Skip to main content
ACL Return to Sport

Updated August 17, 2026

Decision

ACL surgery versus rehabilitation alone

There is a randomized trial on this question, it ran for five years, and the result surprises most people. Here is what it found, what it did not find, and who it applies to.

In the United States the default assumption is that a torn ACL means an operation. For a 17-year-old soccer player that assumption is usually right. As a blanket rule it is not supported by the best available trial, and knowing that changes the conversation you have with your surgeon.

The KANON trial

A randomized trial enrolled 121 young, active adults with an acute ACL tear. One group had early reconstruction plus rehabilitation. The other had rehabilitation with the option of delayed reconstruction if they needed it.

At two years, the change in KOOS4 was 39.2 points with early reconstruction and 39.4 points with rehabilitation plus optional delayed surgery, a difference of 0.2 points (p=0.96). Of the 59 patients assigned to the rehabilitation-first arm, 23 had surgery and 36 did not. At five years, 30 of 59 (51%) in that arm had eventually had surgery, and the mean KOOS4 change was 42.9 points with early reconstruction against 44.9 with optional delayed reconstruction (p=0.54), with no significant differences in any KOOS subscale, SF-36, Tegner score or radiographic osteoarthritis.

What that result actually means

Not that surgery does not work. It means that a strategy of rehabilitation first, with surgery for those who need it, produced the same patient-reported outcomes at two and five years as operating on everyone straight away. About half the rehabilitation-first group ended up having surgery anyway.

The healing finding

A later MRI analysis of the same trial found evidence of ACL healing at two years in 16 of 54 patients (30%) randomised to optional delayed reconstruction, rising to 16 of 30 (53%) among those who never crossed over to surgery. The healed group reported better two-year KOOS Sport and Recreation and quality of life scores than the non-healed, the delayed surgery and the early surgery groups. That is a secondary analysis of one trial, not a treatment protocol, and it is the reason the question is being reopened at all.

Where surgery is the clear answer

  • You want to return to a cutting and pivoting sport at a competitive level.
  • Your knee gives way in daily life or during rehabilitation, which is the crossover trigger in practice.
  • There is a repairable meniscus tear that needs a stable knee to heal.
  • You are young with high sport demands, which is most of the American ACL population.
  • A locked knee or a mechanical block, which is a different urgency entirely.

Where rehabilitation first is a real option

  • Your sport and your job are straight-line: running, cycling, swimming, lifting, most recreational fitness.
  • Your knee is stable during controlled loading and does not give way.
  • You would rather find out how your knee behaves before deciding, knowing that roughly half of the trial's rehabilitation-first group did eventually operate.
  • You are older with lower pivoting demands.

The variable this changes

Choosing rehabilitation first does not mean doing nothing. The trial's rehabilitation arm was supervised rehabilitation, not rest. Deciding not to operate and then not rehabilitating is a third option that no trial supports.

  • What did the MRI show besides the ACL, and does anything else need repairing?
  • Is my knee currently unstable under load, or only on imaging?
  • If we rehabilitate first, what specifically would make us change course, and how long do we give it?
  • What does delaying do to my sport calendar, and what does it do to my meniscus?

Questions people actually ask

Can a torn ACL heal on its own?
An MRI analysis of the KANON trial found evidence of healing at two years in 30% of patients randomised to optional delayed reconstruction, and in 53% of those who never had surgery. That is one secondary analysis of one trial. It is enough to make the question legitimate and not enough to make non-operative management the default.
If I skip surgery, will I get arthritis?
At five years, the KANON trial found no significant difference in radiographic osteoarthritis between early reconstruction and optional delayed reconstruction. Five years is not a lifetime, and this is one trial in young active adults. Anyone who tells you surgery prevents arthritis is going beyond what that data supports.
Is it too late to operate if I waited a year?
Generally no. Delayed reconstruction is routine, and half of the KANON rehabilitation arm had surgery later. Timing matters most when a repairable meniscus tear or repeated giving way is doing damage in the meantime.
Does my insurance care which one I choose?
It cares in the sense that surgery requires prior authorization and rehabilitation is subject to visit limits. Neither route is free, and a conservative route that uses forty physical therapy visits can cost more out of pocket than you expect.