Plica syndrome knee exercises and physiotherapy

What is a plica, and what is plica syndrome of the knee?

A plica is a fold in the lining of the knee joint, left over from how the knee forms before birth. Most knees have at least one. The one that causes trouble is usually the medial patellar plica, a shelf of tissue running down the inner side of the kneecap, and Kuwabara and Fredericson note that medial plicae are both the most common and the most often symptomatic (Kuwabara 2021).

Plica syndrome is the name for pain coming from one of those folds when it becomes thickened or irritated and starts catching between the kneecap and the thigh bone. The pattern described in the literature is pain localized to the fold with snapping, clicking, catching, clunking, grinding or popping inside the knee as it bends and straightens, brought on by repeated bending and straightening or by long spells with the knee held bent, often after a recent increase in activity or a twisting injury (Kuwabara 2021).

Numbers matter here more than usual. In 3,889 knee joints examined during arthroscopic surgery, a medial plica was present in 79.9%, with the four Sakakibara types at 35.2%, 22.4%, 12.3% and 10.0% (Nakayama 2011). Anatomic studies put the figure at 90% of adults having one or more, the vast majority of them incidental (Kuwabara 2021). Having a plica is normal. Having a problem from one is not the same thing, and nothing on a scan report settles which you are.

Is it a plica or a meniscus tear?

Both can catch, both hurt on the inner side, and both turn up in the same age group, so this is the differential that matters most. Kuwabara and Fredericson say plica syndromes are often missed because they are difficult to distinguish from meniscal tears, cartilage injuries and osteochondritic lesions (Kuwabara 2021).

The sharpest distinction we read comes from the randomized trial's own entry criteria. It took people with focal, locatable pain on the inner side of the front of the knee, and excluded anyone with concentric, diffuse pain around the kneecap because of the overlap with kneecap joint problems (Sauer 2022). Focal against diffuse is something you can judge about your own knee: can you put one finger on it, or does it spread around the kneecap?

What points away from a plica is a clear twisting injury with swelling over the following hours, a knee that locks and will not straighten, or giving way under load. Read meniscus tear if that fits, and patellofemoral pain if the ache is spread around or behind the kneecap. Pain a little lower and further round the inner side of the knee, with tenderness on the bone's inner edge, can be pes anserine bursitis. Our pages on inner knee pain and knee clicking and cracking work through the wider lists, and knee giving way covers that symptom on its own. Two of the other folds and pads at the front of the knee have their own pages too, fat pad impingement and quadriceps tendinopathy.

How contested is this diagnosis?

Very, and in four separate ways. You should know all four before you accept or reject the name, because a page that presented this as settled would be wrong and a page that dismissed it would be wrong too.

First, the structure is nearly universal and almost always harmless, which the figures above show. Second, there are no agreed diagnostic criteria. The authors of the randomized trial write that while the potential of medial plicae to cause problems is widely agreed, the incidence of plica syndrome is unknown mostly because diagnostic criteria are lacking, and that estimates of how often it occurs, from small arthroscopic series of people with acute knee pain, range from 3 to 30% (Sauer 2022). They add something sharper in their limitations: it is a diagnosis of exclusion, people with it are rarely referred under that name, and people referred with plica syndrome as the working diagnosis rarely turned out to have it.

Third, even the reference standard is disputed. Stubbings and Smith recommend that future research should use resolution of symptoms after resection of the fold as the standard, instead of what is seen at arthroscopy, while the trial's authors note that arthroscopic appearance remains the gold standard and that their own interpretation of a fold was subjective rather than confirmed under a microscope (Stubbings and Smith 2014, Sauer 2022).

Fourth, the surgeons who treat it do not all take it seriously, and somebody measured that. A survey of the German Knee Society, answered by 238 of 1,065 members with a mean age of 46.5 and 85.3% performing more than 51 arthroscopies a year, found 84% (200 of 238) agreed that plica syndrome or a symptomatic plica can be a partial or entire cause of pain at the front of the knee. In the same survey 105 (44.1%) regularly exchange information about it with colleagues, 99 (41.6%) would prefer to discuss it more often, and 34 (14.3%) would rather not talk to colleagues about the topic at all (Faber 2025). Four in five experienced knee surgeons accept the fold can cause pain, there is no agreed way to diagnose it, and one in seven would rather not discuss it. That is the state of play.

