Physiotherapy vs Surgery for Knee Osteoarthritis | Physio to Home

Should you have knee surgery or try physiotherapy first? The evidence might surprise you. North Tasmania's home physiotherapist compares the outcomes, the risks, and what the research actually says.

Should you have knee surgery or try physiotherapy first? The evidence might surprise you. North Tasmania's home physiotherapist compares the outcomes, the risks, and what the research actually says.

Micheal Ghattas

3/6/2026 · 7 min read

Physiotherapy vs Surgery for Knee Osteoarthritis: What Does the Evidence Actually Say?

By Michael Ghattas, DPT | AHPRA Registered Physiotherapist | 18 Years Experience

Physio to Home, North Tasmania | Last reviewed: March 2026

If you have knee osteoarthritis and your pain has been getting worse, you have probably considered surgery — or been told by someone that surgery is the next step. The question of whether to pursue physiotherapy or surgery for knee OA is one of the most common clinical decisions in musculoskeletal medicine, and the answer the research provides is considerably more nuanced than most people realise.

This article compares the two main surgical options — arthroscopy and total knee replacement — with evidence-based physiotherapy for knee OA, based on the current research literature.

First: There Are Two Very Different Surgical Options

"Knee surgery" for osteoarthritis is not a single intervention. It is important to distinguish between:

Knee arthroscopy — a minimally invasive procedure that "cleans out" the knee joint (washing out debris, trimming torn meniscus tissue, removing loose bodies). This is a widely performed procedure, but its role in osteoarthritis has been fundamentally questioned by high-quality evidence.

Total knee replacement (TKR) — a major reconstructive surgery replacing the articular surfaces of the knee with prosthetic components. This is appropriate for end-stage disease and produces excellent outcomes in correctly selected patients.

The evidence for these two procedures is very different, and conflating them leads to significant confusion in the clinical decision-making process.

The Arthroscopy Evidence: A Paradigm Shift

Knee arthroscopy for osteoarthritis is one of the most studied procedures in orthopaedic medicine — and the evidence against its routine use in OA is among the most consistent findings in the field.

The landmark Moseley trial published in the *New England Journal of Medicine* in 2002 randomised 180 patients with knee OA to arthroscopy, arthroscopic débridement, or sham surgery (skin incision only, no internal procedure). All three groups improved equally — the sham surgery group improved as much as either active surgical group. The authors concluded that the benefits of arthroscopy for OA were entirely placebo effect.

This finding has been replicated. A 2013 Finnish randomised controlled trial by Sihvonen and colleagues published in the *New England Journal of Medicine* compared arthroscopic partial meniscectomy to sham surgery in patients with a degenerative meniscal tear — one of the most common indications for arthroscopy in middle-aged and older adults. After one year, there were no significant differences between real and sham surgery in pain or function.

The 2017 British Medical Journal BMJ Open Access review of arthroscopy for knee OA concluded that arthroscopic surgery for degenerative knee conditions provides no clinically important benefit for long-term pain or function, and carries risks of complications, in patients with or without concomitant osteoarthritis.

Current status: NICE (UK), the American Academy of Orthopaedic Surgeons, and the Australian Commission on Safety and Quality in Health Care have all issued guidance recommending against routine arthroscopy for knee osteoarthritis. It remains indicated in specific acute scenarios (locked knee from a displaced meniscal tear, for example) but is not appropriate for the management of degenerative OA.

Physiotherapy vs Total Knee Replacement: A More Nuanced Picture

Total knee replacement is a very different proposition from arthroscopy. For patients with severe, end-stage knee OA — bone-on-bone joint space loss, severe constant pain limiting all meaningful activity, failure of conservative management — TKR is highly effective. It produces significant, durable improvements in pain and function in appropriately selected patients, and patient satisfaction rates are high.

The question is not whether TKR works. It is: for which patients is TKR the right choice, and at what point does surgery become appropriate?

The OARSI trial (Exercise Versus Total Knee Arthroplasty in Patients 45–70 Years Old, 2019): This landmark Danish trial randomised patients with knee OA aged 45–70 who were eligible for knee replacement to receive either physiotherapy first (12 weeks of supervised exercise) or proceed directly to TKR. At one year:

• 34 of 100 patients in the physiotherapy group subsequently elected to have TKR — but 66% avoided surgery entirely with physiotherapy alone, with equivalent improvements in pain and function at one year compared to the TKR group.
• Both groups improved. The TKR group improved faster, but the physiotherapy group — among those who did not proceed to surgery — achieved equivalent long-term outcomes.

