1 Introduction

Unicondylar knee arthroplasty (UKA) is an established treatment option for symptomatic osteoarthritis isolated to one compartment. Advocates of UKA suggest preservation of bone stock, knee kinematics and accelerated postoperative recovery as reasons for its use.^1,2 Radiographic evidence of osteoarthritis such as joint space narrowing, osteophyte formation and malalignment of lower limb helps determine an indication for joint arthroplasty. There is little hesitation for diagnosis and treatment recommendations when symptoms, physical exam, and radiographs support isolated disease. Likewise, patients are not indicated for UKA when symptoms and signs suggest poly-compartment degeneration because of concerns of failure and conversion to total knee. A conflict arises when the other compartment pathology is not due to degenerative arthritis but instead due to synovitis, cartilage flaps, or mechanical meniscus tears that are responsive to arthroscopic procedures. Though total knee arthroplasty (TKA) is an excellent treatment option for degenerative knee arthritis, a more limited surgery, if effective, may be preferable. Additionally, and increasingly important to our healthcare system, UKA is less costly than TKA.^3–5

Arthroscopic debridement, lavage, partial meniscectomy, and chondroplasty for osteoarthritic knees have had unpredictable outcomes.^6–10 Simultaneous arthroscopic anterior cruciate ligament (ACL) reconstruction and UKA has been successful.^11–14 TKA shortly following arthroscopy has been associated with poorer outcomes and higher complications.¹⁵ There is little literature related to arthroscopic treatment of meniscal tears at the time of UKA.¹⁶ We hypothesize that the simultaneous UKA and arthroscopic treatment of symptomatic meniscus tears and synovitis is an effective treatment, and better than TKA for patients with isolated compartment radiographic disease and bicompartmental symptoms.

2 Material and methods

A retrospective review of our joint registry was carried out after obtaining Institutional Review Board approval. Inclusion criteria consisted of treatment with simultaneous ipsilateral arthroscopy and UKA. Ten patients treated with 11 such simultaneous procedures were identified between May 2004 and November 2013. Two patients were excluded due to lack of follow-up, yielding 8 patients and 9 simultaneous UKA-scope procedures for our case cohort. Our comparison group comprised 12 patients over the same time frame treated with TKA who presented with bicompartmental symptoms but only unicompartmental radiographic disease. One patient in UKA –scope group had the same procedure on both knees staged by one year. All were performed by two fellowship-trained arthroplasty surgeons at a single institution.

Patient demographic data including age, gender, body mass index (BMI), American Society of Anesthesiologist score (ASA), ambulatory status, and range of motion were collected for both groups. In addition, MRI findings were collected for the UKA-scope group. We examined the surgical intervention, operative time, length of stay, physical therapy outcomes, disposition, range of motion, average visual analogue (VAS) pain scores and complications, conversions, or revision surgery for each group. Patient reported outcomes on the KSFS, WOMAC, and SF-12 were analyzed preoperatively and postoperatively at 3 months and yearly thereafter.^17–19 The minimum follow-up was 2 years for both groups.

There were 8 patients and 9 knees in UKA-scope group and 12 patients in TKA group, with each group consisting of 66.7% and 75% females respectively. The mean age was 59.3 (range, 40–84) years in the UKA-scope group and 67 (range, 50–85) years in the TKA group, (p=0.10). Mean body mass index (BMI) was 31.6kg/m² and 33kg/m² respectively for UKA and TKA groups, respectively (p=0.97). Patient demographic characteristics for both groups are shown in Table 1.

In the UKA-scope group five operations were performed on the right and four on the left side. Seven were medial compartment and two were lateral compartment arthroplasties. In one medial arthroplasty, the patellofemoral joint was also replaced. The latter case was not excluded because it fit the thesis that the opposite tibio-femoral compartment pathology but amenable to arthroscopy rather than arthroplasty, and a total knee arthroplasty was avoided. All nine knees were diagnosed with osteoarthritis of one compartment and meniscus pathology in the other compartment. Two knees had previous arthroscopic intervention for meniscus tears. All eight patients had preoperative MRI films revealing meniscus tearing in the non-replacement tibio-femoral compartment (Table 2). Intraoperative arthroscopic findings showed grade 1 to 2 chondromalacia of lateral compartment in six patients, grade 3 chondromalacia of the patellofemoral (PF) in one knee and grade 4 chondromalacia of the PF joint in one knee in which the PF joint was replaced. Seven knees underwent arthroscopic lateral meniscectomy and two underwent medial meniscectomy. Preoperative outcome measure scores, 2 years scores and change in scores for KSFS (Table 3), SF-12 P (Table 4) and WOMAC (pain and stiffness) (Table 5) are shown in tables. There were no statistically significant differences between the baseline scores of the two groups using Student t-test.

A midline incision with medial parapatellar approach was used for TKA and medial UKA. Lateral para-patellar arthrotomy was used for lateral UKA. In UKA-scope group, arthroscopy was performed first without the use of tourniquet. All implants were cemented and utilized a fixed bearing polyethylene liner. The patella was resurfaced in all TKA cases. Rehabilitation for all patients consisted of supervised physical therapy for strengthening and range of motion, weight bearing as tolerated, and assistive gait aid as required.