What did the one randomized trial find?

Sauer and colleagues randomized 48 people with isolated, MRI-confirmed, clinically diagnosed medial plica syndrome to arthroscopic resection of the fold or structured physiotherapy, across four hospitals between 2014 and 2017 (Sauer 2022). Lysholm scores at the start were 65.8 and 66.3.

At 3 months the scores were 80.7 after surgery and 74.9 after physiotherapy, with no significant difference. At 6 months they were 82.3 and 83.1, again with no significant difference, and both groups had improved by more than the 10.1 points the authors treated as the smallest change that matters to a patient. At 1 year the figures were 88.3 and 71.1, not significantly different, on only 21 patients because of a problem inviting people to that visit. At 2 years they were 89.7 and 74.6, a significant difference, on 33 patients, a follow-up rate of 68%.

Read the authors' explanation before you read the 2-year gap as surgery winning. They write that the positive effect of physiotherapy is plausibly tied to continuing to do it, and that deterioration is likely once the disciplined execution of the training protocol is abandoned. In their limitations they record that, although it was not systematically measured, the majority of patients said they had abandoned the protocol after 3 to 6 months.

That is the most actionable sentence in this condition's literature, so here it is without hedging: the exercises matched an operation at six months, and the people whose results slipped were the people who stopped. The program below is not a six-week course.

How to use this program

Start at stage 1 if the knee is catching and irritable. Move up when the current stage feels easy and the knee is no worse the next morning. The trial ran weekly physiotherapy review for 4 weeks and then had people continue on their own, which is a sensible model to copy if you can get a few appointments.

Each exercise page gives a typical starting dose. As a guide, quad sets are often 10 repetitions with a 5 to 10 second hold several times a day, straight leg raises 2 to 3 sets of 10 on each leg, hamstring sets about 10 holds of 5 to 10 seconds, and the stretches 20 to 30 seconds held 2 to 4 times. The standing work in stages 2 and 3 is usually 2 to 3 sets of 8 to 15. Your physio will adjust this.

Some muscle ache afterward is expected and settles by the next day. What should not happen is the catch getting more frequent or the focal pain being worse the next morning, and if that happens, go back a stage for a week rather than stopping altogether.

Kuwabara and Fredericson suggest giving conservative treatment at least 3 months before anything more aggressive is considered, which fits both the trial and the 90-day program in the series from Camanho (Kuwabara 2021, Camanho 2010). After knee surgery, follow your surgical team's plan rather than this page. If you are pregnant, check with your midwife or physio before you start. A child or teenager should be assessed before starting: one case series of symptomatic medial plicae had a mean age of 16.2, so it is not just an adult problem (Hufeland 2019).

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.

Keeping moving between sessions

The trial's physiotherapy arm included a leg press, a stationary bike and an elliptical machine alongside the exercises above (Sauer 2022). We have no exercise page for any of the three, so they are worth mentioning here rather than leaving out.

If you have access to a bike or an elliptical, they are a reasonable way to keep the knee moving and the thigh working between strength sessions. Set the saddle high enough that the knee does not fold up sharply at the top of the pedal stroke, because holding the knee deeply bent for long spells is one of the things the reviews name as a provoker. Start with a short spin rather than a long one and judge it by the next morning.

Does conservative treatment work, and why do the sources disagree?

Three readings of the same question sit in the literature, and at first glance they do not reconcile. The randomized trial found physiotherapy equal to surgery at six months, with both groups beating the threshold for a meaningful change (Sauer 2022). Camanho reported 63 patients given 90 days of strengthening, flexibility work and activity modification, of whom 55 improved, 6 of those later had symptoms return that were never disabling, and 8 went on to arthroscopy (Camanho 2010). Schindler's review of the field, however, concludes that symptomatic plicae may initially be treated with physiotherapy and structured exercise but that success rates are generally low (Schindler 2014).

We read all three to see whether one of them settles it, and the useful finding is that they are not quite answering the same question.

