Targeting Treatment Distally at the Foot for Bilateral Persistent Patellofemoral Pain in a 23-Year-Old

A New Answer to an Old Problem?

Mark Matthews, Bill Vicenzino, Darren A. Rivett

Patient Interview

Ellie was a 23-year-old female who recently commenced working in a hospitality job that involved prolonged hours of standing and walking. She presented to the University of Queensland clinical Sports Injury Rehabilitation and Prevention for Health (SIRPH) research unit with a 10-year history of non-traumatic bilateral anterior knee pain symptoms, with the symptoms in the left knee more severe than the right (Fig. 10.1). Ellie had previously been a gymnast from the age of 6 years, training up to 25–35 hours per week, until the age of 12 years. She then commenced trampolining activities, training up to 6–12 hours per week, until the age of 16 years. Now Ellie worked as a bartender doing shift work for 15–20 hours per week. Outside of work, she led a sedentary lifestyle, with her hobbies including photography and laptop computer work.

Fig. 10.1 Body chart depicting Ellie’s anterior knee pain.

Symptom Behaviour

Since commencing the new job 3 months earlier, her knee symptoms had deteriorated to the extent that she now reported a dull ache at the beginning of the shift which progressed to a tense, cramping, buzzing-like feeling by the end of the shift. Her worst symptoms occurred when ascending stairs, especially after work, with pain increasing after one to two steps, up to an intensity of 5/10 on a pain numerical rating scale (NRS; 0 = no pain; 10 = worst pain imaginable) after one flight. In the previous 7 days, Ellie rated her worst pain as being 8/10 after working more than 8 hours. Her symptoms were also aggravated when sitting for longer than 90 minutes (4/10) or driving a manual car for longer than 30 minutes, which resulted in an uncomfortable ache. Colder weather caused an increase in the knee symptoms, as did a rapid change in room temperature (e.g. when walking in/out of a large refrigerator at work). Throughout the day, Ellie’s symptoms were only aggravated by activity or being in positions of knee flexion for a prolonged period of time.

Symptoms were relieved by avoiding aggravating activities, applying ice for 20 minutes after working and modifying resting knee positions. Ellie wore an elastic knee support to assist in symptom management during work. She reported audible crepitus in the left knee and to a lesser extent in the right knee, with a relieving ‘crack’ felt in the left knee at times after moving out of flexion from prolonged sitting.

Self-Report Forms

During the assessment, Ellie completed the Kujala Anterior Knee Pain Scale (Kujala et al., 1993) scoring 68/100, which indicated a severe restriction in functional abilities due to knee pain. She also completed a Patient-Specific Functional Scale (PSFS) to evaluate her ability to perform individually selected activities (scored from 0 = ‘able to do for as long as I wish’, to 10 = ‘unable to do’) (Stratford, 1995), for which she nominated the activities of walking up/down stairs (3/10), working for greater than 8 hours (5/10) and sitting for more than 1 hour (3/10).

Ellie reported she had seen her local general practitioner for her knee pain and had not undergone any investigations. This medical practitioner essentially advised that the pain would ‘go away’. She had not consulted any other healthcare professionals.

Reasoning Question:

1. Following the patient interview, and considering the chronicity of symptoms, what is your hypothesis regarding the most likely ‘pain type’ (nociceptive, peripheral neuropathic, nociplastic)? What is your reasoning process behind your decision?

Answer to Reasoning Question:

It was hypothesized that Ellie’s pain was most likely to be predominantly of nociceptive origin. Her pain only came on with loading activities of the knee, such as negotiating stairs, and with sustained knee flexion in sitting and driving, suggesting a mechanical load-related cause for her pain. These physical activities are known to particularly increase stress at the patellofemoral joint. Ellie’s report of a long history of persistent symptoms, recent deterioration with increased workloads and moderate level of symptom irritability could also suggest the presence of secondary peripheral sensitivity.

Reasoning Question:

2. Can you please discuss which features of Ellie’s reported history led you to your primary and secondary diagnostic hypotheses?

Answer to Reasoning Question:

The impression following the patient interview was a primary hypothesis of persistent patellofemoral pain, with a secondary hypothesis of fat-pad irritation. The primary hypothesis of persistent patellofemoral pain was supported by the exclusion of findings in Ellie’s history which may be indicative of other pathologies. That is, there was no history of trauma, no mention of symptoms suggestive of ligamentous instability and little likelihood of referral of symptoms from the lumbar spine or hip. Patellofemoral pain is typically aggravated by activities that load the patellofemoral joint (e.g. squatting/crouching, stair ambulation and running) or which involve sustained knee flexion (e.g. prolonged sitting), consistent with the activities that Ellie reported to be painful.

