Cartilage · Knee · Lower limb

Articular cartilage lesions of the knee: a focal lesion is not osteoarthritis beginning

A cartilage defect seen on MRI can be three very different things: an injury in a young person, wear beginning, or nothing at all. One asymptomatic knee in four carries one. What separates those three readings cannot be read on the image, and the whole of the treatment that follows depends on it.

Focal lesion or wear beginning: where does the line run?

This is the question that governs everything else, and it is the one imaging does not settle. Two series, one arthroscopic and one by MRI in healthy knees, together give the markers that allow a decision.

What 25,124 arthroscopies counted

Widuchowski and colleagues analysed every knee arthroscopy performed in their centre from 1989 to 2004, that is 25,124 procedures, classifying cartilage lesions by Outerbridge¹.

Cartilage lesions are found in 60 % of patients. Distributed as follows: focal osteochondral or chondral lesion 67 %, osteoarthritis 29 %, osteochondritis dissecans 2 %, other 1 %. Outerbridge grade II is the most frequent (42 %). The dominant sites are the patellar surface (36 %) and the medial femoral condyle (34 %).

Two figures from this series deserve to be remembered above the others, because they frame the real place of repair surgery. First, 70 % of lesions are not isolated : they accompany another lesion, most often a medial meniscal tear (37 %) or an anterior cruciate ligament injury (36 %). Second, the patients who meet the criteria for repair surgery, that is one to three focal grade III or IV lesions, are only 7 % of the series before the age of 40 and 9 % before 50¹.

The authors themselves conclude cautiously: these patients form a heterogeneous group, and the natural history of cartilage lesions remains unknown, so that the exact number of patients who would benefit from repair cannot be established.

The grade, and what it does not say

Two classifications are in circulation, and it is useful to know which one you are reading. Widuchowski's series uses Outerbridge, the older one, in which grade II is the most represented (42 % of lesions)¹. More recent work uses the classification of the International Cartilage Regeneration and Joint Preservation Society, ICRS, which grades from simple superficial fissuring to exposure of the subchondral bone.

These grades serve to describe a lesion and to establish an indication: repair is discussed for deep lesions, grade III or IV, not for superficial softening. But it is worth knowing from now on what they do not do, because it has been measured and the next chapter returns to it: the ICRS grade does not affect the scores patients report before surgery³. A grade is a description of a joint surface, not a prognosis of how troubled someone will be.

What 5,397 healthy knees show on MRI

The decisive counterpart comes from Culvenor and colleagues, who looked for the prevalence of osteoarthritis features on MRI in asymptomatic and uninjuredadults. Six databases, 63 studies included, 5,397 knees in 4,751 adults, with random-effects meta-analytic estimates².

Overall prevalence of cartilage defects: 24 % (95 % CI 15 to 34). Stratified by age: 11 % before the age of 40 (6 to 17) and 43 % at 40 and over (29 to 57). Meniscal tears follow the same slope: 4 % before 40, 19 % after. Bone marrow lesions affect 18 % of knees, osteophytes 25 %².

The authors' conclusion is an instruction for reading: these images must be interpreted in the context of the clinical presentation, and taken into account as such in the decision.

The table that separates the two situations

From these two series, and from what the patellar dislocation cohort in the next chapter establishes, a grid can be drawn. It rests on no validated test, and that has to be said: it is a synthesis of clinical markers, not a score.

Where do traumatic lesions come from, and in whom should you look for them?

The best-documented mechanism is patellar dislocation, because a consecutive cohort imaged it systematically in the subacute phase. The figures that come out of it change what is done after a first episode.

Patellar dislocation, the best-measured mechanism

Isacsson and colleagues used a consecutive cohort of 1,145 injured knees examined by MRI in a centre with a defined population catchment, to extract the primary lateral patellar dislocations. One hundred and eighty-four cases identified, MRI performed at a median of six days⁴.

