Partial knee replacement recovery: a staged program

How long does partial knee replacement recovery take?

One figure on this page was written for this operation instead of being borrowed from the bigger one, and it is OrthoInfo's. By 6 weeks after the operation, it says, you will most likely have all of your usual daily activities back. Daily living is the limit of that figure, and work, driving and sport have dates of their own further down this page. Before that point it says you begin putting weight on the knee immediately after surgery, and that you may need a walker, cane or crutches for the first several days or weeks. The NHS adds the comparison in one line, that recovery after a partial should be shorter than after a total, and the caution that full recovery from a knee replacement may take several months or longer.

Swelling outlasts the walking. Knowing that in advance saves a lot of worry. The unicompartmental leaflets put it at three months or more (Royal Berkshire) and up to six months to settle completely (King's College Hospital), so a knee that is still puffy in the evening months after the operation is ordinary. The direction matters more than whether it is there at all: swelling that increases from day to day instead of settling belongs in the warning signs at the foot of this page, and swelling that is slowly fading does not.

What the leaflets do about it is unglamorous. Rest with the operated leg up, ideally with the ankle higher than the hip, and support the lower leg on pillows or a stool rather than putting anything under the knee itself, which is the same mistake as the pillow. Keep the ankle pumps going, and use an ice pack wrapped in a damp tea towel instead of straight onto the skin. On how long to leave it on they disagree, from no more than 10 minutes with 20 minutes between applications (Royal Berkshire) to 15 to 20 minutes every two hours (King's College Hospital), so start at the shorter end and use your own team's version if they gave you one. Check that you have full feeling in the leg and no numbness before you use ice at all, because a numb leg cannot tell you it is being burned, and look at the skin before, during and after.

Hospital is short or skipped altogether. OrthoInfo says you will either go home after your surgery or be admitted for an overnight stay, and that people having this operation usually spend little or no time in hospital. The leaflets written for the partial say the same thing without hedging it. The Royal Berkshire's tells you plainly that you can get up and mobilise on the day of the surgery, and lists what you have to manage before you go home: getting from bed to chair and toilet on your own, walking with crutches, the stairs, and your own personal care with little or no help. King's College Hospital's has you standing as soon as the feeling and strength in your legs have come back enough, and says you could be discharged any time up to 8pm the same day. One Scottish trust's physiotherapy page for this operation puts the aim as being up and walking on the day of the operation or the next morning, in wording that covers knee replacement in general (NHS Greater Glasgow and Clyde). That pace is much of the point of the operation, and it is also why people get home before they have properly understood the plan.

Now the part that matters more than any of those numbers. If your team told you to keep the foot off the floor, to put only part of your weight through the leg, or not to bend past a set angle, that instruction governs and nothing here overrides it. Surgeons give those instructions for reasons specific to what they found and what they did, and OrthoInfo's sentence about weight going on immediately describes the usual partial knee replacement, not yours. It is not a permission.

The leaflets for this operation make the same point from both directions. Chelsea and Westminster's says that unless you are advised otherwise you are allowed to put full weight through the operated leg. King's College Hospital's leaves a blank line on its front page for your weight bearing status, and tells anyone who has been put on partial or touch weight bearing to keep the walking aid until that period is up or they are told otherwise. Two leaflets for one operation, and the instruction is still written for the person rather than for the procedure. If what you were told is stricter than what you read on this page, follow what you were told, and ask your team when it changes.

One reason for that sits inside the knee, not in the paperwork. A partial can be fixed to the bone with cement or without it, and the early risks are not identical. In a Danish national study covering 22,080 partial replacements, a break in the shin bone under the implant inside the first 4 months happened after 0.7% of the uncemented operations against 0.2% of the cemented ones, and the risk was concentrated in women, in people aged 70 or over, and in people who are shorter or heavier (Risager and colleagues, 2025). The risk is small either way. It does explain how two people who had the same named operation can go home with different instructions, which is one more reason to follow yours and not a page. If that list of groups describes you, the useful question at your next appointment is whether your implant was cemented and whether that is why they want you on the crutches for as long as they do. None of this is a reason to come off your feet. It is a reason not to hand the sticks back early on the strength of something you read.

What does a unicompartmental knee replacement actually replace?

A knee has three separate compartments that can wear out: the inside, the outside, and the surface behind the kneecap. A partial replacement resurfaces one of them and goes nowhere near the other two. OrthoInfo's sentence on what is left behind is the one to hold on to: the healthy cartilage and bone, as well as all of the ligaments, are preserved. Some units and some patients call it a half knee replacement, and unicompartmental knee replacement is the clinical name for the same operation.

That is why this page exists separately. Your remaining cartilage still has to tolerate load, your own ligaments still have to do their job, and both of those make the early advice a different conversation from the one after a total. Who gets offered which operation, and how the two compare on further surgery and complications, is a bigger question than a rehab page should try to answer in passing, so it has its own page: partial vs total knee replacement.

Where a partial changes the rehab, and where it does not

Three jobs make up the early work after any knee operation. Get the knee straight. Get it to bend. Get the thigh strong enough to control the leg going down a step. A partial does not change that list, and anyone who tells you the rehab is a different discipline is overselling it.

What does change is the pace and the paperwork of restrictions. Weight through the leg usually starts at once instead of being built up. The stay is hours or a night, not days. And the dates on the do-not list are set by your team, not by the published total knee timelines, because some of those dates exist for a knee where the whole joint surface and sometimes the kneecap have been replaced.