The implication: A meaningful proportion of patients who are "ready" for knee replacement will do as well with supervised physiotherapy — at least over a 12-month timeframe. The trial does not suggest that all patients should avoid surgery, but it does support physiotherapy as the appropriate first step before committing to a major irreversible procedure.

Head-to-Head Comparison: What Matters

| Factor | Physiotherapy | Total Knee Replacement |

|---|---|---|

| Pain reduction | Significant, builds over 6–12 weeks | Significant, more immediate |

| Function improvement | Significant, evidence-based | Significant, well-documented |

| Risk | Very low — some post-exercise soreness | Surgical risk, DVT, infection, implant complications |

| Recovery time | None — programme fits around life | 6–12 weeks post-surgical |

| Reversibility | Fully reversible | Irreversible |

| Cost | Low–moderate | High (public wait or private cost) |

| Appropriate for | Mild–severe OA at all stages | End-stage OA after conservative management failure |

| Evidence quality | High (OARSI, Cochrane) | High for appropriate selection |

What Physio First Actually Involves

Physiotherapy for knee OA is not "just exercise." An evidence-based programme includes:

• A thorough assessment identifying specific muscular deficits, gait patterns, and functional limitations
• A progressive lower limb strengthening programme targeting quadriceps and hip abductors — the primary shock absorbers of the knee
• Gait retraining to reduce knee joint load during walking
• Manual therapy to improve joint mobility and reduce pain
• Education about load management, activity modification, and the evidence on exercise and OA
• Outcome measurement at regular intervals to track progress objectively

This is different from a generic home exercise printout. The 12-week supervised programme in the OARSI trial was structured, progressive, and physiotherapist-supervised — and it was this programme, not informal stretching, that produced the outcomes described.

When Surgery Is the Right Choice

Physiotherapy first is not right for everyone. Surgery — specifically TKR — is appropriate when:

• Conservative management including a genuine, committed physiotherapy programme has been completed without adequate relief
• Pain is constant, severe, and significantly limiting quality of life and sleep
• Imaging confirms end-stage joint space loss consistent with reported symptoms
• The patient is medically appropriate for surgery and willing to commit to post-surgical rehabilitation

If you are approaching this decision, a physiotherapy assessment can help clarify where you are in the disease trajectory, whether a structured programme is likely to help, and what realistic expectations look like — information that supports a better-informed conversation with your orthopaedic surgeon.

Frequently Asked Questions

My surgeon says I need a knee replacement. Should I get a second opinion?

A second opinion is always reasonable for elective surgery. It is also worth ensuring that you have completed a genuine trial of physiotherapy before making the decision — most orthopaedic guidelines recommend this. Asking your surgeon specifically about the OARSI trial and the evidence for physio-first approaches is a legitimate clinical conversation.

I've been doing exercises at home for months and my knee is no better. Does that mean I need surgery?

Not necessarily. Home exercise is not the same as supervised physiotherapy. Many patients who have "tried exercise" without adequate supervision, programme progression, or technique correction have not genuinely trialled physiotherapy. A physiotherapy assessment will determine whether there is a genuine evidence-based programme to work through before surgery becomes the appropriate next step.

Does physiotherapy work if my X-ray shows bone-on-bone?

The research — including the OARSI trial — included patients with significant radiological OA, some with severe joint space narrowing. Radiological severity does not reliably predict response to physiotherapy. The muscles surrounding the knee are not visible on X-ray, and their capacity to unload the joint through strengthening is not captured by imaging. A physiotherapy assessment addresses function, not just anatomy.

Book a Knee Assessment in North Tasmania

Whether you are trying to avoid surgery or preparing for it — physiotherapy is the appropriate first and essential step. Physio to Home provides AHPRA-registered knee OA assessment and rehabilitation in your home across North Tasmania.

Book your home assessment today →

About the Author

Michael Ghattas, DPT | AHPRA Registered Physiotherapist | 18 Years Experience

Founder, Physio to Home — North Tasmania's home physiotherapy service.

References

Moseley JB, et al. A controlled trial of arthroscopic surgery for osteoarthritis of the knee. *New England Journal of Medicine*, 2002.

Sihvonen R, et al. Arthroscopic partial meniscectomy versus sham surgery for a degenerative meniscal tear. *New England Journal of Medicine*, 2013.

Skou ST, et al. A randomized, controlled trial of total knee replacement. *New England Journal of Medicine*, 2015.

Fransen M, et al. Exercise for osteoarthritis of the knee. *Cochrane Database of Systematic Reviews*, 2015.

Brignardello-Petersen R, et al. Arthroscopic surgery for degenerative knee disease. *BMJ Open Access*, 2017.