Outer knee pain is one of the most common reasons a runner or a cyclist stops. The typical story is always the same: the first kilometres go well, then a precise burning sensation arrives on the outer side, always more or less at the same point of the route, and running becomes impossible. It goes away with rest; it comes back on the next run.
There is an explanation for this disorder that almost everyone accepts, the band that "rubs" on the bone, and a remedy that almost everyone proposes: rolling the thigh on a foam roller and strengthening the gluteus medius. Research over the last twenty years has called the first into question and scaled back the second, and it is worth knowing this, because it changes what it makes sense to do.
In this guide you will find what the literature really says about outer knee pain: which causes to consider, which signs must not be ignored, what has been documented to work and what we keep repeating out of habit.
⚠ Important: read this before continuing
The exercises and advice in this article are general guidance and may not be suitable for your situation. Every person has a unique clinical history: what works for others could be counterproductive for you.
Want a professional's opinion? Start from the questionnaire: you register and see who will follow you, a physiotherapist registered with the Albo chosen for your case, whom you write to in chat. If you prefer to hear a voice, the first consultation is free and we look at your situation together.
First of all: the signs that come before the exercises
Most lateral knee pain is an overload problem and is resolved by managing the load. However, some situations are not overload, and recognising them straight away matters more than any exercise.
- A hot, swollen, red knee, perhaps with fever or general malaise. It may be acute arthritis or an infection of the joint, and the window for acting is narrow. We discuss this in more depth in the article on a hot and swollen knee.
- After a trauma: you cannot bear weight, or you feel the knee "opening" outwards. A blow received on the inner side can damage the lateral collateral ligament or the posterolateral structures, and in those cases an assessment is needed, not a strengthening programme. To decide whether an X-ray is needed after an acute trauma, doctors use a validated clinical rule (the Ottawa knee rule): in a systematic review of 6 studies and 4,249 adults the pooled sensitivity was 98.5%, meaning that a negative result excluded fracture with good reliability. It is not a rule to apply on your own: it is the reason why, after a real trauma, the visit comes first.
- The knee stays locked and does not fully straighten. A meniscal fragment or a loose body can get caught in the joint: read about a locked knee.
- You cannot lift the tip of the foot, or you have tingling and loss of sensation on the top of the foot. The common peroneal nerve runs superficially, just behind the head of the fibula: an injury to it is a neurological problem and needs to be looked at quickly.
- Night pain that does not calm with rest, swelling that grows for no reason, unexplained weight loss. These are rare situations, but they are exactly the ones in which waiting is not wise.
- A swollen, painful and warm calf together with knee pain, especially after immobilisation or a long journey: a vascular problem must be ruled out.
If none of this applies to you, then yes: you are almost certainly reading about an overload, and that is what the rest of the article is about.
What is really there on the outer side of the knee
The outer side of the knee is crowded, and each structure has a fairly recognisable way of hurting.
The iliotibial band. It is neither a tendon nor a ligament: it is the thickened portion of the fascia lata, the fibrous sheath that wraps the whole thigh. At the bottom it is anchored to the femur by robust fibrous bands, which are not pathological adhesions (they are present in everyone), and it continues down to the tibia. It is the structure most often involved.
The lateral collateral ligament. A thin cord that runs from the femur to the head of the fibula and prevents the knee from opening outwards. It is injured by trauma, not by wear, and gives acute pain with a sense of instability.
The lateral meniscus. More mobile and rounder than the medial meniscus, it tears less often, but when it does the pain is on the joint line, the crease you feel pressing between femur and tibia, and it can be accompanied by clicking or locking.
The popliteus tendon. A deep muscle that "unlocks" the knee at the start of flexion and controls rotation. Its tendinopathy typically causes posterolateral discomfort that gets worse going downhill.
The fat layer under the band. It is the structure almost nobody talks about and is probably the real protagonist: a layer of fatty and loose connective tissue, very vascular and very innervated, lying between the band and the femoral epicondyle. We come back to it in a moment.
The common peroneal nerve. It wraps around the neck of the fibula just below the knee, in a very superficial position. Local compression can cause lateral pain with symptoms running down towards the foot.
