MasterclassHow to manage patellofemoral pain – Understanding the multifactorial nature and treatment options
Introduction
Patellofemoral pain (PFP) is one of the most prevalent conditions within sports medicine, orthopaedic and general practice settings (Baquie & Brukner, 1997; Kannus, Aho, Jarvinen, & Niittymaki, 1987; Taunton et al., 2002). The incidence of PFP varies significantly between specific populations, with figures as low as 3% in established runners and to as high as 43% in naval recruits during basic military training (Boling et al., 2009; Smith et al., 2018; Thijs, Van Tiggelen, Roosen, De Clercq, & Witvrouw, 2007). In adolescents, the incidence of PFP has been reported to be as high as 10% in high school female athletes during the competitive basketball season (Myer et al., 2010). The variability in incidence and high prevalence within specific populations is indicative of a complex multifactorial condition and a lack of consensus on the most appropriate diagnostic criteria (Crossley, Stefanik, et al., 2016; Nunes, Stapait, Kirsten, Noronha, & Santos, 2013).
Symptoms of PFP are characterised by pain around or behind the patella, aggravated by activities that increase load on the patellofemoral joint (PFJ) (e.g. squatting, ascending and descending stairs, prolonged sitting and running) (Collins, Vicenzino, van der Heijden, & van Middelkoop, 2016; Crossley, Stefanik, et al., 2016). Individuals with PFP may also describe or experience crepitus emanating from the PFJ, tenderness on palpation of the patella facets and a small effusion (Crossley, Stefanik, et al., 2016). Although widely researched, the exact source of pain in people with PFP is unclear, with several theoretical models previously described (Dye, 2005; Sanchis-Alfonso, Rosello-Sastre, Monteagudo-Castro, & Esquerdo, 1998). These include altered patellar tracking, resulting in elevated patellofemoral joint contact pressure (Ho, Hu, Colletti, & Powers, 2014; Powers, 2003), and a loss of tissue homeostasis in the surrounding innervated tissues including the synovial lining or fat pad (Dye, 2005).
Based on current evidence and clinical practice, exercise therapy forms the cornerstone of management for PFP (Crossley, Stefanik, et al., 2016; van der Heijden, Lankhorst, van Linschoten, Bierma-Zeinstra, & van Middelkoop, 2015) with efficacy of a multimodal approach having been well reported (Barton, Lack, Hemmings, Tufail, & Morrissey, 2015; Collins, Bisset, Crossley, & Vicenzino, 2012). Despite the reported benefits of multimodal treatment for PFP, longer-term follow up data (5–20 years) indicates that more than 50% of individuals with PFP continue to experience symptoms and unfavourable outcomes (Lankhorst et al., 2016; Nimon, Murray, Sandow, & Goodfellow, 1998; Witvrouw, Danneels, Van Tiggelen, Willems, & Cambier, 2004). The development of a targeted intervention has been proposed to represent an approach that could positively impact long-term treatment outcomes (Crossley, Stefanik, et al., 2016; Glaviano & Saliba, 2016; Powers, Bolgla, Callaghan, Collins, & Sheehan, 2012; Witvrouw et al., 2014).
This masterclass aims to introduce the clinician to currently proposed theories linked to the aetiology of PFP, and the multiple potential factors related to PFP symptom development and persistence. This information will then be used as foundation knowledge to inform clinical assessment and provide insight into potentially effective treatments, along with discussion of areas where greater understanding is still required. Importantly, the masterclass will provide practical tools and resources that can be directly implemented into clinical practice.
Traditional paradigms of pain presentation have been derived from a primary nociceptive pathway directly associated with overload of the PFJ. It has been proposed that alterations in lower limb biomechanics result in maltracking of the patella within the trochlea groove (Fig. 1) (Powers, 2010). Multiple potential mechanisms are thought to result in PFJ maltracking, including movement of the patella relative to the femur, or the femur underneath the patella driven by proximal or distal biomechanical variables (Lee, Morris, & Csintalan, 2003; Witvrouw et al., 2014). Theoretically, this maltracking subsequently increases PFJ stress, initiating nociceptive firing from the densely innervated subchondral bone (Powers, 2003). Evidence related to the validity of this proposed biomechanical model is outlined in the ‘common biomechanical deficits reported in PFP’ section below. Whilst it remains plausible that this represents the primary nociceptive pathway for some individuals, the complexity of the central nervous system within which this peripheral nociception is processed, requires the clinician to be considerate of the individuals psychosocial characteristics in addition to biological theories and concepts (Maclachlan, Collins, Matthews, Hodges, & Vicenzino, 2017).
