Muscle, joint & pain

The Lysholm Knee Score After Ligament Injury

Save

A score out of 100 that a knee patient fills in themselves, first published in 1982 for ligament-surgery follow-up. What the original paper actually establishes is narrower than what gets quoted from it: it sets the range and the direction, and it sets no grading bands and no threshold for meaningful change. Here is what the Lysholm score measures, which version you are probably holding, and how to read your own number honestly.

Last updated: July 2026

Talk to a clinician

Gale can help you find a clinician in your state and request a visit.

Find care →

What is the Lysholm knee score?

The Lysholm Knee Score is a questionnaire you complete yourself about your own knee, and it produces a single number from 0 to 100. Higher is better: the scale runs upward toward less impairment. Jan Lysholm and Jan Gillquist published it in 1982, describing a scoring scale designed for follow-up after knee ligament surgery, with particular emphasis on the evaluation of symptoms of instability 1.

Two parts of that description carry more weight than they look. The first is the direction. On the Lysholm, up is good — a knee reporting 90 is doing better than a knee reporting 60. Plenty of knee and back questionnaires run the other way, so the direction is not something to assume.

The second is the purpose. The Lysholm was not built as a screening test, and it was not built to tell anyone whether a ligament is torn. It was built to follow people afterwards, putting the patient's own account of the knee into the record in a form that could be compared across visits.

The Lysholm is a follow-up instrument, not a diagnostic one. It records what your knee is doing, in your words, converted into a number.

What does the Lysholm score actually measure?

It measures symptoms — how the knee behaves when you use it — with the heaviest emphasis on instability. The 1982 paper operationalises instability as giving way during activity: the moment the knee shifts or buckles under load. In its original validation the scale tracked two things its authors cared about, the patient's own opinion of how the knee functioned, and whether clinical signs of instability were present or absent 1.

The authors tested it against a modified Larson scale in patients with instability and in patients with meniscus lesions. In knees that were genuinely unstable, the new scale returned a significantly lower total score than the comparator did 1. That is the whole point of building an instrument: one that stays flat when the thing you care about is present is not measuring that thing.

Giving way is the knee's own word for instability — a shift, a buckle, or the sense that the joint has moved when it should not have. It is a symptom, and only the person inside the knee can report it.

The paper's title says knee ligament surgery, not ACL surgery. A pcl injury — the ligament behind the knee — sits inside the scale's stated purpose as squarely as a torn ACL does 1. What the score does not do is read your imaging. It has no opinion about how a graft looks, and a graft that looks perfect on film can belong to a knee that gives way on the stairs.

Read the version and the direction before you read the number

A questionnaire score means nothing until you know which version produced it and which way it runs. The Lysholm's original scoring — 0 to 100, higher is better — is the one defined in Lysholm and Gillquist's 1982 paper 1. But instruments get revised, rescored, and republished, and the same name printed on two sheets does not guarantee the two sheets sum the same way.

The oxford knee score is the cautionary example, because the discrepancy is documented. Its 1998 development paper scores each of 12 items from 1 to 5 and sums them to a total of 12 to 60, where lower is better and 60 is the worst symptoms 5. If you have seen an Oxford Knee Score reported on a scale where higher is better, that is a different, later scoring of the same questionnaire — not the one the original paper defines 5. Same questions, inverted meaning. A number carried from one convention to the other without being converted says precisely the opposite of what it means.

Before comparing two scores, confirm they came from the same version of the same questionnaire, scored in the same direction. Otherwise the comparison is arithmetic on unlike things.

The practical form of this is a question worth putting to whoever hands you the sheet: which version is this, and what will you be comparing it against?

Where do the excellent, good, fair, and poor bands come from?

Not from the 1982 paper. The source that created the Lysholm defines the scale and its scoring, and it establishes no severity cutoffs and no grading bands at all 1. If you have seen a Lysholm score sorted into excellent, good, fair, or poor at particular thresholds, those thresholds were added later, by other researchers, and they are not part of the instrument as it was published.

This is not a reason to distrust the bands your clinic uses. It is a reason to understand what a band is. The scale produces a measurement. A band is a judgment laid on top of that measurement — a decision about how much knee trouble counts as good enough — and that decision was made by someone, in some population, for some purpose. The population a threshold was derived in is part of the threshold.

