Knee stiffness after surgery: bending, straightening and what has stalled

Why is my knee stiff after surgery?

Three things are usually going on at once, and none of them starts out as scar tissue. The joint is swollen, and the fluid inside it has to go somewhere, so the knee runs out of room before it reaches the end of the bend. The tissues around the knee are healing, which means they are laying down new material and tightening while they do it. And the thigh muscle has been switched off, by the operation and by the swelling together, so even the movement you still have is hard to use.

A fourth thing then quietly makes it worse. A knee that hurts gets held still, and a knee held still stiffens, which is why the first bends of the morning feel like ground lost overnight and why twenty minutes of moving about usually hands it back.

That is the ordinary version, and it loosens as those four improve. The rest of this page is about telling it apart from the version that does not.

Three kinds of stiffness, and which one you have

One word covers three situations, and only one of them needs a phone call this week.

The first is a knee that feels set after you have been still and frees up once you are moving. That one is expected after an operation, and it settles.

The second is a range that is short of where your team wants it, while still creeping forward from week to week. That is usually not a problem, it is the work, and in the early months it is most of what rehabilitation after a knee operation actually consists of.

One question belongs with that second kind, though, and most people never ask it: what range does your team expect, and by when. Slow improvement only reassures you if it is on course for that number by that date. The timetables further down this page are counted in weeks and in months, not in years, so a knee creeping forward too slowly to get there is worth raising while it is still creeping, not once it has stopped. If nobody has given you a number and a date, ask for both at your next contact, and until you have them go by direction instead: the plateau rule further down this page works without a target.

The third is a range that has stopped changing, or one that is going backward. That is the one to raise with your physio, physical therapist (PT) or surgeon now rather than at your next routine appointment, and the warning lines at the end of this page give the wording.

Judging which you have takes a bit of discipline about measuring. Read your knee over weeks, not days: day to day it tells you about yesterday's walking and this morning's swelling and nothing else. Take your reading at the same time of day, after you have been up and about, or you will keep recording the worst version of your knee and conclude you are going backward. Writing one thing down once a week, the same task at the same time, is enough, and it is what makes the phone call useful if you end up making one.

The straightening, which is the half that gets dropped

Almost everything written about a stiff knee is about the bend. Patients ask about the bend, exercise sheets count the bend, and the number someone wrote down for you at your last appointment was probably the bend. Meanwhile the surgical literature goes the other way: older definitions of arthrofibrosis required a loss of straightening before the label applied at all, which left out people whose problem was pain and a lost bend, while the international consensus definition in use now counts restricted movement in flexion, extension or both (Usher and colleagues, 2019; Kalson and colleagues, 2016).

A few lost degrees sound like nothing, and the long-term figures say otherwise. In 502 people followed for a mean of 14 years after an ACL reconstruction, the factor most strongly linked to lower patient-reported scores was lack of normal knee extension, with normal taken from their own uninvolved knee and hyperextension counted as part of it. Losses of about 3 to 5 degrees were enough to show up, and they counted for more in the people who had also had part of a meniscus removed or had damage to the joint cartilage (Shelbourne and Gray, 2009). Those are results from one operation and one group of patients, so they are not a rule for every knee. They are a reason to stop treating the straightening as the optional exercise.

Stand with one knee slightly bent and hold it there: the thigh has to work the whole time. A knee that will not come flat asks for that on every step you take all day, so the leg tires early and the limp outlasts the pain.

You can see it for yourself at home, and the comparison that counts is with your own other knee, not with a number off a web page. One exception comes first. If your knee is held in a brace, a cast or an immobilizer, leave the measuring to your team and do not take the device off to compare.

Otherwise, lie on your back on a firm bed with both legs out straight and the thigh muscles soft, then prop yourself up on your elbows so you are looking down the length of both legs. Compare the two heels. On the stiff side the heel sits higher off the bed, because the knee is not reaching the surface. Then slide a flat hand under the back of each knee in turn, without pressing down, and compare how much room there is. Sitting with both heels up on a stool and letting both thighs go soft shows you the same difference from the side, and it is the easier one to set up on your own, so use whichever suits you; use the stool if lying flat on your back is uncomfortable or you have been told to avoid it.

