Background and Purpose. Patellofemoral joint problems are the most common overuse injury of the lower extremity, and altered femoral or hip rotation may play a role in patellofemoral pain. The purpose of this case report is to describe the evaluation of and intervention for a patient with asymmetrical hip rotation and patellofemoral pain. Case Description. The patient was a 15-year-old girl with an 8-month history of anterior right knee pain, without known trauma or injury. Prior to intervention, her score on the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) was 24%. Right hip medial (internal) rotation was less than left hip medial rotation, and manual muscle testing showed weakness of the right hip internal rotator and abductor muscles. The intervention was aimed at increasing right hip medial rotation, improving right hip muscle strength (eg, the muscle force exerted by a muscle or a group of muscles to overcome a resistance), and eliminating anterior right knee pain. Outcomes. After 6 visits (14 days), passive left and right hip medial rotations were symmetrical, and her right hip internal rotator and abductor muscle grades were Good plus. Her WOMAC score was 0%. Discussion. The patient had right patellofemoral pain and an uncommon pattern of asymmetrical hip rotation, with diminished hip medial rotation and excessive hip lateral (external) rotation on the right side. The patient's outcomes suggest that femoral or hip joint asymmetry may be related to patellofemoral joint pain.
Patellofemoral joint pain is a very common knee problem physical therapists see in the clinic. Many contributing factors have been suggested as a possible cause of patellofemoral pain, including an increased Q angle, patella alta, abnormal or excessive foot pronation, quadriceps femoris (vastus medialis) muscle weakness, diminished flexibility of the hamstring and rectus femoris muscles, malalignment of the femur, and weakness of the hip.1–7
The role of the lower extremity in the development of patellofemoral pain has recently received increased attention in the literature.6,8–12 Abnormal motion of the tibia and femur in the transverse and frontal planes is believed to have an effect on patellofemoral joint mechanics and, therefore, patellofemoral pain.8,10,13 Ireland et al,11 in a recent article, and Mascal et al,14 in a case report, explained how proximal hip weakness may be related to patellofemoral pain. Powers et al15 reported that diminished femoral rotation may be related to patellofemoral pain, and Carson et al1 proposed that lower-extremity torsional malalignment is a major cause of anterior and peripatellar knee pain. James4 contended that torsional and angular malalignment of the lower extremity has an influence on patellofemoral joint mechanics. He proposed that persistent anteversion of the femoral neck or clinically apparent anteversion of the femoral neck with greater medial (internal) rotation than lateral (external) rotation may result in abnormal mechanics of the patellofemoral joint.4 Although patellofemoral joint pain has frequently been described in patients who have more medial rotation than lateral rotation and who “toe in” and have “squinting” patellae,1,4,16,17 we know of no report that has described patellofemoral pain in patients who have the opposite pattern of hip motion, that is, patients who “toe out” and whose hip lateral rotation is greater than medial rotation. The purpose of this case report is to describe the evaluation of and intervention for a patient who had patellofemoral pain and unilateral asymmetrical hip range of motion, with decreased hip medial rotation and increased hip lateral rotation.
The patient was a 15-year-old girl (height=165.1 cm [65 in], weight=54.4 kg [120 lb]) with an 8-month history of anterior right knee pain. She was referred for physical therapy with a medical diagnosis of patellofemoral pain in the right knee by her primary care physician. She reported that she knew of no injury that caused the knee pain and that the pain was around the kneecap and on the medial side. She did not describe giving way, clicking, or locking. She said that when she sat for more than 1 hour, she developed anterior knee pain (positive “movie-goers'” sign) and that the pain was relieved by extending the knee joint. The worst knee pain developed after basketball practice, when it remained for about 1 to 2 hours and then gradually diminished. Her description of pain was “a throbbing/aching pain around the right kneecap.” She gave no previous history of knee pain and reported no other previous problem in either lower extremity. She said that she regularly slept in the prone position with her right hip laterally rotated, which placed her right hip in maximum lateral rotation.
Running, jumping, and going up and down steps were difficult because of anterior knee pain. She reported that knee pain affected her ability to play basketball by slowing her down. At first, running was not painful, but after running more than 800 m, she usually developed anterior knee pain that was described as more stiffness than sharp pain. Pain also developed when she landed on her feet after jumping. She did not report change in shoes, style of play, or position. She took no medication. An anteroposterior and 30-degree knee flexion lateral radiograph of the knee was taken; however, the radiologist's report indicated no abnormality.
