Patellofemoral Pain CPG

The CPG on patellofemoral pain is both long and dense, so Austin takes us through an information-packed episode with a very helpful practice question at the end to tie it together and explain how you should expect to use this information on the OCS exam.

Use code FIELDGUIDE for 40% off a MedBridge subscription.

Find more resources and subscribe to practice questions at PhysioFieldGuide.com.

Transcript

Hello and welcome back to the OCS Field Guide podcast. Today we are covering the 2019 Patellofemoral Pain clinical practice guideline. 

We’re going to switch things up a little bit this episode and begin with a practice question, then we will answer it at the end of the podcast after we’ve covered the information you need to answer correctly. 

A 14 year old female presents with pain in the front of her L knee that she describes as achy pain under knee cap. She reports that track and field season started two weeks ago and her coach is having her do high jump which she has not done previously and she jumps off her L foot. She is also running the 200m and 400m. She now reports pain with going up and down stairs and when sitting in class in addition to her track and field activities. On NPRS her usual pain is a 4/10 and pain at worst is 7/10 when she goes down stairs. 

Which of the following would be the most appropriate initial treatment?

A. Prefabricated foot orthosis
B. Forward step down exercise
C. Tailored patellar taping
D. Patellar medial glide mobilizations

We’ll come back to this question at the end. 

Let’s get this started with some general prevalence and incidence data. Patellofemoral pain is surprisingly prevalent, with most estimates being right around 25% of the population. Depending on what setting you are looking at, the highest incidence is around 12-19 years old for sports medicine settings, and 50-59 years old for general practice settings, but can occur across the whole lifespan. It is slightly more common in females in all age groups. It is surprisingly recurrent, with rates of return of symptoms of 70-90%. It is NOT a self limiting problem, with over 50% of adults reporting unfavorable outcomes 5-8 years after diagnosis, and 50-56% of adolescents still having pain after 2 years. Clearly, its an area we need to be managing a lot better.  

Let’s move on the clinical presentation. A number of different pathoanatomic features have been implicated over the years in relation to patellofemoral pain such as internal derangement, cartilage softening, and maltracking, but none of them are strongly correlated with developing patellofemoral pain. Rather, the clinical presentation of pain under or around the patella with functional activities such as squatting, stair negotiation, sports participation, prolonged sitting, and walking is most useful in diagnosis of patellofemoral pain after ruling out pathoanatomic diagnosis. To restate that, where and when a patient has pain is useful in diagnosing patellofemoral pain, not specific imaging findings. On that note, let’s talk a little about patellofemoral pain versus patellofamoral osteoarthritis. This is an important take away from this CPG both for the OCS exam and clinical practice, because if you are like me, a large percentage of the patients sent to me for patellofemoral pain are teenagers and are often told terrible things like, “you have arthritis” “you should stop running and playing sports” and “you’ll probably end up needing a knee replacement really young.” And although there is some link between patellofemoral pain and patellofemoral OA, it is not at all causational.  So far, the available prospective studies show that although individuals with patellofemoral pain are likely to have chronic and recurrent issues (especially if not treated early), they are not necessarily more likely to be diagnosed with patellofemoral OA, though more research needs to done to confirm or refute this. The only retrospective study examining a link between patellofemoral OA and adolescent presence of patellofemoral pain was done in those who had received a patellofemoral arthroplasty, and compared to individual receiving partial tibiofemoral arthroplasty, those patients reported a relatively higher rate of adolescent patellofemoral pain, instability, and trauma. Even still, only 22% of individuals who had had a patellofemoral arthroplasty recalled having idiopathic patellofemoral pain. Suffice it to say, current evidence does not support that having patellofemoral pain can predict having patellofemoral OA and need for arthroplasty later in life.