A single reading means little. Do the same check, in the same place, at the same time of day, once a week, and write it down next to the bend. That is what turns a feeling into something your physio can act on, and it is also how you find out that a knee which looks straight to you is a few degrees short, which is the size of loss that showed up in those 14 year results.

The exercises for it are the dull ones, and they are dull in a specific way. The knee extension stretch and the seated heel prop ask you to let the leg hang and wait, building up the time rather than the force, and long sitting knee extension adds the thigh pressing the knee down toward the surface. The first two are the same exercise in two positions, on the floor and between two chairs, so do one of them and not both, whichever one you can get into and out of without help. The hold times on their pages are for one exercise, not a total for the pair. Nothing to count, and nothing visibly improving.

Quad sets and short arc quads are there to make the muscle hold the position once you have won it, because range you cannot hold does not stay.

The pillow is where a lot of that work comes undone. If you are raising the leg for swelling, support it from thigh to heel and leave the knee out straight, rather than putting the pillow under the knee itself. An evening with the knee propped half bent, or a night of it, pulls against everything the day's exercises did.

There is one exception to raising the leg. If your toes are numb, tingling or burning, or the bandage or brace feels tight, do not prop the leg higher than your hip and do not put ice on it, and read the warning signs below for what to do instead.

Scar tissue, and why forcing it is not the answer

"Breaking up scar tissue" is one of the most searched phrases on this subject and one of the most unhelpful, because it describes a knee as a lump of glue and forcing as the solvent.

A review of arthrofibrosis across joints puts the argument the other way round. Exercise itself sets off an inflammatory response; the authors state that aggressive exercise can start or worsen arthrofibrosis, and their own guidance is that intensity has to be matched to the inflammation it produces in that individual, with more inflammation and more fibrosis when the limits are passed (Usher and colleagues, 2019). That is a review arguing from mechanism, not a trial, and they say plainly that the risk factors and the best management are still not well understood. It is enough, though, to retire the idea that pain means progress.

So the test is the morning after, not the moment. Feeling the stretch at the end of the range is part of the exercise working, and an ache for a while afterward goes with it. If that has gone by the next morning, the session was fair. If pain builds with each repetition, or the knee is hotter and more swollen the following day, it asked too much, and the fix is to come back a step rather than abandon the work, because a knee that gets no range work at all stiffens faster than one that gets a little. The pain monitoring model is the clinician-facing version of that rule.

Three things to rule out of your week: weights hung from the shin, a relative pressing down on the knee, and any session you have to grit your teeth through. Your own good leg giving the stiff one a little help, the way the assisted seated knee bend in stage 2 does, is a different thing, because your own leg lets go the instant the knee objects and another person's hands do not. Tissue does get divided deliberately in one place, and that place is an operating room, under anesthetic, by a surgeon who is controlling where the force goes.

How to use this program

Your own instructions come first, every time. If you have been told how much weight to put through the leg, how far to bend the knee, or how long to wear a brace, those limits beat anything on this page, and they are usually tighter after a meniscus repair, a cartilage repair or a fracture than after a simple arthroscopy. If you are in the weeks or months after an operation and nobody gave you limits, or the leaflet has gone missing, do not guess and do not work them out from how the leg feels. Ring your surgical team or your physio and ask them two things: how much weight you may put through the leg, and how far the knee may bend.

If nobody picks up, the number on your discharge letter, the ward or day unit that sent you home, the hospital physiotherapy department and your own doctor are all worth trying, and say when you ring that you have no instructions at all, because that is what moves the call up a list. Until somebody answers, carry on with what you were shown and told to do after your operation and nothing beyond it. Every exercise here asks for your surgeon's or physio's go-ahead first, so stage 1 is where you wait rather than somewhere to start on your own.

Two checks before you start the program. If you have had a hip replacement or other hip surgery, get the go-ahead from your surgeon or physio first, because several of these exercises bend the hip past a right angle and one of them is done sitting on the floor, which means getting down there and back up again. And if your knee operation came with limits on weight or bending that nobody has lifted yet, those limits decide which stage you are on, not how the knee feels.

This page is for adults. A child or teenager who has had a knee operation should follow the plan from their surgeon and physio, and ask them before trying any exercise on this page. If you are pregnant, check with your midwife or physio before you start. Several of these exercises are done lying on your back.