So the useful question is not whether 84 is a good score. It is: good by whose cutoff, drawn from which patients, and does that group describe me? A threshold built around young athletes returning to sport after ligament reconstruction is answering a different question from the one a fifty-year-old asks about walking to work.

The Lysholm gives you a number. Every verdict attached to that number was added afterwards by someone else, and is worth asking about by name.

How much does a Lysholm score have to change to mean something?

The 1982 paper does not answer that. It reports no minimal clinically important difference and no minimal detectable change, and it sets no threshold for meaningful change of any kind 1. That is not a defect — the paper predates the convention of reporting those statistics — but it does mean the instrument cannot, on its own, tell you whether a six-point rise is recovery or noise. Any change threshold you are quoted for the Lysholm comes from later validation work, not from the source.

The minimal detectable change is the smallest shift in a score that exceeds the questionnaire's own measurement error — below it, improvement cannot be distinguished from the ordinary wobble of answering on a different day. The minimal clinically important difference is the smallest shift a patient would actually notice as better. They are separate questions, and a change can clear one while failing the other.

For a sense of how large these figures run on a knee instrument, the ikdc form's development paper states its own directly. A true change in the IKDC Subjective Knee Form's score required 9.0 points 3. That is a minimal detectable change, it belongs to the IKDC, and it does not transfer to the Lysholm — a threshold borrowed from a different questionnaire is a guess wearing a decimal point.

The version of all this that helps at a follow-up visit is a direct question: when you re-score me, what size of change will you treat as real, and where does that figure come from?

Why the Lysholm arrives paired with an activity scale

Because a symptom score can be quietly flattered by a shrinking life. The Lysholm asks how the knee behaves; it does not ask what you are asking the knee to do. Tegner and Lysholm addressed that gap in 1985, developing an activity grading scale incorporating work and sport activities, built explicitly as a complement to functional scoring for evaluating disability after knee ligament injury 2.

The problem it solves is easy to state. A knee that never gives way because its owner quit five-a-side is not the same recovery as a knee that never gives way in a match. Both can report a high symptom score. Only one of them is a knee that got better.

A high symptom score reached by avoiding the activity that provokes symptoms is not recovery. The activity scale exists to catch exactly that.

Read together, the pair says something neither number can say alone: here is how much trouble the knee gives, and here is how much the knee is being asked to do.

How the Lysholm compares to the KOOS, IKDC, and Oxford Knee Score

They are not interchangeable, and they were not built for the same patients. The koos questionnaire is a self-administered measure for knee injury and osteoarthritis, and rather than producing one number it reports five separate subscales: pain, symptoms, daily living, sport and recreation, and knee-related quality of life 4. The ikdc knee score is knee-specific rather than disease-specific, covering symptoms, function, and sports activity across a range of knee problems 3. The oxford knee score was developed and validated for total knee replacement specifically 5.

InstrumentBuilt forShape of the answer
Lysholm (1982)Follow-up after knee ligament surgery, emphasising instability 1One score, 0-100, higher is better 1
KOOS (1998)Knee injury and osteoarthritis 4Five subscales, reported separately 4
IKDC Subjective (2001)Knee problems generally, across diagnoses 3One score; factor analysis found a single dominant component, which is what licenses combining the questions 3
Oxford Knee Score (1998)Total knee replacement 512 items, summed; original scoring 12-60, lower is better 5

The IKDC's paper is unusually candid about its own properties. It was validated by administering the final version alongside the SF-36 to 533 patients with a variety of knee problems, and reports internal consistency of 0.92, test-retest reliability of 0.95, correlations of 0.47 to 0.66 with physical-function measures against only 0.16 to 0.26 with emotional-function measures — evidence it tracks the knee rather than mood — and no meaningful difference in how its items behave by age, sex, or diagnosis 3.

None of this makes one instrument better. The KOOS validation splits the answer because in osteoarthritis, pain and function and quality of life come apart 4. The Oxford Knee Score validation was done in replacement patients, and using it outside that context exceeds what the paper established 5. The Lysholm concentrates on instability because instability is what ligament surgery is about 1.