On to risk factors: here is a potentially surprising list of factors that do NOT put individuals at greater risk for patellofemoral pain: height, weight, percentage body fat, Q-angle, patellar mobility, and static knee valgus. There is conflicting evidence on whether ankle/foot characteristics such as arch height, pronation, hindfoot and forefoot angle are related to development of patellofemoral pain. Now here is the list of factors that ARE related to development of patellofemoral pain: female sex (one study found that female navy cadets were over 2 times as likely to develop patellofemoral pain), sports specialization (where we see females who play only one sport are much more likely to develop patellofemoral pain), and isometric knee extension strength. Cross sectional studies have shown that PFP is also associated with weakness hip extensors, abductors, and external rotators. However, high quality prospective studies have shown that hip weakness is likely a result of patellofemoral pain or are developed along with patellofemoral pain rather than being predictive of patellofemoal pain. Decreased flexibility in the quadriceps, hamstrings, and gastrocnemius is also found in individuals with patellofemoral pain, but not enough research has been done to show if this precedes anterior knee pain or not. Individuals with patellofemoral pain do typically exhibit altered mechanics as well including decreased knee flexion with stair negotiation, and dynamic valgus, or increased frontal plane projection angle, with single leg squat and jump landing. Only one study by Holden et al in 2015 has found dynamic knee valgus in vertical jump landing to be predictive of patellofemoral pain in female athletes. 

To summarize this, if you were asked to address risk factors with a patellofemoral pain prevention plan, you would likely target females, especially adolescent females, and you would focus on educating them to participate in a variety of sports rather than specialization or single sport participation and instruct them in an exercise program that targeted knee extensor strength primarily.

For prognosis, the CPG reads, “individuals with a longer duration of symptoms, higher baseline pain severity, and poorer function were more likely to have negative outcomes or unfavorable recovery.” *end quote* However, no cutoffs for pain, duration, or functional measures have been established at this time. 

For the remainder of the episode I’ll follow the structure of examination and intervention given in the decision making tree provided at the end of the CPG to talk about the rest of the CPG to help you ingrain that structure as well as to give some better organization to the information. The sequential components are: 1. Medical screening (which should include psychological screening); 2. Diagnosis and classification of condition; 3. Determination of irritability stage; 4. Outcome measures; and 5. Intervention strategies.

I know you’re no stranger to medical screening, but I do want to provide list of pertinent medical differential diagnoses to keep in mind. 

• Tumors

• Dislocation

• Septic arthritis

• Arthrofibrosis

• Deep vein thrombosis

• Neurovascular compromise

• Fracture (local and/or at the hip)

• Slipped capital femoral epiphysis in children or adolescents

Of these I think the most potentially tricky one is slipped capital femoral epiphysis, as this can present as knee pain and it fits the age range we are often seeing for patellofemoral pain.  

On to diagnosis. The diagnosis of patellofemoral pain is made primarily with subjective report and functional movements. The most sensitive tests for patellofemoral pain are: reproduction of retropatellar or peripatellar pain with squatting, pain with stair climbing and/or descent, and pain with kneeling. All three of these have moderate to high sensitivity and negative likelihood ratio, but low specificity. Meaning that negatives in these tests should significantly decrease your suspicion of patellofemoral pain syndrome. The most specific test for patellofemoral pain is the eccentric step down test which has moderate specificity (82%) and moderate +LR at 2.3. So, when you put it all together, confident diagnosis of patellofemoral pain syndrome can be made with a combination of these 3 factors. 1. Report of retropatellar or peripatellar pain, 2. That pain is reproducible with squatting, stair climbing and descent, kneeling, prolonged sitting, or other activities that significantly load the patellofemoral joint; and 3. Exclusion of all other possible sources of anterior knee pain. Important consideration for other sources of anterior knee pain include: patellofemoral OA, patellar tendinopathy, patellar subluxation or dislocation, tibial apophysitis, patellar apophysitis, or tibiofemoral issues. Its also important to keep in mind that hip OA can present with a primary complaint of knee pain, and don’t forget to consider the lumbar spine. One of the main differentiators I would look for to differentiate patellofemoral pain from patellofemoral OA are older age and loss of range of motion. 

Now let’s look at other useful exam measures that don’t aid in diagnosis, but will aid in classification and subsequent intervention planning, as each of them will reveal an evidence based objective deficit that has been shown to contribute to this condition. 