The profile is that of an adolescent: median age 16 (interquartile range 14 to 21, range 9 to 47), 41 % women. Overall annual incidence is 14 per 100,000, with a peak between 13 and 15 at 125 per 100,000. Men are older than women at the first episode (17 against 15) and dislocate more often during sport (65 % against 40 %)⁴.

The result that should change practice is this: articular surface lesions are visible on MRI or found at surgery in 75 of the 184 knees, that is 43 %, and they are osteochondral fractures in 32 knees, that is 18 %. The risk does not differ significantly according to whether the dislocation happened during sport or in daily life⁴.

The practical consequence

Faced with a first patellar dislocation in an adolescent, the question is not only future instability: it is whether an osteochondral fracture is present, because a loose fragment or a fixable fragment changes the decision and the timetable. Almost one dislocation in five carries one, and the clinical picture alone does not distinguish them.

Osteochondritis dissecans, the other route

Osteochondritis dissecans is the second main route to an osteochondral lesion in young people, and it follows a different logic. The review by Chau and colleagues, published in the Journal of Bone and Joint Surgery, sets out the frame: it mainly affects active children and young adults, sits at the knee, the elbow or the ankle, and can lead to early osteoarthritis⁵.

The decision hinges on two parameters: skeletal maturity and the stability of the lesion, the latter defined by the presence or absence of a fracture of the articular cartilage and of separation of the subchondral bone, assessed by imaging and arthroscopy. Treatment is typically non-operative for stable lesions in skeletally immature patients, and operative if conservative treatment fails or the lesion is unstable⁵.

The authors note that the origin remains debated (local ischaemia, aberrant endochondral ossification of a secondary subarticular physis, repeated microtrauma, genetic predisposition) and above all that clinical practice guidelines are limited by the scarcity of high-level evidence. A multicentre effort is under way to build reliable classifications and decision algorithms with prognostic value.

One link is worth making here with what the site already documents: two adolescents who developed osteochondritis dissecans following a bone bruise of a femoral condyle are reported in our article on bone bruise. Two cases do not make a frequency, but they justify monitoring a skeletally immature patient after a condylar impaction.

Why does the size of the lesion not tell you how troubled the patient is?

This is one of the most useful results on the subject, and one of those that best protect the patient from an anxiety-inducing account. The link between what the image shows and what the person experiences is weak.

What 90 consecutive patients establish

Randsborg, Årøen and Owesen prospectively collected the preoperative data of 90 consecutive patients scheduled for surgery on an isolated symptomatic cartilage lesion, mean age 33.2, 69 % men, then compared them with those of patients scheduled for ligament reconstruction and for knee replacement³.

First result. There is no significant difference in scores between patients whose lesion measures less than 2 cm² and those whose lesion measures more, with one exception: the KOOS “symptoms” subscale, better in small lesions (62.8 against 51.9, p = 0.005). The correlation between lesion size and Lysholm score exists but is weak, and it disappears after adjustment for age, sex, body mass index and location³.

Second result. The ICRS grade does not affect preoperative scores. Neither depth, nor size, nor location has any notable impact on the symptoms experienced³.

Third result, and the most striking. These patients report symptoms more severe than candidates for ligament reconstruction, and approaching those of candidates for knee replacement³.

What this changes in the consultation

Three direct consequences.

The first is a duty of caution in what is said. Announcing a size in square centimetres or an ICRS grade as though it were a prognosis has no basis: those parameters do not predict what the patient feels today. They have value for establishing a surgical indication, not for predicting an experience.

The second is a duty of recognition. The other face of the same result is that these patients are often far worse off than imagined, up to a level of symptoms close to candidates for a joint replacement. A young adult with a 1 cm² lesion can be severely troubled, and there is nothing exaggerated about it.

The third concerns follow-up. If size does not predict how troubled the patient is, then follow-up is done on functional measures and on patient-reported scores, not on repeated images.

What are the repair procedures actually worth?

The physiotherapist does not set the indication, but receives these patients before and after, and is the first to be asked what the operation will bring. The recent literature is more sober than the reputation of the techniques.