The outer compartment of the joint. It can develop osteoarthritis like the inner one, although less often; bow legs shift the load towards the inside, knock knees towards the outside.
The most common cause, and the wrong explanation we give for it
In runners, outer knee pain is almost a synonym for iliotibial band syndrome. A systematic review on the subject describes it as the most common injury of the outer side of the knee in runners, with an estimated incidence between 5% and 14%; a more recent review puts it at around 10% of all running-related injuries. On this point the data are solid.
On why it hurts, however, what is read almost everywhere is outdated. The historical name, "iliotibial band friction syndrome", describes a band that would slide back and forth over the lateral femoral epicondyle with every flexion and extension, inflaming a bursa. An anatomical study published in the Journal of Anatomy dismantled this account: in 15 cadavers the band was always anchored to the distal femur by fibrous bands, and no bursa was found in any cadaver, in any of the 6 asymptomatic volunteers or in any of the 2 athletes with acute syndrome. The MRI scans showed that at about 30 degrees of flexion the band is compressed against the epicondyle by the internal rotation of the tibia, while in extension it shifts laterally. In the two symptomatic athletes, the signal changes were in the area occupied by fat, under the band.
Put simply: the band does not "roll" over the bone, it cannot, because it is anchored and because it is part of the fascia lata. The sensation of sliding is an illusion produced by the variation in tension between the anterior and posterior fibres. What gets irritated is the innervated fat layer that is squeezed underneath.
This is not academic nitpicking. If the problem is compression and not friction, the idea of "breaking up adhesions" or lengthening the band loses its meaning, while all the strategies that reduce how much and how often that area is compressed gain importance: running volume, cadence, slope, control of the hip during the stance phase.
An important tip: before trying these exercises, make sure you have received a correct diagnosis. People often think they have one problem, but the real cause of the pain is another. Performing the wrong exercises can make things worse instead of better.
The other causes to keep on the table
Attributing every pain on the outer part of the knee to the band is the most common mistake, and it is the one that wastes months.
Lateral meniscus tear. Less frequent than a medial meniscus tear. The pain is on the joint line, often with a sensation of catching, giving way or true locking. In degenerative tears in adults surgery is not the shortcut it appears to be: we discuss this further on.
Lateral collateral ligament and posterolateral structure injury. Traumatic in origin, typically a blow received on the inner side of the knee. Acute pain and instability in varus.
Popliteus tendinopathy. Posterolateral pain that flares going downhill and on downhill runs. It is rare, and precisely for that reason it is often labelled as "the band".
Osteoarthritis of the lateral compartment. More diffuse pain, stiffness on waking, gradual worsening over years rather than weeks.
Meniscal cyst. It can appear as a palpable swelling on the outer side, accompanied by a tear of the meniscus underneath.
Peroneal nerve irritation. Lateral pain with an electric component, tingling, sometimes difficulty lifting the tip of the foot.
Referred pain from the hip. A hip that does not work well can show up as knee pain, and it happens more often than is believed: we have dedicated an article to the connection between hip and knee.
Outer pain without swelling: what it means
This is a clarification that many people add when searching, and it changes the probabilities in a useful way.
The absence of swelling makes the acute injuries that fill the joint with fluid within hours less likely: a recent sprain involving the ligaments, a traumatic meniscal tear, an active joint inflammation.
It does not, however, rule out the most common causes of pain on the outer side, which almost never cause swelling:
- iliotibial band syndrome, typical of runners and cyclists, where the pain appears after a fairly predictable distance and disappears when you stop;
- an overload of the outer compartment, which gives pain under effort and not at rest;
- a pain referred from the hip, felt on the side of the knee but originating elsewhere, which is why a knee that does not respond should be looked at together with the hip;
- a pure overload, with irritated tissue but no structural damage.
In practice: without swelling there is less urgency, but not less need to understand where it comes from. Swelling becomes a signal not to postpone when it appears quickly after a trauma, when it is accompanied by heat and redness, or when the knee locks and does not fully straighten.
Right or left: does the side make a difference?
It is one of the most searched questions and the short answer is no: the right knee and the left knee hurt for the same reasons, and there is no cause that affects only one side. None of the causes listed above has a side preference written into its anatomy.