Tests and Measures
The patient rated her current anterior knee pain as 4 out of 10 on a 0-to-10 scale, where 0 was “no pain” and 10 was “the worst pain possible.” She rated the worst pain as 6 of 10 and the least pain as 0 of 10, usually while at rest. Her score on the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) was 24%. The WOMAC is a self-report measure designed to determine patient response on 3 different functional criteria or subsets: pain, stiffness, and physical function.18 Pain was rated with items asking how severe pain is when walking, when stair climbing, at night, while at rest, and while standing. Stiffness was scored by rating 2 items: morning stiffness and stiffness developing later in the day. Physical function was separated into items including going up and down stairs, rising from a sitting position to a standing position, bending over, walking on flat surfaces, putting on socks, and other activities of daily living tasks. Each item was scored on a scale of 0 (“no pain/problem”) to 10 (“worst pain imaginable”). A score of 24% is considered mild disability, suggesting moderate pain and some mild stiffness. On the pain subcategory, the highest reported score was for stair climbing, which the patient scored as 4 of 10, while walking was scored as 3 of 10. On the stiffness subcategory, morning stiffness was rated as 4 of 10. Finally, on the physical function subcategory, she scored going shopping and going up and down stairs as 3 of 10.
Palpation and examination of the patella.
No knee joint effusion was noted. Examination of the patellofemoral joint for static position while standing and dynamic alignment with walking was performed as described by Crossley and colleagues.19,20 The patella was assessed for glide, mediolateral tilt, anteroposterior tilt, and rotation.20 No abnormalities were noted during testing. Lateral patella tilt (“grasshopper-eye patella”) was assessed at 0, 30, and 90 degrees of knee flexion.1 No lateral tilt of the patella was noted in any position. Slight crepitus was noted when the patella was palpated during active knee motion at about 70 to 80 degrees of flexion on the symptomatic (right) side. Compression of the patellae caused pain on both sides, with pain on the right side worse than pain on the left side. Mild tenderness (2/10 on a verbal 0–10 pain scale) was noted along the right medial retinaculum, and moderate tenderness (4/10 on a verbal pain scale) was noted on the underside of the medial facet of the patella with the knee flexed 30 degrees. Apprehension tests were performed with the knee in full extension and flexed 30 degrees.1 No apprehension or pain was noted when trying to manually displace the patella laterally on either the left or the right side. No tenderness was noted along the medial or lateral knee joint lines.
Posture and alignment.
When standing in a comfortable upright stance with the feet placed shoulders' width apart, genu valgus or varus and knee hyperextension were not observed on either side. Toeing in or out also was not observed, and both patellae appeared to face forward without signs of “squinting” or facing medially.4 Both forefeet appeared normal, with no signs of varus or valgus. On the right side, a talar bulge sign (adducted talar neck)21 was observed while standing upright. A positive talar bulge is a clinical sign of the talar head protruding excessively on the medial side of the foot with the patient weight bearing.21 On finding a positive right talar bulge test, the navicular drop test was performed to determine whether the right medial longitudinal arch was diminished (pes planus) when compared with the left medial longitudinal arch. The navicular drop test measures the difference in height of the navicular bone from the floor to the most lateral prominence of the navicular bone.22 No difference was found when comparing the left and right sides. Patella height was measured to assess for patella alta or baja using the Insall-Salvati method on the right side. The Insall-Salvati method, briefly described, involves taking a measurement from a lateral radiograph of the knee and measures the length of the patella and the patellar tendon. Dividing patellar length by patellar tendon length gives the Insall-Salvati ratio.23 An Insall-Salvati ratio that is less than 1.2 is considered “normal,” and the patient's ratio was 1.0. Some authors,23,24 however, have suggested that this measure is not accurate until skeletal maturity because of difficulty in accurately measuring the patella until complete ossification.
Observation of gait.
Gait was observed while the patient walked 10 m at a self-selected pace on a level surface. When the patient was observed from the rear, the only deviation noted was that the right foot appeared to toe-out during early stance, during midstance, and during swing when compared with the left side. The patient appeared to have normal sagittal-plane motions in the hip, knee, and ankle joints. No other abnormalities were noted.
Range of motion.
Active range of motion of the right knee was full, with 0 to 140 degrees of knee motion. Passive range of motion of the right knee also was 0 to 140 degrees, with no pain at the end of the range. Muscle length tests of the 2 joint hip muscles using the Ober and Thomas tests showed no difference between the left or right rectus femoris muscles, iliotibial band, or hip flexor muscles. Hamstring muscle length assessed by flexing the hip to 90 degrees and passively extending the knee showed no left to right difference. The left and right knees could passively extend to 0 degrees.
