation in the knee joint [7], and thus require total knee arthroplasty (TKA) [8,9]. Such patients are generally accompanied by joint deformity, especially genu valgum [10], which could increase the di culties of TKA and increase the risks of developing postoperative complications such as common peroneal nerve injuries and patellar maltracking [8,9]. Proper intraoperative osteotomy and good soft tissue balancing, as well as gentle and correct procedures, are key factors in uencing the postoperative prosthesis stability and restoration of joint functions in patients with RA who underwent TKA [8,9]. Nevertheless, no studies have yet comprehensively reported the effectiveness of TKA in RA patients with severe osteoporosis and genu valgum.
The aim of this retrospective study was to assess the clinical bene ts of TKA in patients with severe osteoporosis and moderate genu valgum at a single center. The results could provide insights into the management of such patients. The inclusion criteria were: 1) preoperative femorotibial angle of 18°-27°; 2) rst-ever TKA; 3) TKA through the medial parapatellar approach; 4) severe osteoporosis, de ned as mean bone mineral density (BMD) Tvalue measured by dual-energy X-ray of -3.7 ± 0.6 [11]; and 5) follow-up of at least 1 year. The exclusion criteria were: 1) femorotibial angle > 27°; 2) received revision surgery; 3) TKA through the lateral parapatellar approach; 4) dual-energy X-ray absorptiometry (DEXA) showed that the T value of BMD >-3; and 5) poor compliance to follow-up.

Data collection
Demographic (age and sex), anthropometric (height, weight, and body mass index (BMI)), and clinical (symptoms and range of motion (ROM)) were collected from the charts. Genu valgus was classi ed according to the Ellison's valgus deformity classi cation criteria [12]. The Knee Society Scoring System (KSS) was used [13]. The anteroposterior and lateral X-ray images of the knee joint, as well as the loaded full-length X-ray image of bilateral lower extremities, were taken routinely. DEXA (Lunar DPX-NT; GE, Boston, Massachusetts, USA) was used to measure the bone mineral density (BMD) of the lumbar vertebral body, bilateral femoral necks, Ward's trigonum, and greater trochanter of all the patients on the anteroposterior images. All patients had to meet the diagnostic criteria of osteoporosis, as in the Suggested Diagnostic Criteria of Osteoporosis in Chinese People (2nd edition) issued in 2000 [11].

Surgical methods
All operations were conducted by the same team of surgeons, which was led by a chief surgeon with 28 years' experience in orthopedics. The rst assistant was an associate chief surgeon with 15 years' experience in orthopedics. Post-stable prosthesis (Link; Germany) was used for all patients, xed using bone cement. After anesthesia (general anesthesia in 17 patients, and continuous epidural anesthesia in combination with subarachnoid space block anesthesia in 27 patients), the patients were placed in the supine position. Pneumatic tourniquets were applied, with pressure of 45-60 kPa. After continuous use of 90 min, the tourniquets were de ated for 15 min before being in ated again. For all patients, a longitudinal incision at the medial knee was made, and tibial and femoral osteotomies were conducted through the medial parapatellar approach. The highest point at the medial plateau was used as the reference point for the measurement to ensure that the highest thickness of osteotomy was ≤ 14 mm.
Distal femoral osteotomy was conducted with the knee physical valgus angle (KPV) of 5°-7°. For patients with lateral femoral condyle dysplasia or bone defects, the highest osteotomy amount in the distal femur was no more than 16 mm. During joint capsule release, secondary osteotomy was conducted through the tibial plateau or distal femur to increase the exion-extension gap. The residual tibial plateau or defects of femoral condyle after osteotomy were repaired as follows. For peripheral bone defects with depth < 5 mm, bone cement was lled for reparation. For bone mass cystic degeneration of the tibial plateau induced by severe osteoporosis, which was found in 1 patient in this study, long-stem post-stable prosthesis (LINK, Weiliande Orthopaedic Technology co. LTD, Beijing, China) was used for the reparation.
Since simultaneous bilateral TKA involved longer anesthesia, this was only carried out for patients deemed t to undergo such procedure. Otherwise, staged surgery was performed. The Ranawat technique [14] was used for soft tissue release. According to the tension of the lateral soft tissues, step-wise release from posterior-medial to anterior-lateral was conducted. After osteotomy and release of lateral soft tissues were completed, the prosthesis was tested to assess the stability at the medial and lateral knee joint. No patellar replacement was conducted, but the patellofemoral articular surface was reconditioned. Denervation in the surrounding tissues and resection of surrounding osteophyte were also conducted. For patients with patellar maltracking or accompanied with lateral patellar subluxation, placing of the prostheses of femur and tibia slightly at the lateral side, the release of the lateral patellar soft tissues, and tight suturing at the medial patella were conducted to improve the conditions [15].