What does change, and what matters a great deal to the physiotherapist, is why it was that particular one. The side that hurts is almost always the one that has received something extra: the take-off leg if you run or jump, the support leg if you kick, the side you push off with when getting up if you have a hip or ankle that works less well, the side of the foot that lands differently. For those who cycle or drive a lot, position also comes into play: the right outer knee is the first to make itself felt when the pedal or the saddle is set up badly.
So the useful question is not "why the right and not the left", but: what does that knee do more than the other one, in a normal week? The answer is usually in the sporting gesture, in work or in a habit that nobody has ever looked at, and that is also where you intervene, because strengthening works better when you know what you are compensating for.
One exception to keep in mind: if the pain appeared on one side right after a trauma (a twist, a blow, a bad landing), then the side matters very much, and the path is that of the signs in the first section, not that of the exercises.
How the diagnosis is reached
The clinical picture guides a great deal, even before the tests.
Exactly where it hurts. In iliotibial band syndrome the point is very specific: the bony prominence on the outer side, just above the joint line. In meniscal problems the pain is lower, on the joint line. In the popliteus it is further back.
When it hurts. The outer knee pain from the band has a signature that clinicians recognise: it appears after a fairly constant amount of running, settles with rest and restarts at the next session. Many report that it gets worse going downhill and at slow paces, and the explanation given is that in those conditions the knee spends more time around 30 degrees of flexion, the angle at which compression is documented; this is, however, a clinical description and a mechanical inference, not a measured finding in the studies cited here. Meniscal pain is more linked to twisting and squatting; arthritic pain to morning stiffness.
Clinical tests. Ober's test assesses the stiffness of the tensor–band complex, Noble's test reproduces the pain by compressing the area at about 30 degrees, and the varus stress test puts the lateral collateral under tension. They are useful for orientation, but they must be taken for what they are: the systematic review on the subject notes that these tests have not been specifically validated in this patient population, so none of them alone "makes a diagnosis".
Imaging. An X-ray is useful if osteoarthritis, a fracture or an alignment problem is suspected. MRI is indicated when the picture does not add up, when there has been a significant trauma or when the symptoms do not improve: it shows the meniscus, the ligaments and the changes in the tissue under the band. It is not needed in the typical, non-traumatic case, where it risks only revealing incidental findings that frighten without changing the treatment.
What works: hip strengthening, stated honestly
This is where most articles on outer knee pain say something stronger than the data support. The standard line is: "the weak gluteus medius is the cause, strengthen it and you will heal". Let us look at the two parts separately, because they have very different solidity.
The weak gluteus medius as a cause: not demonstrated. The systematic review published in Sports Medicine concludes that aetiological studies offer limited or conflicting evidence, and that it is not clear whether weakness of the hip abductors plays an important role. A 2023 systematic review with meta-analysis in Gait & Posture included 17 articles, of which 10 cross-sectional studies entered the actual meta-analysis, and found lower isometric abductor strength compared with controls only in women runners with ongoing syndrome, not in men, and not in the groups followed prospectively, that is, before the pain appeared. The authors themselves describe this as limited biomechanical evidence, and the fact that the finding concerns those who have the pain now leaves open the possibility that the weakness is a consequence of the pain, not its cause.
Hip strengthening as a treatment: here the evidence exists, and it is more encouraging. A 2024 systematic review selected 13 studies out of 616 records, for a total of 201 participants (five randomised controlled trials, one case-control, one pre-post and six case reports). Strengthening of the hip abductors emerges as the most recurrent strategy, and the effects of the interventions range from a pain reduction between 27% and 100% and a functional improvement between 10% and 57%, over periods of 2-8 weeks. However, the limitations that the authors declare must also be read: the heterogeneity between studies was so great as to prevent a meta-analysis, the participants in total were just over two hundred, and the conclusions concern adult runners.
How do you put these two things together? Like this: we strengthen the hip not because it has been proven that its weakness causes the problem, but because it is the ingredient found in the programmes that have worked, and it is an intervention with practically zero risk. It is a more modest reason than the one you read about online, but it is the true one, and it stops you from concluding that "you have not done enough clamshells" if the pain persists.