Keep one habit from the total knee pathway whichever operation you had. Judge a session by the next morning, not by how it felt at the time. Swelling and soreness arrive late, so the morning after is the honest reading. Two of the unicompartmental leaflets draw the line in almost the same words, and they are worth borrowing: a moderate ache that settles quickly is acceptable, and severe pain that takes hours to settle is not (Royal Berkshire). Chelsea and Westminster's adds swelling to the second half of that, and the test that it gets in the way of your exercises. Sharp pain is a different signal from either, and the Royal Berkshire's asks you to stop the activity straight away if you get it, and to get advice if it does not settle. What that rule cannot tell you apart is an overdone session from an infection around the new joint, because both show up as a knee that is more swollen and more sore. A knee that is also hot or red, a pain that keeps building while you rest, or a temperature puts you outside the training question and into the warning signs at the foot of this page.

If your operation is still ahead of you, the waiting weeks are the cheapest strength you will ever buy. The prehab before joint replacement program is written for them, and so is the knee osteoarthritis program if your team is happy for you to use it while you wait.

What the published evidence on partial knee rehab actually says

Very little, and saying so is more useful than tidying it up. A systematic review looked specifically at early physiotherapy after a first partial knee replacement and found eleven studies with 1,293 people between them, three of them randomized trials and eight observational. What it found was that the dose and content of post-operative physiotherapy was highly variable, with a recent move to minimal physiotherapy input alongside faster discharge (Jenkins and colleagues, 2023). So there is no settled protocol for this operation to hand you, and a page claiming one would be making it up.

The nearest thing to an answer on how much supervision you need comes from a single trial. Fifty-two people were randomly allocated after a partial knee replacement to 6 weeks of outpatient physical therapy or to an unsupervised home program, and the change in range of motion at 6 weeks came out the same in both groups, as did the other measures recorded (Fillingham and colleagues, 2018). The authors put it as self-directed exercise probably suiting most people after this operation, while a subset still benefits from formal therapy. Two of the people who switched from the home program to formal therapy inside 2 weeks later needed the knee freed up under anesthetic, which is the shape of that subset: the knees that were not moving were the ones that needed hands on them.

Hospitals still have to write something down, and three NHS trusts have. The leaflets from Chelsea and Westminster, the Royal Berkshire and King's College Hospital each set out a staged program for this operation with repetitions, frequencies and return-to-activity dates, and they are quoted throughout this page because they are the only material written for the partial and not for the total. They also disagree with each other, on the exercises, on the doses and on several of the dates, which is the variability Jenkins and colleagues measured showing up in public.

So the stages below are not a protocol derived from trials of this operation, because that evidence does not exist. They follow the three leaflets, the other published patient sources named on this page and ordinary physiotherapy practice, and your team's plan still comes first.

How to use this program

Start only with the exercises your team has given you or cleared, and work in the order below. Stage 1 is the first weeks. Move to stage 2 when stage 1 feels manageable and your team is happy with how you are walking, and to stage 3 when you walk steadily and stage 2 feels easy. Those signs come from common practice and are a guide, not a test you pass on a date. Moving on means moving, not adding. A later stage takes over from most of the one before it rather than sitting on top of it, which is how all three leaflets are built: King's College Hospital's moves you from its basic exercises to its intermediate ones rather than asking for both, Chelsea and Westminster's gives its later phase a separate list of its own, and the Royal Berkshire's finishes by asking you to carry on with the exercises you find most beneficial. So let a stage 1 or stage 2 exercise drop away once a harder one has taken its job over, and keep only the bending and the straightening going the whole way through, because those two go backward if you stop. If you find yourself doing every exercise on this page in one day, that is several times what any of these leaflets asks. And if you are months past the operation and an early stage still feels hard, that is a reason to ask for a physiotherapist rather than to push through the stage. The stages are not a test, and being at stage 1 in your third month is not a failure.

Each exercise page gives a typical starting dose. Three NHS leaflets written for this operation give numbers with the right label on them. The Royal Berkshire's starts the day after surgery with a minimum of 5 of each exercise, once or twice a day, holding each for a slow count of 5 to 10, and moves up to 10 of each 3 or 4 times a day if that brings no increase in pain or swelling. King's College Hospital's asks for three sessions a day, starting at 5 repetitions of each and building to 10 to 15. Chelsea and Westminster's spreads the early work through the day rather than stacking it, with about 10 repetitions of the thigh exercises 5 times a day, ankle pumps 20 times every waking hour, and the later standing strength work at 3 to 4 sessions a week.

Two things about those numbers. They are somebody else's starting points rather than your prescription, and your physio will adjust them to your knee and to the plan you were sent home with. And two of the three attach their frequency to the session rather than to the exercise, which is where a program quietly doubles. The Royal Berkshire's asks for a minimum of 5 of each exercise once or twice a day, and King's for three sessions a day, meaning the whole set three times and not each exercise three times. Chelsea and Westminster's works the other way round, with its own frequency written beside each exercise, which is why its numbers look so small: 6 heel slides five times a day, 6 standing knee bends twice a day. A reader who gives every exercise below its own three or four sessions a day ends up doing several times what any of the three asks for. Deliberately, there are no week numbers attached to the stages here. The readable sources for this operation do not support them, and an invented week number is the thing most likely to make a normal recovery look like a failing one.