Further supporting the primary hypothesis was Ellie’s reported audible joint sounds, which is sometimes described in those with patellofemoral pain (Crossley et al., 2016a). It is thought that this noise is the result of aberrant patella motion through the trochlear groove of the femur during flexion and extension of the knee, and it may reflect the integrity of the articular cartilage (Jiang et al., 1993). It has also been suggested that audible grinding noises and/or palpable vibrations may indicate the presence of early osteoarthritic features of the patellofemoral joint on magnetic resonance imaging (MRI) in women without tibiofemoral joint changes (Schiphof et al., 2014).

The secondary hypothesis of fat-pad irritation was supported by the location and description of symptoms (anterior knee, inferior to the patella) and by the provocation of pain during dynamic activities, such as knee extension during stair ascent.

Reasoning Question:

3. It is interesting that cold environments aggravated Ellie’s symptoms, yet she indicated that she used ice for pain relief, which could appear a little contradictory. Are you able to make any comment on this? Was this a consideration in determining your hypothesis regarding the dominant ‘pain type’?

Answer to Reasoning Question:

The pain aggravation induced by cold ambient temperatures is not consistent with our hypothesis of a nociceptive ‘pain type’, but the relief of pain with ice could possibly be consistent with nociceptive pain. A study of patients with patellofemoral pain has reported that those with cold sensitivity indicate higher pain severity, tolerate less physical activity and demonstrate less improvement to lower limb stretching, vastus medialis training and patellar taping treatment (Selfe et al., 2010). Ellie’s presentation did not align well with those reported findings. Perhaps in cold environments, she might have adopted more flexed lower limb postures, which she had reported were provocative of her knee pain. However, this was not explored with her at the time, and so this is purely conjecture. Regarding her use of ice to modulate patellofemoral pain, this could be subserved by a peripheral inhibitory mechanism through cooling effects on nociceptors and small-afferent-fibre function.

Pain is seldom the result of solely peripheral or solely central pathophysiology but is more likely a combination thereof. So it is conceivable that although Ellie’s predominant pain presentation was nociceptive in nature, she could concurrently have had some central nervous system changes (sensitization) due to the long-term nature of her condition.

Clinical Reasoning Commentary:

It is a common clinical reasoning error for the practitioner to only consider the ‘positive’ or supportive clinical findings in the patient examination and to fail to give similar consideration to absent or non-supportive findings in determining likely hypotheses. This was not the case in the clinician’s response to the question of which clinical features supported the primary diagnostic hypothesis of persistent patellofemoral pain where the absence of clinical features indicative of some alternative or competing hypotheses (such as knee ligamentous pathology) was given due weighting in the reasoning process. This suggests that the clinician is actively and simultaneously considering multiple diagnostic hypotheses (tissue/structural; and/or physical impairments) and ordering these based on the presence and absence of features typically to be expected in the associated clinical patterns. Pain type cannot be measured clinically and, as discussed in Chapter 1, needs to be a hypothesis based on pain science and current understanding of expected clinical patterns (see Chapter 2). Although clinical patterns are helpful, they are often not fully validated, features can overlap with other patterns and patients will not necessarily present with every feature. This is nicely illustrated in the reasoning here, where features of a nociceptive-dominant pattern are recognized along with features of central nervous system sensitization.

Physical Examination

Observation

On observation of the lower limb in bipedal stance, the hips were internally rotated, and the feet were pronated, left greater than right. The knees were in hyperextension and appeared normal, with no apparent swelling. Based on the pronated foot posture and knee hyperextension, the Beighton Hypermobility Scale was applied (Boyle et al., 2003), with Ellie scoring 6/9 with bilateral hyperextension of the 5th metacarpophalangeal joints, elbows and knees. This score indicates the presence of generalized joint laxity (Boyle et al., 2003; van der Giessen et al., 2001). Single leg stance resulted in 3/10 retropatellar pain in the left knee only. Performing a small single knee bend on the left leg resulted in 4/10 peripatellar pain, described as an ‘ache’, at approximately 30 degrees of flexion.

Functional Tests

Each functional test was performed either until the onset of pain or performance of 25 pain-free repetitions. These tests included squats (i.e. full deep squat/full knee flexion, onto the balls of the feet, touching the floor with hands either side of the ankles), where Ellie achieved 6/25 repetitions; step-ups onto a 25-cm step at the speed of a metronome set to 96 beats/minute (7/25 repetitions on the left, 18/25 on the right); and step-downs from a 25-cm step (2/25 repetitions on the left, 3/25 on the right).

On active range-of-motion testing with overpressure at the end range, there was full pain-free active range of motion of both knees.