Range work is done little and often rather than saved for one long session in the evening. Each exercise page carries its own starting dose as a guide, and your physio will adjust the numbers, the frequency and which exercises suit your operation. Read those numbers as a starting point for one exercise at a time rather than a total to add up. Each stage holds more exercises than most people need in one day, and if a session leaves the knee more swollen the next morning, the first thing to cut is how many of them you are doing, not the range on the ones you keep. The Aspetar guideline on rehabilitation after ACL reconstruction recommends starting active knee movement immediately after surgery while following any instructions from the surgeon, and reports that early movement can improve both bending and straightening in the early phase whatever the graft, while rating the certainty of that evidence as very low (Kotsifaki and colleagues, 2023).

The stages are not stacked on top of each other either. Stage 2 takes over the range work from stage 1 rather than being added to it, and stage 3 is standing work that goes alongside whatever range work you still need, not instead of it, because a knee that is still short of range keeps working the range.

If you have high blood pressure, breathe steadily through every repetition and never hold your breath. Check with your doctor before you start if you have a heart or lung condition, or blood pressure that is not under control.

The exercise program

The exercises below carry their own stop signs and precautions. Open the line under an exercise and read them before you try it.

Is this arthrofibrosis?

Possibly, and possibly not, and the honest answer is that it is not a word you can apply to your own knee from a web page.

The international consensus definition is deliberately plain: restricted movement in bending, straightening or both, caused by soft tissue fibrosis that was not there before the operation. The same consensus excludes stiffness caused by other things, and the list is worth reading if your knee was not replaced: a bone block to movement, implants that are the wrong size or position, metal hardware, the ligament reconstruction itself, infection, pain and complex regional pain syndrome (Kalson and colleagues, 2016). Read that twice, because the exclusions are the useful part. Arthrofibrosis is what is left when the other causes have been looked for and ruled out. It is not the first explanation for a stiff knee, and it is not something a scan hands you on its own.

The same review that argues against forcing also proposes a distinction patients rarely hear: an active phase, where inflammation and new tissue are still being laid down, and a residual phase, where that process has finished and the joint is simply left stiff (Usher and colleagues, 2019). The authors are clear that more work is needed to tell the two apart. It does explain why a knee can stop changing without anything new going wrong, and why what gets offered at four months is not what gets offered at four years.

How common it is depends almost entirely on who is counting and what definition they used, which is why the numbers in the literature are so far apart. One systematic review describes it as the most common complication after ACL reconstruction and found the definitions varied widely between studies (Ekhtiari and colleagues, 2017). In one published series from two hospitals, 8 of the 166 patients who completed follow-up after an ACL reconstruction, about 5%, met those authors' definition at six weeks, which they had set as losing 15 degrees or more of straightening compared with the other knee (Rushdi and colleagues, 2019). That is the point about definitions in practice: a figure like 5% means nothing until you know what had to be true to be counted. Both of those papers are about one operation, and neither transfers to yours.

You do not need the word in order to act. A range that has stopped moving goes to your team whether anybody has called it arthrofibrosis or not.

Stiffness after an arthroscopy or a meniscus operation

This is the reader most badly served online, because almost everything written about a stiff knee after an operation is written about knee replacements.

The NHS says a joint is likely to be painful and stiff at first after an arthroscopy, and puts full recovery anywhere from around one week to several months. OrthoInfo, the patient information service of the American Academy of Orthopaedic Surgeons, lists knee stiffness among the possible complications of knee arthroscopy. Unless you had a ligament reconstruction, a meniscus repair or cartilage restoration, its figure for getting back to most everyday physical activity is 6 to 8 weeks, and sometimes a good deal sooner than that. It asks you to exercise the knee regularly for several weeks, and notes that a formal physical therapy program may improve the final result. It also gives one timing instruction that often goes unread: contact your doctor if the pain has not started to improve within a few days of the operation.

Two traps here. One is assuming a small operation means a small recovery: the keyhole scars heal fast while the joint inside them still swells, and swelling costs you range on its own. The other is reading a trim and a repair as the same operation. If the meniscus was repaired rather than trimmed, your limits on bending, kneeling and weight are usually tighter and run for longer, and they come from your surgeon, not from a general program. Meniscus tear covers that difference in more detail.