Breathe normally through every repetition rather than holding your breath, especially if you have high blood pressure. Check with your doctor before you start if you have a heart or lung condition, or blood pressure that is not under control. If you have fallen in the past year, feel unsteady on your feet, or take medicines that make you dizzy, keep a hand on a kitchen counter, a handrail or a sturdy chair back for the whole of the standing exercises, practice only when someone else is at home, and ask your doctor or physio about a falls check. If you live alone, that needs a workable version rather than being dropped. Work beside a handrail, a kitchen counter deep enough to take both hands, or a sturdy chair back, whichever is where the exercise happens, keep your phone in a pocket rather than on the worktop, do the standing and balance work at a time of day when somebody could reach you, and tell someone you are about to do it. Leave the first attempt at anything new, and anything done on one leg, for a day when there is someone in the house.

NICE's joint replacement guideline asks a physiotherapist or occupational therapist to get rehabilitation started on the day of the operation where that is possible, and within 24 hours of it at the latest, for anyone who has had a primary hip, knee or shoulder replacement. It says that rehabilitation should take in advice on daily activities, a home exercise program and getting you moving (NICE NG157, recommendation 1.10.1). The guideline nowhere uses the word unicompartmental, and that recommendation does not separate the two knee operations; its knee section recommends offering a choice of partial or total replacement to people whose arthritis sits in the inner compartment alone, so a partial is inside what it means by a knee replacement. The same guideline asks for advice on self-directed rehabilitation before you leave hospital, and that anyone doing it be left with clear goals and a point of contact for advice and support. If nobody has told you what your home program is or who to contact about it, that is worth a phone call rather than a wait.

One last thing before you start. Each exercise below opens a box of its own stop signs and precautions, and those are to read before the first attempt rather than after it. Some of the lines in them are written for the other injuries that use the same exercise, a cast, a broken ankle, a pregnancy, so you will meet advice that has nothing to do with you. That is not a mistake and it does not mean the exercise is wrong for you. Read the knee, leg, chest and dizziness lines in each box, and treat the warning signs at the foot of this page as the list that is yours.

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.

Getting these exercises to work at home

The straightening work is what gets skipped. It is also the half that decides how the leg feels on stairs months later. The knee extension stretch and the seated heel prop achieve nothing while the thigh is still guarding, so rest your hands on the thigh and not on the kneecap, leave the calf unsupported, and allow the leg's own weight to do the work. Both are slow and boring on purpose, which makes them the two to pair with something you were going to watch anyway.

For quad sets, watch the kneecap instead of trusting what the leg reports. It travels a short distance toward your hip as the muscle genuinely switches on, and a roll about as thick as a folded hand towel gives the knee something to press into. A pillow holds it too bent for any of that to happen. In short arc quads the top of the movement is the whole exercise, so reaching properly straight matters more than the count.

Heel slides stick on a bare sheet; a sock on the foot fixes it. For the assisted seated knee bend, choose a seat with room underneath it, and keep an eye on your hips: the moment they travel forward along the seat, the knee has stopped bending and you are stretching nothing.

Seat height decides whether half of stage 2 is even possible. A firm dining chair beats a sofa, and a cushion on top of it can turn an impossible set into a manageable one. Begin at whatever height you can rise from without dragging on your arms, and take it down a little every week or two. For walking with a cane, two details change how it feels: it is held in the hand on your unoperated side and swings forward at the same time as the operated leg, and the handle wants to reach your wrist crease with the arm hanging loose.

What should you not do after a partial knee replacement?

This question gets asked constantly and answered, almost everywhere, with the total knee replacement list. Here is the honest version.

The NHS recovery page names a partial twice, for the shorter recovery and for the shorter driving wait, and gives the rest of its advice for knee replacement without separating the two operations: do not kneel on the new knee until your doctor says you can, do not sit with your legs crossed for the first 6 weeks, avoid twisting the knee, bending down and reaching up as much as possible, and leave household tasks that involve lifting or moving anything heavy for the first 3 months. Healthdirect Australia, writing for the same two operations together, says to avoid activities that involve jumping or twisting while you are recovering.

The three NHS leaflets written for the partial do have a short list of their own, and it is worth more to you than the general one. Do not put a pillow under the knee while you sit or sleep, because it holds the knee bent and keeps it stiff (Chelsea and Westminster, King's College Hospital). Avoid jarring and twisting jobs for 6 to 8 weeks, shopping trolleys and vacuuming included (Royal Berkshire). Leave the bath until the wound is fully healed and you can get out of it again, which is unlikely before 6 weeks, and shower or strip wash until then (Royal Berkshire, King's College Hospital). Do not stop your pain medication abruptly; come off it slowly (Chelsea and Westminster). And do not take up running, jumping or contact sport, which has its own section below.

Treat the general advice as a sensible default rather than as a rule written for your knee. After a partial your own team may lift some of them earlier, and they are the ones who know which compartments they left alone. What is not a default, and does not wait for anybody, is this short list. Do not push on through a knee that is hot, or that is more swollen and more painful every morning rather than settling. Do not give the walking aid back before your team says to; the afternoon that catches people out has a wet pavement, a curb and a supermarket in it. And do not treat a week of no progress in bending or straightening as something to try harder at, because that is the week to make the phone call.

Kneeling, twisting and deep bending

Kneeling first, because it is the one people ask about. The NHS line is that you do not kneel on the new knee until your doctor says you can, and Arthritis UK, formerly Versus Arthritis, says you can try kneeling on a soft surface from about 3 months after the operation and that it may never feel completely comfortable. Neither of those is written for a partial. One of the three unicompartmental leaflets is. The Royal Berkshire's says that kneeling on the new knee will do the replacement no harm, that some people can kneel within a few weeks of surgery and some never find it comfortable, and that the way in is to start on something soft such as a bed once the scar is comfortable, then try a firmer surface with a cushion or a kneeling pad. That is a good deal earlier than 3 months, which is a reason to ask the team that operated rather than to pick whichever of the two figures suits you.