How the score fits into your actual care

As one input among several, and never as the verdict. The APTA and JOSPT clinical practice guideline for knee ligament sprains frames rehabilitation around physical-therapy examination, progressive exercise, and criteria-based return to activity 6. Criteria-based means you advance because you meet a standard, not because a number of weeks has elapsed. A self-reported score is one of those standards. It was never meant to be the only one.

What the questionnaire adds is the part a clinic visit cannot see. Strength testing, hop testing, and an examination all happen in a room, on a good day, with someone watching. The score is about the knee at four in the afternoon on a Tuesday, on the stairs, when nobody is looking.

A score that has not moved is information, not failure. It tells the person planning your rehabilitation that something in the plan needs to change, which is what it is for.

The most useful comparison is almost always your score against your own earlier score, on the same version of the questionnaire — not against a published average from a study population you may have nothing in common with. Which is why answering it accurately matters more than answering it well. Clinicians read your score against what you told them last time, so a number rounded up to sound like a good patient does not make the knee better. It makes the next decision worse.

Common questions

There is no answer to that inside the instrument itself. The paper that created the Lysholm sets a 0 to 100 range where higher is better, and sets no grading bands or cutoffs at all. Any excellent-or-good verdict attached to 85 comes from later research in a particular group of patients. Worth asking your clinician which cutoffs they use and who those cutoffs were drawn from.

You do. It is a self-administered questionnaire, meaning the number reflects your account of your knee rather than a clinician's examination findings. That is deliberate. The original scale was built to track the patient's own opinion of how the knee functioned, alongside whether clinical signs of instability were present. Both matter, and they are gathered separately for a reason.

No. They are two different things by the same authors, meant to be read together. The Lysholm scores symptoms — how much trouble the knee gives you. Tegner and Lysholm's 1985 activity grading scale records work and sport activity, built as a complement to symptom scoring. One says how the knee feels; the other says how hard it is being asked to work.

It depends on the knee. The Lysholm was designed for follow-up after knee ligament surgery and concentrates on instability. The KOOS was built for knee injury and osteoarthritis and reports five separate subscales instead of one total, which matters when pain, daily function, and quality of life are not moving together. Different questions, so they are not substitutes for each other.

You can complete the questionnaire — it was designed for exactly that. Interpreting a single score in isolation is where it stops being useful. The number was built for comparison over time, against your own earlier answers on the same version, and it is read alongside examination and testing rather than instead of them. One score on one day is a data point, not a status.

It means you reported no symptoms on that questionnaire on that day. The Lysholm measures how the knee behaves, not what the tissue is doing. It cannot see a graft, and it does not know your imaging. A high score is genuinely good news about your experience of the knee, and it is not by itself a clearance to return to anything.

Related

Say it back

How would you explain this to someone you love?

Two or three sentences, just as you’d say it. Gale reflects back what you focused on — a mirror, not a quiz.

Talk to a clinician

Gale can help you find a clinician in your state and request a visit.

Find care →

When a knee needs looking at, not scoring

  • —A knee you cannot bear any weight on after an injury, or one that will not straighten because something inside the joint is physically blocking the motion
  • —A knee that is hot, visibly swollen, and painful alongside a fever or feeling generally unwell — this pattern can mean infection inside the joint
  • —New calf pain, swelling, or warmth in the leg after knee surgery or a stretch of immobility, particularly with breathlessness or chest pain
  • —Numbness, pins and needles, or a foot that looks pale or feels cold below an injured or recently operated knee

A hot, swollen knee with fever, a foot that turns pale or numb below the knee, or new calf swelling with breathlessness after knee surgery are emergency-department problems rather than questionnaire problems — call 911 or go to an emergency room the same day.

This article explains what a knee outcome questionnaire measures and how its scores are interpreted in research and clinical follow-up. It is education, not medical advice, and it cannot assess your knee. Scoring decisions, rehabilitation plans, and return-to-activity timing belong to you and the clinician who is examining you.