We could find no dependable figure for how often a knee stays stiff after an arthroscopy or a meniscectomy. Nobody should be giving you one, including this page.

Stiffness after an ACL reconstruction

Here the guidance is clearer, and it starts earlier than most people expect. The Aspetar guideline recommends active knee movement from immediately after surgery within the surgeon's instructions, and it recommends against routine use of machines that bend the knee for you, having found no extra benefit for movement compared with active exercise (Kotsifaki and colleagues, 2023). The same guideline lists full extension among the things to have before returning to running, and full range before returning to sport, while stating openly that there is no research supporting those particular criteria.

Timetables do exist in practice, and they are measured in weeks. One published pathway from two hospitals escalated physiotherapy at six weeks for knees that had not reached their targets, which were 90 degrees of bend and less than a 5 degree lag in straightening, and considered an arthroscopic release at twelve weeks if there was still no improvement; five of their eight affected knees had that release and regained their movement (Rushdi and colleagues, 2019). It is a retrospective report of two centers' own protocol, not a standard anyone has to follow. The reason to know it exists is that it tells you the unit of time your surgeon is thinking in, and it is not years.

One specific cause deserves naming, because it feels different from ordinary tightness. A small nodule of scar tissue at the front of the knee, known as a cyclops lesion, can block the last few degrees of straightening. In an international survey of 250 knee surgeons, symptoms typically appeared within 3 to 6 months of the reconstruction, loss of extension beyond about 5 degrees was consistently the key finding, and initial management was mostly an extension-focused rehabilitation trial before any surgery was considered (D'Ambrosi and colleagues, 2026). If your knee feels blocked, clunky or sore right at the end of straightening rather than simply tight, that is the detail to report, and it is written into the warning lines below. ACL reconstruction rehab carries the staged program.

If you are one, two or five years out, with a knee that stiffens after sitting and aches by the evening, you are reading the right section but you are not in a post-operative pathway any more, and nobody is checking your movement unless you ask. Ask for both bending and straightening to be measured and compared with your other knee. A long-settled knee that has quietly lost a few degrees is worth more attention than it usually gets, for the reasons in the straightening section above, and persistent symptoms years after an injury belong with knee osteoarthritis or patellofemoral pain, not with a stretching program you design for yourself.

Stiffness after a fracture around the knee

Two things make this group different. The bone sets the timetable, not the soft tissue, so your surgeon's instructions about weight and movement are not negotiable in the way a general exercise progression is. The review of tibial plateau complications quoted later in this section lists going onto the leg early among the reported risk factors for the bone joining in the wrong position, which is the plainest reason to hold the limit you were given even on a day the knee feels ready for more (Gálvez-Sirvent and colleagues, 2022). And the stiffness has more possible sources, because the injury affected more than the joint lining.

The readable evidence here is about one fracture in particular. A review of complications after surgery for tibial plateau fractures reports the incidence of knee stiffness as 3% to 18% depending on the series, while the need for further surgery because of it is low, 0% to 5%. It describes the causes as multifactorial, inside the joint as adhesions, inflammation and scarring, and outside it as scarring of the extensor apparatus, the skin and the surrounding soft tissue, with bone irregularity after a malunion also blocking movement. Its listed risk factors include an external fixator, an open fracture, polytrauma, fasciotomies, wound problems that meant longer in a splint, an injury to the extensor apparatus at the front of the knee, and having had more than two operations. The common thread the authors draw out is time spent not moving (Gálvez-Sirvent and colleagues, 2022).

That split between inside the joint and outside it is worth noticing, because it changes what helps. When the block is outside the joint, in the thigh muscle and the scar around it rather than in the lining of the knee, the tightness tends to feel like something tethering the front of the thigh rather than something caught inside the knee, and the bend is often worse when you are lying flat or standing than when you are sitting, because the thigh muscle is stretched over the hip as well as the knee. Stretching the joint harder does not free that kind of block, and the part of the operating list that reaches it is different. So if that is the pattern you notice, describe it to your physio or surgeon in those words rather than treating it as a reason to push harder.