Whenever you get the go-ahead, kneel on a folded towel or a pad and not on rough ground, because breaking the skin over a new joint is a way in for bacteria, and get down and up using your hands and something solid. Kneeling after a knee replacement carries the research.

Twisting does come with a timeframe once you read the leaflets for this operation. The Royal Berkshire's asks you to avoid jarring and twisting activities for 6 to 8 weeks and names the two everybody meets first, pushing a shopping trolley and vacuuming. King's College Hospital's says you may need help with heavier jobs such as vacuuming for up to 12 weeks, and the NHS leaves lifting or moving anything heavy for 3 months. None of them attaches a reason to the twisting itself, so take the instruction and do not let anyone sell you a mechanism for it, including us. What makes it workable is a habit rather than an explanation: pick your feet up and take several small steps to turn, instead of pivoting on the operated leg. Practice that in the kitchen and at the bathroom sink, which are the two places everybody forgets.

Deep bending has a clearer answer than you might expect. The answer is a negative result, not a gap. Three NHS physiotherapy leaflets written for this operation, from Chelsea and Westminster, the Royal Berkshire and King's College Hospital, set no maximum bend anywhere in them. They push the other way. The Royal Berkshire's says the scar tissue tightens over the first 6 weeks, that stretching early is what gets you the most movement you are going to get, and that after 3 months it becomes very difficult to improve flexibility. Both of the angle figures in those leaflets ask for bend rather than cap it: a bridge with the knees bent to at least 90 degrees, and a supported lunge taken to knees close to 90 degrees, the lunge being in the King's booklet rather than in the stages on this page. On where you end up, the Royal Berkshire's says you usually regain the movement you had before the operation and sometimes a little more, and that it may not let you fully squat or sit back on your heels.

So a bend limit after a partial is an instruction somebody has written for your knee, not a general rule of the operation. If your team gave you one, that is your limit, in every standing exercise above and in the bridge. If they gave you none, work the bend daily in the sitting and lying exercises, which is where those leaflets put the stretching, and still keep the mini squat shallow and controlled until you have asked, because the squat is the one that loads the new surface at the same time as bending it.

Driving, work and getting back to activity

Driving has a figure of its own for the partial operation. It is the one most readers come for. The NHS says to wait at least 6 weeks after a total knee replacement, or 3 weeks after a partial one, and to check with your doctor that you are fit to drive. Three weeks is a floor and not a date in your diary, and the hospital leaflets for this operation sit well above it: the Royal Berkshire's tells you not to drive until you can do an emergency stop and puts that at about 6 to 8 weeks, and Chelsea and Westminster's gives 6 to 8 weeks as a rough guide and asks you to tell your insurer about the operation. So 3 weeks is the earliest anybody allows rather than the usual point, and a longer wait from your own unit is the one to follow. Arthritis UK sets the test as being able to control the car safely and perform an emergency stop, and it adds that your insurer is worth a call about cover during recovery. Driving after a knee replacement goes through the braking test and the insurance question.

Long journeys as a passenger are a separate question. The answer there is about clots, not about the knee. The Royal Berkshire's leaflet advises against flying within 6 weeks of a joint replacement and against long haul flights for 3 months, and King's College Hospital's says not to fly for the three months after surgery and to avoid long car journeys for at least 6 to 8 weeks, stopping to move your legs on any that cannot be put off. Ask your own team before you book, and read the clot warning signs at the foot of this page before you travel.

Work does have figures written for this operation, and they are tighter than the general ones. Chelsea and Westminster's table puts sedentary work at 4 weeks and manual jobs at 5 to 6 weeks depending on the job, while its own opening page says office work or work without significant physical activity can start after six weeks. The Royal Berkshire's gives around 4 to 8 weeks for a sedentary job and up to 3 months where the job involves prolonged walking or standing or heavy lifting, and asks you to check the date with the surgeon. The NHS gives about 6 to 12 weeks depending on the job for knee replacement generally. Those overlap without agreeing, including inside one leaflet, which is where it actually stands: the range is real and the date is your surgeon's. What helps an employer is not a date anyway. It is which parts of the job this advice actually touches: kneeling, ladders, driving, lifting, and how long you can stand in one place.

Sport and higher-impact activity need your surgeon. The published picture is genuinely mixed. A review pooling thirteen studies and 1,675 people after a partial replacement found that most were playing sport again afterward, and that high-impact sport made up a bigger share of what they played afterward than the records showed beforehand, 21.1% against 6.3% (Cerasoli and colleagues, 2026). Read it with its own limits in view, because they are substantial. Its headline return rate differs between its abstract and its results, 87.97% and 87.2%. Four of its thirteen studies took their partial replacement group out of mixed groups that also included total replacements. It did not report how long the return took. And the before-and-after comparison is not a measured change: the review says many of the studies it pooled lacked detailed pre-operative sport data, which limits the ability to quantify net changes in sport participation, and that more sports were recorded after surgery than before, so some of that rise is better record keeping rather than more sport. In the same tables, intermediate-impact participation did not rise at all, 22.9% before and 20.4% after. So it tells you that going back to sport after this operation is usual. It does not tell you when, for whom, or how much of the impact figure is real.

