Do minimally invasive approaches affect long-term clinical outcomes of total knee arthroplasty? A cohort analysis with a minimum 10-year follow-up.

To the Editor: Total knee arthroplasty (TKA) is a safe and effective treatment for patients with end-stage knee osteoarthritis, with the conventional medial parapatellar (MP) approach for TKA, good visualization of the knee can be obtained, helping to achieve correct osteotomy, proper ligament balance and good prosthesis position, which are all important for good long-term results.[1] To improve functional recovery and rehabilitation, some studies have adopted a relatively small incision operation to improve operation. Among them, two of the most commonly used procedures are named mini-medial parapatellar (MMP) and quadriceps-sparing (QS) approaches.[2] But other studies suggested that minimally invasive approaches may compromise the advantages of correct osteotomy, proper ligament balance, and good prosthesis position to achieve potentially rapid rehabilitation, potentially sacrificing the long-term clinical outcomes and survival of the prostheses.[3] At present, most of the studies compared different approaches that evaluated the follow-up of no more than 5 years. Thus, it is crucial to obtain the data of long-term clinical comparison. Based on the longitudinal cohort results of previous studies, a follow-up study for more than 10 years was conducted. The cohort study was approved by Peking University Third Hospital Medical Science Research Ethics Committee (No. IRB00006761-2011072) and was consistent with the principles of the Declaration of Helsinki. This study was registered online at the Chinese Clinical Trial Registry website (No. ChiCTR2000039929). Informed consent was obtained from all patients. The inclusion criteria were patients who underwent primary TKA by one senior surgeon between November 2001 and June 2008 in Peking Universtiy Third Hospital. The exclusion criteria were patients with inflammatory arthritis, patients with a history of previous knee surgery, patients with neurological diseases, and patients who received surgery with constrained TKA systems. There were 93 patients (110 TKAs) enrolled in the final study. Data related to patient demographics, including affected side, gender, age, height, weight, body mass index (BMI), preoperative lower limb mechanical axis, Knee Society Score (KSS), range of motion (ROM), and visual analog scale (VAS), were recorded. At the outpatient clinic, two independent observers performed all evaluations. All surgeries were performed by one senior surgeon who was experienced in MMP, QS, and MP TKAs. For all TKAs, the Nexgen Legacy Posterior Stabilized-flex Prosthesis (Zimmer, Warsaw, IN, USA) was implanted in all patients in the three study groups with equal distribution, and a tourniquet was used during the procedure. For the MP and MMP groups, the operation was carried out through a midline incision and MP arthrotomy. A quadriceps split approximately 4 cm above the superior pole of the patella was used in the MP group,[4] and a division of the quadriceps tendon extending 2 to 4 cm above the superior pole of the patella was used in the MMP group.[5] For the QS approach, a medial curved skin incision extending from the superior pole of the patella to the top of the tibial tubercle was made; the quadriceps tendon was divided no more than 2 cm within the vastus medialis obliquus muscle split if its insertion was attached to the medial edge of the patella.[6] During patellar resurfacing, the patella was dislocated and everted in the MP group and subluxed laterally with no eversion in the MMP or QS groups. The patella was removed at the same thickness as the components, and patellar resurfacing was performed in all patients with a cemented, three-pegged, all-polyethylene patellar component. The patellar tracking was examined by the no-thumb test. In each postoperative evaluation, the new KSS was used to assess the knee score, satisfaction score, expectation score, and functional activity score. The Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and VAS scores were also used as patient-reported outcomes. Furthermore, ROM was also evaluated. During the ROM measurements, a goniometer was centered on the lateral femoral condyle with one arm placed along the long axis of the femur pointing to the greater trochanter and the other arm placed along the long axis of the tibia pointing to the lateral malleolus; thus, the ROM was obtained. Meanwhile, the patellofemoral functions were evaluated and recorded. A standardized questionnaire including questions regarding the presence of anterior knee pain (AKP) when climbing stairs, rising from a chair, or exiting an automobile was also asked of all patients as a means of identifying symptoms related to the patellofemoral joint. In addition, a physician-rated Feller patellofemoral score (PFS) and several physical examinations, including a patellar glide test (medial and lateral translations), a grind test (isometric quadriceps contraction with the patella immobilized and the knee in extension, thus forcing the patella into the trochlear groove), and patellar edge tenderness (isometric quadriceps contraction with the patella immobilized and the knee in extension while pressing sites around the patella), of each patient were measured and recorded, as well as the measurement of thigh circumferences (15 cm above the proximal patella). In addition, complications and revisions were determined based on both the inpatient and outpatient medical records of each patient. Preoperative and postoperative long-standing films were obtained with subjects standing barefoot with the patella oriented forward according to Paley's criteria. Routine anteroposterior, lateral, and standard skyline views were obtained thereafter. All the following radiographic measurements were shown in Figure 