In 1999, Scott Dye and colleagues proposed a tissue homeostasis model for the development of PFP, describing how relative supraphysiological over- or under-load on innervated musculoskeletal tissues in the region of the PFJ, may result in symptoms (Dye, Stäubli, Biedert, & Vaupel, 1999) (Fig. 2). It has subsequently been described how a resultant single loading event of sufficient magnitude or a number of repeated loading events of lower magnitude may result in the loss of tissue homeostasis, at least temporarily, and subsequent increase in nociceptive firing (Dye, 2005).
Prior to and following the suggestion of a homeostasis model, a number of potential tissue derived characteristics may be present in those with PFP symptoms (Fulkerson, 1983; Ho et al., 2014; Sanchis-Alfonso et al., 1998; Schoots, Tak, Veenstra, Krebbers, & Bax, 2013; van der Heijden et al., 2016). These included alterations in the neovascularity and thickness of the lateral retinaculum, increased water content of the subchondral bone and articular cartilage stress (Farrokhi, Keyak, & Powers, 2011; Ho et al., 2014; Schoots et al., 2013). The absence of association between changes in cartilage composition and PFP symptoms continues to challenge the assumption that patient's pain can be explained by structural changes in isolation (van der Heijden et al., 2016).
Despite nociceptive activity representing the dominant mechanism of acute pain, the large number of individuals with recurrent or persistent PFP symptoms means that exploration of possible non-mechanical contributors to symptom persistence is warranted (Arendt-Nielsen, Skou, Nielsen, & Petersen, 2015). The biomedical model in isolation possesses inherent limitations within the complex paradigm of human pain perception (Gifford & Butler, 1997) and a treating clinician may need to prioritise assessment of the unique filters through which an individual's nociceptive signals pass.
In some, the non-mechanical amplification of nociceptive signalling that can occur within both the peripheral and central nervous systems (pain sensitisation), may be the predominant driver of pain perception and persistence. In both adolescents (Rathleff, Roos, Olesen, Rasmussen, & Arendt-Nielsen, 2013) and female adults (Pazzinatto et al., 2016; Rathleff, Petersen, Arendt-Nielsen, Thorborg, & Graven-Nielsen, 2015) with persistent PFP symptoms, local and widespread hyperalgesia has been reported, thought to represent altered pain processing. The activities that load the PFJ (e.g. stair negotiation, squatting, walking, running) may generate repetitive nociceptive stimuli, leading to the development and maintenance of pain sensitisation. A recent study provides support for this assumption, as greater running volume was associated with localized and remote pressure hyperalgesia and poorer self-reported knee function in female runners with PFP (Pazzinatto et al., 2017).
Elevated levels of anxiety, depression, catastrophising and fear of movement have also been reported in individuals with PFP, which have the capacity to negatively influence physical function and activity related behaviours (Maclachlan et al., 2017). These observations, in combination with reductions in pain and disability being associated with improvements in psychosocial health (Doménech, Sanchis-Alfonso, & Espejo, 2014), highlight the importance of recognising alterations in a patient's pain processing during clinical examination and integrating these findings into their eventual management plan (Maclachlan et al., 2017).
Proximal, distal and local to the PFJ, biomechanical deficits are evident in people with PFP. In some instances these deficits may exist prior to the development of symptoms, potentially representative of the primary driver of symptoms through resultant altered loading and loss of tissue homeostasis about the PFJ (Dye, 2005). To optimise treatment, individual patient deficits need to be identified and an appropriately tailored intervention programme developed and delivered.
The most frequently investigated muscle local to the PFJ is the quadriceps, with prospective level one evidence indicating PFP development is more common in individuals with reduced knee extension strength (Lankhorst, Bierma-Zeinstra, & van Middelkoop, 2012), level two evidence indicating VMO is delayed in those who develop PFP (Van Tiggelen, Cowan, Coorevits, Duvigneaud, & Witvrouw, 2009) and level one retrospective evidence reporting reduced quadriceps cross sectional area in those with symptoms (Giles, Webster, McClelland, & Cook, 2013). The prospective data is, however, limited to military populations (Boling et al., 2009; Duvigneaud, Bernard, Stevens, Witvrouw, & Van Tiggelen, 2008; Van Tiggelen et al., 2009) and it is currently unclear if quadriceps weakness or VMO activation delays increase the risk of PFP development in non-military populations. Consistent cross-sectional findings indicate decreased knee extension strength exists in various populations with PFP symptoms (Lankhorst, Bierma-Zeinstra, & van Middelkoop, 2013), including recreational runners and physically active individuals. It is important to note that the same quadriceps muscle strength deficits reported in adults with PFP have not been reported to exist in adolescents with PFP, potentially indicating exercise therapy targeting strength in this population may be less beneficial or relevant (Rathleff, Baird, et al., 2013).