Against that sits the registry finding that partial replacements are revised more often than total ones by 10 years, and something more pointed than general surgical caution. Two of the three NHS leaflets written for this exact operation advise against impact, and neither of them puts a date on when the advice stops. The Royal Berkshire's says a unicompartmental replacement is not made for high impact activities such as running and jumping, that walking and swimming are encouraged while sports calling for jogging, jumping or contact are not, and it repeats the same thing in its 3 to 6 month section. King's College Hospital's gives the reason: your new joint is designed for day-to-day living and not for high-impact sport, because that raises the risk it wears out sooner, loosens or breaks. Both also ask you to talk to your surgeon or physiotherapist before you take up anything more demanding.

The caution below is not borrowed from the total knee pages. Here the instruction to follow is the cautious one. If you have had a hip or knee replacement, get the go-ahead from your surgeon before you try running or jogging. Many surgeons advise against running and jogging after a joint replacement for good, not just for the first months, because each step sends impact through the new joint. The same question goes to your surgeon before any jumping or hopping. Many surgeons advise against jumping and hopping after a joint replacement for good, not just for the first months, because each landing sends high impact through the new joint. What the same leaflets allow is a good deal more than walking, and that part gets left out of most impact warnings. The Royal Berkshire's puts golf at 8 to 12 weeks, starting with chipping and putting before you play 9 or 18 holes, and leaves the driving range for up to 3 months because the repetitive twisting aggravates the knee. In its 3 to 6 month section it expects you to be back to sports such as golf, gentle tennis or badminton, and to bowls, gardening and dancing. It asks you to consult your surgeon or physiotherapist before you return to any sport, which is the step to take instead of reading a date off this page. If none of that is what you want the knee for, you are not behind anybody. Most people have this operation for the stairs, the shops and a night's sleep, and the stage 3 work is what buys those; the dates above matter only if you are heading for a particular sport. Cycling and swimming are the usual ways back to general fitness in the meantime, and neither replaces the strength work. Both carry a date in the leaflets for this operation, and the swimming one is about your wound rather than your knee. Wait until the wound is fully healed before you get into a pool, which Chelsea and Westminster's puts at 6 weeks, and it asks for a straight kick rather than breaststroke. It gives a static bike from 4 weeks, a cross trainer or rowing machine from 6 weeks and road cycling from 3 months, and the Royal Berkshire's also starts a static bike at week 4, pedaling or rocking for 5 to 10 minutes.

What if your recovery is not faster than a total?

Plenty of people land here two to six weeks out, holding a total knee timeline they found online and worried by it. Start with why the comparison is slippery. A partial is only possible when the wear sits in one compartment, so the group having one starts with less damaged knees on average. Some of the faster recovery belongs to the operation and some belongs to the knee, which means the sentence "a partial recovers faster" is partly a sentence about who gets a partial. It is not a promise made to you.

Several ordinary things slow a recovery down without anything being wrong. A knee that was very stiff before the operation. A long wait, with the thigh getting weaker through it. The other knee, the back or a hip taking a share of the work. Being less fit going in. None of those is a complication and all of them lengthen the weeks.

The finish line is also further out than most people are told, and the leaflets for this operation say so. Chelsea and Westminster's has improvements carrying on for up to 24 months as the muscles get stronger and the tissues heal. The Royal Berkshire's says joint line pain or tenderness can persist for up to two years after the surgery and that some people are left with some of it. King's College Hospital's puts a full recovery at up to a year. A knee that is still improving at eight months is doing what these sources say it does.

So use markers, not dates. Putting a sock on with nobody helping. Getting into the back of a car. Sleeping through to the morning. Reaching the end of your street and coming home again without having planned where to stop. On an ordinary bad day, drop back a step instead of stopping, and keep the bending and straightening going even when the standing work is left for tomorrow, because a day of nothing costs more than an easy day. A knee that is hot, or more swollen and more painful every morning instead of settling, is not an ordinary bad day and is not one to train through at any level: that one is a phone call.

Two patterns are not a slow recovery and should not be waited out. A knee that is no further on in bending or straightening from one week to the next, or that is getting stiffer, needs your physio or surgical team inside a few days rather than at the next routine appointment, because stiffness is easier to deal with early than late. Pain is the other, and the two year figures above are not a reason to sit with it. Pain that is climbing week by week rather than easing, pain that is there while you rest or that wakes you and is getting worse, or a knee that turns hot, red or more swollen, all go to your surgical team the same day, and the warning signs at the foot of this page say why and what they are looking for. Pain that had settled months or years ago and has come back, or that is creeping up month by month rather than week by week, is a review rather than a same-day call, and that line is down there too. The difference is direction, not speed: slow and still improving is a recovery, and going backward is a question for your team whatever week you are in. For what the sources say about the related worries, stiffness and bending after a knee replacement and how long swelling lasts were written for the total operation, and the general points in them hold.

When to see a physio or doctor

Get help the same day for anything in the warning signs below, and that list is the part of this page to read before you need it. Problems after a knee replacement sets the same signs out by how fast to act and adds who to call and what to have ready. If you cannot get hold of your surgical team on the day, do not let that turn into waiting for your next appointment. Use whatever urgent medical advice there is where you live, an out of hours service, an urgent care center or an emergency department, and tell them you have had a knee replacement and how long ago. The wound, the clot and the infection lines below are the ones that do not keep until morning.