Postoperative management
Prophylactic use of antibiotics was performed routinely after operation. A drainage tube was indwelled in the articular cavity. Compression bandaging was conducted. Cryotherapy with an ice bucket for 48 h was conducted. CPM functional training was started at 24 h after operation, with the ROM increased gradually to reach a exion range of 90° 1 week later. The compression bandaging and the drainage tube were removed 24 h later. Subcutaneous injection of low-molecular-weight heparin calcium injection (Fraxiparine; 0.4 mL/d) was conducted for 14 days. Oral intake of Rivaroxaban (Xarelto) was conducted for two months after discharge to prevent deep venous thrombosis (DVT). A plantar ankle pump was used to prevent lower extremity deep venous thrombosis (LEDVT). The patients were allowed to stand on the oor at 48-72 h after the operation and to receive functional training by walking with the help of walking aid. The suture was removed, and the patients were discharged at 2 weeks after the operation. The patients were followed at 6 weeks, 3 months, 6 months, and 1 year at the outpatient department. The KSS clinical and functional scored, exion-extension ROM of the knee joint, and full-length X-ray image of bilateral lower extremities were assessed. The patients were also guided with functional training during follow-up. Standard anti-osteoporosis therapy was conducted after operation, including long-term intravenous dripping of bisphosphonates.
Statistical analysis SPSS 13.0 (SPSS Inc., Chicago, IL, USA) was used for statistical analysis. Data were described with means ± standard deviation and ranges. The paired t-test was used for comparisons pre/post-surgery. Two-sided P-values < 0.05 were considered statistically signi cant.

Results
Characteristics of the patients (Table 1)  here were eight males with 10 knees involved and 24 females with 34 knees involved. The mean age was 68.5 years (54-82 years), the mean body weight was 68.2 kg (56-108 kg), the mean height was 161 cm (155-177 cm), and the mean BMI was 26.2 ± 4.1 kg/m 2 . Genu valgus in all patients were moderate (18°-27°). The mean disease course of the patients was 8.2 years (3-21 years). The major clinical manifestations were knee joint deformity, pain, and limitation of motion. The pain was mainly at the lateral joint space and patellofemoral joint. The exion-extension ROM of the knee joint was 81.8° ±27.9°, and 17-knee with exion contracture of 11.4°±10.4°. The preoperative KSS was 32.1 ± 9.8, and the functional score was 38.3 ± 9.3. For all patients, the inversion angle measured by X-ray was 23.1°±2.8°. In addition, the lateral articular space of the femorotibial joint was narrower in all patients.

Surgical characteristics
Among the 12 patients with bilateral knees involved, 5 received simultaneous TKA, while the other 7 received staged TKA, with the mean interval between the two surgeries of 8 months. Healing of the incisions by rst intention was achieved in all patients.

Complications
Re-examination with X-ray showed no complications such as the collapse of the tibial plateau, loosening of the prosthesis, and patellar necrosis in the patients (Fig. 1). Common peroneal nerve paralysis appeared after the operation in one patient, for whom nerve nutrition and acupuncture were conducted, and the sensorimotor functions completely recovered 5 months later, but lateral dorsal numbness was still found. Another patient was with periarticular infection at 1 year after the operation. The infection was controlled by a 6-week anti-infectious therapy, and no revision arthroplasty was conducted. Residual genu valgum of 5°-9° was found in three patients, but knee joint functions were good.