Those numbers are for surgery on tibial plateau fractures. They do not describe a kneecap fracture, a thigh bone fracture or any other operation, and the point of this page is that figures from one operation should not be handed to someone who had a different one.

The same review sets out the surgical ladder by how long the stiffness has lasted: manipulation under anesthetic for early cases, under about three months; arthroscopic release once it has run longer than that, and preferably before six months have passed, though the authors add that improvement has also been reported in releases done later than that; and open release for cases that have gone beyond six months or have not responded. Two surgical papers on post-traumatic stiffness more generally put the window in the same place. One states its own practice as considering arthroscopic release between three and six months after the fracture was fixed, in knees that have lost a lot of movement (Kukreja and colleagues, 2019). The other describes arthroscopic release as the standard option from about three months after the injury, or earlier if the range has clearly stopped improving (Rinaldi and colleagues, 2024). So the figures are not tied to one fracture, but all of them are about knees stiff after an injury, and none of them describes a reconstruction or a replacement.

One thing in that literature is not about timing at all, and it is the part a healing fracture most needs to know. The review of open release says the fracture has to have healed before the knee is released, which in practice means a wait of about three to six months, and puts it as a compromise between treating the stiffness and getting the bone to join (Rinaldi and colleagues, 2024).

Read the test rather than the number. What has to be true is that the bone has joined, and three to six months is how long that usually takes rather than a date the operation is booked for. Some fractures join sooner and some take longer, and the tibial plateau review above reports that failure to join, uncommon after these fractures in general, becomes more likely after a high energy injury with a lot of damage to the blood supply and the soft tissue around it (Gálvez-Sirvent and colleagues, 2022). So it is the X-rays that move this along rather than the diary, and the question to ask at a fracture clinic appointment is whether the bone has joined yet, because that is the gate.

That is not the same as nothing being possible before then. The same review describes gentle manipulation, not a release, as an option inside three months in selected cases, and its example is narrow: a break in the shaft of the thigh bone, well away from the knee, fixed with a rod down the inside of the bone, already showing signs of uniting on X-ray, where the contracture is not yet severe and the fixation is unlikely to fail. A plate across the top of the shin bone or the lower end of the thigh bone, right beside the knee, is not that situation, because the manipulation puts its force through the fracture and the metal rather than past them. Which of those applies to your knee depends on your fracture, your X-rays and your metalwork, and it is a surgeon's call rather than a physio's.

A waiting period is also not an empty one. The part of the ladder that is not surgery is what runs inside it, so a wait for the bone to join is a reason to stay in regular contact and have the range measured while it passes, not a reason to leave the conversation until the date comes round.

Read that ladder for what it is. It is not a deadline you have to beat, and it is not a reason to force a healing knee. It is the reason a stalled range gets mentioned early, because the options that need the least surgery sit at the near end of it. A paper on surgical technique for severe post-traumatic stiffness makes the same point from the other side, that what is offered is guided by the original injury, how much movement has been lost, how long it has been since the injury, and the state of the cartilage (Kukreja and colleagues, 2019).

Stiffness after a knee replacement

If your operation was a knee replacement, this page is not your page. That situation has its own expected ranges, its own timetable for manipulation and its own infection warnings, all of which differ from the ones above. Stiff knee after a knee replacement covers the bend, the plateau and what to ask, and total knee replacement rehab carries the staged program and the doses.

How long does knee stiffness last after surgery?

Longer than the leaflet suggests and less predictably than you would like, and anybody giving you a single number for every knee operation is guessing.

The published recovery windows are wide even within one procedure, from about a week to several months after an arthroscopy on the NHS's own account, and larger operations run in months. We found no national guidance putting a date against a particular number of degrees for any of these operations. The week-by-week range targets you will meet online come from individual hospitals and surgeons, which is why two of them can contradict each other without either being wrong for its own patients.

Which is why this page keeps pushing you toward direction rather than duration. A knee that is slowly improving has time. A knee that has stopped has a question that needs answering, and the answer is not another month of the same.

What counts as a plateau, and when should you raise it?

A plateau is a range that is not bending or straightening any further from one week to the next, a knee that is getting stiffer, or movement you had and have lost. Not a bad day. Not a stiff morning. A flat line over weeks.

