1 Introduction

Patellectomy, largely a historical operation, was traditionally performed as a salvage procedure for severely comminuted patella fractures, recurrent patellar dislocation or degenerative arthritis of the patellofemoral joint.^7,10,11 Patients with patellectomy have significant functional limitations with knee instability, pain, abnormal gait, difficulty ascending stairs, and loss of terminal extension.¹³ Poor clinical outcomes have been reported following total knee arthroplasty (TKA) for patients with previous patellectomy.¹

A number of techniques have been described to improve the outcomes of patients with patellectomy undergoing TKA with varying success. These include autograft bone reconstruction using bone graft sewn into the previous anatomical position of the patella with a subsynovial pouch for stabilisation,² allograft reconstruction of the patella using a whole patella-quadriceps tendon allograft,³ and the use of prosthetic trabecular metal patella implant.^8,9 Although some stability can be achieved using patella implant if residual bone is present, early loosening and failure was encountered when soft tissue was used for fixation to the implant.¹²

The aim of this study was to evaluate our consecutive series of patella tubeplasty technique in patients undergoing knee arthroplasty with previous patellectomy describing our surgical technique and clinical outcomes.

2 Methods

This was a retrospective consecutive series of 4 patients with patellectomy who underwent knee arthroplasty (2 primary TKA, 2 revision TKA) and concurrent extensor mechanism reconstruction with patella tubeplasty and a minimum 6 months follow-up. Demographic, surgical, and operative data were collected. The Knee Society Score (KSS) was used at final follow up as an outcome measure.⁶ Clinical and radiographic assessments were taken at 6 weeks, 3 months, and annually thereafter.

Patella Tubeplasty Operative Technique: under spinal anaesthetic with intravenous sedation and prophylactic dose of antibiotics as per local protocol, the patient is placed in the supine position with the knee prepped and draped in the usual fashion. Prior skin incision is marked and used prior to tourniquet inflation, if tourniquet is used, to assess skin vascularity. The wound is then deepened to the level of the extensor mechanism. Due to prior procedures in the knee and previous patellectomy, there is often significant adhesions between the skin and the extensor mechanism particularly proximally involving rectus femoris and vastus medialis obliquus (VMO). With careful dissection, a thick skin and subcutaneous tissue flaps are elevated. The distal end of the extensor mechanism is identified and both the medial and lateral margins of the patellar tendon elevated. Medial arthrotomy is then performed with mid-VMO split 8–10 cm proximal to the midpoint of the joint line to preserve its proximal blood supply and allow its mobilisation (Fig. 1-a). With previous patellectomy, the extensor mechanism is invariably thinned centrally and tends to sublux and track laterally with attenuated medial tissues. The lateral margin of the extensor mechanism is then identified and followed up to the lateral intermuscular spectrum bluntly dissecting vastus lateralis off the intermuscular septum (Fig. 1-b). Vastus lateralis is then separated from the iliotibial band all the way distally to the interval between patella tendon and iliotibial band (Fig. 1-b). This allows sufficient mobilisation of the extensor mechanism and exposure to the knee joint. The planned knee arthroplasty, primary or revision surgery, is then performed in the usual fashion and once completed the extensor mechanism is reconstructed with patella tubeplasty.

Close

Fig. 1 Intraoperative clinical photograph of right knee. (a) Following medial arthrotomy and vastus medialis obliquus release and advancement to the central portion of the extensor mechanism; (b) demonstration of the required lateral release undertaken by lifting the lateral edge of vastus lateralis off the lateral intermuscular septum proximally and distally dividing the lateral patella tendon off iliotibial band.

Export to PPT

The tubeplasty procedure is performed by folding the lateral half of the extensor mechanism tendon under the central half of the rectus portion of the extensor mechanism forming a semi-tube (Fig. 2-a). This results in thickening of the extensor mechanism tendon and central tracking range of flexion. The margin of the reconstruction is sutured using absorbable interrupted sutures such as Vicryl 1/0 and further reinforced using nonabsorbable strong suture such as Ti-Cron™ 2/0. The VMO is then advanced and brought across to the reconstructed (tubed) central portion of the extensor mechanism and held provisionally with towel clips to assess the position for the final suturing. The extensor mechanism reconstruction is then completed with advancement of the VMO into the position which had been tested. The VMO is sutured using nonabsorbable strong suture such as Ti-Cron™ 2/0 to the central portion of the extensor mechanism (Fig. 2-a).

Close

Fig. 2 Intraoperative clinical photograph to demonstrate extensor mechanism tubeplasty procedure (a) where the lateral extensor mechanism tendon is folded and sutured on itself and the VMO is then advanced and sutured to the central part of the patella tendon; (b) following closure of all interval and in knee flexion range with the extensor mechanism tracking centrally.

