Runners Knee: What the Research Says About Patellofemoral Pain, Recovery and Return to Running

Key Takeaways

In a 2021 systematic review of running injuries, patellofemoral pain accounted for the largest share of injuries by specific diagnosis in the prevalence data, 16.7% of reported injuries [1].
It is often persistent: in an observational analysis of data from two randomised trials, 40% of patients reported an unfavourable recovery at 12 months [2].
In a 2016 international consensus, exercise therapy had the strongest support of the treatments reviewed, and the consensus favours combining hip and knee exercises over knee exercises alone [3].
In a randomised trial in runners, adding exercises or gait retraining to education on symptom-guided training-load management did not produce better outcomes than education alone [4].
In my clinical experience, runners knee is often difficult to treat and examination findings are frequently modest; I tend to point patients towards physiotherapy and staying active within the limits of pain rather than inactivity.

Introduction: Why Runners Knee Deserves More Attention Than It Gets

A 2018 systematic review and meta-analysis reported an annual prevalence of patellofemoral pain of 22.7% in the general population and 28.9% in adolescents [5]. The general-population figure in that review came from a single community study, which also reported a higher annual prevalence in women (29.2%) than in men (15.5%) [5].

Patellofemoral pain has traditionally been viewed as self-limiting, but recent studies show that a large proportion of patients still report knee pain at long-term follow-up [6]. The 2016 international consensus statement describes patellofemoral pain as a stubborn condition that, according to growing evidence, can persist for many years [7].

This article covers the definition, causes, diagnosis, long-term outlook and treatment evidence for runners knee.

What Is Runners Knee? The Consensus Definition

The 2016 International Patellofemoral Pain Research Retreat consensus statement names “patellofemoral pain” as the preferred term and lists patellofemoral pain syndrome, chondromalacia patella, anterior knee pain and runner’s knee as synonyms [7].

According to the same consensus, the one core criterion is pain around or behind the kneecap that is aggravated by at least one activity loading the patellofemoral joint while the knee is bent under body weight, such as squatting, stair walking, jogging or running, and hopping or jumping [7]. Supporting but non-essential features include a grinding sensation (crepitus) during knee bending, tenderness when the edges of the kneecap are pressed, a small joint effusion, and pain with sitting, rising from sitting or straightening the knee after sitting [7].

People with a history of kneecap dislocation, or who feel that the kneecap partially slips out, are treated as a separate subgroup with potentially different mechanisms and treatment needs [7].

It is also worth noting that “runner’s knee” is used loosely in everyday language. Running-injury research classifies iliotibial band syndrome as a separate diagnosis from patellofemoral pain [1], so pain on the outside of the knee should not automatically be assumed to be the same condition.

How Common Is Patellofemoral Pain in Runners?

A 2021 systematic review of running-related musculoskeletal injuries reported a mean overall injury incidence of 40.2% and a mean prevalence of 44.6% [1]. Within that review, patellofemoral pain syndrome accounted for 6.3% of injury incidence, the third-highest of any specific pathology, and for 16.7% of injury prevalence, the highest of any specific pathology [1].

The 2016 consensus statement notes that patellofemoral pain frequently brings people to physiotherapy, general practice, orthopaedic and sports medicine clinics [7].

In my clinical experience, runners knee is not diagnosed at the practice particularly often, despite how common it is. From a clinical perspective, the symptoms are fairly non-specific: when the patient is not a runner and the pain is not clearly caused by running, the picture can remain unclear. Pain around the kneecap is nevertheless a fairly common reason for a visit, and I have seen it interpreted as non-specific knee pain. If the knee also shows a small effusion, the problem is often read as non-specific wear-related or overuse-related pain.

One study included in the 2018 meta-analysis suggests that novice runners may be an exposed group, although a single study cannot be generalised to all new runners. In novice female recreational runners, it recorded a 10-week incidence rate of 1080.5 cases per 1,000 person-years [5]. The review authors noted that this was comparable to the rate in conscripted military recruits and interpreted it as consistent with short periods of unaccustomed high activity being a risk factor [5].

What Causes Runners Knee? Load, Mechanics and Risk Factors

The patellofemoral joint carries high loads during running

In a musculoskeletal modelling study of 30 healthy, pain-free recreational runners, the estimated net patellofemoral joint force at the runners’ preferred step rate averaged 5.76 times body weight [8]. When step rate was raised to 110% of preferred at the same speed, peak patellofemoral joint force fell by 14%; when step rate was lowered to 90%, it rose by 15% [8]. Peak knee flexion during stance was the most important predictor of the reduction in joint loading [8].

These forces were model estimates, not direct measurements, and the participants did not have knee pain; the authors themselves state that further work is needed to confirm the effect in people with patellofemoral pain [8].

Training load spikes

In the authors’ interpretation, the 2018 epidemiological review is consistent with short periods of unaccustomed high activity being a risk factor, and they describe training without load “spikes” as one proposed way to manage risk [5]. This is an interpretation rather than a tested causal finding.