Routine follow-up visits and assessments occurred at 2 weeks, 3 months, and yearly thereafter. Patient reported outcomes were recorded at each visit. Preoperative scores, postoperative scores, and change in scores were analyzed. Change in score represents the effect of the intervention as per recent studies.^20–23

3 Results

There were no significant differences between groups for age, American society of anesthesiologists (ASA) score, follow-up after surgery, BMI, or diagnosis (Table 1). The mean follow-up was 3.8 years (SD=2.9, range 0.25-4 years) for the UKA-scope group and 2.4 years (SD=2.5, range 0.25–6 years) for TKA group (p=0.262). There were no significant differences between the preoperative KSFS, WOMAC pain, WOMAC stiffness, WOMAC function, SF-12 mental, or SF-12 physical scores (p>0.128) (Tables 3–5). There were no significant differences between the preoperative range of motion between the two groups (p=0.175). Mean operative time was 131.6min (SD=17.9, range, 105–162) and 115.2min (SD=28.5, range 83–165) respectively for UKA-scope and TKA groups, p=0.019. The mean length of stay in the hospital was 3.1 (SD=0.6, range, 2.2–4.3) and 3.2days (SD=0.5, range, 2.2–4.4) for the UKA-scope and TKA groups respectively. In the UKA-scope group, 7 patients were discharged home, one was discharged to a rehabilitation center, and one was discharged to skilled nursing care facility (SNF). In the TKA group 6 patents went home and 6 patients were discharged to rehabilitation center.

KSFS score was significantly improved in the UKA-scope group at 2 years as compared to the TKA group (83.3 vs 45, p=0.023). SF-12 physical score was significantly improved at 1year and 2 years in the UKA-scope group compared to TKA group (47.2 vs 40.3, p=0.042; 48.3 vs 32.6, p=0.026). WOMAC pain, WOMAC stiffness, and WOMAC function scores were significantly improved in the UKA-scope group at 2- years as compared to TKA group ((98.7 vs 63.8, p=0.030), (90.1 vs 43.8, p=0.013), and (92.3 vs 55.2, p=0.027) respectively). There was no significant difference between the two groups in SF-12 mental component, p>0.076. Range of motion (ROM) at 3 months for the UKA-scope group was better than the TKA group, but it was not statistically significant (124.3 vs 113.8, p=0.058). Likewise, there was no statistical difference in ROM at 1-year and 2-years (p=0.206 and p=0.683 respectively).

At 2 years follow-up, the UKA-scope group had a better improvement in KSFS (28.3 vs −5, p=0.040), SF-12 physical score (13.6 vs 3.0, p=0.026), and WOMAC stiffness score (52.6 vs −6.3, p=0.024). Improvements in SF-12 mental and WOMAC function scores were not statistically significant at 2 years (13.6 vs 3.9, p=0.051), (57.0 vs 18.4, p=0.062) respectively. There was no significance difference in the change in score for WOMAC pain at 2 years (57.5 vs 28.8, p=0.108). Neither group had any complications. There were no infections, revisions, or reoperations in either group.

4 Discussion

Both UKA and TKA lead to improvement in functional outcomes.^25–29 TKA has been reported to be more durable than UKA.²⁴ Advocates of UKA cite better functional outcome over TKA.^30–32 However, patients with isolated compartment radiographic disease but bicompartmental symptoms may not be offered UKA. We found that when these patients have mechanical symptoms related to meniscal tear, a simultaneous arthroscopy and UKA is effective and may be better than TKA. It is unclear why our 12 TKA patients in the comparison group had overall bad outcomes and poorer functional scores than prior to surgery. It is possibly related to the early-grade preoperative osteoarthritis in the opposite compartment and known associations with pain and dissatisfaction.³³ It is more probably an aberrancy of our low number in the group, and no conclusions should be made from these results. Further study and larger cohorts are needed to understand whether bicompartmental symptoms and unicompartmental radiographic disease have poor outcomes after TKA.

There are several limitations to this study. First, this study focuses on a narrow constellation of symptoms, namely bicompartmental pain with unicompartmental radiographic disease. This limits the generalizability of our findings and also explains the small number of subjects in our groups. Further, data of the study spans a decade. This could be a confounding factor with respect to implants, technique, and operative indications but the same confounding variable applies to both groups. Additionally, capturing simultaneous arthroscopy and UKA patients is reliable within our database, but finding appropriate comparison subjects who had TKA was more challenging and required delving into the medical record to identify the specific constellation of symptoms. The TKA group was equally small, which could skew the functional outcome scores for that group. Nonetheless, this retrospective report shows that bicompartmental joint line tenderness need not necessarily exclude UKA as a treatment option. Finally, with our low numbers in each group, we are not powered to determine statistical conclusions of “no difference” between groups. Thus our results should be considered in light of this low n and only our positive statistical findings should contribute to our understanding of these patients.

Arthroscopy after UKA for meniscal pathology and retained cement has been reported with good results.^34–37 Arthroscopy has been used at the time of extra-articular osteotomy to evaluate cartilage and to refine the reconstructive plan.³⁸ Liu et al.¹⁶ reported good outcomes on 24 patients indicated for UKA who had arthroscopy at the beginning of surgery. UKA was aborted in 8 patients due to advanced disease in other compartments, all underwent synovectomy, 5 underwent meniscal intervention, and 11 underwent other arthroscopic procedures. Our study is consistent with these varying reports about the use of arthroscopy as an adjunct procedure. Further, we show that arthroscopy can help expand surgical treatment options to include UKA rather than just TKA. We had no aborted UKAs following arthroscopy, and this was likely due to information obtained from the preop MRI in these patients. We are unable to determine from our database a denominator of how many patients obtained an MRI that caused the surgeon to change from planned UKA-scope to TKA prior to the surgery date.

In conclusion, a viable option for patients who would be candidates for UKA if not for bicompartmental tenderness is to obtain MRI and consider arthroscopy at the time of UKA. In our experience, it is difficult to completely assess the other compartment from a UKA exposure. Therefore, MRI is a useful preoperative study and helps to guide treatment. Simultaneous arthroscopy and UKA is a reasonable treatment option to offer patients, as we found good results in 8 patients who underwent simultaneous UKA and arthroscopy.