Knee Tests

The patella borders were tender to palpation both medially and laterally on the left, with no swelling or joint effusion present. The Hoffa test was conducted to test for fat-pad irritation (Dragoo et al., 2012). The test is designed to irritate the fat pad by applying firm pressure via the thumb inferior to the patella outside the margin of the patellar tendon with the knee in 30 degrees of knee flexion and then in full knee extension (hyperextension). The test is regarded as positive for impingement if pain is produced during the last 10 degrees of extension indicating involvement of the fat pad in the presenting symptoms (Kumar et al., 2007), although little is known about the Hoffa test’s diagnostic properties (Mace et al., 2016). The test was repeated on both the medial and lateral sides of both knees but did not reproduce Ellie’s symptoms. Further testing designed to irritate the fat pad was undertaken, which involved isometric quadriceps contraction in full extension and passive extension overpressure, again with no symptoms reproduced (Dragoo et al., 2012). There was also no pain elicited on firm palpation of the proximal, mid- or distal portions of the patella tendon.

Valgus and varus ligamentous tests of the medial and lateral collateral ligaments, respectively; anterior drawer test and Lachman’s test; posterior drawer test and sag sign; and McMurray’s and Apley’s tests were all negative for both knees, indicating that the ligamentous structures and menisci were not likely to be the source of symptoms. The patellar apprehension sign for instability was also negative. Manual compression of the patella into the trochlear groove at both 0 degrees and 20 degrees of knee flexion was positive for symptom reproduction for the left knee only. Clarke’s test was performed with Ellie lying in supine, with both knees supported in slight flexion (Nijs et al., 2006). The patella was pressed distally (with the therapist’s hand on the superior border of the patella), and she was instructed to gradually perform an isometric contraction of the quadriceps muscle (Malanga et al., 2003). This test is thought to actively compress and stress the articular surfaces of both the patella and the femoral trochlear groove. Reproduction of symptoms is regarded as a positive test and suggestive of a patellofemoral joint disorder, and whilst Ellie tested positive for both knees, this test’s diagnostic utility is questionable (Doberstein et al., 2008). Similarly, the ability of patella mobility testing to assist in diagnosis is marginal, so no assessment of patella translation mobility was conducted (Sweitzer et al., 2010).

Foot Tests

Foot posture index (Redmond et al., 2008), navicular drop (Brody, 1990) and midfoot mobility measurements (McPoil et al., 2009) were recorded. For the foot posture index, the left foot scored +7 and the right +8, indicating a pronated foot posture bilaterally (Redmond et al., 2006). Navicular drop is measured by the change in height of the navicular tuberosity relative to the floor between a subtalar neutral posture and a relaxed stance foot posture. Ellie’s navicular drop was 7 mm on the left and 9 mm on the right. Midfoot mobility is measured by recording the difference between the midfoot width in weight bearing (WB) and non-weight bearing (NWB), and is expressed as midfoot width (MFW) difference (DiffMFW = WB – NWB). Ellie’s midfoot width measurements in weight bearing were 87.7 mm on the left and 87.6 mm on the right, and in non-weight bearing were 75.6 mm on the left and 76.4 mm on the right. Thus, the DiffMFW was 12.1 mm and 11.2 mm on the left and right, respectively. Ellie’s change in midfoot width was more than the 11 mm previously reported to be associated with a greater benefit from foot orthoses intervention (Vicenzino et al., 2010; Mills et al., 2012).

Treatment Direction Test (TDT)

Given the findings on observation, foot posture and mobility testing, a Treatment Direction Test (TDT) was next applied. The TDT has been previously reported (Vicenzino, 2004), however, in brief, it involves applying a physical manipulation (e.g. anti-pronation taping in this case) during the client-specific impairment measure (e.g. pain-free step-ups on a 25-cm step with Ellie). According to Vicenzino (2004), if a significant improvement in the client-specific impairment measure is observed (i.e. ≥75% number of pain-free step-ups), then treatment of the foot with orthoses and exercises would have a high likelihood of success. Ellie achieved nine pain-free step-ups on the left (i.e. her most problematic knee) before the onset of her knee pain. After applying the anti-pronation tape (Fig. 10.2), Ellie was able to achieve 14 pain-free step-ups on the left, suggesting a high probability of a successful outcome with foot orthoses for Ellie.

Fig. 10.2 Anti-pronation taping. A, Low dye technique (just the foot taped). B, Augmented low dye technique (with the lower leg taped).

Ankle Range of Motion

Reduced ankle dorsiflexion range has been previously associated with lower limb pathologies, including an association with aberrant hip patho-mechanics in a single leg squat task in those with patellofemoral pain (Backman et al., 2011; Collins et al., 2014; Rabin et al., 2014; Ota et al., 2014). Ellie’s bent-knee ankle dorsiflexion range was measured using a modified knee-to-wall test (Larsen et al., 2016) (146 mm left and 128 mm right) and also during straight-knee ankle dorsiflexion using an inclinometer placed mid-tibia (48 degrees left and 45 degrees right).

