Partial or total knee replacement for arthritis on just one side of the knee — which is the better choice?

When arthritis is confined to a single compartment of the knee, unicompartmental knee arthroplasty (UKA, also called partial knee replacement) is generally the preferred first-line surgical option for appropriately selected patients. That said, the better choice genuinely depends on individual factors, and understanding the clinical reasoning behind each option is essential to making a sound decision.

The case for UKA rests on several well-established advantages. Because only the diseased compartment is resurfaced, healthy bone and soft tissue are preserved. The cruciate ligaments remain intact, which contributes to more natural proprioception and a knee that many patients describe as feeling closer to their own. Recovery is meaningfully faster — most patients return to activity within 4 to 6 weeks, compared to 8 to 12 weeks following total knee replacement (TKA). Surgical risk is also lower in terms of blood loss and hospital stay. And importantly, a UKA can be converted to a TKA if revision becomes necessary later.

The ideal UKA candidate is generally age 55 or older, with a BMI below 35, isolated medial or lateral compartment arthritis confirmed on imaging, an intact ACL, adequate range of motion, and no inflammatory arthritis.

Total knee replacement becomes the more appropriate choice under a different set of circumstances. Even subtle disease in compartments beyond the primary one predicts UKA failure, so multi-compartment involvement favors TKA. Inflammatory arthritis — rheumatoid arthritis and similar systemic conditions — affects the joint globally and is not well served by a partial procedure. ACL deficiency is a contraindication to UKA, since the implant depends on ligamentous stability. Severe coronal deformity greater than 15 degrees of varus or valgus exceeds what a partial replacement can correct. Higher BMI and high-demand mechanical loading place greater stress on the implant, which favors the durability profile of TKA. For patients who prioritize a single definitive surgery, TKA carries a well-documented 15 to 20 year longevity track record.

One important nuance deserves emphasis. Studies consistently show that UKA carries a higher 10-year revision rate — approximately 15 to 20 percent — compared to roughly 5 to 8 percent for TKA. At the same time, properly selected UKA patients report satisfaction at least comparable to TKA patients, with less awareness of the implant in daily life. The revision risk is real but manageable, and it must be weighed against the functional and recovery advantages that UKA offers in the right candidate.

From a workup standpoint, the decision begins with an orthopedic consultation that includes weight-bearing radiographs to map compartment involvement definitively. MRI adds important information about ACL integrity and the true extent of cartilage loss, confirming or excluding UKA candidacy. Pre-surgical physiotherapy — prehabilitation — in the 6 to 8 weeks before the procedure is supported by strong evidence for improving postoperative outcomes. For patients with a BMI above 35, weight reduction before surgery reduces revision risk for both procedures. A structured postoperative rehabilitation program should begin immediately after surgery, with return to function expected within 4 to 12 weeks depending on the procedure performed.

Surgeon experience is a factor that warrants specific attention. UKA outcomes are meaningfully volume-dependent, and surgeons performing more than 50 UKAs annually demonstrate substantially better results than lower-volume practitioners. Thorough preoperative imaging and rigorous patient selection are the most critical determinants of a successful outcome with this procedure.

The knee is not a simple hinge. It functions as a tricompartmental joint system — medial tibiofemoral, lateral tibiofemoral, and patellofemoral — each contributing distinct arthrokinematic roles during weight-bearing and locomotion. When arthritis is confined to a single compartment, most commonly the medial, the problem is a focal biomechanical failure rather than a global joint failure. That distinction is the entire foundation of the surgical decision, and it determines which movement patterns can realistically be restored, how the kinetic chain adapts, and what the long-term functional ceiling looks like.

When one compartment degenerates, several critical movement dysfunctions emerge. Normal knee mechanics require a precise roll-and-glide relationship between the femoral condyle and tibial plateau. In medial compartment arthritis, the medial femoral condyle loses its cartilage buffer, disrupting the coupled internal tibial rotation that should occur during terminal knee extension — the screw-home mechanism. This alone compromises full extension locking and gait efficiency. Medial compartment loading increases dramatically, often creating a progressive varus deformity that shifts the mechanical axis medially, overloads the medial structures further, and establishes a self-perpetuating cycle of degeneration. The body does not tolerate asymmetric joint pain quietly: ipsilateral hip abductor weakness develops from altered loading strategies, the contralateral hip and lumbar spine absorb excess load from antalgic gait, the ankle and subtalar joint compensate with pronation on the affected side, and arthrogenic muscle inhibition suppresses quadriceps activation, reducing dynamic joint protection.

