In a one-year duration, 196 knees of 150 patients at advanced stages of knee OA received arthroplasty (UKA: 80 knees of 66 patients, TKA: 116 knees of 84 patients) by the first author. In the same year, 16 knees of 16 patients with unknown causes of PPOP after knee arthroplasty (UKA: 2 knees, TKA: 14 knees) were referred from other hospitals for the AMR management. These patients were prospectively followed up as part of an Institutional Review Board Registry. The distribution of age, sex, the main compartment involved, and the type of arthroplasty performed on these knees are listed in Table 1. For the arthroplasty group, the inclusion criteria were advanced primary OA (stage IV or V according to Lyu's clinical classification [16]). UKA was performed for stage IV or V OA involving only one compartment. For stage IV or V OA involving more than one compartment, TKA was undertaken. All patients who received AMR were referred from other institutions with the diagnosis of PPOP of unknown causes after their knee arthroplasty and had been treated conservatively, including physical therapy and medication, for more than one year. Before AMR was performed, evidence of any clearly defined common causes of prosthetic failure including aseptic loosening, instability, progressive patellar arthropathy, infection and recurrent synovitis were ruled out by radiographic and laboratory examinations. Typical symptoms and signs of pain, crepitus, snapping, localized tenderness or palpable band described in previous report [14] confirmed the diagnosis of MAS before surgery.
Table 1
Age, sex and main involved compartment distribution in different type of surgery
| UKA | TKA | AMR |
| Age (SD) No. | F/M (Ratio) | Med./Lat. (Ratio) | Age (SD) No. | F/M (Ratio) | Med./Lat. (Ratio) | Age (SD) No. | F/M (Ratio) | TKA/UKA (Ratio) |
| 72.1 (7.8) 80 | 53/27 (2.0) | 80/0 - | 74.0 (6.6) 116 | 95/21 (4.5) | 104/12 (8.7) | 76.3 (5.7) 16 | 15/1 (15) | 14/2 (7) |
| SD: standard deviation; No.: number; F: female; M: male; Med.: medial compartment; Lat.: lateral compartment |
Surgical procedures for arthroplasty
For UKA, all cases were performed with the same cemented, metal backed fixed bearing implant (ZUK; Zimmer, Warsaw, IN, USA); for TKA, cemented posterior-stabilized implant (NexGen LPS-flex fixed knee system; Zimmer, Warsaw, IN, USA) was used in all cases. All arthroplasty were performed via a straight anterior incision with a medial parapatellar approach. For UKA, a tibia-first extension gap balancing technique was used. For TKA, the technique utilizing intramedullary femoral and extramedullary tibial alignment guides was followed. After the installation of prosthesis, the elimination of existing medial abrasion phenomenon was performed before wound closure. Various severity of pathologic medial plica described in previous literature [17] could be identified in the medial gutter. As shown in Fig. 1 and Additional file 1: Video 1, the thickened medial plica was removed completely from the attachment of genu articularis to the tendon sheath of pes anserinus.
Surgical procedure for arthroscopic medial release
During arthroscopic examination, remnants or fibrosis of medial plica could be identified over the inferio-medial aspect of the patellofemoral joint (PFJ) as shown in Fig. 2a. Tightness of the PFJ and impingment of the fibrotic medial plica were also verified (Fig. 2b). AMR was then performed as shown in Additional file 2: Video 2. The adequacy of the medial release was checked by pushing the tip of the scope under the patella and verifying if the previously tightly closed medial PFJ space could be easily opened and the medial retinaculum was clearly visible when the knee was fully extended (Fig. 2c).
Post-operative Management
Suction drain was used for all patients. Below-knee stockings to prevent thromboembolic disease for both lower limbs were used. Full range of motion and free ambulation were allowed as tolerated. After discharge from the hospital, home exercise programs, including active range of motion (knee hug and knee press) and quadriceps setting, were emphasised.
Follow-up and Evaluation of Outcomes
Regular follow-ups were undertaken monthly for 6 months. Thereafter, patients returned yearly for outcome evaluation. The comparisons of both pre- and post-operative Knee Society score (KSS) and knee injury and osteoarthritis outcome score (KOOS) were used for outcome evaluation. Subjective satisfaction was assessed by direct question using a categorical scale prepared for this study: excellent, free of symptoms, no limitation in activities; good, greatly improved, occasional pain, normal activities; fair, same as pre-operative condition, no improvement; and poor, has
received or considered further operative treatment. The outcome was regarded as satisfactory if subjective satisfaction was rated as “ excellent” or “ good”. The inquiry into subjective satisfaction and the evaluation of KSS and KOOS were conducted by nursing specialists. All investigations focused on individual knees in bilateral cases.
Statistical Evaluation
The data and normality of data distribution were validated, and non-parametric analysis was used in this study. All values were presented with means and standard
deviations. Statistical analysis for comparing preoperative and postoperative KSS and KOOS was performed using the paired t test. P < 0.05 was considered to be statistically significant. All statistical analysis was carried out using JMP, the Statistical Discovery Software (Version 5.0.1.2, SAS Institute Inc., Cary, NC, USA).