The authors give an expert opinion that patients be classified in one of 4 categories, that are based, somewhat loosely, on impairments found in individuals with patellofemoral pain and subsequent treatment strategies. This is an initial attempt at a treatment based classification system for patellofemoral pain. The categories are: 1. Overuse/overload without other impairment. 2. Patellofemoral pain with movement coordination deficits. 3. Patellofemoral pain with muscle performance deficits. And 4. Patellofemoral pain with mobility impairments, which will subdivide into hypermobility (of the foot), and hypomobility (of the patellofemoral joint, hip, and LE muscles). This system has not been validated by a Randomized Controlled Trial, but does help prioritize interventions. Thus, it is especially important with this classification system to remember that patients may meet multiple criteria and may change over the course of care. Also, some of the criteria lead you to interventions that have weaker evidence, and thus it may still be beneficial in those times to include other interventions with higher levels of evidence. 

The first category, overuse/overload without other impairment, will apply best to individuals without significant impairment that have irritated the patellofemoral joint due to training load factors. Two cohorts that are at high risk for developing patellofemoral pain related primarily to significant increase in patellofemoral load are military populations and recreational runners who both often quickly increase magnitude and frequency of loading without proper recovery time. Classification in this category is made when the patient presents with the S/S of patellofemoral pain, history of increase in load magnitude and/or frequency, and has reproduction of anterior knee pain with eccentric step-down test, but no impairments that lead to another classification. 

Classification of patellofemoral pain with movement coordination impairments is made based off noting dynamic valgus on the lateral step down test where a score of greater than 2 points on the quality of movement scale indicates a positive test, and noting frontal plane valgus, also known as frontal plane projection angle, of greater than 10 deg during single leg squat. The lateral step down test is performed from a 20cm step and quality of movement is graded based on 5 criteria: arm strategy (where removal of hands from waist gives +1), trunk alignment (where leaning any direction gives +1), pelvic plane (where loss of horizontal plane of the pelvis gives +1), knee position (where if the tibial tuberocity is medial to the 2nd toe +1 is given, and if the tibial tuberocity is medial to the medial border of the foot +2 is given), and steady stance (where if the patient had to step down with the non-test leg, or if the patient moved the foot side to side, +1 is given. Again, greater than a score of greater than 2 on the lateral step down test, and frontal plane projection angle greater than 10° on the single leg squat are positive criteria for classification with movement coordination deficits. It is important to note that excessive or poorly controlled knee valgus has not been shown to necessarily indicate weakness of LE musculature, but rather lack of motor control in this area and thus should be addressed accordingly. 

Next, classification with Patellofemoral pain with muscle performance deficits is made based hip and knee isometric strength deficits. Each of these strength tests is to be performed with hand-held style dynamometer fixed by a static strap. First, the Hip Stability Isometric Test or HipSIT is recommended as a surrogate for posterolater hip muscle strength. This test is performed in the classic “clamshell exercise” position with the hips at 45°, knees at 90°, and the hip abducted to 20°. Individuals with patellofemoral pain demonstrated 10% reduction in force production compared to individuals healthy knees. Pertinent isometric hip and thigh muscle strength measures include hip abductors, external rotators, and extensors; and knee extensors and flexors. One study by ferber et al was able to derive cutoff scores for responders to hip and thigh strengthening. Strength for each is reported as percentage of body mass. For abductors, the cutoff for men was less than 37% body mass and for women was less than 30%. For hip external rotators, the cutoff for men is less than 13% body mass and for women is less than 17% body mass. For hip extensors, the cutoff for men is less than 28% body mass and for women is less than 30% body mass. And yes, you did hear me right. Apparently, at least in this study, pound for pound women were expected to have stronger hip external rotators and extensors than men. For knee extensors the cutoff for male responders is having less than 44% body mass, and for females is having less than 37% body mass; and for knee flexors there is not an established cutoff. Now, I don’t expect you to have to know verbatim each of those strength ratios, but I mention them specifically to give you a good idea of how strong each of these muscle groups are supposed to be in relation to each other and in relation to body mass. 