Kinematically, individuals with PFP are reported to ascend stairs with reduced peak knee flexion when compared to healthy controls (Crossley, Cowan, Bennell, & McConnell, 2004; de Oliveira Silva, Briani, Pazzinatto, Ferrari, Aragão, & de Azevedo, 2015), which may be reflective of kinesiophobia, or an attempt, conscious or unconscious, to decrease symptoms through a reduction of PFJ and quadriceps loading requirements (Salsich, Brechter, & Powers, 2001).
The relationship between hip strength and symptom development is unclear, with previous prospective studies reporting either increased hip abduction strength (Herbst et al., 2015; Ramskov, Barton, Nielsen, & Rasmussen, 2015) or no association between hip abduction strength and increased risk of future PFP development (Herbst et al., 2015; Ramskov et al., 2015; Rathleff, Rathleff, Crossley, & Barton, 2014). Cross-sectional studies have reported more consistent evidence for decreased hip abduction, extension and external rotation strength (Rathleff et al., 2014), along with a delayed and shorter duration of gluteal muscle activity during functional tasks (Barton, Balachandar, Lack, & Morrissey., 2014), in people with PFP. In addition, emerging evidence indicates rate of force development in females with PFP may also be slower when compared to asymptomatic females (Nunes, Barton, & Serrão, 2017), suggesting that muscle function variables in addition to strength should be of interest to those delivering PFP rehabilitation. Consistent with quadriceps findings, hip muscle weakness is reported to exist in older (15–19 y/o) but not younger (12–16 y/o) adolescents with PFP (Rathleff, Baird, et al., 2013). This indicates proximal muscle deficits may develop as a result of PFP symptoms rather than be a causative factor. Nevertheless, considering potential effects of the hip on knee mechanics, these deficits represent a key treatment target.
Proximal kinematic factors have been reported to be associated with a greater risk of developing PFP. Specifically, increased peak hip internal rotation during a double leg drop landing was reported in a predominantly male military cohort during jump landing who developed PFP compared to those who did not (Boling et al., 2009). Additionally, increased peak hip adduction during running was reported in a group of females who developed PFP compared to those who did not (Noehren, Hamill, & Davis, 2013). This profile of greater hip adduction and internal rotation also appears to exist consistently during running once symptoms have developed (Neal, Barton, Gallie, O'Halloran, & Morrissey, 2016). Interestingly, the same link is not seen during walking, where less peak internal rotation has been reported in people with PFP (Barton, Bonanno, Levinger, & Menz, 2010; Powers, 2003). It is predominantly the female sex where associations between hip biomechanics and PFP are seen, whereas an increase in peak knee adduction has been reported in male runners with PFP compared to female runners with PFP (Willy, Manal, Witvrouw, & Davis, 2012). Overall, the lower limb biomechanical differences between the sexes during high impact tasks have not been fully investigated and further work is required in this field.
Distal to the PFJ, coupling between rearfoot eversion and tibial rotation has been proposed to influence PFJ mechanics (Fig. 3) (Tiberio, 1987), possibly driving symptom development or persistence. Specifically, a theoretical paradigm of prolonged or increased rearfoot eversion, increasing tibial internal rotation and resulting in a compensatory increase in femoral internal rotation to achieve knee extension has been proposed (Tiberio, 1987). The consequence of this coupled movement is proposed to be elevated loading between the lateral femoral condyle and lateral facet of the patella (Lee et al., 2003; Tiberio, 1987). Subsequent investigation of this proposed coupled movement in individuals with PFP has been conflicting (Barton, Levinger, Crossley, Webster, & Menz, 2012; Powers, Chen, Reischl, & Perry, 2002). Individuals with PFP symptoms are reported to demonstrate a more pronated foot posture both statically (Lankhorst et al., 2013) and quasi statically (normalised navicular drop) (Barton et al., 2010) when compared to those without pain. However, the potential impact of these findings on dynamic function is unclear (Barton, Levinger, Crossley, Webster, & Menz, 2011; McPoil & Cornwall, 1996). Prospectively, increased navicular drop has been reported as a risk factor for future symptom development in a military population (Boling et al., 2009; Neal et al., 2014), but the small difference between those who did and did not develop pain (<1mm) means the finding may be of questionable clinical relevance. Overall, the relevance of static alignment remains unclear warranting only limited attention during the assessment of the patient with PFP.