Schindler's paper pooled 23 studies of plica excision, so the meta-analysis in it measured surgery, not exercise. The sentence about conservative success rates being generally low is the author's conclusion drawn from the wider literature, not a pooled result, and the same conclusion goes on to reserve excision for recalcitrant cases and for a fold that has undergone irreversible change. Camanho's series was not restricted to this fold. The commonest complaint in those 63 patients was pain above the kneecap, and the examination finding was pain on compressing the pouch above the kneecap, which points higher in the joint than the medial fold this page is about. The paper never reports how many of the 63 were medial, so it is a series of mixed folds rather than a comparator for this one, and its program was built around terminal knee extension. Sauer's trial is the one that studied the medial fold head to head, and it is the only randomized one.

So the honest summary is this. Exercise is the reasonable first step and all three sources say so. For the medial fold at six months, a trial found it matched an operation. What nobody has established is the long-run success rate of conservative care, and we are not going to invent one by picking a favorite source.

Severity is probably part of why the figures scatter, and one study gives you something to act on. In a series of 117 patients all given 3 months of conservative treatment, 76 did not improve and went to surgery. The ones who failed had a larger fold on MRI, 1.6 mm on average against 0.8 mm, more contact between the fold and the cartilage, more pain on the inner side as the knee moved from bending to straightening, and more snapping (Blanke 2021). Two of those four are measurements you cannot take at home. The other two you can report, and they are a reason to be reassessed rather than to keep grinding away alone. Worth knowing as well: the 41 who did get better without surgery ended up with scores as good as the surgical group's, so succeeding with exercise is not a second-best outcome.

What the operation is, and when it comes up

Arthroscopic resection means trimming the fold out through two small cuts, usually as a day case. Schindler's review pooled 23 studies and 969 patients after open or arthroscopic excision, mean age 25, with trauma considered the cause in 57% and a mean follow-up of 27.5 months: 64% were symptom free, 26% improved and 10% were failures (Schindler 2014). At 10 years, a series of 52 knees in 50 patients reported Lysholm scores rising from 52 to 80, with worse results where cartilage damage was present as well (Paczesny 2021).

Those are reasonable numbers and they are not a reason to skip the exercises, because the trial compared the two head to head and found no difference at six months. We do not set rehabilitation after this operation. If you are having it or have had it, the timings for weight bearing, range of movement and return to sport come from your surgical team.

When to see a physio or doctor

See a physio or doctor if the knee keeps catching, if focal pain on the inner side has not settled after a few weeks of changing what provokes it, or if you are not sure whether this is the fold, the meniscus or the kneecap joint. That last one is worth an appointment on its own, because the trial's authors found this diagnosis was wrong in both directions more often than not. Given that, an examination is worth more than another scan.

If you have done a properly supervised program for 3 months and the pain and snapping are unchanged, that is the point at which the literature starts talking about surgery, and it is a fair thing to raise.

For any of the warning signs below, get help within the time given there.

For physiotherapists

Sauer and colleagues 2022 is the only randomized trial. Its physiotherapy arm was quadriceps strengthening by quadriceps set, straight leg raise, leg press, stationary bike and elliptical, a frequent stretching program for hamstrings, adductors, gastrocnemius and quadriceps, avoidance of activity inducing plica irritation, and weekly physiotherapist review for 4 weeks followed by self-directed work. Note the tension in their own wording: they list a quadriceps set and a straight leg raise, both open chain, and state that open chain exercises that eventually cause plica irritation were avoided. The qualifier may resolve it, and the paragraph is worth reading before you build a protocol on it.

For the examination, Kuwabara and Fredericson describe the Stutter and Hughston tests and require both to be positive. Stubbings and Smith 2014 give the accuracy figures and argue for changing the reference standard away from arthroscopic appearance. Faber 2025 is the survey quantifying how divided the surgical community is.

Blanke 2021 gives the predictors of conservative failure, two radiological and two clinical, and also reports that the 41 of 117 who succeeded conservatively finished with a Lysholm of 96.25 against 95.93 in the operated group, with no difference in return to sport or Tegner. Hufeland 2019 is a prospective case series of 35 patients and 38 knees, mean age 16.2 and 73.7% female, with no control group, which the authors state.