Short of that, contact your physiotherapist or physical therapist, or your surgical team, if your bend or your straightening has not moved for a week or two, if you are still limping once the walking aid has gone, if you cannot get up from a chair without pulling on your arms by the time stage 2 should feel easy, or if you are not confident you are doing this safely on your own. Those are the same grounds NICE gives for being offered supervised rehabilitation rather than doing it alone: difficulty with daily activities, ongoing loss of function, or self-directed rehabilitation that is not getting you to your goals (NG157, recommendation 1.10.5). Asking is not a sign that something has gone wrong. Raise it at your follow-up appointment too if the knee will not straighten fully or bend far enough for stairs and for getting in and out of a car.

If you are older, have fallen before or feel unsteady, the balance and falls prevention program builds on stage 3, and a fall onto a new knee is worth the trouble of preventing. The total knee replacement rehab program and the week by week pages with it were written for a first total knee replacement, so use them for background and follow the plan from the team that did your partial.

For physiotherapists

This page is a general framework after a primary unicompartmental knee replacement and defers to the surgical protocol throughout, including on weight bearing and any flexion ceiling. It deliberately carries no week numbers on its stages. Jenkins and colleagues (2023) reviewed early post-operative physiotherapy after primary unilateral unicompartmental knee replacement and included eleven studies with 1,293 participants, three randomized trials and eight observational studies, reporting that the dose and content of post-operative physiotherapy was highly variable, with a recent move to minimal physiotherapy input alongside rapid recovery and same-day discharge pathways. No included study was at low risk of bias. There is no protocol in that literature to reproduce, and attaching total knee replacement milestones to a unicompartmental knee would misdescribe the operation.

Fillingham and colleagues (2018) randomized 52 patients after unicompartmental knee arthroplasty to 6 weeks of outpatient physical therapy or an unsupervised home exercise program; 25 received outpatient therapy and 22 the home program, and neither the change in range of motion nor the total arc of motion differed significantly at 6 weeks, with no significant difference in the Knee Society Score, KOOS Jr, Lower Extremity Functional Scale or VR-12. Two patients who crossed over from home exercise to formal therapy within 2 weeks subsequently required manipulation under anesthesia, which is the clinically useful part of a small trial: triage on range of motion trajectory rather than on pathway.

Screen as you would after any arthroplasty, for venous thromboembolism, wound and periprosthetic joint infection, a new extensor lag, progressive stiffness and peroneal nerve symptoms, and after any fall for periprosthetic fracture. Fitch and colleagues (2022) is the reason the fracture warning sign also covers the untraumatized knee: their eight periprosthetic tibial plateau fractures after unicompartmental arthroplasty presented at a median of 14 days, range 5 to 52, all as a vertical shear pattern exiting at the metadiaphyseal junction. The trigger on that line is weight-bearing pain rather than a reported injury, and the window is now open-ended with the first 3 months named as the pressing part. Fitch's abstract reports no mechanism either way, so the eight cases cannot be called atraumatic from it, but the wider literature can be. Burger and colleagues (2022) systematically reviewed 81 studies of periprosthetic tibial fracture after unicompartmental arthroplasty and classified 127 fractures by timing and 113 by mechanism: 23 intraoperative, 68 within 3 months, 19 between 4 and 12 months and 17 after 1 year, with 95 of 113 non-traumatic. Thoreau and colleagues (2022) state that the first 3 months are critical because of transient post-operative osteoporosis and local biomechanical change, and that pain on weight bearing signals bone-load imbalance and should be a trigger for unloading, which is the surgical basis for the page telling the reader to stop, unload and ask for an X-ray rather than to watch it. Because 28% of the dated fractures in Burger's series presented after 3 months, the line carries no closing date, which also keeps it consistent with the page's undated same-day line for sudden inability to weight bear.

Four dose points for anyone adjusting the stages. The two extension holds in stage 1 do one job in two positions and their library doses would stack to over an hour a day at the top of both ranges, so the stage text asks for one at a time, and it also caps the day's total at about half an hour and ties the exercise to an extension deficit: the only unicompartmental leaflet carrying it prescribes it for a knee that does not straighten fully because the posterior structures are tight, holding for as long as tolerated, with no duration and no frequency attached (Royal Berkshire), and the 10 minutes 3 to 4 times a day in our library is low-load prolonged stretch dosing for a fixed flexion contracture rather than a routine post-unicompartmental dose. The stage 2 standing strength work follows Chelsea and Westminster's 3 to 4 sessions a week rather than the once or twice daily that several of the individual exercise pages give for general use, because ten exercises at daily top-of-range is about 426 repetitions, counted 2026-11-02, and exceeds anything in the three leaflets. Stage 3 now carries the same frequency with a day between sessions, because it is the same phase in Chelsea and Westminster's terms and is the heaviest work on the page. And the stages are explicitly not additive: all three leaflets progress by replacing or selecting rather than accumulating, and stacking all 25 exercises at top of range comes to roughly 808 repetitions a day before the holds, the stairs and the walking.

Three sourcing notes for reviewers. The numeric dose, the return-to-activity timings and the absence of a flexion ceiling all come from three NHS trust unicompartmental leaflets read in full rather than from total knee material: Chelsea and Westminster (March 2024), the Royal Berkshire (reviewed October 2025) and King's College Hospital (April 2026). None of the three sets a maximum flexion angle; the only angles in them are an inner range prescription, a bridge at or past 90 degrees and a supported lunge near 90 degrees, and the Royal Berkshire's actively pushes early range on the grounds that flexibility is hard to gain after 3 months. Note that the King's booklet carries residual hip wording in several sections, including its driving paragraph, so it is cited here only where the text is knee-specific or generic.