The 2025 trial: microfracture is not superior to debridement

This is the most important result of recent years on the subject, and it comes from a multicentre double-blind randomised trial, level 1. Randsborg and colleagues included 65 patients aged 18 to 50 with an isolated symptomatic femoral cartilage lesion under 2 cm², randomised between microfracture (31) and simple arthroscopic debridement (34), followed for two years. Mean lesion size: 1.2 ± 0.6 cm²⁶.

The primary outcome, the change in KOOS Quality of Life at two years, does not differ: 3.5 points (95 % CI −10.0 to 16.9), p = 0.61. No secondary outcome differs either, at any follow-up point, on a linear mixed model. The complications, however, are not symmetrical: 10 complications in 5 patients in the microfracture arm, against 2 in 2 patients in the debridement arm. The authors' conclusion, word for word: microfracture was not superior to debridement for femoral lesions under 2 cm²⁶.

Two reservations should accompany the result, without weakening it. A sample of 65 patients leaves a wide confidence interval, which does not exclude a moderate benefit from microfracture. And the scope is narrow: femoral lesions under 2 cm², in adults aged 18 to 50. Nothing transfers to large lesions.

Follow-up at fourteen to fifteen years

Knutsen and colleagues followed 80 patients with a chronic, isolated, symptomatic femoral cartilage lesion, without generalised osteoarthritis, enrolled between January 1999 and February 2000 and randomised between autologous chondrocyte implantation and microfracture. They had published the results at 2 years and then at 5 years; in 2016 they published the follow-up at 14 to 15 years, with standardised radiographs⁷.

No significant clinical difference between the two groups. 17 failures in the implantation group against 13 in microfracture, and more total replacements in the implantation group (6 against 3). Patients who had not failed retain a significant improvement over their starting state, but 57 % of the survivors in the implantation group and 48 % of those in the microfracture group have radiographic osteoarthritis (Kellgren-Lawrence 2 or more), a non-significant difference⁷.

The authors' conclusion is unusually blunt for a surgical paper: their results raise serious concerns about the effectiveness of these procedures in delaying osteoarthritis and avoiding further surgery.

The durability of microfracture, measured twice

The systematic review by Mithoefer and colleagues, 28 studies and 3,122 patients, mean follow-up of 41 months with only 5 studies at 5 years or more, gives the picture: microfracture improves knee function in all studies during the first 24 months, but the data on the durability of that gain are contradictory. Defect fill on MRI is highly variable and correlates with the functional result. The limitations noted are the poorly hyaline nature of the repair tissue, the variable volume of cartilage produced, and possible functional deterioration⁸.

Two long-term randomised trials have since measured that deterioration. Volz and colleagues randomised 47 patients between microfracture, sutured AMIC and glued AMIC, and followed them for ten years : all three arms improve over the first two years, then the microfracture arm deteriorates progressively and significantly, while the two AMIC arms remain stable. The MOCART scores are comparable between groups⁹.

Gudas and colleagues randomised 60 athletes of mean age 24.3 between mosaicplasty and microfracture, with a mean follow-up of 10.4 years. Both techniques stay above the preoperative level but decline between 3 and 10 years. At ten years: 15 failures (26 %), of which 4 (14 %) after mosaicplasty and 11 (38 %) after microfracture. Maintenance of the previous level of physical activity is 75 % after mosaicplasty against 37 % after microfracture. Athletes under 25 at surgery keep better scores¹⁰.

What supports each decision, from the most solid to the emptiest

After repair: why the time to weight bearing governs everything

This is the chapter where the physiotherapist has most purchase, and where the literature is, for once, clear. Three successive randomised trials shortened the time to full weight bearing without ever damaging the graft.

The starting point: nobody knew

Ebert and colleagues open their 2012 trial with a statement less than fifteen years old: although structured postoperative rehabilitation after chondrocyte implantation is considered critical, very little data has been published on how to progressively increase weight bearing and exercise after surgery¹². The protocols in use came from caution, not from measurement.