A unicompartmental knee arthroplasty resurfaces only the damaged compartment, preserving the cruciate ligaments, the healthy compartments, and the native joint proprioception network. From a movement science perspective this is profoundly significant. The ACL and PCL are preserved, maintaining the arthrokinematic roll-glide coupling that total knee replacement eliminates or substitutes mechanically. Proprioceptive afferent pathways through the preserved ligaments and menisci remain intact, supporting faster neuromuscular re-education. The natural joint line is maintained, preserving more physiologic patellofemoral mechanics. Bone stock preservation is superior, which matters critically if revision surgery is ever needed. Patients typically achieve faster recovery, greater flexion range, and a more natural-feeling knee postoperatively, and Research suggests unicompartmental knee arthroplasty patients recover early function faster and report less awareness of the implant, though it does not show a guaranteed advantage in higher-level activity.

Total knee replacement is not the wrong choice — it is the right choice for the right patient. The biomechanical indications for choosing it over unicompartmental arthroplasty include bicompartmental or tricompartmental involvement, because preserving compartments that are already significantly degenerated offers no advantage. Significant varus or valgus deformity, generally greater than 10 to 15 degrees, is another indication, as unicompartmental arthroplasty cannot reliably correct substantial coronal plane malalignment. ACL deficiency is a contraindication for most unicompartmental designs because the procedure depends on intact cruciate function for stability. Inflammatory arthritis — rheumatoid, psoriatic — is a systemic joint disease that will progress to involve other compartments. Higher BMI increases unicompartmental failure rates, though the precise threshold remains debated in current literature. Symptomatic patellofemoral arthritis will not be addressed by a unicompartmental procedure.

Rehabilitation after either procedure begins in the acute phase, covering roughly the first 14 days, with the shared priorities of edema control, early range of motion, and quadriceps activation. Quad sets — isometric quadriceps contraction with the knee extended, held 5 seconds — are performed 3 sets of 20 repetitions every 2 hours while awake. Heel slides for range of motion are performed 3 sets of 15 repetitions twice daily in supine, sliding the heel toward the buttock and holding end range for 3 seconds, with a target of 0 to 90 degrees by day 7. Ankle pumps, 20 repetitions every hour, are critical for DVT prevention and calf pump activation. Straight leg raises, 3 sets of 10 twice daily, are introduced only when the quad set is pain-free and no extensor lag is present. Patellar mobilizations — superior-inferior and medial-lateral glides, 30 seconds each direction, three times daily — are essential for preventing infrapatellar contracture and are particularly emphasized after total knee replacement due to the larger surgical exposure. Unicompartmental patients typically achieve 0 to 90 degrees of range of motion faster and with less pain than total knee replacement patients.

The functional restoration phase, spanning roughly weeks 3 through 8, targets symmetric gait, 0 to 120 degrees of range of motion, and elimination of the antalgic pattern. Terminal knee extensions with a resistance band — 3 sets of 15 daily, standing with the band behind the knee, emphasizing full extension lock — re-educate the screw-home mechanism. Mini squats through 0 to 45 degrees, 3 sets of 15 daily, begin bilaterally and progress toward single-leg emphasis, with attention to knee tracking over the second toe. Step-ups beginning at 4 inches and progressing to 8 inches, 3 sets of 12 each leg daily, prioritize eccentric control on descent. Hip abductor strengthening through clamshells and side-lying abduction, 3 sets of 20 daily, is critical for restoring frontal plane stability and reducing medial compartment loading. Stationary cycling begins at 15 minutes and progresses to 30 minutes daily at low resistance, with seat height adjusted to allow comfortable range of motion; it is an excellent modality for arthrokinematic restoration. Balance and proprioception training begins with single-leg stance on a firm surface for 3 sets of 30 seconds and progresses to a foam surface; unicompartmental patients will advance faster here because of preserved ligamentous proprioception.