Two or three weekly readings that come out the same is enough to pick up the phone. You do not need a longer run than that to be taken seriously, and waiting for a longer one is how people arrive at the appointment later than they needed to.

When that is your knee, it goes to your physio or your surgeon within the next few days, not at your next routine appointment. Three separate pieces of the literature point the same way. A systematic review of treatment for arthrofibrosis after ACL reconstruction found that in studies using a step by step approach, more than half of patients were treated successfully without an operation, on mostly low level evidence (Ekhtiari and colleagues, 2017). A 2024 review of knee fibrosis after reconstruction puts non-surgical treatment in the early stages and surgery in the advanced stage (Liang and colleagues, 2024). And the fracture review above sets out a surgical ladder that gets longer the longer the stiffness has been there (Gálvez-Sirvent and colleagues, 2022).

None of that gives you a date. The windows in the literature belong to particular operations and often to single series, and nobody can tell from here where your knee sits in them. That is an argument for raising it now, not for frightening yourself: the cost of mentioning it early is one phone call, and the cost of waiting is that the easiest options are the ones that time takes away.

Make the call easy to act on. Say that the range has not changed for a stated number of weeks, because that sentence is what moves it up a list, and bring the numbers if you have them. Say whether it is the bending, the straightening or both. Mention it if the end of straightening feels blocked or clunky rather than tight. And ask who your point of contact is for rehabilitation problems if nobody has told you, because that question is harder to answer later.

Manipulation and release: what they are and when they are considered

These come up early in any search, usually in frightening form, so it helps to know what they actually are.

A manipulation under anesthetic means the surgeon moves your knee through its range while you are asleep or numb from the waist down, to free tissue that has tightened. A release, also called an arthrolysis or a lysis of adhesions, is keyhole surgery to divide that tissue under direct vision. An open release is not simply the bigger version of that. It is for the tissue a camera inside the joint cannot reach, chiefly the thigh muscle and its tendon stuck down to the healing bone and to the scarred tissue outside the joint, which is why a surgeon tries to work out where the block actually is before choosing (Rinaldi and colleagues, 2024). Which one is considered, if any, depends on your operation, how much movement has been lost, how long it has been lost, and what else is going on in the joint (Kukreja and colleagues, 2019).

Both are operations rather than stretches under another name, and they carry the risks of operations. A review of knee fibrosis after ACL reconstruction lists nerve and blood vessel problems, fractures and loosening of ligaments among the reported complications of manipulation and release (Liang and colleagues, 2024). In knees stiff after an injury the reported risks of a manipulation are listed as a fracture, failure of the metalwork holding a fracture, a torn tendon and damage to the cartilage, which is one reason a surgeon weighs the date against the X-rays rather than the calendar (Rinaldi and colleagues, 2024; Kukreja and colleagues, 2019). The same review says manipulation is offered less often than it used to be, for those reasons, so if yours is not offered that is not somebody being slow about it. That is also the clearest answer to anyone offering to do something similar to your knee without an anesthetic and without an operating room.

Most stiff knees never get this far. In the systematic review above, more than half of the patients in step by step programs did not need an operation at all, and in that two-hospital ACL series, 5 knees out of the 166 followed up went on to a release (Ekhtiari and colleagues, 2017; Rushdi and colleagues, 2019). The published evidence behind all of it is mostly small and retrospective, which your surgeon will be weighing along with your knee, your imaging and what you need the knee to do.

This next part frightens people who were not told about it beforehand. The movement won in the operating room is not the movement you keep. In a systematic review of open release for knees stiff after injury, every one of the thirteen included studies reported some loss of bend from the immediate post-operative period onward, averaging about 13 degrees by the first check and a couple more degrees by final follow-up, while still leaving an average bend above 90 degrees (Rinaldi and colleagues, 2024). A surgical paper on arthroscopic release with manipulation says the same thing more bluntly, that even with routine therapy the early gains may not be maintained, and that controlling pain well enough for the person to do the exercises is what turns movement gained under anesthetic into movement they still have months later (Stiefel and McIntyre, 2017). So the days and weeks straight afterward are the part that decides the result, and a knee that keeps closing up is reported rather than waited out.