References

  1. 1.Lysholm J, Gillquist J. (1982). Evaluation of knee ligament surgery results with special emphasis on use of a scoring scale. The American Journal of Sports Medicine, 10(3):150-154. doi:10.1177/036354658201000306 ✓The origin, purpose, and construction of the Lysholm Knee Scoring Scale: designed for knee ligament surgery follow-up with emphasis on symptoms of instability, operationalised as giving way during activity; its 0-100 scoring with higher indicating better function; its original construct validation against a modified Larson scale in patients with instability and meniscus lesions, returning significantly lower total scores in unstable knees and tracking both the patient's own opinion of function and the presence or absence of clinical instability signs. Also cited for what the paper does NOT establish: no MCID, no MDC, and no severity cutoffs or grading bands.
  2. 2.Tegner Y, Lysholm J. (1985). Rating systems in the evaluation of knee ligament injuries. Clinical Orthopaedics and Related Research, September 1985, issue 198, pages 43-49. doi:10.1097/00003086-198509000-00007 ✓The existence and purpose of the Tegner activity grading scale: that Tegner and Lysholm developed an activity grading scale incorporating work and sport activities as a complement to symptom-based functional scoring, for evaluating disability after knee ligament injury. No responsiveness statistic, MCID, MDC, item count, or score range is attributed to this source.
  3. 3.Irrgang JJ, Anderson AF, Boland AL, Harner CD, Kurosaka M, Neyret P, Richmond JC, Shelborne KD (2001). Development and Validation of the International Knee Documentation Committee Subjective Knee Form. The American Journal of Sports Medicine, 29(5), 600–613. doi:10.1177/03635465010290051301 ✓The IKDC Subjective Knee Form as a knee-specific rather than disease-specific measure of symptoms, function, and sports activity; its validation alongside the SF-36 in 533 patients with a variety of knee problems; internal consistency 0.92 and test-retest reliability 0.95; construct validity correlations of 0.47-0.66 with physical-function measures versus 0.16-0.26 with emotional-function measures; a single dominant component on factor analysis justifying a combined score; absence of meaningful differential item function by age, sex, or diagnosis; and the minimal detectable change of 9.0 points, cited explicitly as an MDC and as belonging to the IKDC rather than the Lysholm. No item count or score range is attributed to this source.
  4. 4.Roos EM, Roos HP, Lohmander LS, Ekdahl C, Beynnon BD (1998). Knee Injury and Osteoarthritis Outcome Score (KOOS)—Development of a Self-Administered Outcome Measure. Journal of Orthopaedic & Sports Physical Therapy. doi:10.2519/jospt.1998.28.2.88The KOOS as a validated self-administered patient-reported outcome for knee injury and osteoarthritis, structured as five separate subscales — pain, symptoms, activities of daily living, sport and recreation, and knee-related quality of life — reported separately rather than as a single total.
  5. 5.Dawson J, Fitzpatrick R, Murray D, Carr A. (1998). Questionnaire on the perceptions of patients about total knee replacement. J Bone Joint Surg Br. 1998;80-B(1):63-69. doi:10.1302/0301-620X.80B1.0800063 ✓The Oxford Knee Score as a 12-item patient-completed questionnaire developed and validated specifically for total knee replacement, and its original scoring as defined in this paper — each item scored 1-5, summing to 12-60, with lower indicating better and 60 the worst symptoms. Cited as the article's worked example that a later, inverted rescoring of the same questionnaire is not the scoring this paper defines, and that use outside the arthroplasty population exceeds what the paper establishes. No MCID or MDC is attributed to this source.
  6. 6.Logerstedt DS, Scalzitti D, Risberg MA, et al. (2017). Knee Stability and Movement Coordination Impairments: Knee Ligament Sprain Revision 2017. Journal of Orthopaedic & Sports Physical Therapy. doi:10.2519/jospt.2017.0303The APTA/JOSPT clinical practice guideline framing of rehabilitation after knee ligament sprain around physical-therapy examination, progressive exercise, and criteria-based return to activity — establishing that a self-reported outcome score is one criterion among several rather than the sole determinant.

6 sources, numbered by first appearance. General health information, not medical advice. AI-assisted editorial content — citations link their sources. Editorial policy

← Back to the health library