The final classification is Patellofemoral pain with mobility impairments. This section is subdivided into hypermobility, which is referring to foot hypermobility not patellar hypermobility which would be beyond the scope of this CPG, and hypomobility, which is referring to hypomobility in the patellofemoral and hip joints, and decreased LE flexibility. For classification as hypermobility, they recommend foot mobility testing by measuring mid foot width in non-weight bearing and weight bearing. A difference of greater than 11mm between NWB and WB is considered hypermobile for this category as those with greater than 11 mm difference demonstrated significantly greater improvements in pain with the use of foot orthoses compared to controls. They also recommend classification with hypermobility with a Foot posture index score of greater than 6. I don’t expect you to need to know this scale perfectly, but know that a higher score indicates a greater pronation and hypermobility. Classification with hypomobility is made with a positive patellar tilt test for tightness of the lateral patellar retinaculum, decreased LE muscle length, and/or decreased hip IR and ER ROM. The patellar tilt test is the only patellar mobility test with as high as low to moderate reliability, which is still not great, while the many other patellar mobility tests have poor to fair reliability and have no diagnostic accuracy for patellofemoral pain. However, one study by Haim et al. did find high specificity of 92% and moderate +LR with this test for individuals with patellofemoral pain, so a positive would be useful for ruling in patellofemoral pain. Cutoffs for decreased muscle length are as follows. For the hamstrings, straight leg raise of less than 79°; for gastroc length ankle dorsiflexion with knee extended of less than 7.4°; for soleus length, ankle dorsiflexion of less than 14.8°; for quadriceps, prone knee flexion of less than 134°; and for iliotibial band, an ober’s test with less than 11° of hip adduction. Finally, there are no established cutoffs for hip IR and ER ROM.

Before we move on, a quick note on other tests that are NOT recommended. You’ll note we have not mentioned anything about trying to visually assess whether the VMO is atrophied, as quad atrophy is only reliable when measured by MRI and is always consistent across all of the vastii groups, not just the VMO. Nor have I mentioned looking to see whether the VMO is firing with the proper speed or quality, as discrepancy in firing pattern of the VMO is only perceptible in EMG studies; not to mention the strong evidence that now exists that one cannot selectively activate the VMO no matter if you squint or squeeze a ball while doing a long arc quad or even with EMG biofeedback, you just can’t; I’ll even get ahead of myself and mention that even electrical stimulation to the VMO or EMG biofeedback to the VMO during quad exercise does not improve outcomes in individuals with patellofemoral pain. So while you are taking the test, don’t select a VMO activation option. 

Now for outcome measures. For patient reported measures, the CPG gives A-level recommendation for the use of the Kajula Anterior Knee Pain Scale and the KOOS-PF, which was developed specifically for patellofemoral pain and osteoarthritis. Both of these are scored out of 100, and are measures of function, with a higher score meaning higher function and less disability. Though I doubt you’ll need to know them for the test, the MCID for the AKPS is 8-10 points and MCID for the KOOS-PF is 14.2. Additionally, they give A-level recommendation for the visual analog scale for worst pain and usual pain, or the VAS for activity, also known as the EPQ, where 6 activities are rated on a visual analog scale and summed for a total score. The 6 activities are: walking, running, squatting, sitting, ascending stairs, and descending stars. For physical performance measures, they give B level recommendation for functional movements that reproduce the patients familiar anterior knee pain or allow assessment of lower limb movement coordination such as squatting, step down, and single leg squat. 

Let’s (finally!) get on to intervention. As I already mentioned, the classification system has not been validated with RCTs and thus is primarily used to help categorize presentations by pertinent objective measures, and prioritize interventions. There is strong A-level recommendation FOR combined interventions, so don’t take this classification to mean you should only perform the interventions listed for that classification. 

For the overuse/overload without other impairment category, remember we are talking about individuals who only have pain because of an increase in training magnitude or frequency. Thus interventions for this category are going to involve calming things down. For this group there is B-level recommendation for patellar taping and expert opinion or F level recommendation for activity modification and relative rest. The patellar taping recommended with best evidence is the tailored McConnel style taping aimed at supporting patellar tilt, glide, and rotation, which has been shown to provide large reduction in pain. Untailored simple medially directed taping produces only immediate small pain reductions. Taping is recommended in combination with exercise intervention to enhance outcome in the short term such as 4 weeks, but may not be beneficial in the long term. While patellar taping is recommended, patellar bracing such as knee brace with patellar block, sleeve, or patellar strap is not. Importantly, taping techniques with the aim of enhancing muscle function such as to improve vastus medialis activation is not recommended in the treatment of patellofemoral pain. 