Export to PPT

The skin/subcutaneous tissue flap is brought across and the deep surface of the subcutaneous tissues sutured to the lateral margin of the extensor mechanism to reduce the dead space on the lateral side of the knee (Fig. 2-b). The remainder of the wound was then closed in layers in the usual fashion.

A knee immobiliser is applied postoperatively and the knee held in extension for 2 weeks allowing weight bearing as tolerated with assisted walking devices. Postoperative rehabilitation programme is then initiated with range of motion initially limited to 45° and minimal active quadriceps and hamstring exercise for the first 3–4 weeks followed by graduated rehabilitation protocol focusing on eliminating extensor lag.

3 Results

Four patients included, 2 males and 2 females with average age at time of surgery 65.5 years (range 58–76). All patients had multiple medical comorbidities, multiple previous knee surgeries (Table 1). The follow up ranged from 0.5 to 13 years. All 4 patients regained function and satisfactory clinical outcomes. Pre- and post-operative range of movement reported (Table 2) with average KSS score at final follow up 84.7 indicating “good” to “excellent” function (range 79–90). None of the patients reported any complications (Figs. 3–8).

Close

Fig. 3 (patient-1): Anteroposterior weight bearing, lateral and skyline radiographs of left knee with failing medial unicompartmental knee arthroplasty and previous patellectomy.

Export to PPT

Close

Fig. 4 (patient-1): Postoperative anteroposterior weight bearing and lateral radiographs at final follow up following revision knee arthroplasty and patella tubeplasty.

Export to PPT

Close

Fig 5 (patient-2): Anteroposterior weight bearing, lateral and skyline radiographs of right knee with advanced post-traumatic osteoarthritis, previous ipsilateral femur fracture, multiple knee surgeries including patellectomy.

Export to PPT

Close

Fig. 6 (patient-2): Post-operative anteroposterior weight bearing, lateral and skyline radiographs at 2.5 years follow up.

Export to PPT

Close

Fig. 7 (patient-3): Post-operative anteroposterior weight bearing, lateral and skyline radiographs of right knee following revision arthroplasty and patella tubeplasty at final follow up.

Export to PPT

Close

Fig. 8 (patient-4): Anteroposterior weight bearing, lateral and skyline radiographs of right knee with primary posterior stabilised components and patella tubeplasty procedure at 13 years follow up.

Export to PPT

4 Discussion

In 1976 Insall described an isolated proximal realignment for patellofemoral joint instability. He reported the results of his proximal “tube” realignment of the patella for chondromalacia patellae in young patients (48 knees).⁵ Since, a number of proximal soft-tissue procedures have been described for dynamic realignment with capsular plication procedures and quadricepsplasty designed to redirect the pull of the quadriceps with lateral release and VMO advancement. The technique described here has been developed by the senior author (JDG) for arthroplasty patients relying on those principles described by Insall to reconstruct the extensor mechanism to track more centrally. The tubeplasty procedure compensate for the thinned central portion of the extensor mechanism tendon.

To our knowledge, this technique has not been described in the context of knee arthroplasty for patients with previous patellectomy. Patients with patellectomy undergoing knee arthroplasty have traditionally had worse clinical outcomes compared to the population of knee arthroplasty patients. Haque et al. reported results of a historical cohort of 134 knees with previous patellectomy from the Mayo clinic between 1985 and 2010 with standard surgical technique and no patelloplasty. They reported higher risk of complications in post-patellectomy knees with instability, delayed healing, and infection. Although risk of revision was similar to non-patellectomy knees.⁴ Further, in a meta-analysis of seven small case-cohort studies largely undertaken in 1980s–90s of primary TKA following patellectomy, Asadollahi et al. reported that patients were less likely to have “excellent” or “good” outcome (OR: 0.3, 95% CI: 0.14 to 0.65). The weighted mean postoperative knee flexion arc was 6.58° less in patients with a previous patellectomy (95% CI: −12.79, −0.37). The risk of complication occurring in a patella-deficient knee was higher, with a pooled OR of 1.97 (95% CI = 1.10 to 3.51).¹

Busfield et al. described the use of a patellar tendon–patella–quadriceps tendon allograft in six primary and three revision arthroplasty cases with previous patellectomy. Infection was the predominant complication (33%) with one patient requiring amputation.³ Kwong et al. reported an 85% complication rate in their series of seven arthroplasties with trabecular metal patellar implants for patients with previous patellectomy.⁸ Our technique has the advantage of relying on normal native tissues with the main aim of improving the extensor mechanism tracking during functional range of motion. Further, it has the added benefit of improving the mechanics of extension, as the tube matures into a thick well-defined structure, increasing the lever arm of the quadriceps.

Our study is limited by the small number of patients included, although this is a rare complex cohort of patients and only similar small case-series studies have been published. Further, our follow up is short. However, our experience suggests that satisfactory clinical outcomes can be achieved with the described technique.

Ethical statement

This research study was conducted retrospectively from data obtained for clinical purposes.