Dynamically, there does not appear to be any differences in peak rear- or forefoot motion between people with and without PFP during running (Noehren, Pohl, Sanchez, Cunningham, & Lattermann, 2012; Powers et al., 2002) or walking (Barton et al., 2011). However, during stair ascent and running, higher rearfoot eversion and a greater percentage of the available pronation range utilised in those with PFP respectively, has been reported (de Oliveira Silva, Barton, Pazzinatto, Briani, & de Azevedo, 2016; de Oliveira Silva, Briani, Pazzinatto, Ferrari, Aragão, de Albuquerque et al., 2015; Rodrigues, TenBroek, & Hamill, 2013). Plantar pressure evaluation in previous literature seems to provide a stronger link between foot function and PFP. Specifically, increased laterally directed pressure distribution at initial foot contact, shorter time to maximal pressure on the 4th metatarsal during running (Thijs, De Clercq, Roosen, & Witvrouw, 2008) and slower maximal velocity of the change in lateromedial direction of the centre of pressure during forefoot contact during walking, have been reported to be associated with high risk of PFP development (Thijs et al., 2007). The mechanistic link between these biomechanical variables and symptom development has not been established, but may be indicative of decreased foot pronation during the loading phase of gait, resulting in reduced load absorption and thus an increased transfer of forces proximally to the patellofemoral joint (Dowling et al., 2014; Neal et al., 2015).
The presentation of biomechanical deficits in people with PFP is both common and varied (Ferrari et al., 2018; Fox, Ferber, Saunders, Osis, & Bonacci, 2017). Following the assessment of proximal, local and distal kinematics during stair ascent, 52% of females presented at least two kinematic alterations and three kinematic alterations were found in 48% of the females with PFP (Fig. 4) (Ferrari et al., 2018). A higher number of kinematic alterations were strongly associated (r = 0.78) with higher levels of pain and lower functional status (Ferrari et al., 2018) and those with chronic PFP have been reported to differ throughout the kinetic chain compared with an acute PFP population and controls (Fox et al., 2017). Therefore, clinicians should carefully assess the movement patterns throughout the kinetic chain during different tasks, as identified biomechanical deficits may indicate a more severe condition and can help guide an individual specific treatment plan.
Section snippets
Assessment
We recently developed a ‘Best Practice Guide’ for managing PFP by combining systematic review findings with qualitative interviews of international experts in order to guide clinical decision-making and inform evidence based treatment delivery (Barton et al., 2015). This synthesis of evidence highlights that in order to effectively manage the multifactorial nature of PFP, consideration of multiple different intervention approaches is required (Table 1) (Barton et al., 2015). To effectively
Treatment
The development and implementation of an individually tailored treatment plan, that incorporates interventions of proven efficacy (Fig. 6), follows the detailed assessment of the individual. Whilst interventions that aim to address specific deficits are advocated, the current recommended physical interventions for the management of PFP include approaches that combine more than one treatment approach (Crossley, Stefanik, et al., 2016). A greater understanding of the potential mechanisms of
Conclusion
This clinical masterclass presents a synthesis of the current evidence relating to PFP symptom development, persistence, assessment and management. It highlights that structural, biomechanical, volume and psychological factors that can contribute to an individual's symptoms in differing amounts. Clinically reasoned paradigms have been described to integrate these features into a deficit focused, individually tailored, rehabilitation plan. The clinician is encouraged to consistently remain
Ethical approval
None.
Funding statement
No funding.
Conflict of interest
None.
References (136)
- et al.
The relationship between rearfoot, tibial and hip kinematics in individuals with patellofemoral pain syndrome
Clin Biomech (Bristol, Avon)
(2012) - et al.
Walking kinematics in individuals with patellofemoral pain syndrome: A case-control study
Gait & Posture
(2011) - et al.
Gait retraining versus foot orthoses for patellofemoral pain: A pilot randomised clinical trial
Journal of Science and Medicine in Sport
(2018) - et al.
Knee flexion during stair ambulation is altered in individuals with patellofemoral pain
Journal of Orthopaedic Research
(2004) - et al.
The mosaic of pathophysiologycausing patellofemoral pain: Therapeutic implications
Operative Techniques in Sports Medicine
(1999) - et al.
Individuals with patellofemoral pain exhibit greater patellofemoral joint stress: A finite element analysis study
Osteoarthritis and Cartilage
(2011) - et al.
Higher pain level and lower functional capacity are associated with the number of altered kinematics in women with patellofemoral pain
Gait & Posture
(2018) - et al.
Fear of movement/(re) injury in chronic pain: A psychometric assessment of the original English version of the Tampa scale for kinesiophobia (TSK)
Pain
(2007) - et al.
The integration of pain sciences into clinical practice
Journal of Hand Therapy
(1997) - et al.
Recreational runners with patellofemoral pain exhibit elevated patella water content
Magnetic Resonance in Imaging
(2014)