For patients classified as patellofemoral pain with movement coordination impairments, the main C level recommendation is for gait and movement retraining. Now this recommendation is based solely on research for running gait retraining in runners, so it does appear to have limited application. Clinicians may use running gait retraining for runners individuals with PFP that includes multiple sessions of curing to adopt a forefoot-strike pattern (for rearfoot-strike runners), cueing to increase running cadence, or cuing to reduce peak hip adduction while running. It’s also important to note that the highest quality RCT that this recommendation is based off of also included patient education on load management (including avoiding hills and reducing run-session volume and increasing session frequency), and the comparison groups of education on load management alone, and combined hip and knee exercise plus load management education where just as effective. Clearly I have a dog in this fight, so I’ll admit my bias, but be careful making changes in strike pattern. Beyond running gait retraining there is not specific movement retraining recommendation for improving mechanics with other functional movements. This research just hasn’t been done. But it is important to note that in the many studies supporting strength training they are typically instructing cuing proper performance with exercise that mimic functional tasks such as squatting and step ups. There is though a B level recommendation against adding visual biofeedback during LE strengthening exercise to improve alignment and found no significant difference in mechanics or outcome with patellofemoral pain. But again, that doesn’t mean that proper instruction and cuing for proper form weren’t given in the non-biofeeback group.

Now on the A-level recommendations. For patients classified with patellofemoral pain with muscle performance deficits, there is A level recommendation for a combination of hip and knee targeted exercise in the treatment of PFP to reduce pain and improve patient reported outcomes and functional performance in the short, medium, and long term. Combined hip and knee exercise is superior to knee targeted exercise alone, and hip exercise should target the posterolateral hip musculature. Both weight bearing and non weight bearing knee targeted exercise has been found effective. There is also some evidence for giving preference to hip-targeted exercise in the early stages of treatment of patellofemoral pain.

For patients classified as mobility impairments in the foot hypermobility category, there is A-level recommendation for prefabricated foot orthoses for those with greater than normal pronation to reduce pain in the short term, up to 6 weeks. This should be combined with exercise therapy. Also there is not sufficient evidence to recommend custom foot orthoses over prefabricated. Although in the CPG this is recommended for individuals with greater than normal pronation, its important to note that there are studies that had success with prefabricated orthosis with greater-than-normal foot mobility, less-than-normal foot mobility, and regardless of foot mobility. So all that to say, you might still look into it regardless of what their foot looks like. This section in the decision tree also includes the B level patellar taping recommendation.

Last, and actually least, are the recommendations with the mobility impairments with hypomobility category. For this category there are only expert opinion or “F” recommendations for LE stretching of shortened muscle groups and for mobilization directed to the lateral patellar retinaculum and surrounding soft tissue. There are no studies that have compared stretching to other established treatment methods, however the studies that support the “combined interventions” recommendation, which received A level recommendation, do include stretching. So, I think its safe to say there is no evidence that supports stretching as stand alone treatment, but rather it can be used alongside established interventions like hip and knee strengthening exercises when those muscle length deficits do exist. With regard to lateral patellar retinaculum mobilization the only studies that included similar treatments found no stand-alone benefit in the treatment of patellofemoral pain, and the only studies that found any benefit were in combination with multimodal treatment including exercise. In fact the CPG gives and A-level recommendation against manual therapy directed at the lumbar spine, tibiofemoral, or patellofemoral joint as a stand alone treatment. So I think the only reason this is included in the decision making tree is that there is evidence for the patellar tilt-test, and they felt the need to include an intervention that addressed this objective deficit. 