Seven exercises for the outer knee
What the literature points to is the ingredient, hip abductor strengthening, not the precise list of exercises: the reviews do not describe a standard protocol. The movements below are the usual ways of training that control, in the frontal and transverse planes, with progression towards loading on one leg. None of them has a demonstrated "right" dose: the numbers are a reasonable starting point, not a prescription.
1. Clamshell. Lying on one side, hips and knees bent, heels together: open the upper knee without letting the pelvis roll back. The temptation to "cheat" by rotating the trunk is strong, and it removes all the point of the exercise. 15-20 repetitions per side.
2. Side-lying hip abduction. Straight leg, in line with the trunk or slightly behind, foot slightly rotated downwards: lift the leg without bringing it forward. 12-15 repetitions.
3. Single-leg bridge. Lying on your back, one foot on the floor and the other leg raised: push the pelvis upwards keeping it level. If the hip on the raised side drops, that is the sign you need to work on it. 8-12 repetitions per side.
4. Monster walk with a band. Band above the knees or at the ankles, knees slightly bent, side steps keeping the tension. 15-20 steps per direction.
5. Controlled lateral step down. Standing on a low step, slowly lower the other foot to lightly touch the floor and come back up. The goal is not the number: it is that the knee does not collapse inwards and the pelvis stays level. 8-10 repetitions per side.
6. Side plank with abduction. Side plank (on the knees if needed) and lift of the upper leg. It combines pelvic control and abduction.
7. Squat and single-leg progression. Controlled partial squat, then lunges, then single-leg squat when the pain allows. It is the bridge between the gym and running: without this step the strengthening stays disconnected from the movement that hurts you.
The practical criterion throughout the process is simple: mild discomfort that settles within a few hours and does not get worse the next day is acceptable; pain that increases during the exercise or leaves you worse the next day says you are asking for too much.
Foam roller and stretching: what they can and cannot do
Foam rolling on the band is probably the most common practice among people with pain on the outer part of the knee, and it deserves a clarification.
You are not lengthening the band. Anatomical studies show that it is a thickening of the fascia lata, connected to the linea aspera of the femur by an intermuscular septum and anchored to the supracondylar region by robust fibrous bands. A structure of this kind does not behave like an elastic to be stretched, and none of the sensations you feel on the roller corresponds to a stable lengthening of the tissue.
This does not mean that rolling is useless: many people report less tension and less discomfort straight afterwards, and it is an effect that has value when it lets you train better. It does mean two practical things. The first is that the effect is temporary and does not replace strength work. The second, more important: rolling right on the point that hurts means compressing tissue that is already compressed. If the mechanism is compression of the fat layer under the band, leaning there with the full weight of the body is the least sensible thing to do. Work instead on the tensor fasciae latae and the glute (at the top of the thigh), on the quadriceps and the vastus lateralis, skipping the bony prominence above the knee.
The same goes for "band stretching": what you really stretch are the tensor fasciae latae and the glute. Including it in your programme is fine, expecting it to solve the problem is not.
Load: the lever almost nobody touches
If compression is the mechanism, then the variable that matters most is how many times a day that compression is repeated. It is also the lever that gets used least, because it is the only one that asks you to give something up in the short term.
Reduce the volume, do not zero it. Complete rest makes the pain pass and changes nothing: when you return, at the same distance as before, it comes back. It is better to drop to an amount that does not provoke symptoms and climb back slowly.
Avoid long descents and sloping trails for a while. Descents increase the time spent in the critical flexion angle; always running in the same direction on a road with a sloped camber loads the two legs asymmetrically.
Try to increase your cadence slightly. Shorter and more frequent steps reduce the joint excursion at each landing. It is a simple change in gait to try; it should be said that on the technical aspect the 2024 review is explicit in describing gait retraining as an emerging strategy, which needs more rigorous studies before it can be recommended with confidence.
The return is gradual, not all or nothing. Alternate running and walking, restart on the flat, increase one factor at a time.
On the bike, similar principles apply, with the addition of the riding position: a saddle that is too high, cleats poorly aligned and low cadences all increase the lateral stress on the knee.
Medicines, injections, surgery
Ice and anti-inflammatories. They can make an acute phase more manageable. They do not modify the cause and do not shorten the pathway; the use of medicines should be agreed with your doctor, especially if you take others.