First trial: 11 weeks against 8

The pivotal trial randomised 62 patients after matrix-induced chondrocyte implantation at the medial or lateral femoral condyle. Both protocols protect the implant for an initial period then increase load in stages; the accelerated group reaches full weight bearing at 8 weeks, the traditional group at 11¹¹.

At three months, the accelerated group achieves better distances on the 6-minute walk test and higher daily activity measured by accelerometry. It reports a significantly better improvement in KOOS pain at 12 weeks. And above all: no patient, whatever the protocol, had an adverse effect on the implant on the three-month MRI.

The biomechanical result is the most telling. Compared with a healthy control group, the knee adduction and flexion moments of the traditional group differ significantly from unaffected subjects, whereas those of the accelerated group do not. In other words, less time spent walking with crutches translates into a faster return to normal gait¹¹.

The same trial at five years

Sixty-three of the 70 patients recruited were reviewed clinically at five years, 58 with high-resolution MRI. All clinical and MRI scores improve significantly over the period. The accelerated group reports significantly less frequent pain on the VAS at five years ; no other difference between the two groups, clinical or radiological¹².

Two observations from this trial deserve to be remembered beyond the question of weight bearing. Patient age and defect size correlate negatively with several MRI scores at five years. And there is no significant correlation between clinical scores and MRI scores : what the image of the graft shows does not say what the patient feels, exactly as preoperatively. At five years, 94 % of patients are satisfied with the relief of their pain and 95 % with their ability to carry out their daily activities¹².

The authors' conclusion is worth quoting: the accelerated protocol is safe and effective, and provides clinical results comparable if not superior throughout follow-up.

Second trial: 8 weeks against 6

Edwards, Ackland and Ebert then tested a further shortening in the tibiofemoral joint. Twenty-eight knees assessed at twelve months, full weight bearing at 6 weeks against 8 weeks, those 8 weeks being explicitly presented as the “current best practice” arising from the earlier work¹⁴.

Both groups improve significantly on all measures. The accelerated group obtains better SF-36 physical scores at 8 weeks and a higher KOOS quality of life at 6 and 12 months. It regains full active knee extension by 4 weeks, against 12 weeks for the other group. No difference in graft quality on MRI (composite MOCART score 3.34 against 3.04), and no patient had an adverse effect on the implant up to twelve months, whatever the protocol¹⁴.

And the five-year follow-up of the six-week protocol

Ebert and colleagues finally followed 35 patients (37 knees) randomised between a return to full weight bearing at 6 weeks (18) and at 8 weeks (19), with a minimum follow-up of five years (5.5 to 7 years), including isokinetic strength and MRI¹³.

No difference between the groups, on any measure. Significant improvement in all reported scores, in 6-minute walk distance, in active flexion and extension range, and in the extensor strength symmetry index. Overall satisfaction: 87.5 % at 6 weeks and 82.4 % at 8 weeks. The composite MRI score declines non-significantly between one year and the final review, with no difference between groups. Two grafts failing on MRI, one patient referred for arthroplasty: 8.1 % failure at a minimum of five years¹³.

The authors conclude that a return to full weight bearing in six weeks gives comparable clinical and MRI results beyond five years without compromising the graft, and that this protocol is faster than any proposed and studied so far.

What these trials allow, and what they do not

What they allow. After matrix-induced chondrocyte implantation at the femoral condyle in adults, a progression reaching full weight bearing in six weeks is supported by three randomised trials, with no deterioration of the graft on MRI at three months, twelve months and five years, with less pain, active extension regained earlier and gait normalised faster.

What they do not allow. Three limitations must accompany that result wherever it is used.