If you cannot tell whether yours has eased back and settled or is still closing, say so rather than deciding alone. It is a question about two measurements a week apart, and it is far easier to answer at two weeks than at two months.

If you have already had one of these procedures, your team's plan for the weeks afterward replaces this page, and the warning lines below include the two that matter most in those weeks.

When to see a physio or surgeon

Get in touch within a few days if your bending or straightening has not changed from week to week, if you are losing movement you had, if the knee feels blocked at the end of straightening, or if pain is stopping you doing the exercises you were given. Get in touch sooner than that, the same day, for anything in the warning list on this page.

It is also reasonable to ask for help at the start rather than at the point of frustration. If nobody has measured your knee, if nobody told you what range to expect for your operation, or if you do not know who to contact when rehabilitation stalls, those are all fair things to ask for at any appointment. If your problem is pain on bending rather than lost bend, knee pain when bending covers the wider picture.

This page is education rather than advice about your knee. Your surgeon and physio have seen your operation, your imaging and your leg, and their plan comes first.

For physiotherapists

The triage that matters on this page is the three-way one, and it is worth making explicit with patients: position-dependent stiffness that resolves with movement, a deficit that is improving on serial measurement, and a plateau or regression. Only the third changes the plan, and only serial measurement at a consistent time of day distinguishes the second from the third.

Give extension its own goal and its own progression rather than treating it as a by-product of flexion work. Shelbourne and Gray's long-term cohort is the argument for holding that line early, and the cyclops survey is the reason to examine the end-range feel rather than only the number (Shelbourne and Gray, 2009; D'Ambrosi and colleagues, 2026). An end-range block, clunk or pain, as distinct from a capsular stretch, is a referral finding.

In the post-traumatic group, separate the intra-articular block from the extra-articular one before deciding how much more range work is reasonable. The same review divides it that way: adhesions, hypertrophic fibrous tissue and a thickened capsule inside the joint, against quadriceps adherent to fracture callus, the aponeurosis and the intermuscular septum, muscle retraction and deep skin adhesions outside it, and says the source of the stiffness determines which procedure is performed (Rinaldi and colleagues, 2024). It matters at the bedside too. An extra-articular contracture, with the deficit worse when the hip is extended than when it is flexed, and a quadriceps that will not glide, does not respond to intra-articular stretching however long you persist, and that pattern earns a surgical opinion rather than another block of therapy. Fixation state belongs in the same conversation, because the union requirement, not the calendar, is what gates a release, and one of the published risks of manipulation in this group is implant failure with loss of reduction, reported where the knee was moved before the fracture had healed (Rinaldi and colleagues, 2024; Gálvez-Sirvent and colleagues, 2022).

Keep complex regional pain syndrome on the list in the post-traumatic group rather than at the end of it. The consensus definition excludes it as a cause of stiffness, and the systematic review of open release says plainly that it is important to identify, that it is hard to separate from post-traumatic stiffness because the two are often interlinked, and that surgery is deferred until the quiet phase if it does not settle on its own (Kalson and colleagues, 2016; Rinaldi and colleagues, 2024). Allodynia, and asymmetry of temperature, sweating or skin appearance, carry more weight here than the range figure does, and they are the one presentation where more stretch is the wrong answer.

Three points for the referral conversation. Reported rates of arthrofibrosis are not comparable across studies because the definitions are not, so a number quoted at a patient needs its definition attached (Kalson and colleagues, 2016; Ekhtiari and colleagues, 2017). The consensus also grades the loss as mild, moderate or severe by degrees of flexion or extension deficit; those bands are research classification and are deliberately not printed here, because "mild" read off a web page is a permission, and the same page cites a cohort in which 3 to 5 degrees of lost extension changed long-term scores (Kalson and colleagues, 2016; Shelbourne and Gray, 2009). And the intervention windows are procedure specific: the post-traumatic ladder, which three of the sources above put in roughly the same place (Gálvez-Sirvent and colleagues, 2022; Kukreja and colleagues, 2019; Rinaldi and colleagues, 2024), does not transfer to a reconstruction or an arthroplasty, the arthroplasty window does not transfer back, and neither should be used to reassure a patient whose operation it does not describe.