Cortisone injection. There is a small randomised controlled trial, conducted on 18 runners with syndrome that had begun less than two weeks earlier: 9 received 40 mg of methylprednisolone acetate plus a local anaesthetic, 9 only the anaesthetic. Compared with the control group, the treated group had a significant reduction in total pain during running from day seven to day fourteen; in the first week, by contrast, the difference between the two groups was only a non-statistically significant trend. It is a real finding, but it must be taken for what it is: eighteen people, two weeks of observation, recent-onset forms. It may help open a window in which to work, it is not the solution.
Meniscal surgery. If you have been found to have a degenerative meniscal tear, the most useful finding is a randomised trial published in the BMJ: 140 adults with a mean age of 49.5 years, with a degenerative tear of the medial meniscus confirmed on MRI and, in 96% of cases, no defined radiographic signs of osteoarthritis, were assigned to 12 weeks of supervised exercise or to arthroscopic partial meniscectomy. At two years no clinically relevant difference emerged in the KOOS4 score (0.9 points, 95% confidence interval from -4.3 to 6.1); in the exercise group thigh muscle strength had improved at three months, and 19% of those assigned to exercise nonetheless went on to surgery during the two years, with no additional benefit. Be careful not to extend this result too far: it concerns the medial meniscus and degenerative tears in adult age, not traumatic tears in young people or true joint locking, where the discussion is different.
Surgery on the band. It is an option reserved for cases that do not respond to months of well-conducted conservative treatment, and it is rarely necessary.
What to expect
In the great majority of cases, outer knee pain from the band improves without surgery: in the studies available the reduction in pain is consistent, but timelines vary a great deal and the research covers windows of 2-8 weeks, so be wary of anyone who promises you a precise date for returning.
The point to take home is this: it is not a structure to be "loosened", it is a tissue that is compressed too many times. The two levers that count are load management and building hip control that holds up the movement, not the roller, not the injection, not waiting. And if after a few weeks of sensible work the picture does not move, the right question is not "should I push harder", but "was the diagnosis correct?" If you run and want to look further into managing your overall load, you will find more in the article on knee pain in running.
If you are living with this problem and want a programme built around your situation, the physiotherapists at Ri-Hub work through teleconsultation: assessment, progressive exercises and periodic review of the load, from your own home.
You might also like:
Good Posture: A Complete Guide to Feeling Well from Head to Toe Postpartum Back Pain: Exercises and Remedies to Get Back in Shape Ankle Sprain: Recovery Times and Full Rehabilitation
Stay in touch with us Join our dedicated Facebook group: we answer all patients, give practical advice and clear up the most common doubts.
Sources
- The functional anatomy of the iliotibial band during flexion and extension of the knee: implications for understanding iliotibial band syndrome, Journal of Anatomy, 2006
- Is iliotibial band syndrome really a friction syndrome? Journal of Science and Medicine in Sport, 2007
- Iliotibial band syndrome in runners: a systematic review, Sports Medicine, 2012
- Lower extremity kinematics during running and hip abductor strength in iliotibial band syndrome: A systematic review and meta-analysis, Gait & Posture, 2023
- Effects of conservative treatment strategies for iliotibial band syndrome on pain and function in runners: a systematic review, Frontiers in Sports and Active Living, 2024
- Local corticosteroid injection in iliotibial band friction syndrome in runners: a randomised controlled trial, British Journal of Sports Medicine, 2004
- Exercise therapy versus arthroscopic partial meniscectomy for degenerative meniscal tear in middle aged patients: randomised controlled trial with two year follow-up, BMJ, 2016
- The accuracy of the Ottawa knee rule to rule out knee fractures: a systematic review, Annals of Internal Medicine, 2004
Last source check: August 2026.
Need a consultation?
Start from the questionnaire on Ri-Hub. A few questions about your case, you register, and see who will follow you: a physiotherapist registered with the Albo, whom you write to in chat whenever you like. If you prefer to talk it through by voice, the first consultation is free, lasts twenty minutes and commits you to nothing: they arrive having already read what you wrote, and together you decide whether and how to proceed.