What prospective studies actually show about risk factors

A 2019 systematic review and meta-analysis pooled 18 prospective studies involving 4818 participants, of whom 483 developed patellofemoral pain [9]. Strong to moderate evidence indicated that age, height, weight, body mass index, body fat and Q angle were not risk factors for future patellofemoral pain [9].

Moderate evidence indicated that quadriceps weakness was a risk factor in military recruits, especially when normalised to BMI (SMD −0.69) [9]. Perhaps surprisingly, moderate evidence indicated that hip weakness was not a risk factor, and in adolescents greater hip abduction strength was associated with future patellofemoral pain (SMD 0.71) [9].

These findings concern risk before symptoms start; they do not address strength changes after pain has developed.

How Patellofemoral Pain Is Diagnosed

The consensus statement describes clinical examination as the cornerstone of diagnosis but notes that no single clinical test is definitive [7]. The best available test is anterior knee pain provoked during a squat: patellofemoral pain is present in 80% of people who test positive [7]. Tenderness on pressing the edges of the kneecap has more limited evidence, with patellofemoral pain present in 71–75% of people with this finding [7].

Patellar grinding and apprehension tests, such as Clarke’s test, have low sensitivity and limited diagnostic accuracy for patellofemoral pain [7].

In practice, examination of a painful knee usually covers range of motion, palpation of the menisci, a McMurray test, testing of the ligaments, checking for patellar ballottement and palpation of the painful areas. In my clinical experience, the findings in these patients are often quite modest: non-specific pain around the kneecap or when the kneecap is pressed, while the rest of the knee may not be particularly painful. The knee is usually not red or notably warm. Swelling can occur, but a red, warm knee may warrant suspicion of something else, for example arthritis.

The consensus lists a small effusion as an additional, non-essential feature [7]. In my own practice, I sometimes see more obvious swelling. I do not aspirate very often unless the knee is markedly swollen, in which case I consider aspiration worthwhile. Often, however, the effusion is small, and in my view aspirating it adds little.

Is it the same as arthritis?

In a 5–8-year follow-up of patients originally diagnosed with patellofemoral pain, 48 of the 50 participants who underwent knee radiographs (98%) showed no signs of radiographic knee osteoarthritis [6]. The sample was small, so this finding should be read with caution.

Runners Knee Recovery Time: What the Long-Term Data Show

In an observational analysis of 310 individuals with patellofemoral pain enrolled in two randomised trials, 55% had an unfavourable recovery at 3 months and 40% at 12 months, irrespective of the treatment received [2]. Knee pain lasting more than 2 months at baseline was the most consistent predictor of poor recovery, followed by an Anterior Knee Pain Scale score below 70/100 [2]. The authors encouraged early intervention with treatments of known efficacy [2].

A later 5–8-year follow-up drew on two randomised trials (310 patients in total) and reached 60 participants (19.3%), and 34 of them (57%) reported an unfavourable recovery [6]. Baseline pain lasting more than 12 months and a lower Anterior Knee Pain Scale score predicted a worse long-term outcome [6]. Fewer than one in five participants responded, although the authors found no differences between responders and non-responders in baseline characteristics or 3- and 12-month outcomes [6].

In my clinical experience, runners knee is often difficult to treat, and patients may suffer from it for a long time. From a clinical perspective, this is in line with the long-term follow-up data above.

Treatment: What the Evidence Supports

Exercise therapy

A Cochrane review of 31 trials with 1690 participants found very low-quality but consistent evidence that exercise therapy may produce a clinically important reduction in pain and improvement in function, and may improve long-term recovery [10]. Compared with control, exercise reduced short-term pain during activity by a mean of 1.46 points on a 0–10 scale (five studies, 375 participants) [10]. Pooled data from three studies (104 participants) favoured hip plus knee exercises over knee exercises alone for short-term pain during activity, with a mean difference of 2.20 points [10]. The review found insufficient evidence to identify the best form of exercise therapy [10].

The 2016 consensus statement reached a similar conclusion: exercise therapy is recommended to reduce pain in the short, medium and long term and to improve function in the medium and long term, and combining hip and knee exercises is recommended in preference to knee exercises alone [3]. Exercise therapy was the only intervention tested on its own that the expert panel supported [3].

Other physical interventions

The same consensus recommended combined interventions (for example exercise together with taping, mobilisation or foot orthoses) to reduce pain in adults in the short and medium term, and foot orthoses for short-term pain relief [3]. It added that the average pain reduction from prefabricated foot orthoses may lack clinical significance because individual responses vary widely [3].

The panel did not recommend patellofemoral, knee or lumbar mobilisations, or electrophysical agents [3]. Whether patellar taping should be a first-line treatment was unclear to the panel, and the role of taping and bracing on their own was considered not yet fully determined [3].

In practice, patients often use some kind of knee support and ask me whether it helps. My own reasoning is that, given how large the forces passing through the knee are, a brace is probably of limited help in damping them; this has not been tested in the studies cited here. In my view, a brace may still warm the knee and help the patient feel more stable, so I consider using one acceptable.

