Etiology of PFPS: A Biomechanical Perspective

So, my last post regarding patellofemoral pain syndrome (VMO? VM-No) stated its prevalence and its misdirected treatment. This next post will help to clear up some of the confusion amongst clinicians as to the cause of PFPS, which should in turn help to drive the most effective treatment strategies.

What causes PFPS? Now, this is a very loaded question. However, over the last couple decades, the etiology of this disorder has become a little more clear…

It all breaks down to a very simple formula: Patellofemoral Joint (PFJ) Reaction Force / PFJ Contact Area.

In general, as an individual goes into deeper ranges of knee flexion, their PFJ contact area increases. During OKC knee flexion the patella glides inferiorly to increase its contact with the femoral trochlea and during the CKC the femur rotates posteriorly causing increased contact with the patella. Due to this variation in loading, the PFJ experiences some of the highest stresses in the body especially during the lower ranges of knee flexion where the contact area is the smallest. In 2002, Heino et al attempted to show that individuals suffering from PFPS did demonstrate altered joint loading and subsequent joint stress. They compared the PFJ reaction force, joint stress, and utilized contact area during free and fast walking between subjects with PFPS and asymptomatic controls. They found that, on average, those with PFPS had increased joint stress, decreased utilized contact area, and decreased joint reaction forces. It was theorized that the decrease in joint reaction force may have been due to the patient’s unwillingness to forcefully load their effected limb. The increased joint stress and decreased contact area seem to implicate this improper loading as a probable cause of the anterior knee pain felt in these patients. This allows us to conclude that the general etiology of PFPS is decreased PFJ contact area and/or increased reaction force → increased chondral stress → increased subchondral bone stress → stimulation of pain receptors → Patellofemoral pain. This pain can be traced back to nociceptive afferent nerves located in the subchondral bone (Biedert et al). This theory of increased cartilage loading and subsequent degeneration was supported by Farrokhi et al in 2011 when they found that patella cartilage thickness was significantly decreased in comparison to asymptomatic controls. Additionally, they found that following an acute bout of exercise, those without PFPS exhibited decreased time to return to baseline cartilaginous thickness.

This is all well and good, but what causes this altered contact area?

This abnormal loading may be caused by non-modifiable morphological abnormalities. The first possible cause is patella alta (PA) which, according to Ward et al, increases the incidence of lateral displacement, lateral tilt, and decreased contact area. In regards to PFJ contact area, they found a statistically significant difference between subjects with PA and those with normal patella vertical displacement at all ranges tested (0°, 20°, 40°, and 60° of knee flexion). This does appear to make biomechanical sense due to the fact that the inferior pole of the patella is thought to make initial contact with the femoral trochlea at approximately 20° of knee flexion and if the patella is positioned more superiorly, initial contact will not occur until deeper ranges of knee flexion, thus decreasing overall PFJ contact area. Another non-modifiable factor is the lateral displacement and resulting decreased contact area caused by trochlear dysplasia. This bony morphological defect results in flattening of the lateral facet of the intercondylar groove, which is typically the most important local factor in controlling excessive lateral patella translation. This can result in not only abnormal loading, but also chronic patellar subluxation and/or dislocation. This flattening of the lateral intercondylar groove can result in up to a 55% decrease in medial patellar stability. Unfortunately, there is very little that we, as therapists, can do to correct these morphological abnormalities, however there are certain factors influencing abnormal patellar tracking that can be altered conservatively.

The static Q-angle has been implicated in patellofemoral pathology for countless years, but how does this correlate to alignment during functional activity? Massada et al found that dynamic Q-angle values were statistically significant in determining individuals who suffer from PFPS even in the absence of a increased static Q-angle. In this study, there was little correlation between static Q-angle and the presence of PFPS. This gives support to the use of this dynamic measure in lieu of its static counterpart. So what factors influence this dynamic Q-angle? This angle has both proximal influences at the hip (excessive adduction and femoral IR) and distal influences at the ankle/foot (excessive pronation and tibial IR). During weight bearing, the femur moves about a fixed patella and therefore excessive femoral IR results in increased contact directed primarily at the lateral facet of the patella (Powers et al). In fact, just 10° of IR can lead to a substantial decrease in PFJ contract area and a 50% increase in joint stress. The figure to the right demonstrates the lateral tilt of the patella during non-weight bearing (A) and the femoral IR resulting in altered contact area during weight bearing (B). Additionally, Souza et al found that females with PFPS demonstrated greater peak hip internal rotation compared to the control group during running, drop jump, and step down. The PFPS group also demonstrated 14% weaker hip abductor strength and 17% weaker hip extensor strength. Wilson et al, Noehren et al, and Nakagawa et al found that individuals presenting with PFPS demonstrated increased hip adduction during running, jumping, and single-leg squats. This adduction creates an increased valgus force about the knee joint, which in turn causes increased loading of the lateral PFJ. Distally, pronation at the subtalor joint can lead to IR of the tibia, which then once again creates an increase in valgus at the PFJ.