On the conservative success disagreement, which earlier drafts carried as three irreconcilable readings: clinical review read all three and the scatter is partly a scope difference, not only an evidence one. Schindler's meta-analysis pooled 23 excision studies, so the "success rates are generally low" sentence is his conclusion from the wider literature rather than a pooled estimate, and it sits beside a recommendation to try physiotherapy first. Camanho's 63 patients presented mainly with suprapatellar pain and were examined by compression of the suprapatellar pouch, so that series is weighted toward the folds above the kneecap. It reports no breakdown by fold type at all, which is why clinical review 2 took out the earlier claim that most of the 63 were not medial: the paper does not support it, and Camanho's own terminology follows Iino's supra-, mid- and infrapatellar scheme, in which the midpatellar fold is the medial patellar plica. The protocol there was built on terminal knee extension, which this program does not copy because the trial's was not. Sauer is the only randomized trial and the only one of the three restricted to the medial fold. The page now says exercise first, matched surgery at six months for the medial fold, and no established long-run conservative success rate, rather than presenting the three as a stand-off.

Common questions

References

1. Sauer S, Karlsen G, Miller L, Storm JO. Medial Plica Syndrome of the Knee: Arthroscopic Plica Resection versus Structured Physiotherapy, A Randomized Controlled Trial. The Surgery Journal (New York). 2022;8(3):e249-e256. https://doi.org/10.1055/s-0042-1756183
2. Schindler OS. 'The Sneaky Plica' revisited: morphology, pathophysiology and treatment of synovial plicae of the knee. Knee Surgery, Sports Traumatology, Arthroscopy. 2014;22(2):247-262. https://doi.org/10.1007/s00167-013-2368-4
3. Nakayama A, Sugita T, Aizawa T, Takahashi A, Honma T. Incidence of medial plica in 3,889 knee joints in the Japanese population. Arthroscopy. 2011;27(11):1523-1527. https://doi.org/10.1016/j.arthro.2011.06.022
4. Stubbings N, Smith T. Diagnostic test accuracy of clinical and radiological assessments for medial patella plica syndrome: a systematic review and meta-analysis. The Knee. 2014;21(2):486-490. https://doi.org/10.1016/j.knee.2013.11.001
5. Faber S, Winkler PW, Henkelmann R, Diermeier T, Petersen W, Balke M, Metzlaff S, Colcuc S, Zimmermann G, Roessler PP, Niethammer TR. Treatment of a medial plica in the knee among German knee surgeons, The Plica Survey. Asia-Pacific Journal of Sports Medicine, Arthroscopy, Rehabilitation and Technology. 2025;40:18-22. https://doi.org/10.1016/j.asmart.2025.01.003
6. Camanho GL. Treatment of pathological synovial plicae of the knee. Clinics (Sao Paulo). 2010;65(3):247-250. https://doi.org/10.1590/S1807-59322010000300002
7. Blanke F, Oehler N, Al Aidarous H, Tischer T, Vogt S, Lenz R. Predictors for an unsuccessful conservative treatment of patients with medial patellar plica syndrome. Archives of Orthopaedic and Trauma Surgery. 2021;141(1):93-98. https://doi.org/10.1007/s00402-020-03646-6
8. Hufeland M, Treder L, Kubo HK, Verde PE, Krauspe R, Patzer T. Symptomatic medial synovial plica of the knee joint: an underestimated pathology in young patients. Archives of Orthopaedic and Trauma Surgery. 2019;139(11):1625-1631. https://doi.org/10.1007/s00402-019-03249-w
9. Paczesny L, Zabrzynski J, Kentzer R, Gryckiewicz S, Lewandowski B, Szwedowski D, Kruczynski J. A 10-Year Follow-up on Arthroscopic Medial Plica Syndrome Treatments with Special Reference to Related Cartilage Injuries. Cartilage. 2021;13(1_suppl):974S-983S. https://doi.org/10.1177/1947603519892310
10. Kuwabara A, Fredericson M. Narrative: Review of Anterior Knee Pain Differential Diagnosis (Other than Patellofemoral Pain). Current Reviews in Musculoskeletal Medicine. 2021;14(3):232-238. https://doi.org/10.1007/s12178-021-09704-9

Written and checked by the PocketPhysio editorial team. Last updated 2026-11-06.

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