The strength and kinetic chain integration phase, covering weeks 8 through 16, targets a limb symmetry index greater than 80 percent and return to community ambulation without an assistive device. Leg press progresses from bilateral to unilateral, 3 sets of 12 with advancing load, through 0 to 70 degrees of range while avoiding deep flexion loading early. Romanian deadlifts, 3 sets of 10, integrate the posterior chain and re-educate the hip hinge pattern. Lateral band walks, 3 sets of 20 steps each direction, activate the gluteus medius and address frontal plane kinetic chain demands. Gait retraining on a treadmill focuses on symmetric step length, heel-toe pattern, and elimination of the Trendelenburg sign. Perturbation training — therapist-applied manual perturbations during single-leg stance — provides an advanced proprioceptive challenge and is particularly valuable for unicompartmental patients returning to sport.

Progression should not be driven by calendar alone. Advancement from the acute to the functional restoration phase requires 0 to 90 degrees of range of motion, no extensor lag on straight leg raise, and ambulation with a single crutch or cane with minimal antalgic pattern. Advancement to the strength and kinetic chain phase requires 0 to 120 degrees of range of motion, symmetric gait without an assistive device, single-leg stance greater than 20 seconds on a firm surface, and an 8-inch step-up with controlled eccentric descent. Return to higher activity requires a limb symmetry index greater than 85 percent on single-leg hop testing, full range of motion of 0 to 130 degrees or greater, and no joint effusion after activity.

When arthritis is genuinely isolated to one compartment, the ACL is intact, the deformity is mild to moderate, and the patient is appropriate by weight and activity profile, unicompartmental knee replacement is biomechanically superior. It preserves the arthrokinematic architecture that makes the knee feel like a knee. The preserved cruciate ligaments maintain roll-glide coupling. The intact proprioceptive network accelerates neuromuscular recovery. The smaller surgical footprint means less quadriceps inhibition, faster range of motion recovery, and a higher functional ceiling.

Total knee replacement remains the gold standard when disease is widespread, deformity is severe, or cruciate integrity is compromised, and it produces excellent outcomes in those appropriate cases.

The critical step is precise imaging and surgical assessment — specifically, weight-bearing full-length alignment films and arthroscopic or MRI evaluation of all three compartments — to confirm that the disease is truly unicompartmental before committing to unicompartmental arthroplasty. A unicompartmental implant placed in a knee with unrecognized lateral compartment disease will fail prematurely. The surgical outcome is only as good as the movement restoration that follows it, which makes early integration of a movement specialist into the rehabilitation process essential from day one.

When arthritis is isolated to one compartment of the knee — most commonly the medial, or inner, side — there is a genuine choice between unicompartmental knee arthroplasty (UKA), the partial replacement, and total knee arthroplasty (TKA). This is one of the most debated decisions in orthopedic surgery, and the answer is genuinely nuanced. The surgical choice profoundly shapes everything that follows in recovery, so it is worth examining the evidence carefully before turning to rehabilitation.

The case for UKA rests on several well-established advantages. It preserves the cruciate ligaments — both the ACL and PCL — which represents a substantial neuromuscular advantage discussed in detail below. It also preserves bone stock in the unaffected compartments. In appropriately selected patients, UKA tends to produce faster early functional recovery than TKA, though the evidence does not set a fixed timetable for either procedure. Preserving the cruciate ligaments keeps knee kinematics closer to the native joint, which may contribute to the lower implant awareness patients report. Blood loss is lower, hospital stay is shorter, and infection risk is reduced. Studies show higher patient satisfaction scores when patients are properly selected, and UKA is easier to convert to TKA later if needed.

The case against UKA — and for TKA — is also real. Some registries show a higher revision rate at 10 to 15 years compared with TKA, though this gap is narrowing with modern implants. UKA requires strict patient selection; if arthritis is progressing to other compartments, the partial replacement will fail sooner. It is technically more demanding, and outcomes are highly surgeon-volume dependent. UKA is not appropriate when the ACL is deficient, when significant deformity exceeding 10 degrees of varus or valgus is present, or when inflammatory arthritis is the underlying diagnosis.