Last, but not least, quadriceps dominance (QD) has been shown to increase the incidence of PFPS due to the subsequent increased PFJ compression. When landing from a jump, a QD individual will have their knees in front of their toes, excessive dorsiflexion, heels off of the ground, and limited hip flexion. This posture puts significant stress on the quadriceps and PFJ while taking stress off of the back extensors and hip extensors. This compensation can be adopted by patients who demonstrate weak hip extensors relative to knee extensors. Compensating in this manner will, in time, lead to quadriceps overuse and increased knee loading, which will lead the patient down the previously mentioned cascade toward PFPS. Pollard et al helped to support this theory when they found that greater utilization of hip extensors during a drop jump activity was associated with decreased knee valgus angles and moments. They also found that individuals who landed in a QD position demonstrated increased knee valgus, increased knee adduction moments, and decreased energy absorption at the knee and hip. These individuals were unable to properly dissipate the high ground reaction forces and ultimately put unnecessary stress on the patellofemoral joint.

Now that we are beginning to better understand this pathology and its risk factors, we can begin to form more efficient and effective conservative treatment strategies. These treatments should be focused on decreasing laterally directed PFJ forces (locally, proximally, and distally), decreasing quadriceps dominance, and maximizing PFJ contact area. Each of these three treatment principles will be discussed in additional blog posts in the coming weeks.

Interested in learning more about PFPS in the meantime? Search for just about any article published by Christopher M. Powers, PhD, PT, FACSM, FAPTA and you should be in pretty good shape…


8 thoughts on “Etiology of PFPS: A Biomechanical Perspective

  1. I think Gary Gray’s analogy here of “the train on the track” referring to the patella tracking is appropriate here.I’ve never found satisfactory outcomes from focusing on medial/lateral joint force couples targeting VMO/VL as a primary strategy.
    Alignment of the “track” by targeting proximal and distal components has always been more fruitful. Nice article, thanks for sharing.


  2. David,
    I completely agree and have begun to wonder why many PTs still focus on the patella as the primary direction of their treatment. During weight-bearing activities, it is the femur moving on a fixed patella, not the other way around. Obviously, during the swing phase of gait and certain other function activities (i.e. kicking a soccer ball) there is a need for correct patellar alignment, but I still feel as though the “track” should be our primary focus.

    Thanks for your comment!


  3. I do not dispute the importance of addressing the joints above and below the knee. However, to ignore the force couple of the VL/ITB vs VMO precludes optimal and long lasting results. Unlike David Fitzgerald, I have gotten excellent and long lasting results by addressing the joints above and below as well well as directly attacking tight VL/ITB and strengthening the VMO with NMES and cce. Research is opening the discussion on OKC but until then, I focus on CKC with NMES and cce focussing on strengthening the VMO in positions = to orientation of the fibers of the VMO between 55-65 degrees as previously stated. Also a lot of research on the use of NMES s/p ACLR which is statistically significant addresses NMES in CKC at 65 degree angles. The use of kinesio tape’s mechanical C-correction has accelerated my results and decreased pain much sooner – not research based – just clinical.


  4. I tip my hat to your quality content. We review your work frequently with my PT’s and interns frequently. In short, while we must not look at the PFPS under a microscope, we do need to assess the patellar tracking mechanism as a component to our evaluation. I spend a considerable amount of time assessing the squat mechanism, as many patient’s squat is not worth squat. Using video with the patient they are able to grasp from all angles their mechanics. We can then use this down the road as proof that their mechanics have improved, and relate this to their pain with functional activity. I look forward to your follow-up blog posts. Take Care, Randy Bauer


  5. Yes, I agree, NMES has shown to be very beneficial for patients post TKR and ACL reconstruction. There has also been a dose-response relationship that has begun to surface (the higher the intensity, the better the benefit). However, if selective activation of the VMO could be achieved, don’t you think the research would have been done by now? There are so many therapists who use this as the cornerstone of their PFPS protocol and there have been stacks of articles devoted to EMG studies during ‘VMO specific’ exercises. None of these studies have shown the ability to recruit the VMO more so than the VL. Until a study demonstrates this, I will continue to be skeptical.

    Like I said before, there is no detrimental effect of performing these VMO exercises (other than the awkwardness of a few techniques). My main complaint is when therapists focus JUST on what’s happening at the knee. The term ‘regional interdependence’ applies here and should be the main focus of treatment. What is happening at the ankle? What is happening at the hip? What is happening in terms of core stability?

    PFPS is multifactorial in nature and our treatment should follow suit. Patellar mobs and VMO exercises won’t cut it.

    Randy, functional testing especially when using video software is incredibly beneficial! Movement and gait analysis can tell us so much about what is causing our patient’s pain. Have you ever used Ubersense ( I have started to use this app in the clinic and it has made assessment much more accurate.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s