1. The hip–knee–ankle angle (HKA) was defined as the angle formed by the intersection of a line from the center of the head of the femur to the knee center landmark and a second line from the center of the ankle talus to the knee center landmark. The lateral distal femoral angle (LDFA), the medial proximal tibial angle (MPTA), the femoral flexion angle, tibial slope angle, the lateral patellar tilt (LPT), and lateral patellar displacement (LPD) were evaluated. The patellar height was determined by the Insall–Salvati (IS) ratio. Aseptic loosening was indicated by the presence of radiolucent lines beyond 2 mm and gross shifting of components that caused subsidence or tilting. HKA, LPT, and LPD were measured before and after surgery. Evaluation of the radiographic assessments was performed using the Picture Archiving and Communication System. The accepted values used in our study for the normal lower limb mechanical axis were (1) 0° ± 3° for HKA; (2) 90° ± 3° for LDFA and MPTA; (3) < 10° for LPT; (4) ≥ −5 mm and <4 mm for LPD; and (5) ≥ 0.8 and ≤ 1.2 for IS ratio. Two orthopedic surgeons skilled in TKA who were blinded to the surgical method performed the measurements, which were recorded with an accuracy of 0.1° and 0.1 mm. To test the reliability of these measurements, all variables were measured twice at one-week intervals. For all measurements, the intraclass correlation coefficient of intraobserver reliability was >0.92, whereas the intraclass correlation coefficient of interobserver reliability was >0.86.Figure 1: Radiographic measurements for knee assessments. (A) The angle of HKA; (B) LDFA and MPTA; (C) FFA and TSA; (D) LPD; (E) LPT. FFA: Femoral flexion angle; HKA: Hip-knee-ankle; LDFA: Lateral distal femoral angle; LPD: Lateral patellar displacement; LPT: Lateral patellar tilt; MPTA: Medial proximal tibial angle; TSA: Tibial slope angle.The descriptive statistics were summarized as the mean and standard deviation. Data with normal distribution and homogeneity of variance would be analyzed using analysis of variance with a post hoc test (Tukey's method) comparing the continuous variances. Otherwise, non-parametric tests were used for inference using the Kruskal–Wallis test. The chi-squared test was used to compare the nominal variables, including gender, affected side, and outliers. The follow-up time was evaluated as a predisposing factor affecting the clinical outcomes using multiple regression analysis. A post hoc power analysis was conducted using G∗Power (3.1.9.4) to determine the power of the study (> 0.95). Statistical analysis was performed using SPSS software (version 26.0; SPSS, Inc, Chicago, IL, USA), and the normal distribution was assessed by the Kolmogorov–Smirnov test. P < 0.05 was considered statistically significant. Overall, 93 patients (110 TKAs) were included for analysis of the clinical results at the last follow-up, which consisted of 28 patients (32 TKAs) in the MMP group, 39 patients (47 TKAs) in the QS group, and 26 patients (31 TKAs) in the MP group. There were no statistically significant differences among the three groups in terms of the affected side, gender, age (at the time points of operation and follow-up), height, weight, BMI, or preoperative measurements, including HKA, LPT, LPD, KSS, ROM, or VAS scores. Clinical evaluations showed that there were no statistically significant differences among the three approaches for TKA. There were no differences in any section or in the total scores of the new KSS or WOMAC. The ROM, VAS scores, and thigh circumference also did not differ among the three groups. Regarding the patellofemoral functions, including AKP, PFS, and patellar physical examinations, no statistically significant differences were found based on the scores or records. In the MP group, one patient developed deep vein thrombosis after surgery, which was suspected based on the swelling of the calf and was confirmed by venous ultrasound. In the QS group, one patient had a deep infection three months after the operation, which was successfully treated after surgical debridement and oral antibiotic therapy. No other intraoperative or postoperative complications were observed during follow-up in the MMP group. For the revision, one patient in the QS group had a revision of patella resurfacing due to patellar component loosening three years after her primary TKA, and one patient in the MMP group underwent a total revised TKA owing to aseptic loosening after 11 years, which was found at the follow-up. No revisions were required in the MP group. No significant differences among the three groups were found regarding complications and reoperation rate (P > 0.05). As for postoperative radiographic results, no statistically significant differences were detected among the three groups in the values of angles or the outliers with regard to lower limb mechanical axis, femoral or tibial component positions in the coronal or sagittal planes, or in patellar position. Longevity is one of the concerns in TKA, and the survivorship of implants is dependent on the patient demographics, the surgical technique, and implant-related factors.[7] In this study, no differences were found in the value or in the outliers of the radiographic assessment angles, which indicated that the MMP and QS approaches did not increase the risk of malalignment and malposition of the components, and did not increase the reoperation rate of TKA. King et al[8] demonstrated that a substantial learning curve (15 procedures) might be required for surgeons before reaching steady results using the QS approach, so we recommend that surgeons gradually decrease quadriceps exposure for patients to gain advantages of the TKA procedure. Funding This research was funded by grants from the National Natural Science Foundation of China (No. 52003008) and the National Key Research and Development Program (No. 2017YFB1303001).