Hip Muscle Strength Tests

Deficits in hip muscle function have been associated with altered movement patterns of the lower limb (Souza and Powers, 2009a, 2009b; Powers, 2010). Recent studies have identified reduced hip muscle strength, particularly of the hip abductors and external rotators, in people with patellofemoral pain compared with an asymptomatic group (Ireland et al., 2003; Robinson and Nee, 2007; Nakagawa et al., 2012). On the basis of this evidence, maximal voluntary isometric hip strength measurements of hip abduction, adduction and external rotation were recorded (in supine lying) using a hand-held dynamometer that was fixated by a belt (Table 10.1).

TABLE 10.1

MAXIMAL VOLUNTARY ISOMETRIC HIP MUSCLE STRENGTH SCORES AT BASELINE
	0 Weeks
	Left	Right
Abduction (N)	71.1	70.2
Adduction (N)	70.7	61.13
External rotation (N)	67.2	64.7

Reasoning Question:

4. Can you explain how the physical examination findings supported/refuted your primary diagnostic hypothesis of persistent patellofemoral pain and your secondary hypothesis of fat-pad irritation? How did your treatment hypothesis of foot orthoses fit with these findings?

Answer to Reasoning Question:

On physical examination, Ellie presented with hyperextended knees and internally rotated femurs in standing. On observation of the knees, there was no evident swelling or enlargement of the fat pad. Ellie tested negative for fat-pad irritation on palpation and on pain reproduction techniques (Hoffa test, isometric quadriceps contraction in full extension and extension overpressure), suggesting the fat pad was not the primary source of pain. Tests were also negative for other local knee pathologies (i.e. ligamentous, tendon, etc.). Most importantly, Ellie’s symptoms were reproduced with techniques that loaded and stressed the patellofemoral joint (squats, step up/down and single leg squats). Ellie also had marked tenderness on the medial and lateral borders of the patellae and symptom reproduction on Clarke’s test.

When physical examination findings were taken into consideration with her patient interview and, importantly, the exclusion of other differential diagnoses, the overall findings were indicative of Ellie having bilateral persistent patellofemoral pain. Based on the findings of pronated foot posture on the foot posture index, DiffMFW ≥11 mm, and a positive response to the TDT, it was decided that foot orthoses would be the initial treatment in managing Ellie’s patellofemoral pain.

Reasoning Question:

5. You performed a comprehensive assessment of foot biomechanics in this patient. Is this an assessment approach you take with all of the patients in your clinic with patellofemoral/knee pain, or were there features in the history and physical examination that led you to pursue that direction, rather than perhaps another approach?

Answer to Reasoning Question:

The focus on the foot assessment was based on Ellie’s report that her most provocative activity was stair climbing, a weight-bearing-under-load task, combined with the initial observation of her marked pronated foot posture. Physical examination of stair walking confirmed it provoked her pain, and correcting her foot posture with anti-pronation taping allowed the patient to perform substantially more steps. These findings led to further examination of foot posture with the foot posture index and measures of midfoot height and weight, which confirmed her feet to be more pronated than normal. If it had not been possible to reproduce Ellie’s pain on stair walking and if there had been no observable pronation of her feet, then the assessment would likely have focussed more on the knee and the hip.

Clinical Reasoning Commentary:

These responses demonstrate how the clinician has come to diagnostic and treatment decisions based on a combination of knowledge/evidence derived from prior experience with similar clinical presentations and also scientific evidence obtained from the published research. Hypotheses tentatively formulated during the patient interview have now been tested in the physical examination to determine whether expected clinical findings are indeed present, based on this previously acquired experiential and empirical data. Impairments were specifically tested to determine their relevance to key presenting symptoms (such as the correction of foot pronation on the knee pain experienced during stair walking) and were not simply assumed to be supportive of the primary structural hypothesis (persistent patellofemoral pain). Similarly, it was not assumed that competing hypotheses (e.g. fat-pad irritation, ligament pathology) were not to be accepted in conjunction with or instead of the primary hypothesis but were each specifically physically tested to ensure their exclusion at this time was appropriate. In the ‘hypothesis category’ framework presented in Chapter 1, assessment and trial correction of foot posture represents reasoning about potential ‘contributing factors’, as might the assessment of femoral posture and hip strength where trial intervention may similarly have had a positive effect. Treatment decisions were therefore based on supportive derived clinical findings and applied scientific evidence built during both the patient interview and the physical examination, as well as the absence of any convincing supportive evidence for competing hypotheses.

Only gold members can continue reading. Log In or Register to continue