Passive hip medial rotation and lateral rotation were measured with the patient in a prone position with the knees flexed to 90 degrees and the pelvis firmly stabilized with one hand. The other hand grasped the ankle and moved until a firm resistance was felt. Another physical therapist measured hip rotation with a 30.48-cm (12-in) plastic goniometer as described by Staheli et al25 and Ellison et al26 while the hip was moved to the end of hip medial and lateral rotation. The reliability of measurements of hip rotation obtained with this method has been estimated to be high (intraclass correlation coefficient [2,1] of .96 for interrater reliability and intraclass correlation coefficient [3,1] of .96 for intrarater reliability).26 Passive medial rotation of the left hip measured 55 degrees, and lateral rotation of the left hip measured 55 degrees. Passive medial rotation of the right hip measured 35 degrees, and lateral rotation of the right hip measured 75 degrees.
Ligament testing was not performed because the patient gave no history of trauma or injury.27 Ankle joint range of motion, specifically dorsiflexion, for its possible contribution in creating subtalar joint pronation, was measured. The left and right ankle joints had 15 degrees of dorsiflexion with the knee fully extended.
Manual muscle testing was performed according to the method of Kendall et al.28 Florence et al29 estimated intrarater reliability of manual muscle test scores for individual muscles to range from κw=.80 to κw=.99 (Cohen weighted kappa) in patients with neuromuscular impairments. Bohannon30 found that manual muscle test grades correlated well with handheld dynamometry scores in 50 patients, suggesting that the methods measure the same variable. The patient's right hip internal rotator and abductor muscles were both graded Good minus. The hip internal and external rotator muscles were tested with the leg vertically at 0 degrees in the prone position. No weakness was found with manual muscle testing of the right knee extensor muscles with the knee held at 0 degrees and at 30 degrees of knee flexion (Normal muscle grade). We did not estimate the reliability of our measurements.
Diagnosis and Prognosis
The physical therapist diagnosis according to the Guide to Physical Therapist Practice31 was impaired joint mobility, muscle performance, and range of motion associated with ligament or other connective tissue disorders (pattern E). We believe the prognosis for this patient was excellent given the patient's young age, no history of trauma, no previous history of knee pain (first episode), and no major impairments (diminished active or passive range of motion of the knee of more than 10°, knee joint effusion, or significant muscle weakness). A search of the literature revealed that no studies have been performed on the prognosis of patellofemoral pain in patients.
The intervention goals were aimed at restoring medial rotation of the right hip, improving right hip muscle strength (muscle grades), and eliminating anterior right knee pain. Strengthening exercises for the right hip abductor and internal rotators were performed while standing using a hip exercise machine (Fig. 1). Five sets of 12 repetitions starting with 9.07 kg (20 lb) and increasing resistance as tolerated with the goals of creating right hip muscle fatigue with only minor pain (2/10) were performed every other day on Monday, Wednesday, and Friday for a total of 6 sessions. The first week the patient stayed with 9.07 kg of resistance, and the second week she advanced to 13.6 kg (30 lb) of resistance; occasionally, she was urged to increase her repetitions to 15 or 20 to fatigue the muscles by the end of the workout. The goals of the exercises were to increase medial rotation of the right hip and to improve internal rotator and abductor muscle strength. The primary internal rotator muscles of the hip area abduct the hip.28 A study by Pare et al32 showed that the internal rotator muscles showed considerably more electromyographic activity during abduction when the foot was toed in. For this reason, exercises to strengthen the hip internal rotator muscles (ie, tensor fasciae latae, gluteus minimus, and anterior fibers of the gluteus medius)28 were done with the patient's hip abducted while trying to hold the leg and foot in a position of medial rotation.
Gentle stretching of the right hip in the direction of medial rotation while the patient was lying prone (with the hip in the anatomically neutral position and the knee flexed to 90°) was performed 6 times using a contract-relax method followed by holding of 3 stretches for 30 seconds at the end of the range. The stretch was performed by stabilizing the pelvis with either a therapy belt or gait belt or by firmly holding the left posterior superior iliac spine while slowly rotating the femur medially by grasping the ankle or foot with the knee flexed to 90 degrees (Fig. 2). The patient regularly slept in the prone position with her right hip laterally rotated, which placed her right hip in maximum lateral rotation. The patient was instructed not to sleep this way and was encouraged to use a pillow between her thighs to try to promote a neutral sleeping posture.
After 6 visits over 14 days, passive medial rotation of the right hip measured 60 degrees (Table). The patient had made an improvement of 25 degrees in right hip medial rotation. Internal rotator muscle strength was tested with the patient in a prone position with the leg vertically aligned and was graded Good plus. Lateral rotation on the right side decreased by 20 degrees following the intervention. Quite often when range of motion is gained in one direction, we have found a decrease in the opposite direction. Her WOMAC score was 0%, 0-to-10 pain scales (now, worst, and best) all scored 0/10, and she had no complaint of pain with activity.
Six months later, the patient came for physical therapy for a shoulder strain, and at that time she reported having no anterior knee pain or any further episodes of knee pain. When examined, her right and left hip medial rotation and lateral rotation were symmetrical, and manual muscle test grades of the right hip internal rotators and abductors were Normal.