Medication and injections in practice

None of the studies cited in this article evaluated anti-inflammatory medication or corticosteroid injections for patellofemoral pain, so the following paragraphs describe my own clinical practice rather than research findings.

In my clinical experience, many patients take anti-inflammatory painkillers (NSAIDs) for their knee pain, and I have often prescribed etoricoxib for patients with knee pain. In my view, however, physiotherapy is the first-line treatment for knee pain, and especially for runners knee, and I try to refer patients to physiotherapy whenever possible. Some patients would prefer a quick fix in the form of an anti-inflammatory drug and a corticosteroid injection, and from time to time patients come to me specifically asking for an injection.

I do occasionally give corticosteroid injections into the knee. In my view, how much an injection actually improves the long-term outlook is difficult to say. In my clinical experience, patients often report a subjective benefit, but whether this reflects a placebo effect or the corticosteroid itself remains an open question for the individual patient. Before an injection, I explain that a knee joint injection always carries risks, including a risk of infection, and that these need to be discussed. For this reason, I do not necessarily recommend injection therapy in many situations. Some patients also feel that other passive treatments help them, but in practice I encourage patients to seek physiotherapy.

Load Management and Gait Retraining in Runners

Education on training loads

The most runner-specific evidence comes from a randomised clinical trial of 69 runners with patellofemoral pain [4]. Runners were assigned to one of three 8-week programmes: education on symptom management and training modification alone, education plus an exercise programme, or education plus gait retraining [4].

All three groups improved similarly at 4, 8 and 20 weeks, with no significant differences between groups in symptoms or function [4]. Only the exercise group increased knee extension strength, and only the gait retraining group increased step rate (by 7.0%) and reduced average vertical loading rate (by 25.4%), without additional clinical benefit over education alone [4]. The authors concluded that education on symptoms and training-load management should be a primary component of treatment for runners [4].

In that trial, the education component asked runners to run more often but with shorter and slower sessions, to avoid downhill and stair running, to allow run–walk intervals, to keep knee pain at no more than 2/10 during running, and to expect pain to return to pre-run levels within 60 minutes without being worse the next morning; distance was then gradually increased before speed and hills were reintroduced [4]. The trial had no untreated control group, so spontaneous improvement cannot be fully excluded [4].

In practice, I tell patients that the best approach is to look after their muscle strength. In the long run, I do not expect this kind of problem to improve with complete inactivity, so in my view continuing physiotherapy and sport is worthwhile. I tend to suggest activities that do not load the knee in the same way, such as swimming if the individual patient finds it comfortable, or cycling if running is not possible and cycling eases the symptoms. I also tell patients that they can move freely within the limits of pain. I do not recommend inactivity, because the problem is often long-lasting, and if it is allowed to take over too much of daily life, it may set off a cascade of inactivity that can lead to other problems.

Increasing step rate

Raising cadence is a popular running modification. Beyond the modelling data above [8], a 2019 case series followed 12 runners with patellofemoral pain who had been selected because they showed specific hip and pelvis movement patterns on 3D gait analysis [11]. After a single session in which step rate was increased by 10% using a metronome, worst pain over the past week fell from an average of 6.2/10 at baseline to 1.0 at 4 weeks and 0.3 at 3 months, and Lower Extremity Functional Scale scores rose from 62.3 to 76.6 and 79.7 [11].

These results come from a small study without a control group, a limitation the authors acknowledge [11]. The 2016 consensus likewise described gait retraining as showing emerging evidence but requiring more rigorous research before it can be considered a recommended intervention [3].

What This Means for Runners

Duration matters. Symptoms lasting more than 2 months were the most consistent predictor of poor 12-month recovery [2].
For runners, managing training load is central. In a randomised trial in runners, adding exercises or gait retraining to education on symptom-guided load management did not produce better outcomes than education alone [4].
Some passive treatments were not recommended. Mobilisations and electrophysical agents were not recommended by the international consensus [3].

If the symptoms do not start to ease, I refer the patient to an orthopaedic surgeon at least once. The orthopaedic surgeon often carries out further differential diagnosis, and an MRI scan may be arranged if the symptoms are difficult and more detailed specialist assessment is needed.

Conclusion: Runners Knee Is Common, Often Persistent and Best Approached With Evidence

Runners knee, or patellofemoral pain, accounted for the largest share of injuries by specific diagnosis in the prevalence data of the 2021 systematic review, 16.7% of reported injuries [1]. It is often persistent: in an observational analysis of data from two randomised trials, 40% of patients still had an unfavourable recovery at one year [2], and longer symptom duration predicts a poorer outlook [2][6].

In my view, the evidence favours exercise therapy, particularly programmes combining hip and knee exercises [3], and, for runners, managing training load [4]. The 2016 consensus did not recommend patellofemoral, knee or lumbar mobilisations or electrophysical agents [3].

This matches what I see in practice: runners knee is often long-lasting and difficult to treat, and while some patients report a subjective benefit from injections or braces, I direct most patients towards physiotherapy and continued activity within the limits of pain.