Discussion
Patients with RA often have severe osteoporosis [5,6], which may seriously affect knee functions [7]. TKA might be required [8,9], but studies have yet comprehensively reported the effectiveness of TKA in RA patients with severe osteoporosis and genu valgum. Therefore, the aim of the present study was to assess the clinical bene ts of TKA in patients with severe osteoporosis and moderate genu valgum at a single center. The results suggest that TKA via the medial parapatellar approach with proper osteotomy and soft tissue balancing could restore the knee function in patients with RA, severe osteoporosis, and genu valgum.
Various criteria are available for the classi cation of genu valgum, such as the Ellison criteria [12], Krackow criteria [16] Ranawat criteria [14], and SOO criteria [17]. Previous studies have reported that patients suitable for TKA are generally Ellison type I genu valgum. In this study, the Ellison classi cation of genu valgum in all the patients was type II, and lateral soft tissue release, medial soft tissue tightening, and application of post-stable prosthesis were conducted. All the patients were with high knee joint stability after the operation.
The operations in all patients were through the medial parapatellar approach. The medial parapatellar articular capsule approach could facilitate the operation procedures and exposure of the operating eld. The complications of operations through the medial parapatellar approach are mainly postoperative joint instability and prosthesis loosening [18][19][20], while the operations through the lateral parapatellar approach are generally accompanied by incision complications [21][22][23][24]. All the patients in this study underwent operations through the medial parapatellar articular capsule approach, and a step-wise release was conducted and achieved good soft-tissue balance.
The mean postoperative FTA of the patients was 171.2°±6.2°, and some patients were still with residual genu valgum. In a previous study about TKA outcomes for patients with genu valgum conducted by Nakano et al. [25], the medial parapatellar approach was used for the treatment of 24 knee joints, of which the preoperative and postoperative FTA of the patients was 172.4°±2.7°. Residual genu valgum was found in some patients, supporting the present study. Femoral valgus resection could be one of the factors leading to residual valgus alignment, and conducting a 3°-5° femoral valgus resection could be performed to prevent insu cient correction [26][27][28]. Therefore, it is applicable to adjust the angle of femoral osteotomy to improve the positioning and alignment of the graft.
The highest point is generally used as the reference point for the measurement in the osteotomy. Bone defects are very common at the lateral tibial plateau in patients with severe genu valgum, which could be repaired by autogenous bone grafting or bone cement lling [29]. In the present study, one patient was with severe lateral tibial defects. We speculated that if the highest lateral point were used as the reference, the osteotomy amount at the medial side would be too important. Therefore, the highest medial point was selected as the reference for osteotomy, and the severe lateral defects were lled with bone cement. After the osteotomy at the distal femur and tibial plateau was completed, the exion-extension gap of the knee joint was assessed, and the lower extremity alignment was calculated to assess the correction of the genu valgum, based on which the thickness of the polyethylene liner was decided, and a second osteotomy could be conducted if necessary. The patella was not replaced routinely, while the surrounding osteophyte was resected, the articular surface was reconditioned to be smooth, and the nerves surrounding the patella were blocked.
In cases of genu valgum deformity, the lateral tissues are generally with different degrees of tension, and thus soft tissue release is among the most important procedures of TKA for genu valgum, leading to postoperative joint stability and the survival of the prosthesis. Nevertheless, soft tissue release has also been acknowledged as the most di cult part of this operation [30]. In the present study, the Ranawat technique [14] was adopted to release the lateral contracture structures step by step, and the exionextension gap balancing and varum-valgum stability were assessed after each step of release to prevent over-releasing. The popliteus tendon was preserved as possible, and the integrity of the lateral stable structures was maintained to prevent postoperative lateral instability. For the loosened medial structures, tight suturing was adopted to improve stability. By using these procedures, good soft tissue balance was obtained, and no joint instability was found after the operation. Bremer et al. [31] also adopted these techniques and avoided the use of semi-restrictive and restrictive prosthesis. Mullaji et al. [32] adopted similar techniques for the release of posterior cruciate ligament and iliotibial band, followed by computer navigated posterosuperior femoral osteotomy, which helped obtain more accurate positions.
For patients found with medial knee joint loosening, enhancing lateral releasing and thickness of padding appropriately, as well as using restrictive padding and tightening the medial ligaments, could help to obtain the medial and lateral balance, which then be followed by the application of post-stable prosthesis. The post-stable prosthesis could facilitate the intraoperative soft tissue release, provide higher internal stability to match the articular surface, and allow the maximum lateral shifting of femoral and tibial prostheses to improve the patellar tracking. Therefore, post-stable prosthesis is recommended for genu valgum patients using non-condyle restrictive prosthesis.
Severe osteoporosis is not an absolute contraindication of TKA, but perioperative treatment and preparation are very important. The frequency of osteoporosis is high in patients with RA [5,6], and TKA in such patients can be highly di cult [8,9]. First, improper position, inappropriate traction, and rough operating during TKA could induce bone fracture [8,9,33]. Therefore, all surgical procedures in osteoporotic patients must be very careful and gentle; the tools, including power saw and drill, should be used instead of a bone knife and bone chisel. The procedures of planting prosthesis, as well as exionextension of the knee joint, should also be careful to avoid intraoperative bone fracture [33]. For patients with underlying bone defects, impaction bone grafting is generally required to improve the xation of the prosthesis and reduce the amount of bone cement used, but the bone mass of osteoporotic patients is generally poor, and suitable bone graft is lacking. Therefore, allogeneic bone is frequently used in such patients.
An important aspect of TKA is guiding the patients to conduct postoperative functional training. In addition to routine postoperative function training, there are still some speci cities in RA patients. For instance, such patients are generally with advanced ages and long disease courses, and their physical condition is often poor. In addition, such patients have also reduced activities. Thus they are generally with different degrees of osteoporosis and muscle atrophy, which could lead to imbalanced strength of the muscles surrounding the knee joint [34,35]. Therefore, training of the quadriceps femoris muscle should be stressed in addition to the training of the exion-extension function of the knee joint.
The present study has limitations. Only 32 patients (44 knees) were included in this study. In addition, this was a single-center study. Furthermore, this study was a retrospective study with a follow-up of 1 year.

Conclusions
In conclusion, TKA via the medial parapatellar approach with proper osteotomy and soft tissue balancing could restore the knee function in patients with RA, severe osteoporosis, and genu valgum. Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Funding
Not applicable Authors' contributions TZ carried out the studies, participated in collecting data, and drafted the manuscript. TZ and NH performed the statistical analysis and participated in its design. YZ, RZ, ML and DZ participated in acquisition, analysis, or interpretation of data and drafted the manuscript. All authors read and approved the nal manuscript. Anteroposterior and lateral X-ray images of the right knee before the operation. (c) Anteroposterior and lateral X-ray images of the right knee immediately after operation.