Common questions

References

1. Kalson NS, Borthwick LA, Mann DA, et al. International consensus on the definition and classification of fibrosis of the knee joint. The Bone and Joint Journal. 2016;98-B(11):1479-1488. https://doi.org/10.1302/0301-620X.98B10.37957
2. Usher KM, Zhu S, Mavropalias G, Carrino JA, Zhao J, Xu J. Pathological mechanisms and therapeutic outlooks for arthrofibrosis. Bone Research. 2019;7:9. https://doi.org/10.1038/s41413-019-0047-x
3. Ekhtiari S, Horner NS, de Sa D, et al. Arthrofibrosis after ACL reconstruction is best treated in a step-wise approach with early recognition and intervention: a systematic review. Knee Surgery, Sports Traumatology, Arthroscopy. 2017;25(12):3929-3937. https://doi.org/10.1007/s00167-017-4482-1
4. Shelbourne KD, Gray T. Minimum 10-year results after anterior cruciate ligament reconstruction: how the loss of normal knee motion compounds other factors related to the development of osteoarthritis after surgery. American Journal of Sports Medicine. 2009;37(3):471-480. https://doi.org/10.1177/0363546508326709
5. Kotsifaki R, Korakakis V, King E, et al. Aspetar clinical practice guideline on rehabilitation after anterior cruciate ligament reconstruction. British Journal of Sports Medicine. 2023;57(9):500-514. https://doi.org/10.1136/bjsports-2022-106158
6. Rushdi I, Sharifudin S, Shukur A. Arthrofibrosis following anterior cruciate ligament reconstruction. Malaysian Orthopaedic Journal. 2019;13(3):34-38. https://doi.org/10.5704/MOJ.1911.006
7. Liang Y, Zhang Q, Fan Y. Research progress of knee fibrosis after anterior cruciate ligament reconstruction. Frontiers in Pharmacology. 2024;15:1493155. https://doi.org/10.3389/fphar.2024.1493155
8. D'Ambrosi R, Helito CP, Della Villa F, et al. Practice patterns, controversies, and proposed management algorithm for cyclops syndrome after anterior cruciate ligament reconstruction: an international survey. Journal of ISAKOS. 2026;19:101165. https://doi.org/10.1016/j.jisako.2026.101165
9. Gálvez-Sirvent E, Ibarzábal-Gil A, Rodríguez-Merchán EC. Complications of the surgical treatment of fractures of the tibial plateau: prevalence, causes, and management. EFORT Open Reviews. 2022;7(8):554-568. https://doi.org/10.1530/EOR-22-0004
10. Kukreja M, Kang J, Curry EJ, Li X. Arthroscopic lysis of adhesions and anterior interval release with manipulation under anesthesia for severe post-traumatic knee stiffness: a simple and reproducible step-by-step guide. Arthroscopy Techniques. 2019;8(5):e429-e435. https://doi.org/10.1016/j.eats.2019.01.005
11. Rinaldi VG, Sassoli I, Fogacci A, et al. Old but gold: is the Judet procedure still a viable option for posttraumatic knee stiffness in 2024? A comprehensive systematic review and meta-analysis. Journal of Experimental Orthopaedics. 2024;11(4):e70079. https://doi.org/10.1002/jeo2.70079
12. Stiefel EC, McIntyre L. Arthroscopic lysis of adhesions for treatment of post-traumatic arthrofibrosis of the knee joint. Arthroscopy Techniques. 2017;6(4):e939-e944. https://doi.org/10.1016/j.eats.2017.03.001
13. Medina O, Arom GA, Yeranosian MG, Petrigliano FA, McAllister DR. Vascular and nerve injury after knee dislocation: a systematic review. Clinical Orthopaedics and Related Research. 2014;472(9):2621-2629. https://doi.org/10.1007/s11999-014-3511-3
14. American Academy of Orthopaedic Surgeons. Knee arthroscopy. OrthoInfo. https://www.orthoinfo.org/en/treatment/knee-arthroscopy/
15. NHS. Arthroscopy. Page last reviewed 7 September 2026. https://www.nhs.uk/conditions/arthroscopy/

Written and checked by the PocketPhysio editorial team. Last updated 2026-11-02.

This page is general education, not a diagnosis or a personal treatment plan. See a physiotherapist or doctor for advice about your own situation.