The purpose of this case report was to describe the physical therapy management of a patient who had an unusual pattern of hip rotation asymmetry and patellofemoral pain. Patients with patellofemoral pain often have signs of lower-extremity torsional malalignment,1,4 usually femoral neck anteversion accompanied by greater medial rotation than lateral rotation.33–38 Our patient was unusual in that she had a unilateral pattern of hip rotation, with hip lateral rotation greater than hip medial rotation. This pattern of hip rotation often is found in patients with femoral neck retroversion.17,34,35,39,40
Our intervention program emphasized restoring hip joint rotation symmetry and strengthening the hip abductor and hip internal rotator muscles. Powers8 suggested that abnormal motion of the tibia and femur in the transverse and frontal planes has an effect on patellofemoral joint mechanics and, therefore, patellofemoral pain. Powers et al41 also found that patients with patellofemoral pain have less femoral medial rotation than control subjects during the early stance phase of gait. Powers et al speculated that patients with patellofemoral pain may laterally rotate their femur as a compensatory strategy to reduce the Q angle. Alternatively, we wonder whether limited hip medial rotation may be a feature of some patients with patellofemoral pain, because we also found limited hip medial rotation during the examination of our patient. However, we did not examine hip rotation during gait. Further studies are needed to assess whether and how hip rotation is altered in patients with patellofemoral pain.
A number of published articles8,9,11,14,42–46 have examined the relationship between altered femoral rotation and patellofemoral or knee joint pain. Lee and colleagues9,10 conducted perhaps the most cogent study that demonstrated the importance of femoral rotation in the development of patellofemoral pain. They showed how altered fixed femoral rotation (eg, anteversion or retroversion) can change the contact pressure on the medial and lateral articular facets of the patella.10 Higher peak patellofemoral contact pressure develops on the medial facet of the patella with a 30-degree lateral rotation deformity (retroversion)10 and on the lateral facet of the patella with a 30-degree medial rotation deformity (anteversion). The study lends support to the notion that femoral rotation asymmetries may be associated with patellofemoral pain.10 Although we did not examine our patient for femoral neck retroversion, our patient had a range-of-motion pattern that is often found in patients with femoral neck retroversion (30° greater hip lateral rotation than medial rotation).
We do not know how our patient developed this pattern of unilateral hip asymmetry. Her history suggests that a sleeping posture may have contributed. Crane16 reported that sleeping or sitting postures in which the hip is held at the end of range can be a cause of asymmetrical hip rotation. We hypothesize that sleeping with her hip in a laterally rotated posture may have contributed to the asymmetry in right hip rotation over time. Other authors,47,48 however, have suggested that people assume end-range postures because they feel comfortable that they already have the necessary range of motion. Although this explanation is plausible, it does not explain how or why our patient developed her specific pattern of asymmetrical hip range of motion.
The improvement in muscle grades in such a short time in this case was remarkable. Perhaps the quick improvement was made possible by diminishing the patient's hip pain and restoring her hip medial rotation. Despite the initial finding of weak internal rotator muscles, we believe that the weakness could have been attributed to the manual muscle testing position. On initial evaluation, lateral rotation of the right hip measured 75 degrees, and medial rotation of the right hip measured 35 degrees. This test position, with the leg vertical, would have placed the internal rotator muscles in a shortened position. According to the length-tension curve, shortening the length of a muscle would put the muscle in a less favorable test position. A muscle tested in a shortened length position could exhibit an apparent muscle weakness simply because of the test position. We hypothesize that after increasing hip medial rotation, this same test position with her leg vertical would place her internal rotator muscles in a more favorable midrange position along the length-tension curve.
After the intervention, we also noticed that her hip range of motion on the right side was now symmetrical when comparing medial and lateral hip rotation. Before the intervention, the patient had diminished medial rotation and increased lateral rotation (Table). We believe that the increase in range of medial hip rotation came from stretching the hip; however, we do not know how or why lateral hip rotation decreased during the intervention. We previously noticed this phenomenon in other patients. We speculate that the improved symmetry in hip rotation was a result of hip muscle strengthening, stretching, and elimination of the extreme lateral rotation hip posture while sleeping.
Future research to examine the role of femoral rotation in patients with patellofemoral pain is needed. Studies could lead to new physical therapy interventions for patients with patellofemoral pain. Lee et al10 have shown that abnormal femoral rotation, whether increased medial or lateral rotation, plays a role in altering the mechanics of the patellofemoral joint and in the development of patellofemoral pain. We believe that future studies to examine the role of hip joint rotation asymmetries in patients with patellofemoral pain are important to undertake.
Both authors provided writing and consultation (including review of manuscript before submission).
- Received February 5, 2005.
- Accepted April 11, 2005.
- Physical Therapy