For patients who are good candidates — isolated single-compartment disease confirmed on weight-bearing X-rays, an intact ACL, correctable deformity, BMI ideally under 35, and access to a surgeon with high UKA volume — UKA is the superior functional choice for most active individuals. The preservation of proprioceptive feedback through the cruciate ligaments alone makes neuromuscular rehabilitation dramatically more successful.

Arthrogenic muscle inhibition (AMI) is the central neuromuscular problem in both knee arthritis and post-surgical recovery. The mechanism is this: joint effusion and pain activate mechanoreceptors in the joint capsule, which reflexively inhibit the quadriceps via the spinal cord — specifically the vastus medialis oblique (VMO). This is not weakness from disuse alone; it is a neurological shutdown that persists even after pain resolves. With UKA, the preserved cruciate ligaments maintain their mechanoreceptor populations, which means proprioceptive feedback loops remain partially intact, AMI resolves faster post-operatively, neuromuscular re-education is more efficient, and patients regain single-leg stability significantly sooner. With TKA, cruciate sacrifice in posterior-stabilized designs eliminates a major proprioceptive input. The brain must essentially remap knee joint position sense from scratch — a process that takes 6 to 12 months and is never fully complete in older patients.

The muscles most affected in either case follow a predictable pattern. The VMO is the primary target of AMI. The gluteus medius undergoes compensatory inhibition from altered gait mechanics. The gastrocnemius and soleus are often overlooked but are critical for terminal knee extension and push-off power. The hip abductors are weakened from antalgic gait patterns that precede surgery.

In the early phase — weeks 1 through 6 post-operatively — the goal is breaking AMI and restoring neuromuscular recruitment, not building strength through load. Volume and neural activation matter more than resistance at this stage.

For quadriceps neuromuscular re-education, quad sets with biofeedback are the foundation. Lying supine with a rolled towel under the knee at 15 to 20 degrees of flexion, the patient contracts the quadriceps and holds for 5 seconds, focusing on feeling the VMO fire just above and medial to the kneecap. The prescription is 3 sets of 20 repetitions, 3 times daily. Low-load isometric contraction at this angle maximally recruits the VMO while minimizing joint stress. Straight leg raises follow: lying supine, the patient braces the quad to lock the knee straight, raises the leg to 45 degrees, holds 2 seconds, and lowers slowly over 3 seconds, performing 3 sets of 15 twice daily. Progression occurs when the patient can complete the movement without a quad lag — the knee bending as the leg is lifted. Terminal knee extensions (TKE) with a resistance band are arguably the single most important early exercise. Standing with a light band looped behind the knee, the patient starts at 30 degrees of flexion and extends to full extension against the band, performing 3 sets of 20 twice daily. This trains the VMO in its functional range and directly counters AMI.

Hip and glute activation is critical and frequently neglected. Clamshells — side-lying with knees bent 45 degrees and feet together, rotating the top knee open — target the gluteus medius to prevent a Trendelenburg gait pattern, and are performed for 3 sets of 20 on each side daily. Supine bridges, with feet flat and hip-width apart, driving through the heels to lift the hips and holding 2 seconds at the top, are performed for 3 sets of 15 daily, progressing to a single-leg bridge when the bilateral version is pain-free.

Once AMI is resolving and the patient can perform TKEs and straight leg raises without compensation, progressive loading begins in the phase spanning weeks 6 through 16. The baseline load for any loaded exercise starts at bodyweight or 50 percent of the estimated one-repetition maximum. Load increases by 10 percent per week, but only if morning swelling is stable — defined as no increase greater than 5 mm from baseline measured at the joint line. If next-day swelling increases measurably, load is reduced by 50 percent and held for 3 days before attempting progression again. Key exercises in this phase include leg press progressing from bilateral to single-leg, starting at 60 degrees of range and advancing to 90 degrees as tolerated; step-ups progressing from a 4-inch to 8-inch to 12-inch box at 3 sets of 12 three times weekly; partial squats progressing to full depth at 3 sets of 10 to 15 three times weekly; and stationary cycling beginning at low resistance for 20 minutes and progressing to 40 minutes at moderate resistance.