As a side note, on the test when you are given a case you should always take into account addressing the objective deficits in the case, rather than just going with the strongest level of evidence. But with patellofemoral pain, I would say you should probably be careful with selecting any intervention OVER hip and knee directed strengthening exercise. The main exception for this might be if the irritability of the condition made the exercise choice given sub-optimal. I’ll repeat what I opened this section saying, the best evidence is for combined intervention, and the studies that found the best evidence for most of the treatments other than exercise combined the intervention with some form of lower extremity strength training. 

To finish up we’ll go through the interventions recommendations again, adding in any not mentioned in the decision making tree, listed from strongest evidence FOR to weakest evidence For, and then from strongest evidence against to weakest evidence against. 

There is A level recommendation for combined hip and knee targeted exercises. With hip exercises targeting posterolateral hip, knee exercises including weightbearing or non-weight bearing knee extension exercise, and preference to hip targeting exercise in early stage of treatment

There is A level recommendation for combining exercise therapy with other interventions such as foot orthoses, patellar taping, patellar mobilizations, and lower-limb stretching. 

There is A level recommendation for prescribing prefabricated foot orthoses for patients with greater than normal pronation to reduce pain in the short term, but should be combined with exercise. 

There is B level recommendation for tailored patellar taping in combination with exercise  to assist in immediate pain reduction, and to enhance outcomes of exercise therapy in the short term.

There is C level recommendation for running gait retraining consisting of multiple sessions of cuing to adopt a forefoot-strike pattern, curing to increase running cadence, or cuing to reduce peak hip adduction while running. 

There is C level recommendation that clinicians may use acupuncture (where they are allowed to practice this) but a strong caution is given that the superiority of acupuncture over placebo or sham is not known. 

There is F level recommendation that patients may use blood flow restriction plus high-repetition knee exercise therapy while monitoring for adverse events, for those with limiting painful resisted knee extension. 

There is F level recommendation for patient education including education on load management, body-weight management when appropriate, the importance of adherence to active treatments like exercise therapy, biomechanics that may contribute to relative overload of the patellofemoral joint, evidence for various treatment options, and kinesiophobia. 

Now for the “against” recommendations.

There is A level recommendation against the use of manual therapy as a stand alone treatment and against the use of dry needling.

There is B level recommendation against the use of patellofemoral knee orthoses or bracing, against the use of EMG biofeedback on vastus medialis activity to augment knee extension exercise therapy for patellofemoral pain, and B level recommendation against the use of visual biofeedback on lower extremity alignment during hip and knee-targeted exercises. 

To finish up let’s revisit our question. A 14 year old female presents with pain in the front of her L knee that she describes as achy pain under knee cap. She reports that track and field season started two weeks ago. Her coach is having her do high jump which she has not done previously and she jumps off her L foot. She is also running the 200m and 400m. She now reports pain with going up and down stairs and when sitting in class in addition to her track and field activities. On NPRS her usual pain is a 4/10 and pain at worst is 7/10 when she goes down stairs. 

Which of the following would be the most appropriate initial treatment?

A. Prefabricated foot orthosis
B. Forward step down exercise
C. Tailored patellar taping
D. Patellar medial glide mobilizations

As questions on the OCS will be, this question has almost all right answers. Option A, prefabricated orthoses has A level evidence, but there is nothing here that indicates the patient has a foot hypermobility issue. Option B, forward step down exercise is tempting because it is a “hip or knee targeted exercise” which has the strongest evidence for patellofemoral pain, however, the patients current level of irritability poses an issue with the high level of patellofemoral loading in a step down exercise. Also, the patient get’s significant pain with that exact activity. Option D is the easiest option to rule out, as there is no evidence that patellar mobility is a problem and because this is not an intervention that has any significant evidence. The correct answer is C. tailored patellar taping. This is the best option for a couple reasons. One, this best matches the stage of irritability, of the options listed, as this pain appears relatively acute in duration and in symptom severity. The case also describes a good example of someone who fits in the overuse/overload category, for which the intervention with the highest level of evidence is patellar taping. It’s also important to note with questions like this that it is not asking for the only treatment you will do, but rather the initial or very first thing you will do. After you taped this patient, you could then have a pain free way to do hip or knee strengthening exercise. 

That wraps up this episode of OCS field guide covering the 2019 patellofemoral pain clinical practice guideline.