• The scope is matrix-induced chondrocyte implantation, at the femoral condyle, in adults. None of these trials concerns microfracture, nor mosaicplasty, nor AMIC. Transferring the timetable to a microfracture is reasoning, not evidence, and that must be said to the patient as to the surgeon.
• The samples are modest and all come from the same Australian team: 62, then 70, then 28, then 37 knees. The consistency of the series is an argument; it does not replace independent replication.
• The protocol does not come down to a date. In both arms of every trial, the implant is protected for an initial period, then load increases in stages. What was shortened is the length of the progression, not the principle of progression.

When can sport be resumed, and at what level?

Two systematic reviews, sixteen years apart, give a consistent answer and a ranking between techniques that the patient has a right to know before being operated on.

What the first review had established

Mithoefer and colleagues pooled 20 studies and 1,363 patients, mean follow-up 42 months. Return to sport is possible for 73 %, with the highest rates after osteochondral grafting. The time varies from 7 to 18 months depending on the technique. An initial return to the pre-injury level concerns 68 % of patients and does not vary significantly between techniques; the maintenance of that level over time concerns 65 %, with the best durability after chondrocyte implantation¹⁵.

The factors influencing that return are identified: the athlete's age, the duration of symptoms before surgery, the level of practice, the size of the lesion and the morphology of the repair tissue. The duration of preoperative symptoms is the only one of those factors that the care pathway has any hold on, and it is an argument for not letting the diagnosis drag.

What the 2025 review adds

Kunze and colleagues took the question up again across 52 studies and 2,387 patients, with random-effects meta-analyses of proportions. Overall return to sport is 80,3 % (95 % CI 73.3 to 86.5)¹⁶.

The ranking between techniques is the new result. For a return to the same level or higher : matrix-induced chondrocyte implantation OR 2.15, osteochondral grafting OR 1.83. Microfracture is the only technique associated with a return to a LOWER level, with an OR of 0.78. The fastest return is obtained after osteochondral grafting: 6.6 ± 2.6 months on average¹⁶.

The authors themselves note a bias that has to be repeated: all the professional athletes in their sample had had a microfracture, which partly confounds level of practice with technique. The association between microfracture and a return to a lower level is nevertheless also found in amateurs. The overall level of evidence of this review is 4.

What place for the physiotherapist before, and without, surgery?

Here we have to be frank about what is missing. All the rehabilitation literature cited in this article concerns the aftermath of surgery. That does not mean there is nothing to be done before, or without.

The gap in the literature, and why it has to be named

There is a striking asymmetry in this file, and the clinician feels it every day without always identifying it. The question “how do you return to weight bearing after a graft” has received three randomised trials. The question “what becomes of a focal cartilage lesion that is not operated on” has received none in this base, and the authors of the largest published arthroscopic series write in black and white that the natural history of cartilage lesions remains unknown¹.

That gap has a direct consequence for what can be promised. Nobody can tell a 30-year-old patient with a symptomatic focal lesion of 1.5 cm² what will become of them in ten years if they are not operated on. We do know, on the other hand, what becomes of them if they are, and it is not triumphant: at fifteen years, about a third of failures and half the survivors with radiographic osteoarthritis, whatever the technique⁷.

The right way to present that to a patient is neither to dramatise nor to give false reassurance: it is to say that there are two unknowns, and that the decision is made on current trouble and on the plan, not on a prediction nobody can make.

What the source base does not contain

None of the sixteen sources in this article establishes that a rehabilitation programme on its own treats a focal cartilage lesion, or that it alters its structural course. Nor is there, in this base, any natural history data on the unoperated focal lesion. That is a gap, and it must be announced rather than filled in with assertions.

What the data nevertheless allow us to hold

Do not treat an image. That is the most direct consequence of the 24 % of cartilage defects in asymptomatic people². A defect discovered incidentally in a patient over 40 is read as a sign of age, not as an indication.

Take the patient's trouble seriously, whatever the size of the lesion. The mirror image of the previous point: when a patient is troubled, they often are very much so, up to symptom levels close to those awaiting a joint replacement³. Playing it down because “the lesion is small” has no basis.