Functional milestones serve as objective gates for progression — time alone does not determine readiness. Quadriceps strength must reach at least 80 percent of the opposite leg, measured by handheld dynamometry or isokinetic testing, before advancing to impact activities. The patient must demonstrate a single-leg squat to 60 degrees without valgus collapse — the knee caving inward — which tests integrated neuromuscular control. Gait symmetry in step length, cadence, and push-off power must be symmetric bilaterally, assessed with a timed 10-meter walk test. Stair descent without a handrail and without lateral trunk shift is required. A Timed Up and Go (TUG) test result of 12 seconds or less is the benchmark for community-level function. Return to sport or high-demand activity requires a limb symmetry index of at least 90 percent on hop testing, including the single-leg hop, triple hop, and crossover hop.

For isolated single-compartment knee arthritis, UKA offers superior neuromuscular outcomes in appropriate candidates — faster AMI resolution, better proprioception, more natural kinematics, and a rehabilitation trajectory that can produce substantial functional gains within the first three to four months, with continued improvement beyond that. Surgical candidacy and surgeon experience are paramount, however; a well-performed TKA will always outperform a poorly-indicated or poorly-executed UKA. Consultation with a surgeon who performs a high volume of both procedures — more than 50 UKAs per year — and who can provide an honest assessment of compartmental disease pattern on weight-bearing X-rays is strongly advisable. Functional goals, including return to normal activities and pain reduction, should be brought explicitly into that conversation. The rehabilitation science strongly favors preserving what works. If the anatomy allows it, preserve it.

In patients with isolated medial or lateral compartment osteoarthritis of the knee, unicompartmental knee replacement (UKA, also called partial knee replacement) produces superior patient-reported outcomes and faster functional recovery compared to total knee replacement (TKA) when patients are appropriately selected — but it carries a higher revision rate over time. Three studies inform this question, with the highest level of evidence being a Level 1 randomized controlled trial.

The TOPKAT randomized controlled trial (Beard D, Davies L, Cook J, et al., Lancet, 2019; PMID 31326135) compared TKA versus UKA in more than 500 patients with medial compartment osteoarthritis over 5 years. UKA patients reported superior knee function, quality of life, and faster return to activity. However, UKA carried higher revision rates — approximately 20% versus approximately 5% for TKA at 5 years. This trial represents the highest-quality evidence directly addressing this question in an isolated medial compartment population.

A systematic review and meta-analysis (Wilson H, Middleton R, Abram S, et al., BMJ, 2019; PMID 30792179) synthesized patient-relevant outcomes across UKA versus TKA studies and confirmed the pattern seen in TOPKAT. When selection criteria were met — specifically, age 55 to 75, BMI below 35, intact ACL, and isolated compartment disease — UKA patients demonstrated higher satisfaction scores, better proprioception, and faster functional recovery. TKA showed lower revision rates but higher rates of anterior knee pain and stiffness. Surgeon volume and experience emerged as the strongest predictor of UKA success. Surgeons performing fewer than 50 UKAs annually have significantly higher revision rates, making this a critical variable in patient decision-making that is often unmeasured.

A 2025 systematic review of early physiotherapy protocols following TKA (Aftab S, Ali H, Saeed B, et al., Musculoskeletal Care; PMID 40635148) adds an important and generalizable finding: structured early rehabilitation beginning within 24 to 48 hours post-operatively significantly improves pain, range of motion, and quality of life outcomes. Although focused on total replacement, the authors' findings apply to both UKA and TKA populations and underscore that post-operative rehabilitation intensity is a modifiable factor affecting success regardless of surgical choice.

Several evidence gaps and caveats deserve attention. The highest-quality evidence from PMID 31326135 and PMID 30792179 focuses specifically on medial compartment arthritis in patients aged 55 to 75 with BMI below 35 and an intact ACL. Outcomes may differ substantially for lateral compartment disease, younger or older patients, higher BMI, or inflammatory arthritis such as rheumatoid arthritis. The TOPKAT trial provides robust 5-year data, but 10- to 15-year durability comparisons remain limited, and UKA revision risk may continue to accumulate beyond the 5-year mark. Alignment of these findings with current American Academy of Orthopaedic Surgeons (AAOS) and American Orthopaedic Society for Sports Medicine (AOSSM) guidelines was not verified in this search and warrants cross-referencing.