The manipulation timing quoted in the stiffness warning sign comes from Hosanee and colleagues (2026), a total knee arthroplasty paper. The sentence is kept whole as recorded house wording, and the qualifying lines after it now carry Rako and colleagues (2026), the first unicompartmental series on this: 140 manipulations in 41,220 Medicare unicompartmental arthroplasties (0.34%) within one year, median 2.02 months, IQR 1.61 to 2.78, which falls inside the 6 to 12 week window without testing whether earlier is better. Incidence is lower than the total knee literature; the timing is concordant.

On return to activity, Cerasoli and colleagues (2026) pooled thirteen studies and 1,675 patients and reported a weighted mean return to sport rate given as 87.97% in the abstract and 87.2% in the results, with high-impact activity making up 6.3% of the sports reported preoperatively against 21.1% postoperatively; four of the included studies drew their unicompartmental arm from mixed cohorts, and time to return was not reported. Those percentages are shares of the sports recorded in each period and not proportions of patients, which is a further reason not to read them as a measured change in behavior. The pre to post comparison should not be read as a measured increase. The authors state that detailed preoperative sport data was missing from many of the studies they included, so a net change in sport participation cannot be quantified from them, and that substantially more sports were reported postoperatively than preoperatively, which may have underrepresented preoperative participation; intermediate-impact participation was 22.9% against 20.4%. The abstract's wording, increased participation in intermediate and high-impact activities, is not supported by the intermediate figures in the results, so this page follows the results. That is participation data in any case, and not a recommendation. The impact caution on this page is now sourced to the unicompartmental leaflets themselves, two of which advise against running, jumping and contact sport with no end date, rather than to OrthoInfo's total knee activity page, and it stands alongside the higher 10-year revision rate in the matched registry analysis (Mohammad and colleagues, 2022). The open-ended wording stands as the page's general advice, on grounds re-derived from the leaflets themselves. The Royal Berkshire's states in its introduction, with no phase and no date attached, that these replacements are not made for high impact activities such as running and jumping, which is a statement about the implant rather than about a stage of recovery, and it repeats the avoidance in its 3 to 6 month section. Correction to an earlier note on this page: that 3 to 6 month section is not the last point at which the leaflet speaks. It has a 6 months to 1 year section after it, which says you should be well enough to continue with your life normally and that joint line pain can persist for up to two years, and which neither repeats nor lifts the impact advice. So the argument is not silence after a date; it is that the prohibition was never dated in the first place and is never withdrawn. King's adds a mechanism, wear, loosening and breakage, that does not expire with a date, and OrthoInfo's activity page for the total operation independently asks readers to avoid high-impact activities such as jumping, jogging or skiing with no end point. Whether it should be lifelong for an individual knee remains that surgeon's call, and the page says so.

Note on the King's booklet, confirmed on surgical review. It is a unicompartmental knee booklet built on a hip template, and the residual hip wording is not confined to the driving paragraph and the discharge checklist. Counted on 2026-11-02 by searching the full text for the string "hip": 15 occurrences, 14 of them hip wording or a cross-reference to hip material and 1 anatomical, which is the instruction to rest the ankle higher than the hip. The 14 sit in 12 separate places: the enhanced recovery paragraph and its NHS cross-reference, the spinal anaesthetic paragraph and its Royal College of Anaesthetists cross-reference, the discharge checklist asking whether you have had a hip X-ray, the outpatient consultant clinic paragraph, the physiotherapy referral sentence about a hip class, the 12 month sentence about a hip feeling more natural, the driving paragraph, the car transfer instruction, the weight advice about pressure on your new hip, and one an earlier note did not name, the "stepping up your exercises" section, which sends readers to online videos of intermediate and advanced exercises for total hip replacements.

The ruling is that the booklet stays as a source and is quoted only where the sentence is knee-specific or carries no body part, never where a figure could have come down the hip template, and never for an exercise it links out to. Every citation to King's on this page was then checked against its own sentence: knee-specific (swelling to six months, the ice regimen applied to the knee, the pillow, the dose sentence, which sits under a heading saying these exercises help you recover after knee surgery, and the supported lunge) or carrying no body part (same-day discharge up to 8pm, standing once feeling and strength in the legs return, full recovery up to a year, no routine onward physiotherapy referral). The trap was the recovery horizon: the booklet's generic "it may take up to a year for you to make a full recovery" is the sentence used, and the hip one about 12 months sits a few paragraphs later. Its driving figure of six weeks is not used, because that sentence names a hip.