Treat what accompanies the lesion. Seven lesions in ten are not isolated¹ : the medial meniscus and the anterior cruciate ligament accompany them one time in three each. The associated injury is often the one for which rehabilitation data exist, and often the one producing part of the trouble. Management of the reconstructed ACL is set out in our dedicated article.

Shorten the time before the decision. The duration of preoperative symptoms is one of the factors influencing return to sport¹⁵. In a young person with a focal picture and a dated mechanism, letting six months of waiting rehabilitation drag on without reassessment is a choice that has a cost.

Prepare for surgery when it is decided on. Strength, range, motor control and understanding of the postoperative timetable: the patient who arrives in theatre with a maintained quadriceps and full extension is not starting from the same place.

Strengthening when load is limited

The postoperative phase poses a precise mechanical problem: during the six to eleven weeks of progressive loading, the quadriceps wastes, and it is exactly the quadriceps that will be needed to return. That is the situation in which strengthening under blood flow restriction finds its indication in principle, since it produces a strength gain at low load intensity. None of the sources in this article has evaluated it in this precise indication: it is reasoned transposition, and it must be presented as such.

What Ebert's trials do measure, on the other hand, deserves to be followed as a target: at five years, the knee extensor strength symmetry index improves significantly in both arms¹³. It is a simple, reproducible measure, and one directly steerable in rehabilitation.

The phases, and the marker that separates them

This division is not a validated protocol. It is a formalisation of what Ebert's three trials describe of their two arms, completed by the criteria they measure. It applies to matrix-induced chondrocyte implantation at the femoral condyle in adults, and to that alone.

Two things deserve emphasis in this table. The first is that the initial protection period has never been shortened : what the trials compressed is the length of the progression, not the principle of protecting the implant at the start. The second is that full active extension is the earliest and clearest marker of the benefit: four weeks against twelve is a two-month gap on a parameter that governs gait, strength and the patient's experience.

How do you apply this in practice?

Three situations cover most of what comes into the clinic: the incidentally discovered defect, the injured knee in a young person, and the postoperative phase.

Situation 1: a cartilage defect discovered on an MRI

The first question is not “what do we do with this lesion” but “does this lesion explain what the patient is complaining of”. Three things point the way: age (43 % of healthy knees over 40 carry one²), whether the involvement is focal or diffuse, and the existence of a dated triggering event.

If the answer leans towards wear, management is that of osteoarthritis, described in knee osteoarthritis, and the associated meniscal question in degenerative meniscal tears. If it leans towards a focal lesion in a young person, a surgical opinion is justified, and the delay matters.

Situation 2: an injured knee in a young person

After a patellar dislocation, remember the 43 % of articular surface lesions and the 18 % of osteochondral fractures⁴ : an MRI in the subacute phase is not a luxury. True locking or a sensation of a loose body should raise the possibility of a free fragment and speed up the pathway.

In a skeletally immature patient, osteochondritis dissecans has its own logic: the decision is made on skeletal maturity and lesion stability, and stable lesions in immature patients are typically treated without surgery⁵.

Situation 3: the postoperative phase

The first thing to establish is which technique was performed, because that is what governs both the timetable and the sporting outlook. The second is the loading protocol prescribed, and the third is whether that protocol has any justification in the technique used.

Three costly errors

First error: transferring the graft timetable to a microfracture in the belief that it rests on evidence. The three accelerated weight-bearing trials all concern matrix-induced chondrocyte implantation. The reasoning may hold, but it has to be named as reasoning.

Second error: lengthening a loading protocol “to be on the safe side”. It is the move that looks risk-free and is not: the traditional group in Ebert's trials has more pain, regains active extension eight weeks later, and keeps an abnormal gait for longer, with no gain at all on the graft¹¹¹⁴.

Third error: promising the previous sporting level without looking at the technique. The overall rate of 80.3 % is encouraging, but it hides a ranking the patient has a right to know, and in which microfracture is the only technique associated with a return to a lower level¹⁶.