Common questions

References

1. American Academy of Orthopaedic Surgeons. Unicompartmental (partial) knee replacement. OrthoInfo. https://www.orthoinfo.org/en/treatment/unicompartmental-knee-replacement/
2. American Academy of Orthopaedic Surgeons. Total knee replacement. OrthoInfo. https://www.orthoinfo.org/en/treatment/total-knee-replacement/
3. American Academy of Orthopaedic Surgeons. Activities after total knee replacement. OrthoInfo. https://www.orthoinfo.org/en/recovery/activities-after-knee-replacement/
4. American Academy of Orthopaedic Surgeons. Infection after joint replacement. OrthoInfo. https://www.orthoinfo.org/en/diseases--conditions/joint-replacement-infection/
5. NHS. Recovering from a knee replacement. Page last reviewed 9 September 2026. https://www.nhs.uk/tests-and-treatments/knee-replacement/recovery/
6. Healthdirect Australia. Knee replacement. https://www.healthdirect.gov.au/knee-replacement
7. Arthritis UK. Knee replacement surgery. https://www.arthritis-uk.org/information-and-support/understanding-arthritis/arthritis-treatments/surgery/knee-replacement-surgery/
8. NHS Greater Glasgow and Clyde. Unicompartmental knee replacement. https://www.nhsggc.scot/?p=75915
9. Chelsea and Westminster Hospital NHS Foundation Trust. Unicompartmental (partial) knee replacement: advice and exercise following your operation. March 2024 update. https://www.chelwest.nhs.uk/your-visit/patient-leaflets/advice-and-exercise-following-ukr-march-2024.pdf
10. Royal Berkshire NHS Foundation Trust. Uni-compartmental knee replacement (UCA) advice and exercises. Reviewed October 2025. https://www.royalberkshire.nhs.uk/leaflets/uni-compartmental-knee-replacement-advice-and-exercises
11. King's College Hospital NHS Foundation Trust. Uni-compartmental knee replacement. April 2026. https://www.kch.nhs.uk/document/uni-compartmental-knee-replacement/
12. Jenkins C, Minns Lowe C, Barker KL. Early post-operative physiotherapy rehabilitation after primary unilateral unicompartmental knee replacement: a systematic review. Physiotherapy. 2023;118:39-53. https://doi.org/10.1016/j.physio.2022.05.003
13. Fillingham YA, Darrith B, Lonner JH, Culvern C, Crizer M, Della Valle CJ. Formal physical therapy may not be necessary after unicompartmental knee arthroplasty: a randomized clinical trial. Journal of Arthroplasty. 2018;33(7S):S93-S99.e3. https://doi.org/10.1016/j.arth.2018.02.049
14. Mohammad HR, Liddle AD, Judge A, Murray DW. A matched comparison of long-term outcomes of total and unicompartmental knee replacements in different ages based on national databases. Journal of Arthroplasty. 2022;37(2):243-251. https://doi.org/10.1016/j.arth.2021.09.018
15. Cerasoli T, Favero A, Coliva F, Fogacci A, Pasquini A, Zaffagnini S, Marcheggiani Muccioli GM. Sport participation after unicompartmental knee arthroplasty: high return rates independent of implant design or technique, a systematic review and meta-analysis. Journal of Experimental Orthopaedics. 2026;13(1):e70514. https://doi.org/10.1002/jeo2.70514
16. Hosanee S, Elebo N, Sikhauli N, Pietrzak JRT. Reluctant knee: understanding and managing stiffness after total knee arthroplasty. International Orthopaedics. 2026;50(9):2337-2346. https://doi.org/10.1007/s00264-026-06905-0
17. Rako K, Park J, Barbera J, Kumar A, Zubizarreta N, Poeran J, Hayden BL. Manipulation under anesthesia after unicompartmental knee arthroplasty: incidence and risk factors. The Knee. 2026;60:104408. https://doi.org/10.1016/j.knee.2026.104408
18. Fitch AA, Terhune EB, Cohn MR, Wright-Chisem J, Weatherford BM, Williams JC. Periprosthetic tibial plateau fractures after unicompartmental knee arthroplasty are successfully treated with open reduction and internal fixation. Orthopedics. 2022;45(5):287-292. https://doi.org/10.3928/01477447-20220425-04
19. Burger JA, Jager T, Dooley MS, Zuiderbaan HA, Kerkhoffs GMMJ, Pearle AD. Comparable incidence of periprosthetic tibial fractures in cementless and cemented unicompartmental knee arthroplasty: a systematic review and meta-analysis. Knee Surgery, Sports Traumatology, Arthroscopy. 2022;30(3):852-874. https://doi.org/10.1007/s00167-021-06449-3
20. Thoreau L, Morcillo Marfil D, Thienpont E. Periprosthetic fractures after medial unicompartmental knee arthroplasty: a narrative review. Archives of Orthopaedic and Trauma Surgery. 2022;142(8):2039-2048. https://doi.org/10.1007/s00402-021-04063-z
21. Risager S, Troelsen A, Viberg B, Henkel C, Lindberg-Larsen M. Risk of early periprosthetic tibial fracture after medial unicompartmental knee arthroplasty with cemented versus cementless fixation: a nationwide cohort study. Journal of Bone and Joint Surgery (American). 2025;107(15):1670-1679. https://doi.org/10.2106/JBJS.24.01538
22. Chartered Society of Physiotherapy. NICE recommends rehab on same day as surgery for joint replacement patients. 10 June 2020. https://www.csp.org.uk/news/2020-06-10-nice-recommends-rehab-same-day-surgery-joint-replacement-patients
23. National Institute for Health and Care Excellence. Joint replacement (primary): hip, knee and shoulder. NICE guideline NG157. Published 4 June 2020. https://www.nice.org.uk/guidance/ng157
24. NHS. Deep vein thrombosis (DVT). Page last reviewed 30 April 2026. https://www.nhs.uk/conditions/deep-vein-thrombosis-dvt/
25. NHS. Pulmonary embolism. Page last reviewed 15 October 2026. https://www.nhs.uk/conditions/pulmonary-embolism/
26. NHS. Sepsis. Page last reviewed 14 May 2026. https://www.nhs.uk/conditions/sepsis/
27. NHS. Anticoagulant medicines. Page last reviewed 9 September 2024. https://www.nhs.uk/conditions/anticoagulant-medicines/
28. NHS. Low blood pressure (hypotension). Page last reviewed 11 July 2023. https://www.nhs.uk/conditions/low-blood-pressure-hypotension/
29. NHS. Falls. https://www.nhs.uk/conditions/falls/

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.