Frequently asked questions

Does a cartilage defect seen on MRI mean osteoarthritis is starting?

Not necessarily. One asymptomatic, never-injured knee in four carries one, and that proportion rises to 43 % after the age of 40². The image must be interpreted in the clinical context, which the authors of that meta-analysis write explicitly.

How do you distinguish a traumatic lesion from early wear?

On four clinical features, not on the image alone: age, the existence of a dated triggering event, the number of lesions, and the state of the neighbouring cartilage. One to three focal grade III or IV lesions in someone under 40 account for only 7 % of a series of 25,124 arthroscopies¹.

Does a large lesion hurt more than a small one?

No, not measurably. Across 90 patients scheduled for surgery, there was no difference in scores between lesions under and over 2 cm² apart from one KOOS subscale, and the ICRS grade does not affect preoperative scores³.

Should a lesion under 2 cm² be operated on?

The question is open, and a recent trial shifts it: in patients aged 18 to 50 with an isolated femoral lesion under 2 cm², microfracture did no better than simple arthroscopic debridement at two years, with more complications⁶. That does not say nothing should be done, it says that microfracture is not the obvious choice it was believed to be in that bracket.

Does surgery prevent osteoarthritis?

Nothing shows that it does. At 14-15 years, 57 % of the survivors in the chondrocyte implantation group and 48 % in the microfracture group have radiographic osteoarthritis, and the authors write that their results raise serious concerns about the effectiveness of these procedures in delaying osteoarthritis⁷.

How long without weight bearing after a cartilage graft?

The best-supported protocol reaches full weight bearing in six weeks, in progressive stages, after matrix-induced chondrocyte implantation at the femoral condyle. Three successive randomised trials shortened that time from 11 to 8 and then from 8 to 6 weeks without ever observing deterioration of the graft on MRI¹¹¹⁴¹³.

Does that timetable also apply after microfracture?

No trial has tested it. The whole of Ebert's series concerns matrix-induced chondrocyte implantation. Applying the same timetable after microfracture is a defensible transposition, but it is reasoning and not evidence, and it must be presented that way.

Can it be faster than six weeks?

Nothing supports that to date. The authors of the last trial write themselves that their six-week protocol is faster than any proposed and studied so far¹³ : there is no data below that.

When can sport be resumed?

Between 7 and 18 months depending on the technique, with an overall return to sport of 80.3 %¹⁶¹⁵. The fastest is obtained after osteochondral grafting, at 6.6 ± 2.6 months on average. These figures are means from series, not an individual timetable.

Does a patellar dislocation justify an MRI?

In an adolescent after a first episode, the figures argue for it: 43 % of those knees have an articular surface lesion and 18 % an osteochondral fracture, with MRI performed at a median of six days in the benchmark cohort⁴. The risk is no lower for a dislocation occurring outside sport.

What can physiotherapy bring without surgery?

We have to be honest: none of the sixteen sources in this article establishes that a rehabilitation programme treats a focal cartilage lesion. What physiotherapy can do is treat what accompanies the lesion (seven lesions in ten are not isolated¹), restore strength and function, prepare for possible surgery, and not let a situation drag on in which the time before a decision matters.

References

Sixteen references, verified one by one through the NCBI E-utilities: a real PMID, journal, issue year, complete author list, and abstract read to check that the source does establish what is attributed to it. The working base is versioned in the repository, with the list of what it does not support.

This page deals with the focal cartilage lesion of the knee and its surgical aftermath. Diffuse wear and its conservative management are the subject of a separate article: knee osteoarthritis. The meniscal question on the same ground is dealt with in degenerative meniscal tears. The subchondral bone and post-traumatic marrow oedema are set out in bone bruise. The ligament injuries that accompany a cartilage lesion one time in three are described in knee sprains and, for the postoperative phase, in rehabilitation after ACL reconstruction. Finally, anterior knee pain, whose differential diagnosis crosses this one, is dealt with in patellofemoral pain syndrome.